Shrinking or expanding access to civic space? The consequences of hate speech, threats and harassment within Swedish civil society

ABSTRACT Previous research has focused on how governments and public authorities can limit civil society organisations’ (CSOs) autonomy by shrinking their operational civic space and their democratic functions. There has been less attention paid to how other external actors and types of pressure might restrict access to civic space and how it can induce coping responses on both an individual and organizational level. This article draws on an interview study with representatives, employees and volunteers in Swedish CSOs and focuses on their strategies on coping with hate speech, threats and harassment or the perceived risk thereof and explores if and how the fear of being subjected to hate speech, threats and harassment restricts civil society actors in their autonomy and affects their democratic functions. Results show that there is a tendency towards withdrawal from public debates by making themselves more anonymous and less accessible, by depoliticizing contentious issues and by disrupting the chain of representation. These coping responses are problematic, as they limit civil society’s ability to act as an independent and critical voice, but the personal costs for being visible in the public sphere are often considered too high.


Introduction
The fact that civil society actors are subjected to hate speech, threats and harassment is a widely-acknowledge problem worldwide.In Sweden, the tone of the debate has become increasingly polarized and harsh, especially on social media (Johansson & Scaramuzzino, 2023;Scaramuzzino & Scaramuzzino, 2021).Attacks against both associations (Swedish Agency for Youth and Civil Society, 2021) and civil society leaders have increased in recent years and several chairpersons have chosen to resign after being subjected to hate speech, threats and harassment (Scaramuzzino & Scaramuzzino, 2021).Resigning because of such situation can be interpreted as an 'exit-strategy' that is used when CONTACT Gabriella Scaramuzzino gabriella.scaramuzzino@soch.lu.seSchool of Social Work, Lund University, Box 23, 22100 Lund, Sweden other strategies no longer are an option (Hirschman, 1970).However, less is known about what other strategies civil society actors use to cope with being subjected to hate speech, threats and harassment and its potential consequences for the individual, organization and society.
A strong, vivid and independent civil society is of great importance for deepening democracy (Warren, 2011) and civil society has at least three important roles in democracy.First, it acts as a 'critical voice' by, for example, influencing and forming public opinion and advocating for different interests.Second, it exists as a 'school of democracy' through fostering participation in democratically-structured organizations.Such participation becomes a practical experience and training in democratic decision-making, meeting techniques and respect for different views.Third, it functions as a 'representative' for both individuals and groups vis-à-vis society.This is done by creating chains of representation from local, regional and national levels (Swedish Government, 2016Government, , 2018)).The relevance of the threat to democracy posed by hate speech, threats and harassment has led the Swedish Government to launch a public investigation on how to strengthen legislation to protect people who have an important societal and democratic role (Swedish Government, 2020).The aim of this article is twofold.First, to explore what coping strategies individuals and organizations use to handle hate speech, threats and harassment.Second, to understand what consequences hate speech, threats and harassment have on Swedish civil society and how it potentially restricts CSO autonomy, access to civic space and affects their democratic functions.
Hate speech, threats and harassment is a broad concept as it includes everything from derogatory comments to physical violence (Scaramuzzino, 2020).This article draws on an interview study with representatives, employees and volunteers in Swedish CSOs.Interviewees had been subjected to, or had seen fellow civil society actors being subjected to, degrading comments, slander, damaged property, sexual harassment, gun threats and homicide.Some of the CSOs included in the sample, also fought against different types of 'structural violence' and 'slow violence' that are more indirect and result in discrimination and inequalities (see e.g., Nixon, 2011;O'Lear, 2021).However, this article will not distinguish between different types of violence because the focus is not on the violence per se, but rather the strategies used in coping with the fear of being subjected to hate speech, threats and harassment and the potential consequences it has for the individual, organization and society.When analyzing potential consequences, it can be problematic to create a hierarchy between different types of violence because degrading comments do not automatically have a less negative impact on victims than physical attacks.In fact, watching others being subjected to hate speech, threats and harassment can also potentially have severe consequences for the individual, organization and society (Scaramuzzino, 2020).This article contributes to the emerging literature by highlighting some different experiences that may have relevance for other civil society actors as well, in their attempts to uphold, defend, or expand their access to civic space.

Autonomy, Access to Civic Space and Individual and Coordinated Coping Responses
There are three main theoretical concepts used in this article, which are also related to each other: (1) autonomy, (2) access to civic space, (3) individual and coordinated coping responses.Scholarly debates on both organizational autonomy and shrinking civic space have tended to understand them as actual, rather than perceived.This article acknowledges that perceived autonomy is intertwined with perceived access to civic space.A central theme in civil society research is the conflict between external control/pressure and organizational autonomy, and these are often presented as dichotomies.However, Arvidson and Linde (2020, p. 210) argue that what is absent in these discussions is that there is always room for discretion and: ' … it refers to both freedom and the controlling boundaries surrounding the space where freedom can be exercised'.Hence, external control/pressure and autonomy are closely linked, and hate speech, threats, and harassment can be understood as an external control/pressure against civil society actors that can affect their autonomy (Scaramuzzino, 2020), but at the same time also their perceived access to civic space.To handle being subjected to hate speech, threats and harassment, individual and coordinated coping responses can be used to uphold, defend, or expand access to civic space.

Autonomy and Access to Civic Space
To be able to understand what consequences hate speech, threats and harassment have to Swedish civil society it is important to address civil society's societal role and democratic functions.According to Alexis de Tocqueville, civil society is an autonomous area which has a special role in fostering democratic citizenship.Civil society is distinct from, and an important counterweight to, both the state and the market (Trägårdh, 2007).In this view, it is problematic when the boundaries between civil society groups, the state, and the market become blurred, and much has been written about the risks of CSOs being coopted by, for example, the government or public authorities; for instance by being dependent on government funding or involved in policy processes (Nayam, 2000).In this view, associations are of great importance for democracy and associational freedom is often emphasized as one of the most important elements in a democratic society.In fact, the central role that civil society actors have in democracy is emphasized both in research and in politics (e.g., Amnå, 2006;Arvidson et al., 2018a;The Swedish Government, 2009;Wijkström & Lundström, 2002).
While studies of professional groups have mostly discussed autonomy as professional discretion exercised at individual and collective level (see e.g., Scaramuzzino, 2020), civil society studies have mostly focused on an organizations' autonomy, in terms of its opportunities for strategic action and service delivery without incurring sanctions related to public authorities (cf.Egdell & Dutton, 2017;Milbourne & Cushman, 2015).A Swedish study shows how privatization and contracting out services to CSOs can result in compromising their function of being a critical voice, as some CSOs choose to hold back their criticism towards the state (Arvidson et al., 2018b) and loosing autonomy.Concerning autonomy, three different types of autonomy can be distinguished (1) individual, (2) organizational and (3) collective autonomy (cf.Scaramuzzino, 2020).While the two first types of autonomy relate to a single actor's or the organization's capacity to uphold its mission without external coercion, the collective dimension relates to the civil society sector or sphere as a whole and its function in society.In the debate about the autonomy of civil society, it is often the critical voice function that is assumed to be threatened.The negative link between loss of autonomy and weakened democracy is mostly taken for granted, but when it concerns hate speech, threats and harassment other functions might be affected as well, such as the capacity of being a 'school of democracy' and building 'chains of representation'.When CSOs, their representatives and members are subjected to hate speech, threats and harassment because they are pursuing their core mission, their autonomy might be at risk.
Previous research has often focused on 'shrinking civic space' or 'closing operational space' for civil society; i.e., CSO's ability to perform their key tasks and core mission.Even if it is not always explicit in the literature, civic space is clearly intertwined with autonomy.By responding to different types of pressure, CSOs can use their discretion to try to both claim and reshape their operational space (van der Borgh & Terwindt, 2012).Scholars have focused on how governments and public authorities use different 'anti-NGO measures' and show how in recent years 'anti-NGO measures' have become part of a broader global trend by restricting the ability of CSOs to organize, advocate and act (Berger-Kern et al., 2021;Buyse, 2018;Ploszka, 2020;Toepler et al., 2020;Tripp, 2019).Research shows how governments have disrupted international funding, introduced legal restrictions and censorship (Chaney et al., 2020;Toepler et al., 2020), and failed to protect certain civil society actors from physical attacks (Chaney & Sahoo, 2020;Ploszka, 2020;van der Borgh & Terwindt, 2012).It is well known that these types of 'anti-NGO 1 measures' have been used in authoritarian and hybrid regimes (Toepler et al., 2020), but they have been used in EU Member States, such as Poland and Hungary (see e.g., Ploszka, 2020).The European Parliament has developed a response against shrinking space for civil society across the world and the issue is described as the heart of EU foreign policy.This means that the development of shrinking civic space is fought with political measures on a supra-national level (Youngs & Echagüe, 2017) but also by CSOs themselves at national and local levels.Where there is an exercise of power there is often also resistance.Studies show that there is strong resistance against anti-NGO measures, and examples of civil society actors having defended or expanded their access to civic space (see e.g., Berger-Kern et al., 2021 for Kenya and Kyrgyzstan) by pushing back against government pressures (see e.g., Vértes et al., 2021 for Lebanon).
However, in a Swedish context, shrinking operational space for civil society has not primarily been understood as coming from pressure from the government or public authorities.Rather, government and public authorities are often seen as guardians of civic space when civil society actors have been subjected to hate speech, threats and harassment by other actors (Scaramuzzino & Scaramuzzino, 2021).Hate speech, threats and harassment against and within civil society is a phenomenon that has been largely recognized by scholars (see e.g., Carmichael & Karamouzian, 2014;Chaney & Sahoo, 2020;Scaramuzzino & Scaramuzzino, 2021), but much of the literature on Swedish civil society actors are reports or maps made by either CSOs themselves or by different authorities.In a survey study, one in ten associations stated that they had been subjected to hate speech and threats during the previous year (Swedish Agency for Youth and Civil Society, 2021) and one in three civil society leaders had been victimized at some point (Scaramuzzino & Scaramuzzino, 2021).Literature indicates that the fear of being victimized tends to prevent Swedish CSOs from carrying out their democratic roles (Larsson & Stjernholm, 2014;Nilsson, 2020;Swedish Agency for Youth and Civil Society, 2021).A mapping of the Swedish youth movement shows how their member organizations have started to 'turn inwards' (Nilsson, 2020).According to the literature, this development does not just concern advocacy oriented CSOs, because hate speech, threats and harassment have also been used to influence and change decisions within the Swedish sports movement (i.e., Hurve & Jonsson, 2018;Nilsson et al., 2015;Swedish Government, 2013;Swedish Sports Confederation, 2019).As in other Nordic countries, sport clubs constitute a significant part of Swedish civil society (Blomqvist Mickelsson, 2022).In recent years, the Swedish sports movement has raised attention to 'settled matches' and 'betting' as an increasing phenomenon where organized crime members threaten, harass and bribe players or referees to manipulate the game and the results (Nilsson et al., 2015;Swedish Sports Confederation, 2019).
In the literature, 'civic space' is often described as something 'actual', instead of 'perceived'.Rather than using concepts such as (actual) 'operational space' or 'shrinking civic space', I will use the concept 'access to civic space' to emphasize both the perceived and the relational aspects of this phenomena.If one civil society actor threatens another civil society actor, one might expand their access to civic space, while at the same time restricting access to civic space for the other.How civil society perceives discretion might also differ greatly, which is why the notion of civic space as something actual must be problematized.

Individual and Coordinated Coping Responses
If there is pressure against civil society actors, it often requires some kind of response.Civil society research has focused much on civil society actors 'strategic action' or 'counter strategies', but the use of theories on 'coping strategies' or 'response strategies' are rare (see e.g., van der Borgh & Terwindt, 2012, pp.1072-1073 for an exception), and is, therefore, a perspective to which this article wishes to contribute.In research on violence and victimization coping strategies are frequently used to understand how people cope with victimization in different ways.Coping often refers to behaviour that protects individuals from being psychologically harmed by difficult experiences, situations, dilemmas and tensions that can occur in relationships.These theories have often been developed within a psychological tradition and tend to focus on individuals (however see Rodríguez et al., 2019 also on collective coping and organizational stress).Within civil society research, there have instead been much focus on 'collective strategic action', i.e., on an organizational level.van der Borgh and Terwindt (2012, p. 1072) argue that it is important to distinguish between individual and coordinated responses.When it comes to analyzing civil society actors, the line between the individual and the organization often becomes blurred, which is why it can be valuable to use theories on coping strategies while attempting to distinguish between individual and coordinated coping responses.
When it comes to coping strategies, scholars have both examined the 'approach'; in other words their direct attempts to address the problem (i.e., seeking support, problem solving) and 'avoidance'; i.e., by managing both the emotional and cognitive reactions resulting from being victimized (see e.g., Cooley et al., 2022).Scholars have also distinguished between 'aggressive coping strategies', for example, by getting through or fighting back, and 'passive coping strategies' (Ricciardelli, 2014) such as withdrawing, self-blaming and self-silencing.Furthermore, cognitive distancing or humour can be used to minimize the problem (see Cooley et al., 2022).In research on cyberbullying victimization, specific 'technical coping strategies' such as blocking a user on social media, reporting instances of abuse to the moderator or social media platform, adjusting privacy settings, avoiding using social media and deleting social media accounts, have all been highlighted as important (Byrne, 2021).In this article I have chosen to call these strategies 'social media responses'.

Method and Empirical Material
This article is based on a qualitative study.Between March 2020 and June 2021, 12 semistructured telephone and video interviews were conducted with elected representatives (chairman and board members), paid employees (managers, project leaders) and volunteers in different Swedish CSOs.One of the interviewees was a representative for a 'security training programme' for Swedish CSOs 2 who had great knowledge on Swedish CSO experiences of and strategies handling hate speech, threats and harassment.Because of the sensitive nature of the topic, and to protect interviewees' anonymity, I omit any specific information about the CSOs that have been included in the sample.
Two sampling methods were used to get in touch with interviewees: goal-oriented sampling and snowball sampling.As a first step, networks of various CSOs and umbrella organizations were contacted and asked to send out a request to their member organizations to participate in the study.Elected representatives, employees and volunteers over the age of 18 could contact the research project if they wanted to participate.When contact was established, snowball sampling was used to get in contact with more potential interviewees.Therefore, the second step was more targeted to ensure that many different types of CSOs were represented in the sample: e.g., adult education, humanitarian, trade unions, sports associations, disability rights, sexuality and reproductive rights as well as refugees and undocumented migrants.However, most of them had been or were active in several CSOs and social movements and had held different positions.The interviewees were mostly women (8 women and 4 men), in different ages, from about 20 to 80 years old.Most of them were active in local branches in different parts of the country.
The interviews lasted between 45 min and an hour and interviewees were asked questions about if they had been victimized, and, if so, what types of hate speech, threats and harassment they as individuals and their association/s had been exposed to, who had exposed them to threats, what strategies they used to cope with these situations, and if and how the fear of getting subjected to hate speech, threats and harassment had affected them.There were also questions about the role of social media in these processes.All interviews were recorded and transcribed for further analysis.A thematic analysis (Nowell et al., 2017) was conducted by first reading transcribed interviews several times to check for recurring themes such as 'more anonymous', 'less accessible', 'withdrawal from public debates', 'depoliticization' and 'engagement'.The analysis was then structured around (1) strategies, and (2) consequences.In the first part of the analysis, analytical concepts were used to analyze different types of 'individual responses' and 'coordinated responses'.The second part focused on three democratic functions of civil society: 'critical voice', 'chain of representation' and 'school of democracy'.The analytical concepts 'autonomy' and 'access to civic space' were used to understand different implications.The last step in the analysis was to explore the relationship between the different strategies used and their consequences, which have been summarized in Table 1 in the concluding part of the article.

The Tension Between Shrinking and Expanding Access to Civic Space
There was a constant tension between using strategies that potentially lead to shrinking access to civic space, and using strategies to expand access to civic space.In fact, one of the most common coping strategies that the interviewees discussed and employed was becoming less visible and less accessible, i.e., 'avoidance'.An interviewee explained how Swedish society had become increasingly polarized, with harsher and more hateful public debates, and it had become more common to encounter open racism.Therefore, their CSO had become more careful to not wear symbols in public spaces, as a way of protecting themselves: Our material is packed in large bags, and today it is rarer that we use bags where there are religious symbols on the outside of the bag.We choose bags that are neutral, where it is just a big bag.We absolutely do.We have talked about; shall we cancel this?No, we should not.It is a conscious decision.But I think that we choose to be more anonymous when we move in the public space (R4).
Hence, as an organization they had made a strategic choice to be more anonymous in public spaces.Another interviewee said that their CSO used to hold open lectures in public places, but: 'We stopped organising them in that way, and, for example, started to demand that you were not welcome if you did not register, and we checked membership at the door' (R2).It was sometimes difficult for interviewees to determine if hate speech, threats and harassment were directed toward them personally, or toward the organization, or both.The blurred lines between the individual and the organization also becomes visible in their coping responses.The theories on coping strategies often focus on an individual's strategies, but in the interviews coping strategies were often described as joint decisions, and were therefore also an 'organisational or collective coping strategy of avoidance'.
Online hate speech was often described as difficult to handle, and in the interviews, there were examples of several 'social media coping strategies', but these can be understood as a subcategory of 'avoidance'.An interviewee explained how their organization had made the strategic choice to neither use social media nor be visible in the media, because of the risk of being subjected to hate speech, threats and harassment: 'We have made a joint decision within our local group to be restrained and do not act on social media, so we have not been subjected to attacks on social media' (R7).Another interviewee said: We have become more closed.During that period and since then, we have, for example, fewer telephone numbers and names on our website, which may make the organisation perceived to be more anonymous.The organisation may not appear as human and cordial as we want it to appear.I know that some (of the staff and members) have wanted to delete their pictures from the website.Many have deleted their names and numbers and sometimes even their e-mail addresses as well and use more general e-mail addresses instead of e-mail addresses with their own name (R2).
The interviewee, who was in charge of security training for Swedish CSOs, presented the issue in the following way: I think that as an organisation, you have to think preventively about how to handle social media.(…) I think you can have strategies and tactics to avoid being exposed (to hate speech, threats, and harassment) on social media, because it is so extremely easy to sit behind an anonymous screen and post hate and threats (…) I think that you have to work more with prevention.For example, that you do not have comment fields.You might create groups for just the members, where you can have your dialogues.Then you have another for the ones outside (R5).
Several of the CSOs had considered and often employed these kinds of 'avoidance' and 'social media coping strategies' and had enforced stronger divisions between member communications and communication with the public.An interviewee, for example, explained that they only shared information about their workshops and lectures on their intranet or private Facebook page for members only: 'we have to make it a little more boring and to make it a little less personal.To be honest, we do not write about our events and how should they then be able to be interested in what we do and our organisation?'(R2).This means that such measures clearly came with a cost for CSOs, which show that the civil society actors were faced with a dilemma: be visible and assume the cost of being subjected to hate speech, threats and harassment or become less visible and assume the cost of not get their message out, and perhaps also lose the chance to recruit new members (see e.g., Johansson & Scaramuzzino, 2019).As representatives are often the ones that carry the organizations' messages, they might be subjected to hate speech, threats and harassment at a personal level because of their engagement in the organization.Sometimes the personal and the organizational costs can therefore become intertwined: After the recent threat that I was subjected to via Facebook.It was first a lot of sexual harassment and then after that I should die.After that incident, I deleted all the information about my children and their activities (…) This is a direct consequence of that I am subjected to threats.I delete things (…) I am not afraid for my own sake, but I am worried that I might, because of my engagement, be exposing my children to danger (…) I choose not to give out [information about] my private life.I choose to be anonymous, and that is sad (R8).Some of the interviewees also said that they had chosen to become anonymous and not share any personal information on social media, simply because the personal costs of being visible was too high.This was especially true for interviewees with families and children living at home.Based on the interviews, it seems like it was easier to talk about organizational than personal costs, especially for civil society leaders: 'I got support that time when I got threatened by telephone, but the fact that I think about if I want to resign as a leader, that is perhaps nothing you share with anyone, because it would affect quite a lot of people' (R2).The interviewee did not want to scare the staff, board members, or volunteers.Another interviewee said: 'When I got information about it (this study), I thought that I never got this question, how interesting!I have not wanted to tell anyone about it.It is so hard to know, should you really worry about it?'(R3).
There were also examples of 'passive coping strategies' i.e., situations of withdrawing from public debates, self-silencing and self-censorship.An interviewee who worked with refugee's rights issues said: 'I have become much more restrictive when it comes to being visible in media when it comes to these types of questions because I know that there are people who do not like it' (R7).Some of the interviewees said that they had hesitated or had not written or published a statement because of the fear of being subjected to hate speech, threats and harassment: 'Yes, I would definitely say that it has happened' (R1).To withdraw from public debates and the depoliticization of certain contentious issues was in several cases described as a joint decision within the association: 'We have spoken about it, and we have decided that we cannot be totally open' (R8).When it came to questions like honour-based violence the interviewee said: 'We are so afraid of being called racists we do not dare to name specific countries' (R8).According to the interviewee, there had been examples of individuals who had been critical about Islam publicly and they had been subjected to threats multiple times.This was why some of the associations that the interviewees represented had made the joint decision not to be as critical as they wished to be: 'I do not want to be subjected to more threats than I already am' (R8).The interviewee said: 'Should I write?Should I write this?No -I think I'll pass' (R8).They did not only use self-silencing and self-censorship responses on social media and in media.The interviewee explained that they were also careful with what they said during lectures and at Almedalen: 3 'I always choose very carefully what I say, and I never say fully what I think because I receive threats (…) I do not feel that you are totally free to say what you want' (R8).This interviewee was afraid that they would lose both opportunities to be invited to collaborations and to get state funding it they were too outspoken regarding honour-based violence.There is a risk that the particular coping responses can contribute to shrinking access to civic space for civil society actors.
However, what seems to be important in deciding what coping responses to use was the nature of the issue and to what extent it was close to the core mission and identity of the organization.An interviewee said: 'What I spontaneously think of, is that lately our organisation has become a little more cautious when we express ourselves' (R2) and 'There are other issues like pornography and prostitution, where we as an organizsation have been pretty cautious' (R2).Another interviewee explained on one occasion, when they had received threats, that there were different opinions as to whether or not they should write an article about it, because some of the members did not want to further aggravate the situation.In the end, they decided not to write the article because it was not related to their core issues and mission: ' … for us, it is most important that we express our opinions and what we are fighting for' (R1).The interviewee further explained: 'But in general, I would say that we have the attitude that we do not have any intention to let themwhat kind of organization are we if we let these forces silence us? (R1).This can be interpreted as this civil society actor using both 'passive coping strategies' and 'resistance coping strategies', depending on whether it was a core issue.If it was about one of their core issues they were often prepared to 'fight' back.
In fact, there were several examples of 'resistance coping strategies' as well.An interviewee, for example, said that after being subjected to hate speech, threats, or harassment s/he personally had been more determined to fight it: 'For some reason, I have been triggered by it."You fools, now I will … " (laughs)' (R3).But resistance coping strategies were described as mostly used to cope with and to fight 'structural violence' and 'slow violence' (see e.g., Nixon, 2011;O'Lear, 2021), for example experiencing that the public space is both hostile and discriminating against people with physical disabilities.An interviewee had organized a collective wheelchair event, to support each other; but also to be more visible in public spaces: I have organised wheelchair groups in town.I invented it (laughs).Guys and girls who went with me out on the town and we watched how to get around and ordered stuff and so on.Much to say, that this is not so strange, I can fix this (R11).
Another frequently-used coping strategy was to 'approach' the problem by collaborating with other actors to fight hate speech, threats and harassment together.It was also common to develop and use preventive strategies before an event or a protest: Before each event we always go through the safety routines because we know we must have safety routines.We have to have a plan in case something should happen, and that's one of the most important things.Before each event we have these numbers you call if something happens, you gather at this place, think about these things, etc. (R1).
These types of coping responses can be interpreted as strategies to defend or even expand access to the civic sphere for civil society actors.

Autonomy and Democratic Functions Under Threat?
The fear of being subjected to hate speech, threats and harassment, and the chosen coping responses, had consequences for CSO's democratic functions as a 'critical voice', 'chain of representation; and as 'a school of democracy'.

Critical Voice
One of the interviewees reflected on their CSO's role and described civil society as a 'salt' and something with a 'sting': Our organisation should be a salt.We should be uncomfortable (…) Maybe not uncomfortable (laughs) but we have to show what is unrighteous in the world and when we do that, we get crap.Thus, it lies in the nature of the organisation itself (R3).
According to this interviewee, the role of society is to be a critical voice and to dare to say and to advocate for issues that are sensitive and against general public opinion.This role of civil society potentially makes them more exposed to hate speech, threats and harassment.The same interviewee argued that if they are subjected to a lot of hate speech, threats, or harassment, it can be seen as proof that they have actually fulfilled their role as a CSO and managed to be a 'salt'.In Sweden, there is a proverb called 'lagom' which means that the right amount is best, or in moderation, but according to the interviewee, civil society should not be or do something 'lagom', rather, they should be 'extraordinary': You do what no one else has dared or done before.You are an innovator.You tighten the bow a little more.If you act like an ordinary business manager in the municipalities, then the civil society sector loses its character (R3).
From this point of view, it is clear that hate speech, threats and harassment can threaten the role of civil society as a critical voice if the people representing the organizations choose to compromise their alleged responsibility to speak up on sensitive issues.The interviewee who was in charge of security training for Swedish CSOs described how this is not only a problem for individuals or CSOs, but a serious concern for democracy: More information is needed about what is happening to both individuals and organisations.We have to talk about what happens to society when an organisation like the Swedish Federation for Lesbian, Gay, Bisexual, Transgender, Queer and Intersex Rights (RFSL) does not dare to participate in Almedalen because they are afraid of the Nordic Resistance Movement (NMR), or whatever organisation it is.Something is happening with democracy which is not positive.We have to talk about it and debate it and then decision-makers have to do something about it.But I think it is important that there are accurate statistics, otherwise there will be someone who says "But look here, there have only been ten such crimes reported this year in this region.It cannot be a huge problem".Then it is not enough to say that we actually think it is a hundred (R5).
As the quote shows, movements such as the Nordic Resistance, a right-wing extremist movement, has become much more visible in the public space making other movements feel less safe.However, from a theoretical perspective, this is not necessarily an example of shrinking civic space per se, but it is certainly an example of shrinking access to civic space for well-established Swedish civil society actors such as RFSL whose main goal issue is to expand LGBTQ + rights and their access to civic space.As the quote shows, incidents are seldom reported to the police, which was also confirmed in other interviews.Incidents were often only reported if someone had damaged the property or if it was considered a severe attack: 'We have always made a police report when there have been direct threats to the office, but as far as I know not when it comes to any messages or comments on internet' (R1).What was put forward as the most important challenge for civil society and democracy according to an interviewee was: 'to protect elected representatives from harassment and threats and mark much more clearly on the penalty scale' (R6).Even though the interviewee was not someone who generally advocated for prison or legal punishment, when people are threatened to silence s/he argued that it should be taken much more seriously: 'Because the goal is to (silence) and people do not have the strength to be representatives and resign' (R6).Hence, hate speech, threats and harassment also affect the chain of representation.

Chain of Representation
An interviewee said that there is a decreasing trend of voices from members and local branches, because of the risk of being victimized: 'It often goes through the communication department and there are perhaps fewer statements from local branches and more from central' (R2).Communicators have become important actors for deciding to whom, how, what and when to communicate.The interviewee further explained: 'Yes, I know that internally, at the federal level, they are very careful about who speaks on a bunch of different issues, based on who used to be subjected to hate campaigns' (R2).Many civil society leaders on a national level have developed coping strategies and often have a communication department that can back them up if and when they are attacked.The empirical material indicates that CSOs who lack the help of communicators, are not as represented in public debates as larger CSOs.Hate speech, threats and harassment might therefore strengthen a dynamic in which the costs of communicating a message require organizations to hire staff that can handle hate speech, threats and harassment.In continuation, it would lead to only the more professionalized organizations' voices being heard.
In the interviewees, women with a leading position were described as particularly subjected to hate speech, threats and harassment (see also Scaramuzzino & Scaramuzzino, 2021).For example: ' … it is something that worries me a lot, and it is threat to women in leading positions' (R3).This might potentially lead to fewer women taking leading positions CSOs because the personal costs are too high.To witness other women being subjected to hate campaigns and reprisals was described as dissuasive.An interviewee, for example, said that she had decided not to be a leading figure anymore, because she had seen what had happened to other females in leading positions advocating for the same issues.It is clear that, if certain groups of civil society leaders are more subjected to hate speech, threats and harassment than others, the representational function of civil society organizations might be compromised when these more vulnerable groups withdraw from representational roles or from being visible in public debate.This can be interpreted as more limited access to the civic sphere for women than for men, when it comes to leading positions in civil society, even in a country with comparatively high gender equality, like Sweden.

School of Democracy
Some threats to autonomy were directed towards another function of civil society, namely as school of democracy.Two of the interviewees, for instance, discussed how match fixing (i.e., when the outcome of a match has been manipulated to make money on betting) has recently become a growing problem in Sweden, and can potentially intimidate players, coaches, referees, club executives from participating in sports: We have to work to prevent this, if we want to have any sports left that are honest, and people who dare to be referees.It is about such a large amount of money and organised crimes, so we have to work with it and take it seriously' (R9).
There were situations where both players and referees had been bribed and then threatened to manipulate the game or the results.The interviewees said that they did not have any statistics to rely on, but their interpretation was that match fixing was mostly connected to men's sports: 'Of course it exists, but I have not heard about any girls or women who have threatened and who tried to manipulate the game.In this case it is criminal men who try to go in and manipulate (R10)'.
Interviewees were also asked if they themselves ever had considered quitting participating in civil society due to the fear of being subjected to hate speech, threats and harassment, and they had different experiences.Some directly said 'no' (R4), 'no, it is strange because it has been tough.It has' (R3) and 'I become a little bit affected, and sad (…) but you have to move on.Otherwise, you cannot work with the thing I do' (R8).Another interviewee also said 'no' but that s/he 'in some situations have felt discomfort' (R7).Other interviewees answered the opposite: 'Yes I have.Partly after I got threatened myself, but also during certain periods when I experienced an increased rise of the Sweden Democrats, 4 that is to say, their political mandates' (R2).Another example was after an interviewee had been subjected to threats by a man who acted aggressively when they were in the city centre: After that, I probably sat and thought like, 'Wow, is this how it is going to be?' Then there were the thoughtsis this something I should continue to do?I ended up saying: 'Yes, I will, because I received such good support from the organisation (…) since then, I have not had any thoughts of leaving because of the hatred.It is very reassuring to know that there is such strong support and it feels a bit like this: "Anything can happen to me because I have my organisation behind me" (R1).
Both collective actions and to feel supported were described as fostering engagement and hence facilitating the capacity of being a 'school of democracy'.In Sweden, there has been much focus on how civil society's function as a critical voice is compromised, but as the analysis shows, other democratic functions are also affected, as well as individual, organizational and collective autonomy.

Conclusions and Discussion
This article has some methodological limitations as it only includes a small number of interviewees, and therefore cannot fully grasp civil society actors' strategies for coping with hate speech, threats and harassment, as well as how it restricts their autonomy and affects their democratic functions.Civil society is a diverse societal sphere with a myriad of different actors with different missions and interests.However, this qualitative study is exploratory and contributes with highlighting some different experiences, in a national context that seldom has been associated with shrinking civic space.Within the literature on shrinking civic space there is a strong focus on how governments and public authorities put different kinds of pressure on civil society actors and restrict their autonomy, discretion, and limit their access to civic space (see e.g., Buyse, 2018;Ploszka, 2020;Toepler et al., 2020;Tripp, 2019;Vértes et al., 2021).This strong focus on government and public authorities is also present in several policy documents on shrinking civic space (see e.g., Youngs & Echagüe, 2017 on the EU response).There is a risk of overlooking the fact that access to civic space can be challenged in different ways and by other actors, and still have similar outcomes.The findings in this article clearly demonstrate how the fear of getting exposed to hate speech, threats and harassment creates different coping responses regardless of who the perpetrators or harassers are.It is therefore important to expand the scope and investigate in what different ways civil society actors are restricted and by whom.This article contributes a relational perspective on 'shrinking civic space' and the findings regarding coping responses and potential consequences used are summarized in Table 1.
This article understands hate speech, threats and harassment to be a form of pressure against civil society actors and has explored different related coping responses employed by these actors.Similar to the Swedish youth movements' mapping of hate speech and threats against their member organizations (Nilsson, 2020), these results show that there is a tendency toward turning inwards due to perceived risk.'Avoidance' and 'passive' coping responses were common, which made CSOs more anonymous, less visible, less accessible and the issues they advocated for less contentious.The often-coordinated response of being more anonymous and less visible in public debates became a difficult dilemma, because to be able to attract public attention and foster engagement they had to be visible and personal on social media and in media (cf.Guo & Saxton, 2020;Johansson & Scaramuzzino, 2019, 2021;Webster, 2014), but the personal costs were often considered too high.The tendency of withdrawing from civic space is worrisome, not only for the individual or the organization, but also from a democratic point of

Speaking collectively (not alone)
. To be a 'salt' i.e., being contentious .Being more visible in the public sphere, but being visible together . Defending or expanding access to civic space .Defending and expanding autonomy

Approach responses
. Report incidents to the police when there are severe attacks .Report to the police when property is damaged .Develop preventive strategies before events . Report incidents to the police when there are severe attacks .Report to the police when property is damaged .Develop preventive strategies before events .Collaborate with other actors .Defending or expanding access to civic space .Defending or expanding autonomy view.However, there are several examples of using 'resistance responses', by collectively speaking up, fighting hate speech, threats and harassment, and by collectively being visible in the public sphere to defend and even expand access to civic space.Findings indicate that, within Swedish society, there is constant tension between using strategies that potentially lead simultaneously to shrinking and to expanding access to civic space.It is important to help Swedish CSOs acknowledge these potential implications and create conditions to build up resilience against hate speech, threats and harassment.Some of the coping responses used by civil society actors tend to collide with CSO's core missions and their expected democratic roles in society and therefore their 'collective autonomy'.For instance, withdrawing from the public sphere collides with the core principle of civil society being an independent and critical voice.The analysis also shows that the fear of being subjected to hate speech, threats and harassment has consequences for the role it plays as a school of democracy and in upholding a chain of representation.In fact, very few of the interviewees were willing to pay the personal costs of being a leading figure.In some of the interviewee's experience, voices from members and local branches seem to be decreasing.Women with a leading position were also described as particularly subject to hate speech, threats and harassment (see also Scaramuzzino & Scaramuzzino, 2021).Accordingly, this could potentially lead to only the more professionalized organizations' voices being heard and perhaps lead to fewer voices from women, i.e., less access to civic space for women.There were, however, other areas, like in the case of sports and match fixing were men seemed to be more targeted.For further research it would be interesting to take a gendered or intersectional perspective on shrinking civic space in Swedish civil society.

Notes
1. Non-governmental organisation.2. In Sweden there were special government grants for extra security-enhancing measures for CSOs working on promoting human rights and safeguarding democracy and whose activities are affected by the fear of violence, threats or harassment due to skin colour, national or ethnic origin, religious beliefs, sexual orientation, transgender identity, or any other similar circumstance.These type of CSOs could apply for financial support to improve their security, as well as attend special training (lectures and workshops) free of charge.The training was a collaboration between the Legal, Financial and Administrative Services Agency, the Swedish Agency for Support for Faith Communities, the Anti-Theft Association, the Fire and Rescue Service and the Police, who were invited as lecturers (Swedish Government, 2018).The present study was supposed to include observations of this security training as well, but because of the COVID-19 pandemic, it had to be cancelled.3. The Almedalen Week is a democratic meeting place where for example parliamentary political parties, CSOs, universities etc. meet to talk and discuss about politics and societal issues (https://almedalsveckan.info/english, 2021-12-14).4. Sweden Democrats is a nationalist, right-wing, social conservative political party with an anti-immigration agenda.

Disclosure Statement
No potential conflict of interest was reported by the author(s).

Table 1 .
Coping responses to hate speech, threats and harassment against civil society actors and potential consequences.
. More determined to fight hate speech, threats and harassment .Organizing collective actions .