Self-censorship is a contested phenomenon, and scholars have so far not arrived at a universally accepted definition, either with regard to its theoretical implications or its empirical investigability (Rantanen, 2013). The most common approach is to perceive self-censorship as a milder form of censorship and as an inevitable side effect of faulty democracies (Bunn, 2015; Kenny and Gross, 2008; Simons and Strovsky, 2006; Tapsell, 2012). Some researchers outright deny the existence of self-censorship in the sense of a distinct category (see, for example, Cook and Heilmann, 2010, 2013). As for Russia, most observers, including Castells (2009), consider self-censorship to be the dominant practice of media control, alongside the state-led production of fake news, the systematic dissemination of misinformation and the continuous clamp down on dissenting media outlets.

With this research, we intend to conceptually advance debates in the sociology of journalism, media control and oligarchisation. Most crucial here is that we contest the idea that self-censorship is nit necessarily coercively imposed on journalists and a direct result of explicit political pressure, interference or fear (Clark and Grech, 2017; Human Rights House Foundation, 2006; Kenny and Gross, 2008; Vartanova, 2011; for a critique of this assumption, see Lee and Chan, 2009; Litvinenko and Toepfl, 2019). Instead, we argue that the triggers of self-censorship are much broader, and its effects are not just of a restraining and prohibitive character, but potentially even experienced as productive (see also Burt, 1998). Research on journalism in authoritarian states has suggested that self-censorship might, in fact, even surreptitiously foster media freedom. As Tong (2009) claims in relation to China, this is because the art of skilled self-censoring helps journalists bypass minefields and avoid pitfalls. One way to do so and yet still get information across is to smuggle topical material into media products which, outwardly, appear inconspicuous in character and content (p. 594).

For a dynamic of this kind to develop, journalists will have to have initially surrendered to obedience, with the expectation of reciprocity – something Weber (1978 [1922]) referred to as typically occurring under charismatic, patrimonial rule: those who are defiant and rebelling are being punished, whereas those demonstrating devotion and subordination might get rewarded. As for journalists this means that only the obedient ones will be able to retain a position that allows them to speak to large audiences. The routinisation of their self-censorship resembles a process described by Elias (1982): most of what we perceive as voluntary compliance was once exposed to external coercion and constraints (Fremdzwänge), before becoming internalised and seemingly self-imposed normative behaviour (Selbstzwänge). After a while, journalists will perceive certain restrain as self-imposed and practice their circumvention subconsciously.

Russian journalists have a name for such expressions of self-censorship; they call it adekvatnost’, a noun deriving from the adjective adekvatnyi, which translates into English as appropriate or reasonable and in our context as ‘a feel for the game’ or ‘a sense of proportion’. We first encountered adekvatnost’ in 2013 when interviewing Russian federal television reporters at various levels, from journalists at the bottom end of the hierarchy to famous media celebrities (Schimpfössl and Yablokov, 2014). Nobody at that time could have anticipated the turbulent events of 2014: the annexation of Crimea, the war in Eastern Ukraine and the rouble crisis of December 2014, all of which had dramatic repercussions for Russia’s media (Kiriya, 2017; Lipman, 2014; Malyutina, 2017). The timing of our first research happened to be serendipitous: 2013 was the last year when journalists were likely to reflect upon and proactively talk about adekvatnost’ rather than be fully occupied with the political events of the day and the immediate impact they had on their work.¹

Most of our respondents saw adekvatnost’ as a mandatory quality of anybody who aspires to become a top journalist. That is, journalists might be highly professional in their craft, but adekvatnost’ distinguished the ‘best’ of them (Schimpfössl and Yablokov, 2017b). As with Boyer’s (2003) observations regarding East Germany (p. 529), this is partly due to the expectation that top journalists quickly understand, ‘in the best spirit of professional ethics’, which questions are not to be asked and which sentences not to be written, without requiring any detailed instructions (see also Lee and Chan, 2009: 127).

This ‘sense for the game’ is not unlike what Pierre Bourdieu called the ‘right’ habitus – the physical embodiment of habits and skills, grounded in an individual’s life experience and dispositions (Bourdieu, 1984; for post-Soviet specifics of habitus see Eyal et al., 1998). It is difficult to teach oneself an ‘appropriate’ habitus; the best ones have it internalised, seemingly ‘instinctively’, like a ‘second nature’. At the lower end of the adekvatnost’ ranking sit those who toe the line too rigidly and are overly anxious about getting things right. Too stringent self-censorship stifles creativity and makes media products lame and boring.² In other words, an outstanding journalist might be daring, flamboyant and quirky, but, in best traditions of Weberian patrimonial reciprocity, clearly knows how far to go, as overdoing such things or getting them wrong would be highly risky in the ever-changing political climate.

Case studies: Hungary and Latvia

By and large, we can observe two slightly diverging developments within Central and Eastern Europe. In one set of countries (among them, Hungary, Poland, Belarus and Russia), once populist leaders secured power, usually with the support of loyal oligarchs, they also secured their control over the media. Another group of countries (among them, the Baltic states and to a certain extent the Czech Republic and Slovakia) have so far proven to be less vulnerable to populism. In these countries, publicly controlled media organisations remain intact and allow for reasonably healthy political competition. The cases we have chosen – Hungary for the first group and Latvia for the second – are characteristic of these two diverging developments. Historically, however, they share a number of similarities. Both countries experienced Soviet occupation during and after the Second World War. Subsequently, Latvia was incorporated into the Soviet Union and Hungary into the USSR-controlled Eastern Bloc.

After 1989, both countries, but particularly Hungary, were initially celebrated as showcases of democratisation (Magyar, 2016). Hungary held the first ‘proper’ elections in 1990 and in the early 1990s Budapest became the home of the European offices of several international organisations. Latvia also transformed its political institutions immediately after it proclaimed independence from the Soviet Union in 1991. In those years of rapid change, state-led censorship became a thing of the past. Journalists proactively shaped the new media landscapes, some working for the newly created independent outlets, others for the modernised successors of the previously state-run media giants (Rožukalne, 2013). Despite these similarities, there were also clear differences in how their media sectors developed (see Jakubowicz and Sükösd, 2008; Metykova and Cisarova, 2009; Mungiu-Pippidi, 2013; Štětka, 2012). Soon after Latvia gained independence from the Soviet Union, it began to absorb many elements of the Scandinavian media model. In Hungary, however, after a brief period of media freedom in the 1990s and early 2000s, the state imposed stricter control over the sector.

The two countries’ trajectories clearly diverged with the onset of the 2008 financial crisis, which hit both economies heavily and drove out advertisers and foreign investors. This opened up opportunities for regional and local businesspeople to buy up struggling media outlets at bargain prices (Urban, 2015). Many previously dynamic outlets gave in not only to political pressure, but also to commercially dubious ventures, such as clickbait (Balčytienė et al., 2015; Lauk and Harro-Loit, 2016). Nevertheless, in Latvia, a reasonable number of independent media organisations have survived, and the country’s public broadcaster has preserved high standards (Rožukalne, 2013). In Hungary, meanwhile, the media landscape has acquired oligarchic traits, and the relationships between owners and employees have become patrimonial (Bajomi-Lázár, 2017b).

The authoritarian tendencies which emerged in Hungary in the aftermath of the 2008 financial crisis resemble those which took place in Russia a little less than a decade earlier (Agh, 2016; Krastev, 2018; Oates, 2013; Schimpfössl and Yablokov, 2017b; Vendil Pallin, 2017). Media control was crucial for Vladimir Putin to cement his power (see his attacks on the largest commercial television channel, NTV. Mickievicz, 2005: 242–264). Hungary’s journalistic community, however, responded very differently to the new censorship and pressure to self-censure than did their Russian colleagues, who, as a previous research into Russia has shown, exhibited great flexibility and creativity in elaborating coping strategies (Schimpfössl and Yablokov, 2014, 2017b).

Journalist interviews

An empirical investigation of individual practices and day-to-day dynamics required a journalist-focused approach; however, not necessarily one relying on qualitative interviews. When researching a potentially sensitive and, in some respects, intangible topic, and given the stigma attached to self-censored journalism, a mixed range of methods would have been the preferred route to collect material. It would have increased the chances of catching things that are likely to remain unsaid in the setting of an interview (Ho, 2008; Kohut, 2000). Interviews can nevertheless generate rich material. It was through interviews that we initially discovered adekvatnost’ – by pure chance, in the context of a project concerned with television content, rather than journalism (Hutchings and Tolz, 2015). Similarly, in our latest interviews with Hungarian and Latvian journalists, the richest accounts we received were in response to questions which were not directly related to self-censorship.

Besides self-censorship, we asked our respondents about the current political and social climate and how they thought this affected the media in general. This included working conditions and unionisation; how economic troubles impact journalists and their work; and how editorial boards interact with their owners. We conducted semi-structured interviews, 15 in Hungary (in December 2017) and 14 in Latvia (in April 2018). When sampling, we recruited journalists from different outlets; state and non-state, public and private, conservative and liberal as well as print, TV, radio and online media.³

In the analysis, we categorised our data into several themes. First, we identified statements indicating how, when and why journalists might censor themselves, according to their own narratives. Those statements appeared most frequently in answers relating to newsroom hierarchies, political influence on the daily work of journalists, and financial pressures. Next, we explored the origins and causes of the dynamics we observed, as experienced by our interviewees.

Hungary

When international investors retreated from Hungary in the aftermath of the 2008 crisis, businessmen close to Viktor Orbán purchased struggling media outlets (Urban, 2016), a process Bajomi-Lázár (2014) calls ‘media colonisation’. The subsequent oligarch-state dominance provided Orbán with a powerful base for self-promotion, something that was hugely decisive for his party’s election victory in 2010 (Magyar, 2016). In addition, Orbán’s government passed changes to the media law which now allows for even more state pressure on the independent media (Polyak and Nagy, 2015).

Not unlike in Russia a decade earlier, also in Hungary the journalistic community became sharply polarised, along the lines of journalists employed by outlets supporting the Orbán regime versus those working for oppositional media. After Orbán’s election victory in 2014 self-censorship became prevalent. According to a study by Urban et al. (2017: 147), by 2015, a third of Hungarian journalists stated that they had concealed or distorted information to protect their jobs. Cautionary measures included holding back any information that might be at odds with one’s outlet’s editorial line so as to avoid any risk of it becoming munition in the hands of competitors and politically adverse media. Such a dynamic in Hungary’s journalism has brought it close to what we see in Turkey where self-censorship is well-rehearsed but is nevertheless painfully coupled with coercion and the fear of punishment (see Yesil, 2014). In both countries – and unlike Russia – self-censorship is exercised in response to what Elias called Fremdzwänge, i.e., journalists perceive the pressure to do so as externally imposed.

Even under circumstances such as evident in Hungary, where journalists perceive self-censorship as inflicted upon them, innovative practices have appeared. A number of respondents told us that they occasionally pass on important material, which they cannot publish themselves, to colleagues from other outlets, both competitive and hostile ones. One of them was Viktor, who works for a liberal online medium: ‘When I have a story I really care about and I know for sure I can’t get it out through my outlet, I’d rather have it published by a colleague elsewhere’. László, another liberal journalist, hinted at an additional motivation for such strategies:

After 1991, Scandinavia sought to establish influence in the Baltics. To bring the region’s media infrastructure up-to-date, Scandinavian media businesses and NGOs provided know how, learning technologies, media training and investments (Balčytienė, 2009). A few of our respondents were critical of this interference and saw the Scandinavians’ main agenda as profit-making rather than investing (Balčytienė and Lauk, 2005; Štětka, 2012). Most, however, assessed their northern neighbours’ activities rather positively. Marta used to work for the daily newspaper Diena, then owned by the Swedish Bonnier group. Until 2008, Diena was Latvia’s flagship media outlet in the market of commercial quality journalism. Marta appreciated the training courses Diena ran and the high ethical standards they introduced. Emilija, a journalist now working for the finance newspaper Ir, remarked that ‘we had Nordic media trainers before we got Nordic banks. They taught us everything: from basic ethics to storytelling and specific examples of business writing’.

After the 2008 financial crisis, Latvian journalists found themselves in a similar situation to their Hungarian colleagues: domestic oligarchs purchased outlets abandoned by international investors (see, for example, Rožukalne, 2012, 2013). The most drastic development came in 2009 when Bonnier sold Diena to an unnamed company registered offshore. Later, it was revealed that the buyers were Latvian business tycoons with a highly dubious reputation (Baltic News Network, 2011). The new owners used their media outlets for their economic and political gains, with little consideration for how this affected journalism. Before long, certain practices re-appeared which had been thought to be features of the past. Most shocking for many was the spread of hidden advertising [dzhinsa], something strongly associated with the shady and unprincipled 1990s.