TEACHER STUDENTS’ LEGITIMATION OF TEACHING ACTIVITIES IN THE L1 CLASSROOM

This study investigates Norwegian L1 teacher students’ legitimation of teaching activities as they retrospectively reflect on internship experiences during their first two years of teacher education. We analyse the legitimation language in qualitative interviews with 20 teacher students using the specialization dimension from Legitimation Code Theory. The findings indicate a diversity in the teacher student’s legitimation codes. The dominant pattern of codes indicate that L1 is legitimized within a horizontal knowledge structure with few opportunities for construing a subject with a unifying goal. A mino rity of the teacher students legitimized L1 by a strong classification motivated in young pupils’ cultural needs in order to partake in a textualized and mediated world. Drawing particularly on Basil Bernstein’s ideas on different knowledge structures and on teaching and acquisition of humanistic subjects, we critically discuss how the interviewees’ legitimation works to classify L1 as a distinct school subject. Using the examples of strong classification of L1, we discuss possible measures to assist teacher students’ development toward professional subject didactic reasoning and legitimation.


INTRODUCTION
How teachers and teacher students talk about their own practice, and how they choose to describe and justify their practice, matters. Teachers' classroom stories contain information about the teller's conceptualization of a subject's specific knowledge, its boundaries, connections and goals, how it is learned, as well as the teller's legitimation. Legitimation implies giving reasons for specific teaching activities as well as the subject's long-term educational objectives, and could derive from different sources like the pupils, the academic discipline, cultural needs and demands, the curriculum, everyday theories, psychology, and so on. Legitimation might be considered a language or a code that functions as a tool for recontextualizations when a subject's body of knowledge is translated into teaching activities (Bernstein, 2003, p. 60;Moore, 2013, p. 155;Nemeth, 2021, p. 52). Regarded as cognitive and linguistic tools legitimations shape and control teaching activities in the classroom, and at the same time provide them significance and meaning. However, legitimation might be hidden, naturalized or self-evident to the actors, and embedded in everyday theories on how the world works and how children learn. In this study we ask: What kind of legitimation codes is used by teacher students to add meaning to their choices and teaching activities in the L1 classroom, and how is the legitimation codes embedded within the L1 subject?
When investigating Norwegian L1 teacher students' legitimations as they retrospectively reflect on internship experiences from their first two years of teacher education, we draw on concepts from a (re)new(ed) field of legitimation-Legitimation Code Theory (LCT) (Maton, 2014). Several contemporary empirical studies make use of LCT in order to analyze and redesign teaching and curricula in higher education (see for instance : Clarence, 2016;Gachago et.al., 2021;Maton & Chen, 2017;McKenna & Boughey, 2022;Rusznyak, 2022). The benefit of using LCT is the inclusion of both a theory of knowledge making it possible to confront challenges of enacting disciplinary pedagogies and working with specific bodies of knowledge (Bratland & Ghami, 2021;Winberg et al., 2020) and of the voice, or the gaze, of the actors of the teaching and learning process. The latter seems to be of particularly relevance in Humanities (Luckett & Hunma, 2014). These studies are situated in contexts similar to the Norwegian where student centered teaching approaches are favored as means for student construction of knowledge. A common backdrop is that a "conflation of pedagogic and epistemic constructivism in enacting these approaches ['inquiry focused', 'authentic', 'dialogic' and 'student-centered' approaches] to teaching and learning, […] may obscure differences between disciplinary knowledges and practices" (Clarence, 2016, p. 123). The studies argue that such teaching strategies are not sufficiently able to address the students' needs for disciplinary framing when reflecting or mediating the discipline's insights and approaches to the "world".
To investigate how L1's body of knowledge is framed and classified in teacher students' internship stories, we make use of one dimension of LCT; the Specialization, that focuses primarily on the nature of a field; that is "what makes this field, and those within it who count themselves as belonging to it 'special'?" (Gachago et al., 2021, p 149). Bernstein's concepts of classification and framing will be used to surface what is valued or worthy of distinctions and what is recognized as specialized practice, that is, "what counts" (Luckett & Hunma, 2014, p. 187) as a legitimate practice in each of the specialization codes. By investigating the legitimation of each code, we emphasize the students gaze as a mechanism for recontextualization of L1's administered body of knowledge into classroom practice.
Legitimation language is not a recent concern among L1 scholars. The Norwegian L1 scholar S. Ongstad (2004, p. 14) calls for a teacher discourse of meta and consciousness: "When you through the what and the how have created an epistemological basis for the teaching, you must be able to legitimate the opt (out)" (p. 167, our translation). Ongstad describes legitimation as the core element of subject didactics and argues that teachers should be expected to give overt reasons for their recontextualization of an academic discipline into classroom activities. It is only through legitimation that teachers can contribute to the development of a profession, Ongstad claims (p. 166). Hence, teacher educators must have two aims: In addition to teaching the subject, they should nurture the why, i.e., a didactic metaperspective of legitimation.
A recent Swedish empirical study points out challenges relating to the legitimation of L1 (Nemeth, 2021). The L1 teachers interviewed for the study consider the subject's specific professional knowledge to be downplayed in the Swedish curricula, where knowledge is constructed as measurable skills. They expressed uncertainty about the significance and meaning of many of the curricula goals, and, to some extent, they consider the curriculum to undermine aspects of L1 that the teachers themselves consider important and meaningful. With reference to Bernstein, Nemeth explains this downplaying of the subject's academic content as a product of L1's horizontal knowledge structures, with blurred boundaries to everyday language and everyday knowledge. Hence, Nemeth calls for a stronger classification of the Swedish L1 subject, and an orientation towards a clearer vertical discourse where knowledge is connected to mental and long-term processes to retain its legitimacy. Nemeth suggests that such a legitimation should derive from reading, intertextuality and deeper understanding of the different layers of texts.

CONTEXT: NORWEGIAN CURRICULUM AND TEACHER EDUCATION
In the following, we argue that the legitimation of teaching activities, particularly in Humanities, is of acute concern in educational contexts influenced by the OECD's framework. The Norwegian curriculum, implemented in 2006 and renewed in 2020, was adjusted to the OEDC's concept of goal-oriented teaching based on skills and cross-curricula competences. This changed the concept of knowledge, where L1's cultural and historical elements were toned down in favour of a highly constructivist knowledge concept (Foros, 2012;Holmberg et al., 2019;Nergård & Penne, 2016).
Replacing a curriculum inspired by E.D. Hirsch's concept of cultural literacy, with common knowledge measures for all pupils and where content and specific texts were listed, the 2006 reform promoted the idea that each subject should define measurable competencies pupils should acquire. The change is explained by the Norwegian Directorate for Education and Training in the preparatory documents for the reform: Norwegian teacher education programmes are closely linked to the curriculum's ideas and ideology, as well as the standards of the Bologna Declaration, which states that programmes and subjects should be presented based on learning outcomes in terms of knowledge, skills and general competence. However, descriptions of goals and outcomes do not explain the nature of the knowledge administered by each discipline, nor how the subject is learned and how it relates to cultural and individual needs.

Knowledge structures, classification and framing
According to Bernstein (2000, p. 163), knowledge is distributed and valued differently within different subjects. Knowledge administrated by subjects such as L1 is characterized by a horizontal structure and is developed by acquiring new segments or discovering new perspectives or connections. The content within these subjects is negotiable and might have weak boundaries to other subject and to everyday knowledge. Horizontal knowledge structures are epistemologically more vulnerable, since their classification is of a more constructive nature than subjects of hierarchical structures where knowledge is mediated through a specialized discourse and insight are gained through general principles and processes towards increasing levels of abstraction (Bernstein, 1999).
The Norwegian L1 curriculum allows teachers great freedom as to "what" should be taught and "how" it should be taught and might be considered what Bernstein calls "weak" in terms of classification and framing. Classification "refers to the degree of insulation between categories of discourses, agents, practices, contexts […]" (Bernstein, 2003, p. 214), and is characterized as strong or weak depending on how isolated the different discourses are from each other. Discourses with a unique identity, voice and specialized rules indicate strong classification, whereas weak classification implies less specialized knowledge and language as well as unclear boundaries in relation to other subjects, everyday knowledge and everyday language. The concept of framing addresses how a pedagogical discourse is accomplished, referring to "the control on the selection, sequencing, pacing, and critical rules of the pedagogic communicative relationship" (Bernstein, 2003, p. 214). Strong framing is characterized by explicit rules and instructions, and a teacher in control of the selection of content, organization of the practice, and pacing and timing. Weak framing, in contrast, allows the learner to take more control, for instance through student-led teaching or symmetrical relations between teacher and student.
As outlined, the Norwegian L1 subject comes across as epistemologically vulnerable both because of the horizontal knowledge structure, but also because of the curricula's recontextualization of knowledge to a selected number of skills and competences (Nergård & Penne, 2016). Hence, the task of legitimation is left to the individual teacher. The shift from a content-based to a competency-based curriculum went together with a change from a prescribed teacher role of knowledge dissemination, relating to content dimension, to a more "invisible" teacher role that emphasised the active and responsible pupil who should construe the subject and its content (Dale, 2010;Foros, 2012;Skarpenes, 2005). This shift thus entails a curriculum where both framing and classification are weakened.
Thus, there is a particular need for legitimation in L1 to support the recontextualization processes in the classroom. On this basis, we argue that teacher students' legitimation is of acute interest for L1 scholars as a source for developing L1 subject didactics and further developing L1 teacher education.

Legitimation Code Theory: A framework of legitimation
Based on Bernstein's work on knowledge structures LCT provides an analytical framework to investigate differences in legitimation of teaching activities by the actors in an education system and make visible the mechanisms at play in the classroom (Maton, 2014). In LCT Bernstein's concepts of classification and framing are reshaped into specialization codes which describe structures of legitimation. According to Maton (2014), action is always directed at something and performed by someone. Thus, it is possible to make an analytical distinction between epistemic relations (ER), expressing the relationship between action and the object of the action, and social relations (SR), expressing the relationship between action and the person performing the action (Maton et al., 2016). Both classification and framing might then be separated in two different axes, ER and SR, and can be given relatively strong or weak emphasis. The different emphasis on epistemic relations and social relations can be visualized as a topological space with four different modalities (Maton et al., 2016): Figure 1. The specialization plane (Maton, 2014, p. 30) The specialization plane ( fig. 1) illustrates those four different positions. The knowledge code is a position where "specialized knowledge, principles or procedures concerning specific objects of study is emphasized as the basis of achievement, and the attributes of actors are downplayed" (Maton & Chen, 2020, p. 38). This position represents strong classification and framing of epistemic relations and weak classification and framing of social relations. The élite code is a position "where legitimacy is based on both possessing specialist knowledge and being the right kind of knower" (Maton & Chen, 2020, p. 38). This position represents strong classification and framing of both epistemic and social relation. The knower code is a position "where specialized knowledge and objects are downplayed and the attributes of actors are emphasized as measures of achievement" (Maton & Chen, 2020, p. 38). This position represents strong classification and framing of social relations combined with weak classification and framing of epistemic relations. The last position is referred to as a relativist code, expressing weak classification and framing of both epistemic relations and social relations.
The specialization codes conceptualize underlying structures and principles in the legitimation of for instance teaching activities and shows how knowledge is distributed and assessed in different ways. The dominant code may differ in different subject, in different classrooms, and in different stages of a curriculum.

Participants and procedures
The data used in this study were collected through individual qualitative interviews with 20 students in the teacher education programme for years 5-10 at one of the largest teacher education institutions in Norway. The informants (4 male and 16 female) were recruited from a group of 58 students at a course in L1 didactics. The selection was a convenience sample, based on availability. The entire group was informed about the research project and what participation would entail. Participation was voluntary, and the students could withdraw at any time. All the students who volunteered were included in the study.
At the time of the interview, the interviewees had completed 60 ECTS of L1, which is compulsory for teaching at the year 5-10, and 45 ECTS of pedagogy. They had also started their training in other optional subjects such as social studies, religion, science, arts and crafts, music and physical education. The interviews were conducted towards the end of the students' second year of the programme, immediately after they had completed their fourth period of internship.
The purpose of the interviews was to gain knowledge about the students' legitimation language while reflecting on classroom experience. We understand linguistic reflection on knowledge and classroom activities as expressions of knowledge that can be applied or translated into action. Thus, the informants' reflections on teaching practice in the classroom will be interpreted as expressions of recontextualization (Bernstein, 2003). The interviews were semi-structured and based on an interview guide (Appendix 1). The interviewer encouraged the interviewees to share teaching experiences from their internships and to reflect on the "what", the "how" and the "why". To motivate the interviewees to reflect on and to legitimate their choices, the interviewer followed up their classroom stories with questions about why they had chosen a specific content or activity, what goals they expected the pupils to achieve, how they carried out the work etc. During the interview, they were explicitly encouraged to comment or reflect on their classroom activities based on the text theory and didactics theory emphasized in their study programme.
The interviews were conducted digitally on Zoom or face-to-face according to the students' preferences and lasted between 30 and 60 minutes. They were all recorded and transcribed, and the analysis of the material was based on the transcripts. The interviews were read and categorized according to the analytical terms by both authors in several stages. We firstly read the texts and made preliminary categorizations, and then discussed and revised the categorisation of each interview during three meetings.

Ethical consideration
The two authors work as teacher educators at the institution where this study was conducted and had previous knowledge of the interviewed students from their first years of study. The data analysis process therefore involved continual and deep selfexamination to enlighten our pre-understanding, role and influence on the process (e.g., Berger, 2015). The research project and data collection process were approved by the Norwegian Centre for Research Data (NSD), and we have used NSD's data management plan throughout the study. The participants signed written informed consent before the interview. After the interview, the recordings were transcribed, and the recordings deleted.

Analysis
We used two analytical approaches to explore the content of the interviews, and in the following we will explain the two approaches in more detail.

Legitimation language and specialization codes
First, we investigated the interviews using the description of epistemic and social relations from Maton's specialization plane. In this phase of the analysis, we used a deductive approach in the interpretation. We marked statements from the interviews where the interviewee emphasized subject-specific knowledge and statements conveying that procedures for learning were explicit to learners as strong epistemic relations (ER+). Contrary statements where the teacher students downplayed subject knowledge were marked as weak epistemic relations (ER-). Likewise, statements implying that procedures are implicit to learners and not significant to learning were coded ER-. Statements in which the student's personal experiences, interests and attitudes were given importance, and statements where individual learners' preferences were explicitly emphasized were coded as strong social relations (SR+). In contrast, statements downplaying individual learners' preferences as not significant were coded as weak social relation (SR-). Also, statements implying that personal experience, characteristics and attitudes were not considered legitimate content were coded SR-. An overview of the translation device showing how the theoretical concepts are realized in the empirical data throughout the analysis is given in Appendix 2. The marked statements were labelled and then counted to explore whether any of the legitimation codes were dominant in each interview.

Legitimation codes in the context of L1
The second analysis seeks to answer the second part of the research question and is an exploration of "the gaze" of the four codes. The analysis substantiates Maton's specialization codes in the context of L1, as the focus is on how knowledge administered by the L1 subject is recontextualized in teaching activities. A discursive approach means searching for expressions of legitimation that fuel the recontextualization process and give meaning and value to the teacher's choices and actions in the L1 classroom. We regard subject didactic legitimation as a language with the potential to create coherence between the teaching activities on the one hand and ideas of L1's potential to meet the pupils' needs, theories of learning and knowledge concepts on the other. The purpose is to investigate whether the legitimation codes form distinct, coherent reasonings and logics specific to L1.

FINDINGS
In this section, we present findings from both analyses, starting with the findings that provide an overview of the students' orientation according to specialization codes. Figure 2 shows the distribution of the interviewees in the specialization plane based on the coding. As the figure shows, there is great diversity in legitimation codes among the interviewees, with the "knower" code as the most dominant. Nine of the teacher students are positioned as knowers. These students highlight the pupils' motivation and interests when choosing content and teaching activities in the classroom. Further, they emphasize the well-being of the students and pay attention to them as individuals by emphasizing their motivation and feelings and facilitating their individual goals. The knowers also turn the attention towards themselves for being the "right person" with the right skills due to, for instance, having "younger siblings", "experience with helping others", or the ability to "see everyone".

The students' orientation regarding epistemic and social relations
The legitimation language of four of the interviewees point to a knowledge code. They emphasize subject content and knowledge as the main point of schooling, " […] simply in order to receive a good formative education", according to one of the interviewees. They describe knowledge as strongly controlled and bound to the teaching profession. On the other hand, they downplay individual adaption and the importance of interests and individual characteristics.
The legitimation language of five of the teacher students matches the élite code. These students emphasize subject-specific knowledge but at the same time point out the importance of individual motivation and interest. As in the knower code, they refer to the importance of having a special kind of competence to be a teacher, and mention "enthusiasm for the subject", the ability to "lead learning processes" and the ability to motivate others for "the subject".
Two of the interviewees' legitimation languages match the last position, the relativist code. These interviewees downplay subject-specific knowledge and subject-specific processes, and, at the same time, downplay personal experience and individual learners' preferences.

Specialization codes in the L1 subject
In the following, we turn to the second part of our research question, the question of the interviewees' legitimation of the L1 subject and how this legitimation is discursively expressed by the students in the four positions. We will present the gaze of one interviewee representing each code based on three aspects: the perceived main object of the L1 subject, the descriptions of the specific pedagogical activities and the reasons given for these activities. The first aspect-the main object-is regarded as a possible mechanism for recontextualization. When using aclantate examples from each code, we describe tendencies in the legitimation of all informants categorized under the same code.

Knowledge code
Common for the knowledge code is the use of a variety of concepts from a professional discourse such as: "Bildung", "literacy", "participance in democratic societies", "cultural knowledge", "identity", and "flexible thinking". These concepts frequently yield meaning to the activities in the classroom but do not specify the subject's knowledge content. The legitimation of the classroom activities seems to derive from a framing based on general pedagogy rather than a classification of L1. In other words, there seems to be a drive for evidence-based knowledge rather than classification of L1 in this code. The concepts used to legitimate teaching activities are for instance "monological and dialogical teaching", "videos, power points, excel, smartboard", "the books and the reading lists" and "primary skills", but sometimes also concepts more specific to L1 such as "interpretive community", "orthography", "primary texts" and "contrasting sources".
When asked about L1 teachers' most important task, the interviewee selected to represent the knowledge code refers to the L1 subject in horizontal segments: "Oh, it's so many things.… It's about literacy. It's about reading. It's about writing. It's about interpretation. It's about source evaluation. It's about critical thinking, which is nearly the same. Reflection. Reasoning. Oral skills." Hence, L1 is not legitimate through a generalized object that would construe a vertical dimension. Rather, legitimation derives from a subject conceptualized as skills and competences, where none is given priority over the other. This conceptualization is in accordance with the goal-oriented Norwegian curriculum where knowledge is conceived as the skills and competences that each subject is responsible for imparting.
When explaining teaching activities, the interviewee refers to procedures that might fit any subject: It's the combination of the teaching-oriented and student-oriented phases that are in focus-acquisition, testing and consolidation, where the acquisition phase is teacher dissemination, the testing phase is the exercises and the consolidation phase is a metaconversation about what has been learned.
The teaching activity he chose to present was the writing of crime stories, and we see a value put on procedures, in this instance "the acquisition phase": We had an introduction where we shared ideas. What characterizes a crime story, the dramaturgic arch, introduction, main part and resolution: The classic characteristics of a crime story. So that was the phase prior to the writing. Then we made use of writing frameworks. Before they started writing, we supervised them, gave them ideas, so they were informed about crime stories, and we showed them some clips from YouTube as well.
The legitimation of the activities connected to the crime genre does not serve to classify L1 as a specific field of knowledge with specific generalized objects, and the interviewee regards the activity as successful since he "felt that many pupils got it".
In this code, professional teaching is a matter of knowledge conceived as an evidence basis for work procedures, and legitimation is related to the structure of action and evaluation of a finished product. The procedures are the dominant mechanisms for the recontextualization and L1's administered knowledge is adjusted to these structures. Hence, the perceived professional knowledge seems to derive from two sources: general pedagogy and the national curriculum, and to a lesser extent from knowledge administered by L1. In this code, L1 is conceptualized as horizontal segments without a vertical dimension that would yield meaning and construe a coherent subject.

Elite code
The legitimation of teaching activities in the élite code is presented both as an integral part of the discourses and expressed explicitly. L1 is presented with a generalized object that gives subject content a vertical dimension. L1's main object is explained in the perspective of a highly textual and mediated contemporary culture, and the recontextualization process derives from this legitimation. All five interviewees regarded text awareness to be a principal object of the L1 subject, related to genre work for the development of contextual awareness. Three of the five students presented additional objects. These were equality as a literacy project regarding socioeconomical differences linked to discourse and critical thinking; linguistic and discursive awareness, expressed through phrases like "language is power"; and the need to approach texts and literature as constructs of meaning that represent possible perspectives on "the world". We will describe the latter of these in more detail to represent the élite code. The interviewee regarded the objective of L1 to be creating awareness of how humans construe the world through stories, expressed in the interview through phrases like "a certain understanding of the world", "to understand the world [through literature]", "other perspectives [than your own]", "be allowed to enter new roles", "you [the pupil] don't have to be yourself". This objective is presented as an existential condition for living in a diverse world where truth is most often a question of perspective:

Today's society has many truths, and truth is hard to find. … giving the pupils knowledge on what the world can contain may make it possible to understand. So, I think L1 is a valuable subject to maintain meaning in being. I think meaning can easily disappear when truth is teared away from us and disappears so easily.
The overall meaning of humanistic subjects is perceived, through text work, to create an awareness that the world can be construed in different ways. The key for such awareness is explained as creating a distance to "yourself", your beliefs and your preferences through the understanding that others might construe the world differently. The development of such a perspective requires meta-conversations in the classroom because the L1 subject and its objectives might appear "vague": The point is that you structure the L1 lesson so that it's possible for the pupil to leave the classroom knowing that "I have learned this or this", and to put what is learned into words.
Such meta-conversations are also regarded as a precondition for engagement in learning activities:

It's about taking learning into your own hands. You want the pupils to amend the subject content, use it. Then something must come from "inside". You don't want to taste ingredients that you don't know. The fruit is there, but how do you peel it? If you don't know, it's hard to peel. But if you get some information about the fruit, know that it should taste like this or that, then you might taste the citrus flavour. Then you dare to open it up and look, to dig for the seed. It nurtures the engagement and takes away the fear of not completing the task.
During internship, the interviewee was assigned the task of teaching the drama genre and saw this as an opportunity to practise three specific aspects. Firstly, it provided the opportunity to invite the pupils to enter roles so as to act and speak from "a different perspective": "If you are not allowed to enter roles from time to time, you will have difficulties learning how to express yourself". Drama also offered the opportunity to separate language and interpretation from "oneself": "You don't always have to stand out as yourself: 'When you were the Empress, you were loud and had a very distinct voice. It was important for you to let her come across as selfobsessed'". Secondly, drama presented an opportunity to investigate the world as it is construed in certain types of narrative genres. The drama genre was also used to study the text as a filmic adaption. The student group working together during internship contacted the film production team, asking for parts of the manuscript, "So we could see how the text was formed, paragraph by paragraph. To understand how this is a distinct kind of text that gave instructions yet in a particular way".
In the élite code, legitimation is often presented unsolicited and expressed explicitly, indicating a high degree of awareness of what has been going on in the classroom. A vertical dimension is construed through work on different texts that nurture textual and contextual awareness. By this, L1 comes across in these discourses as strongly classified, with a strong framing connected to individuals' need for text competence in contemporary cultures. The expressed main object, textual awareness, works as a mechanism for a recontextualization of the subject content into teaching activities that create coherence between the subject's various segments.

Knower code
In the knower code, the classification of L1 is weak. Legitimation in this code has two sources: the pupil and/or the teacher student. The discourse centred around the teacher student particularly provides legitimation when it comes to the subject's "what", as seen in phrases like "What I find interesting …", "I myself have always liked reading [literature]" and "When I learned about reading strategies, it became clear to me…". In this code, teaching becomes a question of recontextualizing the teachers' knowledge and preferences into teaching activity in order to provide the pupils with the insights of the teacher. The construction of a vertical dimension is difficult in this code, as knowledge is relativized and belonging to the teacher. The code that inhibits a pupil-centred legitimation also emphasizes the teacher's knowledge, but this knowledge is relational knowledge, i.e., the pedagogical choices and actions are recontextualizations of the teachers' (proclaimed) knowledge about the pupils, both as individuals but more often as a generalized group. Legitimation of classroom experiences is linked to what can be enjoyable and relatable for the pupils rather than to the L1 subject, and a frequent phrase used to evaluate classroom activities is the distinction between "boring" and "fun": "We have met many pupils who think it's quite boring all of this, so we have always considered how we could make the learning fun", and "We did drama for the fun of it". This code highlights a view of content knowledge that is highly constructive: Knowledge is regarded as something that is construed in the classroom, and as a product of dialogue and cooperation between peers and teachers. Individual facilitation is considered a prerequisite for learning as such adjustments make the pupil feel good and increase motivation. In this voice, the well-being of the pupils is emphasized, and attention to them as individuals is explicitly and repeatedly expressed.
According to the selected interviewee for this code, the most important task for the L1 teacher is "actually to include everyone in the teaching", and the L1 classroom is regarded as a particularly important arena for shaping a classroom community. A strong classroom community is regarded as a precondition for learning: "If there is little companionship in the class, there is little learning". Knowledge is not conceptualized as something that is given within the subject's framework. Instead, a highly sociocultural and constructive perspective on knowledge is promoted: "Students should base their learning on each other, so I'm very concerned with working with partners, groupwork".
The interviewee mentions source evaluation as an important objective in L1 classrooms, though it is not legitimated as a cultural need as such, but through a claimed knowledge of the pupils: Children use social media a lot and expose themselves to lots of information uncritically, like "Yes, so that's how it is! Now I know that" without browsing for more information about it.
When triggered to elaborate on this, the interviewee slightly changes the concept from source evaluation to critical thinking: "Yes, critical thinking! I'm thinking text work, talking about texts, about how credible they are, if you can trust them […] who is communicating and receiving, and things like that". Later in the interview, critical thinking is conceptualized as something that could enhance relational values: "You are able to develop the critical thinking that everyone must respect everyone, to see that things can be different from our own line of reasoning".
The interviewee considered the assigned drama genre as an opportunity to "make the students more comfortable in the classroom, because they have different [social] preconditions". Further, that drama could "contribute to making the classroom community better". The alternative objects "critical thinking" or "text work" are not used to legitimate either the "what" or the "how", and the drama lessons were evaluated as successful because: "What we saw was that pupils that normally wouldn't be so much together, performed the drama together, and that it later wasn't so scary to talk to the others in the class." The recontextualization in the knower code has two sources; what the teacher knows and considers important, and the teacher's knowledge about the pupils. For the latter, we find that L1 is classified as a particular arena for establishing class as a "safe space", downplaying L1 knowledge as significant for the pupils. The interviewees sometimes mentioned objectives of a more general and paramount nature, but these do not provide legitimation for the activities implemented in the classroom, nor are they included in the evaluation of the pedagogical actions.

Relativist code
The characteristics of the two interviewees using relativist code is weak classification and weak framing. All aspects of the teaching presented have pros and cons, whether pupils use printed books or just an iPad and the internet, or the experience of taking an internship physically or on a digital platform (due to : "For me, digital teaching was actually really good. I had to travel for an hour and a half to get to the school, so it made everything easier". Both are particularly preoccupied with the "what", and in many respects, they are opposing the questions of "why": "I should have known more about teaching methods". However, methods are not highlighted as professional procedures: "I feel when you start a classroom discussion that it just evolves". Common for both interviewees is that much of what was going on during the internships is forgotten or just vaguely remembered. Unsolicited, the selected interviewee brought up aspects of the internships that were perceived as negative: "After the lessons, you are supposed to stay behind for a long time, talking about 'Why did you do this or that'. That is of course part of it, but sometimes you choose to do things in a certain way, even though you have not paid particular attention to it." Finding the reason-giving tiring, the interviewee would have preferred the supervised training to be focused on the "what" and the "how": "I would really have liked to know more about teaching methods that could be used for different purposes".
The main object of L1 is presented as the training of basic skills in relation to reading and writing. The accounts of such training are vague with a focus on organizational aspects in training basic skills, rather than text competence, interpretation or contextual awareness: Int: How did you work with reading? K25: I remember we had a reading circle.

Int: What's that?
K25: They sat on the floor together, and then we took turns reading, and they could read as much as they wanted. What book? I don't remember.
Int: But did you read the same book?
K25: We all read one book and listened.
Int: Did you read the whole book?
K25: I think so. We talked about the book, but I don't think we didn't do much more than that.
The same pattern is evident when talking about writing practice: "I remember that we used workstations, we worked with word classes, and then I remember some writing. It was not a long text, maybe a letter or something." When triggered to legitimate the activities, the answer signals little interest in the reason-giving: "I actually thought about that. It might be that the main teacher based it on the curriculum, but that wasn't clarified. Anyway, we didn't pay much attention to it." In this code, legitimation is embedded within the L1 subject only by the notion of reading and writing as basic skills. The professional knowledge in this code does not derive from the L1 subject. Rather, knowledge is related to procedures. Yet in contrast to knowledge code, these methods are denoted by names and not as structures. Such knowledge is regarded as favourable, but not essential. The object of L1 is training in basic skills, and the recontextualization is restricted to making the pupils "do", in terms of making them read and write.

DISCUSSION
The aim of this study is to explore the legitimation language used by teacher students to add meaning to their choices and teaching activities in the L1 classroom, and how the legitimation language is embedded within the L1 subject. We find a great diversity in the teacher students' legitimation language, and all four codes in the LCT's specialisation plane is represented in the data. However, it is only in the élite code that a vertical dimension is embedded in the discourse making a strong classification of L1 possible. Further, in the knower and relativist code epistemic knowledge is downplayed in the legitimation of the teaching activities. For L1 scholars, this is a disappointing result. Nemeth (2021) reports the same tendencies in Sweden as we find in our data. She calls for a stronger classification of the Swedish L1 subject.
Contrary to the intentions of a goal-oriented curricula, our data indicate that the actors' gaze becomes of great significant for the recontextualization process in an educational context such as the Norwegian. A teacher education that is influenced by the compulsory school's OECD-adjusted curricula with a highly constructive knowledge concept and an emphasis on skills and cross-curricula competences, signals both a weak classification and a weak framing of school subjects (Bernstein & Solomon, 1999;Maton, 2014). Adding Bernstein's insight that subjects with horizontal knowledge structures are epistemically vulnerable, also implies that the individual teacher's gaze and legitimation language matters greatly even though curriculum goals are defined. Particularly in the knower and the relativist code, teaching activities are legitimated from vague or "invisible" concepts of knowledge on the expense of epistemic knowledge embedded in L1, though in accordance with the knowledge concept of the curriculum. For instance, in the knower code's this vagueness seems to nurture legitimation based on everyday theories that mediates that knowledge will be developed and that learning will happen once the classroom feels comfortable and safe. In the knowledge code, a vertical discourse is found in general pedagogy rather than in L1, a finding that alludes to Haugen & Hestbek's (2017) study on the differences between higher educational curricula.
In the élite code, the epistemic knowledge, in this case linguistic and textual knowledge, as well as understanding of the function of languages and texts, works to yield meaning and coherence in the L1 subject. However, there is a layer of knowledge beyond the epistemic knowledge that works as a recontextualizing device, where subject knowledge is recontextualized by ideas of what is important for the pupils' life and future. This knowledge is of a more abstract nature than the subject's epistemic knowledge and might be denoted as a meta-perspective or a more generalized perspective where classroom activities are legitimated through acute needs to navigate in and understand a textual world. In line with literacy theory, this meta-perspective could be understood as a critical stance involving "the use of texts to analyze and transform relations of cultural, social and political power" (Luke & Dooley, 2009, p. 1). This means understanding how texts work, in which situations they work, with what consequences, and in whose interests (Luke, 2012). It is also about understanding that texts never give neutral representations of the world but are constructed representations where the text creator, through language and other semiotic resources, conveys specific values and ideologies. Textual competence is linked to opportunities for influence and participation in society. For example, it concerns how linguistic resources and mastery of text can both exclude and provide access in a society (Nicolaysen, 2005). This perspective generates distance and possibilities for verticality in a horizontal knowledge structure, where knowledge is mediated through language, text and discourses. Different from the knowledge code, this code helps to create a coherent, classified L1 subject and, at the same time, to regard learning as specific to the L1 subject.
The élite code is executed in the same educational context as the other codes, but this code classifies L1 through a vertical discourse where subject-specific activities are explained and legitimized by a defined general object for L1. This vertical discourse is related to epistemic relations. The vertical discourse construes a continuum from text work to linguistic, textual and contextual awareness. The code connects the various segments of L1 through the linguistic functions, text competence and interpretation, and of texts' significance in our contemporary culture, thereby establishing literacy competence (Blikstad-Balas, 2023;Veum & Skovholt, 2020). This could be understood as a parallel to Nemeth (2021), where she links the vertical aspect to reading and in-depth understanding of text. To approach texts and literature as meaning constructs representing versions of the world, opens possibilities for understanding that texts work to establish relations of power and how text competences and language might work as a counter-power to textual and institutional power. In the élite code, textual and contextual awareness are linked to access to society or participation in culture, and each small step of recontextualization should serve the development of such awareness.
Our study indicates that there is a need in teacher education to nurture a vertical discourse embedded in L1. Based on our empirical data, it seems that the interviewees who reason within an élite code emphasize and specify teaching activities through a meta-perspective derived from an understanding of what contemporary culture demands of individuals. This discourse differs from the discourse of psychology, intimacy and emotions in social relations, because it is founded on ideas of cultural descriptions and cultural needs. By taking a cultural perspective rather than relying on the curricula or evidence-based knowledge from general pedagogics, a vertical dimension is established in the élite code that appears to yield meaning to the epistemic knowledge administered by L1. Bernstein's code theory aims to make visible the distribution of power and principles of control in the pedagogical discourse. However, in our OECD-adjusted teaching culture, important aspects of power are administered by teachers. Based on our empirical data, we therefore propose that a vertical discourse addressing cultural needs should be emphasized in L1 teacher education. It would entail promoting linguistic, textual and contextual awareness in L1 education both to classify L1 and to empower young individuals that are more textually exposed and mediated than any generations before them. Would you tell me about an educational activity you conducted in the class? -What were the goals of the work? How did you work? How did it work? Did the students "like" it? Why did you choose this specific activity? Would you tell me about an educational activity you conducted which involved reading or writing texts? -What were the goals of the work? How did you work? How did it work? Did the students "like" it? Why did you choose this specific activity and this specific text? What is the most important assignment in the L1-subject or the most important aspect of the subject? What do you think about the connection between the L1-subject at the university and in school? How did you use the new curriculum, LK20? Teaching other subjects What subject did you teach in addition to L1? What is the most important assignment in subject 2? What do you think are the differences between L1 and subject 2? Did you do any work involving texts in subject 2? What do you think are the main differences or similarities in the approach to text in the two subjects? Concluding Questions You have, for now, finished your studies in the L1-subject. How did you experience the different aspect or parts of the subject? What was the significance of the last part about subject didactic and literacy? Who are you as a teacher? Who do you want to be? We just made a google docs presentation like that, with some themes, and we came up with some assignments for the students to do this week.