Affects of verbal hygiene: the impact of language activism at a Swedish high school

This paper explores youths’ language activism at a Swedish senior high school. Following Cameron (Verbal hygiene, Routledge, Florence, 1995), this paper investigates language activism as ‘verbal hygiene’, with a focus on the social dimension of students’ attempts to change how language is used at the school. To capture how the politically motivated language activism came to produce political subjectivities and delineation between social groups, and also to impact the distribution of agency and voice in the local discursive regime, I combine Cameron’s (Verbal hygiene, Routledge, Florence, 1995) notion of verbal hygiene with recent theorizing on affect (Ahmed in The cultural politics of emotion, Edinburgh University Press, Edinburgh, 2004; Sedgwick in Touching feeling: affect, pedagogy, performativity, Duke University Press, Durham, 2003; Wetherell in Affect and emotion: a new social science understanding, Sage, London, 2012). Inspired by studies of semiotic landscapes (Jaworski and Thurlow in Semiotic landscapes: language, image, space, Bloomsbury Academic, London, 2011), the data includes posters and other signage in the high school, as well as recordings of naturally occurring interaction and interviews. The analysis shows how shame is pivotal in the processes of language activism at the school and how this activism, being an emancipatory project, both produces political subjectivities as well as new linguistic normativities and hierarchies.


Introduction
In Sweden, as elsewhere, political activists have turned their attention to issues of everyday language in an effort to draw out biases, sexism, erasures, and structural violence hidden in, and played out in, language use. Matters of, for example, pronoun use and (non-)sexist or (non-)racist language are put into high relief as activists strive towards a more inclusive use of language (e.g. Cameron 1995;Kulick 2014). Following this, activists' focus on language use has become the topic of much political discussion and seems to deeply divide different political groups. There are those who believe that language is a site where social injustices and inequalities are reproduced, justified, and normalized, and on the other side of the debate are those who are frustrated by 'political correctness' and the efforts of 'leftists' to censor free speech. This is a debate that generates strong emotions, and as emotions run high the discussions often seem polarized to the point where nuance or problematization is made difficult.
Within the study of language policy, the relation between language and the political is of course one of the core issues, and insights from the field might be utilized to add nuance to a polarized debate. The potential of language activism to produce social change has been the topic of much investigation, and this paper seeks to offer a novel contribution to academic discussions on contemporary politically motivated language activism through a focus on the affective facets of such activism (Ahmed 2004;Wetherell 2012). Ahmed's (2004) and Wetherell's (2012) theorizing teaches us that affects are performative, and this paper takes an interest in what affect 'do', and what part affect plays in acts of politically motivated language activism. Following McCarty (2010) and Ricento (2000) I use an ethnographic approach to study the micro-level of language policy as I attempt to capture affective and social dimensions of politically motivated language activism. I do this by investigating Swedish high school students' feminist and trans-activist verbal hygiene-that is, students' critical reflections on language use as it relates to gender (Cameron 1995: 9). I will document youths' efforts to change other students' language use as well as analyze the uptake of, and reactions to, such language activism in order to understand how the practices of verbal hygiene play into the social reality for the students and staff at the high school. I trace the potency of the activism to the performative function of affect (Ahmed 2004;Sedgwick 2003;Wetherell 2012), and I will argue that in the setting where my fieldwork took place the affective and performative dimensions of language activism came to play an important role in the creation of political subjectivities, the delineation between social groups, and the distribution of agency and voice in the local discursive regime.
The high school where the fieldwork was conducted stands out in a Swedish context as a school that attracts students invested in social issues. In students' discussions of anti-racist and feminist issues, everyday language use was often regarded as an important site of political struggle. The students' activism was underpinned by a view of language as not merely reflecting the social, but also shaping it, and meta-commentary on language was often guided by an understanding of language as a fundamental tool for achieving social transformation. Although such language ideologies and language activism are found elsewhere, within the walls of this school the practices of politically motivated verbal hygiene were particularly condensed. Students were frequently engaged in discussions of the merits of different words or identity labels. This school thus presents itself as an ideal site to study how politically motivated language activism is played out among youth. With ethnographic methods and an analysis sensitized to affect, it is possible to come closer to understanding the potency and multifunctionality of language activism as it is played out in everyday life, and thus to add nuance to the conversation surrounding the increasingly vexing question of everyday language as a site of political struggle and 'political correctness'.
The article is structured as follows. I start by briefly describing the notion of verbal hygiene (Cameron 1995) and introduce the theorization of affect that informs my analysis of the students' language activism. This is followed by a discussion of the methodological toolkit and a brief introduction to the Swedish context and the school setting where the data were collected. The analysis consists of three subsections. In the first section, 'Shaming on/for the Transgender Day of Remembrance', I analyze a poster produced by the school's LGBTQ association as an example of verbal hygiene at the school. In the second section, 'Learning to speak with a legitimate voice', I study examples of the student's uptake of the on-going verbal hygiene. In the final section, 'Consenting to student 'expertise'', I explore teachers' and staff's efforts to navigate the local discursive regime. The findings are then discussed in the conclusion of the paper.

The affects of verbal hygiene
In her seminal book by the same name, Deborah Cameron explains that verbal hygiene 'comes into being whenever people reflect on language in a critical (in the sense of 'evaluative') way' (Cameron 1995: 9). Verbal hygiene as a notion highlights that '[a]ll attitudes to language and linguistic change are fundamentally ideological' (Cameron 1995: 4) and that the 'impulse to regulate language, control it, make it 'better'' (Cameron 1995: 9) can be found everywhere as people meta-comment on language use. In proposing the notion of verbal hygiene, Cameron disclosed having 'two concerns: to document what particular verbal hygiene practices involve, and to understand what they mean, both for those who engage in them and the wider culture in which they arise ' (1995: 30). Following Cameron, I analyze verbal hygiene within the setting of Bråviken (broːvi:kɛn) high school with an ambition to shed light on how such practices impact the everyday lives of the students. I will also show that such an analysis can tell us something about politically motivated language activism in general and why it evokes such strong emotional reactions. However, to study the social impact of verbal hygiene and do justice to the potential of ethnography to capture the complexities and messiness of the mundane, I will combine Cameron's (1995) ideas with the recent theorization of affect in terms of practices (Wetherell 2012), politics (Ahmed 2004), and performativity (Sedgwick 2003). This allows me to unpack the performative force of language activism and give a more nuanced description of the multiple functions it serves in social context. This interest in affect is shared with other sociolinguists who have recognized its analytical and explanatory value in understanding the relationship between language and social life (e.g. Bucholtz et al. 2018;Kiesling 2018;Milani 2018;Wee 2016). Authors within the field have been inspired by different approaches through which to theorize affect. Some scholars treat affect as an automatically triggered somatic movement beyond, or rather before, discourse. Contrary, instead of regarding affect as pre-discursive, this paper relies heavily on those perspectives that treat affect as always entangled in and inseparable from human meaning-making and discursive practice. Further, as Ahmed (2004) points out, emotions should be considered not for their ontological status but for their ability to do things. Following this call, I interest myself in the performative force of affect and will in the analysis show how the circulation of affect is inseparable from the constitution of subjects. In Ahmed's words, I will 'reflect on the role of emotions in the politicization of subjects' (Ahmed 2004: 171).
In a dialogue with the work of Ahmed (2004), Wetherell (2012) introduces the notion of affective practice to situate the workings of affect in social context and discursive practice. Similarly to Ahmed (2004), she highlights the performative function of affect and states that affective practice is 'onto-formative, meaning that it constitutes subjects and objects' (Wetherell 2012: 159). However, whereas Ahmed´s analysis is at the level of national identity and more general figures like 'the asylum seeker', Wetherell's theorizing can be helpful to study affect as situated in everyday social practices, in the 'everyday human muddle' (Wetherell 2013: 358). Wetherell argues that acknowledging the performative force of affect does not mean that it should be treated as the dominant actor, and human action and meaning making are still to be put at the center of the analysis: 'Rather than affect per se on a pedestal, as the topic, we can become interested in a multi-modal situated event, in a consequential set of sequences in social, cultural and institutional life' (Wetherell 2015: 159). Thus, Wetherell's notion of affective practice sits well with an ethnographic approach because it 'focuses on the emotional as it appears in social life and tries to follow what participants do' (Wetherell 2012: 4). Using ethnographic methods, I study verbal hygiene as affective practices and describe how they generate effects. I will argue that the on-going practices of politically motivated verbal hygiene are part of the production of a local discursive regime in which students' and teachers' language use is understood, valued, and charged with social meaning. Entwined in this production of a discursive regime is an on-going delineation of social groups and the establishing of a common 'we'.
As I will show, shame and shaming play a significant part in the local metadiscursive practices presented here, and so their political potential must be taken into account in the analysis. The affective practice of shaming will be understood as a vital part of the performative force of the language activism at the school. It is important to note that in the framework presented here shame should not be considered as a negative emotion to be excised (Sedgwick 2003). Munt (2007: 3) notes that shame is 'popularly perceived to be an affliction, a toxin to be avoided by good behavior or to be "processed" out of existence into an ideally shame-free future'. Instead, she argues that shame should be seen as an ambivalent, yet potent, emotion with many different potentials (see also Ahmed 2004). Paying closer attention to shame and its performative force in practices of language activism assists us in doing justice to the messiness and complexities of life at the level of everyday experience because '[a]ffective practice is continually dynamic with the potential to move in multiple and divergent directions' (Wetherell 2012:13). An analysis sensitized to the trajectories of affect thus allows for a closer understanding of what is at stake for the teachers and students at the school as language is being regimented and policed. This can help us to shed light on the multifunctionality of the language activism at hand, the formation of subjectivities, and the local redistribution of linguistic agency and voice. Further, taking affective theory as a point of departure (Ahmed 2004;Munt 2007;Wetherell 2012) should make possible a more nuanced and explorative analysis.
As Milani (2015) shows in his analysis of feminist activism in a South African setting, affective theory (e.g. Ahmed 2004) offers an opportunity for an analysis that moves beyond 'a clear-cut, Manichean distinction' between good and bad and towards a more contemplative or 'multifaceted reading' (Milani 2015: 433) that does not have a critical value judgment as its endpoint.
In the operationalization of affective theory, I use analytical concepts found within linguistic ethnography (Copland and Creese 2015) as well as in different strands of discourse analysis, namely multimodal critical discourse analysis (Machin 2007;Machin and van Leeuwen 2016) and analysis of spoken discourse and narratives (Cameron 2001;Georgakopoulou 2007). These strands of discourse analysis offer a set of analytical tools that can be used to unpack the dynamics of the language activism at the high school and capture its affective aspects.

The Swedish context, setting, and fieldwork methods
Sweden has witnessed the breakthrough of several socially inclusive linguistic innovations put forward by language activists or as a result of language planning by linguistic authorities (see Wojahn 2015 for a discussion of language activism in Sweden between 1960 and 2015, and Milles 2011 for an elaboration of feminist language planning in Sweden). One example of such innovations is hen, a gender-neutral pronoun (as a complement to the gendered hon 'she' and han 'he') (Bäck et al. 2017;Lindqvist et al. 2018). Another recent example in Swedish is the use of en (comparable to 'one') as a gender-neutral generic pronoun. The traditional generic pronoun in standard Swedish is man, and arguments have been made that the use of man should be understood as a token of language sexism because it stems from the noun man, meaning 'male person' (for a more extensive discussion, see Milles 2019). These innovations-along with other suggestions for linguistic change put forward with the avowed motive of creating a more inclusive, emancipatory, and just use of language-have not been unanimously celebrated, but have met resistance and led to a debate in Sweden on the merits of 'political correctness' (Wojahn 2015). This debate largely mirrors the one seen in the North American setting, and the arguments against 'political correctness' are fundamentally the same as those described by Cameron (1995) two decades ago. This discussion was also unfolding at the high school where this study was conducted as students and teachers tried to navigate the on-going language activism at the school. Naturally, not all students participated in the processes of verbal hygiene to the same extent, and students and staff reacted to them in different ways. Some students I talked to thought that the political activism and language regimentation had been pushed too far, while others celebrated the school as progressive and inclusive. This is discussed in more detail in the next section as I give a more detailed introduction to the high school in question.

Fieldwork setting
The data presented in this paper were collected during six months of ethnographic fieldwork and consist of field notes, recordings of naturally occurring interactions, and interviews. The high school where the fieldwork was conducted, situated in the center of Stockholm, will be referred to under the pseudonym Bråviken. Sweden has, perhaps in contrast to the image of the Swedish welfare state, one of the most market-liberal school systems in the world (Ambrose 2016). It is built on a voucher system where the schools receive funding in relation to how many students they attract, and students are free to apply to any senior high school in their municipality. In this situation, schools have been trying to carve out their piece of the market in different ways. One strategy is to create their own 'brand image', and Bråviken's reputation for being a politically 'left-wing' school might also be seen as important to attracting a segment of the population of high school students in Stockholm. The school did indeed attract students who were interested in issues of, for example, gender equality, queer rights, anti-racism, and environmental sustainability. Admission to Bråviken, and other high schools in Stockholm, is based on grades, and relatively high grades were required to be admitted to Bråviken at the time of the fieldwork. Students attending high schools in Sweden are aged between 15 and 18.
Bråviken is situated in the district of Södermalm. Historically Södermalm housed the working class, but through a process of gentrification Södermalm is today as attractive on the housing market as any other inner-city district of Stockholm. Further, Södermalm has become something of an emblem in political discourse in Sweden and is often inserted as a symbol for political correctness, hipsters, and a leftwing media establishment. 1 During my fieldwork at the high school, I quickly noted the advanced political discussions and debates taking place in classrooms, in hallways, and at student meetings. I found it interesting that discussions of issues like racism, gender inequality, homophobia, and transphobia drew heavily on discourses found in gender studies and other academic fields. At the high school, social constructivist perspectives were refracted in students' discourse, and everyday language use was often highlighted by the students and regarded as an important site of political struggle and social transformation. Consequently, students' meta-comments on language were often motivated by a desire to create an inclusive, tolerant social space through the regimentation of language. In relation to the school's profile and the students' activism, many students reportedly found this school to be a safe haven where they were free to express their political views or non-normative gender identity in a way that they could not do in other schools. Other students, however, complained that it was 1 3 Affects of verbal hygiene: the impact of language activism… difficult to articulate an opposing political opinion without being at risk of social exclusion.
Further, students' political engagement and language activism were often semiotically observable in the visual landscape of the school. Individual students and student associations produced posters that advertised events or conveyed political messages to the student body. The process of data collection has therefore been inspired by studies of linguistic/semiotic landscapes (Jaworski and Thurlow 2011) and signs in place (Scollon and Scollon 2003). The data include photos of the posters, stickers, and graffiti that formed the visual landscape of the high school. The fieldwork practices were further informed by Stroud and Mpendukana's (2009) call for an ethnography of linguistic landscapes that 'explor [es] in detail how people take up, use, manage and discard, interact with and through, re-contextualized media as they insert signs and artifacts into practices and ideologies of language construction in their everyday interaction' (Stroud and Mpendukana 2009: 382). Following their call, I combine photos of the interior of the high school with transcripts of verbal interactions in my analysis. From a dataset of about 980 photos and 165 h of audio recordings of naturally occurring talk, ethnographic chats, and interviews, as well as field notes, I have selected data excerpts that allow me to discuss the affective dimension of the language activism at the high school and how it produced political subjectivities.
The Bråviken high school is in no way a generic Stockholm high school, if such a thing can be said to exist in the stratified school market of the city, and my aim is not to draw generalizable conclusions based on my analysis. However, the processes of politically motivated verbal hygiene (Cameron 1995) unfolding at the high school can be said to be a condensation of similar processes in other arenas and on other scales. Therefore, by analyzing the affective dimensions of local language activism at Bråviken, I can add to the discussion on similar phenomena elsewhere. As I will show, the affective dimension of the activism functions to create political subjectivities at the same time as it regiments ways of speaking and being. This is demonstrated in the next section, where student language activism as well as responses to it are presented and discussed.

Analysis
The analysis section consists of three sub-sections where the first, 'Shaming on/for the Transgender Day of Remembrance', shows how students' politically motivated verbal hygiene, through the work of shame, produces political subjectivities and creates a local discursive regime in which the stakes for the speakers are high. In the second section, 'Learning to speak with a legitimate voice', I discuss student reactions to the verbal hygiene and present examples of students' efforts to socialize themselves. In the final section, 'Consenting to student 'expertise'', I turn to reactions to the language activism among the school staff and show how the affective aspects of the activism also impact the distribution of voice between staff and students.

Shaming on/for the transgender day of remembrance
The first data excerpt was chosen as an entry point for discussing the meta-linguistic commentary at the school and how it might be articulated and entangled with discourses of emancipation in such a way as to raise the social stakes for speakers. The poster below was on display in the student cafeteria on a noticeboard where the school's student associations advertise upcoming events. The poster does not, however, advertise any particular social event at the school. It was created by the student LGBTQ association, one of many associations active at the school, and it informs students about the Transgender Day of Remembrance internationally observed on 20 November.
The title of the poster (Figure 1) is in English, and drawing on Androutsopoulos' (2012) notion of 'English on top' I argue that the use of English in this context serves 'as a hint (a pointer, or index) to the linguistic practices of certain groups or places that are deemed important for the interpretation of the on-going discourse' (Androutsopoulos 2012: 232). The English title frames the text and highlights the interdiscursive links to a global, or perhaps mainly US, activist discourse. Positioned on the top of the poster, the use of the English title 'Transgender day of remembrance' authenticates the text that follows (see also Martin 2007).
In the second section of the text, the author uses different semiotic resources and their affordances to construct what can be read as one long sentence: 'You contribute to this [arrows pointing to the info box] when you [followed by a list of social actions]'. In the Swedish original, the first part of the sentence 'You contribute to this' is a main clause and the second part 'when you [list of social actions]' is a subordinate clause. The list on the poster offers various endings for the subordinate clause and points to social actions that contribute to the structural violence that trans people are subjected to in Sweden. The social actions that are to be avoided include acts of physical violence ('[physically] abuse and murder us') but are predominantly different types of speech acts (Austin 1962), e.g. 'Misgender/go on about deadnames' and 'Saying fitt-/kukbärare when you mean woman/man'. The list does not present a hierarchy of the different social actions that are to be avoided, and no differentiation is made between 'laughing at transmisogynistic jokes' and 'murder'. Laughing at transmisogynistic jokes is further exemplified on the poster with 'Like haha man in a dress funny'. The fictitious person laughing at a man in a dress is stylized (Jaspers and Van Hoof 2019) to come across as something of a brute-an uncivilized person with a very simple sense of humor.
The use of 'when' (när) as a conjunction should be noted because it implies that the reader is actually performing the actions listed. If the poster had instead used the conjunction 'if' (om), this would have turned the subordinate clause into a conditional clause. There are few nominalizations in the text, which further highlights the reader's agency and responsibility for the social actions. Machin and Mayr (2012:139) argue that nominalizations can erase actors, hiding both the agent and the affected. In contrast, the lack of nominalizations in the poster works to highlight both the agent (the out group/reader) and the affected (the in-group/writer): the poster stipulates that 'You contribute to this', explicitly interpellating the reader as the agent. Further, Machin and Mayr (2012:142) state that nominalizations can remove a sense of time and work to obscure causality. The poster, in contrast, explicitly states the causality: the actions of the out-group lead to the suffering of the ingroup. Further, the word 'YOU' is emphasized by its increased size and boldness in relation to the rest of the text. The writer also uses color as a semiotic resource by outlining 'YOU' in white to make it stand out in relation to its textual environment

Transgender Day of Remembrance
Is a day to honor and remember victims of transphobic violence but also to bring attention to and prevent the structural transphobia that leads to this violence.

Heavy facts
According to the Public Health Agency of Sweden 50% of trans people in Sweden have tried to commit suicide. Many suffer from mental illness and are unemployed. Trans women especially suffer from harassment and violence.
when you: • Laugh at transmisogynistic jokes (Like ha-ha a man in a dress funny) • Use pussies and menstruation as feminist symbols • Misgender/keep using deadnames • Reduce people to their genitals (e.g. to be 'pussy-sexual': trying to figure out what genitals a trans person has got: saying pussy-/cock-bearer 2 when you mean woman/man as in [']cock-bearer have higher salaries[']) • Sexually harass, offend, violently threaten, abuse and murder us!

Figure 1
Poster in the student cafeteria (Van Leeuwen 2006:144). The text ends with 'So stop doing that!', again presupposing that the students reading the poster perform the actions listed.
Three textual elements stand out in regard to size-'Transgender Day of Remembrance', 'YOU contribute to this! when you:', and 'So stop doing that!'. Walking past the sign, these quite brash parts of the text would be the first to be registered. Other parts of the text are less reader friendly and force the interested reader to approach the poster to study it closer. This creates an interesting effect, and some elements of the text compel the reader to move closer to the poster. Simultaneously, other textual elements are 'in your face' and quite intimidating.
The analysis shows that the rhetorical and semiotic resources deployed offer a meta-commentary on the words that students (allegedly) use and misuse, and in so doing conjure two groups, namely the in-group, equipped with knowledge and the authority that derives from it, and the ignorant out-group. According to the poster, the interpellated reader's lack of knowledge is materialized in the reader's everyday speech; by using language in a certain way, the reader participate in structural violence against trans people. This is a message that builds on a theory of language that does not uphold a distinction between speech and action (e.g. Butler 1997). At the high school, meta-comments on language were often guided by an understanding of language as not merely reflecting the social, but also shaping it. In this view, language becomes fundamental for the reproduction of social exclusion, but also an important tool for achieving social transformation. The stakes for the speakers are therefore high-if they do not correct their language, they will continue to be complicit in a structure that leads to mental illness, unemployment, and suicide.
The LGBTQ association used a variety of semiotic resources to produce a discursive regime where students belonging to a vulnerable minority can speak with an authoritative voice on the issue of language use. This snapshot of the on-going language regimentation is also an act of 'calling out' and shaming the students who use language in what is understood as a hurtful way. In line with Ahmed (2004) and Wetherell (2012) and their theorizing of affect, I assume that the act of recruiting emotional states is productive in the sense that it constitutes subjects and groups. In other words, as the LGBTQ association is assembling multimodal recourses to 'call out' and shame their peers, the emotionality of this act of verbal hygiene cannot be bracketed out of the constitution of a group, one with the right not to be targeted by violence or linguistic erasure. This shaming and act of 'calling out' is an affective practice recognizable to the student body that was invested in, or at least knowledgeable of, US activist discourse. Again, it is important to note the intertextual links produced by the use of English and the call to observe the 'The Transgender Day of Remembrance', a practice started by US trans activists. The re-contextualization (Gal 2006) of US activist discourse is important to note because it serves to legitimize not only the text per se, but also the antagonistic and defiant stance that is adopted in shaming the reader for contributing to the suffering of trans people. This interdiscursivity contributes to the 'normative organisation' (Wetherell 2015: 81) of the affective practice at hand and adds authority to the text. Further, as a performative, the affective practice of verbal hygiene 'works by citing norms and conventions that already exist' (Ahmed 2004: 93). That is, the defiant stance and the act of calling out are socially recognizable to the reader through the links to a US activist discourse, and this is part of the performative force of the verbal hygiene.
Shame and shaming can, in relation to language activism, be seen as generating effects in a multifaceted way. Because shame is constitutive of subjectivities, the affective facet of this language regimentation is essential to the activism's force. To unpack this point, it is important to recognize that 'shame both derives from and aims towards sociability' (Sedgwick 2003: 37). To feel ashamed, I need the gaze of someone else. But for shame to stick, the gaze cannot come from a completely unknown other; the witness of my wrongdoing must be someone in whom I am socially invested in and who is socially invested in me. Ahmed (2004: 102) writes that shame is 'a mode of recognition of injustices committed against others', meaning that shame operates only if there is a common understanding of what is unjust. The language activism in the poster presupposes, then, not only that other students at the school are reproducing normativities that harm trans people, but also that (1) the student body at large, just like the LGBTQ association, understands this as unjust, and, importantly, (2) both groups understand that students' everyday language use is at the center of this reproduction of linguistic violence. This means that even though the poster so clearly conjures two groups, the knowledgeable author versus the reader, it also produces a common 'we' that strives for an inclusive, respectful, and safe atmosphere at the school and believes that everyday language is pivotal in this struggle.
This presupposition and organizing of a common 'we' becomes more salient in the next part of the message from the LGBTQ association. The first poster was followed by another, placed beside it on the same noticeboard. Here, the reader is invited to learn how to speak in a more inclusive way. Its starts with the title 'Do this instead', followed by a set of bullet points (Figure 2) 2 To provide the reader with a concrete list of alternative actions and alternative language use can be seen as a gesture towards the commonality between the sender and the reader: that we all agree on the basic assumption that we want all students, including trans people, to feel respected, and that making alterations to our everyday language is a step towards this goal. Note that the reader is also invited to 'call out' and correct others who are complicit in transphobia or misgendering-that is, to reperform the very act of political verbal hygiene that is being performed on them in the moment of reading. The reader of the posters is construed as someone who shares norms and values with the LGBTQ association but who lacks the knowledge and linguistic resources to embody them.
These posters were chosen as examples of the on-going language activism at the high school. Through the analysis of these examples, we can see how such activism produces political subjectivities that claim the right not to be subjected to verbal abuse. It also produces a common 'we' that strives to create an inclusive social space while at the same time generating a discursive regime in which the students' everyday language is strictly regimented. This regimentation of language at the school was often discussed among students as well as staff, and in the next section I will move on to analyze reactions to the language activism. The following excerpts will be examples not of language activism per se, but rather of reactions to it.

Learning to speak with a legitimate voice
Language can be understood as a set of resources, the distribution of which impacts social actors' capacity to be listened to and understood. In any given context, linguistic (or semiotic) resources, be they specific words, jargon, pronunciation, styles, etc., can be employed to speak with a recognizable, enregistered (Agha 2007) voice that embodies values, identity positions, authority, etc. Any practices aimed at linguistic change also serve to reconfigure the distribution and control of these linguistic resources.
Like any social group, the student body was in no way homogeneous and the students reacted in different ways to the on-going language activism. Further, the linguistic resources that were needed to adhere to the different, and at times contradictory, rules of language use suggested by different groups of activists were not evenly distributed among the students. Some students expressed that they had been quite nervous before starting at the school because they had thought that all the other students at the school were much more politically knowledgeable than themselves. In the next excerpt (extract 1), Frida describes how she fell silent in class during her first week as a new student at the school because she did not know what a cis person was.  A cis person is a person whose gender identity matches the sex that the person was assigned at birth. Frida confesses that this term was not part of her vocabulary when she started in her class. In her description of the situation above, she is not directly disciplined or challenged by the other students. Instead, it is she who decides that she lacks the knowledge and linguistic competence required to have the right to speak and to own a voice in the classroom (lines 2-4). The pauses (lines 3 and 4) become a theatrical reenactment and recitation of her own silence in class. They contrast with her description of how the other students seemed to effortlessly display their political awareness by casually using the term cis: 'people just yes cis people lalalala' (line 1). She goes on to explain that she did not dare to ask what cis meant since everyone else seemed to know.
In the initial analysis of the posters for Transgender Day of Remembrance, I argued that verbal hygiene (Cameron 1995) was raising the stakes of language use in the school. In light of this, it is possible that Frida chose to be silent to avoid being 'called out' by other students for her ignorance. However, I believe that Frida's narrative also warrants an analysis that includes the sociability of affect and the desire to be part of a progressive collective that strives to be knowledgeable and inclusive. Frida narrates this as a humorous episode (laughter and smiley voice in lines 1-2), and she does not try to distance herself from the students she cites. Ahmed (2004) writes about the entanglement of shame and love and argues that shame is generated by a failure to live up to a desired ideal. She writes: 'Shame can work as a deterrent: in order to avoid shame, subjects must enter the "contract" of the social bond, by seeking to approximate a social ideal. Shame can also be experienced as the affective cost of not following the scripts of normative existence' (Ahmed 2004: 107;italics in original). If Frida's reported silence is interpreted as a result of feelings of shame, it is not only in a negative sense wherein she is afraid of being called out and shamed by others. Rather, part of the production of silence here is Frida's 'desire to be "like" an other, as well as to be recognized by an other' (Ahmed 2004:106). For Frida these linguistic norms are new, and she does not possess the linguistic resources to follow the local 'scripts of normative existence' (Ahmed 2004: 107). Later in the interview, she aligns with the idea of using hen or en as a means of being more inclusive, and she describes how she, mostly via social media, obtained knowledge of the new linguistic resources she felt she needed. Her story of her falling silent is not so much a critique of the language norms that she failed to live up to as it is a humorous episode exemplifying her own process of learning the register of the knowledgeable activist, of her own language socialization. This is an aspect of the school's language activism and regimentation that is important to note in order to explain such activism's potency. Most students agreed on the moral justification and language ideologies underlying the activists' language norms and were therefore willing to adhere to the activists' suggestions. Hence, some lexical items and linguistic skills, e.g. being able to use cis or the gender-neutral pronoun hen, or mastering the use of the generic pronoun en in everyday speech, become important symbolic assets for those who want to speak with a voice that is considered to embody a progressive world view. Thus, the students develop strategies to incorporate new terms and other linguistic resources they find useful to make their speech more inclusive. As mentioned, the student body was not homogeneous, and there were also students who were skeptical of what they perceived as a disproportionate focus on language use. For those students, stories of other students' efforts to adapt to the linguistic regime could serve as laughable examples of how far students would go in their efforts to socialize to the linguistic norms at the school. In the next excerpt (extract 2), Moa, who at times refers to the school as a 'bubble', is narrating an episode she witnessed on her bus ride to school 1 day.
Extract 2   1 Moa: Yes there were a some first graders on bus number six like next to me on 2 their way to the school and they said like that they should start saying one 3 ('en') instead of one ('man') and as soon as one said one (man) then they would 4 hit that person like to learn okay if you say one ('man') then you get 5 punished so you must say one ('en') instead In the group interview, Moa narrates a small story (Georgakopoulou 2007) about students she reportedly witnessed on a bus ride to school. In a slightly mocking tone, Moa describes how the other students would use an almost Pavlovian strategy in their language socialization. They would help each other to change their use of the generic pronoun from man to en by hitting each other if they used the wrong word. Her story evokes laughter and, contrary to Frida, Moa is using her narrative to make fun of how far language regimentation has been taken by other students. In the interview she explains that she really has nothing against changing man to en, but that she finds it difficult to do so herself in speech. However, she uses her story of the physical discipline of speaking bodies to distance herself from this type of excessive behavior.
We saw the emancipatory potential of the activism in the previous section. Simultaneously, as the reactions of Frida and Moa illustrate, in this discursive regime the mastery of linguistic items like cis or en became important if students wanted to embody an inclusive and politically engaged way of being. Frida's and Moa's reactions exemplify stances taken by students in relation to the on-going language activism. Frida's comment demonstrates the potential of the activism to produce selfimposed silences among students in the process of learning the linguistic 'rules of engagement' at the school, while Moa's comment stands as an example of how some students expressed how the language regimentation could be pushed 'too far'.
In the third and final analytical section, I will show how shame/pride and apologies work to confirm the validity of the language ideologies that underlie the students' activism as I explore the activism as situated within the school's institutional hierarchies. At Bråviken, not only students, but also staff members navigated the discursive regime, and in the following section I give examples of the staff's reactions to the students' activism. The excerpts analyzed next are a photograph of class lists of students posted in the main hallway (Figure 3), and field notes (translated from Swedish) (extract 3) from an encounter with the school principal, who commented on the lists in conversation with the author. The principal pointed out that there were some segments of the lists that were cut out.

Consenting to student 'expertise'
Extract 3: Field notes 'Look here', says the principal and points to a hole in the corner of the lists. 'I've cut this bit out. Guess what I cut out!' I hesitate. My first thought is that it's the names of the teachers that have been cut out by the principal. But their names are still there, so that's not it. 'I don't know,' I reply. 'This is where the numbers of men and women were written,' the principal says. Where the numbers of men and women in each class were written, there is now a hole next to the name of the teacher. Each paper has the same hole. The principal continues: 'We put them up yesterday and I got comments immediately about the fact that not everybody defines themselves as either man or woman.' 'But these categories, they are in your system?' I ask. 'So when you printed the lists they were inserted automatically?' 'Yes, we work a lot with these issues, and still we make mistakes and get comments.' In the school's institutional hierarchy, the principal is at the very top. From this position, the principal comments on the students' language activism and the mistakes made by the school staff as they try to live up to the students' linguistic norms. In an apologetic tone, the principal describes the mistake of posting lists that categorized all students within a binary gender system. The principal describes how this mistake was soon noticed by one of the students at the school, which resulted in a manual erasure of gendered categories from the lists and the linguistic landscape of the school. There is an affective ambivalence in the presentation of the event. On the one hand, the principal is apologetic when providing an account (Buttny 1993) of the mistakes made. On the other hand, the act of telling me about the event-and not without a touch of pride-reveals a desire to portray the staff as knowledgeable and ready to adjust their language use. This ambivalence again points to verbal hygiene's role in processes of producing a common 'we', a speech community sharing moral norms and language ideology; the principal confesses the staff's mistakes and at the same time displays a willingness to improve. Following Sedgwick (2003), I understand shame and pride not as dichotomies, but as entangled in a process of subject formation. The declaration of failure, of not living up to the common linguistic ideal, positions the principal as non-ideal in the local community, yet it simultaneously, and importantly, confirms the principal's identification with that very ideal (Ahmed 2004:108;Foucault 1988).
As mentioned above, some students and teachers made an effort to use en as a generic pronoun in their speech instead of man. In the next excerpt (extract 4), the teacher Erik is talking to a class about how to learn to work together as a group. In this short episode, he uses both the gender-neutral pronoun en and the standard man.

3
Affects of verbal hygiene: the impact of language activism… Extract 4 In speech, en as a generic pronoun was not widely used in the school. However, some students and teachers did try to incorporate it into their everyday speech. When discussing this, they often mentioned the difficulty of doing so. The difficulty can be illustrated by the transcript above, where Erik consistently uses en except for a slip in line 4. Erik is standing in front of a class giving instructions; he is very much on-stage in a Goffmanian sense (Goffman 1959), and in an interview with the author (extract 5) Erik commented on the reaction he receives from the audience of students when he uses en in classroom situations. Before the excerpt below, he had described how he had used en for some time and that it was a choice he made after discussing the issue with his girlfriend. Here, he comments on the reactions from students at the school: Extract 5 Erik is hesitant (there is an extended pause on line 1) as he describes how students react positively to his use of gender-neutral language, and he imitates a tired student who snaps awake in reaction to his use of the pronoun (lines 7-8). He describes students nodding while listening to him speak. In the context of any high school, making students pay attention must be something every teacher strives for. Erik describes how he achieves this by using en instead of man. He believes that this is because the use of en would be a shibboleth indexing a person who has 'reflected' (line 9). He also makes the point that one should not underestimate the 'desire to be heard' and the will to use language that is regarded as legitimate by the students (lines 14-15).
He is here underlining an important aspect of the language regimentation at the school-at this particular social location, some lexical items and linguistic skills, e.g. being able to use the pronoun en in everyday speech, become important symbolic assets for those who want to speak with a voice that is deemed legitimate and worth paying attention to. Mastering the discourse of the politically 'woke' is here described as a prerequisite for having agency and voice, even as a teacher.
One should note the hesitancy with which Erik talks about these issues. In line 1 Erik stops in mid-sentence just as he explains that he 'ha[s] to'. Instead of continuing the sentence and telling the interviewer what it is he has to do, there is an extended pause (3.3 s), and as he continues there is an audible sigh. This hesitancy could be interpreted as an indication that what he is about to say next is not the preferred response to the interview questions. He is in a sense making a confession, a somewhat shameful one, that he too wants to speak with a legitimate voice, and that his use of the pronoun en is a way for him to achieve this. His reluctance to reveal this motivation for his use of en could perhaps be because, by admitting that he changes his way of speaking to adhere to the linguistic norms at the school, he also confirms the power of the students' language activism to rearrange traditional institutional hierarchies: to define a legitimate voice and whether it is worth listening to. However, another analysis is possible if we return to Ahmed's (2004) thoughts on shame as constituting a common 'we', one that recognizes a set of moral principles that we should feel ashamed of being caught breaking. In the case of using en at the school, the common underlying, and often outspoken, motive for changing the use of man to en was a wish to use more inclusive language that did not reproduce the masculine as the norm. This political motivation for changing the use of the pronoun was also expressed by Erik in the interview. In the excerpt, however, he confesses that he has other reasons, perhaps less noble in the eyes of the language activists at the school. Instead of describing, as was more often the case, en as an important tool to deploy as a progressive language user to create an inclusive social space, Erik describes how he can use en to get students to pay attention to him in class. The fact that Erik describes his actions as being a kind of political posturing is a more convincing explanation of his hesitation before giving his account in line 1. He pauses before temporarily placing himself outside of the common 'we' of language activists at the school by admitting to other motives than the shared vision of an inclusive language.
The excerpts above show the principal's and Erik's efforts to navigate the discursive regime conjured by the students' on-going practices of verbal hygiene that regimented language and normative existence (Ahmed 2004) at the school. These snapshots exemplify how the language activism also played into student-teacher relations. The teachers had to become sensitized to the linguistic order produced at the school through students' language regimentation in order to maintain a voice that was regarded as legitimate.

3
Affects of verbal hygiene: the impact of language activism…

Conclusion
Correcting other people´s way of speaking is often a sensitive matter. Language activism tends to generate strong emotions, and this paper has addressed the affective facets of students' acts of verbal hygiene at a Stockholm high school. Following Sedgwick (2003), Ahmed (2004) and Wetherell (2012), I have turned my attention to the performative force of affect and argued that this move is necessary to understand the potency and the multifunctionality of the activism as it is played out in social context. Further, my ethnographic methods have allowed me to study how this micro-level language policy is taken up and reacted to by staff and other students. These acts of verbal hygiene gain their performative force partly from the workings of shame and shaming. I have shown how the activism plays into institutional hierarchies, and I have discussed the ambivalences and silences it produces. As we have seen, processes of verbal hygiene have the potential to create a scene where the linguistic stakes are high and the ability to speak with a legitimate voice is dependent on the mastery of specific lexical items. Simultaneously, through the performative function of affect, the language activism also produces delineations between social groups and creates anxieties about social exclusion and failure to live up to the social ideal.
With this paper, I have tried to do justice to the complexity of the practices under consideration by refraining from either a celebratory description of youths' self-empowerment or a paranoid reading (Sedgwick 2003) of the disciplining of speaking subjects. Such analyses would not be uncalled for, but I believe that it is by highlighting the many functions of language activism that we can come closer to understanding why it has become such a vexing question and so anathema to its adversaries. If we as sociolinguists want to partake in the current discussion on language activism that is unfolding outside of academia, I believe we should utilize our methodologies to add nuance to a polarized debate. We can show how language activism is always situated in, and its effects are contingent on, the context where it is played out. Further, we can show that the emancipatory potential of such language activism does not exclude its potential to create new normativities, hierarchies, and possible exclusions. As Savc (2016) notes in a discussion on queer activism: 'transnational circulation of knowledge on gender and sexuality' is 'emancipatory for many, but it also creates its own disciplinary mechanisms and regimes of truth.'. It is important to attend to such disciplinary mechanisms, not necessarily to criticize the mobilization of shame or practices of calling out, but to acknowledge what Cameron (1995: 163) noted when discussing verbal hygiene: 'there is no language without normativity'. As such, even the most liberative project carries with it new normativities.