Digital teaching as the new normal? Swedish upper secondary teachers’ experiences of emergency remote teaching during the COVID-19 crisis

This paper takes an interest in how schools and teachers dealt with new demands when teaching rapidly went online during school closures related to the COVID-19 pandemic, in what we see as an example of emergency remote teaching. The aim is to make visible how schools and teachers dealt with the demands that they were confronted with while under hard pressure during emergency remote teaching, and what discursive frames are used in upper secondary teachers’ pedagogical considerations. Fifteen teachers of history, mathematics and Swedish (five from each subject) are followed in recurring interviews between April 2020 and September 2020, resulting in a total of 41 interviews. A narrative approach is used in the analysis and results show how teachers made large efforts to maintain teaching in what can be described as a crisis organization. Three main discourses are identified: (a) a strong assessment discourse; (b) a relational discourse; and (c) a compensatory discourse. The findings are discussed in the light of educational policy based on the so-called Nordic model and the idea of one-school-for all, and in relation to what becomes possible to teach as well as what is not possible to do in times of crisis.


Introduction
School systems all over Europe have had to adapt to some level of online remote teaching during the COVID-19 pandemic. According to the United Nations Educational, Scientific and Cultural Organization (UNESCO, 2021), more than half of the world's students still face significant disruptions to their education, where the European average is about 17 weeks of school closures. Also, in Sweden, which is the case in this article, upper secondary schools were forced to shift to full online teaching at extremely short notice in March 2020, a situation that lasted throughout the remainder of the spring semester. In this article, we investigate this period of transition and the challenges that teachers faced as they had to move their teaching from the physical classroom to digital online solutions. Initially, this situation was sometimes described as a 'great on-line learning experiment' (Zimmerman, 2020) that could advance our understanding of digitalisation in education. And certainly, the unexpected and large-scale changes could be seen as a test and perhaps a catalyst for wide-spread digitalisation policies in all of Europe and the western world (Organisation for Economic Co-operation and Development (OECD), 2016; Redecker, 2017;Swedish Ministry of Education, 2017). However, Hodges et al. (2020) argue that there are reasons to be cautious about jumping to conclusions about the potential for development of online teaching based on this unique situation. Instead, they say, the present situation must be understood as a response to a very specific crisis and could best be described as Emergency Remote Teaching (ERT) as it is not teaching that is planned to be online, but 'a temporary shift of instructional delivery to an alternate delivery mode due to crisis circumstances' (Hodges et al., 2020). Bearing this in mind, we still think that the rapid need to mobilise digital solutions that schools were faced with is an event with unique potential to bring new perspectives into research about education in the era of digitalisation. In this paper, we start from an understanding that the development taking place during this specific crisis reveals not only how this emergency situation is managed with the help of digital technology. It also makes visible educational discourses that become influential in decisions during a time of crisis. In this study, the aim is to make visible how schools and teachers dealt with the demands that they were confronted with while under hard pressure during ERT, and what discursive frames are used in upper secondary teachers' pedagogical considerations.

ERT in the context of Swedish upper secondary education
The backdrop of this study is Swedish upper secondary schools. In Sweden, almost all adolescents attend upper secondary school, either in programmes preparing for university studies or in vocational training programmes, and most teachers meet students with a relatively broad variation in terms of social backgrounds and educational needs. Similarly to the situation in other Nordic countries, Swedish education has a long tradition of being considered as a key to the development of a fair and equal society supporting democratic values, welfare and life-long learning under the credo one-school-for-all (Blossing et al., 2014;Klette, 2018). Even though challenged due to reforms regarding school choice, academic performance and the influence of global competition between schools (Lundahl, 2016, Telhaug et al. 2006, the idea of a Nordic model of education is still based on a vision that schools should be inclusive, comprehensive and non-tracked until as high an age as possible. As the solutions to the challenges brought by the crisis are to a large extent digitally mediated, another backdrop is digitalisation of education. In a European perspective, use of and access to information and communications technology (ICT) prevail in Swedish classrooms at a level above average in the OECD countries (OECD, 2020). Previous studies on digitalisation show that digital technologies generally have the potential to dissolve the borders between in-school and out-ofschool learning as the 'when' and 'where' of instruction changes (Erstad and Sefton-Green, 2013;Kumpulainen and Rajala, 2017;Sahlström et al., 2019). Digital tools (platforms, web resources and so on) loosen the frames between formal and informal learning, and hence make it possible to use also tools more associated with leisure (blogs, social media, chat services and so on) for educational purposes. Further, studies of digital educational settings show that students' identities may both coincide and clash when the private sphere and the public sphere overlap (e.g. Kumpulainen and Mikkola, 2014;Vigmo and Lantz-Andersson, 2014). Allen et al. (2020) note that while there is a wide range of digital tools, online teaching seems to lead to a return to more traditional ways of instruction, and Biesta (2019) notices that some of the most popular technology-mediated forms of education -such as TED talks, MOOCs and the numerous professional and amateur instructional videos on YouTube -are all staged in traditional ways, with someone talking and explaining so that others can watch, listen and learn. (50) Knowledge about the effects of the pandemic and ERT are still limited, and mostly based on reports and surveys from official stakeholders like the UNESCO, OECD or national agencies. According to the Swedish Schools Inspectorate (2020), principals of upper secondary schools in Sweden report that the transition to digitally mediated teaching in general worked well. The principals had, over a short period of time during school lockdown, managed to create a reasonably functional structure for their teachers' working situation, as well as for the students' learning. Already available digital tools in schools worked sufficiently well to make the transition from ordinary teaching to ERT. Also, the Swedish National Agency for Education published a report in August 2020, describing preliminary results on how schools have been affected in times of ERT. They conclude that some things have worked fairly well (for instance, the students' grades have not decreased), but the overall impression is still worrying (Swedish National Agency for Education, 2020). The teachers' workload increased, the compensatory assignment of schools had been more difficult to accomplish, and special needs education deteriorated. This pattern corresponds with the overall picture of how the COVID-19 pandemic has challenged education in many OECD countries (Schleicher, 2020).
From the academic field, Sahlberg (2020) discusses the long-term consequences of the pandemic and ERT for education and schooling. He thinks that the need to adapt to some level of online teaching during the COVID-19 pandemic will highlight many of the problems within school systems that we are already aware of, in terms of social inequities, unequal access to digital resources, and a dominance of traditional teaching, which makes school development difficult. He points out two main issues that stakeholders need to address: first, the need to work more on reducing inequalities in education; and second, the need to put more trust in teachers and principals in relation to decisions on how to improve teaching, including how to organise teaching during school closures. These issues, Sahlberg argues, were important before the crisis, and will also continue to be so after it. Hollweck and Doucet (2020), on the other hand, take a more optimistic stance when they discuss the pandemic from a Canadian viewpoint as an opportunity to rethink the future of schooling. They do not wish to go back to 'normal' teaching, since what used to be normal teaching never served all students well and equitably. Williamson et al. (2020) reflect upon how remote forms of teaching during the pandemic have constituted 'frontline emergency service' (107) and point out that it is to be understood as a re-location of digital pedagogies. There are several issues of importance, one of them being how the pandemic has revealed a significant variety in digital access and digital competence across schools and educational systems. This fact, they argue, has to be examined more broadly, also in relation to social inequalities (see also Morgan, 2020). Another aspect that Williamson et al. (2020) bring to the fore concerns how remote teaching during the school lockdown radically changes spatial and temporal relations of teaching and learning. In line with this, Stenliden et al. (2020) highlight, in a preprint of a study based on action research during ERT, how digital solutions intervene in the material framing of teaching and challenge the bodily tools with which interaction and communication are mediated in the ordinary classroom. They argue that if education as a field is going to gain something from this pandemic event, there is a need to understand remote teaching beyond results in terms of what is possible to measure and instead to a higher degree embrace teachers' qualitative experiences and reflective thinking. In line with these researchers, we think that more empirical studies are needed that can deepen our understanding of the consequences of the pandemic, and the present work aims to contribute with such an empirical study from a teacher perspective.

Understanding teachers' narratives about ERT
In this interview study, the rapid transition to online teaching is the reason for the conversations between researchers and teachers, where digitalisation policies form a background and a necessary condition for the transition taking place. This specific period of ERT is the general context of the course of events that teachers and researchers work together to interpret in the interviews. We take a narrative approach (Bamberg, 1997;De Fina et al., 2006;Mishler, 1999) to understanding teachers' experiences of this specific and unique period of time, based on their accounts in a series of interviews. From such a perspective, narratives are understood as practical accomplishments and as acts of performances (Mishler, 1999), that represent a particular occasion in the form of a story. In Mishler's terms, narratives are to be seen as purposeful and reflexive praxis and as 'socially situated actions; identity performances; fusions of form and content' (18). Here, this means that the teachers' re-tellings of their experiences in response to our (the researchers') questions are understood as narratives that take into account a multitude of contextual factors such as how they understand the purpose of the study and interpersonal aspects in the meeting itself. Their narratives could also be constructed in relation to what participants expect to be shared discursive knowledge, or discursive frames, in relation to teaching in this specific situation. This means that in our understanding of the interviews, we must attend to the teachers' stories as more than texts representing a certain content in their experiences, but also take into account dimensions of form and how the teachers' accounts are produced in the specific situation using structures and discourses in the surrounding context. Hence, Mishler's (1999) view on narrative as praxis counters a view on discourses as grand master narratives and structures that speak through a person. Instead, we view discourses as shared understandings, or shared narratives, that people use as social and cultural resources to interpret, make meaning of and perform their lived experiences. This focus will not only help us understand the ERT period as such, but it will also, as Sahlberg (2020) highlights, make visible important knowledge about needs in upper secondary education also beyond the crisis.

The interview study
In line with our narrative approach, the interviews follow the characteristics of qualitative research interviews (e.g. Holstein and Gubrium, 2012;Kvale and Brinkman, 2014), and we have followed teachers of history, mathematics and Swedish at the upper secondary level (five from each subject) in three interviews between April 2020 and September 2020. The selection is based on an availability sampling procedure where we used our networks of schools (for instance, schools connected to teacher education) and asked teachers in these schools to contribute to the study. The teachers worked at 11 different schools (250-2000 students). The schools were situated both in small towns (< 30,000 inhabitants) and in large cities (> 500,000 inhabitants), and had different municipality or independent school organisers.
An interview guide was constructed containing background questions (position and subjects; formal training; years as teacher), and themes guided by six main topics: (a) how the teaching is organised during ERT; (b) possibilities and challenges of using digital tools such as learning management systems during ERT; (c) subject-specific issues regarding teaching during ERT; (d) social aspects of teaching and supporting students during ERT; (e) challenges and possibilities regarding online classroom interaction during ERT; and (f) issues of assessment. Each theme contained a set of sub-questions allowing for the theme to be explored further. The interview guide allowed for a semi-structured design giving a general overview of the interviewees' experiences during ERT, as well as allowing each interviewee to bring up his or her own thoughts. The interview guide was adapted before the second and third interview respectively (questions were reformulated to fit different phases of ERT), but the themes were kept. Each one of the authors was responsible for a group of participants, and conducted all interviews with them. The interviews varied in length between 25 and 50 minutes and the third interview was on average shorter than the first 2. The plan was to interview each teacher twice during spring 2020, at the beginning of the ERT period and after about six weeks of ERT. A follow-up interview was conducted in early September. However, only 12 teachers followed through the whole series, resulting in a total of 41 interviews (see Table 1).
The interviews were conducted via a video-conference tool (Zoom), a choice mainly due to the pandemic restrictions that made it difficult to meet face to face. There is still limited research about video conferences as a method, but Archibald et al. (2019) find that Zoom is a viable tool for qualitative interviews that permits the participants to communicate with each other, as well as show and share documents or other teaching resources. This opportunity was used by some of the interviewees but not all. Using Zoom also made it possible for us to have a broader geographical sample, which would otherwise have been difficult to accomplish due to the pandemic restrictions. A critical aspect of using systems like Zoom, pointed out by Archibald et al., has to do with the safe storage of personal data and the risk of sensitive data becoming accessible to unauthorised persons in cloud-based solutions. However, such risks were minimised in this study. For our analysis we only needed audio recordings, hence we used separate unconnected audio recorders that were placed beside the computer.
Analytical procedure Holstein and Gubrium (2012) state that that the term 'narrative analysis' reflects a wide array of forms and sites (see also Chase, 2011) from a focus on brief utterances and short topical stories to long passages of time as well as from an interest in personal stories and identity to an interest in the ways that stories relate to various conditions of social life and locations. It is an approach that focuses on how storytelling operates in and relates to its social environment, where the point is to ask what is socially accomplished through an account, rather than to seek an objective truth in a narrative. From this perspective, narratives must also be understood as joint products of narrator and listener (Holstein and Gubrium, 2012). In our case, we understand the accounts in the interviews as joint products between us as researchers and the teachers making use of available social and cultural resources in the context (Mishler, 1999) to construe meaning about teaching during ERT. Both in the questions that were posed to the teachers and in the way that they answered, the particular circumstances of the pandemic crisis and the resulting ERT constitute a more or less explicit point of departure.
Applying narrative analysis helped us distinguish how the teachers made sense of their experiences, as well as how this sense-making changed and developed over time, in co-constructed dialogue between the interviewee and the interviewer. As a first step we conducted a thematic content analysis (Braun and Clarke, 2006). The thematic analysis is an iterative process, where we first performed close readings of verbatim transcriptions of the interviews, looking for recurring topics, resulting in 13 initial codes using the software NVivo. The codes were then collated into possible themes in relation to our analytical question 'What pedagogical considerations are present?' All excerpts relevant for each theme were again scrutinised guided by the analytical question, finally resulting in three main themes that each reflected recurring pedagogical problems and considerations in the teachers' accounts of their experiences of ERT. As described above, our narrative approach means that aspects of form and content are intertwined, so the ways of telling something cannot be distinctly separated from what it is about (Mishler, 1999). Holstein and Gubrium (2012) describe this approach in terms of 'the way storytelling operates in and relates to its social environment' (7), meaning that the production of narratives is not only conditioned by, but also shapes their circumstances. In the second analytical step, we therefore investigated the three contentrelated themes with a focus on how the whats in the content of the teachers' tellings of their experiences were shaped in relation to the hows in relation to the researchers' questions as well as the institutional framing. This analytic focus made visible the discursive frames that the teachers used as social and cultural resources to interpret, make meaning of and perform their lived experiences from ERT in the interview situations.

Analysis of pedagogical considerations during ERT
Here in the results section we begin by giving an overview of the interviews that serves as a backdrop for analysis. The analysis is then organised in relation to the three recurring themes that were identified in the first step of the analysis. As described above, we conducted the interviews with the video-conference system Zoom on three separate occasions, forming a trajectory from the beginning of the ERT period during mid-spring to the beginning of the autumn. Both as a reflection of the process and the different questions we posed, the interviews turned out to address different questions as both teachers and researchers successively changed their understanding of the ongoing development.
At the time of the first interviews, in mid-April, use of digital technology was often in focus in the interviews and there was an orientation to merely coping with the situation, finding 'goodenough' ways to move teaching from the physical classroom to digitally mediated spaces. As most schools had already worked with one-to-one solutions with personal computers for the students in combination with some kind of digital platform as an infrastructure for teaching, the teachers considered themselves quite well prepared in terms of technology. Most of them saw themselves as digitally competent and talked about their schools as well equipped with digital infra-structures, programs and devices also before the crisis, except for video-conference systems that for all schools had to be added in order to make online teaching possible.
In the second round of interviews, in May, the teachers had found some kind of everyday normality in the crisis, and they talked about different solutions that they had worked into new routines for teaching. The picture that emerged shows some variation between schools and teachers, in terms of routines for communication, instructions and assignments. It was clear that the teachers used not one but many different platforms and programs in their teaching. At this stage it also became more and more obvious that there were difficulties in keeping some of the students on track, and many teachers described how they tried to get in touch with students through text messages or phone calls outside the lessons. They talked about how digital efforts in combination with time-consuming student contacts resulted in a heavy workload, as they tried to find strategies for how they could be available for their students and at the same time cope with all the different kinds of information that each platform, program or app brought them.
At the time for the third round of interviews, in September, the pandemic situation was calmer and schools opened again. However, many schools still organised different kinds of hybrid forms between ordinary teaching and online teaching to lower the risk. Different groups of students were attending school on different week-days, so sometimes they had to work from home one day a week and sometimes more. Also, some schools tried to provide teaching for students staying at home with mild symptoms, which meant that in some cases teachers were expected to interact with students in the classroom and students participating online simultaneously.
The general impression from the interviews is that the technical shift itself was not that challenging. However, most teachers clearly stated that even if they coped all right with the technology, they did not consider the online teaching to be equally good as teaching conducted in the classroom, and in the third round of interviews they also looked back at the spring as a very stressful period in terms of technology. In the following analysis, organised in three overarching themes, we investigate what pedagogical considerations the teachers talk about as well as how their narratives about this are construed with a focus on the discursive frames that the teachers draw upon to perform their experiences in dialogue with the researchers. The three themes describe considerations in relation to teaching content, classroom interaction and student support.

Theme 1: considerations in relation to teaching content
Our first theme concerns how the teachers dealt with issues of teaching content in relation to ERT. Generally speaking, a coherent picture emerged showing that they did not seem to have made any radical changes concerning content because of the situation. Instead, the teachers put effort into sticking to the plan as much as possible for the end of the spring semester. The challenges they talked about were to a large extent related to getting enough support to grade the students in relation to the curriculum. This is the case, for example, in an interview with one of the teachers of history: In this example, from an interview conducted in early May some weeks into the ERT period, the interviewer specifically asks the teacher about content focus and refers to some examples of core content in the subject of history. In his answer, the teacher states that there were no major changes concerning content and explains this lack of change in relation to the fact that the transition of the teaching to ERT occurred so late in the school year when most of the content areas in the syllabus had already been dealt with. However, he opens up the possibility that it might have been otherwise if the transition had been planned from the beginning of the school year. This opening implicitly indicates an understanding of the transition as a kind of emergency solution that was unexpected and implemented with short notice. In this example, the teacher continues to refer to the content area 'use of history' as a central aspect of the subject that was already sufficiently covered so that there was no need for further student assignments about that. He also refers to his long-term plan for the year, where most of the content -three-quarters -had already been covered. In this narrative, the teacher draws on two discursive resources to make his point that there was no immediate need to make any major changes to the teaching content. First, he refers to the state of emergency as a context and a reason for sticking to the plan without making too many changes. This is recognisable also in many of the other interviews where teaching during this period is talked about in terms of 'being good enough' and characterised by a need to 'cut corners', or that development of new content has to 'wait for another time' as they currently have to prioritise to keep things going. Second, he uses the curriculum to account for the fact that his students had already completed major assignments so that most of the core content was already covered. The significance of the time of the school year is something that several teachers orient to as a reason for their considerations. In the next example, a teacher of Swedish talks about it as a good thing that the school closure came so late during the spring semester and, similarly to the history teacher above, that she had already completed most of the necessary knowledge requirements (that is the Swedish curriculum term for standards set for reaching a passing grade in a school subject). In excerpt 2a, this teacher also refers to the time of ERT, late in the school year, as being important in relation to her pedagogical considerations. She relies on a routine of hers in which she always tries to complete all the necessary requirements from the syllabus before Easter so that she has some months left to support or, as she puts it, 'save' students who are at risk of not passing, and also to give all students the opportunity to improve their grades. Another circumstance that is important in this narrative concerns the national tests in Swedish that are usually given at this time of the year. During the ERT period, the Swedish National Agency for Education decided to withdraw these high-stakes tests that are usually used as standardisation tools for equal and comparable grading across schools nationally. The reason given for withdrawing the tests was that it was considered difficult to guarantee that they could be carried out without risks of cheating. The withdrawal of the national tests is something that many teachers of the subjects Swedish and mathematics talk about as something that affected their teaching, and sometimes also their considerations in relation to content. In this example, the teacher had planned, even before ERT was implemented, to use the weeks after Easter to prepare the students for the national tests, but as they were withdrawn she concentrated on constructing replacement tests in order to make it possible for students who had missed certain subject content to catch up and receive passing grades.
Similarly to the teacher in history, this teacher also uses her routines and long-term planning as contextual resources to explain her experiences in relation to the teaching content. But she also refers to a discourse of assessment that frames her considerations, where the draw-back of the national tests makes it necessary to find other ways to get support for grading in relation to the knowledge requirements in the curriculum. In relation to this she also refers to a compensatory discourse and the teacher's responsibility to help students catch up and give extra support to students at risk of not passing. This interview is also an unusual example from another aspect, as it is one of few that orients to the ongoing situation with the pandemic in the teaching, something that otherwise is very scarce in the interview material as a whole. A bit later in the same interview, as the teacher talks about how she needs to prioritise literature as the students are often reluctant to read novels, she touches on the topic of the crisis: In response to the interviewer's question about whether she will teach literature for the remainder of the semester, the teacher defines this content as important in the replacement tests that she has given in grades 1 and 3, and that was actually something that she has used previously as a preparation for national tests but now sees as a replacement. In this account, the teacher treats the national tests and the course requirements as her first priority in planning her teaching, again using an assessment discourse to account for the fact that the students are given an opportunity to perform well enough to pass their grades. She puts it as a coincidence, 'as it happened', and in her ordinary teaching this involved working with dystopias as a literary genre in third grade. She comments on this as very timely, implicitly referring to the pandemic situation as a dystopian topic. The interviewer then asks explicitly if working with dystopias was a deliberate choice in the given situation or if she would have done it anyway. In her answer, the teacher shows a slight hesitance in her understanding of the question and then turns not to dystopias but to her priority to teach the thirdgrade students in Swedish about academic writing in the form of a brief report. She talks quite extensively about this subject in relation to course requirements, but a bit later the interviewer again brings into the discussion the current situation as maybe being dystopian and now the teacher also develops this topic: In response to the interviewer's elicitation, in excerpt 2c the teacher develops the topic about how dystopian literature is easy to connect to our time where pandemics is one of many issues that are relevant in relation to contemporary global issues. As she tells about this, she treats dystopia as a thematic content that she has worked with before in this way, and that she has found to be important for the students as a way to 'process' questions and reflections about difficult issues which might cause anxiety. As the interview was conducted in late March, quite early in the ERT period and also when the pandemic situation in Sweden still had not peaked, the teacher at this point could not have a full overview of how serious the situation would become and she does not here single out issues related to the pandemic as more relevant than, for example, climate change or food shortage. Still, this example stands out since it is the only interview where a teacher connects subject content to the ongoing pandemic crisis that is the societal reason for implementing ERT. The general picture of pedagogical considerations in relation to content is that there were no big changes due to ERT, as teachers primarily tried to stick to the core content that was already planned and that they used to work with at this time of the school year. All through the interviews, the teachers recurringly use discursive frames related to curricular goals, assessment and grading as resources for explaining their choices and considerations in relation to planning and teaching content. In addition to this assessment discourse, the teachers also draw upon understandings of a compensatory discourse, pointing at a sense of responsibility for giving the most support to students at risk. We also note that a topic that is hardly mentioned at all in the 41 interviews, either by the researchers or by the teachers (the teacher in excerpt 2 is a rare exception), concerns reflections about the ongoing pandemic as teaching content. In a stressful situation, such as the rapid shift to ERT, it is understandable that the teachers did not have much time to reflect upon new possible thematic areas to teach about. But in hindsight, if we take a step back to reflect, it is quite obvious that there are many issues and aspects relating to the pandemic that could have been important to work with in order to support students' understanding of the situation. Hence, connecting subjectspecific knowledge to the ongoing pandemic could be considered an absent discourse in the interviews.

Theme 2: considerations in relation to classroom interaction
The second theme in the analysis concerns how the teachers' considerations in relation to aspects of classroom interaction became important in the transition to ERT. As mentioned above, all schools already used some kind of digital platform and the students all had their own computer, but in order to meet the increasing demands on digitally mediated interaction many teachers added several new programs such as, for example, Google Meet or Zoom as a substitute for real-time classroom interaction in lectures, and also to answer questions from students. One teacher mentions 'written answers to questions in Hangout and also making phone calls to talk to the students individually' as communication resources, and another teacher talks about how Office 365 has added 'a program called Haldor' for administration of school assignments. Yet another teacher introduced the program Discord that he found easier than the official choice of his school and Google Meet for student group discussions, while some teachers of mathematics use programs for visual support such as Whiteboard or GeoGebra. In the following example, a teacher of mathematics describes the challenges he experienced when finding new forms for interaction with the students, mediated through the video-conference system Zoom: This teacher of mathematics tells a story of how he successively changed his understanding of an important aspect that many of the teachers talked about that has to do with the students' possibilities to ask questions during individual work. This teacher describes how he initially did not understand the problem, as he thought of himself as being available during the whole lesson in the joint Zoom room, which could be understood as comparable to how he would be available in the physical classroom. But like many of the other teachers, he also talked to some of the students on the phone and understood that from their point of view, asking for help was difficult. The teacher described how he more and more came to understand that the problem had less to do with being technically available for the students than with the students' unwillingness to display their learning problems to the other students. At first it was only the most accomplished students who asked about advanced problems who were willing to show their difficulties in front of the entire group. As the teacher himself incorporated more digital resources, in this case based on advice from the researcher in the first interview, he created individual breakout rooms for each student so that they could ask for help without having to show their shortcomings to the others. In another example, a teacher of history talks about the problem during ERT that many students do not dare to put on their microphones to ask something. However, this teacher shows a partly different understanding as she finds it problematic in terms of quality that the digitalised teaching becomes too individualised: In this excerpt, the teacher relies on a description of the physical classroom as the ideal where it is possible to pick up things from students and grab opportunities in the moment to explain and sort out knowledge problems in a way that is beneficial for all students in the room. In line with several of the teachers in the data, she describes ERT in terms of a lack of dynamics compared to the physical classroom and that teaching becomes more a question of individual learning projects for each student. In the digital teaching, students are assigned to pose their questions mainly in a chat forum, which this teacher finds rather shallow, and she adds that teaching tends to be more teacher-centred. Drawing on discourses of participation and dialogue as the ideal, a recurring narrative describes teaching during ERT as 'flat', 'less dynamic' and 'more monologic'. Many of the interviewed teachers lack the small talk and non-verbal responses in the classroom that they find important for understanding their students' needs for help and clarifications. However, one teacher in history also points out that in some aspects, teaching before ERT tended to be rather individualised as well, and she describes 'that is how it was before, and it did not change very much when Corona came actually. Cause it was what it was, it was mainly these group exercises that disappeared.' The fact that teaching during the school lockdown is mainly understood as a temporary emergency solution rather than as an opportunity for further digitalisation processes is something that, for example, one of the teachers of Swedish reflects upon in the third interview in September, when schools are open again: Looking back at the ERT period, this teacher of Swedish identifies a pedagogical concern in relation to the length of lectures, where the conditions of online teaching are different compared to lectures in the physical classroom. Looking back, he describes how he had to replace his ordinary lectures with shorter sessions that students could access easily. The teacher claims that this was a personal insight, whereas many of his colleagues maintained very long lectures with questionable technical quality as they only had web cameras and built-in microphones. Here, as well as in other interviews, adapting teaching to ERT is constructed mainly as an individual problem for each teacher. This teacher, like several of the other teachers, talks about how there were very few structured discussions between colleagues about how to maintain teaching quality in the digital classroom. What is also worth noticing is that this is not mentioned as a big problem in any of the interviews, since there seems to be an underlying discursive understanding that teachers are individually responsible for planning their teaching. Besides a few examples of in-service training, collegial collaboration in the implementation of ERT seems to have been rare. From our second theme in the analysis, the general picture from the teachers' narratives is that the physical classroom is seen as superior from an interactional and relational point of view. In their narratives about teaching, the teachers in general refer to themselves as digitally competent and willing to try new technologies for different purposes. Hence, the challenges in accomplishing classroom interaction are not primarily related to technological issues but have more to do with the limitations of digitally mediated interaction compared to how interaction in the physical classroom works. The teachers draw on discursive ideals about promoting student participation and dialogue, where the students can learn from each other. This could be understood in terms of a relational discourse, where some teachers point out the importance not only of verbal interaction, but also of being able to meet the gaze of students, to notice if someone seems to be stuck and to support students by giving feedback and quick responses. The variety and plurality of the different programs, apps and platforms that the teachers refer to when describing their teaching practices during ERT are examples of efforts made to keep and maintain an interactional practice similar to that of a physical classroom.

Theme 3: considerations in relation to student support
A third theme in our analysis is focused on what the teachers tell us about student support during the ERT period. Already in the first interviews this was a concern that many teachers were aware of, and their awareness became more manifest as time went on. In excerpt 6, the interviewer asks about different students in terms of winners or losers when they have to work from home, and also refers to differences between homes. In her answer, the teacher refers to a specific group of students, 'these students', who have difficulties in understanding instructions. The teacher says that despite her efforts to be clear, give visual support and repeat the instructions, some students still have difficulties that have now become more obvious as the students cannot get help from looking at each other, as they can in the physical classroom. Owing to the platform and the ERT situation, the teacher can identify these difficulties in new ways, as, for example, when they are logged into their digital textbooks the teacher can give more active support through the chat. She draws a conclusion that more students than she previously thought have problems understanding oral as well as written instructions, but now she can detect it and contact the students. In this narrative, the teacher draws on a compensatory discourse about the teacher as responsible for identifying and supporting students' special needs. Here, as in many of the interviews, the teacher also (re)produces a narrative about the physical classroom as a place where not only the teacher, but also students can be models that scaffold each other in their school work. The importance of the physical classroom in relation to student support is also shown in the following excerpt from an interview with a teacher of Swedish: A bit differently compared to the previous excerpt, this teacher finds it more difficult to identify students' needs for support during ERT compared to the physical classroom where she could have given more immediate feedback and handled learning problems that occur. She describes a delay in the response, as she cannot see the learning problems until after a lesson and has to take notes on what to follow up with whom after a lesson. In addition to talking about the physical classroom as superior from an interaction point of view, she clearly performs and uses a discourse about working actively to identify and follow up on students' needs, even if they are silent and do not ask for help: Looking back at the spring semester from a slight distance, this teacher talks about how he found out over time that many students did not ask for help even if they needed support, and therefore the teacher actively tried to contact them to ask how they were coping. The interviewer comments that this is something that many teachers seem to have similar experiences with, and asks the teacher if he has any ideas that could explain this. In his explanation the teacher draws on how the digital space is much more interactionally challenging for students to expose themselves in, compared to the physical classroom. The teacher understands this as a question of not only displaying oneself as a student, but also allowing the teacher to get access to the student's home environment. This teacher refers to a pattern of many students being uncomfortable with having the teacher so near them on the screen -almost as if the teacher would tell them to pull themselves together and tidy up their room. Here the teacher uses discourses about asymmetric power relations and boundaries between the institutional role of teachers and the students' private space at home that can be difficult to deal with for both teachers and students. This narrative makes visible how online teaching not only gives the students access to teaching, but also gives the school access to the students' private sphere at home. In parallel with the examples in the second theme, we see here how the teachers emphasise the importance of the classroom as a space where students can learn not only from the teaching, but also from each other. In all examples in this theme, the teachers give examples of how they work actively, in different ways, to find alternative ways to support students in their work. They especially stress how they have to find new ways to make contact with silent students, or students who in different ways show signs of not understanding. These narratives about their efforts draw upon an ideal to even out students' different resources in school work that could be understood in relation to a compensatory discourse strongly rooted in Nordic education.

What major discursive frames are used in the narratives?
The narrative approach to the analysis of teachers' talk of their experiences highlights the discursive frames that the teachers rely on and use when they tell about their pedagogical considerations. These discursive frames consist of formal as well as informal norms, and sometimes show what discourses are considered superior to others. We find that there are three major discursive frames that emerge in the teachers' stories about their teaching practices during ERT: (a) an assessment discourse with a focus on curriculum and grading; (b) a relational discourse departing from the functions of the physical classroom; and (c) a compensatory discourse when it comes to student support. The first discourse, with its focus on assessments, curriculum and grading, is mainly used in narratives about teaching content during ERT and appears to be a discourse that is superior to other grounds for pedagogical considerations about content. The teachers prioritise central content in the syllabi and do not talk about any substantial changes of content because of the shift to ERT or the pandemic. They rely on their existing plans for the school year even if they have to find new digital tools to maintain them. The changes they do refer to have more to do with the withdrawal of the national tests, which some of them rely on for their grading and therefore have to find other ways of obtaining grading material through various assignments and tests. The shift to online teaching challenges ordinary test practices as it makes it difficult to control for cheating, which makes considerations related to testing different from the ordinary.
Second, we also find a strong relational discourse in the teachers' talk about teaching during ERT. The different digital platforms, programs and teaching materials that are brought in and combined can hardly replace the physical possibilities of interaction in the physical classroom. We show how the teachers find it difficult to provide feedback, capture learning problems or visualise and represent a knowledge content in the digital form. We find that they refer to the physical classroom as a complex set of verbal, bodily, material and spatial aspects central to students' learning process and knowledge development (see also Stenliden et al., 2020).
Finally, our results point to a third clearly prominent discourse that is characterised by a compensatory understanding of the teaching assignment. This is shown when teachers talk about how they actively try to find new ways of supporting all students in their learning process in line with the traditions of the Nordic school model and the ideal of 'a school for all ' (Blossing et al., 2014). Here, the idea is to compensate for students' different conditions and needs by providing more support to those who have the greatest need and a readiness to find different paths for different students.

Keeping up the normal: digital resources as substitutes for classroom interaction
This paper is an empirically grounded account of how Swedish upper secondary teachers, from a professional position, dealt with and made meaning of an exceptional situation -the shift to ERT due to the 2020 coronavirus pandemic. In a European perspective, Swedish upper secondary education was relatively well equipped with technology, and in the interviews the teachers talk about how they have access to and use a wide range of different platforms, programs, chat forums and other digital resources in order to keep up the teaching they would have done in the ordinary classroom, had it not been for the pandemic. A short answer to our question in the title would therefore be: no -digital teaching during ERT is not the new normal but rather an effort to keep up the normal, at least what is considered to be the ideal, normal, classroom-based teaching.
In line with Sahlberg (2020), we agree that the challenges and problems that become visible during ERT are not new but represent the state of the current school system with its strengths and weaknesses. The framing discourses that we have shown in our analysis -assessment, relational and compensatory discourses -could all be understood as classical pedagogical challenges that teachers also struggled with before the crisis. The teachers' pedagogical considerations both reflect and form what could be described as a professional discourse where many of the values associated with the so-called Nordic model (Blossing et al., 2014;Klette, 2018;Telhaug et al., 2006) can be recognised. Especially in their use of a relational and compensatory discourse, the teachers emphasise the need to interact with and understand the students in this situation, which seems to be very demanding and time-consuming to do digitally compared to the face-to-face meetings in the physical classrooms. This resonates well with the idea of one-school-for-all where schools and classrooms are not only places for teaching, but also for meetings between students with diverse backgrounds and needs.
As for the strong assessment discourse that we find, we believe that it partly reflects that the shift online occurred at the end of the school year when teachers are normally very occupied with finishing their courses and grades for the students. This assessment discourse can also be understood in the light of the changes that the governance of the Swedish school has undergone in recent decades -a shift from a competence-based curriculum to performance-based goals (Sivesind and Wahlström, 2016). Not only in Sweden, but in all of Europe and the western world, grades and assessment tend to become increasingly important, and, at a policy level, large-scale standardised tests have become important tools for creating accountability and comparison -nationally and internationally (Lingard et al., 2013). Grades also function as a basis for selection in an increasingly marketised school (Lundahl, 2016), and in the teacher interviews it is more from the perspective of the students' need to get passing grades that assessment and grading becomes a challenge. Grading has always been a task for teachers, but what is obvious here is the strong emphasis in almost all interviews on getting enough support for grading in relation to the curriculum and course requirements. Teachers cannot simply use their knowledge about students from previous work, but need 'proof' in terms of test results, oral examinations or other student performances to account for the grades they decide. The ERT situation, when, for example, national tests were drawn back, challenges this discourse and makes it necessary to find new ways of getting the support needed that also includes considerations on how to avoid plagiarism and cheating. In their narratives the teachers make visible how they balance these demands of support for assessment with relational and compensatory concerns for different students, and the complexity this balance means in an all-digital environment.
As this is written, in November 2020, the pandemic seems far from over and a transition to distance education may again be called for. The crisis we have described is thus still a fact. Like several of the reports and evaluations made of the Swedish ERT period in the spring of 2020 (Swedish National Agency for Education, 2020; Swedish Schools Inspectorate, 2020), we can state that the digitisation processes that have been initiated and implemented for a long time were put to the test, and, from a technological point of view, seem to work relatively well. However, as Williamson et al. (2020) argue, there is a need for critical reflections not only on the prospects for digitalisation, but on the long-term consequences for education in general.
Our result is particularly important, we believe, in relation to the digitalisation efforts in the educational sector in Sweden as well as internationally, showing a strong belief in digitalisation as a democratic project promoting equity. Listening to the teachers' experiences, we find it important to emphasise the need to pay more attention to the relational and interactional dimensions of the digitalisation of teaching. We also want to highlight how online teaching changes boundaries between public and private, something we believe should be paid more attention to at a policy level. During the ERT period, new conditions emerge as students participate in teaching from home, via webcams on their personal computers that the school has provided. This shows how teachers are let into the students' home environments through the same video chat system, which in a fundamental way shifts boundaries between the students' private domain and the school's public sphere. We believe that this boundary shift is an important dimension to consider in relation to digitalisation strategies.
A final comment, that we think noteworthy to bring to the table, concerns the very few accounts of changes in relation to teaching content in general and specifically in relation to knowledge about the ongoing pandemic from a subject-specific perspective. Our point here is not to question the teachers' content priorities. In light of the working situation under strong pressure in the initial phase of ERT, it is understandable that they stuck to the plans they had already made and tasks previously used. But we think that it calls for further reflection and exploration that so few expectations are brought up about the global pandemic situation as such as a matter to bring into teaching in the educational debate at large. To what extent should education and teaching be able to be more flexible and make adjustments in relation to contemporary developments in society? Are we stuck in a system where course requirements, student performance and grading criteria frame teaching so strongly that it becomes too difficult for education to address important and contemporary societal challenges, based on academic knowledge? We have no ready answer to these complex questions, but suggest that it is a topic that calls for more research, debate and discussions on many levels.