Who is responsible for verbal violence? : a mixed-methods analysis of hate speech in Polish media discourse

Who is responsible for verbal violence?

A mixed-methods analysis of hate speech in Polish media discourse

Barbara Erling

Master’s thesis

Graduate School of Communication

Master’s programme Communication Science

Supervisor: Michael Hameleers

Date: July, 15 2019

ABSTRACT

The study examines the construction of Polish media discourse which contains hate speech towards out-group members, as well as provides insights into the role journalists play in disseminating verbal forms of violation in the news. Drawing on data from mixed-methods design, the author demonstrates the elements based on which the hate speech discourse was created and indicates the signs of media polarisation in Poland. In addition, the results show high level of desensitisation among Polish journalists and the impact of external and internal pressures on publishing negative news content. The findings are discussed in the light of extant knowledge about partisanship bias and media’s increasing use of negative reporting style.

INTRODUCTION

On January 14, 2019 Paweł Adamowicz, the mayor of Gdańsk, Poland was assassinated in front of hundreds of people while attending an annual charity event. This incident has raised heated debates in media, politics and society regarding the real-life implications of hate speech. This is because, before the assassination Adamowicz was fiercely attracted by the state’s politicians and pro-government media for his liberal approach to politics. In the aftermath of this events, the Council of Media Ethics analysed materials by Polish Broadcasting Company (TVP) which is known for its populistic and anti-elitism stance. The Council found that the authors of the materials broke 5 out of 7 principles recorded in the Code of Media Ethics. The report says that Paweł Adamowicz had not received a chance to refute the allegations in none of the publications which slandered him (REM, 2019).

After gaining power in 2015, right-wing party Law and Justice (PiS) has built up a whole network of propaganda institutions – including TVP – that uses the language of hate towards opposing actors as the weapon in their populistic politics. Along with PiS governance, the right-wing discourse started to gain more prominence, and the media shifted into partisan environment. Literature describes this phenomenon as “journalism of identity” (Deuzen, 2005). The press discovered a business model which appeals to different segments of a fragmented market, and thanks to that citizens became more entrenched in their views as they consume only the media which appeal to their political identity (Jones, 2018). As a result, the media in Poland has become deeply polarised in the past few years and turned partisan.

Scholars suggest a relation between polarization and affect based on partisan identity, and define it as “the extent to which partisans view each other as a disliked out-group”

(Iyengar et al., 2012, p. 1). As mentioned before, since the media are politically polarized, they mirror such aversion towards out-groups too, what consequently contributes to negative bias in media content (Kleinnijenhuis, 2008) and the usage of hate speech as an element of the news construction. Although there is a common agreement on the extensive attendance of negativity and hate speech in contemporary news, it is contested who should be blamed for it. Lengauer et al. (2012) indicate two actors. On one hand there are journalists who use controversy for commercial, political and professional motives. On the other hand, there are politicians framing the discourse accordingly to news values to gain the media attention. This study, however, shifts the focuses only to journalists and the media as the main sources of political knowledge (Owen, 2015), and considers politicians as an out-group member and target of hate speech due to political polarisation and media partisanship.

To better understand how the media discourse containing hate speech was constructed during the 2018 local elections, and to provide insights into the role journalists play in disseminating verbal forms of violation, this paper employs a mixed-methods design. First of all, a qualitative content analysis of Polish news media (N = 50) is employed to provide insights into the discourse used to portray the societal groups which are the most exposed to hate speech. In the next step, the journalists’ perception of hate speech and news evaluation of ani-elitist articles are investigated with an online experiment (N = 121). To structure of the paper is organised as follows. Firstly, the theoretical framework defines key ideas and conceptualise the study. Secondly, the paper provides the explanation of the research methods, and subsequently presents the results. Final section includes discussion in the light of the existing literature, conclusion and areas for further research.

Hate speech discourse and its target

Although the term hate speech is frequently used, there is no agreement on its definition. Bockmann and Turpin-Petrosino’s (2002) emphasise the denigrating and wounding nature of hate speech, which is a form of expression that the perpetrator uses against an object of prejudice. It is oftentimes targeted against minority groups and typically addresses traits of nationality, ethnicity, religion or gender (Nockleby, 2000).

According to Parekh (2001) the permission to propagate hate speech supports a social climate in which specific groups are disparaged and their prejudiced treatment is considered normal. Soral et al. (2018) confirm this issue and suggest that repetitive and frequent exposure to hate speech causes desensitisation to verbal violence and leads to inferior evaluations of the victim and more substantial distancing, ergo growing out-group prejudice.

Moreover, according to the criteria showcased by Matsuda (1989) hate speech is harmful not only because of the words themselves, but also because of the historical, political, and social context in which they are used. The hate speech communication is predominantly simplified by building on the concepts which are embedded in culture and society, hence such communication draws on existing stereotypes, cultural meanings and social beliefs about the targeted groups (Tsesis, 2002).

As people work on mental shortcuts to simplify the cognitive requests of daily interaction (Macrae et al., 1997) such concepts like stereotypes, cultural meanings and social beliefs are nearly inescapable in everyday discourse. Since these concepts, known as cultural products, are inevitably embedded in social structures, this research interprets the targets of hate speech from the perspective of Polish culture.

Bilewicz et al. (2014) outlines seven minority groups who are the most affected by hate speech. These groups include: Jewish; Ukrainians; Roma people; Russians, Black people; Muslims; and sexual minority. The political parties manifestos for the 2018 local

elections focused three minority Ukrainians, Muslims and LGBT movement, and hence these groups gain most of the media attention. Therefore, based on this observation, I this research takes into account: ethnic minority (Ukrainians); religious minority (Muslims); and sexual minority for a better understanding of hate speech in Polish media discourse. Nevertheless, the literature points out that cultural stereotypes and societal beliefs are not the only triggers of hate towards particular groups. The rise of populism introduced a new hate speech target group.

Populism and anti-establishment discourse as a source of hate

The hate speech can be performed by individuals, groups and the media. The latter actor’s performance might be conceptually linked to populism and anti-elitism discourse which is currently visibly present in the media discourse. Populism refers to the people and justifies its action by identifying with them (Jagers & Walgrave, 2007). Moreover, it expresses a boundary to the elites and/or other out-groups. Although, similarly to other terms in the lexicon of political science, populism is highly contested concept, the academic literature offers some key approaches by which populism can be conceptualised.

First approach – populism as an ideology – which can be attributed to Mudde’s (2007) work. Considering its validity in comparative empirical research, this approach has recently become dominant in the literature, especially among political scientists from Europe (Moffitt & Tormey, 2014). However, Ernesto Laclau’s (2005, 2006) idea on thinking about populism from the perspective of a political logic has made a significant impact on social and political theory. The other approach of the ‘Essex School’ has suggested to conceptualise populism as a discourse (Moffitt & Tormey, 2014) yet Weyland’s (2001) definition of populism as a political strategy was established as a starting point for empirical analyses, especially in the literature on populism in Latin America (see Ellner, 2003; Roberts, 2003; 2006).

Nevertheless, in order to conceptualise this study, I will use the definition of populism which is understood as a political communication style. It suggests that populism is a communication frame which addresses its message to people and identifies with them (Taggart, 2000).

Most of the scholars consider anti-elitism as one of the main features of populism (Canovan, 1981; Wievorka, 1993; Taggart, 2000; Mény & Surel, 2002). Both anti-elitist or anti-establishment discourse stresses the gap and antipathy between the elites and the people (Jagers & Walgrave, 2007). Scholars distinguish different types of elites, namely political elites (government, ministers, parties etc.), the state (civil service, administration), media (journalists, editors, media tycoons), intellectuals (universities, scholars) or economic powers (employers, trade unions, capitalists) (Jagers & Walgrave, 2007). Elites are imagined as an enemy who is above ordinary citizens and has the power to pursue only their own interest, consequently they become people’s target of hate.

Although there is no explicit mention of politicians as part of an out-group and the target of hate speech, considering aforementioned literature, and the Polish case of Paweł Adamowicz, I incorporate politicians as another target group of hate in this study.

As people learn about the political and social issues mostly from the press, it is important to understand the mechanisms the media use for producing content. Yet, it is also vital to understand mechanisms of reporting methods on out-group members that the media use and how the hate speech discourse is linguistically constructed. Therefore, RQ1: How are the media constructing news about out-group members when reporting on political issues during the 2018 local election in Poland?

One of the most frequently used hate speech element to slander the out-group members is dehumanisation (Vollhardt et al., 2007) which also has an effect on disinhibition of violence (Bandura, 1990). Hence, a dehumanising narrative of groups or individuals is a sign of hate speech and the peril of future aggression (Vollhardt et al., 2007). Haslam (2006) established a category of different forms of dehumanisation, making division between mechanistic and animistic dehumanisation. The first classification is present when groups or individuals are denied characteristics that determine human nature, such as interpersonal warmth, individuality or activity, and emotional responsiveness (Haslam, 2006). Thereby, the object of ‘mechanistic dehumanisation’ is labelled as cold, inert, superficial, fungible, rigid and passive. The second classification occurs when groups or individuals are denied characteristics that determine humanity (Haslam, 2006). Therefore, ‘animalistic dehumanisation’ appears when the target is labelled by the following criteria: coarseness, irrationality, a lack of culture, amorality or lack of self-restraint and predominance of instincts (Haslam, 2006).

Furthermore, Wiśniewski et al. (2016) while analysing the hate speech in Poland expanded Haslam’s (2006) classification of dehumanisation, adding detestation, aberration, historical bases of hatred, justification of hate speech, criminality and active aggression. Based on aforementioned work of Haslam (2006) and Wiśniewski et al. (2016) I hypothesise that (H1) the media discourse surrounding the 2018 local election in Poland uses dehumanisation to depict minority groups. Although for the sake of this study I classify politicians as members of out-group, they are not treated by the society as minorities. Politicians are social elite and in reality belong to privileged group, hence they are not considered an object of dehumanisation. However, dehumanisation is only one of the elements of hate speech, and it does not assume that politicians are not a target of hate speech.

Media framing and news negativity in covering political policy

To better understand the dynamics of hate speech, this thesis expands our comprehension of hateful comments towards politicians to of strategic framing of political news. Political communication literature often concentrates on two related motifs in media content: 1) negative news frames which portray politics and politicians in an unfavourable light; and 2) cynical coverage, known also as “strategy” coverage which focuses on conflict and “horse race” political aspect, as well as suggests that politicians are not motivated by common good, but rather the power (Trusler & Soroka, 2014). Moreover, journalists tend to cover election campaigns through the strategic game frame that present politicians as players in a contest which focuses only on the power and popularity (Aalberg, Stromback, & de Vreese, 2012). These frames are frequently used in the news coverage of election campaigns (Pedersen, 2014). Consequently, negative political campaigns receives more reports in comparison to positive campaigns (Hansen & Pedersen, 2008). Such a reports oftentimes include hate speech elements in order to intensify the controversy, and hence gain more audience. Considering aforementioned literature, it is expected that (H2) the coverage of the 2018 local elections in Poland focuses on negativity and the use of strategic frames. Since hate speech consists of highly emotive language, it is important to understand the media framing linguistic construction of political events. Research on news framing effects has shown that framing depends on emotional responses of audience (e.g., Aarøe, 2011; Gross, 2008; Nabi, 2003). In other words, it is known that the exposure to news frames might cause emotional reactions (e.g., Kim & Cameron, 2011; Myers et al., 2012), and such reactions can actually operate as mediators on political attitudes, behaviours and opinions (e.g. Gross, 2008; Lecheler, Schuck, & de Vreese, 2013). It is especially interesting to observe the role of emotions in news media framing effects while taking into account political events or topics which are usually linked to a high level of emotive language in the media and cause fierce

public debates (e.g., Myers et al., 2012; Ramasubramanian, 2010). I therefore hypothesis that (H3) the coverage of media during the 2018 Polish local elections will consist of highly emotive language. Although it is scientifically proved that the media use emotionally intensified words to report on politics, it is unknown if they use it consciously or it just became a routine. Hence, the second research question asks: (RQ2) Can journalists recognise hate speech elements?

The hate speech elements, such as the negativity and conflict are essential criteria for a story, not only because it “sells”, but also to meet professional standards of balanced reporting (Schuck et al., 2014). While standards are imposed by the community, journalists individually adopt some roles which they perform in society in order to fulfil their mission. Such roles are called professional role conception, and as scholars indicate, these roles may influence media content that journalists produce.

Journalists professional role conception and its influence on media coverage

Professional role conceptions are considered one of the key concepts in the research of the role journalism fulfil in the society. Journalistic role conceptions is defined as the journalists’ perception of how they are supposed to do their work (Weaver & Wilhoit 1996; Hanitzsch, 2007). The role conception expand across the concept of journalism as an institution and journalist as an individual (Tandoc et al., 2012). Although some studies focused on the roles in isolation from the media content, most of the scholars have turned their attention to the relation between journalists’ perception of their role and their role performance in practice (Shoemaker & Reese, 1996; Vos, 2002; Waisbord, 2013). Since professional role determines communicator’s perception of what is worth transmitting to the audience and how the news story could develop, it proves its direct impact on content

(Shoemaker & Reese, 1996). Prole conceptions and professional attitudes influencing journalists’ behaviour are not only studied theoretically, but also supported empirically.

The first extensive empirical study was accomplished by Johnstone et al. (1972), who interviewed journalists in the United States about conceptions of their roles in the news media. They hence lay the groundwork to other researchers who continued their work in this field (Weaver & Wilhoit, 1996; Weaver, 2007). Hanitzsch (2011) attempted to conceptualise the journalistic professional milieus which crystalize around media workers who share analogous views on journalism’s purpose in society. Thereby, he recognised four global professional milieus of journalists that represent different properties of media workers and the reason why particular media behaviour is emphasised. These milieus are: the populist disseminator, detached watchdog, critical change agent and the opportunist facilitator.

The debate regarding professional role conceptions hold a very relevant position for journalists, because it clarifies the identity of journalistic profession. Moreover, the way journalist follow the professional procedures, which includes the news definition, might amplify the propagandists hate speech voices (Cherian, 2017). Therefore, it is expected that these professional attitudes influence journalists’ behaviour in producing news content, albeit (H4) journalistic role conception of Polish media workers will influence the perception of hate speech elements in the media content. In order to find out the role conceptions of Polish journalists and its influence on content the abovementioned concepts by Hanitzsch (2011) were employed to the Study 2. Although scholars, including Hanitzch (2011) mostly focus on the relation between role conception and news content, it is vital to know specifically why such content turned negative and started to use hate speech elements in the recent years.

To comprehend how hate speech is constructed in the public sphere, we need to understand the surroundings and explanations of the rise of negativity in the coverage. Taking into account the ongoing intensification of negativity in news, academia has expanded the baseline of the prevalence of negativity in the news coverage. The further explanation indicates: 1) changes in journalistic culture and modifications in professional norms; 2) transformation of the relationship between media workers and political PR experts; and finally 3) shift in commercialization and competition in the news business (Lengauer et al., 2012).

In the view of the first explanation, increased level of education among journalists, which leads to more reflexive analysis in the news; widespread shift towards post-materialist attitudes (Blumler and Kavanagh, 1999; Inglehart, 2000); as well as key events, like e.g. Watergate coverage, caused a change in journalists’ views in which negativity moved from a mere news value to a news ideology (Hallin & Mancini, 2004; Kepplinger, 2011). This kind of journalistic strategy – called critical journalism – strives to ensure a constant observation of bothersome aspects of social life and the behaviour of the power holders (Lengauer et al., 2012). Critical journalism for each political statement seeks a counter-statement from an adversary (Lengauer et al., 2012). It transformed the understanding of balanced coverage, and eventually created a more confrontational climate (Westerstahl & Johansson, 1986 in Lengauer et al., 2012). Hence, (H5) the journalists with higher education are more desensitised to hate speech in the coverage.

This brings to the second explanation, namely the impact of political public relations on journalistic work. Aforementioned critical reporting style turns adversarial when journalists sense threat in their reporting options. In the interest of protecting professional integrity and conveying their public image as an independent institution, journalists have become increasingly sceptical toward manipulative message controlled by political

communication experts and pro-active news management (Lengauer et al., 2012). Developing professionalization of politics generated ‘countermeasures’ performed by journalists (Lengauer et al., 2012). It have taken the form of deconstructive, negative and even cynical news (Blumler, 1998; Kerbel, 1999; Zaller, 1999), and led to a ‘démontage of politics’ in Western Europe and the United States public affairs reports (Kepplinger, 1998; Patterson, 1994).

Moreover, criticising politicians not only benefits journalists to project an image of professional autonomy and independence, but also impacts news commercialization. Aggressive and negative approach towards political protagonists operate as a tactic of fostering legitimization of the people, and maximizing the audience (Benson & Hallin, 2007; Dunaway, 2009). It represents negative news as an instrument of economic value in the competition for public’s attention. Conflict-ridden coverage sells better than positive one as it is more catching from human psychology perspective, stimulates interest, brings drama, and is easier to comprehend for wider range of audience (Lengauer et al., 2012). Cohen (2008) found that when news organisations experience greater economic pressures, they increase production of negative news, as the way to gain more audience. Considering two above mentioned concepts of the explanation of the prevalence of negativity in the news coverage it is expected that (H6) the more dependent journalist is on both external and internal pressures the more likely is to publish negative coverage.

In addition to the relationship between media workers and political PR experts, as well as commercialization and competition in the news business partisan bias may also affect whether and how the media report on messages of political actors. Therefore, how media react during election campaigns when issues such as e.g. state subsidies allocation are at stake?

To understand why media use hate speech towards certain groups, I incorporated the concept of partisanship bias which demonstrates the process of unfavouring political actors. Media outlets are argued to systematically report more favourably on some political actors while omitting or even defame others. This is due to the fact that media have a partisan bias towards certain parties (Puglisi & Snyder, 2011). Moreover, partisan media tend to overemphasize disliked political actors and their faults, however they deemphasize scandals and negative stories about actors they favour (Baumgartner & Chaqués Bonafont, 2015). The reason why the media promote some political actor over another may differ. According to Puglisi and Snyder (2011) such partisan gatekeeping bias might be motivated by either demand-side factors or supply-side.

Demand-driven factor is inevitably linked to audience preferences (e.g., Gentzkow & Shapiro, 2006). When it is known that audience on average support certain political parties and prefer to receive news about them, media tend to shape their reporting in order to meet the needs of their consumers (e.g., Iyengar & Hahn, 2009). In other words, media selectively adjust political messages that are popular among their audience. While demand-side factor focuses on consumers need, supply-side factor emphasises the role of political preferences among media actors (Gentzkow & Shapiro, 2006). Such factor reflects an institutionalized bias inside the media system, and this concept is known as political parallelism. According to Hallin and Mancini (2004) political parallelism outlines how media and political systems are connected, and to what extend distinct political orientations are demonstrated in the news. Moreover, journalists tend to be more biased in their reporting style, while there is a high degree of political parallelism (Haselmayer et al., 2017). Such bias might lead to a gatekeeping distort and emphasis of selectivity in news reporting, and in extreme cases to an evident partisan slant (Hallin & Mancini, 2004). Based on the concept of partisanship in the

media, it is expected that (H7) journalist’s political attitudes influence the perception of content elements accommodated in the news.

METHODS

Study 1: Qualitative Content Analysis of the Hate Speech

Before investigating the effects journalists’ perception of hate speech discourse (Study 2), the first study aims to inductively map the elements of news construction, namely dehumanisation target selection, frames adaptation and lexical forms used by the media to report the 2018 local election in Poland in Polish newspapers.

Sampling and Data Collection

This research draws on a content analysis of different Polish media outlets (N=50). The sample includes the most opinion-making newspapers, magazines and online news media in Poland (IMM, 2018). In order to achieve the far ends of a range of hate speech narratives, the selection of media outlets rests on the concept of maximum variation (e.g., Glaser & Strauss, 1976). Hence, the selection includes both broadsheet media outlets and tabloids. It gives a variety of conservative news outlets (Rzeczpospolita, Do Rzeczy, Wpolityce, Wprost) and liberal news outlets (Wyborcza, Newsweek, Onet, Wirtualna Polska). Two tabloids (Fakt, Superexpress) were also included, as tabloids focus more on negativity than other types of outlets (e.g. Lehman-Wilzig & Seletzky, 2010; Stromback et al., 2012).

The content sampling period covers three months before and three months after 2018 Polish local elections, hence the period from July, 21 2018 to January, 21 2019. This time frame was chosen as scholars emphasise that media coverage during campaigns is important for democratic processes (Swanson & Mancini, 1996). Moreover, the media at this time

provide extensive coverage, and the main actors of public discourse – journalists, politicians, citizens – are much more active (De Swert & van Aelst, 2009). Campaigns also bring tensions between different political actors and parties which impact on the language and frames media use to describe political campaigns (van Aelst et al., 2009). The concept of theoretical saturation was used as inclusion criteria (Glaser & Strauss, 1976). Consequently, the six-months period was confronted to the coverage in earlier six-months of 2018 and later six-months of 2019. As these different months did not bring original insights into hate speech in news media coverage, the diversity in sampled period should mirror the discursive framing of the hate speech.

The extracted articles for this study were investigated in respect of only Polish actors and issues, hence some of the initially considered journalistic pieces were rejected as not meeting this requirement. All items were digitally available and dealt with minorities and anti-elitism issues. Therefore, such a search string was used to select items: imigranci* (immigrants), Ukraińcy* (Ukrainians), muzułmanie* (Muslim), geje* (gays) – as the most persecuted minorities in Poland (Bilewicz et al., 2014) – and names of two most likely to win candidates for mayorship of four biggest cities in Poland – Warsaw: Trzaskowski*, Jaki*; Gdańsk: Adamowicz*, Płażyński*; Kraków: Wasserman*, Majchrowski*; Wrocław: Sutryk*, Stachowiak-Różecka*. The units of analysis were examined in line with the tree-step procedure of the grounded theory approach (Glaser & Strauss, 1967).

Data Analysis

The analysis of Polish media discourse was structured based on the coding steps detailed in the grounded theory approach (Glaser & Strauss, 1976). Firstly, the data were analysed using open coding. The samples were coded by using guiding elements such as lexical usage, target selection of media and frame construction. Secondly, as a part of focused

coding, the open codes were merged and classified into categories. Finally, during axial coding procedures, theoretically relevant relationships between categories were established. To provide an example, the construction of dehumanising narration on religious minority was connected with the narration on sexual minority as they both fall under animistic dehumanisation category – herby providing more insights into the news construction on the minorities which are considered as a new threat to native people. The peer debriefing – conducted by an impartial MSc King’s College graduate – was used to examine the validity and adequacy of sampled articles.

Study 1 results

This section presents the elements of news construction containing dehumanisation target selection, frames adaptation and lexical forms. They are considered integrated interpretations of forming hate speech narratives towards ethnic minority (Ukrainians); religious minority (Muslims); and sexual minority and politicians by news media during the 2018 local elections in Poland.

Ideological biases in political news construction

The media, based on stereotypes present in Polish discourse, construct largely negative coverage regarding minorities, however they mostly restrain from explicit blame attributions (RQ2). The most controversial expressions like: “gays suffer from AIDS, they live shorter, molest children they adopt and often are pedophiles” are primarily cited. These findings indicate that, when talking about minorities, the media relies on a hostile language in which negative traits are assigned to Ukrainians, Muslims and LGBT members.

The Polish media discourse of minority groups emphasis explicit difference between liberal and conservative media approach to the topic. Although liberal media try to control the emotions and balance the viewpoints on minorities, de facto they overtake conservative rhetoric and rely on populistic framing of the issue: “immigrants = flood or wave”; “they radicalise Europeans”. This is due to showing irrationality of the right-wing arguments and debunking absurdity of the stories populists disseminate. Liberal outlets stress the problems that are caused by minorities “the threat of Islamic terrorism is not a myth, and after the ISIS performance in recent years, it might take generations to convince people that Muslims are not so scary”, nevertheless they tame their readers with such events by emphasising that most of Muslims are regular people: ”90 percent of Muslims, shake hands with women, let their wives go to work, and the daughters to study, and instead of long beard and a copy of the Koran in their hands, they wears suits or tracksuit, hold briefcase and ride scooters”.

In contrast, conservative outlets problematise the presence of minorities in the society. They define as a threat to native people and warn their readers against destructive influences for Polish culture: “traditional European and Christian identity, even religious symbols, are being washed away under the slogans of alleged tolerance for newcomers”.

Surprisingly, tabloids did not evaluate minorities in a negative way. Although, the titles suggest a shocking development of a story: “Millions of Polish zloty go to migrants from governmental family benefit program”; “The PiS government has brought a quarter of

a million immigrants to Poland! And it encourages others to come”, the body is however

neutral, sometimes even leaning towards liberal. This issue is specifically connected with the

media ownership. On one hand, most of the media in Poland, among others tabloid Fakt, owned by foreign capital are liberal-leaning. On the other hand, although the tabloid Super Express is owned by domestic company, the owner Zbigniew Benbenek is known for his ties to left-wing politicians. Therefore, tabloids which are from its nature people-oriented outlets

do not embrace populism to some extent due to ownership influence. In this sense, the ideological biases of the outlets reflected the divergence in news construction regarding out-group members.

News media’s construction of dehumanization of minorities

As expected, Polish media used dehumanisation as a central theme to depict minority groups (H1). Media use animistic dehumanisation towards both religious and sexual minorities (Haslam, 2006). Specifically, Muslims are portrayed as predominated by instincts, infected by diseases and emotionless “gangs”. Whereas, sexual minorities are seen as amoral and deprived of social norms who “have around 500 partners in their lives”. They are accused of being “sodomites” and “pedophiles” who adopt kids to sexually abuse them. In this case, liberal outlets counteract right-wing discourse, by normalising the image of sexual minority and considering yourself as LGBT+Q “is just as natural as being born heterosexual”.

Per contra, ethnic minorities are mostly described by mechanistic dehumanisation components (Haslam, 2006). The media discourse on Ukrainians focus on two highly emotional and stereotyped narrations, namely economic and security threats. Ukrainians are seen as a threat for economic stability in Poland – they “ran rampat” and are “destroying the market”. Interestingly, historical resentment is very vivid in the coverage which includes Eastern neighbour of Poland. Due to the geographical proximity these two nations have a great part of their history in common. Ukrainians are called “Banderowcy” after the leader of the Organization of Ukrainian Nationalists – Stepan Bandera– who was involved in a Massacres of Poles in Volhynia. News on this ethnic group is frequently connected to this event as if Ukrainians desire to commit the mass execution again, therefore are a threat for the native people. Especially in conservative outlets they are described as enemies to Polish state.

The media’s framing of the political news

As hypothesized the coverage of the 2018 local elections in Poland focuses on negativity and the use of strategic frames (H2). This is particularly seen in the articles regarding political actors and their candidacy in elections. The media tend to denude mistakes of counter competitors, yet express their political sympathy and portray in positive context these politicians whom they favour. Specifically, conservative outlets blame left-wing candidates of anti-Polish approach to politics and acting on behalf of the EU and Germans. This is interpreted that liberal politicians do not care about the common good, but are motivated by power. Albeit, liberal outlets have a tendency to sneer right-wing candidates and point out their lack of competences in governing the state. At the same time, tabloids restrain from taking a political stance and write more balanced and neutral coverage.

Moreover, liberal and conservative newspapers are cynical towards candidates coming from the opposing fractions to their political line and use negative frames towards them. They refer to politicians as liars; defrauders; and dependent on external influences. Although the media while reporting on elections largely use conflict frames, the horse race metaphor that consists of strategic frame is rather absent.

Emotive language as reader’s mental shortcut and message simplifier

As stated in (H3), the coverage of media surrounding the 2018 local elections consists of highly emotive language. The analysis of lexical layer of the items demonstrate that the language mostly evoke rather negative emotions. By strategically employing negative language, the media influence the readers perception of the events surrounding elections in general, and actors in particular. The media used phrases such as: “being boiled of hatred towards Muslims”; or “thousands of Poles have lost their life savings”.

Moreover, newspapers in their coverage tend to use radical discourse, which form new words or phrases that are usually conglomerate of two or more words and together create new, mainly negative meaning, e.g. “homopropaganda” (*homosexualilty + *propaganda); “totalna opozycja” (*total + *opposition); “gejoterroryzm” (*gay + *terrorism); “bandersyn” (*Bandera + *son). Therefore, such a linguistic shortcut increases the emotions of the reader and simplify the message.

Study 2: The perception of hate speech among Polish journalists

Based on the qualitative findings of Study 1, the experimental study was created. Specifically, the central elements such as frames, language and actors of the coverage were manipulated to investigate the perception of hate speech among Polish journalists.

Design

Building on the central content features identified in Study 1, I relied on a between-subject experimental design with one experimental condition and a control condition. Participants were randomly exposed to (1) an anti-politician message which contains hostile statements regarding his speech at the final campaign meeting; and (2) a control condition that did not use hate speech, negativity and hostility towards actors presented in the condition. Participants were randomly assigned to one of the two conditions.

Sample

The survey-embedded experiment was completed by 121 participants (completion rate 64.0%). The journalists were be recruited by sending an e-mail invitation for study to the

representative sample of editorial offices. In total 277 e-mail were sent. The invitation email contained the link to the questionnaire, administered through Qualtrics.

The composition of the sample was narrowed down to participants who 1) are currently working in Polish news media organisation and deal with news production process; and 2) the participant needs to speak Polish in order to understand the questionnaire.

The definition of journalists in this study was based on Weaver and Wilhoit (1994) as “those who [have] responsibility for the preparation or transmission of news stories or other information—all full‐time reporters, writers, correspondents, columnists, news people and editors” (p. 248). Although this definition does not take into account online journalists and interns, these two were incorporated to the sample of the study as they also have a responsibility for the preparation or transmission of news stories.

The mean age of participants was 34.0 years (SD = 10.53). 44.05 % was female. 87.6 % of all participants had higher education, and 43.8 % have obtained a degree in journalism studies.

Procedure

The experiment was infixed in an online survey. By clicking the invitation link, participants were forwarded to questionnaire and briefed about the procedures of the study. Given informed consent, they started the pre-treatment survey. They completed a number of items on demographics; media environment they are working for; political predispositions; professional role conception and reporting style. Subsequently, they entered the treatment block. They were informed that they would read a randomly selected news item made by one of the Polish news outlets. They were asked to read the article carefully since they would be asked questions afterwards. Following the cover story, the piece was shown, however, considering media workers’ high level of knowledge regarding news media in Poland, the

block was pure, descriptive text and did not resemble any online news environment. In the following reading section (average reading time = 43.7 seconds, SD =33.08) participants were sent to the post-treatment survey which contained the dependent variables components and manipulation checks. The final block of the questionnaire enclosed a debriefing that stated the scope of the experimental study.

Since the survey was in Polish language, it was carefully translated, and the wording of questions was discussed with researchers from University of Warsaw who were experienced in studying media discourse and political communication as well as linguistics and rhetoric.

Independent variables

Based on the insights from Study 1, I created a piece of news article including the elements of hate speech, namely hostile expressions toward political actor and his behaviour. The article also embedded conflict frame context. The stimuli emphasised the polarisation between current government and opposing party. Since it was already discussed that journalists themselves do not use overheated, malicious language, two citations of politicians from the opposition party were embraced. First quote built on disgrace, shame, stigma of ignorance and lack of understanding dramatic Polish historical events. Second quote was based on accusations of lying and manipulating voters; political misbehaviour; and expression which pointed out that this politician is “a cancer of Polish democracy”. A control condition reported on the same event, but did not include any opinionated phrases. It was rather a purely neutral report of the happening. Stimuli are included in appendix.

In the post-treatment section I measured likeliness of publishing seen article; attitudes towards displayed item; and journalists’ recognition of hate speech. Specifically, participants had to estimate the likeliness of previously shown article appearance in the medium they work for on bipolar 7-point scale. Next, they had rate their position towards the article they have seen on the pairs: credible-deceptive, important-meaningless, biased-unbiased,

interesting-uninteresting, well-written-badly-written, and relevant-irrelevant (Cronbach’s α = .78, M =

4,1, SD = 1.48). The items were measured on 7-point scales that formed a reliable unidimensional construct. The latter one – recognition of hate speech – was measured by answering closed question with three options – yes, no, I do not know.

Manipulation check

The manipulation check aimed to verify journalists understanding of different types of news. Article with condition was marked as opinionated piece, whereas article with control condition was considered informative. The manipulation was successful F(2, 274) = 48.73,

p<0.001). Although 12 participants did not recognize the right message, since this is a

marginal number, they were not excluded from further analysis.

Study 2 results

Recognition of hate speech

In order to explore whether Polish journalists can distinguish elements of hate speech (RQ2), Pearson’s chi-square was computed. Since 76% of journalists did not recognised hate speech in the article, the results disclosed no statistical significance χ2(2) = .905, p = .636.

Relation between journalistic role conception influences the perception of hate speech elements in the media content

In order to explore the expected influence of journalistic role conception on the perception of hate speech elements in the media content (H4) two-way ANOVA was carried out. The results demonstrate that the main effect is statistically not significant (F(6, 120) = 1.31, p = .258, η² = 0.004) as well as the interaction effect between condition and journalistic role conception (F(2,120) = .225, p = .779, η² = 17.51).

The influence of education on hate desensitisation

It was hypothesised that journalists with higher education are more desensitised to hate speech in the coverage (H5). To investigate the relations between the factors a two-way ANOVA was conducted. First the education variable was recoded into two groups – 0 for journalists with lower education and 1 for journalists higher education. The results of the test shows that there is no significant effect between education and desensitisation to hate speech (F(3, 120) = 1.80, p = .151, η² = 0.002), however the interaction effect between the condition and education turned out to be significant (F(1, 120) = 4.29, p = .041, η² = 1.26).

The relation between dependence and likeliness of publishing negative coverage

A linear regression was calculated to predict the likeliness of publishing negative article based on internal and external pressures on journalists (H6). The regression demonstrates that the effect is approaching significance (F(1, 101) = 3.61, p = .060) with an

R2 of .035. The analysis demonstrates that the likeliness of publishing negative article is equal to 2.869 + .243 - .837. Internal pressures increases by .243 and decreases by .837 for external pressures. Internal pressures are significant predictors of the likeliness of publishing negative article, whereas external pressures are not significant.

Figure 1. Relation between internal and external pressures, and likeliness of publishing

Coefficientsa Model Unstandardized B Coefficients Std. Error Standardized Coefficients Beta t Sig. 1 (Constant) 2.869 .452 6.344 .000 Internal pressures .243 .128 .186 1.901 .060 2 (Constant) 3.092 .469 6.595 .000 Internal pressures .278 .129 .212 2.159 .033 External pressures -.837 .512 -.161 -1.636 .105

a. Dependent Variable: likeliness of publishing negative coverage

The relation between political attitudes and news values

It was expected that journalist’s political attitudes influence the perception of content elements accommodated in the news (H7). A two-way ANOVA demonstrates that the main effect is statistically not significant (F(3, 74) = 1.45, p = .237, η² = 0.059) as well as the interaction effect between political attitudes and condition (F(1,74) = 2.92, p = .092, η² = 0.002).

DISCUSSION AND CONCLUSION

The discussion regarding hate speech in Polish public discourse has sparked in the immediate aftermaths of the death of Paweł Adamowicz, liberal mayor of city Gdańsk. Politicians, the media and citizens realised that the dissemination of hateful comments may

lead to tragic events, and as a result the society decided to recognise the actors involved in heating up the discourse. There were two influential groups blamed for such a performance: politicians and journalists. In this study I particularly focus on journalists’ and the media role as a hate disseminator, however the politicians were considered as an inexplicit target of hate, together with minorities – explicit target of hate – which usually suffer from hate the most.

The first study relied on qualitative content analyses of Polish online newspapers – liberal, conservative and tabloids. The ideological biases of analysed outlets reflected the discrepancy in the news construction. The key findings demonstrate that the news interpretation of the out-group members was overtly hostile and negative. Specifically, by using dehumanisation the media introduce minorities as a threat to native people who are predominated by instincts (religious minority); deprived of social norms (sexual minority); as well as cold and degenerated (ethnic minority). This actually indicates the misleading definition of hate speech, since unlike hate speech towards politicians, it rather results from contempt. Bilewicz et al. (2017) have already stressed the difference of erroneous interpretation of the negative remarks towards minorities and called this phenomenon

contempt speech. According to Bilewicz et al. (2017) the speech of contempt is built on

degrading emotions that block any empathic reactions and objectify another human being. Such an interpretation is also compatible to the findings of Study 1.

Furthermore, the political news were reported by using strategic framing, and the coverage relied mainly on presenting permanent conflict between political actors. This was accompanied by highly emotive language which mostly evoked rather negative emotions. The analysis additionally revealed the widespread contestation among news outlets based on the political leanings. Newspapers produced favourable news about the ideas or actors they support, whereas opposing ones were presented in unfavourable light. It consequently led to biased coverage surrounding the 2018 Polish local elections. Additionally, the media

discourse is deeply polarised, and hence divides the society into two opposing fractions. Such a behaviour contributes to the echo chambers (Sunstein, 2009) and filter bubbles persistence (e.g., Pariser, 2011).

The factors which cause the media dissemination of hate speech and the reason why the press do it is however scientifically unexplored. To investigate the aforementioned issues, Study 2 relied on an online experiment which aimed to examine the perception of hate speech among journalists in Poland as well as the mechanisms underlying this phenomenon. As a central finding, this research demonstrated that Polish journalists do not recognize elements of hate speech, what in consequence might influence the news production. Since journalists cannot identify the hate speech, it means that they unconsciously use it in their coverage. This finding can be connected to the desensitization to hate speech which is caused by overexposure of negativity and forms of verbal violence that journalists deal with in their daily work. Literature suggest that repetitive and frequent exposure to hate speech result in desensitization to such expressions and consequently leads to lower evaluation and greater distancing to the target of hate (Soral et al., 2017). Thus delivering the news with hate speech increases out-group prejudice among citizens.

The research on journalistic role conception indicates that the roles determine communicator’s perception of what is worth transmitting to the audience and how the news story should be developed, hence it proves direct impact on content (e.g. Shoemaker & Reese, 1996; Waisbord, 2013). Based on these premises of research, it was assumed that there will be an interconnection between role conception and the identification of the media content elements, however the analysis proved no association. Moreover, contrary to the expectations, there was no significant relation between journalists’ education and recognition of hate speech in the coverage. This is due to the sample limitation, because 88% of the respondents had

higher education. Although interaction effect is significant, one of the groups was substantially smaller so the result should be approached cautiously.

The commercialisation of the news is deeply established in current trend of media organisation management. The literature indicates that the economic pressures lead to increasing production of negative news (D’arma, 2018). Based on this assumption, it was found that the hypothesis was partially supported in the study. The relationship between the external and internal pressures on journalists was very weak. This might be due to news media business approach which tend to focus on freelancers whose labour is cheaper and who have selective comprehension of the situation in editorial office. In this case rejecting journalist news proposal is not considered as a pressure rather than a news selection process.

Although Study 1 demonstrated evident polarisation of the media, the experimental study results did not support the finding. Apparently, journalists’ political attitudes do not influence their evaluation on news elements, which were supposed to reflect their political leaning. This again might be related to the desensitisation process and higher objectivity of news they read in comparison to the news they produce.

The studies have some limitations. First, the content analysis zoomed in on the online newspapers exclusively. There were not included other types of media, such as television or social media platforms which are very influential sources of information. Taking into account other media might allow to include more complexed interpretation of the media discourse. Second, the experiment had homogenies sample, which in some cases did not allow for statistical interference of the results. For the further study, I would suggest the more in-depth research which includes the analysis of the coverage of specific group of journalists who subsequently in experimental study would be confronted with their own news items.

Despite these limitations, I believe that this study has offered important insights into the hate speech discourse comprehension, and understanding of consequences of the dissemination of verbal violence towards out-group members.

REFERENCES

Aarøe, L. (2011). Investigating frame strength: The case of episodic and thematic frames.

Political communication, 28(2), pp. 207-226.

Adam, S., Eugster, B., Antl-Wittenberg, E., Azrout, R., Möller, J., de Vreese, C., Kritzinger, S. (2019). News media’s position-taking regarding the European Union: The synchronization of mass media’s reporting and commentating in the 2014 European Parliament elections.

Journal of European Public Policy, 26(1), pp. 44-62.

Arcan, H. E. (2013). Interrupted Social Peace: Hate Speech in Turkish Media. The IAFOR

Journal of Media, Communication and Film, 1(1), pp. 43-56.

Bader, V. (2014). Free Speech or Non-Discrimination as Trump? Reflections on

Contextualised Reasonable Balancing and Its Limits. Journal of Ethnic and Migration Studies 40 (2), pp. 320–338.

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of general psychology, 5(4), pp. 323-370.

Bonafont, L. C., Baumgartner, F. R., & Palau, A. (2015). Agenda dynamics in Spain.

Springer.

Benson, R., & Hallin, D. C. (2007). How states, markets and globalization shape the news: The French and US national press, 1965-97. European Journal of Communication, 22(1), pp. 27-48.

Blumler, J. G. (1998). Public spheres in contention: Reflections from Britain, 1997. In Wie die

medien die welt erschaffen und wie die menschen darin leben (pp. 83-101). VS Verlag für

Sozialwissenschaften.

Blumler, J. G., & Kavanagh, D. (1999). The third age of political communication: Influences and features. Political communication, 16(3), pp. 209-230.

Boeckman, R. J. & Turpin-Petrosino, C. (2002), Understanding the Harm of Hate Crime, Journal of Social Issues, vol. 58, no. 2, pp. 207-225.

Branton, R. P., & Dunaway, J. (2009). Slanted newspaper coverage of immigration: The importance of economics and geography. Policy Studies Journal, 37(2), pp. 257-273.

Cacioppo, J. T., & Gardner, W. L. (1999). Emotion. Annual review of psychology, 50(1), pp. 191-214.

Canovan, M. (1981). Populism. London: London Junction.

Carraro, L., & Castelli, L. (2010). The implicit and explicit effects of negative political campaigns: Is the source really blamed?. Political Psychology, 31(4), pp. 617-645. Cherian, G. (2017). Hate Speech: A Dilemma for Journalists the World Over. Ethical

Journalism Network. pp. 29-33.

Cohen, B. (1963).The press and foreign policy. Princeton, NJ: Princeton University Press. Council of Europe Committee of Ministers (1997), Recommendation no. R (97) 20, (Retrived from www.coe.int/t/dghl/standardsetting/media/doc/cm/rec(1997)020&expmem_EN.asp) Council of Europe CETS No. 185 Convention on Cybercrime, opened for signature Nov. 23, 2001. (Retrived from http://conventions.coe.int/ Treaty/en/Treaties/Html/185.htm)

Cowan, G. & Khatchadourian, D. (2003),‘Empathy, Ways of Knowing, and Interdependence as Mediators of Gender Differences, Attitudes Toward Hate Speech and Freedom of Speech,’

Psychology of Women Quarterly, vol. 27, no. 4, pp. 300-308.

Deuze, M. (2005). What is journalism? Professional identity and ideology of journalists reconsidered. Journalism, 6(4), 442-464.

Donsbach, W., & Patterson, T. (2004). Political news journalists: Partisanship,

professionalism, and political roles. Cambridge University Press. Comparing political

communication: Theories, cases and challenges, 251-270.

Ellner, S. (2003) ‘The Contrasting Variants of the Populism of Hugo Chávez and Alberto Fujimori’, Journal of Latin American Studies, 35 (1), pp. 139–62.

Geer, J. G. (2012). The news media and the rise of negativity in presidential campaigns. PS:

Political Science & Politics, 45(3), pp. 422-427.

Gentzkow, M., & Shapiro, J. M. (2006). Media bias and reputation. Journal of political

Economy, 114(2), pp. 280-316.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

Hallin, D. C., & Mancini, P. (2004). Comparing media systems: Three models of media and politics. Cambridge University Press.

Hanitzsch, T. (2007). Deconstructing journalism culture: Toward a universal theory. Communication theory, 17(4), 367-385.

Hanitzsch, T. (2011). Populist disseminators, detached watchdogs, critical change agents and opportunist facilitators: Professional milieus, the journalistic field and autonomy in 18 countries. International Communication Gazette, 73(6), 477-494

Hanitzsch, T., Hanusch, F., Mellado, C., Anikina, M., Berganza, R., Cangoz, I., ... & Virginia Moreira, S. (2011). Mapping journalism cultures across nations: A comparative study of 18 countries. Journalism Studies, 12(3), 273-293.

Haselmayer, M., Wagner, M., & Meyer, T. M. (2017). Partisan bias in message selection: Media gatekeeping of party press releases. Political communication, 34(3), pp. 367-384. Haslam, N. (2006). Dehumanization: An integrative review. Personality and social

psychology review, 10(3), pp. 252-264.

Hernández, T. K. (2011). Hate speech and the language of racism in Latin America: a lens for reconsidering global hate speech restrictions and legislation models. U. Pa. J. Int'l L., 32, 805. Hjarvard, S. 2008. The mediatization of society. A theory of the media as agents of social and cultural change. Nordicom Review 29(2): 105–134.

Inglehart, R. (2000). Globalization and postmodern values. Washington Quarterly, 23(1), 215-228.

Iyengar, S., & Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of communication, 59(1), pp. 19-39.

Jagers, J., & Walgrave, S. (2007). Populism as political communication style: An empirical

study of political parties' discourse in Belgium. European Journal of Political Research, 46(3), pp. 319-345. Iyengar, S., Sood, G., & Lelkes, Y. (2012). Affect, not ideology: A social identity perspective

on polarization. Public Opinion Quarterly , 76(3), 405–431.

Janowitz, M. (1975). Professional models in journalism: The gatekeeper and the advocate. Journalism quarterly, 52(4), 618-626.

Johnstone, J. W., Slawski, E. J., & Bowman, W. W. (1972). The professional values of American newsmen. Public opinion quarterly, 36(4), 522-540.

Jones, G. (2018). Trends and Analysis of the Media Landscape in Poland: A Brief History of Polish Media. Poland Media Report.

Kerbel, M.R. (1999). Remote & Controlled: Media Politics in a Cynical Age . 2d ed. Boulder,

Kim, H. J., & Cameron, G. T. (2011). Emotions matter in crisis: The role of anger and sadness in the publics’ response to crisis news framing and corporate crisis response.

Communication Research, 38(6), pp. 826-855.

Kleinnijenhuis, J. (2008). Negativity. The international encyclopedia of communication. Krotz, F. (2009) ‘Mediatization: A Concept With Which to Grasp Media and Societal Change’, in Lundby, K. (ed.) Mediatization. Concept, Changes, Consequences. pp. 21-40. Laclau, E. (2005) On Populist Reason. London: Verso.

Laclau, E. (2006). Why Constructing a People is the Main Task of Radical Politics. Critical

Inquiry, 32 (4), pp. 646–80.

Lecheler, S., Schuck, A. R., & de Vreese, C. H. (2013). Dealing with feelings: Positive and negative discrete emotions as mediators of news framing effects. The European Journal of

Communication Research, 2013, Vol.38(2), pp.189-209.

Lengauer, G., Esser, F., & Berganza, R. (2012). Negativity in political news: A review of concepts, operationalizations and key findings. Journalism, 13(2), 179-202.

Macrae, C. N., Bodenhausen, G. V., Milne, A. B., Thorn, T. M., & Castelli, L. (1997). On the activation of social stereotypes: The moderating role of processing objectives. Journal of

Experimental Social Psychology, 33(5), pp. 471-489.

Matsuda, M. J. (1989). Public response to racist speech: Considering the victim’s story. In

Words that wound, pp. 17-51. Routledge.

Maussen, M., & Grillo, R. (2014). Regulation of speech in multicultural societies: Introduction. Journal of Ethnic and Migration Studies, 40(2), pp. 174-193.

Mellado, C. (2013). Professional roles in news content: Six models of journalistic role performance. In ICA’s 63rd Annual Conference Challenging Communication Research. Mény, Y. & Surel, Y. (2002). Democracies and the populist challenge. Basingstoke:

Palgrave.

Moffitt, B., & Tormey, S. (2014). Rethinking populism: Politics, mediatisation and political style. Political Studies, 62(2), pp. 381-397.

Mudde, C. (2007) Populist Radical Right Parties in Europe. Cambridge: Cambridge

University Press.

Myers, T. A., Nisbet, M. C., Maibach, E. W., & Leiserowitz, A. A. (2012). A public health frame arouses hopeful emotions about climate change. Climatic change, 113(3-4), pp. 1105-1112.

Nabi, R. L. (2003). Exploring the framing effects of emotion: Do discrete emotions

differentially influence information accessibility, information seeking, and policy preference?

Nockleby, J. T. (2000). Hate speech. Encyclopedia of the American constitution, 3, 1277-1279.

Oster, J. (2013). Theory and doctrine of ‘media freedom as a legal concept. Journal of Media

Law, 5(1), pp. 57-78.

Owen, D. (2015). The New Media’s Role in Politics. Georgetown University. Patterson, T. E. (1994). The news media: An effective political actor?. Political

Communication, 14(4), pp 445-455.

Parekh, B. (2001). Rethinking multiculturalism: Cultural diversity and political theory.

Ethnicities, 1(1), pp. 109-115.

Puglisi, R., & Snyder Jr, J. M. (2011). Newspaper coverage of political scandals. The Journal

of Politics, 73(3), pp. 931-950.

Ramasubramanian, S. (2010). Television viewing, racial attitudes, and policy preferences: Exploring the role of social identity and intergroup emotions in influencing support for affirmative action. Communication Monographs, 77(1), pp. 102-120.

Reicher, S., Haslam, S. A., & Rath, R. (2008). Making a virtue of evil: A five step social identity model of the development of collective hate. Social and Personality Psychology

Compass, 2, pp. 1313–1344. https://doi.org/10.1111/j.1751- 9004.2008.00113.x

Roberts, K. M. (2003) ‘Social Correlates of Party System Demise and Populist Resurgence in Venezuela’, Latin American Politics and Society, 45 (3), pp. 35–57.

Roberts, K. M. (2006) ‘Populism, Political Conflict, and Grass-Roots Organization in Latin America’, Comparative Politics, 38 (2), pp. 127–48.

Rozin, P., & Royzman, E. B. (2001). Negativity bias, negativity dominance, and contagion.

Personality and social psychology review, 5(4), pp. 296-320.

Shoemaker, P. J., & Reese, S. D. (1996). Mediating the message. White Plains, NY: Longman. pp. 781-795.

Shoemaker, P. J., & Vos, T. (2009). Gatekeeping theory. Routledge.

Soral, W., Bilewicz, M., & Winiewski, M. (2018). Exposure to hate speech increases prejudice through desensitization. Aggressive behavior, 44(2), pp. 136-146.

Soroka, S., McAdams, S. (2015). News, politics, and negativity. Political Communication,

32(1), pp. 1-22.

Starck, K., & Soloski, J. (1977). Effect of reporter predisposition in covering controversial story. Journalism Quarterly, 54(1), 120-125.

Strömbäck, J. (2008). Four phases of mediatization: An analysis of the mediatization of politics. The international journal of press/politics, 13(3), pp. 228-246.

Taggart, P. (2000). Populism. Birmingham: Open University Press.

Tandoc Jr, E. C., Hellmueller, L., & Vos, T. P. (2013). Mind the gap: Between journalistic role conception and role enactment. Journalism Practice, 7(5), pp. 539-554.

Thompson, T. & Ungerleider, C.S. (2004), Single Sex Schooling: Final Report, Canadian Centre for Knowledge Mobilisation, Vancouver.

Tsesis, A. (2002), Destructive Messages: How Hate Speech Paves the Way for Harmful Social Movements, New York University Press, New York.

Van Dijk, T. A. (2012). The role of the press in the reproduction of racism. In Migrations:

interdisciplinary perspectives. Springer. pp. 15-29.

Van Aelst, P., Vliegenthart, R., & Boydstun, A. E. (2018). The whole world is watching: Comparing European and United States news coverage of the US 2008 and 2016 elections.

International Journal of Communication, 12, 23.

Van Noorloos, M. (2014). The Politicisation of Hate Speech Bans in the Twenty-first Century Netherlands: Law in a Changing Context. Journal of Ethnic and Migration Studies 40 (2), pp. 249–265.

Vollhardt, J., Coutin, M., Staub, E., Weiss, G., & Deflander, J. (2007). Deconstructing hate speech in the DRC: A psychological media sensitization campaign. Journal of Hate Studies,

5(15), pp. 15-35.

Vos, T. (2002). Role enactment: The influence of journalists’ role conceptions on news content. In 85th AEJMC Conference, Miami, Florida.

Waisbord, S. (2013). Reinventing professionalism: Journalism and news in global perspective. John Wiley & Sons.

Weaver, D. H., & Wilhoit, G. C. (1996). The American journalist in the 1990s: US news people at the end of an era. Psychology Press.

Weaver, D. H. (2007). Thoughts on agenda setting, framing, and priming. Journal of

communication, 57(1), 142-147.

Weyland, K. (2001). Clarifying a Contested Concept: Populism in the Study of Latin American Politics. Comparative Politics, 34 (1), pp. 1–22.

Wiśniewski, M., Hansen, K., Bilewicz, M., Soral, W., Świderska, A., & Bulska, D. (2017). Mowa nienawiści, mowa pogardy. Raport z badania przemocy werbalnej wobec grup

mniejszościowych, Fundacja Batorego/Centrum Badań nad Uprzedzeniami UW, Warszawa.

Wievorka, M. (1993). La démocratie à l’épreuve: nationalisme, populisme et ethnicité. Paris:

La découverte.

Zaller, J. (1999). A theory of media politics: How the interests of politicians, journalists, and citizens shape the news. University of Chicago Press.