Erasmus Mundus M.A.
Journalism, Media and Globalisation
Specialization: Media and Politics
THE CASE OF DENK:
Expanding the protest paradigm to new political parties
by
Jesse Joannes Beentjes
Student ID: 10175008
Master’s Thesis
Graduate School of Communication
Master’s programme in Communication Science
Supervisor/Examiner: dr. mr. J.H.P. van Spanje
2
Table of contents
Abstract p. 3
Introduction p. 3
Description of the case: DENK and its background p. 5
Theoretical framework p. 5
The hierarchy of influences model p. 5
Framing theory p. 7
The protest paradigm p. 7
Hypotheses p. 8 Methodology p. 10 Operationalization p. 11 Reliability p. 13 Limitations p. 13 Results p. 14 Discussion p. 17 Conclusion p. 19 Acknowledgements p. 21 List of references p. 22 Appendix 1: Codebook p. 27
3
[ABSTRACT]
The 2017 Dutch elections were characterized by a considerable number of new political parties. New political parties are fundamental to pluralist democracies. Their performance and organization have been studied extensively, but analyses of media coverage remain scarce. This paper takes the newspaper coverage of the five national Dutch newspapers (n= 449) on the new Dutch political party DENK as the subject of a quantitative content analysis. As such, it measures to what extent the protest paradigm was used to frame this party. Was the attention of the party’s main agenda drawn away towards either deviant, threatening or impotent frames? This paper connects the protest paradigm to Shoemaker and Reese’s hierarchy of influences model, in order to give an overview of factors influencing the frames. As such, it also investigates if this concept, which was previously reserved for protest movements, is an effective way of categorising media frames of new political parties. By doing so, this paper takes a new step towards the formulation of a new theory on media framing of new political parties
Keywords: protest paradigm, new political parties, the Netherlands, elections, hierarchy of influences,
4
Introduction
On the day of the 2017 Dutch parliamentary elections, a problem was identified by many voters: the large amount of parties on the voting ballots made it hard to fold the paper properly, in order to vote (RTL Nieuws, 15 March 2017). This was caused by the fact that no less than 28 parties were running for parliament, half of them ‘new parties’ (Kiesraad, 2017). Only two of them eventually made it into parliament, a score that fits into earlier research on the amount of new political parties entering the Dutch Tweede Kamer (Krouwel and Lucardie, 2008).
New political parties are a fundamental feature of vivid pluralist democracies. The demand for these parties is related to social heterogeneity and social or value change; they emerge because traditional parties have failed to absorb new issues in their programmes (Sikk, 2012; Hug, 2000; Muller-Rommel, 2002). Seemingly marginal politicians and groups can quickly catalyse powerful institutional changes, once the global government changes (Henson and Kopstein, 2005, 91). Previous research has brought forward a number of conditions for new parties to be successful, that are related to the party’s origin (roots in civil society, initial programmatic profile) and its developmental process (organisational strength, leadership potential and the occurrence of defections) (Beyens et al., 2015; Bolleyer and Bytzek, 2016).
This paper takes into consideration another factor that is important for the profile and performance of new political parties: media coverage and framing. It takes DENK, one of the two political parties to make it to parliament in the 2017 Dutch elections, as a case study. Coverage from the five main Dutch daily newspapers of this party (n=449) was the subject of a quantitative content analysis, measuring the presence of the protest paradigm. This term was originally coined to describe a set of media frames in coverage of protest movements (Lee, 2014). For the first time, this paper expands the paradigm to the realm of new political parties, and as such the paradigm figures in the main research question of this paper:
RQ1: To what extent is the protest paradigm present in the media coverage of the new Dutch political party DENK?
The theoretical framework that forms the backbone of this study, combines framing theory and the protest paradigm as its manifestation with Shoemaker and Reese’s model of the hierarchy of influences, in order to explain the outcome of this study. The multiculturalist agenda of DENK is also taken into consideration, as the
5
framework relies on on Cottle’s (2000) claim that multiculturalism impacts on all levels of editorial decision-making.
The societal relevance of this study lies in the fact that new political parties challenge and change the traditional landscape in both ‘new’ and old democracies all around the world (Selb and Pituctin, 2010). Examples are the Spanish Podemos, the Italian Movimento Cinque Stelle and recently En Marche!, the party of the French president Emmanuel Macron. All these parties have affected the democracies they operate in, and should therefore be taken seriously in all their aspects, including media coverage.
Academic analysis of media framing of new political parties is very scarce; while new parties as a whole are an increasingly popular subject of study (Sikk, 2012; Engler, 2015; Zons, 2015; Beyens et al., 2015), framing analysis of political parties in the media focusses on old political parties or remains limited to new right-wing anti-immigration parties, such as the Dutch PVV (Baumgartner and Chaqués-Bonafant, 2015; Mazzoleni, 2003; Boomgaarden and Vliegenthart, 2007; Bos et al., 2008). This paper steps into that void, by taking a non-right wing, explicitly pro-immigrant party as a case study. The findings of this paper add media framing as another factor to the already existing conditions for a new political party’s success, and set a step towards the formulation of a new theory regarding media coverage of new political parties.
Description of the case: DENK and its background
The political party DENK is an example of the third category of new parties as defined by Krouwel and Lucardie (2008): a split-party. It was founded in February 2015 by two former members of the Labour Party (PvdA). Turkish-Dutch members of parliament Tunahan Kuzu and Selçuk Öztürk first were a fraction, the Group Kuzu-Öztürk, that stayed in the parliament on the seats that formerly belonged to the Labour-fraction: they were so-called seat-thieves (‘zetelrovers’) (Du Pré, 2016). Having left the Labour party because of a major disagreement on integration and multiculturalism, these two topics became the main content of DENK’s agenda. This is already reflected in the name of the party: ‘denk’ is not only the Dutch first person singular and imperative of the verb ‘to think’, it is also Turkish for ‘equality’. According to its motto, ‘DENK is a party for the assertive New Dutch’.
6
The party was controversial from its launch in February 2015, because of its aggressive political style and continuously displayed distrust in the media. Besides of this, the fact that Kuzu and Öztürk refused to acknowledge the Armenian genocide or to take distance from the Turkish president Erdogan and his repression of the press and opposition raised questions about their loyalty to the Netherlands. For a short period, famous Surinamese-Dutch DJ Sylvana Simons joined the party, which brought again a lot of media attention for the party, only to leave the party a couple of months later, according to herself because of an increasing discontent with DENK’s ‘too conservative’ direction (Nieuwsuur, 2017). During the 2017 elections, DENK won three seats in parliament.
Theoretical framework
This study uses three main theories, namely the protest paradigm, the hierarchy of influences model and framing theory. In the following paragraphs, each individual theory used in this study will be shortly discussed, after which a line will be drawn between the three.
The hierarchy of influences model
The influential hierarchy of influences-model is not a formal theory, as it would be hard to incorporate all the different levels of analysis present in the model in one single study (Li, 2016). The model is more of a ‘theoretical umbrella’, or a tool get an overview of all the different implications, individuals and institutions that together shape the represented reality in the media (Li, 2016).
The model is built up out of five concentric circles, in which “media content” is the core, being hierarchically influenced by all outer layers. Each level has its features and role-players, but together, they form a ‘media society’. In this media society, each individual layer is interrelated to the others, rather than that there are strict boundaries. The model serves as a tool to see the interconnectedness of the media with culture, other organizations and institutions (Shoemaker and Reese, 2014). The model is laid out in the following layers, from small to large: individual, routines, organizational, extra-media (institutional) and ideological (socio-cultural) (Shoemaker and Reese, 2014; Reese, 2007).
7
that identifies that them. This can be as gender, sexual orientation, race, political preference and age, but also a journalist’s professional role conception, educational background and ethics (Shoemaker and Reese, 2014). The second level of the routines, emerges over a longer period of time in the newsroom, where patterns and routines take shape that influence content or decide whether this will be content at all, as White’s well-known “Mr Gates”-principle shows (White, 1950).
Thirdly, the organizational level, comprises the internal structure of a media organization, for example enterprise policy, but also the ownership of the organization (Shoemaker and Reese, 2014) and the way in which this directly or indirectly influences the content. The fourth level, the extra-media level or social institutional level, looks at other social institutions that influence media content, such as media regulations in countries or developments in society. Finally, the ideological level or socio-cultural level looks at the way ideological forces shape media content (Shoemaker and Reese , 2014). The levels can be prioritized in terms of importance and influence, according to Reese, but there is no final answer to the question how this exactly should be done (Reese, 2007). Reese suggests a bottom-up building up of the model, in which everything starts at the individual journalist, who is dealing with the outer layers. He also mentions that it could work the other way around, in which the ideological level can be prior to any journalistic actions taken. In the discussion section, there will be a reflection of what is the most plausible way of prioritizing the levels of influence on the extent to which the protest paradigm was present in the coverage of DENK, based on the findings of the content analysis.
Framing theory
One of the characteristics of media content is framing. This broad concept that is used in multiple disciplines, but means with regards to media content the process which ‘sets the frame in which citizens discuss public events (Tuchman, 1978; Entman, 1993; Pan and Kosicki, 1993; Vraga et al., 2010) ) The frame makes the public actively classify, organize and interpret the information (Goffmann, 1974). Framing takes place through the use of metaphors, examples, catchphrases, depictions and visual images. In other words: the frame shapes the story, which then again shapes the reception of the article. Media coverage can have direct effects on political opinion-making through direct exposure, according to several studies (Norris et al., 1999). A party’s election results can
8
depend on visibility and tone in the media. Based on a research in Denmark, a country that is comparable to the Netherlands in terms of media system and a history of coalition governments, Hopmann et al. conclude that the more visible and the more positive the tone toward a given party is, the more voters are inclined to vote for this party (Hopmann et al., 2010). However, they set forward a number of conditions for the strength of media effects: the amount of visibility only seemed to influence undecided voters, while the tone of the coverage did not influence voting behaviour at all (Hopmann et al., 2010). They also conclude that the effects of news coverage on voting behaviour is only visible when talking about the cumulative information environment; in other words: by the repetition of discourse in the news and interpersonal communication about the news, media coverage influences voting behaviour.
Research on the effects of media and media framing on the perceptions of new political parties in the Netherlands is so far somewhat limited to right-wing (populist) parties (Bos et al., 2011). In a 2011 study, Bos et al. concluded that the strength of media effects for right-wing populist party leaders on the public’s perception was not significantly different from other party leaders, arguing that these leaders are just as dependent on media as leaders of other parties (Bos et al., 2011).
The protest paradigm
One set of frames that has been identified from research on media coverage of protests and protest movements since the 1980s, is the protest paradigm. In short, it describes the media’s tendency to marginalize movements by drawing attention away from the core issues raised by these movements (Weaver and Scacco, 2012; Reul et al., 2016). Research indicates that this bias is structurally in favour of the status quo, as they frame protest movements as being deviant, threatening or impotent (Shoemaker, 1999; Lee, 2014). This process delegitimizes in a way the protest or protest movement (McLeod, 2007).
The deviant-component of the protest paradigm, firstly, falls in line with what Stuart Hall has called the process of ‘othering’, in which certain characteristics of groups within society are repeatedly emphasized, which creates a majority-minority dynamic with the ‘we’ or insiders and ‘they’ or others (Hall, 1982). The threatening component of the protest paradigm can be explained as media putting emphasis on violence or non-conformism with (for example) democracy. The impotent component of the paradigm
9
finally puts emphasis on the protest’s failure to achieve its goals or maintain an effective internal organization.
This paper expands the protest paradigm from the realm of protest movements to new political parties, as the frames present within the protest paradigm, can also be projected on news coverage of new political parties, such as DENK. Reul et al. conclude that much of the reasons of the occurrence of what is called the protest paradigm, can be traced back to the level of journalistic routine in the hierarchy of influences model (Shoemaker and Reese, 2014). Professional journalistic routines are developed in response to an array of practical considerations and constraints, such as information availability, audience expectations and a medium’s limitations. `
Connecting the abovementioned concepts, the theoretical argumentation of this paper can be summarized as the following. The frames that are used to map the way in which DENK has been framed and that might influence the public’s opinion on DENK and its voting behaviour, fit into the protest paradigm. An explanation for the way in which DENK has been framed, can be found on multiple levels of the hierarchy of influences-model. Summing up, a theoretical line can be drawn from the composition of a newsroom to the framing of new political parties in the media. This framing then again influences electoral behaviour and as such the performance of new parties such as DENK.
Hypotheses
The main research question of this paper asks to what extent the protest paradigm is present in the Dutch newspaper coverage of DENK. With the challenges of the multicultural society defined as DENK’s core issues, the ways in which the protest paradigm draws away the attention of these issues still need to be specified. Based on the findings of previous studies on the frequency of the paradigm that state that the protest paradigm is used in defence of the status quo, the first main hypothesis of this paper is as follows:
H1: In most of the articles, the protest paradigm is present.
‘Most’ here means that in more than 50 percent of the articles analysed, the frames employed to describe DENK fit into the protest paradigm. However, as will be discussed furtherly in the methodology section, in order to measure the protest paradigm, it needs to be split up and furtherly defined: when are we talking about the protest paradigm
10
exactly? As the protest paradigm is about the marginalization of protest movements by drawing attention away from their core issues, these core issues first need to be specified in the case of DENK, based on its party programme (Weaver and Scacco, 2012).
The ‘inclusive society’ is defined by DENK itself as its core topic in its 2017 programme for the election, referring to a multicultural society in which problems that migrants and second or third generation immigrants face are taken seriously (the programme called for the introduction of an anti-racism-police) (DENK, 2016). This ‘inclusive society’ in DENK’s discourse can be reduced to four main issues: anti-discrimination, migration, anti-racism and integration. Whenever these four issues from DENK’s core agenda are addressed, we cannot label an article as using the protest paradigm. In order to support the first hypothesis, the majority of articles must not address these issues. Therefore, the first sub-hypothesis reads the following:
Sub-H1: The majority of the articles does not address DENK’s main agenda (migration, racism and discrimination).
Again, the majority in this case is defined as more than 50 per cent. The paradigm itself, when found present in the articles, however, is still to be split up. It consists of three main parts: the party or movement is either deviant, threatening or impotent. A condition for speaking about the protest paradigm is that at least one of these three components is used to frame DENK, when its core agenda is not mentioned. So, in order to be able to support H1, this second sub-hypothesis needs to be supported as well: Sub-H2: All articles that do not address DENK, use at least one frame from the deviant, threatening or impotent categories within the protest paradigm. In order to support the hypothesis that the majority of articles uses the protest paradigm, not only the majority of articles must not address DENK’s main agenda, within this majority, at least one of the protest paradigm-frames must be used.
Given the fact that DENK in essence is a political party and not a protest movement with physical rallies, we can argue that it is more likely for the party to be framed as deviant or impotent, than as threatening. When putting up a ranking of the frequencies, a second hypothesis therefore expects that the deviant and impotent components score higher than the threatening one, regardless whether the protest paradigm occurs in the majority of articles or not. In this case, ‘most articles’ is defined in percentages.
11
‘deviant’ and ‘impotent’ frame , rather than the ‘threatening’ frame.
Furthermore, some expectations can be raised about the general tone of the articles about DENK. According to Lee (2014), the protest paradigm is generally used in favour of the status quo. Therefore, it is the expectation that DENK, a party that challenges the traditional political landscape by entering the arena, is framed more negatively than positively.
H3: The majority of articles has an only negative or more negative than positive tone towards DENK.
Methodology
Figure 1 Amount of articles mentioning DENK per newspaper (n=449)
For this paper, 688 newspaper articles were gathered through the database LexisNexis, using the searching terms ‘DENK’, ‘Kuzu’ and ‘politiek’. The five main national Dutch newspapers were studied: namely De Telegraaf, Algemeen Dagblad, de
Volkskrant, NRC Handelsblad and Trouw. The period used was February 9, 2015, the date
on which the party DENK officially launched its new name, to March 14 including, 2017, the date of the Dutch parliamentary elections.1 Out of these articles, a sample was
constructed (n=449), based on Q1: Is the political party DENK mentioned in this article? This large difference can be explained simply by the fact that the word ‘denk’ is also the first-person singular of the verb ‘to think’ (denken) in the Dutch language.
The sample articles were classified on date and on the type of article. The largest percentage of articles (29,6%) was formed by background articles, followed by news articles (19,2%), columns/editorials (16,5%), readers’ letters (10%) and opinion pieces (6,5%). All newspaper content that appeared after performing the search was studied,
1
The exact term used was: (((("DENK") AND ("Kuzu") OR ("Politiek"))) and Date(geq(09/02/2015) and leq(15/03/2017)))
12
regardless the type of article. News articles, background pieces, editorials and opinion pieces were taken into consideration, as well as readers’ letters. There was no further statistical sampling performed for this study, as the entire population was analysed.
Operationalization
As discussed in the hypotheses-section above, in order to measure whether the protest paradigm was present in an article or not, the concept had to be broken down in roughly four components. Seen as the protest paradigm presents a set of frames that are fit for protest movements, the paradigm needs to be reconceptualised for the framing of new political parties in the media
Firstly, there is the question whether or not the main agenda of the party was discussed in the article or not. DENK takes an explicit pro-migrant standpoint as its starting point and has a strong anti-racism and anti-discrimination agenda. The ‘inclusive society’ is defined by DENK itself as its core topic in its 2017 programme for the election, referring to a multicultural society in which migrant problems are taken seriously (the party programme called for the introduction of an anti-racism police). This definition forms the basis of the measure that tests whether DENK’s core issues are mentioned or not: is DENK being connected towards issues regarding as integration, migration, racism or discrimination (Q7). This measure helps to find evidence for H2: The majority of the articles does not address DENK’s main agenda (migration, racism and discrimination).
Then, the three main components of the protest paradigm need to be operationalized. Whether the party was framed as threatening, or in other definitions of the protest paradigm violent, was operationalized in the following three measures. Firstly, the question whether the party is called extremist or radical (Q5). This is quite clear: extremism and radicalism forms a threat to the status-quo. Secondly, by the question whether the party was called anti-democratic (Q9), in which the threatening component is the question whether the party is considered a ‘threat’ to democracy. The third variable is the question whether the party is described as authoritarian or dishonest (which is again threatening).
The impotent component of the protest paradigm was operationalized through only one variable, measuring whether the party is called a party without political experience (Q11), which has implications for the organizational success of the party.
13
Finally, the deviant component of the protest paradigm was operationalized in another set of measures. Firstly, by measuring whether the party is called a new political party (Q3). This is a rather neutral frame, but it still ‘others’ the party, by putting it outside of the traditional political landscape. A second way of measuring this, related to the previous measure, asks whether DENK is called a (protest) movement instead of a political party (Q4), which sounds less serious than ‘party’, even though the party is running for parliament.
A third operationalization to measure the deviance of the party is Q12, that checks whether DENK is called a populist party or not, which again puts the party in a certain corner. A fourth variable is rather case-specific and focusses on the question whether the party DENK is being negatively connected to Turkey in any way, by for example mentioning the fact that the party refuses to acknowledge the Armenian genocide or take distance from the authoritarian measures by president Erdogan (Q8). Another case-specific operationalization of the protest paradigm was used in Q6, a variable to measure whether Sylvana Simons is mentioned or not, which provides the chance of looking into the question whether an article is only about Simons and not about the core issues of the party. H3: The majority of articles has an only negative or more negative than positive tone towards DENK was operationalized in a measure that coded the general tone of the article on a scale from 1 (only negative) to 5 (only positive) (Q2).
But to what extent do these frames have to be present, in order to talk about the protest paradigm dominating the newspaper coverage of DENK? The protest paradigm can only appear when the core issues of the party are not addressed at all (Q7), but any of the other variables is, as is reflected in the second hypothesis.
Reliability
Krippendorff’s Alpha-tests (Kalpha) were performed on 10 per cent (n=46) of the articles, comparing the coded results of the author and those of a second coder. A score of Kalpha=,8 or higher was considered desirable for a good reliability test (De Swert, 2012). Most of the variables tested had such a score, with only three variables resulting in Kalpha<,8. Q9 (Is DENK called anti-democratic?) gave a Kalpha=,333, while Q10 (Is the party described as authoritarian or dishonest in this article?) had a Kalpha=,500. As these two variables scored below the minimum of .60 as defined by De Swert, they were,
14
unfortunately, left out of further analysis.
Q8 (Is DENK negatively connected to Turkey in any way) scored also relatively low: Kalpha=,716. This was above the minimum, and therefore the variable was still analysed. An explanation for this lower Kalpha-score may lie in the fact that during evaluation with the second coder, it appeared that the author had not coded this variable positively when the article had mentioned ‘Turkish-Dutch’ citizens (which would be Q7), while the second coder had. Further evaluation showed that the codebook had been confusing, due to the nature of the term ‘is described as’ in Q10, that was considered too vague by the second coder. It remained unclear why Q9 did not give a satisfactory KALPHA score.
Limitations
Due to the size of the sample and some case-specific variables, this study is difficult to generalize to other Dutch parties and international contexts. Given the fact that this was a one-party case-study, comparisons are also hard to draw between the coverage of DENK and that of other parties, due to the fact that the aforementioned case specific variables were used (The mentioning of Sylvana Simons and the connection of DENK to Turkey). However, if these variables were to be left out of comparison, the findings of this paper can be used for such a comparison.
As in this paper the protest paradigm was used as a starting point for the codebook, most of the variables were conceptualized from this principle. This has resulted in a codebook that is rather negative, in fact only in the positive scores in the tone of the article-variable (Q2), there was space for a positive frame of DENK, which is not very specific. During the coding, it appeared that there were a couple of ways in which DENK was framed as ‘a welcome addition to the political landscape’, as ‘an asset’, as ‘giving voice to a minority that was not fully represented yet’ or as ‘an interesting democratic experiment.’ This is lacking from the analysis, which makes it more difficult to put the findings of this study into perspective. The reality of the articles would perhaps have looked different in this analysis when a broader scope of frames had been taken into consideration.
A second note has to be made on Q12, the question whether DENK was being called a populist party in the article. During coding, it appeared that this barely happened explicitly. However, the comparison between DENK and the right-wing party
15
of politician Geert Wilders (Freedom Party/PVV) was drawn multiple times, comparing DENK’s approach to mainstream media to the PVV’s press policy, for example. As such, the party was called a populist party implicitly a lot more than it appears now in the data. Also the party’s hostility towards certain media outlets and its refusal to admit some newspaper to their meetings was also not taken into consideration.
Results
In order to support H1: in most of the articles, the protest paradigm is present, the two sub-hypotheses need to be answered first. As can be seen in the table below, the variable that scored the highest t was the ‘DENK as an immigrant /multiculturalist party’-frame (Q7), which was used in the majority of articles analysed: 57,2%. The second most frequently appearing frame was the ‘new party frame’ (Q3), that appeared in 42,5% of the cases, followed by the frame of DENK as a Turkish party, which appeared in 35,4% of the articles. As the variable that measured the presence of the core issues of DENK in the article was found in the majority of the cases, sub-H1: The majority of the articles does not address DENK’s main agenda (migration, racism and discrimination), had to be rejected, as this was simply not the case.
FRAME FREQUENCY PERCENT
Q3: DENK as a new party 191 42,5
Q4: DENK as a (protest) movement 81 18,0
Q5: DENK as extremist/radical 26 5,8
Q6: Is Sylvana Simons mentioned? 110 24,5
Q7: DENK as anti-racism/discrimination/pro-migrant 257 57,2
Q8: DENK connected to Turkey 159 35,4
Q11: DENK lacking political experience 14 3,1
Q12: DENK being a populist party 12 2,7
Table 1: Frequencies of the variables (n = 449).
In order to find evidence for the second sub-hypothesis Sub-H2: All articles that do not
address DENK, use at least one frame from the deviant, threatening or impotent categories within the protest paradigm, we need to look into those cases in which Q7,
16
percentages of the occurrence of each variable in the articles that do not mention DENK’s core issues. In 47 of these cases (24%) none of the frames of the frames of the protest paradigm was used. Therefore, the second sub-hypothesis should be rejected as well, as it was only in 76% that at least one frame from the protest paradigm was used, and not in all of them, as the hypothesis expected. This automatically leads to the rejection of H1 as a whole as well, as both the two sub-hypotheses were rejected. Only in 144 out of 449 articles, DENK’s core issues were not mentioned, but one of the protest paradigm-frames was used instead. The protest paradigm was therefore only present in 32% of the articles, which does not approach the majority of articles at all.
Variable Amount Percentage
Q3: DENK as a new party 76 39,6
Q4: DENK as a (protest) movement 24 12,5 Q5: DENK as extremist/radical 3 1,6 Q6: Is Sylvana Simons mentioned? 32 16,7 Q8: DENK connected to Turkey 60 31,3 Q11: DENK lacking political experience 7 3,6 Q12: DENK being a populist party 4 2,1
Table 2: Frequencies of protest-paradigm frames when Q7 = 0 (n=192)
Even though the protest paradigm was not found present in the majority of articles, the second hypothesis H2: Most articles in which the protest paradigm is used, employ the
‘deviant’ and ‘impotent’ frame, rather than the ‘threatening’ frame, can still be tested.
Overall, two frames that fit into the ‘deviant’-set of frames in the protest paradigm, occur the most often: Q3 (new party) appears in 42,5% of the cases, while Q8 (Turkey-connection) comes in second with 35,4%. This, however does not tell us anything yet, as these numbers still include articles that do address DENK’s core issues and as such cannot use the protest paradigm. Table 4 shows the amount and percentage of articles that did not address DENK’s core issues (Q7), but that did use one of the other frames.
As the table shows, there is no difference in the ranking of cases in which DENK’s main agenda is not mentioned, but another frame is, and the general outcome of the analysis (see table 3). There is a high degree of overlapping, as together these variables together have a cumulative percentage of 107,4%. Q3, DENK as a new party, still is the highest (39,6%), followed by DENK being connected to Turkey (31,1%). The other higher scores also fit into the ‘deviance’ category, namely Q4: DENK as a protest movement (12,5%) and the mentioning of Sylvana Simons (16,7%). The impotence and threatening components of the protest paradigm score far
17
lower. Therefore, H2: Most articles in which the protest paradigm is used, employ the ‘deviant’ and ‘impotent’ frame, rather than the ‘threatening’ frame is supported.
Figure 1: Evaluation/tone (n=449)
When it comes to the third hypothesis, H3: The majority of articles has an only negative or
more negative than positive evaluation towards DENK, the bar chart in figure 1 above shows
that 33,2% of the articles was coded in balance, whereas in 25,8% of the articles, no evaluation was made about DENK whatsoever. 9,8% of the articles had an only negative evaluation, whereas in 25,8% of the articles, DENK was evaluated more negatively than positively. The positive evaluations score (far) lower: 6,5% evaluated DENK more positively than negatively, whereas only 2,2% had a purely positive evaluation of the party. As for the entire population (n=449), the majority of articles was coded either balanced or not evaluating DENK, H3 had to be rejected as well. Lee’s (2014) claim that the protest paradigm supports the status quo by negatively framing protest movements does not seem to work for DENK.
Discussion
Even though only one out of the three hypotheses of this study is supported by this analysis, the protest paradigm was still found in roughly a third of the articles coded: 32% did not mention DENK’s main agenda, but used a frame from the protest-paradigm instead. Within those articles, the frames that fit into the deviant-category of the paradigm were found most often. In order to find possible explanations for this, we need to zoom out somewhat and go back into the hierarchy of influences-model (Shoemaker and Reese, 2014).
According to Reul et al., the reasons for the protest paradigm to occur can largely be found on the level of journalistic routines within this model. If the application of the protest
18
paradigm on new political parties is to be followed, this would also result in ‘a new political party’ being framed as being impotent, dangerous or deviant, simply because this is ‘the way’ professional newsrooms deal with new political parties. This explanation however ignores the fact that the core issues of DENK involve the multicultural society. Multiculturalism is generally seen to determine recognition of cultural diversity, even though the normative implications for thinking about societies consisting of cultures vary in different parts of the world (Parekh, 1997). DENK has taken multiculturalism as one of its core ideals, strongly advocating for (ethnic) minority rights and the acceptance of migrants having, for example, multiple passports. This is also recognized by the majority of newspaper articles that mention the multiculturalist agenda of DENK.
Multiculturalism as an ideology directly challenges the notion of journalism operating outside society, and impacts on all levels of editorial decision-making (Cottle, 2000). If this was to be translated to the hierarchy of influences-model again, the levels that would be affected would be respectively the individual, routines and organizational levels, as these are the levels editorial-decision making takes place on (Shoemaker and Reese, 2014). Following this line of argument, the fact that more articles had a negative than a positive tone towards DENK and the fact that a considerable amount of articles did not address the party’s core issues can be derived to the individual, routines and organizational levels.
A fundamental question here is the inclusiveness of the journalistic organizations, that affect also the routine- and, obviously, individual level. When the individual journalist is ethnically Dutch, the issues that DENK wants to put on the agenda are harder to understand. When it is the newspaper’s routine to pay more attention to issues that attract the ethnically Dutch population, again DENK’s issues are framed in a certain way (as ‘deviant’). This also applies to the organizational level: if the newsroom does not include journalists that have a non-Dutch background, DENK will always be portrayed from the perspective of the ethnically non-Dutch journalist.
The inclusiveness of the predominantly white, male newsrooms has been problematized in multiple ways (Deuze, 2005). When one follows the modernist principle of the journalist as a provider of public service (namely information) that helps the electorate to take part in democracy, news organizations have to provide an representation of society, in which non-whites or migrants are not represented as ‘the Other’ (Hallinn, 1992; Hall, 1982).
Besides of the new party frame, ‘DENK as a Turkish party’ also scores high, which can be classified as both being a ‘danger’ and ‘deviant’ within the protest paradigm. A possible explanation for this, when following Hallinn and Hall’s argumentation, is the abovementioned fact that most Dutch news organisations employ predominantly white staff. As such, the migrant-issues of DENK are by definition ‘other’, and are seen by the individual journalist and
19
the organization as ‘deviant’ or ‘outsider’, due to a lack of identification.
If this explanation is to be followed, questions can be raised about the effectiveness of using the protest paradigm for explaining newspaper coverage of DENK: even when the core issues of the party are discussed, the party is still framed as ‘deviant’ or ‘other’ by ethnically Dutch journalists due to its multiculturalist agenda. The assumption that the protest paradigm can be directly applied to new political parties needs to be amended, at least in the case of DENK. Due to the ethnic or multicultural dimension of DENK, it is rather difficult to speak about the occurrence of the protest paradigm frame, as the party by its nature is already framed as ‘other’. However, another explanation can also be found on the ideological and extra-media levels, that correlate with the abovementioned issues regarding the inclusiveness of the newsroom. These both coincide with the events in Turkey in 2015-2017, in which president Erdogan repressed the free press and became a more authoritarian ruler. Discussions about ‘Turkey’s long arm’ infiltrating other countries were held regularly in the media: this term also appeared in many of the articles about DENK. This extra-media force may have shaped the framing of the party, as it refused to take distance from Erdogan’s policy. The same is true for the ideological level; as Erdogan’s authoritarian policies do not fit into the parliamentary democracy’s ideals in the Netherlands, this also influences the coverage of DENK, as the Dutch media operate within the democratic corporatist media-system (Hallinn and Mancini, 2004).
Conclusion
This paper started off with the main research question to what extent the protest paradigm was present in the Dutch press coverage of DENK. For the first time, the protest paradigm was applied to the framing of a new political party. The answer to the main research question is rather straightforward: the protest paradigm was present in 32% of the articles on DENK that appeared in the Dutch media, when the presence of the protest paradigm is defined as not mentioning the party’s main agenda but using at least one frame from the protest paradigm’s set of frames instead. Two out of the three hypotheses had to be rejected: it was not the majority of articles that used the protest paradigm (H1) nor was the majority of the articles more negative than positive or only negative towards the party (H3). Most of the articles used a frame that fitted into the ‘deviant’ category of the protest paradigm, which led to the support of H2.
This paper has provided valuable insights on the effectiveness of using this set of frames for framing analysis of new political parties. The findings of this paper are not easily applicable to other Dutch new political parties, due to the fact that it was a one party-case study with some specific characteristics, such as the multiculturalist component of the party’s agenda. However,
20
future studies that will apply the set of frames of the protest paradigm to other new (Dutch) political parties, will make it possible to compare the extent to which DENK was framed within the protest paradigm to the other parties. This will allow for the identification of similarities and differences, through which a more final answer can be given to the question to what extent the protest paradigm can be applied to the realm of new political parties. As such, this paper has contributed to the formulation of a new theory of media framing of new political parties. Another future study could shed light on what would describe the way in which DENK was framed in the Dutch media the best, when the protest paradigm does not seem to be fully useful for the party. A new quantitative content analysis could test a set of frames that are more in line with the theory of ‘Othering’ which was rather present in this study, and also take into consideration the limitations discussed earlier: due to the fact that the protest paradigm was taken as a starting point for the codebook, there was no space for positive evaluations of DENK. This new analysis could paint a fuller picture of the party’s coverage and framing. Finally, comparative studies between this paper’s findings on DENK and other pro-migrant parties within Europe or worldwide could also contribute to the formulation of a more effective theory to explain the dynamics in newspaper framing of such parties.
21
ACKNOWLEDGEMENTS
Writing a M.A. thesis is a tough job, especially when one can only work under pressure and chooses to use a methodology that is new and, in a way, terrifying. Therefore, I would like to thank my supervisor Joost van Spanje wholeheartedly, for supporting and guiding me during the intensive process, taking the time to answer my questions, stimulating me to experiment and pushing me to deliver my work on time. Besides of academical support, emotional support is of equal importance when one finds oneself struggling with numbers and concepts and SPSS-sheets. My dear Steven, love of my life, thank you for tolerating my temper and providing all help possible. Many thanks go to my amazing and inspiring friends that were never hesitant to advise, listen, moan or to bring me a coffee when very needed, in no particular order Hannah, Ali, Aimée, Mari, Pablo, Lisette, Maris-Ka, Charlotte, Anne, Djuna, Kims, Bart and Pia. My final thanks go to my dear parents, for supporting my academic development and always being interested in what I do. THANK YOU.
22
List of references
Bardoel, J. (1996). Beyond Journalism. A Profession between Information Society and Civil Society, European Journal of Communication 11(3), 283-302.
Baumgartner, F., & Chaqués-Bonafont, L. (2015). All News is Bad News. Newspapers Coverage of Political Parties in Spain. Political Communication, 32(2), 268-291.
Beyens, S., Lucardie, P., Deschouwer, K. (2015). The Life and Death of New Political Parties in the Low Countries. West European Politics, 8 August 2015, 1-20.
Bolleyer, N. and E. Bytzek, 2016: New Party Performance after Breakthrough: Party Origin, Building and Leadership, Party Politics, doi:10.1177/1354068815626604
Boomgaarden, H. G., & Vliegenthart, R. (2007). Explaining the Rise of Anti-Immigrant Parties: The Role of News Media Content in the Netherlands. Electoral Studies,
26(2), 404-417.
Bos, L., Brug, W. van der, & De Vreese, C. H. (2011). How the Media Shape Perceptions of Right-wing Populist Leaders. Political Communication, 28(2), 182-206.
Brüggemann, M., Engesser, S., Büchel, F., Humprecht, E., & Castro, L. (2014). Hallin and Mancini Revisited: Four Empirical Types of Western Media Systems. Journal of
Communication, 64(6), 1037-1065.
Brants, K., De Vreese, C.H., Möller, J., & Praag, P. van (2010). The Real Spiral of Cynicism? Symbiosis and Mistrust between Politicians and Journalists. The International
Journal Press/Politics, 15(1), 25-40.
Chong, D., & Druckman, J. N. (2007). Framing Theory. Annual Review of Political
Science, 1(10), 103-126.
Cottle, S. (2000). Media research and ethnic minorities, in: S. Cottle (ed.), Ethnic
Minorities and the Media: Changing Cultural Boundaries pp. 1-30. Ballmoor: Open
University Press.
DENK (2016), Denkend aan Nederland, Verkiezingsprogramma DENK 2017-2021. Retrieved from: https://www.bewegingdenk.nl/
wpcontent/uploads/2016/11/Verkiezingsprogramma_DENK_2017-2021.pdf De Vreese, C. (2012). New Avenues for Framing Research. American Behavioural
Scientist, 56(3), 365-375.
23
Krippendorff’s Alpha, University of Amsterdam, retrieved from:
http://www.polcomm.org/wp-content/uploads/ICR01022012.pdf
Deuze, M. (2005). What is Journalism? Professional Identity and Ideology of Journalists Reconsidered, Journalism 6(4), 442-464.
Du Pré, R. (2016). Tweede Kamer Opent Aanval op Zetelrovers zoals DENK, de
Volkskrant, 22 June 2016.
Ellian, A. (2016). DENK Wil Burger Heropvoeden. Waar Zagen We Dat Eerder? Elsevier,
16 November 2016.
Engler, S. (2015). Corruption and Electoral Support for New Political Parties in Central and Eastern Europe. West European Politics, 22 October 2015, 1-27. Entman, R. (1993). Framing: Towards a Clarification of a Fractured Paradigm, Journal of
Communication 43(4), 51-58.
Ferree, M. M., Gamson, W., Gerhard, J., & Rucht, D. (2002). Four Models of the Public Sphere in Modern Democracies. Theory and Society, 31(3), 289-324.
Goffmann, E. (1974). Frame Analysis: an Essay on the Organization of Experience Cambridge MA: Harvard University Press.
Grotz, F. and Weber, T. (2016). ‘New parties, Information Uncertainty and Government Formation: Evidence from Central and Eastern Europe’, European Political Science
Review 8(3), 449-472.
Hall, S. (1982). The Rediscovery of Ideology: Return of the Repressed in Media Studies, in: Gurevitch, M., Bennet, T., Curran, J. and Woollacott, J. (eds.), Culture, Society
and the Media, pp. 56-90. London: Methuen.
Hallinn, D.C. (1992). The Passing of the ‘High Modernism’ of American Journalism,
Journal of Communication 42(3), 14-25.
Hallin, D. C., & Mancini, P. (2004). Comparing Media Systems: Three Models of Media and
Politics. Cambridge: Cambridge University Press.
Hänggli, R. (2012). Key Factors in Frame Building. Model Specification. American
Behavioral Scientist, 56(3), 300-317.
Harcup, T. and O’Neill, D. (2001). What is News? Galtung and Ruge Revisited, Journalism
Studies 2(2),
Henson, S.E. and Kopstein, J.S. (2005) ‘Regime Type and Diffusion in Comparative Politics Methodology’, Canadian Journal of Political Science 38 (69-99).
24
News Coverage: How Visibility and Tone Influence Party Choice. Political
Communication, 27(4), 389-405.
Hug, S. (2000), Studying the Electoral Success of New Political Parties: A Methodological Note, Party Politics 6(2), 187-197.
Kiesraad (2017), Officiële Uitslag Tweede Kamerverkiezingen 15 Maart 2017, 21 March 2017, retrieved from:
https://www.kiesraad.nl/actueel/nieuws/2017/03/20/officiele-uitslag-tweede-kamerverkiezing-15-maart-2017
Krouwel, A., & Lucardie, P. (2008). Waiting in the Wings: New Parties in the Netherlands. Acta Politica, 43(2-3), 278-307.
Kuhn, R. and Nielsen, R.M. (2014). Political Journalism in Western Europe: Change and Continuity, in: Kuhn, R., and Nielsen, R.M. (eds.), Political Journalism in Transition:
Western Europe in a Comparative Perspective (pp. 1-26). London: I.B. Tauris & Co.
Ltd.
Lago, I. and Martinez, F. (2011), Why New Parties? Party Politics 17(1), 3-20.
Lee, F.L.F. (2014), Triggering the Protest Paradigm: Examining Factors Affecting News Coverage of Protests, International Journal of Communication 8, 2725-2746. Li, H. (2016), Mediating the Message in the 21st Century (Book review of the third
edition), Asian Journal of Communication 26(3), 298-300.
Mazzoleni, G. (2003). The Media and the Growth of Neo-populism in Contemporary Democracies. In: G. Mazzoleni, J. Stewart, & B. Horsfield (Eds.), The Media and
Neo-populism: A Contemporary Comparative Analysis (pp. 1-20). Westport, CT:
Praeger.
McLeod, D. (2007). News Coverage and Social Protest: How the News Media’s Protest Paradigm Exacerbates Social Conflict, Journal of Conflict Resolution vol. 2007 (1),
185-194.
Missfelder, P. (2013). Young People’s Engagement In ‘New’ Parties, European View
12(2), 243-248.
Muller-Rommel, F. (ed.), New Politics in Western Europe, The Rise and Success of Green
Parties and Alternative Lists, Boulder CO: Westview Press.
Nieuwsuur (2017), Denk: Anti-discriminatie, Opstandig en op het Randje, 28 February 2017, retrieved from: http://nos.nl/nieuwsuur/artikel/2160533-denk
25
Patterson, T., & Donsbach, W. (1996). News Decisions: Journalists as partisan actors. Political Communication, 13(4), 455-468.
Pan, Z. and Kosicki, G.M. (1993). Framing Analysis: an Approach to News Discourse,
Political Communication, 10(3), 55-75.
Praag, P. van (2005). De Veranderende Nederlandse Campagnecultuur. In K. Brants & P. van Praag (Eds.), Politiek en Media in Verwarring: De Verkiezingscampagnes in het
Lange Jaar 2002 (pp.21-43). Amsterdam: Het Spinhuis.
Reese, S. (2007). Journalism Research and the Hierarchy of Influences Model: A Global Perspective, Brazilian Journalism Research 3(2), 29-41.
Reul, R., Paulussen, S., Raeijmakers, D., Van der Steen, L., Maeseele, P. (2016). Professional Journalistic Routines and the Protest Paradigm: The Big Potato Swap in Traditional and Alternative Media, Journalism, 8 March 2016, 1-18.
Shoemaker, P. and Reese, S. (2014). Mediating the Message in the 21st Century, a Media Sociology Perspective. New York: Routledge.
Sikk, A. (2012). Newness as a Winning Formula in New Political Parties. Party Politics,
18(4), 465-486.
Strömbäck, J. (2005). In Search of a Standard: Four Models of Democracy and Their Normative Implications for Journalism. Journalism Studies, 6(3): 331-45
RTL Nieuws (2015), ‘Hadden Jullie Ook Zo’n Moeite met dat XXL-stembiljet?!’, 15 March 2017, retrieved from: https://www.rtlnieuws.nl/opmerkelijk/hadden-jullie-ook-zon-moeite-met-dat-xxl-stembiljet
Selb, P. and Pituctin, S. (2010). Methodological Issues in the Study of New Parties Entry and Electoral success, Party Politics 16(2), 147-170.
Tuchman, G. (1978). Making News: A study in the Construction of Reality New York NY: Free press.
Vraga, E. K., Carr, D. J., Nytes, J. P., & Shah, D. V. (2010). Precision vs. Realism on the Framing Continuum: Understanding the Underpinnings of Message Effects. Political Communication, 27(1), 1-19.
Weaver, D.A. and Scacco, J.M. (2012), Revisiting the Protest Paradigm: the Tea Party as Filtered Through Prime-time Cable News, The International Journal of
Press/Politics 18(1).
White, D. (1950). The ‘Gate Keeper’: A Case Study in the Selection of News, Journalism
26
Zelizer, B. and Allan, S. (2012), Keywords in News and Journalism Studies, Maidenhead and New York NY: Open University Press, McGraw Hill.
Zons, G. (2015). The Influence of Programmatic Diversity on the Formation of New Political Parties. Party Politics 21(6), 919-929.
27
Appendix 1: Codebook
CODEBOOK
Quantitative content analysis
Newspaper framing of DENK
Third versionApril 27, 2017
J.J. Beentjes
Universiteit van Amsterdam
Erasmus Mundus M.A. Journalism, Media and Globalization Master’s thesis
The articles for this study were gathered through the online-database LexisNexis, using the following search-term: (((("DENK") AND ("Kuzu") OR ("Politiek"))) and Date(geq(09/02/2015) and leq(15/03/2017)))
DEFINING: Newspaper article
A newspaper article is defined as journalistic, non-advertorial content of one of the following Dutch national newspapers: NRC Handelsblad, de Volkskrant, De Telegraaf, Trouw and
Algemeen Dagblad. There is no minimum or maximum length for the articles, that were gathered through Lexis Nexis, from 9 February, 2015 to 15 March, 2017.
Q1: Is the political party DENK mentioned in this article?
0 No (stop coding) 1 Yes (continue with Q2)
NP1 Date (format: 20170315) NP2 Newspaper:
28 1. NRC Handelsblad 2. De Volkskrant 3. Algemeen Dagblad 4. De Telegraaf 5. Trouw NP3: Type of story
1. News story (fact based report with the basic questions about the event answered: what, who, when, where, how)
2. Commentary/editorial (column. Opinion-based piece by a journalist)
3. Editorial comment (an opinion piece by the editor-in-chief of the newspaper, often commenting on a particular incident or the previous week)
4. Background-story (an analytical piece providing background information on the news, that goes beyond the factual news)
5. Interview (a pure question/answer-based piece)
6. Reportage (coverage of an incident where the journalist was present in real-time) 7. Portrait (non-news based portrayal of a person, organization or location)
8. Announcement (a short announcement of an article to be found at a different place in the newspaper)
9. Photo/infographic (only a picture/infographic with a description) 10. Opinion piece (longer opinion piece)
11. Letter (short letter to the newspaper, often called ‘Lezersbrief’) 12. Other
Q2: Is the political party DENK explicitly evaluated in this article in any way (in quotes, commentary, analysis, paraphrases etc.)? If so, how?
0: Unclear/no positive or negative evaluations
1. Only negative
2. More negative than positive 3. Neutral
4. More positive than negative 5. Only positive.
29 0. No 1. Yes
Q4: Is the party called a protest party in this article?
0. No 1. Yes
Q5: Is the party called extremist or radical?
0. No 1. Yes
Q6: Is the politician Sylvana Simons mentioned in this article?
0. No 1. Yes
Q7: Is DENK’s core agenda mentioned? (I.e.: is the party being connected in the article towards multiculturalism, migration, (anti)-racism, (anti)-discrimination or diversity?)
0. No 1. Yes
Q8: Is DENK being connected to Turkey in any way (i.e. is the Armenian genocide mentioned, president Erdogan, the coup of 2017 mentioned or the metaphor ‘the long arm of Turkey’ used?)
0. No 1. Yes
Q9: Is DENK called anti-democratic?
0. No 1. Yes
Q10 Is DENK described as authoritarian or dishonest in this article?
0. No 1. Yes
30 0. No 1. Yes
Q12 Is DENK called a populist party?
0. No 1. Yes
Title
