Politicization of migration in the countermedia style : A computational and qualitative analysis of populist discourse

Politicization of migration in the countermedia style: A computational and qualitative analysis of populist discourse Tuukka Ylä-Anttila a, ⇑ , Gwenaëlle Bauvois b , Niko Pyrhönen b a Centre for Sociology of Democracy (CSD), Tampere University, Hämeentie 72E, 00550 Helsinki, Finland b Centre for Research on Ethnic Relations and Nationalism (CEREN), Swedish School of Social Science, University of Helsinki, PB 16 (Snellmansgatan 12), 00014, Finland 1. Introduction Online partisan news outlets are a thus-far scarcely studied phenomenon made possible by changes in technology and the media sphere that give users the possibility to not only consume but produce their own media content (Atton, 2006; Chadwick, 2013; Ritzer and Jurgenson, 2010). While this development was initially seen as a welcome democratization of the public sphere, it has led to a crisis of trust in mainstream knowledgeproduction industries and a proliferation of polarizing partisan information often at the expense of factual accuracy, resulting in what some have described as the ‘post-truth era’ (D’Ancona, 2017; Sunstein, 2017). Such websites have been called ‘fake news sites’, ‘alternative media’ or ‘partisan news sites’ (Atton, 2006; Moffitt, 2018), but these terms are problematic for various reasons. First, the concept of ‘fake news’ was quickly politicized – most notably by Donald Trump – to which the concept lends itself too easily. This is because not all content published by these sites is ‘fake’, as in ‘made up’. Rather, they typically report on real events, but present them with an ultrapartisan bias. Moreover, political issues are, to a large degree, matters of interpretation. This ambiguity can be used to resist the label ‘fake news’ on messages one is politically sympathetic towards, while simultaneously brandishing it against one’s opponents. The concept of alternative media, on the other hand, tends to have positive normative connotations, as in providing a (supposedly) much-needed alternative to mainstream views. We believe our concept of countermedia to be useful in understanding these sites: they are media outlets, but also tend to explicitly oppose ‘the (mainstream) media’, as well as the establishment more generally (however ambiguously defined), in a populist style (Moffitt, 2016). Online communication has been central to populist successes all over the world, particularly combined with a derisive, even conspiratorial view of the mainstream media as part of the ‘elite’ the populists oppose (Engesser et al., 2017). We argue that countermedia employ a particular political style (Moffitt, 2016) to differentiate themselves from the mainstream, as do populist politicians. Following Moffitt, we define political style as repertoires of action (Swidler, 1986) used to politicize issues in the public – that is, how politics is done. Style is analytically distinct from the substantive content of political claims. However, content and style are obviously interrelated, and certain content is particularly suited to be expressed via certain styles – as we will show. Style is not only discursive, if discourse is understood as ‘social use of language’ which ‘contributes to the ‘‘construction” of social reality’ (Fairclough and Fairclough, 2012: 78), since style is not only linguistic but also embodied and performed visually and physically. For example, populist leaders dress and act publicly in certain stylistic ways, whether strategically or not, to claim to represent a particular constituency – ‘the common people’ – presenting themselves as a ‘man of the people’ (Moffitt, 2016: 28–50). The importance of style is particularly amplified in today’s hybridly mediated public spheres (Chadwick, 2013), in which spectacle is vital for success. However, in the case at hand, we study style performed in discourse. Scholars of political style often see discourse – understood as language use – as an ‘element of political style’ (Ostiguy, 2017:3), but understanding discourse merely as language use that is part of ‘style’ largely forgoes the broader implications of discourses as constitutive of systems of knowledge that uphold constellations of power with concrete outcomes. For example, it has been argued that discursive representations of Muslims as a detestable outgroup in online discussions contribute to ‘fueling hatred’ and may even ‘encourage physical attacks’ (Törnberg and Törnberg, 2016: 134). This is why we focus on what we call the discursive style of a right-wing populist countermedium, the Finnish website MV Media (WTF Media) 1 , and its role in the wider political and media sphere, particularly in the politicization of migration: the process whereby anti-immigration political actors raise the issue of migration to the public sphere, make political claims, spread information and appeal to emotions, particularly to a sense of crisis, to create and utilize collective political agency to oppose immigration (Horsti, 2015; Luhtakallio, 2012; Moffitt, 2016). As such, our study of discursive style takes cues both from earlier studies on political style (e.g. Moffitt, 2016; Ostiguy, 2017) and of political discourse (e.g. Atton, 2006; Horsti, 2015; Törnberg and Törnberg, 2016; Wodak, 2015). The focus on discourse is taken because of the texthttps://doi.org/10.1016/j.dcm.2019.100326 2211-6958/Ó2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). ⇑ Corresponding author. E-mail addresses: [email protected] (T. Ylä-Anttila), gwenaelle.bau- [email protected] (G. Bauvois), [email protected] (N. Pyrhönen). 1 The website was first named ‘Mitä Vittua?! -lehti’, literally ‘What The Fuck?! News’, later simply known as MV Media, which we translate as WTF Media. Discourse, Context & Media 32 (2019) 100326 Contents lists available at ScienceDirect Discourse, Context & Media journal homepage: www.elsevier.com/locate/dcm based nature of countermedia and to broaden the ‘stylistic’ viewpoint to better incorporate the discursive construction of social reality. The focus on style is employed because stylistic repertoires are central to populist mobilization. We conduct a mixed-methods text analysis, combining computational (Törnberg and Törnberg, 2016) and discourse-analytical (Fairclough and Fairclough, 2012) methodology, complemented with literature on populist style (Moffitt, 2016), to answer the questions: What defines the countermedia discursive style and how does it differ from mainstream media style? Is the countermedia style also used in the arena of ‘high politics’ (i.e. Parliament), and if so, how and by whom? The Finnish case is particularly relevant because after decades of political stability and coalition governments by moderate parties, the political system was shocked in 2011 by the explosive success of the radical right-wing populist Finns Party, in 2015 by their inclusion in government, and in 2017 by their split into a moderate and a hard-line anti-immigration faction. During these years, the meteoric rise of right-wing populism has been strongly fuelled by online anti-immigration activism (Hatakka, 2017), a network that includes WTF Media. And immigration has indeed been successfully politicized, as it has become one of the central polarizing issues of Finnish politics along a liberal–conservative axis (Westinen, 2015). The analysis proceeds in four phases: (1) we computationally identify key expressions in WTF’s discursive style by comparing it to mainstream media; (2) we qualitatively analyse WTF’s discursive style; (3) we search for these countermedia-typical expressions in the Finnish parliament and quantify their prevalence to assess the relevance of the countermedia style for politics; (4) we qualitatively analyse how Finnish parliamentarians use the countermedia style to discuss immigration. The results of our analysis show that the countermedium in question politicizes immigration in the characteristically populist style of performing crisis; that is, it attempts to politicize migration and frame it as a threat, to ultimately affect public opinion. Furthermore, together with its political allies on the populist right, WTF Media seems to be somewhat successful, judging from the prevalence of countermedia-style expressions in parliament, particularly by the right-wing populist Finns Party. But when politicians use this style in the highpolitical arena of parliament, they employ what we call a ‘politician’s filter’ in order to appear credible (see Ostiguy, 2017); this is a balancing act, however, if they also wish to retain the ‘man of the people’ appeal of the populist style. 2. Data This study is based on three datasets: a countermedia dataset (WTF), a mainstream media dataset for comparison (HS; introduced below), and parliamentarians’ addresses to the floor in plenary sessions of the Finnish parliament. WTF is the most influential right-wing populist countermedium in Finland (Ylä-Anttila, 2018), similar to, if less professional than, Breitbart News in the USA, and comparable to Fdesouche in France. WTF Media quickly rose to the top 20 most popular Finnish websites in 2015, despite being branded by the mainstream media as ‘fake news’ or a ‘hate site’ that should not be trusted (Kaleva, 2015; YLE, 2016). It resembles a tabloid magazine, but does not adhere to journalistic standards, and frequently publishes unconfirmed rumours of crimes committed by immigrants as news. Other pieces are based on police reports or mainstream media stories and describe actual events; by no means is all the content ‘made up’. The site has an explicit anti-immigration agenda coupled with a vulgar style, claiming to uncover ‘the ugly truth’ about immigration that ‘the mainstream media has silenced’. The importance of WTF Media for Finnish public debates on immigration and ‘post-truth politics’ – that is, challenging knowledge authorities – has been significant. We included in our analysis all articles published by WTF Media between 1 January 2015 and 31 December 2017 that were available on its site, mvlehti.net, on 20 January 2018. The article titles and body text were extracted using the BeautifulSoup Python library, discarding formatting, images and links. The resulting dataset contained 2,059,592 words in 15,015 articles. For purposes of comparison, we used the full archive of all content published in Helsingin Sanomat (HS), Finland’s main daily newspaper 2 , during the same timeframe (2015–2017). This data was kindly provided to us by the newspaper itself and totalled 26,025,060 words in 131,957 articles. This dataset enabled us to compare WTF to arguably the most mainstream news media outlet in Finland. The third dataset, the minutes of the plenary sessions of the Finnish parliament, was provided by the Language Bank of Finland. This data was not available for 2017; the latest published minutes are from summer 2016. However, the year 2015 is contained in full (which was a particularly important year for the migration debate, as the number of asylum-seekers peaked in September 2015 at more than 30-fold that of a typical month). The third dataset contains the full transcripts of Parliament between January 2015 and June 2016 – that is, 183 parliamentary sessions, 6,819 speeches, and 3,354,519 words. 3. Methods We take a mixed-methods approach to text analysis, which combines computational methods with qualitative discourse analysis (e.g. Bail, 2014; Hillard et al., 2008; Laaksonen et al., 2017; Törnberg and Törnberg, 2016). The role of computational methods is not to replace qualitative analysis, but to complement it. The aim is to combine the ‘thick description’ (Geertz, 1973) of close reading with the representativeness, reproducibility and quantifiability of computational ‘distant reading’ (Babones, 2016) in a ‘best of both worlds’ approach. By computing frequencies of word sequences we can describe the numerical properties of a text (such as the prevalence of a particular expression), and then select samples for qualitative analysis in a representative and reproducible way, avoiding some of the human bias in sampling. This two-stage analysis was conducted first for the countermedia dataset and then for the parliamentary dataset, making for a total of four distinct analysis phases, described below. First, using Python’s Natural Language Toolkit, we identified countermedia-specific expressions used in debates on migration by comparing the frequencies of n-grams (sequences of words) in WTF Media and HS, and selected a sample of the most typical WTF posts based on this comparison. Second, we conducted a qualitative discourse analysis of the sample informed by Moffitt’s (2016) work on political style, using the qualitative analysis software Atlas.TI. To ensure sensitivity to the data, we utilize the principle of ‘multiple coding’ (Barbour, 2001), that is, inductive and reflexive collaborative coding, in which each researcher closely read an equally-sized portion of the sample, then met to discuss the themes and recurring stylistic repertoires we found, compare and harmonize the coding scheme, and finally re-coded their portion of material in accordance with the commonly agreed features of interest. Third, we searched for these countermedia expressions (n-grams) in the parliamentary dataset and described their 2 Helsingin Sanomat occupies an exceptionally important position in the Finnish media landscape since its circulation is more than three times that of the nearest competitor, Aamulehti, and in fact largest in all of the Nordic countries. 2T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 frequencies over time and by party. Finally, we analyse the usage of these expressions in parliament, again based on an inductive, collaborative and reflexive qualitative discourse analysis in Atlas.TI 3 . 4. Phase 1: Computational analysis of WTF Media N-grams are simply sequences of words in a corpus of text – that is, words that occur in a particular order. Bigrams are sequences of two words: for example, the bigram ‘computational analysis’ occurs once in the heading above. A count of such frequencies gives a general idea of the most common themes and tones in a text. A significant part of WTF Media’s content seems to be reporting on crimes allegedly committed by migrants, particularly sexual assaults. The most common bigrams in WTF were ‘police report’ (2201 occurrences), and ‘source: police’ (2164 occurrences) (see Table 1). We also calculated frequencies for trigrams (three-word sequences), but only use bigrams in the analysis for the sake of simplicity, since the results using trigrams were very similar. To identify the kinds of expressions that are typical of WTF Media in particular, we compared the WTF bigram frequencies to the HS dataset, representing mainstream media text (see Bossetta, 2017; Daniels, 2017 for methodological analogues). Contrary to WTF, the most common bigrams in HS are predictably not about crime and immigration, but general, neutral and descriptive in nature: last year, million euros, large part, billion euros, next year, and so on. We computed a list of the 100,000 most common bigrams in WTF and looked up the relative frequency (occurrences per total words in dataset) of each bigram in HS to see which were disproportionately represented in WTF. A sample of this comparison can be found in Table 2. Many bigrams were such that their mere appearance could not be considered indicative of discourse typical to WTF: politicians’ names, for example, as well as general expressions related to crime and police. Furthermore, our primary interest was in the politicization of migration. Thus, we turned to the interpretive side of our mixed-methods approach. With the help of a research assistant, we considered two lists: the 400 most common bigrams in WTF that did not appear in HS at all, and the 400 most common bigrams in WTF that appeared much more often in WTF than in HS, sorted by ratio (Table 2). From these lists, we selected those that were related to immigration or nationalism as determined by at least three of four coders. Proper nouns (the names of individuals and organizations) were excluded. The result of this qualitative selection is a list of the 67 most countermedia-typical bigrams used for discussing migration (Table 3). To reiterate, we first computationally listed bigrams in order of how much more common they were in our countermedia data than in our mainstream media data, then qualitatively selected the bigrams related to the politicization of immigration. These expressions reflect how WTF discusses immigration: through vocabularies of crime and legal status, as well as the nation, nationalism, terrorism, and Islam. The bigrams capture both thematic content and style: they do not just refer to a particular issue (immigration), but include certain interpretations of those issues – a focus on illegality, crime, race, the nation, and gender. For the qualitative phase, we took a sample of 27 WTF Media articles by selecting those in which one or more of the 67 bigrams in Table 3 occur at least 12 times. This is based on an elbow cut-off of the frequency of the immigration-related, countermedia-typical bigrams in WTF articles. The bigrams were concentrated in certain articles, and it was those that we selected for our qualitative sample, leaving out the ‘long tail’ of articles in which the bigrams were mentioned only sparsely. This sample, used in Phase 2, consists of 28,031 words. 5. Phase 2: Qualitative analysis of the countermedia style In this analysis, we argue that the countermedia discursive style forms a seedbed for political mobilization based on collective identification as disentitled, disenfranchised and victimized, the resulting resentment of which is often harnessed to support populist politics (Hochschild, 2016; Kriesi et al., 2006). Below, we analyse five elements of the countermedia style: depictions of internal enemies, a dualist stance towards authorities, struggles over concepts, majority victimization, and vulgarity. To start with, one dominant feature of the countermedia style is its descriptions of ‘internal enemies’, a tool typical to populism. In the case of WTF, this means above all the so-called suvakit, ‘toletards’ (a portmanteau of ‘tolerant’ and ‘retard’): the liberal ‘useful idiots’ who tolerate immigration. WTF argues that priests, academics, green–left politicians and of course pro-migration activists are internal enemies who facilitate the attack supposedly being carried out by external enemies (i.e. Muslim immigrants, by and large). The political green–left wing, in collaboration with the unpatriotic liberal bourgeoisie, are going to rot the education system, too. (WTF 10 Nov 2017) 4 3 Codebooks available upon request. Table 1 The fifteen most common two-word sequences (bigrams) in WTF Media, excluding proper nouns, translated into English. Inflected forms of words were lemmatized, that is, ‘police request’ includes ‘police requests’, ‘police requested’ etc. Bigram Frequency police report 2201 source police 2164 last year 854 social media 700 photo police 692 police request 659 large share 652 to catch [a criminal] 510 photo [of the] day 493 Finland first 491 illegally [in the] country 471 time between 458 video [of the] day 455 sexual abuse 434 young man 425 Table 2 Comparison of relative frequencies of bigrams in WTF and HS. Inflected forms of words were lemmatized, that is, ‘police publish’ includes ‘police publishes’, ‘police published’ etc. Bigram Relative frequency in WTF Relative frequency in HS Ratio request report 0.0001792 0.0000015 123 police inform 0.0010687 0.0000090 119 inform police 0.0001262 0.0000011 117 steal property 0.0000646 0.0000011 58 request information 0.0000816 0.0000015 56 can notify 0.0001316 0.0000024 55 photo large 0.0000471 0.0000009 53 police publish 0.0001554 0.0000030 53 time between 0.0000583 0.0000011 52 4 All quotations have been translated from Finnish by the authors. WTF often republishes other sites’ posts verbatim, and does not always make this clear. We assume content nevertheless reflects WTF Media’s positions. T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 3 As is typical of populist argumentation, elites make up one category of internal enemies in our countermedia sample, and include government ministers, EU officials and the mainstream media. Politicians are accused of being unable to criticize Islam because of their excessive ‘tactfulness’ (WTF 30 May 2017), for example – an explicit denunciation of the ‘high’ political style, that is, wellbehaved professionalism, which Ostiguy (2017) notes is the opposite of populist political style. In arguments blaming elites, the nature of the countermedia as populist opposition media (both towards the establishment and towards the mainstream media) is made clear. This church propaganda about aid for migrants is advocated by the Church Resources Agency, led by millionaire Jukka Alho, who became known, when he was CEO of the Itella logistics corporation from 2009 to 2011, for receiving over a million euros in pension bonuses. (WTF 31 Dec 2016) The category of internal enemies often overlaps with that of authorities, such as the police and the mainstream media, to which WTFhasadualiststance:they are respectedin principlebutstrongly condemned when they are perceived as failing to protect the interests of the people. For example, the mainstream media were often accused by WTF of not being politically neutral. This is noteworthy, considering that WTF itself is decidedly non-neutral. WTF seems to see itself as fulfilling a different function than the mainstream media. This is an acknowledgement of the importance of the independent press in principle, though its current form is accused of being corrupt. Particularly YLE, the Finnish national broadcaster, is singled out as an instrument of ‘propaganda’: YLE is learning to operate as effectively as the Soviet media, which always checked the facts and told the people ‘what they should think about this issue’. (WTF 23 Aug 2017) Similarly, WTF gives a central role to the police and their fight against ‘illegal immigration’ and the resulting threats to safety. Such stories proliferate in WTF and are presented in an overwhelmingly positive manner. According to WTF, the police are protecting the country against ‘infiltrated terrorist agents’, but their efforts are thwarted by a lack of support from indifferent or corrupt legislators: ‘Authorities are powerless until the legislation is updated’ (WTF 29 May 2017). In several articles, WTF defends police officers who have spoken publicly about immigrants in a derogatory and discriminatory manner, and accusations of hate speech are viewed as ‘stupidities’ (WTF 26 Sep 2017) that undermine police work. The police are also contrasted with the church, which is accused of ‘irresponsibly’ protecting immigrants despite the danger they represent to society and to native Finns. And police officers that speak out in favour of immigration or multiculturalism are strongly denounced as ‘political’; from WTF’s point of view, they are traitors of the people. Political debates are always also about the concepts and connotations used in discourse, as well as the power relations and conceptions of justice carried by these concepts. WTF engages in this conceptual struggle not only implicitly, by using a particular style which furthers its representations and agendas, but also explicitly, such as in articles openly discussing what terms should be used to refer to migrants ‘realistically’ – that is, to emphasize features that are important for an anti-immigration agenda. As we showed previously, a typical framing is that of ‘illegal’ immigrants, which is preferred to ‘undocumented’ by WTF because it is said to be more ‘accurate’. Despite this appeal to accuracy, WTF often refers to immigrants as ‘illegal’ without providing any information about the actual legal status of the person in question. The following quotation demonstrates this conceptual struggle. Note the internal enemy definition, as well. Samuli Suonpää, who has risen to the position of communications planner in the [Finnish Evangelical Lutheran] Church by scheming and personal connections [...] incorrectly uses the word ‘undocumented’, even though in fact many illegals do have documents, just not the right documents to make it legal for them to stay in the country. (WTF 23 Mar 2017) WTF commonly frames the majority as being victimized by minorities, above all immigrants, who ‘cut in line’ (Hochschild, 2016). Asylum seekers are portrayed as ‘terrorists in disguise’ whose victims are the ‘Finnish people’. Aside from actual violent terrorism, this theme is typically linked to welfare redistribution through unsourced anecdotes marked by bombastic hyperbole: This is equality à la Titanic [...] family men take along fertile women and herds of children. [...] The road to a five-star waterfront penthouse opens easily in municipalities’ residential services, as the guidelines perverted by the green–leftists put those presenting themselves as the most wretched at the front of the line. (WTF 4 Oct 2015) Our sample is rife with similar urban tales of welfare abuse, such as a refugee woman ordering 300,000 euros worth of products by mail to a German reception centre (WTF 24 Jul 2016) and the Finnish Ministry of the Interior planning preferential treatment in social services for ISIS fighters (WTF 10 Nov 2017). Such anecdotes narrate majority-victimizing welfare abuse as enabled not only by internal enemies – particularly the toletards – but also by the unwitting benevolent majority itself. Finally, WTF uses an explicit, imaginatively vulgar style to reinforce the general tone of ridicule, resentment and threat. For example, a post argued that elites harness ‘a demagogy that cries about otherness and diversity’ to help in ‘fingering the minorities ombudsman under her skirt’ (WTF 4 Oct 2015), and another one Table 3 The most WTF-typical bigrams related to immigration or nationalism, proper nouns excluded, translated into English and sorted into thematic categories. Migration and refugees Crime and illegality Nation and nationalism country stay illegal migrant own people asylum-seeker tell country illegally Finnish interest persecution flee illegally country country people political asylum illegally Finland Finnish person country stay illegal immigrant Finnish justice deport country illegal intruder Western society stay Finland illegal immigration Finnish taxpayer return homeland deport country Finland people open border illegal incomer taxpayer money country intrude Finnish safety against Finnish let [into] country illegally stay Finnish tradition immigration multiculturalism illegally country national resistance close border illegal Muslim Finnish against multicultural society illegal dweller Sweden migration [office] reception-center chief Islam and terrorism Race and racism Gender and ethnicity Islamic terrorism against incite foreign man young Muslim group against foreign-born man Islamic country ethnic background dark-skinned man peace religion incite group white woman Islam religion anti racist white person political Islam ethnic profiling Allahu Akbar Muslim do religion culture Muslim Christian violent radicalize Helsinki Muslim with Muslim 4T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 identified ‘toletard priestesses’ who ‘sordidly sob’ as ‘authorities devise decisive control measures’ against illegal immigration (WTF 3 Apr 2017). As illustrated, the vulgarity often comes with misogyny. The ‘richness [of multiculturalism] can only be measured in the numbers of bodies mutilated by Muslim terrorists’, and once it is too late to take action, the only thing left to do will be to ‘eat shit and die’ (WTF 24 Jul 2016). To conclude, our analysis illustrates how the countermedia style constructs an overarching sense of disenfranchisement and malcontent, which is central to populist mobilization (e.g. Bonikowski, 2017; Hochschild, 2016; Kriesi et al., 2006), through elements such as internal enemies, a dualist stance towards authorities, conceptual struggles, majority-victimization, and vulgarity. Some of these elements are extensively documented previously in the literature on anti-immigrant discourse (e.g. on enemy constructions see Törnberg and Törnberg, 2016), others, such as vulgar style, have thus far received little attention and may be somewhat specific to the case at hand. 6. Phase 3: Finding the countermedia style in parliament In the next phase, we looked at whether Finnish politicians use expressions similar to those of WTF Media – and if so, which politicians and when. This was done by searching for the countermediatypical immigration-related bigrams in a third dataset consisting of politicians’ speeches. For this purpose, we used the minutes of the plenary sessions of the Finnish parliament between January 2015 and June 2016, which were provided by the Language Bank of Finland. This timeframe spans two governments: the rainbow coalition led by the moderate-right National Coalition until summer 2015, and the right-wing coalition led by the Centre Party, including the right-wing populist Finns Party, from summer 2015 onwards. 44 of the 67 countermedia-typical bigrams were indeed used even in the high-political arena of Parliament – perhaps even surprisingly often (Fig. 1) – but the 23 bigrams that were not used 5 include some of the most explicit references to Muslims and skin colour. This suggests that even fiercely anti-immigrant or antiIslam politicians filter their speech in Parliament in order to not appear explicitly racist, as we argue below. The frequency of countermedia-typical phrases peaked in January of 2015, when the parliament debated a government proposal to extend public healthcare provision to undocumented migrants (a proposal which did not pass). It is notable that the Finns Party was in opposition at the time and obviously heavily opposed the proposal. Previous research has shown that politicians tend to intensify their populist appeals when in opposition (Bonikowski and Gidron, 2016; Ernst et al., 2017). However, the ‘refugee crisis’ of autumn 2015 is also visible as a momentary increase in countermedia-typical expressions in Parliament in October, November and December of 2015, as seen in Fig. 1. As shown in Fig. 2, the anti-immigration right-wing populist Finns Party strongly dominates in the usage of countermediastyle bigrams about immigration. However, it should be noted that not every mention of ‘illegal immigration’, for instance, is an antiimmigration expression. Some MPs use the expressions to refer to the discourse of their opponents. In the case of the Left Alliance, for example, the use of countermedia expressions is largely an artefact of MP Anna Kontula’s speeches in which she criticizes the Finns Party for referring to people as ‘illegals’. Still, Fig. 2. gives an overall impression of which parties most often use this language, and from the qualitative analysis which follows, it is evident that the vast majority of such usage is indeed anti-immigration discourse, rather than criticism of it. The qualitative sample for Phase 4 was selected similarly to the selection of the Phase 2 sample. A search of the 6820 total addresses to the floor delivered between January 2015 and June 2016 for the 67 countermedia-style bigrams resulted in 298 addresses. We then sorted these addresses by the number of countermedia-style bigrams they contained and identified an elbow cut-off at three bigrams per address. Using this criterion, we selected a sample of 21 addresses that is manageable in scope for qualitative analysis (24,257 words), and roughly matches the size of the Phase 2 qualitative corpus (28,031 words). The distribution of sample addresses across political parties is strongly Finns Party -dominated, as is the usage of countermedia-style expressions in the broader dataset, indicating that right-wing populist MPs use the countermedia style for debating immigration much more commonly than other MPs (in the qualitative subsample of 21 addresses, 13 are by Finns Party MPs, 3 by National Coalition MPs, 3 by Social Democrats MPs, 1 by a Centre Party MP and 1 by a Left Alliance MP). As a final indicator before moving on to the qualitative analysis, we looked at the density of countermedia-style bigrams in parliaFig. 1. Frequency of countermedia-style immigration-related expressions in the Finnish parliament January 2015–June 2016. 5 The 23 bigrams we did not find in parliamentary speeches were Allahu Akbar, Helsinki Muslim, Islamic terrorism, national resistance, against incite, incite group, illegal intruder, illegal Muslim, with Muslim, Muslim Christian, Muslim do, young Muslim, political Islam, peace religion, Sweden migration [office], Finnish tradition, dark-skinned man, foreign man, foreign-born man, white person, white woman, against group, violent radicalize. 13 59 46 98 21 35 17 0 20 40 60 80 100 120 Democrats Naonal Coalion Others Social Finns Party Green Party Centre Party Alliance Fig. 2. Use of countermedia-style expressions about immigration in the Finnish parliament by party, January 2015–June 2016. T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 5 mentary speeches and WTF Media to determine how much more common such expressions are in countermedia than in Parliament. The 524 instances of countermedia-style bigrams in the WTF Media subsample made up 3.7% of the subsample’s 28,031 words; the density was roughly six times that of the parliamentary speeches (82 bigrams in 24,527 words, or 0.7%), even though both subsamples contain those documents in which the bigrams occur most often. Additionally, as noted above, 23 of the 67 bigrams are not present in the parliamentary data at all. This suggests that MPs use a ‘politician’s filter’ when discussing immigration to appear credible. However, there is considerable variation in how strict this filter is, as we will show below. 7. Phase 4: Qualitative analysis of the countermedia style in parliament The immigration-related phenomena in the parliamentary sample are thematically very similar to the WTF Media sample. However, partly because the parliamentary sample includes the January 2015 debate over healthcare for undocumented migrants, the parliamentary data exhibits a more concentrated focus on the ‘illegal’ status of certain migrants. The countermedia style does not typically define a complete address to the parliamentary floor, but rather manifests in concentrated passages in the addresses, often preceded and followed by more typical political language aimed at mainstream audiences. This ‘politician’s filter’ is used to restrain the use of countermedia style in the parliament, rendering the addresses more ‘ordinary’ and allowing politicians to simultaneously appeal to both radical and moderate constituents. The authoritarian emphasis on control and safety is perhaps the most notable commonality in how the countermedia style of debating immigration is harnessed in both WTF and in Parliament. The Finns Party MPs are not alone in underlining the importance of authorities’ resources to control immigration, with mainstream politicians also using the language of ‘illegality’. The following excerpt exemplifies how using a strict politician’s filter allows an MP to present himself as a level-headed proponent of the bare minimum measures necessary to ensure the authorities’ capabilities to maintain control: The healthcare system should at least acknowledge that [...] the person in question is illegally residing in the country, and at least at some stage [...] convey this information to the authorities responsible for monitoring legal or illegal residence. (Mauri Pekkarinen, Centre Party, 28 Jan 2015) The Finns Party was the most vocal in invoking citizens’ fears, presenting themselves as the upholders of safety. As in WTF Media, parliamentary addresses by Finns Party MPs characterized migrants as criminals, positioning Finnish women and children as the most vulnerable members of society and at risk of becoming victims of violence perpetrated by ‘foreigners’ and ‘illegal’ immigrants: The people are worried over whether it is safe to walk the streets [...] [H]ow can we protect our children, our girls and women on the streets, without having the law act as a deterrent for people who do not respect this kind of legislation? (Mika Niikko, Finns Party, 11 Feb 2016) The control and safety elements of the countermedia style are commonly linked to taking a stand against the profoundly different immigrant Other, who is identified as the main source of insecurity. Such paternalistic and authoritarian language on security is common in essentialist nationalism, which may also seek to explain the immigrant’s role in the alleged erosion of safety as deriving from irreconcilable value systems: They have completely dissimilar human values and a different understanding of how society should be built around here. (Mika Niikko, Finns Party, 24 May 2016) Performing crisis (Moffitt, 2016) is a prominent feature of the countermedia style through which both WTF Media and populist politicians commonly articulate urgent threats to safety. As it is based on alarmist depictions of immigration-related threats, such claims are not easily softened by the politician’s filter. Finns Party MPs, in particular, are heavily engaged in not only reacting to the ‘refugee crisis’, but actually performing it through references to ‘immigrant crime’ in all its forms – drugs, terrorism, rape and murder: We have had enough with the rapes and robberies. Now is the time for measures. Next spring will be too late. Then the landslide of incomers will be uncontrollable, if we don’t take serious action now. Soon we, too, will have slums where violence, drugs and terror reign supreme [...] Currently, we are integrating ISIS fighters here, even those who have been identified as perpetrators of 11 murders in that country of theirs. (Teuvo Hakkarainen, Finns Party, 15 Dec 2015) The crisis being performed is not restricted to crime alone, but also incorporates threats to safety vis-à-vis social and economic conditions, articulating a slippery slope whereby an influx of innumerable immigrants will bring an end to the whole system of welfare redistribution: [T]he whole system cannot function in any way if the paperless are given even a temporary residence permit, because this would mean that the paperless can access our blissful income support, for instance. And how many paperless migrants, declining to return to their home countries, are there in Europe? One million, two – who knows how many million? We cannot resolve these things so that we lower the threshold so low that no one needs to adhere to any laws here. (Mika Niikko, Finns Party, 11 Feb 2016) Occasionally, politicians from the mainstream right employ a similar discursive strategy of instilling doubt and worry. In these cases, the politician’s filter can manifest in verbose expressions of his or her utmost reluctance to perform crisis: I, myself, do try to avoid painting pictures of unnecessary threats, but as we already know today, next year we will have thousands of people with negative asylum decisions who will stay in Finland – people residing illegally in Finland. I don’t think it can be regarded as an unnecessarily emphasized threat that the more expansive the social and health services they have, [...] the harder it is to make them leave Finland voluntarily. (Ben Zyskowicz, National Coalition, 2 Dec 2015) The above excerpt illustrates that many features of the countermedia style, such as the performance of crisis and support for authoritarian control, often have a distinct connection to the moral, nationalist dimension of the welfare state (Pyrhönen, 2015). This anti-immigration discourse is based on the assertion that the welfare state and its benefits should apply to Finns alone – whether defined via citizenship or some other standard of belonging to the Finnish nation. While the moral condemnation of ‘welfare abuse’ features highly in both WTF and in Parliament, the subject matter of the parliamentary debates – health care and social services – creates even more room for anecdotal evidence of non-Finns ‘cutting in 6T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 line’ (Hochschild, 2016). The parliamentary debates on expanding the provision of publicly-funded healthcare to irregular migrants highlight a focus on moral deservingness, in which MPs assert that – unlike immigrants – Finns have a moral right to access these state-provided services. ‘Illegal’ immigrant status is commonly presented as a moral justification for considering migrants undeserving of economically costly healthcare, and, paradoxically, the fact that granting a migrant legal status would allow ‘unjust’ access to these services is used as justification for not granting such status: Public services are not meant for illegal people who could come here and enjoy the same level of health services as Finnish citizens and those legally residing in the country. (Vesa-Matti Saarakkala, Finns Party, 13 Mar 2015) Similar arguments that build on the notion of undeservingness appear commonly in both WTF and in Parliament. However, in the latter context, the politician’s filter frames welfare nationalist arguments in a more technocratic and rationalist manner than the pure countermedia style would allow. In this argument, the cost of welfare is addressed in economic terms, thereby obscuring the intimate connection between the normative (deservingness) and rationalist (economic) realms. Accordingly, parliamentary addresses by traditional centre–right politicians tend to embrace the economic frame as ‘neutral’ and ‘objective’, as opposed to the right-wing populists’ explicitly moralistic framing. The politician’s filter also affects how immigrants are depicted in welfare nationalist passages. Parliamentary addresses emphasizing the moral dimension of welfare nationalism are common among Finns Party MPs, who argue – exactly like the WTF Media articles do – that immigrants are ‘intruders’ and, as such, morally reprehensible. When MPs from other parties use welfare nationalist arguments, they are often filtered through an economic frame that portrays immigrants simply as rational human beings. According to this claim, these people – just like any other people – will invariably take advantage of welfare services unless the MPs devise specific legislation to prevent this: It is claimed that expanding welfare provision to undocumented immigrants wouldn’t constitute an incentive for immigrants to come to Finland. But there are no studies to prove such a claim. I don’t believe this would result in whole populations coming here, but I do believe people will do what’s rational and logical for themselves. (Ben Zyskowicz, National Coalition, 28 Jan 2015) As such, while the ‘raw’ and ‘filtered’ flavours of welfare nationalism are both used to present anti-immigration arguments, they propagate different perceptions of the humanity of migrants. Right-wing populist politicians favour the raw variant, arguing that the autochthonous population is profoundly different from the immigrants who prey on the vulnerabilities of the Finnish system and the Finnish population alike. The politician-filtered framing favoured by centre–right MPs, on the other hand, presents the economically erosive effect of the basic human self-interest of migrants as a reason to curb immigration. Finally, there is one prevalent feature of the countermedia style in the parliamentary corpus that is completely monopolized by the right-wing populist Finns Party: the practice of positioning political opponents as internal enemies of ‘the people’. The explicit manifestations of this enemy category refer to either specific politicians or to the whole political process, and are conveyed through expressions of outrage that are hard to differentiate from passages in WTF Media: The green–left fools have spent the whole autumn yelling about every single budget cut in the parliament. Never have they said a word against the half a billion euros of extra funding that is currently being spent to maintain the infiltrators’ benefits. (Teuvo Hakkarainen, Finns Party, 15 Dec 2015) The Finns do not accept the proposed bill, and we regard it as an outright expression of ridicule towards the Finnish people [...] [T]he impudence of bringing this kind of proposal here is incomprehensible. (Hanna Mäntylä, Finns Party, 28 Jan 2015) This positioning is also employed in a more implicit fashion, whereby a self-depreciating ‘low style’ underlines the MP’s own role as a person ‘of the people’, an underdog challenger to ‘the elite’: Listening to these addresses coming from the mouths of professional politicians, we bow as they use words fancy enough to certainly bring even a telephone pole to its knees right here. They refer to small children and such things [...] This parliament has lost touch with the citizens, the ordinary people. (Teuvo Hakkarainen, Finns Party, 28 Jan 2015) Occasionally, MPs reluctant to employ this position themselves nevertheless lend support to those who do adopt this position: MP Hakkarainen’s address was the kind of speech from a man of the people that comes directly from the heart and from the deepest of human emotions, and in this sense I respect that someone calls things by their proper names, even if those kinds of people may sometimes be misinterpreted. (Mika Niikko, Finns Party, 11 Feb 2016) This method of harnessing the countermedia style allows politicians to both endorse and transcend the ‘lowest’ registers (Ostiguy 2017) of the countermedia style, positioning themselves simultaneously as men of the people and as respectable politicians. 8. Conclusions This paper has utilised computational and qualitative discourse analysis to examine the discursive style of a right-wing populist countermedium, WTF, and its role in the Finnish public and political spheres. We have argued that the concept of countermedia is useful for understanding hyper-politicized online news outlets that engage in a type of ‘citizen journalism’ and position themselves in opposition to the mainstream media and the broader establishment, attempting to influence public opinion by discursive means. And we have shown, both qualitatively and quantitatively, that the countermedia style is indeed used even in parliamentary debates, which should highlight the relevance of studying countermedia. In Phase 1, we identified the countermedia style of debating immigration by comparing the bigram frequencies of WTF and HS, and found that WTF Media discusses migration mostly in conjunction with crime and illegality, the police, and safety, in a crisisperforming populist style. In Phase 2, we conducted a qualitative, interpretive analysis of the countermedia style identified in Phase 1, finding an overarching incitement of sense of disenfranchisement using the elements of internal enemies, a dualist stance towards authorities, conceptual struggles, majority victimization, and vulgarity. In Phase 3, we identified and measured the use of countermedia-typical expressions in the Finnish parliament. We found that the frequency of usage peaked in January 2015 with the debate over migrant healthcare, with a second wave of usage in late autumn 2015, co-occurring with the peak in asylum applications. The Finns Party MPs used this language more often than others. In Phase 4, we qualitatively analysed the usage of countermedia-typical expressions in Parliament. We found that T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326 7 politicians employ, to varying degrees, a ‘filter’ in harnessing the countermedia style. Politicians wishing to channel the power of populism and present themselves as being ‘of the people’ face a balancing act: it is not possible for an MP to fully embrace the scaremongering, crisis-performing populist countermedia style and appear credible at the same time, at least not in the Finnish context at this point in time. Some politicians come fairly close, however. Finally, some caveats. We do not imply that the countermediastyle expressions used by MPs would necessarily be drawn directly from the countermedia, although surely some politicians do read these websites and appropriate their messages. Rather, there is a more widely used discursive style of politicizing migration that both the countermedia and right-wing populist politicians tap into and utilize (see e.g. Horsti, 2015; Törnberg and Törnberg, 2016; Wodak, 2015). This study provides further evidence that this discursive style is fairly international and can be applied not only to overtly political communication (e.g. party or activist group platforms) but also to re-frame news events to create hyper-partisan news media. Further research could assess some of the shortcomings of this study; for example, to what extent are the features found here specific to a) countermedia as a format as opposed to other forms of communication, b) Finland as a polity and a national public sphere? Such assessments could be done by comparing countermedia not only to mainstream media, as we have done, but to other forms of anti-immigration communication, and Finnish countermedia to other countries. Also, one could study the possible change in discursive style over time. Our perception certainly is that ever-more transgressive and radical language is becoming more and more common even on the high-political arena in Finland, as elsewhere. Funding This work was funded by the Helsingin Sanomat Foundation (grant 201700133), the Academy of Finland (grant 296969) and the Alfred Kordelin Foundation. Declaration of competing interest None References Atton, C., 2006. Far-right media on the internet: culture, discourse and power. New Media Soc. 8 (4), 573–587. Babones, S., 2016. Interpretive quantitative methods for the social sciences. Sociology 50 (3), 453–469. Bail, C.A., 2014. The cultural environment: measuring culture with big data. Theory Soc. 43 (3), 465–482. Barbour, R.S., 2001. Checklists for improving rigour in qualitative research: a case of the tail wagging the dog? BMJ 322 (7294), 1115–1117. Bonikowski, B., 2017. Ethno-nationalist populism and the mobilization of collective resentment. British J. Sociol. 68 (S1). Bonikowski, B., Gidron, N., 2016. The populist style in American politics: presidential campaign discourse, 1952–1996. Soc. Forces 94 (4), 1593–1621. Bossetta, M., 2017. Fighting fire with fire: mainstream adoption of the populist political style in the 2014 Europe debates between Nick Clegg and Nigel Farage. British J. Politics International Relations 19 (4), 715–734. Chadwick, A., 2013. The hybrid media system. politics and power. Oxford University Press, Oxford. D’Ancona, M., 2017. Post-truth. the new war on truth and how to fight back. Ebury Press, London. Daniels M., 2017. The Language of Hip Hop. Available at: (accessed 20 March 2018). <https://pudding.cool/2017/09/hip-hop-words/>. Engesser, S., Fawzi, N., Larsson, A.O., 2017. Populist online communication: introduction to the special issue. Informat. Commun. Soc. 20 (9), 1279–1292. Ernst, N., Engesser, S., Büchel, F., et al., 2017. Extreme parties and populism: an analysis of Facebook and Twitter across six countries. Informat. Commun. Soc. 20 (9), 1347–1364. Fairclough, I., Fairclough, N., 2012. Political discourse analysis: a method for advanced students. Routledge, London. Geertz, C., 1973. The Interpretation of cultures: selected essays. Basic Books, New York. Hatakka, N., 2017. When logics of party politics and online activism collide: the populist Finns Party’s identity under negotiation. New Media Soc. 19 (12), 2022–2038. Hillard, D., Purpura, S., Wilkerson, J., 2008. Computer-assisted topic classification for mixed-methods social science research. J. Inf. Technol. Polit 4 (4), 31–46. Hochschild, A.R., 2016. Strangers in their own land: anger and mourning on the american right. The New Press, New York. Horsti, K., 2015. Techno-cultural opportunities: the anti-immigration movement in the Finnish mediascape. Patterns Prejudice 49 (4), 343–366. Kaleva., 28 Mar 2015. Vihasivusto tienaa isoja tuloja tunnettujen suomalaisyritysten mainoksilla. (accessed 15 Nov 2018). <http://www.kaleva. fi/uutiset/kotimaa/vihasivusto-tienaa-isoja-tuloja-tunnettujensuomalaisyritys ten-mainoksilla/693237/>. Kriesi, H., Grande, E., Lachat, R., et al., 2006. Globalization and the transformation of the national political space: six European countries compared. European J. Political Res. 45 (6), 921–956. Laaksonen, S.-M., Nelimarkka, M., Tuokko, M., et al., 2017. Working the fields of big data: using big-data-augmented online ethnography to study candidate– candidate interaction at election time. J. Informat. Technol. Polit. Luhtakallio, E., 2012. Practicing Democracy. Local Activism and Politics in France and Finland, Palgrave. Moffitt, B., 2016. The global rise of populism: performance, political style, and representation. Stanford University Press, Stanford. Moffitt, B., 2018. Populism and media in Western Europe. In: de la Torre, C. (Ed.), Routledge Handbook of Global Populism. Routledge, London, pp. 235–248. Ostiguy, P., 2017. Populism: a socio-cultural approach. The Oxford Handbook of Populism. Oxford: Oxford University Press. Pyrhönen, N., 2015. The true colors of Finnish welfare nationalism: consolidation of neo-populist advocacy as a resonant collective identity through mobilization of exclusionary narratives of blue-and-white solidarity. University of Helsinki, Helsinki. Ritzer, G., Jurgenson, N., 2010. Production, consumption, prosumption: the nature of capitalism in the age of the digital ‘prosumer’. J. Consumer Culture 10 (1), 13– 36. Sunstein, C.R., 2017. #republic. Princeton University Press, Princeton. Swidler, A., 1986. Culture in action: symbols and strategies. Am. Sociol. Rev. 51 (2), 273–286. Törnberg, A., Törnberg, P., 2016. Muslims in social media discourse: combining topic modeling and critical discourse analysis. Discourse, Context Media 13, 132–142. Westinen, J., 2015. Cleavages in contemporary Finland. a study on party-voter ties. Åbo Akademi University Press, Turku. Wodak, R., 2015. The politics of fear. what right-wing populist discourses mean. SAGE, London. YLE., 16 Sep 2016. Valheenpaljastaja: Varoituslista valemedioista – älä luota näihin sivustoihin. (accessed 15 Nov 2018). <https://yle.fi/aihe/artikkeli/2016/09/ 16/valheenpaljastaja-varoituslistavalemedioista-ala-luota-naihin-sivustoihin>. Ylä-Anttila, T., 2018. Populist knowledge: ‘post-truth’ repertoires of contesting epistemic authorities. Eur. J. Cult. and Political Sociology 5 (4), 356–388. 8T. Ylä-Anttila et al. / Discourse, Context & Media 32 (2019) 100326