With its own national language, Luxembourgish, and three languages of administration, French, German and Luxembourgish, Luxembourg has long been a very multilingual country. The nature of this multilingualism is now changing, due to the rising proportion of migrants in the country, who now make up 43% of the resident population. The changing demographic profile of Luxembourg is reflected in a diversification of language practices within this already highly multilingual context. This article focuses on one group of people who exemplify these changes, 24 students of diverse national and language backgrounds at the University of Luxembourg. Using data from a language diaries exercise conducted as part of an introductory course on multilingualism, we examine the reported multilingual practices of the students and consider what these language practices reveal about traditional and newer forms of multilingualism in Luxembourg.
Keywords: multilingualism, multilingual practices, language diaries, migration, Luxembourg, university students
Introduction: multilingualism in Luxembourg
Luxembourg is a small country with a geographical size of 2,586 square kilometres and a population of 511,800 (Statec 2011). Bordered by France, Belgium and Germany, Luxembourg is also located on the border of the Romance and Germanic language families, which has had an important influence on its unique language situation. Traditionally, French and German (and, earlier, Latin) were used for written functions. Luxembourgish, a Germanic language variety similar to the Moselle Franconian varieties spoken across the borders in France and Germany, functioned mainly as a spoken language, with a written literature beginning in the nineteenth century. It was not until 1984 that a language law formalised the existing language situation, making Luxembourg officially trilingual. Since then, Luxembourgish has been the national language of Luxembourg, French the language of the law, and French, German and Luxembourgish all languages of administration.
In contrast to some multilingual countries in Europe where multilingualism is territorially based (e.g. Belgium and Switzerland), Luxembourg’s languages have traditionally been used in different domains. Hoffmann (1979) described the Luxembourgish language situation as triglossic, with German and French used mostly in written domains and Luxembourgish in spoken domains. French, for example, was the language chosen for street signs, German was for a long time the main language of the print media, and Luxembourgish was traditionally the language used for spoken interactions at all levels of society, from government interactions to shopping. But language practices in Luxembourg have also varied significantly within domains, with education a prime example. Whereas Luxembourgish is the official language of instruction in pre-school education, German takes up this role in primary school, and French is introduced gradually from the second year onwards. Both French and German are used as languages of instruction in secondary education, with French predominating in the lycée classique and German in the lycée techniquei. English is a compulsory subject from secondary school onwards. Luxembourgish, on the other hand, has very little official presence in the education system (one hour of instruction a week in primary school), but is reportedly widely used informally by both teachers and students. On leaving the school system, therefore, students are theoretically quadrilingual in Luxembourgish, German, French and English, although in practice their level of competence in the different languages varies considerably according to ethnic or class background (Weber 2009, Davis 1994). Until recent years, if a student chose to go to university, this would mean studying in another country (usually Belgium, France or Germany), as Luxembourg did not have its own university. Since the creation of the University of Luxembourg in 2003, however, students can undertake degree studies within Luxembourg. The university has an explicitly multilingual orientationii and attracts a high proportion of both local and international students from across Europe and around the worldiii.
If there has long been a high degree of multilingualism in Luxembourg, the nature of this multilingualism is now changing, due to important demographic changes within the country. The favourable economic situation of Luxembourg has resulted in increasing migration in recent decades, to the point that people of non-Luxembourgish origin now make up 43% of the resident population. This includes several different groups of migrants. The first relates to historical waves of long-term migration, including Italians and Portuguese, the latter of whom now make up 16% of the resident population (Statec 2011). Next there are migrants who come to work at EU institutions and multinational companies, often only for a temporary period. Finally, there are daily migrants (cross-border workers), who live in Belgium, France and Germany and work in Luxembourg, and who currently make up 42% of the workforce (Statec 2011). The presence of these groups of relative newcomers is having a significant impact on the languages used in Luxembourg. One of the most important changes in this regard is an increased use of French overall. As the majority of cross-border workers come from France and French-speaking Belgium, French, which was traditionally largely a written language in Luxembourg, is now used extensively in spoken domains and has ‘come to be used increasingly as a lingua franca among people living and working in Luxembourg’ (Horner & Weber 2008: 87). Meanwhile, English has assumed an increasingly important place within many of the multinational companies based in Luxembourg, and there has been a hypothesised decrease in the prevalence of German in some workplaces (see Klein 2007), so that now ‘German is not nearly as important as French and English in many sectors of the job market’ (Horner and Weber 2008: 97). These evolutions go alongside a growing salience of the Luxembourgish language as a marker of group membership within the Luxembourgish population, which is particularly notable in its increasing use as a written language (Horner & Weber 2008: 86). Horner and Weber (2008) relate the recent drive to promote Luxembourgish to fears among some of the local population that the language is under threat due to the increasing presence of ‘foreigners’. Whatever position one takes on this debate, these sociolinguistic developments challenge the traditional triglossic picture of Luxembourg and pave the way for the development of new forms of multilingualism in the country.
The University of Luxembourg, with its mixed population of local and international students, presents in some ways a microcosm of the increasing linguistic diversity of Luxembourg. With this in mind, this article examines the language practices of a group of first-year students at the University of Luxembourg to answer the question ‘what are the multilingual practices adopted by these students, and what do these practices suggest about changing forms of multilingualism in Luxembourg?’
Methodology: language diaries
The method used for this study is that of language diaries. Jones et al (2000) highlight several advantages of language diaries for investigating multilingual practices, in contrast to other methods such as semi-structured interviews (and, for that matter, questionnaires). First, interviews tend to involve the researcher leading the discussion and determining what is relevant for the participants to provide in terms of data about their language practices, thereby imposing their own agenda and assigning the participants a somewhat passive role. Participants can assume more control in a language diary, being responsible themselves for selecting what counts as data and what is relevant. Secondly, interview questions about participants’ language practices often result in quite general responses, because it is difficult for participants to think on the spot of specific instances of language use. This can result in data that is ‘too decontextualized, abstract and generalised’ (Jones et al 2000: 323) for the researcher's purposes. In language diaries participants are able to describe instances of specific language use at specific times based on their actual daily behavior. If subsequent interviews about the diaries are involved, this allows the researcher to move from ‘the specific to the general, rather than from [the] general to the specific’ (Jones et al 2000: 344). Of course, language diaries cannot possibly capture all the language practices a participant engages in during the period of keeping the diary, and what is written is necessarily a subjective selection of these practices. Furthermore, participants’ language diaries relate only to that particular day or those particular days that they complete the diary, and may not be generalizable to their habitual language practices. Nevertheless, this snapshot of language use provided may come closer to capturing the variation inherent in ‘authentic’ language use.
The specific model for the language diaries used in this instance was based on that used by Jones et al (2000) to investigate multilingual literacies in Britain. The authors report on two studies, one focusing on 12 Welsh/English bilinguals in north-east Wales, looking at their literacy practices in public sector institutions, and one focusing on 6 Gujarati/English bilinguals in Leicester, studying their literacy practices in workplace settings. In both cases, the researchers gave the participants a language grid to fill out the languages they used for spoken (listening and talking) and written (reading and writing) purposes in the course of two to four weeks, depending on the study. This was then followed up with interviews with the participants about their language diaries, as well as observation of some of their daily activities, as part of a longer-term ethnographic project. The current study differs from that of Jones et al (2000) in several respects. Firstly, the studies described above focused on multilingual practices taking place in quite different contexts. The Welsh/English study involved the use of an indigenous minority language in a bilingual environment, and the Gujarati/English study involved the use of a migrant minority language in an English dominant environment. The context for the present study is a much more multilingual setting overall. Secondly, whereas Jones et al (2000) were focusing primarily on literacy practices, i.e. participants’ relationships with written texts (although also considering speaking and listening practices in relation to these), our interest is equally in spoken and written practices. This is not, therefore, a study of multilingual literacies in particular, but of multilingual practices in general. Thirdly, both studies reported in Jones et al (2000) were in-depth ethnographic projects, with the Welsh/English study involving field-work spanning two years. Our study focused on a single day in our students’ lives and, unlike these other studies, cannot hope to capture more than a glimpse of their longer term multilingual practices. Finally, unlike Jones et al (2000) we did not engage in detailed individual interviews with our participants about their language diaries, although we did engage in other activities to build on their diary accounts, as described below.
The participants of the current study were 24 first and second year university students from an introductory course on multilingualism at the University of Luxembourg in 2012, which we jointly taught within the English section of the ‘Bachelor of European Cultures’ program. Within this course the students were required to submit written exercises to an online student forum each week in preparation for the subsequent lecture. The exercise reported on here was undertaken for the lecture on ‘multilingual literacies’. The week prior to the lecture students were asked to prepare a language diary, picking one day in the course of the week and noting down ‘the languages [they used] for reading and writing, speaking and listening at different moments of the day’. We provided them with a grid based on that used by Jones et al (2000), as follows:
[insert Table 1 here]
Twenty-four students submitted a language diary (out of a total class of twenty-five). All of these students completed their diary in English, which we expected, given that the language of the course was English. As in the studies reported in Jones et al (2000), the students varied in the amount of detail they provided, with most diaries involving brief bullet point descriptions of activities undertaken and the languages used (examples will be shown below). When the students had uploaded their diaries onto the discussion forum, we analysed the diaries to look for patterns and themes in the students’ language practices. We then brought the diaries along to class, where they were integrated into the lecture on multilingual literacies. At the start of the class we asked the students to interview each other in groups of two about their language diaries, picking up on aspects they found interesting in each other’s diary. Later in the class we presented to the students patterns that we had identified in the diaries as a whole and asked the students if they recognised themselves in our descriptions. Finally, at the end of the class we led a group discussion, asking what the students had found interesting or surprising about doing the language diaries themselves and what had they learnt or noticed about their own multilingual practices. The students were not aware that their diaries would subsequently be used for a research exercise, as in fact neither were we - it was on the basis of the interesting results and discussion that we decided to write this article. We then sought the consent of the students for the use of their diaries for this purpose, and provided them with a draft of the article in time for the last class of the course. As preparation for the class we asked them to prepare critical feedback on the article and to suggest corrections and changes to our analysis of their practices. Only minor revisions were suggested by the students, and these were subsequently made. As well as a research exercise, therefore, this was a teaching exercise, offering the students the possibility to reflect on and contribute to research about their own multilingual practices.
Before moving on to the results of the analysis, it is important to consider in more detail the national and language backgrounds of the students. From teaching the class, we were aware that the students had a mixture of backgrounds, some appearing to have a longstanding Luxembourgish family history, some of migrant origin but resident in Luxembourg for part or all of their lives, some more recent migrants, and some international exchange students. Before conducting this exercise we estimated that approximately half the class was of Luxembourgish nationality, and the other half of different national/language backgrounds. To have a clearer picture, once we had undertaken an initial analysis of the language diaries we asked the students to fill out a table indicating their nationality (or nationalities), if they had any further nationalities in their recent family background, and the first language (or languages) they learnt at homeiv. On the basis of the results we separated the students into three broad categories. The first category was students with strongly Luxembourgish backgrounds, who had Luxembourgish nationality, no other nationalities in their recent family history, and Luxembourgish as their first language. The second category was students with migrant backgrounds. This was a more varied category, including students with Luxembourgish somewhere in their nationality/recent family background/first language(s), but with other nationalities and languages too. Examples of students in this category include:
a student with Luxembourgish nationality, an Italian father, a Swiss and Luxembourgish mother, and Luxembourgish as a first language;
a student with Luxembourgish and French nationality, Dutch and Polish nationalities in her recent family background, and French as a first language;
a student with Luxembourgish nationality, Dutch and Polish nationalities in her recent family background, and Dutch and Luxembourgish as first languages; and
a student with French and Portuguese nationality, Brazilian nationality in his recent family background, and French, Portuguese and Luxembourgish as first languages.
The third category was students with international backgrounds, who had no Luxembourgish nationality, no Luxembourgish family history and did not have Luxembourgish as their first language. To our surprise, the results for the number of students in each category were rather different from our expectations, as summarized in Table 2 below:
[insert Table 2 here]
While we had correctly estimated the proportion of students with no Luxembourgish background at all, we had radically underestimated the proportion of students with migrant backgrounds in their recent family history, and concomitantly overestimated the number of students with what we had considered to be strongly Luxembourgish backgrounds. As well as further foregrounding the national and linguistic diversity of the class, this raises the slippery question of what it actually means to be a ‘Luxembourger’ (see Horner and Weber 2008), a question we will return to later.
Results: student’s multilingual practices
The analysis of the language diaries revealed several interesting themes in relation to the students’ multilingual practices. The results of the analysis are described below, also incorporating further material gained from the discussion about the diaries in the class on multilingual literacies.
High degree of multilingualism
The most striking result of this study was the highly multilingual practices adopted by every student in the class. One student for example reported listening to the radio in German, speaking Polish to her family, reading a newspaper in French, speaking English at work, and surfing the internet in Luxembourgish, in the course of one day. Several students reported using several languages at the same time, for example one referred to surfing the internet in English, Luxembourgish and German. There was no monolingual language diary, with every student using at least four languages in the course of one day, and some up to six. French, English, Luxembourgish and German were the most commonly used languages overall, with French and English most frequently used by the students as a whole. Whereas the prominence of French is likely to relate to its frequent use throughout Luxembourg, the strong presence of English is likely to relate in part to the students studying for an English major (or at least taking English courses) and reporting on their daily activities during the course of a week at university. Other languages actively used by the students included Polish, Latvian, Slovak, Portuguese, Russian, Dutch, Romanian, Hungarian, Farsi, and Albanian. Still further languages appeared in receptive form, for example listening to Italian in a restaurant, Arabic at a party, and Korean in a movie.
There is a possible effect of the social desirability bias in the very high degree of multilingualism reported. The students knew the other students and us as lecturers would be reading their diaries, and they also knew that this exercise related to a class on multilingual literacies. These two factors may have led them to over-report their multilingual practices. As this exercise was undertaken in the context of a semester-long course relating to multilingualism, however, we had the opportunity to learn a lot about the students’ multilingual practices over time, and the results for this exercise align with the highly multilingual practices they reported in other contexts throughout the course. Moreover, this high degree of multilingualism reported is not surprising given the multilingual context of Luxembourg. In the class discussion one student commented that, in doing her language diary, she had realised that one simply cannot avoid being multilingual in Luxembourg, that it is in fact impossible to stick to one language alone. This was in contrast to the experience of a cross-border student who lives in Germany and studies in Luxembourg. This student noticed a marked difference between his language practices when he was in Germany (where he largely uses German exclusively) and in Luxembourg, where he uses a range of further languages extensively.
Languages by domain
Some languages were associated with some domains more than others. English was prominent in surfing the internet, listening to music, at university, playing video games, and conversations at parties. German was often associated with listening to the radio, reading the newspaper, and watching television. Luxembourgish was frequently used for interacting with friends and family, and in text messages and the new media (using Facebook, chatting online, surfing the internet, etc). French was used for administrative purposes, at university, while shopping, reading the news, and to interact with the general public (e.g. wait staff, bus drivers). French generally occupied a broader range of domains than other languages, underlying its widespread use in everyday life in Luxembourg. Language practices also varied according to the relative formality of domains. More formal domains were generally associated with the use of one or two languages. For example, students often reported using English, French or German at university for contributing to lectures, reading books for their course, writing homework, taking notes, etc. Informal domains were associated with the use of a broader range of languages. For example, one student surfed the internet in English, Polish, Luxembourgish, German and French. While some languages such as French and English were used in both formal and informal domains, some languages were used mostly in informal domains (e.g. Luxembourgish and Slovak for text messages).
Spoken versus written distinction
There was also a distinction between languages used for speaking and listening compared to reading and writing. Some languages were used mostly for speaking and listening and less for reading and writing (for example Farsi and Russian). Other languages were used for both (for example French and English). The speaking and listening practices were also more multilingual overall; while students described remaining in one language for reading books and articles for example, they might describe watching television or listening to the radio in multiple languages in the course of an evening. An example of this is provided in Table 3. This student mainly reads and writes in English, with some reading in French and German, but speaks in Latvian, English and French, and listens to Latvian, German, Arabic, English and French. A further finding in relation to speaking versus writing was that students often engaged in written practices in one language while simultaneously engaging in spoken practices in another (or vice versa). In these cases what could appear to be a monolingual written literacy practice (e.g. reading English essays for university on the train) turns out to have been a multilingual practice (reading English essays for university while listening to Luxembourgish and French train announcements). A combination of reading and writing in one language while speaking in another was also noted in the Gujarati/English study reported in Jones et al (2000), and shows the importance of considering speaking and listening in addition to reading and writing even within studies primarily about written literacy.
[insert Table 3 here]
The distinction between speaking and writing is especially relevant for Luxembourgish. Although norms for standard written Luxembourgish do exist, these are not widely known, as students are not formally taught to read and write in Luxembourgish at school. Many Luxembourgers continue to perceive Luxembourgish as largely a spoken language and are not confident in writing it. Nevertheless, use of written Luxembourgish is increasing, particularly in the new media and among young people (see e.g. Wagner 2011). The language diaries show that the students tend to use Luxembourgish for speaking more than writing, and those that do use written Luxembourgish tend to do so for less formal written domains such as text messages or chatting with friends online. This is not universally the case, however. In the class discussion, a Luxembourgish student whose mother is a Luxembourgish teacher noted that he was familiar with norms for standard written Luxembourgish and referred to helping his mother to correct written Luxembourgish tests in his diary. This spoken/written distinction can also work in the opposite direction, with students possessing skills in reading/writing that they do not possess in speaking/listening. In the class discussion, a student noted a disparity in her listening and reading comprehension of Portuguese; she claimed to not understand a word of Portuguese when listening to people speaking it on the bus, but she can make out the meaning when she sees it written down.
Passive versus active practices
There was a further distinction between the reported active versus passive language practices of the students. In the class discussion, one student noted a distinction between her passive and active competence in Luxembourgish and German – she frequently listens to and reads Luxembourgish and German, but she would not be able to speak or write these languages. Another student, with Luxembourgish as a first language, noted that he had discovered reading skills in a language he was not aware he could understand: on Facebook he could make out the meaning of his friend’s status updates in Dutch despite not realising he had skills in this language. This example is likely to be related to the similarities between Luxembourgish and Dutch as Germanic languages, and the widespread knowledge of French and German in particular in the class suggests similar experiences would also apply to other students in the class in relation to Romance and Germanic languages.
Another feature of the diaries was their lack of explicit mention of mixed language practices such as code-switching. Some of the diaries noted several languages used during the course of one activity, e.g. ‘speaking in English and Latvian’ at a party or ‘I speak to my friends in Luxembourgish, but I use French to talk to the waitresses’ at a restaurant. This implicitly suggests code-switching, but it is not specific enough to indicate, for example, whether switches occurred within an utterance. None of the students specifically referred to such mixing in their diaries. From the students’ comments during the course and our own observations in class, we have strong reason to believe that most did engage in such forms of code-switching. Their lack of presence in the diaries may relate to a perceived illegitimacy of such practices among some students, and/or to a methodological limitation of our presentation of the diary template, namely the use of the heading ‘what language’ rather than ‘what language(s)’v. Starks and Lee (2010) represents a good example of an alternative approach to designing a language diary template that specifically indicates mixed varieties. In their study of Korean migrants in New Zealand, the diary template suggested the options Korean (K), English (E), Korean with some English (KE) and English with some Korean (EK). A similar approach would be useful in further such studies in the Luxembourg context.
Traditional and new forms of multilingualism
Perhaps the most interesting result from the analysis of the diaries relates to the theme of changing forms of multilingualism in Luxembourg. The language practices reported in the diaries enabled a general division of the students into three groups, which all relate to either traditional or new forms of multilingualism in the country.
Fehlen (2011: 152-162) describes a defined set of language skills as constituting the ‘legitimate multilingual language competence’ of Luxembourgers, demanding a subtle combination of proficiency at different levels of the three main languages of Luxembourg. This combination is: a deep knowledge of standard Luxembourgish, with borrowings from French to indicate French competence; a good knowledge of standard French acquired at school, with written French being at least as important as oral French; and knowledge of German, but not to a high level, as German is mainly to be used for writing down what is thought or said in Luxembourgish. To this traditional vision of competence is added proficiency in English as the first really foreign language, and a requirement for speakers to use their varied proficiency in these languages to accommodate to the language of their addressee. According to Fehlen, this perception of legitimate competence is firmly embedded in the mindset of Luxembourgers and long-term migrants, if often imperfectly achieved in practice. Many of the students’ language diary reflected use of Luxembourgish, French, German and English in socially appropriate and expected ways. For example, Table 4 shows a student reading a newspaper and watching television in German, speaking to her family and listening to the radio in Luxembourgish, using French to speak to shop assistants and waiters, and reading university material and listening to music in English. To a local, her diary reads as very ‘Luxembourgish’, particularly in relation to her media choices (the Luxemburger Wort newspaper, Eldoradio), shopping habits (Cactus, a Luxembourg-owned supermarket), description of attempts to use Luxembourgish in everyday interactions (‘the language I use when speaking to shop assistants or teller [sic] is mostly French, but I initially try and see if they speak Luxembourgish’) and reference to using Luxembourgish with her mother ‘because it is our mother tongue’. The presence of English as part of this model is a relatively new phenomenon in Luxembourg as a whole, but clearly well-established in the language practices of many of the students. This is likely to relate to the education level of the students – the further along students go in their education, the more likely they are to encounter this language, and to meet people with whom it serves as the only common language of interaction.
In the language diaries, this ‘Luxembourgish model’ of multilingualism is associated both with the category of students with strongly Luxembourgish backgrounds (N=2) and many of the students within the category of students with migrant backgrounds (N=8). With one exception, these students all stated Luxembourgish as (one of) their first language(s). These students, having gone through the Luxembourgish education system prior to entering university, have had the opportunity to formally acquire the set of language skills it requires. But the fact that many students with migrant backgrounds fall into this group again gives us cause to reflect upon what it means to be a ‘Luxembourger’ when it comes to language practices. Woolard (1989), in her book on the politics of language and ethnicity in Catalonia, discusses what it is to be Catalan as opposed to Castilian, and after dismissing various other potential defining features, concludes that it is speaking Catalan that allows someone to be Catalan, and that non-Catalans can effectively become Catalan by using Catalan in appropriate ways (1989: 62). The results from this analysis suggest that engaging in language practices in line with the Luxembourgish model, rather than a ‘pure’ Luxembourgish background, may also be a defining feature of what it means to be a ‘Luxembourger’ in Luxembourgvi. It is our observations of these language practices over the course of the semester that are likely to have led us to underestimate the proportion of students in our class with other national and linguistic backgrounds – because to our eyes (and ears) they clearly came across as ‘Luxembourgers’. It is important to note, however, that not all students who considered themselves to be Luxembourgish recognized themselves within this category of practices. One student claimed to seldom use German, even for domains that continue to be German-dominant for most people with Luxembourgish backgrounds, such as watching television, and claimed to use French and English to a much greater extent. Another student, while fitting the Luxembourgish model in terms of Luxembourgish, French and English, similarly showed no use of German in her language diary. Elsewhere in the course, this student spoke of her difficult experiences while learning German at school, which resulted in feelings of failure and frustration and her subsequent avoidance of the language. This underlines the aspirational aspect of this model; while representing a sociolinguistic norm, it does not describe the linguistic realities of some students.
[insert Table 4 here]
The Luxembourgish model plus migrant languages
The language of diaries of a further group of students showed that they engaged in the language practices associated with the Luxembourg model in addition to further languages associated with a migrant background. For example, Table 5 shows the language practices of a student of Russian descent, who uses French, Luxembourgish, German and English for a range of spoken and written purposes in line with social expectations in Luxembourg, but uses Russian to speak to his father. It is this use of languages other than Luxembourgish in the home/family domain that is the defining feature of this category. Other examples include the use of French, Dutch, Polish, Portuguese, German and Farsi in the home. In the cases of French and German, this is not an instance of a language addition to the Luxembourgish model, but rather the use of one of the languages included in the model but in an unexpected domain (the home). Many of these students also use their additional languages for informal written domains, e.g. use of Dutch, Farsi, Portuguese and Polish on the internet. This second group is made up of those students with a migrant background who are not included in the previous category (N=8). Three of these students have a non-Luxembourgish first language (Farsi, Russian, French), three have Luxembourgish as a first language in addition to another language or languages (Dutch, French, Portuguese) and one has Luxembourgish as a first language, but a Polish family history. These students’ language practices both reflect the traditional form of multilingualism in Luxembourg and emerging forms of multilingualism. Some of these emerging forms are likely to have more of an impact on the general multilingual environment of Luxembourg than others. For example, given the high proportion of people of Portuguese origin in Luxembourg, Portuguese is likely to have more of an impact on the linguistic environment in Luxembourg than Russian.
[insert Table 5 here]
The international model
The third model of multilingualism identifiable in the language diaries is the international model. The language diaries in this group showed frequent use of French and English for interactions in Luxembourg, as well as the use of languages related to an international background for communicating with friends and family (see Table 3 above). Five of these students used Slovak, Polish, Romanian, Hungarian and Dutch, respectively, in addition to French and English. One used mainly French, with the addition of Latvian. These practices are associated with the category of students with international backgrounds (i.e. no Luxembourgish national or language history) (N=6). This encompasses two different sub-groups: more recent migrants to Luxembourg, and students from across Europe studying in Luxembourg within the context of a student exchange. These students’ diaries thus involve the use of a wider range of languages than those used by the Luxembourgish group as a whole (e.g. Latvian, Slovak) but less use of some of the languages used in Luxembourg by the Luxembourgish students (e.g. little Luxembourgish or German). Their language practices can therefore be seen as diverging from the ‘legitimate multilingual language competence’ of Luxembourg, and represent quite a new form of multilingualism in Luxembourg. Although use of these languages will fluctuate with the coming and going of the exchange students, the presence in general of these students may have a longer-term impact on the Luxembourg language situation, given that the highly international nature of this group favours the use of a lingua franca, which – as in the case of multinational corporations and the EU institutions in Luxembourg – is often English.
Conclusion: Changing forms of multilingualism in Luxembourg
This article has reported on the multilingual language practices of a group of students at the University of Luxembourg with a range of national and language backgrounds, who reflect a similar diversity within the population of Luxembourg as a whole. The results of the language diary exercise undertaken with the students reveal a striking degree of multilingualism in the students’ language practices. The languages used by the students vary depending on domains (formal versus informal), channel (speaking versus writing), skill level (active versus passive competence), and their own social background (nationalities, first languages). The multilingual practices reported by the students take a range of different forms, some fitting within the model of multilingualism traditionally associated with Luxembourg, some extending this model, and some departing significantly from it. Fears are sometimes expressed that the increasing presence of migrants in Luxembourg presents dangers to the survival of other languages, Luxembourgish in particular. The results of this exercise, however, suggest that Luxembourgish is alive and well in the language practices of many of the students, in addition to the use of a wide range of further languages, depending on the students’ personal migration histories. The considerable diversity of language practices in combination with the students’ descriptions of their varied national and language backgrounds suggest the presence of hybrid national and language identities among the students, calling into question conceptions of what it means to be a ‘Luxembourger’, and underlining the increasingly diverse nature of Luxembourg society itself.
One limitation of this study is that it relies entirely on the validity of the language diaries that were provided by the participants and is therefore based on self-reported data rather than observations or recordings of actual language practices. The collection of the diaries in a teaching context is also likely to have influenced to some degree the validity of the diaries as research data, by for example encouraging over-reporting of multilingual practices. Such limitations could be minimised in future studies by incorporating a more ethnographic approach, and undertaking the research outside of a teaching settingvii. Nevertheless, the study has provided suggestive findings on changing multilingual practices among young people in Luxembourg, opening up directions for future further research using complementary methods.
Davis, K. A. 1994. Language Planning in Multilingual Contexts: Policies, Communities, and Schools in Luxembourg. Amsterdam: John Benjamins.
Fehlen, F. 2011. Le statut du français sur le marché linguistique du Luxembourg. Le choix de la langue comme enjeu d’un champ scientifique en devenir. In: P. Gilles and M. Wagner (eds.): Bausteine der Luxemburgistik. Frankfurt a. M: Peter Lang: 151–175.
Hoffmann, F. 1979. Sprachen in Luxemburg. Wiesbaden: Franz Steiner.
Horner, K. 2004. Negotiating the language-identity link: Media discourse and nationbuilding in Luxembourg. PhD dissertation, State University of New York at Buffalo. Ann Arbor: UMI.
Horner, K. and J. Weber 2008. The language situation in Luxembourg. Current Issues in Language Planning 9, 1: 69-128.
Jones, K., M. Martin-Jones and A. Bhatt 2000. In M. Martin-Jones and K. Jones (eds.) Multilingual Literacies: Reading and Writing Different Worlds. Amsterdam/Philadelphia: John Benjamins: 319-351.
Klein, C. 2007. The Valuation of Plurilingual Competences in an Open European Labour Market. International Journal of Multilingualism 4:4: 262- 281.
Wagner, M. (2011). Luxembourgish on Facebook: Language ideologies and discourse types on group pages. Minority languages and the Social Web. Peter Lang: 39-52.
Weber, Jean-Jacques 2009. Multilingualism, Education and Change. Frankfurt: Peter Lang.
Woolard, K. 1989. Double talk: Bilingualism and the Politics of Ethnicity in Catalonia. Stanford: Stanford University Press.
Starks, D. and J. Lee 2010. Rethinking methodology: what language diaries can offer to the study of code choice. Language Awareness 19,4: 233-248.
Statec 2011. Luxembourg in Figures 2011. Luxembourg : STATEC (Institut national de la statistique et des études économiques).
Speaking / Listening
Table 1. Language diary grid
Around half the class (12)
Around a quarter of the class (6)
Around a quarter of the class (6)
Actual number of students
Table 2. National and language backgrounds of students
Speaking / Listening
1. having breakfast with some roommate’s friends from Latvia
2. discussing the study questions for a test with some friends in the train
10. meeting other people in my residence, joining their party
Reading instructions of a game in German
Talking in English, listening to English and French
Listening to music in French, talking in English
Listening to conversations in Latvian, German, Arabic, English; speaking in English and Latvian
Table 3. Example of language use in speaking versus writing
Speaking / Listening
When I get up, I listen to the radio while having breakfast or I am reading the newspaper.
Going out with the dog while listening to some music.
The Newspaper I read is mostly in German (=Luxemburger Wort)
The Radio I am listening to is in luxemburgish (=Eldoradio)
The Music I am listening to is mostly in English, but I also listen to some German or French artists.
Going to "Cactus" to do some grocery shopping.
Visiting my grandmother.
Reading a book.
The products I buy are mostly labeled in French or German, except some products from luxembourgish territory.
It is in English, mainly because I have to read it for one of my classes, but I tend, in general, to read mostly in English.
The language I use when speaking to shop assistants or teller is mostly French, but I initially try and see if they speak Luxembourgish.
We talk to eachother in Luxembourgish.
Having dinner with my parents.
Watching some Television.
Going to a pub with some friends where a band is playing.
We converse in Luxembourgish, because it is our mothertongue.
Mostly German channels, like RTL, Pro7, etc.
I speak to my friends in Luxembourgish, but I use French to talk to the waitresses.
The music that the band is playing is in English.
Table 4. Language diary reflecting the Luxembourgish model
Speaking / Listening
I woke up and listened to the radio.
I met a director and a cameraman to discuss the shooting of a documentary.
We wrote each other text messages in Luxembourgish
I was listening to the Luxembourgish radio stations( RTL,Eldoradio).
I spoke with both Luxembourgish.
We went to Subway for lunch.
The actors arrived.
The shooting of the documentary.
After we finished we went to a pub.
The menu was in French and English.
The contracts were in French.
I ordered my sandwich by using French.
I spoke with them Luxembourgish and German.
During the shoot I spoke mainly Luxembourgish with the cast and crew.
The waitress spoke Luxembourgish and so I used Luxembourgish to order a drink.
I arrived at home.
I watched a little TV.
I watched a movie.
I read a book for class.
The ads of the channels were mostly German or English.
I use Russian to speak with my father.
I watched mainly German or English television.
Table 5. Language diary by student with a migrant background
i Luxembourg has a two-track secondary education system, where students can go to either the lycée classique (classical secondary school), for a more academically-oriented education, or the lycée technique (technical secondary school), for a more vocationally-oriented education.
ii http://wwwfr.uni.lu/etudiants/informations_utiles_de_a_a_z/multilinguisme, accessed 27 April 2012
iii International students made up 2,370 (41.6%) of the total of 5,700 students in 2012. (http://wwwfr.uni.lu/universite/presentation/chiffres_cle, last accessed 27 April 2012).
Whereas, in some other parts of the world, the term ‘ethnicity’ would be used here, use of the term ‘nationality’ is a convention in Luxembourg. Although these terms arguably have quite different meanings (one can have a much broader range of ethnic affiliations than passports), we felt that using ‘ethnicity’ would be likely to confuse the students.
v We thank one of the anonymous reviewers for making this point.
Note that Horner (2004) draws a similar conclusion from a comparison between the Luxembourg and Catalan language situations, claiming that ‘as is the case in Catalonia, language currently constitutes the most salient symbol of ethnic group identity in Luxembourg’.
vii Starks and Lee (2010: 244) claim that language diaries can be particularly useful as pilots for future ethnographic research. The same could be said about the present research.