Content uploaded by Ana-Maria Bliuc
Author content
All content in this area was uploaded by Ana-Maria Bliuc on Mar 07, 2019
Content may be subject to copyright.
1
Collective Identity Changes in Far-right Online
Communities: The Role of Offline Intergroup Conflict
Ana-Maria Bliuc, Western Sydney University
John Betts, Monash University
Matteo Vergani, Deakin University
Muhammad Iqbal, Victoria University
Kevin Dunn, Western Sydney University
Abstract
The role of online communities in shaping behaviours ‘in real life’ (IRL) is well-established,
however, less is known about how such communities are affected by offline events. Thus, we
investigate here the ways in which the collective identity of a far-right online community is
affected by offline intergroup conflict. That is, we examine over 14 years of online
communication between members of a national division of the global white supremacist
community Stormfront.org. We analyse members’ language use and discourse before and
after significant intergroup conflict in 2015, that is, local racist riots in Sydney, Australia. We
found that the riots were associated with significant changes in the collective beliefs of the
community (as captured by members’ most salient concerns and group norms), emotions and
consensus within the community. Overall, the effects of the local riots were manifest in a
reinvigorated sense of purpose for the far-right community with a stronger anti-Muslim
agenda.
2
Collective Identity Changes in Far-right Online
Communities: The Role of Offline Intergroup Conflict
Introduction
We live in a networked world where online communities are increasingly impactful ‘in real
life’ (IRL); most obviously, they enable people globally to connect under shared ideologies
and mobilise to achieve collective goals regardless of geographical distances and national
borders (Wojcieszak, 2010). Many of these ideologically-driven communities function by
nurturing networks of supporters who share the aim of achieving social change through
commitment to a common cause (such as gun control, animal rights, marriage equality, etc.).
The impact of online communities in the real-world has been extensively studied; for
instance, we have evidence that engagement with online communities affects interpersonal
relationships, personal empowerment, and recovery capital in online communities of support
(Barak et al., 2008; Best et al., 2018; Bliuc et al., 2017; Ishii and Ogasahara, 2006).
Engagement in politically-driven online communities was also shown to be connected with
‘offline’ collective action (Alberici et al., 2016; Lee et al., 2017; McGarty et al., 2014; Odağ
et al., 2016). However, we know far less about the potential effects of significant socio-
political events ‘in real life’ on online communities. In particular, in the domain of politics,
we have recently witnessed a global reinvigoration of the far-right movement, arguably aided
by online communication both within (ideologically-driven) online communities and between
these communities and their supporters (Awan, 2014; Bliuc et al., 2018b; Jakubowicz et al.,
2017; Lawrence and Boydstun, 2017; Oksanen et al., 2018; Zhang et al., 2017). That is, the
popularity of far-right political groups in Western democracies have been steadily on the rise,
with many of these groups moving to mainstream politics through winning parliamentary
representation in Western nations such as Australia, Netherlands, France, Austria, Denmark,
and Sweden (Bos, et al., 2017; Inglehart & Norris, 2016).
Thus, in this research, we focus on examining associations between events ‘in real
life’ and the internal dynamics of online communities from the extreme of the political
spectrum. In particular, we examine how far-right online communities may be affected by
historical significant socio-political events underpinned by intergroup conflict. To test
whether and how such events are associated with the socio-psychological processes in far-
right online communities, we examine changes occurring over a period of 14 years in the
collective identity of Stormfront Downunder, the Australian division of the major global
white supremacist community Stormfront.org. In Australia, Stormfront Downunder has been
for a long time a key online communication channel for the far-right movement and their
sympathisers (Jakubowicz et al., 2017). By collective identity, in this instance we refer to the
specific social identity content (including beliefs, values, norms and attitudes) associated to a
group membership (Livingstone and Haslam, 2008) which implies difference in how we see
the world rather than how we are (in terms of fixed social categories that define us).
We focus on Stormfront because it constitutes one of the most influential far-right
online communities in the world with branches in Europe, South Africa and Australia;
3
notably, notorious far-right terrorists Anders Breivik (responsible for the 2011 Norway
attacks) and Dylann Roof (responsible for the 2015 South Carolina attack) were linked to
Stormfront. More recently, in August 2017, Stormfront.org was shut-down in response to the
violence during the Unite the Right rally in Charlottesville, Virginia, which resulted in the
death of one counter-protester (Southern Poverty Law Centre, 2017). While the offline
impact of Stormfront is well-evidenced (see also Beirich, 2015), there is no research directly
investigating how significant socio-political events ‘in real life’ may impact this powerful
online community.
From a theoretical viewpoint, in the field of intergroup relations, how groups create
and transform the content of their collective identities is not very well-understood (Turner et
al., 2001). From research on opposing groups formed around mutually exclusive beliefs about
how the world should be - opinion-based groups such as climate change sceptics and
believers or supporters of political opponents (Bliuc et al., 2007; 2015; 2018a; Postmes,
2015), we know that intergroup conflict can increase group commitment. Thus, in a sense,
groups are empowered by conflict with opposing groups. Applying these ideas to online
communities, we go beyond investigating increased group commitment as an outcome of
intergroup conflict to examine whether intergroup conflict ‘in real life’ is associated with
changes in the collective identity of the online far-right community Stormfront Downunder.
In this case, we focus on the Australian context for two main reasons. First, in this particular
setting of Australian politics, we have a clear case of pronounced intergroup conflict which is
of significance for the local far-right community: the Cronulla Race Riots which occurred in
Sydney in December 2005. The riots represent a powerful illustration of sharp ideological
division and intergroup conflict (Bliuc et al., 2012). As the variations in levels of online
activity in Stormfront Downunder illustrated in Figure 1 show, the 2005 riots coincide with a
significant reinvigoration of the online community with an increased number of people
joining at that point and immediately after. The number of posts per week on the forum
stayed at levels significantly higher than the pre-riots levels for a long period of time (taking
approximately 3 years to return to the pre-2005 levels).
4
Figure 1. Weekly activity levels in Stormfront Downunder forum (2001 – 2015) reveals peak
members’ activity in the aftermath of the local race riots.
To identify change over time in the collective identity of Stormfront Downunder, we
examine the evolution of collective beliefs and emotions between 2001, when the online
community was formed in Australia, and 2015, a decade after the race riots, as reflected by
several indicators. Thus, we consider group members’ concerns and group norms
(representing the collective beliefs of the online community) as emerging from the
communication between members, collective emotions in the community and levels of
consensus within the community.
Group concerns, norms, emotions and consensus are captured by analysing
communication before and after the Cronulla race riots (pre- and post-December 2005). We
seek to establish whether the effects of the riots go beyond changes in members’ levels of
engagement in the online community (as captured by increased activity levels illustrated in
Figure 1), in the particular political context of Australia at that time; we predict that the local
race riots would produce changes in the collective beliefs of the online community - captured
as focus on particular concerns and content of group norms and mirrored by corresponding
changes in emotions and levels of consensus within the community.
Stormfront as an ideologically-driven online community
We propose that communities such as Stormfront are essentially opinion-based groups – that
is, psychological groups formed around a shared opinions, ideologies or worldviews (Bliuc et
al. 2007; 2015; 2018a). The collective identities of people identifying with these communities
are derived from specific versions of a national identity, but which are based on an
ideological view characterised by the justification of white privilege and associated values.
Previous research in Australia showed that the Cronulla race riots were crucial in the
rhetorical activation of national identity among White supremacists (Dunn 2009; Noble,
2009; Noble and Poynting, 2010). The Cronulla Riots were a key moment in racist politics of
Australia. At the time, people of non-Anglo appearances, who happened to be in the Cronulla
area, were abused and bashed by angry white (mostly young male) crowds. The young
appropriated songs and icons of Anglo-ness, and through these performances, a racially
privileged nationalism was asserted. The first riot was followed by subsequent incivility by a
handful of Lebanese-Australian youths, as retaliatory responses, and this provided great
material for racialised media attacks (Dunn, 2009). The official inquiry into the riots blamed
media coverage, especially talk radio, for inciting and inflaming local tensions into this
nationally shocking event. In keeping with an earlier refusal to condemn racist politics (anti-
Asian in that case) the Australian Prime Minister, John Howard, denied that the riots were
racist acts.
In the current research, we use the race riots as a reference point in our analysis. We
propose that they represent a clear illustration of intergroup conflict between two
ideologically-driven groups, broadly defined as supporters and opponents of multiculturalism
in Australia. The group consisting of opponents of multiculturalism (or supporters of the
riots) is ideologically aligned to the online community Stormfront Downunder, which means
that the riots likely represent a significant intergroup conflict for the online community that
we study.
Political context
In this article, our aim is to track not just the rate of activity and what it indicates
about breadth of association in the online community, but also examine changes in collective
5
social psychological indicators as reflected in the nature of the online contributions. These
changes are examined in relation to the local race riots which in turn occurred in particular
socio-political context. Specifically, the riots represent a materialisation of a political climate
in which racist politics had been achieving unheralded success within Australian mainstream
politics. From the mid-1990s the leader of the nationalistic One Nation Party, Pauline Hanson
had achieved electoral success in the Federal Parliament (in a lower house seat) and even
more success in the Queensland State Parliament. The One Nation Party had deployed a
particularly anti-Asian theme in their political rhetoric, eschewing the more traditional far-
right emphasis on Jews and Blacks. They referred to ghetto formation and cultural
incompatibility and de-emphasised ‘race’ per se, as was typical of the emerging new racist
movements globally (Dunn, 2009). This reveals the constructed and dynamic foci of racist
political movements. When Pauline Hanson came to power, the Australian Prime Minister at
the time (John Howard) did not condemn her racist attacks on Asian-Australians and this was
indicative of a growing permissiveness around overt and stridently racist political and media
commentary which later inflected with anti-Muslim attacks (overtly deployed for assumed
political gain, and strongly supported by sections of the tabloid and ‘infotainment’ media). A
decade later, Pauline Hanson re-entered the Australian Parliament, as a Senator for the State
of Queensland, and made exactly the same maiden speech as a decade before, only swapping
the words “Asian” for “Muslims”. She was tapping into the emerging Islamophobia in
western nations and using new racist arguments that she had seen as successful ten years
before (Briskman, 2017; Kamp et al., 2017).
Current Study
We analyse the language used in online social interactions by members of the online
community in the form of their posts and comments on the forum from 2001to 2015. Thus,
all user posts to the Stormfront Downunder (https://www.stormfront.org/forum/f38/) from
2001 to 2015 were extracted for the analysis. In total 75,795 posts were analysed, comprising
of 73,275 member posts and 2,520 anonymous posts (made under 5,489 threads). The data
extracted from each post consisted of a user identifier (or anonymous), thread identifier,
temporal stamp (time and date), and textual content.
Our research design aims at identifying differences in the language use and discourse
in the online community Stormfront Downunder before and after local race riots. We use
both quantitative and qualitative methods to examine the changes in collective identity of the
community as expressed in members’ communication in the form of the main concerns of the
group and group norms. In addition to these indicators, we also investigate emotions and
levels of consensus in the community. Specifically, we examine three language functions as
described in Table 1.
Table 1. Study design: methodological approach, research techniques and samples.
Function of
the language
Object of the
analysis
Technique used in
the analysis
Sample
Collective
beliefs as main
concerns of
the group
Most
frequently
used nouns
before and
after the riots
Natural Language
Processing (NLP)
approach consisting of
extracting and
comparing the most
used nouns before and
after the riots
Posts and comments before and after the
riots
6
Collective
beliefs as
group norms
Context and
content of the
posts before
and after the
riots
Qualitative thematic
analysis
The 100 most quoted posts in the forum
between 2001 and 2015. Our decision to
analyse the top quoted posts is based on
the assumption that these posts are likely
to capture content that is central to the
group’s collective identity; these posts
are drivers of intra-group interaction, so
they are likely to reflect the
development and change occurring in
the group’s normative content.
Emotions and
levels of
consensus)
All elements
of the
language
(including
nouns, verbs
and function
words)
Computerised
linguistic analysis
using Linguistic
Inquiry Word Count
(LIWC, Pennebaker et
al., 2007; 2015)
Posts and comments published 6 months
before and 6 months after the riots.
Results
Main concerns of the online community
Based on the analysis of changes in word usage, the riots seem to mark a clear qualitative
transformation in the main concerns of the community members. Before the riots, language
use reveals a focus on topics of discussion typical of white supremacist communities world-
wide. This includes discussions around ‘being white’, that is, race and colour, the Jewish
people, women and government, as well as around anti-Black and anti-Asian content. This is
evidenced in Table 2, which shows the most frequently used nouns in the timeframe
commencing at the beginning of the forum in 2002 through to the end of December 2005 (the
date of the riots). By contrast, comparing language use before and after the riots, the focus on
the geographic location of the riots, Cronulla (and the beach) and the prominence of concerns
with specific ethnic groups and especially followers of Islam becomes evident. Table 3 shows
the 50 most frequently used nouns having the greatest proportional change in use from the 6
months pre-riots to the 6 months post-riots. Average word frequency pre- and post- riots as
well as during the week of the riots are shown.
Table 2. The 50 most frequently used nouns occurring in the forum (excluding common
speech) prior to the riots ordered by rank. Variants of some nouns have been grouped as
indicated, *included aussies
Rank
Noun
Rank
Noun
Rank
Noun
Rank
Noun
Rank
Noun
1
white(s)
11
asian(s)
21
war
31
population
41
national
2
australia(n)(s)*
12
women
22
things
32
children
42
media
3
people
13
work
23
live
33
society
43
help
4
jew(s)(ish)
14
government
24
men
34
sydney
44
immigration
5
country(ies)
15
thing
25
nation
35
problem
45
family
6
time
16
day
26
life
36
thought
46
racial
7
world
17
fact
27
place
37
year
47
european
8
race
18
man
28
culture
38
point
48
power
9
years
19
something
29
money
39
hope
49
someone
10
black(s)
20
anyone
30
anything
40
support
50
group
7
Table 3. The 50 words having the greatest proportional increase in use during the 6 months
before and after the riots, ranked by increase. Average word frequency per week is shown
pre- and post-riots as well as for the week of the riots. Word variants have been grouped as
indicated.
The focus on the riots is evidenced by the prevalence of the term ‘Cronulla’ which
was used 190 times during the week of the riots and approximately 20 times per week, on
average, after the riots. Other words showing significant increases in usage also reflect higher
concerns with race, religion and culture. The key cultural groups that underpin the ideological
conflict figure prominently (“lebanese”, “greek”, “anglo”, “aussie”), as well as an increased
Word
Pre-
Riots
Riot
Week
Post
Riots
Increase
Word
Pre-
Riots
Riot
Week
Post
Riots
Increase
cronulla
0.04
190
20.04
521.00
immigration
5.00
48
16.42
3.28
anglo
0.38
11
14.31
37.20
forum
2.81
19
9.19
3.27
flag
1.23
67
21.50
17.47
media
6.85
160
21.85
3.19
leb(s)(anese)
2.77
364
41.81
15.10
news
3.42
68
10.88
3.18
greeks
0.81
30
11.19
13.86
north
2.27
12
7.15
3.15
pride
2.38
45
15.27
6.40
yeah
3.96
25
12.46
3.15
nsw
1.58
19
9.73
6.17
please
3.96
27
12.38
3.13
movie
1.69
3
10.19
6.02
chinese
2.88
1
8.88
3.08
eastern
1.81
93
9.65
5.34
lets
3.46
37
10.23
2.96
smoking
1.62
2
8.54
5.29
racism
2.65
30
7.81
2.94
violence
1.73
92
8.92
5.16
local
3.19
21
9.35
2.93
police
6.42
184
30.42
4.74
forget
2.38
10
6.81
2.85
middle
2.77
97
12.92
4.67
muslim(s)
19.19
134
54.27
2.83
islam(ic)
4.96
28
22.23
4.48
aboriginal(s)
10.54
12
29.69
2.82
beach
1.46
88
6.27
4.29
culture
8.38
32
23.65
2.82
stuff
2.92
16
11.88
4.07
evil
2.38
9
6.73
2.82
aussie(s)
13.65
344
53.92
3.95
read
6.35
25
17.88
2.82
ethnic
2.88
23
11.12
3.85
community
4.12
46
11.50
2.79
gangs
1.92
66
7.27
3.78
within
3.19
30
8.81
2.76
link
1.88
17
7.00
3.71
scum
2.73
87
7.50
2.75
attacks
1.88
29
6.92
3.67
especially
3.65
18
9.96
2.73
europe
4.12
44
14.38
3.50
general
2.54
38
6.92
2.73
multiculturalism
4.54
39
15.81
3.48
came
5.23
39
13.92
2.66
freedom
2.23
19
7.69
3.45
happened
3.27
31
8.58
2.62
game
3.96
3
13.38
3.38
soon
2.62
23
6.85
2.62
8
concern about Islam. The major offline change across that time axis was the Cronulla riots.
No other major changes occurred associated with the other nouns, other than how they were
implicated in the riots. The results (Tables 1-2; Figure 1) identify how the riots are the
independent variable during this change in online content and focus. By comparison
‘Cronulla’ was used, on average, 0.04 times per week (or once during 6 months) before the
riots, making it clear that the riots represent one of the key focus areas of the community.
Transformation of the group norms
To examine how group norms might change overtime in relation to the local riots, we
analysed the 100 most quoted posts in the Stormfront Downunder between 2001 and 2015.
After the posts were extracted and divided into two categories (i.e., posts made before the
riots, and posts made after the riots), we analysed their content using an approach derived
from thematic analysis (Clarke and Braun, 2013). Based on this analysis, two main thematic
categories in the content of group norms were identified:
a) Normative content about group boundaries – establishing ‘who are our allies’ and
‘who are our enemies’;
b) Normative content about collective self-definition – establishing ‘who we are as a
group’.
While the two broad thematic categories were found in both posts before and after the
riots – suggesting similar concerns around group boundaries and self-definition - we found
differences in how these normative categories were framed, that is, the arguments used to
construct these norms are different. Table 4 shows the key thematic categories before and
after the riots including some of the most illustrative extracts for each of these categories.
Norms about group boundaries. Our analysis suggests that before the riots members of the
online community were mostly preoccupied with establishing group boundaries by clarifying
the group attitudes towards perceived allies such as the British royals (see extract 1 in Table
4), and by using essentialist arguments – establishing group membership based on an
‘evidence-based’ assessment of Australians’ ethnic origins. The content on norms about
group boundaries before the riots seems to reflect a positioning of the group as an Anglo-
Saxon nationalist group, and generally, the discussions about group boundaries is both more
positively framed (in relation to allies rather than enemies) and more inward looking/oriented
(identifying essentialist criteria for group membership).
However, the content discussed after the riots reflects a more specific positioning not
in relation to allies, but to outgroups which are in general viewed as enemies of the group.
After the riots, group discussions more directly refer to Muslim Australians. Specific
references to the riots are made to build an argument for exclusion based on cultural
incompatibilities (extract 2, Table 4). Ironically, an argument for strategic group
inclusiveness is also made, that is, that group boundaries should become more flexible so that
southern Europeans who share white supremacist (and specifically anti-Muslim) values can
join the group. The argument is framed around shared ideological aspects such as whiteness,
Christianity, and the perceived common goal of stopping any culturally non-white migration
(extract 4).
Norms about collective self-definition. Before the riots, discussions of group self-definition
seem to be taken again on a more inward focus, that is, the members of the community focus
on group validation in relation to, for instance, appropriate markers of group identity
(physical appearance and grooming that would conform to group prototypes) and broader
ideological positioning in relation to beliefs about racism (extract 5).
9
After the riots, however, the discussions are focused on ideological positioning of the
group, but this being done in a more specified manner as, for instance, vis-a-vis particular
outgroups, e.g., Asians and Muslims (extract 6). The extracts also illustrate differences in
tone - the discussion about Asian outgroups (extract 6) is more neutral in tone and includes
no direct hostility, while discussions about Middle Easterners and Muslims (extracts 2, 8, and
9) are arguably more negative and emotionally charged. It is interesting to observe that the
framing of the arguments changed, so after the riots the group has a stronger outgroup focus.
This aligns with the accumulating Islamophobia of these times, and the later emergence of
other events, such as Pauline Hanson’s return to power on anti-Muslim politics. It is also a
strong reflection of the increasing political permissiveness of racism at the time, and of a
decreased inhibition at being blatantly racist in the public sphere.
Table 4. Illustrative quotes suggesting changes in the main thematic categories in the most
quoted posts in the forum before and after the riots
10
Thematic categories before the riots
Thematic categories after the riots
Norms about group boundaries: attitudes to allies
(support for British royalty)
1.“(…) As we enter the 21st century Prince Harry is
heading off to Iraq once he graduates from Sandhurst
military college, to be with the guys on the front line.
(…) I'm writing to PM Howard to request that 5-6
Aussie special forces cadre (perhaps more) have the
honour of accompanying our prince into the field, as
brothers in arms. That way (…) he will have Australia at
his side in more than just spirit.”
Norms about group boundaries: attitudes to specific
outgroups (Muslims)
2. “ (…) As you can see already the backlash by the
lebanese has been far more brutal and savage than that
dealt by a small number of angry Australians on
sunday at Cronulla. This was expected, their culture
and psyche is more force and revenge oriented, they
strive to enforce their male ego and establish
dominance, coupled with the belief of superiority and
cultural strength. (…) Basically they are not flexible
and tolerant enough to conform to the Australian way
of life.(…)”
Norms about group boundaries: essentialist arguments
3. “It has taken me a surprisingly long time to gather
together and verify enough statistics to make an attempt
at calculating Australia's ethnic origins. The
classifications I have used are mainly my own and I
realise there may be some people who will consider
them incorrect; but I've done my best. (…)”
Norms about group boundaries – making the identity
more inclusive for like-minded people
4. “it is time for all the Maltese, the Italians, the
Greeks, the Croats etc. to start joining these forums. is
time to start coming together in a way we all have
never really done before. because whats happenning to
the whites in Australia is basically happenning in your
own countries of origin, of which i am sure alot of you
have family over there. time for the anglos to ditch the
use of the word wog and its time for mediterraneans to
become more associated with the anglos. and together
we can say to the gutless leaders of this country, NO
MORE NON-WHITES! EUROPEAN AND WHITE
MIGRATION ONLY! or else we can all cop what
Malta is copping (…)
Response: (…) Most of these people are Christians and
have no time for the muslims. (…) They have given
much to Australia and Australia has been good in
return. It is time to think of ONE NATION, ONE
RACE, NO MORE NON-WHITE IMMIGRATION
AND DEFINITELY NO NON-WHITE REFUGEES.”
Norms about self-definition - physical appearance
(visible markers of identity)
5.“I regard my self as white national but I keep my hair,
I am 42 with short hair combed to the side. I have seen
quite a few that shave there heads I have seen pictures of
baron. What about you guys and for you guys that do
shave your heads what are your motives for doing that.”
Norms about self-definition – ideological positioning in
relation to outgroups (Asian)
6. “Are proud white Australians against non whites
living here? or are we just unhappy with immigrants
not conforming to our way of life and our Aussie
values? I'm currently working with an asian who likes
to have a beer and watch the footy, and who I think is
not such a bad bloke, he has good work ethics and
supports his family well. Does this mean I'm starting to
soften my attitude? Or do I just believe that all are
welcome as long as they adopt Australian values?”
Norms about self-definition – broad ideological
positioning
7. “What is Racism? Racism is the belief that human
races exist, that these races possess different abilities and
different characters, or natures, and that some races are
better at some things than other races. Racists do not
want to live in multi-racial societies. Instead, they want
to live among their own kind.”
Norms about self-definition: ideological positioning in
relation to outgroups (anti-Muslim)
8. “the aussie flag has just been banned from sydney
schools because apparently it "entices racial hatred" yet
the muslim "hijab" is not banned”
9. “This identical conflict happened once before in
Spain, our White people forced into all out victory in
war or lose and be enslaved by Arab Muslims.(…)”
11
Collective emotions and consensus in the online the community
Table 5 shows an extract from the computerised linguistic analysis (LIWC analysis,
Pennebaker et al, 2007; 2015) for the 50 most active users over the period one year prior to
one-year post riots. To increase the reliability of the analysis, only posts containing more than
20 words were included in the analysis (Boyd et al., 2016). As Table 5 shows, computerised
linguistic analysis conducted on online communication before and after the riots shows
significant changes in the categories capturing negative emotions such as ‘anger’ and ‘swear
words’ increased in frequency of use, while the overall category ‘negative emotions’ stayed
the same. Computerised linguistic analysis shows that the internal levels of consensus as
indicated by variations in the linguistic category ‘assent’ stayed the same, while ‘negation’
and ‘differentiation’ categories significantly decreased (see also Table 5).
Table 5. LIWC word categories and their increase in their usage after the riots.
Discussion
This study provides evidence that the collective identity of the far-right online community
Stormfront Downunder changes in relation significant intergroup conflict ‘in real life’.
Changes in the groups’ main concerns and norms, collective emotions and levels of
consensus within the community are examined using a mixed-method approach that
combines the use of natural language processing tools, thematic analysis, and computerised
linguistic analysis. Our analysis reveals that significant transformations in the collective
identity of the online community can be observed following local racist riots underpinned by
sharp ideological conflict. However, the first and most significant finding regards the amount
times the word ‘Cronulla’ was used in the forum in the week and aftermath of the riots
indicating that this was a significant event for Stormfront, even though it was not an event of
their making.
Second, our analysis of the most used nouns before and after the riots indicates that
there is a change from a focus on generic, exclusivist topics which are broadly aligned to
(global) white supremacy principles (such as race, nationalistic values, outgroups such as
Jews, etc.) to a more crystalised focus on (local) specific ethnic and religious out-groups. The
focus on religious or cultural groups has been an emphasis of the new racism politics, but
hitherto in Australian that had been on Asians, criticised as culturally inferior of incompatible
(Dunn et al., 2004). The increased use of words related to particular outgroups (such as
“Lebs”, “wogs”, and “Muslims”) was accompanied by an increase in words suggestive of
violence such as “bash”, “violence”, and swear words. Before the riots, we can observe a
more general focus in a way that could apply to most white supremacist groups in the
Western world – it is only the mention of Australia that can pinpoint the community to a
geographical location. When examining the nouns that had the greatest increase in use after
LIWC
Categories
%
Change
t
p
SE
Means
Pre-riots
SD Pre-
riots
Mean
Post-
riots
SD Post-
riots
Anger
9.742
2.685
0.004
0.034
0.874
1.405
0.965
1.563
Swear words
29.822
4.836
0.000
0.017
0.220
0.668
0.302
0.845
Negative affect
2.881
1.201
0.115
0.052
2.121
2.209
2.183
2.272
Assent
3.104
0.468
0.320
0.018
0.267
0.778
0.275
0.771
Differentiation
-5.248
-2.854
0.002
0.059
3.311
2.521
3.143
2.535
12
the riots, we can immediately see that more specific references to ethnic and religious groups
coupled with aggressively charged words (e.g., “gangs”, “evil”, “scum”), indicating a clear
change, not only in the main concerns of the group, but also in the general affective tone of
communication. The commentary was picking up on an emergent Islamophobia that had also
underpinned the media and political foment that lead up to and followed the riots. Pauline
Hanson would adopt the same strategy in five years-time on her return to Parliament. The
riots were a crucible for this change in language by Stormfront, which facilitated a growth in
online membership and activity.
Thirdly, the thematic analysis of the 100 most quoted posts reveals changes in the
group norms, in particular, in relation to how group boundaries are conceptualised and how
group members define themselves as a collective entity. The analysis suggests that before the
riots, the community would predominantly discuss norms about group boundaries in relation
to attitudes to allies (‘who is on our side’ type of argument), and essentialist criteria of group
inclusion (Haslam et al., 2006). Collective self-definitions were mostly concerned with
establishing what visible markers of identity are required from members and arguments
around the ideological bases of group membership (in relation to racism and other ideological
points relevant for the group). The normative content has evolved after the riots to be more
aligned with a highly focalised, unambiguous oppositional stance (against outgroup/s)
evident when debating both norms about group boundaries and collective self-definition. The
findings of this analysis are consistent with the previously discussed findings and suggest that
the community has become more about strictly defined group boundaries designed to keep
people out, rather than let people in (with the exception of the strategical argument of
opening the group for ideologically aligned people who can provide strength by numbers),
and self-definition through differentiation from outgroups. The identity content also becomes
more clearly centred on opposition to Muslims.
Finally, the findings of the computerised linguistic analysis regarding changes in
collective emotions and levels of consensus are mixed. That is, while the overall category
‘negative emotions’ did not change, it is significant that anger increased. In LIWC, the
‘negative emotions’ category includes anxiety and sadness related words which are
suggestive of passivity and resignation. On the contrary, anger is often associated with a
sense of injustice on the group’s behalf which, together with commitment to the group and
perceived group efficacy, is a precursor of collective action mobilisation (Duncan, 2012; Van
Zomeren et al., 2008). Thus, it is plausible that anger as a collective emotion has had an
energising role for the group members which over time was reflected in more sustained real-
world activity which could have been associated with the gradual increase in far-right active
groups in Australia and their levels of activity (Rydgren, 2018).
In relation to levels of consensus, while there was little change in the use of ‘assent’
words, ‘differentiation’ words decreased. These findings show that levels of agreement in the
online communication roughly stayed the same before and after the riots, but levels of
disagreement decreased. This is a subtle difference, but nevertheless it indicates a climate
characterised by increased consensus. Thus, after the riots, the community seems to have
experienced less disagreement, thus likely stronger attraction between members. This is
consistent with findings of stronger structural cohesion of the group in the aftermath of the
riots (Bliuc et al., 2016) and provides support for the argument that communication within
information bubbles characterised by ideological homophily can increase polarization and
radicalisation (Kaakinen et al., 2018).
Conclusion
13
The current study focuses on the effects of a local socio-political event on a specific
ideologically-driven online community. A large amount of data (i.e., communications
between group members over 14 years) was analysed using mixed methods, but to be fully
confident about generalising these findings to other online communities, similar analyses
should be conducted in other groups. In the Stormfront Downunder, the riots produced
unprecedented levels of activity; no other offline events with effects close in magnitude were
identified in the timeframe of the study. However, in other communities it may be that more
than one key event affects the community, so examining these effects in future research could
tell us more about how online communities are shaped by offline events and how in turn
these communities may affect behaviours and collective action in the ‘real-world’.
Overall, our findings suggest that the collective identity of this particular online
community studied has significantly changed in the aftermath of strong intergroup conflict in
the real-world. Our analysis shows strong associations of collective beliefs, emotions and
levels of consensus in the online community to events ‘in real life’ and ultimately to the local
political climate at the time, suggesting that that far-right online groups are sensitive (and
perhaps reactive) to what happens in society, in particular, if the events are directly relevant
to the group. The results suggest that the riots played a key role in unifying and giving
purpose to supporters of white supremacism in Australia by helping them clarify and
crystallise their collective beliefs and enhance their reach.
References
Alberici AI and Milesi P (2016) Online discussion, politicized identity, and collective
action. Group Processes and Intergroup Relations 19: 43-59.
Awan I (2014) Islamophobia and Twitter: A typology of online hate against Muslims on
social media. Policy and Internet 6: 133-150.
Barak A, Boniel-Nissim M and Suler J (2008) Fostering empowerment in online support
groups. Computers in Human Behavior 24: 1867-1883.
Beirich H (2015) 20 Years of Hate. Southern Poverty Law Center Intelligence Reports;
https://www.splcenter.org/fighting-hate/intelligence-report/2015/20-years-hate.
Best D, Bliuc AM, Iqbal M, Upton K and Hodgkins S (2018) Mapping social identity change
in online networks of addiction recovery. Addiction Research and Theory 26: 1-11.
Bliuc AM, Best D, Iqbal M and Upton K (2017) Building addiction recovery capital through
online participation in a recovery community. Social Science and Medicine 193: 110-
117.
Bliuc AM, Betts J, Faulkner N, Vergani M, Chow RJ and Iqbal M. Group Cohesion in Online
White Supremacist Communities – Integrating Social Network and Computerised
Linguistic Analyses (2016) Vox-Pol Mid-Project conference: Taking Stock of
Research on Violent Online Political Extremism.
Bliuc AM, McGarty C, Badea C and Boza M (2018a) Enacting collective support for the
European integration: Participation in pro‐integration action and preference for
specific transnational acculturation strategies. Journal of Community and Applied
Social Psychology 28:15-28.
Bliuc AM, McGarty C, Hartley L and Muntele Hendres D (2012) Manipulating national
identity: the strategic use of rhetoric by supporters and opponents of the ‘Cronulla
riots’ in Australia. Ethnic and Racial Studies 35: 2174-2194.
Bliuc AM, McGarty C, Reynolds K and Muntele D (2007) Opinion‐based group membership
as a predictor of commitment to political action. European Journal of Social
Psychology 37: 19-32.
14
Bliuc AM, McGarty C, Thomas EF, Lala G, Berndsen M and Misajon R (2015) Public
division about climate change rooted in conflicting socio-political identities. Nature
Climate Change 5: 226-229.
Bliuc AM, Faulkner N, Jakubowicz A and McGarty C (2018b) Online networks of racial
hate: A systematic review of 10 years of research on cyber-racism. Computers in
Human Behavior 87: 75-86.
Bos L, Sheets P and Boomgaarden HG (2018) The Role of Implicit Attitudes in Populist
Radical‐Right Support. Political Psychology, 39: 69-87.
Boyd RL, Wilson SR, Pennebaker JW, Kosinski M, Stillwell DJ and Mihalcea R (2015)
Values in words: Using language to evaluate and understand personal values. Paper
presented at the ICWSM.
Briskman, L. (2017). The politics of state-sanctioned violence in Australia: racialized
constructions of nation. In R. Mason (Ed.), Legacies of Violence: Rendering the
Unspeakable Past in Modern Australia (pp. 11-29).
Clarke V and Braun V (2013) Teaching thematic analysis: Overcoming challenges and
developing strategies for effective learning. The Psychologist 26: 120-123.
Duncan LE (2012) The psychology of collective action. In The Oxford handbook of
personality and social psychology.
Dunn KM (2009) Performing Australian nationalisms at Cronulla. Lines in the sand: The
Cronulla riots, multiculturalism and national belonging, pp.76-94.
Dunn KM, Forrest J, Burnley I and McDonald A (2004) Constructing racism in Australia.
Australian Journal of Social Issues 39: 409-430.
Haslam N, Bastian B, Bain P and Kashima Y (2006) Psychological essentialism, implicit
theories, and intergroup relations. Group Processes and Intergroup Relations 9: 63-
76.
Inglehart R and Norris P (2016) Trump, Brexit, and the rise of populism: Economic have-
nots and cultural backlash. HKS Faculty Research Working Paper Series RWP: 16-
026.
Ishii K and Ogasahara M (2006) Links between real and online networks: A comparative
study of online communities in Japan and Korea. CyberPsychology, Behavior &
Social Networking 10: 252-257.
Kaakinen M, Sirola A, Savolainen I and Oksanen A (2018) Shared identity and shared
information in social media: development and validation of the identity bubble
reinforcement scale. Media Psychology, doi.org/10.1080/15213269.2018.1544910
Kamp A, Alam O, Blair K and Dunn K (2017) Australians’ views on cultural diversity,
nation and migration, 2015-16. Cosmopolitan Civil Societies: an Interdisciplinary
Journal 9: 61-84.
Jakubowicz A, Dunn KM, Mason G, Paradies Y, Bliuc A-M, Bahfen, N, Atie, R, Connelly K
(2017) Cyber Racism and Community Resilience, Palgrave Macmillan: Basingstoke.
Lawrence RG and Boydstun AE (2017) What we should really be asking about media
attention to Trump. Political Communication 34: 150-153.
Livingstone A and Haslam SA (2008) The importance of social identity content in a setting
of chronic social conflict: Understanding intergroup relations in Northern Ireland.
British Journal of Social Psychology 47: 1-21.
Lee FL, Chen HT and Chan M (2017) Social media use and university students’ participation
in a large-scale protest campaign: The case of Hong Kong’s Umbrella
Movement. Telematics and Informatics 34: 457-469.
McGarty C, Thomas EF, Lala G, Smith LG and Bliuc AM (2014) New technologies, new
identities, and the growth of mass opposition in the Arab Spring. Political Psychology
35: 725-740.
15
Noble G and Poynting S (2010) White lines: The intercultural politics of everyday movement
in social spaces. Journal of Intercultural Studies 31: 489-505.
Noble G (2009) Lines in the sand: The Cronulla riots, multiculturalism and national
belonging. Institute of Criminology Press.
Odağ Ö, Uluğ ÖM and Solak N (2016) “Everyday I’m çapuling”: Identity and collective
action through social network sites in the Gezi Park protests in Turkey. Journal of
Media Psychology: Theories, Methods, and Applications 28: 148-156.
Oksanen A, Kaakinen M, Minkkinen J, Räsänen P, Enjolras B and Steen-Johnsen K (2018)
Perceived Societal Fear and Cyberhate after the November 2015 Paris Terrorist
Attacks. Terrorism and Political Violence, 1-20.
Pennebaker J W, Booth R and Francis ME (2007) Linguistic inquiry and word count: LIWC
[Computer software]. Austin, TX: liwc. Net.
Postmes T (2015) Psychology: Climate change and group dynamics. Nature Climate Change
5:195.
Rydgren, J. (2018) The Oxford Handbook of the Radical Right. Oxford University Press.
Southern Poverty Law Centre (2017) Waning Storm: Stormfront.org Loses Its Domain.
https://www.splcenter.org/hatewatch/2017/08/29/waning-storm-stormfrontorg-loses-
its-domain.
Turner JC and Reynolds KJ (2001) The social identity perspective in intergroup relations:
Theories, themes, and controversies. Blackwell handbook of social psychology:
Intergroup processes 4: 133-152.
Van Zomeren M, Postmes T and Spears R (2008) Toward an integrative social identity model
of collective action: A quantitative research synthesis of three socio-psychological
perspectives. Psychological Bulletin 134: 504-535.
Wojcieszak, M (2010) ‘Don’t talk to me’: effects of ideologically homogeneous online
groups and politically dissimilar offline ties on extremism. New Media and Society
12: 637-655.
Zhang Y, Wells C, Wang S and Rohe K (2017) Attention and amplification in the hybrid
media system: The composition and activity of Donald Trump’s Twitter following
during the 2016 presidential election. New Media & Society 20: 3161-3182.
Acknowledgments
Funding: The authors are grateful to the Australian Research Council, Australian
Human Rights Commission, and the Victorian Health Promotion Foundation for
funding this project.
Competing interests: There are no competing interests for any of the authors.
A preview of this full-text is provided by SAGE Publications Inc.
Content available from New Media & Society
This content is subject to copyright.
