University of Amsterdam
Populism on Social Networking Sites: Bulgarian
politicians during the 2014 European Parliament
Election Campaign
Lora Karamfilova (10608745) Email: lora.karamfilova@gmail.com
Supervisor: Dr. Sanne Kruikemeier Graduate School of Communication
M.Sc. Political Communication June, 27th, 2014
1 Abstract:
This article addresses the presence of traits of populism on the social media platforms Facebook and Twitter. The time period analysed is the campaign for the 2014 European parliament elections. The studied political subjects are four Bulgarian parties and their leaders. Subsequently, the focus is on the empirical evidence gathered through a manual content analysis of Facebook and Twitter posts during the campaign. As a result, it was discovered that indeed the populist parties and their leaders are more present and are sharing more negative content online towards the governing elite than conventional parties do. However, when referring to the average people, the two sets of political actors did not have significant differences. Neither were they very distinct in relation to the party leader being represented as charismatic men. An essential finding was the bold contrast in the amount of online activity between the studied subjects, which led to an overrepresentation of the soft populist leader of party GERB Boyko Borissov.
2 Table of Contents
Introduction ... 3
Theoretical framework ... 5
What is populism? ... 5
Social media and populism ... 9
The case under study: Populism in Bulgaria ... 11
Methodology ... 12
Content analysis and code book ... 12
Inter coder reliability ... 13
Results and Findings ... 14
Discussion and Implications ... 20
Limitations of the Study ... 24
Conclusions ... 25
Appendix A Inter coder Reliability ... 26
Appendix B Code Book ... 26
Appendix C Screen shots ... 31
3
Introduction
There is an existing consensus among social scientists that when it comes to electoral success populist political leaders are highly dependent on the media coverage they receive (Bos et al, 2011). However, up until now, not enough attention has been paid to the specific ways parties and politicians represent themselves through the use of social networking sites. This topic has been rather neglected not only by academics in political and communication science, but also by sociologists and policymakers. To the contrary, there have been extensive findings in regards to traditional media channels such as newspapers and the manners by which they portray populists (e.g. Bos et al, 2010; Elrich, 2005).
Some recent empirical evidence suggests that online media is particularly effective when populist messages are planned to be sent in a direct way straight to the voters (Groshek & Engelbert, 2012). Moreover, statements that appear on social media platforms such as Facebook, Twitter or YouTube are capable of carrying the primary ideas of political actors and those posts are not merely interpreted by journalism gatekeepers. As opposed to the print and broadcast media outlets, social networking sites give a so-called unfiltered and
unregulated version of the political talk that circulates in a given society between politicians and citizens.
The purpose of this research paper is to study the usage of populist styles on social networking sites. This will be achieved through the gathering of up-to-date relevant factual data on populist political rhetoric on two social media platforms. More specifically, it will be analysed whether Bulgarian politicians use populist rhetoric on Facebook and Twitter during the 2014 European parliament election campaign. Taking into account that on social media politicians have the powers to personally choose the format and length of their texts and the accompanying visuals, the results will add valuable input when assessing the
4
Looking through the prism of Bulgaria, a country with a relatively short democratic history, this case study makes the research inductive and unique in nature. After a year-long protests and social instability in 2013, the European parliament elections are the first
campaign opportunity for Bulgarian politicians to communicate to their nation. Many former communist countries in Central and Eastern Europe are still struggling to adapt new liberal governing practices, which in the end turns into a fertile ground for the emergence of populist parties and a more emotional and provocative political talk (Zaslove, 2008; Cholova and de Waelle, 2011). External socio-economic factors such as globalization, technological advances, diplomatic relations with international institutions and low incomes are undoubtedly adding up to the dynamic landscape in those countries. Hence, the findings of this paper will not only contribute to the theoretical background on the concept of populism, but will also shed light on the political arena in a “young” democratic European member-state.
In order to navigate the research, firstly, a broader perspective will be applied, covering the overall content of the rhetoric. The research question is formulated as follows: Research question 1: To what extent do Bulgarian politicians apply populist discourse in their Facebook and Twitter messages during an election campaign?
Second of all, a comparison will be made between parties that have already been defined as populists and the conventional structures. For many years now, social science scholars have questioned whether populism is merely linked to the radical right-wing extremist actors that are on the rise in Europe, or if it has also spread to the mainstream political subjects (Zaslove, 2008, p.320). Common social issues across the old continent such as high taxes, immigration, health care and crime are fruitful sources for a profuse populist rhetoric not only for
extremists, but for all of the political groups thriving to obtain more power and popularity among the electorate. Hence, the contrast will be drawn in order to see whether populist traits have become omnipresent in the online rhetoric of politicians.
5
Research question 2: How does the political discourse on Facebook and Twitter differ between populist parties and mainstream parties?
In the following section the concept of populism will be positioned within the existing theoretical framework, by looking at the findings and debates of scholars in the political communication field.
Theoretical framework
What is populism?
The concept of populism is not a novelty in the field of political science. Nevertheless, it is still provoking severe disagreements among academics whether populism is a political ideology, a certain mentality or simply a rhetorical style (Zaslove, 2008, p.322). This societal phenomenon is a subject for investigation across the globe, now more than ever due to the blurring nature of political parties, economic stagnation and the rapid rise and unexpected electoral success of extreme right-wing nationalistic parties.
Subsequently, many definitions of populism have been provided in respect to the political, communication and social connotations of the word (e.g. Canovan, 2004; Laclau, 2005; Arditi, 2010; Barr, 2009; Hawkins, 2003). Given the research questions outlined in the previous section, here the focus is placed on populism as a political discursive style and henceforth a communication tool. Cholova and De Waelle (2011) outline the main traits of populism according to the behaviour of politicians in the media. Some establishing criteria include how often political actors are present in news slots or political talk shows, if they have a personal contact (in the sense of a discussion) with the voters, if they adapt negative rhetoric towards opponents and whether they offer short-term solutions to problems without any clear vision for the future (p.26).
6
Typically, communication researchers have analysed the way politicians address the big issues in their societies in a populist manner by studying the way they speak or behave with the voters. Laclau (2005) builds a discursive model of populism which is not limited merely to verbal language and linguistics, but also includes the different set of interactions that politicians enact with their colleagues, opponents and with the electorate. The author argues that through the building of an imaginary fence within a given society, populists tend to divide the population into two groups, the common people and the powerful elite.
Narrowing this theory down, the conceptualization of Jagers and Walgrave (2007) is considered applicable for the purpose of this paper as it looks closely at the content of the discourse interrelated with the main subject of populism- the people:
“…a political communication style of political actors that refers to the people” (p. 322)
In today’s media-dependent political sphere the core link between electorate and politicians is the media (Blumer and Kavanagh, 1999). Thus, political subjects are highly subordinate to how the “fourth estate” is picturing them to the wider audience, and hence, whether this image will lead to electoral success or not. Some scholars (e.g. Bos et al., 2011; Gurov & Zankina, 2013) argue that this mediatisation has also affected the way populism transformed towards a more personalised and emotional political discourse in general. The interconnectedness of populism and the traditional media has attracted substantive academic attention (Bos et al 2011; Elrich, 2005; Burack & Snyder-Hall, 2012). Heated television debates and tabloid-framed political scandals are not a novelty in the modern campaigning programs.
The more a politician is visible in the media, the more his or her image becomes a part of the public’s agenda. In order to be successful in elections, the party members and
7
trustworthy and legitimate (Bos, 2011). A vivid example of a populist leader with limitless air time is the former Venezuelan president Hugo Chávez. His television show “Alò Presidente!” became a case study and gave substantive data for the political scholars to investigate how populist leaders talk to their supporters. Plentiful media coverage, emotional speeches and disclosure of personal information were just few of the techniques Hugo Chávez used to gain electoral support (Hawkins, 2003). The president positioned himself within the social class of the average Venezuelans by emphasizing his Indian roots that make him a common man. Hugo Chávez succeeded in several election campaigns also because of his regular appearance on the show “Alò Presidente!” sometimes continuing for more than four hours. Thus, in this paper it is expected for the populist parties and their leaders to be more present on social media than conventional actors (H1).
One of the most prominent discursive techniques of populist leaders deliberated by social scientists who have studied this phenomenon is the “us versus them” approach (Cholova and de Waelle, 2011). Scholars have identified this method as one which has the aim to implicitly divide the society into two homogeneous groups and then oppose them to one another (Mudde 2004; Schedler, 1996; Bos, 2011). The first group is the common people, or those who deserve better governance and transparent rule and the other group is the
governing elite. Jagers and Walgrave (2007) discuss this as a “communication frame” which enables the positioning of the populist politicians in the role of the chosen ones, speaking in the name of the people. Furthermore, Zaslove (2008) defines the average people as
‘non-plural, virtuous, and homogeneous groups that are part of the ‘everyday’ and the ‘normal’ core of the country, or the region in question’ (p.323).
In this sense, the populist parties and leaders turn out to be saviours of the majority who simultaneously differentiate themselves from the dominant class and any extremist groups. Populist leaders also promise, if elected, to quickly improve the living conditions for
8
the common men and women, usually without unambiguously giving suggestions as to how this would happen (Groshek & Engelbert, 2012, 184). Moreover, populists tend to speak on behalf of the silent homogeneous majority of the nation (Zaslove, 2008). By using metaphors and allegory to refer to the people as a whole, those politicians rarely explicitly state who and what they represent. Consequently, the populist parties and their leaders are expected to refer more often to the average people in their online social media discourse than the conventional political subjects (H2).
On the other hand, Zaslove (2008) suggests that the “dangerous others” are not only the government executives, but could also stand for political competitors or large economic corporations for instance (p. 323). Usually the elite groups are blamed to be responsible for all the misfortunes that occurred to the general population (Jagers and Walgrave, 2007, p. 324). For this reason, it is expected that the populist politicians will use more negative content on Facebook and Twitter towards their opponents compared to the mainstream leaders (H3). More concretely, the nationalistic extreme-right party is assumed to have more messages against the governing elite than the centre-right soft populist party included in this study (H3a).
Another important aspect of the conceptualization of populism is the charisma of the party leaders (Cannon, 2008; Barr, 2009, Hawkins, 2003). Cannon (2008) argues that populist leaders are mainly perceived by the society as “macho men, with an autocratic, authoritarian bent.” (p.740). A special niche in the research field is reserved for right-wing extreme party leaders. Bos (2011) states that those parties are best known and become the centre of media attention mainly because of the charismatic person around which the party structure is organized and not due to the program of the group (p. 184). Thus, it is hypothesized that the leaders of populist parties will be more often depicting themselves online as macho men, compared to their conventional counterparts (H4).
9
In sum, it is considered that populism is a communication style adopted by political actors in a certain setting. The overall scientific aim of this article is to test whether this conceptualisation works in an empirical context with the major incentives being social media platforms and election campaigns.
Social media and populism
So far, academics have concentrated on the traditional media’s portrayal of populist leaders, which consequently is a product selected and defined by journalists, editors and news producers, just to name a few. This media product is packaged in a certain way in order to be “sold” to the consumers, or audiences, under the pressure of deadlines and ratings (van der Wurff & Schonbach, 2011, p.416). Social networking sites such as Facebook and Twitter give political figures the power to freely address their electorate without any framing on behalf of the media industry and its gatekeepers. Most importantly, it enables a more personal and horizontal interactive communication between politicians and citizens (Kruikemeier, 2014, p.10).
Considering populism as anti-elitist, promoting direct democracy, and focused on the presumed homogeneous will of the average citizens (Abts and Rummens, 2007), recent developments in the communication technologies fit these prerequisites better than traditional media such as newspapers or radio. Particularly taking into account online media’s power to overcome time, anonymity and distance restraints, now participation in Habermas’s public sphere is easier than ever. The mobilizing effect of election news online has received a great attention by political communication scholars regarding political involvement and voting turnout (e.g., Quintelier & Vissers, 2008; Scheufele & Nisbet, 2002; Shah, McLeod, & Yoon, 2001; Kruikemeier, 2014). Specifically in relation to the young voters, Bakker and de
10
elections than traditional media do. Therefore, the possibilities for democratic discourse and legitimization of politicians have spread beyond television and newspapers and entered a somewhat unregulated online arena of communication that can directly connect politicians with citizens (Zaslove, 2008). Social networking sites have given parties and candidates the extraordinary opportunity of self-representation.
Sharing opinions, posting pictures and videos, organizing events and entering debates with the voters are just several of the main tools which politicians could use online. By intertwining user-generated content and populist behaviour, Groshek and Engelbert (2012) argue that social networking sites are crucial for the direct reach for voters and stimulation of positive media coverage for those politicians. Their cross-national comparative research of the Dutch PVV and the US Tea Party shows that online media facilitate and expand the
possibilities of these populists to both distance themselves from the elite and the extremists and at the same time gives them the power to influence the political order and connect with people. In the context of an electoral campaign, social media platforms not only have started to play a crucial role for persuading voters, but they have become a rather standard practice in political campaigning (Kruikemeier, 2014, p.11). This is why, from an academic point of view, it is essential to look more in depth at the relation between social media platforms and populism in the context of a campaign.
The concrete goal of the current study is to see how populist and mainstream parties and leaders in Bulgaria address the electorate through Facebook and Twitter. The content of the social media pages will be studied through quantitative content analysis. Nevertheless, the research will not only take into account parties that are already categorized as populist, but rather it will look at the four major parties in Bulgaria. They are composed of one soft populist party, one right-wing extremist nationalistic party and two mainstream parties. The latter two also form the current government coalition in the country.
11 The case under study: Populism in Bulgaria
The parties included for analyses are the first four in the Bulgarian national parliament according to the seats they won in the last general elections in May 2013. The nationalistic Ataka party and the soft populist party Citizens for European Development of Bulgaria GERB are opposed to two mainstream structures established directly after the fall of the communist regime in 1989 (Bulgarian Socialist Party and Movements for Rights and Freedoms). Even though BSP and the MRF are not defined as populists, it is crucial to compare them to Ataka and GERB as during election campaigns even the old-established parties tend to apply populist rhetoric (Cholova and De Waelle, 2011; Pitrova, 2007). Given that the ultimate goal of all of the four political subjects is electoral success, it could be presumed that diverse range of campaigning styles will be applied with the purpose of attracting voters.
Nevertheless, it has to be stressed that both political party Ataka and GERB do not fully meet one of the popularly accepted characteristics of populist formations- to be a newly established political entity (Bos et al., 2011). The radical right-wing group was formed in 2005 and it bases its political rhetoric on nationalist, xenophobic and anti-EU integration statements (Stefanova, 2009). On the other hand, Citizens for European development of Bulgaria was established in 2006 and is characterized as a “soft” populist or “European” populist party (Novakovic, 2010; Gurov & Zankina, 2013). Boyko Borissov’s simplistic political discourse directed towards the average citizens, together with his looks of a strong, alpha male, contributed to the perception of him as a populist leader who is “one of the
ordinary people” (Smilov, 2011, p.23). Novakovic (2010) points out that a core element of the 2009 campaign of Boyko Borissov was the anti-establishment rhetoric. In the current
government, GERB and Ataka are playing the role of opposition to the coalition cabinet of MRF and BSP. Thus, we expect the soft populist Borissov and the radical nationalist Siderov to apply more populist rhetoric in comparison to the conventional BSP and the MRF.
12
Methodology
For this study a narrowed sampling frame was built in order to obtain the most
objective and yet precise sample of social media messages of politicians. Given the digital and boundaries-free character of this medium and its tight interconnectedness to other media channels, it was essential to accurately determine what kind of content will be analysed and is to be excluded. First of all, only Facebook and Twitter posts which contain a verbal message were selected. This permitted for the inclusion of videos, but no still images which are not accompanied by a written text. Due to feasibility constraints of the study, only the first layer of the data is analysed. This indicated that external hyperlinks were not included in the sample. In addition, some of the actors and parties turned out to be overrepresented in the final sample than others, but this difference is explained thoroughly in the results section. Comments from Facebook “friends” and “followers” were not taken into consideration, neither retweets. The sample had to be kept as narrow as possible due to time restraints and the nature of this study which is aiming at investigating solely content coming from
politicians and targeting citizens.
An obstacle for obtaining an objective and concise population was the substantive amount of fake profiles and fan pages of politicians and political parties. So the correct and real profiles were traced down through the official websites of the parties or the politicians. As a result, the profiles of the party leader together with the one of the party were the only ones selected.
Content analysis and code book
For the purpose of this study a small-scale code book was developed (see appendix B). This measurement instrument is built on the theoretical ground of populism presented in the first section of the paper. Thus, it consists of questions about anti-elitist talk, reference to the
13
people and image of the party leader. The code book is divided into two main sections. The first one contains seven questions on the overall information about the units of analysis’ length, date, source and format. The second part is composed of eleven tailored-specific questions, which are presenting the concept of populism put into practice based on literature and previous research. In total, the coders had to give answers to eighteen questions for each Facebook or Twitter post. This was done after a short training period when details were explained and doubts were clarified. All in all, a sample of 243 Facebook posts and 87 tweets were included for the whole official period of the campaign- from the 25th of April till the 24th of May (European elections, 2014, n.p.).
Inter coder reliability
To make sure the coding is reliable two independent Bulgarian native speakers were assigned to conduct the coding. A total number of thirty messages were double-coded which composes a 9 % of the whole population. All of the studied Facebook posts are saved as print screen shots (see appendix C), while the Twitter posts were generated through the “All My Tweets” software program1
. The reliability of the findings for each of the studied variables was tested with the use of Krippendorff’s Alpha formula which ensures that any agreement did not occur by chance.
The detailed calculations of the inter coder reliability are put in appendix A of this paper. Krippendorff’s Alpha value for each variable is above 0.6, which is a good reliability in scientific terms, taking into account that the score of the complete agreement is 1 and the complete disagreement is -1.
1
This program generates all previous tweets from a given user, no matter the time period. The results are shown on one single page, where the text of the tweet is presented together with a hyperlink to the actual posting.
14
Results and Findings
First of all, in the primary hypothesis of this study it was expected that the populist parties and their leaders will have more abundant social media presence than the mainstream politicians and party members. We will see how active each of the studied political actors was on Facebook and Twitter during the campaign, what type of messages they have sent and what social media channel they chose more often. This will put the general results of the study into a broader context, enabling to better understand how the online campaigning of the four biggest Bulgarian parties was structured.
Looking at the results per political actor, we see that 65.2% of all the studied online messages are shared by the soft populists and their leader together with the extreme-right nationalistic populist group. This allows us to accept the first hypothesis. Surprisingly enough, the party leader of the extreme-right populist party did not have a Twitter account and throughout the campaign did not post anything on his official Facebook fan page. A personal profile of the nationalist Volen Siderov was also absent from Facebook.
Consequently, the leader of the soft populists Boyko Borissov and his party are the source of 48.2% from the total number of the 330 units of analysis.
The only two party leaders active on both social media are the soft populist and the leader of the mainstream socialist party Sergei Stanishev. An independent samples T-test shows that on average Borissov’s messages have a mean length of 309.95, while the
conventional leader’s posts had a mean of 124.08. This is an indication for either word count or length of video in seconds. Those results are statistically significant as the p value of the Chi Square Test is .00 (below the .05 threshold). It is safe to say that the most active political subject in this study is the leader of the soft populist party.
15
All in all, the preferred social medium by the political subjects was Facebook. 73.6% of all the studied campaign messages were shared on this platform, compared to the 26.4% that were posted on Twitter. Moreover, the four parties and their leaders were mostly
communicating to the supporters through messages accompanied by external hyperlinks. This format of political content took 45.2% of the total sample. On the other hand, only 8.8% of the posts were purely written-based.
The majority of the online content provided on Facebook and Twitter, was therefore not originally created to be solely a political dialogue on the social media, but rather the two websites were used as a means for spreading ideas originating from traditional media. Usually, the external hyperlinks were leading to websites where there were news articles uploaded or interviews with a party member conducted by a professional journalist.
As supported by substantive theory and previously gathered empirical data, in the second hypothesis, it was presumed that the populists will refer more often to the average people than the conventional parties. Nonetheless, the two sets of parties have very similar results. Together the soft populists and the extreme-right nationalists referred to the people in 55.3 % of their online messages to the electorate. In comparison, the mainstream parties BSP and MRF had a result of 52.2 %. This difference is insignificant in statistical terms as the p value of the conducted Chi Square Test is .581 (lower than .05). The second variable which was measured in relation to the political discourse towards the common men was whether the politician or party show care for the people’s needs. The populists were demonstrating
empathy for the problems of the Bulgarians in 43.3 % of the studied online content, while the mainstream groups and their leaders in 42.6%. Once again the p value is higher than .05 (it is .910), which makes the results not significant. Thus, the second hypothesis can be rejected.
16
The third outlined hypothesis suggested that the soft populists and the extreme-right nationalists would have more negative messages towards the governing elite and the other parties in the elections than the mainstream politicians. This expectation was operationalized with two questions in the constructed code book, asking whether the actor is criticizing other parties, or whether he/she accuses the current or previous government for a specific issue in the society. The proposition that the populists will show more negativity towards the other parties and politicians was confirmed. 48.4% of all the messages of the populist parties and leaders were found to be negative. Blaming, critiquing and showing mistrust in the decisions of others were some of the traits of negativity I was looking for in the online messages.
On the other hand, the two conventional parties had only 13.9% negativity in their overall content.
“I am the guarantee that the price of the electricity will not go up! If [the government] do increase it, I will make the ruling government resign!” Volen Siderov, May 16th, Facebook
The politician with the highest number of criticizing messages is the soft populist leader Boyko Borissov. Furthermore, out of the two populist parties, it has been discovered that GERB is the structure with the higher number of condemning messages towards the
17
governing elite. Out of all the content the party has posted online, 50.9 % was critiquing the mainstream BSP/MRF coalition, while the extreme-right nationalists have posted content consisting of 14.3% negative messages.
Thus, the anti-government rhetoric was not dominated by the nationalists, as expected in hypothesis 3a, but by the soft populists and their leader. Thus, hypothesis 3a can be
18 1Political subject describing the overall situation in the country as being critical, hopeless or dangerous
This finding is also significant in scientific terms as the conducted Chi Square Test once again scored a p value of .00. Consequently, we could conclude that in relation to the second aspect of populism (anti-elitist rhetoric) indeed the two mainstream parties are overall less negative than the populists.
“With every day that is passing, this government is leading the country towards a condition from which Bulgaria could not get out afterwards. Every single day… If they keep going like this [the government], at the end they will drop like ripe pears from a tree.” Boyko Borissov,
May 7th, 2014, Facebook
Finally, support was not found for the fourth and last hypothesis of this paper, where it was suggested that the populist leaders will be presented as charismatic macho men more often than the first party members of the conventional parties. Indeed, in 54.8% of the cases the populists were described as having charisma, while the mainstream leaders scored 43.4%. The politician presented the most as a person with special appearance, who is capable of solving quickly the social issues in the country, is the soft populist Boyko Borissov.
19 2Political subject proposing simple and fast solutions to long-term social issues in the country
However, not in line with the expectations, the results show that the leader of the conventional BSP party and the European socialists, Sergei Stanishev, was presented as having charisma more often than the leader of the nationalistic right-wing populist party. It has to be stressed, that the other conventional leader included in the sample was not depicted as being charismatic in any of the studied messages. This illustrates that the encountered differences between populists and non-populist leaders are entirely based on the self-portrayal of the mainstream socialist party leader and the two studied populists. Nonetheless, those results were not statistically significant as the Chi Square Test p value is .195 which is above the standard of .05. Hence, the last hypothesis is not supported either.
An important conclusion, still, is that the mainstream socialist leader Sergei Stanishev was also self-portraying himself in a more personal way, as a man with magnetism and attractive personality for the voters. On the 9th of May, the leader of the socialists has shared on his Facebook profile an album with pictures from his visit to the small village of
20
“The grandmothers from the village introduced me to their grandsons. Later that day, I visited the temple “St. Ioan Bogoslov” for which we allocated 10 thousand leva for the cleansing of the murals and reconstruction of the iconostasis” Sergei Stanishev, 9th of May,
2014, Facebook
To sum up, two out of the four presented hypotheses were confirmed and two were rejected. First of all, as expected, it was proven that populists are more active online than mainstream politicians. The actor with the most numerous, as well as the longest, Facebook and Twitter posts was the soft populist Boyko Borissov. Second of all, it was also confirmed that Borissov, the extreme-right nationalist leader and their parties are more negative on their social media platforms than the leaders of the two conventional groups. However, hypotheses two and four were rejected due to statistically insignificant differences. Namely, it was not verified that the populists will refer to the average citizens more often than the mainstream political structures. Also, the populists’ first party men were not found to be self-portrayed as being more charismatic than the mainstream leaders. The theoretical and practical
implications of these findings will be discussed thoroughly in the following section.
Discussion and Implications
In the context of an economic crisis and instable political situation in Europe, the 2014 European parliament election campaign turned out to be a fertile ground for polemics and deliberation on behalf of politicians. The Bulgarian political domain once again proved itself as a “populist-friendly” arena of promises and negative campaigning (Cholova & De Waelle, 2011). The current research paper looked particularly at online rhetoric of two populist and two conventional parties. First of all, the findings confirmed Zaslove’s (2008) statement that populism is a concept very complicated to grasp. As it was proven, populists do not refer to the average citizens more often than the mainstream parties. Substantive number of academics
21
(e.g. Jagers and Walgrave, 2007; Zaslove, 2008, Laclau, 2005) argue that populist politicians try to play the role of “chosen ones”, representing the common will. However, the results presented in the previous section of this paper, did not support this claim. Bulgarian populist and mainstream parties were somewhat similar when addressing the average citizens.
On the other hand, findings proved that as expected, populists will be more negative than conventional actors. It should be highlighted that negative campaigning, euroscepticism (as seen in the case of the Bulgarian radical right-wing party) and populism are three
completely different ideas which could often be confused and perceived as the same concept (Pitrova, 2007). Jagers and Walgrave (2007) driven by this complex nature of populism, confirmed also by their findings, try to narrow down the term as they divide it into four sub-groups, depending on the closeness to the original definition of the classical thick populism. Applying this ranking system also to online political discourse could be a strong turning point for any future research. It could give more precision and reliability to the results.
Furthermore, it was discovered that as a whole Bulgarian politicians chose the social media platform Twitter way less than Facebook as a communicative channel for reaching voters. One explanation could be that Bulgarian citizens are still not comfortable with Twitter, given that the number of Twitter users in the country for 2012 was approximately 300 000, 22% of which profiles were active (Capital, 2012, n.p.). For a comparison, the Facebook users were more than 2 million for the same time period. Therefore, the results of this article might not be valid and reliable in a country with a higher Twitter penetration such as the
Netherlands for instance. A cross-national comparison might shed some light on this distinction.
All in all, there was a sharp contrast encountered in the amount of social media presence between the soft populist leader and his party and the other actors in this study. In
22
addition, a big part of the online content posted by his party was shared from the original messages coming from the leader’s profile. Henceforth, the nature of the posts was similar and this reinforced the already exiting populist connotations that were present primarily. This overrepresentation, however, is not perceived as a drawback of the sampling in the study, but rather as an indicator for the figure of the party leader and his salience in the media. As Bos (2011) argues, being a visible actor in the public’s agenda plays an important role for the politician’s image as a reliable and trustworthy leader.
As mentioned earlier, 45.2% of all the studied social media texts were accompanied by an external hyperlink, directing the reader to another website. Usually, those were containing the full article or statement made by the politician or the party in question. This takes us to the conclusion that the primary channel for which those messages were created for is neither Facebook nor Twitter. The latter two social networking sites play the role of an additional, secondary tool for dissemination of political ideas. Moreover, given the rather low usage of SNS by some of the studied political subjects, it could be argued that Facebook and Twitter are still not considered as a powerful and convincing media by the political actors in Bulgaria. Once again, a repetition of this study in a country with higher Internet usage might look different.
This paper assumed a more intense self-portrayal of populist leaders as charismatic men. Specifically, previous research showed that right-wing extreme parties were mainly recognized and are becoming the centre of media attention because of the charismatic person in charge of the group (Bos, 2011, p.184). Nevertheless, the social media content coming from Bulgarian politicians did not show statistically significant differences between populists and mainstream leaders. This finding could be linked to discussions in the communication field about the vivid presence of personalization in social media for political purposes (Kruikemeier, 2014, p.19). The personal information about politicians circulating online
23
makes the boundaries between populists and mainstream leaders blurred within the context of social networking sites. Thus, the concept of charisma as a trait of populism has to be
analysed more thoroughly in the future.
Surprisingly, in contrast to the assumptions made in the beginning of this paper, the leader of the radical right-wing party was not active on either Facebook or Twitter. It is crucial to stress, however, that the findings of this research are merely limited to the social networking sites’ field. At the same time, the nationalist leader has ample presence in other media. The party has its own television channel, newspaper outlet, blog and online forum. Moreover, the political structure was found to be the only party with eurosceptic political discourse. A cross-media investigation could build upon the theoretical implications of this research as it would reveal the difference between the ways politicians use the traditional and the new media for sending populist messages.
The second most active leader in the sample was the conventional politician Sergei Stanishev, who is also in charge of the European Socialist Party. The Bulgarian socialist party however, was not present either on Facebook or on Twitter. Most importantly, the findings supported the hypothesis that the populists will be more negative than the conventional parties and their leaders. The results were also statistically significant. Critiquing messages were measured not only towards the ruling elite, but also against other political structures as a whole. This was done in order to control for differences, due to the fact that the two
mainstream parties are part of the ruling coalition and obviously they will not be criticizing themselves. This finding could be a stimulating starting point for a future research that will look in depth into negative campaigning online. Further research should consider different timeframes as well, for instance comparing campaign with non-campaign periods. Also, it ought to include social media platforms beyond Twitter and Facebook.
24
Ultimately, a part of any subsequent analyses on the topic of populism, would be studying the effects of the political populist messages on the voters. Measuring electoral outcomes or voting turnout is one option to test the effects on people after being exposed to populism on social networking sites. Taking into account the impact of social networking sites on the political involvement especially of young citizens (Bakker & de Vreese, 2011), it is essential for political scientists to analyse whether those unexperienced voters are exposed to populism online and, if so, what the consequences are.
Limitations of the Study
The reliability is the major limitation of the manual content analysis since it is not done by technology, but human beings. Every researcher has his or her own personal background, view point and belief system which undoubtedly influences the way he reads a particular text and understands its meaning. Nevertheless, the inter coder reliability calculated in the methods section of this paper makes sure that all deviances are encountered and
scientifically justified.
Moreover, as any other study, the manual content analysis as a method of investigation has certain drawbacks in relation to the operationalization and conceptualization of the topic. A highly complex and controversial notion such as populism is challenging not only for defining, as seen in the literature review section of this paper, but also imposes obstacles for measuring. For this reason the created code book was built based on the four major traits of populism- abundance, reference to the people, anti-elitism and charismatic leadership.
The lack of sufficient amount of empirical data is often mentioned as a disadvantage of the manual content analysis. Ultimately, this study would have looked more complete if it had included all the parties and candidates in Bulgaria who were running for the elections. This would have grasped the holistic image of the country’s political landscape before the
25
European Parliament elections. Another possibility for extending the research and thus increasing the data is the inclusion of more social networking sites such as YouTube for instance.
Conclusions
The self-representation of party leaders in the social networking sites is a topic which will continue to evolve, given the constantly growing social and political importance of websites such as Twitter and Facebook. The presented research adds empirical findings to the contradictory nature of populism and its application on social media. Even though this is not an exploratory research it adds hands-on data on populism and social media platforms. It was discovered that the soft populist party and its leader are more active and negative on Facebook and Twitter than any of the other studied subjects. However, when referring to the average people- a major trait of populism, the two sets of political actors did not have significant differences. Neither were they distinct in relation to the party leader being represented as a charismatic man, also identified by scholars as significant trait of populism.
The fours studied political actors apply populist discourse in their Facebook and Twitter messages to very different extents. Importantly, noticeable differences were found within the two sets of parties- the populists and the mainstream subjects. This only confirms the arguments that the concept of populism is very complex and in order to be studied, a combination of scientific methods will be the most appropriate. Nevertheless, the conclusions of the paper advance the knowledge in the field about the nature of the communication directed from politicians to the voters during a campaign through user-generated media and how those self-determined media spaces have facilitated shifts in media legitimation and political representation in a young democratic country like Bulgaria.
26 Appendix A Inter coder Reliability
Krippendorff’s Alpha (nominal)
Percent Agreement K’Alpha N agreements N cases
Variable 1 90 0.798 27 30 Variable 2 96.7 % 0.913 29 30 Variable 3 100 % Undefined* 30 30 Variable 4 100 % 1 30 30 Variable 5 93.3 % 0.832 28 30 Variable 6 100 % 1 30 30 Variable 7 96.7 % 0.655 29 30 Variable 8 100 % 1 30 30 Variable 9 100 % 1 30 30 Variable 10 96.7 % 0.841 29 30
Appendix B Code Book
Project: “Populism on social networking sites: Bulgarian politicians during the 2014 European parliament election campaign”
Author: Lora Karamfilova (lora.karamfilova@gmail.com) Date: 08.05.2014
27
This code book has the aim to verify the occurrence of populist tendencies by tracing down selected attributes of populism by conducting a manual quantitative analysis.
A/GENERAL INSTRUCTIONS
1. You will be given a number of Facebook posts and tweets to code
2. Follow the code book and answer the provided questions, by selecting an answer and inserting the response into a Microsoft Office Excel file
3. In case of any uncertainties or questions regarding the coding process, contact the author of this code book
4. Selecting items to be coded
a. The unit of analysis in this project is communication messages on Facebook and Twitter. Everything to appear on the Facebook walls and Twitter pages of the pre-selected parties and the party leaders shall be coded. However, still images without a supporting written text will not be selected
b. Each unit will have its authentic value for the final analysis. The information could take the form of text or audio visual clips. External hyperlinks to news stories, statements of politicians, etc., will not be included as separate items c. The language of the messages is Bulgarian, thus the coders are only native
Bulgarian speakers. Comments of followers/friends of the party/leader will not be analysed. The data is collected from the 25th of April 2014 till the 25th of May 2014. It is stored in two primary ways: using the tool “all my tweets” and by making screen shots of Facebook wall posts.
B/Overall information 1. Coder ID 2. Item ID
3. Date of Facebook post or Tweet
4. Type of SNS (1= Twitter; 2=Facebook)
5. Source (1= BSP; 2= GERB; 3= MRF; 4= Ataka; 5= Boyko Borissov; 6= Sergei Stanishev; 7= Lyutvi Mestan; 8= Volen Siderov)
6. Type of item (1= text only; 2= video clip; 3= picture and text; 4= text with link) Note: Video clip with text must be coded as video clip only
28
7. Number of words/ Length of video in seconds C/ITEM INFORMATION
1. What is the overall tone of the message? 0= Neutral
1= Positive 2= Negative
Note/Example: An example of positive tone would look like: “С много песни и танци партията отбеляза празника в град Пловдив”. Neutral would be: “Женското дружество на партията стартира предизборната кампания”. Negative: “Ако не се промени нещо, българите ще продължат да бъдат най-бедните европейци”
2. Does the author/s refer to the average people? 0= No
1= Yes
Note/Example: The author/s can mention “the people” directly with words such as “the people”, “the citizens”, “the Bulgarians”, “Bulgaria”, etc. They can also refer to “the people” more indirectly. This is the case if they use terms such as “our nation”, “the electorate”, “we”, etc.
3. Does the author/s show care for the people’s needs? 0= No
1= Yes
Note/ Example: „Ние знаем от какво се нуждаят хората и ще им го осигурим“
4. Does the author/s show mistrust/negativism towards any of the European institutions? 0= No
29 Note/Example: “Членството на България в Европейския съюз не е довело до нищо добро за страната”
5. Does the author depict the current or near future situation in the country as being critical/tragic?
0= No 1= Yes
Note/Example: “Ситуацията в страната достигна дъното, по-лошо не сме виждали“ 6. Does the author/s present the party leader as a problem solver?
0= No 1= Yes
Note/Example: „Аз отивам на изборите, за да защитя правата на онеправданите българи“
7. Does the author/s use intensifiers such as “definitely” or “certainly”? 0= No
1= Yes
Note/Example: „Партията със сигурност влиза в кампанията с ясна стратегия за бъдещето на България“
8. Does the author/s promise fast and decisive measures if being elected? 0= No
1= Yes
Note/Example: „ Държавата ще бъде на крака до 400 дни“
9. Does the author/s criticize other parties/ third actors in any way? 0= No
30 Note/Example: „Управляващите са измамници и лъжци“
10. Is the party leader presented as a charismatic man? 0= No
1= Yes
Note/Example: He/ She is portrayed as a special human with exceptional personal qualities (e.g. he/ she shows confidence, empathy, sense of humour; physical appearance or strength are important part of the overall message of the text/video)
11. Is it suggested that the current/previous government is responsible for an issue in the country?
0= No 1= Yes
Note: This includes also reference to the previous government, composed of parties that are different from the authors’ own one. In this way traits of populism in the content of the current ruling party could be detected.
D/List of coders 1. Lora Karamfilova 2. Loreta Bahtchevanova
E/ List of parties and leaders 1. Ataka- Volen Siderov
2. GERB- Boyko Borissov 3. BSP- Sergei Stanishev 4. MRF- Lyutvi Mestan
31 Appendix C Screen shots
32 Appendix D Reference List
Abts, K. & Rummens, S. (2007). Populism versus democracy. Political Studies, 55(1), pp. 405–424
Alexandrova, E. (2011). Metamorphoses of Civil Society and Politics: From Ganko’s Café to Facebook, in G. Lozanov & O. Spassov, eds., Media and Politics. Foundation Media
Democracy, pp. 102-118
Bakarjieva, M. (2011). Reconfiguring the Mediapolis: New media and civic agency. New Media and Society, 14(1), pp.63-79
33
Bakker, T. P., & de Vreese, C. H. (2011). Good news for the future? Young people, Internet use, and political participation. Communication Research, 38(4), pp. 451-470.
Barr, R. R. (2009). Populists, outsiders and anti-establishment politics. Party Politics, 15(1), pp. 29-48.
Blumler, J. G., & Kavanagh, D. (1999). The third age of political communication: Influences and features. Political Communication, 16(3), pp. 209-230.
Bos, L. (2012). Public Images of right-wing populist leaders: The Role of the Media. Amsterdam: Ipskamp Drukkers, Enschede
Bos, L., van der Brug, W. & de Vreese, C. H. (2011) How the Media Shape Perceptions of Right-Wing Populist Leaders. Political Communication, 28(2), pp.182-206
Burack, C. & Snyder-Hall, C. (2012). Right-wing populism and the Media. New Political Science, 34(4), pp.439-454
Cannon, B. (2008). Class/Race Polarisation in Venezuela and the Electoral Success of Hugo Chavez: a break with the past or the song remains the same?, Third World Quarterly, 29(4), pp. 731-748
Canovan, M. (2004). Populism for political theorists? Journal of Political Ideologies, 9(3), pp. 241-252
Capital (2012). “Almost 300 thousand Bulgarians use Twitter”. Retrieved June, 20, 2014 from
http://www.capital.bg/biznes/media_i_reklama/2012/02/03/1758598_blizo_300_hiliadi_dushi _ot_bulgariia_polzvat_twitter/
De Waele, J. & Cholova, B. (2011). Bulgaria: A fertile ground for populism? Slovak Journal of Political Science (1), pp. 25-54
34
Elrich, F. (2005) Caracteristicas y efectos del discurso autocentrado en “Alò Presidente!”, Boletin de Lingüística, 24 (1), pp. 5-32
European Parliament (2014). European elections 2014. Retrieved May, 10th from
http://www.elections2014.eu/en
Groshek, J. & Engelbert, J. (2012). Double differentiation in a cross-national comparison of populist political movements and online media uses in the United States and the Netherlands. New Media and Society, 15(2), pp.183-202
Gurov, B. & Zankina, E. (2013). Populism and the construction of political charisma: Post-transition politics in Bulgaria. Problems of Post-Communism, 60 (1), pp. 3-17
Hawkins, K. (2003). Populism in Venezuela: The Rise of Chavismo. Third World Quarterly, 24 (6), pp. 1137-1160
Laclau, E. (2005). On Populist Reason. New York: Verso
Merolla, J.L.& Zechmeiste, E.J.(2011). The Nature, Determinants, and Consequences of Chávez’s Charisma: Evidence From a Study of Venezuelan Public Opinion. Comparative Political Studies, 44(1), pp.28-54
Mudde, C.(2004). The populist zeitgeist. Government and Opposition, 39(1), pp. 542-564
Kruikemeier, S. (2014). Getting Connected: The Effects of Online Political Communication on Citizens’ Political Involvement. Amsterdam: GVO drukkers & vormgevers B.V.
Novakovic, I. (2010). “European” and “Extreme” populists in the same row: the new government of republic of Bulgaria. Western Balkans Security Observer, 17, pp. 63-76 Pitrova, M. (2007). Populism in the first European elections in the Czech Republic. European Integration Online Papers, 11(2), pp. 1-12
35
Quintelier, E., & Vissers, S. (2008). The effect of Internet use on political participation: An analysis of survey results for 16-year-olds in Belgium. Social Science Computer Review, 26, pp. 411-427
Raycheva, L., & Dimitrova, D. V. (2008). ‘Election news coverage in Bulgaria’, in J. Strömbäck & L. Kaid, eds., The handbook of election news coverage around the world. Routledge, 341-56.
Rooduijn, M. & Pauwels, T. (2011). Measuring Populism: Comparing Two Methods of Content Analysis, West European Politics, 34(6), pp.1272-1283
Schedler, A. (1996). Anti-political establishment parties. Party Politics, 2, pp. 291–312 Scheufele, D. A., & Nisbet, M. C. (2002). Being a citizen online: New opportunities and dead ends. Harvard International Journal of Press/Politics, 7(3), pp. 55-75.
Shah, D. V., McLeod, J. M., & Yoon, S. (2001). Communication, context, and community: An exploration of print, broadcast and Internet influences. Communication Research, 28, pp. 464-506
Smilov, D. (2008). “Bulgaria.” In Populist Politics and Liberal Democracy in Central and Eastern Europe, eds. Meseznikov, Gyarfasova, and Smilov, pp.15–38
Smilov, D. (2011). Rule of Law and the Rise of Populism: A case study of post-accession Bulgaria. CAS Sofia Working Paper Series, 4(1), pp. 1-49
Stefanova, B. (2009). Ethnic nationalism, social structure, and political agency: explaining electoral support for the radical right in Bulgaria. Ethnic and Racial Studies 32(9), pp. 1534-1556
36
Stefanova, B. (2012). Between ethnopolitics and liberal centrism: the Movement for Rights and Freedoms in the mainstream of Bulgarian party politics, Nationalities Papers: The Journal of Nationalism and Ethnicity, 40(5), pp.767-782
Taggart, P.(2000). Populism. Buckingham, England: Open University Press
Van der Wurff, R. & Schönbach, K. (2011). Between Profession and Audience. Journalism Studies, 12(4), pp. 407-422
Vergeer, M. , Hermans, L. & Cunha, C. (2012). Web campaigning in the 2009 European parliament elections: A cross-national comparative analysis. New Media and Society, 15 (1), pp.128-148
Zaslove, A. (2008). Here to stay? Populism as a new party type. European Review, 16(3), pp. 319–336
