Lone Wolves, Copycats and The Chans: How Anonymous, Far-Right Online Communities and Their Vernacular Practices Radicalize Lone Wolf Terrorists

(1)

LONE WOLVES, COPYCATS AND THE CHANS

how anonymous, far-right online communities and

their vernacular practices radicalize lone wolf terrorists

Supervisor: dr. D. de Zeeuw Second Reader: prof. dr. R.A. Rogers

17 July 2020 University of Amsterdam

Media Studies Referencing: MLA, 8th edition

Tiede Bart

MA Thesis

(2)

List of Figures

Introduction

Figure 1 Brenton Tarrant’s announcement post on 8chan 6 Figure 2 A post on 8chan linking to archived 8chan threads 11

Chapter 2

Figure 3 Facebook post made by ‘Incel’ Alek Minassian 29

Chapter 3

Figure 4 Mentions of perpetrator names on the chans 40 Figure 5 Memes picturing Brenton Tarrant as a Saint 41 Figure 6 ‘Brenton Tarrant Memetic Warfare’ thread on 8chan 42 Figure 7 Rituals for honoring Tarrant 43 Figure 8 Announcement post made by Philip Manshaus on Endchan 45 Figure 9 OP accuses Brenton Tarrant of being a Mossad agent 48 Figure 10 OP mocks John Earnest 49 Figure 11 ‘Virgin’ Earnest vs. ‘Chad’ Tarrant 51 Figure 12 Assessments of Poway Synagogue shooter John Earnest 51 Figure 13 Schadenfreude after Balliets attack on the Halle Synagogue 52 Figure 14 Meme mocking Stephan Balliet 52

Chapter 4

Figure 15 Gamified stills of Tarrant’s livestream 56 Figure 16 Custom-made game maps resembling the Al Noor Mosque 57 Figure 17 Excerpt of the Q&A section in Tarrant’s manifesto 59 Figure 18 Anon responds to post stating that “optics will backfire” 60 Figure 19 Scornful laughter after Australian news outlet falls for troll 61 Figure 20 Altered images of YouTube celebrity PewDiePie 62 Figure 21 ‘Glownigger’ meme 63 Figure 22 Direct address to federal agents 63 Figure 23 Anon addresses ‘glowniggers’ 64 Figure 24 ‘In minecraft’ as an irony disclaimer 65 Figure 25 Mirrors of the Christchurch livestream footage 66

(4)

Abstract

On 15 March 2019, an individual named Brenton Tarrant committed a deadly terrorist attack on two Christchurch mosques, killing 51 worshippers and injuring many more. Tarrant announced his deeds in a post on the anonymous imageboard 8chan, hoping to inspire others with his actions. Within several months, four copycats followed his lead. All of the perpetrators, commonly referred to as ‘lone wolves’, display striking similarities, both in terms of ideological motivations, as well as their modus operandi. Each of these attacks was announced on one of ‘the chans’; a range of anonymous, loosely-knit forums at the fringe of online culture. A closer look at these platforms reveals a sphere of far-right extremism that is dedicated to discussions and memes of these lone wolves, as well as the encouragement of copycat attacks. Although a ‘digital turn’ of lone wolf terrorism has amplified the phenomenon at an unprecedented scale, what exactly goes on within these extremist communities has received little academic attention. The aim of this research is to understand how the chans’ social dynamics and vernacular practices enable the radicalization and encouragement of community members to act out lone wolf terrorist attacks. First, I will combine a literature review of studies covering lone wolf terrorism with a literature review of works on chan culture, to gain an understanding of the historical context and characteristics of both. This will shine light on the socio-technical affordances that make communities so effective at enabling extremist practices, as well as the chans’ pivot to the far-right, that turned the innovative meme factory that once was, into a dark, politicized hate machine. The second half of this research contains case studies of text and image posts discussing five lone wolf terrorists, scraped from 4chan and 8chan. Through qualitative analysis, I will uncover the deliberate ways of encouraging copycats, and show the chans’ ways of shaping a collective identity. I found this happens through signalling cultural knowledge, and through the veneration and ridiculing of lone wolves: some perpetrators get canonized into the chans’ hagiography of ‘Saints’, others are the target of derisive laughter. I will then examine the various strategies of ‘memetic warfare’ developed by these communities. Through intricate, gamified tactics, chan users ensure the production, dissemination and preservation of their extremist content. By connecting these case studies to existing work on lone wolf terrorism and chan culture, I aim to uncover the social and vernacular processes that underlie extremism and radicalization within these anonymous communities. By doing so, my aim is to contribute to interdisciplinary research on lone wolf terrorism and chan culture, and to offer insights that may be of practical use for detecting and counteracting similar forms of online extremism. virtual

(5)

Acknowledgements

I would like to express my gratitude towards those that helped me reach this final phase of my degree. First of all, I would like to thank my supervisor Daniël de Zeeuw, who has been of great help in guiding me through this writing process: Not only by providing valuable feedback, but also by being considerate when a global pandemic threatened to complicate things. I would also like to thank my good friend Sem: for proofreading my early ramblings, and for designing the cover and timeline for this thesis. Lastly, I am very grateful to my parents: for their unconditional love and support, and their never ending encouragement for me to pursue my academic interests.

(6)

Introduction

On 15 March 2019, an Australian 28 year old male by the name of Brenton Tarrant posted a message to an online imageboard called 8chan. In this post, Tarrant announces that it is ‘time to stop shitposting and time to make a real life effort post’ (fig. 1). As became apparent shortly after the message appeared on 8chan, Tarrant’s ‘real life effort post’ resulted in the loss of 51 innocent lives. His mass killing in two Christchurch mosques did not just become known through news coverage; he broadcasted most of it live on Facebook, allowing viewers to witness the tragedy in first person view . In his 74 page manifesto, which is linked in the 8chan post, Tarrant elaborates on his core beliefs and goals for the attack, based on white supremacist conspiracy theories, and expresses his hope to inspire others with his actions.

Figure 1. The post on 8chan in which Brenton Tarrant announced the mass killing he acted out shortly after ( Anon #1 ).

Unfortunately, others did follow Tarrant’s lead. In the months after the Christchurch massacre, four terrorist attacks followed that showed many similarities to the way Tarrant operated: John Earnest’s attack on the Poway Synagogue in California; Patrick Crusius’ attack on a Walmart Supercenter in El Paso, Texas; Philip Manshaus’ attack on a mosque in Baerum, Norway and lastly Stephan Balliet’s attack on a Synagogue in Halle, Germany. The perpetrators had many things in common: Similar far-right extremist ideologies, similar modi operandi , similar identities, similar manifestos, but perhaps most striking: all of them announced their deeds online, on similar anonymous imageboards–together making up an online sphere which I will refer to as ‘the chans’. Moreover, the many obscure memes and cultural references made by these perpetrators hint at a deeper connection to the communities on these platforms.

(7)

Individuals engaging in mass violence like the aforementioned tragedies are commonly referred to as lone

wolves . Research into lone wolf terrorism is scarce (Spaaij 855), and the process of radicalization remains poorly understood (Borum et al. 390). How and under which circumstances individuals radicalize, and what kind of social dynamics convince them to turn to violent action, are questions that warrant investigation (Spaaij 867; Holt et al. 85). Although the Internet and its virtual communities are now considered highly important in enabling the radicalization process (Holt et al. 83; Borum et al. 394; Gable & Jackson 2; Ganesh 35; Schuurman et al. 771), what exactly happens within such extremist communities has received little academic attention (Quent qtd. in Maxwill). My aim is thus to find out what exactly goes on within this hateful, dark corner of the Internet, that motivates individuals to commit such heinous acts. An approach from the angle of media studies is of particular value here, since it offers the capabilities to understand the socio-technical specificities of the platforms, as well as the online culture they accommodate.

The main objective of this research is to understand how the chans’ social dynamics and vernacular practices enable the radicalization and encouragement of community members to act out lone wolf terrorist attacks. I will do so through combining a literature review of studies covering lone wolves and chan culture, with case studies of posts on 4chan and 8chan discussing the above mentioned lone wolf terrorists. By examining the characteristics of this new breed of lone wolf, and the subcultural specificities of the platforms that served as its breeding ground, I aim to provide insight into the way virtual communities of support enable and amplify violent extremism. Through qualitative analysis of text and image posts made by lone wolves and their (fan) followings on the chans, I attempt to uncover the social and vernacular processes that underlie extremism and radicalization within these anonymous communities.

In the first chapter, I will investigate the definitions and main characteristics of a lone wolf. I will explore the historical context and various notions of the ‘lone wolf’ concept by consulting existing research on the matter. This exploration will reveal a connection to earlier white supremacist movements, both in terms of ideology and mode of organization. Next, I will address what I describe as the myth of lone wolf terrorism: contrary to what the term implies, lone wolves are rarely truly ‘lone’ and radicalization is an inherently social process. This will underline the critical role played by communities of support such as the one central to this research. Finally, I describe the implications of the digital turn of lone wolf terrorism, to highlight the particular affordances that make the Internet such an effective space for amplifying extremist communities.

(8)

In the second chapter, I connect the outcome of the first chapter’s investigation to a literature review covering the specific platforms that harbor the extremist communities studied here. The anonymous imageboards central to my case studies make up an online sphere defined by ‘chan culture’. By studying the origins and major developments of ‘the chans’, I aim to uncover their subcultural specificities that allowed for extremist communities to thrive. I will do so by initially exploring the chans’ cultural background and platform affordances, followed by studying the developments that politicized the chans, to the point of its /pol/ communities becoming ‘dangerously right-wing’ (Tuters & Hagen 16). I argue that the irony-ridden meme factory that these platforms once were, turned into a breeding ground for far-right ideologues. The goal of this second chapter is to illuminate the path that led to this outcome.

The two chapters covering the case studies will each have their own focus. Chapter 3 covers the interplay and social dynamics between lone wolves, copycats and their following on the chans. Seeing as virtual extremist communities defy conventional extremist organization models, it is productive to look at the new ways these communities take shape. I will show how collective identity is formed through memes as shared cultural capital and through posts in which lone wolves are negotiated as representatives of the communities. Some lone wolves are venerated through ritualized practices, in what has been termed a

dark fandom (Broll 792). Other perpetrators are rejected and ridiculed, often for the purpose of entertainment at another’s expense known as ‘the lulz’. Then, in order to understand the concerning rise of copycat attacks, I will examine the logic behind the process of radical encouragement. I have found this process to be highly deliberate: through personal and practical calls to arms, fellow community members are encouraged to follow the example lone wolves aim to set.

In the last chapter, I will analyze the various strategies of memetic warfare underlying the production, dissemination and preservation of extremist content. Here, I will examine what the gamification of terrorism entails, a concept that has recently been argued from various angles (Ayyadi; Schlegel; Evans, “El Paso”; Mackintosh; Macklin, “Christchurch” 19; Ware 10). I have found how through gamified strategies, users have developed effective tactics to collectively engage in their hateful practices, all while remaining elusive to law enforcement. In games of trolling and trickery with media and law enforcement, chan users have developed playful tactics to ensure the spreading and preservation of their hateful content. This chapter will uncover some of these calculated strategies: following the logic that in order to combat such extremist communities, one must first understand their innovative ways of spreading hate.

The research presented here thus connects existing work on lone wolves from terrorism studies and adjacent fields to various studies on chan culture. The combined concepts and theories on lone wolves

(9)

and chan culture, will form the base for the case studies of five lone wolves that I conduct in Chapter 3 and 4. In these case studies, I analyze online posts made by the perpetrators and their online followings. I used 4CAT (Peeters & Hagen) to gather relevant text and images posted on 4chan and 8chan in the year since the initial shooting in Christchurch. The resulting datasets were used as a starting point for the selection of threads and individual posts that I further analyzed in my case study chapters. Through close reading of user posts and visual analyses of attached memes, I uncover the social dynamics and vernacular practices that characterize these extremist communities.

By studying these far-right communities that invoked multiple mass killings, I aim to contribute to interdisciplinary research on lone wolves, as well as public debates on the increasingly imminent risks of digital hate culture (Ganesh 30). Examining social dynamics within these communities can help to develop a better understanding of how they emerge and how members encourage one another; potentially allowing for the development of strategies aimed at preventing such malignant practices. Uncovering some of the intricate tactics developed by the chans in their battle against media and law enforcement could prove to be useful in nullifying their effects by strategizing ways to prevent playing into their hands. The goal is thus twofold; I aim to contribute to academic work on the understudied phenomenon of (online) lone wolf terrorism, and simultaneously offer insights that may be of practical use to those working to counteract similar forms of online extremism.

Datasets

I initially queried archived 4chan and 8chan posts from 4CAT (Peeters & Hagen), a tool that scrapes a variety of online sources and provides various modules for analyzing them. To gather the posts discussing the lone wolves central to my case studies, I queried the tool for the names (and nicknames where applicable) of the five perpetrators. The time period for the analyzed posts ranges from the day of the Christchurch shooting (15 March 2019) until the time of this research: exactly one year later (15 march 2020). As 8chan was taken offline on 5 August 2019 (shortly after the El Paso shooting) no data after this date is available. The queries for the used datasets are listed below: 1

1_{During this research, 8kun.top emerged as a successor to 8chan, offering a platform to new discussions on the lone}

(10)

Case 4CAT Query Starting date End date Results 4chan

Results 8chan

Brenton Tarrant Brenton | Tarrant 15-03-2019 15-3-2020 37.158 10.084

John Earnest John Earnest 2 _27-04-2019 _15-3-2020 ₄₇₁ ₁₄₀

Patrick Crusius Crusius 03-08-2019 15-3-2020 1019 76

Philip Manshaus Manshaus 09-08-2019 15-3-2020 175 n/a

Stephan Balliet Balliet | Doorcuck 3 _09-10-2019 _15-3-2020 ₇₀₅ _n/a

Table 1. 4CAT queries, date ranges and result counts.

The resulting datasets were used as a starting point, providing insight into basic quantitative information, such as post volume per platform and post frequency over time. Manually reading through the posts allowed me to take note of recurring themes, which I then queried within the data sets to gauge their prevalence. This method of snowballing keywords provided me with a sense for the most central topics and patterns within discussions surrounding lone wolves on the chans. This ultimately allowed me to search more directly for the most recurring themes and the most striking issues, which I analyzed qualitatively in Chapters 3 and 4. Every post, thread or image referenced throughout this research, is included in an attached ‘Anonology’ in appendix A , which details post metadata and provides URLs to 4 archived copies of the posts.

Visual elements (pictures, memes, infographics) make up an important part of chan culture; in the specific communities studied here, this is no less true. As such, I have frequently included screenshots and images posted on the chans to exemplify my findings. While images attached to posts made on 4chan are available through the 4plebs archive, this is not the case for those posted on 8chan. Because of this, I have made use of alternative methods where possible, to obtain archived copies of 8chan posts that included images (or at least their thumbnails). In addition to screenshots included in news coverage, 8chan users have unknowingly contributed to this data collection process by archiving notable threads themselves. Using web capturing tools like Archive.today, users manually store copies of threads they deem important, or would like to save for other purposes (fig. 2). I am aware that this sample could be biased,

2_{In order to minimize false positives, the query was [John Earnest] instead of [John | Earnest]. In 4CAT’s query}

syntax, this notation is used to only match posts that contain “John” and “Earnest”, rather than one of the two. In the case of Brenton Tarrant, both his first and last name were unique enough to use the [Brenton | Tarrant] syntax.

3_{In the case of Stephan Balliet, the keyword ‘doorcuck’ was added, having found after an initial, exploratory search}

that this was a common nickname used to refer to him. In the other cases, none of the identified nicknames were prevalent enough to warrant including them in the query.

(11)

as it concerns a collection curated by the extremist communities themselves. Having compared the archived threads to my datasets however, I have found them to give a fair representation of the discussions being held at large.

Figure 2. Example of a post listing links to archived 8chan threads using Archive.today’s web capture tool. This screenshot is cropped: the original post contains URLs to 28 different threads. ( Anon #2 ).

Throughout these chapters, I make use of various accounts of the lone wolf attacks studied here, to support my own findings and observations and provide additional context where necessary. Macklin (“El Paso” 2; “Christchurch” 19), Bogost (n.p.) and Evans (“Shitposting”; “El Paso”) provide detailed accounts of the attacks, as well as their offline and online aftermath. Additionally, I frequently refer to parts of the manifestos that four out of five perpetrators left behind. These documents have been described as “[...] powerful [acts] of propaganda designed to deliver an explanatory narrative, an ideological justification, a tactical lesson, and a call to arms for others to follow” (Macklin, “El Paso” 2).

As I will show in the following case studies, these manifestos are a critical part of the ways lone wolves communicate; not only to the public, but also addressing a specific audience on the chans. Likewise, anons on the chans often discuss the contents of these manifestos. As such, excerpts of the manifestos are frequently included to illustrate my findings.

(12)

Ethical Concerns

Conducting such a detailed analysis of extremist content comes with the risk of giving the studied communities what they pursue: “the oxygen of amplification” (Phillips, Oxygen 14). Indeed, Chapter 4 highlights how users within these communities develop intricate strategies to spread their extremist narrative through news outlets and other ‘mainstream’ channels. I contend, however, that this should not be a reason to let such harmful online activity go unchecked, as to not give extremists the attention they are after. In this regard I follow De Zeeuw and Tuters, who argue that before taking on such an opponent, “[...] one should understand their world-view and motivation” (227). Rather than taking a defeatist stance, the goal here is instead to develop a greater awareness and sensitivity to online extremism (De Zeeuw & Tuters 225), as an important first step in combating these violent practices.

The process of manually going through the archived posts is labor-intensive, but comes with the benefit of developing an understanding of the vernacular language, coded statements and cultural references present in so many of the posts. Online encyclopedias, (urban) dictionaries and other repositories describing the lore of the chans have been valuable resources for verifying definitions of obscure terms and references. I have attached a glossary in appendix B, offering basic definitions and etymologies of the terms appearing in this research. In my analysis, I use some of these terms in order to clarify their meaning and contextual use. I am aware of the problematic nature of some of these derogatory, or outright racist terms. They are, however, a vital part of the studied communities, and I believe there is value in getting a solid understanding of their use. As such, I do not censor any materials displayed in my case studies, but do wish to warn readers that these contain upsetting, racist or generally hateful language and imagery.

(13)

1. The Myth of Lone Wolf Terrorism

The aim of this first chapter is to explore the historical context and definitions of lone wolves and related concepts, to ultimately examine the implications of the ‘digital turn’ of extremist communities. As discussions on lone wolves often have great difficulty clarifying what ‘lone wolf terrorism’ entails, beyond the perpetrator acting individually (Gable & Jackson 80), I will first venture into the history and defining features of the ‘lone wolf’ concept by drawing upon work from the field of terrorism studies. Historical accounts reveal the phenomenon’s roots within older white supremacist groups, as well as the major developments it has been through since (Spaaij 569; Gable & Jackson 83; Borum et al. 390). Spaaij (854), Bakker and De Graaf (2) and Van Buuren (1) provide definitions of the concept and analyze its main features and patterns. Looking at the motivations for individuals to engage in an act of terror (Fein and Vossekuil 321; Borum et al. 390) will show that these are usually an interwoven, complex bricolage of personal grievances and ideological beliefs (Spaaij 862). It is thus not surprising that the term ‘lone wolf’ is a contested one, seeing as individual cases often contradict each other and are not easily captured in binary classifications.

After exploring various notions and characteristics of the lone wolf phenomenon, I will proceed to unravel the myth of lone wolf terrorism. By engaging in a more critical examination of the terms ‘lone wolf’ and ‘self-radicalization’, I will highlight their deceiving, paradoxical nature; lone wolves are rarely ever truly ‘lone’ (Schuurman et al. 771), and radicalization is an inherently social process (Holt et al. 83; Johnson 101). Next, I will focus on the space that can now be considered the epicentre of extremist socialization and organization: the Internet. Communities shifting to the virtual make use of digital affordances that enable and amplify their extremist practices. Ganesh addresses the features of digital hate

culture that make extremist communities so elusive when they congregate online (30), and studies on

dark fandoms (Broll 795) shine light on the cultural practices of online communities celebrating the acts of mass killers (Paton 226; Raitanen & Oksanen 206). I will examine the techno-social affordances that enable these virtual communities to disseminate their extremist content, while remaining highly elusive. Finally, I will synthesize the various discussed notions and characteristics of lone wolf terrorism into a definition that will be used throughout this research.

(14)

1a. Historical Context and Definitions

Despite concerns about the recent rise of lone wolf terrorist attacks (Borum et al. 395), the phenomenon has a much longer history. Popular use of the term ‘lone wolf’ likely originates in American white supremacist movements around the 1990s (Spaaij 859). Louis Beam, an American White Nationalist, promoted the tactic of ‘leaderless resistance’ (Borum et al. 390). His idea was to have actors operate independently, without reporting to a single leader for direction or instruction (Beam qtd. in Spaaij 859), imagining that this organization method would be more difficult to detect than conventional terrorism (Spaaij 859; Gable & Jackson 83). This concept was later adopted by the white supremacist group the White Aryan Resistance (WAR). WAR Founder Alex Curtis proposed a distributed network of ‘lone wolves’, which he imagined as racist warriors attacking the government or other targets in “daily, anonymous acts” (“Alex Curtis”). Curtis implemented a ‘Lone Wolf Point System’ on his website to award actors based on the ‘importance of their victims’ (ADL). The 1998 FBI investigation in this matter was called “Operation Lone Wolf” (Borum et al. 390). This is the first known case in which the perpetrator is described as a lone wolf.

When defining lone wolf terrorism, most theorists stress the individual nature of the act, as opposed to the networked processes of large terrorist organizations. The Dutch security service COT defines a ‘lone operator terrorist’ as a person operating individually, not belonging to an organized terrorist group or network (COT qtd. in Van Buuren 1); similar to the notion alluding to ‘a person who acts on his or her own without orders from–or even connections to–an organisation” (Burton & Stewart; Bakker & De Graaf 2) Kaplan et al. also name independence of any movement, leader or network of support as a key characteristic (8), and Spaaij uses the term to refer to a “[...] terrorist attack or campaign [...] [resulting] from solitary action during which the direct influence or support of others, even those sympathetic to the cause, is absent” (Spaaij 856). In most notions of the phenomenon, this aspect of acting in solitude is considered as the most central characteristic.

A well known case that exemplifies the individual nature of the act is ‘Unabomber’ Ted Kaczynski, who mailed explosives to his targets, killing and injuring several. Kaczynski is often included in historical accounts of lone wolf terrorism, as he is believed to have acted in complete isolation: He chose targets, created and deployed his explosives all by himself (Chase). Kaczynski was also the first high-profile terrorist to publish his motivations in the form of a manifesto (Berger). In his writings, called “Industrial Society and Its Future”, Kaczynski argues that the development of large scale technologies are

(15)

detrimental to human freedom, causing social disruption and psychological suffering (Kaczynski 1). Justifying his bombings, Kazcynski states: “In order to get our message before the public with some chance of making a lasting impression, we’ve had to kill people (12).” Since Kaczynski’s innovation of 5 leaving behind a manifesto, many following terrorists have done the same (Berger; Ware 3), including most of the cases studied in this research.

Although Kaczynski’s manifesto could be interpreted in multiple ways, the Unabomber’s motives seem predominantly politically charged. Such ideological motivations politicize an act of violence, turning it into something generally understood as terrorism : “an extreme form of politics”, which is never committed for the self-interest of the perpetrator alone (Gable & Jackson 81). This particular use of violence is aimed at creating a culture of fear in order to advance political, religious or ideological causes (Gable & Jackson 81). Some theorists see this political dimension of the act as a requirement, arguing that the label of ‘lone wolf’ should only be applied to acts that are “politically or religiously motivated and aim to influence public opinion or political decision-making” (Bakker & De Graaf 2).

Stated political motivations may however hide deeper, personal reasons. In the case of the Unabomber, Spaaij states that Kaczynski’s motives were also “inextricably related to personal resentment”, and speaks of a bricolage , “combining elements of different ideological traditions” (862). Such a mixture of motivations is a common theme for lone wolves. There is great difficulty in differentiating between politically motivated lone wolves and mass murderers acting motivated by personal grievances or mental health problems (Malkki 186). Even when perpetrators leave written statements justifying their actions, stated motivations may or may not correspond to actual motivations (Borum et al. 395): Fein and Vossekuil found that some perpetrators appeared to have disguised their attack in a veil of political rhetoric in order to give legitimacy to increase the visibility of their deed (333), in pursuit of mostly notoriety or fame (Van Buuren 12). ‘Murderers in search of a cause’ like these communicate a political goal to transform their act into something that is justifiable (Van Buuren 12; Malkki 204), or–in their eyes–honorable. This is important to keep in mind when gauging the intent of a lone wolf attack: a seemingly political manifesto may hide deeper, personal grievances.

Such personal grievances are often the focus of analyses of lone wolves. Both academic research as well as news coverage often concern problems in the perpetrator’s personal lives. Having trouble socializing, romantic rejection and social isolation seem to be recurring themes in the lives of lone wolves. An

(16)

individual who feels ostracized by their environment may desperately force themselves into visibility through an act of terror. Van Buuren calls this a form of performative violence , which is not necessarily directed against the world, but more so a cry for attention from audiences to recognize and acknowledge the actor’s existence and uniqueness (18). Findings in previous research also suggest that lone wolf terrorists are more likely to suffer from some form of psychopathology (Spaaij 862), certainly when compared to group actor terrorists (Gable & Jackson 80). Grandiosity and entitlement are the most occurring elements in the mental illnesses lone wolf terrorists suffer from (Van Buuren 10). It is emphasized however, that mental disorders should not be regarded as the primary reason for a lone wolf attack, as a range of sophistication and attention is displayed in the planning phase (Fein en Vossekuil 331; Ware 7). A mental illness is thus rarely a ‘master motivation’ (Borum et al. 395). While one should certainly include any mental disorders into the equation when studying lone wolf terrorism, these should not be understood as an excuse, or the main reason for an attack.

Media are often eager to come up with headlines pointing out a perpetrator’s past of being bullied, being radicalized by religion or having played violent video games. This type of checklist-analysis is a somewhat desperate attempt to come up with simple answers to the complex questions these inhumane deeds raise. Likewise, from an academic angle, there have been efforts to systematically analyze lone wolf terrorism (Borum et al. 391).While it is tempting to come up with strictly demarcated categories for classifying different types of individual terrorists, they rarely prove productive when applied to the hybrid reality found in actual case studies. In most cases, lone wolf terrorists’ motivational patterns “[...] tend to involve complex constellations of ideas and feelings that change over time” (Spaaij 866). While there is no academic consensus on the definition of a lone wolf, most theorists do seem to agree on one ‘law’: The key reasons for a lone wolf to commit a terrorist attack are complex, dynamic, and multidimensional.

1b. The Myth of the Self-Radicalized Lone Wolf

The following section will be an effort to unravel the myth of the self-radicalized lone wolf. I will review critical examinations of the term ‘lone wolf’, arguing that this type of terrorist may not be so ‘lone’ after all. Many theorists agree that a lone wolf rarely operates in complete isolation, stressing the importance of social ties and the Internet in this radicalization process (Holt et al. 83; Borum et al. 394; Gable & Jackson 2; Ganesh 35; Schuurman et al. 771). The term ‘self-radicalization’ is equally problematic, since radicalization is generally understood as an inherently social process (Holt et al. 83; Johnson 101). Despite the somewhat misleading terms, I will ultimately justify the use of the term ‘lone wolf’ in this

(17)

research as a reference to a historical tradition of white supremacy tactics, rather than a sound theoretical concept.

Holt et al. demonstrate how lone wolves are not necessarily true ‘loners’, taking into account the ways the Internet and social ties facilitate the radicalization process (83). Like the idea of lone wolf terrorrism, the concept of self-radicalization is often disputed. Radicalization is described as a ‘social process of affective networking’ (Johnson 101), involving exposure to radical ideas, either offline through peers and family, or online via websites, forums and social media (Holt et al. 84). Radicalization can be seen as a form of ‘subcultural indoctrination’ where individuals are introduced to an extremist ideology, and to the ways these beliefs are to be expressed (Holt et al. 88). As such, the concept of ‘self-radicalization’ presents itself as a bit of paradox: How can one self-radicalize, when the process of radicalization is inherently social?

Like the paradoxical concept of self-radicalization, the ‘lone wolf’ concept warrants further critical examination. The fact that the Unabomber acted in complete isolation is, contrary to what one might expect, rather atypical for a lone wolf. Although the wording ‘lone’ implies an isolated perpetrator, many definitions of the term leave room for potential support or inspiration from friends or networks. The Canadian Security Service for instance, characterizes a lone wolf as “inspired by a terrorist ideology or organization to conduct attacks” (qtd in. Borum et al. 391). Several theorists follow a similar train of thought, and include individuals who are inspired by extremist movements or ideologies in their understanding of a lone wolf (Bakker & De Graaf 2; Spaaij 856). While a lone wolf may not be officially part of a terrorist organization, it is possible–or likely even–that they find their ideological justification within those spheres.

In a report of 40 case studies of far-right lone wolves in the United Kingdom, Jackson and Gable found that perpetrators were motivated, and found the required information to carry out their attacks, through ideological networks: “Many of them used the internet to associate with like minded people, exchange ideas and obtain information about how to acquire weapons or make explosives and deadly chemicals” (Gable & Jackson 2). The authors state that the far-right terrorists in their report are “[...] connected with, influenced by, and often helped by organisations whose beliefs they share” (Gable & Jackson 7). These networks of support provide structures that legitimize extremism, and offer a violence-encouraging community (Gable & Jackson 80). Although these contacts may not engage in terrorist acts themselves, they do enable lone wolves through providing a community of support (Gable & Jackson 85). While contemporary lone wolves may not be members of a traditional, hierarchical extremist network,

(18)

characterizing them as socially isolated would be a misconception, as the only phase of their terrorist act that happens truly ‘alone’ seems to be the final one: the attack itself.

The lone wolf theory has little acceptance in Western European police services, and a senior FBI officer states the only genuine ‘lone wolf’ they had come across was Unabomber Ted Kaczynski (qtd. in Gable & Jackson 13). Jackson concludes that the term ‘lone wolf’ is thus of limited value, since these individual operators are often (in)directly supported by others (Gable & Jackson 7). When it comes to using this term, Jackson proposes to take into account its historical context. Rather than using it as a synonym for solo actor terrorism, the term ‘lone wolf’ should be used to specify terrorism inspired by far-right, leaderless resistance style ideas (Gable & Jackson 80; 87). This specific notion is certainly applicable to the individuals studied in this research: All of the lone wolves covered in my case studies recycle ideas from older White Supremacy movements, and follow modes of organization similar to the style of leaderless resistance. Even though the term ‘lone wolf’ is limited and contested, I opt for using it with this specific historical context in mind.

1c. The Digital Turn of Lone Wolf Terrorism

In the last section of this chapter, I will address what I describe as the ‘digital turn’ of lone wolf terrorism, which enabled and amplified extremist communities at an unprecedented magnitude. The critical role the Internet plays in harboring extremist communities is now often acknowledged in studies on lone wolf terrorism (Gable & Jackson 85; Quent qtd. in Maxwill; Borum et al 394; Schuurman et al. 771; Berger). Ganesh stresses that the Internet plays an important role across different strands of extremism by “enculturing individuals into extreme communities” (35): “There is a striking relationship between [...] online groups, the discourse of digital hate culture, and the growth of hate crimes (Ganesh 43).” While many acknowledge the paramount role of the Internet in contemporary lone wolf terrorism, most of the analyses of this new digital breeding ground remain speculative. To begin mapping this uncharted territory, I will explore the aspects of online communities that allow extremist communities to thrive. Before turning to the specific platforms and communities central to this research, I will first examine more generally the digital affordances that allow extremist communities to flourish online. I will argue that the Internet and its social platforms facilitate unparalleled abilities for spreading and discussing extremist content on a global scale. Additionally, the communities that form online can offer lone wolves the sense of significance and belonging many are seeking. Finally, decentralized modes of organization and unique social dynamics make these extremist networks hard to detect, combat and dismantle.

(19)

The first and foremost affordance offered by online communities is their ability to spread extremist narratives. The ease at which one can study and spread extremist views online has been described as concerning (Cohen et al. 246). Ganesh argues the democratizing force of social media comes with the risk of ‘clever groups of anonymous amateurs’ with the power to disseminate extreme views, bigotry and propaganda (31). Anyone with a desire for violence against civilian targets can easily find a ‘cause’ online (Borum et al. 394). Additionally, it has become increasingly easy to produce and spread user-generated content online, easing (international) cooperation between like minded groups that would not have had the same opportunities offline (Froio & Ganesh 93). As I will exemplify throughout my case studies, extremist communities employ innovative ways to not only spread, but also normalize extremist rhetoric within mainstream audiences.

In addition to spreading extremist ideologies, online communities can also offer practical help to extremists. Cohen and colleagues explain how critical information, such as bomb-making tutorials and geographical information, significantly lowers the ‘capability threshold’ for individuals to carry out advanced attacks (Cohen et al. 346). The 1999 London Nail bomber David Copeland, for example, looked up instructions for making bombs through online sources (Spaaij 863). Van Buuren argues the availability of extremist readings and DIY instructions may have furthered the growth of the ‘autodidactic extremist’ (4). We can take from this that extremist online communities have a critical educational function for lone wolf terrorists: they can provide them with practical information required to carry out an attack.

Another key feature of extremist online communities is fulfilling the need for significance through a sense of belonging. As discussed in the previous section, radicalization does not happen in isolation, but is a highly social process. Whereas in the pre-digital era a sense of belonging was found in a traditional terrorist organization, this feeling of being part of something bigger can now be provided by online means. It is now easier than ever for already-troubled individuals to seek out leaderless resistance style communities of like minded individuals (Kaplan et al. 6), who will gladly provide them with the feeling of being part of something bigger, giving meaning to their lives and producing identity and self-understanding (Van Buuren 14). These anonymous, online communities may especially lure in individuals who are reluctant to participate in offline society (Van Buuren 14). Group identification can then lead to collectivistic values, making individuals more prone to self-sacrifice (Cohen et al. 250). This sense of significance offered by online extremist communities should not be underestimated: after all, if an individual sacrifices him or herself to avenge or benefit a group of anonymous strangers, then that group may well play a big, influential role in their personal life (Borum et al. 394).

(20)

As will become apparent in my case studies, such a community of support can take on the form of what Broll coined a dark fandom , in which fans of those who have perpetrated heinous acts come together (792). Studies on this phenomenon, which has been linked to the growth of the Internet, have so far exclusively focused on fans of school shooters (Broll 802; 795). Members of these online subcultures share common interests, participate in giving perpetrators fame and theorize on their possible motivations (Raitanen & Oksanen 195) They imitate each other's media practices and engage in collective ritualistic activities (Paton 204; 221). Potential copycats, seriously considering an attack of their own, likely make up a small part of these communities (Raitanen & Oksanen 204). Although school shooters make up a specific ‘breed’ of lone wolves that come with their own characteristics, the cultural practices undertaken by their fan followings show striking similarities to the communities studied here. This makes the dark

fandom concept a useful frame of reference that I will return to throughout my case studies.

A last problematic aspect of online extremist communities is their elusiveness. The strategy of leaderless resistance through online means, proposed by white supremacist Louis Beam, has proven its worth: Beam likely never imagined that the Internet would provide an “almost perfect medium” for remotely commanding detached, anonymous lone wolves (Burton & Stewart). The Internet empowers extremists to reach their following, without the risks of interpersonal exchanges that could incriminate them and alarm authorities (Borum et al. 394). While the Internet is certainly not a lawless space (Lessig 5), the loose-ties-structures and distributed networks of online extremist communities can make them hard to govern (Kaplan et al. 8). Ganesh argues that their decentralized structure, elusiveness and use of coded language make them ‘ungovernable’ (36). More often than not these communities are very aware of their strong ability (and necessity) to evade scrutiny by authorities. In my case studies, I have found this cat-and-mouse-game with law enforcement to be a recurring theme. New network topologies presented by extremist online communities are not easily captured by traditional models for terrorist organizations. If we wish to understand (and ultimately govern) these new modes of organization, it is productive to study the way these collective identities are formed and their elusive ways of self-preservation.

I have argued here that the Internet has played a crucial role in enabling and amplifying the radicalization process of lone wolves throughout the last decades. Extremist views are now readily available to anyone seeking a cause for a violent attack. DIY-guides for explosives and other sources of practical information empower autodidact lone wolves. Troubled loners who feel ostracized can collectively socialize, granting them a sense of belonging as well as encouragement for acting upon their grievances. These communities can take on the form of dark fandoms , where perpetrators are praised and potential copycats lurk. Lastly, new, swarm-like modes of organization ensure self-preservation through a high degree of elusiveness.

(21)

Combined, these features allow for online communities to function as fertile breeding grounds for lone wolf terrorists.

Conclusion

In this chapter I have explored the history and definitions of the lone wolf concept, in order to ultimately connect these to the development I describe as the digital turn of lone wolf terrorism. The origins of the term 'lone wolf' reveal a historical context of white supremacy movements dating back to the 90s, that deliberately promoted such decentralization tactics following the logic that they would be harder to detect by law enforcement. The various notions of love wolves seem to agree on what I described as a 'law of hybridity', posing that motivations that lead lone wolves to act are usually a complex mix of political beliefs and personal grievances. Mental disorders often play a role, but should not be considered as the main reason for a lone wolf attack. While the term lone wolf implies that the perpetrator is socially isolated, generally only the final act of violence happens alone. Networks of support legitimise extremism, and encourage troubled individuals to resort to violence. These communities now increasingly take shape on virtual platforms, where they offer radicalized members a sense of community, while amplifying the ability to spread extremism and avoid the scrutiny of law enforcement.

One should be prudent with the term lone wolf, as to avoid using it as a default blanket term for any puzzling act of violence committed by an individual. Arguably, the term ‘lone wolf’ can be misleading, implying an isolated perpetrator; and therefore failing to account for potential social ties that influenced them. Additionally, some suggest avoiding the term ‘lone wolf’ to prevent glorifying or empowering the attackers (Borum et al. 390). With these considerations in mind, I still opt for using the term, due to the connection to its historical context that is once again highly relevant. I will follow Gable & Jackson’s suggestion of using the term as a synonym for solo actor terrorism inspired by far-right, leaderless resistance style ideas (80; 87). This specific definition is applicable to all of the individuals studied in this research. As I will show in the case studies conducted in this research, the lone wolves and their followings studied here can be considered a resurgence of older white supremacy movements, both in terms of ideology and mode of organization. Having considered the historical context, various notions and characteristics of lone wolf terrorism, I will offer a definition for the term that I will be using throughout this research. A ‘lone wolf’ is here understood as: an individual engaging in an act of mass violence, usually motivated by a combination of ideological and personal reasons; often influenced or inspired by an extremist community . This definition may not apply to lone wolves within other contexts. It is

(22)

presented here as a workable concept that is referred to in this research when the term ‘lone wolf’ is mentioned.

(23)

2. Chan Culture

Towards the end of the first chapter I have described the digital turn in lone wolf terrorism, and examined the affordances offered by online means of communication that amplify extremist communities. This chapter will continue this investigation by looking at the specific platforms and their communities that harbored the five lone wolves central to my case studies. The platforms I am referring to are anonymous imageboards, including (but not limited to) 4chan and 8chan. This range of similarly structured platforms make up a specific online subculture I will refer to as chan culture. My notion of ‘the chans’ is similar to what De Zeeuw and Tuters coined the Deep Vernacular Web: “a loosely-knit network of forums and sites that exist at the margins of the platformized web”, in which ‘vernacular’ is conceptualized as “[...] a group-based discourse, highly innovative of novel forms of expression that are often intentionally obscure to those outside of the in-group” (216).

The most popular and well-known platform within chan culture is 4chan, which over the years has been joined by many more fringe imageboards like 7chan, 8chan and Endchan. Although I will mostly refer to these communities as a whole, the chan communities do not form a homogenous entity; I will be pointing out key differences between the platforms as they present themselves. For the sake of brevity, throughout this research I will be referring to the whole of the communities on these imageboards as ‘the chans’. The two platforms I will be looking at specifically are the (in)famous 4chan and its lesser known, more extremist offspring 8chan. Although none of the lone wolves’ announcement posts appeared on 4chan, the majority of posts and memes that followed (analyzed in the case studies) originate from these imageboards.

In the first half of this chapter, I will explore the cultural background and platform affordances that made the chans a fertile breeding ground for extremist communities. Works on chan culture provide valuable resources for understanding these platforms’ subcultural specificities. Beran provides a detailed account of the platform’s origin stories (ch. 4). I will highlight some of the chans’–initially mostly innocent– vernacular practices (De Zeeuw & Tuters 215) that formed the base for a more dark and toxic culture to emerge. The range of works on chan culture I consult address the platform’s technical affordances (Bernstein et al. 56); ‘lulz’ and trolling (Phillips, This Is Why 99); /pol/’s content characteristics (Hine et al. 7; Tuters & Hagen 1), and the way collective identity is formed through memes (Nissenbaum & Shifman 483). Not only do these aspects define chan culture, they turn out to be equally important to the extremist communities studied in this research.

(24)

The second half of this chapter describes the chans’ pivot to the far-right. I will point out a shift in chan culture from what was originally an ironic and lulz-ridden meme factory where ‘nothing is to be taken seriously’, towards an Incel- and Nazi-infested sphere in which hateful ideologies have become rampant–and are now taken very seriously. The rise of alt-right groups on the chans (Nagle ch. 1) came with an abundance of conspiracy theories (Cosentino 79) and frequent strategic use of the chans’ ironic style (Tuters 37; Nagle ch. 2; May & Feldman 26). Addressing these developments will help to understand how these particular communities have welcomed and amplified extremism: not just in terms of the platforms’ socio-technical affordances, but also in terms of the specific culture that spawned this hateful environment, ultimately leading to multiple (mass) killings.

2a. The Subversive Meme Factory

The biggest social media platforms have been known to be facing hate speech and extremist content: Facebook, Twitter and Reddit have all implemented policies and advanced moderation systems to combat hateful content (Alba et al.; “Terrorism and Violent”; Breland). Fringe platforms like Gab and Telegram have been more permissive in this regard, resulting in an influx of extremist communities on their platforms (Roose, “On Gab”; Venkataramakrishnan). The platforms most notorious for taking pride in their lack of moderation, and which consequently take the cake when it comes to tolerating hateful content and extremism, are the chans. These imageboards have been repeatedly linked to lone wolf attacks, and all perpetrators studied in this research announced their attacks on one of these imageboards.

The most popular and notorious platform at the root of chan culture, is 4chan. This anonymous, imageboard-style forum was founded by Christopher ‘moot’ Poole in 2003. Moot created the website on a whim to exchange anime pictures with online friends (Beran ch. 4), copying the format of the popular Japanese discussion forum Futaba, or 2chan (Bernstein et al. 52). 4chan offered a platform to a niche 6 community from the fringe of otaku-culture, mainly focused on anime and fetish porn (Beran ch. 3). The imageboard rapidly grew to be one of the most popular sites on the web: as a ‘prolific meme factory’ (Bernstein et al. 50), 4chan simultaneously defined, and became defined by, mainstream culture (Beran ch. 5). 4chan described itself as a pile of garbage posting with no rules, a practice which has come to be known as shitposting (Beran ch. 5; Burton; Evans, “Shitposting”).

(25)

Anonymity is one of chan culture’s most defining features, setting the imageboards apart from social media and most other online discussion forums. The vast majority of 4chan’s posts are made anonymously (Bernstein et al. 55), leading users to refer to themselves individually as ‘anons’, or ‘Anonymous’ as a collective (De Zeeuw & Tuters 217). Since everyone is ‘Anonymous’, this likely shapes a strong feeling of community among 4chan’s many users (Bernstein et al. 56). The chan’s vernaculars have been described as “intentionally obscure to those outside of the in-group” (De Zeeuw & Tuters 216), driving away many-but creating strong social ties to those that decide to stay (Tuters 40). The chan’s anonymity by default has made for some unique social dynamics and collective identity formation processes.

Chan culture, where users are anonymous by default, has developed its own unique way of forming and iterating a collective identity. Users communicate their status as a member of the community through vernacular and memetic literacy (Bernstein et al. 56; Nissenbaum & Shifman 484). To put it in the chans’ own vernacular language: so-called ‘normies’ (signifying the out-group) are identified by their display of lacking vernacular knowledge, upon which they will be told to either ‘lurk moar’ (increasing their subcultural knowledge) or ‘gtfo’ (leave the premises). Nissenbaum and Shifman argue that within web-based communities like these, memes are used to create ‘membership-based distinctions’ (484). These may be especially vital to anonymity based communities like the chans, due to the absence of the usual social identity cues (Nissenbaum & Shifman 488). Drawing on Bourdieu’s concept of cultural capital, Nissenbaum and Shifman argue that knowledge about the community culture determines who is a member and who is not (488). “Memes thus function as part of a culture, contributing to the set of ideas around which communities gather and act” (Nissenbaum & Shifman 485). As I will argue throughout my case studies, this procedure for signifying and negotiating a collective identity through memetic and vernacular literacy is central to lone wolves and their followings on the chans.

Bernstein et al. argue 4chan’s anonymity also comes with an ‘online disinhibition effect’ (Suler 321), 7 which could increase participation, openness, and creativity, at the risk of causing antisocial behaviour (Bernstein et al. 55) Indeed, quantitative analysis shows that 12% of the posts on 4chan’s /pol/ board contain hateful terms (Hine et al. 7). Like anonymity, the offensiveness of content on the chans is a defining characteristic of chan culture. Content on the ‘random’ board /b/’s content is described as crude, cynical, irreverent and intentionally offensive (Nissenbaum & Shifman 487); the often racist, sexist, homophobic language is part of the group identity (Bernstein et al. 53). To give two notable examples that

7_{According to this concept by Suler, while online, “some people self-disclose or act out more frequently or intensely}

Lone Wolves, Copycats and The Chans: How Anonymous, Far-Right Online Communities and Their Vernacular Practices Radicalize Lone Wolf Terrorists