Digital aftermaths of terror

(1)

Digital aftermaths of terror

Reactions to terrorist attacks on Twitter

Moa Eriksson Krutrök

Department of Sociology Umeå 2020

(2)

This work is protected by the Swedish Copyright Legislation (Act 1960:729) Dissertation for PhD

ISBN: 978-91-7855-255-9 (print) ISBN: 978-91-7855-256-6 (pdf) ISSN: 1104-2508

Cover: Jennie Hägglund & Joakim Bergqvist

Electronic version available at: http://umu.diva-portal.org/

Printed by: CityPrint i Norr AB Umeå, Sweden 2020

(3)

Abstract

Background. This thesis explores digital public reactions to terrorist attacks, and specifically how discourses on terrorism become articulated on social media.

The overarching aim of the thesis is to explore the reactions spread on social media following attacks, and how these digital platforms may alter or transform the ways in which such attacks are collectively understood and interpreted. It explores issues relating to Twitter as a platform for backchannel meaning making, for downplaying fear mongering practices, the ways different terrorist attacks become co-articulated, as well as the divided attention of networked publics during terror-related events.

Methods. The thesis is made up of four individual papers, each exploring digital reactions to different terrorist attacks on Twitter. The cases include the Utøya massacre in Norway 2011, the Stockholm lorry attack in 2017, 12 different attacks occurring between 2015-2017 in Europe (multi-case study), and the terrorist attack in Sinai, Egypt and the subsequent false alarm about a suspected terrorist attack on Oxford Street Underground station in London, UK, occuring on the same day in 2017. A combination of methods used in this thesis include network analysis of hashtag co-occurrences, and thematic analysis of prominent themes in tweets, using discourse theory as its analytical framework in the qualitative readings.

Results. The findings of the thesis suggest that discourses on terrorism on social media are shaped by the specific logics present in the connective action of networked publics. The political action of digital audiences is performed in a personalised way that shapes terrorism discourse. How terrorism is understood is influenced by collective ideas of how societies become affected by trauma or resilience, and digital communities engage in downplaying or amplifying practices for such articulations. Prevailing spatial and temporal contexts also matter for the ways in which terrorist attacks are understood and reimagined.

The (real or perceived) closeness to the attack, the interrelatedness of attacks, and aspects of virality of information plays a role in how terror discourses become articulated in digital spaces.

Conclusion. This thesis makes theoretical and empirical contributions to the understanding of how digital audiences articulate terrorism discourse in the aftermath of terrorist attacks. It stresses that emotive personal expressions on social media should be understood as a highly politicised reaction, relating to a vast range of issues, such as immigration control, nationalism, or an increased prevalence of everyday racism. On social media, audiences may both amplify or downplay attention to specific terrorist attacks, and, as such, may choose to provide (or not provide) attention. Social media may thus function as a space for everyday political action in times of terrorism.

Keywords. Social media, terrorist attacks, discourse, networked publics, Twitter.

(6)

(7)

List of original papers in the thesis

I. Eriksson M (2016) Managing collective trauma on social media: the role of Twitter after the 2011 Norway attack. Media, Culture & Society. 38(3):

365-380.

II. Eriksson M (2018) Pizza, beer, and kittens: Negotiating cultural trauma discourses on Twitter in the wake of the 2017 Stockholm attack. New Media & Society. 20(11): 3980-3996.

III. Eriksson Krutrök M and Lindgren S (2018) Continued contexts of terror:

Analysing temporal patterns of hashtag co-occurrence as discursive articulations. Social Media + Society. 4(4): 1-11.

IV. Eriksson Krutrök M, Lindgren S and Merrill S (Manuscript) Divided attention: exploring socio-technical factors in responses to terror on Twitter. Submitted to journal.

(8)

(9)

Acknowledgments

Writing a thesis may seem lonely at times, but I’ve been given an immense amount of support during this process. I was lucky to be able to share my PhD journey, an office and a PhD podcast with Sara Kalucza from day one. Thank you for being there in good times and bad, both in front of, and beyond, the microphone. My best office buds Mathilda Åkerlund and Meffrey Jitchell have provided my last years as a PhD candidate with so much laughter and plenty good beers after hours. Thank you both for being your amazing selves and Jeff, for misspelling my name in your acknowledgement, sike! A huge thank you to everyone in the PhD gang at the department who have supported me throughout, and especially Sanna Hardell, Camilla Carbin and Magdalena Sjöberg for sharing the PhD mom experience. Thank you Sofia Wård, Gunilla Renström, Erika Antill and Maria Abrahamsson in the administrative staff for all the technical support and Daniel Larsson for providing me with teaching opportunities.

This PhD project was funded by the Marcus and Amalia Wallenberg Foundation, without which none of this would have been possible, and I would never have been offered this position in the first place. For this, I am immensely humbled. I was lucky to partake in different research collaborations during my time as a PhD candidate. I have been a proud member of the board of DIGSUM, the Centre for Digital Social Research in Umeå, since its start in 2015. It continuously brings together researchers from adjacent fields who have inspired me throughout my time as a PhD candidate, and who continue to inspire me. Additionally, the project Advancing Social Media Studies brought together The Communication and Media Research Institute (CAMRI) at Westminster University in London and DIGSUM, and connected me with an amazing team of researchers. This provided me with great insights to pressing digital research issues.

As a PhD candidate, I’ve attended courses that have helped me in shaping this thesis. Some of these courses have been offered at other universities. First, the Essex Summer School in Social Science Data Analysis at University of Essex provided me both with an increased understanding of methods, and a social setting that helped immensely in my first year as a PhD candidate. Second, the CCI Digital Methods Summer School hosted by the Digital Media Research Center (DMRC) at Queensland University of Technology (QUT) gave me tools to think broader about the potential of digital methods. Third, I attended a course in Discourse Theory and Methods at Uppsala University at the same time as I restarted the work on this kappa (introductory chapter), which helped immensely in ways to rethink some important concepts. Thank you for this.

(10)

A sincere thank you to my parents and siblings who have supported me immensely in this process. Mom, you have been my number one cheerleader from day one, and you’ve kept telling me that ‘if anyone can do this, it’s you’, which has meant so much for me to hear in times of doubt. Thank you, dear friends who have comforted me in times of need and given me space to completely fall apart in order to keep on going. Thank you Jennie Hägglund, my fellow härjajämtska, for the beautiful artwork on this cover. An extended thank you to all in HärjaJämt for supporting me in my PhD and life endeavors, especially Frida Kjellin who told me back in 2013 that ‘the next PhD defense party I’ll attend will be yours’ while we were still working as kitchen staff at defense parties, and at that point, I couldn’t dare to dream about that. I still can’t believe in any form of celebration after this whole thing is over due to covid-19, of course. But that’s a whole other story.

I’m grateful to Samuel Merrill for pushing me to achieve more than I thought I was able to as my midway seminar reader. Even though you made me realize I had to rework and start anew on oh so many things (gör om, gör rätt), without your insights, this thesis would not have become what it did, so sincerely thank you Sam. My readers Anna-Britt Coe and Malcolm Fairbrother provided me with great comments in the very end, which was much appreciated. Thank you Nelly Wahlfort and my older and wiser sister Li for your meticulous proofreading skills, and Joakim Bergqvist for your last minute design support. Thank you Johanna Sumiala for being the perfect reader at my end seminar. Your attentive reading is what really pushed me across the finish line, and your comments mattered so much for how this final product ended up.

Thank you Ragnar Lundström, my co-supervisor, for keeping me grounded and for always seeing the big picture when I become nearsighted. You’ve taught me not to sweat the small stuff. Thank you for always having my back and being my emotional anchor during my time as a PhD candidate. My deepest gratitude is directed to my supervisor Simon Lindgren for persuading me to accept the offer of this PhD position when I almost chickened out. Thank you for believing in me, you always have, even when I’ve doubted myself the most. Thank you for letting me trial and error, get back on the horse, and try again. Thank you for being such an inspiration and mentor and for being my professional safety net. If not for you, I would never have moved to Umeå, written this thesis, found my husband and become a mother two times over.

And on that note, thank you, thank you, thank you my lovely Viktor for being my best friend, husband and co-parent. We’ve always been each other’s number one fans and you truly make me become my best self. I love you. And to Zakarias and Allan, thank you for giving the loveliest hugs when I’ve needed them the most.

Ok, mommy’s done now.

(11)

Introduction and aim

A series of high-profile terrorist attacks has set the prevailing threat of terrorism front and centre in the public consciousness in Europe in recent years. The public reactions after attacks such as the Utøya massacre, the Brussels suicide bombings, the Charlie Hebdo attack, and the Manchester Arena bombing, only to name a few, have been vast and diverse, and with the increased importance of social media in individuals’ lives, these reactions have been prominent on social media platforms. While terrorism research has focused primarily on the societal causes and outcomes of terrorism, studies on the societal upheaval and public reactions in the aftermath of terrorism has not been given as much attention. Truc (2018) has proposed that, what he calls, a ‘sociology of terrorism’ should extend the research on terrorism onto questions concerning what connects us as collectives in the aftermath of terrorist attacks, including the digital expressions of such collective responses.

On social media, networked publics (boyd, 2010) engage in public discourse on pressing issues, such as the prevalence of terrorism. Such publics collectively shape, as well as dispute, ideas on how terrorist attacks should be understood and interpreted, and, as such, social media platforms provide an important space for both information dissemination and articulation of networked discourses on terrorism. This may shape collective ideas on other societal issues such as public policy, foreign affairs, or immigration, to name a few. In this way, networked discourses on terrorism may have political, as well as social, outcomes.

While social media platforms have become important spaces for relating to, and negotiating, the issue of terrorism, they have also become a space for spreading rumours and unverified information in times of crisis, making them unreliable at times. Terrorist attacks are acute events, which are often initially characterised by a scarcity of information as the news is still breaking, creating ‘information holes’ which networked publics often attempt to fill in real-time (Liu, Fraustino and Jin, 2016). These unverified accounts shared on social media may spread and become included in news updates from more conventional news streams, for example where tweets get included in the ‘news tickers’ of live television news broadcasts, and newspapers may integrate Twitter feeds into their websites as part of their unfolding reporting of such events (Jang et al., 2019; Loke and Grimm, 2017; Mare, 2013; Simons, 2016).

The viral character of social media content, however, makes it a potential liability in the spread of unverified, even misleading, information. In addition to this, information spread on social media may be biased in a number of ways. In the

‘attention economy’ (Goldhaber, 1997) present in digital spaces, information on

(12)

specific terrorist attacks is competing for attention on social media platforms. As such, information may be overflowing in the midst of certain attacks while not providing such attention to other attacks. This amplification of certain information over others is complex, as algorithms may push specific information to become more visible. However, this is also done by social media users themselves in their tweeting (and retweeting) practices. Attention biases are additionally caused by the (real of perceived) closeness to specific terrorist attacks, which may shape the reaction on social media platforms. Such closeness may both be spatial, where attacks in our vicinity is considered more important, and temporal, where attacks are understood in practices of referencing back to previous terror incidents. These contexts matter to the ways terrorism is politicised and understood.

This thesis aims to explore the reactions’ spread on social media following terrorist attacks, and how these digital platforms may alter or transform the ways in which such attacks are collectively understood and interpreted. As such, these reactions are understood within this thesis as different articulations of terrorism discourses. The thesis focuses on two research questions, which will be addressed through the thesis’ articles. These are formulated as follows:

- How are terrorism discourses articulated on social media following terrorist attacks, and how do these articulations relate to prevailing societal discourses on terror?

- How are spatial and temporal contexts shaping social media articulations of terrorism discourses after terrorist attacks?

This thesis takes a case studies approach, looking specifically at a selection of different, mostly European, terrorist attacks which have taken place in recent years. Beginning with a duo of single case studies on the Utøya massacre in Norway 2011 (Paper 1) and on the Stockholm lorry attack in 2017 (Paper 2), focus has been placed on the discourses forming online following these specific attacks.

These articles are followed by two cross-event studies, first, looking at 12 different attacks occurring between 2015-2017 in Europe (Paper 3), and, finally, a study focusing on the Sinai attack and the false alarm on Oxford street in London occurring on the same day in 2017 (Paper 4). These four studies point to different aspects of these digital public reactions to terrorism regarding, for example, the use of Twitter as a platform for backchannel meaning making, for downplaying fear mongering practices, the ways different terrorist attacks become co- articulated, as well as the divided attention of networked publics during terror- related events.

Methodologically, this thesis uses a combination of methods, such as network analysis of hashtag co-occurrences and thematic analysis of prominent themes

(13)

within the tweets themselves, in order to develop a more detailed understanding of the discursive articulations taking place on social media following terrorist attacks. In the four specific studies, different cases of terrorist attacks are analysed, where social media has been prominent platforms in terms of shaping how these attacks have become discursively articulated. For ethical and practical reasons, Twitter was chosen as the sole platform used. The one-platform focus allows for greater uniformity in the studies, but also its own focus due to its platform specificity. In relation to online discursive articulations, Twitter has been a platform prominent for political, often heated, discussion, where few accounts are set to private settings, unlike for example Facebook and Instagram which may be used in a more private manner, closed off to those outside of the individual’s personal network. As such, the content published on Twitter is more often than not intended for a larger audience, and is more focused on sharing opinions on a range of different topics, including those related to terrorism.

The different parts of this introductory chapter are organised in the following manner. In the first section, the conceptual frameworks of terrorism and social media are outlined, followed by a section on the previous research relating to the role of social media in society, and the ways in which terrorism discourses become articulated. In the third section, the analytical framework and methodology are outlined, concluding with an ethical discussion on social media research. The fourth section presents the results of the thesis, as presented in the four articles, after which the articles themselves are described in brief. The fifth and final section of this introductory text presents the discussion and concluding takeaway of the four individual studies and their implications as a whole. Directly following this introductory chapter, the reader will find the four papers of this thesis in full.

(14)

Key concepts: social media and terrorism

While social media has been given an increased role in contemporary society since the early 2000s, so has the prevalence of terrorism in the Western world been given an increased public attention, as both large-scale attacks have increased in Europe, and is ever-changing in range and form since the September 11 attacks in 2001. Both social media and the current state of terrorism could be considered ‘new’, but what are their actual implications for society? The following parts will provide such a critical perspective on these two concepts by, first, providing an introduction to social media’s role in society, followed by a critical review of the concept of terrorism in previous research.

The role of social media in society

While this thesis focuses on Twitter as the sole social media platform for public digital reactions to terrorism, the term ‘social media’ may imply a wide range of platforms with their own specificity and features. Social media is a term which tends to be limited to platforms in which social networks are created, upheld and maintained, such as Facebook, Twitter, Instagram or LinkedIn (or older examples such as MySpace or mIRC). Twitter, specifically, is a micro-blogging platform which allows users to ‘tweet’, i.e. publish texts containing 280 characters (double from the previous 140 characters up until 2017) that has grown substantially in recent years and is now an important tool for publicly sharing information and opinions about politics and news events. Social media communication has been understood by Castells (2009) as mass self- communication, i.e. both self-regulated through individuals’ capacity to make conscious choices about which content is being consumed and the ways in which it is shared and spread, as well as being a form of mass communication in larger networks of individuals. This form of digital communication has resulted in shifts in power structures of society at large, according to Castells, by increasing freedom and autonomy for actors within these networks. The question whether or not social media may have emancipatory functions or not has developed over time within the internet research community, as early scholars saw its potentials for moving beyond social classifications such as gender or race (see for example Turkle, 1995), while later scholars have stressed the fact that prejudices persist in these digital platforms, even to the point of being embedded in the platforms themselves, for example in algorithmic processes (Noble, 2018).

Generally, social media is no longer considered a ‘new’ phenomenon, but the platforms are constantly evolving. However, while the platforms may change, the idea that there is a disparity between the ‘offline’ and the ‘online’ world, is considered by most researchers a stale definition of how to understand digital spaces. When I refer to social media practices, I do this in line with Highfield’s

(15)

(2016: 7) definition where digital articulations of political issues should not be

‘divorced from wider contextual factors which affect individuals’ and groups’

experiences of politics and the internet’, and because of this, social media and societal politics should be understood as highly interlinked. As explained by Couldry and Hepp (2016), the social world is based on various mediatisations of social reality. Through the increased use of technological devices in our communication practices, and the ways in which these platforms are accessible at any time and any place, our social realities are continuously shaped by this form of mediated communication. As such, social media platforms constitute the spaces in which these social realities are articulated, contested, and upheld.

According to Couldry and Hepp (2016: 30), we are currently living in an age of

‘deep mediatisation’, where our social worlds are interconnected with technologically driven communication, and

[a]s a result, the ways in which we make sense of the world phenomenologically become necessarily entangled with the constraints, affordances and power- relations that are features of media as infrastructures for communication.

This thesis is interested in how digital publics construct, negotiate and upheld discourse on terrorism in the midst of specific terrorist attacks. As such, this thesis is interested in the ways in which meaning is created in digital spaces.

However, these spaces are themselves dependent on the networks of individual users and how social media content is ordered, not merely algorithmically. One such way of ordering content is the use of hashtags, where tweets are ‘tagged’ as an incentive to situate them within larger threads of connected tweets. This feature, which was initiated by Twitter users themselves, was subsequently integrated into Twitter’s architecture as a metadata indicator aimed to coordinate discussions and simplify the finding, following, and listing of messages related to a specific theme. As such, hashtags are a kind of ‘folksonomy’, a tagging system emerging from the free social tagging of information and objects. As put by Vander Wal (2007: para. 9):

The value in this external tagging is derived from people using their own vocabulary and adding explicit meaning, which may come from inferred understanding of the information/object. People are not so much categorizing, as providing a means to connect items (placing hooks) to provide their meaning in their own understanding.

Through the uses of hashtags different publics are formed, in what Bruns and Burgess (2011) calls an ‘ad hoc public’, or in the terminology of Rambukkana (2015), a ‘hashtag public’, where temporarily formed collectives consolidate over a specific issue by using these hashtags as ‘a space, an event and a network’ to engage with (Sauter and Bruns, 2015: 47). The hashtag thus enables a large

(16)

number of Twitter users, who are otherwise uncoordinated, to organise discourses around a particular event or issue. Users thus come together through an ‘ambient affiliation’ (Zappavigna, 2011), that is, by bonding around the evolving topics of interest that are rendered findable by emerging hashtags, instead of connecting users through the more traditional follower/followee networks or through formal organisations.

The public character of Twitter has made it an important platform for negotiating discourses prevalent in the mainstream media. Networked publics (boyd, 2010:

39) both constitute a ‘space constructed through networked technologies’, as well as ‘the imagined collective that emerges as a result of people, technology, and practice’. As such, these collectives are shaping discourse on political issues in digital spaces in what Maireder and Ausserhofer (2013) has called a networked discourse. Twitter has been understood as a space for so-called affective news streams that are ‘blending humour, news sharing, opinion expression, and emotion [which] is reflective of the affective patterns of interpersonal conversations’ (Papacharissi and de Fatima Oliveira, 2012: 278). In an increasingly digital life of breaking news, networked publics are able to collectively join in the discursive negotiations following acute events, and some call this aspect of digital commentary a form of hybridity in news events, where events are not only constructed through mainstream media formulations but also in the ways in which publics are engaging with these events in digital spaces (see for example Sumiala et al., 2018).

Hashtags are neither registered nor controlled, which means that they are never retired from public use. At the same time, however, the role and meaning of any particular hashtag is all but static and might change considerably over time (Bruns and Burgess, 2015). Indeed, meaning-production through and around hashtags occurs dynamically, as put by (Rambukkana, 2015: 3):

[H]ashtag-mediated discursive assemblages are neither simply the reflection of pre-existing discourse formations nor do they create them out of digital aether.

Rather, they are nodes in the becoming of distributed discussions in which their very materiality as performative utterances is deeply implicated.

But as a platform for contesting information, negotiating meaning and transmitting own ideas, this counterhegemonic function of Twitter, as an opposing power to the ways in which terrorism is collectively understood, should not be overlooked. Thus, Twitter may function as a disruptive platform against hegemonic information forming in, for example, more traditional media outlets.

Hegemonic meaning making is thus disrupted by networked publics by such decentralised and inclusive functions of digital platforms. And while social media continuously becomes undermined as a distraction to individuals in everyday life,

(17)

this disruptive power of social media should not be overlooked. In an interview with the blog Culture Digitally, Zizi Papacharissi (n.y.) says:

What is defined as distraction is simply attention to something else. Labeling a certain form of content consumption/production as distraction suggests we de facto view it as disruptive. I am not suggesting that all forms of distraction are beneficial, but I am simply arguing that we examine the cultural meaning and context of distraction more closely.

Certainly, social media content is often understood as a form of banal form of communication. Selfies and social media content with focus on everyday aspects of the self may however be an expression of larger societal issues. Highfield (2016:

10) has called this a form of everyday politics taking place on social media, which concerns the ways in which ‘everyday and banal communication online is extended to, remixed and appropriated for political themes’. As such, social media communication may point to issues beyond the surface level engagements and should be understood as a ‘means to study cultural conditions’ (Rogers, 2013:

xiv).

Networked publics’ articulations are the main focus for this thesis, but tweets are also published by other senders than members of the general public. It is important to reflect upon the diversity of those present in this multitude of voices from different parts of the social hierarchy. The heterogeneity of these digital audiences means that they are neither unified nor consistent. People may have the ability to be heard through these online spaces, but social hierarchies persist, and people who generally tend to be heard in the general society may still be the ones who are most heard in digital spaces. There is also the constant presence of non-individual voices on these social media platforms, such as corporations, mainstream media, or political organisations, which may also shape, negotiate, and influence these networked discourses. These voices are present on Twitter, but they are not the focal point of this thesis. However, due to the persistence of social hierarchies, digitally shaped discourse should not be seen as unanimously bottom-up or inherently different from discourses formed in other spaces, such as in the mainstream media.

The concept of terrorism

This thesis examines discourses concerning specific terrorist attacks on digital platforms. Terrorism, as it is outplayed through these acts of violence, lies at the heart of the study, however, by referring to specific violent acts as terrorism is in itself a form of social construction of these events (Jackson et al., 2011). In agreement with Homolar and Rodriguez-Merino (2019: 2), I too claim that ‘the labeling of violence as terrorism is a political act’. In addressing the political climate after 9/11, Derrida (Borradori, 2004: 105) also made this point, saying:

(18)

Semantic instability, irreducible trouble spots on the borders between concepts, indecision in the very concept of the border: all this must not only be analyzed as a speculative disorder, a conceptual chaos or zone of passing turbulence in public or political language. We must also recognize here strategies and relations of force. The dominant power is the one that manages to impose and, thus, to legitimate, indeed to legalize (for it is always a question of law) on a national or world stage, the terminology and thus the interpretation that best suits it in a given situation.

On a phenomenological level, my own labelling of these attacks as terrorism should therefore not be understated, and as such, the rationale behind this categorisation is important to pinpoint. Specifically, these labels have been dependent on them having been socially and collectively understood and construed as acts of ‘terrorism’. Now, this is not a clear-cut definition by any means. There is always a ‘war of words’ in relation to terrorism (Turk, 2004: 272), where ideas of terrorism play a great part in how they are discursively articulated.

However, Homolar and Rodriguez-Merino (2019) have developed a set of

‘common-sense characteristics’ of what is generally understood as terrorism, where violence against civilians, the status of the victims, the arbitrariness and context of the violence, and the political aim of the violence are all constituent parts in the ideas of what ‘terrorism’ commonly constitutes. Critics have underlined the supposed connection between terrorism studies and Western state policies, proclaiming that these studies are essentially part of a ‘terrorism industry’, where both researchers and policy makers benefit when issues of terror deterrence are given an increased attention (Jackson et al., 2011: 13).

As stated by Jackson et al. (2011: 3), ‘[t]errorism is not a self-evident, exceptional category of political violence. Rather, it is a social construction - a linguistic term or label that is applied to certain acts through a range of specific political, legal and academic processes.’ This label has often been used as a way of delegitimising oppositional warfare, while state violence itself gets overlooked, even though the type of violence, number of and characteristics of casualties, and other factors may be virtually the same. For Butler (2004), the use of the term ‘terrorist’ is equally problematic, as it gets exploited for political reasons by state actors at war with various independence movements. Jackson et al. (2011) propose that terrorism researchers see past this actor-based definition of terrorism and instead understand these acts of violence as a form of strategy, employed also by state actors. This distinction provides a basis for a hegemonic labelling of specific acts of terror as terrorism while similar violence, performed by Western state actors, is not. Often, these events are framed as acts of terrorism in the media quite early depending on their specific qualities. As suggested by Butler (2004: 4) the terrorism term is used as an articulation of a hegemonic position by state actors as sufferer of violence, and adds:

(19)

It is one thing to suffer violence and quite another to use that fact to ground a framework in which one’s injury authorizes limitless aggression against targets that may or may not be related to the sources of one’s own suffering.

As will be addressed within the studies on specific terrorist attacks, the label of terrorism is always framed and constructed in different ways in society, and, however alike these attacks may seem on paper, their implications may be completely different because of this articulatory practice. They become understood as acts of terrorism by states affected by the violence, in order to delegitimise it, and mainstream media is co-creating it as such by creating a form of media event of the incident. Liebes (1998) has called this form of broadcast a

‘disaster marathon’, where the continuation of reporting, interviews with experts and politicians speculating on all aspects of the incident, and the general implications of the event itself is key. As such, our perception of violent events is blurred by collective ideas of terrorism, mass violence, and acute events through these mainstream media articulations. Indeed, Derrida (quoted in Borradori, 2004: 108) rhetorically asked ‘what would “September 11” have been without television?’. The actual reporting on different instances of terrorism may also be different depending on regional contexts. In their comparative analysis of newspapers’ reporting on terrorism in the US and the UK, Papacharissi and de Fatima Oliveira (2012) found that the UK based papers’ reporting were more prone to be infused with trauma and emotion compared to those in the US.

Meanwhile, US papers relied more on US based sources of information, from government and military sources for example, while the UK selected sources from a wider and more global perspective, which ultimately helps in shaping the policy solutions to terrorism presented in the US. A study by Nellis and Savage (2012) also shows that consuming news of terrorism may influence individuals’

perception of the threat of terrorism. In light of this, the discourses on terrorism in relation to different terrorist attacks are formed by mainstream media articulations as well as in the general public, and this ultimately shapes the ways in which such attacks are understood and historicised.

(20)

Networked discourses on terrorism

This thesis is concerned with the ways networked publics react to, and negotiate the significance of, terrorist attacks on Twitter. These reactions may be shaped in a number of ways, and the following two sections of this introductory chapter deals with the existing body of research on these types of reactions, in relation to the two research questions specified above. The first section deals with the articulation of terrorism discourses, and the way it may become negotiated in digital spaces. The second section deals specifically with the ways in which spatial and temporal contexts matter for these types of articulations of terrorism in relation to terrorist attacks.

Articulations and negotiations of terror discourse

In the midst of terrorist attacks, networked publics react to the news of terrorism on social media as spaces for, for example, finding information, sharing condolences, and coming together as a collective. An important feature of social media during crises is sharing information and connecting with others (Burnap et al., 2014; Gupta & Kumaguru, 2012; Liu et al., 2016; Oltenau et al., 2015;

Buntain et al., 2016) and social media has been studied after, for example, the Queensland floods (Bruns et al., 2012) and the Christchurch earthquakes (Bruns and Burgess, 2013). When studying how social media is used during disasters, Liu et al. (2016) found that people seeking information online about disasters did so with the intent to connect with others, in addition to following government updates about possible evacuations. Pressure has increasingly been put on social media sites to prevent them from becoming a space for sharing, for example, live streamed videos of attacks, such as during the Christchurch mosque shootings (McKenzie, 2019), and social media users have indeed used these platforms for critiquing mainstream media for their reporting on terrorism (Lazreg, 2019).

These reactions consist of the affective responses to the specific attacks, as well as by the prevailing political contexts. Terrorist attacks may constitute a shock to society as a whole, however, the reactions to terrorism may have other political motives. What is deemed ‘traumatic’ and not may thus be shaped by political incentives. This may also be done by state powers, as attempts to uphold positions of power. Later critiqued by Butler (2004: 29) for his incentive to start a violent conflict, George W. Bush famously addressed the nation saying after 9/11:

These acts of mass murder were intended to frighten our nation into chaos and retreat. But they have failed. Our country is strong.

(21)

As such, resilience may be expressed by officials in order to gain political leverage.

Fear, distraught or trauma may however be important for other forms of discourse, for example by right-wing populists.

Reactions to terrorism on social media may however be resilience focused as a political action against narratives of fear, such as the #ikwilhelpen hashtag after the Brussels terrorist attacks in 2016, or aimed to propose an open border policy in the aftermath of terrorism with the #portesouvertes hashtag after the terrorist attacks in Paris in November 2015 (Onorati et al., 2016). Hashtags such as these show the potential for social media platforms to function as a balance to the risk- centred and sensationalist reporting on such attacks. Digital initiatives on social media platforms may as such be understood as attempts by networked publics to downplay articulations of fear, and ultimately work as a political opposition to terrorism, which aims at striking fear in societies. The tendency for sensationalist reporting by mainstream media, and the spread of fear that such reporting creates, is akin to the goal of terrorism itself, which, through the words of Zygmunt Bauman (2006: 124), is the ultimate goal of terrorism itself:

In a vicious loop, the threat of terrorism turns itself into an inspiration for more terrorism, spilling on its way ever greater volumes of terror and even larger masses of terrorized people - the two products which terrorist acts, deriving their name from precisely such an intention, are bent on producing and plot to produce. One may say that the people who are terrorized are the terrorists’ most reliable, even though unwilling, allies. The ‘understandable desire for security’, always ready and waiting to be played on by a crafty and astute exploiter, and now whipped up by scattered and apparently unforeseeable acts of terror, proves in the end to be the main resource on which terror may count to gather momentum.

The sociological notion of cultural trauma focuses on the trauma experience as culturally and socially constructed. In the words of Alexander (2012: 3), ‘[t]he lives lost and pains experienced are individual facts; shared trauma depends on collective processes of cultural interpretation.’ How these events may be interpreted depends on different factors, such as nearness to the event, closeness with the victims, or identification with the persons represented. As described by Eyerman (2011: 12):

[A cultural trauma] can be defined as a discursive response to a tear in the social fabric, where the foundations of an established collective identity are shaken by a traumatic occurrence and are in need of renarration and repair.

However, trauma is not merely formed in the contexts of individual instances of terror. It depends both on the historical understandings present in the cultural contexts in which the crime is committed and the imagined potentials of similar

(22)

events. On his understanding of the ways in which 9/11 can be understood as traumatising, Derrida (Borradori, 2004: 97) says, ‘[t]raumatism is produced by the future, of the to come, by the threat of the worst to come, rather than an aggression that is “over and done with”.’ Given the recency of the term, it is no surprise that certain historical incidents are reformulated as ’cultural trauma’

even years after they occurred. One example is the death of the 96 football supporters at the Hillsborough stadium in the UK in 1989, which, through its depiction in mainstream media, has subsequently been labelled as cultural trauma (Hughson and Spaaij, 2011).

According to Sztompka (2000), a traumatic incident must have four specific characteristics to be formulated as a cultural trauma, which formalise the scope and structure of how these events are understood. First, the incident must appear to arise suddenly, and ended just as quickly. Second, the substance and scope of the event must be radical and deep, as it must affect the core of society, such as social values or the institutional fabric of society. The coherence of cultural life also needs to be disorganised as an effect of a traumatic incident as to be considered a cultural trauma, leading up to what Sztompka (2000: 453) calls a cultural disorientation. This affects those attached to the specific culture to lose a social and cultural tie to others, as well as affecting their patterns of acting and thinking. The third characteristic is the narrative of having been struck by trauma from something positioned outside of society. This does not imply that, for example, the perpetrator must be a non-resident, but that the incident is not considered to have been contributed by society itself, and thereby is seen as external. Fourth, and last, it must be conceived as an unpredicted and shocking event, something that the public could not have expected to happen. Events carrying these characteristics may however not necessarily evoke trauma, even though they might, and in many instances, do.

As will be later detailed in the papers in the thesis, the expressions of fear, trauma or resilience may however have deeper political incentives which may influence such reactions. As put forth by Fassin and Rechtman (2009), our contemporary ideas of trauma are influenced by political notions of solidarity and the expected outcomes of being affected by traumatic events. However, the historical construction and the political uses of the word ‘trauma’ as a natural reaction to unnatural events, may influence the ways in which victims are seen as ‘passive recipients of the label “traumatised”’ (Fassin and Rechtman, 2009: xi), independent of individual dispositions, differences in expression, as well as the nature of the event itself. As such, the affectivity in these digital expressions may not merely entail the experienced emotion of individuals in a digital space, but rather, the affectiveness of social media reactions to terrorism may be shaped by political incentive related to general discourses on political issues within different contexts.

(23)

The ways that grief is expressed on social media platforms is however often subjected to criticism. These informal norms shaped around expressions of grief on social media platforms are often a cause for debate in the midst of acute events.

Gach et al. (2017) found that emotional ‘grief policing’, i.e. the attempt to enforce ideals of how public grieving should be done, in relation to celebrity death, often circled around notions of how displayed emotionality could be understood as disproportionate in relation to the former relationship between the griever and the celebrity. This notion of legitimacy in individuals’ grief expression can also be understood as a form of ‘grief hypejacking’, i.e. the exploitation of grief narratives for the sake of becoming visible and gaining attention (Abidin, 2018). The conventions inflicted on individuals thus shape the, what Burgess et al. (2018) calls, social media rituals of mourning. The rituals present in these repsonses may include, for example, sharing song lyrics and images in the midst of the death of David Bowie rather than news updates. These rituals, thus, shape conventions on how to grieve in digital spaces.

The norms inflicted on grief expressions online reflect deeper issues of fundamental ideas of public grief and the ways in which publics are expected to react to acute events. These discussions concerning grief etiquette online, and the ways in which these digital forms of grief are expressed, tie into larger discussions relating to ideas of authenticity of self-expression. ‘Slacktivism’ is a term similarly used as a tool to disregard efforts of online communities in raising awareness and extending their activism to online spaces. But, as argued by Jackson et al. (2020:

xxxii), what they call a ‘hashtag activism’ may, instead, have ‘far-reaching influence, moving debates about identity politics, inequality, violence, and citizenship from the margins to the center and into places as crucial as presidential agendas’. Ideas of ‘grief hypejacking’ reproduce this derogatory narrative of online inauthenticity of self.

But expressions of grief may also imply additional aspects of post-terror tweeting.

In the mainstream media, ‘disaster marathons’ put emphasis on discourses of fear and social disruption, which emphasises what Glassner (1999) has called a

‘culture of fear’. Media exposure to news of terrorism may influence a societal and psychological sense of threat in a variety of ways (Altheide, 2006; Hopwood and Schuette, 2017) and while scholars have believed that an increased media literacy might reduce this sense of threat, it has not been shown to be as effective as once thought (Bergan and Lee, 2018). However, this fear may spread on social media platforms in the aftermath of terrorism, leading to a range of societal (and political) outcomes. The link between the cyberhate spread on social media platforms after incidents of terror and a perceived societal fear after the November 2015 terrorist attack in Paris has been explored by Oksanen et al.

(2018), showing that hateful communication online after terrorist attacks may contribute to a climate of fear and an increased collective uncertainty. In a report

(24)

by Gallup in 2017, general concerns about immigration in 14 European countries was found to correlate with a rise of an experienced sense of threat of terrorism, showing that threat is a political issue relating to political views of immigration issues (Crabtree and Kluch, 2017), which have been given an increased status in the European political climate in the midst of the so called refugee crisis in 2015- 2016, relating to war and terror in Syria and adjacent countries.

The way of sharing and connecting over terrorism or acute events on social media may be understood as a following the logic of connective action, as it has been outlined by Bennett and Segerberg (2012). This form of political action entails sharing personalised content on social media platforms in a connected and networked way, where ‘personalised action frames’ provide an individualised form of political expression. In contrast to collective action, individuals who are connecting over political issues in digital spaces are not ascribing to collective ideas and common identifications in general, but instead, these political action frames are individual and personalised. However, as proposed by Pond and Lewis (2019), the culturally situated logic of this form of political action should not be understated. In this sense, political action may not merely be understood by its level of connectivity.

In this thesis, the reactions on social media to the news of terrorism have been understood as a form of counteraction of a culture of fear and trauma, often used in right-wing circles for the purpose of framing issues of, for example, immigration control. However, while the second paper in the thesis is focused on this downplaying reaction on Twitter after the Stockholm lorry attack in 2017, this societal reaction is also inclusive of other accounts. For people present at the central pedestrian street where the truck hit, this was of no doubt a traumatic experience. I too was calling friends and relatives and felt scared. When researching how publics react to news of terrorism, these dimensions are always in the back of my mind when thinking about the ways in which the uncertainty affect us, and how collectives understand and interpret terrorist attacks. They are in many ways personal. They feel close. But the digital expressions of these reactions tell more than the mere personal accounts. This may further respond to why terrorist attacks which may result in several casualties and physical damages, such as the 2017 Stockholm lorry attack, may not be articulated as particularly

‘traumatic’. The contextual framing of the digital responses to terrorism is, as such, dependent on the cultural and ideological predispositions which are reflected in the connective action of networked publics.

Spatial and temporal contexts of terror discourse

While digital media spaces allow for news of terrorism to move quicker, become decentralised and spread all over the world, this does not mean that structural

(25)

divisions are non-existent in a digital society. Social media platforms reflect the society of which it is sprung, which means that they are laden with social injustice and potential harm. Whether or not social media may have emancipatory functions or not has split the internet research community, as some scholars see its potentials for moving beyond social classifications such as gender or race, while others stress the prejudices which persist in these online communication platforms, as well as the prejudices which are present in the platforms designs and embedded in algorithms (Noble, 2018).

The attention economy (Goldhaber, 1997) of social media is based on the ways in which certain issues get neglected while others are spread. Social media platforms have become important in this development, where public attention may be brought to underreported issues, such as sexual misconduct, leading up to the

#metoo movement. The specific affordances present on digital platforms, and the ways in which these platforms are used, may contribute to enhancing, amplifying, or transforming social contexts and lay the foundation for social change (Lindgren, 2017). Thus, these spaces may function as platforms for gaining attention previously not provided by mainstream media. In relation to terrorist attacks, this economy is upheld by the provided attention of networked publics to some attacks over others, thus questioning their relevance in people’s lives.

We collectively relate to terror in regard to different contextual factors, such as geographical closeness or the most recent occasions of terror which are ‘close’ in memory. Zelizer (2017: 139) has proposed that terror is understood through the lense of past events, in a kind of mnemonic scheme of terror events:

mnemonic schemes allow for a strategic broadening and narrowing of the category of terror. Particularly in response to events otherwise seen as ordinary and quotidian, on one hand, or atypical and noteworthy, on the other, memory acts as an equalizer that accommodates terror’s flexible discursive invocation. It does so, paradoxically, by playing against what is seen and thickening engagement with the event, strategically engaging cues of invisibility that introduce particular temporal flows to help explain what transpires.

Hashtags may contribute in building such mnemonic schemes, where older attacks are referenced in the conversations concerning newer attacks using the old attack hashtags, such as #berlinattack or #prayforberlin which were widely used in relation to the lorry attack at a local Christmas market in Berlin in December 2016. These types of attack hashtags can later be reiterated when newer attacks occur, as a form of discursive practice. As such, both temporal and spatial contexts matter for how terrorism is understood and articulated.

(26)

Societal reactions to terrorism become laden with emotional expressions and personalised accounts of the attacks themselves, and this expression of emotionality is both a personal and a social practice. By sending condolences, sharing grief and showing solidarity, digital communities are able to find their own social media practices for reacting to, and dealing with, terrorist attacks.

Grief expression may in this way be a form of action, which, in itself, is conditioned on cultural contexts and social conditions. But even though the loss of any life would presumably be a loss for all, this may not be the case when expressing grief after terrorist attacks. As asked by Butler (2004: 20), ‘what makes for a grievable life?’. The expression of grief over the loss of human life is not merely a personal experience of grief. It is a social response to terrorism, and through this response, attention is both given and directed on social media to specific instances of terror. As such, public displays of grief are also political acts.

As emphasised by Truc (2018) in his critique of the theory of cultural trauma, the

‘feeling of closeness’ to certain victims over others is not merely dependent on geographical closeness. He emphasises that while a spatial distance may be great, a social distance is generally not. And the ways trauma is experienced as a highly personal experience should not be understated, he says, in the ways it currently is done by cultural trauma researchers. Ultimately, this is a question of solidarity, of which an individual’s and a collective’s connectedness is difficult to pinpoint, and this may depend on cultural contexts, and may also shift over time. However, this vulnerability to some victims over others may be amplified in times of terror, Butler (2004: 39) says:

Although I am insisting on referring to a common human vulnerability, one that emerges with life itself, various terror alerts that go out over the media authorize and heighten racial hysteria in which fear is directed anywhere and nowhere, in which individuals are asked to be on guard but not told what to be on guard against; so everyone is free to imagine and identify the source of terror.

The reactions to terrorism by digital audiences to different acts of terrorism is, as such, contextually bound and not equal depending on material factors such as the perpetrators motive, association to terrorist organisations, number of casualties, or type of violence, etc. Spatial and temporal contexts matter for the ways in which acts of terror are interpreted and understood by different collectives.

Relating newer attacks to previous ones is one such practice. For example, terrorist attacks occurring after 9/11 have occasionally been called their ‘own’

9/11, for example ‘Spain’s 9/11’ after the 2004 Madrid train bombings. The ways in which attacks become unified in this way show their relational status. This relationability of attacks also cement its impact for the targeted country in a very discursive way. As specific countries are not given this same status, Truc (2018:

56) states that:

(27)

The analogy thus draws on an opposition between ‘us’ and ‘them’ and carries with it, whether we like it or not, the idea of a war of civilizations. In this respect, it expresses a Western-centred view, which contradicts the desire to promote a cosmopolitan point of view.

The disparity in attention given by networked publics to news of terrorism indicates the ways in which an international audience fails to amplify news events from other cultural contexts as well as failing to express grief over victims of terrorism. In light of this, expressing grief (and choosing not to) can be understood as a form of political action, where specific discourses of fear is employed to steer the conversation away from, or towards, specific political issues, such as immigration policy.

Acute events in general commonly generate a specific type of communication on social media, where for example intense sharing URLs is common, for the sake of updating others on the situation (Bruns et al., 2016). But social media may also become a space for spreading rumours and misinformation, and it is important to reflect upon the ways in which both technological and socio-political factors play into the spreading of false, or simply unverified information. When terrorist attacks, or other acute events, occur, they are often followed by a scarcity of information, creating what has been referred to as ‘information holes’ which publics attempt to fill in real-time on social media platforms (Liu, Fraustino and Jin, 2016). And, as put by Beckett (2016, para. 16), these accounts are integrated into the mainstream media reporting on terrorism, even though they often lack verification:

Social media amplifies the communicative scale and impact of terrorism, and it adds to the misinformation and emotional responses to terror events.

Journalists using social media as a platform or a source do not always maintain the best editorial standards. Social media has changed the very nature of news around terror, for example, by providing imagery, eyewitness accounts, and live video. But it can also deceive, distort, and distract.

While the scarcity in information is not new, the socially mediated practice of attempting to fill them in the least amount of time possible, is new, but instead of finding a scapegoat for the practice, it should be important to recognise the platforms themselves as built for creating virality. First, they provide users of an easy access for sharing content quickly using the retweet button, and second, the way content is algorithmically ordered provides a timeline where what is most amplified, gets the most attention. In the case of terrorist attacks, these two features may help in drowning out relevant information to the advantage of more accurate, but less juicy, reports.

(28)

In this way, public responses to terrorism on social media may be dependent on many factors, where closeness to the attack, the interrelatedness of attacks, and aspects of virality of information plays a role in how terror discourses become articulated.

Digital aftermaths of terror