DOI 10.24053/ FLuL-2023-00022 52 • Heft 2 J ULES B UENDGENS -K OSTEN , F REDERIK C ORNILLIE , S HANNON S AURO * Teaching (multi)literacies, supporting multilingual identities Plurilingual writing in an interactive fiction project Abstract. The EFL classroom can be a place in which students develop target language skills and overarching plurilingual competencies, but also their multilingual identities. Digital games - including interactive fiction (IF) - may play a role in this context, as participation in digital games and gaming practices has been claimed to afford identity work. This paper is based on a follow-up study for the “FanTALES” Erasmus+ project. Drawing on IF stories created in a pedagogic intervention and on follow-up focus group interviews, it finds that multilingual storytelling in an interactive fiction context was challenging for students, even though they self-assessed their productive plurilingual competencies as fairly high, and that the writing task itself was only partially successful in creating a ‘translanguaging space’ in which all linguistic resources could be used and valued. 1. Introduction Teaching language and teaching literacy are intertwined, as they both involve preparing learners to use a wide range of semiotic systems, productively and receptively. Or, as the D OUGLAS F IR G ROUP put it: “language learning is semiotic learning” (2016: 27). In this article, we draw on the N EW L ONDON G ROUP ’s notion of multiliteracy, the ability to use a wide range of semiotic resources in a wide range of settings. In their article, which suggested a reconceptualization of literacy and an approach to teaching multiliteracies, they draw on two sources of societal change for their arguments: “the * Addresses for correspondence: Dr. Jules B UENDGENS -K OSTEN , IEAS, Goethe-Universität Frankfurt, Norbert-Wollheim-Platz 1, 60323 F RANKFURT / M. E-Mail: buendgens-kosten@em.uni-frankfurt.de Research areas: Computer-assisted language learning, Multi/ plurilingualism, Inclusive education Dr. Frederik C ORNILLIE , Itec, KU Leuven & imec, Etienne Sabbelaan 51 box 7800, 8500 K ORTRIJK E-Mail: frederik.cornillie@kuleuven.be Research areas: Computer-assisted language learning, Instructed second language acquisition, Games and play Associate Prof. Dr. Shannon S AURO , Department of Education, University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, B ALTIMORE , MD 21250, USA E-Mail: ssauro@umbc.edu Research areas: Fan practices, Computer-assisted language learning, Virtual exchange Teaching (multi)literacies, supporting multilingual identities 73 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 context of our culturally and linguistically diverse and increasingly globalized societies, for the multifarious cultures that interrelate and the plurality of texts that circulate” as well as “the burgeoning variety of text forms associated with information and multimedia technologies” (N EW L ONDON G ROUP 1996: 61). These two dimensions might - partially - be discussed under the labels of ‘multilingualism/ plurilingualism’ and ‘digitalization’. In this article, we will look at an pedagogicall intervention in the English as a foreign language (EFL) classroom that is situated in the intersection between digitalization and plurilingual practices, and at the impacts of this intervention on both identity and plurilingual competencies. In the next section, therefore, we will discuss storytelling practices that lie at the foundation of this intervention. 1.1 Storytelling in the EFL classroom: Interactive fiction In the EFL/ ESL classroom, creative writing and storytelling have been used to support the development of language skills and literary knowledge (cf. S AURO / S UNDMARK 2016). One type of creative writing of increasing interest to EFL instructors is interactive fiction (IF), playful narrative texts with which the reader can interact in some way, for example by playing the role of a character or by manipulating other elements of the world depicted in the story through mechanics that are associated with the design of games, often resulting in a non-linear narrative experience (for a more restrictive definition, see P LOTKIN 2011). While IF predates the digital age (cf. choose-your-own-adventure books), the reading and writing of IF is typically mediated by digital technologies. Moreover, accessible authoring tools like Twine (twinery.org) have put the creation of IF within the hands of hobbyist writers, and are also being adopted for educational purposes, including for the teaching of foreign languages in multilingual classrooms (e.g. C ORNILLIE et al. 2021). 1.2 Storytelling in the EFL classroom: Multilingual storytelling Writing multilingual texts deserves a space in the foreign language classroom for two main reasons: Being able to draw on multiple languages to transport a message, for rhetorical effects or humor, is a relevant language skill (C OUNCIL OF E UROPE 2018: 161-162), and opportunities to draw on a broad range of language skills might impact the (multilingual) identity of learners. Identity, understood here as “the way a person understands his or her relationship to the world, how that relationship is constructed across time and space, and how the person understands possibilities for the future” (N ORTON 2013: 4), is an important concept in the teaching of EFL, as it contributes directly and indirectly to language learning success. There is a growing awareness that looking at identity with a focus on only one language at a time might not give the full picture. D ÖRNYEI ’s popular “L2 self system” model, which focuses on a person’s self guides, i.e. self-directed standards or acquired guides for being” (H IGGINS 1987 p. 321), for example, has been 74 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 extended from L2 contexts to “situations where bilinguals/ multilinguals learn additional languages, or where monolinguals are learning two or more L2s” (H ENRY 2017: 549). This system of relevant, interacting self-guides forms part of learners’ Multilingual Identity System (H ENRY 2017, 553), or, shortly, their multilingual identity. Henry discusses how the Ideal Multilingual Self emerges from the interaction between different Ideal Selves related to specific languages, and how different higher-order self guides might impact language learning in different ways (H ENRY 2017, 552-558). Multilingual identity, i.e. aspects of a person’s identity related to being or becoming a speaker of several languages (as opposed to aspects of identity related to each specific language, seen in isolation), has been found to influence foreign language learning outcomes (H ENRY / T HORSEN 2018) and pedagogical interventions can impact multilingual identity and related concepts such as plurilingual self-aspirations (F ORBES et al. 2021). Multilingual writing is relevant to identity as well. P RADA (2022) describes a university-level Spanish composition class for Spanish (heritage language/ native language) speakers, in which students engaged in digital storytelling with a focus on bilingualism. P RADA argues that digital stories can serve as (digital) translanguaging spaces (defined by W EI 2011 as spaces for translanguaging, and spaces created by translanguaging), supporting student empowerment. C UMMINS (2007) uses the term “identity texts” to refer to (often bilingual) texts where learners “invest their identities in these texts […] that then hold a mirror up to students in which their identities are reflected back in a positive light.” (C UMMINS 2007: 60). He pointed out that digital formats, with their ability to reach a broader audience, can amplify the impact of multilingual writing (C UMMINS 2007: 71). Research on fanfiction has pointed to identity as a motivating factor for multilingual creative writing in the digital wilds. This is illustrated in the case study of Nanako (T HORNE / B LACK 2010) and a case study of teenage Finnish fanfiction (L EPPAÄNEN 2007). Similarly, participation in digital games and gaming practices has been claimed to afford identity work. B ARAB / G RESALFI / I NGRAM -G OBLE (2010) explore the relation between playing narrative digital games and identity development in educational contexts. They have tested the hypothesis that ‘transformational play’, requiring a player’s agentic and consequential role in simulated and narratively rich experiences, aids both conceptual understanding in complex problem solving activities and the development of “potentially more enduring outcomes, such as the development of identity or affiliation” (B ARAB / G RESALFI / I NGRAM -G OBLE 2010: 525). B UENDGENS - K OSTEN (2022: 139) argues that this notion of “transformational play” might also be conceptually transferred to apps and games for language learning, where “language learners can play around with language and language use, and experience themselves as somebody who uses language(s) for specific communicative or non-communicative goals, achieving outcomes within the game world”. R EINHARDT (2019: 129) conceptualizes identity work in gameful language learning and teaching as “the actions and behaviors integrating one’s newly forming L2 identity into existing identities”, for example by shifting between the player’s out-of-game stance, their stance as a Teaching (multi)literacies, supporting multilingual identities 75 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 player, and the stance of their avatar. IF, as a form of digital game, equally affords learners such identity work. Specifically, by playing the role of characters using multiple languages in meaningful ways or by envisioning multilingual roles for readers, learners can engage in identity work that contributes to their multilingual identity. In this paper, we argue that creating a translanguaging space (W EI 2011) in EFL settings through writing and reading multilingual IF has the potential to support learners’ higher order self-guides related to multilingual identity, by supporting learners in developing more elaborate future selves and supporting an Ideal Multilingual Self connected to all languages in a learner’s repertoire (including those that learners plan to study in the future). 1.3 Research questions Accordingly, this project investigates the following research questions in the context of an EFL classroom intervention that involved reading and writing multilingual IF based on concepts and materials from the FanTALES project (S AURO / B UENDGENS - K OSTEN / C ORNILLIE 2020): • Does a pedagogical intervention based on the reading and writing of multilingual IF have the potential to impact plurilingual competencies (in the sense of “Building on a plurilingual repertoire” and “Plurilingual comprehension”, C OUNCIL OF E UROPE 2018: 160-162)? • Does a pedagogical intervention based on the reading and writing of multilingual IF provide a context for engagement with multilingual self-aspirations and/ or multilingual identities? 2. Methods The study used a pretest/ posttest design with a delayed treatment control group (see Fig. 1,  page 76), enriched by a collection of stories and other materials created during the pedagogical intervention, and by focus group discussions with students and interviews with teachers. 76 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 Fig. 1: Research design Fig. 2: Structure of the intervention The five-lesson intervention (see Fig. 2) included lessons raising awareness for languages within students’ communities and lives, and in the book they were reading, introducing students to IF (including a IF sample story featuring multilingual storytelling), time for multilingual IF writing with the digital authoring tool Twine, and for peer feedback and revision. Homework included observation and reflection tasks (such as linguistic soundscaping activities), or writing text sections (such as imagining the linguistic landscape and soundscape at different settings mentioned in the books). In small groups, students created an IF based on the book they were reading (grade 9: Coast to Coast by David Fermer, grade 10: 13 Reasons Why by Jay Asher, both chosen by their teachers). Students were asked to focus on a situation in which a character has to make an important, meaningful choice, to demonstrate their understanding of the story, setting and characters, and to include descriptions and dialogues in their Teaching (multi)literacies, supporting multilingual identities 77 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 IF. Additionally, they were asked to include three to four choices for the player, and a maximum of three to four endings. This requirement and a sample structure that students could use were provided to reduce the risk of combinatorial explosion, where each passage branches into two or three new passages, resulting in a large number of passages but short play times. Students were instructed to include at least three different languages, “in a way that makes sense for the story” and were provided with suggestions. 2.1 Sample The study took place at a comprehensive school (Gesamtschule) in Rhineland-Palatinate. Four classes, two 9th grade and two 10th grade classes (78 students), and their 2 teachers, participated. The average age in classes A and B (9th grade) was 14.8; in classes C and D (10th grade) 15.9 and 16.0. In a demographic questionnaire, students were asked to identify all the languages that played a role in their lives, and to provide additional information on this. 70 students identified German as (one of their) mother tongues (Muttersprache), three students Russian, two students each Moroccan (Arabic) and Turkish, one student each Albanian, English, Italian, Persian, Romanian, Hungarian. A larger number of languages were identified as playing a role in their lives, though, encompassing both school and non-school languages, including, for example, Arabic, French, Italian, Japanese, and Latin. Students’ self-assessment of plurilingual competencies - at least referring to some of the languages in their repertoire - was fairly high, corresponding to an average selfassessment between C1 and C2 levels. 54.5% of students (1 NA) out of the 78 had prior experience reading multilingual stories, but only 22.4% (2 NAs) had experience writing them. 2.2 Data collection & Instruments This study was approved by Rhineland Palatinate’s ADD and data protection office. All participants were over the age of 14 and provided written informed consent. 2.2.1 Demographic questionnaire Students provided their age and gender, their most recent grades in German, English and, if applicable, a second foreign language. Language use was assessed using a variant of H AUKÅS / S TORTO / T IURIKOVA ’s (2021) items on language use habits (translated, reformatted, with additional items). Plurilingual competence was assessed using a selection of translated, simplified descriptors from the Companion Volume scale “Building on plurilingual repertoire” (C OUNCIL OF E UROPE 2018: 162). Self-developed items included items on language study (current/ desired), experience with media in different languages and in multilingual formats, and with transformative works. 78 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 2.2.2 Pretest/ Pre-Posttest/ Posttest/ delayed Posttest & Feedback questionnaire Pretest, pre-posttest, posttest, delayed posttest used identical items, focusing on students’ identity (or using identity related constructs) in relationship to English, and to multilingualism. As these will not be discussed in this paper, details are omitted for brevity. This also applies to the feedback-questionnaire, which used a range of selfdeveloped items to ask students about the experience of writing and reading multilingual IF. 2.2.3 Stories Stories were collected by the teacher using the school’s learning management system and transmitted to the researchers in a pre-anonymized format, using pseudonymous codes to identify learners/ authors (grade 9 only). This pseudonymization was doublechecked by one of the researchers. 2.2.4 Focus groups & interviews A self-selected sub-group from each class participated in the focus groups, which were conducted on school grounds by one of the researchers. Four focus groups with four to six students each took place, lasting between 13 and 30 minutes each. Discussions were audio-recorded. The teachers were also interviewed. These interviews are beyond the scope of this paper. 2.3 Analysis procedure Source files of students’ texts were collected in the authoring tool Twine and descriptive statistics were extracted: number of words and of passages. The word-to-passage ratio was computed to indicate the extent to which the story included descriptive narrative text. 1 Such narrative stretches of text preceding choices provide context for the player which may explain why certain choices are meaningful for the player character. A low word-to-passage ratio can result in a rather meaningless point-and-click reading experience. Segments of multilingual storytelling were identified in the stories, and bottom-up codes were developed to qualify how students made use of multiple languages in the segments. These codes comprised: which languages were used, how they were used, and whether the use of more than one language appeared in the choices or in other parts of the text. Two researchers independently applied the developed codes to the data and found their codes to match. They also identified ‘edge cases’ that could qual- 1 The word count procedure in Twine does not distinguish between words that appear in the story and markup or other programming code. However, the amount of code which students used in their stories was negligible. Teaching (multi)literacies, supporting multilingual identities 79 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 ify either as use of different languages as part of multilingual storytelling, or as typographical deviations or instances of L1 interference. The focus groups were transcribed in standard orthography, with emphasis indicated by capitalization. A rater identified segments related to multilingualism or multilingual storytelling. The unit of coding was at the level of ‘meaning segments’, i.e. units of discourse (partial sentence to multi-utterance exchanges) that contained an identifiable idea or statement. For content analysis, descriptive codes were developed bottom up from the data by two researchers. Where researchers developed codes that differed not only in wording but in substance (11 codes), they discussed their codes and agreed on shared codes. One researcher structured these codes and allocated code categories, which were adapted based on feedback from a second researcher (2 adaptations at the level of code categories). 3. Results 3.1 Stories The groups in grade 9 produced 20 IF stories in total, one per group of two or three students. One story not based on the assigned book was omitted from the analysis. Average story length was 206 words, (minimum 23, maximum 413). The average number of passages was 11 (minimum 1, maximum 18). The ratio of the number of words to the number of passages was 22.3 on average (minimum 7.1, maximum 41.3). To compare, the words-to-passages ratio of the most highly-rated work published on the popular Interactive Fiction Database, for which the source code was available at the time of writing (A NTHROPY 2013) - thus allowing the inspection of its statistics - is 68.8. To illustrate, Excerpt 1 and Excerpt 2 (  page 80) contain the introductory passages to the stories with the highest and lowest words-to-passages ratio, respectively. In Excerpt 1, the preceding narration contextualizes the choice for the reader to a certain extent, and also creates an interesting conflict for the reader: travelling by foot can be more dangerous if the boy is at the perils of nature, but so could driving if the car needs to pass checkpoints where the boy is likely to be questioned by the Taliban. In this excerpt, the students demonstrate some understanding of the setting of the story and the main character Bashir, as well as considerate use of a choice. Both the author and the reader need to critically think through the options by devising or carefully reading the context of the interactive choices in the text. 80 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 Start You are Bashir and you live in Afghanitan. It’s a dangours land and you are a seventeen year’s old boy who flees the country. What are you doing to flee the land from the talibans? [[ You are traveling by foot to the coast ]] [[ You are driving to teh coast of Pakistan ]] Excerpt 1: Introductory passage to story with the highest words-to-passages ratio (text between square brackets represents an option within a choice) In contrast, Excerpt 2, also from an introduction passage, provides no background on the characters or setting in which the story takes place, resulting in a rather meaningless reading experience (unless readers draw on their knowledge of the book). Dad find Bashir in the hut [[ call the police ]] [[ helps bashir ]] Excerpt 2: Introductory passage to story with the lowest words-to-passages ratio Overall, and contrary to the task instructions, students designed few meaningful and important choices, and did not demonstrate through their design of choices to what extent they comprehended the fictional text that formed the basis of their writing task. As for the multilingual nature of the stories, 12 out of the 19 analyzed stories were found to incorporate at least one language other than English. 5 stories used two or more, thus meeting the requirement of the assignment (see Section 2). Table 1 comprises an overview of the languages used. Language other than English Used in … stories German 9 Spanish 5 French 2 Romanian 1 Russian 1 Table 1: Languages other than English used for multilingual storytelling Teaching (multi)literacies, supporting multilingual identities 81 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 The use of languages other than English typically appeared at a single point in the story text and was limited to a word, multi-word unit or sentence. Analysis revealed 26 segments drawing on more than one language (see Table 2), of which three appeared in the choices, and 23 in the rest of the text. Three of these 26 segments were identified as edge cases, as they included deviations from the primary target language (English) that can best be described as typographical errors or L1 interference. In the other cases, the majority of integrating an additional language occurred in direct speech (13 segments), which was always a full sentence, one of which occurred with mediation. In five segments, German was used for what could be called ‘interface text’: instructions to the player about the state of the game (a full sentence or a part of it). Use of more than one language also appeared in narrative descriptions (four segments). Finally, multilingual storytelling could be connected to player choice, where the reader can decide to act in one language or another language, with different effects. This was least frequent, found in only one segment in our corpus. Ways of integrating more than one language in the story Number of segments Example(s) in descriptions (full sentence) 1 So you call the police. Du überlegst ob du Bashir verraten sollst oder ob du ihm und cooper einfach in Ruhe lassen sollst. in descriptions (part of sentence) 3 You will stay in a Jugendherberge and take the next ferry. in direct speech, with mediation 1 You ask a man with a Russian accent, If he can take you to the coast he says “Да, я отвезу тебя туда, это в пути” (Yes, I’ll drive you there, it’s on the way) in direct speech, without mediation 12 He loose and his last words were “merde”. in interface text 5 Hans you leave him. Next day you read the news paper. You see that the police found a dead stranded Man. Game Over du Lappen go back home. [[ versuche es noch einmal ]] 82 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 Ways of integrating more than one language in the story Number of segments Example(s) in player choice 1 yes he gives you the mobilephone but you need to do some sexual things for him [[ do the sexual things ]] [[ matarol y robar el telefono ]] edge case 3 You take him to the shed and hide him. You go in and take something to trink. Table 2: Segments including use of more than one language Looking at the story data qualitatively, and similar to the students’ overall random use of choices as a narrative device, use of multiple languages typically did not make much sense for the story. The prototypical case in direct speech is where a secondary character communicates in another language than English and appears to be monolingual, without any backstory that can explain why the character is unable or unwilling to talk in a language known to the main character (see Excerpt 3). Further, it is worth noting that in partial-sentence multilingual descriptions, changes from English to German may be the result of a vocabulary gap in English, instead of a device to achieve a particular narrative effect. You stay in Nauru and you meet a group of criminals, they say to you “du verdammter Veräter gib uns sofort das Geld” they mistake you for someone else, but since you don’t know German you don’t understand, so the boys beat you up and you die Excerpt 3: Segment of multilingual storytelling comprising a monolingually-acting character One story features an example of multilingual storytelling that is more meaningful and imaginative, and plays with the use of choices and multilingualism in an intertwined way. In this story, the reader subsequently plays the role of Bashir and Cooper. When playing as Cooper, they find out that Bashir is fond of learning languages, and can extend the story as long as they keep asking Bashir to demonstrate his knowledge of other languages. When the reader chooses not to engage with Bashir’s love of languages, the story ends sooner, with a message stating that the player loses the game. 3.2 Focus groups 98 segments were identified and coded, resulting in 83 codes across three domains: Writing multilingual stories (64 segments; 50 codes), Reading multilingual stories (31 segments; 30 codes), and Other (three segments; three codes). The codes were sorted into code-categories listed in Table 3. Teaching (multi)literacies, supporting multilingual identities 83 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 Domain Code-category Number of codes Number of segments Writing multilingual stories Decisions for/ against specific languages 9 19 Factors influencing integration/ nonintegration of languages beyond English 26 29 Ways of integrating languages beyond English 11 12 Relationship between book and unit/ story 2 2 Descriptions of languages integrated 2 2 Reading multilingual stories Expectations & norms 10 10 Dealing with non-comprehension 7 8 (Emotional) responses to reading multilingual stories 6 6 Ease of/ lack of comprehension 4 4 Description of stories/ languages in stories 2 2 Other 1 1 Other - 3 3 Table 3: Codes and categories In this paper, we will only refer to codes from the domain “Writing multilingual stories”, focusing on its first three code-categories. 3.2.1 Decisions for/ against specific languages During the focus groups, volunteers from all four classes (9th and 10th grade groups) shared their reasoning process for using or not-using specific languages. Several students identified English as default (without specifying what exactly made English a default, two segments), or, more specifically, English as default in the EFL classroom (two segments): 84 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 Also, bei mir war es jetzt so, da wir jetzt im Englischunterricht sind, habe ich jetzt die diese ganze, also die ganze Seite halt auf Englisch so gut wie gemacht. (FG4) English also played a prominent role in statements that framed language choice as based on competence in a specific language, mentioned alongside German, Spanish, Romanian and French (five segments). One group also made a statement that might be seen as the corollary of this code: they explained that they used English due to a lack of skill in other languages (one segment). Also important was the status of languages as school languages, either as current (three segments) or as future (two segments) foreign languages. Ehm, wir haben überlegt was für eine Sprache man nehmen könnte, und dann haben wir überlegt, was für Sprachen wir jetzt im nächsten Schuljahr lernen müssten, und da war dann bei uns in der Gruppen halt zwei Leute, die Spanisch lernen müssen, haben wir uns halt auf Spanisch geeinigt. (FG1) An interesting example is the use of Russian, which was mentioned only in one focus group, by one student. He explained that the use of Russian was motivated by a lack of language skills in his group beyond German and English. As they did not speak any other languages, they had to choose between all languages they did not speak as their third language. The student explained further why Russian specifically was chosen: Nein, wir haben, wir konnten nur Deutsch und Englisch, deswegen haben wir Russisch genommen, weil es gerade durch den Ukraine Krieg aufgeleuchtet hat. (FG2) Notably, later in the focus groups, students mentioned languages they speak at home, which were not used in their stories (and were not mentioned, therefore, in the discussion of why they used the languages they did use). 3.2.2. Factors influencing integration/ non-integration of languages beyond English The next code-category helps us understand the factors that supported or hindered integration of languages beyond English from the perspective of students. A range of statements made might best be summarized as falling under the vagaries of teaching reality. Two codes (two segments) credit time constraints for failure to include several languages. Several codes - nearly all from a single focus group - refer to forgetting the requirements to include several languages (nine codes, eleven segments) or to mislaid/ lost instructions (one code, one segment). As the majority of stories from grade 9 included more than one language, this might not have been a typical experience. Task factors were referred to both as justification for including additional languages (as this was a task requirement) and for not including them (non-comprehension of task) (two codes, two segments). Also, ehm, ich kann, glaube ich, für unsere ganze Klasse sprechen, wir haben das nicht so ganz kapiert, sagen wir das mal so. (FG3) Teaching (multi)literacies, supporting multilingual identities 85 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 Beyond the vagaries of teaching reality, four codes (four segments) refer directly to absence of language competence and how this relates (usually: negatively) to including additional languages. This absence of language competence should not be misunderstood as an inability to tap into resources that would have allowed the inclusion of additional languages. One student acknowledged the option of using translation tools for multilingual writing (and argues against this, due to concerns of this not contributing to language competence), while one student cited that it would have been awkward to use a language only they spoke: Bei uns in der Gruppe war eigentlich nur Englisch, eeh weil ja der einzigste, der bei uns in der Gruppe noch ne andere Sprache konnte, war ich, und dann bisschen komisch gewesen so. Ja. (FG3) Of special interest are those codes that cite the story or storytelling in general as reasons to include or not include languages beyond English. Four codes (four segments) clearly fall into this category, all stemming from the same focus group. Factors cited against including additional languages are focus on plot, rather than on languages, and it not making sense within a plot to have additional languages. The idea to draft a plot that would have allowed all this, or to include characters or situations that would have supported use of multiple languages, was not mentioned. These were things, though, that pop up in another code-category (“Ways of integrating languages beyond English”), which will be discussed below. Finally, one code/ one segment each could be identified that referred either to the needs of readers (the confusion readers might experience who do not understand all languages in the text) or to the logic of the source text (which was monolingual). Overall, many more codes from this code-category were about justifying noninclusion, with only one code explaining inclusion of languages besides English (citing the formal task requirement to do so), while a number of codes might be seen as neutral, providing additional information on these points without stating explicitly whether this supported or hindered including additional languages. 3.2.3 Ways of integrating languages beyond English Four codes (four segments) were identified that connected language choice to the characters in the stories - all from one focus group, and, more specifically, relating to the same IF. This group had created a new character, a German exchange student in the US who was studying Spanish at school, adding her to the existing roster of characters and using her as a device to integrate both languages into the story. In several interviews, the interviewer asked explicitly about connections between the branching structure allowed by Twine and its integration into the game, through player choices/ story branching connected to languages. They did not succeed to elicit examples for this from any of the focus groups, to whom this seemed to be a novel idea they had not considered in writing their stories (even though this was demonstrated in the sample IF). 86 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 Two codes (two segments) were identified that connected language choice to setting: In one story, the visual landscape was used to introduce Spanish through a Spanishlanguage advertising board. Even though one of the preparatory lessons looked specifically at languages in California, the decision to use Spanish was not based on their knowledge of local language use, but on their own prospective language studies. It is surprising that there are few comments regarding drawing upon characters and setting to introduce languages. Preparatory activities (learning about languages in the book’s settings, an activity in which students imagined which languages they could see or hear in specific book locations, a linguistic soundscaping activity), as well as the sample story and the task description prompted exactly these kinds of storytelling. Instead, other ways to integrate languages were sought for - such as using the narrator/ game interface for adding additional language (one code, one segment) - or the task instruction was ignored. 2 4. Discussion This paper aimed to investigate how a pedagogical intervention based on multilingual IF can teach (multi-)literacies, including plurilingual competences, and support the development of plurilingual identity by creating a context for the use of learners’ full repertoires in social settings (i.e. translanguaging spaces). We will structure the conclusion along these lines. 4.1 Teaching (multi-)literacies: Promoting plurilingual competencies in an IF setting This project aimed to provide a space for developing multiliteracies, with a focus on storytelling that draws on a range of linguistic resources and that utilizes the narrative opportunities of IF. The stories analyzed contained few examples of effective use of more than one language in the sense of CEFR’s Companion Volume’s “Building on plurilingual repertoire” (C OUNCIL OF E UROPE 2018). This is surprising when compared against students’ self-assessment of productive plurilingual competencies, where the majority of students had agreed with can-do statements describing e.g. the ability to use a fitting word from another language, even if this requires explanations, to switch to a language an interlocutor feels more comfortable in, or the ability to draw on similarities and differences between metaphors and proverbs for humor or rhetorical effect. If the selfassessment correctly reflects the students’ ability to draw on their plurilingual reper- 2 In discussions of the stories read as part of peer feedback, a broader range of using words or sentences in the narration were mentioned and critically assessed. Students were well able to critique the multilingual storytelling strategies of others, even if they themselves had struggled or avoided engaging in multilingual storytelling themselves. Teaching (multi)literacies, supporting multilingual identities 87 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 toire, then they did (rarely) succeed in applying these skills to their IFs. The present data is insufficient to determine if the task itself was unsuitable to tap into (and provide a space for practicing) plurilingual competencies, or if the self-assessment of students’ plurilingual competencies was inaccurate maybe borne out of a lack of experience with tasks that required application of these plurilingual competencies. The overall challenge involved in creating literary IF might have been a contributing factor as well, as evidenced by many stories with minimal narrative text beyond the choices, and little information provided on characters and their motivations, on places and their features. Some of the stories lacked the canvas for integrating some multilingual practices (e.g. the absence of dialogues). Against this backdrop, it is not surprising that the choice structure of IF was utilized so little for plurilingual storytelling. Letting a player choose in which language to respond (like modelled in the sample story) would have been an obvious way of combining the requirements for IFs and for plurilingual storytelling. The only IF approaching this is the story discussed above, in which the player can decide to keep encouraging a character to demonstrate their language skills. The player, though, is taking up the role of the encourager, not of the plurilingual speaker. In the focus groups, too, we had many indications that students struggled with the language requirement. This difficulty in imagining how multilingual storytelling could look like seems to have been insufficiently supported by a range of preparatory activities, from a sample story that modeled possible ways to include additional languages, to linguistic landscape/ soundscape focused activities or learning about multilingualism in the US/ California and in Australia/ Darwin. For students, the abstract knowledge “Spanish is a frequently spoken language in California”, did not translate into the ability to use this in storytelling. That this was not mitigated even by the presence of multilingual students in the group will be taken up in the next section. 4.2 Supporting identities Even though this project revealed aspects of language related identity (e.g. students seeing themselves as future learners of French), it is unclear in how far it succeeded in supporting these, in the sense of contributing to increasingly elaborate future selves, or to a stable Ideal Multilingual Self. While the intervention had the potential to create a (digital) translanguaging space (W EI 2011) from the perspective of the task designers, student perspectives diverged on this point. One element in this might be related to the specific context of school, with its associated language policies and habitus (cf. G OGOLIN 2008), including those specific to the EFL setting (e.g. English as default). Of special interest in this context is the selection of languages in the stories, including considerations of which languages were not included. Even if the task partially succeeded in creating a translanguaging space, not all linguistic resources were brought into it equally. The stories drew on English, German, Spanish, French, Romanian and Russian. Italian, Albanian and Turkish were identified as mother tongues by these students, but not used in the stories, nor were other languages iden- 88 Jules Buendgens-Kosten, Frederik Cornillie, Shannon Sauro DOI 10.24053/ FLuL-2023-0022 52 • Heft 2 tified as playing a role in students’ lives within the group. Focus group data suggests that students were aware of the option of using L1s, but did not generally utilize this option, with few exceptions. Of note, the potential for drawing on the linguistic repertoire of multilingual students within the group was perceived, albeit not utilized, by people who did not identify themselves as multilingual: Aber ich glaube so (unv.), wenn wir diese zwei Stunden noch gehabt HÄTTen, die dann jetzt gefehlt haben, eh, und dass wir dann noch mal ein bisschen mehr drauf eingegangen wären mit diesen MEHRsprachigen, weil ja bei uns auch mehrere Leute gibt, die auch mehrere Sprachen sprechen (FG1) Naturally, multilingual identity also encompasses languages formally studied at school. It is interesting in this context that several students in grade 9 pointed towards Spanish as an element of their ‘prospective multilingualism’. The fact remains, though, that the majority of stories analyzed that used additional languages did not seem to reflect many of the practices associated with multilingualism in authentic contexts of language use, such as conversational code-switching or strategic code-choice depending on context and interlocutor(s). When students did draw on their full linguistic repertoire, they seemed to do so primarily in order to meet formal task requirements, rather than as a means to reflect their own or their community’s language practices in their stories. 5. Conclusion Multilingual storytelling appears to be a non-trivial challenge for EFL learners at German secondary schools. In this study, we found indications that the writing of multilingual IF might create a space for identity work. It also laid bare the need for intensive scaffolding, including more intensive work with text models and more support in transferring out-of-school language use experiences to the writing activities. References A NTHROPY , Anna (2013): And the Robot Horse You Rode in On. Interactive Fiction Database. Retrieved from: https: / / ifdb.org/ viewgame? id=enoy8dnveyic48ei. Last access 27.02.2023 B ARAB , Sasha A. / G RESALFI , Melissa / I NGRAM -G OBLE , Adam (2010): “Transformational play: Using games to position person, content, and context”. In: Educational Researcher 39, 525-536. C ORNILLIE , Frederik / B UENDGENS -Kosten, Jules / S AURO , Shannon / VAN DER V EKEN , Joeri (2021): “‘There’s always an option’: Collaborative writing of multilingual interactive fanfiction in a foreign language class”. In: CALICO Journal 38, 17-41. C OUNCIL OF E UROPE (2018): Common European Framework of Reference for Languages: Learning, Teaching Assessment: Companion Volume with New Descriptors. C UMMINS , Jim (2007): “Identity texts: The imaginative construction of self through multiliteracies pedagogy”. In: G ARCÍA , Ofelia / S KUTNABB -K ANGAS , Tove / T ORRES -G UZMÁN , Maria E. (Ed.): Teaching (multi)literacies, supporting multilingual identities 89 52 • Heft 2 DOI 10.24053/ FLuL-2023-0022 Imagining Multilingual Schools. Languages in Education and Glocalization. Clevedon: Multilingual Matters, 51-68. T HE D OUGLAS F IR G ROUP (2016): “A transdisciplinary framework for SLA in a multilingual world”. In: The Modern Language Journal 100, 19-47. F ORBES , Karen / E VANS , Michael / F ISHER , Linda / G AYTON , Angela / L IU , Yongcan / R UTGERS , Dieuwerke (2021): “Developing a multilingual identity in the languages classroom: The influence of an identity-based pedagogical intervention”. In: The Language Learning Journal 49, 433-451. G OGOLIN , Ingrid (2008): Der monolinguale Habitus der multilingualen Schule. Münster: Waxmann. H AUKÅS , Åsta / S TORTO , André / T IURIKOVA , Irina (2021): “Developing and validating a questionnaire on young learners’ multilingualism and multilingual identity”. In: The Language Learning Journal 49, 404-419. H ENRY , Alastair (2017): “L2 motivation and multilingual identities”. In: The Modern Language Journal 101, 548-565. H ENRY , Alastair / T HORSEN , Cecilia (2018): “The ideal multilingual self: Validity, influences on motivation, and role in a multilingual education”. In: International Journal of Multilingualism 15, 349-364. H IGGINS , E. T. (1987). “Self-discrepancy: A theory relating self and affect”. In: Psychological Review, 94, 319-340. L EPPÄNEN , Sirpa (2007): “Youth language in media contexts: Insights into the functions of English in Finland”. In: World Englishes, 26(2), 149-169. https: / / doi.org/ 10.1111/ j.1467-971X.20 07.00499.x N EW L ONDON G ROUP (1996): “A pedagogy of multiliteracies: Designing social futures”. In: Harvard Educational Review, 60-93. P LOTKIN , Andrew (2011): “Characterizing, if not defining, interactive fiction”. In: J ACKSON -M EAD , Kevin / W HEELER , J. Robinson (Ed.): IF Theory Reader. Boston, MA: Transcript On Press, 59- 66. P RADA , Josh (2022): “Articulating translanguaging as pedagogy of empowerment for racialized, language-minoritized bilinguals: From concepto to proyecto through digital storytelling”. In: TESL Canada Journal 38, 171-185. R EINHARDT , Jonathon (2019): Gameful Second and Foreign Language Teaching and Learning: Theory, Research, and Practice. Cham: Palgrave Macmillan. S AURO , Shannon / S UNDMARK , Björn (2016): “Report from Middle-Earth: Fan fiction tasks in the EFL classroom”. In: ELT Journal 70, 414-423. S AURO , Shannon / B UENDGENS -K OSTEN , Jules / C ORNILLIE , Frederik (2020): “Storytelling for the foreign language classroom”. In: Foreign Language Annals 53, 329-337. T HORNE , Steven L. / B LACK , Rebecca W. (2011): “Identity and interaction in internet-mediated contexts”. In: H IGGINS , Christina. (Ed.), Identity Formation in Globalizing Contexts: Language Learning in the New Millenium. New York: Mouton de Gruyter, 257-278. W EI , Li (2011): “Moment Analysis and translanguaging space: Discursive construction of identities by multilingual Chinese youth in Britain”. In: Journal of Pragmatics 43, 1222-1235.