Ellemijke Donner (10734198) Bachelor thesis political science
‘Emotions and Beliefs in policy making’ Universiteit van Amsterdam
29-01-2017
Rosa Sanchez Salgado Second reader: Rob van Es 9477 words
Contact: ellemijke@hotmail.com
Voters, followers and likes –
personalization of politics in the Digital
Era: the case of Jesse Klaver.
1
Table of contents
1. Introduction ... 2
2. Literature review ... 4
2.1 Social media and personalization of politics ... 4
2.2 Identifying personalization of politics ... 5
2.2.1. Adding emotionalization of politics ... 6
2.3 Implications of personalization of politics ... 7
3. Theoretical framework ... 9
3.1 Model of personalization ... 9
3.2 Expectations ... 11
4. Data ... 12
4.1 Explanation used data. ... 12
4.2 Method of Analysis ... 12 4.2.1. Coding of Twitter ... 15 4.2.2. Coding of Facebook ... 16 4.2.3. Coding of Instagram ... 17 4.3 Data Analysis ... 18 4.3.1 Moving average ... 18 4.3.2. Individualization ... 20 4.3.3. Privatization ... 21 4.3.4. Emotionalization ... 23 5. Conclusion ... 27 5.1 Conclusion ... 27
5.2 Discussion and reflection ... 28
6. Appendix ... 30
I. Codingscheme ... 30
II. Interview with Leon Boelens ... 31
III. Outliers on moving average ... 41
2
1. Introduction
Past elections, GroenLinks had an historical win - they went from 4 to 14 seats in the parliament. The Dutch media stated this win as a ‘reward for the modern, un-Dutch campaign’ (de Witt Wijnen 2017). Un-Dutch in this sense means American, means personalized – a more personal and more glamorous approach (Jinek 2016). Jesse Klaver is the party leader of GroenLinks and he indeed used new ways of connecting with the voters. Klaver and GroenLinks made thorough use of new media like Twitter, Facebook, YouTube, Instagram and even WhatsApp. They were inspired by the campaigns of Obama and Sanders – and it seemed to have worked out pretty well.
In this thesis, I want to contribute to understanding what processes led to this enormous win. One clear tendency in political campaigning nowadays is the use of social media (Kruikemeier et al 2014; Enli and Skogerbø 2013). As Leon Boelens, social media spokesman of GroenLinks and Klaver said: ‘with just social media you will never win, but without it you will never win either’ (Boelens 2017). Another notable tendency concerns the personalization of politics (Van Aelst et al 2011; Bjerling 2012; Van Santen & Van Zoonen 2009; Voerman 2007). This process of personalization shows us a paradox: where the media named the campaign of Klaver modern, un-Dutch and personalized – the campaign fits perfectly in a trend of personalization in the Netherlands (Kriesi 2011: 825). Though, the combination of the intense use of social media, special focus on Klaver instead of the party and the enormous win, make the case of Klaver in the 2017 elections a special one. To research if personalization was really used in the campaign, and to what extent, the research question of this paper will be: to what extent do the social media accounts of Klaver include
personalization? The sub-questions will focus on three different aspects of contemporary political
communication: individualization, privatization and emotionalization. The sub-questions will be the following: do the social media accounts of Jesse Klaver include individualization? Do the social
media accounts of Jesse Klaver include privatization? And: do the social media accounts of Jesse Klaver involve emotionalization? Few recent research concerning personalization of politics in
combination with social media usage exists. Less research has been done referring to the visual representation of personalization on social media (Adam and Maier 2010). This also has to do with a process described by Kruikemeier (et al 2014: 108); when political research involves new
technologies, a research that has been done two years ago can already be outdated.
Facebook has 10,4 million Dutch users, Instagram 3,2 million and Twitter 2,6 million. This means that more than half (!) of all Dutch people are connected online. And as the voters are online, it is important for politicians to be there as well. All Dutch political parties have websites and social media channels like Facebook and Twitter. When focussing on the channels of GroenLinks, the green left party of the Netherlands, it is striking that on all their channels they have less followers than
3 Klaver. The Instagram of GroenLinks even consists of nine pictures, together showing the face of Klaver with the line: #stemvoorverandering. This demonstrates that personalization is deeply incorporated in the political arena. Though this personalization of politics might not be without danger. As the Netherlands consists of a party-centred system and not a candidate-centred system, this might have influence on the way the Dutch democratic system works. Even more: with the personalization of politics, the most popular candidate might win over the most capable. When taking this and the existing literature gap in mind, it shows that research concerning personalization of politics and the usage of social media is extremely relevant.
To be able to answer the asked questions, first a literature review will follow. In this review, the contemporary debates surrounding personalization of politics will be addressed. Secondly, the theoretical framework will be discussed and a combination of two frequently used models
concerning personalization will be made. Than the data analysis will follow. The data used to
investigate the personalization on the channels of Klaver are online accessible for everyone. Finally, a conclusion will follow.
4
2. Literature review
To answer the research question of this paper, a literature review will follow. This review consists of three parts: social media and personalization of politics, identifying personalization of politics (including emotionalization) and the implications of personalization.
2.1 Social media and personalization of politics
The internet has become a vast part in our daily live – as a famous saying goes: ‘the world lies at our
fingertips’. This constant stream of connectedness and information also plays an increasing role in
the communication between voters and politicians. Nowadays we can barely think of a time where one didn’t have a smartphone, but when we go back just 20 years – (political) life was very different in the Netherlands. In the campaign of 1998, internet was just a small, extra way of campaigning. Especially because there were not many people using it. Television, that reached millions of people, was way more important. But this changed just a few years later. In the campaigns of 2002 and 2003 blogs and websites were thoroughly used. The communication happening through those channels was fairly traditional - political parties put their statements on it and that was that. There was not much interaction. Even for the campaign of 2006 this was still the case. But since then, there is a trend where parties are using other channels as YouTube, Hyves (a Dutch substitute for Facebook), Twitter and Facebook (Kruikemeier et al 2014: 109). Nowadays, even 10,4 million Dutch people use Facebook, 3,2 million use Instagram and 2,6 million use Twitter (Van der Veer et al 2017: 6).
The communication through these new media channels is more personal and more interactive than it used to be with traditional media (radio, television and newspapers). The information online usually exists of references to party politics or personal characteristics of politicians and is directly communicated with the followers. The voter, as follower, is reached in a different and more simple manner. By communicating this way, politicians are able to make a personal connection with voters (Kruikemeier et al 2014: 121). In other words: technologies have eroded the barriers between the public and the private sphere (Van Aelst et al 2011: 206). Leon Boelens1, social media spokesman for
GroenLinks and Klaver states that through social media, it is easier to show a personal side of a politician to the voters (Boelens 2017). Enli and Skogerbø underline this: ‘social media such as
Facebook and Twitter (…) represent semi-public, semi-private spaces for self-representation where borders between offline personal and online mediated relations are blurred. They allow politicians (and voters) to stage their public and private roles, and to shift between them seamlessly and more or less consciously and strategically’ (2013: 759). Politics get more and more personalized. This
personalization of politics is a long-term ongoing process where political communication has become
1 References to an interview with Leon Boelens, online spokesman of GroenLinks and Klaver, will be made. This interview is done on 20-12-2017 and can be found in Appendix II.
5 increasingly focused on personalities and personal traits of politicians (Enli & Skogerbø 2013: 758). Social media like Twitter and Facebook provide new impetus to the personalization of politics (Enli & Skogerbø 2013: 757). They create opportunities for intimate relations with voters (idem: 758). As sites like Facebook and Twitter place the focus on the individual rather than on the political party, they expand the political arena for increased personalized campaigning (idem: 758). Voerman notes that this personalization of politics, strengthened by the intense use of social media, will change the political system as we know it. In his paper Voerman concludes that the future of politics lays more with the politician than with the political party – and due to this trend, the focus on personality, emotions and charisma of the politician will increase (2007: 22). Note that there are two sides to this coin: as the personalization of politics grows, so does the politicization of popular culture. TV dramas, game and talk shows – they all compete for the most exciting and attracting political news (idem: 759). This focus on a person instead of a party might be dangerous: as the most popular candidate might win over the most capable one.
2.2 Identifying personalization of politics
This (potentially dangerous) process of personalization of politics, is not so agreed upon in the political science area. There is no consensus on the definition of personalization; and scholars do not agree on the very existing of personalization (Karoven 2010: 2; Voerman 2007: 17). Kriesi did a comparative research in six European countries, and hardly found any evidence of a personalization trend in recent decades: except for in the Netherlands (Kriesi 2011: 825). And as Voerman notes: ‘whether they [the researchers] are right or not: politics sees personalization as reality and creates its
own reality, helped by the media’ (2007: 17). Thus, even if the existing evidence is not convincing, it
has to be accepted that a process of personalization is happening (an example: during the Dutch elections of 2002, two third of the election coverage concerned politicians instead of political parties – which does show an increase of personalization when compared to 1994 and 1998 (ibid)). But how is this process of personalization defined? Some scholars speak of personalization, others of
personification, orientation towards personae, individualization or Americanisation, and again others of privatisation. Enli and Skogerbø (2013: 758) for example define personalization of politics as a long-term ongoing process where political communication has become increasingly focused on personalities and personal traits of politicians. This is a very broad definition of the concept of personalization – and one that not every scholar agrees upon. There is no clear consensus about the concept of personalization.
According to Van Aelst, Sheafer and Stanyer, the rise of candidate-centred politics is seen as the result of two interconnected factors. The first being the weakening of traditional affective ties between voters and parties, the second the changing media environment. The concept of
6 personalization is used in relation to the behaviour of voters, political actors and the media (Van Aelst et al 2011: 204). Van Aelst (et al 2011: 205) defines two forms of personalization,
individualization and privatization. Individualization - the first form of personalization – is defined as
‘a focus on individual politicans as central actors in the political arena, including their ideas, capacities and policies’. This implies a shift in media visibility from parties to individual politicians.
The second form of personalization is privatization: ‘[privatization] implies a shift in media focus from
the politician as occupier of a public role to the politician as a private individual, as a person distinct from their public role’ (Van Aelst et al 2011: 205). Adam and Maier address personalization as a
longitudinal process as well, and identify two positions: the first refers to a stronger focus on politicians (instead of parties), the second refers to a focus on personal, non-political characteristics that become more relevant (Adam and Maier 2010: 216). Kriesi (2011) also addresses this distinction. Bjerling (2012: 45) defines these two reoccurring positions as a hard and a soft side of
personalization: the hard side being the trend of individualization, the soft side the increased importance of personal matters and characteristics. Brants and Santen (2014: 149) address, as they call it, two affective components: popularisation and personalization. These components aim to win voters over in an unsecure voter environment by addressing what interests and is liked by the voters (ibid). They define popularising as the portraying of the politician as the ‘normal’ men and
personalization as emphasizing the political and personal qualities and emotions of a politician (ibid). Whilst these different authors refer to the same processes, they name them differently. Van Aelst (et al 2011: 206) note that although there may not be consensus about the exact naming of the process of personalization, there is consensus about three issues. First, that personalization refers to a trend, a process of change over time. Second, that the personalization of political news is a
consequence of new media technologies and the strategies of political actors. Third, that this process is multi-dimensional.
2.2.1. Adding emotionalization of politics
While there is an existing consensus about those three aspects, one aspect of political
communication has not (yet) been incorporated: emotionalization. Although most scholars do refer to it between the lines, not all of them go as far as to recognize it as an extra dimension of
personalization. Richards (2004: 340) does state that if political leaders do not show emotions in their political communication, the democratic deficit will grow. This deficit exists because in the modern era, popular culture is ‘about the expression and management of emotion (…) the incursion
into political experience of the values of popular culture means that we now seek certain kinds of emotionalized experience from politics that we have not done in the past’ (ibid). Therefore, to be able
to connect with the voters, politicians will have to show emotional qualities – or at least use
7 what he calls a cultural transformation, where politics more and more include the personal, with psychological considerations and emotionality (Richards 2004: 348). Voerman also refers to this ‘new emotionality’ as a change in society, where nowadays there is less of a taboo on the showing of personal emotions and feelings; on television and on internet people reveal their deepest feelings – as do politicians (2007: 20). By showing more of their personal and emotional side, a politician tries to identify with the world of the voters (Voerman 2007: 20). Richards names these processes as ‘emotional labor’, which means that ‘politicians must present themselves as individual persons of a
particular emotional makeup, who (…) can offer themselves as social-emotional, expressive leaders as well as task-orientated leaders’ (2004: 348). Bjerling also refers to this trend of emotionalization in
society and politics: ‘from having been passionate about principles, policies and ideology, the party
leaders are over time increasingly often portrayed in ways and situations signalling affectionate relations between them and the people within their personal spheres’ (2012: 204). The most
elaborate study that includes emotionalization is the one done by Van Santen and Van Zoonen (2009). Their literature study concerns the key concepts of personalization – and concluded that a definition of personalization should exist of three pillars: individualization (focus on individual political characteristics), privatization (focus on private life) and emotionalization (interests for personal feelings and experiences) (idem: 170).
2.3 Implications of personalization of politics
But what does this personalization of politics (including individualization, privatization and
emotionalization) imply? Why does it happen and above all: does it work? As noted earlier Enli and Skogerbø (2013) and Voerman (2007) address that personalization of politics might have implications in the political area. The media compete for the most exciting news – whereby they especially focus on the politicians and their personal lives (in the Dutch election campaign of 2002, two third of the coverage was referring to the candidate, whereas one third referred to the party (Voerman 2007: 17)). This focus on the candidate might lead to a political system where not the most capable, but the most popular candidate wins the elections. Also, the role of political parties is inclined to change. During the pillarization, the Dutch political parties were anchored in the social and religious life and had a membership rate of 15 % of the Dutch people (Voerman 2007: 16) – nowadays a political party is more an organisation that facilitates the political process and nomination of candidates, with a membership rate that has declined to 2,5 % (Voerman 2007: 22; Parlement en Politiek 2016). Media focus on the private person of politicans because, apparently, this is what voters like to see. Political parties acknowledge this trend and they also shift their focus to one main candidate. This works because voters like a political leader they feel they can trust, a leader with which they can identify (Bjerling 2012: 6; Voerman 2007: 15; Richards 2004: 340; Boelens 2017). As Bjerling states: ‘the kind
8
of leader that we today look for is one that reminds us of our ‘good selves’; the kind of person that we entrust with the running of our tenant organisation; the kind of parent who we believe is most likely to protest if our kids are served cold food at school’ (2012: 201).
To reach a point where people feel like they can trust a political leader, this leader must put in the earlier named ‘emotional labor’ to (partly) close the emotional deficit and show that they are made out of flesh and blood, by showing their personal side. Politicians do conform to the noted trend of personalization in the Netherlands (Kriesi 2011: 825), and for example show more of their family life to their voters (Bjerling 2012: 202). Other research shows that smiling on pictures triggers positive reactions with voters (Sullivan and Masters 1988). After this research little other research has been done about personification with data from images and video’s (Adam and Maier 2010). Bjerling (2012: 186) confirms this and states that very little is known about the visual manifestations of personalization (in his research, the images show the most personalization of the investigated ways). So while it is known and accepted that there is an ongoing trend of personalization in the Netherlands, especially since the use of social media is increasing, there is still a lot of research to be done when it comes to visual manifestations and personalization of politics.
9
3. Theoretical framework
Now that the existing debates concerning the personalization of politics have been addressed, the focus shifts to the theoretical framework, where the model of Van Aelst (et al 2011) will be explained and expanded with the theories of Van Santen and Van Zoonen (2009) and Richards (2004). Finally, some expectations concerning the data will be defined.
The data of this qualitative study are gathered from the case of Jesse Klaver. This case can be seen as a unique case because of several reasons. First, because GroenLinks had a historical win past elections. Second, they made such thorough use of social media and even What’sApp, which hadn’t been done before. The research concerns a single-case design, with a single unit of analysis. The case being a unique one, the results are not transferable to other contexts. They do however give insight in the relation between personalization and social media – which is of importance for future
research.
3.1 Model of personalization
To be able to contribute to the existing literature about personalization, a more elaborate review of personalization will follow. As noted in the literature review Van Aelst (et al 2011) shows an existing consensus about three points: first, that personalization refers to a trend, a process of change over time. Second, that the personalization of political news is a consequence of both media technologies and the strategies of political actors. Third, that this process is multi-dimensional. Their study aims to make a broad useable operationalisation, it has a substantially literature review and displays a wide knowledge of the subject. Their model is used for this research. The operationalisation of Van Aelst et al refers to a shift over time, and this research only checks if the stated process occur on the social media channels of Klaver. By doing this, it is accepted that these shifts concerning personalization appear over time and the model of Van Aelst et al is used to research how these different process of personalization develop.
Van Aelst (et al 2011: 209) divide the concept of personalization in two dimensions with subcategories. The first dimension consists out of individualization and privatisation, as shown in figure 1. Individualization refers to a heightened focus on individual politicians and a diminished focus on parties (idem: 209). Privatization refers to the attention that is being paid to personal, non-political characteristics. The second dimension of individualization exists of general visibility
(politicians in the news, opposed to parties) and concentrated visibility (shift to a limited number of political leaders). The second dimension of privatization exists of the attention on the personal life on the one hand and the attention on more personal characteristics on the other (ibid). For the personal life, Van Aelst et al propose four index factors, referring to different types of personalization. These include: coverage of the family, past life and upbringing, leisure time and love life (2001: 213).
10
Figure 1. Model Van Aelst et al 2011: 207.
Though, as noted in the literature review, there appears to be a lack of reference to emotion, or use of emotion in politics and political communication in the model of Van Aelst (et al 2011). For this research to be complete, to include the ‘new emotionality’ of Richards (2004) and the emotion aspect of personalization as discovered by Van Santen and Van Zoonen (2009) a dimension should be added to the scheme of Van Aelst et al. This is shown in figure 2. Emotionalization will in this new model be the same as the emotionalization used by Van Santen and Van Zoonen: emotionalization as interests for personal feelings and experiences. This emotionalization also refers to the use of emotions in political communication. Again, reference is made to the shift concerning personal experience and the use of emotions in political communication. This because just as individualization and privatization of politics, emotionalization is an on-going process (Richards 2004; Bjerling 2012; Voerman 1988).
11
3.2 Expectations
In this study, the data will be gathered from the social media platforms of Jesse Klaver. By focussing on the personal channels of Klaver, the aspect of general visibility is already incorporated in the analysis. It is thus expected that his personal channels do include a lot of individualization, and especially concentrated visibility – which is supposed to show Klaver in the role of leader. Leon Boelens, social media spokesman of GroenLinks, state that they consciously focus on Klaver as a leader, because: ‘he is a leader that people like to follow’ (Boelens 2017).
When referring to privatization, the shift of the public to the personal is meant. Privatization consists of the focus on the personal characteristics and personal life of a politician. As it has shown to be difficult to separate political traits from non-political traits (Bjerling 2012: 38) and as this study focusses on the personal channels of Klaver, there will be no attention paid to this part of
privatization (as this reference lies most with others writing about a politician, a politician will not refer to his or her own characteristics as much). When concerning personal life, previous studies have found an increase in the showing of the family (Bjerling 2012: 202). It is therefore expected that Klaver also refers to his personal life.
When referring to emotionalization in political communication, most people will think of populist right-wing parties – and their messages of hate, anger and fear. But which emotions can we then expect from left wing parties? According to Boelens, GroenLinks especially focusses on a different message: one consisting out of hope and happiness. It is thus expected that the emotion words referring to hope and happiness occur the most. As earlier research showed that visuals showing politicians smiling evoked positive responses from voter, we expect to find the same results (Sullivan and Masters 1988).
12
4. Data
In this chapter, the data analysis will follow. First, explanation of the used data is given. The method of analysis follows, explaining how the used coding scheme works and on what literature the emotion words from the coding scheme is based. An explanation of the moving average follows. Examples of the different treated social media platforms will pursue. After this, each sub-question is addressed separately.
4.1 Explanation used data.
The used data is gathered, coded and analysed in a time period of three months. The data consist of 132 tweets (from 21-09-2016 till 12-07-2017), 97 Facebook statuses (from 21-11-2016 till 16-5-2017) and 78 Instagram posts (from 21-09-2016 till 4-7-2016). The data include the campaign period, which was officially from the 15th of December 2016 until the 15th of March 2017. To include the (possible)
fore- and aftermath of the elections, earlier and later posts were included. The tweets are from the account of Jesse Klaver (www.twitter.com/JesseKlaver), the Facebook statuses are from the
Facebook of Jesse Klaver (https://www.facebook.com/jesseklaver) and the Instagram posts are from Klavers personal Instagram (https://www.instagram.com/jesseklaver/). All these channels are accessible for everyone.
4.2 Method of Analysis
The coding scheme can be found in appendix I and will be explained here. The coding concerning individualization was focussed on the appearance of Klaver in video’s or picture’s – was it Klaver alone, with somebody else or was it a picture/video without Klaver. It also included whether or not Klaver was portrayed as a leader (this is defined as: standing alone in front of a mass of people or being interviewed in the role of party leader). The coding of individualization was just focussed on the visual part of the communication, this has not yet been done before. By focussing on the personal channels of Klaver, here the biggest part of individualization is included.
For the coding concerning privatization, four categories are used: coverage of the family, past life and upbringing, leisure time and love life.
For the coding of the data concerning emotionalization the focus was mostly on the text. A coding scheme was used, consisting of the primary emotion words (1 to 6) and an extra category (7) containing the words verandering, eerlijk and empathie, as these words reoccurred frequently. There was also a category 8 added, containing emotion words that seem like emotion words but do not fit the framework used. Later, this category proved not to be usable, as there were too many different words in it to be able to make statements about it. There was also a code to state if Klaver was smiling (only indisputable smiles where recognized as such) – to see if this really evoked more likes.
13 On all three studied channels the number of likes is stated. This can thus be seen as a
constant factor in the three different channels. By incorporating the number of likes in the data analysis, it is possible to see how many likes picture’s, video’s or emotion words evoked. This gives us a bit of insight in what works best to get the most likes. Likes are important for the social media accounts, as it is shown to people who do not follow Klaver that other persons like his messages. This way, the reach of a post is bigger if a post gets many likes.
For the Twitter, Facebook and Instagram analysis the emotion words framework by Shaver (et al 1987) is used, as this is a frequently used and exhaustive one. For this study, we accept these emotion words and do not address the debate about what emotions actually are – as this is not of necessary in this research. The focus is on political communication and the use of emotion words – not on what emotions actually are. For their research, Shaver et al asked 112 students to participate in a prototypicality-rating. First, they compiled a list of 213 emotion names (like worry, excitement, guilt) by combining existing lists of emotion names. Then they asked the students to rate these 213 words between 1 (I definitely would not call this an emotion) to 4 (I definitely would call this an emotion). After this, 135 words were left that had an average rating of above 2.75 and could thus be accepted as emotion words. When the word list was set, they did another test to be able to sort the words under the right emotions (like love, joy, hate and anger). This time 100 students were asked to sort out the words in different categories. Note that no limitations were given whatsoever. After these tests, Shaver et al came to three different dimensions of emotions. Primary, secondary and tertiary emotions. Primary emotions in their framework are love, joy, surprise, anger, sadness and fear. Secondary emotions for example for love would be affection, lust and longing. Tertiary
emotions for lust would be arousal, desire, lust, passion and infatuation. Their framework is shown in table one. In the coding scheme, just the primary emotions were used – but only words that could be found in the table were coded.
14
Primary Emotion Secondary Emotion Tertiary Emotion
Love Affection Adoration, affection, love, fondness,
liking, attraction, caring, tenderness, compassion, sentimentality
Lust Arousal, desire, lust, passion,
infatuation
Longing Longing
Joy Cheerfulness Amusement, bliss, cheerfulness,
gaiety, glee, jolliness, joviality, joy, delight, enjoyment, gladness, happiness, jubilation, elation, satisfaction, ecstasy, euphoria
Zest Enthusiasm, zeal, zest, excitement, thrill, exhilaration
Contentment Contentment, pleasure
Pride Pride, triumph
Optimism Eagerness, hope, optimism
Enthrallment Enthrallment, rapture
Relief Relief
Surprise Surprise Amazement, surprise, astonishment
Anger Irritation Aggravation, irritation, agitation,
annoyance, grouchiness, grumpiness
Exasperation Exasperation, frustration
Rage Anger, rage, outrage, fury, wrath, hostility, ferocity, bitterness, hate, loathing, scorn, spite, vengefulness, dislike, resentment
Disgust Disgust, revulsion, contempt
Envy Envy, jealousy
Torment Torment
Sadness Suffering Agony, suffering, hurt, anguish
Sadness Depression, despair, hopelessness, gloom, glumness, sadness, unhappiness, grief, sorrow, woe, misery, melancholy
Disappointment Dismay, disappointment, displeasure
15 Neglect Alienation, isolation, neglect,
loneliness, rejection, homesickness, defeat, dejection, insecurity, embarrassment, humiliation, insult
Sympathy Pity, sympathy
Fear Horror Alarm, shock, fear, fright, horror,
terror, panic, hysteria, mortification
Nervousness Anxiety, nervousness, tenseness, uneasiness, apprehension, worry, distress, dread.
Table 1 - Table from ‘Emotion Knowledge: Further Exploration of a Prototype Approach’ by Shaver et al 1987.
4.2.1. Coding of Twitter
For the coding of the Tweets of Klaver, the main focus was on the text, but appendixes were also included. Picture one shows a Tweet of Klaver including different words that fit in the coding scheme. The word zin (can be translated in excitement), fits in category 2 (joy). The words eerlijker and end
empatischer, fit in 7.3 respectively 7.2. The appendix shows a retweet of GroenLinks – but the
pictures are very small and therefore not prominent enough to be coded.
16
4.2.2. Coding of Facebook
The coding of the Facebook posts exists of the written text in the post and the appendix of the post – which is a little film fragment or a picture almost every time. These appendixes are also analysed and included in the coding of the post. This is done because, according to Boelens, people on Facebook watch a video clip on average for one minute and a half (Boelens 2017). If a clip lasts longer than two and a half minutes (which happened five times on 97 posts), I did not include it, as it can be expected that most people who saw the post did not watch the video. Below an example of the coding of a Facebook is shown:
Post: Afgelopen zaterdag hadden we met GroenLinks
een geweldig (2) congres. Daar presenteerden we
onze nieuwe koers voor Nederland. Ons antwoord op het rechts-populisme moet een radicale aanval zijn op ongelijkheid en onzekerheid (6). Ik schreef daarover dit stuk: bit.ly/koerswijziging-nl. Een offensieve agenda voor een eerlijker (7.3) en empathischer (7.2) land. Zo kunnen we bouwen aan een land waarin we elkaar weer vertrouwen (8)en we samen welvaart maken en delen.
Het kernwoord isvertrouwen (8). In een klimaat van polarisatie, toenemende tegenstellingen en dreigend terrorisme is dat de opdracht: vertrouwen herstellen. De aanslag in Berlijn laat zien dat
veiligheidsmaatregelen nodig zijn om geweld te voorkomen. Maar dat is niet genoeg. We moeten als samenleving ook ons gevoel van verbondenheid versterken. Als Rutte en Wilders in het voorjaar de verkiezingen winnen, dan verliest Nederland. Dan nemen de tegenstellingen en de ongelijkheid toe. Dan neemt het wantrouwen (8) toe. Ik vecht ervoor dat we na 15 maart een linkse en progressieve coalitie kunnen vormen. Met de SP, de PvdA, D66 en de christelijke partijen. Ik ben daarom blij dat de PvdA een linksere koers lijkt in te slaan. Tegelijkertijd voel ik enig ongemak (6). Ik heb de afgelopen jaren harde debatten gevoerd met het kabinet over de toegenomen ongelijkheid, de bezuinigingen op de ouderenzorg, de belastingontwijking van grote bedrijven. Iedereen voelt aan dat er iets wringt als Lodewijk Asscher in harde bewoordingen afstand neemt van de VVD, zonder verantwoording af te leggen over de innige samenwerking met Mark Rutte. Dat tast de geloofwaardigheid van de politiek aan. Volgende week is het Kerst. Voor mij zijn dat altijd dagen van bezinning en geniet (2) ik van het samenzijn met mijn familie. De beste dagen van het jaar. Even niets doen en opladen voor de spannende verkiezingscampagne in 2017. Dat wordt een optimistisch (2)
jaar, een jaar van verandering (7.1). Daar gaan we hard aan werken. Ik wens je alle liefde (1) en geluk. - Jesse
Bijlage: foto Klaver als leider (1).
17
4.2.3. Coding of Instagram
Instagram is a social media platform that evolves around pictures mostly. The text written next to the pictures is most of the time very short and referring to the pictures. Emoticons are also used. If an emoticon was a happy face (:-)) or a heart (<3) it was coded as 2 respectively 1. The Instagram post below gives an example of a text where only the word verandering (7.1) fits in the coding scheme, and he uses it twice. It is thus also coded twice. The image shows a clear picture of a leader with his support in the back (1). Note that with Instagram, when a video is posted, it is not possible to see the number of likes. Therefore, the posts with videos were left out of the analysis.
18
4.3 Data Analysis
With the analysed data, the number of likes were also included. As noted earlier: likes are important for the online campaign, as posts with a lot of likes have a greater reach than posts with less likes. Thus, by obtaining a lot of likes, posts spread on the online platforms (as explained earlier also to people who are not interested in GroenLinks or Klaver). In this research a dimension was added that shows the number of likes compared to the average number of likes; to see what kind of posts work well in obtaining likes. It must be noted that the average GroenLinks voter is between 18 and 34 years old, 54 % is higher educated and 61 % identifies as female (NOS 2017). These numbers are not representative for the Dutch voters.
4.3.1 Moving average
The number of likes of the posts increased over time (and decreased after the elections) - for example, in 2016 400 likes on Instagram was a fairly good score, but in 2017 it wasn’t. Thus, to be able to make sensible statements about the number of likes, the moving average was calculated – for each different channel. The moving average is the average number of likes at one point, calculated by the number of likes of four posts before and after a certain post. Graph number 1 shows the moving average for Facebook, graph number 2 shows it for Twitter and graph number 3 for Instagram.
Graph 1 - moving average of Facebook.
0 5000 10000 15000 20000 25000 30000 1 6 111621263136414651566166717681869196 Reeks1 moving avg 4
19
Graph 2 - moving average of Twitter.
Graph 3 - moving average Instagram.
These graphs already show a few notable things. First: there is one post that is an outlier on Facebook and Instagram (it was not posted on Twitter). This post refers to the passing away of Klavers mother. This is a message that evoked sympathy and people showed their support by giving likes. The other posts on Facebook that were outliers (number 49 and 84) are presented in the Appendix III with the rest of the outliers. They concern a post about the party program of GroenLinks and a post on the 15th of March (election day), that encourages people to vote on GroenLinks. The
outliers on Twitter are referring to Trump and Rutte in a negative way. These Tweets are responding to actuality and therefore evoke responses (Boelens 2017). Most striking are the outliers on
Instagram. Each outlier on Instagram shows Klaver with a family member. Number 36 is showing 0 200 400 600 800 1000 1200 1400 1600 1 9 17 25 33 41 49 57 65 73 81 89 97 10 5 11 3 12 1 12 9 Reeks1 moving avg 4 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 likes moving avg
20 Klaver with his two children, ready to go for a swim. Number 63 shows Klaver and his wife after election day, celebrating in front of a mass of people. Thus, only when it comes to the Instagram posts we see similarities in the posts that get most likes.
4.3.2. Individualization
After nominating the average outliers, the focus shifts to the sub-questions. Starting with: do the social media accounts of Jesse Klaver include individualization? Individualization refers to a
heightened focus on individuals and a diminished focus on parties. Individualization exists of general visibility (shift to individual politicians) and concentrated visibility (shift to leaders). The general visibility is already incorporated by the focus on the channels of Klaver. Concentrated visibility is measured by seeing if Klaver is displayed as a leader.
4.3.2.1. Data analysis individualization
The data show that of the 97 Facebook posts, only 5 do not include any visual appendix. Of these 92 visual appendixes, Klaver is shown alone 24 times and together with others 39 times. When he appears on these visual appendixes, it is as a leader 44 times. An appearance as a leader is coded when 1) Klaver is standing in front of a mass of people, speaking to them 2) Klaver is being
interviewed in the role of party leader – as these situations show the image of a leader. On Twitter there is a visual appendix in 84 of the 132 tweets. Klaver appears alone 20 times and with others 28 times. He is portrayed as a leader 28 times. On Instagram, there are visual appendixes on every post, as this is the main interest of the channel. He is portrayed alone 25 times and with others 38 times. Of these, he is portrayed as a leader 13 times. One example is shown in picture 4, where Klaver portrayed as a leader, almost worshipped by his ‘followers’.
21
Social media platform Number of
appearances as leader Percentage of likes +/- average Facebook 44 - 19 % Twitter 28 - 5 % Instagram 9 (+4) - 6 % Total 85
Table 2 - appearances as leader.
Table 2 shows the number of posts where Klaver appears as a leader, and how well these posts do when comparing the likes to the moving average. The percentages show that it can be stated that Klavers followers do not like the posts where he appears as a leader. On Facebook these posts even do 19 % worse than the average post.
4.3.2.2. Results individualization
A fair amount of individualization was expected – in the sense of concentrated visibility – on the personal channels of Klaver. In the total of 254 posts, Klaver is portrayed as a leader (in the category of concentrated visibility) 85 times. This is in 33 % of the posts. Relatively he is portrayed as a leader the most on Facebook (48 %, against 33 % on Twitter and 17 % on Instagram). This might have to with what Boelens said about the difference in use of the channels. He stated that Twitter is used to reach journalists, Instagram for the young voter and Facebook mostly for the followers of GroenLinks and potential voters (Boelens 2017). As noted, there is a case of concentrated visibility. At the same time, when comparing likes to the average, it shows that the representations of Klaver as a leader do not evoke great numbers of likes, it even shows a negative trend. The portraying of Klaver as party leader over the portraying of Klaver as part of the political party is a consequence of new
technologies and social media. Due to these trends, political communication has become increasingly focused on personalities (Enli & Skogerbø 2013: 758).
4.3.3. Privatization
Do the social media accounts of Jesse Klaver include privatization? A short recapitulation: privatization happens when the focus shifts from the public to the personal. This happens in two ways: the focus shifts to non-political traits (personal characteristics) or/and the focus shifts to private life and personal interests (personal life). When privatization happens through a focus on the personal life, this happens in four ways: coverage of family life, past life, leisure time and love life.
4.3.3.1. Data analysis privatization
One striking thing concerning privatization is that it appears mostly on Instagram. On Twitter on the contrary, in all the 132 analysed Tweets, he only references to his private life once. This tweet contains a picture of a football game on television and the text: ‘come on Ajax!’. There are no tweets in the analysed data where Klaver refers to his personal characteristics. Out of all his Facebook posts, he refers to his personal life only twice. Once when he is attacked about his background – and he has
22 to ‘defend’ himself. The other time is when his mother passed away – to thank everyone for the nice messages. In the analysed data of the Facebook posts there are no posts where he refers to his personal characteristics. On the earlier named Instagram posts, Klaver refers to his personal life 13 times in 78 posts. Five of these posts are referring to his family life, one is referring to his family and his past (this is the same post as on Facebook, concerning his mother), five are concerning his leisure time (playing football and cooking, mostly) and two are about his wife.
Personal Number Percentage of likes +/- than
average
Coverage of family 5 26 %
Past life 0 0 %
Leisure time 5 19 %
Love life 2 5 %
Combination family and past life
1 35 %
Total 13
Table 3 - personal references Klaver.
Table 3 shows that all the posts concerning pictures referring to Klavers personal life do really well on Instagram – they all score more likes than the average. Striking is the one outlier post
combining his family life with his past life got 35 % more likes than the average at that time. This is the post concerning the passing away of his mother, and is thus an extremely rare case (as there is also just one post).
The other reference to the personal, family life, is also a topic that scores well when focussing on the likes. See for example picture seven, on which Klaver got 3,245 likes – the post before got 1,598 likes and the one after got 1,081 likes. References to leisure time are also doing quite well.
Picture 7 – Klaver with his children. Picture 6 – Klaver with campaigner.
23
4.3.3.2. Results privatization
Previous research showed an increase in references to personal life – but only when referring to visual representations (Bjerling 2012: 202), thus it was expected that references to personal life by Klaver were present as well. The results again show a difference in the use of the social media channels. Twitter and Facebook are not used to show a private side of Klaver, however this is done on Instagram frequently. This is in line with previous research, as Instagram evolves around visual communication. It also explains why other authors do not find such strong evidence for privatization, as they do not pay special attention to pictures or focus on Instagram. The followers of Klaver on Instagram do like the personal posts – better than the normal ones. This might has to do with the fact that voters like ‘passionate politicians’ and leaders they feel they can identify with (Bjerling 2012: 6).
4.3.4. Emotionalization
Now we have seen to what extent the social media accounts of Jesse Klaver involve privatization, our focus shifts to the same question about emotionalization. Emotionalization in this paper is referred to as the use of emotions in the political communication.
4.3.4.1. Data analysis emotionalization
Code Facebook word use Facebook % of likes +/- moving average Instagram word use Instagram % of likes +/- moving average Twitter word use Twitter % of likes +/- moving average Total use of word Score in % on weighted average 1 – love 11 30 % 7 9 % 3 0 % 21 19 % 2 – joy 28 8 % 24 14 % 15 72 % 67 25 % 2.1 - hope 29 24 % 2 10 % 2 23 % 33 23 % 3 – surprise 1 127 % 0 0 % 0 0 % 1 127 % 4 – anger 11 3 % 2 0 % 7 16 % 20 7 % 5 – sadness 13 - 5 % 1 35 % 6 - 29 % 20 - 10 % 6 - fear 41 10 % 2 - 7 % 4 60 % 47 14 % 7.1 - change 69 - 2 % 32 - 4 % 37 - 15 % 138 - 6 % 7.2 - empathy 28 24 % 7 - 16 % 7 - 5 % 42 13 % 7.3 - honesty 15 - 12 % 5 - 16 % 6 - 25 % 26 - 16 %
Table 4 – data analysis emotion words.
Table 4 shows us the number of words used for the coded categories and how a post did when comparing the number of likes to the average. The number of the total use of the emotion words
24 (categories 1 to 6), shows that the joy category scores high. Hope has been made a separate
category, as there were so many references to hope – but when put with the joy category, joy even reaches 100 mentions. This is a lot, especially in comparison with the other emotion words
categories. Surprise on the other side, is only mentioned once. Surprise is left out of the analysis, as one mention is not applicable for analysis. The category that is mentioned most after joy is fear – which is not in line with what is expected when hearing Boelens, as he stated that GroenLinks and Klaver referred to positive emotions instead of negative ones. Though, what is notable about the posts where there is being spoken of fear emotions, is that love, joy and/or hope emotions are almost always set against this fear. Some examples of this are shown in table 5. If a message includes a negative emotion like fear or anger, an emotion of love or joy will be put against it, as the
alternative. The alternative of these positive emotions is then connected to GroenLinks. Which implies, that if you do not want to vote for fear, you have to vote for positive emotions – connected with GroenLinks.
Facebook post number 26:
Facebook post number 68:
Facebook post number 83:
25 Looking at the words change, empathy and honesty (which were added because of their frequent appearances), it appears that change was used a lot. In the 307 analysed posts, change occurred 138 times, which is almost in half of all the social media posts. The main message of the Klaver campaign was: tijd voor verandering, which can be translated in ‘time for change’. As this message of change was propagated a lot, and included in half of all the Tweets, Facebook and Instagram posts – it is not of use to make statements about it. The categories 7.1, 7.2 and 7.3 will therefore be no further explored.
On Facebook, the category that is most used is joy (combined with hope) with 57 references, followed by fear with 41 references. This in line with the earlier findings. It is also notable that the only emotion category receiving less likes than the average, is sadness. Hope and love on the other side, are emotions that do get peoples likes on Facebook.
On Instagram a slightly different tendency is noticed, where references to joy have been made way more than references to other emotions. These also perform best of the emotions – they get 14 % more likes than the moving average. Note that the one post is left out where sadness was included out of the analysis, as this is the earlier treated post concerning Klavers mother.
On Twitter, references to the joy category have also been made the most. This fits in the tendency which was seen with Facebook and Instagram. It appeared that these get the most likes, even 72 % more than the average tweet. Sadness scores minus likes (again), and fear evokes even 60 % more likes than average.
Table 6 – appearances smiling.
Table six shows the number of appearances where Klaver is smiling. Smiling is only coded when it was undoubtable a smile – as shown in picture 6. Most of these smiling appearances are made on Instagram. It is striking that on Twitter there are only three pictures that show Klaver smiling; but these got 112 % more likes than the average. On Facebook and Instagram these smiling appearances also got more likes.
Social media platform Number of
appearances smiling Percentage of likes +/- average Facebook 10 10 % Twitter 3 112 % Instagram 17 10 % Total 30
26
4.3.4.2. Results emotionalization
In the literature review it was addressed that political leaders need to show emotions to voters to not let the democratic deficit grow (Richards 2004: 340). The online strategy of GroenLinks includes the use of emotions, especially positive ones. References to hope and happiness were expected. Also, when a visual representation of Klaver smiling is shown, it is expected to evoke positive responses: expressed in a high number of likes. The data analysis shows that using emotions in posts (except for
sadness), elicits responses (in likes) from the followers. Joy is the category that works best – people
‘like’ this emotion. Followed by hope and love. Sadness doesn’t work, as people do not respond to this by liking the posts. Fear on the other hand is getting 14 % more likes than posts without this word, but this could be because it is often put in one post with the emotions like love and joy. It was expected that images showing Klaver smiling were more liked than pictures without it, and this is confirmed by the data analysis of Klaver smiling. A possible explanation for these findings is that voters prefer a political leader they feel they can trust. This feeling of trust is partly evoked by showing emotions (Richards 2004: 340) – like smiling on pictures. The use of emotions in political communication fits in the trend of new emotionality, where emotions are a big part of daily life and ways of communicating (idem: 348).
27
5. Conclusion
In this chapter, the conclusion of the research is given. Afterwards, a discussion and reflection follow.
5.1 Conclusion
The traditional relation that political parties and voters had, is languishing (Van Aelst et al 2011: 204; Voerman 2007: 16; Parlement en Politiek 2016). New technologies and social media have eroded the barriers between the public and the private spheres (Van Aelst et al 2011: 206). Political parties put politicians forward and use new strategies to win voters in an unsecure voter environment. The existing literature does not address these strategies of personalization in combination with social media – as technology develops quickly and new platforms keep appearing. To make a small
contribution to filling the literature gap, a case-study on Jesse Klaver, party leader of GroenLinks, was done. The research question of this paper was: to what extent do the social media accounts of Jesse
Klaver include personalization? To be able to answer this question, sub-questions were formed and
a literature review was done. The literature review revealed a discussion about what personalization of politics actually was and how it should be defined. The study of Van Aelst (et al 2011) gave a certain revelation as they noted consensus under the scholars about three issues: personalization should be seen as a process of change over time, personalization is a consequence of new media technologies and the process of personalization is multi-dimensional. Building on these consensuses, Van Aelst et al formed a model of personalization. This model consisted of individualization and privatization. Emotionalization was added – as this also appeared to be of crucial importance to personalization.
The data analysed to answer the research question consisted of the social media channels of Jesse Klaver in a time period that included all the posts of the 15th of December 2016 to the 15th of
March 2017. Different data were gathered, concerning all different aspects of personalization. The moving average was calculated as well, to be able to make statements about the number of likes on certain posts. Previous research displayed a lack of research on the visual representations of
personalization – a process that is especially strong on the social media accounts of politicians. In this research, the aim was to highlight the process of personalization in politics and the role social media have in this.
Al three sub questions concerned the different aspects of personalization: individualization, privatization and emotionalization. The data analysis revealed that all three parts of personalization were present on the social media accounts of Klaver. The sub questions (do the social media channels of Jesse Klaver include individualization/privatization/emotionalization?) can thus be answered with yes, they do. Focussing on individualization, the social media of GroenLinks showed a shift to individual politicians (general visibility). Subsequently, the social media of Klaver included
28 concentrated visibility (portraying of Klaver as leader) in one third of all the posts. Individualization was found on all the studied social media. Privatization on the other hand, was only found on the Instagram of Klaver. These personal Instagram posts get more likes than the average posts of Klaver. Emotionalization was found on all the different channels of Klaver. The data analysis showed that using emotions in posts (except for sadness) elicits responses (in likes) from the followers – especially positive emotions like joy and hope. But, against the expectations, Klaver also referred to fear
frequently. When giving these messages (containing fear) a closer look, it showed that the emotion of fear was almost always set against positive emotions. These positive emotions were then linked with GroenLinks. This way, people were encouraged to vote for positive emotions (over negative ones) by voting for GroenLinks. This trend of positivity was also found in the visual appearances of Klaver. Klaver was portrayed smiling in 30 posts, and these posts also evoked more likes than the average posts.
The voter profile of GroenLinks differs from the average Dutch voter. It can be expected that people following Klaver on social media are also the people who vote for GroenLinks. Therefore, the results of the analysed data might differ from what would be transferable to other political parties. Even so, concerning the case GroenLinks (and partially other green left parties) the research contains some interesting findings. First, portraying the party leader as leader does not work well when referring to the number of likes. Second, showing the private side of a politician receives a lot of likes. Third, positive emotions work well when aiming for a greater reach of a topic. This research showed that in the unique case of Klaver, a trend of personalization was discovered. The research did not focus on the shift over time of personalization, but it can be accepted that it also concerns a shift in time, as social media are a relatively new phenomena.
5.2 Discussion and reflection
As this research was focussed on the unique case of Jesse Klaver, the results are not transferable to other cases – except maybe for other Green Left parties in western Europe. The followers of the social media of Klaver are described as the same people who follow Klaver on Instagram, but this might not be the case. Unfortunately, no access was given to these data by GroenLinks. The number of given likes was treated as a method of measuring how well a post performed, but there was no psychological substantiation for this assumption. However, the number of likes is important for the reach of a post – and is thus still important. The time measured was a bit more than three months and was chosen around election times. Election times might differ from when a party is seated in parliament or as opposition. This research did an analysis of personalization on the personal accounts of Klaver, which is a different approach than what has been done before (earlier studies focussed on
29 the media reporting about politicians). But as this is the Digital Era, where social media are
omnipresent, it seemed that this new approach was needed.
For further research, it would be interesting to make a comparative study of GroenLinks and a populist party, to see how the different supporters of that party respond to the messages posted by the party. Also, more research on emotions in political communication should be treated, as this was not addressed sufficient in this thesis.
30
6. Appendix
I. Codingscheme
Love 1 Joy 2 Hope 2.1 Surprise 3 Anger 4 Sadness 5 Fear 6 Verandering 7.1 Empathie 7.2 Eerlijk 7.3 Overig 8Picture Klaver alone 1.1
Picture Klaver and others 1.2
Picture without Klaver 1.3
Video Klaver alone 2.1
Video Klaver and others 2.2
Video without Klaver 2.3
Klaver portrayed as leader 1
Klaver not portrayed as leader 0
Klaver portrayed smiling 1
Klaver not portrayed smiling 0
Reference to family life 1
Reference to past life or upbringing 2
Reference to leisure time 3
31
II. Interview with Leon Boelens
Semigestructureerd interview met Leon Boelens, online woordvoerder van GroenLinks, op 20-12-2017. Interviewer (Ellemijke Donner) wordt aangeduid als E, geïnterviewde als L. Het interview is leesbaar gemaakt – dat wil zeggen termen als ‘uh, haha, ehm’ zijn eruit gehaald en kromme zinnen zijn omgevormd.
L: kom maar op!
E: heb jij nog vragen over mijn onderzoek?
L: nou wat is je vraag?
E: mijn hoofdvraag is nu: hoe heeft Klaver emoties gebruikt in de campagne? Maar ik denk dus dat dat nog iets aangepast gaat worden naar meer de personalisatie, omdat ik dat eigenlijk interessanter vindt.
L: ja, en wat is dan de definitie daarvan?
E: je hebt drie dimensies in personalisering; individualisering en privatisering. Individualisering is dat het veel meer op de persoon is in plaats van de partij (zoals ten tijde van verzuiling), privatisering is echt over de niet-politieke karakteristieken en het persoonlijke leven. Mijn focus ligt meer op privatisering, dat vind ik heel interessant. Zo weet ik dat Jesse Klaver ukelele speelt en een vrouw en twee kinderen heeft, maar ik weet niet wie de derde persoon op de lijst van GroenLinks is.
L: ja, ja, precies.
E: dus daar wil ik het nu heel graag over doen.
L: oké, grappig, ja.
E: en dan vooral, wat voor een rol social media daarin speelt, daar is een heel groot gat in de literatuur.
L: ja, oké, cool
E: kan jij vertellen wat jij precies doet?
L: mijn titel is online woordvoerder, en dat houdt in dat ik de online kanalen voor Jesse en
GroenLinks vanuit de fractie beheer, dat ik daar eindverantwoordelijk voor ben, die beheer er dingen op plaats.
E: en doe je dat alleen of heb je een team?
L: nou, we hebben twee soorten teams, het landelijk campagne bureau zit in Utrecht,
daar zitten vooral heel veel mensen die zijn bezig met de campagne. Ik overleg met hen over de dingen die we online doen. En met de medewerker in Brussel, omdat er ook Europees nieuws op dezelfde kanalen komt. We overleggen veel met elkaar over hoe we het beste resultaat kunnen behalen. Verder hebben we nog een vormgever, beeldvoerder, assistent beeldvoerder, die ook bezig zijn met oké hoe krijgen we die dingen in beeld. Eigenlijk moet ik de teams een beetje aansturen.
32 L: zo zou je het kunnen noemen, online eindredactie. Ja, niets gaat online zonder dat ik dat weet.
E: oké, en ik zag dat bijvoorbeeld op de Instagram van GroenLinks een hele grote foto van Jesse - en ook dat hij op alle kanalen meer volgers heeft dan GroenLinks zelf. Is dat een trend die jij veel terugziet in de Nederlandse politiek? Dus dat het zo veel meer over de politicus als persoon gaat?
L: ja, zeker, het is makkelijker schrijven over iemand. Ik denk dat daar een wetmatigheid in zit, in de media, het is makkelijker schrijver over een mens dan over een niet mens, een mens doet dingen, een partij is een geheel van wat mensen vinden. In de politiek nu zie je dat er in bijna alle partijen met mensen wordt geschoven, behalve dan misschien bij de Christen Unie, maar als je ziet hoe Klaas Dijkhoff naar voren is geschoven, Thierry Baudet nu naar voren wordt geschoven… Ik verwacht bijvoorbeeld ook dat ze bij de SP veel meer met Lillian Marijnissen gaan pochen, om het even zo te zeggen. Omdat je overal ziet dat de identiteit van een politicus veel belangrijker is, ook door social media. Je kan daar je persoonlijke kant veel meer laten zien.
E: maar dat is wel een ontwikkeling van de laatste jaren, of denk je dat altijd wel zo was?
L: ik denk dat het altijd wel zo is geweest, mensen kunnen eerder politici opnoemen dan politieke partijen. Ik denk dat het altijd wel zo is geweest. Maar het is vooral zo dat wanneer je een leider hebt, onthouden mensen de leider. Daarom willen mensen het ook goed doen, je onthoudt eerder een persoon, daar identificeer je je ook meer mee dan met een partij.
E: zijn jullie er actief mee bezig, zoals die foto’s van Jesse op de GroenLinks Instagram, om Jesse bijna een merk te maken?
L: ik wil het geen merk noemen, maar we zijn er natuurlijk wel mee bezig. Natuurlijk, wij zien ook dat dat werkt in de campagne. Toen Jesse het stokje overnam van Bram, onze vorige fractievoorzitter, zagen we meteen dat Jesse een hele grote aantrekkingskracht had op nieuwe kiezers, jonge mensen. Dat had GroenLinks daarvoor niet, het had heel erg het imago van geitenwollensokken
boomknuffelaars die veganistisch eten. Jesse daarentegen was jong, had idealen, was fanatiek en heel veel mensen hadden zoiets van die gast - dat vind ik wel een mooie gast. Hoe meer mensen hem leerden kennen, hoe meer mensen zich aansloten bij GroenLinks. Je zou gek zijn als je daar geen gebruik van zou maken. We hebben daar erg op ingehaakt in de campagne. Jesse is een leider, iemand waar mensen achter willen staan. Hij spreekt mensen aan.
E: als persoon of als politicus?
L: ik weet niet of die scheiding van toepassing is. Je wil politiek ook in bepaalde mate menselijk maken. Je wil dingen laten zien van zijn persoonlijke leven, tegelijkertijd je wil ook laten zien dat hij de touwtjes in handen heeft - leiding kan geven. Hij moet iemand zijn die je vertrouwt als leider van een land of partij. Ik geloof niet dat wij een heel duidelijk onderscheid gemaakt hebben tussen de mens en de politicus. De politiek wil je humaniseren, om te laten dat het ook ‘jou’ raakt in plaats van
33 dat je het als een abstract iets in het midden wil laten. Om te laten zien dat ‘mensen’ die keuze moeten maken, dus zo probeer je die werelden samen te brengen als het ware.
E: dus misschien meer opdat de kiezers zich met Jesse kunnen identificeren?
L: nou kijk, via sociale media is het heel makkelijk om de achterban te bereiken. Sociale media werkt gewoon veel beter als je het schrijft vanuit een persoon. Een persoon heeft een mening, een partij niet, een partij heeft een standpunt. Je kan niet zeggen: ‘dat vind ik belachelijk’ namens GroenLinks. In zoverre is het makkelijker om te spreken vanuit een mens.
E: merk je dat Jesse zijn persoonlijke berichten, bijvoorbeeld de Instagram post waarin hij zijn zoon naar school brengt, dat die meer respons krijgen dan standpunten over bijvoorbeeld belasting?
L: je moet een onderscheid maken tussen verschillende standpunten die je hebt, zo weten wij dat sociaal culturele standpunten bij veel mensen het best werken. Dingen als papa-dag, ik weet niet of je dat hebt gezien? Dat hij bij Jinek met Johan Derksen zit, dat filmpje is een van de best lopende filmpjes op zijn Instagram ooit. Dat heeft iets van 80.000 views.
[L laat filmpje papa-dag op de Instagram van Jesse Klaver zien]
L: ja hier zie je dus heel duidelijk dat hij een standpunt inneemt en wordt aangevallen. Vervolgens maakt hij er als persoon, als vader er politiek van. Dit heeft niets te maken met politiek en 100 procent met Jesse als persoon. Hij legt geen politiek standpunt uit, maar impliciet straalt hij wel een waarde uit die wij als politieke partij willen uitstralen. Daarom moet je denk ik niet denken in welke onderwerpen het grootste bereik halen, maar iets abstracts zoals dividendbelasting koppelen aan waarden. Door die dividendbelasting tegenover docenten en vrachtwachten chauffeurs te zetten, laten we zien dat de regering de mensen in de steek laat en kiest voor multinationals in plaats van mensen. Opnieuw wordt gekozen voor de waarden, of dat doen wij in ieder geval, en je ziet dat dat ook mensen aantrekt
E: en focus je nu ook op bepaalde aspecten van zijn leven?
L: eigenlijk niet.
E: op alles?
L: ja als hij wat tegenkomt wat hij leuk vindt posten we dat. Hij vindt koken bijvoorbeeld heel leuk, dus dan doen we daar wat mee. Maar dat is dan meer omdat hij toevallig iets aan het bakken is, en voetbal vindt hij erg leuk, dus daar doen we dan ook wat mee.
E: maar er zit niet echt een strategie achter?
L: nee, ik denk ook niet dat dat werkt. Je moet wel iets hebben wat wel echt is. Die neppe dingen daar trappen mensen niet in. Nu denken al heel veel mensen dat het nep is, laat staan wanneer je echt neppe dingen gaat doen. Stel hypothetisch; je zou wat nep doen, en dat komt uit, dan kan je
