Narratives of Teachers and Teacher Unions in Swedish Facebook Rebellion Groups

This study examines narratives about the teaching profession and teacher unions that Swedish teachers jointly produce in two t eachers’ rebellion groups on Facebook, which is followed by a total of around 20,000 teachers. A sample of 33 posts and 2,445 comments were analysed using a narrative approach. The findings highlight narratives in which teachers wish to return to “the good old days”, struggle with everyday frustrations, call for a strike as an immediate solution, and describe hypothetical futures presenting the opportunity for proactive action and call teacher unions to dialogue rather than wait for them to satisfy the teac hers’ demands.


Introduction
The focus of this study is on the teaching profession and on how teachers themselves relate to Swedish teacher unions (TUs) as part of shaping the teaching profession. To approach the teaching profession, this study investigates narratives about the teaching profession and Swedish TUs jointly produced by Swedish teachers in two teachers' rebellion groups on Facebook. Teacher unions were one of several topics discussed in these groups in the spring of 2021. We became curious about how teachers presented TUs and how they related to the actions-or lack thereof-attributed to TUs.
Since the late nineteenth century, Swedish teachers have been members of various TUs. Until recently, approximately 120,000 Swedish primary and secondary school teachers, accounting for approximately 85% of all teachers, were members of two large TUs (Mörck, 2021). On 1 January 2023, these TUs merged into a single union. Unlike their English and American counterparts, which traditionally take a more activist position, Swedish TUs, with less activist strategies, have moved in the opposite direction, establishing tight relations with policymakers. This has made Swedish TUs "a part of the same establishment they are to negotiate with" (Lilja, 2014, p. 118). The development of TUs is also part of a more general transformation of the Swedish welfare state from the late 20th century characterised by initiatives for decentralisation, municipalisation and marketisation of the national school system. The TUs efforts to mobilise and professionalise the teachers thus became dependent on the good will of the state and the economic situation in the country (Ringarp, 2011). As elaborated later in this article, one might say that Swedish TUs are influential in gathering members but not so powerful in presenting the teachers' case.
Ongoing research shows that teachers are burdened by a heavy workload, which has been increased and changed by assignments such as documenting, reporting, and even advertising (Lundström, 2015;Samuelsson et al., 2018). Teachers' working conditions, working environments, and autonomy are considered important, and in this era of rapid policymaking, recent education reforms in Sweden have "reconstituted teachers' work and professionalism" (Hardy et al., 2019, p. 362) and have made teachers feel constrained and controlled by a range of demands.
The empirical data for this study came from discussions in two Swedish rebellion groups on Facebook. Their members are mainly Swedish teachers who discuss issues related to teachers' work and working conditions in primary and secondary schools. We view the phenomenon of rebellion groups in the broader context of teachers' changing conditions and their strife to improve them through such new collective formations. The rebellion aspect refers to activism for issues concerning teachers' work and working conditions, mainly through joint online discussions, narrating, and seminars. In this study, adopting a narrative approach (Georgakopoulou, 2013;Linde, 2009), we specifically explored narratives about TUs that teachers jointly produce in Facebook rebellion groups. What could these narratives tell us about Swedish teachers and their working conditions? Further, based on previous knowledge (see Löfdahl Hultman et al., 2022) of rebellion groups' dissatisfaction with TUs, we searched for "bets" (Boje, 2011)-suggestions about possible new or different narratives about TUs and their relation to the teaching profession. Such narratives can help distinguish stories about teachers' past and present working conditions and their experiences of TUs' role and support, as well as stories containing bets on hypothetical futures that can influence TUs and their future role. The guiding research questions were as follows: What past and present experiences of TUs' role and support emerge from discussions in teachers' Facebook rebellion groups?
What is the significance of these narratives for the Swedish teaching profession?

TUs and the teaching profession
TUs seem to have an ambivalent position. They play an important role in the professionalisation of the teaching profession, but teachers' trust in them is questioned due to their position as policy actors. Issues highlighted in research focusing on teachers' reasons for becoming union members include changes in working conditions, higher salary levels, teachers' ideological views, and-not least-unions' moral legitimacy in terms of what kind of professional support they provide (Fidan & Erkocak, 2015;Popiel, 2013). Several researchers have pointed to a need to shift unions' moral centre of gravity to better understand and address the issues that are most important to their members (Bascia, 2008;Popiel, 2013). Bascia (2008) found few TUs in the United States that could "honestly claim broad, sustained, and positive relationships with their members" (p. 95) and argued that teachers' perceptions of their unions were obscured by a lack of organisational visibility, issues that were too narrow, and a sense that unions served the needs of only a select group of teachers.
In Nordic countries, TUs seem to take on increasingly proactive roles in policy development and professional quality. For example, the idea of a partnership between employers and employees is a strategy aimed at strengthening and renewing TUs. However, studying policy documents from The Union of Education Norway, Bie-Drivdal (2018) found the dual role of local representatives as both watchdogs and agents of change to be problematic. While documents called for proactive voices and local representatives were expected to cooperate with employers, problem-solving and quality assurance work seemed to be more important.
In Sweden, TUs have played important roles as collective agents striving for better conditions, such as higher salaries, control over working hours, and participation in decision-making (Ringarp, 2011). However, unions have been closely dependent on political initiatives concerning aspects such as the principalship of schools, the organisation of working hours, shifting national curriculums, and reforms to strengthen the impact of detailed governance by the state. In short, the professionalism developed through the relationship between Swedish teachers, their unions, and the state can be described as "professionalism from above" (Evetts, 2003), 4 closely linked to and dependent on mechanisms of governance and control. Swedish TUs and teachers' status as professionals are also a matter of politics. The failure or success of initiatives for professionalisation is not just a matter of mobilisation within TUs, but just as much a matter of the relation to the welfare state. Although Swedish TUs have historically striven to enhance teachers' professionalism (Ringarp, 2011), several studies have shown that their alliances with employers counteract what the teachers themselves consider important aspects of their professionalism. Lilja (2013) found that TUs failed to support the teachers, and instead of challenging politicians about teachers' working conditions, they chose to collaborate with them. Studying Swedish TUs' magazine editorials, Krantz and Fritzén (2022) found that Swedish TUs adapted and defended their positions with respect to the prevalent political, economic, and ideological conditions, leading to several contradictions and inconsistencies. Accordingly, the authors argued that unions should assume a more proactive role and that a "union's argumentation should be based on the professional conditions that its members work under" (Krantz & Fritzén, 2022, p. 268). In line with this argument, we aimed to explore the potential of discussions about TUs in Facebook rebellion groups to serve as initiatives for "professionalism from below" (Vaidyanathan, 2012). According to Vaidyanathan, such professionalism has a different meaning from that of professionalism from above in the sense that it weakens the teachers' sense of control or accountability and allows more intrinsic rights and values to be expressed by employees-in this case, teachers.
In Sweden, as in other Nordic countries, "collective agreements have a much more prominent position than legislation and other state regulations" (Kjellberg, 2022, p. 3). We believe that this has strongly influenced the Swedish teaching profession's confidence in unions' ability to contribute to improving working conditions. Simultaneously, it is reasonable to assume that teachers blame their unions when working conditions deteriorate and that they will find new arenas where the fight for improved working conditions can be carried out from within. The way in which teachers organise themselves in Facebook rebellion groups is similar to a "proactive bottom-up model of collective organization", as opposed to the top-down servicing model of union officials who rely on legislative procedures (de Turberville, 2004, p. 777). Thus, narratives about TUs shaped in Facebook groups may carry nuances and a potential bet on ways for teachers and TUs to shape a more independent form of professionalism.

Data and methodology
Over 15 weeks in the spring of 2021, we followed two Swedish Facebook groups: Teachers' Rebellion (TR) (Lärarupproret, n.d.) and School Rebellion (SR) (Skolupproret, n.d.). The groups mainly aim to discuss issues related to teachers' work and improve working conditions in schools. Even if not all teachers post, there is a large number of teachers who follow the groups and who are possibly affected. TR was established in 2018 and has more than 15,000 members, while SR was established in the spring of 2021, steadily amassing 4,000 members during the data collection period. We followed TR for seven weeks and SR for 10 weeks, with a two-week overlap. During the data collection period, TR and SR had, on average, 350 and 650 posts per month, respectively. TR is a closed group for members only, although its broad membership makes it a more or less public network. SR was formed by TR members as a public group in pursuit of more activism and a wider audience (Löfdahl Hultman et al., 2022). At any rate, the two groups largely overlap in terms of both members and aims, and their discussions-and thus the data collected from both-are similar.
We were inspired by "netnography" (Kozinets, 2019) and the clarifying steps through which mere information is transformed into data. Unlike traditional ethnography, netnography covers a vast amount of data that researchers do not collect themselves, which means that the procedural operations must be clearly stated. Our choice of focus reflected what we, as researchers, were interested in and what we were looking for in the fluidity of posts and comments in the two Facebook groups. We employed a mix of investigative and immersive procedural operations. Our study was investigative in that we selected our data from a vast body of informational traces created in the communication between members of the two Facebook groups. Thus, we did not produce our own data from our research questions. but selected data to be included in our research. Our study was also immersive in that it was based on a set of choices that we considered to be rich, deep, and extensively descriptive and informative. A further choice is that we have included poorly described posts in a way that might represent silence in our data. In our immersive operations, we strove to preserve the context of the data-for example, the decisions made by Swedish TUs during that period or other events in society that may have had an impact. We added the number of comments or reflections on the lack of comments on a post, and we took notes on the numbers and symbols of likes (angry, love, thumbs up, thumbs down, etc.). We recorded our reflections on the kinds of content and arguments that we found and our initial ideas of the narratives emerging from the data in what Kozinets terms a "netnographer's immersion journal" (2019 p. 194), which is similar to a traditional ethnographer's field notes. In a similar way, our reflections based on the immersion journal are not presented in the results but constitutes part of our knowledge of the field. This study was reviewed by the Ethics Committee of Karlstad University, Sweden (Dnr: 2020/917). No sensitive personal data were collected. Personal data were handled in accordance with the legal basis of general interest (Article 14, 5b GDPR). To conceal any identifying features, such as names, schools, and exact dates of posts, we chose not to specify which of the two groups the quotes are from. We repeatedly informed the members of the Facebook groups about our research and their right to opt not to participate via administrators and direct posts. One person chose not to participate. No information or comments related to that person were downloaded, saved or used in the analysis.
Our sample consisted of 33 posts with a total of 2,455 comments regarding TUs in some way. Each post had from 0 to 324 comments as well as likes (thumbs up) and various emojis. Likes commonly ranged from 50 to 100, although there were also posts with more than 200. The use of emojis was less frequent-for example, one post received 15 angry emojis and 15 crying emojis. Comments on posts, in turn, received new comments and likes. Some posts were closed to further comments and likes, as the administrators deemed the discussions unethical or hateful. The sample contained posts with discussions and/or mere comments that could allow us to interpret how members related to TUs and what they wanted TUs to achieve in terms of working conditions.
One of the researchers read the thousands of posts and comments saved for analysis. Almost every day over the data collection period, approximately two hours were spent scrutinising the posts and comments. Some posts may have been lost during the operation process. However, the researcher read almost all the posts and scrolled through the comments. After 15 weeks, we reached what is often described in traditional ethnography as ethnographic saturation. New posts appeared to concern the same overarching issues, although they were authored by different persons and were sometimes related to different contextual aspects. Comments appeared in threads several days after the posts were made. For this reason, we downloaded and saved data after some days, when no or few new comments appeared. Posts that we did not save may have concerned important teacher-related issues (e.g. mobile phones in classrooms, teacher education, and students' achievements) but were not related to TUs. We reflected on the extent to which our small sample drawn from a large number of posts could be described as "cherry-picking"-that is, selecting data that can yield certain desired results. Our curiosity about what narratives were retold or were new and what bets on a hypothetical future we might observe led us to rule out this possibility.
The netnographic approach focuses not only on the contextual aspects of social media but also on societal issues that may affect what is discussed. The data collection in this study took place some months before and after a new three-year agreement between employers and the two main Swedish TUs. During these weeks, a report was released (Mörck, 2021) detailing the possibility of a merger into a single union for all teachers. This period also coincided with a third wave of the Covid-19 pandemic in Sweden, during which the spread of infection in schools increased. Although the role of TUs in general was a main topic for discussion in the Facebook groups during this period, these societal issues were also frequently discussed.
By following the two Facebook groups, we gained an understanding of the complexity of teachers' relationships with TUs. There were calls for more or different support and new ways for TUs to represent the teachers. This contextual knowledge of teachers' relationships with TUs guided us in searching for narratives about the difficulties, challenges, and opportunities for teachers and the Swedish school.

Analytical frames and methods for analysing the data
Our analytical framework was based on research on professionalism and digital and social media, and a narrative approach was used for analysing the data. To interpret our results in terms of teacher professionalism, we relied on Evetts' (2010) notion that professionalism is 7 not static but in a state of constant change as the conditions of a profession change. In line with Evetts (2010) and Löfgren and Wieslander (2020), we contend that both occupational and organisational forms of professionalism, emphasising agendas of internal collegial control and external accountability respectively, are involved in shaping teachers' professional identities. Professionalism in relation to Swedish TUs is a matter of teachers' trust in unions' ability to identify conditions that need to be improved. The trade union movement at the national level can best be associated with an organisational form of professionalism, although renewal and democratisation efforts are also made at both the national and international levels. Also, models of collective bargaining occur at both the central and local levels (Kjellberg, 2019), and renewal strategies are made for developing an organising culture in teacher unionism, such as empowering teachers at the grassroots level (Gavin, 2022), thereby raising teachers' voices and exerting influence. Despite initiatives at the central TU level, a collegial form of professionalism that takes teachers' voices into account is lacking. Stevenson and Gilliland (2015) argued for a new kind of democratic professionalism that acknowledges teachers' skills and the complexity of teaching and gives TUs as democratic organisations a central role in representing teachers' voices.
We view the two Facebook groups as networked publics (Boyd, 2010) and the stories taking shape in them as networked narratives (Wang et al., 2017) in the sense that they are surrounded by technical resources, likes, and shared posts. According to Khazraee and Novak (2018), being part of a social media group contributes to shaping identities and communities that influence what is discussed by the members. The working conditions discussed by the teachers may be both conditional and limited by the fact that a new post can quickly make an earlier post less visible (Boyd, 2010). To some extent, to gain an understanding of positioning among teachers, we searched for protagonists to identify the alliances created within and outside the Facebook groups, as well as their conceptual antagonists (Khazraee & Novak, 2018), to determine the groupings or organisations outside (or within) the groups that teachers targeted.
In order to analyse the data, we used analytical concepts drawn from a narrative research tradition. In line with narrative research on social media (Georgakopoulou, 2013;Georgalou, 2017;Page, 2010), we also regard the two Facebook groups as arenas where small stories are shaped and negotiated through discussions about TUs. Typically, small stories on Facebook are open-ended and ongoing, as users contribute various posts asynchronously. They are short and "do not necessarily fulfil prototypical criteria of narrative inquiry such as beginningmiddle-end, a complicating event, and a clear evaluation of events" (Georgalou, 2017, p. 86). The ongoing storytelling in a multidimensional network of Facebook friends makes time and temporality multilayered, involving a sense of nowness, cyclical time, and hypothetical futures (Georgalou, 2017). Particular moments are given meaning through references to a common past, and cyclical repetitions open up for immediate as well as hypothetical futures. To mark what we saw as exclusive hypothetical futures in the data, we adopted Boje's (2011) concept of antenarrative. According to Boje, "ante" refers to a bet and a before; therefore, an antenarrative is a "bet on the future" (p. 1). To comprehend an antenarrative in relation to other narratives, one must understand it as a description of a narrative that might become, but does not yet possess linearity, coherence, or stability. This understanding enabled us to investigate the open-ended small stories as bets or suggestions for new stories about the relationships between TUs and the teachers.
We supplemented our small story analysis with the framework of institutional narratives (Linde, 2009). This allowed us to investigate how the teachers positioned themselves through stories about TUs and about school as an institution in change. According to Linde (2009), institutional narratives consist of a canon of stories about an institution. Some of these stories have an everyday character, while others-the retold ones-are well known to most teachers because they are frequently and repeatedly told memories from the past that represent what Linde calls the core memory of an institution. Retold stories become influential, as the teachers who tell them position themselves through a certain kind of argumentation in relation to the Swedish school.
In our analysis, we searched for small stories, institutional narratives, and especially retold stories related to TUs. We explored the content that different individuals told in various ways and how it was related to different past and present aspects. However, the main aim of the analysis was to examine whether there were any bets pointing to a new or different narrative about the teaching profession and its relation to TUs in the massive body of stories shaped in the Facebook rebellion groups.

Results: Narratives of a teaching profession on its way
The analysis revealed stories of temporal dimensions in which experiences of the past and present related to TUs and wishes and hopes about TUs told and retold by teachers contributed to forming a narrative about the Swedish school and the meaning of being a teacher. This narrative provides a picture of a teaching profession that is on its way somewhere, or at least wishes it were, but seems to be stuck in the "everyday swamp".
The overarching narrative consisted of three distinct narratives. The first narrative, Back to the future, was about taking advantage of and further developing the specific tasks and benefits that once formed the basis of one's choice to become a teacher. This can be seen as an institutional narrative (Linde, 2009). The second narrative, Everyday frustrations, concerned the current situation of the Swedish school and consisted of small stories (Georgakopoulou, 2013) about a school system in a poor condition, with exhausted teachers and students not receiving the help and support to which they were entitled. The third narrative, The new future, consisted of small stories about a hypothetical future (Georgalou, 2017) that we interpreted as a bet on what the teaching profession could and would be, on new thinking, and on changing status positions.
In each temporal dimension, TUs played a key role as possible agents of change. TUs in general were referred to in quite negative terms, such as cowardly and uninterested, while local union representatives were depicted in a more positive light, as they were also close colleagues.
The quotes presented below were translated from Swedish into English as faithfully as possible. As the narratives emerged from different posts, we do not provide the numbers of likes or the kinds of emojis received by posts and comments.

Back to the future
Users posted and commented on issues that had existed long before some of them were educated and became teachers. Several stories positively recalled more regulated working hours, usually referred to as part of an old agreement on the teaching obligation (Swedish: undervisningsskyldighet) that was in effect until the beginning of the 1990s. The teaching obligation presented an image, or a core memory (Linde, 2009), of the Swedish school system in an era before new public management. This was a time when the state, rather than municipalities, was responsible for the schools and teachers had almost equal teaching hours, while salaries were linked to years of experience and responsibilities. These stories evoked carefree times of a well-functioning school and respect for teachers and their assignments. Below are some examples of such retold stories: No more reforms, please. The only way forward is a return to what previously worked: teaching obligation, equal salaries, teacher autonomy, sufficient resources, solid teaching materials, and teachers' own working hours (Swedish: förtroendearbetstid) without meaningless meetings. Of course, this is a fantasy. The all-powerful winner never concedes anything without being forced to do so, and the holy agreement on cooperation has become a restriction that the union has lost the will to even try to get rid of. We still join in central negotiations like an abused wife in front of the husband who declares his good intentions while he continues to beat her.
The proof of this tragic metaphor was the last main union agreement in 2018. Not even the pandemic, during which we are exposed daily to a threat to life, has been enough for the unions to realise and acknowledge their complete failure. Not a word has been uttered to demand that the union bosses step down. The battle may have been lost, our profession eroded, but sacrificing even more is not a viable option. Please realise that the winners are not even afraid of the consequences of a lost generation despite a shortage of teachers.
Strike, my colleagues, is the only weapon powerful enough now! These quotations capture oft-repeated stories about what was better in the past and how TUs' failures led to the erosion of the teaching profession, typically ending with a call for a strike as the only solution. These calls for a strike seemed to reinforce the notion of TUs' failure to fulfil their task of creating good conditions for teachers. Traditionally, strikes have been a weapon used by trade unions to press their demands, but these stories presented TUs as overplayed, and it was the teachers themselves that called for a strike because Tus were unable to negotiate a good agreement. This institutional narrative about the past was an attempt to give members an agentive position as strikers, which, however, was not very powerful.
The prospect of resigned acceptance was present in several stories. It was about teachers who must be free on their vacations and in their spare time-something that had been possible before but had been made impossible by the latest agreements, which gave employers more power to decide teachers' working hours and tasks. Tus had caused the profession's alleged erosion. Changes were called for to enable Tus to reach an agreement with clearly defined working periods and reasonable conditions: Recent agreements have not benefited teachers. The union has failed to assert itself as strongly as it claims and has caused a divided, egocentric corps of teachers. Teachers who think more about their own leave than about those they are there for-the students-drive the profession's legitimacy to the bottom. I want to be able to be a professional teacher, with clear tasks, clear working periods, and reasonable conditions to complete the assignments.
The final quote from this narrative represents the retold stories about the "good old days" and TUs' role: You can change the absolutely basic premise (for being a teacher) and return to what was actually good. It is about making the teaching profession attractive. TUs need to talk about this. There is a lot that the TUs could do if they wanted to-if they represented us teachers. Tough words, but there is a lot to fight for.
The Back to the future narrative included strong stories about Tus' failure to preserve what was good. Constituting an institutional narrative (Linde, 2009), these were retold stories of great importance looking back at "the good old days" in a cyclical way, expressing hope for a hypothetical better future that took the past into account. However, this future was not particularly innovative. In this narrative, Tus were depicted as general or central organisations with no close relationship to the teachers.

Everyday frustrations
This narrative was characterised by frustration with deplorable TUs that failed miserably in everything. Such small stories (Georgakopoulou, 2013) were about TUs that were supposed to fight on behalf of the teachers but did so in quite another way than the teachers wanted. A frequently told story dealt with the question of whether to be or not to be active in a TU. This narrative highlighted the differences between unions at the central level and the local level represented by colleagues at school, which complicated teachers' relationships with their unions. Several conversations, or even arguments, consisted of small stories about the advantages and disadvantages of multiple unions, such as "several different unions are harmful" or "let's finally unite in a single strong TU." During the data collection, a new main agreement was presented, which was immediately rejected by most group members as wretched and paltry, as represented by the following excerpt from a longer thread: When agreements are vague and uninterpretable, they are seldom to our advantage. Clearer rules are needed, especially in New Public Management systems, in which everything must be made more efficient and in which the individual must bear the blame for not being good/efficient/skilled enough, even if it is the conditions that are wrong. Why do the unions not fight centrally? Locally, it is like banging your head against the wall of local politicians and their suggested cuts if there are no clear rules in the agreements.
In the analyses, it became clear that the narrative context and the interactions between the teachers were significant. Teachers who saw advantages and progress in the new agreement were well aware of their minority opinions. One of the few posts arguing that the new agreement included benefits for the teachers started as follows: "Can I write something positive about the main agreement?" and ended: "Now, every local representative needs to familiarise themselves with the new agreement and be able to use it to the benefit of the employees." This post received more likes than usual (70 thumbs up and smileys), and the thread included comments expressing both surprise ("Can the employer still do business as usual?") and curiosity ("Tell me more; in which paragraph?"). The discussion revealed that most teachers had not read the agreement; nevertheless, negative small stories dominated and were followed by complaints about TUs in general. The argumentation was the same as in the Back to the future narrative, in which "strike" was a buzzword used in almost every post. However, besides complaints, the threads also included repeated calls for teachers to become involved in TUs, which would otherwise become depleted and weaker. However, in response to a call to become involved, one teacher stressed the crippling burden that this would place on teachers in their frustrating everyday situation.
In the Everyday frustrations narrative, the small stories presented both employers and TUs as the teachers' antagonists, while the protagonists were the local unions and colleagues. The narrative revealed a kind of ambivalence: Teachers and local TUs were presented as neglected and helpless, creating an image of TUs divided between local and central tasks. Thus, TUs played a dual and tricky role, with TUs at the central level-incompetent and evil-on the one hand and their representatives at the local level on the other, trapped in the big system and hardly able to get the TUs involved in the teachers' everyday issues. However, there were also voices in these stories suggesting a change in this respect.

The new future
The small stories shaping this narrative were not as common, but some stories directly concerned a new future in which things might be different and a kind of bottom-up or insider's professionalism might emerge. These small stories formed an antenarrative in which strikes were no longer considered a relevant means of improvement. In the new future, digitalisation created new possibilities for decision-making and action. Besides the call for more active engagement in union activities, small stories exemplified how to make TUs more involved in teachers' work by simply inviting them to participate in teachers' activities. The following quotations represent a small story of activism as a means of change: A willingness to "do something" or a feeling that the unions are not doing anything is not enough. Such demands spread easily in all directions and are sometimes contradictory. As the most basic thing, we need to try to agree on a platform with the most important, well-elaborated requirements. I agree that physical encounters are fundamental. A step we can take even now, during pandemic times, is to hold local meetings. In larger cities, there may be several local meetings. The first task will be to discuss the content and possible actions for the platform. Nothing prevents these meetings and discussions from being open to those who prefer to participate remotely within the local groups' areas.
Even if we become several hundreds or even thousands, we are still a minority in relation to the number of union members. A central and ever-present effort must therefore be to present our platforms and actions and offer the unions the chance to participate in the actions that we ourselves may initiate.
Don't you think that, with this focus, we could take an important step forward?
This small story can be considered a hypothetical future (Georgalou, 2017)-a clear bet on how to move forward by organising in ways that attract TUs, but without specifying the content to negotiate.
A small story emerged from one post, and comments on it centred on trust in digital tools. The story elaborated on possible digital demonstrations and concerned about how employers would react to such novelties. A more concrete idea about using digital tools was to let all members take part in an agreement and sign it digitally before it came into effect, as expressed in the following thread: Let the teachers [members] make a decision on the agreement before it is signed instead of the agents at the central TU. There can be a reference group, and we can have digital voting. There are great opportunities to democratise the agreement process using the web.
The discussion continued with the possibility and suitability of voting and it was pointed out that few teachers seemed to be sufficiently familiar with this new digital agenda. Nevertheless, another teacher predicted that TUs could even be democratised in this way: Most teachers are dissatisfied with how the TUs work today. If it is technically possible to let the members participate, we should do it. It feels like those negotiating at the central level control much more than they are allowed to. Democratise the TUs! Some small stories depicted a school free from stress and anxiety, in which teachers had time for their assignments: It is about "staring time". This means time to do just that-stare, walk, run, and sort impressions and events without having to document, reply to emails, and refer to ordinances or plans. It would be seen as extended wellness for employees, I think. If you knew that there was room for staring time without anxiety about yet-to-be-graded tests, unfilled matrices, or uncopied documents, I think that many more tasks would be done easier and faster. It is so sad that we always have to learn the hard way, even though there is so much collective knowledge of the opposite.
Two replies to this post amounted to what we interpret as a call not only for the teachers but also for the TUs to take it into account: So wise! It would be an investment in the future. In addition, it may be possible to have a two-teacher system.
It is not about being a superhero but just about being a regular civil servant.
The bet suggested in this antenarrative was that a new kind of democratic teacher professionalism might be possible if new techniques and social media were used. Although the content of this professionalism was vague, it involved elements of sustainability. The new professional teacher was presented as someone who had time to stare. This kind of professionalism was based on the idea that the teacher could decide what needed to be done and what to prioritise without having to be a superhero. Perhaps we can call it professionalism based on respect for teachers' discrete knowledge of how to handle professional tasks with colleagues.

Discussion
In this study, we have been investigating narratives about the teaching profession and Swedish TUs jointly produced by Swedish teachers in two teachers' rebellion groups on Facebook. Our findings show that members of Swedish teachers' rebellion groups on Facebook are as dissatisfied with TUs as previously suggested (Bascia, 2008;Fidan & Erkocak, 2015). The various narratives describe disappointed teachers who struggle with their everyday frustrations, demanding a return to "the good old days" and calling for a strike as an immediate solution. However, there is also a narrative about a bet, a new hypothetical future in which proactive teachers invite the unions to talks rather than wait for the unions' hitherto insufficient efforts to satisfy their demands. This quest for a more communicative relationship between TUs and their members that acknowledges the complexities of teachers' everyday work is in line with a more democratic kind of professionalism that ensures that teachers are heard and represented by their unions (Stevenson & Gilliland, 2015).
What can these narratives tell us about the Swedish teaching profession? We understand them as expressions emerging from layers of institutional memories (Linde, 2009) of the Swedish school and teachers' working conditions. The present strained conditions (Lundström, 2015;Samuelsson et al., 2018) are contrasted with a sometimes romanticised view of past conditions. However, these institutional memories might hinder, rather than support, the teaching profession. Everyone probably knows that it is impossible to return to the previous status, and the narrative Back to the future is an expression of what might be different because it was once different-though in another societal context. Teachers and their relationships with TUs do not seem to have kept up with the present time of "fast policies" (Hardy et al., 2019). Who, then, struggles against time? Is it the teachers, the unions, or both? Maybe the basic idea of a relationship with the unions has been lost along with the loss of teachers' autonomy. Perhaps teachers simply do not have the same ability-or strength-to look ahead and make constructive suggestions when they are stuck in the "everyday swamp", leaving them with longings and wishes while calling for strikes. From a more general point of view, and in line with Ringarp (2011) this case illustrates how economic conditions and the recent development of the welfare state in Sweden limits the possibilities for TUs as well as teachers to raise their voice in the discussion of teacher professionalism.
The relationship between teachers and TUs seems fragile, and the way in which TUs are described hardly contributes to recruiting new leaders. In the long term, this might lead to an even more centrally controlled TU allied with employers and policymakers, with an outsider's perspective on teachers' working conditions (Lilja, 2013) that reinforces professionalism from above (Evetts, 2003). The question often asked in the groups' discussions about whether one should be a TU member is highly relevant not only for TUs as organisations but also for the Swedish school system, teachers, pupils (not least), and society at large. If teachers merely long for "the good old days" without taking responsibility, there is a risk that the "everyday swamp" that many teachers testify to will remain. We argue that the Swedish school system and teaching staff need to preserve the union tradition and unlock the potential of the relations between teachers and TUs based on respect and responsibility, as suggested in The new future narrative. This is a hypothetical future that holds what we see as a possible bet. This bet is based on the idea that teachers involved in Facebook rebellion groups can trust TUs' ability to balance loyalty demands based on NPM and external control with the teachers' need for improved everyday working conditions defined on a more collegial basis from within (Evetts, 2010;Löfgren & Wieslander, 2020). Such trust, at its best, could become a fertile soil for reflective teacher professionalism, which can be resilient to the negative effects of NPM in the Swedish education system if TUs are ready to listen carefully to what needs to be done according to the teachers. However, teachers in Facebook rebellion groups also need to be clearer in their message if this bet is to become something more powerful than a mere statement on a social media platform.
A related concern is who will represent the teachers in the fight for improved working conditions in the long term. Will it be a democratically appointed TU representative or selfappointed administrators and loud voices in teachers' rebellion groups in social media? The question that arises is whether all teachers' voices can be represented, as in The new future narrative, or whether a certain vision of the professional will be presented, developed from teachers' experiences of being members of a teachers' Facebook group. Perhaps such a bet has the potential to lead to increased democratic engagement among teachers-in both TUs and rebellion groups-who want to work for healthier and more sustainable working conditions. This, however, seems difficult to accomplish without a clearer picture of what should be in focus and what the actual problem is. The bet identified in this study is more of a suggestion on how virtual collaborative resources can act as a vehicle for past and present experiences, thus functioning as part of an infrastructure that can shape a more autonomous teaching profession with more control over working conditions. The narrative approach used in this study contributes to visualising the temporal dimensions of teachers' relations with TUs. Without an understanding of the historical layers depicted in the narratives and teachers' everyday frustrations, it is hardly possible to sketch a new hypothetical future. Temporality is another aspect of the data derived from the Facebook discussions. A post from a rebellion group member represents what Georgalou (2017) terms an immediate future "that constitutes part of a description or evaluation of the present" (p. 94) at the moment it is written. However, such a post might be a description of the past, as another member commented a few hours/days later. The cyclical time emerging from our data is both a strength and a limitation, as it is easy to become lost between the nowness and the hypothetical future. We are aware that some posts may have been hidden behind other posts and thus may have been less visible in our data. However, the concept of retold stories guarantees that we, in our analyses, used the small stories that best revealed the teachers' relations to TUs. Using the concept of bet contributed to a more nuanced understanding of temporality and enabled us to present a narrative containing ideas that might otherwise have been left outside our results as too naïve or unrealistic.