My plan for the Netherlands is called Liberation.
And liberation begins with stating the facts.
Geert Wilders (2016)


## Introduction

In the context of his ongoing political argument that Muslims in the Netherlands are the major threat to its society and identity, Dutch right-wing politician Geert Wilders tweeted the phrase ‘liberation begins with stating the facts’. In doing so, he stressed the necessity of factual information for dealing with (political) issues of a larger scope, while questioning the truthfulness and factuality of traditional institutions and representatives in the government. So, what are the facts that need to be stated? How does he voice this important information and where does he find the necessary information that is apparently missing in the debate?

Wilders took office in 2006 when his party Partij voor de Vrijheid (PVV), or Freedom Party, gained nine of the 150 seats in the House of Representatives. As of the latest elections, in 2017, the PVV holds 20 seats, which means they are the second largest party in the Netherlands (Kiesraad, 2017). Wilders’ main talking points, immigration and Islam, and his strong opinions, for instance his appeal to close the Dutch borders or to ban the Koran, often cause controversy. In making these statements, Wilders fits the description of a classic populist, who emphasizes sovereignty and nativism of the people, identifies and attacks the elite, and who engages in the practice of othering a group of people (Aalberg et al., 2016; Jagers & Walgrave, 2007; Bracciale & Martella, 2017). During his 2017 general election campaign, Wilders avoided the majority of debates organised by mainstream media publications but instead used Twitter as his main outlet and means of communication. He has become, both online and offline, a mediagenic flag bearer for a global rise of new right-wing populist parties, also called the new right.

In this paper we examine claim-making on social media by populist politicians, using Wilders’ Twitter feed as a case study. With the global rise of new right movements, there is a need to come to terms with their unique discourse, which seems to be (at least partly) located outside of the traditional mass media of newspapers and television, and has moved to the new media, namely the online media of social networking sites and blogs. In his arguments, Wilders regularly cites media sources and (official) statistics to support his claim that Dutch society and identity need to be ‘liberated’ from Islam and immigration. We examine the ways in which this politician makes use of facts, statistics, and numbers to support his claims, by analyzing his atypical excerpt of his Twitter activity during his campaign for the Dutch elections for the House of Representatives in 2017. We investigate how Wilders employs numbers to make his arguments appear rational, and how statistics are used to rationalize discrimination.

We use the model of contentious politics, as introduced by McAdam, Tilly and Tarrow (2001) and further defined by Tilly and Tarrow (2015), to study and explain the way Wilders makes claims on Twitter and how this can be placed within his broader political objectives. Framing theory, as introduced by Robert D. Benford and David A. Snow (2000), will be incorporated into this model of contentious politics to elaborate on how Wilders phrases the statements he makes and how he uses statistics and numbers, and to explain how he rationalizes and legitimizes discriminating claims about Dutch immigrant populations.

Our paper presents an analysis of a fundamental aspect of Wilders’ communications, namely the way in which he rationalizes and legitimizes discriminating claims, examined through a mixed method approach of quantitative data analysis and a qualitative critical reading of Wilders’ Twitter timeline. Our main research question is: How did Geert Wilders frame his political claims, specifically about race and ethnicity, through statistics, numbers, and facts on Twitter during his campaign for the 2017 elections for the Dutch House of Representatives? Our aim is to closely examine how politicians can frame a topic to suit their own purposes, and how this is achieved when politicians move to the new media sphere where their views are seemingly less frequently challenged and their statements less frequently verified by the traditional media, which of old was the Fourth Estate1 of the political system. Subsequently in this article we focus on the frame of objectivity through the presentation of ‘factual’ information like statistics. How do these frames – of the factual – work online, and what do they rationalize? Through the concept of information bricolage, we aim to make claims on whether framing practices – in this case, Wilders’ – can be considered new, contentious, and/or transgressive. Ultimately, we hope to contribute to a broader understanding of the ontologically interwoven relationship between politicians and media. Now that all over the western world new populist right wing movements have arisen, while at the same time the media’s gatekeeper role has begun to weaken, we wonder how new right political communication is shaped in this new, and increasingly important, media environment.2

## Contentious Politics

Contentious politics is a theoretical approach to studying political claim-making and will be instrumental to us in explaining Geert Wilders’ political communication. Contentious politics is explained by Tilly and Tarrow as involving:

Interactions in which actors make claims bearing on other actors’ interests, leading to coordinated efforts on behalf of shared interests or programs, in which governments are involved as targets, initiators of claims, or third parties (2015: 7).

Here, contention is a process of social construction, constituted by multiple episodes in which parties are interacting. Actors, in our case politicians, aim to create a justification for action, either by proposing policy or through the media activity in itself, by strategically influencing the dynamics of opportunities, resources, frames, and repertoires in a public way (McAdam et al., 2001: 5). This framework lends itself particularly well in the investigation of social movements that violently and disruptively oppose power structures from totalitarian regimes to capitalism as a whole (Tarrow, 2011). However, McAdam et al. (2001: 6) explain that contentious politics can just as equally be used to analyze conventional political conflicts that have a transgressive element. Transgressive, in this context, can either mean that some of the parties are not formally constituted political actors, or that some of the means used in the contention are unconventional. ‘Transgressive’ can also signify that some of the means used in the contention of the conventional are (very) innovative, such as, in our case, the use of social media. In this case, forms of contention take place within democratic frameworks. This ‘democratic contention’ as Charles Tilly explains:

Takes place in or adjacent to the regimes’ prescribed and tolerated forms of political participation; public meetings, for example, provide vehicles for both established power holders and dissident groups, while election campaigns offer opportunities for electoral, para-electoral, and counter-electoral claim making by organized critics (2004: 30).

These types of situations will perhaps not always stand out as transgressive or contentious, because some of the parties are not formally constituted political actors, nor are they typically taking place solely within the formal political arena (McAdam et al., 2001: 7–8). The type of contentious politics that we will be investigating is done by an elected political actor, but happens outside of the traditional realm of state politics. This is not in contradiction with the theory as:

Contentious politics operates partly within (yet in tension with) official, prescribed politics; it depends on a degree of accommodation with the structure of domination, the deft use of prevailing cultural conventions, and an affirmation — sometimes sincere, sometimes strategic — of existing channels of inclusion (O’Brien, 2003: 54).

With reference to the existing channels of inclusion the academic field of contentious politics has not ignored the rise of new media. Bennett and Segerberg (2012: 749), affirm that ‘there is increasing coordination of action by organizations and individuals using digital media to create networks, structure activities, and communicate their views directly to the world’. A better understanding of the role of social media in contentious politics is ‘essential if we are to attain a critical perspective on some of the prominent forms of public engagement in the digital age’ (Bennett and Segerberg, 2012: 762).

## Politics and digital media

Social media, Twitter in particular, has become an important site for political debate (e.g. Schäfer, 2016) and practices of contentious politics (e.g. Vicari, 2013; Fábrega & Sajuria, 2014; Buettner & Buettner, 2016). Regular political debates on social media have until now mainly been researched in terms of communication strategies of political actors (e.g. Broersma and Graham, 2012; Graham et al., 2013) or in terms of the networks that exist between political actors such as politicians and journalists (e.g. Schäfer et al., 2012; Verweij, 2012). However, some authors have suggested that in the unstable conditions faced by many democracies, digital and social media, mixed with citizen discontent, offer the potential for individuals to quickly attain prominence. These leaders do so by creating hybrid repertoires, switching between social movement modes and conventional political modes of communication (e.g. Chadwick, 2017; Bennett et al., 2018).

Generally, political actors depend on the media for the spreading of their message and the shaping of their image (Sheafer, 2001). However, this relationship is dialectical, because the media cannot ignore what is newsworthy. The populist politicians who contests the status quo politics with radical and aggressive language and emotive rhetoric become newsworthy (Mazzoleni, 2003: 6–7). So, politicians who employ populist rhetoric and style can acquire a very high prominence, thereby exploiting political mediatization by adopting a language that suits the requirements of media (Taguieff, 2003). On the other hand, radicalism can also reduce media attention, since certain media brands do not want to be associated with political extremes. In addition, a display of outlandish behavior may threaten the public image of a political actor, losing ‘authoritativeness’ and thereby credibility as a knowledgeable, persuasive, and ultimately effective political force. A political outsider needs to balance generating media prominence with appearing authoritative (Bos & Van der Brug, 2010: 142–46). Therefore the populist politicians have to display a rationality in order to be perceived convincing to the electorate. The use of ‘objective’ knowledge like statistics is one way of achieving this.

Due to the rise of social media, the public sphere is highly fragmented. The public sphere is an ecosystem of interconnected spheres of public awareness, media platforms, audiences, and agendas, with niches inhabited by a range of professionalized political organizations (Mazzoleni & Schulz, 1999). What is now called the fourth age of political communication is characterized by the explosion of different digital media platforms, an overload of information, networked communication, and new hierarchies (Blumler, 2016). The ability to find one’s electorate in such a digital ecosystem, and create legitimacy through a combination of social and mass media, is the major challenge for political organizations (Aagaard, 2016).

More recently attention has been given to the role of technical and political factors interwoven in the design of platforms such as Facebook and Twitter and how their affordances shape possibilities for communication (see Van Dijck, 2013; Van Dijck et al., 2016). An important characteristic of social media platforms is that they provide political actors with the opportunity to bypass the gatekeepers of the traditional media and their professional norms and news values (Bracciale & Martella, 2017). In addition to the fact that politicians can now speak directly to ‘the people’, a longstanding wish of the populist, it seems that the affordances of social media, among them the absence of the gatekeepers and news value filters, make social media particularly suitable for populist politics. Social media tends to contribute to populism due to its preference for simplified messages and personalized frame, its fragmentation of information, and the ease of message dissemination (Engesser et al, 2016).

In a study of the news dispersion of political actors, De Winkel & Wieringa (forthcoming) discuss framing practices during the 2017 election for the Dutch House of Representatives. They draw conclusions regarding the power balance between media and politicians – and thereby the fourth democratic pillar – suggesting we should redirect our attention from filter bubbles (see Pariser, 2011) and echo chambers (see Wallsten, 2005) to framing. They note that in the new media ecology, power is differently distributed, with traditional media losing its control over the news. The media have to share their control of a message’s frame with other parties, such as politicians. The messages of the politicians show ‘high levels of framing […] of themselves, their message and their opponents’, media outlets and their content (De Winkel & Wieringa, forthcoming).

With an increasing percentage of our population getting their news from their social media feeds (see Hermida et al., 2012; Lenhart et al., 2010) mediated political practices – like framing – take on a different meaning. Conventional patterns of political engagement, interaction and information gatekeeping no longer apply (Coleman et al., 2016). As such, it is valuable to reassess what framing means in a contemporary media ecology.

## Framing theory

The topic of framing has been widely discussed in academia, involving scholars from a variety of backgrounds such as political science, conflict studies, sociology, media studies, journalism, and social psychology. A broad definition is given by communication scholar Jörg Matthes:

The key idea is that strategic actors, journalists, and audiences do not simply reflect or transport the political and social realities. In contrast, politics, issues, and events are subject to different patterns of selections and interpretations. These interpretations of issues are negotiated, contested, and modified over time. In light of this, frames are selective views on issues — views that construct reality in a certain way leading to different evaluations and recommendations (2012: 459).

Most research into framing has focussed on political frames in general (e.g. Hanggli & Kriesi, 2012; Slothuus & De Vreese, 2010) and (political) framing by the media (e.g. Wasike, 2013), yet there has been little attention to the currently widespread phenomenon of the framing of media by politicians (e.g. De Winkel & Wieringa, forthcoming). Reformatting the core concept of framing is necessary in the disrupted field of relationships between media, democratic institutions, and publics (Bennett & Pfetsch, 2018).

Framing is also defined as a mechanism within contentious politics. Snow and Benford (1988) have introduced and defined this process, and use the verb ‘framing’ to refer to the active process of meaning construction. Framing is an element of contentious politics, in the sense that it ‘involves the generation of interpretive frames that not only differ from existing ones but that may also challenge them’ (Benford & Snow, 2000: 614). It is the process of strategic issue position which can manifest in media input and could aim to shape voter behavior (Schemer, Wirth & Matthes, 2012).

The process of framing can be split into several different kinds of frames and stages. Diagnostic framing is the identification of a problem and the attribution of a problem. It focuses on blame and responsibility. This type of framing can be used to set up boundaries between what is good or evil, and between the in-group and the out-group (i.e. the ‘true’ Dutch as opposed to Dutch citizens with a migration background). The stage of diagnostic framing is sometimes also defined as the agenda setting stage (Entman, 2007). Prognostic framing is articulating a proposed solution to the problem. It answers the question of ‘what can be done to solve this problem?’. Motivational framing provides a rationale for engaging in ameliorative action. These types of frames stress the severity and urgency of the problem and aims to motivate the audience. In the case of politics, a call to action could be to go and vote for a specific party, which is of course within the scope of good democratic and political practice. However these three successive steps can also be used for a variety of undemocratic, toxic, and/or damaging actions including violence.

## Information bricolage

If we turn our attention back to the political sphere, it was found that Dutch politicians selectively pick media content and messages which support their standpoints (De Winkel & Wieringa, forthcoming; Wieringa, 2017). Social media allows its users to easily decontextualize articles from their original source and its target audience, meaning messages are easily picked and framed. We propose ‘information bricolage’ as a term to describe the combination of technologically enabled decontextualization, selectively-picking, and subsequent framing.

The term ‘bricoleur’ was coined by Claude Lévi-Strauss, who defined it as making:

[…] do with ‘whatever is at hand’, that is to say with a set of tools and materials which is always finite and is also heterogeneous because what it contains bears no relation to the current project, or indeed to any particular project, but is the contingent result of all the occasions there have been to renew or enrich the stock or to maintain it with the remains of previous constructions or destructions (1962: 17).

Thus, in our understanding an ‘information bricoleur’ is someone who uses and remixes existing pieces of content. While keeping the article itself intact, the collage of items can create a new or different meaning for the whole, and possibly emphasizing other particular aspects of its parts. Or, as Mark Deuze describes the term with respect to the digital context, it describes ‘the remixing, reconstructing, and reusing of separate artifacts, actions, ideas, signs, symbols, and styles in order to create new insights or meanings’ (2006: 70).

As such, bricolage is seen as a central principle of emerging forms of digital culture (Deuze, 2006). In the present case of Geert Wilders’ new media behavior, the term refers to the selection and framing process he engages in to create a ‘news feed’, which conveniently underlines his own views, in order to express those views to others. This includes bestowing blame and responsibility, proposing a solution to a problem, or providing a rationale for engaging in action. By appropriating items that are currently debated in a diverse set of media outlets, grouping them, and making them part of the same narrative, information bricolage makes sure the deployed narrative effectively resonates with the intended audience. The news that is selected is already part of the daily experience, narratives, and cultural resonance of a target audience. The impact of such a practice is heavily connected to the way power is distributed in and through a (new) media reality. Changes in such a landscape can elevate the platform of new or transgressive practices and actors, while subverting conventional ones.

## Geert Wilders’ communication strategy

What is so special about Geert Wilders? To begin with, the media attention on him often has more to do with his absence than his presence. Wilders makes very few appearances in mainstream televised debates, and when he does it is rarely in a setting where he can be critiqued: he generally avoids debates with opposition figures or interviews in which he can expect a critical interviewer. His traditional media performances are mainly limited to entertainment talk shows, showbiz or gossip outlets, and media who share his political views. This absence from political platforms that feature journalistic inquiry and political debate could hurt his authoritativeness and he seems to only make exceptions to his general media strategy when he is offered a lot of time, attention or validation, perhaps helping him to keep control of his ‘frames’.

Geert Wilders’ absence from the national political debate on the mainstream and conventional stages is further underscored by the general absence of debatable party positions: as the PVV has no extensive party program, election program, or annual budget proposal. Wilders’ pattern of avoiding the traditional mediatized democratic process can be interpreted as a distrust of traditional media, further underlined by his own statements (see Figure 1).

Figure 1

Tweet by Wilders featuring cartoon in which Donald Trump bypasses the traditional media by using social media to reach his fans.

In Figures 2 and 3 we see that both Geert Wilders and the PVV4 are most exemplary of the category of broadcaster. The PVV has the least replies of all political parties that have significant Twitter activity. Other striking examples of broadcasters are for instance Forum voor Democratie (a new rival populist far right party) and DENK (a new populist Islamic party). At the level of the individual account, we see that again the PVV candidates can generally be typified as broadcasters (i.e. a very low amount of their tweets are communicative replies). One marked exception to this rule is Dion Graus, who is the sole communicator among the party’s candidates.

Figure 2

Replies compared to total amount of tweets (party level).3

Figure 3

Replies compared to total amount of tweets (individual level – PVV highlighted).

Geert Wilders hardly ever seems to enter in dialogue with his constituency, neither through the mainstream media nor his Twitter feed. This behavior is in line with the practice of monological communication typical for (old) mass media but is not the default mode for new media. Wilders’ tweeting of the cartoon (Figure 1) was very apt, as the man in the chair is at the receiving end of the politician’s messages. In sum, we find that the context in which Wilders’ tweets are sent is one reminiscent of the ‘old’ media: a one-way broadcast, which allows for a heavily controlled frame of the message. In following section, we will discuss what kind of content Geert Wilders spreads using his Twitter account, specifically the content that has the appearance of objectivity or factuality, like statistics.

## Tweeting Numbers: The Appropriation of Statistics

We have qualitatively analyzed Wilders’ Twitter feed during the campaign period which started on 01 January 2017 up to and including 15 March 2017. As the entirety of tweets and their qualitative analyses cannot be discussed in the context of this paper, we will report on specific tweets with the aid of screenshots either made on 01 May 2017 or in the first two weeks of January 2018, which are illustrative of the general tendencies found in the politician’s communication regarding the use of statistics.

On 04 January 2017, Wilders tweeted a quote from an article by newspaper De Telegraaf (Figure 4).5 In this article, De Telegraaf refers to an interview in another newspaper, AD, with a sociologist named Ruud Koopmans.6 Ruud Koopmans says in this interview in AD: ‘Of Muslims of Turkish and Moroccan descent in The Netherlands, 45% must be perceived as fundamentalist’ (translation by the authors).7 Here a clear misconception can be seen in the transferal of Koopmans’ quote from AD to De Telegraaf. De Telegraaf did not include that Koopmans was talking about Muslims of Moroccan and Turkish descent, so the percentage mentioned is actually incorrect.

Figure 4

Tweet by Geert Wilders.

Another tweet written by Wilders, on 21 February 2017, linked to an article from local news agency RTV Rijnmond (Figure 5).9 The article is about research carried out by Erasmus University, asking 1,200 women about their experiences with intimidation. According to the article by RTV Rijnmond, the women were asked about the background of the offenders, who they stated to be mainly Moroccan (32%). However, in the research report by Erasmus University, there is no mention of this percentage.10 Quite the opposite: it is explicitly stated that the researchers did not ask the women to define the ethnic background of the offender, only to state whether the offender was from the same ethnic group as the woman herself or from a different group (Fisher & Sprado, 2017: 38). When the offender was from a different ethnic group, the researchers did not ask which group they belonged to. For some reason this percentage started to be circulated in media from all over the political spectrum.11

Figure 5

Tweet by Geert Wilders: “According to women, most offenders are Moroccans (32%). Almost all women in Rotterdam encounter intimidation on the streets”.8

Both these percentage-based tweets make claims about people with a Moroccan descent in the Netherlands. When looking at these tweets through the theoretical lens of framing theory, we can identify the three types of framing. Through diagnostic framing Wilders focuses on blame and identifies a problem. In the case of the first tweet, fundamentalism is presented as the problem. According to Wilders’ tweet, almost half of the Moroccans in the Netherlands are fundamentalist, thus contributing greatly to this problem. In the second tweet, Wilders states that most offenders are Moroccan and therefore they are the main cause of this problem. With prognostic framing he proposes a solution to the problems by ‘draining the swamp’ – a decidedly Trumpian reference12 – in the Netherlands after the national elections. The implication seems to be that Wilders will get rid of Moroccans when his party is elected. Within the same tweet Wilders calls for action (motivational framing), asking his audience to vote for his party. The framing of both tweets relies heavily on the numbers, as does rationalization of his rhetoric.

Another category of Wilders’ tweets deals with the appropriation of statistics (Figure 6). In this example Wilders draws upon the various Dutch statistical agencies such as Centraal Bureau voor de Statistiek (CBS) or the Central Bureau of Statistics Netherlands.

Figure 6

Tweet by Geert Wilders: ‘60,000 additional non-western immigrants in welfare since 2012. Thanks to Rutte the Netherlands functions as the ATM of many immigrants’.13

The website of CBS offers a system called Statline, which gives users the possibility to generate graphs and tables themselves. This also creates an opportunity to frame information in such a way that it fits one’s argument best. The tweet above shows a graph generated by Wilders (or his team) using the Statline tool. In it, Wilders makes use of the categories of ‘autochthone’, ‘Western allochtone’, and ‘non-Western allochtone’.14 Through everyday use this dichotomy that literally refers to people from the same and different soil – or in common parlance, their ‘migration background’ – has become ‘a racial discourse […] where “everyone” knows, and understands, tacitly, the unspoken text’ (Yanow and Van der Haar, 2013: 250). Importantly, this distinction obscures the fact that the both the autochtones and allochtones in question are Dutch citizens.

The CBS’ use of these categories is problematic and inconsistent.15 Officially, the division is due to ‘differences in socio-economic and cultural position’ (Keij, 2000: 24; authors’ translation). However, as noted by Kees Groenendijk (2007: 105), other factors also play an important role: the distinction ‘between Western and Non-Western allochthones is evidently based on political criteria, namely welfare level, geographical or cultural proximity of the country of origin and assumptions about the problematic character of the group’. With ‘the problematic character of the group’, Groenendijk is explaining the difference between the Netherlands’ former colonies, where people from Suriname, Aruba, and the Dutch Antilles (all considered non-Western) have been viewed as problem groups, while most people from Indonesia assimilated into Dutch society without any major problems.16

Here, these arbitrary categories are used to paint a very particular picture. One could also visualize the numbers of people receiving benefits from CBS as totals per year only. Thereby, one bypasses the terms of (non-)Western allochtone and autochthone to paint instead a picture of Dutch citizens receiving benefits. This prompts the reader to ask different questions. Instead of relating the amount of social welfare received to cultural background, one could think of economic reasons for the increase in welfare spending.

In his tweet (Figure 6) Wilders draws heavily on diagnostic framing; more specifically, the message attributes blame. As can be seen in the CBS graph featured in Wilders’ tweet, in the period between 2012 and 2016, there has been an increase in the number of people that have applied for and received social benefits from the state. The increased welfare spending can be seen as a problem by political actors and citizens. Wilders focuses on the number of non-Western immigrants in these statistics and attributes blame to this group of people for the rise in social benefits costs. This is framed negatively, painting the Netherlands as ‘the ATM’ for non-Western immigrants. With the phrase, Wilders is implying it is easy for these people to get money from the state, without working or doing anything in return. By attaching official statistics to the tweet and using it as underlying evidence for his claim, Wilders adds impact and credibility to his message.

In contrast to Geert Wilders his usage of statistics, there are other cases in which he actively denounces the calculations of statistical semi-governmental agencies (Figure 7). Here, Wilders refers to an opinion article in NRC by Ewald Engelen, financial geography columnist for NRC and professor at the University of Amsterdam’s Social and Behavioral Sciences faculty. Engelen’s article highlights the troubling nature of the Netherlands’ Bureau for Economic Policy Analysis (CPB), within the Dutch political landscape.17 The article explains that the programs on which political parties base their projections will never be executed in their entirety, as the Dutch political system is rooted in coalitions, meaning that all the parties have to make compromises, thus the projection in itself is considered meaningless by Engelen. Moreover, he writes that the projections are steering the parties in their campaigns, as they model their political program on what will create a favorable projection by the CPB. Additionally, he notes that the model the CPB uses does not allow for radical changes such as a big turn towards sustainability (as the Party for the Animals, PvdD, proposes), or withdrawing from the Euro (such as the PVV proposes).

Figure 7

Tweet by Geert Wilders: ‘Indeed, do not fall in the trap of the grey and false reality of the CPB. PVV-vision crystal clear: control of our own country, our own money + our own borders!’.18

While such concerns and reservations about the CPB’s methods and models are arguably valid, the way in which Wilders denounces the CPB is of interest here. Wilders reproduces and adds the catchphrase ‘don’t fall for it’ to his (citation)retweet of the NRC tweet to underscore his own point, while it is very debatable whether the initial user of the phrase Ewald Engelen – or the NRC – actually agree with his frame and message.

Wilders strategically makes use of official government statistics to make his messages more credible and legitimate. He uses them when they add to his claims, but he questions their validity when they are detrimental to his claims. In this way, governmental statistics become a tool in Wilders’ framing strategy. This is a pressing example of the selective picking of messages, and the harnessing of an information bricolage. Where in the previous example he incorrectly uses the statistics of the CBS to convey his message, he denounced another semi-governmental data crunching institute, the CPB, only twelve days earlier as ‘fake’. Messages are used and are declared as true when they subscribe to Wilders’ ideas, but messages of the same nature or similar sources are false, clueless, and traitorous when they do not. The same goes for the media (brand) that sends them.

## Rationalization of discrimination through the ‘objectivity’ of numbers

The presentation of numbers in order to create an air of objectivity is common practice. In his work on the pursuit of objectivity through quantification, Ted Porter states that an excess of objectivity ‘crushes individual subjects, demeans minority cultures, devalues artistic creativity and discredits genuine democratic political participation’ (Porter, 1996: 3). The absolutism Wilders offers in his tweets exists within a certain system and rationality that combines an assumed objectivity of statistics with a dataset in which a colonial history and unfair economic relations between countries has trickled down. This excess of objectivity is instrumental for Wilders in backing up his claims by relying on (misinterpreted) research reports and official statistical institutions. Framing societal issues in relation to the ethnic background of Dutch citizens, however, predates the statistical analysis and is not devoid of subjectivity, but instead can be considered as highly political. In addition to this the collection of data always depends on availability and priority, which have no inherent rational basis.

Oscar Gandy explains that ‘we can’t ignore the ways in which historical factors have led us to include race, and racial proxies in predictive and explanatory models even where their inclusion made little sense at all’ (2016: 63). Even for ‘objective’ statistics, political choices have been made with regard to the indicators, measuring scales, and definitions, although:

There is a tendency to think about rationality in terms of a continuum; one that moves from an idealized intelligence — a difference engine that engages in rapid computation, without errors in calculation, and more critically, without any systematic bias introduced by irrational emotional distractions. On the other end of the continuum we find the sometimes slow, sometimes fast, error prone, easily distracted, and routinely distorted information processing by humans (Gandy, 2016: 58).

The socio-economic statistics with regard to migrants lends itself to right-wing framing. With his ostensible rationality in using statistics, and in combination with his belligerent tone, Wilders comes across as rather ideological. This seems to be a deliberate strategy: he defines a vote against his party as something zealous, as ‘resistance’.19 Wilders draws freely from a pool of figures and selective statistics to underscore his own program. Moreover, by creating an absolutist image by using these numbers, he attempts to rationalize discrimination.

## Conclusion

Geert Wilders’ is a good example of a contentious politician, in that he is among a select group of politicians, predominantly populist and new right, that uses social media to broadcast to the electorate. By strategically limiting the access to him, both traditional media and the electorate are essentially forced to meet him on his social media, where he controls the frame and broadcasts solely his message. He is thereby in contention with conventional media institutions and conventional political parties. This is transgressive in terms of the typical practices of entering in a dialogue or debate with your opponents and the people you represent alike, but atypical for Wilders, who is not a non-traditional political actor, having been a member of the parliament for over twenty years.20 The problematic and therefore controversial nature of this political communication is particularly striking when combined with our concept of information bricolage. In this article we coin the term ‘information bricolage’ referring to a combination of decontextualization, selective picking of information, and framing which makes it possible for the politician to be the online gatekeeper of the media instead of the other way around.

De Winkel and Wieringa (forthcoming) warned about the potential dangers of the shifting balance of power between media and politics. They noted that the Fourth Estate’s traditional role has become more precarious now that the Dutch citizens are increasingly turning to Facebook and Twitter for their political news and opinions. Dutch citizens do not fear the ‘filter bubble’ (see Pariser, 2011), which is when everybody encounters only opinions they already share, meaning that the online media diet shows confirmation bias. Nor do De Winkel and Wieringa distinguish an ‘echo chamber’ (see Wallsten, 2005), where any specific online media simply echoes the particular mainstream media it supports. Competing political parties and conflicting political ideologies and ideas do meet online, in an abundance. Likewise, different constituencies meet in social media spaces. We do, however, subscribe to De Winkel and Wieringa’s (forthcoming) concerns on the distribution of power online, specifically when employed to the subversion of the fourth pillar of democracy. We add to this concern the absence of a debate embedded in a neutral or shared space, or in the presence of referee or authoritative entity that gets acknowledged as transcending the party politics. This would entail that Wilders leaves the controlled space of his own Twitter feed, interacting with other users on other digital spaces where he (potentially) has less control over the frame. He justifies the absence of political debate with the motivational frame that alludes to an anti-Wilders bias, where he identifies not only the other political parties as opponents, but also the mainstream press and other institutions that like the CBS, as either politically colored, out of touch with reality, or dishonest. This is where ‘stating the facts’ creates a paradox for Wilders. The air of objectivity is both criticized by pointing out the politics of statistical agencies, while at the same time using them to underline his opinions on migrants. In doing so Wilders legitimizes and rationalizes discrimination based on national origins.

## Discussion and future research

We recommend several strands of possible further research which we believe can be situated along four lines of inquiry. The first would be to pursue more singular case-studies. The second would be to compare political actors’ social media framing strategies in particular national contexts. The third would be to compare different types of political actors (i.e. ‘broadcaster’/’communicator’). The final strand we like to propose is to compare politicians, whose ideological convictions overlap, on an international level.

For future research we recommend to combine the insights of this paper with the insights of the above mentioned De Winkel and Wieringa (forthcoming) article along with Schäfer, Overheul, and Boeschoten’s (2012) insights on broadcasting politicians. Possible questions for research include: Are the online political media practices typical for Wilders, or are they comparable to the practices of other parties in national and international new right or left media sphere? Do they fall within a populist tradition, transcending the right-wing left-wing binary, or are they the transgressive element of the new right only?

## Notes

1. ‘Fourth Estate’ refers to the public press or news media. It appears as such in Thomas Carlyle’s 1841 book On Heroes and Hero Worship which encompasses a series of lectures Carlyle given the previous year. In this work he claims that Edmund Burke said ‘here were Three Estates in Parliament; but, in the Reporters’ Gallery yonder, there sat a Fourth Estate more important far than they all’ (1993: 41, emphasis in the original). [^]
2. According to DeFleur, journalists and editors are gatekeepers of the news because they allow a portion of it to be run, while other potential news stories are left out (2010). [^]
3. Here all candidates of the party plus the youth party accounts are taken together. Data was collected from 01 January 2017 up to and including 15 March 2017. [^]
4. In the case of Geert Wilder and the PVV, the person and the party are largely the same entity on social media. The party has no party account on Twitter: when navigating to the party account on the PVV’s website, one automatically gets redirected to Geert Wilders’ account. Similarly, if one wants to know more about ‘the party’, the PVV website redirects the user to Wilders’ personal website. In non-digital life Geert Wilders is, officially, the only member of his party. [^]
5. 2017 45% Marokkanen fundamentalistisch. De Telegraaf, 04 January. Available at: http://www.telegraaf.nl/binnenland/27362883/__45__Marokkanen_fundamentalistisch__.html?utm_source=t.co&utm_medium=referral&utm_campaign=twitterfeed&apw_campaign=aaa1fa005122e97e6203e2c79357da21 [Last accessed 26 May 2017]. [^]
6. 2017 Het Westen moet zich veel weerbaarder opstellen. AD, 04 January. Available at: http://www.ad.nl/nieuws/het-westen-moet-zich-veel-weerbaarder-opstellen~a6ac99b0/ [Last accessed 26 May 2017]. [^]
7. Original text: ‘Van de moslims van Turkse en Marokkaanse komaf in Nederland moet 45 procent als ‘fundamentalistisch’ worden beschouwd’. [^]
8. Translation by the authors. [^]
9. 2017 Bijna alle vrouwen in Rotterdam hebben last van straatintimidatie. RTV Rijnmond, 21 February. Available at: http://www.rijnmond.nl/nieuws/152015/Bijna-alle-vrouwen-in-Rotterdam-hebben-last-van-straatintimidatie [Last accessed 26 May 2017]. [^]
10. Fisher & Sprado, February 2017, Seksuele straatintimidatie in Rotterdam. Available at: https://www.rotterdam.nl/wonen-leven/seksuele-straatintimidatie/Onderzoek-EUR-straatintimidatie.pdf [Last accessed 16 October 2018]. [^]
11. The authors have not been able to trace the source of this misinformation. [^]
12. Alter, C and Beckwith, R T 2017 Draining the swamp?, 17 January. Available at: http://time.com/donald-trump-drain-swamp/ [last accessed 19 January 2018]. [^]
13. ‘60,000 additional non-western immigrants in welfare since 2012. Thanks to Rutte the Netherlands functions as the ATM of many immigrants’. The table is called: ‘People receiving welfare; personal characteristics’ and shows the number of people in welfare in 2012 and 2016, split up by the default categorization of autochtone, western immigrant, and non-western immigrant (translation by the authors). [^]
14. Until 2017, these terms were used in the Netherlands to refer to the migration background of Dutch citizens. Allochthone and autochthone linguistically link ‘ethnicity’ with birth soil in the form of nationality. The Greek word chtõn means earth and allo and the prefixes auto and allo refer to ‘same’ and ‘different’ respectively, creating the words autochthon and allochthon (Yanow and Van der Haar, 2013: 237). [^]
15. CBS determines characteristics by first looking at whether or not both the parents of a person have the Dutch nationality. If that is the case, their offspring is considered to have Dutch origins, even if they (the children) are not born in the Netherlands. Offspring of two parents with Dutch origins are considered autochthones. The origins of allochthones are in principle derived from their own country of birth. If their country of birth is the Netherlands, the country of birth of the mother is chosen as a person’s origin. If the mother’s country of birth is the Netherlands, then the country of birth of the father is used as origin for their children (Bovens et al., 2016: 23). Within the category of allochthones, there is a subdivision in Western allochthones and non-Western allochthones. People from Europe (excluding Turkey), North America, Oceania, Japan, and Indonesia are considered Western. People from Turkey, Africa, Latin America, Asia (excluding Japan), Suriname, the Netherlands Antilles, and Aruba are considered non-Western. [^]
16. The reason for this can be found in the different social and economic circumstances in which these groups settled in the Netherlands. Where people from Indonesia immigrated in the 1950s, a time of a big economic boom and with the plan to stay indefinitely, people from Suriname immigrated in the beginning of the 1980s in the middle of a recession (Bovenkerk, 1978: 13–14). This difference in situation has severely shaped the image Dutch society constructed about different groups and has resulted in the construction of a race-ethnic hierarchy. [^]
17. Engelen, E 2017 CPB rekent de politiek door: trap er niet in, 15 February. Available at: https://www.nrc.nl/nieuws/2017/02/15/cpb-rekent-de-politiek-door-trap-er-niet-in-6713016-a1546179 [last accessed 19 January 2018]. [^]
18. Translation by the authors. [^]
19. Language of war is a discourse often used as a tool in framing. By using words such as ‘war’, ‘betray’, ‘revolution’, and ‘defend’, Wilders stresses the importance and relevance of his claims. It supports the prognostic and motivational aspects of his framing strategy, in the sense that it gives more urgency to the solution Wilders proposes, namely ‘no more Islam’. It legitimizes drastic measures, because it intensifies the problem. It acts as a powerful incentive to the readership to do something. What the readers should do is not explained by Wilders in the tweet above. He does bring forward a call to act in the previous tweets about ‘de-islamizing’, however this is rather vague. These kinds of messages could lead to more discrimination against immigrants or even violence. [^]
20. He was a member of the centrist-right VVD (The People’s Party for Freedom and Democracy) from 1989 until 2004, before formed his own party. Bunnik, A 2012 The rise and fall of Geert Wilders?, 12 September. Available at: https://www.aljazeera.com/indepth/opinion/2012/09/201291192756398991.html [Last accessed 27 February 2018]. [^]

## Competing Interests

The authors have no competing interests to declare.

## References

Aagaard, P 2016 The fourth age of political communication. Nordicum Mediterraneum, 11(3). Available at: https://doaj.org/article/bc2cbd793f5f4f52856709c7978b62bc [Last accessed 18 October 2018].

Aalberg, T, Esser, F, Reinemann, C, Stromback, J and de Vreese, C H (Eds.) 2016 Populist political communication in Europe. Routledge: New York. DOI:  http://doi.org/10.4324/9781315623016

Benford, R D and Snow, D A 2000 Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1): 611–639. DOI:  http://doi.org/10.1146/annurev.soc.26.1.611

Bennett, W L and Pfetsch, B 2018 Rethinking political communication in a time of disrupted public spheres. Journal of Communication, 68(2): 243–253. DOI:  http://doi.org/10.1093/joc/jqx017

Bennett, W L and Segerberg, A 2012 The logic of connective action: Digital media and the personalization of contentious politics. Information, Communication & Society, 15(5): 739–768. DOI:  http://doi.org/10.1080/1369118X.2012.670661

Bennett, W L, Segerberg, A and Knüpfer, C B 2018 The democratic interface: Technology, political organization, and diverging patterns of electoral representation. Information, Communication & Society, 21(11): 1655–1680. DOI:  http://doi.org/10.1080/1369118X.2017.1348533

Blumler, J G 2016 The fourth age of political communication. Politiques de communication, 1(6): 19–30. Available at: https://www.cairn.info/revue-politiques-de-communication-2016-1-page-19.htm [Last accessed 18 October 2018].

Bos, L and Van der Brug, W 2010 Public images of leaders of anti-immigration parties: Perceptions of legitimacy and effectiveness. Party Politics, 16(6): 777–799. DOI:  http://doi.org/10.1177/1354068809346004

Bovenkerk, F 1978 Rasdiscriminatie in Nederland. In: Bovenkerk, F (Ed.), Omdat Zij Anders Zijn: Patronen van Rasdiscriminatie in Nederland, pp. 9–30. Amsterdam: Boom Koninklijke Uitgevers.

Bovens, M, Bokhorst, M, Jennissen, R and Engbersen, G 2016 Migratie en classificatie: Naar een meervoudig migratie-idioom. Den Haag: Wetenschappelijke Raad voor het Regeringsbeleid.

Bracciale, R and Martella, A 2017 Define the populist political communication style: The case of Italian political leaders on Twitter. Information, Communication & Society, 20(9): 1310–1329. DOI:  http://doi.org/10.1080/1369118X.2017.1328522

Broersma, M and Graham, T 2012 Social media as beat: Tweets as a news source during the 2010 British and Dutch elections. Journalism Practice, 6(3): 403–419. DOI:  http://doi.org/10.1080/17512786.2012.663626

Buettner, R and Buettner, K 2016 A systematic literature review of twitter research from a socio-political revolution perspective. In: 49th Hawaii International Conference on System Sciences, Kauai, H I (Ed.), pp. 2206–2215. on 5–8 January. DOI:  http://doi.org/10.1109/HICSS.2016.277

Carlyle, T 1993 On Heroes, Hero-Worship, and the Heroic in History. Berkeley & Los Angeles: University of California Press.

Chadwick, A 2017 The Hybrid Media System: Politics and Power. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/oso/9780190696726.001.0001

Coleman, S, Thumim, N, Birchall, C, Firmstone, J, Moss, G, Parry, K and Blumler, J G 2016 The Mediated City: The News in a Post-Industrial Context. London: Zed Books.

De Winkel, T and Wieringa, M 2019 ‘The Politician As Editor-in-Chief: News Distribution on Social Media by Candidates for the 2017 Election for the Dutch House of Representatives.’ Media and Communication. Under review.

DeFleur, M L 2010 Gatekeeping theory. In: Mass communication theories: Explaining origins, processes, and effects, pp. 148–158. New York: Allyn & Bacon.

Deuze, M 2006 ‘Participation, Remediation, Bricolage: Considering Principal Components of a Digital Culture’. The Information Society, 22(2): 63–75. DOI:  http://doi.org/10.1080/01972240600567170

Engesser, S, Ernst, N, Esser, F and Büchel, F 2016 Populism and social media: How politicians spread a fragmented ideology. Information, Communication & Society, 20(8): 1109–1126. DOI:  http://doi.org/10.1080/1369118X.2016.1207697

Entman, R M 2007 Framing bias: Media in the distribution of power. Journal of communication, 57(1): 163–173. DOI:  http://doi.org/10.1111/j.1460-2466.2006.00336.x

Fábrega, J and Sajuria, J 2014 The formation of political discourse within online networks: the case of the occupy movement. International Journal of Organisational Design and Engineering, 3(3). DOI:  http://doi.org/10.1504/IJODE.2014.065094

Gandy, O H 2016 Coming to Terms with Chance: Engaging Rational Discrimination and Cumulative Disadvantage. London: Routledge. DOI:  http://doi.org/10.4324/9781315572758

Graham, T, Broersma, M, Hazelhoff, K and Van ’T Haar, G 2013 Between broadcasting political messages and interacting with voters: The use of Twitter during the 2010 UK general election campaign. Information, Communication & Society, 16(5): 692–716. DOI:  http://doi.org/10.1080/1369118x.2013.785581

Groenendijk, K 2007 Allochtonen of burgers: Definitiemacht in debat en wetgeving over immigranten. In: Duyvendak, J W, Engbersen, G and Teeuwen, M (Eds.), Macht en verantwoordelijkheid: Essays voor Kees Schuyt, pp. 101–112. Amsterdam: Amsterdam University Press.

Hanggli, R and Kriesi, H 2012 Frame Construction and Frame Promotion (Strategic Framing Choices). American Behavioral Scientist, 56(3): 260–278. DOI:  http://doi.org/10.1177/0002764211426325

Hermida, A, Fletcher, F, Korell, D and Logan, D 2012 Share, Like, Recommend. Journalism Studies, 13(5–6): 815–824. DOI:  http://doi.org/10.1080/1461670X.2012.664430

Jagers, J and Walgrave, S 2007 Populism as political communication style: An empirical study of political parties’ discourse in Belgium. European Journal of Political Research, 46(3): 319–345. DOI:  http://doi.org/10.1111/j.1475-6765.2006.00690.x

Keij, I 2000 Standaarddefinitie Allochtonen. Index, pp. 24–25. November.

Kiesraad 2017 Official results of the elections to the House of Representatives on 15 March 2017, 21 March. Available at: https://english.kiesraad.nl/latest-news/news/2017/03/20/official-results-of-the-elections-to-the-house-of-representatives-on-15-march-2017 [Last accessed 02 February 2018].

Lenhart, A, Purcell, K, Smith, A and Zickuhr, K 2010 Social Media & Mobile Internet Use among Teens and Young Adults. Washington, DC: PEW Internet and American Life Project.

Lévi-Strauss, C 1962 The Savage Mind. Chicago: University of Chicago Press.

Matthes, J 2012 Framing politics: An integrative approach. American Behavioral Scientist, 56(3): 247–259. DOI:  http://doi.org/10.1177/0002764211426324

Mazzoleni, G 2003 The media and the growth of neo-populism in contemporary democracies. In: Mazzoleni, G, Stewart, J and Horsfield, B (Eds.), The media and neo-populism: A contemporary comparative analysis, pp. 1–20. Westport, CT: Praeger Publishers.

Mazzoleni, G and Schulz, W 1999 “Mediatization” of politics: A challenge for democracy? Political communication, 16(3): 247–261. DOI:  http://doi.org/10.1080/105846099198613

McAdam, D, Tarrow, S and Tilly, C 2001 Dynamics of Contention. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511805431

O’Brien, K 2003 Neither transgressive nor contained: Boundary-spanning contention in China. Mobilization: An International Quarterly, 8(1): 51–64. DOI:  http://doi.org/10.17813/maiq.8.1.6543n6m111722u42

Pariser, E 2011 The Filter Bubble: What the Internet is hiding from you. London: Penguin.

Porter, T M 1996 Trust in numbers: The pursuit of objectivity in science and public life. Princeton: Princeton University Press. DOI:  http://doi.org/10.1515/9781400821617

Schäfer, M T 2016 Unstable (Counter)Publics. Online Platforms as Hybrid Forums for Socio-Political Debate. In: Baxman, I, Beyes, T and Pias, C (Eds.), Social Media – New Masses, pp. 267–284. Chicago: Chicago University Press.

Schäfer, M T, Overheul, N and Boeschoten, T 2012 Politiek in 140 tekens: Een analyse van twitterende Nederlandse politici. In: Van ’t Hof, C, Timmer, J and van Est, R (Eds.), Voorgeprogrammeerd: Hoe Internet Ons Leven Leidt, pp. 1–17. Den Haag: Boom.

Schemer, C, Wirth, W and Matthes, J 2012 Value resonance and value framing effects on voting intentions in direct-democratic campaigns. American Behavioral Scientist, 56(3): 334–352. DOI:  http://doi.org/10.1177/0002764211426329

Sheafer, T 2001 Charismatic skill and media legitimacy: An actor-centered approach to understanding the political communication competition. Communication Research, 28(6): 711–736. DOI:  http://doi.org/10.1177/009365001028006001

Slothuus, R and De Vreese, C H 2010 Political parties, motivated reasoning, and issue framing effects. The Journal of Politics, 72(3): 630–645. DOI:  http://doi.org/10.1017/S002238161000006X

Snow, D A and Benford, R D 1988 Ideology, Frame Resonance and Movement Participation. International Social Movement Research, 1(1): 197–217.

Taguieff, P A 2003 L’illusione populista. Milan: B. Mondadori.

Tarrow, S G 2011 Power in movement: Social movements and contentious politics. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511973529

Tilly, C 2004 Contention and democracy in Europe, 1650–2000. Cambridge: Cambridge University Press.

Tilly, C and Tarrow, S 2015 Contentious Politics. Oxford: Oxford University Press.

Van Dijck, J 2013 The Culture of Connectivity: A Critical History of Social Media. Oxford: Oxford University Press. DOI:  http://doi.org/10.1093/acprof:oso/9780199970773.001.0001

Van Dijck, J, Poell, T and De Waal, M 2016 De Platformsamenleving: Strijd om Publieke Waarden in een Online Wereld. Amsterdam: Amsterdam University Press. DOI:  http://doi.org/10.5117/9789462984615

Verweij, P 2012 Twitter links between politicians and journalists. Journalism Practice, 6(5–6): 680–691. DOI:  http://doi.org/10.1080/17512786.2012.667272

Vicari, S 2013 Public reasoning around social contention: A case study of Twitter use in the Italian mobilization for global change. Current Sociology, 61(4): 474–490. DOI:  http://doi.org/10.1177/0011392113479747

Wallsten, K 2005 Political blogs and the bloggers who blog them: Is the political blogosphere and echo chamber. In: American Political Science Association’s Annual Meeting, pp. 1–4. Washington, DC, 1 September.

Wasike, B S 2013 Framing news in 140 characters: How social media editors frame the news and interact with audiences via Twitter. Global Media Journal: Canadian Edition, 6(1): 5–23.

Wieringa, M 2017 ‘F’ is for fake news/filter bubbles/framing: How Politicians Frame Media (Content) on Facebook & Twitter during the Dutch Tweede Kamer Elections of 2017. Unpublished thesis (PhD) Utrecht University.

Wieringa, M, de Winkel, T and Lewis, C 2017 Wie is de waakhond op sociale media? Onderzoek naar de rol van traditionele media en politici op sociale media in verkiezingstijd, 20 April. Available at: https://www.burgemeesters.nl/content/rapport-wie-de-waakhond-op-sociale-media#.W8mB9U1lKM9 [Last accessed 18 October 2018].

Wilders, G 2016 Wilders’ Plan: Time for Liberation, 8 November. Available at: https://geertwilders.nl/index.php/94-english/2015-wilders-plan-time-for-liberation [Last accessed 01 May 2017].

Yanow, D and Van der Haar, M 2013 People out of place: Allochthony and autochthony in the Netherlands’ identity discourse—metaphors and categories in action. Journal of International Relations and Development, 16(2): 227–261. DOI:  http://doi.org/10.1057/jird.2012.13