Keywords

10.1 Introduction

This chapter sheds light on the language use of Sweden Finns, Finland Swedes, Finns in Estonia, and Finnish L2 speakers of English. More specifically, we examine how representatives of these groups conceptualise and talk about the world around them using what we call mixed compound nouns—complex nouns consisting of two or more words taken from different languages. One example of a mixed compound in our data is the word bilkatsastus (‘car inspection’), where the head noun katsastus (‘inspection’) is in Finnish and the modifier bil (‘car’) is in Swedish. Mixed compounds are also known as hybrids or hybrid compounds (Chrystal, 1988; Graedler, 1998; Mickwitz, 2010).

Compounds are notoriously difficult to define linguistically (Lieber & Štekauer, 2009), but one crucial property is the combination of lexemes into larger words (Booij, 2005, p. 75). In this chapter, we concentrate on endocentric mixed compound nouns. Like in the example above, endocentric compounds consist of two nouns: a modifying noun and a head noun (in this order in the languages concerned in this chapter). More specifically, endocentric compound nouns are word constructions ‘in which the compound as a whole is a hyponym of its head element’ (Lieber, 2009, p. 89). To illustrate this with the example of bilkatsastus (‘car inspection’) above, there can be many kinds of inspections, but by adding a modifier and using the compound noun car inspection, the meaning of the hyperonymous or superordinate word inspection is narrowed down and specified as a certain kind of inspection, the hyponymous or subordinate word car inspection. Creating a (mixed) compound can be a random, ad hoc phenomenon, but overall, compounding is useful, as it is one of the most prolific ways of creating fresh expressions and neologisms.

Compounding is thus one of the main strategies for creating words for new concepts in a language (cf. Graedler, 1998, p. 201). Compounding is a productive way to form new words also in Swedish and Finnish. To take an example, there are 36 words in a list of Swedish neologisms, Nyordslistan 2021, compiled by The Language Council of Sweden and Språktidningen, a popular Swedish-language magazine (see Institute for Language & Folklore, 2021). 34 words out of the 36 on the list are nouns. Furthermore, 26 words on the list are compounds. Similarly, out of the 75 neologisms listed by the Finnish language section of the Institute for the Languages of Finland (2021), 71 were nouns and 63 compound nouns. Mixed compound nouns, where one constituent is taken from another language, further enhance the possibilities of fine-tuning the expression. In the Institute for the Languages of Finland’s (2021) list of Finnish neologisms, there are mixed English-Finnish compounds such as cut out -mekko (‘cut-out dress’), pile-takki (‘pile coat’), and Y2K-muoti (‘Y2K fashion’).

All languages constantly need new words for naming new objects and phenomena. Nouns label and categorise all kinds of phenomena in the world: people, things, places, ideas, and emotions. In that sense, they are a powerful tool of human understanding and perception. The words we choose tell something about how we see and evaluate the reality around us. To take a well-known example, it makes a big difference whether a violence maker is called a terrorist or a freedom fighter. Through our word choices, we also invite our interlocutors to interpret the world in a certain way. To a large extent, then, meaning is negotiated in interaction. As Johnstone (2002, p. 45) puts it, ‘every linguistic choice – every choice about how to produce discourse, but also every choice about how to interpret it – is a choice about how the world is to be divided up and explained’. Linguistic choices can thus be seen as potentially strategic, and choices about how to name entities in the world are part of the strategy (Johnstone, 2002, pp. 45, 48−49).

Using words and other linguistic features from different named languages or varieties evoke different associations of values in communication such as solidarity and prestige. As Jørgensen et al. (2011, p. 29) state, ‘[s]uch an association may be an important quality of any given [linguistic] feature, and one which speakers may know and use as they speak’. Mixing languages can be used as a communicative or social strategy for reasons like showing speaker involvement, marking group identity, displaying status or expertise, or even excluding someone from communication (Grosjean, 2010, pp. 53−54).

Our word choices also carry echoes of our life history in terms of age, national belonging, ethnicity, education, and so on. The words we choose in communication reflect our background and previous encounters with language. According to Blommaert and Backus (2013, pp. 13–16), linguistic repertoires and means of communicating are individually and biographically organised complexes of linguistic resources that follow the rhythm of language users’ actual lives: With age, education, and other stages of life, we learn new patterns of using language, while some older ones may wane. According to them, the meaning-making practices of people are also increasingly characterised by (super)diversity. Different forms of mixing and blending are very common, and people can no longer automatically be put into specific national, ethnic, or sociocultural groups and identities based on their language use. Linguistic diversity and hybrid language identities are increasingly seen as a given assumption rather than a deviation from the ideal of monolingualism (cf. Hall & Nilep, 2015, p. 611).

It is from this communicative perspective that we examine our data. We seek to answer the following questions:

  1. 1.

    What kind of referents and textual contexts attract the use of mixed compound nouns in the data?

  2. 2.

    How do the mixed compounds in the data reflect the language of the surrounding culture, community, or speaker?

The overall aim of this study is to increase the understanding of how bilingual individuals make use of their linguistic resources to make meaning in their bilingual lives. Mixing languages is a typical feature of bilinguals’ interaction and a central competence in their language proficiency and intercultural competence (Council of Europe, 2020).

In this chapter, we examine data from closely related languages (Finnish–Estonian) and languages that are typologically far apart from each other (Finnish–Swedish and Finnish–English). Language mixing in the language pairs and in the data has been studied previously, but only scarcely from the point of view of mixed compounds (see Frick, 2009, 2013; Kolu, 2017). We bring together data from different language pairs in an examination of mixed compounding. We study if there are similarities in how bilinguals in our study organise and make use of their lexical repertoire irrespective of language boundaries when creating compound nouns.

10.2 Data

In order to demonstrate a wide distribution of mixed compounds, we have chosen a data collection that is very diverse, not only representing different language pairs but also different modes of communication (spoken, written, computer-mediated) in different genres (audio-recorded conversations, private diary, email messages, online video game, humorous podcast) and informants of different ages. The data used in this study includes datasets from five different sources: (1) bilingual (Finnish/Swedish) adolescents living near the Finland–Sweden border, (2) Finland Swedes, represented here by data from an elderly woman’s diary, (3) Finnish university students in Estonia, (4) native Finnish speakers in Finland, and (5) a humorous Finnish podcast. Table 10.1 shows an overview of the data. The details of the data are explained in the following chapters.

Table 10.1 Overview of the data
Full size table

10.2.1 Recorded Conversations Between Bilingual Adolescents in Haparanda, Sweden

The Haparanda data consists of over 19 hours of video- and audio-recorded informal group and pair conversations. Haparanda, located on the Swedish–Finnish border in Tornedalen (Torne River Valley) in the northern part of Sweden, is a small municipality of about 10,000 inhabitants. According to Statistics Sweden (2022), over 40 per cent of the inhabitants of Haparanda were born abroad and 82 per cent of them were born in Finland. But an even higher percentage of inhabitants has a Finnish background, that is, they were either born in Finland or have at least one parent or grandparent from Finland. Migration from Finland has strengthened the position of Finnish in Haparanda. Consequently, most schoolchildren currently have Finnish as their first language (see also Ruotsala, 2014). Many inhabitants in Haparanda speak Finnish or Meänkieli as their first language. Meänkieli is the official name of the language spoken in Tornedalen. While it is close to the dialects of Northern Finnish, Meänkieli has been subject to significant influence from Swedish and therefore has a number of Swedish features, such as modern loan words.

Currently, the towns of Tornio in Finland and Haparanda in Sweden share one commercial centre, and the border is extremely busy, with over 14 million annual crossings (Business Tornio, 2023). It is common for inhabitants in the border region to study, work, take part in different activities, and meet relatives and friends on both sides of the border.

The data was collected in 2014–2019 among 14–19-year-old bilingual adolescents at three junior high schools and one senior secondary school. One of the junior high schools was officially a Swedish compulsory school, whereas two of them were bilingual schools in which pupils could choose bilingual tuition. Most of the pupils in the bilingual junior high schools had a Finnish background, that is, at least one of their parents or grandparents were born in Finland. The aim of the curriculum in these two bilingual schools was to develop functional bilingualism. The only senior secondary school in Haparanda is Swedish-speaking.

31 adolescents in total participated in the recordings in Haparanda. All participants in the recordings were from Finnish-speaking or multilingual families, and they were by their own definition bilingual. In most families, both parents were Finnish- or Meänkieli-speaking, and both Finnish and Swedish were used at their homes. The conversations were recorded outside of the classroom at schools in Haparanda. The interlocutors were not told what language to use or what subjects to talk about, just to talk as ‘normally’ they could. The conversations turned out to cover such subjects as school, future plans, hobbies, travelling, and music. Finnish was the base language in 13 of the 18 conversations, but the bilingual adolescents frequently switched from Finnish to Swedish, but not so much from Swedish to Finnish.

Previous studies (Kolu, 2017, 2018, 2020) of the bilingual adolescents’ conversations in Haparanda revealed patterns and functions in the interlocutors’ use of linguistic resources and translanguaging practices in interactions outside the classroom. The participants fluidly and flexibly made use of their language resources, including words, grammar, and discourse markers as well as discourse practices. For example, the interlocutors used Swedish words when they referred to school subjects, although they were speaking in Finnish. This is understandable considering that instruction is given mainly in Swedish, and the Swedish national curriculum is followed at schools in Haparanda.

We found a total of 21 mixed compounds in the informal conversations.

10.2.2 The Private Diary of an Elderly Finland Swede in Northern Finland

The diary data consists of 795 diary entries from the years 1995−2004, written in Swedish by an academically educated, bilingual person in northern Finland. She lived in the city of Oulu, which is predominantly Finnish-speaking and has only a handful of Swedish speakers. Only 0.2 per cent of the city population have registered themselves as Swedish speakers (Statistics Finland, n.d.). As a result, she was influenced daily by the Finnish language in her life. In addition, her family language in adult life was Finnish: She spoke Finnish with her husband and children. One of the writers of this article knew the diary writer personally, and we have the permission of her close relatives to use her diary for research purposes.

The total number of graphic words in the diary is 40,855, excluding the dates. The average length per entry is 50 words. The shortest entries consist of only a couple of words, whereas the longest comprise several hundred words.

The most important named languages used in the diary are Swedish (in its Finland-Swedish variety) and Finnish, but sometimes even English and other languages are used in the text. According to af Hällström-Reijonen (2012, p. 15), the Finland-Swedish variety of the Swedish language is an over-regional variety of Swedish, and it is characteristic of Swedish-speaking Finns in speech and writing. Standardised Finland-Swedish is almost identical to standardised Swedish in Sweden, except for pronunciation and some official words and expressions that denote Finnish phenomena lacking a counterpart in Sweden. Finland-Swedish language usage, on the other hand, is a broader concept, including dialectal words and expressions, slang, loan words (notably from Finnish), and other less acceptable language features from a normative point of view (af Hällström-Reijonen, 2012, pp. 15, 79). The diary seems to have been written in what could be called a bilingual language mode (Grosjean, 2007) or translanguaging mode (Otheguy et al., 2015). The base language is Swedish, but the text is speckled with linguistic features that dissolve conventional language boundaries, for example language mixing (Kosunen, 2016, 2017, 2019). The number of Finnish–Swedish mixed compounds in the diary is 15.

10.2.3 Computer-Mediated Conversations Among Finnish Students in Estonia

In Estonia, approximately 1000 email messages were collected from a Finnish student association’s mailing lists and another 1000 from a mailing list of a small group of medical students from Finland. The data was collected in the town of Tartu, which is a popular place to study among Finns and other nationalities. Estonian is the sole official language in Estonia and the most common first language of the inhabitants of Tartu.

The linguistic backgrounds of many of the participants were rather plurilingual. For instance, out of the 11 students of medicine who gave permission to use their emails, seven were from monolingual Finnish families, and of these three had attended a school where a foreign language was the medium of instruction at some point in their late childhood or adolescence. Four participants were by their own definition bilingual from early childhood, either with a bilingual family background or because they had lived in a bilingual community. The bilingual participants all had Finnish as one of their native languages, although three of them were not from a Finnish-speaking family, and two had never had Finnish as the language of instruction in school.

At the time of data collection (early 2000s), most subjects at the University of Tartu were taught only in Estonian, but international medical students studied in English for two years, after which they switched to studying in Estonian. For this reason, the medical students’ mailing list (see Frick, 2008) contained a lot of language mixing to both Estonian and English. The Finnish student association’s mailing lists gave approximately 150 and the medical students’ list approximately 400 cases of Finnish–Estonian language mixing. In addition, approximately 50 cases of Finnish–Estonian language mixing were collected from private electronic messages between Finns (mainly students) residing in Tartu (Frick, 2008, 2009, 2013).

We collected 41 Estonian–Finnish and 34 Finnish–English mixed compounds from this medical students’ mailing list.

10.2.4 Native Finnish Speakers in Finland

Data was also collected from native Finnish speakers who have acquired English as a foreign language in school. English is by far the most common second language in Finland. In 2011, approximately 66 per cent of students in grades 1–6 and 99 per cent of students in grades 7–9 studied English in school (Official Statistics of Finland, 2012). This has resulted in widespread Finnish–English language mixing and the acquisition of new loan words from English.

One subset of the native Finnish speakers’ data was from a group of male friends in their 20s who played an online video game (PlayerUnknown’s Battlegrounds; see Heinilä, 2019, 2022). The Finnish–English collection was further expanded with a well-known Finnish podcast Kaverin puolesta kyselen ‘Asking for a friend’, whose hosts are two female native Finnish speakers (see Koivisto, 2021).

Prior studies show that native Finnish speakers use English words and expressions even in conversations among themselves. Since English is considered a language that ‘everyone knows’, even people who do not identify as Finnish–English bilinguals use it (see, e.g., Leppänen et al., 2008; Räisänen & Karjalainen, 2018).

We collected five English–Finnish mixed compounds from the video game data and eight from the podcasts.

10.3 Background

According to Busch (2012, p. 520), ‘the meanings that speakers attribute to languages and linguistic practices are linked with personal experience and life trajectories, especially with the way in which linguistic resources are experienced in the context of discursive constructions of national, ethnic, and social affiliation/non-affiliation’. With multilinguals, their diverse language background often shows in a language practice where they alternate between and mix different languages. The practice has been given various terms such as code-switching (e.g., Bullock & Toribio, 2009; Gardner-Chloros, 2009; Gumperz, 1977; Poplack, 1980), code-mixing (Muysken, 2000), polylingual practice (Jørgensen, 2008), translingual practice (Canagarajah, 2013), and translanguaging (e.g., García & Li Wei, 2014; Otheguy et al., 2015; Pennycook, 2016). In addition, the Common European Framework of Reference for Languages or CEFR (Council of Europe, 2020) has introduced the concepts of cross-linguistic and cross-cultural mediation, the ability to enable communication between people who do not share a language and who come from different cultures. In this chapter, we use the term language mixing in a neutral way without any predetermined theoretical assumptions. By language mixing we refer to all forms of language use where two or more named languages or language varieties are simultaneously used and intertwined in a communicative situation: between sentences, within sentences, and within words. In many language mixing situations, single lexical items appear in a text that is otherwise seemingly monolingual. We will use the term base language to refer to the main language of the text and inserted language to refer to the language used in language mixing (for a discussion on base language, see Muysken, 2000, pp. 64 − 69).

As Gardner-Chloros (2009, pp. 42–43) notes, there are several social factors that contribute to the use of language mixing in language contact situations. She names three main groups of social factors that have an impact on if and how language mixing occurs:

  1. 1.

    Factors that affect all the speakers in a particular community, for example in terms of prestige and power relations of the different language resources;

  2. 2.

    Factors that are connected to the speakers, for example in terms of attitudes, self-perception, and affiliation to different groups; and

  3. 3.

    Factors ‘within the conversations’, that is, language mixing as a pragmatic discourse practice available to bilinguals.

The language choices in language mixing can act as social indexes that hint to the interlocutors that the topic in question is linked to the language the speaker changes into on any of the above-mentioned levels (community, speaker, or conversation). For instance, in a Finnish email message, the use of the English word exam might indicate that English is a high-prestige language in the community, that the speaker perceives themselves as an English speaker or that the exam in question will be held in English. The finding that linguistic variables act as social indexes in interactional situations has already been described by Gumperz (1982, p. 202). Myers-Scotton’s (e.g., 1983, 2000) ‘markedness’ theory on the social indexicality of language mixing has been criticised for presenting social indexes as predetermined and omnipresent instead of interactionally emergent and negotiable (see, e.g., Frick, 2013, p. 70; Li Wei, 1998). Social indexicality is often linked to what we call cultural specificity: Certain phenomena only exist or are known to the speaker in one or the other culture and therefore attract the use of the corresponding language when talked about (see, e.g., Kolu, 2017). For instance,  in a Swedish school where the Swedish curriculum is taught, bilingual pupils use Swedish words for school subjects, like SO (Samhällsorienterande ämnen) ‘social studies’ and NO (Naturorienterande ämnen) ‘natural sciences’, even when they are speaking Finnish (Kolu, 2017). As García (2009, p. 48) notes, bilinguals usually have differentiated use and proficiency in their language resources which they use to communicate, as they have been exposed to various language experiences and practices. Consequently, the language settings have an impact on adolescents’ language use.

There are also cases of unmarked language mixing with no apparent motivation (Myers-Scotton, 1983). There are discourse and linguistic practices among groups of bilinguals and multilinguals where the interlocutors make use of their common linguistic resources in their interactions with each other (Kolu, 2020; McCormick, 2002). The language resources are often intrinsic and associated with the speakers’ identity (Kolu, 2020). They are also used as in-group markers by adolescents (Henricson, 2015; Lehtonen, 2015; Kolu, 2017). The notions of linguistic superdiversity, translanguaging, mobile resources, and heteroglossia are often used as umbrella terms for the use of multiple language resources that seem random on the textual level (Blackledge & Creese, 2020; Blommaert, 2010; Pennycook, 2016). The superdiversity perspective stresses the dynamic use of language resources and the ‘normality’ of language mixing. Many recent inquiries take the stance that bilingual language use should not be viewed as ‘plural monolingualism’ (Makoni & Pennycook, 2012, pp. 444−446) or ‘dual linguistic competence’ (Otheguy et al., 2015, p. 298) but as evidence of dynamic linguistic practices that deconstruct our understanding of what ‘languages’ are. The idea is that language mixing can be unmotivated by the expressive needs of the speaker and just unfold as neutrally as any monolingual text would (see, e.g., Blommaert, 2015; Blommaert & Rampton, 2011).

As bipartite units, mixed compounds offer a fertile ground for the study of language mixing. As Alexiadou (2020) states, the research of (mixed) compounds in language contact situations gives us information about the units of language mixing as well as the organisation of the mental lexicon of bilingual speakers. Mixed compounding has been seen as evidence that the bilingual lexicon is integrated, as bilinguals can use language materials from both languages when constructing compound structures (e.g., Alexiadou, 2020; Brysbaert, 1998; Graedler, 1998; Putnam et al., 2018; van Heuven et al., 1998). As Otheguy et al. (2018, p. 3) argue, ‘[t]he myriad linguistic features mastered by bilinguals (phonemes, words, constructions, rules, etc.) occupy a single, undifferentiated cognitive terrain that is not fenced off into anything like the two areas suggested by the two socially named languages’. On the other hand, we suggest that bilingual individuals attach some meanings of the words to the particular social groups and cultures in which they participate or have participated as a consequence of their own life histories, language backgrounds, and personal experiences.

10.4 Analysis

In this section, we discuss what kind of referents and textual contexts attract the use of mixed compound nouns in our data. In addition, we discuss how mixed compounds in the data reflect the surrounding culture or its norms, the community of speakers, and the speaker’s affiliations or communicative motivations. The discussion is organised into subsections according to the referents and textual functions of the mixed compounds. In the data, we found words denoting places, food, family, and other community- and context-specific terminology. Some of these words fill a lexical gap (see Baker, 2011) in one language or are culturally specific. Others are used in contexts where the participant talks about events where the inserted language is, was, or will be used.

10.4.1 Places

Prior studies of language mixing have shown that toponyms are typically uttered in their original form, that is, in the language of the surrounding community (see, e.g., Frick, 2003, p. 14; 2013, p. 16). Our data shows, however, that mixed compounds can also be used to refer to places. The places named in our data include both semi-translated forms of a commonly known place name and more sporadic references to places.

In Excerpt (1), the mixed compound hampurilais+restaurang-iin ‘to [a] hamburger restaurant’ is used in a conversation where two participants are discussing where they are going to do their on-the-job training for school.

(1) Haparanda

mää en ainakaan mee minnekään hampurilaisrestaurangiin

‘At least, I am not going to any hamburger restaurant’

The compound is formed of the modifier in Finnish: hampurilais ‘hamburger’ and the head restaurang ‘restaurant’ in Swedish. The word is inflected according to the base language Finnish and its grammar. The Finnish illative ending -(i)in indicates movement into something.

The language choices in the word hampurilaisrestaurang reflect the everyday language use and environment the participants live in. The towns of Haparanda and Tornio on the Finland–Sweden border share a town centre, and therefore the border crossing-point is almost imperceptible. Linguistic diversity is commonplace in the twin city, and both Swedish and Finnish can be used, for instance, in restaurants such as the one talked about in Excerpt (1).

Other examples of place names in the data include Meditsiini+talo ‘Medicine Building’, which is a mixed compound found in the dataset of Finnish students in Estonia. Meditsiinitalo refers to the main building of the Faculty of Medicine in Tartu, which is commonly known as Meditsiinihoone in Estonian. The compound is formed by translating the head of an originally Estonian compound noun into Finnish: The modifier meditsiini is in Estonian and the head talo ‘building’ in Finnish. The head is a translation of the original hoone.

In the diary data, the word ammatti+koulu+butiken ‘the vocational school shop’ is used to refer to a bakery shop at the vocational school in Oulu. Ammattikoulubutiken is a three-part compound with the Finnish modifier ammattikoulu ‘vocational school’ and Swedish head butiken ‘the shop’. The use of Finnish in the modifier of the word is motivated by the school being a Finnish one.

10.4.2 Community-Specific Terminology

Baker (2011, p. 108) explains that people may have lexical gaps if they use different languages in different domains: ‘A young person may, for instance, switch from the home language to the language used in school to talk about a subject such as mathematics or computers’. In those cases, the gap is in the individual’s linguistic repertoire. We find, however, that trying to explain all domain-motivated language mixing with the lexical gap theory shows the analyst’s ‘monolingual bias’ (see Auer, 2007). A bilingual individual may choose to mix languages even when they would know how to communicate monolingually.

There are, nevertheless, lexical gaps that encompass the whole linguistic system—that is, words that do not exist or are not commonly used in all languages. Such words often refer to community-specific phenomena. In this case, it is typical that the modifier of the compound fills a lexical gap in the other language. For instance, the term kommunaalimaksut ‘utility costs’ is used by Finns in Estonia to refer to building maintenance costs that differ essentially from the respective system in Finland. The head of the compound is a bilingual homophone (Fin. maksut ~ Est. maksud), and the modifier kommunaali is in Estonian.

Kommunaalimaksut reflects a conventionalised multilingual discourse practice in the Estonian Finnish community. Similarly, the bilingual adolescents in Haparanda use the word SO-luokka ‘a classroom for social studies’. The modifier of the compound SO is Swedish and stands for the school subject samhällsorienterande ämnen ‘social studies’. There is not one equivalent school subject for ‘SO’ in Finnish; instead, the topics are covered in several subjects (history, religion, and geography) in the Finnish school system. The head of the compound luokka is the Finnish word for ‘class’. SO-luokka is thus an example of both a specific place name and of terminology that is community-specific.

The words kommunaalimaksut and SO-luokka also reflect the power relations of the languages used in the two respective communities. Estonian and Swedish as the official languages of Estonia and Sweden respectively are used when referring to community-related topics, indicating that the societal structures function in these two languages. The local minority language, in this case Finnish, is not used by the authorities, and the minority language speakers adapt to this instead of, for instance, translating the terms into their own language.

We also find, in the Finnish podcast data, an example of a lexical gap in that is not the result of Finnish being used as a minority language but, instead, in a context where reference is made to a foreign institution. The mixed compound in question is high school -elokuva ‘high school movie’, in which high school refers to the United States’ school system. The school system in Finland differs from the US one, which is why there is a lexical gap in the Finnish language regarding the term, and replacing high school with a Finnish word would not work. High school films may also be regarded as their own film genre. The term is used in Finland to describe films where the main teen characters attend an American high school. Furthermore, the mixed compound high school -elokuva has a special connotation which is associated with teenagers in the USA.

10.4.3 Food

Food-related words are another group that seems to trigger language mixing often due to the cultural specificity of food items (Frick, 2013, p. 16). In our data, there are mixed compounds that refer to both culture-specific and non-specific food-related items and topics. Excerpt (2) shows a culturally non-specific food-related mixed compound in the diary data. The example is shown in its sentential context, which is in Swedish.

(2) Diary

Först var det sill och kirjolohipiroger.

‘First there was herring and rainbow trout pastries.’

In Excerpt (2) the writer of the diary tells about a party and the food service there. Kirjo+lohi+piroger is a three-part compound noun consisting of the Swedish word pirog (‘pastry’) as the head and the Finnish compound word kirjo+lohi (‘rainbow trout’) as the modifier. The motivation for using a mixed compound in the context can be interpreted from the community perspective. The party the diary author refers to was with Finnish-speaking friends, and so the language of the situation is echoed in the diary text. Moreover, expressing the meaning in Swedish would have resulted in a slightly more complex compound consisting of four parts: regn+bågs+lax+pirog. Thus, the economy of expression might also have played a role in choosing Finnish as the language of the modifier.

Other food-related mixed compounds in the data include fika+juttu ‘coffee break thing’ from the Swedish fika ‘coffee break’ and Finnish juttu ‘thing’ and juustu+bileiden ‘cheese party, pl gen’ from the Estonian juustu ‘cheese, gen’ and Finnish bileiden ‘party, pl gen’.

10.4.4 Studies

Students often refer to study-related topics with mixed compounds, in which the modifier is in the language of the studies while the head is in the language of the surrounding text. This is common for both the bilingual pupils speaking Finnish as a base language in Haparanda (five study-related examples) and the Finnish university students in Tartu (53 study-related examples). In the other datasets, studies are not among the topics of conversation.

Rebase+meininki ‘freshman activities’ in Excerpt (3) refers to student activities for freshmen. The example is taken from an email by an older Finnish student to first-year Finnish medical students in Estonia.

(3) Email

Tahtoisin vain kertoa teille, että minusta jokaisen kannattaa ottaa huomenna osaa rebasemeininkiin.

‘I just wanted to tell you that I think everyone should participate in the freshman activities tomorrow.’

The email, which Excerpt (3) is taken from, is in Finnish, as is the head of the compound meininki, which is inflected with the Finnish illative suffix -in. The head rebase ‘freshman, gen.’ is in Estonian (see Frick, 2009, for a more extensive analysis of the example). Although the Finnish medical students started their studies in English, extra-curricular activities such as the ones talked about in this email were organised for the whole student faculty and were mainly held in Estonian.

In Excerpt (4) another medical student writes about a reexamination.

(4) Email

sori mut on pakko pitää cranial nerve uusinta torstaina klo 9 koska pe on kemian suullinen ja sit pitäisi lukea tätä histologiaakin.

‘Sorry but I have to take the cranial nerve reexamination on Thursday at 9 AM, because I have the chemistry oral [exam] on Friday and I need to study histology as well.’

The mixed compound cranial nerve uusinta consists of an English modifier and Finnish head (uusinta ‘reexamination’). The normative spelling of the word would be cranial nerve -uusinta.

These excerpts show how the language of the modifiers in the compounds reflects the language of the events talked about; freshman activities are held in Estonian and the cranial nerve exam in English. The language choice acts as an intertextual cue and social index (see, e.g., Frick, 2013, p. 70) that bridges the real-life events to the text here-and-now and adds authenticity to the referential potential of the word. In Excerpt (4) the writer refers to three different study topics: cranial nerves, chemistry, and histology, but only one of them is in English. There is no definite reason why one word gets written in a different language than the surrounding text and others do not, but researchers in language mixing have found certain tendencies in language mixing patterns. For instance, in Excerpt (4), the anatomical term cranial nerve is a more specific term than chemistry and histology, which refer to areas of science. Semantic specificity such as this may motivate language mixing (see Backus, 2001).

Other study-related mixed compounds from the Finnish medical students’ email messages include exam+päivät (Eng. exam+Fin. päivät) ‘exam days’, keskkonna+kysymykset (Est. keskkonna+Fin. kysymykset) ‘environment questions’, and nursing opettaja (Eng. nursing+Fin. opettaja) ‘nursing teacher’. There are also study-related mixed compounds, such as reklam+juttu (Swe. ‘advertisement’+Fin. ‘thing’), in the data we collected from the Finnish students in Haparanda.

10.4.5 Other Context-Specific Activities

Context-specific language selection is not limited to the field of studies. Excerpt (5) is taken from Heinilä (2022). It reflects the context-specific use of the compound quarryporukka ‘quarry team’.

(5)

Mihihä se quarryporukka sitte kerkes juosta voi että.

‘I wonder where the quarry team ran, oh my’

The word quarryporukka consists of the English modifier quarry and Finnish head porukka. It is used by a native Finnish speaker when playing a video game with other native Finnish speakers. When playing the game, the participants in this dataset tend to refer to quarries with the English term instead of the Finnish equivalent louhos, which they would use in other contexts. For them, the English word seems to be a conventionalised way to refer, specifically, to quarries within the game.

10.4.6 Family Vocabulary

In the diary data, we find examples of what could be called a familylect (see Søndergaard, 1991). One example of family vocabulary is the word Koppi-bil in the diary data, as seen in Excerpt (6). A literal English translation of the word is ‘booth car’. The word refers to a small van car owned by one of the family members.

(6) Diary

Just som jag ätit kom Koppi-bilen

‘As soon as I had eaten, the booth car came’

The head noun bilen (‘car’) in the example is in Swedish and the modifier koppi (‘booth’) in Finnish. The Standard Swedish word for the referent is skåp+bil (literally ‘cupboard car’), whereas Finland-Swedes often use the word paket+bil (literally ‘packet car’), echoing the Standard Finnish word paketti+auto (literally ‘packet car’). In the family, however, the car was always called koppiauto. As the word has the quality of a proper noun, and as the diary writer was used to language mixing in her text, it is somewhat surprising that she does not use the all-Finnish compound koppi+auto but creates a mixed compound instead.

10.4.7 Other

The compounds discussed in this section include both words whose monolingual counterparts have a conventionalised meaning and are frequently used, as well as ones that are novel compounds that serve the interactional situation here-and-now but are not conventionalised in either bilingual or monolingual language use.

One example of a mixed compound noun whose monolingual counterparts have a conventionalised meaning is found in the diary data: bil+katsastus, meaning ‘car inspection’. The head noun katsastus (‘inspection’) is in Finnish, and the modifier bil (‘car’) is in Swedish. From the analyst’s point of view, it is hard to interpret why a mixed compound was used, especially as the writer also uses the respective conventionalised Swedish noun bil+besiktning once elsewhere in the text.

In the diary data we also found the word bastu+patio ‘sauna porch’, which is used to refer to the porch of the sauna building at the family’s summer place. The word consists of the Swedish modifier bastu ‘sauna’ and the Finnish head patio ‘porch’. (The word patio is a Spanish loan word in Finnish and can mean either ‘inner yard’ or ‘porch’.) Compared to the two previous examples of place names, bastupatio shows a reversed pattern of combining the compound lexemes: In meditsiinitalo and ammattikoulubutiken (discussed in Sect. 10.4.1), the generic head noun is expressed in the language of the surrounding text, and the specifying modifier in the language that echoes the local circumstances. In bastupatio, the head noun is expressed in the local language and the modifier in the base language of the text.

Bilkatsastus ‘car inspection’, pesumasiina ‘washing machine’, and bastupatio ‘sauna porch’ are all general words that are not associated strongly with one or another culture, nor are they related to a domain that would be more associated with one of the languages. Other such mixed compounds that have conventionalised monolingual counterparts include läpi+paistavat (‘trans+parent’ cf. Fin. läpi+näkyvät, pl.; Est. läbi+paistvad) and sminkka+pussit (‘make-up bags’, Swe. smink ‘makeup’+Fin. pussit ‘bags’).

Overall, compounding is a productive way to facilitate the integration of foreign elements into the base language (Graedler, 1998, p. 202; Söderberg, 1983, p. 14). One pattern in compound formation found in the Haparanda data is to combine Swedish elements with semantically relatively empty Finnish nouns (cf. Kolu, 2017, p. 61; see also Söderberg, 1983, p. 14). The compounds reklamhomma (‘advertising thing’), fikajutut (‘coffee things’), and plastjutut (‘plastic things’) are formed with the modifiers reklam (‘advertising’), fika (‘coffee’), and plast (‘plastic’) in the inserted language, Swedish, and the heads homma (‘thing’) and jutut (‘things’) in Finnish. Compound forms make it easier to insert Swedish nouns into the Finnish syntactic environment. Similar examples were also found in the Stockholm data collected for another study (Kolu, 2017). Correspondingly, in the Helsinki data (Kolu, 2017, p. 61), where the base language is Swedish, the young Swedish-speaking Finns inflect the Finnish noun juttu in the Swedish plural by adding the Swedish plural endings -n or -r in order to integrate the compound word into the Swedish syntactic environment: skolhälsojuttun ‘school health thing’ and kärleksjuttur ‘love things’.

(7) Haparanda

sit me saadaan ne fikajutut

Then we receive those coffee things

(8) (cf. Stockholm, Kolu, 2017)

sit mun pitää tehdä se svenskajuttu

Then I have to do that Swedish thing

In Excerpt (7) and (8), the mixed compounds had no conventionalised equivalent in any language. The exact meaning of these words must be interpreted from the context. Many of the words in this group end in the Finnish head noun juttu (‘thing’). Without knowing the context, it is often impossible to determine what ‘thing’ the speaker is referring to. In Excerpt (7), the speaker refers to the gift cards they had received. In Excerpt (8), the student refers to a school assignment in Swedish that she must do. Similarly, one of the Haparanda schoolchildren uses the mixed compound band+juttu (‘ribbon/tape/band thing’) to refer to a plastic strip used in skiing centres to mark the sides of the skiing slope. The language selection of the modifier, the Swedish band, is peculiar, because the main language of the conversation is Finnish and the skiing centre in question was also located in a Finnish language-dominant area of Finland. This example, among others, shows that mixed compounds are not always motivated by an association with one or the other language.

10.5 Summary and Discussion

In this chapter, we analysed and discussed examples of compound formation in different contact situations, that is, in the language use of Sweden Finns, Finland Swedes, Finns in Estonia, and Finnish L2 speakers of English. The study of compounds in language contact provides insights into the mental lexicon and multilingual discourse practices of bilingual speakers.

The data set represents not only different language pairs but also different forms of communication (spoken, written, computer-mediated) in different genres (audio-recorded conversations, private diary, emails, online video game, humorous podcast) and different ages of informants. The data comprises naturally occurring written and spoken texts and pair interviews with people whose one language is Finnish and other language is either Swedish, Estonian, or English. All texts except the one collected from native Finnish speakers in Finland had a base language that was different from the dominant language of the surrounding society: The bilingual adolescents in Haparanda, Sweden, who used both Finnish and Swedish but mainly Finnish in the recorded conversations, lived in a Swedish-dominant town and went to a bilingual or Swedish-speaking school. The diary of our Finland-Swedish participant was written in Swedish, but the main language used in the town was Finnish. The Finnish students in Estonia wrote their email messages mainly in Finnish but lived in a town where Estonian was the dominant language, and they studied in English and Estonian.

Specifically, we found evidence for the view that bilingual speakers make use of their multiple language resources when forming compound words (see also Kolu, 2020; McCormick, 2002). In addition, our examples show that the bilingual lexicon seems to be integrated and dynamic, as bilinguals can choose material from both languages when compounding words (see also Alexiadou, 2020). This provides further evidence for the view that bilinguals do not consider their languages as separate and isolated systems. As Jørgensen et al. (2011, p. 29) state: ‘Speakers use features and not languages. Features may be associated with specific languages (or specific categories which are called languages). Such an association may be an important quality of any given feature, and one which speakers may know and use as they speak.’

The examples discussed in this section show that word-internal language mixing in the form of mixed compounds happens in many different multilingual communities in which Finnish is one of the languages spoken. Speakers typically use a pattern in which the head of the compound reflects the language of the surrounding text while the modifier is used in another language. This is true for 113 out of the 125 mixed compounds in our dataset. In some cases, the language pairs studied here share underlying concepts but lexicalise them differently (skåp+bil, literally ‘cupboard car’ in Swedish and paket+bil, literally ‘packet car’ in Finland Swedish), while in other cases, the mixed compounds in the data fill lexical gaps in the language of the surrounding text. For instance, Finnish does not have words to denote all the referents related to building maintenance in Estonia or the Swedish or US school systems. In the case of Finnish in Estonia and Sweden, this reflects the weaker status of Finnish and the fact that the societal structures function predominantly in the majority language. Such examples were not found in the Finland Swedish diary data, which may reflect the stronger position of Swedish in Finland compared to that of Finnish in Estonia and Sweden. More data and research on the topic are needed, however, before any definite conclusions can be drawn on the matter.

Very often, these mixed compound nouns reflect the multilingual everyday life and discourse practices of the speakers, especially objects, events, and topics that, in the speaker’s everyday life, are more linked to the inserted language. Thus, the inserted-language modifiers act as social or cultural indices, ‘belonging’ more to the life lived in that language. This is not only true for texts written in a minority language but also for many of the English insertions in the native Finnish speakers’ Finnish texts. English insertions reflect both the widespread knowledge of English in Finland and the use of English in different life areas such as video games.

Earlier studies show that compounds may have an expressive and pejorative function, for instance in slang vocabulary (cf. lovebrud ‘girlfriend’ English–Swedish, dumass ‘stupid’ Swedish–English; Kotsinas, 2002). We did not detect such usage in our data. Instead, the referents of the mixed compounds in our data were rather mundane and affectively neutral. The mixed compounds in our data were used to refer to places, food, school-related topics, and other everyday matters.

All in all, it is impossible to unequivocally say why people mix languages and create mixed compound nouns. Language use can fulfil several functions simultaneously, and mixing languages gives the message an enriching ambiguity that allows for many different interpretations.