Fostering the “Intercultural Reader”? An Empirical Study of Socio-Cultural Approaches to EFL Literature

ABSTRACT The work presented here is a qualitative study of socio-cultural approaches to literary texts in a selection of upper secondary English as a Foreign Language (EFL) classrooms in Norway. The article explores opportunities and challenges related to the fostering of competent “intercultural readers” in foreign language (FL) educational contexts by examining how notions of interculturality are implicated in the teaching materials and classroom discourse. The analysis indicates that the examined text interpretation processes rely on a complex interplay between literary texts, tasks and classroom participants in such respect. Two particular strands of the analysis, pertaining to how issues of intertextuality and the emotional dimension of literary reading play a role in the data, are highlighted. The article concludes by discussing the didactic implications of findings.


Introduction
In a context of foreign language (FL) education, the literary medium has always been linked to culture. However, its role has varied over the years according to changing views of culture and language learning (Fenner, 2011;Hall, 2005). The 1990s represented a paradigm shift in such respect, as the focus turned away from teaching about foreign cultures to promoting the language learners' intercultural competence; i.e. their ability to recognise similarities and differences between cultures and to successfully interact in intercultural encounters (Larzén-Östermark, 2008). This led scholars to accentuate FL literature as a springboard for learners' reflections on the perspectivity of individual viewpoints (e.g. Bredella, 2006;Fenner, 2011;Greek, 2008;Kramsch, 1993). At the same time, there has been an increasing emphasis on standardisation and testability in all subjects and levels of education, with FL curricula becoming progressively utilitarian (Bloemert, Jansen, & van de Grift, 2016;Byram, 2010;Tornberg, 2013). Potential practical consequences of such dichotomies may be that literature is given a marginalised place in the classroom and treated as no more than a means to an end, for example to promote reading skills, or to teach about certain linguistic or cultural phenomena (Fenner, 2011;Paran, 2010;Pulverness, 2014). Accordingly, aesthetic and analytical approaches to the literary text may be neglected, and there may not be much room for the kind of personal involvement and critical thinking that intercultural learning requires. Researchers within the field of FL didactics have therefore pointed out the importance of integrating language, culture and literature in language education (Burwitz-Meltzer, 2007;Hall, 2005;Lütge, 2012). Contributing to the debate on how education can adapt to the demands of twenty-first century society, scholars have also addressed a need to bridge the gap between "old" and "new" media in the FL classroom due to young people's negative attitudes towards reading literature (Habegger-Conti, 2015).
In the light of such concerns, there is a need for empirical research on how literature is included in school settings. The work presented here provides insight into this realm through its focus on socio-cultural approaches to literary narrative texts in a selection of upper secondary English as a foreign language (EFL) classrooms in Norway. The study, which is qualitative in nature, is part of a PhD project that explores the theoretical and practical ramifications of working with Bildung, intercultural competence and literary reading as educational aims in the EFL subject. Using the Model of the Intercultural Reader (MIR) (Hoff, 2016) to analyse teaching materials and classroom discourse, the overarching aim of the study is to explore the ways in which the intercultural dimension is implicated in the participants' engagement with English literature. The study investigates how teaching materials influence processes of text interpretation in such respect, in addition to examining the impact of peer and teacher interaction in this equation. The article highlights two particular strands of the analysis, pertaining to the role of intertextual exploration and emotional reader response in the examined data. The rationale for focusing on these particular analytical strands is that they are suited to illustrate both trends and varieties in the material. Accordingly, the article discusses the following research questions: how do teaching materials and classroom participants shape the ways in which notions of intertextuality and emotional reader response are involved in socio-cultural approaches to English literature? Which didactic opportunities and challenges related to the fostering of "intercultural readers" are implied by the findings?

Intercultural Competence and Literature in FL Education
As a theoretical concept, intercultural competence cannot be tied down to one, authoritative definition. 1 However, Byram's (1997) model of intercultural communicative competence (ICC) has been particularly influential in the context of FL education. This model presents the qualities of a competent "intercultural speaker" who acts as a mediator between different worldviews in order to establish mutual respect and understanding. The "intercultural speaker's" qualities are described as a set of knowledge, skills, attitudes and dispositions to act (see Byram, 1997, pp. 57-64).
In recent years, scholars have pointed out that Byram's model does not adequately reflect the complexities of our contemporary world. For example, the model's representation of cultural identity as tied to a particular language or country has been criticised (Dervin, 2010(Dervin, , 2015Ros i Solé, 2013). As processes of globalisation, migration and technological developments have contributed to current perceptions of "culture" and "identity" as dynamic and fluid concepts (see Davcheva & Fay, 2016;Holliday, 2011), it has been argued that intercultural competence must be understood as more than the ability to communicate successfully across cultural boundaries or the empathetic tolerance of otherness (Kramsch, 2011). If we acknowledge that individuals represent multiple and composite identities which come into play in various ways in different contexts, we must simultaneously recognise intercultural communication as a challenging undertaking dependent on the ability to navigate conflict, contradictions, complexity and ambiguity (Dervin, 2015;Hoff, 2016;Holliday, 2011;Kramsch, 2011). A consequence of this is that when dealing with intercultural competence as an educational goal, the primary aim may not be to help learners to adapt certain behaviours and attitudes in order to establish effective communication, but to lay the ground for learning processes which allow them to go beneath the surface of discourse and appearances and to explore questions to which there are no clear-cut answers (Dervin, 2015;Hoff, 2016;Kramsch, 2011).
The present article focuses on literary reading as a potential factor in this endeavour. While a number of scholarly works have theorised the role of FL literature as a medium for intercultural explorations (e.g. Bredella, 2006;Greek, 2008;Hoff, 2016;Kramsch, 1993;MacDonald, Dasli, & Ibrahim, 2009), less is known about whether and how such concerns are taken into account in current reading practices in FL education. Some research on classroom practice related to the development of intercultural competence has emerged (e.g. Harbon & Moloney, 2013), but studies focusing on the 1 See e.g. Deardorff (2004), Dervin (2010), and Perry and Southwell (2011) for an overview of different theoretical definitions and models.
function of literary reading in this undertaking have either been conducted in higher education and/ or been approached from an action research perspective, relying on specifically designed teaching materials or the researcher as practitioner in the classroom (e.g. Burwitz-Meltzer, 2001;Gomez, 2012;Porto, 2014;Rodríguez & Puyal, 2012;Thyberg, 2012). Other research projects have explored teacher cognition about intercultural competence (Jokikokko, 2005;Larzén-Östermark, 2008;Moloney, 2013;Sercu et al., 2005) and the teaching of literature (Bloemert et al., 2016), respectively. Because these studies rely on self-report instruments such as questionnaires and interviews, they provide valuable insight into practitioners' perspectives, but it should be recognised that teachers' beliefs about own practice do not necessarily correspond with what they are actually doing in the classroom (Borg, 2006).
Similarly, studies of teaching materials (e.g. Eide, 2012;Lund, 2007) can point to potential consequences for pedagogical practice, but they cannot provide insight into how this material is used. Accordingly, while certain possibilities for intercultural learning may be identified in the tasks or discussion prompts which serve as a point of departure for the learners' engagement with literature, it is important to recognise that tasks on paper and tasks "in action" are two very different issues (Kumaravadivelu, 2007). Learners within the same classroom may perform a given task differently, and the teacher's approach to guiding the learners in this endeavour is also a deciding factor (Devlieger & Goossens, 2007;Harbon & Moloney, 2013). One factor in this equation is the extent to which the classroom discourse is allowed to take place as what Barnes (1976) labels "exploratory talk", i.e. the "active formulation and reformulation of hypotheses, engagement with understanding of the material, and […] the thought processes that enable that understanding" (Samuda & Bygate, 2008, p. 33). Moreover, because literary reading is a deeply personal process (Iser, 1978;Rosenblatt, 1994), any classroom work on literature which engages the learners' perspectives will result in a multitude of different interpretations, and these are in turn further influenced by the interplay between learners and learners as well as learners and teacher (Fish, 1980;Ibsen, 2000;Thyberg, 2012). With a foundation in Vygotsky's (1986) view of learning as co-constructed through social interaction, the present study thus aims to fill a gap in the research on literary reading and intercultural pedagogy by exploring how literary texts, tasks and classroom participants correlate in shaping the ways in which notions of interculturality are implicated in socio-cultural processes of text interpretation.

The Model of the Intercultural Reader (MIR)
The theoretical framework chosen for the study is the MIR (presented here in Figure 1), a descriptive and prescriptive model for the reading of FL literature (in Hoff, 2016). The rationale for using this particular model to analyse the data is that it allows for in-depth insight into aspects of text interpretation about which previous models have been unclear. For instance, while Byram's model of ICC includes the "intercultural speaker's" engagement with FL texts, it does not bring attention to "what distinguishes processes of text interpretation, and particularly the reading of literary texts, from other forms of intercultural communication" (Hoff, 2016, p. 59). Reconceptualising Byram's (1997) concept of the "intercultural speaker", the MIR focuses specifically on the qualities of the "intercultural reader" and is thus suited to analyse classroom practice related to literary reading.
The MIR is based on an understanding of reading as a negotiative, communicative experience (cf. hermeneutic theory and the reader reception tradition of literary theory, e.g. Gadamer, 1996;Iser, 1978), and consequently rests on the premise that the reading of FL texts inherently involves processes of intercultural communication. Furthermore, the model aims to take into account the types of concerns that have been addressed in sub-section 2.1, illustrating the multiformity these communicative processes may take when the competent "intercultural reader" engages with FL literature.
The illustration earlier shows how the text interpretation process may transcend notions of time and space by operating at three, interlinked levels of communication. Level 1 entails the "intercultural reader's" interaction with the FL text and its inherent literary voices, whereas level 2 involves her consideration of how a variety of other readers might communicate with the text. Finally, level 3 requires her to reflect on how the text may communicate with other texts. At all three levels, "the intercultural reader's" emotions as well as her cognition are involved as she considers the effects of the narrative style and structure of the text as well as the different cultural, social and historical subject positions of text(s) and reader(s). In this sense, the reading of FL literature can take place as a multifaceted process that may contribute to learners' understanding of intercultural communication as a "complex, changing and conflictual endeavor" (Kramsch, 2011, p. 359). This presupposes, however, that the teacher helps them to explore ambiguities in the reader-text relationship "from a variety of different vantage positions involving all of the three levels of communication described in the model" (Hoff, 2016, p. 64).
In order to gain insight into whether and how such concerns play a role in current reading practices in Norwegian upper secondary EFL education, the MIR was used to analyse teaching materials and classroom discourse. As the present article focuses upon two particular aspects of the analysis, namely the role of intertextual explorations and emotional reader response in the examined material, the theoretical foundation for these components of the model will be outlined before the material and methods of the study are presented and discussed.

Intertextuality and Emotion
As described in section 2.2, level 3 of the MIR involves the reader's consideration of how the literary text may communicate with other texts; i.e. it entails identifying and reflecting on aspects of intertextuality. A term originally coined by Kristeva (1986), intertextuality relies on the Bakthinian view that "[e]ach word tastes of the context and contexts in which it has lived its socially charged life; all words and forms are populated by intentions" (Bakhtin, 2006, p. 293). This means that a text cannot be understood as a self-sufficient whole, but rather as "the site of an intersection of numberless other texts, and existing only through its relations to other texts" (Abrams, 1999, p. 317). The exploration of intertextual relationships may play an important role in the development of learners' intercultural competence, as this can promote their awareness of the "constant interplay between multiple voices in discourse and society" (Dervin, 2015, p. 51). A didactic challenge in such respect, however, is that this type of inquiry demands from learners a high level of analytical thinking in order for them to recognise parallels between different texts, which in turn relies on their having established a rich and diverse "cultural capital" (Bourdieu, 1991) from which to draw such parallels.
The other analytical strand dealt with in this article concerns the emotional dimension of literary reading. The ability to experience literature aesthetically is an aspect of literary competence that is highlighted in Rosenblatt's (1994) transactional theory. It also plays a part in the "intercultural reader's" engagement with FL literature, as the subject's emotional and personal involvement is essential to the development of intercultural competence (Byram, 1997(Byram, , 2010Kramsch, 1993;Narancic-Kovac & Kaltenbacher, 2006). When a reader responds to literature at an affective level, her primary concern is with "what happens during the reading event" (Rosenblatt, 1994, p. 24). In other words, the reader pays attention to feelings and thought processes brought about by the encounter with the text. Research has indicated that the human capacity to empathise with people who hold beliefs and desires different from one's own may be promoted through the very exposure to literary texts (Bredella, 2006;Kidd & Castano, 2013). However, it is important to keep in mind that the development of intercultural competence is not an automatic process. Indeed, a FL text may also serve to confirm, or even enhance, any prejudiced attitudes and stereotypical views that learners might have of people from other cultures unless such perceptions are brought out in the open and challenged in the classroom (Hoff, 2013). Furthermore, if literary reading is to contribute to learners' ability to handle complexity, conflict and ambiguity in intercultural communicative situations, the aim cannot merely be for them to express empathy with the literary characters that they are reading about, but also to explore feelings of confusion, tension and ambivalence that they may experience during reading, and to reflect on the reasons why they react the way that they do (Hoff, 2016;Kramsch, 2011;Thyberg, 2012).

Research Design
In order to reach sufficient depth in the analysis, a qualitative approach was deemed the best course of action. The number of informants is thus limited, represented by a collective case study (see Stake in Creswell, 2008, p. 477) of four embedded units. Although the observed classroom work was based on different literary texts and sets of tasks in each case, a cross-unit provision was that learner group constellations constituted the central approach to working with the material. To varying degrees and in different ways, these peer group interactions were further supported by plenary discourse and teacher input.
The study follows an instrumental case study design by providing insight into an issue or phenomenon rather than focusing on the particularities of the case itself, which is characteristic of intrinsic case studies (Stake, 1994, p. 237). Consequently, the study does not concentrate on the experiences of single students or take into account motivational and developmental factors that may have governed individual learners' performance, nor does it aim to assess the teachers' professional skills. Rather, it is concerned with expounding the complex ways in which socio-cultural processes of text interpretation can unfold in the FL classroom. The objective of using the MIR to analyse the material was therefore not to label the examined teaching practices as "good" or "bad" as far as the fostering of "intercultural readers" is concerned, but to illuminate some of the didactic opportunities and challenges related to this undertaking.

Context and Participants
The empirical data collected and analysed in the study is investigated as manifestations of a specific discourse, i.e. the teaching of literature at the first year of Norwegian upper secondary EFL education (VG1). The material must be understood in context with the guidelines provided by the Norwegian National Curriculum for the English subject, which highlight literature as a medium holding a special potential for promoting "a deeper understanding of others and of oneself" (Norwegian Directorate for Education and Training, 2013). In this sense the curriculum can be said to represent a holistic perspective, emphasising affective factors and the personal development of learners, in addition to linking the reading of English literature to the very cornerstone of intercultural competence: the relationship between Self and Other (Hoff, 2014). However, alongside these overarching objectives, the curriculum also prescribes level-specific competence aims which tend to focus on skills and knowledge that can be assessed and tested. For the VG1 level, this entails the ability to "discuss and elaborate on different types of English language literary texts" (Norwegian Directorate for Education and Training, 2013), but what it means to "discuss and elaborate on" literature is not explained. Furthermore, the curriculum's description of reading as one of five basic skills in the subject includes the ability "to understand, explore, discuss, learn from and to reflect upon different types of information" (2013), thus leaving out aesthetic dimensions of reading as well as the ability to "interpret" a text. Accordingly, the Norwegian EFL curriculum represents both possibilities and limitations as regards its potential to generate teaching practices which may foster competent "intercultural readers" in the English classroom.
Four VG1 EFL classes, from three different upper secondary schools in the western region of Norway, participated in the project. The teachers, two females and two males, ranged in age from the early 30s to the mid 50s, and had diverse educational backgrounds as well as teaching experience. The learners were in the 16-17 age range. As peer group discussions constituted the central approach to working with the texts in all four classrooms, a focus group consisting of three to five learners was selected in each case. With the aim of including heterogenic, mixed ability participants in the study, the focus group members were recruited through purposive selection (Onwuegbuzie & Leech, 2007) in the pre-observation phase. This was done through a questionnaire that gauged the learners' attitudes to, and experiences with, reading and working with English literature. Participation in the study was voluntary, and all learners and teachers were assigned pseudonyms in order to protect their privacy. Informed consent was given by the participants, and they were free to withdraw from the project at any point during the process.

Texts and Tasks
In the observed lessons, the classes worked with the following texts: The discussion prompts that served as a springboard for the classroom discourse were a mixture of textbook tasks, material taken from online teaching resources, and tasks specifically designed by the teacher and/or colleagues. The texts and tasks were chosen by the participating teachers, who were given no guidelines by the researcher.

Data Collection and Analytical Procedures
The data was collected during the spring and autumn semesters of 2015. The number of lessons observed varied according to how much time was spent working with the text in question, but no less than five lessons were observed in each class. All of the lessons were video taped, but the focus group discussions were audio taped rather than filmed, as the researcher deemed this to be a less intrusive approach. The recorded material was subsequently transcribed.
As previously indicated, categories derived from the MIR were used to analyse the material. A coding sheet (Appendix A) was developed in order to be able to identify patterns in the data. The categories of primary relevance to the analytical strands discussed in this article are the following: . Level 3 communication: identifying and reflecting on other texts with which the level 1 text shares aspects of intertextuality, either indirectly (and even unintentionally) through a similarity of plot or theme, or more directly by intential referencing to another text through allusions, quotations or parody. 2 In this case, the primary text was a level 2 rather than a level 1 text, as it constituted a reader's (i.e. the film director Baz Luhrman's) interpretation of Shakespeare's original play. The learners also read an excerpt of Shakespeare's text ("The Balcony Scene").
. Emotional reader response: responding to the text at an affective and personal level, for instance by identifying with a literary character, linking aspects of the text to own experience, exploring feelings brought about by encounter with the text.
Due to the interlinked nature of all components of the model, both of these categories intrinsically involve the reader's cognitive consideration of the effects of the author's narrative choices and the cultural/social/historical subject positions of different literary voices, readers and texts (see Figure 1).
Each set of tasks was first analysed separately. Next, the correlation between the task potentials and the focus group responses was examined. This was a rather challenging analytical process, as some of the groups did not always discuss the task questions in chronological order, and in some cases they involved elements of text interpretation that were not implied by the tasks. Nevertheless, an attempt was made to sort the group responses into sequences linked to corresponding task questions. 3 Following this stage of the analysis, the impact of teacher input and full class interaction was considered. Since this discourse could not always be linked directly to the tasks, the primary aim of this part of the analysis was to examine whether these interactions touched upon aspects of the MIR which the tasks and/or peer group dialogues did not.
Finally, the complete set of data pertaining to each case was treated as an individual sub-unit of analysis, allowing the four sub-units to be compared and contrasted through a process of cross-analysis. The aim of this cross-analysis was to examine whether some overall trends in the data could be identified in spite of the fact that the classes had worked with different texts and tasks.

Overall Results
The findings presented and discussed here relate to trends emerging from the cross-unit analysis of the material. However, the analysis disclosed not only cross-unit differences but also great varieties within each case. After an initial overview, therefore, the two aforementioned focus points of the analysis are examined in more detail, and an in-depth discussion of specific examples from all four sub-units is given in order to allow for a nuanced understanding of the phenomenon at hand. 4 A key conclusion is that the examined teaching material allowed learners to take into account a wide range of different literary voices, readers and texts as they worked with English literature in class. Most of the task sets could thus be said to fulfil an important premise for their development of "intercultural readers". The analysis further revealed a considerable correlation between task potentials and the actual processes of text interpretation which unfolded in the classroom, in the sense that task sets involving several of the categories derived from the MIR tended to stimulate more thorough and multifaceted processes of text interpretation in the peer group discussions than those which involved a limited range of categories. However, as the examples discussed in sections 4.2-4.3 will demonstrate, learners also moved beyond the potentials identified in the tasks, or they overlooked or struggled to fulfil them. The analysis also indicated that social interaction at peer group level as well as plenary level, in addition to teacher input, had both enhancing and undermining effects in such respect. Finally, the literary text itself could not be ignored as a central, influential factor. In the following, findings related to level 3 communication and emotional reader response will be presented and discussed in more detail.

Level 3 Communication (Intertextuality)
A trend identified in the task sets was that they primarily involved levels 1 and 2 of the MIR, as there was a focus on the learners' communication with different literary voices inherent in the text as well as their consideration of other readers' interpretations of it. 5 Only one of the four task sets opened up, quite implicitly, for level 3 communication. As part of their engagement with Romeo + Juliet, the learners in Case C created PhotoStories in the form of news segments reporting from one of the scenes of the film. These multimodal stories combined still images with audio recordings of the learners acting as television (TV) news anchors and field reporters, in addition to incorporating citations from Shakespeare's original text. Recontextualising both Shakespeare's play and Baz Luhrman's film adaptation by playing with the conventions of the TV news genre, the learner texts could thus be said to involve multiple layers of intertextuality. Prompted by a task to discuss how they experienced the process of creating these texts, student C2m appeared to recognise how his success in this endeavour depended on his ability to observe and comment upon the literary characters' actions and motivations from an objective, distanced point of view: You have to kind of, see the situation from an outside perspective. […] You don't know why they they did what they did. You just have to report the fact of what happened.
In other words, this learner recognised how his alternative representation of the literary characters and events had an effect on how he related to them. However, student C2m's peers did not follow up on this insightful observation, and instead the group discussion turned to technical issues concerning the multi-modal PhotoStory format. In the subsequent full-class discussion, learners also expressed concern about the amount of time they had spent figuring out the technicalities of the software, arguing that this got in the way of what they perceived to be the true potential of the PhotoStories: the opportunity to develop "artistic integrities" and to show "how good you are to use the English language" (plenary students CXm and CYm).
In the example discussed earlier, student C2m's inclination to involve level 3 communication was neither acknowledged by his peers in the group discussion nor allowed further exploration in the plenary session. This could indicate that the teacher shared the learners' understanding of the PhotoStory task as one which mainly involved creative and digital skills rather than the ability to reflect on intertextual relationships. However, it should also be acknowledged that the teacher was not privy to what was said in the group discussions, and therefore could not encourage further exploration of the issue of intertextuality when it was raised by student C2m.
However, teacher Cf did initiate another exchange related to level 3 communication. In response to her enquiry about whether they were familiar with any other texts than Romeo + Juliet within the love story genre, the learners referred to motion pictures and novels such as Titanic, The Fault in Our Stars, and the Twilight series. A discussion of genre conventions related to this kind of text followed, upon which teacher Cf noted Shakespeare's prevalent influence on filmmakers and authors who "use his plays as a starting point for telling their story […]". What this teacher directed the learners' attention to here, then, was how texts bear traces of other texts, voices and discourses (cf. Bakhtin, 2006;Dervin, 2015;Hoff, 2016;Kramsch, 2011). However, this did not evolve into a consideration of how these texts represent different narrative choices as well as a variety of cultural, social and historical subject positions. According to the MIR, such considerations are essential to an understanding of how issues of interculturality are involved in level 3 communication (Hoff, 2016, p. 63).
The examples mentioned earlier have dealt with level 3 communication incited by the tasks or the teacher. An interesting find identified in the focus group responses, however, was that some of the learners touched upon aspects of intertextuality even though they were not explicitly prompted to do so. In Case A, the futuristic, utopian/dystopian setting and representation of political power in Kurt 5 Level 2 communication in the examined data involved alternative versions of the level 1 text in the form of film adaptations or cartoon summaries, audio recordings of "expert" readers discussing the text and how it was received at the time of its publication, or interpretative statements about the text incorporated in the discussion prompts.
Vonnegut's short story "Harrison Bergeron" appeared to activate the learners' knowledge of similar texts. Without referring to the level 1 text, student A1f pointed out that, "[t]he problem with creating a perfect society is … it could easily become … Brave New World … 1984", whereas student A3m offered his own idea of a perfect world: " […] like Star Wars, but without the wars". Further on, when reflecting on one of the characters in "Harrison Bergeron", the latter student was compelled to talk about a TV series he knew: It might be that the Handicap General really is … more worth than everyone else, and she … just wants to bring everyone else down, so that she can be the boss. She might be a genius, really. Does anyone watch House of Cards? […] It's so good. And it's the first time I really understood … people with power … Because before I always thought, power is boring … but now it's like … "I want power!" As the citations earlier illustrate, the learners did not explicitly acknowledge any link between the different texts or reflect on why they had suddenly been thinking about them. In other words, their inclination to involve aspects of intertextuality appeared to be a subconscious reaction as much as a deliberate attempt to engage with "Harrison Bergeron" at an analytical level.
Such an effort could, however, be identified in Case B, as one focus group participant noted how "The lottery" " […] reminds me of The Hunger Games, but I don't know why" (student B1f). A bit further along in the discussion, her fellow group members elaborated on this observation. As they reflected on how Jackson's story projected the image of a utopian society, yet everything might not be quite as it seemed on the surface, student B3m followed this up by giving examples of other, comparable texts: For example, that Hunger Games … that's dystopia. I haven't seen the movie, but … Or Divergent. Have you seen Divergent? That's a dystopian society. Like, people are divided into what kind of personalities they are […].
By bringing up a range of movies with which his discussion partners might already be familiar, this learner drew upon his peers' "cultural capital" (Bourdieu, 1991) in the text interpretation process. He functioned, in Vygotsky (1986) terms, as an "expert" who mediated his peers' communication with the level 1 text by placing it within a text sphere that they were comfortable navigating. In doing so, he made the literary text more accessible to them while at the same time adding a level of complexity to the reading process by bringing attention to how these texts relate to one another.
A final trend identified in the material is that all cross-unit instances of learner-initiated involvement of level 3 communication took place at group level, not in subsequent plenary discussions or in exchanges with the teacher. One possible reason for this could be that the learners primarily associated discussions at group level with "exploratory talk" (see section 2.1). This assumption is further supported by the fact that the subsequent full class discourse tended to focus on issues specifically raised by the tasks, serving as summary sessions in which the groups were incited by the teacher to offer their respective responses to the discussion prompts. Accordingly, the learners may have been more concerned with producing "final draft talk" in the form of "well-shaped" utterances (Barnes, 1976, p. 108) at this stage than with bringing up issues that had intrigued them but that they somehow regarded as less "on topic".
Overall, the findings indicate some missed potentials in terms of level 3 communication according to the MIR. While the analysis revealed an inclination among the learners to touch upon aspects of intertextuality in the text interpretation process, their awareness of what they were doing seemed to vary greatly. The findings also indicate that the teachers were not particularly concerned with this aspect of text interpretation, which might have something to do with the fact that it was not a focus point in the tasks. Accordingly, the data provides limited insight into how the juxtaposition of different texts may involve an important aspect of the competent "intercultural reader's" engagement with FL literature: the consideration of how "alternate narrative choices and subject positions" (Hoff, 2016, p. 63) may contribute to a more nuanced understanding of the level 1 text.

Emotional Reader Response
The analysis of the task sets revealed a clear tendency to elicit the learners' cognitive rather than emotional response to the text, and this was generally reflected in the learner discussions. However, what also became evident through the analysis was that the presence of an emotional dimension in the learner responses depended on a complex and unpredictable interplay between text, tasks and classroom participants. For instance, a discrepancy between task potential and actual reader response could be identified in some instances where the learners were specifically prompted to respond to the text at an emotional level. When asked to discuss how they were affected by the ending of Romeo + Juliet, the focus group learners in Case C seemed to indicate that the text had not made an emotional impact on them: Student C1f: The ending shows how much they care for each other, and it's sad. But why should they, why should Romeo take that poison just because, because she is dead? That is// Student C2m: //I also think it's a bit unrealistic that he acted so quickly, before he even knew that she was dead. I mean, that's just how the story is, I guess.
Although student C1f noted that the ending was depressing due to the tragic fate of the main characters, she described this as if observing the text from a stance of critical detachment (" […] it's sad") rather than focusing on any emotions the text had invoked within her (e.g. "it made me sad"). This might indicate that she realised how she was supposed to feel, but that the text had not made a genuine, emotional impact on her. Focus group member C2m echoed this sentiment in his response. While he appeared to acknowledge that the characters' decisions and actions might be typical of this particular text genre (" […] that's just how the story is, I guess"), his inability to relate to the characters at a personal level ultimately prevented an affective dimension from having an impact on his reading of the text. It is pertinent to note that his reluctance appeared to be caused by the very "otherness" of the text. Indeed, the learners' view of the Shakespearian drama as "unrealistic" or "far fetched" was voiced repeatedly throughout the peer group dialogue, and as such it corresponds with previous research on adolescent readers, which has found that this demographic prefers realistic fiction that they can relate to their own experiences (Appleyard, 1991;Zwillenberg, 2015). Although the subsequent plenary discussion did include the perspectives of a few other learners who found that the ending made them "sad" or "annoyed", it is not likely that this made any profound impact on those who were unable to relate to the text and its characters at an affective level. In contrast, the following exchange taken from the peer group discussion of "The lottery" shows how learners' identification with literary characters is not a prerequisite for their willingess to engage with FL literature at an emotional level. Responding to the narrative's depiction of an annual ritual, the brutal stoning of one of the community members of a small village, the learners expressed how the actions of the literary characters stimulated feelings of repulsion, shock and inquisitiveness in them: Student B3m: Like it's perfectly normal to kill someone once a year. And the children takes a part, and it's a big feast … killing … 'Yeah!' Student B4f: But the children had to kill her … I don't know, it's just really messed up. To kill your mother in front of everyone else, it's just really// Student B3m: //Yeah, that was quite shocking … The old lady gave her son a stone for the stoning … that was … Student B1f: That's sick. Student B3m: That was one of the most shocking parts. Student B1f: And I don't understand how the children … doesn't get affected.
[…] Student B4f: They don't really know what normal is … because for them, this is normal. Student B3m: It's weird. We'd never kill someone just to keep the population down.
Taking the form of a spontaneous reaction to the text rather than the answer to a specific discussion prompt, this exchange is notable because it shows how feelings of conflict and ambivalence served to fuel the learners' personal involvement with the text. However, it also illustrates how this caused them to distance themselves from the literary characters' actions. Their attitude may have been further enhanced by the fact that they engaged with the text at a literal rather than a metaphorical level. Although teacher Bm pointed out that it was possible to interpret "The lottery" as a critique of "other issues that are in American society … for example blind faith … following the leader", the learners were not asked to reflect on whether the story could be said to constitute a realistic representation of life in post-World War II (WWII), small-town United States. Moreover, when discussing the text as social critique, the learners did not take into account whether such "blind faith" is a typically American phenomenon, nor did they consider what the political climate in the United States was like at the time of the short story's publication (e.g. it might be interpreted as an allegory for the witch-hunt processes caused by McCarthyism). Instead, the plenary discussion, which preceded the group work in this case, included an attempt to relate the story to present-day concerns, leading the classroom participants to drawing parallels to how stoning is used as a form of punishment in "Islamic cultures" (plenary student BXf) in "the Middle East" (plenary student B2m).
What the earlier example illustrates is that, rather than promoting "a deeper understanding of others and oneself" (Norwegian Directorate for Education and Training, 2013), the presence of an emotional dimension in the learners' responses contributed to an "us versus them" attitude which was enhanced rather than challenged through the group and plenary discussions. One way for the teacher to counter such responses would be to prompt the learners to critically evaluate their own, personal experiences and dispositions. Young individuals may be well acquainted with the psychological mechanisms of mob mentality. This could have been used as a point of departure for a discussion of such questions as: are there situations in which you might have acted similarly? To what extent can the story be said to depict universal aspects of the human condition? Does this affect how you relate to the literary characters?
Still, it is important to acknowledge that the learners' emotional response was not only brought about by the tasks but the literary text itself. It should be noted that the task set guiding the learners' engagement with "The lottery" did help them to recognise how the text manipulated their emotions by drawing their attention to how "the use of words is never innocent" (Dervin, 2015, p. 106). For instance, they were incited to discuss the following questions: how does Jackson lull us into thinking that this is just an ordinary story with an ordinary town? What do you understand to be the writer's own attitude toward the lottery and the stoning? These discussion prompts encouraged the learners to explore how their perception of the literary voices operating on the surface of the text was shaped by more elusive voices such as the narrator and the implied author. As one student put it, "So it's like the happy setting that [Jackson] creates … uh, this makes it seem, like, even more dystopian, because [the literary characters] are happy because of something that's … that you're not supposed to be happy for" (student B3m). In other words, the learners discovered aspects of polysemy in their communication with text that were not immediately recognisable to them.
In contrast, Case A serves to illustrate how any absence of emotion in the learners' responses may not necessarily have been the result of a lack of focus on this aspect in the tasks but a direct effect of the narrative style and structure of the text. Learners in this group read Kurt Vonnegut Jr's short story "Harrison Bergeron", a text that depicts an absurd and inhuman, apocalyptic society. As it culminates with the murder of the protagonist, the plot shares some similarities with "The lottery". However, rather than creating sensations of tension, shock and disbelief, "Harrison Bergeron" can be said to have a numbing effect on the reader. Stylistically influenced by the author's early work as a journalist, the story is narrated through short, reporter-style sentences devoid of affective language. This is further emphasised by the fictional characters' inability to experience any feelings. Although they were asked about how the main characters respond to the murder of their son, the learners in Case A did not note the characters' apathy, but instead answered such questions matter-of-factly, noting that "Hazel cries but doesn't remember why" (student A2f) and "Yeah, and then George tells her to think about something else" (student A3m). As they were not urged to consider the narrative tone of the text, an important element of ambiguity in "Harrison Bergeron" was thus left unexplored. The learners' communication with the text can therefore be said to have been, if not unsuccessful, then at the very least, incomplete.
In a final illustrative example, one of the Case D tasks brought attention to how any given reader's response to the text is not governed by the text alone but also by what the reader brings to the interpretation process (cf. Iser, 1978;Rosenblatt, 1994). Readers of Animal Farm were prompted to reflect on the following statement and questions: Mollie is a traitor; she could have changed her ways and given more thought to common interest than to her own. I feel sorry for Mollie, an ideal society should be able to include members like her. She is not to blame. Characterise Mollie and comment on the two statements above. To what extent do you sympathise with her and/or to what extent do you condemn her?
After a brief deliberation of how Mollie was perceived by the other animal characters in the book, the focus group members moved swiftly on to the next question. As teacher Dm walked around the classroom listening to the various group discussions, he asked this group of learners how they felt about Mollie. Student D1f answered by seeing both sides of the coin: We feel that, she kind of, could've made a bigger effort to adapt to the society […] but, kind of understand that she wanted to keep doing what she liked: to wear ribbons and eat sugar. Because it doesn't really affect the other animals, that she wears ribbons.
Teacher Dm followed up this observation by spurring a discussion of the values Mollie might represent, and by asking the learners to consider these in light of the different societal groups partaking in the Russian Revolution, for which Animal Farm can be interpreted as an allegory.
After this interlude, they revisited the original question: to what extent did the learners sympathise with or condemn Mollie? The following exchange ensued: Student D2m: Well, it's sort of a question if you're for or against communism// Teacher Dm: //Yeah// Student D2m: //So, and I think most of us are against. Student D1f: Yeah. Teacher Dm: Yeah. And you could, sort of, instead if communism, say, sort of any totalitarian regime, where everything is decided for you, should you be able to keep some individuality? Should student D3f be allowed to be student D3f, or should she, sort of, be part of one, singular mass?
This example is notable for a number of reasons. First, it illustrates the crucial role of the teacher in recognising task potentials and helping learners to establish the necessary background knowledge to be able to fulfil such potentials. By linking Animal Farm to the Russian Revolution as well as implicitly placing it within a larger ideological context of individualism, democratic ideals and human rights, teacher Dm incited the learners not only to take into account other historical and cultural points of view but also to reflect on their own dispositions and values. Notions of human rights must be dealt with in a sensitive and nuanced manner to avoid a hegemonic and Eurocentric understanding of the concept (Dervin, 2015), and in such respect it is worth noting that although the learners ultimately tended to empathise with the free-spirited Mollie, they came to this conclusion after having demonstrated a willingness to understand other perspectives. The exchange shows how the learners, with some assistance from the teacher, were able to look beyond actions and words (Hoff, 2016;Kramsch, 2011) and identify implicit ideologies in the task's interpretative statements about a literary character. Accordingly, the discussion promoted the learners' awareness of how their response to a given character when reading literature is influenced by their own cultural, social and historical points of view, and also how this might cause other readers to respond differently (" […] it's sort of a question if you're for or against communism"). As such, the learners' deliberation of this task may have helped them to regard issues of interculturality as an instrinsic aspect of their communication with FL texts.

Conclusion
This article has discussed a study that relied on the MIR to examine reading practices in four upper secondary EFL classrooms in Norway. With a focus on findings related to two particular aspects of the MIR, the aim of the article was to answer the following research questions: how do teaching materials and classroom participants shape the ways in which notions of intertextuality and emotional reader response are involved in socio-cultural approaches to English literature? Which didactic opportunitites and challenges related to the fostering of "intercultural readers" are implied by the findings? In response to the first research question, the article has provided in-depth insight into how tasks, literary texts and classroom participants interrelate in rather unpredictable ways in shaping processes of text interpretation. For instance, whereas only one of the four task sets involved level 3 communication, the focus group discussions indicated an inclination among the learners to involve aspects of intertextuality, albeit to varying degrees of awareness and analytical intent. Furthermore, the analysis disclosed a discrepancy between task potentials and actual learner response when it comes to the emotional dimension of literary reading. The examples that were discussed served to illustrate how the complexities of the literary medium contributed to this outcome.
The article has also discussed the crucial role of the teacher in recognising task and text potentials as well as in acknowledging and challenging student utterances when working with literature in the classroom. While the peer group deliberations displayed both insightful and problematic observations concerning intercultural matters on the part of the learners, the article also showed how social interaction at peer group and plenary level did not always allow for further explorations of these observations. This was particularly evident in how level 3 communication was represented in the data, as aspects of intertextuality could not be said to be dealt with in a way that would promote the learners' competence as "intercultural readers". In this connection the findings point to the importance of including "exploratory talk" at all levels of classroom interaction, allowing the teacher insight into particular aspects of text interpretation which are not implied by the tasks but which nevertheless may be involved in the learners' engagement with the text.
The second research question aimed to illuminate some of the didactic opportunities and challenges related to the fostering of "intercultural readers". Although the qualitative and limited scope of the investigation does not allow the researcher to generalise on the basis of findings, the analysis carries some implications for teaching practice. First of all, the tendency among some of these learners to touch upon level 3 communication in their encounter with English literature demonstrates that issues of intertextuality may not be too demanding for readers at this level to consider; in fact, it may be a natural part of how they go about creating meaning when they engage with texts. Accordingly, explicit attention to intertextual issues may give the teacher useful insight into the learners' pre-existing "cultural capital" (Bourdieu, 1991), opening up for opportunitites to place the level 1 within a text sphere that they are already comfortable navigating. The task for the teacher in such respect is to make the learners aware of how they often refer to other texts as part of their engagement with the level 1 text. Next, if the consideration of intertextual issues is to contribute to their competence as "intercultural readers", they must be encouraged to compare and contrast the narrative styles and structures as well as the different cultural, historical subject positions represented in the texts. Such exploration may provide insight into the "constant interplay between multiple voices in discourse and society" (Dervin, 2015, p. 51), in addition to promoting an understanding of "culture" as a complex and ever-evolving concept. In this connection it is worth noting that a majority of the level 3 texts represented in my data were multimodal, digital texts (i.e. PhotoStory, TV series and films). Accordingly, paying attention to level 3 of the MIR when engaging with literature in FL educational settings may potentially help to bridge what Habegger-Conti (2015) refers to as the gap between "old" and "new" media. Further research of classroom practice based on the MIR is needed in order to explore this potentially beneficial aspect of level 3 communication.
Furthermore, as the learners' emotional response to the texts ran the gamut from empathy to indifference, confusion and repulsion, the examined material illustrates the importance of not only encouraging learners to respond to literature at an emotional level, but also to urge them to critically explore these emotionsor lack thereoffrom a critical distance. Otherwise, significant aspects of ambiguity and ambivalence in their communication with the text may be left unexamined, and their encounter with FL literature may contribute to enhancing stereotypical views of foreign cultures rather than challenging them. As the discussion of findings has indicated, this entails not only the learners' evaluation of own and other readers' outlook, but also their investigation of the convoluted ways in which the literary text shapes their emotions. In other words, they must be encouraged to go beneath the surface of discourse and appearences (cf. Dervin, 2015;Hoff, 2016;Kramsch, 2011).
As a final point, the practical examples discussed in this article have elucidated why the reading of FL literature must be understood as a multidimensional form of intercultural communication that entails navigating conflict, complexity and ambiguity. Moreover, although it has focused upon findings related to two particular aspects of the MIR, the article has provided insight into how the model may be used to evaluate, and, by extension inform, classroom practice. This section has already addressed the need to further explore the didactic potentials of involving level 3 communication in pedagogical approaches to FL literature. Qualitative interviews with teachers and learners could also provide insight into the classroom participants' experiences related to the fostering of "intercultural readers" in FL educational settings.

Disclosure Statement
No potential conflict of interest was reported by the author.