Content uploaded by Ryan Scrivens
Author content
All content in this area was uploaded by Ryan Scrivens on Mar 11, 2021
Content may be subject to copyright.
Comparing the Online Posting Behaviors of Violent and Non-Violent Right-Wing
Extremists
Ryan Scrivens
School of Criminal Justice, Michigan State University
East Lansing, Michigan, USA
rscriv@msu.edu
Thomas W. Wojciechowski
School of Criminal Justice, Michigan State University
East Lansing, Michigan, USA
Joshua D. Freilich
Department of Criminal Justice, John Jay College of Criminal Justice, CUNY
New York City, New York, USA
Steven M. Chermak
School of Criminal Justice, Michigan State University
East Lansing, Michigan, USA
Richard Frank
School of Criminology, Simon Fraser University
Burnaby, British Columbia, Canada
2
Disclosure Statement
No potential conflict of interest was reported by the authors.
3
Comparing the Online Posting Behaviors of Violent and Non-Violent Right-Wing
Extremists
Abstract:
Despite the ongoing need for researchers, practitioners, and policymakers to identify and assess
the online activities of violent extremists prior to their engagement in violence offline, little is
empirically known about their online behaviors generally or differences in their posting
behaviors compared to non-violent extremists who share similar ideological beliefs particularly.
In this study, we drew from a unique sample of violent and non-violent right-wing extremists to
compare their posting behaviors within a sub-forum of the largest white supremacy web-forum.
Analyses for the current study proceeded in three phases. First, we plotted the average posting
trajectory for users in the sample, followed by an assessment of the rates at which they stayed
active or went dormant in the sub-forum. We then used logistic regression to examine whether
specific posting behaviors were characteristic of users’ violence status. The results highlight a
number of noteworthy differences in the posting behaviors of violent and non-violent right-wing
extremists, many of which may inform future risk factor frameworks used by law enforcement
and intelligence agencies to identify credible threats online. We conclude with a discussion of
the implications of this analysis, its limitations and avenues for future research.
Introduction
This study examines the posting behavioral patterns of both violent and non-violent right-wing
extremists (RWEs) within a sub-forum of the largest white supremacy web-forum, Stormfront.
We investigate if violent RWEs are differentially active in one of the largest and well-known
4
online spaces of the extreme right compared to non-violent RWEs. This study represents an
original contribution to the academic literature on violent online political extremism on two key
fronts.
First, many researchers, practitioners, and policymakers continue to raise questions about
the role of the Internet in facilitating violent extremism and terrorism.
1
Questions often surround
the impact of the offenders’ consumption of and networking around violent extremist online
content in their acceptance of extremist ideology and/or their decision to engage in violent
extremism and terrorism.
2
Understandably, increased attention has been given to identifying
violent extremists online prior to their engagement in offline violence and scrutinizing their
online presence.
3
A growing interest on these issues notwithstanding, few empirically grounded
analyses identify which online users have engaged in violent extremism offline and in turn
explore their digital footprints. Second, some empirical literature is emerging on the posting
behaviors of extremists online,
4
but no work has identified differences in posting patterns of
those who share extreme ideological beliefs but are violent or non-violent in the offline world. In
fact, current research does not fully capture nuanced differences between violent and non-violent
extremists in general.
5
Despite claims from scholars about the importance of making such
comparisons,
6
terrorism and extremism studies tend to focus on ideologically motivated violence
and oftentimes overlook other crime forms and other types of extremists.
7
Sarma,
8
for example,
in his assessment of the challenges associated with conducting risk assessments to combat
violent extremism concluded that “effective risk assessment tools need to be able to distinguish
between individuals who are on violent and non-violent trajectories. This, in turn, requires, a
body of research that has isolated indicators that are sensitive to the different processes.”
Importantly, our unique data includes a comparison group of non-violent RWEs which helps us
5
to build an understanding of how violent extremists – violent RWEs in particular – engage the
Internet compared to non-violent counterparts.
Examining the online behaviors of violent extremists
Like most of us, violent extremists often leave a digital footprint behind. Notable examples
include Anders Breivik, the Norwegian far-right terrorist convicted of killing 77 people in 2011,
who was a registered member of a white supremacy web-forum and had ties to a far-right wing
social media site;
9
Dylann Roof, the 21-year-old who murdered nine Black parishioners in
Charleston, South Carolina in 2015, allegedly posted messages on a RWE website;
10
and 28-
year-old Australian Brenton Tarrant who, before killing 50 people in two Christchurch, New
Zealand mosques in 2019 and live-streaming his attack, announced his intentions on 8chan and
produced a ‘manifesto’ linked on the website.
11
In these cases, questions are oftentimes raised
about whether such violent individuals can be identified online prior to their attacks offline.
These questions sit at the top of the priority list for many researchers, practitioners, and
policymakers.
12
At recent specialized terrorism conferences – including, for example, the 2018
VOX-Pol Conference on ‘Violent Extremism, Terrorism, and the Internet: Present and Future
Trends’ – the nexus between the on- and offline worlds of violent extremists dominates much of
the discussion.
13
Similarly, workshops hosted by law enforcement, the defence community, and
similar agencies – such as the Swedish Defence Research Agency’s Seminar Day on ‘Violent
Extremism: Methods, Tools and Techniques for Detecting and Analyzing Violent Extremism’ –
focus much of their attention on gaining insight into how the online discussions, behaviors and
actions of violent extremists can spill over into the offline realm.
14
Others, including policy-
makers, are concerned that high and increasing levels of always-on Internet access and the
6
production and wide dissemination of large amounts of violent extremist content online may
have violent radicalizing effects.
15
Although important concerns, there remains a significant need
to first identify individuals online who are violent extremists offline.
Offline behavior of violent extremists is linked to their online engagement.
16
There is not
a significant body of research in this area, but what little evidence does exist suggests that we
should not conceive violent extremism as an offline versus online dichotomy. To illustrate, Gill
and Corner
17
looked at the behavioral underpinnings of lone-actor terrorists since 1990. The
results suggested that while the number of lone-actor terrorist plots remained stable over time,
the growth in the Internet altered their means of radicalization and attack learning. The Internet
therefore acts as a substitute for other factors such as intelligence gathering and attack planning,
not necessarily a force enabler. Further, according to Gill and Corner, there was a significant
positive correlation between those who virtually interacted with co-ideologues and those who
interacted with co-ideologues face-to-face.
Building on Gill and Corner,
18
Gill and colleagues
19
examined the online behaviors of
223 convicted United Kingdom (U.K.)-based terrorists. Those who learned online were 4.39
times more likely than those who did not learn online to have experienced non-virtual network
activity and 3.17 times more likely to have experienced non-virtual place interaction. Of those
who plotted an attack, the individuals who attended training camps were also significantly more
likely than those who did not attend training camps to have learned online. Additionally, the
evidence also suggested that communicating with co-ideologues online was significantly more
likely to have been accompanied by face-to-face interactions with non-violent co-ideologues.
Those who communicated online were 3.89 times more likely to have experienced non-virtual
network activity and 3.17 times more likely to have experienced non-virtual place interaction. Of
7
those who plotted an attack, the individuals who attended training camps were also significantly
more likely to have communicated online. This may be due to the compartmentalization of tasks
noted by Gill.
20
For example, individuals tended to learn about a specific necessary task online
(e.g., bomb-making), but then found a different instrumentalization in their offline interactions
with co-ideologues (e.g., the justification of bombing a particular target).
Holbrook and Taylor
21
focused on pre-arrest media usage of five case studies of U.K.-
based terrorists. In all five cases, a belief pathway precipitated any operational action where all
actors were active participants in the consumption, discussion and distribution of extremist
materials before any physical involvement in attack planning. Subjects consumed a diverse range
of media across a number of platforms and interacted online in chatrooms as well as offline by
copying compact discs (CDs) of extremist content for one another.
Lastly, Gaudette and colleagues
22
conducted in-depth interviews with former violent
RWEs on their use of the Internet during their involvement in violent extremism and identified
an important interaction between their on- and offline worlds which were intertwined with
extremist activities, identities, and a need for security. Here the Internet served as a gateway for
individuals to engage in violent extremist activities offline, connecting adherents in the online
world to the offline world, often through the online promotion of offline events (e.g., concerts,
rallies, protests, and gatherings). Gaudette and colleagues
23
also found that most of study
participants were concerned about their on- and offline security during their involvement in
violent extremist groups, noting that they modified their on- and offline behaviors to avoid
detection and infiltration from law enforcement and anti-fascist groups. Interestingly, though, is
that, despite their security concerns, most of them – unlike the newer generation of violent
RWEs who are active and communicate anonymously in various encrypted online spaces –
8
maintained the same identities in both their on- and offline worlds and displayed their roles in the
movement (e.g., as recruiters or promoters) similarly in both worlds. Discussed within this
context was how the Internet was flooded with “Net Nazis” or “Internet Warriors” (i.e.,
adherents who are very active online but will not meet others offline). Regardless, this emerging
evidence base remains in its infancy and requires further exploration.
Comparing behaviors of violent and non-violent extremists
Reviews of the terrorism and violent extremism research literature regularly highlight the paucity
of primary data (be it, open-, closed- or researcher generated data) and basic methodological
requirements that inform analyses, which too often eschew using statistical analyses – thus
lagging behind analogous fields.
24
Most of this work consisted of what Lum and colleagues
25
describe as “thought pieces, theoretical discussions, or opinions.” The most recent review,
however, offers a more positive outlook. Schuurman
26
analyzed over 3,000 articles published in
leading terrorism-specific journals between 2007 and 2016 and found that over half used some
form of primary sources. Schuurman
27
also identified a slow but steady upward trend in the use
of statistical analyses in terrorism and extremism studies in recent years. Other recent systematic
reviews of the scientific knowledge base of factors associated with engagement in terrorism
shows a similar improvement in empiricism over time.
28
Although researchers have made great strides in the last ten years to develop empirical-
based research in terrorism and extremism studies,
29
this growing body of research oftentimes
lack comparison groups.
30
This is light of scholars arguing that subset comparisons are an
important step in the empirical study of terrorists and violent extremists.
31
Fortunately, a few
studies have compared the offline behaviors of violent and non-violent extremists, with research
9
largely focusing on the organizational level and compared behaviors of far-right hate groups in
the United States (U.S.). Chermak and colleagues,
32
for example, examined the organizational
differences between violent and non-violent hate groups and found that violent groups were
more likely to be older, larger, have conflicts with other organizations, and have charismatic
leaders. In addition, groups that published ideological literature were significantly less likely to
be violent. Similarly, Asal and colleagues
33
compared variations in organizational characteristics
of violent and non-violent hate groups over time and found groups that had been previously
involved in violence were strongly linked to other groups, and larger groups were more likely to
be violent organizations. They also found that groups that advocated for extreme ideological
principles were also significantly more likely to be violent.
Some studies have also investigated offender-level behaviors of violent and non-violent
extremists. For example, LaFree and colleagues
34
compared violent and non-violent criminal
extremists in the U.S. Here they applied criminology’s social control and social learning theories
and analyzed data from the Profiles of Radicalized Individuals in the U.S. (PIRUS) database.
LaFree and colleagues
35
found that a lack of stable employment, extremist peers, criminal record,
and a history of mental illness were significantly associated with violent extremism. On the other
hand, education, marital status and extremist family members were not significant. Becker
36
also
used PIRUS data and similarly found that attributes representing constructs from both social
control and social learning theories accounted for differences across violent and non-violent
extremists. Jasko and colleagues
37
also used PIRUS data and investigated the impact of the
psychological quest for significance model on extremist offending. They found that offenders
who experienced economic loss of significance, social loss of significance, and had radicalized
friends were more likely to commit violent ideologically motivated crimes. Knight and
10
colleagues
38
compared the behavioral outcomes of 40 convicted U.K.-based extremists across
four extremist types (violent lone, non-violent lone, violent group member, and non-violent
group member). While the authors found few differences in behavioral outcomes across
extremist type, they found that many more violent extremists felt a sense of superiority but a
sense of underachievement, were rejected by others, experienced a personal crisis and a
perceived personal responsibility to act. Kerodal and colleagues
39
used data from the U.S.
Extremist Crime Database (ECDB) and compared far-right violent extremists who committed a
fatal attack to non-violent far-right extremists who committed a financial crime. The authors
found that conspiratorial, anti-government and anti-tax beliefs were positively associated with
risk of financial crimes, while xenophobic, survivalist and anti-gun control beliefs were
positively associated with the risk of violent crimes. In light of these important contributions, far
less is empirically known about the online behaviors of violent and non-violent extremists.
Researchers, practitioners, and policymakers have paid close attention to the presence of
terrorists and extremists online in recent years, with a particular emphasis on the digital activities
of the extreme right.
40
This is largely the result of ongoing reports that some violent RWEs and
terrorists were active online prior to their attacks.
41
It should come as little surprise, then,
researchers have focused on the activities of RWEs on various platforms, including on websites
and discussion forums,
42
mainstream social media sites including Facebook,
43
Twitter,
44
and
YouTube,
45
fringe platforms including 4chan
46
and Gab
47
, and digital applications such as
TikTok
48
and Telegram.
49
But these studies, like terrorism and extremism research in general,
lack comparison groups, despite a significant need to focus on comparative analysis and consider
how violent extremists are different than non-violent extremists.
50
In other words, empirical
11
research has overlooked differences in the online activities of those who share extreme
ideological beliefs but are violent or non-violent offline.
An exhaustive search using dedicated academic research databases produced just one
empirical study on the online behaviors of violent and non-violent extremists. Holt and
colleagues
51
examined the underlying theoretical assumptions evident in radicalization models
through a case-study analysis of violent and non-violent extremists, which included an in-depth
analysis of the on- and offline behaviors of four extremists. The findings provide initial support
for aspects of social control, particularly in the role of peers in facilitating social relationships
that increase radicalization and engender criminality. By comparison, there was less support for
the role of social learning as peers communicated definitions supportive of radical ideologies but
did not consistently serve as sources of imitation nor play much of a role with respect to
differential reinforcement for behavior. It is important to note, though, that what we generally
know about the online behaviors of violent and non-violent extremists is limited. We have little
scholarship on the posting behavioral patterns of violent and non-violent extremists and fewer
efforts to methodologically and systematically analyze their online posting behavior. There can
be little doubt that an assessment is needed.
Current study
Data and sample
We analyzed all postings made by a sample of violent and non-violent RWEs in the open access
sections of Stormfront Canada, a Canadian-themed sub-forum of the largest white supremacy
discussion forum Stormfront. It is the oldest racial hate site and discussion forum used by
members of the RWE movement. Stormfront is also one of the most influential RWE forums in
12
the world.
52
From the mid-1990s, the extreme right became increasingly reliant on web-forums
to facilitate movement expansion by spreading propaganda and connecting with like-minded
individuals, both within and beyond national borders.
53
It was around this time that Stormfront
transformed from a website into a forum with an array of sub-sections addressing a variety of
topics, including an ‘International’ section composed of a range of geographically and
linguistically bounded sub-forums (e.g., Stormfront Europe, Stormfront Downunder, and
Stormfront Canada). Stormfront has also served as a “funnel site” for the RWE movement,
wherein forum users have been targeted by other RWE users for the purpose of recruiting new
members into violent offline groups, including online forums hosted by the Hammerskins, Blood
& Honour, and various Ku Klux Klan (KKK) branches.
54
Today, Stormfront has just shy of
357,000 ‘members’ and contains over 13.5 million posts.
Understandably, Stormfront has been the focus of much research attention since its
inception, including an assessment of recruitment efforts by forum users,
55
the formation of a
virtual community
56
and collective identity there,
57
the extent to which Stormfront is connected
to other racial hate sites
58
, and how Stormfront discourse is less virulent and more palatable to
readers.
59
Although a number of emerging digital spaces have been adopted by the extreme right
in recent years, Stormfront continues to be a valuable online space for researchers to assess
behavioral posting patterns. To illustrate, recent efforts have been made to examine RWE
posting behaviors found on the platform,
60
the development of user activity and extremist
language there
61
, the impact of presidential election results on Stormfront posting behaviors
62
,
and the ways in which the collective identity of the extreme right takes shape over time
63
and is
affected by offline intergroup conflict on the forum.
64
However, research in this space has yet to
13
identify differences in posting patterns of Stormfront users who share extreme ideological beliefs
but are violent or non-violent in the offline world.
Data collection and sampling efforts proceeded in two stages. First, all open source
content on Stormfront Canada was captured using a custom-written computer program that was
designed to collect vast amounts of information online.
65
In total, the web-crawler extracted
approximately 125,000 sub-forum posts made by approximately 7,000 authors between
September 12, 2001 and October 29, 2017.
66
Several forum attributes were also collected during
this process, including: (1) usernames and a unique identification number for each author; (2) the
date that each author joined the forum; (3) all posted content for each author in the sub-forum;
(4) the date and time for each sub-forum posting; (5) a tally of the total number of posts for each
author in the sub-forum; and, (6) a tally of the total number of posts for each author in the
broader forum.
Second, to identify users in the sub-forum who were violent RWEs offline and those who
were non-violent RWEs offline, a former violent extremist
67
who was actively involved in the
North American RWE movement for more than 10 years – both in recruitment and leadership
roles, predominantly in violent racist skinhead groups – voluntarily reviewed a list of users who
posted in the sub-forum and then selected those who matched one of the two user types.
68
This
identification process was done under the supervision of the lead researcher of this project,
wherein each time the former identified a user of interest, they were asked to explain in as much
detail possible why the user was identified as a violent or non-violent RWE. The former was also
asked to provide examples of the activities that each user engaged in as well as their association
with or connection to each identified user. This was done to verify the authenticity of each user
identified by the former. A total of 49 violent and 50 non-violent RWEs were identified from a
14
list of approximately 7,000 usernames. Each of these identified usernames represented a unique
forum user and there was no evidence of extremists in our sample using multiple usernames.
This was confirmed by the former, whose connections were strong enough to be able to link
individuals to usernames. The online content for each user was then identified in the sub-forum
data: 12,617 posts from the violent users and 17,659 posts from the non-violent users.
69
The
sample included 30,276 posts, with the first post made on September 1, 2004 and the last post
made on October 29, 2017.
RWEs who were identified for the current study were actively involved in right-wing
extremism. In particular, they were those who – like all extremists – structure their beliefs on the
basis that the success and survival of the in-group is inseparable from the negative acts of an out-
group and, in turn, they are willing to assume both an offensive and defensive stance in the name
of the success and survival of the in-group.
70
RWEs in the current study were thus characterized
as a racially, ethnically, and/or sexually defined nationalism, which is typically framed in terms
of white power and/or white identity (i.e., the in-group) that is grounded in xenophobic and
exclusionary understandings of the perceived threats posed by some combination of non-whites,
Jews, Muslims, immigrants, refugees, members of the LGBTQ community, and feminists (i.e.,
the out-group(s)).
71
However, violent RWEs that were identified for the current study were those
who committed several acts of known physical violence, including violent attacks against
minorities and anti-racist groups. Violence in this regard aligned closely with Bjørgo and
Ravndal’s
72
understanding of RWE violence, which they describe as “violent attacks whose
target selection is based on extreme-right beliefs and corresponding enemy categories—
immigrants, minorities, political opponents, or governments [...] [or] spontaneous violence.”
Conversely, non-violent RWEs identified for the current study were those who were actively
15
involved in RWE movements and activities offline, including – but not limited to – rallies,
marches, protests, postering and flyering campaigns, and group meetings and gatherings, but did
not engage in physical violence in any known capacity.
Measures
Total forum posts
The total number of posts within the broader Stormfront forum was included in the analysis and
was measured as a frequency count. The presence of outliers in this measure was assessed using
the bacon function in Stata and outliers were identified as being any value greater than 5,938
posts.
73
As a result, 5,938 total posts for a given user was assigned as the ceiling category for the
recoded variable and 13 users were assigned membership to this ceiling category.
74
Total sub-forum posts
The total number of posts that users made in the sub-forum was also included as a variable in the
analysis. This was also measured as a frequency count. The presence of outliers was also
assessed in Stata and outliers were identified as being any values greater than 510 posts. As a
result, 510 sub-forum posts for a given user was assigned as the ceiling category and 12 users
were categorized in this top code category.
75
Posts per month in the sub-forum
Another variable assessed in the analysis was the number of sub-forum posts per month for user
in the sample. This was simply a count of the total number of posts that were generated in the
16
sub-forum each month, which was a calculation of the first and last dates on which individual
users posted in the sub-forum.
Initiation month in the sub-forum
We examined the month that each user made their first posted in the sub-forum relative to the
lifespan of the sub-forum. To illustrate, for users who posted their first message in the sub-forum
during the first month that the sub-forum became active, they would be assigned a code of “1”
for this variable, whereas those who first posted in the sub-forum the following month would be
assigned a code of “2”, and so on.
Active months in the sub-forum
The number of months that each user was active in the sub-forum was assessed. This acted as a
time variable, as it allowed us to delineate posting behavior across the months following the date
of users’ first post in the sub-forum. Here individual posts were tracked, coded and analyzed
across a time period of 54 months for each user, beginning at the time of their first post in the
sub-forum. A period of 54 months was chosen as the time span because it was the median
number of months between users’ first and last post in the sub-forum.
Dormancy status in the sub-forum
This variable explored the posting status of users as either “active” or “dormant”. Dormancy,
which was defined as a state of no longer being an active poster in the sub-forum, was
operationalized as a span of 24 consecutive months in which users generated zero new posts in
the sub-forum following their first post in the sub-forum. A time period of 24 months was chosen
17
as the threshold because it was the mean number of consecutive months that users generated zero
new posts between their first and last month of active posting behavior in the sub-forum within
the 54 months that were tracked for the analysis. This variable was then coded in two forms. The
first was a variable that delineated users who were dormant at any given wave from those who
were not (0 = no; 1 = yes). The second variable delineated users who were identified as ever
going dormant during the study period from those who did not (0 = no; 1 = yes).
76
Violence status
Whether or not users were violent RWEs was also assessed. As previously noted, a former
extremist who was familiar with users’ status as a violent or non-violent RWE provided
information regarding users’ violence status. This was operationalized as a binary variable which
delineated violent and non-violent users into two groups (0 = non-violent; 1 = violent).
Analytic strategy
Analyses for the current study proceeded in three phases. The first phase involved the descriptive
plotting of posting frequency across the 54-month study period. This involved plotting the
average posting trajectory for the full sample and then delineating posting behavior by violence
status.
The second phase of the analysis involved a survival analysis of the dormancy rates
across the study period, which entailed a censoring of users from the analyses once they had
gone dormant in the sub-forum. Doing so provided the rates at which users either stayed active
throughout the entire study period or went dormant at some point based on the criteria outlined
18
above. Like the first phase of the analysis, we examined the full sample and then delineated users
based on their violence status.
The final phase of the analysis utilized logistic regression to examine the relevance of key
variables that are associated with users’ violence status. Table 1 provides descriptive information
of all variables included in the analyses. Logistic regression was selected as the model of choice
because of the binary nature of the dependent variable (i.e., violence status). The results are
described in the form of odds ratios (OR), which are interpreted as the predicted change in the
odds of being a violent user based on a one unit increase in an independent variable of interest,
net of all other variables in the model. All analyses were conducted using Stata/MP 16.1.
Table 1. Descriptive information of the sample.
Mean/proportion
Standard
deviation
Minimum
Maximum
Total forum posts
1762.556
2164.406
3
5938
Total sub-forum posts
142.606
198.066
1
510
Posts per month in the
sub-forum
4.142
14.896
0
282
Initiation month in the
sub-forum
25.687
13.772
1
53
Dormancy status in the
sub-forum
(0 = no; 1 = yes)
.525
N/A
0
1
Violence status
(0 = non-violent; 1 =
violent)
.495
N/A
0
1
19
Results
Figure 1 describes the average posting trajectory for the full sample, collectively. These analyses
indicated a general negative trend in posting frequency following a sharp spike in posting
frequency after month one of posting activity in the sub-forum. Throughout the entire study
period, the average posting frequency declined to near zero rates at the final month of the 54-
month study period.
Figure 2 describes these same analyses delineated by violence status, which painted a
much more complex picture. While the non-violent user group showed some spikes in posting
frequency throughout the measurable study period, the average posting frequency of this group
remained relatively stable across time. Conversely, there was much greater variance in the
posting frequency of the violent user group. Here they generally demonstrated an early spike in
average posting frequency up until about month 13. Following month 13, the average posting
frequency of this group immediately declined sharply and then demonstrated a steady decline
across the remainder of time in the study period. However, it should be noted that 95%
confidence intervals were not plotted because of a lack of statistical significance in average
posting frequency between these groups at any point in the study period.
20
Figure 1. Average number of posts per month in the sub-forum for the full sample.
Figure 2. Average number of posts per month in the sub-forum by violence status.
The second phase of the analysis involved the use of survival analysis to assess the
dormancy rate of the full sample and the violence status groups. Figure 3 illustrates the survival
21
rate for the full sample. These results revealed a steady dropout rate of users throughout the
entirety of the study period, with approximately 50 percent of the full sample surviving to the
end.
77
Figure 4 describes survival rates for the violent vs. non-violent groups. These results
indicated that the violent user group demonstrated quicker time to dormancy than the non-violent
user group throughout the entire study period. These differences in survival rates were
statistically significant for months 18 to 20. Approximately 55 percent of the non-violent group
survived to the end of the study period whereas only about 40 percent of the violent group
survived. Despite the 15 percent difference in final survival rates, this difference was not
statistically significant.
Figure 3. Survival analysis of dormant rate in the sub-forum for the full sample.
22
Figure 4. Survival analysis of dormant rate in the sub-forum by violence status.
The final phase of the analysis entailed the use of logistic regression to examine covariate
effects of violence status, which is described in Table 2. The results indicated that only total
forum posts emerged as significantly associated with violence status, wherein having less total
forum posts throughout the entire study period was associated with greater odds of being a
violent user (OR = .9997; p ≤ .01). Interestingly, total sub-forum posts, dormancy status and
initiation month in the sub-forum were not significantly associated with violence status.
23
Table 2. Logistic regression model of covariate effects on odds of being a violent user.
Odds ratio
p-value
95% confidence interval
Total forum posts
.9997
.010
.9995
.9999
Total sub-forum
posts
1.0002
.290
.9988
1.0041
Initiation month in
the sub-forum
1.5675
.425
.5191
4.734
Dormancy status in
the sub-forum
(0 = no; 1 = yes)
.9863
.395
.9554
1.0181
Constant
1.4808
.532
.4323
5.0730
Discussion
This study examined the posting behavioral patterns of violent and non-violent RWEs within the
open access sections of a sub-forum of Stormfront. We assessed the posting trajectories and
survival rate of each user and user type as well as whether specific posting behaviors were
characteristic of users’ violence status. Several conclusions can be drawn from this study.
Interestingly, there was a general decline in posting behavior over time for the RWEs in
the study. There was also a steady dropout in posting frequency for approximately half of the
sample throughout the entirety of the study period. This is a noteworthy set of findings because it
suggests that RWEs in the current study – whether violent or non-violent offline – generally do
not increase in their posting behavior over time. This finding is in contrast to empirical research
which found a steady increase in user participation in RWE online spaces
78
as well as studies that
conceptualize digital platforms of the extreme right as those that polarize members’ opinions
over time.
79
Our study findings do, however, align with other empirical research which suggests
that the majority of those who post in RWE online spaces, whether in collective extremist forums
24
for a general audience such as Stormfront
80
or violent RWE forums
81
, tend to desist in and tail of
their posting behavior over time. However, our findings may be a symptom of the study sample:
each user was either a violent or non-violent RWEs and were actively involved in right-wing
extremism. It is reasonable to assume, then, that posting in an open access sub-forum may have
become a concern to the users over time, given the offline activities they were engaged in (e.g.,
violence, illegal activities, organizing rallies, recruiting, etc.). Empirical research on the online
behaviors of RWEs suggests those who are actively involved in RWE activities offline are
oftentimes concerned that law enforcement officials and anti-racist groups are monitoring their
online activities and may modify their posting behaviors to avoid detection.
82
With this in mind,
it may be the case that the RWEs in the current study were concerned that, by posting content in
an online space that can be publicly viewed, they may be putting themselves in a vulnerable
position and could become the subject of an investigation from anti-hate watch-organizations or
even law enforcement. Some of the users in the sample may have also been incarcerated during
part(s) of study period and were unable to post messages then. Regardless, these findings require
further exploration.
From a policy perspective, the results of the current study suggest that analysts who are
searching for signs of violent extremists online should be less concerned about investigating
high-frequency posters and more concerned about those who post fewer messages online. This is
because violent RWEs in the current study posted more messages in the sub-forum in the
beginning of their posting careers than the non-violent RWE group but posted less than the non-
violent group in the later portions of their posting careers. Violent RWEs also desisted in posting
frequency at a much quicker rate than their non-violent counterpart and the only observed
characteristic of the violent status was having less total number of posts in the broader forum. In
25
other words, violent RWEs in the current study tended to be those who were much less active
online than non-violent RWEs. This set of findings is supported by empirical research which
found that violent members of RWE movements are largely clandestine, often paranoid because
of the violence they engage in, and for this reason are concerned about revealing their
identities.
83
Similar to previous research that highlights how some violent RWEs alter their
online behaviors to avoid detection from law enforcement,
84
the waning and inactive posting
behavior of the violent RWEs in our sample may reflect this security concern. This finding
comes as little surprise, given that similar tactics have been adopted by a new generation of
RWE who in recent years have exploited various encrypted online platforms and messaging apps
to avoid being tracked and detected.
85
While our analyses provide unique insight into the differences in posting behaviors
between violent and non-violent users online, there remain several noteworthy limitations. The
first of these pertains to the small sample size. With a sample of just 99 RWE users, analyses
were somewhat limited. The sample was limited to a group of violent and non-violent RWEs
who were known to one former RWE, which biases our sample and restricted our ability to
identify and analyze the posting behaviors of an array of violent and non-violent RWEs found
online. In addition, the relatively small sample size likely contributed to a power problem that
was manifested in some of the null findings despite relatively large differences between the
violent and non-violent RWE user groups. While this tempers the findings and their implications,
this should spur future research to gather a more robust sample in order to provide a stronger
understanding of the posting behaviors of violent and non-violent RWEs.
A second limitation of this study pertains to the violence status variable. While this is
very novel data presented for the current study, the variable lacks the specificity in measurement
26
that allows for the establishment of temporal ordering with independent variables of interest. In
other words, it is unclear if changes in posting behaviors are driving changes in violent offending
or vice versa, and our analysis is instead limited to being understood as associations only. Future
research should therefore pinpoint exact moments in time that individuals engaged in violence
offline and then assess their posting behaviors both before and after the act of violence. Doing so
may shed light on whether specific posting behaviors in the online world escalate to violent
actions in the offline world.
A third limitation pertains to the data itself. While this type of data is novel in its capacity
to provide details about the posting behaviors of violent and non-violent RWEs, it is limited in
the information provided by one former extremist. There remain many unanswered questions
about the characteristics of those identified as violent or non-violent in our study. Unlike the
growing body of research that has drawn distinctions between the offline activities and behaviors
of violent and non-violent extremists,
86
our data does not include information on key
characteristics identified in this literature such as an individual’s employment status, criminal
records, history of mental illness, extremist/radicalized peers, and types of grievances, among
many others. Future research is therefore needed to assess whether these characteristics (and
others) mirror our sample of violent or non-violent extremists as well as whether certain
characteristics drive differential posting behavior.
Lastly, while the novelty of the data in the current study cannot be understated, their
validity is based on one key former RWE informant. Future research should further verify the
authenticity of each RWE user by identifying the birth and given name of each individual in the
sample. These names could then be triangulated with open source intelligence (e.g., media
reports, court documents, terrorism databases, and social media accounts on each user) or with
27
the Extremist Crime Database (ECDB), which currently includes over 500 data points on nearly
1,000 violent and 1,500 non-violent RWEs
87
Triangulating these data may further inform law
enforcement and intelligence communities on the online discussions that may result in offline
extremist violence as well as shed light on the nexus between the on- and offline worlds of
violent and non-violent extremists.
Notes
1
See Maura Conway, “Determining the Role of the Internet in Violent Extremism and
Terrorism: Six Suggestions for Progressing Research,” Studies in Conflict & Terrorism 40, no. 1
(2017): 77-98.
2
See Ryan Scrivens, Paul Gill, and Maura Conway, “The Role of the Internet in Facilitating
Violent Extremism and Terrorism: Suggestions for Progressing Research,” in Thomas J. Holt
and Adam Bossler, Eds., The Palgrave Handbook of International Cybercrime and
Cyberdeviance (London, UK: Palgrave, 2020), pp. 1-22.
3
Joel Brynielsson, Andreas Horndahl, Fredik Johansson, Lisa Kaati, Christian Mårtenson, and
Pontus Svenson, “Analysis of Weak Signals for Detecting Lone Wolf Terrorists,” Security
Informatics 2, no. 11 (2013): 1-15; Katie Cohen, Fredik Johansson, Lisa Kaati, and Jonas C.
Mork, “Detecting Linguistic Markers for Radical Violence in Social Media,” Terrorism and
Political Violence 26, no. 1 (2014): 246-256; Lisa Kaati, Amendra Shrestha, and Katie Cohen,
“Linguistic Analysis of Lone Offender Manifestos,” Proceedings of the 2016 IEEE International
Conference on Cybercrime and Computer Forensics, Vancouver, BC, Canada.
4
Examples include Bennett Kleinberg, Isabelle van der Vegt, and Paul Gill, “The Temporal
Evolution of a Far‑Right Forum,” Journal of Computational Social Science. Ahead of Print, 1-
23; Ryan Scrivens, Garth Davies, and Richard Frank, “Searching for Signs of Extremism on the
Web: An Introduction to Sentiment-based Identification of Radical Authors,” Behavioral
Sciences of Terrorism and Political Aggression 10, no. 1 (2018): 39-59; Ryan Scrivens, Garth
Davies, and Richard Frank, “Measuring the Evolution of Radical Right-Wing Posting Behaviors
Online,” Deviant Behavior 41, no 2. (2020): 216-232; Ryan Scrivens, “Exploring Radical Right-
Wing Posting Behaviors Online,” Deviant Behavior. Ahead of Print, 1-15; Ryan Scrivens,
Thomas W. Wojciechowski, and Richard Frank, “Examining the Developmental Pathways of
Online Posting Behavior in Violent Right-Wing Extremist Forums, Terrorism and Political
Violence. Ahead of Print, 1-18.
5
Freilich and LaFree, “Criminology Theory and Terrorism”; Kiran M. Sarma, “Risk Assessment
and the Prevention of Radicalization from Nonviolence Into Terrorism,” American Psychologist
72, no. 3 (2017): 278-288; Allison G. Smith, Risk Factors and Indicators Associated with
Radicalization to Terrorism in the United States: What Research Supported by the National
Institute of Justice Tells Us (Washington DC: National Institute of Justice, 2018); Allison G.
Smith, How Radicalization to Terrorism Occurs in the United States: What Research Sponsored
by the National Institute of Justice Tells Us (Washington DC: National Institute of Justice, 2018).
6
Victor Asal, Steven M. Chermak, Sarah Fitzgerald, and Joshua D. Freilich, “Organizational-
Level Characteristics in Right-Wing Extremist Groups in the United States Over Time,”
28
Criminal Justice Review 45, no. 2 (2016): 250-266; Steven Chermak, Joshua Freilich, and
Michael Suttmoeller, “The Organizational Dynamics of Far-Right Hate Groups in the United
States: Comparing Violent to Nonviolent Organizations,” Studies in Conflict & Terrorism 36, no.
3 (2013): 193-218; Joshua D. Freilich and Gary LaFree, “Criminology Theory and Terrorism:
Introduction to the Special Issue,” Terrorism and Political Violence 27, no. 1 (2015): 1-15; Gary
LaFree, Michael A. Jensen, Patrick A. James, and Aaron Safer-Lichtenstein, “Correlates of
Violent Political Extremism in the United States,” Criminology 56, no. 2 (2018): 233-268; Clark
McCauley and Sophia Moskalenko, Friction: How Radicalization Happens to Them and Us
(New York, NY: Oxford University Press, 2011); John Monahan, “The Individual Risk
Assessment of Terrorism,” Psychology, Public Policy, and Law 18, no. 2 (2012): 167-205.
7
Joshua D. Freilich, Steven M. Chermak, and Joseph Simone, “Surveying American State Police
Agencies About Terrorism Threats, Terrorism Sources, and Terrorism Definitions,” Terrorism
and Political Violence 21, no. 3 (2009): 450-475; John Horgan, Neil Shortland, Suzzette
Abbasciano, Shaun Walsh, “Actions Speak Louder Than Words: A Behavioral Analysis of 183
Individuals Convicted for Terrorist Offenses in the United States from 1995 to 2012,” Journal of
Forensic Sciences 61, no. 5 (2016): 1228-1237; Sarah Knight, David Keatley, and Katie
Woodward, “Comparing the Different Behavioral Outcomes of Extremism: A Comparison of
Violent and Non-Violent Extremists, Acting Alone or as Part of a Group, Studies in Conflict and
Terrorism. Ahead of print, 1-22; Alex P. Schmid and Albert J. Jongman, Political Terrorism: A
New Guide to Actors, Authors, Concepts, Data Bases, Theories and Literature (New Brunswick,
NJ: Transaction Books, 1988); Leonard Weinberg, Ami Pedahzur, and Sivan Hirsch-Hoefler,
“The Challenges of Conceptualizing Terrorism,” Terrorism and Political Violence 16, no. 4
(2004): 777-794.
8
Sarma, “Risk Assessment and the Prevention of Radicalization from Nonviolence Into
Terrorism,” p. 279.
9
Jamie Bartlett and Mark Littler, Inside the EDL: Populist Politics in the Digital Era (London:
Demos, 2011).
10
Southern Poverty Law Center, “Dylann Roof May Have Been a Regular Commenter at Neo-
Nazi Website The Daily Stormer,” June 22, 2015,
https://www.splcenter.org/hatewatch/2015/06/21/dylann-roof-may-have-been-regular-
commenter-neo-nazi-website-daily-stormer (accessed 29 October, 2020).
11
Maura Conway, Ryan Scrivens and Logan Macnair, “Right-Wing Extremists’ Persistent
Online Presence: History and Contemporary Trends,” The International Centre for Counter-
Terrorism – The Hague 10(2019): 1-24.
12
Conway, “Determining the Role of the Internet in Violent Extremism and Terrorism.”
13
For more information on the conference, visit: https://www.voxpol.eu/events/vox-pol-third-
biennial-conference/.
14
For more information on the workshop, visit:
http://www.it.uu.se/research/arenas/security/security-seminars/ELIAS2.
15
Charlie Edwards and Luke Gribbon, “Pathways to Violent Extremism in the Digital Era,”
RUSI Journal 158, no. 5 (2013): 40-47; David H. Gray, and Albon Head, “The Importance of the
Internet to the Post-Modern Terrorist and its Role as a Form of Safe Haven,” European Journal
of Scientific Research 25, no. 3 (2009): 396-404; Anne Stenersen, “The Internet: A Virtual
Training Camp?,” Terrorism and Political Violence 20, no. 2 (2008): 215-233.
29
16
Paul Gill, Emily Corner, Maura Conway, Amy Thornton, Mia Bloom, and John Horgan,
“Terrorist Use of the Internet by the Numbers: Quantifying Behaviors, Patterns, and Processes,”
Criminology and Public Policy 16, no. 1 (2017): 99-117.
17
Paul Gill and Emily Corner, “Lone-Actor Terrorist Use of the Internet and Behavioural
Correlates,” in Lee Jarvis, Stuart Macdonald, and Thomas M. Chen, Eds., Terrorism Online:
Politics, Law, Technology and Unconventional Violence (London, UK: Routledge, 2015), pp.
35-53.
18
Ibid.
19
Gill et al., “Terrorist Use of the Internet by the Numbers.”
20
Paul Gill, Lone-Actor Terrorists: A Behavioural Analysis (London, UK: Routledge, 2015).
21
Donald Holbrook and Max Taylor, “Terrorism as Process Narratives: A Study of Pre-Arrest
Media Usage and the Emergence of Pathways to Engagement,” Terrorism and Political Violence
31, no. 6 (2019): 1307-1326.
22
Tiana Gaudette, Ryan Scrivens, and Vivek Venkatesh, “The Role of the Internet in Facilitating
Violent Extremism: Insights from Former Right-Wing Extremists,” Terrorism and Political
Violence. Ahead of Print, 1-18.
23
Ibid.
24
Bruce Hoffman, “Current Research on Terrorism and Low-Intensity Conflict,” Studies in
Conflict & Terrorism 15 (1992): 25-37; Ariel Merari, “Academic Research and Government
Policy on Terrorism,” in Clark McCauley, Ed., Terrorism Research and Public Policy (London,
UK: Frank Cass, 1991), pp. 88-102; Reuben Miller, “The Literature of Terrorism,” Terrorism 11
(1988): 63-87; Schmid and Jongman, Political Terrorism; Andrew Silke, “The Devil You Know:
Continuing Problems with Research on Terrorism,” Terrorism and Political Violence 13, no. 4
(2001);” Andrew Silke, “An Introduction to Terrorism Research,” in Andrew Silke, Ed.,
Research on Terrorism: Trends, Achievements and Failures (London, UK: Routledge, 2004), pp.
1-29; Andrew Silke, “Research on Terrorism: A Review of the Impact of 9/11 and the Global
War on Terrorism,” in Hsinchu Chen, Edna Reid, Joshua Sinai, Andrew Silke, and Boaz Ganor,
Eds., Terrorism Infomatics: Knowledge Management and Data Mining for Homeland Security
(Boston, MA: Springer, 2008), pp. 27-50.
25
Cynthia Lum, Leslie W. Kennedy, and Alison J. Shirley, “Are Counter-Terrorism Strategies
Effective? The Results of the Campbell Systematic Review on Counter-Terrorism Evaluation
Research,” Journal of Experimental Criminology 2 (2006): 492.
26
Bart Schuurman, “Research on Terrorism, 2007–2016: A Review of Data, Methods, and
Authorship,” Terrorism and Political Violence 32, no. 5 (2020): 1011-1026.
27
Ibid.
28
See Sarah L. Desmarais, Joseph Simons-Rudolph, Christine S. Brugh, Eileen Schilling, and
Chad Hoggan, “The State of Scientific Knowledge Regarding Factors Associated with
Terrorism,” Journal of Threat Assessment and Management 4, no. 4 (2017): 180-209.
29
See Bart Schuurman, “Research on Terrorism, 2007–2016.”
30
Smith, Risk Factors and Indicators Associated with Radicalization to Terrorism in the United
States; Smith, How Radicalization to Terrorism Occurs in the United States.
31
See Freilich and LaFree, “Criminology Theory and Terrorism;” see also Horgan et al.,
“Actions Speak Louder Than Words.”
32
Chermak et al., “The Organizational Dynamics of Far-Right Hate Groups in the United
States.”
30
33
Asal et al., “Organizational-Level Characteristics in Right-Wing Extremist Groups in the
United States Over Time.”
34
LaFree et al., “Correlates of Violent Political Extremism in the United States.”
35
Ibid.
36
Michael H. Becker, M. “When Extremists Become Violent: Examining the Association
Between Social Control, Social Learning, and Engagement in Violent Extremism,” Studies in
Conflict & Terrorism. Ahead of Print, 1-21.
37
Katarzyna Jasko, Gary LaFree, and Arie Kruglanski, “Quest for Significance and Violent
Extremism: The Case of Domestic Radicalization,” Political Psychology 38, no. 5 (2017): 815-
831.
38
Knight et al., “Comparing the Different Behavioral Outcomes of Extremism.”
39
Ashmini G. Kerodal, Joshua D. Freilich, and Steven M. Chermak. “Commitment to Extremist
Ideology: Using Factor Analysis to Move Beyond Binary Measures of Extremism,” Studies in
Conflict & Terrorism 39, no. 7-8 (2016): 687-711.
40
Conway et al., “Right-Wing Extremists’ Persistent Online Presence”; Thomas J. Holt, Joshua
D. Freilich, and Steven M. Chermak, “Examining the Online Expression of Ideology among Far-
Right Extremist Forum Users,” Terrorism and Political Violence. Ahead of Print, 1-21.
41
See, for example, Southern Poverty Law Center, “White Homicide Worldwide,” April 1, 2014.
https://www.splcenter.org/20140331/white-homicide-worldwide (accessed 29 October 2020).
42
Les Back, “Aryans Reading Adorno: Cyber-Culture and Twenty-First Century Racism,”
Ethnic and Racial Studies 25, no 4. (2002): 628-651; Ana-Maria Bliuc, John Betts, Matteo
Vergani, Muhammad Iqbal, and Kevin Dunn, “Collective Identity Changes in Far-Right Online
Communities: The Role of Offline Intergroup Conflict,” New Media and Society 21, no. 8
(2019):1770–1786; Val Burris, Emery Smith, E., and Ann Strahm, “White Supremacist
Networks on the Internet,” Sociological Focus 33, no 2. (2000): 215-235; Willem De Koster, and
Dick Houtman, “‘Stormfront is Like a Second Home to Me,’” Information, Communication and
Society 11, no 8. (2008): 1155-1176; Robert Futrell and Pete Simi, “Free Spaces, Collective
Identity, and the Persistence of U.S. White Power Activism,” Social Problems 51, no 1. (2004):
16-42; Holt et al., “Examining the Online Expression of Ideology among Far-Right Extremist
Forum Users”; Scrivens et al., “Measuring the Evolution of Radical Right-Wing Posting
Behaviors Online”; Scrivens, “Exploring Radical Right-Wing Posting Behaviors Online”;
Magdalena Wojcieszak, “‘Don’t Talk to Me’: Effects of Ideological Homogenous Online Groups
and Politically Dissimilar Offline Ties on Extremism,” New Media and Society 12, no. 4(2010):
637-655.
43
Mattias Ekman, “Anti-Refugee Mobilization in Social Media: The Case of Soldiers of Odin,”
Social Media and Society 4, no. 1 (2018): 1-11; Lella Nouri and Nuria Lorenzo-Dus, “Investigating
Reclaim Australia and Britain First’s Use of Social Media: Developing a New Model of Imagined
Political Communities Online,” Journal for Deradicalization 18(2019): 1-37; Sebastian Stier, Lisa
Posch, Arnim Bleier, and Markus Strohmaier, “When Populists Become Popular: Comparing
Facebook Use by the Right-Wing Movement Pegida and German Political Parties,” Information,
Communication and Society 20, no. 9 (2017): 1365-1388. Ryan Scrivens and Amarnath
Amarasingam, “Haters Gonna “Like”: Exploring Canadian Far-Right Extremism on Facebook,”
in Digital Extremisms: Readings in Violence, Radicalisation and Extremism in the Online Space,
edited by Mark Littler and Benjamin Lee (London, UK: Palgrave, 2020), 63-89.
44
Reem Ahmed and Daniela Pisoiu, “Uniting the Far Right: How the Far-Right Extremist, New
Right, and Populist Frames Overlap on Twitter – a German Case Study,” European Societies.
31
Ahead of Print, 1-23; J. M. Berger, Nazis vs. ISIS on Twitter: A Comparative Study of White
Nationalist and ISIS Online Social Media Networks (Washington, DC: The George Washington
University Program on Extremism, 2016); J. M. Berger and Bill Strathearn, Who Matters Online:
Measuring Influence, Evaluating Content and Countering Violent Extremism in Online Social
Networks (London, UK: The International Centre for the Study of Radicalisation and Political
Violence, 2013); Pete Burnap and Matthew L. Williams, “Cyber Hate Speech on Twitter: An
Application of Machine Classification and Statistical Modeling for Policy and Decision,” Policy
and Internet 7, no. 2 (2015): 223-242; Roderick Graham, “Inter-Ideological Mingling: White
Extremist Ideology Entering the Mainstream on Twitter,” Sociological Spectrum 36, no. 1
(2016): 24-36.
45
Mattias Ekman, “The Dark Side of Online Activism: Swedish Right-Wing Extremist Video
Activism on YouTube,” MedieKultur: Journal of Media and Communication Research 30, no.
56 (2014): 21-34; Kevin Munger and Joseph Philips, Right-Wing YouTube: A Supply and
Demand Perspective, The International Journal of Press/Politics. Ahead of Print, 1-34; Derek
O’Callaghan, Derek Greene, Maura Conway, Joe Carthy, and Pádraig Cunningham, “Down the
(White) Rabbit Hole: The Extreme Right and Online Recommender Systems,” Social Science
Computer Review 33, no 4. (2014): 1-20.
46
Savvas Finkelstein, Joel Zannettou, Barry Bradlyn, and Jeremy Blackburn, “A Quantitative
Approach to Understanding Online Antisemitism,” arXiv:1809.01644, 2018; Antonis Papasavva,
Savvas Zannettou, Elimiano De Cristofaro, Gianluca Stringhini, and Jeremy Blackburn, “Raiders
of the Lost Kek: 3.5 Years of Augmented 4chan Posts from the Politically Incorrect Board,”
arXiv:2001.07487, 2020.
47
Savvas Zannettou, Barry Bradlyn, Elimiano De Cristofaro, Haewoon Kwak, Michael
Sirivianos, Gianluca Stringini, and Jeremy Blackburn, “What is Gab: A Bastion of Free Speech
or an Alt-Right Echo Chamber,” Proceedings of the WWW ’18: Companion Proceedings of The
Web Conference 2018’ Yuchen Zhou, Mark Dredze, David A. Broniatowski, and William D.
Adler, “Elites and Foreign Actors Among the Alt-Right: The Gab Social Media Platform,” First
Monday 24(2019).
48
Gabriel Weimann and Natalie Masri, “Research Note: Spreading Hate on TikTok,” Studies in
Conflict & Terrorism. Ahead of Print, 1-14.
49
Jakob Guhl and Jacob Davey, A Safe Space to Hate: White Supremacist Mobilisation on
Telegram (London, UK: Institute for Strategic Dialogue, 2020); Aleksandra Urman and Stefan
Katz, “What They Do in the Shadows: Examining the Far-right Networks on Telegram”,
Information, Communication & Society. Ahead of Print, 1-20.
50
Freilich and LaFree, “Criminology Theory and Terrorism”; Joshua D. Freilich, Steven M.
Chermak, and Jeff Gruenewald, “The Future of Terrorism Research: A Review Essay,”
International Journal of Comparative and Applied Criminal Justice 39, no. 4 (2015): 353-369.
51
Thomas J. Holt, Joshua D. Freilich, Steven M. Chermak, and Gary LaFree, “Examining the
Utility of Social Control and Social learning in the Radicalization of Violent and Nonviolent
Extremists,” Dynamics of Asymmetric Conflict 11, no. 3 (2018): 125-148.
52
Bliuc et al., “Collective Identity Changes in Far-Right Online Communities”; Pete Simi and
Robert Futrell, American Swastika: Inside the White Power Movement’s Hidden Spaces of Hate
(Second Edition) (Lanham, MD: Rowman and Littlefield Publishers, 2015).
53
See Jessie Daniels, Cyber Racism: White Supremacy Online and the New Attack on Civil
Rights (Lanham, MD: Rowman and Littlefield Publishers, 2009); see also Conway et al., “Right-
Wing Extremists’ Persistent Online Presence.”
32
54
Bradley Galloway and Ryan Scrivens, “The Hidden Face of Hate Groups Online: An Insider’s
Perspective,” VOX-Pol Network of Excellence Blog, January 3, 2018.
https://www.voxpol.eu/hidden-face-hate-groups-online-formers-perspective (accessed 4
December 2020).
55
See W. Chris Hale, “Extremism on the World Wide Web: A Research Review,” Criminal
Justice Studies 25, no. 4 (2010): 343-356; see also Christopher J. Lennings, Krestina L. Amon,
Heidi Brummert, and Nicholas J. Lennings, “Grooming for Terror: The Internet and Young
People,” Psychiatry, Psychology and Law 17, no. 3 (2010): 424-437; Meghan Wong, Richard
Frank, and Russell Allsup, “The Supremacy of Online White Supremacists: An Analysis of
Online Discussions of White Supremacists,” Information and Communications Technology Law
24, no. 1 (2015): 41-73.
56
See Back, “Aryans Reading Adorno”; see also Lorraine Bowman-Grieve, “Exploring
“Stormfront:” A Virtual Community of the Radical Right,” Studies in Conflict and Terrorism 32,
no. 11 (2009): 989-1007; see also Willem De Koster and Dick Houtman, “‘Stormfront is Like a
Second Home to Me:’ On Virtual Community Formation by Right-Wing Extremists,”
Information, Communication and Society 11, no. 8 (2008): 1155-1176.
57
See Futrell and Simi, “Free Spaces, Collective Identity, and the Persistence of U.S. White
Power Activism;” see also Barbara Perry and Ryan Scrivens, “White Pride Worldwide:
Constructing Global Identities Online,” in Jennifer Schweppe and Mark Walters, Eds., The
Globalisation of Hate: Internationalising Hate Crime (New York, NY: Oxford University Press,
2016), pp. 65-78.
58
See Burris et al., “White Supremacist Networks on the Internet;” see also Phyllis B.
Gerstenfeld, Diana R. Grant, and Chau-Pu Chiang, “Hate Online: A Content Analysis of
Extremist Internet Sites,” Analysis of Social Issues and Public Policy 3, no. 1 (2003): 29-44.
59
See Daniels, Cyber Racism; see also Priscilla M. Meddaugh and Jack Kay, “Hate Speech or
‘Reasonable Racism?’ The Other in Stormfront,” Journal of Mass Media Ethics 24, no. 4 (2009):
251-268.
60
Scrivens, “Exploring Radical Right-Wing Posting Behaviors Online.”
61
Kleinberg et al., “The Temporal Evolution of a Far‑Right Forum.”
62
Ryan Scrivens, George W. Burruss, Thomas J. Holt, Steven M. Chermak, Joshua D. Freilich,
and Richard Frank, “Triggered by Defeat or Victory? Assessing the Impact of Presidential
Election Results on Extreme Right-Wing Mobilization Online,” Deviant Behavior. Ahead of
Print, 1-16.
63
Scrivens et al., “Measuring the Evolution of Radical Right-Wing Posting Behaviors Online.”
64
Bliuc et al., “Collective Identity Changes in Far-Right Online Communities.”
65
For more information on the web-crawler, see Ryan Scrivens, Tiana Gaudette, Garth Davies,
and Richard Frank, “Searching for Extremist Content Online Using The Dark Crawler and
Sentiment Analysis,” in Mathieu Deflem and Derek M. D. Silva, Eds., Methods of Criminology
and Criminal Justice Research (Bingley, UK: Emerald Publishing, 2019), pp. 179-194.
66
September 12, 2001 was simply date that the sub-forum went live online. Based on our
assessment of the first messages posted on the sub-forum, it would appear as though Stormfront
Canada was not launched in response to the 9/11 terror attacks.
67
By former violent extremists, we refer to individuals who, at one time in their lives, subscribed
to and/or perpetuated violence in the name of a particular extremist ideology and have since
publicly and/or privately denounced violence in the name of a particular extremist ideology. In
33
short, they no longer identify themselves as adherents of a particular extremist ideology or are
affiliated with an extremist group or movement.
68
Data collection efforts followed the proper ethical procedures for conducting research
involving human participants. Here the former extremist was informed that their participation in
the study was entirely voluntary. They were also informed that they had the right to decline to
answer questions or to end the interview/withdraw from the study at any time. In addition, the
former was informed that they would not be identified by name in any publication, and that all
data collected from the interview would be de-identified for the purpose of ensuring participant
anonymity. One in-person interview was conducted with the former in June 2017 and was
approximately 10 hours in length. The interview was audio-recorded and transcribed.
69
This study was not an indictment of this sub-forum itself. The sub-forum was selected because
it was an online space that the former extremist actively participated in during his involvement in
violent RWE, meaning that they were familiar with the users who posted there and could identify
individuals who the former knew were violent or non-violent RWEs in the offline world.
70
See J. M. Berger, Extremism (Cambridge, MA: The MIT Press, 2018).
71
See Conway et al., “Right-Wing Extremists’ Persistent Online Presence.”
72
Tore Bjørgo and Jacob Aasland Ravndal, “Extreme-Right Violence and Terrorism: Concepts,
Patterns, and Responses,” The International Centre for Counter-Terrorism – The Hague
10(2019): p. 5.
73
The bacon function in Stata is a useful and parsimonious means of detecting outliers. This
function calculates Mahalanobis distances of a subset of observations from the whole sample and
any values outside of a designated threshold identified by a corrected percentile of a χ2
distribution are identified as outliers. For additional details about this process, see Sylvain
Weber, “Bacon: An Effective Way to Detect Outliers in Multivariate Data Using Stata (and
Mata),” The Stata Journal 10, no. 3 (2010): 331-338.
74
The highest value in the data was identified at 56,284 total posts for a single user.
75
The highest value in the data was identified at 3,839 sub-forum posts for a single user.
76
One user posted zero sub-forum posts during the 54 months that their activity was tracked.
This user was coded as being dormant during this time.
77
Analyses stop at month 30 because this is the latest point in the study period that dormancy
could be achieved based on the definition of dormancy being 24 consecutive months of zero new
posts and a 54-month study period.
78
Scrivens et al., “Measuring the Evolution of Radical Right-Wing Posting Behaviors Online”;
Wojcieszak, “‘Don’t Talk to Me’.”
79
Manuela Caiani and Patricia Kröll, “The Transnationalization of the Extreme Right and the
use of the Internet,” International Journal of Comparative and Applied Criminal Justice 39, no
4. (2014): 331-351; Futrell and Simi, “Free Spaces, Collective Identity, and the Persistence of
U.S. White Power Activism.”
80
Kleinberg et al., “The Temporal Evolution of a Far‑Right Forum.”
81
Scrivens et al., “Examining the Developmental Pathways of Online Posting Behavior in
Violent Right-Wing Extremist Forums.”
82
See Gaudette et al., “The Role of the Internet in Facilitating Violent Extremism.”
83
Kathleen M. Blee and Kimberly A. Creasap, “Conservative and Right-Wing Movements,”
Annual Review of Sociology 36 (2010): 269-286; Jeff Gruenewald, Joshua D. Freilich, and
Steven M. Chermak, “An Overview of the Domestic Far-Right and Its Criminal Activities,” in
Barbara Perry and Randy Blazak, Eds., Hate Crime: Issues and Perspectives, Vol. 4 Offenders
34
(New York, NY: Praeger, 2009), pp. 1-22; Mark S. Hamm, American Skinheads: The
Criminology and Control of Hate Crime (Westport, CT: Praeger, 1993); Barbara Perry and Ryan
Scrivens, Right-Wing Extremism in Canada (Cham, Switzerland: Palgrave, 2019); Simi and
Futrell, American Swastika.
84
Gaudette et al., “The Role of the Internet in Facilitating Violent Extremism.”
85
See Conway et al., “Right-Wing Extremists’ Persistent Online Presence.”
86
Horgan et al., “Actions Speak Louder Than Words”; Jasko et al., “Quest for Significance and
Violent Extremism”; Knight et al., “Comparing the Different Behavioral Outcomes of
Extremism”; LaFree et al., “Correlates of Violent Political Extremism in the United States.”
87
For more information on the ECDB, see Joshua D. Freilich, Steven M. Chermak, Jeff
Gruenewald, William S. Parkin, and Brent R. Klein, “Patterns of Fatal Extreme-Right Crime in
the United States”, Perspectives on Terrorism 12, no. 6 (2018): 38-51.
