Turning talk into text: the representation of contemporary urban vernaculars in Swedish fiction

2.1 Swedish contemporary urban vernaculars

In Sweden, migration began to increase rapidly in the second half of the 20th century, during which time the country also underwent a rapid process of urbanization. This led to growing housing shortages in metropolitan areas (e.g., Hall and Vidén 2005). In 1965, the Social Democratic government therefore decided to launch a large-scale national housing program, which came to be popularly known as the “Million Homes Program,” because of its mission of building a million new homes over the course of ten years. The label is still used today to refer to the housing areas that were built during this era, often in the urban peripheries, with ultramodern high-rise buildings as their trademark (e.g., Ericsson et al. 2000). Over time, due to the housing policies and practices in Sweden, these areas have become increasingly migrant-rich and class-stratified (Hübinette and Lundström 2020). In the media, they are often stereotypically associated with social problems and used as a projection screen for criticizing migration (e.g., Milani 2020).

In the 1980s, the linguist Kotsinas (1988; see also 1994) began to investigate the language practices of youths growing up in Rinkeby, a Million Homes area in northwest Stockholm with a large migrant population. She was mainly interested in their use of certain innovative, non-standard language features. Two decades later, a large-scale research program was launched aiming to gain a deeper understanding of the typical ways youths from these areas used Swedish (e.g., Källström and Lindberg 2011). Around the same time, similar projects also emerged in other parts of Europe (for an overview see, e.g., Kern and Selting 2011; Nortier and Svendsen 2015; Quist and Svendsen 2010).

The early research mainly focused on describing linguistic features perceived as “typical” of the spoken registers of youths from these settings, for example their frequent use of certain lexical register emblems, such as slang words and expressions originating in different migrant languages, their use of the nominal and verbal suffix –ish (e.g., “gärish” [‘girl-ish’], “händish” [‘happens-ish’]), as well as extensive use of certain discourse markers (e.g., “du vet,” [‘you know’]) and (masculine) vocative address terms (e.g., Bijvoet and Senter 2021; Ganuza et al. 2020). Foreign-accented speech and XSV word order (in contexts requiring subject-verb inversion in standard Swedish) were also described as “typical” of Swedish CUV. Furthermore, perception studies and attitudinal studies found that the use of emblematic CUV register items was negatively evaluated by most listener groups and indexically linked to perceptions of youth, toughness, and masculinity (for an overview, see Bijvoet and Senter 2021).

In 2006, the former Swedish magazine, Gringo, tried to launch the label “miljonsvenska” (lit. ‘Million Swedish’) to refer to Swedish CUV, in order to emphasize the geographical boundedness of CUV to the Million Homes areas. This attempt gave rise to a heated debate in one of Sweden’s larger daily newspapers (e.g., Milani 2010), but the label never really caught on.

While the linguistic practices associated with CUV have mainly been viewed by researchers as signs of creativity and innovation, they have often been linked to discourses of deficit in other contexts. In the Swedish media, CUV are often positioned in opposition to standard Swedish, being described as “bad,” “deviant” and “worrisome” language practices, indicative of deficits in Swedish (e.g., Stroud 2004). Million Homes areas and the use of CUV have also frequently been linked to stories about criminality, sexism and educational failure (e.g., Jonsson et al. 2020; Milani 2010; Milani and Jonsson 2011). In these accounts, CUV tend to function as icons of ethnic otherness, which are associated with various sorts of social problems.

Nevertheless, the Million Homes areas and CUV have also attracted increased attention and interest over the years, not least in popular culture. At the turn of the twenty-first century, the Swedish literary field was eagerly awaiting new authors who would be able to narrate stories about Sweden as a multicultural nation-state and depict the “new” ways of speaking Swedish associated with the multilingual urban settings (Nilsson 2010; see also Beschnitt 2013; Källström 2005, 2010). Debut authors such as Leiva Wenger (2001), Khemiri (2003), and Bakhtiari (2005) published their first novels during this time. They were commercially successful, won high critical acclaim, and their innovative use of language was lauded by many critics.

2.2 Literary representation

In this article, we focus on the literary representation of certain spoken registers. Even if the authors might be striving for orality and authenticity, we view their representations as selected and constructed images of language use (see also Ganuza and Rydell 2023). In comparison to spontaneous oral communication, literary representations, for example, are always more ordered and tidied up (Bublitz 2017: 238). Authors tend to select only a subset of the features that they associate with orality and with specific registers, and written language is rather underdeveloped in its capacity to represent the subtle differences that can exist between registers (Bublitz 2017; Bucholtz 2000). In agreement with Lopez and Bucholtz (2017: 4), we view authenticity as a “jointly produced semiotic effect,” that is, as the result of an interplay between the author’s attempt at depicting language in a realistic way and the audience’s presumed ratification of it as successful.

Here, we are interested in identifying the linguistic features that are employed to evoke the idea of Swedish CUV, and how these add to the creation of symbolic authenticity of the novels. For instance, the narratives of all four selected novels are grounded in explicit and specific configurations of time and space, which are anchored in the biographical time of the protagonists. The novels also use first-person narration and are largely dialogue-driven (see Section 3). All of these aspects increase the perception of the novels reflecting reality in a realistic way (Bakhtin 1981). Furthermore, both the dialogue and the narration are characterized by extensive use of colloquial language, including certain features associated with orality, which contributes to blurring the division between the dialogue and the narration, and to aligning the narrator with the fictional characters’ perspectives (Djenar and Ewing 2015: 110). Källström (2010: 150) describes these strategies as typical of literary performances aimed at indexing realism and authenticity.

2.3 Register formation

Agha (2007) views register formation as a result of repeated co-occurrence of linguistic elements and patterns, which are paired with certain social practices, roles, places and value attributions. He uses the notion of enregisterment to describe the processes through which “performable signs,” for example certain linguistic features, come to be grouped together and recognized “as belonging to distinct, differentially valorized semiotic registers by a population” (Agha 2007: 81). Through repeated practice, a register can thus become intimately tied to a particular social persona, or “characterological figure” (Agha 2003). The latter refers to the idea of an “imagined typical user” of the register, which is linked to value-laden ways of being and acting (Agha 2007; see also Ilbury 2019).

According to Agha (2003, 2007), mass media and literature play important roles in the production, circulation, uptake and resignification of registers, and in making characterological figures recognizable to larger audiences. Jonsson et al. (2020) have, for example, illustrated how the young male CUV speaker involved in criminality has become a popular and widely recognized characterological figure in Swedish popular culture. For authors, depicting specific linguistic traits can be an efficient way to build up characters and make them identifiable to readers (see also Ganuza and Rydell 2023). Through the characters’ and the narrator’s metapragmatic comments about language, the author can also influence how readers interpret the depicted language practices. Moreover, in fictional accounts, registers often acquire important social meanings by being contrasted with other registers (Androutsopoulos 2010, 2012). The reader’s ability to recognize registers and their semiotic displays and differences, however, also depends on prior familiarity (Agha 2007: 15).

Agha (2007) also speaks of indexical icons, which arise when individual linguistic features or behaviors come to be perceived as individual semiotic signs of a register. At this point, it may be sufficient to deploy one or only a few icons to evoke a range of indexicalities linked to that register (see also Ilbury 2019). In spoken language, this is evident, for example, when speakers use stylized language for symbolic and rhetorical effect. Rampton (2009: 149) defines stylization as

reflexive communicative action in which speakers produce specially marked and often exaggerated representations of languages, dialects and styles that lie outside their own habitual repertoire.

In the fictional accounts studied here, it is interesting to examine how spoken language is represented in writing so that it will be perceived as authentic and part of the character’s own language repertoire, in contrast to how it is depicted when not forming part of the character’s habitual language. Moreover, as mentioned in the introduction, analyses of language variation in fiction need to consider not only the linguistic features that are employed, but also the available linguistic options that the authors choose not to exploit, and the possible reasons why this is so (Simpson 1993: 8).

4.1 Indexical register icons

The protagonists of all four novels are depicted as frequent users of linguistic features that have typically been associated with spoken Swedish CUV (see Section 2.1). This adds to the novels’ sense of location in the Million Homes areas, and to the impression that they represent reality in a realistic way. In particular, the characters are depicted as using certain CUV lexical indexical register icons (Agha 2007). For example, all of them use slang words and expressions originating in different migrant languages (for example, in Turkish, Arabic, Bosnian/Croatian/Serbian and Spanish, which are among the larger migrant languages spoken in Sweden). Example (1), from the novel SPS, features a short dialogue between the protagonist Carlos and his friend Vic. It illustrates how both characters employ CUV lexical icons in the depicted interaction, such as “bazz” [‘fuck’], “siktelan” [‘fuck off’], “guzz” [‘girl, chick’], and “bax(a)” [‘steal’, ‘hustle’, ‘snitch’].

The protagonists in all four novels are also depicted as using the nominal masculine vocatives “bror,”/“brorsan”/“brosh” [∼ ‘bro,’ ‘brother,’ ‘bruh,’ different variants are preferred in the different novels], and “mannen” [∼ ‘man’]. Abundant use of these vocatives has been associated with Swedish CUV, and are sometimes used in humorous stylizations of CUV speakers in popular cultural outputs (e.g., Jonsson et al. 2020). They are often described as typical of spoken youth language, and important for creating rapport between speakers (e.g., Smalley 2015). The vocatives are particularly common in SPS and FTL, where they are used exclusively for depicting interactions between the male characters. One could accordingly argue that they contribute to reinforce the discursively constructed link between CUV and masculinity (e.g., Milani and Jonsson 2011), which is also reinforced by the thematic content of the depicted interactions, as in Example (1).

In IDB, however, “bror” is also used in dialogues between the female characters. The protagonists of SPS and FTL are furthermore depicted as frequently using the discourse marker “du vet” [‘you know’], as well as the nominal and verbal suffix –ish (e.g., “beknish,” [‘push-ish’], FTL p. 163), both of which have been linked to Swedish CUV (e.g., Ganuza et al. 2020). The protagonist in SPS is also depicted as using the first-person pronoun “benim” on several occasions, a loan word originally from Turkish that has been described as a new ego-honorific pronoun in Swedish CUV (Young 2021). It occurs once in Mizeria as well.

The CUV lexical indexical register icons are used by the protagonists of all four novels, and by most other characters of the same age and social background. They are depicted as being mutually intelligible for these characters. They are not generally used by the older characters, such as the protagonists’ parents. Hence, their use is indexically linked to youthfulness and in-group membership, and they are important for the construction of the characterological figure of the young person from the Million Homes areas. Importantly, the lexical register icons appear in dialogues and inner monologues, and sometimes also in the narration of the stories. They are not visually marked as non-standard, nor are they generally commented on. That is to say, they are depicted as forming part of the young characters’ authentic voices (Rampton 2009) and their everyday Swedish.

Words and expressions of English origin are relatively common in the four novels as well. Contributions in English are also neither visually marked nor commented on. Hence, they are similarly depicted as forming part of the fictional characters’ everyday Swedish (see Example (2)).

In general, no explanations are given of colloquial slang or words and phrases in English, indicating that readers are expected to be familiar with them, or to be able to follow the narrative in spite of them.

FTL differs in this regard, since it includes a two-page slang glossary at the end of the novel. The glossary implies that the author (or publisher) suspects that some readers might need the support of a word list to be able to follow the story. It constructs the language used in the novel as being in need of translation. Interestingly, the glossary contains only a selection of the slang words used in the novel. It does not contain several of the most commonly used slang words that have previously been associated with Swedish CUV, such as “guzz,” [‘girl’], “keff” [‘bad’], “lack” [‘angry’] and “chilla” [‘take it easy’], perhaps because these are common in Swedish popular culture today and are also used by many individuals who do not come from Million Homes areas.^[5] The glossary also omits older Swedish slang words that frequently appear in the novel, like “gola” [‘snitch’], “knega” [‘work’], or “polare,” [‘buddy], and slang words of English origin, like “bitch,” “cash,” and “stash.” Above all, the glossary contains a variety of swear words (many related to sex), and street terms for drugs and drug dealing. Hence, the glossary has been created with a particular reader in mind, one who is familiar with some sorts of slang but not others. Through its selection of words and expressions, one could argue that the glossary contributes to create a rather stereotypical indexical link between CUV, sexism and criminality.

Word order variation is another stylistic resource employed strategically in three of the four novels. As mentioned in Section 2.1, the use of XSV has been described as emblematic of spoken Swedish CUV. Like other Germanic languages except for English, Swedish is often described as a “verb-second” (V2) language, due to its preference for placing the finite verb in the second position of declarative clauses. Accordingly, when a declarative clause begins with something other than the subject, subject-verb inversion typically occurs (Ex. “Då sprang dom” (XVS), lit. ‘Then ran they’). It is also used to distinguish between declarative and interrogative clauses. In the Comprehensive Grammar of Modern Swedish (SAG 1999), the V2 rule is presented as robust, with only a few known exceptions.

Research on spoken Swedish CUV has shown, however, that Swedish allows for more word order variation than the descriptive grammar acknowledges (e.g., Ganuza 2008). In fact, some speakers from the multilingual urban settings produce relatively many examples of XSV (e.g., “Då dom sprang” (XSV) lit. ‘Then they ran’), at least in their everyday Swedish with peers of the same age and social background (e.g., Ganuza 2008; see also Senter 2022). Moreover, research on Swedish second-language acquisition has shown that XSV is also a common feature of Swedish learner language (e.g., Ganuza 2008).

The four novels vary with respect to whether or not, and how, they employ XSV. In Mizeria, XSV does not occur at all. In SPS, the main protagonist Carlos is depicted as a frequent user of CUV lexical register icons, but is never depicted as using XSV. Instead, XSV is reserved for representing adult second-language speaker styles (see Section 4.3). In IDB and FTL, by contrast, XSV is used frequently in the representation of the protagonists’ language. Both Samira, in IDB, and Khosro and Morad, in FTL, are depicted as switching between XVS and XSV throughout the novels, as do all of their close friends. Examples (3a)–(3c) illustrate how Morad switches between XSV and XVS in the same paragraph.

In both FTL and IDB, the characters produce standard XVS more often than XSV, and XVS is found in all contexts. XSV occurs mainly in dialogues, and sometimes in the narration of the protagonists’ inner thoughts. It rarely occurs in descriptive narration, however.

Descriptions of spoken Swedish CUV often include impressionistic observations of a distinctive “suburban-sounding”^[6] staccato rhythm, which indexically links the speaker to the urban peripheries (Bijvoet and Senter 2021; Young 2022). In the four novels, however, pronunciation is rarely used to characterize the language of the speakers who are portrayed as having grown up in Sweden in the Million Homes areas. Their pronunciation is not visually marked as non-standard, and is seldom remarked on explicitly (the exception being a few occasions in SPS when a minor character is described as having “a real-Swede accent” (p. 103), “a ghetto-sounding accent” (p. 221), or “an exaggeratedly Swedish accent” (p. 259)). Instead, marked pronunciation, and explicit comments about pronunciation, tend to be reserved for the characters portrayed as immigrants and as adult second-language speakers of Swedish (see Section 4.3).

4.2 Metapragmatic comments about language

The narrator’s and the fictional characters’ metapragmatic comments about language are important dimensions in the process of attributing meaning and value to the depicted language practices, and in getting readers to pay attention to them and interpret them in certain ways (e.g., Agha 2007). Explicit comments about language are, however, relatively infrequent in the four novels (which stands in contrast to what we have found in the larger corpus, Ganuza and Rydell 2023). When they do occur, they rarely concern the protagonists’ ways of using Swedish, but rather focus on other, minor characters’ ways of speaking (see Section 4.3).

One exception is found in SPS, when Carlos moves with his family to Jordbro, after having lived in a mid-sized Swedish town for a couple of years. As he settles into Jordbro, he is depicted as going through a transformation that affects the way he talks, dresses and acts (see Example (4)).

This example evokes several common stereotypes about individuals from the Million Homes areas (not speaking “clean-sounding Swedish,” dressing like “true artists,” and being hustlers) – that is, it calls upon a specific characterological figure, which contributes to enregistering ways of speaking, being and dressing associated with that figure (Agha 2007). This new characterological figure is built by way of contrast with Carlos’s former self. Importantly, Carlos also seems to take pride in his quick transformation, that is, it is imbued with mostly positive connotations.

Another example occurs in FTL, when the protagonist Khosro starts dating a girl who lives with her family in the city center. She dislikes him hanging out with his criminal friends, and she sometimes remarks on his Swedish. For instance, she complains about his frequent use of “abuu,” a multifunctional word typically associated with CUV, which she claims makes him sound “so ghetto” (see Example (5)).

Importantly, the girlfriend’s way of pronouncing the word is depicted differently than Khosro’s (abuu vs aboo), and later in the dialogue, Khosro even corrects her pronunciation. This marks her contribution as less authentic than his, and conveys to the reader that while “aboo” forms part of Khosro’s habitual repertoire, it does not belong to his girlfriend’s everyday Swedish. The distinction also contributes to accentuating the difference between the two characters in terms of one being from the suburbs and the other from the city center.

At one point in IDB, the protagonist Samira and her friend Amina are in a meeting with their school principal. Unlike in the rest of the text, the way they address the principal is visually marked in italics (see Example (6)), and the first-person narrator remarks that the girls address the principal as “svennigt” [‘Swedishly’] as they can.

This contrasts with how the characters are depicted as speaking elsewhere. The use of italics, together with the remark and choice of lexis, marks this as an instance of stylization, as a performance by which Samira and Amina put on a voice of another (Rampton 2009). The depicted interaction showcases their ability to adjust their language to accommodate the principal, but also illustrates how they distance themselves from the voice of the principal, and from how they speak more standard-like in interaction with her. The contrast with standard Swedish in this example thus mainly serves to attribute positive value to their everyday Swedish.

4.3 Contrasts with adult second-language speaker styles

In general, pronunciation is rarely used to characterize the protagonists’ ways of speaking in the four novels. By contrast, the pronunciation of the characters who are portrayed as immigrants and as adult second-language speakers of Swedish is often visually marked as non-standard, with the use of unconventional spelling to suggest their variant pronunciations (see (7a)–(7c) from the novel SPS).

The examples in (7) typify aspects of Swedish pronunciation that are well known to be difficult for adult second-language learners (e.g., Abrahamsson 2009), such as the minimal contrast between the vowels “y” and “i” (Example (7a): “mikke” instead of “mycket”), and “ö” and “u” (Example (7c): “buuurja” instead of “börja”), the minimal contrast between the consonants “v” and “b” (Example (7b): “barfö” instead of “varför”), as well as the difference in duration between Swedish short and long vowels (Example (7b): “hellatidden” instead of “hela tiden”). Partly different unconventional spellings are used with the different characters, thus indexing their different foreign accents. For example, the character in (7b) is depicted as using final consonant deletion, while the character in (7c) uses elongated vowels. This impression is also reinforced by the characters’ names (Vlatko, Bafetimbi Faal, and Stanislav) and the narrator’s comments about their origins (“Yugo,” “Gambian,” and “Russian,” respectively).

Representations of adult Swedish second-language speaker styles are most frequent in the novels SPS and FTL. In addition to visual representation of the characters’ foreign-accented speech, the narrator of SPS repeatedly comments on their (deficient) Swedish: “despite his bad Swedish” (p. 55), “in his Yugo accent” (p. 113), “his Swedish got even worse” (p. 114), “He hardly spoke Swedish” (p. 188), “with his peculiar accent” (p. 209). This stands in contrast to the general lack of commentary about the protagonist’s Swedish, except when he first moves to Jordbro (see Example (4)).

In addition to foreign accents, the characters who are positioned as second-language speakers of Swedish are often represented as using simplified morphology and syntax and non-idiomatic expressions. This is visible, for example, in FTL when Khosro’s mother is depicted as speaking Swedish on the phone (see Example (8)). It contrasts with how she is depicted as speaking when using Farsi, which is textually represented by standard Swedish together with a remark that it is uttered in Farsi (a commonly used strategy for representing multilingualism in literature, e.g., Tidigs 2014).

In Example (8), the mother’s incompetence in Swedish is reinforced by the image of Khosro’s younger siblings roaring with laughter at her “SFI Swedish” (SFI is an acronym for “Swedish for Immigrants,” and refers to classes in basic Swedish for adult second-language learners). The comical framing of the mother’s use of Swedish is an image that recurs in several other novels in our larger study. That is, adult second-language speakers are often represented as comical and laughable.

The marking of the characters’ Swedish as non-standard, the remarks about their deficient Swedish, and the comical framing of their use of the language are thus semiotic resources employed to construct the characterological figure of the adult second-language speaker. This contrasts with the construction of the characterological figure of the young CUV speaker, who is depicted as competent, cool and street-wise. The juxtaposition of the two characterological figures contributes to indexing the CUV speaker character with positive values.

It is mainly secondary and minor characters who are positioned as second-language speakers of Swedish, often the protagonists’ parents. These characters are never central to the storyline, but they represent an important linguistic contrast to the protagonists. While the parents, and other adult migrants, are depicted as struggling to learn and use Swedish, and as being comical when they attempt to use it, the protagonists are mostly depicted as competent with respect to Swedish. For example, in SPS, the narrator explicitly says that whereas Carlos’ parents struggled to learn Swedish, he acquired it easily, becoming fluent in less than a year (p. 165). In Mizeria, the parents are described as always having to rely on their children to communicate with Swedish authorities, and therefore feeling inferior and outdone by them. Struggling with Swedish is thus enregistered as part and parcel of the characterological figure of the adult second-language speaker, but not with that of the young CUV speaker.