“Are Journalists Traitors of the State, Really?”—Self-Censorship Development during the Russian–Ukrainian War: The Case of Latvian PSM

Abstract

Media self-censorship related to war and military conflict is usually analysed by evaluating the journalistic practices of the countries involved in the war. The objective of this study is to explore how the self-censorship of Latvian public service media (Latvian Radio and Latvian Television) employees developed in response to changes in the internal socio-political discourse after Russia’s full-scale invasion in Ukraine, because of which the Latvian PSM found themselves in the crossfire of long-term criticism and attacks by politicians and audience representatives. Employing semi-structured interviews and qualitative content analysis, this study analyses the perception of self-censorship at all levels (journalists, producers, programme hosts, editors) (15 informants), factors that influence the development of self-censorship, informants’ coping strategies, and the impact of self-censorship on PSM content (78 items of content on various channels and platforms). The conceptual framework of this study is based on Bourdieu’s field theory and Spiral of Silence Theory, exploring how self-censorship affects journalists’ professional habitus, social capital, and agency. The results of this study show that, although Latvia is not involved in the nearby war, politician- and audience member-driven self-censorship affects PSM platforms’ daily agenda, source selection, and editorial line, reducing the diversity and pluralism of PSM content.

1. Introduction

Self-censorship is present in all levels of relationships where the assumed price of information disclosure is higher than the individual or common good achieved if the information were published (Bar-Tal 2017). Self-censorship in the media is connected to basic issues regarding freedom of expression.

Normally, media self-censorship is approached from the viewpoint of media representatives’ self-perception. Research analysing the impact of self-censorship on current content is much scarcer. Our study focuses on the comprehensive and deep-rooted change in Latvian PSM (Latvian Radio (LR) and Latvian Television (LTV)) employees’ perception of self-censorship after the beginning of Russia’s war in Ukraine and analyses war-related PSM content. Back in 2021, Latvian journalists did not find the problem of self-censorship topical (Rožukalne and Ozoliņa 2022, 27 May). Soon after Russia’s full-scale attack on Ukraine, Latvian PSM journalists began expressing publicly and explicitly (Krejere 2023, 9 June) that their job is influenced by self-censorship brought on by high-level politicians (Brauna 2022, 4 April) and national-conservative-politics-supporting social network users whose communication techniques correspond to those used by a digital “trolls army” and populistic political communication activities (Brauna 2023, 18 May; Ozoliņa and Rožukalne 2022, 4 October). Though the Latvian PSM have clearly defined their editorial and personal stance and condemnation of the aggressor country, media representatives have frequently been accused of supporting war, called “kremlins” and “the fifth column”, and have had their loyalty, patriotism, and professionalism doubted.

These trends were later recognised in an open letter by the Editorial Council of Latvian Radio (Latvijas Radio 2024, 5 April), expressing concern about the limitations of freedom of speech caused by constant attacks on journalists and the politicisation of public media reform. This letter was discussed in an event organised by the President of the State (Latvijas Valsts Prezidents 2024, 8 April). The concern about press freedom was also supported by the head of the Baltic investigative journalism centre “Re:Baltica” Sanita Jemberga (2024, 11 April), emphasising that politicians are increasingly turning against journalists, “filing complaints” with various law enforcement institutions.

This situation mainly refers not to the performance of the media, and PSM in particular, but to the long-standing “salient cleavage” in Latvia between ethnic Latvians and the Russian-speaking part of Latvia’s society (Auers 2013, p. 92), which can be observed in the structure of political parties, in voters’ preferences, and in cultural and ethnic identities. This cleavage reflects the consequences of the Soviet occupation, when the ethnic structure of Latvia changed after the influx of a large number of residents of other Soviet countries after the Second World War; many years later, some of the Russian-speaking inhabitants did not support the restoration of Latvia’s independence. Thus, in the context of the crisis caused by Russia’s full-scale invasion of Ukraine, certain political forces use problems of the disintegrated society (Auers 2023) to attack professional journalists with populist arguments.

The objective of this study is to analyse which aspects determined the change in the self-censorship climate in PSM newsrooms in 2022 and 2023 and how the perception of self-censorship influenced the content Latvian PSM dedicated to the war in Ukraine.

Latvian PSM platforms are the country’s largest media organisations. Their work is addressed to a linguistically divided audience, since one-third of the population uses media in Russian. The Latvian PSM are insufficiently funded, but the audience’s trust in them is slightly higher (51%) than the EU average (49%) (Eurobarometer 2022). The audience highly values PSM platforms’ contribution to cultural development, whereas the assessment of democracy is insufficient (Jõesaar et al. 2022) as the PSM encounter challenges targeting minority and youth audiences.

The political influence on PSM content and operation has been found regularly in academic studies (Beitika 2016; Dimants 2022; Rožukalne 2013). It has manifested itself in various forms, from chronic underfunding (SEPLP 2022a) and politicisation in the appointment of the supervisory body or the board members to judging media personnel and their management in “oligarchs’ conversations” (Ir 2017, 29 June).

The constant political influence on PSM platforms’ activity can be explained by historical factors that have influenced the political culture of Latvia, as well as the understanding of politicians and the part of society in the role of the media in democracy. LTV and LR are the heirs of broadcasting organisations from the time of Soviet occupation (1945–1991). Since the end of the 1980s, they have played a vital role in setting the mood of Latvian society that, after the long occupation, it is possible to regain an independent state. After regaining independence, former Soviet media employees and communist party functionaries were also involved in the process of the transformation of former “state” media into public media and their supervising institutions. In the Latvian PSM development, political influence can be defined as an indirect intervention implemented ideologically through structural weaknesses of the public media at the systemic level, related to the supervision and management system and the financing system (Beitika 2016, p. 246).

The PSM’s development partly reflects Latvia’s media system, which has been described as hybrid liberal (Dobek-Ostrowska 2015) or, using Hallin and Mancini (2004) models, hybrid and suffering from the lack of a dominant paradigm (Skudra et al. 2014). Elements of liberal (the free market as the primary regulating force), democratic corporate (weak media professionalisation, attempts to strengthen PSM), and polarised pluralistic models (researchers especially stress the relatively strong political parallelism) are characteristic of the Latvian media environment.

The relationship between the PSM and the audience is complicated. Jānis Juzefovičs (2019) concluded that the use of traditional media has been decreasing since 2010 in Latvia, while public support for PSM is increasing. The researcher assumes that the increase in support among the loyal audience can be attributed to the geopolitical crisis and upheavals in the information environment in recent years in Latvia and globally, which have increased concerns about the manipulation of information and led to the political polarisation of society. The author also points out that regarding the use and perception of PSM in Latvia, not only ethnolinguistic and age differences can be observed, but also a socio-economic gap (Juzefovičs 2019).

Latvian PSM have difficulties coping with the digital transformation-driven diversity of potential audiences, especially in the context of increasing media personalisation (fragmentation), Juzefovičs (2019) emphasises. This can seriously threaten the legitimacy of the media in public perception in the long term. A 2022 study (SEPLP 2022b) also showed that part of the audience, firstly, does not recognise the content of PSM, and, secondly, associates it with the influence of the “state” and therefore distrusts it.

After Russia launched its war in Ukraine, the Latvian political and social discourse regarded the geopolitical issues as “our war” and believed that “Ukrainians are also fighting for our freedom” (LV 2023, 24 February). PSM journalists went to Ukraine regularly and provided information from the warzone, representing the patriotic and peace journalism approach (Ginosar 2015; Mazahir 2013).

Conceptual Framework

This study on professionals’ self-censorship in Latvian PSM uses the definition by Bar-Tal (2017) that self-censorship is “a deliberate and voluntary suppression of information if no formal obstacles are placed” (Bar-Tal 2017, p. 60). The definition is supplemented to fit media job specifics: self-censorship in the media is characterised by a practice of deciding to avoid publishing information partially or fully available to the journalist, though editorial considerations and/or professional principles include the duty to provide information or opinions important to public.

The free exchange of information plays an essential role in ensuring freedom of speech, the transparency of public processes, open and critical decision-making, the development of tolerance, and the creation of open relations between various groups of society, as per Bar-Tal et al. (2017). However, Bar-Tal (2017) recognises that self-censorship is sometimes necessary to maintain security, public welfare, and the protection of certain groups. This refers also to cases of conflict when, to avoid societal polarisation and dysfunction in different societies and contexts, arguments justifying self-censorship arise (Bar-Tal et al. 2017; Sharvit et al. 2018).

The analysis of the situation of the PSM’s activity led to the definition of the following research problems:

• Latvia is not a participant in a military conflict. Therefore, the PSM’s situation differs from the media self-censorship described in studies (Robinson et al. 2009; Springer et al. 2022) explaining self-censorship arising from warfare or a state’s participation in military conflict. Latvia, however, is located relatively close to the territory of war, it is involved in supporting Ukraine, and it suffers the social, political, and economic consequences of the geopolitical crisis. Thus, it is essential to understand if and how PSM journalists and editors interpret perceived influences (individual, organisational, societal) in relation to self-censorship in the situation of a nearby military conflict-related crisis.
• Although usually media self-censorship in Eastern European countries is related to media owners’ or leaders’ influence (Schimpfössl and Yablokov 2020), in the context of the war in Ukraine, Latvian PSM’s journalists link the rise in self-censorship to external circumstances—personal or public attacks by various players. We are therefore interested in the dynamic of self-censorship development in PSM newsrooms after Russia’s invasion in Ukraine and how self-censorship impacts the content of PSM.

To describe the research problems, we use two theoretical approaches. Bourdieu’s field theory is useful for understanding journalists’ self-perception (Bourdieu 2005). For self-censorship development analysis, the Spiral of Silence Theory (Noelle-Neumann 1993) is used to see how self-censorship influences journalists’ habitus and the role of the agent. We are interested in whether and how compromising the set of rules and practices essential to journalism, accumulated over a long time, leads to self-censorship and affects the content offered to the public.

Based on selected theories and the created definition, we have set the following main research questions:

What perceived influences determined the perception of self-censorship in the Latvian PSM at the time of the war in Ukraine?

How is self-censorship influencing the representation of war in the PSM?

We contribute to the discussion of whether censorship can exist outside clear legal and political censorship regimes. We intend to add to the studies that examine the development of self-censorship in the context of today’s multi-layered geopolitical and democracy crisis, which affects/changes the professional self-perception of journalists and the role of journalism as a discursive institution in society. Furthermore, our research explores the perceived influences of self-censorship and attempts to assess their impact on the content produced by journalists. This study involved interviews with PSM employees and a content analysis of the PSM’s content related to war.

2. Literature Review

Self-censorship is a characteristic of authoritarian regimes. In democratic societies, self-censorship arises in situations of war and conflict where journalists’ work can tread on a group of society or a general opinion. In such cases, self-censorship comes from fear of social sanctions imposed by persons or groups who do not wish the information or opinion to become public. In political discourse, the request to restrict information is often linked to security or state integrity interests.

Examples of self-censorship by media professionals created by military conflicts and major crises have been analysed in the assessment of the adaptive strategies of journalists in Russian-occupied Crimea to Russian-imposed media censorship after 2014 (Zeveleva 2020); in the publishing of caricatures of the prophet Muhammad in the Danish press (Cook and Heilmann 2013); and in journalists’ attempts, through self-censorship, to mobilise society during the war in Lebanon (Bar-Tal 2017).

2.1. The Individual and Social Contexts of Self-Censorship

As opposed to censorship with its precise definition, self-censorship is much harder to identify as it happens covertly (Kis 1986; Reese 2001). It is created by a set of various internal and external (social, political, economic, cultural, organisational, individual) circumstances. Self-censorship is researched more at the level of family or an organisation but not enough at the level of society, even though it impacts mass media as culture agents (Bar-Tal 2017; Afifi and Steuber 2009).

We study the causes of self-censorship in a situation where media professionals do not have to adapt to formal restrictions but adhere to pressure similar to external censorship. Thus, media professionals come up with unwritten rules to be able to continue working professionally (Schimpfössl et al. 2020; Schimpfössl and Yablokov 2020; Zeveleva 2020). In such cases, the external influence is a diffuse, decentralised form of power (Bunn 2015), created by private actors rather than external, repressive power from the state.

Analysing self-censorship at the level of media and in a social–cultural context, we view media as creators of culture, considering the central social theory question that requires a link between human agency and social structure, developed in Pierre Bourdieu’s (1986) field theory, in which the relationships between media as a field and other cultural production fields are analysed (Hilgers and Mangez 2015). Journalism is a field on its own acting upon other fields, and its actions have far-reaching social and political consequences (Bourdieu 1996; Couldry 2003; Fligstein and McAdam 2012).

We chose Bourdieu’s field theory to describe the multilevel, dynamic relationship that affects the professional field of journalism and the journalistic habitus. In Bourdieu’s view, social reality is made up of several fields of social activity, or separated but interconnected fields (Bourdieu 1993). Each field is characterised by the logic of internal operation and the degree of autonomy. The existence of each field is linked to changing relations between fields, which are determined by the agents in each field and their social positions. Under the influence of other fields, the available resources of each specific field and the opportunities to dominate in interaction with other fields may change. According to Bourdieu (2005), journalism is a relatively autonomous field of social activity which, like other fields, is characterised by internal laws and functioning logic, but it does not exist and cannot be evaluated as independent of external laws and influencing processes. Journalism is a part of the structure of the cultural field analysed by Bourdieu, the existence and position of which can be evaluated in wider social power structures (Bourdieu 2011). By illuminating both the visible and invisible journalistic processes that determine what is reported and how it is reported, Bourdieu sees journalism as occupying a dominant position in the cultural field because it holds a monopoly on the means of large-scale information production and dissemination (Bourdieu 1993).

The basis of Bourdieu’s theory is the concept of habitus, which at the same time determines the result of both the structure of society and can be perceived as a promoter of changes in the structure of society. To characterise the field of journalism and its relations with other fields, the journalistic habitus is explained (Bourdieu and Wacquant 1992) as a set of specific rules and practices accumulated over a long period of time, as an understanding of the rules of operation and unique functions creating specific forms of symbolic capital. Journalistic habitus is characterised as dependent on practice or role, or the perception of one’s role. When developing the understanding of habitus in journalism, Meyen and Riesmeyer (2012) include both socio-demographic socialisation and current operating conditions, as well as operational goals, professional ethics, audience understanding, and media effects.

We are interested in the dynamics of the journalistic field in the context of geopolitical events. Bourdieu’s habitus concept, which refers to the “rules of the game” or “sense of the game” (Bourdieu and Wacquant 1992, p. 128), is used for understanding professional self-perception in relation to autonomy and agents’ struggle for dominance in “the battlefield” (Bourdieu and Wacquant 1992, p. 17). Journalism habitus is defined by relationships between fields; they are unstable and in constant flux between settled and unsettled rules (Fligstein and McAdam 2012; Zeveleva 2020).

Regarding journalists, Bourdieu interprets self-censorship as a fear of losing habitus, which characterises their work rules and their internalisation in professional routines and practice (Šulmane 2011). Habitus includes media professionals’ symbolic capital, i.e., it places journalists’ work in the societal structure between politics, economics, and culture. Although the journalistic professional field is dynamic and open to external influence, media professionals try to preserve long-standing journalistic values (Papanagnou 2023), which form the basis of habitus.

During times of change, new requirements influence the habitus, forcing a reassessment of the real and imagined symbolic culture capital of a journalist (Bourdieu 2005). In the process of structural and social changes, important agents like journalists who choose to remain silent or restrict their activity can face the risk of losing a part of their previously accumulated social, cultural, and economic capital (Šulmane 2011). The actor and agency approach provides the opportunity to analyse not only the journalism field but also the changes in media professionals’ agency (Papanagnou 2023). We use Archer’s (2003) idea that the role of an agent is not only on the plane of social and cultural influence, but that a personal agent’s role is active and reflexive to a much greater extent than it is passive and instrumental.

Media professionals’ self-censorship characterises a situation where a gatekeeper restricts free information flow (Harcup and O’Neil 2017; Shoemaker and Reese 2014). If self-censorship manifests in a journalist’s decision not to publish important content or a part of it (Cook and Heilmann 2013), this determines the social influence of self-censorship. This study focuses on the formation of self-censorship in PSM, whose independence is protected by a duty defined in the regulation (Puppis and Ali 2023) to reach different groups of society and encourage the free exchange of information.

The self-censorship dilemma is described as a process when a person knows they have access to vital information, but its disclosure could overstep another principle, dogma, norm, ideology, or value (Neuwirth et al. 2007; Noelle-Neumann 1993). The decision to self-censor always comes from an assessment of pros and cons and gains and losses in case the information is disclosed. If the losses outweigh the gains, a decision in favour of self-censorship can be taken (Bar-Tal 2017).

In our study, we seek answers to the question of who censors what. In media practice, content creation takes place collectively, but the journalist must make many decisions individually. This raises the question of private and public self-censorship (Cook and Heilmann 2013). If a society has agreed on the need for self-censorship and public pressure is exerted, individuals can internalise the influence of public censorship and, as a result, censor themselves. In our study, we analyse situations where no formal censorship exists, and bring up the question of what social norms can lead to other norms acquiring the status of censorship.

2.2. Self-Censorship and Attacks on Journalists

Analysing the growing number of attacks on journalists in the online and offline environment, it shows that attacks have spread from conflict-ridden environments to prosperous and democratic countries (Baroni et al. 2022). Attacks have acquired an everyday characteristic. They diminish the quality of journalism and restrict media freedom (Miller and Lewis 2022). Offline and online attacks are frequently experienced by journalists who tackle controversial issues; the most frequent victims are women and representatives of ethnic minorities. Attacks can be verbal aggression, cyber-attacks, physical assault, and stalking or persecution (Baroni et al. 2022). A Flemish journalist survey concluded that one in six journalists had experienced attacks by politicians, entrepreneurs, or citizens. In the USA, contempt toward the media is expressed by the phrases “fake news” or “lamestream media”. In Germany, the word “Lügenpresse” or “lying press”, which has been known since the Third Reich, has been brought up again. The term appeared in public discussion when German media started to publish content on the annexation of Crimea in 2014 (Koliska and Assmann 2021).

According to the “Worlds of Journalism Study” data on Latvia, one in three Latvian journalists encounters insults, condemnation, and public discrediting daily (Rožukalne and Ozoliņa 2022).

2.3. Self-Perceived Influences and Self-Censorship

Since we are interested in the influence of self-censorship on PSM content creation, we must separate objective influences and self-perceived influences on journalists’ content. Objective events do not always evoke the desire of a journalist to nix their duty to publish important information (Jungblut and Hoxha 2017). External influences on the decision to self-censor are not direct, immediate, or linear, as objective events or processes do not always align with their perception.

Self-censorship motivated by the social environment is explained in the Spiral of Silence theory (Noelle-Neumann 1993). Its core idea is that a society or group isolates individuals who deviate from accepted norms in their opinions, statements, or other actions. It is particularly difficult to assess the audience’s perception or public opinion and conclude, based on separate critical remarks or even attacks, that the audience does not support the journalists’ interpretation of a certain topic. For fear of social isolation (Noelle-Neumann 1974, 1993), the journalist can decide in favour of self-censorship and avoid publishing certain perspectives (Neuwirth et al. 2007). Journalists can encounter similar hardships concerning editorial stances (Skjerdal 2010). These risks impact the individual’s opportunity to express their opinion if it does not conform to the opinion of the majority (Moy et al. 2001). Therefore, social pressure defines external influences, whereas assumptions about the climate of the external environment can determine the individual’s reaction to social pressure.

In communication science, the study of the impact on journalism includes multiple levels of influence associated with a hierarchical structure. They are measured as objective impact generators and typically include individual, organisational, and societal impact levels. Some authors, evaluating the impact on journalists in detail, link them to the levels of practice and the ideology or system (Shoemaker and Reese 1996), cultural forces (Preston 2009), or international influence (McQuail 2000).

Hanitzsch et al. (2010) propose a model of self-perceived influence, emphasising that journalists’ own perceptions of the influence on journalism ethics do not necessarily correspond to models of objective influence. We use Hanitzsch et al.’s (2010) conceptualisation of self-perceived influences on journalists’ work to find out how the perception of the external environment influences the decision to self-censor as well as the strategies for overcoming it. By reviewing six journalistic influences which can produce self-censorship as defined by Hanitzsch et al. (2010)—professional, procedural, organisational, reference group-based, economic, and political self-censorship—we use three for data analysis: organisational, reference group (colleagues), and political influences.

3. Method

This study employs two methods of data collection and analysis: semi-structured qualitative interviews with PSM employees and qualitative content analysis of content in the PSM related to the war in Ukraine.

The interview informant sample was created to represent the self-censorship perception of PSM employees involved in different levels of content production. The group of the selected respondents reflects the situation characteristic of Eastern and Central European countries, i.e., that the structure of media employees is mostly feminine (Šulmane and Uzule 2018). The positions of Latvian PSM content creators, from the management of news services and commissioning editors to the chief editors, channel directors, and programme hosts, are filled by women, while the interviewed investigative journalists and discussion hosts are men.

The sample includes Latvian and Russian-language editorial offices of all Latvian PSM platforms (Radio, TV, and digital): journalists and investigative journalists, news producers, news and discussion programme hosts, and editors of different levels. Five informants represent Russian newsrooms: two respondents work at the Latvian Radio channel LR4, and three respondents represent a multimedia newsroom (which prepares content for LTV7 and rus.lsm.lv) and the management of the newsroom of rus.lsm.lv (the PSM’s digital platform in Russian). Selective sampling was used to pick the informants, since the participants must have personal experience regarding the research topic and must be able to express their opinion on it (Mārtinsone et al. 2016).

In this study (2022 and 2023), we conducted 15 qualitative semi-structured interviews, obtaining informed consent beforehand. The interview structure consisted of the questions included in the WJS global study methodology (socio-demographic data, working condition, independence and autonomy, understanding of professional roles and ethics, etc.) and a set of questions about self-censorship (understanding of self-censorship, evaluation of its internal and external causes, reaction to self-censorship in individual and at the level of the organisation, employees’ coping options and mechanisms, assessment of the impact of self-censorship on the content) (see Appendix B).

To process the interview data, reflexive thematic analysis was used (Braun and Clarke 2019). A code map of the interview questions/themes was created to conduct the thematic analysis of the interviews. For each question/theme, sub-codes were created through which all interviews were analysed. In the Section 4, the informants are anonymised (Appendix A).

To assess the quality of content dedicated to the war in Ukraine, the Public Electronic Mass Media Council (PEMMC) ordered an independent expert evaluation in April of 2022. It was based on a code table (see Appendix C) created using the criteria of qualitative content analysis and assessed the compliance of topics, information sources, stances, genre diversity, and content with the core professional principles and ethical norms of professional journalism. Given that part of the content analysis (LR, LR4, LTV7, rus.lsm.lv, FB accounts of PSM) was carried out by one of the authors of this article (further Expert 1), the other review (LTV1, rus.lsm.lv, TikTok accounts of PSM) was used as secondary data (Sprudzāne 2022) (further Expert 2).

The content analysis sample was created based on random selection, analysing Ukraine war-related content from 24 February to 24 March 2022, by evaluating the informative and analytical programmes on every Tuesday every week in April and May of 2022. The content analysis was carried out on 78 content units of PSM (news and current affairs programmes, interviews and discussion programmes, documentary films, social media posts in Latvian and Russian on Facebook and TikTok).

To find out whether self-censorship influenced the decisions related to the content, the results of the content analysis on different PSM channels and platforms were compared with the conditions mentioned in the interviews, which could implicitly indicate self-censorship in the choice of topics, sources, and opinions.

The content of the following channels and platforms was evaluated: LTV1 (Morning Panorama, one special news release, daytime news, Panorama, Today’s Question, and night time news), LR1 (3 news releases from 6:00 a.m. to 11:00 a.m., lunch programme, and afternoon programme from 4:00 p.m. to 6:00 p.m., including news, news in plain language, and/or a review of daily events), LR4 (3 news releases from 6:00 a.m. to 11:00 a.m., the programme Ceгoдня [Today] at 1:00 p.m., 3 news releases from 12:00 p.m. to 6:00 p.m., the programme Обзoр дня (Daily events), rus.lsm.lv (including the news feed of the given day, multimedia programmes Сегoдня (Today), Сегoдня вечерoм (Evening Today), TЧK (frequently a special edition of news during the analysed period), and the social media accounts of this platform. It is important to note that rus.lsm.lv integrates a large part of the content broadcast in Russian on linear platforms; its analysis includes news and discussion programmes broadcast on LTV7 and LR4.

4. Results

This chapter describes the dynamics of the self-censorship perception of PSM employees, the internal (editorial) and external (audiences, institutions, politicians) drivers of self-censorship, strategies to overcome self-censorship, and the impact of self-censorship on editorial choices. Finally, content analysis data on the PSM’s content covering the war in Ukraine are compiled.

4.1. Perception of Self-Censorship

Self-censorship in media editorial work is identified in the context of psychological, professional, and managerial considerations. Regardless of work experience and professional position, PSM employees characterise self-censorship as a disturbance hindering decision-making. It is an all-encompassing sense of insecurity and a fear of publishing specific topics and opinions or choosing information sources. Self-censorship is mostly linked to consequences that could arise after publication. It restricts journalists’ autonomy and editorial independence. The development of self-censorship is influenced by various crises which force a re-evaluation of the core principles of journalism and professional routines. The nearby war in Ukraine is identified as a new and deep-reaching crisis.

Informant 2: “The war was something new for all of us, we didn’t know what to do. The main concern was how to work so that we don’t drive the enemy and so that we don’t lose our humanism-based position. I come from news journalism and if it were possible, I would interview Putin. Even though many of my colleagues think that it shouldn’t be done”.

Informant 1: “It means I am much more careful in using my words, accents, phrases, because I am worried about what reactions could arise. It is a huge sense of insecurity, a fear of making a mistake”.

PSM representatives analyse self-censorship as a problem that has developed gradually since the annexation of Crimea in 2014 and after Russia’s full-scale invasion of Ukraine in 2022. Informants perceive self-censorship as a dynamic phenomenon whose presence is noticed daily.

Informant 7: “When we began broadcasting [in Russian] in 2014, we were much more courageous. We weren’t afraid. Yes, there were complaints, but we’ve since become much, much more cautious”.

Self-censorship is related to controversial socio-political issues, such as the relations between representatives of different ethnicities and cultures in Latvia, the usage of the official language, and the society’s attitude toward the Russian war, migration, and culture.

Informant 1: “It began during the pandemic when we tried to back up each word with specialist-provided information. Currently, it is much harder, as it is all about opinions. People are concerned about these issues, like the Russian culture or the renaming of Soviet-era streets, they discuss, but we report too little on it”.

The informants characterise self-censorship as a reaction to judgement of their job, of them as an individual, or of PSM in general.

4.2. Self-Censorship Promoting Factors

External factors promoting self-censorship can be divided into two groups: audience and politicians (sometimes officials). The audience’s influence manifests itself in media users’ comments, direct messages, and personal attacks.

All informants mention comments on the platform X (formerly Twitter) as the most common format of attacks. The comments discuss the journalists’ professional activity, private life, and appearance. Informant 11 emphasises that these attacks seem so loud as the other party keeps silent for fear of being “nailed to the Twitter shame pole”.

PSM platforms as an institution are the recipient of more frequent and more severe attacks than individual journalists. Attacks on the WhatsApp platform are on the rise. Implicit threats and hate speech appear in texts.

Informant 12: “You’re being sent pictures of killed Ukrainian soldiers with a text “You’re next”!”.

The viewpoint of leading employees and journalists differs. A journalist (Informant 11) says that “attacks are no longer private; they are carried out by large groups of people”. PSM management representatives admit that the problem is topical, but they encounter it rarely. An editor (Informant 13) stressed the following: “I myself rarely encounter self-censorship and I give great freedom to my colleagues”. Leaders emphasised that they cannot know for sure if a journalist can self-censor without anyone noticing.

Several informants think that politicians’ public attacks have a greater impact and encourage audience representatives to turn against journalists. This especially concerns Russian-language PSM newsrooms.

Informant 5: “A certain political party is not satisfied with our newsrooms work in general; they don’t agree with the existence of our [Russian-language channel]”.

The interviews also show the media supervisory institution’s attitude as a determining factor in the constant presence of self-censorship.

Informant 7: “PSM supervisor’s attitude toward us is objective, most of the council supports us. But the chair of the media regulator [National Electronic Mass Media Council] follows us closely, he has a personal dislike toward us. There used to be complaints about every interview for no reason. You know, it’s horrible, simply horrible”.

Informant 9 explains that politicians create fake accounts to create an illusion that a big part of society supports them. However, officials are also not afraid to speak in their own name. Politicians can disturb journalists’ work, hinder the flow of information to society, and actively turn against journalists. Informant 12 describes politicians as the most active fuellers of hate toward journalists.

Informant 12: “I think in Latvia the greatest concern is that politicians and opinion leaders, for the sake of their own selfish goals, heat up the atmosphere and turn against the media as the root of all evil. At one point, someone will lose their mind and shoot someone live, like in the USA, they’ll come into the television [building] and shoot everyone they see”.

Respondents conclude that politicians’ impunity has no consequences. One of the respondents mentioned helplessness in the fight against politicians’ public attacks on PSM:

“We are a sort of a soft target. You can attack us; we cannot ignore you. We must give word to all political forces. We also cannot fight dirty”.

However, the audience’s criticism makes PSM employees evaluate their work using professional criteria. The fact that someone is displeased by a topic or opinion does not create self-censorship. It is driven, rather, by personal attacks, blaming journalists, accusing them of not being patriotic enough, or stating that varied information is a risk to the security of the state.

Informant 2: “We are “kremlins”, Putin’s servants, threats to national security, representatives of state enemies. We must worry not just about every word, but every letter. We were one of the first who transitioned to the spelling of Ukrainian cities in accordance with the Ukrainian pronunciation. But God forbid, a guest on a show accidentally says Kiev instead of Kyiv, there will be a wave of new complaints and attacks”.

Russian-language editorial office representatives add politicians’ attitudes and changes in media politics, which jeopardise the future existence of PSM’s Russian-language service, to the factors promoting self-censorship.

Informant 12 thinks that the war in Ukraine has substantially changed the work of Russian-language media. She says that the Latvian Radio (LR) board’s support for Russian-language media is low. The minority audience in Latvia is very “split, multi-layered, and diverse”. LR4 receives a lot of attacks and criticism from the audience. Many of the attacks include politicians’ opinions that Russian-broadcasting PSM channels must be restricted or shut down.

Informant 7: “Some say that it’s time to stop paying for this pro-Kremlin organisation [Russian-language PSMs’ editorial offices]. For some, we are pro-American sell-outs, but for some Latvians, we are pro-kremlins with a totally pro-Kremlin narrative”.

Commenting on the internal factors of self-censorship, some informants said that there could be sensitive issues which are decided not to be talked about to avoid further attacks. More often, they talked about the refusal of information sources to participate in programmes or express an opinion because they do not wish to be publicly attacked and humiliated. The self-censorship of sources also drives journalists’ and editorials’ self-censorship.

Informant 7: “Our sources feel the pressure. People more and more often refuse to speak Russian to us on air. They excuse themselves and say I can, but I won’t because someone will screenshot it and put on social media with the caption “speaks the language of Rasha”. Nobody wants to be Twitter-famous for several days, up to their neck in the dirt”.

However, the self-censorship issue is taken seriously and decided upon in daily work at different levels of the PSM’s newsrooms.

Informant 3: “I disagree that we promote self-censorship on an editorial level. On the contrary, I try to encourage and convince [others] to speak freely and bravely about complicated issues”.

Journalists and editors have observed that, for the second year, mostly anonymous users have been acting in the style of a troll factory and trying to discredit the PSM. The informants analyse the political attacks in detail, they know their causes and actors, and they recognise representatives of different parties as aggressive anti-media discourse representatives.

Informant 5: “There are forces who work in accordance with Kremlin methodology, who keep “yanking” us in personal entries on Facebook and in comments that we don’t reflect the real issues Russians face in Latvia, that we are a government-paid media. They try to humiliate us, call us sell-outs, and liken us to Russian propagandists—Kiselyov, Solovyov. They purposefully denigrate public media. These are tactics and politics built on schemes of other countries—to attack PSM. These parties purposefully discredit public media, this is also done by activists ‘fed’ by these parties and working on social media platforms”.

Informant 2: “The comments show a clearly politicised wing that has created a situation that we are insulted constantly”.

Self-censorship in the PSM has been observed for a long time; it is perceived differently at different employment levels, and the strategies for overcoming it also vary.

4.3. Self-Censorship Coping Mechanisms

The most common sign of self-censorship is silence to protect oneself from new attacks. It involves avoiding the audience’s comments, refusing legal help in case of public defamation, the desire to leave one’s job, and influence on content creation practices. Coping mechanisms for self-censorship can be split into three categories: resistance and support, silence, and apathy. Individually, a person chooses avoidance, but colleagues feel empathy and try to support them, for instance, by protecting them from information about attacks targeted at the person.

Informant 7: “To be honest, we don’t [fight self-censorship]. Maybe in separate discussions or topics. At some point, one gets angry—why are we afraid? We’ll respond to complaints if any! If there is no anger, we do nothing. We accept this as fact. And you think—maybe it’s best to just show kitties? The longer the war lasts, the longer there will be self-censorship, it will settle, and in the end—what if we come to censorship?”

Most interviewees acknowledge that the self-censorship context has both “yes” and “no” factors. “Yes” aspects mean that people feel pressure; “no” refers to professional self-respect and free expression of opinions.

Conversations with colleagues are mentioned as important in creating the sense that you are not the only one experiencing attacks. Seasoned journalists take the attacks lighter, whereas young journalists might even consider whether they should stay in the profession.

Informant 2: “I try not to appear on social networks. I could post the flowers of my garden and still get cursed at, harassed. Our opponents are bothered both by our words and the fact that we are brave enough to speak”.

Informant 12 observed that many of the attacks come from fake accounts. She concludes that there are fewer attacks on LTV than observations on social platforms suggest. To avoid attacks and self-censorship, professional procedures are changed.

4.4. Effects of Self-Censorship on Editorial Processes

Public denigration of journalists’ work and their professionalism changes the relationships in editorial offices. During this time, several colleagues, especially young journalists, and editors, have left their jobs, with one of the reasons being external pressure, humiliation, and attacks coming from the audience and politicians.

Informant 4: “I have grown a rather thick skin. I understand that the opinion of us forced upon us is not reasonable. But I think about how to present negative and proven information about Ukraine. Maybe I shouldn’t place news about corruption in Ukraine in the most visible spot? In my newsroom, a person asked for help because colleague couldn’t handle the commenters’ pressure. Now a journalist who wanted to and could have worked for the news, is archiving content”.

Media representatives create new practices and “terminologies”, signifying new routines previously not used in editorial offices, making self-censorship a self-explanatory part of everyday life.

Informant 7: “We evaluate all pros and cons [of a topic]. Sometimes colleagues convince me, try to stop me, say stop, but sometimes they say—no, I guess we need to self-censor”.

Informant 1: “I feel pressure; it is responsibility. I think we are poorly skilled in discussion about sensitive topics, we fear conclusions on which side we stand”.

In several cases, external pressure and editorial discussions have improved the work, influencing the editorial line.

Informant 4: “We improved headlines and other parts of content; they are more precise. But there are some topics where we are overly cautious. That we are tired of it being so hard to work is not a good argument if we want to, for example, report on the migrant crisis on the Latvian-Belarusian border”.

Self-censorship has influenced all important aspects of editorial work: information analysis, the choice of topics and sources, the presentation of information, and everyday management. It determines strategic decisions about the daily agenda of the PSM.

4.5. War in Ukraine-Related Content Analysis in PSM

This subsection presents the main conclusions of two experts’ reviews (Kažoka 2022; Sprudzāne 2022), which analyse the content of PSM related to Russia’s full-scale invasion of Ukraine, paying attention to trends related to changes in the domestic political discourse in attitudes towards polarising issues such as ethnic relations, the status of Latvian and Russian languages in public communication, attitudes towards coverage of events in Russia, etc.

The main conclusion of the two experts is that the content of the public media both in the first days of the conflict and in the later weeks has been diverse and developed according to the principles of professional journalism. This applies especially to the coverage of war events and the content produced by war reporters. Expert 1 explains the differences in war journalism, where values and ethical principles must be clearly defined, and new skills and concepts must be learned literally within a few hours. Even special “dictionaries” for unambiguous reporting must be developed. Journalists have to learn the basic principles of conflict-sensitive journalism and trauma-informed journalism and be able to apply them in practice.

The assessment did not identify editorial incompetence or breaches of journalism ethics. However, though this was not the goal of the assessment, data implicitly pointing to the presence of self-censorship were observed. It was observed in the choice of sources and topics related to war on the Latvian and Russian-language PSM’s platforms. Comparing LR1 (a Latvian-language Radio channel) to LR4 (a Russian-language Radio channel), LR1 had a lesser variety of topics and sources. The Latvian-language war-related content is dominated by official sources and experts.

This practice could be explained in two ways. On the one hand, Russian newsrooms have always noticed that their audience is more interested in Russia’s events. On the other hand, several cases were recorded (according to Informant 2) when the earlier approach to source selection (the Kremlin’s official press releases without context and information from Russian official news agencies, as their preparation does not require additional resources) was difficult to accept given the geopolitical circumstances. Moreover, the war situation forced a reassessment of whether the earlier assumptions that the Latvian-speaking audience is not interested in Russia’s events and that the Russian-speaking audience has special, distinct interests are still relevant.

Expert analysis finds that Russian PSM’s editorial staff (LR4, rus.lsm.lv), since the very first days of the war, have featured eyewitnesses and ever-new experts. Russian editorials’ representatives (LTV7 and rus.lsm.lv) more often start complex discussions and look for discourses of problems formerly not talked about in society. LR4 news also covers more events in Russia.

In the Latvian-language content, some important issues are not addressed. For example, content creators avoid speaking about the sense of threat felt by Russian-speaking residents of Latvia. Content targeted towards the Latvian audience does not include sources concerned about the splitting of society and talking about how tough and scary it is to be Russian at a time when the word “Russia” has become toxic. Some cases (a couple of interviews) on LR1 attempt to discuss controversial issues on how the attitude towards representatives of ethnic minorities, specifically Russians and citizens of the Russian Federation living in Latvia, could change. However, at the beginning of the war, the editorial offices for each language followed the existing professional routines. Both looked for ways to adapt to the new situation when media content was influenced not only by the course of the war, but also by the impact of the war on the understanding of domestic socio- political issues.

In some cases, criticism of sensationalised headlines appears in the review of LR1 content. For instance, the mentioned message is not found in the content of the article, and certain topics are covered one-sidedly. Analysing the framing of Ukrainian refugee topics on LR1, Expert 1 emphasises the risk of stereotyping when, for example, talking about refugees and their future intentions can promote the classification of refugees into good and bad positions, and indicates the possibility of the model minority myth development.

Differences are observed within topics one could label “Russian” topics. For example, LR4 news paid attention to events regarding citizens of the Russian Federation and beneficiaries of pensions in Latvia, whereas LR1 did not. Similar data were identified looking at other topics. Latvian PSM channels have insufficiently reported on the Russians of Latvia, their stances, and their condemnation of Russia as an aggressor, which is well reflected throughout the Russian-language PSM content.

Comparing LR1 with channels broadcasting in Russian, Latvian newsrooms more clearly express the editorial position in relation to the war. In the Russian-language content, more attention is paid to the position of Latvia’s Russians and their worries about being blamed for Russian aggression. The stance of Latvian minorities on war issues appears much more clearly on Russian-language channels. The news about the condemnation of the war signed by 120 Latvian Russian writers (rus.lsm.lv, 1.03.2022) and news on the protest of the employees of M. Chekhov’s Riga Russian Theatre at the Russian Embassy (rus.lsm.lv, 8.03.2022) have not been sufficiently reported in the Latvian-language media.

In Latvian newsrooms, war reporting is on the social and political agenda, whereas the Russian-language content, which also includes general political and war news, prioritises the “everyman” perspective. Evaluating discourses, Expert 2 (Sprudzāne 2022) emphasises the difference in emotional attachment. Contrary to LTV1, rus.lsm.lv journalists let the interviewees focus on witnesses’ and Ukrainian residents’ emotions about current events. Expert 2 opines that LTV1 and rus.lsm.lv lack reportages (which are an integral part of war journalism), as well as sequential information about frontline locations and their direction/movement.

These differences point to journalists selecting topics on a principle of “ethnic belonging”. The war-related content created by Latvian newsrooms shows a relative avoidance of issues which could provoke outrage in the audience; therefore, journalists for news and current affairs programmes on LR1 and LTV1 mostly rely on the expertise and responsibility of authorised sources. As a result, some of the Latvian resident groups’ opinions are not represented in the broadcasts targeted towards the Latvian audience, thus diminishing the diversity of content.

5. Discussion

In our study, self-censorship is a reaction to external censorship or pressure to restrict the diversity of information and opinions offered by the media. There is no formal censoring. Self-censorship manifests itself as individual and group reactions to a public censorship regime. In our study, the censor and the censored are different agents, as the pressure to restrict free discussion about sensitive issues, interpreting them as security threats, is inspired by influential politicians and exerted by an audience that is active online.

Thus, we consider that in the current geopolitical situation, the main factors influencing self-censorship can be found at the political and societal level. Analysing the hierarchies of perceived influences (Reese 2001; Shoemaker and Reese 2014), it should be concluded that influential politicians are at the highest level, followed by active representatives of society who use the possibilities of social networking platforms to multiply politicians’ attacks on the PSM and other professional media. The presence of institutional influence is important because the attitude of the Latvian media regulator, which is not free from political influence (Ostrovskis 2023, 23 May), towards the PSM appears in the interviews as a factor promoting self-censorship. The place of the organisational level of influence in the hierarchy is debatable. PSM organisations also receive attacks; they have the most direct opportunity to influence the work of journalists, but although they want to, they are not able to protect PSM professionals from self-censorship, and in some cases, they accept it at the professional routine level and the individual level. At the same time, we found that self-censorship is fluctuating and case-dependent. The influence of reference groups or colleagues (Hanitzsch et al. 2010) is of great importance when looking for ways to resist self-censorship and reduce the risk that its presence is permanent.

Analysing the political level of influences, we assume that turning against the media is caused by imagined media effects and a simplified understanding of public communication. Contrary to what studies suggest, in the view of censorship promoters, media influence is similar to a “magic bullet”, assuming that the audience’s behaviour is determined by media content. In this approach, the audience is viewed as passive and unable to resist; therefore, it is safer not to present ‘dangerous’ information. This principle guides Latvian media politics regarding ethnic minorities (Kruk and Skulte 2022). This is significant in the circumstances of the war in Ukraine, with a political decision made to stop content creation in the Russian language as of 2026 for the sake of “public security and a unified information space” (Latvijas Vēstnesis 2023). Efforts for PSM instrumentalisation, which have been topical since Latvia reclaimed its independence (Beitika 2016) and which restrict pluralism, continue.

In a modern media ecosystem, the role of society as an influencer of the media agenda (Olsen et al. 2022) increases audience engagement and serves as proof of media democratisation. In our study, attacks on journalists, as an audience activity, can leave an impact on PSM content’s quality and harm the interests of society. In addition, both politicians and online commenters follow a populistic slogan, finding a culprit for societal problems in the media, as well as an “enemy of nation” (Alvares and Dahlgren 2016; Rožukalne 2020) in the context of the war in Ukraine, as the discussion about the PSM is one-sided, aggressive, and emotionally charged (Hameleers et al. 2017). At the same time, attacks and unfounded accusations of the PSM, through which political capital is gained and the PSM’s position is weakened, resemble the same Kremlin manipulation techniques, the danger and impact of which on the public sphere (Hanley 2022) are constantly discussed politically and academically.

The third issue is the response to self-censorship and the opportunities to overcome it. An open discussion about self-censorship could be one of the ways to resist the pressure felt by PSM journalists. It ensures both reflexivity (Archer 2003), which helps journalists evaluate their role as agents, and adds social context to the understanding of self-censorship (Caetano 2015). However, PSM editorial offices have not created strategies to protect the professional integrity and legitimacy of journalism as a discursive institution (Hanitzsch and Vos 2017).

Our research did not identify cases of self-censorship related to the fear of not obtaining information. We identified self-censorship similar to cases when journalists and editors, for example, in cases of military conflict, think they must ensure the integrity of society and support the government (Bar-Tal 2017).

According to the Spiral of Silence theory, the media and journalists define the climate of the concealment of information and opinions in society as directly and indirectly determining what to be silent about, why to be silent, and who should be silent. Our data show that the Spiral of Silence effect in relation to self-censorship development in PSM is unstable, inconsistent, and is evolving in contradictory directions. Its development is reduced by regular attempts to overcome external pressure-driven self-censorship in various PSM newsrooms and other professional media. The Silence spiral effect’s contradictory nature is determined by the professional habitus of journalists and the efforts of media professionals to preserve and strengthen their cultural capital.

The perception and practices of self-censorship affect both the level of the journalistic habitus opus operatum or the work situation and one’s position in the profession and the habitus as a modus operandi (Meyen and Riesmeyer 2012). They are propelling the necessity to solve internal and external conflicts that influence the goal of journalism and to understand everyday practices in relation to oneself, the audience, and society.

The influence of self-censorship on content diversity and pluralism is identified both in interviews and content analysis. As a result of self-censorship caused by the social environment, the traditional journalist habitus is doubted, and in a crisis, the proportion of PSM journalists’ professional field gradually reduces. At the same time, self-censorship, even in cases when employees disagree with restrictions at the individual level, diminishes PSM journalists’ role as agents.

6. Conclusions

In our study, we analysed the question of what social and individual factors drove self-censorship in Latvian PSM after Russia’s full-scale invasion of Ukraine and if it impacted war-related content. The presence of self-censorship is not an individual experience. Self-censorship is being discussed in editorial offices; it influences professional performance and employee wellbeing. Self-censorship decisions are made both at the individual and newsroom level, mostly refusing to cover certain topics or sources.

Content analysis in the first months of the war shows both the professional adaptation process and the uncertainty created by the new crisis, which was also influenced by the efforts to aggressively denigrate PSM’s content production approaches.

Self-censorship affects news, analytical, and investigative journalism editorial staff members. It is a substantial part of PSM content, as most of the PSM’s budget consists of financing for news and analytical content creation (SEPLP 2023). Self-censorship was also promoted by the information sources—their refusal to publicly represent a particular topic, expertise, or opinion. Though there are studies analysing how the minority can convince the majority, or the fact that self-censorship usually regards true information whose publication can cause harm (Bar-Tal 2017), in our study, self-censorship is attributed to opinions about controversial social–political issues identifying polarisation of public discourse.

The main cause of self-censorship is public attacks on PSM. Journalists identify certain actors and their goal to justify a demand for a one-sided, political stance-appropriate interpretation of events and selectiveness of opinions. To achieve this, journalists or newsrooms are discredited. This creates constant anxiety and takes up psychological, emotional, and time resources.

Research participants interpreted these attacks as attempts from the outside to intervene in the professional decision-making process before publication and connected them with attempts to silence journalists and their sources, limiting the diversity of content and opinions.

A positive effect was also detected. Editorial offices reviewed their editorial stances, assessing whether and what alterations were necessary in response to changes in the public discourse after Russia’s invasion of Ukraine. The doubting of professionalism has made editorial offices and journalists review professional procedures related to preciseness.

The assessment of attacking actors and their influence on self-censorship differs greatly. Attacks coming from social network users are more frequent, but they are “unimportant” or “unfounded” unless they express personal humiliation or threat. A greater influence is attributed to political decisions and attacks by politicians or politicised activists. The fact of an attack or the status of the attacker is not the main driver of self-censorship. Self-censorship is caused by the content of the attacks related to doubting media employee professionalism and influencing journalists’ confidence in their habitus. Reproach toward journalists is created by the attackers’ conviction that in war circumstances, there are topics and opinions that should be restricted since their distribution can “pose threats” or “help the enemy” and influence society’s integrity and security.

Attacks on PSM are not related to the editorial stance declared by the media or content analysis but are based on spreading suspicions and seeking and adjusting hateful discourse. PSM professionals encounter hardships debunking these attacks and adapt via methods of self-censorship.

Therefore, PSM employees prioritise changes in the political climate during a crisis over professional principles when interpreting information. As a result, the diversity of opinions in PSM diminishes; important issues are not represented comprehensively, and the public discussion is simplified, influencing the cultural and social capital of journalists.

The results of this study show that social and professional identity can play two roles simultaneously. On the one hand, when encountering constant harassment, the desire to protect the interests of a group can lead toward a decision to self-censor. On the other hand, the group support can help one overcome the social pressure driving self-censorship. Collective identity surpasses the individual’s cognitive–emotional processes, as per Bar-Tal (2017), when it comes to self-censorship decisions. Editors cooperate with the leadership of the media platform, separating reasonable criticism of professional quality from unfounded accusations of treason or posing a threat to national security.

Russian-language PSM representatives experience self-censorship deeper in the context of risks of being shut down. It is not just an individual burden but an existential accountability for the entire editorial office. Russian-language journalists emphasise their being “between a rock and a hard place” as they receive reproach both for “supporting Kremlin” and insufficient representation of the Latvian Russian community’s interests. Journalists sense emotions akin to being self-despising when abiding to social control contrary to professionalism.

The impact of external attacks, is, however, asymmetrical. Russian newsroom staff receive more attacks regarding disloyalty, but their endurance in not giving in to self-censorship is greater. They conform to the pressure to a lesser extent and do not deny themselves discussion about complex war-related topics. Latvian editorial office employees restrict themselves much more, explaining this with the self-censorship of sources. This aspect relates to different professional routines. In Latvian-language media, official sources are used in disproportional volumes, and the content does not have enough minority representatives (Kruk and Skulte 2022). The discourse is more frequently “about them” and not “with them”. That is why in a situation of socio-political turbulence, Latvian-language content creators are more dependent on political opinion leaders. On the Russian-language PSM channels, everyday sources include more minority representatives.

Media representatives do not see a clear solution to the self-censorship problem; they notice moments of vigour within themselves, followed by a relaxation period of fatigue and a focus on “safe” content.

Given that information disorder has become a global threat to democracy and Latvian society also lived under the influence of constant Russian war propaganda for a long time, politicised attacks on PSM reduce their capacity to create reliable and pluralistic content and find remedies to combat disinformation (Horowitz et al. 2021). Even more, threats and fears make it difficult for PSM to consistently fulfil their fundamental purpose (Moore 2024) of delivering quality, verified, and unbiased news at a time when the public’s need for trustworthy information is growing.

These findings can be used both in the analysis of the activities of PSM in countries living in near-war conditions and in the study of self-censorship caused by the constant attacks and professional discrediting of journalists. Our results help one understand the consequences of regular attacks on journalists discussed in various studies that identify changes in journalistic cultures (Carlson et al. 2021) around the world.

6.1. Limitations of This Study

Although we obtained a broad spectrum of data which refer to the assessment of the situation of self-censorship of Latvian PSM employees in the situation of the Russia–Ukraine war, our study has specific limitations. Firstly, our content analysis data represent a short period: the first months after Russia’s full-scale invasion of Ukraine. Secondly, during the research, there were transformations both in the external political and public discourse regarding the role of PSM and in PSM employees’ own understanding of professional functions in the context of a nearby military conflict. We did not have the opportunity to fully analyse the interaction and dynamics of these changes. Thirdly, the Latvian PSM self-censorship case study is based on conclusions obtained using qualitative data analysis methods; thus, they are not generalisable.

6.2. Further Trajectories of This Study

Our study took place in a very turbulent period, which means that all the most important aspects (the perception of self-censorship, coping mechanisms, and the impact on content) were constantly changing. Thus, its subsequent trajectories mark at least two directions. One is related to issues of media management and will focus on researching media organisations to understand how they deal with the consequences of attacks on journalists and editors with the aim of preventing the reduction in the diversity of content and the narrowing of the boundaries of the freedom of expression. The second direction is an in-depth analysis of the content about war in Ukraine, trying to test how and in what ways self-censorship affects the content over a longer period, possibly changing the professional habitus of journalism.