Mass Murderer or International Terrorist?
Framing of Lone Actor Attacks by Dutch Newspapers
Christie Ebeling
s2096994
Master ThesisCrisis and Security Management
Faculty of Governance and Global Affairs Leiden University
Supervised by Dr. P.G.M. Aarten June 10th, 2018. The Hague.
Abstract
“Without knowing much, if anything, about the particular people, groups, issues, or even places involved, the terrorist and anti-terrorist frame allows us to quickly sort out, interpret, categorize, and
evaluate these conflicts” (Norris, Kern and Just, 2003, p. 11).
The media can frame acts of terrorism in multiple ways: by adding or leaving out keywords, phrases, sources and stereotyped images, they present the issue through a certain lens (Entman, 1993). The way a terrorist event is framed can affect the public perception on acts of terrorism. This paper examined how lone actor attacks committed by perpetrators with different ideological backgrounds were framed in five Dutch national newspapers. Through a framing analysis, 178 newspaper articles on two lone actor attacks, on by a right-wing extremist perpetrator, and the other by an Islamic perpetrator were examined in depth. Results show that Dutch newspapers connected the Islamic perpetrator to international terrorism, focusing disproportionately on his religion to ‘other’ him. In contrast, the right-wing extremist perpetrator was less frequently connected to terrorism, and was mainly ‘othered’ based on his mental health. These findings can contribute to the segregation of certain groups in today’s society, as well as on the risk perception and corresponding behavior of the public.
Acknowledgements
I would like to express my sincere gratitude to my supervisor Dr. Pauline Aarten, as well as to Jessica Sciarone for the time they took to provide me with constructive feedback and useful guidance throughout the whole thesis-writing process. I am also grateful to my parents for all their support, and to Lewis, who never complained about my complaining. Lastly, a special mention goes to Olivia, my thesis-partner in crime; I am glad we were in this together.
Contents
1. Introduction ……… p. 5 1.1 Research Question ………. p. 6 1.2 Societal & Academic Relevance ………p. 6 1.3 Reading Guide ………p. 7 2. Body of Knowledge ……… p. 8 2.1 Terrorism and the Media ……… p. 8 2.2 Media Framing ………..…... p. 9 2.2.1 Frames and Framing ……… p. 9 2.2.2 Which Factors Influence Frames? ………. p. 11 2.2.3 Frames and Terrorism ……… p. 12 2.2.4 The Effects of Framing ……… p. 13 2.3 Previous Empirical Research….………. p. 13 2.3.1 Framing Terrorism in the United States………. p. 13 2.3.2 Framing Terrorism in the Netherlands……….. p. 14 2.4 Lone Actor Terrorists……… p. 15 2.4.1 Islamic & Right-wing Extremist Lone Actors ……… p. 16 2.5 Summary……….. p. 17 3. Methodology……… p. 18 3.1 Operationalization ………. p. 18 3.2 Sampling Methods ………. p. 18 3.2.1 Sampling of Lone Actor Attacks ………p. 18 3.2.2 Sampling of Dutch Newspapers ……….………p. 20 3.3 Method ……….………. p. 21 4. Results ……… p. 26 4.1 Which Frames Were Used?... p. 26 4.1.1 The Bouhlel-case in Nice……… p. 27 4.1.2 The Breivik-case in Oslo and Utoya ………. p. 29 4.2 The Differences and Similarities in Framing………. p. 32 4.21. Framing of the Perpetrator ……… p. 33 4.2.2 Framing of the Incident ………. p. 37 4.2.3 Framing of the Victims ………. P. 38 4.2.4 The Subsequent Discourse ……… p. 39 4.3 Differences in Framing by Popular and Quality Newspapers ……….. p. 41 4.4 Explaining the Differences and Similarities………... p. 43 5. Discussion & Conclusion ……….. p. 45 5.1 Conclusion: Framing of Lone Actor Attacks in the Media ………. p. 45 5.2 Implications of the Findings ……….. p. 46 5.3 Limitations & Further Research ……… p. 46 References ……… p. 48 Appendix ………. p. 62
1. Introduction
On October 1st, 2017, Stephen Paddock shot into a crowd of festivalgoers from a hotel room in Las Vegas,
killing 59 people and injuring more than 500. The 64 –year-old American was responsible for what has been called the ‘deadliest mass shooting in modern U.S. history’ (Ruiz-Grossman, 2017). Days after the attack, U.S. media received much criticism on their coverage of the Las Vegas shooting; it was supposedly very different than coverage of other recent shootings in America, due to the fact that the shooter was a white, non-Muslim man (Beydoun, 2017; King, 2017; Ruiz-Grossman, 2017). For example, white attackers are often described as mentally troubled, while in the case of Islamic attackers, Islam is almost always seen as the root of the problem (Ruiz-Grossman, 2017). A study by Powell (2011) confirms this finding; when the perpetrator of an attack is a non-Muslim, media humanize and individualize the
perpetrator and focus on the possible reasons the attacker could have to commit the act. However, when a Muslim has committed the attack, the act is depicted as a part of Islamic jihadism and terrorism (Powell, 2011). This paper aims to find out whether this framing of lone actor attacks is also the case in a European country; the Netherlands.
Studies on framing of terrorism in the United States have found that the way terrorist attacks are covered and framed in the media can have an effect on the public perception of acts of terrorism (Patrick, 2014). Through applying certain frames, media can highlight bits of information in order to promote a particular problem or recommend countermeasures (Entman, 1993). The audience perceives reality through such frames, which can possibly affect the audience’s perception of terrorism in general as well as the threat it poses (Powell, 2011). Thus, the way media frame the perpetrator or victims of an attack can have an influence on the public opinion of terrorism, as well as on their perception of the magnitude of the threat and from whom that threat is coming.
Even though the Netherlands has not had to deal with a major terrorist attack on Dutch soil so far, research shows that more than 60 percent of the Dutch population was worried about terrorism in the Netherlands in 2017 (NCTV, 2017a). Moreover, 40 percent even feels that the terrorism threat has an impact on their personal safety (NCTV, 2017a), and also the government considers a terrorist attack to be a realistic threat; the threat level of the Netherlands is 4 out of 5 (NCTV, 2017b). Other Western European countries such as France, Belgium and the United Kingdom did see terrorist attacks take place on their soil; attacks that were not only covered heavily in the countries own media, but also by newspapers and television channels in the whole of Europe. Acts of terrorism already have a definite newsworthiness due to their dramatic and sensational nature, but the interest in the attack increases when it takes place in a country nearby (Patrick, 2014).It is therefore not strange that Dutch media pay much attention to terrorist attacks in other (European) countries.
1.1 Research Question
For this thesis, the following research question was developed:
“To what extent are there differences in the framing of lone actor terrorists with different ideological backgrounds in Dutch national newspapers?”
As shown, the research question focuses on the news framing of one particular type of terrorism; attacks perpetrated by lone actors. More specifically, the thesis will take a closer look at the framing of two major recent lone actor attacks in Western Europe; one attack perpetrated by a right-wing extremist (the 2011 attack by Anders Breivik in Oslo and Utoya) and one attack perpetrated by an Islamic perpetrator (the 2016 vehicle-attack by Mohamed Bouhlel in Nice). In order to examine the differences and similarities in framing, the thesis used discourse analysis to analyze the media coverage of five national newspapers in the Netherlands, namely het Algemeen Dagblad, de Volkskrant, de Telegraaf, NRC Handelsblad and
Trouw. The study will look at both differences in frames between the two cases, as well as differences in frames used by popular versus quality newspapers. In order to structure the analysis as much as possible, the following sub-questions are presented:
SQ1: “Which frames were used by the five Dutch newspapers to describe the two lone actor attacks?”
SQ2: “What are the differences and similarities between the framing of the two cases?” SQ3: “Do popular newspapers use different frames to describe the attacks than quality newspapers?”
SQ4: “What could explain the differences and similarities in framing?” 1.2 Societal & Academic Relevance
Due to the fact that news media are accessible and widely available, most of the public relies on newspapers, television, and online news websites for information about terrorism (Chermak &
Gruenewald, 2006). Therefore, the way that newspapers cover terrorist attacks can have an influence on the public opinion on terrorism, “including what people learn about any terrorist event, how they evaluate the main actors and issues under contention, and how far this coverage affects public concerns and perceptions of the risks and threats of further terrorist acts” (Norris, Kern & Just, 2003). Thus, the way media frame lone actor terrorists can have an effect on the public opinion of certain groups in society and can even lead to an “us-versus-them” mentality if the public believes that terrorism is equal to this group. It is therefore relevant to look at the way news media in the Netherlands frame attacks committed by perpetrators of different ideological backgrounds. Lastly, Liem, Kuipers & Sciarone (2018) mention that visibility of recent terrorist attacks can increase the risk perception of the public, which can lead to a
change in behavior. For example, after the London Bombings in 2005, almost a third of the inhabitants of London wanted to avoid the public transport services after the attack (Liem, Kuipers & Sciarone, 2018). Media coverage and framing can thus have a direct influence on the risk perception of the public and their behavior.
Academically, this study addresses a gap in the literature on the framing of terrorism, specifically lone actor terrorism which is an upcoming field of research. Even though multiple studies have been conducted that examine the framing of terrorism by news media, most of these studies focused on the United States (Nacos & Torres-Reyna, 2003; Nevalsky, 2015; Papacharissi & Oliveira, 2008; Powell, 2011; Schaefer, 2003). In contrast, framing studies that focus on the Netherlands are scarce, even though there are a number of studies that focus on the framing of Islam, Syria fighters and immigration/integration (Berbers et al., 2015; d’Heanens & Bink, 2006; Vliegenthart & Roggeband, 2007). Thus, research into the framing of terrorism in the Netherlands will add to the existing body of knowledge that covers media framing of terrorism in the Western world. It will also contribute to an understanding of the coverage of terrorism by multiple Dutch national newspapers and thus contributes to the field of terrorism studies as well as the field of media studies. In addition, since research into terrorism has become a dominant aspect in the field of security studies since 9/11, the present study also contributes to this field of research.
1.3 Reading Guide
The thesis will proceed as follows. The next section will discuss the theoretical foundation that the study is based on; it will elaborate on the complex relationship between mass media and terrorism, and will address the characteristics and motivations of lone actor attacks specifically. Furthermore, it will explore the concept of framing as well as reflect on previous relevant studies in this field. Then, the
methodological section will elaborate on the chosen research design and the selected cases. For this thesis, a discourse analysis was performed in order to take an in-depth look at the media framing by Dutch newspapers of two selected lone actor attacks in Europe. Subsequently, the results section will discuss the findings of the discourse analysis and will compare the results of the two cases. Lastly, in the concluding section, an answer to the research question posed in the introduction of this paper will be provided.
2. Body of Knowledge
The following chapter will discuss the most relevant literature on which the current study builds, and consists of four main sections. The first section discusses the complex relationship between terrorism and the media in general. The second section addresses the concept of framing more in-depth; it focuses on the different definitions of ‘frames’ and ‘framing,’ the factors that can influence frames, which frames are related to terrorism and what the effects of framing can be. Thirdly, previous empirical research on media framing of terrorism is discussed, followed by a short overview of the concept of ‘lone actors’ and the differences between Islamic and right-wing extremist actors. Lastly, a short summary of the discussed topics will be provided.
2.1 Terrorism and the Media
The act of terrorism has a communicative nature: “Its [terrorism] main goal is not to harm … the immediate victims but to send an intimidating message to a target population, state, or organization” (Morin, 2016, p. 988). Many scholars have written about terrorism as a communication strategy and the role that mass media play in this communication (e.g. Powell, 2011; Nacos, 2007; Norris, Kern & Just, 2003; Weimann, 2005). Nacos (2007) speaks of mass-mediated terrorism, Weimann (2005) mentions
media-oriented terrorism; both refer to the increasing awareness of terrorist organizations of the role that
mass media plays in society and the growing importance they attribute to publicity and mass media. Schmid (1983) identified four roles that the media can play in a terrorist attack: as a communication platform, as an escalator of fear, as an attention grabber and as a channel for intimidation and propaganda. In line with these roles, Nacos (2007) argues that terrorists typically strive for four specific
media-dependent goals: (1) attention and awareness of various audiences, (2) recognition of their motives, (3) respect and sympathy of their followers and (4) a quasi-legitimate status. Excessive reporting on terrorist threats and attacks by the media contributes to obtaining all the aforementioned goals. For example, Nacos (2007) argues that “the perpetrators do not necessarily have to do the explaining [of their political causes and motives] themselves – the media do it for them” (p. 21). Thus, mass media are an excellent way for terrorists to reach their target audience; since terrorists per definition attack to get a political point across, it is important to the terrorist cause that their message is heard (Camphuijsen & Vissers, 2012; Nacos, 2003). After all, if their political statement is not publicized, the terrorists’ goal is not reached. On the other hand, the media also benefit from terrorist attacks. Some scholars speak of a symbiotic relationship between the media and terrorists; the latter seek attention and publicity for their message and want to influence the general public, while the media tries to sell their product with stories containing dramatic, emotional and sometimes disturbing aspects (Camphuijsen & Vissers, 2012; Morin, 2016;
Wardlaw, 1981). Stories that are exciting, sensational, dramatic or emotional are often the kind of stories that readers are interested in (Camphuijsen & Vissers, 2012). Terrorist attacks tick all those boxes, and therefore have a definite newsworthiness. Through media coverage of terrorism, terrorism does not only affect the direct victims, it also impacts the everyday lives of the public that reads the newspaper, watches television, or listens to the radio (Morin, 2016); the perception of the threat and the risk of a terrorist attack plays a role in the impact of terrorism as well (Liem, Kuipers & Sciarone, 2018).
2.2 Media Framing
When reporting on terrorism-events, media consciously or unconsciously choose a specific lens through which they present the issue. Thus, communicators make decisions of how to frame a topic, and manifest this in a text by adding or leaving out keywords, phrases, sources and stereotyped images; what is not included in the news is just as important as the information that is included (Entman, 1993). The following section will take a closer look at the concept of framing. It discusses what framing is, how frames are chosen, which frames already exist and the possible effects of framing.
2.2.1 Frames and Framing
Even though the concept of framing is increasingly gaining popularity amongst scholars, there is no consensus on a definition of framing or on how to apply the concept (De Vreese, 2005). Some scholars place emphasis on the organizing character of frames. Gamson & Modigliani (1987) give the following definition: “A frame is a central organizing idea or story line that provides meaning to an unfolding strip of events (...)” (p. 376). According to Gamson & Modigliani, a frame can point out the essence of the issue at hand. Tuchman (1978) also addresses this organizing aspect of framing: “The news frame
organizes everyday reality and (…) is an essential feature of news” (p. 193). This definition is in line with the definition that Reese (2001) puts forward: “frames are organizing principles that are socially shared and persistent over time, that work symbolically to meaningfully structure the world” (p. 11). In other words, Reese (2001) considers frames to be tools, which can be used to give meaning to the reality and to define situations. He thus not only emphasizes the organization aspect of frames, but also gives frames a social function. Neumann, Just & Crigler (1992) also argue that frames can be seen as tools “which media and individuals rely on to convey, interpret and evaluate information” (p. 60).
Other scholars emphasize the communicative aspect of framing. De Vreese (2005) argues that “communication is not static, but rather a dynamic process that involves frame-building (how frames emerge) and frame-setting (the interplay between media frames and audience predispositions)” (p. 51). Entman (1993) defines frames as “highlighting some bits of information about an item that is the subject of a communication, thereby elevating them in salience” (p. 53). In other words, some pieces of
specific way or associating them with culturally familiar symbols (Entman, 1993). Lastly, Brüggeman (2014) also mentions how “framing focuses on communication as a process of making sense of the world” (p. 61).
Two media processes are related to the concept of framing, namely agenda setting and priming. The main difference between framing on the one hand, and agenda setting and priming on the other, is that the latter two are concerned with issues; the focus lies on the importance attributed to certain issues by the audience (Scheufele & Tewksbury, 2007; Van Gorp, 2007b). Framing on the other hand, distinguishes clearly between issues and frames: multiple frames can be used to cover an issue (Van Gorp, 2007b). Moreover, framing is thus more concerned with the way an issue is presented instead of the issue itself (De Vreese, 2005; Scheufele & Tewksbury, 2007).
Frame Locations
According to Entman (1993), frames have various locations within the communication process. The first location is the communicator, who makes a conscious or unconscious decision to frame a topic in a certain way. Secondly, the text contains certain words, images, sources and sentences that contribute to the frames used. Thirdly, the perception of the receiver can be guided through frames, and lastly, the culture refers to frames that are commonly used in the discourse of a certain social grouping. Thus framing is not
necessarily only concerned with the frames used by journalists, but also with frames that individuals possess. Scheufele (1999) defines this as the difference between media frames and individual frames. He argues that individual frames are “mentally sorted clusters of ideas that guide individuals’ processing of information” (p. 107). Yet, Entman (1993) mentions that framing in all four locations can serve similar functions of selection and highlighting of pieces of information.
De Vreese (2005) sorts the process of framing into 3 distinct levels. Firstly, frame-building refers to the conscious or unconscious choices a communicator makes to cover the news event. The outcome of this stage are the frames that are present in a text. Then, frame-setting is related to the receivers predispositions and prior knowledge. This can influence how frames affect learning, interpretation and the evaluation of events (De Vreese, 2005). Thus, the process on the first level takes place at the location of the
communicator as defined by Entman (1993), while the second level takes place at the location of the receiver. The consequences of framing are present on two levels: the individual and the societal level. On
the individual level, a consequence can be a change in attitudes towards a topic as an effect of certain frames the individual has been exposed to. On a societal level, the way frames are perceived may contribute to policy-making and collective actions (De Vreese, 2005).
Frames are represented by ‘frame packages,’ which Van Gorp (2007b) defines as “a cluster of logical organized devices that function as an identity kit for a frame” (p. 64). Van Gorp (2007b) argues that a frame package consists of three different parts: framing devices, reasoning devices, and “an implicit cultural phenomenon that displays the package as a whole” (p. 64). Framing devices are for example metaphors, descriptions, word choice, arguments and visual images that display the frame that is used in the text. Reasoning devices refer to statements concerning causes, consequences and justification, while the cultural phenomenon can be a narrative, a value or an archetype (Van Gorp, 2007b).
Types of Frames
De Vreese (2005) argues that there are two different types of frames. The first category is the
issue-specific frame, which relates to only issue-specific topics or events. The second category is the generic frame,
which is more abstract and can relate to multiple different topics even over time and in different cultural contexts (De Vreese, 2005). The categories are not exclusive concepts, and can even complement each other (Brüggeman, 2014). Examples of generic frames are for example the human interest frame, which brings an emotion angle to the story, and the responsibility frame, which attributes responsibility for the problem at hand (Semetko & Valkenburg, 2000).
Iyengar and Simon (1993) distinguish episodic framing and thematic framing: whereas the first concept refers to topic-specific coverage, often live and visually appealing, the second category is more concerned with background information and interpretation.The difference in focus can have an influence on the public perception of the responsibility for the issue; if a story is framed episodically, the public will attribute the responsibility to specific individuals, whereas with a thematically framed news story, the responsibility will be attributed to society as a whole (Berbers et al, 2016). Furthermore, episodic frames often occur directly after an event, while thematic frames generally follows after a while (Powell, 2011). 2.2.2 Which factors influence frames?
Framing of news stories can be influenced by different factors, which Brüggeman (2014) divides into three levels: the individual level, the organization level and the societat level. The different levels can overlap and can influence each other (Brüggeman, 2014). The first level, the individual level, exists of frames that each journalist already obtains to make sense of the world. Thus, journalists interpret issues on the basis of these frames, which can influence the frames a journalist uses to cover the news story.
Scheufele (1999) mentions that there are at least five factors that potentially influence the way journalists frame an issue: ideological or political orientations of the journalist, social norms and values, organization structures, journalistic routines and pressures of interest groups. The first two factors fall under
Brüggeman’s (2014) ‘individual level,’ while ‘organization structures’ and ‘journalistic routines’ belong to the second level of factors that influence the chosen frame: the organizational level. Scheufele &
Tewksbury (2007) emphasize that frames should not be understood as a tool for journalists to deceive the audience or spin a story, but rather to make it easier for the audience to understand the issue.
The second level, the organization level, refers to the frame patterns of the newsroom. In other words, the “dominant editorial policy on a specific issue” (Brüggeman, 2014, p. 69) can influence the news frames chosen. De Vreese (2005) refers to these editorial policies and news values specific to a certain media outlet as ‘internal factors’ that influence the frames. Besides that, if a journalist writes multiple articles for a specific section in the newspaper, he is also more likely to use frames that are related to that section (Brüggeman, 2014). The third level, the societal level, is linked to cultural values that can influence the frames used in the news coverage (Berbers et al., 2016; Brüggeman, 2014). Furthermore, the audience can also influence the frames used; in a newspaper with a certain ideological background different frames might be used than in a newspaper with a more neutral background (Brüggeman, 2014). Besides that, positive feedback of audiences can also have an effect on the frames used by journalists. Especially now that it is possible for individuals to respond to a news story immediately via social media or elsewhere online, a journalist can be influenced by this response when covering a similar event or issue (Brüggeman, 2014).
2.2.3 Frames and Terrorism
Previous studies on the framing of terrorism have found multiple frames concerning terrorism. Buitrago (2013) discusses two main frames: (1) the Western values and the need to defend them-frame, and (2) the
frame of control and surveillance. Another frame that is often mentioned is the War on Terrorism-frame,
(Campbell, 2017; Morin, 2016; Woods, 2011). Morin (2016) argues that “within this frame, the acts of individual Muslims become associated with the acts of terrorist groups whether the individual is associated with them or not” (p. 989). Campbell (2017) connects this frame to another frame: the
Orientalism-frame: this frame holds that the West has a narrow and shallow image of Muslim and Arab
identities, which leads to a negative representation of a big and diverse population. Furthermore, within this frame, the differences between the East and the West are very much exaggerated: this frame is also called the “Muslim Other-frame” (Campbell, 2017). Powell (2011) identified a related frame in American media coverage of terrorism: the war of Islam on the United States frame, which explains the motivation for Islamic attacks on the United States.
Brinson & Stohl (2012) identified two competing frames in their study on the media coverage of the 2005 London attacks. The domestic frame defined the act as a homegrown problem – often suggesting that the incident is isolated and random, - while the international frame connected the problem to international terrorist organizations – relating the incident to the aforementioned “war on terror” (Brinson & Stohl, 2012). Morin (2016) analyzed why media use either one of two specific frames when covering a violent
incident: the crime frame and the terrorism frame. Papacharissi & Oliveira (2008) found a military frame in which the incident was connected to the overall military strategy of the United States, and a diplomatic
frame, which focused on a diplomatic point of view regarding the terrorist event. Patrick (2014) identified
a human interest frame, referring to personal stories regarding the terrorist attack. Lastly, Woods (2011) mentions multiple frames that are used in the coverage of terrorism, namely the radical Islamic frame and the nuclear frame.
2.2.4 The Effects of Framing
Entman (1993) stated that by framing, media “define problems – determine what an agent is doing with what costs and benefits; diagnose causes – what are the reasons for the problem; make moral judgments – they evaluate the agents, problem and effects; and suggest remedies – offer solutions to the problem” (p. 52). As a result, framing can have multiple effects on the public.
De Vreese, Boomgaarden & Semetko (2011) argue that news framing can have both an indirect effect as well as a direct effect. The indirect effect is related to belief importance. This means that framing can change the perceived importance of some aspects of an issue. Entman (1993) also mentions this: according to him, media can direct the attention of the public to particular aspects of the reality, automatically meaning that attention is taken away from other aspects. Thus, framing highlights bits of information, making other aspects seem less important (De Vreese, Boomgaarden & Semetko, 2011). The direct effect of framing affects the belief content. This means that news frames can present an individual with new considerations; “by presenting information [in a specific way], individuals might become aware of connections between beliefs and topics they had not thought about before” (De Vreese, Boomgaarde & Semetko, 2011, p. 182). Thus, not only the importance of the existing beliefs of the individual is altered, but the content of the beliefs themselves are also changed. The effects that De Vreese, Boomgaarde & Semetko (2011) address both take place at the individual level, but De Vreese (2005) argues that there can also be an effect on the societal level: the way frames are perceived may contribute to policy-making an collective actions.
2.3 Previous Empirical Research
Multiple studies have been conducted in the past that examined how news media frame terrorism or topics related to terrorism. The most relevant studies are discussed below in order to position the current research in the standing body of knowledge.
2.3.1 Framing Terrorism in the United States
Most of the studies that are concerned with the framing of terrorism focus on the United States. Morin (2016) studied the media coverage of two deadly shooting incidents in the United States perpetrated by
lone actors and identified various rhetorical strategies that newspapers use to frame acts of violence as either terrorism or crime. An example of such a strategy is the “othering” of the perpetrator. “Othering” refers to the construction of the identity of an individual “against an implicit backdrop of what is
considered ‘the norm,’ and ‘normal’” (Morin, 2016, p. 944). Morin (2016) found that, when an incident is framed as a terrorist attack, newspapers othered the perpetrator based on his religious and ethnic
background. When an incident is framed as a crime, othering of the perpetrator focused on his mental abnormalities.
A study by Nevalsky (2015) analyzed American news coverage of terrorist attacks in Paris, France, and Borno, Nigeria, in order to include the effect of the geographic location of an attack on the news coverage. He found that there were significant differences between the coverage of both attacks. Most noticeably, the American media covered the attack in France with detailed and sympathetic stories, while the attack in Nigeria was framed negatively by overgeneralizing. Patrick (2014) also focused on two attacks that took place in different geographic locations and found that the coverage of an attack in the West compared to an attack in the East contributes to Islamophobia and an us-versus-them mentality.
Powel (2011) focused on domestic and international terrorist events on U.S. soil since the terrorist attack of 9/11. Domestic terrorism is defined as an act against the government of a country perpetrated by a citizen of that country, while international terrorism is linked to international terrorist organizations. Powell (2011) concluded that domestic terrorism is often framed as a minor threat that is committed by mentally ill individuals, while international terrorism is framed as a much bigger threat, perpetrated by Islam working together against “Christian America”.
Similarly to Powell (2011), Kearns, Betus & Lemieux (2017) also pay attention to the portrayal of Muslims and Islam with regards to terrorism. They found that attacks by Muslim perpetrators received an average of 449% more coverage than other attacks. They suggest that more representative media coverage could adjust the perception of the public, making it more in line with reality. Only the first study
mentioned above looked specifically at lone actor attacks, but did not include differences in ideological motivation.
2.3.2. Framing Terrorism in the Netherlands
Most of the framing studies that are conducted with regards to Dutch media focus on the framing of Muslims and Islam. A research by Ruigrok and van Atteveldt (2007) found that the events of 9/11 created a strong framework of Muslims as terrorists in the Dutch national newspapers de Volkskrant and de
(2007) showed that in Dutch media, geographical proximity of an attack is still an important factor when it comes to the coverage of such an event.
A study by d’Haenens & Bink (2006) studied the coverage of Islam in Dutch press, specifically in het
Algemeen Dagblad, and found that it is often associated with negative events, despite the fact that
reporting on Islam is fairly balanced. They also make a distinction between foreign Islam and Dutch Islam, in which foreign Islam is more often associated with terrorism than Dutch Islam – however, this changed after the murder of Theo van Gogh.
Similar to d’Haenes & Bink (2006), Roggeband & Vliegenthart (2007) conducted a study on media frames regarding immigration and immigrants in the Netherlands. They found that out of the five identified frames, only one placed immigrants in a positive light. The other four frames were pointed towards aspects of immigrants as inferior to Dutch culture. Roggeband & Vliegenthart (2007) also found that Muslim culture and Islam are specifically presented as threatening to Dutch norms and values. Thus, most framing studies that focus on the Netherlands analyze the representation of minority groups or ethnicities in the Dutch newspaper. Furthermore, most of these studies only focus on one or two
newspapers. The current research aims to contribute to the body of knowledge by analyzing framing of lone actor terrorists in multiple Dutch newspapers.
2.4 Lone Actor Terrorists
The phenomenon of lone actor terrorism increased significantly during the past three decades in both the United States and Europe (Ellis et al., 2016). Moreover, a popular conception is that lone actors might increasingly replace the traditional terrorist groups and networks (Berntzen & Sandberg, 2014). One explanation for the rise in lone actor attacks is that it is easier for lone actors to spread their message around the world relatively quickly, since the media and, more specifically, the Internet have a growing influence in society (Feldman, 2013; Stern, 2003).
There has been much ambiguity in scholarly literature with regards to the definition of lone actor
terrorists. For this thesis, lone actor terrorists are identified as “individuals that (a) decide, plan and carry out violent attacks individually, (b) do not belong to or follow orders from an organized terrorist group or network, and (c) handle out of ideological, political, or religious motivations” (Bakker & de Graaf, 2011; Barnes, 2012; Bates, 2012; Feldman, 2013; Spaaij, 2010; Spaaij & Hamm, 2015). Thus, lone actors might sympathize with extremist organizations or movements, by definition they are not part of these
organizations (Spaaij, 2010).Furthermore, Spaaij (2010) argues that the distinction between a lone actor terrorist and a “lone assassin” – he mentions political assassinations carried out by individuals – can be found in “the presence of a broader political, ideological, or religious cause that informs the actions of the
lone actor” (p. 857). Spaaij (2010) mentions that most lone actors of whom the ideological motivation was known, are inspired by White supremacy or Islamism.
In general, it is extremely difficult for counterterrorism officials, police and intelligence communities to deal with lone actor terrorists (Bakker & De Graaf, 2011). One often-mentioned reason for this is that lone actors work individually, which means that they do not communicate with others about their plans. This makes it difficult to detect these individuals and predict lone actor violence (Bakker & de Graaf, 2011; Michael, 2014).
2.4.1 Islamic lone actors vs. Right-wing extremist lone actors
Ellis et al. (2016) published a series of reports on lone actor terrorism and terrorists. They found that three out of four lone actor attacks in the past fifteen years in Europe were committed by right-wing and
religiously inspired1 perpetrators; religiously inspired perpetrators account for 28 percent, while
right-wing perpetrators account for 33 percent. Ellis et al. (2016) note that “given the intense public focus on religiously inspired terrorism, the finding that right-wing extremists account for a similar proportion of perpetrators within the database is particularly significant” (p. 11).
Ellis et al. (2016) found that right wing perpetrators often refer to immigration policy, patriotism and defending their country from “Islamism” as motivation for their acts. The last one fits in with a wider trend within broader right-wing extremism to define their cause as defense against the perceived threat of Islam. Religiously inspired perpetrators on the other hand mention revenge for political action by the West and religious extremist principles like violent jihad as justification for their actions (Ellis et al, 2016). As for personal characteristics of both perpetrator groups, a significant difference in age profiles was found; the majority of religiously inspired perpetrators are younger than 25, and the number of
perpetrators decreases as age increases. For right-wing perpetrators, the opposite is true: the majority of those perpetrators is at least forty years old (Ellis et al, 2016). Furthermore, it was found that right-wing perpetrators are more likely to play a politically active role than religiously-inspired lone actors. Lastly, a significant difference between the two types of perpetrators can be found with regards to social isolation; only 9 percent of religiously-inspired perpetrators were in some way socially isolated, in contrast to 33 percent of right-wing extremists (Ellis et al., 2016).
1 Ellis et al. (2016) adopted the categories used by Europol in its ‘Situation and Trend’ reports. Religiously inspired
terrorism is defined by Europol as “perpetrated by individuals, groups, network or organizations that evoke religion to justify their actions. Groups inspired by or affiliated with al-Qaeda belong to this category” (Europol, 2015, p. 47).
2.5 Summary
The media have a mutually beneficial relationship with terrorists, whereby terrorists use media to
communicate their message, and the media sell their product with the dramatic and sensational stories that terrorism offers. Media report on terrorism-events by consciously or unconsciously choosing a specific lens, or frames, through which they portray the issue. These frames can be influenced by different levels and located in various locations, and can ultimately have an effect on the beliefs and perceptions of the public. The framing of a right-wing extremist attack and an Islamic attack can therefore have an influence on the opinions and perceptions that the public has of different groups in society.
Even though previous empirical research on the framing of terrorism took place in the United States, previous research on framing in the Netherlands is scarce, especially with regards to terrorism. The current study therefore focuses on the framing of a specific type of terrorism: terrorist attacks perpetrated by lone actors with different ideological backgrounds in the Netherlands.
3. Methodology
3.1 OperationalizationA number of concepts need to be operationalized in order to examine the differences between the framing of Islamic perpetrators and right-wing extremist perpetrators.
“Framing” refers to the selection of some aspects of a perceived reality to make them more noticeable in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described (Entman, 1993).
“Lone actor terrorism” refers to individuals that (a) decide, plan and carry out violent attacks individually, (b) do not belong to or follow orders from an organized terrorist group or network, and (c) act out of ideological, political, or religious motivations (Bakker & de Graaf, 2011; Barnes, 2012; Bates, 2012; Feldman, 2013; Spaaij, 2010; Spaaij & Hamm, 2015).
An “Islamic perpetrator” is understood as an extremist Muslim that threatens, prepares or commits acts of violence in society in order to cause fear to a population, influence political decision-making or bring about social changes (AIVD, 2017a).
A “right-wing extremist perpetrator” is understood as an individual that undertakes actions that threaten the democratic legal order based on ideas of xenophobia, hatred of foreign (culture) elements or ultra-nationalism (AIVD, 2017b).
3.2 Sampling Methods
In order to realize the research goals set out in this paper, the coverage of five national Dutch newspapers on one right-wing lone actor attack and one Islamic lone actor attack in Europe will be analyzed. The choice to look into one right-wing extremist attack and one Islamic attack was made because these forms of terrorism are the most frequent in Europe in recent years, and thus form the main terrorism threat (Europol, 2017). The following sections elaborate on the sampling methods for the chosen attacks and the chosen newspapers.
3.2.1 Sampling of Lone Actor Attacks
Due to feasibility issues, this research will focus on only two terrorist incidents perpetrated by a lone actor in Europe. Even though the Netherlands did not experience a major terrorist attack on Dutch soil so far, it can be argued that attacks in other countries in Europe can affect the public perception of Dutch citizens. After the 9/11 attacks in the United States, citizens in Europe started to realize that terrorism was not only a threat to Middle Eastern countries or Asia, but that it had arrived to the Western world as well (Teoman, 2017). Moreover, as Vivienne Walt (2016) argued after the Paris attacks of November 2015; “The Paris
attacks were not aimed at a particular nationality, allegiance or even faith (several victims were Muslim). It targeted a way of life: ours.” Especially in Europe, where the Schengen Agreement omitted the internal borders of the European Union and therefore the physical boundaries between countries and cultures, attacks in one European country can be perceived as an attack on the Western or European “way of life” by citizens of other European countries.
The first selected case is the 2011 attack in Norway by Anders Breivik (hereafter referred to as the Breivik-attack). On July 22, a bomb went off in Oslo, killing 8 people and injuring 17. Two hours later, Breivik arrived at the island Utøya, where the annual youth camp of the Norwegian Labour Party took place. He shot 69 campers and wounded 66 more. After police arrived on the island, Breivik surrendered and was arrested. Breivik is a far-right extremist, who sees Islam and “cultural Marxism” as the enemy according to the manifesto he distributed on the day of the attack. According to the definition of ‘lone actor terrorism,’ Breivik can be considered a lone actor since he decided, planned and carried out his attack individually without belonging to an organized terrorist group. Furthermore, he was obvioursly inspired by ideological and political motivations. Moreover, multiple scholars treat Anders Breivik as a lone actor in their studies on lone actor terrorism (Bakker & de Graaf, 2001; Ellis et al. 2016). Lastly, Breivik fits the profile of a right-wing lone actor as defined by Ellis et al. (2016); he was politically active and his motivation was related to the immigration policy and the defense of his country from Islamism. The second case is the 2016 vehicle attack in Nice by Mohamed Lahouaiej-Bouhlel (hereafter referred to as the Bouhlel-attack). On the 14th of July, the French holiday Bastille Day, Bouhlel drove a cargo truck
into the crowds of people celebrating on the Promenade des Anglais. 86 people died during the attack, while 458 people were injured. Bouhlel was shot and killed by the police. Investigation shows that Bouhlel was a radical Muslim, who was interested in the extreme jihadist movement. Europol classified the attack as jihadist terrorism (Europol, 2017). Opinions on whether Bouhlel can be classified as a lone actor are divided: after the attack, Islamic State claimed the attack as they did after many other attacks. However, no clear connections between Bouhlel and IS or any other Islamic terrorist group have been found (Bloom, 2016). Therefore, the case has been regarded as a lone-actor terrorist attack, also by scholars (Perry, Hasisi & Perry, 2017).
In a “most-similar” comparative design, it is necessary for the cases to be similar enough across relevant variables in order to draw valid conclusions (Schaefer, 2003). Entman (1991) notes that, when analyzing news coverage, a comparative case study can help to discover inevitable biases and framing that
journalists use to form a narrative. However, Schaefer (2003) mentions that “the cases do not have to be identical but merely have ‘comparable journalistic potential’ (Entman, 1989)” (p. 94). Even though the modus operandi of both attacks were different, these particular events were selected because they can be
regarded as similar for journalistic purposes. Both cases are considered as two of the most extreme lone actor attacks to date in Europe: “in terms of effectiveness, Anders Breivik as well as the perpetrator in Nice, Mohamed Lahouaiej-Bouhlel, are outliers in the data set producing a lot more fatalities than other lone actors” (Schlegel, 2017). Since one of the factors that predicts media coverage as well as the prominence of the coverage seems to be the amount of fatalities and injuries, the two selected cases thus share the same definite newsworthiness (Weimann & Brosius, 1991). Moreover, both attack were innovative during the time of the incident. The Breivik attack was one of the first massive and deadly terrorist attacks in Europe since the bombings in London, and on top of that targeted and killed teenagers on a summer camp. The attack in Nice was one of the first major vehicle-ramming attacks in Europe, and targeted all kinds of people as opposed to one certain group. Thus, both cases were shocking and had characteristics that were new at the time of the attack. Therefore, both attacks received a lot of media attention in the Netherlands, as well as worldwide.
3.2.2 Sampling of Dutch Newspapers
Newspapers were selected as the type of media to analyze for two reasons. First, newspapers are easily accessible and have often been used in previous studies on framing. Second, Schaefer (2003) notes that “newspapers tend to have more in-depth content than many other media” (p. 94). This means that there is more opportunity for frames to emerge in this in-depth coverage.
The five Dutch newspapers that have been selected are de Volkskrant, de Telegraaf, het Algemeen
Dagblad (AD), Trouw and NRC Handelsblad (NRC). These five national newspapers were chosen
because they rank among the highest countrywide circulation and are accessible both in print as well as online (Bakker, 2018). Printed newspapers in the Netherlands reach 44% of Dutch citizens of 13 years and older on a daily basis. Together, the five selected newspaper reach almost 30% of Dutch citizens.
Furthermore, the selected newspapers are representative for the newspaper landscape in the Netherlands:
De Telegraaf is a more popular and sensationalist newspaper, de Volkskrant is a serious newspaper for
professionals, the NRC is a serious paper without popular stories, Trouw pays relatively more attention to religion and backgrounds, and the AD is easy to read and aimed at a broad general audience (Bakker & Vasterman, 2008; Boukes & Vliegenthart, 2017; Semetko & Valkenburg, 2000). Generally, the Telegraaf and the AD are considered to be “popular” newspapers, while the Volkskrant, Trouw and the NRC are regarded as “quality” newspapers (Boukes & Vliegenthart, 2017). Choosing a relatively wide range of newspaper types increases the amount of people that are reached by the news stories and who might be susceptible to the specific framing of the attacks.
The articles used in the analysis were retrieved by searching for different keywords in the headline and leaders in LexisNexis. For the Breivik attack, the keywords ‘Breivik,’ ‘Oslo,’ ‘Noorwegen’ and ‘Utoya’
were used. The last three keywords are important in order to create a complete overview of articles, since in the initial response after the attack it was not clear yet who the perpetrator was. For the attack in Nice, the terms ‘Nice,’ ‘aanslag’ and ‘boulevard’ were used, as well as a part of the last name of the attacker, since the pilot study showed that the media called the attacker ‘Bouhlel’. The choice to make use of the LexisNexis research database to retrieve the articles for analysis was made because the database has a reputation of quality, is easily accessible, and contains a comprehensive archive of articles in all major and minor newspapers of the Netherlands.
3.3 Method
To examine the way lone actor attacks are framed, this research used a comparative two-case study design, in which it will focus on two extreme lone actor attacks in Europe. In order to examine the selected articles, the research used a discourse analysis. The choice for qualitative research was made because it allows the researcher to conduct an in-depth analysis of the news frames of the two lone actor attacks. Even though the main method of this study is qualitative research, the current study used
quantitative research as well in order to answer sub-question 3: a chi-square statistical test was performed using SPSS to examine whether the differences found were statistically significant.
There are several approaches to perform a discourse analysis with regards to framing; the most important ones are the deductive approach and the inductive approach (Semetko & Valkenburg, 2000). The
inductive approach focuses on the analysis of news articles without any predefined frames in mind. Thus, frames surface during the course of analysis. An advantage of this approach is that it potentially identifies other frames than the ones already discovered in previous research. The deductive approach makes use of predefined frames while analyzing a text. The main advantage of this approach is the transparency that it offers.
The analysis in this study used a combination of the above-mentioned approaches. By relying on previous empirical studies on the framing of terrorism in the media, predetermined frames were established and kept in mind during the inductive phase. During the analysis, an additional frames were created inductively2.
Pilot Study
Before the start of the analysis, two short pilot studies were conducted. The objective of the first pilot study was to determine the time span of the articles that were analyzed. In the pilot study, the volume of the news coverage on the Breivik-attack in the selected newspapers was examined. By searching for
2 Since the current study predominantly relied on inductively found frames, it is difficult to form hypotheses. It was
multiple keywords in the headline & lead of the articles, the results showed that the peak of articles on the Breivik-attack took place four days after the attack. From day 5, the number of articles started to decrease, with the biggest decrease after 10 days.
Following the pilot study, it was therefore decided to analyze articles that appeared in the newspapers within a time span of ten days, in order to gather most of the relevant news articles. This means that for the Breivik-attack, articles published between July 22, 2011 and July 31, 2011 were consulted. For the Bouhlel-attack, the timespan of the articles was between July 14, 2016 and July 23, 2016. Besides increasing the feasibility for this research and the fact that this period contains the most concentrated coverage of the events, this time span also still allowed the research to focus on the initial response to the terrorist events. Due to the fact that uncertainty is a big factor in this initial phase, it is likely that framing takes place in this stage specifically. The study concerns both episodic and thematic frames, as it looks at the initial response in the couple of days after the attacks (episodic), as well as the articles that are published up until ten days after the incidents (thematic).
A second pilot study was conducted in order to test a number of predetermined frames that were created deductively by reviewing previous studies on news framing of terrorism. A total of 10 news articles (one per newspaper for each attack) were randomly selected and coded. This deductive process ultimately lead to a set of four main coding categories based on studies by Powell (2011) and Morin (2016): (1) the perpetrator, (2) the incident, (3) the victims, and (4) the subsequent discourse.
Coding Procedure
In order to collect the volume of articles that was needed for analysis, the timespan of the selected articles was determined in the first pilot study. After retrieving all the available articles published in the ten days after the attacks, the selection was further filtered based on the content of the articles. Opinion articles, letters to the editor, interviews and commentaries were excluded from the body of articles that was
selected for analysis, as well as duplicate and irrelevant articles. The articles that were selected all covered either the attack or the perpetrator, and the main criteria was that the attack or the perpetrator were the incentive for writing the article. Therefore, the focus of the analysis is more on framing of news reporting that is considered ‘factual’ instead of opinion pieces. The final sample consisted of 178 news articles published in the ten days after both attacks. The majority of the articles covered the Breivik attack (103 articles 58%), while 75 news articles covered the Bouhlel attack (41%). The division of the articles amongst the different newspapers can be found in table 1. Furthermore, the complete list of the analyzed news articles is included in the appendix.
Table 1
Division of Newspaper Articles per Newspaper per Case
Newspaper Type Newspaper Bouhlel-case Breivik-case
Popular Newspapers ‘t Algemeen Dagblad 14 24
De Telegraaf 15 17
Quality Newspapers NRC Handelsblad 19 19
Trouw 14 23
De Volkskrant 13 20
Total 75 103
The analysis consisted of two levels: the coding categories and the frames. On the first level, all
information was coded in the selected articles that related to one of the four coding categories established in the second pilot study: “the perpetrator,” “the incident,” the victims,” and “the subsequent discourse.” In order to find all the relevant information, all selected articles were read through multiple times by the researcher, and all words or phrases that referred to one of the four coding categories were highlighted per category.
Table 2
Explanation of the Coding Categories
Then, the second level zoomed in on the different frames used in order to describe the perpetrator, the incident, the victims and the subsequent discourse. These frames were created both deductively as well as inductively. Ultimately, the following frames were found: the us versus them frame, the personal
perpetrator frame, the terrorism frame, the human interest frame and the control & surveillance frame.
Coding Category Explanation
The Perpetrator All the information in the news articles that refers to the perpetrator: description of the perpetrator, the labels used to refer to the perpetrator, additional information on his background, family, social relations, mental health and motivations.
The Incident All the information in the news articles that concerned the incident: mainly the labels that are used to refer to the incident, and the description of what exactly happened.
The Victims All the information in the news articles that discusses the victims: eye-witness stories, response of the community, and stories of surviving victims of the attack.
The Subsequent Discourse All the information that concerned the discourse that the attack sparked: why did the event take place, how was it possible, how can it be
Table 2 and table 3 show the meaning of the coding categories and the specific frames. In some articles, multiple frames were found; in others, only one frame was present.
Table 3
Explanation of the Corresponding Frames
Lastly, for some frames, sub-categories were found that addressed a specific topic related to the frame. These sub-categories allowed for an even more in-depth comparison of the framing of the two cases. Under the personal perpetrator frame, three sub-categories were distinguished: the label that was used to refer to the perpetrator, the motivation of the perpetrator, and the characteristics and mental health of the perpetrator. For the terrorism frame, two sub-categories were established: the labelling of the incident, and the connection to domestic and international terrorism. For the human interest frame, only one
sub-category was added: the labels used to refer to the victims. Lastly, two sub-categories were established under the control & surveillance frame: the political climate and the future threat. Table 4 shows the meaning of the sub-categories, while table 5 summarizes the different levels of analysis.
Table 4
Explanation of the Sub-categories Belonging to the Frames
Frames Sub-categories Explanation
Us versus Them - -
Personal Perpetrator Frame
Labelling All the labels used to refer to the perpetrator.
Motivation All the information that identifies the motivation of the perpetrator to commit the attack according to the news article.
Characteristics & Mental Health
All the information on the characteristics of the perpetrator, including his psychological health. Terrorism Frame Labelling All the labels used to refer to the incident. Corresponding Frames Explanation
The Us versus Them Frame Distinguishes between “our” norms and values (often Western values), and the ones of the perpetrator or his religious or political background. The Personal Perpetrator
Frame
Discusses the personal life, mental health and characteristics of the perpetrator, as well as quotes of family members or neighbors. The Terrorism Frame Describes the act as a terrorist attack or relates it to terrorism. The Human Interest Frame Refers to personal stories regarding the incident; attack details by
witnesses and victims, as well as the response of the community. The Control & Surveillance
Frame
Refers to measures that can or should be taken in order to prevent another attack from happening. Also refers to criticism on police or intelligence-agencies and the implications of the incident for the political situation of the country.
Domestic/International Terrorism
Information that identified the incident as either
domestic terrorism (the incident is random and isolated) or international terrorism (the incident is connected to international terrorist organizations).
Human Interest Frame
Labelling All labels that referred to the victims. Control &
Surveillance Frame
Political Climate Information that referred to the political climate related to the attack (policies on immigration and
radicalization).
Future threat All discourse that concerned a future threat.
During the whole coding process, all articles were analyzed and manually coded using Atlas.ti. Since the coding was done by only one researcher, intercoder reliability does not have to be accounted for. The results of the framing analysis will be presented in the next section of this thesis, and will be compared per incident in order to find out whether the events were framed differently by the news media.
Table 5
The Different Levels of Analysis: Coding Category, Frames and Sub-categories
Coding Category Corresponding Frames Sub-categories
Perpetrator Us versus Them Frame -
Personal Perpetrator Frame Labelling of the Perpetrator Motivation
Characteristics & Mental Health Incident Terrorism Frame Labelling of the Incident
Domestic/International Terrorism Victims The Human Interest Frame Labelling of the Victims
Subsequent Discourse The Control & Surveillance Frame Political Climate Future Threat
4. Results
Following the analysis of the articles, differences between the frequencies of the main coding categories per case were revealed. Table 6 shows the frequency of each coding category in the news coverage on each case. Even though the same percentage of coverage concerned the subsequent discourse, there are differences between the amount of coverage on the perpetrator, the incident and the victims: whereas the coverage on Bouhlel focused more on the incident and the victims, the Breivik-case cast more attention to the perpetrator. The results section will take a closer look at these differences and similarities by zooming in on the different frames used in each category, as well as the sub-categories belonging to the frames. Table 6
Division of the Frequency of the Coding Categories per Case (Percentage of Total Articles per Case)3
The following section is divided into four parts, corresponding to the sub-questions that were posed in the introduction. The first part will explain the frames used by the five Dutch newspapers to describe the two lone actor attacks. The second part will discuss the differences and similarities in framing that were found between the two cases, followed by a discussion of the differences in framing by popular and quality newspapers. Lastly, a number of explanations for the identified differences and similarities will be presented.
4.1 “Which frames were used by the five Dutch newspapers to describe the two lone actor attacks?” In order to assess the framing of both cases and the differences and similarities between them, this section will first address the frames that the Dutch newspapers used to describe both cases. Firstly, the frames per coding category will be discussed for the Bouhlel-case, followed by the frames used in the coverage of the Breivik-case. The next section then will look further into the differences and similarities between the two cases. Table 7 shows the division of the frames per case.
3 All percentages in the tables in this section refer to the percentage of the total number of articles per case in
which the frame was present.
Coding Category Bouhlel-case Breivik-case
Perpetrator 26 (35%) 48 (47%)
Incident 18 (24%) 14 (14%)
Victims 24 (32%) 16 (16%)
4.1.1 The Bouhlel-case in Nice
The most dominant frame identified in the coverage on the Bouhlel-attack was the control & surveillance
frame. This frame was reflected in three types of narratives: (1) criticism on the police and the government
Table 7
Division of Frequency of the Frames per Case (Percentage of Total Articles per Case)
for not taking enough measures and not being able to prevent the attack, (2) a more general discussion of future threats and measures in order to prevent another attack from happening and (3) the implication that France is at war with terrorism.
As for the first type of narrative, a number of articles mentioned criticism on the police. Some articles discussed the fact that the police and the government were accused of only employing one police car and 64 policemen in order to take care of the security of 40.000 people (Jager, Trouw, 2016e). Moreover, the French minister of Home Affairs was accused of lying about the fact that the national police was guarding the promenade, despite the statement of the government that the national police was present a couple of hundred of meters away from the boulevard. Besides criticism on the police with regards to this specific event, the attack also triggered much discussion on the government in general and their counterterrorism policies (Vermaas, NRC, 2016b; Schoonhoven, Telegraaf, 2016). While after the attacks in Paris in 2015, the country came together as one after president Hollande called for unity, after the Nice attack “his words start to become worn out” (Vermaas, NRC, 2016a).
The second type of narrative relates to the discourse on future threats. Some articles discussed the threat of the truck as a weapon of attack, as well as the measures that could and should be taken in order to prevent another vehicle attack from happening (Bahara & Feenstra, Volkskrant, 2016; Beemsterboer, NRC, 2016). Radicalization was also often mentioned in relation to future threats: many articles discussed how France, and in particular Nice, is a breeding ground for radicalization and Syria-goers (Bijlsma, Telegraaf, 2016; Vlasblom, NRC, 2016). One article discussed the way that young men radicalize and are persuaded to go to Syria by a well-known jihadist in specific neighborhoods in Nice (Bijlsma, Telegraaf, 2016). Most of the radicalization discourse was also related to the threat of international terrorist organizations, iFn
Coding Categories Frames Bouhlel-case Breivik-case
Perpetrator Us versus Them Frame 11 (15%) 3 (3%) Personal Perpetrator Frame 15 (20%) 45 (44%)
Incident Terrorism Frame 18 (24%) 14 (14%)
Victims Human Interest Frame 24 (32%) 16 (16%)
particular Islamic State. Other articles discussed the European counterterrorism strategy, the threat of lone-wolf terrorists and the security at events in the Netherlands.
The last type of narrative refers to the implication that France is at war with terrorism: politicians mainly use war-terms to describe the attack (Vermaas, NRC, 2016b). It also refers to tensions in the French society. It was often mentioned that France is strongly polarized; if right-wing extremists would take revenge on Muslims for this attack, then that could be the start of a civil war in France (Jager, Trouw, 2016d; Vermaas, NRC, 2016c).
Another dominant frame is the human interest frame, which gives a personal spin to the stories. Many articles contained statements of what happened given by eyewitnesses, as well as stories about the victims. A number of articles emphasized that many of the victims were children, for example an article called “The children’s hospital is full of drama’s that are difficult to process”4 (Jager, Trouw, 2016a).
Furthermore, this frame also placed some focus on the Netherlands by specifically covering the stories of Dutch eyewitnesses.
Most instances of the terrorism frame linked the incident to international terrorist organizations such as the Islamic State or Al-Qaeda. Two days after the attack it was still unclear whether Mohamed Bouhlel acted on his own, whether he was only inspired by the ideas of the Islamic State or whether he had committed the attack on behalf of Islamic State. Nonetheless, after the attack many newspaper articles immediately questioned whether there was a link between the perpetrator and a terrorist organization, especially because the attack was carried out with a vehicle; a method that the leaders of IS and Al-Qaeda have promoted in the past. Moreover, many articles did not only suggest a link between the incident and international terrorist organizations, but also discussed the possible reasons for an attack on France soil by the Islamic State (“Crimineel Ja,” Trouw, 2016; “Aanslag met Truck,” NRC, 2016). They also compared Bouhlel to other lone actor terrorists that were inspired by Islamic State: “Bouhlel is the result of the ‘freelance jihad,’ just like Mateen and Abballa. These jihadists do not receive orders from the Middle-East, but are inspired by the call of IS to kill as many Western people as possible (..)” (Giesen,
Volkskrant, 2016b). Moreover, there was only one article that mentioned ‘domestic terrorism;’ this article, just like the others, questioned whether Bouhlel acted on behalf of Islamic State or out of his own
initiative, but concluded that “in any case, it was homegrown terror, just like Paris and Brussels”
(Rosman, AD, 2016). However, domestic terrorism is often presented as an isolated attack, which was not the case in the coverage on Bouhlel.
