Monitoring Political Extremism through Social Media: taking the temperature of intolerance in Spain
Ariadna M. Fernández
Grote Bickerstraat 58 F3 1013 KS (Amsterdam) / andairamf@gmail.com Supervisor: Anat Ben David
Second Reader: Bernhard Rieder
MA-‐ New Media and Digital Culture. University of Amsterdam 27 June 2014
CHAPTER 1. Introduction 4
1.1. Literature Review 9
1.1.1. Facebook as an Object of Study for Social Science Research 9 1.1.2. Internet and Politics 12
1.1.2.1 Politics and Social Media Platforms 12
1.1.2.2 Social Media Production Practices and Media Strategies 12 1.1.2.3. Extreme right Parties’ Use of the Internet, from the 1990s to 2014 14
1.1.3. The Internet as a Comfort Place to Incite Hatred 15
1.1.3.1. What is Hate Speech? Intolerance and Racism on the Internet 15
1.1.3.2. Countermeasures 16
1.1.3.3. Regulating Hate Speech in Spain 18
CHAPTER 2. Case Study: Spanish extreme-‐right Political Parties and Social Media 19
2.1. The Upsurge of Extreme Right Populism in Europe. Spain as an Exception 19
2.2. Previous research on Spanish extreme right parties and the Internet 21 2.3. Research Questions 23
CHAPTER 3. Methodology 23
3.1. Introductory Notes 23 3.2. Dataset 25
3.2.1. Description of the Dataset 25
3.2.2. Explanation of the Gathering Process 29 5.2.3. Cleaning and Formatting the Data 31 3.3. Content Analysis 31
3.3.1. Word frequencies and Co-‐occurrences 31 3.3.2. Image and Link Analysis 33
3.4. Network Analysis 37
CHAPTER 4. Findings 39
4.1. General Overview 39 4.2. Content analysis 41
4.2.1. Most common Words Used in Discourses 41
4.2.1. Clusters of Polemic Issues, their Frequencies and Co-‐occurrences 42 4.2.1.1. Immigration 42
4.2.1.2. Catalan independence process 46 4.2.1.3. Insults and its Target 50
4.2.2. Image and Link Analysis 51
4.2.2.1. Categorization 1: Type of Content 52 4.2.2.2. Categorization 2: Purpose of Posting 57 4.2.2.3. Categorization 3: Aesthetics Strategy 60 4.2.3. Network Analysis 61
4.2.3.1. Networks of Likes 61 4.2.3.2. Network of Users 67
CHAPTER 5. Discussion 69 CHAPTER 6. Conclusion 75 7. Acknowledgments 78 8. Works Cited 79 9. Appendix 91
CHAPTER 1. Introduction
Over the last decade, right-‐wing populist and extreme right parties have increased their success in countries where they have traditionally had little support at the polls (Wodak). In the European elections in May 2014, the xenophobic right-‐wing populist party, Front National, won in France with 26% of the vote (eldiario.es). Five years before, this party obtained just 5% of the electorate’s support (Mora). Similarly, other right-‐wing populist and extreme right parties have improved their results significantly in the U.K., Denmark, Austria, Hungary, Poland, Greece and Germany; but, Spain is the exception. Spanish extreme-‐right political parties are considered weaker and less influential than in other European countries because of their poor electoral results and their low popularity (Caiani and Parenti 721). In the European elections in 2014, the extreme right coalition, La España en Marcha (LEM), obtained 16,879 votes, which accounted for 0.1% of the vote (El País). In Spain, there is a wide spectrum of fringe parties with this trajectory, eight of which are included in this study: Democracia Nacional (DN), Alianza Nacional (AN), FE-‐La Falange and Nudo Patriota Español (NPe), which constitute the LEM coalition, and Alternativa Española (AES), Movimiento Social Republicano (MSR), España 2000 and Plataforma per Catalunya (PxC). Most of them are nostalgic fascists, and the only two that have obtained electoral success in recent years are España 2000 and PxC, which have discourses that are more aligned with new European right-‐wing populism and, thus, focus on immigration (Casals, “La ultraderecha española” 172).
Nevertheless, the Internet and social networking platforms represent a powerful alternative political arena for fringe parties to increase their profiles through self-‐promotion (Caiani and Parenti 721), as well as to reach new audiences, balance their underrepresentation in traditional media and create a stronger sense of community (Jackson and Lilleker 239). In addition, the Internet is home to hate ideologies (Chan et al. 2). In Spain, a country deeply hit by the economic and financial crisis, hate speech on the Web is increasing at an alarming rate
(ECRI 22), and there seems to be a general tolerance for freedom of expression with no limits (ECRI 23). Neo-‐Nazi groups in Spain have set up more than 200 websites, blogs and forums (ECRI 22) and extreme right parties have radicalized their public statements (both online and offline) against immigrants since the economic crisis started in 2008 (Benedí 2).
In May 2014, limiting freedom of expression on social networking platforms was a hot topic in the Spanish media. That month, a political representative of the governing party, the Partido Popular (PP), was murdered in public and some Twitter users celebrated her death (De la Fuente). Similarly, the Israeli basketball team, Maccabi, was the victim of anti-‐Semitic insults on Twitter when they beat Real Madrid to win the Euroleague title (García). As a result of these incidents, the Spanish Minister of the Interior said that his department was going to pursue tweets that incited hate and violence, and was rethinking additional legal means to prosecute this kind of behaviour on the Internet (Garea). Nevertheless, this move was criticized by other political parties, like the one that governs Catalonia (the north-‐eastern region of Spain), arguing that the minister only prosecuted hate speech that targeted the State or the governing party, while other collectives were victims of abuse on social networking platforms without any reaction from the central executive administration (La Vanguardia). The debate over whether the Spanish legislation should have specific mechanisms to target this kind of abuse on the Internet is still open. Social networking platforms have largely become “the curators of public discourse” (Gillespie 347), as well as the most effective way to incite hatred. This has created new challenges for tackling the spread of racist discourses and the incitement of violence against minorities and collectives on these platforms (Benedí 3).
An important array of scholars has studied hate speech and racism on the Web (Atton; Chau; Chan et al.; Levin; Whine; Gerstenfeld et al.; Warner and Hirschberg; Schafer; Burris et al.). One group has approached hate speech from a historical perspective by outlining how extremists have used the Internet to improve efficiency, stimulate action and spread racist propaganda (Levin; Whine). Another group has used empirical research to prove these claims by
showing that the Internet is a powerful tool for extremists to reach new audiences, recruit new members and build communities (Gerstenfeld et al.; Schafer), in addition to inciting violence offline (Chan et al.). Other scholars have identified hierarchical mechanisms in the construction of racist discourse by extreme right political parties on their websites (Atton) and revealed how extremists avoid “dirty word” filters when performing hate speech online (Warner and Hirschberg). Others have focused on the connections between extremist communities in blogs (Chau) and on the Web more generally (Burris et al.). Likewise, there is existing literature on political online media tactics that ranges from the study of political parties’ social media strategies (Jackson and Lilleker), to the influence of national development, to political culture, to the type of political actors in the production practices of websites during an electoral campaign period (Foot et al.). Scholars have also researched online action through the study of web spheres, which are defined as different sets of digital resources across multiple websites that revolve around a subject and are often connected by hyperlinks (Foot and Schneider). Finally, academics have reviewed the communication strategies of the Spanish extreme right on the Internet (Caiani and Parenti; Sánchez and Rodríguez; Rogers, “R.W”), thus portraying a fragmented political spectrum that has mainly used this medium to inform and spread propaganda.
However, there has been relatively little scholarly work on intolerance in social networking platforms in general, particularly in Spain. Hence, the purpose of this study is to help to fill this gap and enrich the study of extreme right parties’ behaviours on the Internet, by focusing specifically on their Facebook activity. To do so, an empirical study was performed by means of extracting and analysing data from the Facebook pages of nine political parties over a period of five years, from 2009 to 2013. By employing different research methods, this thesis aims to answer the following questions: To what extent have intolerance and hate speech risen in the Facebook pages of extreme right parties in Spain over the last three to five years? What kinds of media strategy have these parties employed on social networking platforms and how does extremist propaganda (texts, videos, photos, links) contribute to the increase of intolerance? Finally, how are these
political parties interlinked to each other and with other socio-‐political actors on Facebook? Furthermore, by way of contrast and to enable comparison, this study also examines the Facebook activity of the main right-‐wing political party in Spain, the Partido Popular (PP), which brings together a wide range of voters, including the majority of the extreme right-‐wing electorate (Ferrándiz). The idea here is to see whether there are similarities between the Spanish extreme right and the PP in terms of their strategies to increase their numbers of Facebook followers. This study aims to explain how communities are created on Facebook by analysing the scope, the use of language and the materials that are shared by these parties. It also aims to explain the socio-‐political and semantic dynamics of the parties. In this way, Facebook will be used as a barometer to measure the level of intolerance on social media.
This study uses the digital methods approach, which uses digital tools to conduct online research (Roger, E.V 5). The Internet presents a challenge for social science researchers who wish to study the new social formations that result from interactions between people online (Hine 1). The Web is considered a source of cultural and social information (Rogers, E.V 8), and in order to understand these interactions, scholars have developed new social science research methods that are more suited to this new medium. These are the so-‐called “digital methods”, which refer to a new research methodology that uses technology to study the Internet and online culture and focuses on the medium-‐specific dynamics (Rogers, E.V 5). Medium-‐specific dynamics are used to study the web through its native digital objects of research, i.e., those born in the new medium, such as websites, the link, the hyperlink, search engine algorithms and social networking platforms (Rogers, E.V 5). These native digital objects are interesting for social science research not only for the social formations that are embedded within them, but also as objects of study themselves, in the sense that one can explore their technical specificities and how this influences the interactions they host. The aim of this research is to examine a native digital object Facebook through a digital methods approach in order to study the social interactions between Spanish extreme right political parties and their users, which are mediated by this social networking platform. Therefore, this method contributes to
understanding the functioning of these actors as a whole, since online interactions complement, rather than replace, the offline reality (Rogers, E.V 6).
The first phase of social-‐networking platforms research was based on “demographic” data, in the sense that scholars were interested in the information that users wrote in their profiles to describe and present themselves, such as their ages, locations and genders (boyd and Ellison). This is because the first social networking platforms were “profile-‐centric sites” that targeted specific demographics (boyd and Ellison 216). The personal data obtained through this stage of social-‐networking platform research was similar to that found through surveys, questionnaires or interviews, which have been used traditionally by social science researchers when studying society (Rogers, E.V 25). Nevertheless, with the advent of new social networking platforms, such as Facebook, which allows users to connect with each other and express preferences in different ways, a new kind of personal data based on interactions, interests and tastes has attracted the interest of researchers (Rogers, E.V 25). Richard Rogers called this latter research approach “post-‐demographic,” meaning social media platforms, as native digital objects and in comparison to standard datasets, provided users with more space to express their personal preferences (Rogers, “P.D” 155). Therefore, this is a post-‐demographic study because it relies on the data embedded on Facebook in terms of users’ “tastes and especially taste relationships” rather than just their personal information (Rogers, E.V 25). This type of research shows what users do, rather than who they are; therefore, post-‐demographic data is relevant to the research questions presented in this study since the idea is to recognise patterns in the activity of the Spanish extreme right on Facebook, as well as their connections on this platform. An analysis of demographic data alone would not have provided this information. Additionally, the methodology of this research is framed within an emerging field called “data science”, which uses a computational approach to study a wide range of social, political and humanitarian fields (Jockers 4).
In the following sections, I will introduce Facebook as an object of study for social sciences, briefly trace the roots of the extreme right parties’ use of the
Web, describe political communication practices on the Web 1.0 and Web 2.0, and explain what hate speech is and the extent to which it finds a comfortable home on the Internet. Finally, I will describe previous research related to the extreme right in Spain and their online practices, before presenting my research questions.
1.1 Literature Review
1.1.1 Facebook as an Object of Study for Social Science Research
Mark Zuckerberg created Facebook in 2004 when he was studying at Harvard University, and within one month, half of the Harvard student body was a member (Phillips; Markoff). With an average of 757 million daily active users as of December 2013 (Facebook annual report), Facebook is changing the way society interacts online, especially in comparison with the Web 1.0 where information was published to be found rather than to promote participation (Song 251). In 2012, the social networking platform was the most visited website in the U.S. for the third consecutive year, followed by Google (Tatham). That same year, it administered 2.5 billion pieces of content, 500 terabytes of data, 2.7 billion “likes” and 300 million photos per day (Constine). In 2008, a Spanish version of Facebook was created, which explains why most of the political parties analysed in this study joined the platform in the years that followed.
There is considerable literature on the impact that Facebook had on social life and on research (Wilson et al.; Webb et al.), but a small group of scholars have also researched its intrinsic characteristics and its corporate social media logic (Gerlitz and Helmond; Langlois and Elmer). Langlois and Elmer have studied the new forms of power embedded within Facebook. They consider it a mistake to see social media platforms as simple tools to facilitate communication because they also monetize and encode all shared information (2). According to them, researchers should study social networking platforms from a critical standpoint and be aware of their own logics and tactics, which influence and reshape the users’ acts of communication (5). This is what they call the “double articulation”
of digital objects, meaning that communicative acts that happen at one level simultaneously create new articulations (e.g., economic, technological, cultural and social logics) at another level (5). In other words, Facebook works in the back-‐end of the platform to transform the users’ interaction processes into valuable data (6). Gerlitz and Helmond are aligned with this critical approach in their article, “The like economy: Social buttons and the data-‐intensive web,” which examines Facebook’s expansion through its social buttons and the data flows they empower (1349). Their main argument is that, while the open web measures user engagement through links and hits (Rogers, “O.I.N.W” 200),1 Facebook recentralizes data from the open web through its external “like” and “share” buttons and, thus, creates a “like economy,” in which the social has economic value (1349). In this new context, where Facebook redirects the activity in the open web to its platform, the online interaction experience of the user is also reshaped. As Gerlitz and Helmond explain, the more social plugins embedded in a website, the more it is open to being shaped by the activities of Facebook users (1354). Therefore, people will experience these websites in a personalized way because social plugins offer recommendations that are built on the users’ activities and their contacts. Therefore, Facebook and the open web are becoming more interlinked because what happens in one space influences the other (1354). In this way, communities moved by similar interests can interact more easily and create a loop that is based on recommendations and inputs from like-‐minded people. To draw on Langlois and Elmer, and Gerlitz and Helmond, the decentralized presence of platform features across the web facilitates the creation of atmospheres in which users are prone to behave in a certain way rather than in another. Likewise, in his book “The Filter Bubble” Eli Pariser discusses the consequences of the increasing personalization of the web. The Internet shows users what they want to see rather than a variety of information (8). In this sense, Facebook’s algorithm, which tracks what users like and the links they click on, recommends similar information based on prior interests (8). This creates individualized and personalized “filter bubbles,” in
which the platform filters the information that is most “relevant” to the user, according to their previous online behavior (9).
These perspectives speak to the politics of platforms (Gillespie) and constitute critical research regarding how Facebook’s technological structures have social and economic effects, as well as how this is conditional on the social interactions that the platform hosts. As an example, social plugins for criticising or disliking on Facebook do not exist, which reinforces the idea that the platform decides what types of social activities can be executed (Gerlitz and Helmond 1360). Although I agree with Langlois and Elmer that Facebook data is not a transparent representation of human behavior but, rather, represents the articulation of participatory and corporate logics (14) (e.g., one cannot “dislike” a post), for the purpose of this research, I will consider Facebook data as a partial indicator of the conduct of extreme right Spanish political parties without entering into a debate about the platform quantification and reshaping of this data. Nonetheless, I will study Facebook as a multi-‐layered digital object that presents three main characteristics that were revealed by Langlois and Elmer: (a) The social media platform as a media object with a semantic layer, in which a discourse analysis can be performed; (b) as a network object that connects different kinds of information, especially through the “like” button; (c) and as a phatic object (Miller) that establishes specific kinds of relationships among users. Facebook as a phatic object particularly characterizes the critical approach of this thesis. I will study the extent to which Facebook contributes to creating a hate speech climate in the extreme right’s political pages by creating micro-‐worlds of recommendations based on the users’ posts and acts of “liking,” but also as a result of its EdgeRank algorithm, which uses an unknown logic to suggest news items and interesting content for specific groups of users.
So far, I have described Facebook as a digital object (Rogers) and examined how its technical characteristics influence the social relations embedded within it. In the next section, I expound the relationship between politics and the Internet, moving from the political parties’ practices on social networking platforms to a historical outline of extremists’ use of the Web.
1.1.2 Internet and Politics
1.1.2.1. Politics and Social Media Platforms
In the mid-‐1990s, the Internet gained political significance by offering people the possibility to learn about relevant political affairs and get involved in political action (Bimber 133). The Internet reconfigured the information flow of political participation and the organization of interests, for instance, by reinforcing existing communities (Bimber 134). Likewise, political parties have profited from the Internet to further their own interests through political organizing and communication (Bimber 154). Historically, political parties have used the Web as a means of communication and information (Gibson and Ward), and the explosion of immediately available information has contributed to the growth and maintenance of political identity (Borgida 472). Today, the use of social networking platforms has become a feature of political engagement for many citizens (Rainie et al.). The shift from the Web 1.0 to the Web 2.0 has allowed users to coproduce content rather than being passive observers (O’Reilly 13), which creates options for higher grades of interaction between political parties and the public (Esteve and Borge 2), as well as the possibility to build networks within a community of users (O’Reilly 16). In this regard, political parties have lost some control over their online communication flows in social networking platforms, while that of the users has increased (Esteve and Borge 2). These new kinds of interactions on the Web 2.0 have pushed political parties to rethink their communication strategies on social networking platforms, which is explained further in the next section.
1.1.2.2. Social Media Production Practices and Media Strategies
Political communication on the Web 1.0 has historically been more about information provision and persuasive communication than encouraging participation (Jackson and Lilleker 235). With the shift to the Web 2.0 and its promise to bring potential changes in terms of participation and co-‐production (O’Reilly 33), political parties are readapting their media strategies to this new
paradigm, which, unlike the Web 1.0, is less compatible with the traditional top-‐ down, elite-‐to-‐mass-‐media tactics in politics (Jackson and Lilleker 233). Previous research has examined web strategies within a political context on the Web 1.0 (Foot et al., Foot and Schneider) and has developed a theoretical framework that is useful for framing social media strategies in this context. In their article, “Electoral Web Production Practices in Cross-‐national Perspective,” Foot et al. analysed four communicative functions of websites for political purposes: informing, involving, connecting and mobilizing citizens (40). In this research, they studied the extent to which different factors influenced the political uses of digital information and communication technologies. These dimensions included economic, technical and political development, as well as political culture and genre effects (45). The latter refers to a website that is produced by the same type of socio-‐political actors (e.g., political party or non-‐governmental association) and often reflects similarities in “form” and “function” for the producers and visitors that are associated with the genre of the site (44). The main point of their findings is that the same types of political actors have strong similarities in their web production practices at a cross-‐national level, rather than different actors having the same online tactics as a result of being in the same country (52). This finding is explained using the concept of web spheres—a termed coined by Foot and Schneider in their article, “Web Sphere Analysis: An Approach to Studying Online Action,” which refers to the different units of analysis that one can extract from the web in order to explore it better. These units are formed by texts, features, links and sites that are produced by different socio-‐political actors around a certain issue (Foot and Schneider 1). Furthermore, these web spheres can become potential structures for online action (1). According to Foot and Schneider, interactions within a web sphere are relevant for their multiple social, political and cultural dimensions (1), which can be extrapolated to the public traces left by the users on different Facebook pages. In this regard, Jackson and Lilleker have studied how political parties in the U.K. have embedded their communications strategies in Web 2.0 and the extent to which this has been done to encourage interactivity between users (233). Far from wanting to increase user participation, Jackson and Lilleker argue that political parties have found a new space for self-‐promotion in Web 2.0 (234). In
fact, minor and fringe parties are usually the ones that tend to promote a higher grade of interaction in comparison with mainstream parties (240), which merely use social media platforms to inform (241). Moreover, for some minor and fringe political parties, social networking platforms are useful to balance their underrepresentation in traditional media (239). In sum, Web 2.0 offers the possibility to create and share information, build networks and interact. However, there is little evidence of a substantial change in organizational thinking in terms of the political parties’ apparatuses, especially for mainstream parties in which top-‐down communication tactics still prevails (257).
1.1.2.3. Extreme Right Parties’ use of the Internet From the 1980s to 2014
Since the beginning of the Internet, extreme right parties have also harnessed the potential of this new medium (Levin 958). The far-‐reaching features of the Web make it an attractive medium for extremists to spread their hate ideologies and reach new audiences (Berlet). The first computer bulletin board system (BBS) used by the extreme right, Liberty Bell Net, was created in 1983 by the neo-‐Nazi publisher George Dietz and was used to spread propaganda against Jews and Blacks (Berlet; Levin 960). With easy access to the Internet in the 1990s, hate groups started to go online (Berlet) and reproduced the media strategies of the first Nazi BBS, as well as similar initiatives that had begun a decade earlier (Levin 963). In general, the purpose was to stigmatize and undermine the “enemies” of the Aryan race (Levin 962). The BBS transformed into so-‐called discussion boards—online forums through which bigots could encourage violence among people with whom they had no direct connection (965), thus reaching new audiences. The next period of right-‐wing online extremism began in 1995 with the launch of the major “hate site”, stormfront.org, run by Nazi Don Black (1996) and which became a model for other extremist websites in the following decade.
Roger Eatwell explains that the rise of the Internet’s popularity among extremists is an outcome of the fact that the Internet is a low-‐cost technology with a high potential to present and spread information, to customize messages
for specific audiences, to create a sense of community and to reach audiences across jurisdictional boundaries (16). Additionally, the extreme right uses the Internet to rally supporters and incite violence (Whine 243), as well as to build transnational contacts (Chase-‐Dunn and Boswell; Petit; Bennett). Today, more than 10,000 websites promote hate worldwide (Simon Wiesenthal Center), 600 of which are hosted in Spain (Ibarra). In addition to websites, extremists are also using social networking platforms to spread their propaganda (Simon Wiesenthal Center), as well as adapting their communications strategies to avoid being accused of inciting violence through the use of hate speech on these platforms (Warner and Hirschberg). In the next section, I will expand on this and explain what hate speech is. I will also explore the challenges that the Internet and social media bring in terms of limiting freedom of expression, both internationally and domestically, as in the case of Spain.
1.1.3. The Internet as a Comfortable Place to Incite Hatred
1.1.3.1. What is Hate Speech? Intolerance and Racism on the Internet
Although researchers do not share a universal definition of hate speech, the Council of Europe defines it as an offense that covers “all forms of expression which spread, incite, promote or justify racial hatred, xenophobia, anti-‐Semitism or other forms of hatred based on intolerance, including: intolerance expressed by aggressive nationalism and ethnocentrism, discrimination and hostility against minorities, migrants and people of immigrant origin” (Council of Europe). Hate speech on the Internet is a rising problem that, from the 2000s onwards and with the explosion of social media, is gaining a special relevance. According to a report by the European Network Against Racism (ENAR), the Internet presents new challenges for tackling the spread of racist discourses, especially in social networks, such as Facebook and Twitter, which are often used to propagate stereotypes and xenophobic ideas. This concern is usually related to how it clashes with the right to freedom of expression, which is often used by those who advocate hatred to legitimate offenses, particularly against minorities (Gascón 310). The Internet, which considered a public forum that
should enable freedom of expression, lacks a standardized regulation for hate speech (Levin 971). Thus, polemic content that triggers anger, hatred or offense has a home in this new medium (Levin 974). In the following sections, I outline some of the countermeasures that have been adopted to fight hate speech both offline and online at the international level and specifically in Spain.
1.1.3.2. Countermeasures
Regulating expression on the Internet is a difficult task that depends on multiple factors (Levin 971), one of which being the fact that the Web is ruled by different national laws that contradict and overlap each other (Arellano). Furthermore, the general norm is usually minimal intervention (Moretón 5). In 2001, the Council of Europe, in collaboration with other countries, such as the United States, signed the first Convention on Cybercrime,2 which regulates different typologies of crimes through computer networks, but left aside the criminalization of hate speech on the Internet. The American Society of International Law argued in an article that the Convention was useless for preventing “unlawful hosting,” which occurs when people write offensive statements on the Internet in one state and locate their servers in another country that has more permissive hate-‐speech laws (973). Nevertheless, after passing this regulation, a group of experts from the Council of Europe proceeded to write an additional protocol3 to the Convention concerning the criminalisation of acts of racist and xenophobic nature committed through the Web. Although this was an attempt to join forces to fight intolerance and discrimination online, governments usually delegate the control of user content to the Internet service providers (ISPs), which limit the freedom of expression in the terms of their service agreements (Levin 977; Brown 1). The problem, however, is that ISP requirements are more flexible than those established by law and they seek limited liability for what users say (Gillespie; Brown). Similarly, social networking platforms also regulate hate speech in their terms of use and the
2 http://conventions.coe.int/Treaty/en/Treaties/Html/185.htm
criteria to shut down a page depends on the platform, unlike a judge who would force them to block certain profiles. Facebook, for instance, advocates for the free circulation of information with some limits, such as the respect to the dignity of persons, no incitement of violence and no distribution of illegal content that breaks copyright laws (Facebook community standards). In its terms of service, it also refers to hate speech:
“Facebook does not permit hate speech, but distinguishes between serious and humorous speech. While we encourage you to challenge ideas, institutions, events, and practices, we do not permit individuals or groups to attack others based on their race, ethnicity, national origin, religion, sex, gender, sexual orientation, disability or medical condition” (Facebook).
Mark Zuckerberg’s company uses a specific blocking system for each case. For instance, Nazi content is forbidden in Germany, but allowed in the United States; therefore, Facebook acts in accordance with each country’s legislation in terms of removing undesirable pages (Arellano 2018). Normally, Facebook initiates its procedure for shutting down a page after someone has denounced it. Users can report content they consider to be harmful under the following criteria: self-‐ harm, bullying and harassment, hate speech, violent graphic content, nudity and pornography, identity and privacy, intellectual-‐property-‐regulated goods, phishing and spam (Facebook). The Spanish parliament approved the Law of Information Society Services and Electronic Commerce (LSSI)4 in 2012, which allows judges to shut down Spanish sites and social media profiles, as well as block access to websites hosted on foreign servers that do not obey national laws (Scheeres). The following section briefly reviews the Spanish government’s regulations concerning hate speech.
4 Which is the implementation of the European Directive on electronic commerce (Dir.
2000/31/EC) and some of the provisions of the Directive on Privacy and Electronic Communications (Dir. 2002/58/EC) into the Spanish law (epic.org).
1.1.3.3. Regulating Hate Speech in Spain
The Spanish Penal Code contains some articles that categorize behaviours that fall under the category of hate speech5 (Gascón 311), as well as different legal figures that typify crimes of injuries, incitement to commit a crime, or exaltation of violence with sentences that are sufficiently dissuasive (Queralt). At the same time, the Spanish constitution recognizes freedom of expression in article 20.1, which is often used to justify racial hatred. The Constitutional Tribunal has ruled several times in favour of victims of hate speech, arguing that racist discourse goes beyond freedom of expression when it attacks other rights that are also worthy of protection under the law, such as human dignity6 (Gascón 317). However, there are differences between the ways that various legal institutions interpret the law. For instance, the Spanish Constitutional Court takes into account the “potential danger” of hate speech, while the Supreme Court demands “concrete actions” derived from hate speech in order to prosecute the perpetrators (Gascón 331). Likewise, the European Court of Human Rights (ECHR) has made several pronouncements against the protection of the Holocaust denial discourse.7 In general, hate speech is a complex issue and the variability of the application of this penal nature—which depends on the criteria of the judge or the tribunal—is an obstacle for the protection of those facing discrimination (Gascón 336). Therefore, regulating hate speech is a difficult task, not only because of the legal precedents that put freedom of expression in conflict with other fundamental rights, but also because of the differences in the definition and protection of hate speech in various international and national laws (Moretón 4). Likewise, it must be remembered that hate in itself is not a crime; instead, provocation of violence, which is criminal if it is against people and things, is the decisive factor (Queralt). Hence, deciding whether danger is
5 Articles 510 and following of the Penal Code from 1995
6 Protected in article 10 of the Spanish Constitution, which derives from the right to honour
contemplated in article 18.1
implied in hate speech or not is the challenge that judges face, and is an issue that legislators must explore further.
In this sense, the term “cultural violence,” as coined by Johan Galtung, is relevant for exploring when hate speech begins and ends, and when there is an incitement of violence. For Galtung, cultural violence occurs in “those aspects of culture (…) that can be used to justify or legitimize violence” in its direct (an event, an aggression) or structural (processes of domination) manifestations (291). Some occidental cultural features, such as the tendency to individualise and discriminate (301) or nationalistic feelings, which are based on ideas of being a “chosen people” and the devaluation of the “other” (298), contribute to the rationalisation of violence against “those lower down” (302). In this sense, Galtung’s triangle of violence—consisting of direct, structural and cultural violence—is useful for understanding the repeated stigmatisation of immigrants in social media. This type of harassment could be considered cultural violence that might be easily transmitted to the other corners of the triangle (302). Extreme right parties in Spain are deeply nationalistic and tend to discriminate against immigrants and anyone who attempts to unify Spain (Casals, “La ultraderecha española” 164). The particularities of these groups are discussed in the next section in order to provide context for the empirical findings concerning their use of hate speech on Facebook that are presented in the sections that follow.
CHAPTER 2. Case Study: Spanish extreme-‐right Political Parties and Social Media
2.1. The Upsurge of Extreme Right Populism in Europe: Spain as an Exception
Right-‐wing populist and extreme right parties started to gain prominence after the fall of the Iron Curtain in 1989 when fear of an influx of foreigners and the possible loss of jobs became significant (Wodak). In 2014, as a result of the continuing economic and financial crisis in the EU that started in 2008, citizens
became worried again about losing their jobs and quality of life; a fear that was combined with a strong inclination towards Euroscepticism (Wodak). This scenario has helped right-‐wing populist and extremist parties to gain sympathizers because they use the “politics of fear” combine with an anti-‐ immigration and anti-‐Establishment discourse (Wodak). Lately, as was confirmed by the results of the latest European elections in 2014, there has been an advance of the far and radical right throughout Europe, with Spain being the exception.
Spanish extreme right parties have not succeeded at the polls in this country despite a favourable conjunction: institutional discredit, political disaffection toward mainstream political parties, an economic crisis and high levels of unemployment (Casals, “La ultradreta espanyola”). These political parties have been profoundly divided and their electoral results have been insignificant since the first democratic elections in Spain after Francisco Franco’s dictatorship (1939–1975) (Casals, “La ultraderecha española” 147). However, two of the political parties included in this study—both of which were founded in the 2000s—do have significant political representation at the local level. The PxC, which is known for its Islamophobic agenda, has 67 political representatives in local administrations, and España 2000, based in Valencia, has four councillors in Valencia, and another in Alcalá de Henares, Madrid.
The electoral weakness of the rest of the extreme right could be due to three main factors that are shared by two other southern-‐European countries: Greece and Portugal. Firstly, the memory of the dictatorship in these countries had a profound impact in society and extreme right parties have no figured out how to renew their ideological discourses (Casals, “La ultradreta espanyola”). Secondly, these countries have experienced a democratization process since the dictatorial period, which has been accompanied by expansive social policies that have delayed citizens’ political disaffection. Third, new right-‐wing political parties have appeared in all of these countries and have gathered the majority of the extreme right electorate (Casals, “La ultradreta espanyola”).
Currently, the majoritarian right-‐wing political party in Spain is the PP, which embraces voters from the extreme right that would normally be part of the majoritarian traditional conservative party in other countries (Férrandiz). In 2013, for example, the Spanish press published photos of the new generation of the PP, brandishing Francoist symbols8 (Giménez). Moreover, a Spanish nationalism sentiment and an antagonism to all the movements that go against the unity of Spain are common features of the extreme right and the PP (Navarro). The purpose of this study is to examine whether these similarities can also be found on Facebook. Scholars have previously studied the Spanish extreme right on the Internet, which is briefly explained in the next section; however, the comparison with the PP in this study sheds further light on the existing literature.
2.2. Previous Research on Spanish Extreme Right Political Parties and the Internet
In their article, “The Spanish Extreme Right and the Internet”, Manuela Caiani and Linda Parenti portray the online organizational structure of the right wing in Spain9 through a content and network analysis of its websites (719). They found that the Spanish extreme right is weakly connected through the web, at both the international and local levels (724) and that it uses the Internet merely to diffuse information and, less often, propaganda (728). Furthermore, these groups commonly use the Web to launch anti-‐immigration campaigns and to spread messages in favour of the unity of Spain (731). Extremists are not competent at using the Internet to mobilize their followers (730). Rather, they prefer to focus their online practices on presenting their identities, albeit very divided, inarticulate and old-‐fashioned ones (732). The content on such websites generally expresses a "nostalgic and closed political culture”, as well as hate symbolism, which appeared in half of the sample they analysed (732).
8 http://www.publico.es/politica/462835/el-‐lider-‐de-‐las-‐juventudes-‐del-‐pp-‐de-‐xativa-‐hace-‐el-‐
saludo-‐fascista
![Letter to the editor: "Nasal high-flow versus non-invasive ventilation in patients with chronic hypercapnic COPD" [Letter]](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAAAICRAEAOw==)