1 4 PhD revisited: Content in Nordic pupil narratives in instructed EFL A Norwegian Perspective 1

In this chapter, the results of a doctoral study (Larsen, 2009) of thematic content in Nordic pupil narratives in EFL 2 are presented. A literary analysis of pupil texts focussing on intertextual references was applied. The findings suggest that the pupils’ narratives formed a discourse in which these narratives explored the power aspects of the parent–child relationship, rebelled against teachers, and ridiculed the perfectionism ideals of popular culture. Implications for teaching English in Norway and suggestions for further research will also be discussed.


INTRODUCTION
The English language has developed into a language that the individual citizen needs to master in a number of fields, not only professionally and socially, but also in the personal sphere. Worldwide, pupils with cultural backgrounds other than the typically British or American ones use English as a means of communication (Graddol, 2006). The emphasis on "communicative competence" (Savignon, 2000), which has influenced EFL learning and teaching during recent decades, has made the pupils' roles constantly more active and makes their contribution meaningful in a new way. Pupils' EFL narratives take on traditional mother tongue qualities of self-development and the content of their narratives thus gain importance. According to Barthes (1966), "narrative is international, transhistorical, transcultural: it is simply there, like life itself" (p. 46); pupils' narrative texts could be seen as part of this collection of narratives. What pupils communicate in their texts becomes relevant; and, in this respect, the phenomenon of intertextuality is important. As Lodge (1992) puts it, "some theorists believe that intertextuality is the very condition of literature, that all texts are woven from the tissues of other texts, whether their authors know it or not" (pp. 98-99). In an intertextual approach to language, the individual writer is seen to contribute to what has already been written by other writers in previous texts; something novel is formulated from another position by the pupil writer. Pupil narratives may be seen to form a discourse by young people who are in the role of Nordic EFL pupils when they write English narratives.
Within the broader field of social constructivism, the doctoral study reported here put forward two main research questions: 1. What characterises the discourse that is developed in the narratives? 2. What aspects of identity are expressed in this discourse?
Previously, this interest in thematic and discursive content was not at issue in English didactics. According to Tornberg (2000), it was rather the language-practising potential of pupil texts that had so far been emphasised within modern foreign language teaching (p. 132). Substantiating the content-wise meaning seemed somewhat unusual, and this reflected the relative novelty of the research topic. The reason for studying pupil narratives in this way was based on the need to enhance knowledge about what pupils communicated in their role as EFL pupils.

THEORY
The theoretical basis involved multidisciplinary research areas within pedagogy and literature studies, and it was relevant to integrate perspectives from different research disciplines. The actual writing was seen as both an individual (Piaget, 1923) and a social act (Vygotsky, 1962). The ideas of some poststructuralist theorists who see language as intertextual (Kristeva, 1974) and dialogic (Bakhtin, 1981), and who see discourse as institutionalised practices in communication (Foucault, 1986), were seen as relevant. The study leaned upon the concept "discourse" not as defined in linguistics, but as it is defined in social science as an institutionalised way of communication characterised by the absence or presence of power. Scollon and Scollon (1995) distinguish between two major categories of discourse relevant to the understanding of pupil discourses; firstly, "those into which one becomes a member through the natural processes of birth and growth within a family and a community (one's gender and one's generation, for example)" (p. 136); and, secondly, those one enters for utilitarian purposes, such as one's professional specialization. In the second category, individuals are seen as able to choose how they express their membership in various group identities.
According to Foucault (1977Foucault ( , 1985, this freedom of choice is restricted. The pupil identity is formed, by not only the individual pupils' background, but also the role that pupils are objectified into. What is possible within a particular discourse is defined by the rules of the discourse system in which any identity is located. The concept "identity" is ambiguous; it denotes the ego-identity of individuals in psychology (Erikson, 1958), but is in sociological studies to be understood, "not as the core of an individual's being, or as a set of fixed characteristics, but as changing, fluid and multiple" (Hawkins, 2004, p. 18). Nevertheless, the identity of someone who communicates his or her emotions, cognition and knowledge, is inherent in that person's voice whether in spoken or written form (Scollon & Scollon, 1995). The personal identity or signature of the writer will rub off in some way (Ivanic, 1998). Within this view, identity can be perceived as both constructed and negotiated in language; the pupils negotiate their role with each other and with society in general.
The composite relationship between the text and its writer and reader is multi-faceted. The notions of text versus intertext step up the complexity of this relationship. Kristeva (1974) coined the concept "intertextuality", where she sees the text in relationship to two axes, the horizontal one and the vertical one. She charts a threedimensional textual space as intersecting planes whose three coordinates of dialogue are: 1) the writing subject, 2) the addressee, and 3) exterior texts. The horizontal axis relates the writer of the text (the writing subject) to the reader of the text (the addressee). The vertical axis relates the text to other texts. The words in the text belong to both writing subject and addressee. The meaning of the words in the text is oriented towards a preceding or synchronic literary corpus. Any text is the absorption and transformation of other texts and constructed as a mosaic of quotations, "the term inter-textuality denotes this transposition of one (or several) sign system(s) into another" (Kristeva, 1974, pp. 59-60). The intertextual nature of any new text implies that the new text is a conversion of at least one text already written.
The pupil narratives in my doctoral study belong in such a textual universe. The influences from other texts are thus central in the analysis of the pupils' narratives. A Bakhtinian perspective on language allows the identification of novel elements in pupil texts. The doctoral study used the somewhat inaccurate term "novel elements" to denote the alterations that the pupil text made to elements in the intertexts, where the novel is seen, not only as something new, but also as a force that operates in language (Bakhtin, 1981). The novel "was formed and matured in the genres of familiar speech, found in conversational talk language (genres that are as yet little studied) and also certain folkloric and low literary genres" (Bakhtin, 1981, p. 50). Given that any text is in essence intertextual, the somewhat parodic nature of any textual production is thus given emphasis in a Bakhtinian view. Pupil narratives may be seen as stratified into what Bakhtin would perhaps have called a pupils' professional jargon, a pupil discourse. In this research connection, a nonmother-tongue discourse, which is fictional and narrative, is formed. The term "the low language of contemporaneity", applied by Bakhtin (p. 21), denotes the type of authentic language that this may invoke. The term "speaking subject" may elucidate how identity operates in narrative texts. "Behind the narrator's story we read a second story, the author's story, he is the one who tells us how the narrator tells stories, and also tells us about the narrator himself" (Bakhtin, 1981, p. 15). The speaking subject, who is behind the narrator, is here defined as the author of the narrative (i.e., the pupil participants in my doctoral study). When the narrative is seen as an intertextual text, the speaking subject may involve the positions of several subjects in the production of the narrative; the speaking subject may constitute multiple subjectivities.

REVIEW
Most of the studies of both Nordic and other EFL pupil narratives that had been carried out before my doctoral study concentrated on the linguistic aspects of writing (Tornberg, 2000). Linnarud (1986) studied vocabulary in Swedish upper-secondary level texts and claimed that poor lexicon was to blame when Swedish pupils wrote "dull and uninteresting compositions with repetition of high fre-quency lexical items" (p. 42). In Albrechtsen, Evensen, Lindeberg, and Linnarud (1991) the various aspects of Scandinavian pupil writing with regard to discourselevel properties, such as superstructure and cohesion, were examined, and Lee (2003) compared their results to Chinese English as a Second Language (ESL) students' discourse development, finding a consistent pattern of superstructure even though their first languages are different. Rahbek (2005) compared the outcome of text exchange for pupils who exchanged texts internationally, to texts written in an ordinary classroom context. She found that the " authenticity and the demands for a tangible product" (Rahbek, 2005, p. 220) that developed through international interaction with peers was beneficial. Backlund (2005) studied gender contrasts in the use of adjectives and found that pupils who took part in an international exchange program "improved in their ability to write descriptively" (p. 59). In Larsen (2005), the thematic content of pupils' text production was interpreted with a focus on "what the pupils say in English, as opposed to focusing on, for instance, their range of vocabulary, their grammar or their spelling" (p. 121). Flognfeldt (2005) examined the use of lexical chunks in pupil narratives and found, for example, that "there is a great chance for a pupil to learn idiomatic English if he or she internalises chunks like make a mistake and What does ___ [NP] look like? as functional wholes rather than through item-by-item processing" (p. 86). Guldal and Raaen (2007) studied linguistic characteristics of Norwegian seventh graders' written English.
Broadly speaking, the linguistic aspects, as opposed to the meaning of the genuine content of EFL texts, were dealt with in the above-mentioned studies. Apparently, analysis of substantial content of pupil texts in EFL was not a well-documented field outside the Nordic countries either (Tornberg, 2000). Within an EFL paradigm, the focus on linguistic, and not thematic content, was prevalent.

METHODOLOGY
In the doctoral study, a qualitative document analysis design in which pupil narratives were approached as literary texts was developed. First, quantitative data were collected from a text corpus in the Nordplus project Teaching and Learning English (TALE) 3 , (Hansson, Kjartansson, Larsen, & Lassen, 2005). Qualitative methods were then used in the assessment of the various text types. They involved thematic and literary analysis.

RESEARCH DESIGN
In the doctoral study, a mixed-method, qualitative-dominant research design was used consisting of document analysis methods that were purposeful for the different parts of the study. Determining which of the text submissions were narratives called for an initial quantitative approach to the data (Grønmo, 1982). From an estimated 400 texts in the open fictional genre, 15 pupil texts were randomly selected for further study. Because of deviation in submission procedures applied by the approximately four hundred 12-14-year-old pupil users, it was difficult to determine the exact number of texts and categorize the different tasks that constituted the doctoral study's material; some texts were submitted as Messenger files, not as Word documents in attachments. Firstly, thematic analysis was applied by the researcher to delineate the material. The number of target texts was reduced from a corpus of 5,000 pupil entries -consisting of messages, narratives and response to narrativesto a corpus of 400 pupil narratives (see Table 14.1). Secondly, the 178 narrative text types in the corpus (i.e., fictional and biographical) were evaluated to select the textual contributions in which the level of pupils' own content was high. The fictional narratives provided more pupil-produced content than the biographical narratives, which were often cliché-ridden and standardized. Open fictional narratives provided a higher level of the pupils' own content than hometown descriptions and chain stories, which were often standardized. Pupils' open fictional narratives were selected for further study, and 15 narratives were randomly selected. Thirdly, a literary analysis was used by the researcher to interpret the 15 selected narratives.
In the literary analysis, each pupil text was juxtaposed to its relevant intertext(s) (see Table 14.2); the analysis was then focused on an interpretation of novel elements in the pupil's narrative (see the theory section). These novel elements were seen as an important part of the pupil writer's own content. In the doctoral study, the wide term "element" was applied to account for the variety of literary devices, for example, character portrayal, setting, plot and motif that this could involve. The change that the pupil made to an established element of the intertextual reference was seen as significant. Each text was interpreted on its own terms; the pupil texts were not compared to each other, only to the intertexts.

MATERIAL
The writing sessions in which the pupils had written the 15 narratives that comprised the material of the doctoral study took place in various Nordic classrooms. Following an introduction by a student teacher, the texts were written on computers and exchanged in an international e-network in transnational groups, each comprising three pupils with different first languages (L1). In the following, three pupil narratives from the sample are presented; (1) target text 4, Pou's Comeback, (2) target text 10, The hippo, and (3) target text 15, The three trolls.
Step Butt… one minute after they started play in the show a very, very big bomb exploded. Every living thing on the planet earth died and the planed vent to very, Very small pieces.

The hippo
Once uppon a time there was a hippo that wanted to marry a hippo girl so he went to a girl and said "möööööööööööh" and the girl said " im so not interested " so he said the same thing to every girl in the river but nobody wanted to marry him. Then when he saddly swimmed in his privet pond he heard some bullets and ran to the river. He saw hippos runing in every direction and behind them where some hunters so the hippo stod in there way and burpd. He burpd so hard the hunters flew to china. Now the hippo is surrounded by girl hippos that want to marry him so he marryd the fatest hippo and they lived happily ever after.

ETHICAL CONSIDERATIONS
To evaluate the relationship between the narrative and the pupil writer was difficult, and an ethical dilemma was related to the information the narrative provided about its writer. Rather than being perceived as informative about their individual writers, the narratives were seen to provide information about a pupil discourse. My interpretation did not necessarily provide information about what the pupil writer had intended to tell; as researcher, I could hardly prove that my interpretation was correct or true, but by showing the procedures for the interpretation, I substantiated that it was trustworthy. Thus, the text analysis was not meant to be a reconstruction of what the individual pupil had meant to narrate, but more what the text communicated when applying the method of interpretation developed in the doctoral study.

The three trolls
Once upon a time there were three, very big and ugly trolls! The trolls were left from there family and they were all alone in the scary forest! Their task was to find the witch, because she knows were the parents of the trolls was, but the three trolls was not so nice as you think they are! They didn't like their parents because they were mean with the trolls, and they want to kill them! But they have to find the witch first! She lived in the scary forest! They start their task, and till the end they have find the house, the house was big and little scary. They knock on the door, and there is no answer, they knock on the door for the second time and there is still no answer. They knock on the door for the third time and finally the witch open the door, and they ask were their parents are, and she answer: they live in a cottage, deep done in the forest! They say they have to begin the trip. And of they go. After four hours with Woking they finally see the house! They knock on the cottage, and some big trolls open the big door! And they ask, "who are you"? They answer "we are your children, when we were baby trolls, you was not so nice to us! And we are for revenge! And it was then the fights start! And a while after that their parents was dead!!! And they have got their revenge! They wanted to find the witch, and give her something, because she help the trolls to find their parents! They walked to her house, and they knock on the door, and she opens the door, and they go " thank you for helping us, and they gave her some magic boots! And they began the travel home! And finally they got their revenge.. And they live as trolls happy ever after!!!!!!! FINISH!!!!!!!!!!!!!!..

FINDINGS
Popular culture constituted a common frame of reference in the texts. This was evident in the multitude of intertextual reference to films and TV shows, for example, The Karate Kid and Teletubbies. Presumably, screen and game versions of literary works were also part of the intertextual framework included. Although biblical texts and literary heritage texts were also referred to, what can somewhat summarily be called a "global entertainment culture" dominated this framework (see Table 14.2). The literary interpretation of pupil narratives 4, 10 and 15 are presented in the following:

LITERARY ANALYSIS OF TARGET TEXT 4: POU'S COMEBACK
The headline is succeeded by a grey shadow repeating the headline and thereby pointing to the denotative meaning of the word "comeback". The shadow headline suggests that this is a parody of something well known already, a shadow narrative slightly altered. The pupil text refers to the Teletubbies quite explicitly and uses almost identical names for the dolls, Tingky Vinky, Lala and Pou. The deviant spelling might be misprints or deliberate modifications. The BBC television show about the four Teletubbies, Tinky Winky, Laa-Laa, Po and Dipsy -the fourth teletubby is not mentioned by name in the pupil's version -was produced from 1997 to 2001. Teletubbies features mechanical figures who repeat simple activities. For example, Laa-Laa throws a ball repeatedly, and the phrase, "Laa-Laa throws the ball", is repeated while one sees the action performed by the doll.
In the pupil's narrative version, the idea of the original work is parodied. So when Pou was in Mexico, "she had to walk… so she walked and walked" (lines 4-5). The repetitious activity typical of the action in the original TV series is slightly exaggerated: "She walked for years" (line 5). The BBC's Teletubbies are nice entertainment figures aimed at entertaining babies and preschool children, but the pupil's teletubbies are corrupted power-seeking scoundrels whose actions are described, for example, in the following way: "Lala had her plan to take over da world" (line 9). The use of the word "da", taken from the hip-hop culture's modern slang instead of "the", underlines the rebellious attitude portrayed in the narrative. The innocent Teletubbies are turned into prostitutes and terrorists, and although the tone is ironic, the narrative is rather depressing. Pou is kicked out because of her ugly clothes, and when she tries to get a decent job, she fails. She earns her money through prostitution to get new clothes.
In the BBC's Teletubbies, the dolls do not wear specific clothing. The novel element in the pupil narrative, of having modern designed clothes on the teletubbies instead of the original jumpsuits may allude to the focus on dress and individuality in modern media. The promise of success if one dresses stylishly, and the finding of one's own personality through clothing is ridiculed; Pou has to sell her body in order to get clothes that will fulfil this desire. The narrative seems to deal with the idyllic portrayal of society that some TV series for children present. The text satirises the escapist content of modern media and disassociates itself from society as it is portrayed in unrealistic TV shows such as Teletubbies. The irony is sus-tained to the very end. There is no happy ending to this story. The mimicking of the repetitious style is maintained throughout when the "very, very big bomb exploded" (line 20), and the planet turns into "very, Very, small pieces" (lines 21-22). In "Pou's Comeback", the superficial focus on appearance does not pay. When the girl is corrupted into selling sex to promote her career, everything else in society is corrupted as well, and the world falls apart.

LITERARY ANALYSIS OF TARGET TEXT 10: THE HIPPO
The title denotes the big, fat African mammal, and the main character is a fat male hippo. The romantic topic of courtship is suggested in the introductory line. The hippo wants to marry a hippo girl, but when he tries courting, he is constantly rejected by the opposite sex. The modern slang of negation, "im so not interested" (line 3), locates the hippos in the younger generation of modern society (Drange, Kotsinas, & Stenstrøm, 2002). His pick up line "möööööööööööh" (line 3) is not appreciated by his fellow hippos. The fact that there is no particular hippo girl he fancies, and that he tries to pick up all of them, helps portray an unromantic hero who is desperate in his search for a mate. Luckily for the hippo, a moment of crisis occurs in the hippo gathering: an opportunity to demonstrate his true heroic nature occurs. Hunters attack the hippos, and panic breaks out. The hippo gets an opportunity to show off. He burps heavily, but his malodorous burp is so powerful that it blows the hunters to China. By chasing the hunters away, the hippo saves his fellow hippos. Afterwards, he is surrounded by female admirers.
Alliterations such as in the misspelled introduction, "Once uppon a time there was a hippo", are frequent, and regardless of their faultiness they help to create a poetic style. The moment of crisis is conspicuously alliterative: "Then when he sadly swimmed in his privet pond he heard some bullets and ran to the river". "Sadly swimmed", "privet pond", "pond he heard" and "ran to the river" (lines 5-7) are sequences of alliterative language that may soften the vigorous edge of the action.
This narrative seemingly has no particular intertext, but does refer to predominant plots in popular culture in general: the male has to carry out a heroic deed to attract the attention of the female. However, the main character does not fit into the formula of the archetypal romantic hero (Radway, 1984). He lives in hippoland, a different society, like a hippy gathering, perhaps. In the alternative culture, other values apply. In hippo-land, there is no point in being the slimmest hippogirl around; quite the opposite is the ideal here. Fatness is approved of and made into a positive attribute. The novel element of twisting the significance of corpulence affects the way in which identity can be constructed and negotiated. By por-traying an alternative setting, this narrative suggests that fatness can be attractive. The hippo chooses the fattest hippo-girl. His choice is self-evident in hippo-land.

LITERARY ANALYSIS OF TARGET TEXT 15: THE THREE TROLLS
The title may connote Tolkien's character Bilbo Baggins' encounter with the three trolls in The Hobbit. Bilbo Baggins somewhat unwillingly accompanies Gandalf and the thirteen dwarfs to reclaim a stolen fortune from the dragon Smaug. On their way to the Lonely Mountains, they encounter various threatening creatures, and among them are the three trolls. Bilbo Baggins is taken prisoner by the three trolls and has to tell stories to be released.
The pupil's text depicts the story of three siblings who are searching for their parents. The trolls are "very big and ugly" (line 1) and have been left by their family and are "alone in the scary forest" (line 2). Thus they may earn sympathy, even though they are trolls. They want to find the witch because she knows where their parents are. Connotations of Hansel and Gretel in the wood create an impression of their underlying wish to become united with their parents in a happy ending. The trolls disapprove of their parents because they were mean to the trolls when they were little. The three trolls seek revenge, and they actually want to kill their parents.
The novel element of dealing with the parent-child relationship in a troll universe makes it possible to elaborate on aggression and hatred in a safe frame. They need the assistance of the witch, who lives in the spooky forest. The witch knows where their parents reside. "After four hours with Woking they finally see the house" (lines 11-12). The parents do not recognise their children: "who are you" (line 13). The crisis implied by not being recognised forms a climax in the narrative. Who are we when we are not even recognisable to our own parents? The three trolls are full of revenge: "we are your children, when we were baby trolls, you was not so nice to us" (lines 13-14). The confrontation in the doorway ends with a fight between the parents and their children. To be vindictive is not necessarily an honourable sentiment; however, this is a troll's universe with its own ethical codes. Human ethics are adjusted to suit the troll world. One is led to feel that justice is done. This is possible since the three siblings are not children, but trolls. Here it is possible to play with identity, put on a tail, become trolls and deviate from common ethics.

SUMMARY OF THE LITERARY ANALYSIS
The persistent demand for beauty from popular culture was dealt with in several narratives. "The ugly princess" (Target text 1), for example, can be read as an example of revolt against the beauty ideal and portrays a society that does not accept girls who are not beautiful. Quite the opposite is portrayed in "Philip the fox" (Target text 11). Philip is excluded because he is too beautiful. In "The fairy tale about the big fat boy" (Target text 9), the main character is excluded because he is too fat. There is no way that he can go back home before he has lost weight.
Most narratives dealt with the relational sides of human existence. Different family situations were portrayed. The single parent family was often favoured, but the conventional family, with two parents and two children, was not represented at all. The modern extended family was also absent. Some narratives dealt with the absence of one or both parents. The sibling relationships in the narratives were genuinely happy; siblings were often wandering about on their own with little parental support. Brothers and sisters were joined in their fight against possible threats that grown-ups in different roles represented. Friendship was portrayed as demanding but generally worthwhile. Identity was explored mainly through the portrayal of a multitude of different main characters where the individual's struggle to fit in was a frequent topic. The individual's role in the group was often portrayed. On the surface, the demands from the group with regard to individual exterior success were portrayed as merciless.
Both male and female heroes struggled to fulfil individual goals. Their heroic deeds were often portrayed with humour and intelligence and questioned conventional gender roles. As to the portrayal of heroes, they operated both on an individual basis or as part of a group. The heroes and heroines struggled to obtain acceptance, peace, love, even revenge, but heroic deeds were ridiculed and parodied in the narratives. The traditional simplification of good/bad conflicts between princesses and witches, heroes and villains were confused. In "Hero -monster" (Target text 14), the question of distinguishing between good and bad was challenged through irony. Although narratives such as "The biggest battle" (Target text 12) confirm the belief in the heroic deed, several narratives' playful approach to this motif implied an ironic distance. Finally, the distribution of power between generations was also dealt with in several narratives.

DISCUSSIONS: CONTRIBUTIONS TO THE ENGLISH DIDACTICS FIELD
The two main research questions in this doctoral thesis were: What characterised the discourse that was developed in the narratives? What aspects of identity were expressed in this discourse?

EMPIRICAL CONTRIBUTIONS
The main empirical contribution in this doctoral thesis is that the narratives suggest that the pupils did not necessarily turn to the mother tongue to write about topics from the personal sphere, but did so using English. The pupil narratives formed a discourse where certain aspects of identity were explored. For example, the narratives idealised the relationship between siblings and stressed the importance of friends. Moreover, the narratives explored the power aspects of the parent-child relationship. The pupils seemed to make English into their other tongue, a language which is not their mother tongue, but which still provided them with a different position for the exploration of identity. One might expect texts to be less locally marked since their intertext basis was fairy tales, hence identity was almost denationalised. The pupils' frames of reference constituted content that is readily expressible in English in such a way that the non-local and non-national interlocutor or reader can understand.
The pupils wrote their image of the generic narrative in another language, which in Bakhtinian terminology could be referred to as other-languageness. The notion of other-languageness may imply that the pupils parody the narrative genre in which they write their narratives. It may also suggest that the language they employ is different, as "one's own language is never a single language: in it there are always survivals of the past and a potential for other-languageness" (Bakhtin, 1981, p. 66). The pupils write in the global language English, which, to these pupils, is a language that is not their mother tongue. Although Nordic EFL pupils do not necessarily engage in "a deeply involved participation in alien culture and languages" (Bakhtin, 1981, p. 369) when they write narratives in English, they write in another language from a different perspective from that of their mother tongue. When pupils write in another language, it may provide another outlook on the world. Their texts are part of an intertextual repositioning which involves a different perspective. Their texts are intertextual according to Kristeva's (1974) definition of the term; a possibility to take on a different position and to formulate something novel within the given genre is thus provided. The concept of otherlanguageness may be applied to the EFL pupils' narratives to suggest this other perspective that is involved in their non-mother-tongue writing.

THEORETICAL CONTRIBUTION
The main theoretical contribution concerned the identification of a Nordic EFL pupil discourse. The doctoral study involved multidisciplinary research areas within pedagogy and literature studies, and it produced knowledge about the thematic con-tent of pupil narratives. It also demonstrated that pupils' language-practising texts could be viewed as literary texts to be analysed with literary methods. The interpretation of pupil texts and their various intertext(s) substantiated that the pupil narratives could be perceived to form a discourse, a Nordic EFL pupil discourse.
When the pupils wrote narratives in English, the expressions of identity seemingly moved away from the national and the local into the direction of global popular culture and the personal sphere. Thus, a Foucaultian discourse, in which the fight for power is central, was formed (Foucault, 1986). In general, the grown-up parental world was often made into an enemy. "A day in the school" might be read as a protest against the ageing and feminised teaching profession. In this fight for power, the good/bad distinction was zero-grounded. Troll children can do terrible things to gain power; the end justifies the means. In order to become independent of parental power, anything goes.
It was symptomatic that national, regional and local identities were de-emphasised. Primarily, the relational aspects of human existence were dealt with. The pupil narratives addressed topics that were significant to the authors' age group; a generational pupil discourse was formed. The identity that was formulated in such settings was disengaged from nationality. It was disengaged from a specific Nordic culture. The exploration of local identity was likewise difficult to trace in the narratives. The international, global language did not promote the constitution of national identity.

METHODOLOGICAL CONTRIBUTION
The main methodological contribution concerns the method developed in this doctoral study, which attempted to identify novel elements as the interface between the pupil's text and its intertext(s). The novel elements were seen to provide insights into the pupil-produced contribution in the narrative. The identification of the novel elements in the pupil narrative provided information about which aspects of identity were expressed. In this research context, the pupils wrote their novel content in EFL. Pupils' intertextual references in the narratives involved a new or novel mode of expression of, for instance, motifs and characters.

IMPLICATIONS FOR TEACHING ENGLISH
The pupils in this doctoral study seemingly reflected on matters of genuine interest to themselves and fellow pupil readers in their narrative discourse. Thus, the networked text exchange may be beneficial for teaching English as L2, not only to develop pupil proficiency in English, but also to develop as human beings. The participating pupils' textual contributions were influenced by intertexts, rendering it important for English teachers to pay attention to which selection of texts is at their pupils' disposal in the school setting. Vestli (2008) discusses the development of the didactics of literature in foreign languages and argues in favour of an action-and production-oriented approach to literature. Such an approach would include practising creative writing to achieve what she calls "literary competence", here understood as part of pupils' cultural formation, such as the ability to read and interpret a literary text and recognise various literary devices (p. 14).
With regard to the content of texts that pupils read in school, English teachers' search for high quality literature suitable for pupils will possibly continue. The canonical texts from cultural heritage have seemingly paved the way for pupils' literary production, but input from popular culture is overpowering. To include more popular culture in the English school subject's reading list is perhaps not required since such input is overwhelming in society, regardless of what happens in the classroom. However, pupils may be able to master this input overflow. They are not entirely mesmerised by popular culture. The literary interpretation in this doctoral study demonstrated that pupils do manage to resist popular culture content. For instance, if given the opportunity in English lessons, they may construct a narrative discourse together in which criticism of such content appears.

SUGGESTIONS FOR FUTURE RESEARCH
Firstly, the pupil narratives were shared with peers, teachers, and student teachers; however, a subsequent study of response activity in the same corpus (Larsen, 2012) showed that there was little evidence of pupils paying attention to peer and teacher responses to their linguistic performance. The pupils were to some extent more open to making minor revisions to the thematic content, but even the response on content was generally ignored in subsequent versions of the texts. Further research into the response phase in pupils' written narratives is important, considering the amount of time and energy spent on making it worthwhile for both teachers and fellow peers. There is a need for more research into the form that literary feedback might take.
Secondly, the doctoral study suggests that learning and teaching critical analysis of a great variety of texts may be worthwhile. Future studies in this field might address questions related to the quality of the input. Does the intertextual nature of pupil writing require high quality model texts if one aims to support a high level of language proficiency and promote pupils' development as human beings? It is also relevant to study if and how pupils may benefit from deep learning (Kunnskapsdepartementet, 2016) about the origin of the literary formulas they are surrounded by. Another point worth researching more fully is the learning and teaching of critical analysis. The fact that a number of pupils in their texts indirectly expressed a critical attitude to their parents needs to be approached critically.
Furthermore, a comparison of narrative content in pupils' texts in L1 and L2 might study whether there are differences in the type of thematic content that is taken up in the L1 versus in L2 English. The analysis in the doctoral study suggested that pupils took the opportunity to question the power of grown-ups in society. They were critical towards both teachers and parents. To study whether there are qualities in L2 English language narrative practices that allow pupils to express criticism would be an interesting research question to pursue.