Disinformation and the Brussels bubble: EU correspondents’ concerns and competences in a digital age

Abstract

European Union (EU) political actors have been heavily affected by the so-called disinformation crisis, leading to intense worries about how EU citizens may be guaranteed access to trustworthy information in the years to come. While there is increasing research on how EU officials, platforms, and political parties react to the threat of disinformation, less attention has been paid to how another crucial group of actors in Brussels copes with this threat: EU correspondents. This paper presents one of the first empirical observations of EU correspondents’ perceptions of the disinformation crisis. We conducted a survey with Brussels-based correspondents ahead of a politicized 2019 European Parliament election campaign, and asked: What are (a) the concerns and (b) self-perceived competences these correspondents have in dealing with disinformation? Our study offers two main take-aways: First, we recommend further studying EU journalism as a complex organism whose concerns and competences are influenced by intra-European differences in press freedom and journalistic professionalisation. Second, disinformation studies in the EU context must no longer neglect the challenges professional journalism faces when aiming to stop false content.

Introduction

During recent years, policy-makers have become increasingly concerned with the threat of disinformation across the European Union (EU). Their concerns are fuelled by data protection scandals (e.g., Cambridge Analytica), involving new technologies that enable malicious actors to exploit such technologies to spread falsehoods. The latter have not only been regarded a particular danger during EU elections (e.g., Scott, 2019), but appears to be a persisting challenge as the world battles the spread of online disinformation related to the COVID-19 pandemic or Ukraine war (e.g., Basch et al., 2021; Patel and Erickson, 2022).

Naturally, the threat of disinformation, i.e., “forms of false, inaccurate, or misleading information designed, presented and promoted to intentionally cause public harm or for profit” (European Commission, 2018a: 3), is highly relevant for journalism - a profession in an almost permanent mode of ‘crisis’ (Frau-Meigs, 2022, 2022, 2022 Pickard, 2020). A growing body of research focuses on citizens’ abilities to identify mis- and disinformation (e.g., Corbu et al., 2020; Hameleers et al., 2021). This research is increasingly mirrored in journalism studies, where a number of studies focus on how journalists themselves build up new skills and resilience when faced with disinformation during their work (e.g., Duffy and Tan Rui Si, 2018; Graves et al., 2016; Saldana and Vu, 2022; Shapals and Bruns, 2022). This literature is highly relevant as journalists are centrally affected by the emergence of disinformation in two ways: first, it poses a threat to the credibility and trustworthiness of their profession as a whole – both when disinformation is shared online, and when political actors label journalistic work as “fake news” (e.g., Egelhofer & Lecheler, 2019); second, news journalists are increasingly pressurized to detect disinformation in their daily working routines in times of fast-changing news environments and of the wealth of information distributed via social media channels (e.g., Graves et al., 2016; Saldana and Vu, 2022). In this paper, we focus on the latter challenge.

Today, journalists across the globe must make informed choices as to which sources are reliable in a constant digital information flow (Barnoy and Reich, 2021). The pressure this creates for journalists is amplified for Brussels journalists, because of the characteristics of this particular group of journalists and the political environment in which they operate. As we will elaborate below, EU journalists are well-connected within the ‘Brussels bubble’. This leads them to rely heavily on official elite sources, which are increasingly professionalising their press relations. Yet, Brussels correspondents tend to act as generalists rather than specialists, who are forced to make the complex and distant subject matter of EU politics interesting to their audiences back home. All of this takes place against the rising politicisation of EU affairs in national contexts and increasing pressure from policy-makers and citizens to counteract disinformation.

The aim of this paper is thus to shed light on (a) the concerns and (b) self-perceived competences of EU correspondents in dealing with disinformation in their work. Thereby, our contribution to the existing literature is two-fold. First, it adds to the literature on the Brussels press corps, which has not been amended considerably in recent years. While there is research examining the role of understanding, working conditions and approaches to covering EU affairs (e.g., Gattermann, 2011; Frangonikolopoulos and Papapadopoulou, 2021; Raeymaeckers et al., 2007), most of these studies relied on in-depth interviews. Others presented a general approach to Brussels journalism through different forms of survey research with a selective or limited number of Brussels correspondents (Statham, 2008; Terzis and Harding, 2014). Our paper also relies on a relatively small survey sample (N=83) despite a response rate of 13.91%, which is in line with comparable research (Terzis and Harding, 2014) and relates to variation in other samples of European journalists (Fengler et al., 2015). However, we extend existing research by assessing journalistic perceptions and self-perceived competences concerning disinformation at the beginning of a highly politicised 2019 EP election campaign. We also offer an agenda for future research, that is structured along two main arguments: First, we see a great need to further study EU journalism as a complex organism whose concerns for citizen exposure to disinformation and competences are influenced by intra-European differences in press freedom and journalistic professionalisation. Second, we conclude that disinformation studies in the EU context have spent too much time focusing on extra-journalistic fact-checking organisations and citizen resilience, thereby neglecting the challenges professional journalism faces when aiming to stop false content.

Journalistic verification in the digital age

The speed of the current news cycle, economic bottlenecks, as well as the professionalisation of political PR and message-control, have put journalists under immense pressure to deliver political information as fast as possible (e.g., Diekerhof, 2021), while still adhering to quality indicators, such as objectivity and independence, in times of “post-truth” politics (Poulakidakos et al., 2018). However, where journalistic production has long relied on personal conversations with trusted sources to confirm a new piece of information, there are now questions of geo-location, big data, decloaking false visuals and deepfakes, or acquiring eyewitness information from inaccessible parts of the world via secure online communication (e.g., Thomson et al., 2022). This means that, as convenient and immediate as the Internet might be for conducting information searches, it also confronts journalists with an overwhelming amount of information, offered to them through the algorithmic lens of personalised search engines, newsfeeds, deepfakes, and social bots (Basch et al., 2021).

A competent response to this complexity can be defined by advanced levels of (a) knowledge and (b) skills in verification – as well as expedient (c) ethics and professional attitudes. However, research shows that all three factors may require improvement: Many journalists lack the most basic knowledge on where to find out who/what is a reliable source online as evidenced by the limited verification of online information (Schifferes et al., 2014, Duffy and Tan Rui Si, 2018; Shapals and Bruns, 2022). This also suggests that journalists might simply not have the skills or access to necessary automated tools to check information (Himma-Kadakas and Ojamets, 2022; Thomson et al., 2022). Beyond that, the threat of disinformation has inspired other information intermediaries, such as fact-checking NGOs or social media platforms, to take on the role of verifying and moderating disinformation online (Cherubini and Graves, 2016; Hameleers, 2022).

Unfortunately, a ‘verification crisis’ comes at a time when the existence of journalism as a profession itself is already in question. With digitalisation has come economic transformation, ownership concentration, increased precariousness of news work, as well as new business models blending marketing activities with political news production, challenging the traditional boundaries of the newsroom (e.g., Alexander et al., 2016; Pickard, 2020). Large international tech companies, such as Google and Meta, are not only taking over the task of distributing political information but also influence the selection and relevance of news (Bell et al., 2017). Similarly, journalists are grappling with several emerging political and social upheavals, such as increased polarisation among the public and the rise of populist political parties that are fundamentally suspicious of established news media (Van Aelst et al., 2017). All in all, there is the worry that journalistic ecosystems in many European countries may soon no longer be able to perform the tasks so important to upholding democracy.

Verification competences and concerns in the brussels press corps

In what way are EU correspondents potentially affected by these developments? Once among the largest press corps in the world (Raeymaeckers et al., 2007), the Brussels press corps is now shrinking (Council of the European Union, 2022; Terzis and Harding, 2014). It has traditionally been described as “the first public of Europe” (Baisnée, 2002: 110), due to its direct and exclusive access to information available in the EU capital and its close interaction with colleagues from different news-gathering traditions. In other words, EU journalists are part of the so-called ‘Brussels bubble’, which also comprises policy-makers, advisors, lobbyists, and pundits. This is reflected in the heavy reliance on official information sources in line with churnalism practices (Lorenz, 2017) and the dominance of European and national political elites in the news coverage of European affairs – both in traditional media and in online news (e.g., Michailidou et al., 2014; Peter and De Vreese, 2004). This would suggest that Brussels-based journalists consider themselves less vulnerable to disinformation. Then again, journalists are faced with a large and fragmented information network across EU institutions, where they are forced to act as generalists covering complex political topics (Chronaki and Frangonikolopoulos, 2020; Gattermann, 2011; Raeymaeckers et al., 2007). This requires sophisticated verification skills from EU journalists. Furthermore, Brussels-based journalists have to convince their editors back home of the newsworthiness of a certain EU story (e.g., Frangonikolopoulos and Papapadopoulou, 2021). This entails that EU correspondents are rather selective in their news coverage and may be less familiar with those developments that are not of direct relevance to their domestic audience. This, in turn, may undermine their verification competences.

These challenges are exacerbated by the increasing politicisation of EU affairs (e.g., Hutter and Grande, 2014), which goes hand in hand with an increase in EU news coverage (e.g., Lecheler, 2008). Likewise, communication work within the EU institutions has become increasingly professionalised, taking a more active stance in controlling how journalists access sources and information from the EU (e.g., Laursen and Valentini, 2015). This means that journalists working in the Brussels bubble are now under high pressure to produce news and that competition for information among them is growing (Raeymaeckers et al., 2007). We know from current research that such pressures also increase the vulnerability to disinformation (e.g., Himma-Kadakas and Ojamets, 2022), which thus may also apply to EU journalists.

Nonetheless, European affairs are still considered less important than domestic issues on the daily news agenda. EP elections exemplify this, which is why they have traditionally been described as ‘second-order’ elections compared to national elections (Reif and Schmitt, 1980), which still holds despite increasing politicisation (e.g., Ehin and Talving, 2021). This suggests that disinformation may be seen as a less pressing issue during EP election campaigns, simply because these elections are considered less important – with the consequence that journalists may also be less worried about its influence on their work. What speaks against this hypothesis is that the EU has taken great effort in addressing the perceived threat of disinformation, thereby pushing its salience in Brussels. At the same time, EU citizens consider journalists to be most responsible for stopping the spread of online disinformation – before national authorities, citizens themselves, online social networks, and EU institutions (European Commission, 2018b).

In sum, there is reason to assume that EU journalists will feel vulnerable to the challenges of the disinformation crisis. Importantly, in this study, we distinguish between their perceptions of their competency in dealing with this crisis, as well as their wider assumptions regarding the threat the crisis holds for EU citizens and the profession. We ask:

RQ1:

What are the (a) concerns and (b) self-perceived competences EU correspondents have in dealing with disinformation?

Determinants of journalistic competency perceptions

Perceptions of own (lack of) competency and that of others in dealing with disinformation will most likely not be universal but differ among EU journalists (see Shapals and Bruns, 2022). First, there are likely differences between the countries in which the audiences of Brussels correspondents reside. For example, while actual data are scarce, there are studies that at least assume that the supply of disinformation is greater in some EU member states than in others, reflected in varying levels of public worry about its influence (e.g., Hameleers et al., 2021).

One argument related to this (perceived) supply of disinformation is media ownership and press freedom, as those who own or control media outlets can influence the information reported and disseminated in their outlet (Mattoni and Ceccobelli, 2018). This is particularly important when considering illiberal democracies and regions with compromised media freedom – which are a reality within the EU today (Gimson, 2019; Jakubowicz, 2018). Here, state control is used to decide which messages are reported or to prevent journalists from correcting and fact-checking false content (see Gerli et al., 2018). This means that by controlling public information flows in their countries, these actors can ensure widespread distribution of disinformation.

A second determinant may be allocated at the organisational level, in particular the type of media concerned. Media markets have undergone a tremendous shift during the last two decades, with a new form of digital journalism disrupting traditional media production. Digital production has changed the channels through which journalistic reporting is disseminated, the technological tools with which it is produced, and the expectations that society has regarding the performance of journalism (Steensen and Westlund, 2020). Importantly, journalistic production in digital media has differing focal points compared to their colleagues working in traditional media. The former, if working in networked investigative or data journalism, AI tools and verification devices are more often used in everyday work (e.g., Stray, 2019). Furthermore, Salamon (2020) suggests that freelancers expertly use digital tools during the production process, which may suggest higher levels of literacy vis-à-vis disinformation.

Third, research also shows that journalists working for digital media tend to work in more precarious conditions (Gollmitzer, 2014; Hayes and Silke, 2018). EU journalism in general has suffered from the adverse consequences of the financial crisis (Chronaki and Frangonikolopoulos, 2020). It is not entirely clear whether economic uncertainty and the lack of stable source networks and newsroom backing that comes with it may challenge their competency perceptions negatively. Economic uncertainty and greater pressure to produce a high number of stories every day to make ends meet may push those journalists away from more thorough verification techniques and towards a higher reliance on agency news, social media sources, and fast-produced stories. For example, Gollmitzer (2014) finds that freelance journalists value the professional standards and identity of journalism, but that the pressure of atypical work forces them to abandon these standards or suffer serious consequences for their mental and physical well-being (Sherwood and O’Donnell, 2016). In sum, we consider three possible and interconnected determinants for concerns about disinformation:

RQ2:

Do concerns and self-perceived competences in dealing with disinformation of EU correspondents differ according to (a) level of press freedom, (b) media type, and (c) journalists’ employment status?

Method

Sample

To address our research questions, we designed an online survey for Brussels correspondents. We compiled a list of relevant media based on the longitudinal media content analysis conducted on the European Parliament (EP) elections between 1999 and 2009 (Banducci et al., 2014), which included a sample of main newspapers and broadcasters in 27 EU countries (Croatia joined in 2013). We amended these media by additional newspapers, broadcasters and online media, alongside pan-European and global media that cover European affairs. We gathered contact details of individual journalists from media or journalists’ websites. As it is not always evident whether journalists indeed work in Brussels, we initially included journalists who regularly report on EU affairs. ¹ This resulted in a sample of 539 individual journalists from all EU member states and beyond (see Appendix 1). We obtained ethical approval from the Ethical Committee at University of Vienna prior to the fieldwork. Invitations to take part in the survey were sent out individually; invitees were informed about the purpose of the survey and provided with information about the data collection, processing and analysis.

The fieldwork was carried out 4 months before the 2019 EP elections, which took place between 23 and 26 May. The survey was launched on 9 January 2019 by contacting a small sample of correspondents to test the survey. On January 16, the survey invitation was sent to all remaining contacts. We sent four reminders in total (22 and 29 January, 12 and 27 February) and a number of contacts in EU politics sent out the invite to their networks. The last response was recorded on 7 March 2019. A total of 75 respondents completed the survey (response rate: 13.91%). Appendix 1 provides an overview of the distribution across countries.

The sample is not representative of all Brussels correspondents. We were unable to recruit journalists from eight EU member states (Cyprus, Estonia, Ireland, Lithuania, Malta, Poland, Romania, Slovakia) and recorded a very low response rate for several other countries, including some of those in which press freedom has been increasingly constrained in recent years (Bulgaria, Croatia, Hungary). Likewise, journalists from large countries like France and Germany were rather reluctant to fill in our survey, while others were comparatively more willing to respond, particularly from smaller countries (for a similar variation, see Fengler et al., 2015). In the absence of a central and open register of Brussels correspondents, we built our own database of correspondents working for major news outlets. However, the population was unknown to us. An overview published by Politico Europe in May 2020 (Gehrke, 2020) indicates that the number of correspondents per country based in Brussels was nonetheless often similar to our records in 2018 when we compiled the database, including Bulgaria, Cyprus, Germany, Italy, Romania, Slovakia, and Slovenia (see Appendix 2). Yet, Politico Europe has recorded a much higher number of correspondents for Belgium, France, Greece, and the UK, and a somewhat higher number for Hungary, the Netherlands, and Portugal (note that the 2020 figures by Politico Europe are also higher than the ones by the Council of the European Union (2022)). Our results are thus not necessarily generalisable to other journalists within the same country. In other cases, Politico Europe recorded fewer correspondents compared to the number of journalists we contacted (Estonia, Ireland, Lithuania, Malta). The low response rate in these cases can thus be explained by the fact that the population was smaller already, if not non-existent. This, of course, limits our ability to study the attitudes of these journalists (see Appendix 3 for a comprehensive overview of the sample/survey).

Analysis and measures

To test the influence of the determinants of concerns on and self-perceived competences in dealing with disinformation, the relatively small sample provides us with a challenge relating to statistical power: some analyses are conducted on a sample of only 60 journalists. To tackle this issue, we combine two analytical strategies: in the paper we present the results of difference tests between the groups, while in the appendix we present the results of regression analyses. However, these latter results should be interpreted with great caution, as the dataset is likely to be underpowered to detect significant effects. As the dependent variables in these analyses are highly skewed and fail to comply with the assumptions underlying OLS regression analysis, we recoded them into dummy variables and ran logistic regression analysis.

The survey included several questions related to concerns journalists have about disinformation and their self-perceived ability to deal with this issue (see Appendix 4). These include two open-answer questions asking about the most important problem a) the EU and b) journalism as a profession is facing at the moment (in the logistic regression, we distinguish between those that mentioned disinformation and those that did not). In addition, we asked two questions about how likely respondents think it would be that a) citizens across the EU and b) citizens in the main country of distribution of their news outlet will be exposed to impactful amounts of disinformation during the upcoming EP election campaign (7-point scales; in the regression analyses we distinguish between low (0-3) and high scores (4-6)). Furthermore, we asked how confident they are to be able to distinguish between disinformation and truthful information during the news production process (7-point scale; in the regression analyses we distinguish between scores (0-3) and high scores (4-6)).

To determine the level of press freedom, we used the World Press Freedom Index 2019 of Reporters without Borders ² for the main country of distribution of journalist’s news outlets. We recoded the countries into countries with high press freedom (original scores between 85 and 100 points) and countries with a lower press freedom situation (original scores below 85). In our sample, 30 journalists (36%) work in countries with high press freedom and 32 (39%) in countries with lower press freedom (21 journalists, 25%, work for international news outlets and are not included in these analyses). The type of news outlet was recoded into traditional media (including tv, radio, newspaper, magazine and news agency) for 64 journalists (77%) and digital media (including online news and mixed news outlets) for 19 journalists (23%). The employment status of journalists was recoded into those who have a contract (full-time or part-time, 64 journalists, 77%) and those 19 journalists (23%) who work freelance.

Results

Concerns about disinformation

To examine whether journalists working on EU affairs are concerned about the problem of disinformation, we first asked respondents an open question about their opinion on the most important problem the EU is facing at the moment (n=80). Only one journalist in our sample believes that disinformation is the biggest problem in Europe, which also renders studying the determinants of this attitude obsolete. Most worry about the rise of populism in the EU (mentioned in 20% of answers) and the EU’s legitimacy (18.75%). Other often-mentioned problems are the relation between nation-states and between the EU and the rest of the world (‘geopolitics’; 11.25%); the functioning of EU institutions (8.75%); Brexit (8.75%); and migration (8.75%). All in all, even though journalists might worry about the spread and impact of disinformation, they generally did not believe this is the most important pressing issue for the EU prior to the 2019 EP elections (only mentioned in 1.25% of answers, see Figure A1 for full overview).

Secondly, we asked respondents an open question about their opinion on the most important problem that journalism as a profession is facing at the moment (n=81). By far the most often named problem is the current business model of media companies: journalists are very worried about their funding and resources (mentioned in 35.80% of answers). A regularly posted comment by journalists was that there is simply not enough money to run media organisations and that the pressure to produce news very fast is high, which leaves them with too little time and resources to dig deep into a story.

However, disinformation is the second most mentioned problem journalism as a profession is dealing with (24.69%). ³ Related to this, some answers mention digitalisation (6.17%), and social media (8.64%), which are thought to affect the quality of the journalistic output and the spread of disinformation. Another expressed worry is the declining trust in the media (13.58%, see Figure A2). Egelhofer & Lecheler (2019) argue that declining trust would be in part also related to disinformation: the rise of both the fake news genre (disinformation spread with the intention to deceive) and the fake news label (the instrumentalization of the term by politicians to discredit media outlets) in public discourse impacts citizen’s trust in media organisations.

To answer RQ2, the country’s level of freedom of press appears unrelated to whether journalists perceive disinformation as the most important problem or not: The results show that journalists from countries with high press freedom indicate more often that disinformation is the most important problem journalism is facing at the moment (M = 0.32; SD = 0.48) than journalists from countries with lower levels of press freedom (M = 0.19; SD = 0.40), but this difference is not statistically significant (t = 1.19; p = .12). The media outlet journalists work for does matter: journalists from traditional media outlets more often indicate disinformation to be the most important problem (M = 0.31, SD = 0.46) than journalists working for digital media outlets (M = 0.05; SD = 0.23) (t = 2.29; p = .01). ⁴ Employment status of the journalist does not seem to affect indicating disinformation as the most important problem in journalism, as the difference between journalists working contract-based (M=0.29; SD=0.46) do not statistically significantly differ (t=1.52; p=.06) from journalists working freelance (M = 0.11; SD = 0.32).

To dive deeper into the perceived problem of disinformation by journalists, we asked them directly how likely they think it is that a) citizens across the EU and b) citizens in the main country of distribution of their news outlet will be exposed to impactful amounts of disinformation during the 2019 EP election campaign. Results for the perceived exposure of disinformation during the 2019 EP election campaign for citizens across Europe (n = 82). It show that journalists worry about this a lot. The average score is 4.93 (SD = 1.16), indicating that journalists strongly believe that citizens across Europe will be exposed to disinformation during the 2019 EP election campaign (see Figure A3).

We tested whether there are differences in worries about the exposure to disinformation during the EP election campaign for citizens across Europe between journalists from countries with high and lower levels of press freedom. The results show that journalists from countries with high levels of press freedom are significantly less worried about this exposure (M = 4.76; SD = 1.21) than journalists from countries with lower levels of press freedom (M = 5.23; SD = 0.91) (t = -1.69; p = .048). ⁵ In addition, we examined whether journalists working for digital news media outlets (M = 5.00; SD = 1.00) are more worried about this exposure than journalists that work for traditional media outlets (M = 4.90; SD = 1.21) and the findings show that the difference is non-existent (t = -0.31; p = .62). Additionally, we studied the influence of employment status on this worry and show that journalists that work contract-based (M = 4.94; SD = 1.22) do not significantly differ (t=0.16; p=.44) from journalists that work freelance (M = 4.89; SD = 0.96).

A similar question focuses on citizens in the main country of distribution of their news outlet instead of citizens across Europe. The results show that journalists worry quite a lot about this as well, although a bit less than for the exposure to disinformation for citizens across Europe (t = 5.52; p < .01): the average score for the perceived exposure to disinformation during the 2019 EP election campaign for citizens living in the main country of distribution of their media outlets is 4.27 (SD = 1.44; see Figure A4).

Journalists from countries with lower levels of press freedom (M = 4.84; SD = 1.14) are significantly more worried about the exposure to disinformation in their home countries than journalists from countries with high press freedom (M = 3.72; SD = 1.41) (t = -3.42; p < .01). ⁶ Journalists working for traditional (M = 4.33; SD1.39) and digital news outlets (M = 4.05; SD1.61) are both equally worried about this issue (t = 0.74; p = .23). The findings also do not show a significant difference (t = 1.27; p = .10) between journalists that work on a contract-basis (M = 4.38; SD = 1.50) and freelancers (M = 3.89; SD = 1.18).

Competences to detect disinformation

To answer the second part of our research questions, relating to the self-perceived competence of journalists to detect disinformation, we asked the question how confident journalists are that they are able to distinguish between disinformation and truthful information during the news production process (7-point scale, 0 = not at all confident and 6 = extremely confident to make this distinction, n = 82). Results show that journalists in general are quite confident that they can detect disinformation (M = 4.78; SD = 1.10, see Figure A5). This is somewhat surprising considering that they are also quite worried about the exposure to disinformation by citizens across Europe and in their home countries. Apparently, they trust themselves to detect disinformation, but do not trust their colleagues to do the same.

In addition, we tested whether there are differences in the self-perceived ability to detect disinformation in the news production process between journalists from countries with high (M = 4.62; SD = 1.12) and lower levels of press freedom (M = 4.81; SD = 1.20); these differences are not significant (t = -0.64; p = .26). We do find that journalists working for digital media outlets (M=5.26; SD = 0.81) have more trust in their ability to detect disinformation than journalists working for traditional media (M = 4.63; SD = 1.14) (t = -2.23; p = .01). ⁷ Lastly, journalists working contract-based (M = 4.75; SD = 1.17) are equally convinced that they can detect disinformation (t = -0.47; p = .32) as journalists working freelance (M = 4.89; SD = 1.10). The analyses in Appendix 13 do reveal that there is a negative effect of age: older journalists are less convinced that they can detect disinformation than younger journalists.

Discussion

Our results shows the great need to further understand EU journalism as a complex organism whose concerns for citizen exposure to disinformation and competences is influenced by intra-European differences in press freedom and journalistic professionalisation.

First, we show that EU correspondents are rather concerned about funding and resources, disinformation and declining media trust affecting the profession. However, they do not consider these problems part of the challenges that the EU would be currently facing. This suggests that Brussels correspondents either believe that the professional problems they identify relate to national traditions and should perhaps also be addressed at the national level, or they perceive these problems as being so universal that solving them would be beyond the scope of EU policy-making. This stands in contrasts to several efforts at the EU level to combat disinformation across the EU, such as the 2018 report by the High Level Expert Group on Fake News and Online Disinformation on the request of the European Commission. ⁸ The findings that press freedom in journalists’ home countries does neither matter for considering disinformation a problem for the profession nor their self-perceived ability to detect disinformation, lend support to the assumption that challenges pertaining to disinformation in the profession are perceived to be rather universal than being context-specific. Future research has to enquire what journalists would suggest in terms of solutions to these major challenges and which stakeholders, including platforms, policy-makers, civil society organisations and others, can help to combat the problems associated with disinformation in the profession. To do so, journalism scholars – potentially in collaboration with European journalism associations – need to find a way to gather larger and more representative samples of Brussels-based journalists from different countries as their input is crucial to understanding their assessment and recommendations to tackle disinformation in the EU. This is particularly important as the fight against disinformation is additionally challenged by the fast-changing nature of how falsehoods are produced, disseminated, and debunked online and offline. For instance, during recent years, there is growing concern regarding the potential threat of visual disinformation for both users and journalists (Ferreira, 2022). Here, it is not only the often-discussed “deepfake” that may be a concern, but the increased use of decontextualised photos and videos in digital sources. Both decontextualization as a relatively low-tech and deepfakes as advanced forms of producing visual disinformation may only be verifiable with the use of advanced verification tools – many of which are not available to news journalists across the EU at this moment (e.g., Thomson et al., 2022).

Second, future research should provide more insights into the relationship between concerns and competences in relation to disinformation and the professional background of journalists. While we found that the employment status of EU correspondents hardly plays a role for any sort of concern over disinformation and their own competences to detect disinformation, our findings show that those working for traditional media are slightly more concerned about disinformation as a threat to the profession than those working for digital media. This suggests that having work experience in a context where disinformation is more likely to be present makes the problem less abstract and more manageable from a professional point of view (see Ekström et al., 2020). Our findings related to the competences of EU correspondents lend further support to this assumption: journalists working for digital media tend to feel more confident when it comes to detecting disinformation than their colleagues from traditional media. We require more research into the strategies and skills journalists have developed to cope with the increase in disinformation and what role contemporary journalism education plays alongside training and resources provided by media organisations at home. For EU journalism profession-wide training programmes, in which more versatile journalists share experiences and best practices with respect to verification methods, may be particularly beneficial. Such programmes must be coordinated by professional organisations independent of governmental and state actors. The EU should support them financially; similar calls have already been expressed by the European Federation of Journalists. ⁹

Third, we show that, although journalists are rather worried about the spread of and citizen exposure to disinformation, EU correspondents feel rather confident about their own abilities to detect disinformation. This provides evidence for the applicability of the so-called third-person effect (e.g., Corbu et al., 2020; Lee and Coleman, 2020): EU correspondents consider disinformation to be a problem for the profession and European citizens but are less worried about their own abilities to deal with disinformation. This seems to be the case for journalists across the board as there are no differences between journalists working for traditional and digital media when it comes to their concerns over exposure to disinformation by EU and national citizens. Thereby, we add another building stone to the systematic empirical examination of the so-called ‘disinformation crisis’ in mediated democracies. Although research shows that, due to their professional training, journalists should react differently to media influences (e.g., McGregor and Molyneux, 2020), such a discrepancy likely mirrors that of citizens, ascribing a high threat level to disinformation, but feeling that this threat can be mitigated in a digital media environment. Future studies may focus on studying the effects of exposure to disinformation indicators on both journalistic and citizen behaviour during the news production and news use process.

Fourth, our study revealed that different levels of press freedom in journalists’ home countries affect the extent to which they believe that both EU citizens and fellow nationals are affected by disinformation. Here, journalists from countries with comparatively high levels of press freedom tend to be less concerned than journalists from countries where press freedom is more constrained. Although we have to be cautious with their interpretation given that we have few observations from countries in which press freedom is severely under threat, the results suggest that journalists are aware of the (increasingly) problematic situation in their own countries: restrictions to press freedom can be the result of various factors, including harsher media laws, restrictions in media ownership, hindering media organisations in their work as seen in violent attacks by protestors against journalists, and threats against individual journalists. All of these increase the likelihood for self-censorship (e.g., Gerli et al., 2018) and the erosion of journalistic autonomy (e.g., Císarová and Kotišová, 2022) and provide more space for illegitimate news to spread (Bayer et al., 2019). Our initial results are therefore plausible and strongly suggest the need for more research on the interaction of individual journalistic perceptions of skills and independence in reference to disinformation (see also, e.g., Saldana and Vu, 2022).

Reporters Without Boarders (2021) warned that press freedom is increasingly at risk in Europe as even some of the traditionally high-ranked countries saw their rates dropping during the COVID-19 pandemic, such as Germany and Luxembourg. It was also the first time that European Commission President Ursula von der Leyen focused on media freedom in her State of the Union speech on 15 September 2021 by calling on EU member states to improve journalists’ safety. ¹⁰ Since then, the pandemic and the Russian war against Ukraine have exacerbated the threat that disinformation poses to EU democracy. While we still wait for more academic evidence on how this has exactly impacted (EU) journalism, we can assume that fears about the spread of disinformation, as well as actual disinformation, have increased (e.g., Erlich and Garner, 2023; Hanley et al., 2022), which makes a research enquiry into the consequences for the work of EU journalists even more pressing. These crises have also led more citizens, incited by populists and conspiracy theorists in times of crisis, to challenge the legitimacy of legacy media across Europe (Egelhofer et al., 2022; Wright, 2021), which has led to violence against journalists and further erosion of press freedom in countries such as the Netherlands, Italy or Greece. ¹¹ Journalists play a central role in safeguarding the EU’s accountability and democratic legitimacy (Frangonikolopoulos and Papapadopoulou, 2021: 1124). Our results suggest that journalists are aware that declining press freedom and the threats posed by disinformation are interlinked, and we therefore recommend that both EU policy-makers and journalism scholars pay more attention to their interplay.

In sum, our study represents a new avenue of research for disinformation studies in the EU context, which has previously spent much time focusing on extra-journalistic fact-checking organisations and citizen resilience, thereby neglecting the challenges professional journalism faces when aiming to stop false content.

Supplemental Material

Supplemental Material - Disinformation and the Brussels bubble: EU correspondents’ concerns and competences in a digital age

Supplemental Material for Disinformation and the Brussels bubble: EU correspondents’ concerns and competences in a digital age by Sophie Lecheler, Katjana Gattermann and Loes Aaldering in Journalism