Preventing Radicalisation,or Forcing Ideological Consent? Crises,Uncertainty Reduction and Collective Identity Negotiation in Collaborative Communities of Online Third Spaces

Abstract

With regard to the recent accumulation of crisis events in Europe, scholars stress the potential of online spaces to reduce people’s uncertainty through collective identity formation, providing their members with emotional support and recognition. However, polarisation and radicalisation are mentioned as possible negative consequences of online community formation in crisis times. “Third spaces” are described as online forums where political talk emerges against the background of the primary purpose of the environment – exchanging ideas or goods and mutual support. Our article explores how the “purpose-based” collective identity of the participants of a parenting forum, a typical “third space,” is challenged with their potentially conflicting political identities in the context of strongly polarising issues such as the COVID-19 pandemic and the Russian-Ukrainian conflict. A qualitative analysis of nearly 2,500 contributions confirms that, to a certain extent, a third space constitutes an environment of cooperation and support even during major societal crises and provides space to negotiate conflicting ideological identities. However, broader societal polarisation is reflected in the forum which clearly sets boundaries to diversity, strictly excluding opinions considered extreme or unacceptable by the majority. Proponents of minority opinions are thus pushed to the margins, where they may radicalise.

Introduction

This article is inspired by recent studies that connect crisis times and rising societal uncertainty with the creation of affective communities in the online sphere and the resulting societal fragmentation or even polarisation (Hogg, 2023; Martinsson & Ericson, 2023; Törnberg, 2022; Törnberg & Törnberg, 2024). Although these studies approach the topic from different theoretical perspectives including identity theory (Hogg, 2023; Martinsson & Ericson, 2023) and discursive community formation (Törnberg, 2022; Törnberg & Törnberg, 2024), they describe a similar process: crisis times cause rising self-uncertainty which nurtures people’s need to belong to a distinct group with a clearly defined collective identity. Identification with a collective identity with sharp boundaries, despite its possible positive effects for the individuals (Hogg, 2023), can fuel fragmentation, threaten democratic institutions and constitute “a metacrisis” if it occurs in a context “characterized by intense negative feelings between partisan groups” (Törnberg, 2022, p. 1). In the era of digital communication and social network sites, various authors agree that digital media may drive affective polarisation and political radicalisation and lead to a “breakdown of social cohesion” (Törnberg, 2022, p. 1).

Our motivation for the research presented here stems from two sources. First, we were inspired by the ambivalent role of online collective identity formation described in the literature. The fact that it can have a positive role (providing members with emotional support and recognition, reducing uncertainty) as well as the negative impact on societies described above stimulated us to trace these tendencies in empirical data, particularly in online communities and their political debates. Second, our long-term interest in political communication in so-called third spaces, that is, specific online communities gathering in primarily non-political online forums, yet discussing politics intensely (Vochocová & Rosenfeldová, 2019; Wright, 2012), motivated our interest in how the above-described effects of online community formations are manifested in such an environment. Moreover, we wanted to examine how potentially conflicting collective identities are negotiated in third spaces. The starting point of our research design is the fact that at least two potential collective identities are at play when regular participants of an online interest forum discuss politics. The first identity, which we call “purpose-based,” is derived from the collaborative purpose of online environments where people exchange ideas or goods, share hobbies or provide mutual support. In the case of a parenting forum, the focus of our study and a typical instance of such third spaces, their primary identity is that of a parent or someone interested in parenting and exchanging ideas related to child care. The second identity, which we label “ideology-based,” however, is expressed in the political debates on controversial and polarising issues on such forums. It can be expected that people sharing the identity of a parent or a person interested in parenting are divided in their political opinions on issues such as the war in Ukraine or the COVID-19 pandemic.

Thus, in this article, we present our findings related to our interest in how the primary, “purpose-based” collective identity of participants in a parenting forum is challenged by the potentially conflicting political identities of the group members in times of crises that significantly polarise European societies. To study the discursive negotiation of the “purpose-based” and “ideology-based” identities, we analysed discussion threads related to the COVID-19 pandemic and the war in Ukraine on the popular Czech parenting forum eMimino, employing qualitative exploratory analysis tools from Grounded Theory. Our aim was to identify patterns of such negotiation, as well as uncertainty-reduction strategies, using the example of two different major crisis events.

The national context of this study is the Czech Republic, a country in the Central and Eastern Europe region (CEE) associated with democratic backsliding (Macková et al., 2023). It has an unstable, fragmented multiparty and proportional voting system with significant voter volatility. Because of that, political polarisation does not necessarily follow partisan divides and “public opinion is fragmented into smaller groups regarding various issues,” with political antagonism being rather a result of identity-based culture wars (Macková et al., 2023, p. 56).

Uncertainty-Identity Theory, Crises and Discursive Collective Identity Formation

In the last decades, at least two sources of the rising societal polarisation across liberal democracies of the Western type can be traced – large societal crises (climate change, war conflicts and related immigration, pandemics) dividing people in terms of the appropriate response to such crises, and the rapid development of interactive social media and digital discussion platforms, where the existing divides are further shaped through users’ activity and algorithms. Online spaces are largely associated with the identity work of online media users, which may result in positive outcomes on the individual level during crisis times, as well as rather negative consequences for society (Hogg, 2023; Martinsson & Ericson, 2023; Törnberg, 2022; Törnberg & Törnberg, 2024).

Studies generally connect social and digital media during crises with positive effects on users’ psychology, as they facilitate group identification and provide members with emotional support and recognition (Martinsson & Ericson, 2023). Hogg (2023) describes the crucial role of group identification during large societal crises, as it reduces heightened uncertainty by providing people with “a sense of who we are that prescribes what we should think, feel and do” (Hogg, 2023, p. 828). Such identification with a group, “readily available” on social media, also provides “consensual validation of our worldview and sense of self, which further reduces uncertainty” (Hogg, 2023, pp. 828–832). Similarly, Martinsson and Ericson (2023) argue that in times of crisis, a desire to belong to a community that provides its members with recognition is higher, including the need for affective, emotional identification which may occur both offline and online. The authors are mainly interested in the process of collective identity formation enabled by the performativity of digital technologies, stressing that they significantly increase the intensity and volume of connections to an extent which would have been impossible a few years ago. During times of crisis, they argue, emotions and affects accumulated during longer periods are mobilised, resulting in a more intense process of discursive formation of shared, collective identity in the online sphere, often constructed in opposition to “them” or the “Other.” They thus associate online connectivity and performativity also with the political power to transform “the social fabric through connectedness” (Martinsson & Ericson, 2023, p. 17).

Hence, besides the psychological benefits for individual participants, the discursive formation of collective online identities can result in negative phenomena and dangers for democratic societies, such as societal fragmentation or even (affective) polarisation (Törnberg, 2022; Törnberg & Törnberg, 2024), dissemination of populist thoughts, and nurturing of intolerance towards different groups (Hogg, 2023). Törnberg and Törnberg (2024) approach the expression of opinions in the online sphere as identity building and conclude that members of online discussion groups become socialised into the community through “conversational rituals that instil in people a sense of social membership and intersubjectivity, contained in the elaboration of a shared discourse, within which certain beliefs become sacred and unquestionable.” They label the process as “discursive community formation” (Törnberg & Törnberg, 2024, p. 6). According to them, social media echo chambers may indeed support political radicalisation or at least societal polarisation. Törnberg (2022) explicitly connects the rise in “affective polarization,” that is, intense negative feelings between groups in society, with digital media. By driving “a sorting of differences,” he argues, they significantly contribute to “a breakdown of social cohesion,” representing “a severe societal risk, threatening democratic institutions and constituting a metacrisis, reducing our capacity to respond to pressing societal challenges such as climate change, pandemics, or rising inequality” (Törnberg, 2022, p. 1).

Hogg (2023) similarly warns of the possible negative consequences of the formation of online echo chambers in crisis times. From the perspective of uncertainty-identity theory, he describes the process in which individual identity uncertainty produced during various crises may lead to the creation of “identity silos,” groups that strongly identify “with ethnocentric, populist and xenophobic” ideas and values. Such identification with distinctive, homogeneous and often intolerant groups, Hogg argues, is a very effective way of reducing self-uncertainty and feelings of marginalisation, and such “identity confirmation” is “readily available in the era of social media and widespread internet access” (Hogg, 2023, p. 832). By communicating with fellow group members or observing their communication with others, Hogg specifies, people learn about the group’s identity to which they can conform to be accepted. This process also largely divides societies into polarised factions hostile towards each other (Hogg, 2023, p. 832).

The above-sketched concerns about the possible negative consequences of online identity confirmation and collective identity building in online spaces are, however, associated mainly with fringe, radicalised online spaces in which relatively homogeneous groups interact. In our research, we are interested in online spaces which cannot be defined as homogeneous in terms of the ideology, opinions or ethical values of their members, simply because they are rather broad, inclusive, mainstream (discussion) forums primarily aimed at helping members exchange practical advice and information. We dedicate the next section to summarising recent findings related to these online spaces, referred to in the literature as “third spaces,” to explain why we consider them specific in terms of online collective identity building.

Third Spaces as Inclusive Sites Reducing Polarisation

Building on Oldenburg’s (1999) notion of the third place, Wright’s (2012) concept of the third space refers to a primarily non-political online discussion environment where political talk emerges. Whereas Oldenburg conceived third places as informal offline settings beyond home and work – such as cafés, pubs or community centres – where people gather and foster social ties, Wright (2012) and subsequent scholars extend the concept to digital contexts. Third spaces are not explicitly political in purpose; people do not visit them to engage in politics. They become political when personal life and political issues intersect during everyday conversations (Wright et al., 2015).

Research on discussions in third spaces reveals specific characteristics, particularly in contrast to primarily political online environments. Some authors report a relatively high discursive quality of political talk in third spaces (Vochocová & Rosenfeldová, 2019; Wojcieszak & Mutz, 2009; Wright et al., 2017), referring to a high level of rationality or a low incidence of uncivil reactions (Vochocová & Rosenfeldová, 2019; Wright et al., 2017). Sun et al. (2022), examining Chinese lifestyle and current affairs forums, argue that these function as open and inclusive “third space” where citizens engage in political debate. They emphasise that to understand the nature of political conversations in third spaces, it is necessary to focus on communicative acts other than those traditionally considered central to deliberative reasoning, mainly complaining and sharing personal concerns and interests. Pennington (2018) even sees third spaces as sites of resistance, with liberatory potential, “where hegemonic and normative understandings of the world may be challenged” (Pennington, 2018, p. 620).

Of particular interest to our article are findings on the heterogeneity of views in these online communities and its impact on fragmentation and polarisation. Wright (2012) explains that because people do not primarily visit third spaces to discuss politics, they can be considered politically inclusive. He cites Graham and Hajru (2011, p. 29), who argue that “fragmentation theory makes little sense once we move beyond politically oriented communication landscape” and concludes that instead of polarisation, third spaces “may actually facilitate a broader range of information sharing and debate” (Wright, 2012). According to Steinkuehler and Williams (2006), participation in third environments contributes to forming social relationships, which usually expose individuals to diverse worldviews. Similarly, Wojcieszak and Mutz (2009) found that political talk in non-political forums was less polarised than in explicitly political ones. Studies have also found significant interactions between people with different views, i.e., high levels of cross-cutting discussion (Vochocová & Rosenfeldová, 2019; Wright et al., 2017).

In this context, we were interested in how the discussion unfolds in such third spaces during the two crisis events examined in this article. Third spaces are represented mainly by online (interest) forums that often operate collaboratively, providing a space to ask questions, share experiences, express frustrations, and build connections (Pedersen & Smithson, 2013). Parenting forums are a typical example of such third spaces (Wright et al., 2017). Research shows that parents increasingly engage in online media to seek information and support (Goh & Chi, 2017). Here, they can access help they once would have received from their families and friends, as well as alternative sources of information (Madge & O’Connor, 2006). Digital spaces of this kind become increasingly important during times of crisis, providing participants, including parents, with emotional support and a space for sharing information (Hooper et al., 2022). Studies of parents’ online activities highlight how technology helped them cope with isolation and uncertainty and maintain their mental well-being during the COVID-19 pandemic (Nowland et al., 2024). For these reasons, we consider third spaces focusing on parenting a particularly suitable environment for our inquiry.

Methodology

Since we were interested in how different identities are discursively negotiated, we deliberately chose the most visited Czech parenting forum eMimino.cz (“eBaby”) (www.spir.cz). In the Czech context, it represents one of the most widespread examples of online third spaces. It is not a narrowly specialised community, but rather, a widely used forum where topics related to family life and everyday concerns are discussed. Participants share experiences and support each other on parenting issues, but simultaneously, agree or disagree on political topics. As we focus on crisis periods, we decided to analyse discussions related to the COVID-19 pandemic and the Russian-Ukrainian war. We first searched for all discussions containing at least one of the keywords (“Ukraine,” “Russia,” “COVID,” “coronavirus”) and removed those not related to the crises under study. This produced hundreds of discussion threads, most with only a few comments. To capture the debate dynamics, we purposively selected three discussion threads within both crises, (1) where we expected conflict and (2) which belonged to the most extensive discussions in terms of contributions. Specifically, these were three COVID-19 vaccination-related discussions (“Are there ex-COVID supporters who have changed their minds?,” “COVID vaccination YES/NO,” “Reasons why you don’t want to be vaccinated for COVID”) and three discussions related to the war in Ukraine (“Why is the willingness to help Ukraine weakening?,” “Do you mind refugees from Ukraine?,” “Help Ukraine”). Within each thread, we then analysed the first and last ten pages of contributions to capture various phases of the discussion. In total, we analysed nearly 2,500 comments. The forum is publicly accessible and can be viewed without registration or identification. In our analysis, we treated forum contributions as public data and anonymised them by removing all usernames.

We analysed the data employing an in-depth, exploratory, inductive approach of Grounded Theory (Charmaz, 2006; Strauss & Corbin, 1998), namely open and axial coding, leading first to identifying main concepts and, second, to their categorisation. We selected Grounded Theory procedures because our research, unlike most previous studies on the issues, is of qualitative nature, concerns some previously unexplored areas and aims to provide in-depth insight into the interactions of contributors to a third space. Grounded Theory was chosen because it allows for an in-depth, flexible, inductive exploration of new topics and investigative research questions. It is particularly suited for developing a nuanced, data-driven understanding of complex social phenomena, without preconceived hypotheses. Moreover, it is specifically aimed at studying social interactions and processes (Charmaz, 2006; Strauss & Corbin, 1998), allowing us, together with its well-conceptualised coding scheme, to uncover emergent patterns without imposing preconceived frameworks. The iterative process of coding and category development facilitates capturing the complexity of participants’ negotiations while, at the same time, preventing the influence of potential researcher bias on the analysis. It thus makes it an ideal tool for our research purpose.

There were two coders, the authors of this article. Both started with open coding, during which they systematically closely read the data, breaking comments into fundamental components to develop emerging concepts and identify patterns and themes related to cooperation, support, boundary-setting, and radicalisation. Descriptive labels assigned to these concepts then constituted a basis for constructing and iteratively refining categories and a model in which the two most general categories emerged inductively, each developed by subcategories. The presented system of categorisation is a result of the negotiation of the emerging concepts and categories related to two main research questions. Grounded Theory enabled us to build a framework grounded in the participants’ actual discourse, accurately reflecting their perspectives, thus providing a robust foundation for understanding the dynamic negotiation of identities in the online community in times of societal polarisation.

The first research question is a general one and aims at exploring the various ways in which the forum participants discursively deal with the fact that their ideological approaches may differ significantly in relation to the two discussed crises, i.e., that their opinion on the discussed issues can be conflicting:

• Research Question 1 (RQ1): How are dissenting ideological identities negotiated in the parenting forum?

The second research question is more focused on the mitigation of possibly emerging conflict in the online discussion:

• Research Question 2 (RQ2): What strategies mitigating conflict stemming from different ideological positions of the participants are employed?

Results

Our results indicate that the analysed parenting forum represents an ideologically diverse environment connecting people with very different positions. Despite a situation in which – based on our qualified estimation – pro-vaccination and pro-Ukraine stances dominate the environment, mainly in the later phases of the two crises, those who refuse vaccination and aid to Ukraine also constitute an important part of the forum.

In response to our RQ1, our analysis revealed that forum participants approach dissenting ideological identities in two general ways – (a) by strictly excluding positions identified as unacceptable (such as “anti-vaxxers” or “Russophiles,” as labelled in the forum) and (b) by employing various strategies mitigating conflict in those cases when disagreement is not perceived as being from “extreme,” unacceptable positions. The latter comprises, in response to our RQ2, the acknowledgement of the right to emotions and worries of the ideological opponents and generally their right to their opinion, the patient explanation of one’s perspective by offering arguments, sharing expertise, information, and personal experience, mitigating emerging conflict, avoidance of clearly conflicting issues, and explicit discursive strengthening of the collective identity of the forum members.

In the following sections, we describe and interpret the above summarised strategies and illustrate our findings with quotations from our sample. The scheme below shows the system of categories (two general strategies as a result of the final stage of the axial coding) and their subcategories (emerging in the later phases of open coding and during the axial coding) resulting from our explorative analysis of the data.

Scheme 1

• RQ1: How are dissenting ideological identities negotiated in the parenting forum?

• Exclusion of positions identified as unacceptable within the forum

• • rejection of alleged Russia supporters (“Russophiles”)

• • intolerance to anonymous contributions

• Strategies mitigating conflict (RQ2: What strategies mitigating conflict stemming from different ideological positions are employed?)

Exclusion of Positions Identified as Unacceptable Within the Forum: Eliminating Disinformation, “Anti-Vax,” “Russophiles,” and Anonymous Contributions

First and foremost, it is important to stress that, in addressing our RQ1, not all ideological positions are negotiated in the most popular Czech parenting forum. Some perspectives are strongly rejected by the forum participants and proponents of such opinions are clearly informed that they are not welcome in the discussion.

Generally, there is a very low tolerance for spreading disinformation and conspiracies in the forum or disseminating sources perceived as producing or promoting disinformation, both regarding health issues, mainly vaccination, and the war in Ukraine, revealing a low potential for radicalisation of political views (see Törnberg & Törnberg, 2024, for comparison). Contributions of this sort are typically refused by other discussants, who employ different approaches, ranging from sarcasm and mockery (“Where have you found this? Your sources are funny”; “Do you really believe that after vaccination, your body becomes magnetic?! Still not working in my case, they probably just vaccinated me with vitamin C, do you know where I can complain about this?”) to clear designation of the sources as disinformation and their strict verbal refusal (“Are you really sharing a link to your personal Gmail box? Like really? You have it all from the chain disinformation emails like every anti-vax.”). Such reactions often include verbal attacks and labelling (“Kremlinbots are totally blunt.”; “I think this bullshit is fabricated by Russian agents and every single goose on eMimino is repeating it.”) and a strong refusal of further discussion (“I am really not interested in information sources provided by a Russian bug.”).

Regarding disinformation and conspiracies, we observed an interesting trend in the data. In the early stages of both crises, a relatively patient approach, containing explanations and even educational attempts was typical for the contributions. Discussants explained the character of the sources and advised those who shared them to “not look for conspiracy everywhere” and “use your common sense”:

Vk.com? Really? Do you know what kind of website that is? It is a Russian version of Facebook.

However, as the crises progressed, and especially after some measures taken by the government, such as obligatory vaccination in the case of COVID-19, the discussion became more heated and strict in refusing “hoax and rubbish about microchips in vaccines” or “the anti-vax league.” Discussants began to be highly sensitive about any indicator of an anti-vaccination stance, such as the mention of an “experimental vaccine”:

Anytime anyone mentions an ‘experimental vaccine’, it is immediately clear which way the wind is blowing. You are like a religious sect.

The same was observed in the case of contributors identified as pro-Russian in the discussions on the war in Ukraine. Whereas the refusal of pro-Russian stances was already noticeable in the discussions a few days before the war started and participants expressing them were labelled as “hired red troops” (the association of Russia with Soviets and the Red Army) or as a “courtyard gallery plenary of comrades from the anti-imperialist cell,” recalling the Cold War period, after the start of the war, such discussants were rejected and labelled as a rule:

Stop screwing around and go prove your Russophilia on the front in Belarus. [in reaction to a discussant identifying as male]

This is what local Bolsheviks are going to claim.

Besides the clear refusal of mis- and disinformation, which may be interpreted as an effort to protect the environment and community from manipulation, expressing astonishment at the intensity of Russia support in the discussion is also frequent (“How can there be so many Russophiles among young women? I don’t get it.”). The prevalence of the action against pro-Russian stances is also demonstrated in discussants’ frequent complaints that they were falsely identified and labelled as pro-Russian:

Just because I have a different or neutral opinion does not mean I am a Russophile, as you immediately label me here in the forum.

As we can see, whereas most of the strict boundary-setting comments aim to prevent the spread of disinformation, it is not only clear misinformation, disinformation or conspiracy theories that are policed in the forum. In the case of the Russian aggression against Ukraine, any sympathies for Russia are clearly discursively refused and unconditional support for Ukraine is expected. This includes strict rejection of realistic concerns that stress the economic struggle of the domestic (Czech) population as a consequence of the COVID-19 pandemic and the war in Ukraine. Discussants who bring up the issue of their poor economic situation are subject to reactions that stress how high the living conditions in the Czech Republic are (“I don’t understand how someone safe and warm at home, having a place to sleep and food to eat, a place to wash themselves, with no one shooting at them, can complain.”). Much to our surprise, there are even harsher comments denying any possibility of comparing the situation of Czech and Ukrainian mothers (“It’s like when older, independent individuals whimper that someone cares about those more vulnerable, weaker and those whose world literally broke down.”) and a lot of comments express anger and have the character of uncivil reactions and even mocking and shaming of the lower social status, stressing that people in the Czech Republic are responsible for their own situation:

We can clearly see what trash you are. You want social welfare to cover your debts. You caused your situation on your own. But people from Ukraine didn’t.

Discussants do not hesitate to publicly identify not only those who are spreading disinformation (“This particular hoax is being spread by the ‘little vampire’ nickname here in the forum.”), but also those who expressed support for Russia or vaccine hesitancy in the past. Some even add screenshots of the history of particular contributors and their opinions from the past to generally discredit their contributions – thanks to an affordance of the eMimino parenting forum enabling the easy tracking of the discussion history of individual members:

I remember you, you are the one complaining about the Covid-19 measures. So it’s clear and logical. An anti-vaxxer who also refuses help for Ukraine.

Such contributions seem to influence the discussion and the discussants’ expectations strongly and lead to a situation in which those with marginalised perspectives (typically those supporting Russia or refusing various COVID-19 measures incl. vaccination) either stop expressing them in the discussion (“I have my own opinion but I will keep it for myself.”), are pushed to the margins or forced to start their own discussion thread (“The comrades from the red troops are already discussing the issue in their own thread out there.”). Those who decide to not give up their opinion often switch to the anonymous mode. However, anonymous contributions are refused as another unacceptable strategy:

Why an anonymous contribution? I do not see any reason for abusing the anonymity mode.

It is typically suggested in the comments that anonymous contributions are dubious, and it is speculated that they come from subjects outside the forum with the aim to disrupt the environment:

You throw that shrapnel anonymously and you think I will join the discussion?

We can thus read such comments as a strategy to protect the discussion environment from what is perceived (and rejected) as interference in the forum’s culture from outside:

OK, so she deliberately started a provocative discussion, anonymously, and she doesn’t even bother to contribute, she’s just having fun. Or maybe the owner of the forum has his hand in it? To stir it up here a bit again?

Strategies Mitigating Conflict Between Differing Ideological Positions

In seeking to answer our RQ2, we found that forum participants engage in several strategies to mitigate potential conflict arising from their differing ideological positions. As long as their opponents’ views do not take positions perceived as too extreme or otherwise unacceptable, they acknowledge each other’s right to express their concerns, emotions, and opinions:

Yes, you have the right to such an opinion and I fully accept it; I don’t get [vaccinated] against the flu either and I don’t inject my children with everything that is available . . .

Despite their differing views, we noted that the discussants shared their concerns about COVID-19 or vaccination, as well as the possible consequences of the war in Ukraine, and comforted each other (“I’m scared too. But the probability is just too small . . .”). In the forum discussions, especially in the early days of both crises, it was emphasised that “everybody is entitled to their concerns, without the need for some righteous people to constantly and continuously put them down, ridicule them, insult them, lurk in discussions, bash them over the head with everything they have ever written anywhere and give each other hearts”.

In addition to sharing concerns and emotions, the need to respect opposing views was strongly emphasised, both within and outside the forum (“The fact that you have a different opinion does not give you the right to lecture another self-righteous person.”). Discussants often referred to experiences from their families and surroundings, where conflict arose precisely because of differing views, especially on COVID-19 vaccination, and the inability to discuss the topic. The discussants seemed keen to prevent such conflicts within the safe environment of the forum:

Unfortunately, many people (from both camps) see these discussions as an opportunity to stomp the person of the opposite opinion into the ground and spit venom on them.

Some discussants even appreciated good argumentation despite the differences in positions on the topic under discussion:

Thanks for your contributions. I enjoyed reading them back in that mega-thread while it was still somewhat readable. I’d guess we’re each significantly different politically, but I appreciate that at least someone here is repeatedly bringing up the psychological aspect of things.

One of the distinctive features of the discussion threads we analysed (especially compared to more politically oriented spaces such as online news discussion sections – Coe et al., 2014) was the considerable patience exhibited by the forum participants in exchanging views with their opponents. They politely and patiently explained their perspectives, offered arguments and provided information:

The vaccine was developed by private companies, and within a year, all the focus was on it, and what will go to the EU, if approved, has been adequately tested on tens of thousands of people (except children, the vaccine is not for them).

Humour was also occasionally used to persuade opponents and defuse potential conflict, and sometimes discussants tried to persuade and draw their opponents into the debate by asking questions about alternative solutions to the situation, but still, in a very patient way:

You probably wouldn’t want your grandparents infected by a nurse who refuses to be vaccinated. The risk to old people is just too high . . .

The patience to explain and present more and more arguments often made exchanges very long. Leaving the discussion, as in this case, was thus very rare: “In other words, you can’t say anything against UA, but you can say something against our people, [. . .] I am not going to argue more on this topic here. I will not convince anyone.”

The two strategies described above reflect the tendency of people in third spaces to provide each other with emotional support and recognition identified in previous research (Wright, 2012). However, despite the patience in the discussions and the emphasis on respect for others’ opinions, we observed the tendency among participants to avoid potentially conflicting topics. It turned out that, especially in the later stages of the discussion threads, when pro-Ukrainian or pro-vaccination positions dominated the debates, discussants with minority positions may have been afraid to express their views. “I come into contact with refugees daily. And I have my own opinion about them, but I keep it to myself.”

This self-censorship of minority views is also evidenced by the existence of one entire discussion thread, which we included in our sample, the founder of which calls for “the few debaters, who I know there are and who think that the help to Ukraine is going too far, to also express themselves,” and especially the numerous reactions to this introductory post: “You have the courage!”; “It’s about to get ugly here.”

However, we also noticed a tendency to avoid conflicting topics in cases where the discussion between two opponents was already underway but had reached a stalemate, and one of the debaters decided to end the exchange:

I don’t think you and I are going to agree this time. And the reason is that you believe this fanatical propaganda and I am basing my opinion on what I have been drilled with at school and what I have passed exams on. It’s damn hard for me just to throw that information overboard.

From the analysed thread, it is clear that the two discussants know each other from earlier discussions on other topics. Thus, group and ideological identities come into conflict here, and to maintain good relations, the debate is terminated.

Last but not least, there is also a complete avoidance of polarising topics. In the forum discussions, participants share their personal experiences of how they prefer to avoid discussing conflicting issues related to the coronavirus pandemic or the war in Ukraine outside the forum to prevent destroying good relationships, especially within the family:

My husband and I are not vaccinated, the rest of the family is, and we don’t discuss this topic at all.

Nevertheless, the avoidance of conflicting topics also occurs on the forum itself, both by the discussants (“Can you talk about anything else?”), and by the forum administrators (“We will think about how to deal with such topics that continue to stir up negative emotions and promote the spread of hoaxes, hatred and envy.”). Previous research stresses that third spaces play a positive role in democratic deliberation, because it is harder to avoid political talk there due to relatively strong ties, a sense of community identity, and the ability to facilitate a wide range of weak ties (Wright et al., 2017). Third spaces are thus considered a new opportunity for deliberation because in other spaces, people generally avoid talking about politics or “talk only in ‘safe’ places or on ‘safe’ subjects where they perceive others to hold similar views” (Wright et al., 2017, p. 4). Our data reveal that in case of conflicting issues, even in third spaces, explicit avoidance of political talk or its specific aspects is employed as a strategy to protect the forum ties and to prevent conflict in the online community.

Forum participants also share expertise or information while negotiating their positions on the issues discussed. Some of them were themselves experts, especially regarding the coronavirus pandemic, as some doctors also participated in the forum discussions:

Women come to me with long Covid, it’s horrible, damaged lungs [. . .] short of breath, heaviness in the chest, they are scattered and unable to function . . . [. . .] they have depression . . . [. . .] I don’t want this, so I’ll get vaccinated when possible.

However, expert opinions also come from other experts indirectly. Thus, discussants share the opinions and advice they receive from their doctors with others on the forum:

Today I received a reply, in case anyone was interested . . . “I would not hesitate at all to vaccinate my daughter and I would definitely have her vaccinated.”

Discussants not only share expertise and information, but also actively seek advice: “Wow, so please let me know later how it went at the diabetes clinic.” Many thus view the discussion forum as a safe environment, a trusted source of information and a pool of knowledge. Its credibility may stem from debaters knowing each other from discussions on other, non-political topics and having established good relationships.

Besides sharing expertise and information, participants also use their own experiences to support their positions and share them with others, adding to the usual functions of online forums in difficult times, as previously described (Hooper et al., 2022; Madge & O’Connor, 2006). During the pandemic, the most frequent were positive experiences with vaccination (“No viruses, no colds, and no diarrhoea. One day my vaccinated arm was cold. That’s all.”), or negative experiences with the disease itself:

My brother, a healthy guy, 40+, died. He had no other health issues, not even obesity, embolisms.

Regarding the Russian-Ukrainian conflict, people shared their positive experiences with Ukrainian refugees or with Ukrainians in general:

. . .we used to employ Ukrainians, and they were fine. They come to us all the time, and I have two in my family.

It was mentioned earlier that extreme positions, disinformation and anonymous comments were strictly rejected in the forum. This, together with the gradual dominance of pro-Ukrainian and pro-vaccination positions in the discussions, meant that discussants holding minority views tended to mitigate any emerging conflict by revealing their ideological position to avoid being condemned by the majority. They emphasised that despite holding a minority position, they were not among “the others” (“Just to prevent you from considering me part of an anti-vax sect, I want to inform you that I even have a few optional vaccines. . .”)

For the same reason, in the case of the war in Ukraine, discussants often supplemented their reservations about Ukrainians by emphasising their active assistance to Ukrainian refugees:

I don’t have a problem with them, we’re quite active in helping them ourselves, but I imagine a lot of people will start to have a problem. Helping is the right thing to do, on the other hand people are usually willing to help to the extent that it interferes with their personal standard of living.

We also noticed that discussants reinforced the collective forum identity by discursively negotiating who “we” are here on the forum (“The other thing is that before we, the common plebs, get the vaccine, maybe half a year will pass”). Implicit as well as direct comparisons of eMimino with other environments were also frequent, contributing to the shared feeling of a better discussion environment on eMimino:

I read the discussions on, for example, idnes [online mainstream news website] . . . tell me that there are no such people . . . really . . . some people are acting like dumbasses, don’t they understand that the war is really around the corner.

To sum up, our analysis of the most influential parenting forum, treated here as a typical example of a third space, revealed that whereas the political debates in the forum related to major recent societal crises provide a lot of space for cross-cutting discussion, that is, discussion among participants of different opinions and their negotiation, there are also clearly set boundaries to diversity of opinion. Unlike in typical fringe online spaces connected with radicalisation of political views and spreading of disinformation (Törnberg & Törnberg, 2024), this third space discursively excludes opinions perceived as “extreme” by the majority. Opinions perceived by other discussants as connected to political (mainly the pro-Russian propaganda) and health-related disinformation and conspiracies (regarding vaccination, etc.) are strictly refused. So are, however, contributors who do not directly spread false information, but are identified as ideological supporters of Russia or vaccine-hesitant, mainly as the crises progress. Nevertheless, in many cases, dissenting ideological views are negotiated in the parenting forum with explicitly mentioned regard to the shared identity of the forum participants who are there mainly to cooperate, exchange information and get advice (RQ1). In response to RQ2, certain norms are pushed, such as respect for oppositional opinion (unless identified as organised political action/trolling), and issues which are possibly too polarising are avoided explicitly. The collective identity is discursively formed by negotiating questions such as “who are we,” “how are we different from other forums” or “what do we do or not do here in the forum as a rule,” revealing the potential of online connectivity to influence power- and social relations “through connectedness” (Martinsson & Ericson, 2023).

Discussion and Conclusions

Our article contributes to the general study of online political communication, as well as to the more specific fields studying the role of third spaces in political communication in various ways. First, we document, based on an unprecedented in-depth analysis of third space participants’ interactions, how the primary purpose of the forum probably influences the fact that the specific third space we studied can be considered relatively friendly in comparison with traditionally polarising online political spaces, such as online news discussion sections (Coe et al., 2014). It provides space for a lot of patient cross-cutting discussion, i.e., debate and negotiations among participants of different opinions. Participants of dissenting political perspectives typically provide each other with emotional support and recognition (Wright, 2012). We can thus conclude that the primary collective identity of the forum participants, which we call the “purpose-based” identity, clearly influences participants’ reactions to dissenting political identities – a lot of informational as well as emotional work is done to maintain relationships.

Second, however, our main findings place certain limits on the generally positive conclusion that third spaces (typically leisure-focused discussion forums) represent environments in which the “nonpolitically motivated exchanges expose participants to dissimilar perspectives,” and thus specifically stimulate cross-cutting discourse that can’t be observed to such an extent in some more political environments (Wojcieszak & Mutz, 2009, p. 50). While we can agree that social relationships formed in the forum facilitate peaceful discussion among individuals of diverse worldviews to a certain extent (Steinkuehler & Williams, 2006) and that the debate is definitely less polarised than in explicitly political spaces (Wojcieszak & Mutz, 2009), we do not consider this third space immune to further polarisation in the case of already very polarised political topics connected with large societal crises. Boundaries to diversity are clearly set in the forum in such cases, either by preventively avoiding conflicting topics, or by disciplining some specific views. The strict exclusion of disinformation and misinformation doesn’t need to be interpreted as refusal of ideological disagreement, but can rather be considered mere boundary policing, contesting possible strategic manipulation in the forum. However, our data document that, mainly as the crises develop, opinions and political perspectives which do not contain false information are considered equally unacceptable by the majority in the forum, if they are identified as not reflecting the general political identity or position of the forum – i.e., if they are rather pro-Russian and anti-vaccine. We were particularly surprised by the hatred towards and shaming of socially marginalised Czech mothers, the participants of the forum whose economic concerns were strictly rejected by the majority of discussants once they mentioned that the help for Ukraine may come at the cost of the domestic poor. Based on a content analysis, we can only speculate that these attacks may have been influenced by the fact that their authors consider the contributions mentioning the economic struggle of Czechs as organised manipulation, not sincere expressions of worries of real members of the third space community. In light of these findings, it is worth considering the role of so-called “super-participants” (Wright et al., 2017) or “regulars” (Oldenburg, 1999) as one of the defining features of third spaces. Previous research emphasises their positive contribution to the quality of discussion (Wright et al., 2017). Our data, however, suggest a more ambivalent picture: while these dominant participants help police boundaries and thus protect the environment against disinformation and misinformation, they may also suppress perspectives that diverge from the prevailing political identity or position of the forum. In doing so, they risk limiting the diversity of viewpoints and undermining what Wright et al. (2017) regard as a key value of third spaces – exposure to disagreement.

With our knowledge of the context, however, we tend to interpret the intensity of the rejection of any seemingly pro-Russian position or anti-vaccination stance as a consequence of the strong polarisation of the general Czech public in case of these two highly dividing crises, which also reflect a broader societal polarisation. Regarding the war in Ukraine and the pandemic, two opinion camps occur in the mediated public sphere, each representing a set of worldviews and arguments strongly conflicting with one another. Hence, whereas the parenting forum facilitates group identification and provides its participants with a discursively negotiated shared collective identity, it does not provide such a service to everyone. During the analysed crises, the collective identity of the forum participants was built, among other things, on the exclusion of political identities which can be generally described as anti-system or anti-Western. These identities are typically more vulnerable to radicalisation, as Hogg (2023) explains, because they are pushed to the margins by the majority. They thus feel greater self-uncertainty to which they react by going “to extremes to try to win” the alternative group’s trust (Hogg, 2023, p. 832). We have witnessed in our data a tendency similar to what Hogg (2023) describes as people choosing their “own on-line ‘echo chamber’ to validate their world view and social identity” (Hogg, 2023, p. 832): participants of the parenting forum who were pushed away from the large threads discussing the war in Ukraine or the COVID-19 pandemic founded their own discussions in which they were isolated from the opposing majority. Such discussions were frequently mentioned and further mocked in the parenting forum’s discussion mainstream. We thus conclude that besides helping many of its participants gain recognition, be heard and treated with respect, third spaces such as the parenting forum we analysed also discursively create an environment close to ideological hegemony, informing participants clearly about positions that are not accepted.