Colloquial and Academic Language in Interaction: Students’ Linguistic Strategies During a Collaborative Task in History Class Colloquial and Academic Language in Interaction

In recent years there has been a growing interest in the topic of academic language in the context of educational success of (mostly) immigrant children . However, most studies focused on academic language only and did not consider the role of other linguistic resources such as colloquial language for the development of academic language . In this paper, we will discuss the interplay between both registers by presenting an analysis of students’ (n = 3) utterances during a collaborative task in history class . Data was collected in a College for Higher Vocational Education (upper secondary level; ISCS 10) in Vienna where one of the researchers was working as a history teacher . The interaction analysis and interpretation focus on strategies students employ to solve the exercises. The following strategies are identified and explained: mutual explanation for a better understanding of, e . g ., technical terms, collective planning and monitoring of the writing process as well as orientation along the structure of other texts . Since colloquial and academic language seem interrelated in these strategies, this paper combines the concept of academic language and linguistic repertoire .

In recent years intensive discussions have revolved around obstacles for the academic success of students whose language use does not comply with language use at school, focusing especially on immigrant students and students from families with low socio-economic status (Cummins, 2013;Gogolin & Lange, 2011;Schleppegrell, 2004) . In this context, supporting the acquisition of the language/s of schooling is often seen as a key factor, whereas a special focus lies on the academic register of the particular language/s of schooling that are necessary to master school related tasks (Cummins, 2008;Quehl & Trapp, 2013, pp . 13-25;Schleppegrell, 2004, p . 21) .
In this paper, we examine the interplay between academic language 1 and colloquial language during collaborative processes in a linguistically very diverse class by analysing two interactions . These interactions were recorded during a group-work 1 The term "academic language" refers to the academic registers of the particular language/s of schooling which, according to Cummins (2008), are important to academic success, in German "Bildungssprache" (Gogolin & Duarte, 2016). We follow Halliday's (1978) notion of register as a situation-specific variation of language. phase during a history lesson in a College for Vocational Education (upper secondary level; ISCS 10; Berufsbildende Höhere Schule) in Vienna, Austria . The students recorded themselves with their mobile phones during a collaborative task and sent the recordings to their teacher, who is also one of the authors of this paper (for further information, see section 3) .
The purpose of this study was to better understand the linguistic practices and especially the role of academic language in an interactional setting in which the teacher does not intervene unless the students ask for it explicitly . The following section provides the theoretical background of our study, focusing on two core concepts: "academic language" and "linguistic repertoire" . The underlying assumption of this study is that academic language, conceptualised as a register, is not isolated from other linguistic resources in the repertoire of speakers . The analysis represents a first step to combine these concepts from an interactional usage-centred perspective .

Core concepts
As mentioned above, two theoretical concepts inform our analysis of the data: the notions of academic language and linguistic repertoire . The synthesis of these two concepts forms the analytical lens for the examination of the data .

Academic language
When children enter the school system, they encounter linguistic practices that differ from the linguistic practices of family communication in aspects of explicitness, complexity and cognitive demand (Schleppegrell, 2004, pp . 7-16) . Children growing up with the language of schooling in this stage can rely on their colloquial language as being a good basis to cope with this new linguistic demands during their first years in primary school (Michalak, Lemke & Goeke, 2015, p . 49), multilingual children on the other hand experience not only the differences in explicitness, complexity and cognitive demand but also a gap between the multilingual practices of their everyday life and the monolingual practices in school . These different challenges need to be taken note of, especially as the language children need for school-related tasks becomes more complex over the years (Schleppegrell, 2004, pp . 1-4) . This part of the linguistic repertoire -the academic language -is especially important for performing in formal education (Gogolin & Lange, 2011;Heller & Morek, 2015), as it is a register used for "presenting information in highly structured ways, and in ways that enable the author/speaker to take an assertive, expert stance toward the information presented" (Schleppegrell, 2001, p . 451) .
The concept of academic language, "Bildungssprache" in German, is based on Halliday's (1994) Functional Grammar, Bernstein's (1971) work on class-specific linguistic socialisation as well as on the concept of basic interpersonal communicative skills (BICS) and cognitive academic language proficiency (CALP) established by Cummins (2008, p . 71): […] in order to draw educators' attention to the timelines and challenges that second language learners encounter as they attempt to catch up to their peers in academic aspects of the school language. BICS refers to conversational fluency in a language while CALP refers to students' ability to understand and express, in both oral and written modes, concepts and ideas that are relevant to success in school .
While Cummins (1979Cummins ( , 2008 focuses on school environments, Michalak, Lemke, and Goeke (2015) find academic language to be important for knowledge transfer in every educational context . Heller and Morek (2015) further distinguish three different functions of academic language, namely academic language "as a medium of knowledge transmission (communicative function); […] as a tool for thinking (epistemic function); [and] […] as a ticket and visiting card (socio-symbolic function)" (Heller & Morek, 2015, p . 175;italics in original) . These three functions imply that academic language is much more than specialised lexical knowledge, as it also includes special (linguistic) performance abilities such as knowledge of certain school-specific text genres (Feilke, 2014;Heller & Morek, 2015;Schleppegrell, 2004, pp . 82-112) .
Key characteristics of oral and written academic language are aspects of written language mode . These aspects are explicitness, decontextualisation and complexity with specific features on a lexical, syntactical and discursive level such as nominalisations and elaborate noun phrases, complex clauses and specific stylistic standards (Heller & Morek, 2015, p . 176;Herzog-Punzenberger & Schnell, 2012, p . 234;Koch & Oesterreicher, 1985) . Table 1 gives a rough overview of different features of academic language according to Heller and Morek (2015, p . 176): To succeed at school, students need a high level of academic proficiency in the language/s of schooling (Schleppegrell, 2004, pp . 39-42) and schools are responsible for supporting students in acquiring it (Gogolin & Duarte, 2016, p . 480) but seem to fail to achieve this goal in more and more cases . In Austria, for example, this weakness is indicated by the fact that the number of children who drop out of school is significantly higher amongst children who speak languages other than German at home than among children who grow up with German, even taking into account differences in students' socio-economic status (Herzog-Punzenberger & Schnell, 2012, pp . 252-255;Vetter, 2015, pp . 238-239) . Thus, many experts argue that academic language should be considered at all school levels and in all subjects (Gogolin & Lange, 2011;Quehl & Trapp, 2013;Brandt & Gogolin, 2016) . 2

Linguistic repertoire
The analytic scope of this study also implies the sociolinguistic concept of linguistic repertoire in order to contextualise academic language within other linguistic resources incorporated by the students whose interactions we examine .
The term "linguistic repertoire", also termed "verbal repertoire", has undergone distinct developments and theoretical framings: It was introduced by John Gumperz (1964) and initially focused on the situated employment and functions of all linguistic resources -for example named languages, dialects or registersduring interaction in a given speech community . These linguistic resources are seen as fluidly interrelated and linguistic behaviour in interactions and language choice as restricted by grammatical and social norms: Ultimately, it is the individual who makes the decision, but his freedom to select is always subject both to grammatical and social restraints. Grammatical restraints relate to the intelligibility of sentences; social restraints relate to their acceptability . (Gumperz, 1964, p . 138) Recent publications (Blommaert & Backus, 2013;Busch, 2013) include general societal and global developments to elaborate a differently nuanced definition. Backus and Blommaert (2013) base their understanding of linguistic repertoire on the concept of super-diversity (Vertovec, 2007) and the acknowledgement that 2 However, other languages that students bring with them should also be considered, especially as "multilingual resources offer the potential to support […] school-based learning [and] it is difficult to make the most effective use of these resources in an education system that assumes a monolingual and monocultural bias" (French, 2016, p . 229) . communicative practices can no longer be analysed through the lens of earlier understandings of language (Blommaert & Backus,p . 14) . As a result, they call for usage-based approaches to communication . Patterns of language use and learning these days are more flexible or polycentric. 3 The consequences of their perspective are the acknowledgement that knowing a language never means knowing "all the resources of language" (Blommaert & Backus, 2013, p . 15) and that repertoires are "individual, biographically organized complexes of resources" that are tied to various learning contexts (Blommaert & Backus,p. 21) and influence the ways individuals use their linguistic resources .
Apart from Blommaert's and Backus' (2013) understanding, Busch's (2013) redefinition of the notion of linguistic repertoire implies a poststructuralist perspective . Distancing herself from merely observing the repertoire as part of interactions whose "rules and conventions" (Busch, 2013, p. 22) are examined, she implies the subjective perspective, in German "Spracherleben" in three dimensions: the embodied, the emotional and the historical-political dimension (Busch, 2013, pp . 22-23) . As a result, she argues that a particular methodology is necessary to access the intertwined relationship between linguistic resources and their deeper subjective meanings . In her work, she combines a biographical and multimodal approach to do so (Busch, 2013) .
As the analysis of the data is conducted from an interactional point of view, it is not possible to include the subjective and ideological embeddedness that are highlighted in more recent definitions of the linguistic repertoire, although such an approach would have great potential and could open a wide field of reflection. This study represents a first step in this direction -starting from the interactional approach.

The empirical study -sample, context and research questions
for this aspect when planning learning processes in class . Hence, two interrelated questions guided the research process: 1. Which strategies of language use do students employ in reading and writing school-specific texts in a group-work task? 2. How do these strategies reflect the intertwined relationship between academic language and colloquial language within students' linguistic repertoire?
The data for this study was collected in May 2016 at a College for Higher Vocational Education (upper secondary level; ISCS 10) in Vienna, Austria . As in many Viennese schools, the observed class is linguistically diverse . Apart from German, other languages observed included Albanian, Bosnian/Croatian/Serbian, Macedonian, Hungarian, Italian, Tagalog, Punjabi and Slovakian. At the time of recording, there were 26 students (between 16 and 17 years of age) in the class and they were asked to complete a collaborative task in groups of two or three . The students themselves recorded their conversations with one mobile phone per group . The teacher, who is one of the authors of this paper, informed the students who wanted to participate and their parents several weeks in advance that the data would be used for academic purposes only . As the classroom is "polycentric" (Blommaert & Jie, 2006, p . 35) and there were several interactions going on simultaneously, the teacher -who was at the same time the researcher -considered it useful for each group to make a separate audio recording . Furthermore, the mobile phone as a recording tool was regarded as less intrusive and more compatible with the students' habits than, for example, a Dictaphone . Moreover, the students voluntarily handed in their recordings . The teacher/researcher received four recordings via e-mail, two of which were analysed . These two were chosen, because they were recorded from the beginning to the end of the group-work process . Although the teacher had explained that all parts of the interaction had to be recorded, two out of four recordings contained only the results of their work and not the working process . The chosen recordings, which we refer to as recording 1 (Daniela, Marion, and Silvia 4 ) and recording 2 (Milana and Manuel), are 25:08 and 25:22 minutes long, respectively, and were transcribed using EXMARaLDA 5 following Hoffmann-Riem's (1984) conventions 6 . In these simple transcripts, "paraverbal and non-verbal elements of communication are usually omitted . The focus of simple transcripts lies on readability (Dresing, Pehl, & Schmieder, 2015, p . 23) . As we aimed to identify how the students gain a better understanding of content-related knowledge, we decided to approach categorisation inductively. The two categories formed refer to the identified students' strategies of language use and the interaction between academic and colloquial language therein . For interaction analysis, we selected conversational sequences in the two recordings that contained (often mutual) explanations of concepts, technical terms, 4 All student names have been replaced with pseudonyms . image content and extracts from texts (strategy/category 1) as well as collaborative development of the text (strategy/category 2) . 7 In section 4 we present our main findings relying on selected conversational sequences from recording 1.
Concerning the content of the history class, students had to deal with a group-work task consisting of several partial tasks on "lèse-majesté" (in German: "Majestätsbeleidigung") in Austria in the 19 th and 20 th century . The task was taken from the website habsburger .net, launched by Austrian historians, which provides a large number of tasks and texts for history lessons . The teacher's choice was based on a specific principle of history didactics -to relate present political themes and the students' life-world to historical content (Bergmann, 2012) -in this case the ongoing election campaign for the second round of the Austrian presidential election . Moreover, as the task aimed at questioning authoritarian power, the teacher/researcher considered it an appropriate document for this purpose . From a language-centered point of view, the multimodal task consisted of reading various types of texts written in academic language, including one original text segment from the beginning of the 20 th century and explanations written in a style similar to regular text books in history class . Moreover, there was also one partial task that demanded Internet research . Therefore, the teacher/researcher assumed that it would be necessary for the students to deal with different text genres in academic language and to "translate" their impressions and findings from the texts into spoken language and vice versa.
The chosen task 8 consisted of five different parts with various objectives: First, the students should -with the help of the Internet -define and contextualise the notion of "lèse-majesté" and find out which punishment is imposed upon an offender . Second, they should compare portraits of the Austrian Emperor Franz Joseph I and the then 9 Austrian president Heinz Fischer in order to discuss who seems more reverent and, thus, to give a "translation" from visual impressions into spoken language. The third task contained an excerpt with the description of a "lèse-majesté" and ended with the question: "Should people rise when they hear a national anthem?" In the fourth part, students should, with the help of the Internet, find out how the notions "Hump-Dump" and "Kurti" relate to the former Austrian presidents Kurt Waldheim and Thomas Klestil. 10 Finally, the last task leads to a test with the title "Culturally open? Gestures and facial expressions", where pupils had to link certain gestures and facial expressions to nations, states and/or continents in the 7 For further reading on strategy use with an emphasis on writing see Budde and Michalak (2017, pp . 26-27) . 8 http://www.habsburger.net/de/unterricht/module/eine-majestaetsbeleidigung [21/12/2016] 9 At that time, the campaign for the presidential elections had just started. Heinz Fischer's mandate ended in July 2016 . 10 "Kurti" is the name of a song (album: Burli, 1987) in which the Austrian band Erste Allgemeine Verunsicherung criticises Kurt Waldheim i. a. for allegedly not being able to remember what he had done as a Wehrmacht officer during World War II. Kurt Waldheim intended to file charges against the band for defamation (in German: "Ehrenbeleidigung") . In 2000, the politician Hilmar Kabas from the right-wing Austrian Freedom Party (FPÖ) called president Thomas Klestil "Lump" (in English: "rascal") . After being criticised, he claimed to the APA (Austrian Press Agency) that he had not said "Lump" but "Hump" or "Dump", neither of which exists////has meaning in German .
world . The solution of the test showed that the same gestures and facial expressions are used in different countries around the world but differ in their meaning . When the teacher explained the group-work task, she explicitly encouraged the students to use languages other than German, the officially approved classroom language in a history class (besides English at this school) . Nevertheless, none of the students used other languages . Recording 2 (Milana and Manuel) contains some utterances in English that are popular in German youth language . Apart from that, Milana and Manuel shared Serbian as a common language, but did not use it during the recorded interaction .

Results
We identified two main strategies of language use in the data: mutual explanations of sequences in academic language and collaborative development of the text . Strategies of language use in this context are verbal and nonverbal actions to cope with problems in comprehension or in communication (Demme, 2010) .

Mutual explanations for a better understanding of the task
Cummins (2008) sees collaborative learning as an important tool to help students improve their academic language skills, because talking about texts can help students to "internalize and more fully comprehend the academic language they find in their extensive reading of the text" (pp . 79-80) . Indeed, our data provides various sequences in which students help each other by explaining specialised terminology or whole passages of texts they come across while solving the task together . Extract 1 shows an example of such a mutual explanation: Extract 1 (recording 1): No no . It is not as a word . . you were insulted by saying lèse-majesté. It was just if you disdained someone, that you exposed him, treason today . 6 Daniela: Ja .
This example shows that the technical term lèse-majesté is not quite clear to Marion (lines 2 and 4), so the others try to explain it to her . However, it is not only the term that is unclear, but also the concept of lèse-majesté as a whole. This passage is a good example for how "linguistic and content knowledge […] have to be seen as a didactic entity" (Handt & Weis, 2015, p. 76) and that students need to understand the concepts behind technical terms in order to better internalise the terms themselves. To explain the concept of lèse-majesté to Marion, Silvia (line 5) applies the strategies of contextualisation ("It was just if you disdained someone, that you exposed him") and comparison with a contemporary concept ("treason today") .
The explanations are not limited to the words and passages of the input text the students read in order to accomplish the task, but also cover the task instructions themselves: Extract 2 (recording 1): Not anymore at all … So it means ((reads)) is in a monarchy the wilful insult or ass-assault that is committed against a reigning monarch . It is a transgression against the inviolability of the holder of the state sovereignty that in constitutional monarchies is constitutionally codified.
In this extract, Marion misunderstands the task instructions (line 2), as she thinks she needs to explain the term "lèse-majesté" instead of just researching information about its historical context . Only when Daniela (line 4) and Silvia (line 7) explain the task in their own words she understands that they need to find the definition online and starts reading the article about "lèse-majesté" on Wikipedia 11 . This leads to the first step towards the solution of the task by Silvia, summarising the passage (expression 10) previously read aloud by Marion (expression 9) in her own words . To better understand the importance of these processes, Gallin's and Ruff's (2010) differentiation (based on Wagenschein, 1980) seems very useful: They distinguish between the language of understanding (Sprache des Verstehens), a language form that correlates with the linguistic resources available to the learner (e . g . colloquial language), and the language of the understood (Sprache des Verstandenen), which is explicit, decontextualised and complex -academic language . Velasco and García (2014) also point to the importance of the interaction between different registers, stating that "[a]dding and integrating new linguistic resources cannot be done without reference to those linguistic resources the child [or student] already has" (21) . Extracts 2 and 3 analysed above are good examples for how students can help each other to bridge the gap between already-known and as-yet-unknown linguistic structures, that is between their colloquial language "of understanding" and their academic language "of the understood", respectively . Beese and Roll (2015, p. 53) understand "writing as a thinking tool". When writing, students have to assess, arrange and relate their knowledge . This process -decelerated through the writing -can lead to a deepened reflection of the content and thus to its better understanding . In order to successfully complete a writing task, students need to be able to realise several aspects . Firstly, they have to identify the text type adequate to the task and/or expected by the teacher . Secondly, they need to recall the corresponding text model, and, thirdly, they have to write a linguistically and structurally adequate text (Bachmann, 2014) . Feilke (2014) states that the quality of a text is not only based on its individual creative and linguistic arrangement but depends very much on the knowledge of text types, text structures and writing strategies . The four examples presented below were recorded during a collaborative writing task and illustrate writing strategies . They seem to support Storch's (2005, p. 168) findings that collaborative writing results in texts that show "greater grammatical accuracy and linguistic complexity" and are "more succinct" .

Collaborative development of the text
In the recordings, several strategies for developing a linguistically and structurally adequate text can be identified. On the structural level, one such strategy is orientation along the structure and characteristics of another text: Extract 3 (recording 1): In extract 3, Marion sees the input text, an article from Wikipedia, as a text model for their own text about "lèse-majesté" and suggests adopting the element of quotations for their own text, a stylistic element one could see as characteristic of academic texts . Another strategy on the structural level is the collective planning and monitoring of the writing process as shown in extract 4: Extract 4 (recording 1): Oh yes where we took our sources from .
And we have to add our sources .
In extract 4, Silvia starts checking whether their text includes all required information (line 1) . Initiated by Silvia, Daniela starts thinking about what might still be missing as well and consequently wants to add one piece of historic information (line 2) and reminds the others that the task instructions explicitly require them to indicate their sources (line 4) .
An important writing strategy on the linguistic level can be observed in extract 5: Extract 5 (recording 1): 1 Silvia: Verbrechen der Majestätsbeleidigung wurde auch gegenüber Gott verwendet .
The felony of lèse-majesté was also used in relation to God . 2 Marion: Also es ist dasselbe, es wurde gleichgesetzt .
So it is the same, it was equated .
In this case, Silvia summarises one passage of the input text in her own words (line 1) and Marion 'translates' the colloquial phrase "so it is the same" into "it was equated" (line 2), a phrase more suitable for a text in academic contexts .
A similar process can be observed in extract 6, in which Silvia and Daniela start comparing a portrait of Emperor Franz Josef I to a portrait of the Austrian president Heinz Fischer: Extract 6 (recording 1): Silvia (line 1) notices that both pictures show upper parts of the body and defines these kinds of visual representations as portraits, a content word we would argue attributable to academic language rather than to colloquial language . This process of increasing specification is essential for the production of academic texts (Brandt & Gogolin, 2016, pp. 28-29). We would see extract 5 and extract 6 as good examples for mutual explanation and collaborative development of the text moving back and forth along the continuum between Wagenschein's (1980) colloquial language of understanding (Sprache des Verstehens) and the academic language of the understood (Sprache des Verstandenen) .

Discussion
The scope of this study are students' communicative practices in a group-work task, focusing on their strategies of language use and negotiation of meaning in writing school-specific texts, specifically the interaction between students' colloquial and academic language within their linguistic repertoire . The examination of selected sequences recorded during a history lesson in a College for Vocational Education (upper secondary level; ISCS 10; Berufsbildende Höhere Schule) in Vienna shows that language use in this specific group-work situation is neither academic nor colloquialthe students regularly change their way of speaking and in doing so move along the continuum of academic and colloquial language . Frequently used strategies to gain a better understanding of the sometimes linguistically challenging historical input texts are mutual explanations of technical terms and the use of colloquial language as well as contextualisation to gain a better understanding of concepts in general . The students also apply similar strategies in their collaborative writing process, for example when they "translate" colloquial phrases into phrases more suitable for a text in academic contexts . Other strategies they apply are the collective planning and monitoring of the writing process as well as orientation along the structures of other texts . Both strategies support Feilke's (2014) and Bachmann's (2014) calls for a stronger focus on text models and writing procedures (Schreibprozeduren in German) in education, meaning that text conventions need to be made transparent and explicitly practised in school .
These results call for a greater recognition of the role of colloquial language in the acquisition of academic language and the understanding of topic-specific contents . Thus, if teachers were more aware of the different roles colloquial and academic language play in the acquisition and the organisation of knowledge, they could facilitate the understanding and the acquisition of academic language by actively calling on students' movement back and forth the continuum between col-loquial and academic language for creating meaning . The concepts of the language of understanding (Sprache des Verstehens) and the language of the understood (Sprache des Verstandenen) as well as their interdependence have proven very useful in this context (Gallin & Ruf, 2010) .
However, our results should be regarded as pieces of a bigger puzzle . Further analysis of interactions in collaborative processes is needed in order to gain a better understanding of the intertwined relations between academic language and other linguistic resources combined in a repertoire . Gumperz (1964, p . 138) stated that the repertoire, conceptualised as an arsenal, "[…] provides the weapons of everyday communication [and that] speakers choose among this arsenal in accordance with the meaning they wish to convey" . For the school context, our results indicate that students might also choose in accordance with the cognitive process they want to achieve .
Finally, conceptualising academic language as a dimension of a large and dynamic linguistic repertoire requires discussing the implementation of new research methods. When highlighting the subjective and ideological dimensions of the linguistic repertoire (see section 2 .2 .), e .g . (auto)biographical methods may need to be implemented in the research process . Such an approach would also allow us to react to criticism that especially register-based research that seeks to systematically describe academic language tends to ignore socio-symbolic functions and the link between language use and social positioning (Heller & Morek, 2015, p . 179) . In conclusion, it should be emphasised that this is merely one example of how theoretical concepts from applied linguistics and pedagogy can be combined in order to better understand communicative practices in learning contexts .