Between civic virtue and vice: Self-censorship of political views on social media among Norwegian young adults
| 12 mar 2024
Publicado en línea: 12 mar 2024
Páginas: 152 - 172
DOI: https://doi.org/10.2478/nor-2024-0011
Palabras clave
© 2024 Lukas Mozdeika et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Under deep mediatisation, conditions for civic debate and expression have been heavily shaped by algorithmic digital media systems that enable and valorise certain forms of expression at the expense of others. The turn toward more visual social media platforms in the recent decade roughly coincided with the rise of an expressive form of citizenship shifting away from conventional party-politics concerns to matters of sociocultural identity, climate, and social justice (Banaji & Mejias, 2020; Svensson, 2011; see also Murru, 2017). Although young adults have more opportunities for civic engagement and expression than ever before, the proportion between vocal citizens and passive audiences remains generally skewed. While relatively small active groups leverage disproportionate visibility online, oftentimes resorting to derogatory language, the use of social media for political expression by the majority of Internet users in Scandinavia, and particularly the Norwegian context, has been theorised in terms of lurking (Jensen & Schwartz, 2021), inhibition (Sakariassen & Meijer, 2021), and self-limitation or self-restraint (Steen-Johnsen & Enjolras, 2016). Recent studies on freedom of speech and hateful expression in Norway likewise confirm the discrepancy between active small groups and a moderate “silent majority” (Jensen & Schwartz, 2021) that prefers a spectatorship role to interactive engagement in online spaces. While this discrepancy is not alarming, the observed gender, political, and ethnicity biases deserve scholarly attention for their longer-term effects in shaping the development of the public sphere (Mangset et al., 2022).
It is noteworthy that the commonplace charge of political apathy among young adults no longer holds today, as young people demonstrate keen interest in various political issues and find their own ways – media channels, audiences, and styles – for engagement, marking an observed generational shift in media practices (Steen-Johnsen et al., 2021). Social psychology–inspired literature on discursive citizenship norms recognises how self-expressive ways of fashioning civic identities arise out of disillusionment with institutional politics, calling into question normative categories of the informed or dutiful citizen (Gagrčin, 2023). Wary of moral panics surrounding changing citizenship norms, media and communication scholars have instead welcomed new forms of civic agency enabled by digital media and saw this techno-cultural shift as an opportunity to rethink the normative assumptions behind the notions of active political participation and citizenship (Banaji & Mejias, 2020). Lurking, for example, has been redefined in terms of listening in digital environments, calling for further studies regarding its civic relevance (Crawford, 2011; Jensen & Schwartz, 2021). Small acts of engagement, referring to social media affordances of liking, sharing, and commenting, alongside the disparaging terms of slacktivism or clicktivism, have similarly been vindicated for lowering the participation threshold among previously passive audiences (Picone et al., 2019) and promising to mobilise publics beyond the virtual screens (Dennis, 2019). However, the research on expressive citizenship has tended to confound new forms of political expression with changing normative understandings of citizenship falling short of adequate theorical account (Lane, 2020). Consequently, a sense of motivational ambivalence observed among some social media users seems to elude a recent reconceptualisation of “good citizenship” centred on identity and political self-concepts (Gagrčin, 2023: 6).
In this article, I continue to grapple with the question of how civic identity and participation in the digital public sphere is conditioned by deep mediatisation from a different and yet insufficiently explored perspective in media studies: self-censorship. Whether invoked in the pejorative to lament the lack of freedom of speech, or exalted as a civic virtue of moderation or tactfulness, the very idea of self-censorship calls into question different motivations for withholding one’s views. While the difficulty of operationalising this concept in empirical research has resulted in the use of alternative concepts, I suggest that we should source its strength precisely in its conceptual and moral ambiguity, serving us a heuristic tool for probing the changes in mediatised civic culture. Although morally ambivalent and politically charged (Horton, 2011), this notion helps to bring into sharper focus the democratic stakes of digitally mediatised, expressive forms of participation. As we see in the following section, it complements previous research on hesitancy, lurking, and inhibition by calling into question the different normative standpoints for assessing rationales for restraining or expressing one’s voice in the public sphere (Steen-Johnsen & Enjolras, 2016). The critical-theoretical impulse of this article approaches the optimistic ethos and hasty analogies – such as lurking to listening – with caution. Many reasons might explain young citizens’ preference for lurking and self-restraint online, ranging from a hostile debate climate and polarised interlocutors’ views to the questionable fit of algorithmic digital platforms as venues for civic expression. The proliferation of different social media platforms challenges the idea of the digital public sphere in the singular. Their visual focus casts doubt on the relevance of civic debate and argumentation as the main means of political engagement. Although social media platforms hardly discourage the exchange of ideas, digital ethnographers have observed how even robust arguments are prone to context collapse, raising social stakes for voicing one’s opinions in digital environments (Marwick & boyd, 2011; Phillips & Milner, 2017). More general changes introduced by social media, such as ephemeral temporality, unbinds interactions from real time–space, altering users’ perceptions of the public and private domains and strategies for engagement (Schwarz, 2021).
Focusing on the Norwegian context, this qualitative inquiry probes young adults’ rationale for withholding their views online in light of changing communicative dynamics on social media platforms. The main research questions are as follows:
In the following section, I situate my research questions within the recent shift towards the expressive paradigm of citizenship. Methodological approach, data collection, and sampling strategy are outlined in the methods part. The results section presents the key observations from the interviews. In the final discussion, I recast my findings, considering the question of how users’ perceptions of social media affordances come to bear on the normative status of the digital public sphere and digital civic culture.
The following section takes a brief detour into theories of deliberation and participation to flesh out the normative assumptions behind the notion of self-censorship that inform this study. Jakob Svensson’s (2011) argument about the turn from an instrumental to expressive paradigm of citizenship serves as a theoretical backdrop to this end. Inspired by Brennan and Lomasky’s (1993) study of expressive voting, expressive rationality, for Svensson, typifies highly digitalised contemporary civic culture, which may be understood in Dahlgren’s (2014: 257) terms as “the cultural conditions that facilitate participation [and which] can be promoted or impeded, depending on circumstances and the forces at play”. Although generic, the bivalent role of culture stressed by this definition in tandem with circumstances we shall associate with social media affordances brings us to the core dilemma within the competing theories of deliberation and citizenship. To rehearse briefly, rational deliberation requires citizens’ capacity for self-critique and restraint in order to open individual interests and beliefs for revision in search of consensus or the common good. This legitimacy condition – or transformative feature of deliberation, as phrased by Lowe-Walker (2018: 150) – not only presumes a high motivational threshold at odds with lived reality for most, but it is especially problematic in intercultural dialogues that might effectively “endanger, rather than protect, minority culture” (Lowe-Walker, 2018: 152). Subjecting cultural minorities to the demands of the deliberative process requires opening their beliefs and convictions to revision, thus endangering their cultural identity or otherwise compelling them to participate in bad faith (Lowe-Walker, 2018). Taking into account these limitations, some scholars have sought an alternative standpoint in Habermas’s distinction between communicative and instrumental rationality that shifts the locus of the legitimacy of public reason and debate onto inter-subjective processes of meaning-making (Svensson, 2011). And yet, the attempt to salvage the high-rational ground of civic discussion commits Habermas to an ideal speech situation, falling short of a realistic account of how and for what reasons people do engage in the public realm. At the risk of over-generalisation, the gap between reality and hypothetical, ideal conditions so often remains the decisive point in critiques of deliberative theories, whether of instrumental or communicative bent. Following Svensson (2011), we might transpose this gap into a motivational lack that afflicts citizens before rational deliberation even takes place. This is where digital media, and particularly social media platforms, come to play a pivotal role in driving the shift from instrumental or communicative toward expressive rationality. They do so on two counts, echoing the limitations of the demanding deliberative theories: First, they serve as sites for identity formation and self-reflexivity, and second, they provide sources of motivation for engagement and self-expression, which while not necessarily deliberative – but reflexive (Svensson, 2011) – do not foreclose but allegedly encourage desirable forms of participation and deliberation within or outside digital environments (Dennis, 2019). The expressive turn, in other words, is premised on strengthening the civic culture as a precondition for participation not limited to rational debate. Discovering meaning through self-expression and self-reflexivity are thought of as legitimate ends of political participation and citizenship in their own right, mitigating against motivational lack and vulnerability of citizens with lower social capital, such as cultural and ethnic minorities (Murru, 2017; Svensson, 2011).
To further assess the premise of expressive rationality, this inquiry probes gestures of avoidance, instead of motivations, of political expression online as more telling aspects of media practices. Under interpretative scrutiny, they can reveal different dispositions among the young adults regarding their perceived (mis) fit of social media as a public arena for expressing their voice and engaging in dialogue. As a working definition, self-censorship implies an act of “[refraining] from expressing something that we believe it would not be unreasonable to express: we feel pressurised, but not that we have no real choice in the matter” (Horton, 2011: 99). As such, it is a peculiar concept denoting “a choice that is made by the self-censor not to express whatever is being self-censored” (Horton, 2011: 97). To make this a genuine choice, it must entail a wish to express oneself against certain – censorial – reasons not to. Although deliberate and conscious, this choice centres on the main subject as both the author and the instrument of censorship, for otherwise, self-censorship would blur into the ordinary form of censorship (Horton, 2011). Indeed, this is what often occurs in rhetorical appeals to self-censorship lamenting a lack of freedom of speech on the pretext of political correctness or some other effective or purely imagined cause of censorship (Nilsson, 2021). We shall notice how the locus of censoring agency here shifts from the self to some other entity or cause while the self plays a merely instrumental role in censoring oneself, or, alternatively, instrumental-ising self-censorship in the rhetorical act made on behalf of others, supposed self-censors. On the other hand, once we shift the focus to the self as both the author and the instrument of censorship, the censorial aspect seems to recede from the picture. In this view, self-censorship can more easily be reduced to forms of “self-restraint” or “self-control” that fail to connote censorial reasons for which self-censorship is usually deemed controversial and morally reprehensible (Horton, 2011: 99). For Horton (2011: 99), this is an interesting ambivalence that renders self-censorship susceptible of “collapse into no more than either a form of self-restraint or of straightforward censorship”.
We can now see how this ambivalence figures into the normative dilemmas discussed previously. Self-restraint denotes a positive norm (Steen-Johnsen & Enjolras, 2016) or a requisite for legitimate discussion (Svensson, 2011: 48) that extends into civil culture in the form of social etiquette and civic virtues such as discretion (Worden, 2019), interpretative tact (White, 2006: 455), moderation (Eisenberg, 2000), and courtesy (Horton, 2011). From the standpoint of expressive rationality, however, these virtues might seem superfluous at best and disingenuous at worst, if, for instance, deliberative encounters and opportunities for earnest dialogue are tactfully avoided through so-called virtue signalling or dismissed as undeserving of debate. This hints at a worrisome prospect lamented by Hannah Arendt, among other political commentators, who foresaw the risk of an insidious form of self-censorship fuelled by a general climate of complicity, operating “without any application of force, without any terror, [whereby] nothing at all happens – and yet the whole thing eats its way farther and deeper into the society” (Arendt et al., 1993: 210). Such a dire prospect is left open by Horton’s (2011: 97) admission that although centred on deliberate choice, acts of self-censorship can be routinised into “unreflective disposition, a habit, a kind of second nature”. As such, I consider them potentially pervasive phenomenon on the Internet, although not necessarily as alarming as some are led to believe. Certain rational conceptions of civility “imply a duty to self-censor on some occasions” (Horton, 2011: 99), while rhetorical appeals to self-censorship are often made on behalf of others with exaggerated presumptions of censorship. Rather, I would like to consider the possibility that the tension inherent in the Janus-faced character of self-censorship concerns the relationship that young people might have with digital technology itself, being acutely aware of both its benefits and limitations for expressing their political views. As we shall see, this awareness helps explain their media practices and the very expectations and motivations with which they enter, or err not to, the digital public sphere. While not necessarily opposed to the expressive rationality thesis, two aspects of self-censorship remain at odds with it: First, it does not imply a mere lack of motivation, given the wish and reasons to express oneself, and second, it remains prone to devolve into an unreflective, habitual predisposition antithetical to self-reflection, a central feature of expressive rationality.
A recent large-scale study on freedom of expression in Norway found starker polarisation of political opinion among online debaters relative to the general population (Mangset et al., 2022). The observed incivility and lack of tolerance that is pervasive in online environments risks exaggerating differences in political attitudes among citizens and biasing our perception of existing political polarisation. Consequently, this might lead others to withhold their opinion or withdraw from the digital public sphere, a risk that particularly affects young adults with minority backgrounds who are disproportionately exposed to hostility and hate speech online. While these risks are not peculiar to a Scandinavian context, and have been subject to theories of the spiral of silence and chilling effects (e.g., Johansson, 2018), the cultural preconditions for self-expression or censorship typically remain outside their scope, partly because the existing motivation for expressing oneself is usually taken as an independent variable leveraged for determining expression-inhibiting factors in large-sample empirical studies (Stoycheff, 2016). Given the relative abundance of quantitative research, in this article, I take an abductive-hermeneutic approach, treating “conclusiveness of an interpretation” with caution (Feil & Olteanu, 2018: 211). At its core, abduction is a “metaphorical process” reliant on the hermeneutics of interpretation for drawing analogies and generating productive hypotheses (Feil & Olteanu, 2018: 213). Unlike more conventional methodologies of empirical research, the goal here is to attain a more integral perspective on the research question at hand that could facilitate further theory development and generate novel normative insights.
As part of a larger research project on youth participation and media literacy in digital dialogue spaces (U-YouPa), this study draws from twelve qualitative in-depth interviews conducted throughout 2022 and early 2023. Ten out of the twelve interviews were conducted individually, while the other two were small-scale semi-structured focus group discussions involving three participants each. Interviews were audio-recorded with prior approval from NSD (Norwegian Centre for Research Data) for sensitive data collection, and they varied in length from 40 to 140 minutes, with a mean of 63 minutes. The general target profile aimed at 18–25-year-olds based in the Oslo area. The sampling strategy sought to facilitate comparative analysis within the overall focus of the project and therefore was highly selective, aiming at internal diversity based on two main criteria: ethnic-cultural background and political views, distinguished in terms of ethnic majority and minority and progressive or conservative in orientation, respectively. Half of the participants were people of colour, and 9 out of 16 of the sample represent minority cultural-ethnic groups. Gender balance was equal.
Overview of the sample
| Mark | 25 | m | Majority | Low | Centre-left |
| Jim | 24 | m | Majority | High | Green |
| Edvard | 19 | m | Minority | High | Right |
| Elen | 20 | f | Minority | Low | Left |
| Rebecca | 19 | f | Minority | High | Left |
| Lian | 19 | f | Minority | Low | Left |
| Inju | 25 | m | Minority | High | Green |
| Nora | 18 | f | Majority | High | Left |
| Jakob | 25 | m | Minority | High | Right |
| William | 23 | m | Minority | High | Left |
| Anders | 18 | m | Majority | High | Right |
| Kiran | 22 | f | Minority | Low | Neutral |
| Frida | 25 | f | Minority | High | Left |
| Sofia | 23 | f | Majority | Moderate | Centre-left |
| Sara | 23 | f | Majority | High | Left |
| Isaak | 23 | m | Majority | Moderate | Centre-left |
The use of social media played a crucial role in recruitment, yet the decision-making did not follow explicit rationale but relied on intuitions grounded in real-world situations, not least by getting to know potential interviewees before inviting them for an interview. While internally varied, the sample is externally homogeneous to the extent that I sought persons with political interests and concerns who might have some reasons or inclinations for self-censorship on social media. Roughly half of the interviewees admitted having recently disconnected or severely limited their social media use, despite this not being part of the recruitment consideration and hence an unanticipated, surprising finding in its own right. Personal network connections and snowballing served as the main recruitment strategies. Prior to and during the research, I participated in political and interest-based organisations (Extinction Rebellion Norway; FAKS – the association for all conservative students; and the technology-free-choice organisation Eksist!) of my personal accord, which helped me gain an emic perspective on the local civic culture, of considerable importance given my Lithuanian national background. In addition, it served as a recruitment tactic, as roughly half the respondents were recruited from these organisations. Most interview invitations were gladly accepted and only three never received a response. The interviewees were remunerated with gift cards in value of 300 Norwegian kroner.
An interview guide was used to tease out different assumptions on the use of social media for the expression of opinion and dialogue, gradually shifting towards reasons and motivations for self-censorship or hesitation. Questions such as the following were posed at different points during the interviews:
How often do you find yourself inclined to respond in a comment section, starting to type a message, but for some reason hesitate? Why? Do you prefer expressing your political opinions to your friends rather than the general public on social media? How did you adopt your political views? Have they ever changed?
This way of steering the discursive logic of conversation proved fruitful in bringing different context-considerations to bear on the reasons for not expressing oneself or participating in discussions and debates online. I partially relied on the free association narrative method (Hollway & Jefferson, 2008) to invoke and juxtapose different rationales and scenarios brought up by other interviewees, or I even revealed my own hypotheses and assumptions where I felt they were due. This proved to be a fruitful strategy in line with the abductive-hermeneutic approach, eliciting more nuanced and thoughtful observations where the main questions did not. Interviewees would contribute with confirmative or counterexamples moderating my own assumptions and complementing it with phenomenological detail drawn from their own experience. However, shared discursive tropes, in contrast to individual experiences, played no lesser role in establishing a common ground, which later proved crucial for enabling comparative analysis. Interviewees of different political orientations would often invoke common frames of reference, expressing a surprising amount of agreement, yet they diverged in further conclusions drawn from them. Although an unconventional strategy, revealing my own thought process in a dialogic way helped to attain the interviewees’ trust, which was not easy given the politically sensitive topics addressed.
The data were coded and thematically analysed in several turns, or hermeneutic cycles, to facilitate interpretation in light of theoretical assumptions. In the second coding stage, the emerging patterns were clustered into five themes for narrative presentation: low expectations, limitations, and risks; self-expression as an opportunity turned into a burden; political divides and resistance strategies; cultural minorities and ideological transformations; and divergent motivations behind lurking. Despite the limited generalisability of the chosen methodology, reflections articulated by some interviewees may be expected to be shared, in different degrees, among groups represented by the sample. Even if modest generalisations remain unavoidable, they are not the aim of the abductive-hermeneutic approach, however, and should be treated as working hypotheses. The ambiguity that makes self-censorship an interesting, difficult-to-pin-down concept calls for contextualised analysis interested in the peculiarity of the possibly widespread phenomenon, and not its reach.
Well into their way of adulthood, most interviewees demonstrated mature, if somewhat despairing and contradictory, understandings of social media as a digital public space, including its potentials and limitations. There was a notable sense of disillusionment with social media behind a common choice of platforms tailored around specific needs and desires that held across political divides and cultural-ethnic differences of the sample. Namely, few participants perceived any benefit of contributing to discussions and debates, although reading others’ comments defined their style of daily digital media use, typified by swift jumps from story headings to comment sections for scanning users’ reactions, which they often found more interesting than the article itself. As put by a self-identified lurker:
I enjoy reading them [comments], because often when I see this provoking story, then I look at commentary section to see if people have reacted in the way I thought they would react.
Even when the anticipated reaction or argument was not there, there was little incentive for expressing it. Lurking remains a common, arguably important practice even among young adults who actively engage with political issues and debates outside social media platforms. Although having participated in online discussions in the past, most interviewees either saw no point or foresaw risks in doing so currently. This kept them at bay for reasons that often set identity- and argument-based ends of online engagement apart, as seen from the following reflections:
Sometimes, I look at comments or posts and I think of a comment, and then I write it up but usually, I just end up deleting it. So I think of some things, but then I think either I don’t want to spend my time arguing with strangers or I think nobody will care about it anyways. So it’s partly both, because I don’t care about what other people write and naturally other people will not care about what I write. I don’t see the point of engaging in a discussion, partly because I don’t think it’s a real discussion […], partly because I know people will be able to see that I messaged, participated in the discussion. Also, my friends will get a notification that I participated in this discussion. I don’t want my friends to see it.
Sara’s description of online debates as somehow unreal and yet with undesirable social effects that deter her from engagement bespeaks an ambivalent, yet common attitude succinctly summarised by another respondent:
Many young people who are introduced to politics through social media will feel that it’s still important to have these discussions, but there is no point to it.
The use of the adverb “still” hints at awareness among digital natives of bygone forms of discussion that are somehow no longer valued or palpable on social media platforms, despite their latent civic concern. The seeming disinterestedness might lead us to conclude that self-censorship is not really the case where motivations for expression seem to be low. Yet, we shall first consider the entry costs which require users to accept the “terms and conditions of engagement” on social media, a phrase used evocatively by Mark, by then a disengaged social media user. “The consequences or benefits of being on social media press people posting on” believed Edvard, referring to the online post cycle, a temporal threshold enticing participants to constantly maintain their profile feed to stay relevant for the algorithm. This is unlike the “bygone” forum discussions, still prominent on Reddit and Discord platforms, whose pseudonymity provides relief from the identity-burden, yet for this very reason generates a sense of pointlessness. Neither saves from frustration: Not following up an argument spells out internal loss or defeat that could even lead to depression – for not caring enough to defend your views – despite knowing that nobody cares in the first place. Picking arguments on someone’s wall, on the other hand, will most likely be perceived as transgression or offense rather than an invitation for discussion, as reasoned by other respondents.
For those who found them acceptable and willing to put in the effort, the terms and conditions were gradually internalised and normalised, while social media sceptics in the sample identified these very norms as the reasons for inhibition and potentially disconnection:
When I see others posting things on Facebook is a bit like show off, […] posting seems to admit that now I know that you know – so you must read it. I don’t want to pressure people into this.
Nora, sceptic of social media just like Edvard and just as civically concerned, believed status updates are used as “shields” that excuse users from engaging in discussions or standing up for their own opinions:
I don’t want to post something on social media just to show that I am a good person and I have these opinions. I want to post political opinions because I think they are important opinions [...]. You take away the complete value of being politically engaged when everybody has to do it even though nobody cares.
Although at first sight implausible, the last clause of Nora’s reflection reaffirms the exhibitionist tendency identified by Sofia as the driving force behind pervasive conformism on social media platforms:
Affordance is not just an invitation, sometimes it can be a kind of pressure, that others should want to do the same things in a way. There is something very exhibitionist about social media.
While most interviewees agreed on the centrality of identity and “having one’s face behind one’s views”, they drew starkly different conclusions regarding social media affordances used to express them. While Sofia likened reliance on them to a “game of caring”, as did Isaak, who dismissed it as “cheap way of showing other people you care […] rather than [actually] expressing your opinion”, a relatively active social media user of progressive leanings, Frida, vindicated virtue-signalling as inherent in human sociality, which is merely amplified and rendered more transparent by social media affordances:
It’s about signalling; I think everyone does it. The tone of voice of how you say feminism – that is also signalling […]. So yeah, I think that’s part of the visual social media. The language you use says something where you are politically and which side you are on.
For the sceptics, however, it was less about signalling particular beliefs or taking sides but the very premise of signalling that was morally reprehensible, as admitted by Sofia’s fear of being brainwashed by slogans, not arguments. Isaak corroborated this:
I know that me posting about war – switching my profile picture to blue and yellow will not change anything about the war. The only reason I would be doing that is just to signal to people what I care about, and I don’t want people to think I am the kind of person to signal my beliefs. I don’t do that.
Reliance on social media affordances such as flagging one’s profile picture for expressing political affiliations provoked contentious responses, mirroring different styles of social media use across the political divides. As summarised by Edvard:
Progressives have done incredibly good things – helping minorities to have a better life, but it has gone so far that its either you are with us or against us […]. They have a thing that you are either racist or anti-racist, and if you don’t define yourself as anti-racist then you are a racist.
Although firmly on the left, Nora echoed Edvard’s observation, referring to the debate climate on social media during Black Lives Matter protests:
This kind of understanding that you had to post something on social media and if you didn’t, you were either a bad person or you didn’t care, or maybe you were a racist after all. […] We can say that you should be posting something that most people agree is generally a good cause, but what about when it’s not, when its right-wing teenagers on Reddit talking about how you should be posting nazi propaganda on media […]. And that is the exact same mentality, we just don’t view it like that because it’s a different moral perspective.
Generally, the motivational ambiguity of self-censorship oscillating between the mere exercise of discretion and aversion to real or imagined censorship (Horton, 2011) falls along partisan lines in the context of this study. While less active users situated to the left on the political spectrum tended to internalise their reactions and avoid confrontation, conservatives more often projected their frustrations onto political opponents and resorted to self-censorship as a discursive trope on behalf of the silent majority, a scapegoat on both sides of the political divide. Surprisingly, all respondents agreed about the general lack of civility resulting in a hostile debate climate online. Similarly, they all recognised the high stakes of losing an argument and remained averse to criticism, as admitted by Edvard:
Although I spoke on debates, even I am scared of criticism. Why? Maybe because we are all supposed to be perfect and if you say something stupid... I feel like generation Z is so used to laughing, of mocking people on the Internet. We bully people but we don’t like being bullied ourselves, hence we don’t say anything.
Conservatives showed more willingness to excuse hateful expressions as a corollary of free speech not to be contained by arbitrary norms, lest it leads to censorship and silence. Sarcastic humour and meme parodies of political adversaries is therefore a prominent expressive strategy among right-wingers, granting deniability to their claims under the pretence of freedom of expression. Although respondents seemed to recognise that they reproduced the climate they themselves recognised as hostile, ironic humour was weaponised as a politicised strategy for testing the limits of speech performed with a certain internal distance or even indifference, as inferred from admissions that it is often unworthy of time even if part of their daily media use. Progressives, on the other hand, tended to dismiss political memes as disingenuous and took comfort in standing for an inclusive majority, admitting that user numbers do make a difference on social media platforms not least by securing group loyalties via any means of expression afforded. There is nothing new or particularly alarming in the digital public sphere, according to Frida, as discussions about political and cultural issues are justifiably difficult and controversial. She explains the hostile tone of brash users by their own insecurities and lack of voice, thus affirming the centrality of identity in deliberations when real peoples’ identities are at stake. Without delving further into the controversies of cultural identity politics, suffice it to say that group loyalties and corresponding styles of expression are crucial for understanding polarisation dynamics online.
More worryingly, young adults of minority background fall outside this dynamic, perceiving the digital public sphere as an inhospitable place replete with controversies that keep them away from engagement and open discussions. Elen, of Muslim religious-cultural background, described the perceived risks of speaking her mind on social media, affirming the spiral of silence theory:
In Norway you have to be careful what you say, everyone does it – you have to be careful – forsiktig – what you talk or else they will attack you. You have to be true. That’s why we never debate with other people. You hesitate sending that message for some reason: Yesss! You get bashed, I don’t want to get bashed. Cause on TikTok people get dragged, if you don’t follow the majority and they will actually search you up, they will come after you. I’m from a small culture and I don’t wanna go through that.
Nor do young adults of minority background feel particularly welcome to partake in discussions on institutional or local political issues, as revealed by Elen’s friend, Rebecca, who described her impression of Norwegian civic culture as inert:
It’s just boring. Norwegians don’t really talk about the people they vote for. It’s anonymous. I don’t know if it’s good or bad. Sometimes I want to hear peoples’ ideas but it’s like NO!
Kristin, of Pakistani ethnic-cultural background who has lived in Norway for five years, corroborated, often word for word, these views, admitting that taking interest in politics in Norway, compared with her native countries Canada and Pakistan, is challenging because people do not generally express themselves. For Elen, Rebecca, their friend Lian, and Kristin, social media offered a safe space of socialisation that they preferred to keep within their cultural milieu, with little intention of actively partaking in public debates. If anything, it is the need to adapt to prevailing expectations in their daily lives that motivated their social media use as a retreat from situations prone to censoring their views or preferences, a point corroborated by a student from Turkey, William, who noticed that Norwegians are much more vocal on private social media channels than real-world environments. Importantly, monitoring others’ opinions and staying informed on political matters that affect their lives remained crucial ends of their social media use, making them highly receptive and largely informed citizens.
In contrast to female respondents, Edvard and Jakob, also of ethnic-minority background, became politically engaged through exposure to debates both on and outside social media, which made them question the left-wing consensus prevailing in their family and social milieu. Jakob told me how his refusal to partake in an LGBTQIA+ initiative at his educational institution drew him into a controversy, a decisive turning point in solidifying his conservative standing. Private messages of support that Jakob received from other peers confessing that the stakes of speaking out were too high made him realise that identity politics, otherwise of little actual relevance, seem inconsequential until it affects one personally, forcing the picking of sides:
There are issues that are going to have effect on your life and those are not identity politics. You don’t want to take a position until you have it straight in your face. I was against this identity politics until it was at my doorstep.
Such decisive turning points in the formation of political outlook among young adults bring to the fore authentic communicative situations where personal identity and deliberation intertwine. It is in real-world rather than virtual environments that respondents came closest to revising their political beliefs and moral convictions. Sara noted how taking up imaginary opponents’ positions in discussions with her granddad helped her fine-tune her views on ethical and political issues:
I’m not necessarily saying it because I mean it, because it’s my honest opinion, but it is to try it out as an idea. It’s built on trust. The other will not think that this is my identity. Or this is like my real opinion.
Echoing the reasons for which Elen, Rebecca, and Lian tended to keep deep conversations in their private circle, Sofia traced the difficulty of assuming trust to the lack of courtesy on social media platforms:
On touchy subjects I find myself telling to my friends: I know I sound crazy but sometimes I think this and this and this, and I feel I would never put this on my wall or write that article about it because I’m not sure enough that I stand for it […]. I feel there is something about having that courtesy to say that I am not after you, we are doing this together. You don’t see this a lot on social media. […] The worst thing is being criticised for something you know is wrong, it’s the worst feeling in the world – I am an idiot.
“You know is wrong” is instructive for understanding the missing element of courtesy Sofia gleaned from Sara’s role-playing example. If the transformative quality of deliberation requires one to open one’s views to revision, thus endangering vulnerable identities, the reverse also seems true in the sense that revising one’s views in conditions of trust helps There is always this extra layer: Am I doing this because I mean it or am I putting too much into what other people are going to think, and I get into this existentialist crisis – who am I really, what am I doing, or what I want other people to watch. I prefer posting [opinion pieces] so that I can talk to people who actually read it than just people who watch.
Awareness of various risks and pressures involved in expressing one’s opinion publicly online is commonplace among the young adults, outweighing the reasons for participation in discussions and free expression. Feelings of attachment and addiction to social media platforms frequently alluded to by the respondents cloud the perceived civic potentials of these online spaces, whereby considerations of disconnection, for those who are not disconnected yet, can be understood as attempts to re-assert civic agency by prioritising more authentic forms of communication, be it offline or by writing opinion pieces for news outlets. The choice of disconnection, however, is often deferred for reasons that affirm lurking as the default user state online, a state which conceals a remaining unease regarding one’s own motivations and intents as measured against the prevailing norms and affordances of social media platforms:
It’s not good for me to be that much on social media. It’s a waste of time mostly. Because most of it is scrolling, not engaging. Kind of mindless, and I don’t like that.
While information-searching and sentiment-monitoring remain legitimate aims behind lurking, lacking any censorial intent, the admissions of wasting time and “mindless scrolling” mirrors the motivational ambivalence of self-censorship discussed previously. Lurking, or retreating from active participation in online discussions, is a common response motivated by perceived poor quality of dialogue, interlocutors’ disingenuity, and the asynchronous nature of online conversation generating never-ending post-threads that either gradually exhaust or deter users. Respondents showed an acute awareness of how digital traces can draw them into unwanted debates where opinions are endlessly contested, distorted, and defamed, creating a hostile debate climate and fuelling mistrust, which forecloses the possibility of an open political discussion. Forgetting the real person behind piling-up post threads, as one respondent put it, is a constant risk online, and yet the want of a specific thought or argument – rather than one’s expression – to be out there often motivates interventions in debates that remain unactualised, as the risks of getting it wrong are too high when one’s identity is at stake. It is as if the Internet got caught up in the transition from the anonymous free space of reasons to one totally enveloped around identity claims, where expressions are judged as more or less deserving by reference to group loyalties. The general, although exaggerated, perception of this dynamic was described by Elen, a left-wing progressive of minority background, as one where “everyone is cancelling everyone”, referring to the so-called cultural wars deemed replete on the Internet.
While some channel the sense of frustration to picking futile rhetorical battles or feeding on moral outrage and vitriolic joking among like-minded communities, others withdraw further into the margins of the digital public sphere using social media strategically and sparingly. As hinted by Sara’s anticipation of reactions in the comment section, lurking can aid in self-reflection, yet such reflexivity rarely leads anywhere. The sense of pointlessness or alienation revealed by the interviewees should be seen in the light of the valorisation of quick affective reactions driving much of the trivial provocative content on social media platforms. Such content is rarely amenable to civic debate but invites further clicking, performed either in consideration of the algorithm for tailoring one’s feed or for others’ voyeuristic gaze, as demanded by the character of the social situation shaped by social media affordances (Schwarz, 2021). While highly enjoyable for some, other users choose to opt out, deterred by accruing expectations of performing one’s better self, which they perceive as disingenuous and at odds with earnest dialogue.
Even if descriptively more adequate than the deliberative democratic theories, the expressive rationality thesis leaves crucial questions about its civic merits intact, arguably more so two decades after the rise of social media platforms. For Svensson and other media scholars, the descriptive adequacy of concepts “in tune with the observed phenomena” trumps more demanding criteria for qualifying forms of participation and debate as civil or political proper (Murru, 2017: 58; see also Sakariassen & Meijer, 2021: 506). This entails a limitation, I argue, of the expressive citizenship thesis and calls for more conceptual work before empirical findings could swing the remaining quandary: Is expressive rationality sufficient for citizenship, or a precursor to discussion among persons with different preferences, cultural identities, and deep-seated beliefs? Rather than construing citizenship as expressive, reflexive, and networked (Svensson, 2011) on empirical grounds, we might recast the incivility and inhibition online as limitations of expressive civic culture shaped by affordances of social media platforms which encourage certain forms of expression at the expense of others, or else lead to silence. Notably, the possibility of the expressive turn generating self-censorship or withdrawal as its side effects calls into question different normative ideals subsumed by the comprehensive category of expressive rationality, rather than their idealism per se.
Although I cannot develop this further, my basic point of contention is that expressive citizenship as a “more comprehensive and more adequate perspective” (Svensson, 2011: 51) blurs the distinctions between expression, communicative rationality, and deliberation rather than appreciating their respective differences that could explain the ongoing transformation of the digital public sphere. It is unlikely that the expressive turn absolves citizens from the requirements for self-critique and listening prescribed by deliberative theories. Rather, what we are witnessing in the shift away from the dutiful or communicative paradigms of citizenship is not so much a denial of the demands they entail, but the disillusionment and reproach of those who fail to live up to them despite the common awareness that the majority do (Svensson, 2011). This might either take a form of civic apathy or “interpassive” tendencies in civic culture (van Oenen, 2010), or it might take extra-discursive, antagonistic character as manifest in the so-called culture wars and the rise of affective political polarisation. The former gives fewer reasons to worry about self-censorship than the latter scenario, but these should be seen, in any case, as interrelated tendencies beset by processes of mediatisation.
I invite the reader to explore this theoretical junction further, as different normative standpoints can help us better diagnose the discontents in civic culture that have become more palpable since the writings of the expressive turn around a decade ago. One takeaway of this study concerns social media affordances that offer means of expression derided by sceptics in terms of slacktivism, clicktivism, or by extension, virtue-signalling. Isaak’s aversion to signalling beliefs on social media serves as a corrective to the expressive rationality thesis. The principled insistence that he is not the kind of person to signal his political views unsettles Svensson’s (2011: 50) equivocation of signing an online petition to signalling that “I am the kind of person who supports popular democratic movements” or who “is environmentally concerned”. What we tend to overlook here is the performative or perlocutionary function of social media affordances in speech-act theory terms. Signalling one’s political views communicates more than intended, that is, belief in signalling itself, or by analogy, over-identifying oneself with one’s social media profile. In this case, principled silence or inhibition might imply not so much listening but discontent or “silent protest” (van Oenen, 2010: 302): users’ preference for civic norms tailored to dialogue that have little purchase or applicability in virtual environments.
As noted by Nora, abstaining from signalling her affiliations during political protests will likely be read as support for opponents, compelling her to perform – signal! – her civic identity in a way that feels disingenuous to her. In contrast, perfectly content lurkers more often rely on social media affordances as a way to avoid confrontations perceived as unpleasant or potentially exhausting, representing civic apathy as the flipside of the dilemma. This brings us back to the motivational ambiguity of self-censorship oscillating between civic virtue and vice that I started with.
Identity-based vindication of social media affordances trades the civic virtue of standing behind one’s views into the vice of virtue signalling by refusing to make claims inviting potential disagreement and thus excluding adversaries. The classic Internet adage – “don’t feed trolls” – affirms Nora’s analogy of status updates to shields of protection. On the other hand, anonymous discussions often premised on ideals of objectivity yet furnished by vitriolic memes and collective joking effectively bypass one’s identity, often creating a hostile debate climate, especially for vulnerable participants. It thereby trades a civic virtue of a better argument into a vice of malevolent intent of trolling. Neither qualifies as a fully-fledged self-censorship but the respective omissions – of one’s identity markers or one’s arguments or political views – help in understanding the reasons for which the majority of users censor their opinions online: The presumed disingenuity of others motivates the very preemptive acts of self-restraint that further propel such a self-perpetuating dynamic.
This emerging pattern offers an empirical backdrop for understanding social media’s role in transforming civic culture in ways that both facilitate and inhibit participation, “depending on circumstances and the forces at play” (Dahlgren, 2014: 257). Indeed, rational reasoning grounded in specific situations demonstrated by interviewees dovetails with existing empirical findings on inhibition and lurking as highly circumstantial and platform-specific (Sakariassen & Meijer, 2021; Steen-Johnsen & Enjolras, 2016). What escapes such users’ rationale, however, is the very normative basis of the public civic sphere: their keen interest in others’ opinions more than the content itself, yet unwillingness to contribute by dismissing it as unworthy of time or effort that envelops into a contradiction. It is as if their very awareness of the potentials and risks of social media platforms makes a genuinely civic – curious, concerned, unassuming – engagement into an irrational or unprofitable choice. Such irrational choice, however, merely identifies the gap or fissure between the demands of the platform society and experiences of its citizens, which calls for further critical attention.
By redescribing citizenship in terms that conform with existing realities, in the case of expressive citizenship, conventional empirical studies risk succumbing to the status quo bias and falling short of a critical standpoint that is empirically informed and adequately theorised. Defining political expression as an “interpersonal communication process” (Lane et al., 2021: 339), for example, misses out on the capacity for self-restraint and listening required for reciprocal, meaningful discourse or deliberation. While perhaps too demanding, these aspects remain no less important than individual forms of self-expression afforded by social media platforms. Such recognition draws us back into conceptual debates which an empirically inclined reader might have thought we had already left behind, yet ones that cannot be resolved through empirical studies alone.
The limitations of low generalisability of the hermeneutic-abductive approach turns into its strength of identifying significant moments of ambivalence that expose structural or systemic contradictions in culture. If the aim of abductive-hermeneutic inquiry is to reconstruct general phenomena by exploiting idiosyncratic individual perspectives, its insights can only complement more conventional sociological approaches negatively, by problematising cases that often remain obscured by formal – extensive yet planar – typologies. This promises a more nuanced understanding of the “connective dynamics” (Schwarz, 2021) of the digital public sphere that can inform normative theorising and offer a measuring rod for the expressive turn as an ideal, not just a painfully familiar reality. The arguments offered in this article are far from decisive, but in an abductive vein are provisional and suggestive for further research directions. The disconnection from social media platforms and the turn to mainstream or alternative media might seem a potentially laudable change in relatively high media-trust societies such as Norway. We should remain attentive to the risk, however, of the fragmenting digital public sphere and polarisation fuelled by a discursive vacuum that disproportionately affects citizens with vulnerable identities, such as young adults belonging to ethnic-minority groups. The presumption of self-censorship on behalf of others, on the other hand, will likely remain a potent rhetorical tool in the political discourse, hence the need for a more nuanced understanding of it and related concepts in media and communication studies.