"Euro crisis and Europe, the trending topics of 2013", Politicisation of Europe in Tweets of Dutch Politicians during Cabinet Rutte II

MARIJE TIMMERMANS

12435430

DHR. DR. MATTHIJS LOK

MW. DR. YOLANDA RODRÍGUEZ PÉREZ

“EURO CRISIS AND EUROPE,

THE TRENDING TOPICS OF

2013”

POLITICISATION OF EUROPE IN TWEETS OF

DUTCH POLITICIANS DURING CABINET RUTTE

II

MA THESIS IN EUROPEAN STUDIES

GRADUATE SCHOOL FOR HUMANITIES

UNIVERSITEIT VAN AMSTERDAM

Table of contents

1 Introduction... 3

2 Theoretical framework...6

2.1 The European public sphere(s) and politicisation...6

2.2 Public sphere and ‘Web 2.0’...11

2.3 Politicians on Twitter...14

2.4 Reciprocity between citizens and politicians...16

3 Method: digital humanities and political thought...18

3.1 Analysing social media data...18

3.2 Selecting useful Twitter data...19

3.3 For every question the right digital method...21

3.4 Studying political language...22

3.4.1 Political thought...22

3.4.2 Framing analysis...23

3.4.3 Useful frames...24

4 First look at the data... 26

4.1 Europeanisation of tweets?...27

4.2 When do politicians tweet about Europe?...28

5 What words do politicians use when they tweet about Europe?...31

5.1 VVD... 31

5.2 PvdA... 32

5.3 Oppositional parties...32

5.4 Politicisation of Europe in the use of words...35

6 Peak 1: November 2012 – June 2013...36

6.1 Words used during peak 1...37

6.2 Sentiments used during peak 1...38

6.3 Frames used during peak 1...40

6.4 Politicisation during peak 1...41

7 Peak 2: October 2013 – May 2014...44

7.1 Words used during peak 2...45

7.2 Sentiments used during peak 2...47

7.3 Frames used during peak 2...48

7.4 Politicisation during peak 2...50

8.1 Words used during peak 3...53

8.2 Sentiments used during peak 3...55

8.3 Frames used during peak 3...56

8.4 Politicisation during peak 3...58

9 Conclusion... 59

9.1 Further research...62

10 Bibliography... 63

11 Appendix... 0

11.1 List of tweets from politicians...1

1 Introduction

After the results of the European elections, scholars and newspapers declared

that there were signs of a transformation in European Union politics. The

number of parties that joined the European Parliament elections was higher

than ever in most countries and the percentage of citizens that voted was almost

51%, the highest participation rate since 1994. This shows that there is more

interest in European politics.

_{Moreover, the results of the elections show that}

votes for the ruling parties, the European People’s party (EPP) and the Socialists

& Democrats (S&D), declined. This means that the European Parliament is

going to become more fragmented with different parties and opinions, and thus

more polarised.

However, it most likely that the politicisation of European Union issues

already happened before this years’ elections. This thesis will look at the

politicisation of issues on a European level during the governing period of the

Rutte II cabinet.

_{This coalition was composed of the conservative liberal party}

(Volkspartij voor Vrijheid en Democractie; hereafter VVD) and the labour party

(Partij van de Arbeid; hereafter PvdA) and governed between 5 November 2012

and 26 October 2017. In the period that VVD and PvdA were in the government,

three important events happened that are related to the European Union. First,

the last waves of the euro crisis that started in 2009 hit the Netherlands.

Secondly, the ongoing war in Syria led to many refugees at the borders of the

European Union. Thirdly, on 6 April 2016 the

Ukraine - European Union

Association Agreement referendum

was held. These challenges dealt in some

way or the other with the European Union and stirred a lot of debate in the

Netherlands about the functioning of the European Union as well.

1

Lydie Cabane, ‘European Elections: Does Politicisation Really Matter?’, Leiden Safety

and Security Blog, published 28 May 2019,

https://www.leidensafetyandsecurityblog.nl/articles/european-elections-does-politicisation-really-matter.

2

Jon Henley, ‘A Fractured European Parliament May Be Just What the EU Needs’, The

Guardian, 26 May 2019, sec. Politics,

https://www.theguardian.com/politics/2019/may/26/a-fractured-european-parliament-may-be-just-what-the-eu-needs.

3

See for other forms of European politicisation Pieter De Wilde, ‘No Polity for Old Politics?

A Framework for Analyzing the Politicization of European Integration’, Journal of European

Integration 33, no. 5 (September 2011): 560,

While the challenges faced by the Rutte II cabinet were related to the

European Union, they were hotly debated in national politics as well. This means

that these issues got lot of media attention. This issue salience and visibility in

the media is one dimension of politicisation. A second dimension that is needed

for politicisation is an increasing number of actors that are speaking and

debating about European issues. A third dimension is the polarisation of this

debate, meaning that the political actors have different positions in this debate.

While the first and the second are a surety in this period, because the issues

related to European politics were debated both in Dutch parliament and in the

media, the third dimension is harder to put a finger on, as that is a matter of

how politicians and the media communicate about these issues.

To see if politicisation of European issues already took place in during the

Rutte II cabinet, the following research question is designed: “To what extent

were tweets concerning the European Union or issues on a European level from

Dutch politicians during the Rutte II cabinet politicised?” This means that this

research will analyse tweets from the fourteen most popular Dutch politicians on

Twitter during this period. In doing so, three factors will be studied: what kind

of words politicians use, which sentiments about European issues or European

integration are captured in a tweet, and which frames are used to express these

sentiments about Europe.

Because issue salience is necessary, this thesis will first study when

politicians tweeted about Europe. Together with an overview of the most

frequently used words, this will also shine a light on which issues politicians do

not tweet about. After establishing peaks in which European issues are more

present than normal, these peaks will function as case studies to see which

words, sentiments and frames are being used by politicians when they tweet

about Europe.

In this research, the social media platform Twitter has been chosen as the

channel of communication between politicians and the electorate. Since the

emergence of social network sites, Twitter is the most widely used platform by

politicians for informing the public about policies, campaigning and even having

conversations with citizens.

_{Moreover, Dutch politicians and the Dutch}

4

Swen Hutter, Edgar Grande, and Hanspeter Kriesi, Politicising Europe: Integration and

Mass Politics (Cambridge, 2016), 8, http://dx.doi.org/10.1017/CBO9781316422991.

5

Paul Nulty et al., ‘Social Media and Political Communication in the 2014 Elections to the

European Parliament’, Electoral Studies 44 (1 December 2016): 429,

population were early adopters of the Twitter communication service.

_{It is a}

frequently used medium in the Netherlands, 26% of the Dutch population 15

years old or older having used this medium in 2013.

In addition, the relatively new communication platform Twitter offers us a

new way to study political discourse and communication strategies. In so-called

tweets of 140 characters (since November 2017 the limit is raised to 280

characters), politicians are required to compose compact messages, in which

they can use hashtags to link the tweet to a specific topic and make use of

@mentions to mention another user in their tweet. Moreover, they can retweet

messages from other users to share a tweet.

The chosen politicians have more followers on Twitter than the official

accounts of their parties, so their tweets have a larger scope of impact than

tweets of party accounts. National politicians also have more visibility than their

colleagues of the European parliament and a bigger influence in shaping the

public opinion.

_{Since Twitter is also medium on which politicians share}

personalised content, it raises the question how they communicate about Europe

on Twitter and to what extent this communication is politicised or not.

6

Maurice Vergeer, ‘Twitter and Political Campaigning’, Sociology Compass 9, no. 9

(2015): 746, https://doi.org/10.1111/soc4.12294.

7

Tjaša Redek and Uroš Godnov, ‘Twitter as a Political Tool in EU Countries during the

Economic Crisis, a Comparative Text-Mining Analysis’, Društvena Istraživanja: Časopis Za

Opća Društvena Pitanja 27, no. 4 (2018): 694.

8

Ruud Koopmans and Paul Statham, eds., The Making of a European Public Sphere:

Media Discourse and Political Contention, Communication, Society and Politics (New York:

Cambridge University Press, 2010), 65.

2 Theoretical framework

This research will combine theory of the public sphere and more specifically the

European public spheres with Twitter and communication by politicians. In this

chapter, literature about these topics will be reviewed. Because the topic is

extensive, the chapter is divided in four themes. The first paragraph will clarify

the notions of the public sphere and politicisation in relation to Europe and how

these notions are intertwined. Second, literature on how social media or Web

2.0 has changed the public sphere will be reviewed. Third, what is known about

how politicians communicate on Twitter will be outlined. Fourth, literature on

reciprocity between citizens and politicians is discussed and more important,

how politicians on social media broaden the public sphere is considered.

2.1 The European public sphere(s) and politicisation

Since the very influential work of Jürgen Habermas from the 1960’s and its

translation in English in the 1990s, the public sphere is a prominent subject of

study. According to Habermas, the creation of a civil society network and the

creation of newspapers provided the means for a reason-based, public opinion.

In contrast to before the eighteenth century, the public sphere was in principle

open for all to discuss public issues and was protected from church and state

powers. Written in the 1960s, Habermas continued his theory to the twentieth

century, where mass-media appeared in the forms of daily newspapers, radio

and television and became influential opinion-leading institutions.

_{As national}

democracies become more intertwined with European level politics, the

questions arise whether the public sphere could also emerge, or is already in

9

Jürgen Habermas, The Structural Transformation of the Public Sphere: An Inquiry into a

Category of Bourgeois Society, trans. Thomas Burger, 10. print, Studies in Contemporary

German Social Thought (Cambridge, Mass: MIT Press, 1999).

existence, on the European level and second, if a European public sphere is

needed for a functioning European Union.

According to the book by Ruud Koopmans and Paul Statham, the

European Union needs a public sphere for European integration for a few

reasons. First, because of the democratic values that are part of the EU. The

public sphere is a component of the democratic ideal of the European union.

Moreover, the decisions reached in European politics and the way the decisions

are developed should be transparent in order to increase citizens’ awareness of

European affairs.

_{Second, it also has an instrumental value, meaning that the}

European public sphere is a legitimizing force for European integration and

shows not only its democratic values but can also communicate the economic or

geopolitical benefits of belonging to the EU family.

A third way that the European public sphere is important for European

integration is that public communication creates a ‘space’ of relationships

among citizens. Participating in a public sphere with other citizens, allows for

the forming of a ‘social imaginary’ that gives cultural form to integration, and in

line with the idea of Anderson’s imagined communities

_{, it enables citizens to}

imagine the European community. As both Jürgen Habermas and Craig Calhoun

argue, a European public sphere can help establish a thicker solidarity between

citizens of Europe that can make the European project a successful union.

Moreover, politicisation of public debates does the same when dialogues and

contestations among citizens of different European countries about different

European-level topics occur.

Another debate is if a European public sphere is needed for the

functioning of the European Union. In other words, is identification as a

European citizen needed before the establishment of the European public

sphere, or does a European identity follows from a European public sphere? As

10

See also Patrick Bijsmans, ‘The Commission, the Politics of Information, and the

European Public Sphere’, in The Politics of Information: The Case of the European Union,

ed. Tannelie Blom and Sophie Vanhoonacker, European Administrative Governance Series

(London: Palgrave Macmillan UK, 2014), 181, https://doi.org/10.1057/9781137325419_12.

11

Koopmans and Statham, The Making of a European Public Sphere, 28.

12

Benedict R. O’G Anderson, Imagined Communities: Reflections on the Origin and

Spread of Nationalism, Rev. and extended ed (London ; New York: Verso, 1991).

13

Craig Calhoun, ‘The Democratic Integration of Europe’, in Europe without Borders:

Remapping Territory, Citizenship and Identity in a Transnational Age, ed. Mabel Berezin

and Martin Schain (Balitmore, 2003), 248; Jürgen Habermas, The Inclusion of the Other:

Studies in Political Theory, Studies in Contemporary German Social Thought (Cambridge,

Mass: MIT PRess, 1998), 160.

14

Mauro Barisione and Asimina Michailidou, eds., Social Media and European Politics:

Rethinking Power and Legitimacy in the Digital Era, Palgrave Studies in European Political

Sociology (London, United Kingdom: Palgrave Macmillan, 2017), 4.

mentioned above, Habermas and Calhoun emphasize the importance of

communication in Europe for European integration. Koopmans and Statham also

see a vibrant European public sphere as a requirement for European

integration.

_{Thomas Risse stresses that Europeanisation and politicisation of}

European public sphere leads to more identification levels with the EU

community.

_{His argument is the following: first, he sees that communication}

in national and issue-specific public spheres are more and more about European

issues; they are Europeanised. Second, European issues are being more and

more politicized both on the EU and the national level. He defines politicisation

as ‘an increase in polarization of opinions, interests, or values and the extent to

which they are publicly advanced towards the process of policy formulation

within the EU.’

In other words, a European public sphere requires not always EU-positive

voices. The opposite is true: politicisation means that a plurality of voices is

being heard, also the more extreme ones. Politicisation in the form of positive

and negative reporting about the EU both increase political identification levels

with the EU, if the issue is discussed as a matter of concern for European

citizens instead of one nation against the other. For example, during the

economic crisis, the identification levels with the EU actually increased. People

that only identified themselves with their nation-state decreased between 2005

and 2013, and people that identified themselves both nationally and European

increased.

_{Moreover, the politicisation of the euro crisis is not principally}

driven by power transfer and conflicts in integrating new member states, but is

about how to redistribute wealth within and across member states. So, the euro

crisis not only led to higher levels of politicisation, it also changed the nature of

politicisation.

This thesis does not see politicisation with its critical voices as a threat to

European integration. On the contrary, politicisation could work for all different

kinds of opinions of European integration. It could also give supporters of

European integration spaces to speak in favour of the European Union. In

addition, critical voices among politicisation in the European public sphere could

lead to changes within the EU, which is something else than leaving the

15

Koopmans and Statham, The Making of a European Public Sphere.

16

Thomas Risse, ed., European Public Spheres: Politics Is Back, Contemporary European

Politics (Cambridge ; New York: Cambridge University Press, 2015).

17

Risse, European Public Spheres, 13.

18

Risse, European Public Spheres, 156–57.

European Union all together.

_{Risse argues that if issues are framed as a}

common European issue, than politicization can lead to useful transnational

communication that can benefit the European democracy, but when issues are

seen as a burden for the national community, caused by the European Union, it

can bring about re-nationalisation of public spheres.

Francisco Pérez argues, on the other hand, that scholars of the European

public sphere focus to much on the media themselves instead of the reason for a

communications gap in Europe. In his book

Political Communication in Europe

,

he writes that the European public sphere dysfunctions because of the lack of

domestication and politicisation in European politics. In other words, first, the

EU is seen as a distant institution and lacks the “we-feeling” and second, its

politics is based on bureaucrats rather than politicisation, which he sees as

political clashes that are familiar to national politics.

_{In short, Pérez’s point is}

that there is need for European demos that could become political engaged,

before a functioning European public sphere can develop.

For the concept of politicisation, Pérez uses the politics theory of Carl

Schmitt. Carl Schmitt argued that liberalism is always in search for consensus

and is based on technocracy and made politics a political. Instead of focusing

conflict and antagonism, liberal politics focuses on economics and ethics.

_Pérez

follows Mouffes agonistic model of democracy, in which the friend-enemy

relation of Schmitt’s antagonism is transformed into ideological contestation in

party politics.

_{Pérez agrees with Schmitt’s critiques of liberalism, and}

according to him the EU is too rationalist, technocratic and apolitical. According

to Pérez’s the notion of politicisation, EU is barely politicised because its

agnosticism is not based on ideological left-right divisions, but it’s a dispute

between national sovereignty and Brussels.

This research would define politicisation on the basis of ideological

left-right divisions per se. It embodies all forms of political conflict on European

integration in public debates, related to the question of whether or not (further)

integration should take place and also debates on what kind of European Union

20

Hutter, Grande, and Kriesi, 5.

21

Thomas Risse, ‘No Demos? Identities and Public Spheres in the Euro Crisis’, Journal of

Common Market Studies 52, no. 6 (2014): 1213.

22

Francisco Pérez, Political Communication in Europe. The Cultural and Structural Limits

of the European Public Sphere (Basingstoke: Palgrave Macmillan, 2013).

23

Carl Schmitt, The Concept of the Political, trans. Georg Schwab, Expanded ed

(Chicago: University of Chicago Press, 2007).

24

Chantal Mouffe, The Democratic Paradox (London and New York, 2000).

25

Pérez, Political Communication in Europe. The Cultural and Structural Limits of the

European Public Sphere, 28.

is wanted.

_{European decision making is not limited to the European elites}

anymore and has been brought to the arena of mass politics.

Following Edgar Grande and Swen Hutter’s definition, politicisation

should be studied in the dimensions of visibility of issues in the public sphere,

the actors involved in speaking about these issues, and the polarisation of

opinions, interests and values among actors.

_{Another important dimension of}

the pollicisation is the important role of the media.

_{Debates about European}

issues do not have to be classified in the traditional left/right scope of

socioeconomic issues. Political conflicts could also arise on national sovereignty,

national or European identity and problems with solidarity and cooperation

among nations.

Pérez sees the European public sphere as one united public sphere, in the

form of a national public sphere. Most scholars abandoned the idea of one

European public sphere in the meaning of one transnational media platform.

Risse argues that there is no such thing as “the European public sphere”. There

are multiple European public spheres that overlap and interact and must be

seen as continuing communication flows within Europe.

_{According to him,}

three dimensions of European public sphere can be identified. ‘First, European

and EU issues, policies and actors are sufficiently visible in the various public

spheres. Second, fellow Europeans are present in the various national and

issue-specific public spheres (as both speakers and audiences). Third, common

European themes and issues are addressed using similar frames of references or

making claims across borders.’

Although the title of their book mentions “a European public sphere”

rather than multiple public spheres, Koopmans and Statham have the same

understanding of the European public sphere. They stress that national circuits

of communication within the national space should open themselves up to one

26

Eric Miklin, ‘From “Sleeping Giant” to Left-Right Politicization? National Party

Competition on the EU and the Euro Crisis’, JCMS: Journal of Common Market Studies 52,

no. 6 (November 2014): 1200, https://doi.org/10.1111/jcms.12188.

27

Liesbeth Hooghe and Gary Marks, ‘A Postfunctionalist Theory of European Integration:

From Permissive Consensus to Constraining Dissensus.’, British Journal of Political

Science, 39, 1 (2009): 13.

28

Swen Hutter, Edgar Grande en Hanspeter Kriesi, Politicising Europe: integration and

mass politics (zp 2016) 8.

29

Barisione and Michailidou, Social Media and European Politics, 4; Koopmans and

Statham, The Making of a European Public Sphere, 4.

30

Hutter, Grande, and Kriesi, Politicising Europe, 10.

31

Risse, European Public Spheres, 31.

another while remaining intact.

_{According to them, a European public sphere}

should contain three aspects. First, visibility of European decision making to

citizens, second, it should be accessible to the public, and third, it should have

the same politicisation as national public spheres.

This thesis will adopt the notion of the multiple European public spheres

and will focus on the Dutch Europeanised sphere specifically. This idea of

multiple European public spheres also emphasizes the idea of the social

imaginary of Europe that the Europeanisation of the national publics construct.

By allowing European level-issues into the national public sphere, it will help

citizens to imagine a European community. Therefore, it is not needed to have a

European demos already functioning as such, as Pèrez argues.

The question of multiple public spheres is also relevant for the scholars of

the public sphere and the internet. Since the popularization of the internet, the

collective nature of mass-media has shifted to a more individualised model, in

line with modern society. According to social and political scientists Mauro

Barisione and Asimina Michailidou, the public sphere on the internet, social

media and mobile communication functions more as ‘networked individualism’,

where individuals participate loosely in diverse personal networks.

_{As Patrick}

Bijsmans argues, the all-encompassing space of the public sphere was always

more an idealist concept than reality, but since the internet, the divisions in the

public sphere have become more visible.

_{This is why Axel Bruns and Tim}

Highfield suggest to ‘embrace a more complex, dynamic, and multifaceted

model that allows for connections and overlaps between a multitude of

coexisting public spheres’ in their work.

_{So the idea of one public sphere is}

giving away to a multiplicity of public spheres, both in studies on the impact of

the internet on the public sphere and globalisation and European integration.

2.2 Public sphere and ‘Web 2.0’

In studying the public sphere on the internet, the following question is relevant:

does the internet allow more democratization? The ‘Web 2.0’, as social media

33

Koopmans and Statham, The Making of a European Public Sphere, 5.

34

Barisione and Michailidou, Social Media and European Politics, 11.

35

Bijsmans, ‘The Commission, the Politics of Information, and the European Public

Sphere’.

36

Axel Bruns and Tim Highfield, ‘Is Habermas on Twitter?’, in The Routledge Companion

to Social Media and Politics, ed. Axel Bruns et al., 1st ed. (New York, NY: Routledge,

2016.: Routledge, 2015), 59,

https://www.taylorfrancis.com/books/9781317506560/chapters/10.4324/9781315716299-5.

websites are also called, allow for a more participated user experience that is

focused on collaboration and interaction. Initially, both scholars and the media

businesses themselves were optimistic about the interactive aspect of new

media. For example, the Web 2.0 promised grassroots journalism, a broad range

of information content that could be easily created by the media users

themselves and internet platforms dedicated to political participation and

political issues, functioning both as a space for opinion expression and resource

for citizen mobilization.

_{In other words, the Web 2.0 was seen as an ideal}

public sphere, where every media user could equally share content.

This view is also shared by Barisione and Michailidou. They stress the

ongoing structural transformation of the public sphere. In their book

Social

Media and European Politics

they emphasize the significance of ‘digital social

networks of communication both as platforms that enable citizens, organised

civil society and social movements to exert ‘counter-power ’ in the network

society and as organising agents for ‘digitally networked action’.

_{The Web 2.0}

offers more democratization in the form of a platform for power for normal

citizens and a platform for participation in civil society.

On the other hand, this utopian vision of the web received critique. The

Web 2.0 should not be seen as a source of new movements that are willing to

participate, but rather as a factor that gave these movements that already

existed new visibility. Moreover, the democratisation force of grassroots

journalism has been criticized, since grassroots journalism on the internet is a

specific form of amateur production that does not meet standards and thus

cannot replace professional journalism.

Christian Fuchs adds to this that the business side of the Web 2.0 should

also be considered when studying the democratic benefits of social media. In the

end, the media business promotes the democratisation force of social media to

make money by connecting user data with advertisement.

_{However, consumer}

targeting has always been part of the business models for traditional mass

37

Barisione and Michailidou, Social Media and European Politics, 8; Alexandra Bardan,

‘Insights into the Dialogic Communication on the “Debating Europe” Internet Channels’,

Styles of Communication 9, no. 1 (2017): 57.

38

Barisione and Michailidou, Social Media and European Politics, 4.

39

Bardan, ‘Insights into the Dialogic Communication on the “Debating Europe” Internet

Channels’, 58.

40

Christian Fuchs, ‘Digital Prosumption Labour on Social Media in the Context of the

Capitalist Regime of Time’, Time & Society 23, no. 1 (2013): 97–123.

media as well, although data analytics offers more precise and tailor-made

advertisement for Web 2.0 media consumers.

Moreover, media scientist José van Dijck argues that social media network

sites are not mere facilitating social connections. Instead, the algorithms of Web

2.0 play an important role in shaping what users see and how they interact.

For example, Twitter is programmed to organize search results and timelines,

making more popular users with more popular tweets appear at the top. This

means that not all users on Twitter are equal, and that Twitter is not a neutral

platform.

Furthermore, political scientist W Lance Bennet argues that social

fragmentation and decline of group loyalties make politics more and more

focused on the personal and individual.

_{Rasmussen argues that the}

development of Web 2.0 democratised the public sphere in such a way that it

became too diverse to provide legitimacy. In other words, the different

communicative platforms on the internet offers less guidance for politics and is

more focused on personal expression and individuality. Moreover, the diversity

of platforms combined with the anonymity and the informal styles of

communication, can lead to ‘polarisation of debates, isolation of issue-based

groups, unequal participation, lack of responsiveness and respect in debates and

uncivil and hateful speech.’

_{In other words: not a place for civil and}

rational-based communication of public issues. Furthermore, some scholars argue that

social media is more focused on the popular voices and opinions than on actual

dialogue. Again, this phenomenon is not new. In traditional mass media, a small

number of media personalities appear on tv or in the newspapers, and only a

small number of opinions are covered.

The Web 2.0 has changed the way we communicate in the twenty-first

century fundamentally. Since the traditional idea of the public sphere

transformed into overlapping and coexisting public spheres, it is self-evident

41

José van Dijck, Thomas Poell, and Martijn de Waal, The Platform Society (New York:

Oxford University Press, 2018), 11.

42

José van Dijck, The Culture of Connectivity: A Critical History of Social Media (Oxford ;

New York: Oxford University Press, 2013), 6.

43

Dijck, 75.

44

W. Lance Bennett, ‘The Personalization of Politics: Political Identity, Social Media, and

Changing Patterns of Participation’, The ANNALS of the American Academy of Political and

Social Science 644, no. 1 (1 November 2012): 20–39,

https://doi.org/10.1177/0002716212451428.

45

Terje Rasmussen, ‘Internet-Based Media, Europe and the Political Public Sphere’,

Media, Culture & Society 35, no. 1 (2013): 99.

46

Thomas Poell and José van Dijck, ‘Understanding Social Media Logic’, Media and

Communication 1, no. 1 (2013): 6.

that more platforms and publics emerge, based on personal preferences, where

political issues are debated. This could also be a good thing, as Fraser argued,

since the concept the public sphere presupposes social equality, but in reality,

opinions from minorities are often not heard in public debate. For them, the

multiple public spheres that are existing on the Web 2.0 could offer a solution.

While democratization of the public sphere by the Web 2.0 may have the

disadvantage of coming with uncivility and hate speech, it nevertheless is a step

closer the ideal of a participatory parity of the public sphere.

The upsides and the downsides of the Web 2.0 spark another debate, a

debate on how new these particular praises and critiques are. As José van Dijck

and Thomas Poell have argued in their research on social media logic, social

media and its norms, strategies, mechanisms and economies has changed the

conditions and rules of social interaction. The social media logic is compared to

mass media logic, defined by David Altheide and Robert Snow in the late 1970s.

Important elements of the mass media logic is that media presents the world as

a natural stream of events. However, this stream is not natural since media are

renewing topics so the audience keeps being interested. Secondly, media

presents itself as neutral, instead of information filters through which some

people get more covering than others. This neutrality is obtained by neutral

reporters and voicing different opinions to give a representative programme.

Moreover, Van Dijck and Poell also emphasize that since cable television in the

1980s programmes and television channels appeared for niche audiences and

culture became more and more commercialised, meaning that information and

opinions from the media were more infused with advertisement.

_{In other}

words, critique on self-proclaimed neutrality of media platforms also existed

before the Web 2.0, as well as isolated audiences and the effect of business

models on content of (social) media.

What is different in social media logic from mass media logic however is

programmability, popularity, connectivity and datafication. This does not mean

that these logics did not apply to mass media, only that they have become more

important or work differently than in more traditional media. First,

programmers used to alter the audience experience as media as a continuous

flow. In social media logic, this is not only done by programmers of a platform in

47

Nancy Fraser, ‘Rethinking the Public Sphere: A Contribution to the Critique of Actually

Existing Democracy’, Social Text 25, no. 26 (1990): 66.

48

Robert Snow and David Altheide, Media Logic (Beverly Hills, 1979).

the form of an algorithm, but by users as well. Second, media personalities

dominated the faces of the mass media. Social media claim to have a more

egalitarian and democratic appeal, still, some media professionals or experts

have more authority than ordinary users and algorithms increase popularity of

already popular users. Third, traditional media connected content with

audiences based on age, geography, gender and hobbies. Social media use

connectivity to connect users to content, advertisers, and other users.

_{In this}

sense, we should not only think of human connectedness on social media, but

also about automated connectivity.

2.3 Politicians on Twitter

Since social media are here to stay, politicians are part of the ‘Twittersphere’ as

well. This has implications for how politicians communicate on Twitter.

Scholarly focus has been mainly on politicians Twitter behaviour during

campaigning.

_{This is not surprising, since Twitter is mostly used during the}

elections.

Why is Twitter so convenient to use for politicians? According to social

media expert Maurice Vergeer, Twitter is a non-reciprocal social medium, in

contrast with the reciprocal relations on Facebook or LinkedIn. This means that

Twitter could also be used for broadcasting messages, just like the more

traditional mass media would do. Further reasons for the Twitter usage by

politicians are that Twitter quickly became a mobile application for

smart-phones, its communication style resembles the familiar SMS communication,

and Twitter data can be easily used on other web platforms or in research.

Since Twitter is used by politicians, this could have an influence on

politics. Since political parties also use social media platforms for campaigning,

they no longer necessarily need the mass media for visibility of their campaign.

This should benefit small parties with small funding, that no longer need the

expensive mass media.

_{According to Ulrike Klinger and Jakob Svensson, social}

50

Poell and Van Dijck, ‘Understanding Social Media Logic’.

51

José van Dijck, The culture of connectivity: a critical history of social media (Oxford ;

New York 2013) 12.

52

For example: Sanne Kruikemeier et al., ‘Nederlandse Politici Op Twitter: Wie,

Waarover, Wanneer En Met Welk Effect?’, Tijdschrift Voor Communicatiewetenschap 43,

no. 1 (2015): 4–22; Tod Graham, Dan Jackson, and Marcel Broersma, ‘New Platform, Old

Habits? Candidates’ Use of Twitter during the 2010 British and Dutch General Election

Campaigns’, New Media & Society 18, no. 5 (2016): 765–83.

53

Vergeer, ‘Twitter and Political Campaigning’, 755.

54

Vergeer, 746.

media platforms operate with a different logic than traditional mass media,

because it depends on the connectivity of politicians’ messages, other media and

the audience. For one, the network is important because the one platform can

link to other political contents.

_{Moreover, since politicians seek maximum}

attention, it is important for them to establish an audience by being connected

to many others on social media. Another implication is the necessity to go viral,

by sharing positive, personalized and emotional content.

Sanne Kruikemeier and colleagues also emphasize that a lot of politicians’

tweets during election campaign are personal and contain emotions.

Consequence is that this personalized style of communication is used in order to

get individual support. This strategy works, as shown by the study of

Kruikemeier and colleagues, which indicates that the Dutch politicians who used

Twitter in 2010 got more personal votes than their colleagues who did not use

Twitter during campaigning.

_{This supports the thesis that Twitter is an}

important tool for personal image branding, meaning that constructing

favourable self-images is an aspect of social media. This could be seen as a

counterargument of the democratizing force of social media.

_{However, as van}

Dijck and Poell have shown, broadcast producers as well have used emotions

and feelings as a part of media logic to attract a broader and returning

audience.

_{In addition, Vergeer has suggested that Twitter communication can}

lead to a view of politicians as being more approachable.

Another related effect of Twitter and other social media is that citizens

demand more of their representatives. As politicians personal accounts are used

for communication of information, politicians have to be good at explaining their

decisions.

_{Moreover, the “always-on” logic of social media, pressure politicians}

56

Ulrike Klinger and Jakob Svensson, ‘The Emergence of Network Media Logic in Political

Communication: A Theoretical Approach’, New Media & Society 17, no. 8 (2015): 1241–

57; see also: Graham, Jackson, and Broersma, ‘New Platform, Old Habits? Candidates’ Use

of Twitter during the 2010 British and Dutch General Election Campaigns’, 778.

57

Klinger and Svensson, ‘The Emergence of Network Media Logic in Political

Communication: A Theoretical Approach’.

58

Kruikemeier et al., ‘Nederlandse Politici Op Twitter: Wie, Waarover, Wanneer En Met

Welk Effect?’

59

Bennett, ‘The Personalization of Politics’.

60

Poell and Van Dijck, ‘Understanding Social Media Logic’, 4.

61

Maurice Vergeer, Liesbeth Hermans, and Steven Sams, ‘Is the Voter Only a Tweet

Away? Micro Blogging during the 2009 European Parliament Election Campaign in the

Netherlands’, First Monday 16, no. 8 (13 July 2011),

https://doi.org/10.5210/fm.v16i8.3540.

62

Eva Majewski, ‘Online Politics for Citizens in the Twenty-First Century’, European View

14 (2015): 77–83.

to update regularly and also interact more with their electorate than during

election time alone.

For politicians, Twitter is an important tool for establishing

communication between the voters and the voted. Because Twitter and other

social media platforms are relatively new, and require a different media logic,

there is still a lot of uncertainty in how politicians exactly communicate, during

elections but more so during every day politics. This research will shed a light

on how politicians communicate to the Dutch public on a specific topic:

European integration and European-level issues. In order to be able to see how

Dutch politicians communicate European issues, it will be beneficial to see if

they only correspond to, or if they correspond with the Dutch electorate.

2.4 Reciprocity between citizens and politicians

A contested effect of Twitter usage by politicians on Twitter is whether Twitter

allows for a dialogue or is merely an “echo-chamber”. In other words, do social

media foster the public sphere on the internet, or is it a space where only the

most popular opinions and voices are heard. Related to this discussion is the

question if politicians simply use social media as additional medium for

transmitting information to citizens or for communication with citizens.

Ideally, the social networks of the Web 2.0 support interaction between a

politician and their followers. In practice communication between citizens and

politicians on Twitter do not exist on a large scale. In studying national politics

on Twitter, most studies found that politicians mainly interact with other

politicians, journalists and activists, and that there is little space for the normal

citizen to participate in this debate.

Eva Majewsi argues that interaction between these two groups remains

mostly limited to the run-up to elections or for specific issues. Moreover, people

on Twitter use a sort of “screen name”, this avoids ownership of one’s ideas,

leads to speaking with less social restraint

_{and puts the politicians in a}

situation of having to respond to an anonymous mass of people. Thirdly, it

should be considered that communication on Twitter can be seen by everyone

63

Anders Olof Larsson, ‘The EU Parliament on Twitter—Assessing the Permanent Online

Practices of Parliamentarians’, Journal of Information Technology & Politics 12, no. 2 (3

April 2015): 151, https://doi.org/10.1080/19331681.2014.994158.

64

Axel Bruns and Tim Highfield, ‘Political Networks on Twitter: Tweeting the Queensland

State Election’, Information, Communication & Society 16, no. 5 (2013): 667–91.

65

See also: Rasmussen, ‘Internet-Based Media, Europe and the Political Public Sphere’,

99.

and thus requires close attention when writing a response. It demands a lot of

time and thus resources to reply to citizens.

Other studies support that during elections politicians interact more with

citizens on Twitter. Rebekah Tromble showed that a reciprocal relation between

citizens and politicians exists and benefits the democratic decision-making.

Citizens do not expect a direct response to their post aimed at politicians, they

are used to the top-down communication. However, when a politician responds,

it will bring him or her public goodwill.

_{From a politician’s perspective, they}

are more likely to respond when the message is in a neutral or positive tone.

Both Tromble and Todd Graham and colleagues show that Dutch

politicians are having dialogues on Twitter on a regularly basis. On average,

28,8% of their (re)tweets are part of a conversation.

_{The data gathered during}

the 2010 elections in both the United Kingdom and the Netherlands showed that

the Dutch politicians embrace the interactive potential of Twitter and the public

discourse and reciprocity. In comparison to Great Britain, Dutch politicians are

relatively more active on Twitter than their British colleagues. According to the

authors, a reason for this is that a large number of Dutch people were already

early on active on earlier social media sites.

_{This information is important to}

keep in mind when studying politicians’ tweets about Europe, and to what

extent these tweets are being read and have impact on the Dutch digital public

sphere.

66

Majewski, ‘Online Politics for Citizens in the Twenty-First Century’, 79.

67

Rebekah Tromble, ‘Thanks for (Actually) Responding! How Citizen Demand Shapes

Politicians’ Interactive Practices on Twitter’, New Media & Society 20, no. 2 (2018): 681,

https://doi.org/10.1177/1461444816669158.

68

Tromble, 689.

69

Tromble, 688; Graham, Jackson, and Broersma, ‘New Platform, Old Habits?

Candidates’ Use of Twitter during the 2010 British and Dutch General Election

Campaigns’, 778.

70

Graham, Jackson, and Broersma, ‘New Platform, Old Habits? Candidates’ Use of

Twitter during the 2010 British and Dutch General Election Campaigns’.

3 Method: digital humanities and political thought

Social media is a relatively new field of study in humanities that leads to a more

complete understanding of human behaviour, interaction and thought in this era

of Web 2.0. As described above, Web 2.0 moved away from communication

driven by media providers to many-to-many communication, and therefor offers

more interaction between users, content and advertisement. Although studying

the internet seems like a complicated field that is more related to scholars of IT,

Web 2.0 should be seen as just another form of media that could be studied by

humanities scholars and students. The research of social media could be done

by analogic methods like discourse analysis. However, since social media is

made of data, additional results could be abstracted from social media as well

and enrich research. Tracing patterns, in politicians’ tweets for example, only

get real meaning if they are combined with contextualised research.

_This

research will combine methods of traditional humanities with methods of digital

humanities.

3.1 Analysing social media data

One method that this research will adopt is data mining and data text mining.

Data mining could be explained as extracting meaningful, but not easily

obtainable information from a data set by computing the data.

_{Social media}

data has three characteristics that makes it a difficult object for study. Social

media contain lots of data, social media data have unwanted data as well, for

example spam and third, social media data is dynamic, it changes and is updated

frequently.

_{Data mining is used by researcher to reveal insights to social media}

data that could not be obtained otherwise and thus data mining can give a

better understanding of large data sets.

A difference can be made between text mining and data mining. Since this

research uses both methods, a short description will be given of each method.

Data mining uses metadata of an object to tell something about the world,

71

Huub Wijffjes, ‘Digital Humanities and Media History: A Challenge for Historical

Newspaper Research’, Tijdschrift Voor Mediageschiedenis 20, no. 1 (2017): 4–24.

72

Charu C. Aggarwal, ed., Social Network Data Analytics (New York, NY: Springer, 2011),

327.

outside the text collection itself.

_{The data mining method is used in this}

research to look when politicians tweet about Europe

.

In other words, not the

text of the tweet is leading in this part of the research, the metadata of ‘created

at’ is used to give insights in when politicians most frequently tweet about

Europe and which tweets are the most popular.

The second method of text data mining does not make use of meta data

and looks at text alone. Hearst defines text mining as ‘a process of exploratory

data analysis that leads to the discovery of heretofore unknown information, or

to answer to questions for which the answer is not currently known’.

_This

method helps to identify what exactly politicians talk about, when they tweet

about Europe, by analysing word frequencies in the text of a tweet.

3.2 Selecting useful Twitter data

Before the methods of data mining and text mining can be performed, the data

in the form of tweets is selected and filtered, to get a data set suitable for the

topic of this thesis. Filters that are applied are the time range, the language of

tweets, the terms that indicate that a tweet is dealing with European or EU

issues, and the number and names of politicians. The first filter is the time range

of the data. Tweets between 5 November 2012 and 26 October 2017 have been

scraped and cover the Rutte II cabinet period and includes the demissionary

period of the cabinet (starting from 14 March 2017).

The second filtering aspect is the language of tweets. Since this research

is looking at the European and EU issues, policies and actors that are

sufficiently visible in the Dutch political Twittersphere, it involves tweets about

Europe that are aimed at the Dutch citizens. Therefore, only tweets in Dutch are

selected. This means that double tweets of prime minster Rutte who translates

his Dutch tweets in English as well, are avoided. Minister Dijsselbloem was both

minister of finance and president of the Eurogroup. His tweets about Europe are

aimed at EU citizens and are mostly in English, and thus not all Dijsselbloem’s

tweets concerning Europe are included in this data set.

Third, it was decided to only scrape tweets that include terms that are

linked to Europe and EU, instead of scraping all the data of a politician during

the Rutte II period. The following words were chosen to filter tweets: 'EU',

'Europa', 'Brussel', 'Europees', 'Europese', 'Europeaan', 'Europeanen' 'Euro'.

74

Marti Hearst, ‘Untangling Text Data Mining’, University of California, Berkeley, 1999,

http://people.ischool.berkeley.edu/~hearst/papers/acl99/acl99-tdm.html.

These terms have been chosen because they make sure to cover both issues on

an EU-level and take into account more general ideas about a European society

and European integration. For such a large data set of tweets from 15 politicians

in four years preselecting tweets is necessary, because it is impossible to select

by reading this large number of tweets. However, I am aware that this choice of

words that is used for selection is arbitrary. Moreover, a consequence of

filtering before scraping means that the data sets that are being used possibly

left out some of the tweets relating to Europe that do not mention the terms

above. On the other hand, it selects tweets that contain one of these words, but

has nothing to do with European society, politics or integration. It could for

example mention Brussels as a municipality or a European sports championship.

Fourth, for this research the tweets from fifteen politicians are being

analysed. The fifteen politicians are chosen based on the number of followers

they had, measured on 21 March 2016, by the website Toponderzoek.

_The

downside of this measurement is that the artificial and inactive followers are

also recorded. For example, according to the web tool Twitteraudit, 55 % of

Geert Wilders’ followers are fake or inactive, based on the number of tweets,

date of the last tweet, and ratio of followers to friends.

_{This high inactive}

follower percentage also applies to less contested politicians. Moreover, other

methods could also determine how popular someone is on Twitter. For example,

the number of followers could be measured next to the impact of the tweets

themselves, for example how many times a tweet is retweeted and thus

dispersed to non-followers. Another method that can decide Twitter popularity,

is looking at the discussions that a person was involved in by looking at the

replies and mentions a Twitter profile received.

Although the method of establishing a ranking of popular politicians is not

perfect, the top fifteen most popular politicians are useful for this research. It

contains a politician from every political party in the Netherlands in 2016,

except for ChristenUnie and Denk, a small new political party that was founded

in 2015. It also shows a good divide between the governmental parties VVD and

PvdA and the oppositional parties. The oppositional parties are the right-wing

populist party (Partij voor de Vrijheid; hereafter named PVV), the progressive

liberals (Democraten66; hereafter named D66), the Socialist Party

76

Toponderzoek, ‘De top 32 politici met het grootst aantal Twitter-volgers’, 21 April

2016,

https://toponderzoek.nl/2016/03/21/de-top-32-politici-met-de-grootst-aantal-volgers-op-twitter/.

77

Twitteraudit, ‘How many of your followers are real?’, accessed 14 March 2019,

https://www.twitteraudit.com/geertwilderspvv.

(Socialistische Partij; hereafter named SP), the Party for the Animals (Partij voor

de Dieren; hereafter named PvdD), the Christian democrats (Christelijk

Democratisch Appèl; hereafter named CDA), Green Left (GroenLinks; hereafter

named GL) and the Christian reformed party (Staatskundig Gereformeerde

Party; hereafter named SGP). Seven politicians are from the governmental

parties and eight are from the oppositional parties. What is evident from this

dataset, however, is that PvdA had many supporters on Twitter, five politicians

are in this ranking, in contrast with only two of the other governmental party,

VVD. Another observation by looking at these politicians is that only three of the

fifteen politicians in this ranking are women. This could have to do with the

unequal divide of the space for women in the public debate as Fraser argues.

3.3 For every question the right digital method

After selecting which data is useful for this thesis, the tweets are scraped

through the Application Programming Interface (API) of Twitter. This is a fairly

easy process. Access through the Twitter API is granted when you apply for a

developer account. Several personal keys will give will give access to the Twitter

API. Then, the tweets are downloaded, converted in more easily readable csv

and cleaned from unreadable symbols.

A useful data set containing tweets about Europe from fifteen politicians

during the Rutte II cabinet is set up, and then the real computing part of the

research can begin. Which method of data analysis is requested depends on the

sub-question. The first question ‘When do politicians tweet about Europe’

requires the analysis of metadata of when the tweets are composed. To answer

which words politicians are used, data text mining is used as a method. To

answer what sentiments and what frames politicians use, tweets will be

critically analysed and labelled.

For looking which words politicians use when they tweet about Europe,

the Natural Language Toolkit (NLTK) is used. This powerful tool for the

programming language python is especially developed for programming for

language processing.

_{The text of the tweets (lines) are divided in tokens. A}

token is specific sequence of characters that form a word. This research is only

interested in words, so symbols as punctuation or hashtags are filtered out. In

other words, distinct words are extracted from tweets. After filtering Dutch stop

78

Fraser, ‘Rethinking the Public Sphere: A Contribution to the Critique of Actually

Existing Democracy’.

words, a frequency distribution is made to give a list of the frequency that the

tokens appeared in the set of tweets. The word frequency is then visualised in a

word frequency table of the top twenty or top forty most used words.

Since one aspect of politicisation is issue salience, a bar chart of when

politicians tweet about Europe is made in the visualising program Tableau. If

there are peaks in which there is a lot of communication about Europe, these

peaks will serve as case studies. For finding out what sentiments and frames are

being used in communication on European issues by Dutch politicians Tableau is

used for making pie charts and bar charts.

After finding the peaks in the time chart, tweets in this period are being

labelled manually with the following labels: p (positive), n (negative), i

(informative), c (critical), - (no sentiment on European issues or European

integration in general, or impossible to detect the right sentiment based on the

message of the tweet). To distinguish between negative and critical, the

following criterium is used: if the tweet is negative about European-level issues

or is negative about European integration in general, without nuancing or

showing ideas to transform the European Union from within, a tweet is labelled

as a negative tweet. Is a tweet negative about a European-level issue or

European integration in general, but nuances this criticism in the same tweet, or

does show the possibility to transform the European Union from within, this is

labelled as a critical tweet.

_{By labelling the sentiments of the tweets within a}

specific peak, a quick overview of sentiments that are used by politicians when

they tweet about Europe. Moreover, since these peaks probably lead to the

salience of a specific topic, this will help to understand different sentiments that

are communicated on a specific European issue on Twitter.

3.4 Studying political language

In studying politicians’ tweets, two methods are useful: the political thought

method as used by Quentin Skinner and John Pocock and others from the

Cambridge School and framing analysis. The first is more focused on how actors

and societies thought at a specific historical period and what language they used

in expressing these thoughts. The latter is useful in studying the rhetorical

techniques of political language.

80

These criteria are used to distinguish between Euroskepticism and Eurocriticism in

Hans Vollaard en Gerrit Voerman, ‘Nederlandse partijen over Europese integratie’,

Internationale Spectator 71 (2017), 82.

3.4.1 Political thought

Skinner sees texts or speech as acts performed in history, and he argues that

these texts needed to be related to the discourse of the specific time, as well as

the linguistic context of the period. A historian must seek the author’s intention

for writing the text, and thus it is necessary to look for relations between

different statements within the same historical context. According to Skinner,

understanding a text has two components. It should encompass what the author

intended to mean and how this meaning was intended to be taken, because the

text itself has the intention of communication of an idea.

_{According to Pocock,}

historians must understand how, why and what languages are being used,

especially since the language of politics are intertwined in theory and practice.

This means that historians should study the political language both in how this

relates to the societies’ tradition and experience.

Since this research will also study political thought, although of more

recent subjects, Skinner and Pocock’s method will be useful. As the study of

Twitter and especially political communication on Twitter is fairly recent, it will

give new insights of how political thought is dispersed on this social media

platform. Moreover, an important lesson that could be learned from the political

thought method is that studying political communication on Twitter should take

into account not only the message of the tweet, but also, as Skinner emphasized,

it should consider how this meaning was intended to be taken. This could be

done not by studying the text tweet alone, or studying the social context alone,

but by combining these two dimensions. Moreover, studying political tweets

should both encompass the linguistic tradition of the idea communicated in the

tweet and how this text fits in the political practice.

3.4.2 Framing analysis

The second method of framing analysis is frequently used in media studies. In

short, a frame is a feature of an issue or thought that is highlighted. A frame

includes facts and are valanced, meaning that they support a certain point of

view. In a broader explanation a frame refers to the habits of communication in

news, strategic communication choices in mass movement organizations and

81

Quentin Skinner, ‘Meaning and Understanding in the History of Ideas’, History and

Theory 8, no. 1 (1969): 3–53; Richard Whatmore, ‘Quentin Skinner and the Relevance of

Intellectual History’, in A Companion to Intellectual History, 1st ed. (New York, 2016), 99–

112.

82

Kenneth Sheppard, ‘J.G.A. Pocock as an Intellectual Historian’, in A Companion to

Intellectual History, 1st ed. (New York, 2016), 114.

opinion makers, and to how individuals organize facts. Framing is the act of

choosing and applying a certain frame to an issue.

Ece Özlem Atikcan studied the impact of political language upon public

opinion of European integration, by looking at the agenda setting, frames and

outcomes of six EU referendum votes for the European Constitution of 2005.

She states that framing strategies had an influence on the outcomes of the

referendum polls. Atikcan defines framing as giving relative weight to one

dimension of an issue, in order to let individuals focus on these considerations

when forming an opinion. This is not the same as persuasion, it changes the

importance attached to a certain aspect of the issue so individuals might change

their minds or get a stronger opinion of something they already believed in.

_In

the book of Brisione and Michailidou framing is one of the methods used to

research the online politics concerning the European Union. They define

framing as ‘highlighting particular aspects of political decisions or connecting

them to some issues, and not others.’

_{Also Risse sees framing as an important}

aspect of European politics. He argues that different European national

politicians use similar frames in talking about the European Union, and for him

this is an argument of the existence of the public sphere.

Complimentary to framing theory, psychological research uses the

relational frame theory. Relational frame theory sees understanding of things

only possible within a relational frame, that comes from the context of discourse

in which an object is placed. Following Maranke Wieringa, relational frame

theory is useful for studying social media, because whereas framing theory

mostly focuses on the media that are creating and using frames, relational frame

theory argues that not only media, but politicians and citizens use framing

strategy as well.

_{Moreover, previous study showed that political parties are}

very influential in giving frames to political conflict over Europe.

83

Thomas E. Nelson, ‘Issue Framing’, in The Oxford Handbook of American Public

Opinion and the Media, ed. George C. Edwards, Lawrence R. Jacobs, and Robert Y. Shapiro

(Oxford, New York, 2011), 189–203,

http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199545636.001.0001/

oxfordhb-9780199545636-e-12?rskey=DDnGTC&result=3#oxfordhb-9780199545636-bibliography-12.

84

Ece Özlem Atikcan, Framing the European Union: The Power of Political Arguments in

Shaping European Integration (Cambridge: Cambridge Univ. Press, 2015), 17–18.

85

Barisione and Michailidou, Social Media and European Politics, 7.

86

Risse, European Public Spheres, 10.

87

Maranke Wieringa, ‘“F” Is for Fake News / Filter Bubbles / Framing: How Politicians

Frame Media (Content) on Facebook & Twitter during the Elections of 2017’ (Research

master thesis, Utrecht University, 2017), 12.