Securitising information in European borders: how can democracies balance openness with curtailing Russian malign information influence?

ABSTRACT

The EU views malign information influence (MII) as a threat, and this has been enhanced with the Russian invasion of Ukraine resulting in a ban on several Russian media. This article adds to research on the dilemmas of democracies in combatting MII, by consulting the public on whether the ban was a proportionate and effective measure and if it stained the EU’s moral authority. Combining focus groups with a survey of the Swedish population, we found robust support for the ban – slightly stronger among women and increasing with age – and little backlash towards the EU. Respondents supported active countermeasures yet recognised problems with curtailing the free flow of information. The results indicate a need for refinement of an ideal-type model of countermeasures to MII (Hellman and Wagnsson 2017). We suggest a new category – “fortifying” – that highlights shared state/individual responsibility and the strengthening of citizens, authorities and the media. We conclude by suggesting a need for additional research on how to balance values of freedom and security and whether public support of bans relies on the precarious geopolitical situation resulting from the Russian invasion of Ukraine or if securitisation of information is an enduring trend.

KEYWORDS:

• Censorship
• malign information influence
• EU
• Russia
• securitisation

Introduction

To what extent do the public in democratic states think state institutions should take action in countering malign information influence (MII)? This study investigates this question by asking how Swedish citizens perceive and respond to the EU’s bans on the Russian state-sponsored outlets RT and Sputnik. In March 2022, the European Union (EU) took the controversial step of banning RT and Sputnik from broadcasting within EU borders (Council of the European Union 2022) as a response to Russia’s invasion of Ukraine. The move was met with considerable criticism from several media organisations (e.g. European Federation of Journalists 2022; the International Press Institute 2022), who admonished the union for reneging on democratic values and emphasised the difference between governmental intervention and media blocks already implemented by independent regulators. Indeed, while this was not the first instance of Russian state media being blocked in Europe – in the past, an array of European states have issued individual bans on Russian state-owned media outlets, including Germany in 2021 (Murray and Marrow 2021) – it has been the most prominent, comprehensive and controversial.

Democratic state institutions have many reasons to view the threat that malign information influence (MII) – the projection of antagonistic narratives intended to harm media controlled by foreign governments (Wagnsson 2020) – poses as a pressing security challenge that may warrant decisive measures. Most concerns regarding MII centre around hostile governments, such as Russia or China, exploiting the information domain to foment societal tensions, undermine the capabilities of state institutions and interfere with electoral processes in foreign states (Hall Jamieson 2018, Elshehawy et al. 2021, Hoyle et al. 2022; Wagnsson and Barzanje 2021). In addition, as Szostek and Orlova (2023, pp. 20–21) argue, MII may serve to “ … eroding public support for democracy; via distorting perceptions of truth and thereby hindering rational debate and via weakening the morale needed to fuel resistance and defence of the democratic state in the case of physical attack”. Indeed, Freelon and Wells have stated that the threat posed to democracy by such tactics is “the defining political communication topic of our time” (2020, p. 145). On a political level, the threat presented by MII has already been widely acknowledged, and before the ban’s implementation, the EU had already implemented an array of different countermeasures to allay the effects of MII (European Council 2015, France 24 2017, European Commission 2018).

The controversy that the EU’s ban stirred is reflective of the uncomfortable paradox that state institutions face, wherein they must balance actively constraining the spread of potentially harmful content with upholding foundational democratic norms that they represent. There are significant implications attached to striking the right balance here. As discussed, allowing MII to proliferate freely threatens the stability of European societies. Yet, to be seen as violating the freedom of expression and information – the right to receive information without interference (United Nations General Assembly 1948)¹ – can invite criticisms of hypocrisy. The EU has defined itself as a normative power, based around the promotion of democratic values and fundamental freedoms (Manners 2002). Bjola (2018) points to accountability – the requirement that political actors open themselves to public scrutiny – as essential to an actor building the normative standing necessary to retain moral authority while taking decisive action against MII. Notably, then, the decision to ban RT and Sputnik was taken without public debate or the involvement of the European public. In this light, the sudden implementation of these restrictions without any public discussion might be seen to compromise the EU’s normative standing, and consequently, might damage the legitimacy of the EU in the eyes of the European public.

As this predicament demonstrates, European citizens play a central and significant role in the mitigation of MII – both as the receivers of the potentially harmful information and as the pillars that state institutions rely on for support and legitimacy. To what extent state institutions should take action to counter MII is, therefore, a thorny political and ethical question, and one that also motivates a closer examination of the public’s perceptions and reactions. Thus far, very little research has shone a light on how “ordinary” European citizens think about MII (n.b. Wagnsson 2020). Yet gaining an understanding of the views and opinions of the European public regarding the possible approaches to countering (Russian) MII, including their thoughts about the EU ban, is key to assessing the legitimacy of the ban (henceforth labelled “proportionality”), its effectiveness and consequences for the EU’s moral authority.

Dilemmas for democracies when responding to MII

The question of what the optimal approach for state institutions seeking to counter MII lingers despite growing scholarly and political attention (Bjola and Pamment 2019, Szostek and Orlova 2023). Szostek and Orlova (2023, p. 20) argue that Ukrainian bans on Russian media were indeed proportionate since the banned media threatened public support for democracy, distorted perceptions of truth and prevented rational debate. In contrast to Szostek and Orlova, we engage with the issue of proportionality and other dilemmas from the point of view of public support. We spell out the significance of citizens’ views on countermeasures to MII by discussing the three considerations mentioned above.

The first consideration is indeed proportionality (Bjola 2018), i.e. countermeasures need to be adjusted to the magnitude and nature of the threat. Proportionality should be assessed in connection with public opinion since generally speaking, the democratic state is dependent on public support for its actions. Moreover, it is particularly dependent on public consent on countermeasures to MII, for the successful employment of such measures. If the public disapproves of censorship, this might complicate implementation since people will try to bypass it (Hobbs and Roberts 2018). Bans are also particularly problematic for public support because excessively aggressive countermeasures that violate freedom of expression might prompt negative reactions from the public if seen as clashing with core democratic ideas. Lack of, or weak, countermeasures might, on the other hand, yield negative responses from citizens who expect the state to defend democratic values threatened by MII. Different approaches to MII may be seen as more or less proportionate, with censorship requiring the strongest justification. Here, the approaches may only be seen as morally permissible if proportionate in the sense that they are offered as a “lesser evil”, that is, if the harm potentially inflicted by MII is seen as more damaging than the harm caused by the countermeasure.

The second consideration is the prospective effectiveness of the countermeasure. When looking specifically at digital censorship, most research converges on the idea that such a practice is often largely effective in reducing access to restricted content. A case in point is the effectiveness of the EU’s ban; the Institute for Strategic Dialogue (2022) showed that EU traffic to sites associated with Rossiya Segodnya or RT domains was reduced by an average of 74%, and in many countries, up to 90%. Furthermore, Golovchenko (2022) demonstrates that a Ukrainian ban substantially decreased Ukrainians’ online activity on the Russian-sponsored network VKontakte and that this was primarily due to the phenomenon of friction, that is making online access and consumption more cumbersome. Yet some of the most compelling research has shown that while censorship indeed often successfully reduces consumption and sharing of undesirable content, it can frequently also lead to “backfire effects”, including stimulating a desire to seek out the prohibited information and increasing negative sentiments towards the “prohibitor”. These effects are demonstrative of “reactance” (Brehm 1966, 1989), the psychological concept describing the motivation to re-establish previously free behaviours that have subsequently been forbidden.

Several studies have examined the backfire effects associated with audience reactance in the context of media censorship. Hobbs and Roberts (2018), for example, describe how an abrupt ban on Instagram by the Chinese state incentivised millions of citizens to not only acquire virtual private networks to circumvent the ban but also propell them to further seek political content that was previously not interesting to them. Similarly, research from Turkey shows that the perceived threat to media freedom was associated with an increase in online political information-seeking behaviour as citizens in closed media environments attempt to mitigate this prohibition (Behrouzian 2016). Nabi (2014) provided evidence that the censoring of YouTube in Turkey and Pakistan was ultimately ineffective, with users acquiring software to help them circumvent the ban, and interest in the censored content spiking after the block was implemented. Further extant research has shown that censorship or the perceived infringement of freedoms can lead to the derogation of those that people perceive as the “threateners” – those prohibiting the behaviour. Studies have shown that the perception of a threat to, or the actual loss of, freedom can increase anger and hostility towards the source of the threat/loss to freedom (Steindl et al., 2015) and can affect perceptions of how domineering the source appears (LaVoie et al. 2017). Prohibiting freedoms can also damage the prohibitor's perceived credibility (Silvia 2006). Results such as these provide more grounding to examine how European citizens reflect on the EU’s ban on Russian state-sponsored media, the EU as a political institution and the mitigation of MII more broadly. In conclusion, we argue that also from the point of view of effectiveness, paying attention to public views, is key, since high levels of support for bans will reduce the risk of backfire effects described above.

The third consideration is moral authority. According to Bjola (2018, p. 312), to uphold moral authority, a polity needs to act responsibly, carefully considering the situation at hand, and acting only if the polity faces a likely risk of being harmed. We argue that moral authority needs to be assessed in connection with public opinion. This is particularly relevant and sensitive in the case of the EU, which draws its legitimacy from claims to be a “normative power” that sustains basic principles including democracy and fundamental freedoms (Manners 2002). Actions that infringe on the freedom of expression, such as bans, might erode citizens’ support for the EU and result in accusations of hypocrisy.

Finally, to problematise a variety of potential countermeasures including normative dilemmas, Hellman and Wagnsson (2017) categorised different approaches to MII, exemplifying countermeasures taken by different European states, consolidating this into four overarching models of ideal-type responses to MII. These models consisted of the direct and explicit confronting of antagonistic narratives with overt counternarratives, naturalising, that is projecting a favoured national narrative, ignoring the antagonistic narration and finally, blocking the source(s) of the antagonistic narration (see further below). The authors outlined how these four distinct approaches differ in their engagement with the MII content, and the extent to which they focus on a foreign or domestic audience. In this, the models demonstrate the variety of approaches that democratic states have available to them and the different considerations they must weigh. The ensuing investigation will provide a basis for a revision of this ideal type model in connection with the three considerations outlined above.

The current study

A key contribution of the current study is the novel approach of scrutinising countermeasure approaches from the bottom-up perspective. While Hellman and Wagnsson (2017) have examined the approaches taken to mitigate MII, this was from a top-down perspective: how can and do democratic governments respond to MII? Gaining a better understanding of how the public sees various response models is important, given the aforementioned central role that the European citizens play in the success of response models and because their support is key to EU cohesion, legitimacy and defence of the EU as a liberal democratic polity. Research has shown that “ordinary” European – in this case, Swedish – citizens are also keenly aware of the threat posed by MII (Wagnsson 2020), yet until now, analyses of European citizens’ views regarding mitigation strategies have been mostly absent from academic research and discourse. The current article seeks to remedy this, interviewing Swedish citizens directly about their attitudes towards the EU’s ban on Russian state-sponsored media, and mitigation strategies more broadly and conducting a large-scale survey.

Selecting Sweden as the focus country for this research is notable for several reasons. We choose to explore citizens’ views in a state where support for the ban is among the highest in the EU (European Commission 2023), to find out how this strong support is rationalised by the citizens. Furthermore, Swedish support for censorship is intriguing. Freedom of information has been part of the Constitution since 1766 when Sweden was the first country in the world to pass a law protecting the freedom of the press and of information (Strömbäck et al. 2012, Nordin 2023). Lastly, as noted before, of the few articles that have examined the perspectives of “ordinary citizens” on the problem of MII, one has focussed on Sweden (Wagnsson 2020), and so continuing to examine the views of Swedish citizens provides the opportunity to build on this research.

In sum, the article contributes to research on MII by analysing public views on the mitigation of the problem, and more specifically, whether citizens view censorship as acceptable, despite its downsides. Our analysis of public views of measures will also inform the revision of the ideal type model of how to counter information warfare proposed by Hellman and Wagnsson (2017).

Research design

In this research, we apply a mixed-methods methodology, combining eight focus groups, with a large-scale survey with a representative sample of the Swedish population. The focus groups were conducted to uncover respondents’ (own) reasoning and views on MII, their reactions to the EU’s ban, including (for instance) potential implications for attitudes on the prohibitor (the EU) and their opinions on other potential countermeasures. Two of the authors and two research assistants took part in conducting the focus group interviews, and subsequently read and systematised the transcripts. Recurring views and opinions were noted as were general patterns as well as individual differences. Drawing on the findings, a survey was constructed that was used to probe the results in a larger population (N:1000).

Method of focus groups

The focus group interviews were held on 21 and 22 March 2023, roughly a year after the introduction of the EU’s ban. In total, 65 respondents participated in the 8 focus groups, with 7–11 respondents in each group. The groups were divided by age and gender, comprising four male and four female focus groups in the age ranges 18–29 (group 1–2), 30–45 (group 3–4), 46–60 (group 5–6) and 61-older (group 7–8), the oldest respondents being in their eighties. The respondents were recruited through a leading Swedish analysis and research agency, Novus, using its Sweden Panel, consisting of approximately 50,000 randomly recruited panel members from the entire country, with no possibility for self-registration. This research received approval from the Swedish Ethical Review Committee on 7 February 2023, file number 2022-07177-01.

The focus groups were conducted digitally in a 60-minute session, using Teams Software. The individuals participated online and did not reveal their faces or voices, participating under terms of full confidentiality and informed consent. They decided under what first name they would participate, yet to ensure full anonymity, all names have been changed to fictive names while keeping gender and age unchanged. The interviews were held in Swedish and led by the person in charge at Novus, with the presence of two authors and two student assistants, who were allowed to ask follow-up questions when needed. For this article, the lead author has translated questions and quotes into English.

The sessions began by inquiring about the respondents’ awareness and perspectives on the issues related to foreign information influence. This included discussing which actors might be engaged in such influence and whether they were familiar with or had consumed RT and Sputnik. Subsequently, the following information was provided to ensure everyone had a common starting point for the subsequent discussions: “RT and Sputnik are Russian state-financed international media companies to disseminate pro-Russian messages to audiences in other countries”.

Next, respondents were asked if they were aware of the EU’s decision to block RT and Sputnik. The following information was then shared:

In February 2022, the EU decided to block RT/Sputnik from broadcasting until the aggression against Ukraine ceases and Russia discontinues the dissemination of “disinformation” and engagement in “information manipulation” against the EU and its member states. The EU also justified this decision by stating that RT and Sputnik are not independent media but rather weapons utilized by the Kremlin.

Subsequently, respondents were asked to express their opinions on the EU’s decision, its justification for the ban and potential consequences for their views of the EU. They were also asked if the information about the ban had heightened their curiosity about the Russian outlets. The subsequent discussions focussed on respondents’ viewpoints regarding blocking as a method and alternative countermeasures. Finally, the interviewees were allowed to rank the four potential countermeasures suggested in previous research (Hellman and Wagnsson 2017) using a five-grade scale.

• To block/censor as the EU has done (blocking)

• To do nothing, but trust that our society can maintain a stable democracy with the institutions and media that are already in place (ignoring)

• To respond to propaganda/disinformation with “counternarratives” that expose lies and distorted descriptions of reality (countering)

• To spread information about Sweden and Swedish society without commenting or directly confronting lies and disinformation (naturalising)

The complete list of questions and alternative responses is available on the online repository.

Results of focus groups

Awareness of the threat, ignorance of the ban

The focus group discussions revealed that about a quarter of the respondents had read or listened to RT or Sputnik on some occasions, while only a handful had done so after the ban. Nevertheless, the analysis of the discussions exposed that respondents were mindful of the problem of MII and Russia as a key agent. With a few exceptions, the interviewees described MII as threatening. In addition to Russia, respondents mentioned China, the US, Iran and Turkey. One respondent answered rather typically: “I’ve heard that Russia is a big actor in that field. China as well and other communist dictatorships work like that, I assume” (Ofelia, group 5). These findings are in line with previous research (Wagnsson 2020).

Discussing potential negative consequences of MII, respondents displayed awareness of the risk of MII causing confusion about what is true and false and spurring polarisation and instigating divisions. Quite a few mentioned MII destabilising the US, with one respondent for instance saying:

… sometimes false articles in the US result in upheaval and deaths. I think it harms democracies because it yields division and confusion about what is true. And the greater number of actors involved, the more difficult it is to know what is true, as everyone shares their theories and arguments about why their theory is right. This creates what seems like small camps everywhere. (Sebastian, group 4).

The possible reduction of media and governmental trust, as well as the general undermining of European states stretching from weakening of trust to disorder and violence, was also discussed. One respondent, for instance, said: “Russians aim to create chaos and disorder in the EU among other places, and of course, it has an effect” (Juha, group 6). Another respondent discussed effects including the potential for violence, stating: “It may induce people to act in a way that they would not have done, for example, voting in a particular way, burning the Quran As this is a quote, demonstrating, committing acts of violence against other people et cetera” (Marianne, group 5). The discussion among four young women in group 1 is illustrative of the diversity of concerns being mentioned:

• − Yes, it can harm the state, the country and individuals. (Mirabelle)

• − It can lead to extreme opinions and polarisation, which brings about risks. (Vanna)

• − The consequence will be polarisation, ignorance and so on. (Annie)

• − Conspiracy theories have become a large part of the latest election in the US … (Johanna)

• − RT and Sputnik disseminate disinformation about Sweden and the EU, which leads to a weakening of trust, which can affect the EU’s power for example. (Annie)

The very high degree of concern with the MII led us to expect that respondents were aware of the ban on RT and Sputnik. However, merely a third of the respondents reported awareness about the blocking. Astonishment was a common reaction when being informed as part of the focus group session. One respondent asked: “How so? Is it also valid in Sweden?” (Marianne, group 5). Others were surprised and provoked, one, for instance, stating: “No, I did not know. That sounds worse than the potential propaganda they are spreading” (Johannes, group 2). Asked whether the acquired awareness resulted in an increased interest in RT and Sputnik, many sharply denied such feelings, while others declared their curiosity, with one respondent stating: “I did know about them before, but blocks are always of interest to break with” (Jim, group 2). Another one said: “The two-year-old in me just became curious. Not as much regarding the content but regarding whether it is possible to circumvent the block” (Oscar, group 2). Reactions in group five displayed a broad diversity of reactions:

• − Absolutely not. (Petronella)

• − No. Not to read myself, but it would be interesting if people for instance in Sweden knowledgeable in Russia and defence would read and discuss, for example, broadcasting the discussion on the radio. (Marianne)

• − Absolutely more curious! (Maud)

• − Would have been a bit interesting to read what they write. (Kristin)

• − I am curious! (Ofelia)

• − Yes, it would be interesting to compare their reporting on an event with that of another outlet. (Moa)

• − No, information is being spread by both Swedish and foreign journalists on site and I believe more in their reporting. (Carin)

Support for the ban and the EU

In general, despite low awareness, support for the ban was very strong, with roughly two-thirds recognising the need for blocking. One respondent explained that even though he viewed the blocking as a kind of “censorship”, he regarded it as a measure necessary to stop manipulation and “alternative facts”, adding that societal support seemed to be strong: “ … most people I talked to support the blocking” (Ingemar, group 8). Another respondent linked the ban to the ongoing war and the need for firm measures: “I think that it is positive precisely because it was done as a standpoint against the war” (Amanda, group 3). A third one said: “It is only positive. Right now, I cannot think of anything negative” (Eva, group 7). However, some called for greater transparency, such as a respondent in group 4:

If the EU had stated forcefully that this has happened and explained why they did it, it would have felt better. It does seem like the information has reached most people anyway. But if the EU had blocked without providing a justification, it would definitely have been a case where the EU violates its principles and values, I think. Information is a good thing. It reduces the number of questions. (Sebastian).

Thus, the respondent was satisfied with the EU’s justification but wished that the message had been spread more broadly to the public to enhance transparency. Yet, when asked about whether the ban impacted trust in the EU or national authorities, only a few referred to a weakening of trust. Three respondents in group 7, for instance, replied:

• − I have confidence in the EU. (Maj)

• − My confidence in the EU is not affected. Sometimes I really question how the EU acts, but in this case, I have confidence. (Aina)

• − Unaffected. (Gun-Britt)

Not only did most interviewees support the ban, but quite a few specifically expressed their appreciation of the EU’s power of action in this regard. Rationalising their backing of the blocking, many respondents referred to the war in Ukraine as requiring strong and effective joint measures. Two respondents in group 6, for instance, supported one another showing appreciation for the EU’s move:

• − Even though I am somewhat hesitant as to the effect of the blocking, I am a bit happy that the EU makes such a powerful move. (Tord)

• − I think that the EU is the right forum for taking the decision. (Juha)

• − Yes, otherwise it contributes to the divisions that Russia tries to accomplish. (Tord)

A key issue in the group discussions was whether the responsibility to shield against MII should be placed on the individual or the state. A few viewed the problem/threat entirely as one that the individual must resist and counter, whereas most regarded it as a social or structural problem of a kind that one single individual cannot defend against, thus legitimising collective measures. Yet, respondents commonly thought that whereas the state must provide tools – for example, education in source criticism or issuing warning labels – individuals must also assume responsibility. Many respondents were attentive to the need to protect individuals who are not fully educated in source criticism. For example, one participant stated: “Because many are still not being source-critical, the state and the EU must act and block the flow of information” (Ofelia, group 5). Another respondent said: “It is not a human right to poison us. People are more stupid than we think and the consequences for society override the potential benefits of allowing anything” (Max, group 4). A third one stated “ … making the entire population in a country learn source criticism in this situation appears impossible so better to block when in need of a full-scale solution” (Natalie, group 3).

Even though most respondents were supportive of the ban given the circumstances, a nuanced picture emerges when analysing discussions in detail. One respondent argued: “Who knows what is true or false? Who really stands above it all and decides what kind of information is to be spread?” (Moa, group 5). Another one highlighted the risk of the implementation of additional media bans (Solveig, group 7). Several young men in group 2 were particularly prone to assume an individualistic perspective and to focus on the potential negative effects that bans might have on society and on their own lives. One respondent argued, “It is our right to read any news we want to read and to be able to create our own opinion, no EU should be allowed to decide what we think” (Hans). Another respondent stated: “I don’t think that the EU should decide what media are allowed or not” (Ronnie). A third one expressed concern about consequences for the EU and its development: “The argument is comprehensible, but the one who fights monsters must watch out for turning into one” (Oscar). A respondent in another group worried about the consequences of the state not trusting its citizens: “What does it mean to assume that people cannot take responsibility on their own, people who are supposed to sustain democracy?” (Jonas, group 8).

Finally, in general, the ban was seen as a temporary measure. One respondent, for instance, discusses how the ban is ok for the time being, but not in the long run and only as long as there is an ongoing war: “It is good as a temporary measure. There is a problem with freedom of speech I believe” (Annie, group 1). Similar concerns with censorship and worries about freedom of speech occurred among respondents in all groups, although to varying degrees. The banning of media outlets was generally regarded as a short-term solution to security problems and many respondents raised concerns about threats against democracy and freedom of speech.

Strong support for alternative measures

Asked about potential alternative measures, in addition to or in place of blocking, respondents came up with a variety of ideas. Some, who were clearly ambivalent about censorship, referred to the democratic system as ideally being capable of enduring such threats without extreme measures: “In the best of worlds, democratic systems should stand firm even when facing ‘fake news’ and propaganda” (Ingemar, group 8). Yet, most respondents suggested active countermeasures. Many wished for the boosting of circulation of easily available factual and reliable information in society, as an alternative to, but not directly confronting Russian information influence. One respondent said: “The absolutely best is to spread knowledge and concrete facts, but sometimes it might not be enough” (Milla, group 1). The strengthening of truthful media reporting and journalism was frequently mentioned as part of such endeavours. One respondent, for instance, stated the need for qualitative journalism supplementing foreign information influence, advocating for “sound and reliable reporting” (Lena, group 3).

Others proposed directing attention to the problem, for instance, through information, such as governmental messaging including public discussions among experts. One respondent said: “Debates among specialists. Perhaps information from the Swedish Civil Contingencies Agency on a regular basis, such as weekly press conferences” (Kurt, group 4). Some respondents advocated placing more responsibility on journalism, such as when one respondent suggested: “Meta news? That different media oversee Russian news from a source critical perspective?” (Kjell, group 6). Another one said: “In the best of worlds we would have a 24/7 commentator providing an unbiased version of what comes out of such media outlets” (Tord, group 6). There were also proposals to respond through counter-narratives or counterpropaganda: “Perhaps we ought to increase domestic ‘counterpropaganda’ as a response, instead (of blocking)” (Tobias, group 6). Another one phrased such countering efforts as “counterinformation” (Fabian, group 8).

Another frequent suggestion was the addition of warning flags, red labels or warning texts, on false media content and media outlets, including social media content. Some were aware of this already being done by Twitter and YouTube. One respondent said: “If the blocking was not on, I would like to see warning messages prior to visiting the internet home pages and continual information on the TV channels”. (Johan, group 4).

Finally, the most frequent suggestion was the strengthening of education in source criticism and individuals acting sensibly. This was brought up again and again in the discussions. One respondent, for instance, suggested: “Education and individuals limiting their use of social media, for example not using social media to consume news” (Maja, group 3). Nevertheless, some were doubtful as to the feasibility of applying source criticism in practice:

The problem is that we have access to an infinite amount of information, but to analyze a source as reasonable/unreasonable takes so much more time and energy than just keep on scrolling, after having seen and been affected by the misinformation. (Erik, group 2).

In line with these nuanced discussions on countermeasures, many respondents favoured options other than blocking when given the chance to rank four countermeasures. This exercise resulted in the following ranking:

1. Responding through counternarratives

2. Spreading information about Sweden

3. Blocking

4. Doing nothing

When analysing the ranking in detail, we noted a slight gender difference, with males being slightly more negative to blocking, placing it as number four on average, thus as the last resort, whereas women placed it as number three on average. Furthermore, seven men but only one woman placed “doing nothing” as the top alternative.

Discussion of focus group results

The analysis of the focus group discussions shows that most respondents were surprised yet supportive when being informed of the EU’s decision to ban RT and Sputnik. Also, for most respondents, the blocking did not undermine confidence in the EU, but rather the opposite, even if – from a democratic point of view – the EU ought to have spread the information about the ban’s existence in a better way. Respondents often linked their support for the ban and for the EU’s actions to the Russian full-scale invasion of Ukraine. The considerable and widespread awareness of MII as a threat, and of Russia as a key agent, that was uncovered in the analysis of the focus group discussions rhymes with previous research (Wagnsson 2020). Furthermore, the use of RT and Sputnik among respondents dropped after the ban.

Nevertheless, an important takeaway is some evidence among focus group respondents of backfire effects, with increased curiosity and a few reactions in line with the aforementioned reactance model, which predicts negative reactions to the loss of behavioural freedom (Brehm 1966, 1989). For instance, in the young men’s group, a handful of respondents were vocally critical of the ban and tended towards placing the responsibility to discern MII on the individual. This stands in contrast to the generally prevailing opinion among respondents that due to the difficulty of discerning information influence, responsibilities cannot be placed entirely on the individual, but the state must act to assist citizens in this regard. Nevertheless, most respondents accepted the ban as a necessary measure given the problematic circumstances.

Many thought of the ban as a provisional measure that would remain in place until the end of Russia’s full-scale invasion of Ukraine. When given the chance to discuss the problem of bans in general in-depth, many preferred other measures. Asked to rank the four ideal type measures – confronting, naturalising, ignoring and blocking – the first two alternatives were more popular than blocking. The discussions also exposed a variety of additional ideas of what could substitute blocking. Respondents, for instance, called for warning labels, increased responsibility and activity on the part of domestic media organisations, and information and educational efforts arranged by the state and authorities. In the ensuing survey, we will test the attractiveness of such measures in a larger population.

Method of survey

Sample

The respondents in the survey were recruited by the research agency Novus, from a randomly selected web panel (no opportunity for self-recruitment) and consisted of a representative random sample of the population. The response rate was 51 per cent, with a total number of 1007 respondents participating out of 1975 people invited, aged between 18 and 89 (527 men; M_age= 54 years). The age distribution of the respondents was fairly similar to that of the total population, with a slight under-representation of the youngest age groups 18–29 and 30–49 years and slight overrepresentation of the older age groups 50–64 and 65–79 years, caused by non-response. Weights have been applied to compensate for the disproportionate stratification. This research received approval from the Swedish Ethical Review Committee on 14 June 2023, file number 2023-03063-01.

Survey design, variables and procedure

The survey was conducted online between 14 and 31 July 2023 using Novus’ survey software. Respondents were first informed of the EU’s ban, including the official justification. Findings from the focus group discussions were then probed by asking respondents to indicate responses to the following items (full questions and scales can be found in the supplementary materials):

Awareness

Awareness of the ban was measured through a simple yes-no question: “Did you know that the EU has decided that RT and Sputnik are to be blocked within the EU?”.

Support for the ban

The extent to which respondents supported the EU’s decision to block RT and Sputnik was measured with one item, with respondents indicating their support on a scale of 1 (completely distance yourself from) to 6 (fully support).

Support for ban’s lift

Participants’ willingness to have the EU lift the ban after the end of the Russian full-scale invasion of Ukraine was measured with one item, with respondents indicating support on a scale of 1 (completely distance yourself from) to 6 (fully support).

Satisfaction with public discussion

The extent to which respondents felt the ban has been sufficiently discussed/debated in Sweden was measured with one item, where respondents indicated satisfaction on a scale of 1 (there is no need to discuss it) to 6 (should have been discussed much more).

Curiosity

Whether knowledge about the ban increased respondents’ appetite for consumption was measured with one item, on a scale of 1 (not curious at all) to 4 (much more curious).

Impact on trust in the EU

Whether respondents’ knowledge about the ban impacted their trust in the EU was measured with one item, on a scale of 1 (greatly reduced my trust in the EU) to 5 (greatly increased my trust in the EU).

Consumption

Respondents’ level of consumption of RT/Sputnik before and after the ban was measured by respondents indicating how often they accessed the sites: at least once a day/week/month, more rarely, or never.

Support for alternative methods

Since the survey’s primary purpose was to probe the legitimacy of the EU’s ban, the attractiveness of the ban was finally tested by asking respondents to rank the attractivity of bans in comparison to seven alternative actions. Respondents were thus presented with descriptions of the four ideal type measures suggested in previous research (Hellman and Wagnsson 2017) plus four additional measures that we formulated drawing on the focus group discussions. According to Hellman and Wagnsson (2017), the ideal type measure of “naturalising” refers to the outward projection of narratives by governmental agencies and media outlets as part of public diplomacy. In the survey, respondents were informed about naturalising being a measure directed outward to an external audience. Focus group respondents also suggested the spread of reliable information domestically. We, therefore, added such an alternative in the survey. Both variants highlight the need for factual information to be spread, but without directly responding to malign narratives and projection of counternarratives.

The complete list of questions and alternative responses is available on the online repository.

Survey results

Awareness of the ban

The majority of respondents were unaware that the EU had implemented the ban on RT and Sputnik, with about 70% saying they were unaware. About a quarter of respondents (27%) were aware of the block.

Support and public discussion of the ban

The average level of support for the EU’s decision to block RT and Sputnik was 4.47 (SD = 1.05), indicating a relatively high level of support. A significant majority of respondents (81%) fully support the EU’s decision to block RT and Sputnik. 14% of respondents partially support the decision. Only 5% of respondents either partially or completely distance themselves from the decision. 39% of respondents would fully or partially support lifting the blockade once the full-scale Russian invasion of Ukraine ends. 21% of respondents are neutral, neither supporting nor distancing themselves from lifting the blockade. 27% of respondents take some level of distance from lifting the blockade. The level of support for the ban was significantly correlated with age (r = .151, p < .001), meaning support generally increased as participants got older. An independent-samples t-test indicated that women (M = 4.57, SD = 0.86) were significantly more supportive of the ban than men (M = 4.39, SD = 1.18), t(930.45) = −2.83, p = .005.

When it comes to discussing the blockade, the average level of agreement that the ban had been sufficiently discussed was 2.64 (SD = 1.25), which is relatively low. The majority of respondents indicated that they desired more discourse (in total: 58%). 12% of respondents believed that the blocking of RT and Sputnik should have been discussed much more, while around a quarter (28%) felt it should have been discussed more. Only 16% believed enough has been discussed, and 10% think there is no need to discuss it further. This data suggests that there is a varied perception of whether the issue has been adequately discussed. An independent samples t-test indicated that the extent of satisfaction with the discourse around the ban differed significantly between genders, with men (M = 2.77, SD = 1.34) significantly more satisfied than women (M = 2.47, SD = 1.11), t(743.7) = 3.35, p < .001.

Impact on trust in the EU

The decision to block RT and Sputnik greatly increased trust in the EU for 12% of respondents, and to some extent for 26%. Around half (45%) of the respondents felt it did not impact their trust in the EU in any way. Few respondents felt their trust in the EU had decreased (5% in total). 11% of respondents are unsure about the impact on their trust. The extent of the impact on trust in the EU was significantly associated with age (r = .127, p < .001), meaning the older participants were, the more positive the impact on trust was.

Curiosity in RT/Sputnik content and reading habits

The average level of curiosity in RT/Sputnik content was 3.48. The data suggest that a relatively sizeable minority of respondents expressed an increased curiosity about RT or Sputnik content in response to the knowledge of the ban (in total: 37%), while the majority maintained their lack of curiosity (58%). Curiosity was significantly associated with age (r = .106, p < .001), meaning curiosity increased with participants age.

Most respondents (84%) had never read RT or Sputnik. Only a small percentage of respondents reported reading RT or Sputnik at least once a day (0%), every week (1%) or every month (1%). 10% of respondents have read RT/Sputnik on some occasions. After the EU’s blockade, 5% of respondents reported reading RT or Sputnik. Most respondents (91%) did not read RT or Sputnik after the blockade. Reading habits after the blockade seem to be consistent with the pre-blockade habits with a small percentage accessing the site. However, readership had decreased to 5%.

Methods to counter MII

Regarding the methods available, most respondents favoured either media training or strengthening the capacity of Swedish media to directly respond to foreign malign messaging. Following this, introducing warning texts and blocking or censoring the information, and public diplomacy garnered sizeable support from the respondents, in that order. Relying on existing social institutions without taking action was the least popular method, with only a quarter of respondents supporting this method. Full results can be seen in Table 1.

Table 1. Methods ranked by full or partial support of respondents.

Methods (ranked)	Full or partial support (%)
1. Train people in source criticism and digital information management	89
2. Strengthen the capacity of Swedish authorities and/or journalists to directly respond to disinformation/propaganda with stories that depict the event	83
3. Introduce warning texts on pages on the internet that contain state-funded disinformation/propaganda (e.g. RT and Sputnik)	75
4. Block/censor disinformation as the EU has done against RT and Sputnik	64
5. Strengthen the capacity of Swedish authorities and/or journalists to portray events accurately, but without informing about, or directly responding to disinformation	54
6. Strengthen the capacity of Swedish authorities and/or journalists to inform about and shed light on disinformation/propaganda but without responding to disinformation	54
7. Spread information about Sweden and Swedish society to an international audience through, for example, public diplomacy, but without responding	47
8. Rely on existing social institutions and media without taking action	28

Concluding discussion

This study was born out of a desire to expose public reactions to bans, and to understand how citizens rationalise their support for censorship. The survey confirms that despite low awareness of the ban, support is very strong and there is almost no backlash in terms of loss of trust in the EU. This was illustrated in the focus groups where many displayed true contentment with the EU’s resolve. They generally linked the need for the ban to the Russian full-scale invasion of Ukraine and saw it as a short-term measure. The survey confirmed that almost 40 per cent wished for the ban to be lifted after the end of the war.

In line with the focus groups, the survey confirms that consumption decreased after the ban, dropping from 10 to 5 per cent. Nevertheless, close to 4 out of 10 survey respondents stated that they felt curious about RT and Sputnik upon learning about the ban. These results mirror the mixed reactions in the focus group discussions, with some expressing a lack of interest or even aversion to the thought of seeking out these outlets, yet others express curiosity. This points to the possibility that the ban might have been less effective had there been more public discourse about its implementation. Nevertheless, a majority (58 per cent) of survey respondents wished for more debate, which is in accordance with the many remarks made in the focus groups about the need for public debate. This is an example of a dilemma for authorities when considering using bans while upholding a reasonable degree of transparency and accountability.

There were some notable, and perhaps counterintuitive, differences with regard to age and gender. Women were more supportive of the ban in comparison with men – but were simultaneously less satisfied with the public discussion about the ban than men. Meanwhile, age was significantly positively associated with both support for the ban and curiosity about the censored content. Furthermore, the older the participants were, the stronger the impact of the ban on their trust in the EU.

Our findings indicate that there is a strong securitisation of information – in the sense that respondents believe in firm and engaging countermeasures to curb the spread of Russian-sponsored information including censorship – that appears to be linked to the view of Russian MII as a threat. Such threat perceptions correspond with findings of previous research indicating that a majority of Swedes perceive MII as a threat to core values, including national security (Wagnsson 2020, p. 403), and with a recent survey demonstrating that 64% of Swedes believe that the war in Ukraine poses a significant threat to Sweden (Bricker 2023).

The securitisation of information is somewhat stronger among women and increases with age. The stronger securitisation among women is notable given what has previously been documented about women in the US being less prone than men to support the use of military force (Eichenberg 2003) and forceful actions in response to terrorism (Eichenberg 2022, p. 233), and about Swedish women being more prone to cut defence budgets (Ydén et al 2023, p. 137). The association between age and support for the ban is not necessarily surprising, given research has shown that older people tend to be more supportive of banning potentially harmful media (e.g. Wan and Youn 2004). The relationship with curiosity, however, is surprising given that younger people have been consistently more rebellious in the face of censorship (Lemaire 2023). Future research may examine these differences further. It might be particularly interesting to investigate if the stronger approval for censorship in older generations relates to the degree of digital dexterity: are younger people more accepting of unfettered media consumption due to higher online competence and consumption of online (including foreign language) media?

All in all, then, was the ban on RT and Sputnik a proportionate or disproportionate measure (Bjola 2018) according to the public? Our study shows robust public support and weak indications of backlash, with only a minority of consumers being attracted to the prohibited media. The study demonstrates that Swedish citizens, in a situation where individuals identify a distinct threat, view a ban as reasonable. Security concerns were then prioritised over liberal freedoms. We thus argue that citizens are aligned with the idea of the ban as a proportionate measure.

There are good grounds to argue that the three considerations or requirements outlined above were fulfilled as far as public opinion is concerned. The ban was apprehended as proportionate, the public support laid a solid ground for the effectiveness of the ban by facilitating its implementation and the acceptance and widespread appreciation of the EU’s resolve indicates that the ban rather strengthened than stained the EU’s moral authority.

Nevertheless, there are nuances to this conclusion. Discussing bans in general, many were conscious of the need to balance values of freedom and security and discussed the problem in nuanced ways. Respondents suggested and supported alternative countermeasures to bans. We identified three main trends: First, the more engaging the measures were, the stronger the support. Second, there was support for the strengthening of offensive and defensive measures. Third, responsibility was placed on the state and the individual.

Returning to the original study by Hellman and Wagnsson (2017), Sweden was, at the time, seen as being in favour of “ignoring” as the chief strategic response to MII. Yet, the authors did not include the perspective of the public in their analysis. The current study shows that Hellman and Wagnsson’s model holds well, but there is room for some revision based on the results and the three trends mentioned above. We propose adding a new category of countermeasures, “fortifying”, that encompasses measures that increase the resilience of society.

Revisiting the models, we find that as in the original model, “confronting” is still about responding with counter-narratives directly confronting antagonistic narratives. “Blocking” remains a defensive strategy that is “ … inward-looking and protective, in the sense that it aims to safeguard the national strategic narrative but without promoting it outwards” (Hellman and Wagnsson 2017, p. 161). Yet, it is broader than the original ideal-type including not only censorship but also more moderate measures such as warning flags on dubious media outlets. Both these measures, which rely on action from authorities, have very strong societal support.

The outward-looking approach that is about projecting a favoured narrative to external audiences, telling the story of the “self” without denigrating and contrasting the self to an “other” labelled “naturalising” lacked strong appeal among respondents. This might be due to the hardened geopolitical situation in Europe and weakened beliefs in soft power and diplomatic measures.

The ideal type “ignoring”, placing trust in existing institutions and media organisations without taking any particular action, was the least popular alternative (Hellman and Wagnsson 2017, p. 162). This strategy does not correspond to the public’s current expectations or views. Respondents felt that democratic institutions must not be left to themselves to carry on as before. On the contrary, various supportive measures were suggested to strengthen public institutions and the media.

This leads us to the new addition, the ideal-type model labelled “fortifying”, which puts existing institutions and citizens at the centre and enhances their ability to defend against MII. Fortifying encompasses inward-looking and outward-looking features, for instance, educating citizens in source criticism and identifying and targeting antagonistic messaging from external powers. Individual responsibility is held forth as significant and necessary. Various activities and measures are suggested that all point to the defence of civil society and empowering citizens to build resilience. We place fortifying at the right-hand side in Figure 1, next to ignoring. Within this strategy, the degree to which it engages with the antagonistic messaging varies. For example, when authorities highlight antagonistic narratives, fortifying measures work to shield citizens from MII, by making them vigilant. Yet, fortifying differs from “confronting” by not engaging directly with the adversary in creating counter-narratives. It centres on protection rather than the projection of counter-narratives (Figure 2).

Figure 1. Engaging and disengaging strategies in information warfare, support for each model in percentage *89% supported education in source criticism, 54% supported strengthening the capacity of Swedish authorities and/or journalists to portray events accurately and inform about and shed light on disinformation/propaganda.

Figure 2. Ideal type strategies in information warfare.

A key observation obtained by consulting members of society is the emphasis that they place on state and personal responsibility. Many argue that one cannot leave the entire responsibility to the authorities but keep stressing that personal responsibility is important. Yet, few respondents think that dealing with MII is entirely up to the individual. Even those critical of the ban in the focus group discussions recognised the need for authorities to assist individuals in shielding against MII. There is a sense of a need for shared responsibility and action. This was something not properly contended with in the earlier study by Hellman and Wagnsson (2017). Furthermore, we found stronger support for engaging than disengaging strategies. Taking into account the present geopolitical situation in Europe, and the respondents’ tendencies to link the need for a ban to the Russian invasion of Ukraine, we refer to this as a “securitisation of information”.

This research increases our understanding of the legitimacy of countermeasures to MII in democratic states, yet it comes with a few limitations. First, to enable an inclusive discussion, also among focus group participants who lacked prior knowledge about RT/Sputnik and/or the ban, we informed respondents about these platforms being Russian state-financed media outlets set up to further Russian messaging abroad, and about the EU’s justification for the ban. This might have heightened threat perceptions and support for the ban. Second, we call for future research that takes into account a wider, potentially more negative, range of views on the appropriateness of bans on foreign media. Even though supporters of all parties across the political spectrum were represented in the focus groups, future research should centre attention on supporters of political parties at the fringes of the political spectrum, who are generally more supportive of Russia (Golosov 2020). Similarly, a quarter of focus group respondents had read or listened to RT or Sputnik, but only a handful stated consumption after the ban. To better understand citizens critical to bans, we thus call for research pinpointing active RT/Sputnik consumers. Third, this study focuses on Sweden, where support for the ban is comparatively high, which limits the generalisability of the results. Nevertheless, according to a recent Eurobarometer, support for the ban was 60% or higher in 20 of the EU’s member states. An average of 66 per cent of EU citizens did support the ban (European Commission 2023). In this sense, Sweden can be considered representative of EU member states. Yet, if we want to know more about citizens with critical and more nuanced opinions about the ban, we should explore public opinion in states with comparatively low support for the ban.

In conclusion, more research is needed to determine to what extent information is being securitised more broadly across Europe and whether such securitisation is contingent on the Russian full-scale invasion of Ukraine or has another explanation. Indeed, the public support of blocking and active countermeasures against MII could indicate an enduring change of perceptions whereby the projection of information is seen as a potentially existential threat that requires extraordinary measures.

Acknowledgements

The authors would like to Linnea Berkhahn-Lindholm and Sara Norbäck for excellent research assistance.

Disclosure statement

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

Additional information

Funding

This work was supported by Försvarsmakten.

Notes

1 This right was encoded in Article 19 of the UN’s Declaration of Human Rights in 1948: “Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.”