Content uploaded by Anne Birgitta Nilsen
Author content
All content in this area was uploaded by Anne Birgitta Nilsen on Dec 17, 2022
Content may be subject to copyright.
ESSACHESS – Journal for Communication Studies
Article received on August 20, 2021. Article accepted on May 16, 2022.
Conflict of Interest: The author(s) declare(s) no conflict of interest.
Cite: PATON, N., NILSEN, A. B., DECHESNE, M., SAKELLARIOU, A., HELM,
G., & SALORD, T. (2022). The European Far Right and Islamist Extremism on
Twitter: From Radicalisation to Political Participation. ESSACHESS.
https://doi.org/10.21409/VAQQ-S725
Grant HELM
PhD Candidate, University of
Lancaster, UNITED KINGDOM
e-mail: grant@moonshotteam.com
Engineer, INREA,
Tristan SALORD
Research
FRANCE
e-mail: tristan.salord@inrae.fr
Guillaume CABANAC
Professor, University of Toulouse,
FRANCE
e-mail: guillaume.cabanac@univ-tlse3.fr
Nathalie PATON
PhD, School for Advanced Studies in
the Social Sciences (EHESS),
FRANCE
e-mail: nathalie.paton@gmail.com
Anne Birgitta NILSEN Professor,
Oslo Metropolitan University
(Oslomet), NORWAY
e-mail: abin@oslomet.no
Mark DECHESNE
Professor, University of Leiden,
NETHERLANDS
e-mail: m.dechesne@fgga.leidenuniv.nl
Alexandros SAKELLARIOU
Professor, Panteion University
(PUA), GREECE
e-mail: sociology.panteion@gmail.com
The European Far Right
and Islamist Extremism on
Twitter: From
Radicalisation to Political
Participation
ESSACHESS –
Journal for Communication Studies
Volume 15 Issue 2(30), p. 13-46
© The Author(s) 2022
Reprints and Permission:
© ESSACHESS
https://www.essachess.com/
DOI: 10.21409/essachess.1775-352x
PATON ET AL.
14
Abstract: This article explores the results of a study on media participation on Twitter
in 2018/2019 perceived as contributing to far right and Islamist radical ideologies, in
7 European countries. By combining online ethnography and big data approaches, we
see that online far right extremism in Europe is active and ranks are growing, while
Islamist extremism has been incapacitated in sharing controversial forms of
expression. We describe how the far right uses Twitter as a means for political
activism, while Islamist extremists offer lifestyle information, related to local
branches of faith, using Twitter as a storefront that re-routes users to other platforms.
We consider resources for action, notably on the far right. World leaders influence
followers online and support a global conversation between users that paves the way
to a far right European milieu thriving on Twitter. To conclude, we delve into the
notions of radicalisation and political participation to emphasise the asymmetry
between the two forms of media participation in respect to academic discourses and
big tech and States’ practises.
Keywords: far right extremism, Europe, social media, Islamist extremism, political
participation
***
L’extrême droite et l’extrémisme islamiste sur Twitter : de la radicalisation à la
participation politique
Résumé : Cet article explore les résultats d'une enquête sur Twitter entre 2018-2019
portant sur la participation médiatique aux idéologies de l'extrême droite et de
l'extrémisme islamiste, dans 7 pays européens. En combinant des approches
d'ethnographie en ligne et de big data, nous montons que l'extrême droite est active,
tandis que l'extrémisme islamiste est mis à mal dans sa capacité à exprimer des idées
controversées. Nous décrivons les deux répertoires d'action : l'extrême droite utilise
Twitter comme moyen d'activisme politique ; les islamistes extrémistes y partagent
des informations relatives à un style de vie, en s’appuyant sur des interprétations
locales de la foi, et emploient Twitter comme une vitrine. Puis, nous considérons les
ressources à l’action susceptibles d’amener les utilisateurs à la radicalisation,
notamment pour l’extrême droite. Nous expliquons comment l'identité de groupe, liée
à une conversation globale, est soutenue par les dirigeants mondiaux, et cela ouvre la
voie à un milieu européen d'extrême droite. Pour conclure, nous approfondissons les
notions de radicalisation et de participation politique pour souligner que l’asymétrie
tient certes aux participations médiatiques des citoyens, mais aussi aux discours
académiques et aux pratiques des big tech comme des États.
Mots-clés : extrême droite, extrémisme islamiste, Europe, réseaux sociaux,
participation politique
***
ESSACHESS vol. 15, no. 2(30) / 2022
15
Introduction
Studying Islamist Extremism and the Far Right in a Cross-country Perspective
This paper explores the results of a two-year study on Twitter investigating citizens’
media participation in 2018/2019, perceived as contributing to far right and Islamist
extremism ideologies, in seven European countries (Belgium, France, Germany,
Greece, the Netherlands, Norway and the United Kingdom) to further understand the
role of social media in the process of engagement in radical causes.
In little under a decade, we have witnessed a rise of the far right in major
democracies with their election at heads of State, such as in the United States and
Brazil. Such surges of extremism, while not limited to anti-Islamist arguments, are
nonetheless fuelled by them throughout Western countries. In parallel, Islamist
extremism has made its way into the daily lives of European citizens with a number
of terrorist attacks over the last twenty years, as well as the highly-mediatized
departure of European citizens to Syria and Iraq to fight for the Islamic State. These
general tendencies are discussed in this paper based on field observations from seven
European countries.
To understand how political extremism shapes the contemporary world, research
benefits from comparative perspectives. We do so by targeting a group of European
countries, currently experiencing strong radicalisation trends with an increase in far
right votes and/or Islamist extremism. Dealing with political participation within the
European setting provides a standpoint to observe common trends regarding how the
issues of radicalisation and extremism emerge, as well as how they are perceived in
relation to a more global geopolitical landscape.
Islamist extremism and the far right are amalgamated here in the expression
radicalisation. Like extremism, radicalisation is a relational concept reflecting
distance from moderate, mainstream or status quo positions. However, radicalisation
and extremism are not equated in this paper. Rather radicalisation is defined here as
“the process by which individuals or groups come to embrace attitudes, or engage in
actions, that support violence in the pursuit of extremist causes.”
1
This makes its
definition path-dependent upon what is understood as ‘extremism’. Extremism refers
to beliefs and behaviour that seek a conformist, homogenous society based on
ideological tenets achieved, maintained or defended by force, violence, hierarchy and
the prioritisation of uniformity over diversity and collective over individual goals. In
this sense, extremism is rather a belief system, while radicalisation is a process leading
to extremism. Both, however, are understood as forms of political participation. They
are forms of collective action leaning towards destruction, but nonetheless, they fall
within the spectrum of engagement and political practices. In this paper, we intend to
1
The definition of the concept as exploited in the H2020 DARE project can be found at the following link:
http://www.dare-h2020.org/concepts.html
PATON ET AL.
16
discuss radicalisation in Europe in a cross-country perspective to better understand
the making of contemporary forms of political participation in its most extreme
expressions.
The role of the Internet in the phenomena of radicalisation
It is a well-known fact that far right and Islamist extremist organisations alike have
used big tech platforms, such as Facebook, YouTube or Twitter, to rally their
audiences, engage new recruits and build a larger political movement.
The notion of targeted ‘recruitment’ on social network sites, and possible ‘self-
radicalisation’ through one's' Internet uses, triggered a debate starting in the years
2005/2010 in line with the idea that violence spread online. Youth were identified as
a target for ‘fertilisers’ or ‘radicalisers’ (O’Neill & McGrory, 2006; Hoffman, 2009).
Papers evoked the strategies deployed by Al Qaeda (Cheong & Halverson, 2010;
Venhaus, 2010) or ISIS to encourage western-born Muslims to identify with Islamist
extremism (Picart, 2015; Saleh, 2014). Since then, some of the dominant Islamist
terrorist organisations have diminished in force in the field, especially with the
collapse of ISIS territorially in 2018. Simultaneously, takedown campaigns have
become commonplace online, preventing any actual direct recruitment, causing the
whole academic debate to move away from what still takes place online on major
platforms in support of Islamist extremism.
In general, concern in academia has shifted from Islamist extremism to media
participation of the far right in social networking sites (Medina-Serrano et al. 2018;
Neudert et al., 2017), and its role in the radicalisation process (Hassan et al. 2018;
Koehler,, 2014). Beyond academic debates, national states take these threats more
seriously, as is the case in the Netherlands, for example, where the Dutch security
services have expressed concern about the rise of the far right and the role that social
media has in the spreading of extreme-right content (AIVD, 2018).
Whatever the ideological standpoint, social networking platforms provide access
to digital arenas of public debate and large communities of practices that could pave
the way to the development of networks of like-minded people (Postmes et al. 2014).
In an era when counter-hegemonic groups may emerge through counter-narratives,
when censorship from gatekeepers is weakened and know-how in terms of political
tactics, techniques, and strategies is shared openly and freely (Paton, 2015), it is
important to pursue the scientific understanding of how social media provide means
for radical ideologies to spread. We do so by focusing on digital uses on Twitter. The
decision to focus on Twitter is based on a burgeoning body of academic literature
highlighting the volume (Berger & Morgan, 2015) and geographic segmentation
(Berger, 2018) of radical Twitter networks.
ESSACHESS vol. 15, no. 2(30) / 2022
17
1. Research Questions and Methodology
1.1. Research Questions and Structure
This paper contributes to this area of research by tackling four levels of enquiry before
summing up findings in a final conclusion. First, we briefly describe the state of online
participation for far right and Islamist extremism users to determine whether both
strands of extremism are active on Twitter in 2018/2019, and if so, whether their ranks
are growing.
Then, moving away from observations regarding whether something is taking
place, a second section identifies what is taking place. This part deals with the main
‘repertoires of action’ (Tilly, 2013) - meaning the dominant forms of collective action
- for each ideology. Nowadays, civic action is increasingly convergent, combining
online and offline activities, and relies on distributed trust across horizontally linked
citizens, given that citizens favour collective discussions through horizontal and
decentralised networks over claims from a single authoritative information source
(Chadwick, 2007; Meraz & Papacharissi, 2013; Figeac et al., 2021). Studying
repertoires of action is a way to understand the inherent logics behind media
participation for each ideological strand, and characterise the main features of each
group - assuming that, and to the extent that, either or both strands create an actual
group.
Thirdly, we question whether Twitter offers online resources for radicalisation.
We will consider to what extent it is possible to support claims that the Internet
harbours radicalisation triggers. In doing so, we tackle questions such as: what makes
these forms of digital sociability attractive? What do participants identify with? Who
are those setting the tone and driving conversations? These questions informing
resources for action find answers through three levels of enquiries pertaining to
discussions about influencers, topics and networks of conversation.
Finally, we reflect upon the uses of the expressions “radicalisation” and “political
participation.” Radical ideologies are not a set of characteristics, but an interactive
process between hegemonic and counter-hegemonic movements, strongly dependent
on the context and the situation of the symbolic productions. This section captures
how similar phenomena can be framed in different lights to the extent of invalidating
one, while downplaying the democratic threats of the other.
1.2. Target Groups: Personal Accounts from Seven European Countries
The target groups for this study are accounts, set up by people generally born and
raised in seven European countries (Belgium, France, Germany, Greece, the
Netherlands, Norway and the United Kingdom), currently experiencing strong
radicalisation trends (Figure 1). However, it was not possible to find data for Islamist
extremism in Greece and therefore no research was carried out on Islamist extremism
in Greece.
PATON ET AL.
18
Figure 1. Map of the countries investigated
Given that the goal was to compare the political engagement of private citizens, only
open personal accounts of engaged citizens were selected. Hence, the lack of political
leaders, organisations and prominent figures amongst the accounts in our study. The
research design called for 50 ‘personal’ accounts for each gender (women, n=25; men,
n=25) per stream of radicalisation, amounting to a total of 100 personal accounts per
country.
In respect to GDPR laws, we did not collect sensitive data. In selecting accounts
manually during the ethnographic phase of the study, we were cautious to select
accounts in which people specified their country of origin.
1.3. Criteria for the Selection of Accounts
Accounts were selected through an initial phase of direct observation, i.e. online
ethnography (from Sept.-Dec. 2018). 20+ criteria were used to single out accounts
(Table 2); accounts had to correspond to at least 5 of these criteria to be considered as
relevant to the study.
Table 2. List of criteria used by researchers to select accounts
List of criteria
- Anti-immigration
- Ultra-nationalism
- Superiority
- Focus on purity
- Violence
- Authoritarianism
(hierarchical structure
with a strong leader)
- Anti-democratic
- Victimhood (e.g. unjustly
- Religious
fundamentalism
(Catholic, Orthodox
or Islamic
extremism)
- Anti-politically
ESSACHESS vol. 15, no. 2(30) / 2022
19
- Misogyny
- Ultra-nationalism
- Promotion of gender roles
- Conspiracy
imprisoned)
- Militaristic
- Anti-system
- Hate speech
- Salafism
- Racism
correct
- Anti-semitism
- Martyrdom
- Jihad
Additional techniques were used, notably: interviews with experts from the field,
snowballing, and automated searches based on keywords.
1.4. Research Methods and Composition of the Sample
The mixed-method approach adopted for this study combined digital ethnography
and big data techniques and was implemented across 3 phases. During the first phase,
from September to December 2018, researchers conducted direct ethnography and
collected 712 Twitter accounts. Of these 712 accounts, 116 subsequently became
unavailable, leaving 596 accounts in the final database (see Figure 2).
Figure 2. Full sample within the database with numbers of accounts, tweets and
retweets
The Twitter pages were scraped as far back as possible, collecting posts from 2
May 2009 to 14 February 2019. While accounts were scraped without a time limit, it
is important to point out that the greater part of activity was in the last two years, as
the table clearly shows (Figure 3).
Figure 3. Table of activity of the whole sample over time (from 2010 to 2019)
PATON ET AL.
20
Phase 2 was a phase dedicated to data retrieval, anonymisation, the creation of a
database and the preparation of the collected material for analysis. In addition to
anonymisation through automatised processes, all personally identifiable information
was removed. Typically, the examples used are translated from the native language
into English and rewritten in such a way that the Twitter users could not be identified.
The images presented are not taken from the actual Twitter accounts of those in our
sample but instead are representative of the same genre and symbolic content as that
expressed in their Twitter accounts.
Phase 3 led to the analysis of data sets and observations retrieved during phase 1
and 2. To exploit ethnographic observations and datasets, a mixed method approach
was adopted. The approaches employed to analyse the collected material vary and
will be systematically specified in each subsection throughout the paper.
2. From Bold Positions to the Lack of Controversial Expression
When we started conducting ethnographic research on Twitter, we immediately
identified prevalent differences between the far right and the Islamist extremist
samples regarding levels of engagement and presence of extremism. To present this
phenomenon in greater detail, we appraise levels of activity and homogeneity within
samples by combining qualitative interpretations of ethnographic observations to
quantitative interpretations of the timelines of publication and patterns of behaviour.
Then, to identify whether activity is scattered or follows a similar pattern, we rely on
a quantitative approach of the composition of the sample. Finally, to examine the
public expressions of opinions and ponder levels of engagement in extremism, we
look at semiotic material of personal accounts (i.e. images, text, memes, banners, gifs)
that allow to build collective identity.
2.1. Contrast in Levels of Activity
Our field observations of Twitter allow to assert that far right Twitter activity
increased over the period of study, while Islamist extremist Twitter activity was faint.
It was easy for researchers to find far right accounts, indifferently for all countries.
These accounts were active and presented common characteristics pertaining to a set
of signifiers, topics of discussion, people they were interacting with. Yet, there was a
lack of activity amongst the Islamist accounts and they could not be grouped in a
similar manner.
Numbers pertaining to online activity support this general idea. In the figures
below, we can see that the far right is more active on Twitter with nearly 6,000
messages within a month (on the right, Figure 4), while the Islamist extremist activity
never surpasses 600 messages in a month (on the left, Figure 5). Speaking of activity
in this context refers to tweets and retweets. Furthermore, the far right activity is
higher in recent months but non-existent before; activity even spikes in the last months
ESSACHESS vol. 15, no. 2(30) / 2022
21
of the observation. This general upsurge is due to the fact these are very recent
accounts, created between 2017 and 2018, and ranks are growing (Figure 4); whereas
in the case of Islamist extremism, activity had greatly diminished at the time of the
study, leading us to think accounts are rather old and indicative of past tendencies
(Figure 5). Our findings fall in line with the literature regarding departure from the
platform (Conway et al., 2017).
Figure 4 & 5. Twitter timeline of activity for Islamist extremism (on the left) and
the far right (on the right)
Another element allows us to defend the idea that the far right is growing in ranks,
whereas Islamist extremism activity is a shadow of the past. In the table below, we
can see the tweet activity for both ideologies side-by-side; the first half of the graph
contains box-plots addressing tweet levels of activity for Islamist extremism; on the
second half of the graph are box-plots illustrating tweet levels for the far right (Figure
6). Box plots represent the distribution and the dispersion of a full dataset. The more
the rectangle is bottom-skewed, the smaller the activity is; reversely, the more the
rectangle is top-skewed, the more activity there is. Dots that appear outside of the
surface of the rectangle are indicative of activity that is not inline with the rest of the
sample.
Figure 6. Composition of the samples (distribution and dispersion)
In this figure, we see that in the case of Islamist extremist accounts - and this is true
for all countries (with the possible exception of France) - plotboxes are low-skewed,
PATON ET AL.
22
meaning most individuals have barely published on Twitter, with the exception of
some random accounts. Significantly, there is no common behaviour, activity is
scattered and lacks uniformity. Contributors appear poorly related to one another.
Comparatively, the far right activity seems to be homogeneous, pertaining to common
behaviour amongst the users (with the possible exception of accounts within the
Norwegian sample that appear to be outside of the scope of a dominant pattern of
behaviour). Connectivity seems high.
2.2. Identities Portrayed: Like-minded or Scattered
Let us now further consider engagement in extremism by means of how identities
are portrayed by qualitatively appreciating semiotic material of personal accounts.
Throughout Europe, Twitter users’ biographies, display of names, and profile
pictures on the far right - despite wide variation across individual accounts - are
commonly a primary space for endogenous labelling. Labels offer evidence of
generalised acceptance of hate speech and a high level of tolerance for radical views
and opinions that lays the groundwork for tolerance for extremism. In the UK for
example, pictures declare ideological allegiance through the use of prominent neo-
Nazi and white-supremacist symbolism, such as the sun cross and the British Union
of Fascists flag.
Throughout Europe, the similarly large presence of labels given to those outside
of the established in-group aligns with academic theories on the formation of identities
and the importance of binaries in establishing group boundaries. External labels and
terms such as ‘Muslim’, ‘barbarian’, ‘Jew’, ‘rapist’, ‘attacker’ suggest that the
perceived threat from the out-group is used to define the identity and position of the
in-group. In general, the sample refers to binaries such as ‘Christian’ and ‘Muslim’,
‘peace’ and ‘invasion’, ‘religion’ and ‘cult’. We can see this same dynamic of ‘us vs
them’ which essentialises relative differences between groups, naturalising qualitative
dichotomies from ‘sacred’ to ‘barbaric’, ‘heritage’ to ‘genocide’, and designating
everything on the outside as evil and threatening and everything inside as good,
virtuous and desirable. In short, one observes a link-minded posture throughout
accounts displaying strong signs of extremism underpinned by multiple forms of far
right expressions. There are preliminary signs of violent extremism.
Comparatively, the performed identities of the Islamist extremist sample are
diverse in composition. There is considerable variation in the expression of religious
opinions as much throughout Europe as within the countries of the same sample. In
certain countries, notably Germany, Norway and the UK, we found random accounts
displaying signs of support for jihadist-Salafism. The vast majority of the accounts
however, all countries alike, predominantly supported Salafi purism, with the notable
exception of the UK.
Purist Salafism is exhibited as much in Twitter users’ biographies, display of
names, and profile pictures as in messages. Quietist Salafism is not indicative of
ESSACHESS vol. 15, no. 2(30) / 2022
23
engagement in processes of radicalisation and defence of violent extremism
(Wiktorowicz, 2006), as purists focus on non-violent methods of propagation,
purification and education. Only closer consideration of messages could allow us to
demonstrate engagement in violent extremism.
2.3. The Forms of Expression: From Ostentatious to Lack of Expression of
Opinions
One criterion for selecting the sample was the direct or indirect call for violence.
In the case of the far right, it is easy to find numerous accounts that fit the criteria,
either explicitly or implicitly shared violent content targeting minorities or
immigrants. We found that violent intent is often expressed and hate speech is an
integral part of digital conversations. For example, in France, it was easy to spot the
following type of imagery and texts that openly stigmatise immigrants and/or
expressed radical opinions in terms of how to treat immigrants and Muslims in
particular (Images 1, 2).
Image 1.
The user states: “May I present to you the
future classmates of your children.”
Underneath this claim, there is a video, excerpt
from an ISIS propaganda video, with a toddler
killing a man on his knees at bullet point.
Image 2.
“We come in peace”
Likewise, in Greece, tweets offer similar content (Images 3-5).
We even found explicit calls to kill in Norway, as in the following tweet: “Cannot be
integrated. Kill all of those who do not go home. Bomb women and children
#visomstøtterlisthaug#frp.”
As for the Islamist sample, we did not find enough accounts that openly supported
or encouraged violence, to allow us to keep this as the main criterion of selection
PATON ET AL.
24
(again with the exception of the UK). In very rare cases, there is direct support for
martyrdom, for instance, reflected in the image below (Image 6) found in the Dutch
sample and the associated quote by Azzam, one of the most influential jihadi
ideologues, sometimes wrongfully rumoured to be the founder of the predecessor of
Al-Qaeda.
Image 6. Direct support for martyrdom
ESSACHESS vol. 15, no. 2(30) / 2022
25
Some of the tweeters also show very explicit videos or photos of the mistreatment of
Muslims in the world. Below is a screenshot of a video (Image 7) showing how the
Rohingya are burnt alive, and even dragged into the fire, taken from the Dutch sample.
Image 7. Screenshot of a video of a Rohingya being burnt alive
However, the contrast between both ideological strands in expressing violent intent
or violent material is striking; participation on the far right showcases violent content,
some of which could be assimilated to calls for violence, whereas Islamist extremists’
participation offers very little to no explicit reference to, nor expresses support for,
Islamist extremism.
3. Repertoires of Action: From Activism to Digital Marketing
Digital tools, by providing means to engage in political action for change, allow
private citizens to take their grievances, concerns, beliefs, and ideals into their own
hands and facilitate collective action (Norris, 2002). Here, in a second section of
results, we define the manners in which each sample participates online by means of
a repertoire of action and identify the main features distinguishing these forms of
collective action. We do so by focusing on how users from the sample engage online
and what justifications they provide to explain their contribution.
3.1. Far Right: Engaging in Media Activism to Support a Cause
On the far right, contributors use Twitter to express their political opinions and
promote their ideological views in the public sphere. Users engage in a classical form
of media activism, consolidated around five main features:
i) Anti-Islam/anti-immigration
Twitter is used for political ends to engage in a ‘cause’, that of combatting Islam
and immigration in Europe. Anti-immigration and anti-Islam discourses are abundant
and at the core of participation. In Germany, people discuss how Europe is committing
PATON ET AL.
26
suicide through its immigration and refugee policies. In Belgium, users post explicit
images. In the first example below, we can see a man saying something in Arabic and
a woman asking “what” in English. The response to her question appears to be
physical brutality and an imposed head scarf (Image 8). In the next image, the captions
reads “kids’ daycare” with the picture showing a ticking timebomb in the hands of a
Muslim man (Image 9).
In an example from the Dutch sample, the prospect of Sharia Law - as the future of
The Netherlands - is portrayed as a threat in an image meant to alarm followers (Image
10). The text associated with this image reads: “These are only the “incidents” that
make the news. The reality is much worse. Not everybody reports to the police after
being beaten. As a victim, you are perceived as the perpetrator.”
Image 10. A picture used to support the idea Islam is a threat
ii) Identity under threat
There is an excessive focus on depicting collective identity under threat, with
violations and injustices described as structural rather than incidental. A common
theme concerns the sentiment that a constructed national culture is threatened by
immigration in general and Islam in particular. This is expressed in the following
ESSACHESS vol. 15, no. 2(30) / 2022
27
picture with the header ‘changing Netherlands’ in Dutch (Image 11). We see how the
population has changed to wearing headscarves in 2017 and to being mostly
immigrants.
Image 11. Picture illustrating ‘changing Netherlands’
Perceived threats pertaining to immigration, ‘Islamisation’, and the gradual
devaluation and disappearance of national culture and identity are reflected in an
obsession with crimes committed by immigrants and Jihadist terrorist attacks. In the
UK, references to rape, grooming and Rotherham (referring to the Rotherham child
sexual exploitation case) are numerous. Similarly, terms that refer to terrorist attacks
in London, Paris, and Manchester, which are synonymised with their ‘Muslim’
attackers, are used as examples of the dangers of immigration and the incompatibility
between Islam and ‘Western’ values.
iii) National history, culture or religion are considered as the basis for a new
societal order
The flip side of these threats is the emergence of a new horizon composed of
national glory, cultural unity and religious uniformity. Media engagement is meant to
restore past greatness.
Part of the far right sample claims to defend a socially constructed ‘heritage’, one
of Christianity, white-skinned people who share an identical historical background.
This translates into pushing forward (ultra)nationalism and patriotism. For many
PATON ET AL.
28
countries, this means using flags to present themselves. In Greece, accounts showcase
the flag of the Byzantine Empire with the double-headed eagle as well as one with a
Christian cross, offering symbols of past glorious empires and Christianity. In the
Netherlands, a Dutch variation of the national flag is shown containing the acronym
NL4NL, which refers to “the Netherlands for the Dutch.” In the UK, Norway, Greece,
France and Germany, Celtic crosses are employed (Image 12), given the narrative
context, as white power/pro-Aryan movement symbols (Miller Idriss, 2017) and
support for Stormfront (an online community of white nationalists) (Image 13).
Also, we note a number of references to the relation between people’s skin colour and
belonging to the nation. For example, in the Norwegian sample, there are historical
images of national costumes with Nazi symbols, Vikings, Norwegian nature and the
Nordic Resistance Movement’s flag (Image 14).
Image 14. Soldiers in brown military uniforms
iv) Decline of institutions
The fourth feature is based on a ‘re-information’ strategy (Jammet & Guidi, 2017).
To properly understand the implications behind this communicative strategy, one
must consider a broader common conception within the far right sample: the State,
ESSACHESS vol. 15, no. 2(30) / 2022
29
the educational system and the media contribute to, or fail to, address the threats
mentioned hereinabove. Given the failure of these traditional institutions, users take
it upon themselves to inform others, with ‘adequate’ news sources to counter ‘fake’
news from the corrupted mainstream media.
Figure 7. The most shared domains in tweets
In France, as we can see in the tab summarising the most shared domains in tweets,
the most common news sources in tweets are those of well-known far right online
news sources (Figure 7).
v) Level of radicalism is extraordinary
The political nature of this format of participation is different from ‘ordinary’
political stances insofar as the level of radicalism is high; this repertoire openly
stigmatises minorities and can be a vehicle for the expression of violent content as we
have already established. It is not a form of clicktivism or slacktivism as sometimes
described in literature, but instead, a strong claim to intervene in everyday politics and
act offline, at the very least, during election times.
3.2. Islamist Extremism: Storefront to Promote a Brand of Faith
The Islamist extremist accounts are employed in an outbound manner to broadcast
religious ideological material. They focus on preaching, religious lifestyle tips,
scriptures and sermons pertaining to a Muslim way of life. This does not fully
PATON ET AL.
30
eliminate the activist dimension of this repertoire of action, nor does it negate its
extremist nature. Simply, this repertoire serves primarily to publicise religious beliefs.
We must mention the fact that accounts may fit only one of the characteristics
outlined below. With that being said, there are five main features:
i) Quietist Salafism
Posts are predominantly representative of quietist Salafism as we have established
hereinabove.
ii) Preaching, delivering faith-oriented lifestyle tips and broadcasting religious
scriptures
This repertoire is centred around preaching and sharing religious lifestyle tips,
scriptures and sermons pertaining to a Muslim way of life. Typically, we can spot
messages in the image header of personal pages. For example, one header from the
UK reads: ‘The Salif should strive to walk in the footsteps of the salaf as Salih’.
Within conversations, much talk is about how to lead a good life by following the
Prophet. Terms refer to phrases and concepts familiar to Muslims, including terms
such as ‘Koran’, ‘Qur’an’, ‘Allah’, ‘recite’, ‘pray’, ‘angel’, ‘mercy’, ‘deen’ (religion),
‘dua’ (supplication to Allah), ‘the prophet Muhammad’, ‘prophetic hadith’ (the
prophetic sayings and teachings of the Prophet) and ‘God’.
Accounts can promote their own interpretation of faith and create enough of an
online audience to become an influencer. Typically, such influencers within the
Islamist Belgian sample, for example, offer day-to-day lifestyle tips, publish self-
reflective or inspirational blog posts, summed up in a single quote, and present
organisational advice on how to stay committed or reach personal development goals
in relation to faith (Images 15-17). These accounts can represent lone individuals (aka
influencers) or brands related to specific Mosques and schools of faith.
ESSACHESS vol. 15, no. 2(30) / 2022
31
Image 17. Excerpt from the Koran
iii) Twitter is exploited as a storefront that re-routes users to other online platforms
The third feature of this repertoire is related to the fact that Twitter is exploited as
a storefront that re-routes users to other online platforms, such as chat rooms where
conversations can take place, storage spaces where files can be retrieved,
crowdfunding sites where money can be collected, or more commonly, blogs with
information on branches of faith in relation to local Mosques.
We note examples of schools using Twitter to promote religious scriptures and
sermons with the clear intention of showcasing what they do locally, thus using
Twitter as a platform to generate an audience and bring attention to their activities, as
is the case in France. In the first example below, a local Mosque promotes a previous
speaking event (Image 18); in the next example, the tweet shares information about a
series of conferences (Image 19). The last message pertains to religious teachings of
a spiritual leader (Image 20).
PATON ET AL.
32
In this sense, microblogging to promote ideologically-oriented faith material is a
means to build an audience in order to pursue notoriety as a digital influencer. They
may also sell associated products (e.g. books, speaking events, tours). This repertoire
of action can be related to ‘educational entrepreneurs’, i.e. a new ‘offer’ - one
consisting of tweets, YouTube videos, blog posts - based on ‘traditional’ material,
typically classical religious material (Van Praet, 2018).
iv) Users take public stands to denounce terrorism and acts of violence
If we did spot a handful of accounts promoting jihadi-type material throughout
Europe, the general tendency is that users do not condone violence, quite the contrary:
they take public stands to denounce terrorism and acts of violence. Contributors argue
against terrorism or radical forms of political Islam. They denounce infamous political
figures or organisations supporting radical Islamism by providing links to anti-
extremism leaflets or arguing against the reasoning behind terrorism, explaining how
such activities are misguided and go against Islam. This phenomenon is noticeable in
a tweet claiming “Khalid Yasin” - a well-known controversial British Salafist
preacher - is misguided. As stated in the tweet from the UK, his preachings go against
an “authentic” version of the Islamic faith:
“Several testimonies in this article serve to demonstrate the fallacious nature of
Khalid Yasin, along with the fact his preachings are dangerous.”
Another tweet openly warns against Al-Qaeda and ISIS organisations by linking
to an educational leaflet explaining why these groups must be avoided:
“Get your new leaflet now: Warnings against terrorist activities of ISIS & Al-
Qaeda”
In a final example, we can see in the “about” section of a user’s profile that the
contributor adopts the opposite position to one of violence, by portraying Islam as a
ESSACHESS vol. 15, no. 2(30) / 2022
33
religion of peace:
“Allah calls for the Home of Peace and guides whom he wishes to the righteous
path.”
v) This repertoire is one of religious fundamentalism dabbling in political
extremism
The last feature situates this repertoire of action within the range of political
extremism and slightly contradicts the fourth feature. Signs of religious
fundamentalism exclude the possibility of considering this repertoire as apolitical and
exempt from any sort of extremism or political consideration. In this respect, we can
qualify this repertoire as one of religious fundamentalism dabbling in political
extremism.
4. Resources for Action
To further our understanding of the role of social media in the engagement in
radical causes, we now discuss the resources for action. We see that there is little to
no matter for discussion of resources for action for Islamist extremism, apart from
out-of-the ordinary ISIS type activity. On the far right, however, we demonstrate that
group identity paves the way to radicalisation, knowing that group identity is
supported by world leaders and related to a global conversation.
4.1. Pro-ISIS vs. anti-ISIS Online Power Struggle: A Doorway into ISIS
Propaganda Material?
Direct observation over several months led to witnessing bot-related operations
opposing pro-ISIS and anti-ISIS accounts. These accounts are of a different nature
than the ones we set out to study in this paper as they pertain to highly-technical
operations related to a digital war between audiences and counter-audiences. Lively
and very active online groups – part of the greater ecosystem of what is happening on
Twitter and more generally illustrative of contemporary trends on the web – fuelled
by a power struggle opposing pro-ISIS accounts to a global army of anti-ISIS users,
takes place on a day-to-day basis. Pop-up accounts attempt to create holes in
censorship, knowing that accounts will be deleted within hours, at best days (Image
21). These accounts are ostensibly supportive of ISIS. They contain links to other
websites on which one can find official propaganda-type material (access to official
ISIS journals, like issues of Rumiyah). On the other hand, there are private citizens
(a.k.a. ‘hunters’), without a united political front, who actively moderate social media
websites (Image 22). This power struggle is tied to the use of online bots and
automatic tracking systems. ‘Hunters’ act as gatekeepers and safeguards of the web,
by tracking and flagging ISIS-type content, to get deviant accounts taken down by the
Commercial Content Moderation workers, i.e. workers responsible for judging what
is acceptable and what should be removed.
PATON ET AL.
34
Even though the ISIS-type accounts exemplify how radicalisation is supported by
social-technical means, they are out of scope of the samples. It is impossible to know
who is behind these publications, whether we were dealing with private citizens, fan
boys or girls, leaders of opinion or actual members of ISIS. Yet, it is important to keep
in mind that such trends exist on Twitter. Of equal importance, the links shared in the
pop-accounts are means for radicalisation.
However, on the Islamist extremist end of the spectrum, means for radicalisation
stop there. Our focus will therefore be on what takes place on the far right from this
point.
4.2. How the Far Right uses of Twitter paves the Way for Radicalisation
Here, we will see that radicalisation is favoured by media participation in three
ways: 1) influencers 2) topics of conversations 3) networks of conversation.
i) Influencers: an endorsement for far-right ideologies
Influencers play a crucial role in persuading others (Berry, Keller 2003). Studying
influence patterns is therefore meant to pinpoint which individuals shape
conversations and impact public debates in Europe through their popularity in digital
threads. What we apprehend as influencers is the quantitative consideration of
influentiality, established by measuring a ratio between followers/follow and
responses to original tweets (Cha et al. 2010).
Research shows a small number of highly visible political leaders (Figure 8).
Trump, but also Bolsonaro, Salvini and Farage, have had a considerable impact on the
debates. They are not ranked according to their level of approval; their influence
derives from how much noise they generate in the public sphere, or how much
attention they garner in public debates. The most influential Twitter user from our
data in all the countries was, by far, former President Donald Trump. The impact of
Trump is likely due to a combination of factors including his expansive follower base,
regular activity, controversial tweets and large presence of bots associated with his
account. Without in-depth analysis of tweets that resonated and the nature of replies,
ESSACHESS vol. 15, no. 2(30) / 2022
35
it is impossible to say confidently how these factors contributed and exactly how
Trump has influenced the activity of our sample.
In Greece, the top fifteen influencers include: the British far right digital influencer
Paul Joseph Watson (cf. Prison Planet); the US extreme-right wing, celebrity-actor
James Woods; Voice of Europe, a now banned alternative news-outlet, broadcasting
far-right ideologies; the far-right British vlogger, Peter Sweden; the right-wing
extremist Italian party, Lega Nord; the (then) Minister of the Interior, Salvini; the
official account of the US White House; Trump’s wife, Melania; the Brazilian
President, Bolsonaro; Netanyahu, the prime minister of Israel; and Adonis Georgiadis,
an openly anti-leftist and anti-communist Greek politician, well-known for his anti-
Semitic views. Detailing the Greek case allows us to exemplify the fact that top
influencers are transnational figures, well-known for their far right positions; concerns
are not nationally bound but expressed within a far-right digital circle. The uniformity
of such figures presupposes the existence of a European far right ‘milieu’. The term
milieu is understood as a space where radical/extreme messages are encountered
online via people who participate in radical/extreme activities.
Figure 8. List of top influencers per country
PATON ET AL.
36
What is also important to emphasise is that those who could be spotted as
potentially contributing to online radicalisation can be well-known world leaders, not
simply obscure organisations, underground digital influences or online gurus. When
the needle of the boundaries of what is acceptable to say in the public sphere is moved
by legitimate figures, the work of moderators operating to restrict extremism is put to
the test and democracy with it.
ii) Narratives federating people and creating a sense of togetherness
If we turn to conversations, we can spot narratives uniting people. To demonstrate
this we examined the tweets and retweets by means of an automatic statistical lexical
analysis and then compared the main themes for each country. The method used is
based on a descending hierarchical classification; this operation was carried out
through a factorial analysis of correspondences. To complete this analysis, we used
the Reinert method (Reinert, 1983, 1990) implemented in the free software
IRaMuTeQ (Ratinaud, 2014). Below is an example of this type of analysis, taken from
the Belgian sample (Figure 9). The results - presented in a dendrogramme - offer a
segmentation of the online conversations into clusters, underlining the main lexical
themes of the corpus. Information on the size of each cluster and the overrepresented
lexicon in each cluster also appears.
Figure 9. Dendrogramme for the far right Belgian sample
In comparing the individual samples of the seven countries, evidence shows that
conversations are global. The main themes go back to what was discussed in the
section on repertoires of action. They deal with anti-immigration and anti-Islam;
strong leaders and far right unity in the West; anti-establishment and anti-Europe
discourses. There is excessive focus on depicting collective identity under threat.
ESSACHESS vol. 15, no. 2(30) / 2022
37
Perceived threats pertain to immigration, ‘Islamisation’, and the gradual devaluation
and disappearance of national culture and identity. The State, the educational system
and the media are perceived as a single entity (a.k.a. ‘the establishment’), failing to
address the threats. Representatives of the perceived current ruling class are mocked
and derided via caricature and hate speech, most notably political leaders, judges and
media figures. These ‘anti’ features are similar to the characteristics of the alt-right
movement described by VOX-Pol in a report describing the Twitter activities of this
movement (Berger, 2018). Failures are seen as due to dilution of national political
authority, EU membership, and political correctness in media and education that
blindly promote equality regardless of differences. Together, this appears as a
concerted effort by left-wing politicians, mainstream media, and the educational
system to cover the true extent of the threat posed by immigration and Islam. This
global conversation revolves around national history, culture or religion; such matters
are considered as the basis for a new societal order. Many references to historical
national heroes and images of a glorious national or European past are made,
underlining the perceived (sometimes racial) purity that existed and is currently
threatened. All these topics dip into double standards, victimhood and unfair treatment
sentiments, and highlight identity issues.
The fact conversations were global on the far right leads us to support the idea that
the consolidation of nationalist and conservative positions are nowadays being forged
on an international scale. There is probably some sense of unity that goes beyond
national contexts, supporting evidence of an online digital far-right milieu at the
European level.
iii) A far-right milieu related to a global conversation
Network analyses for the seven countries completes our understanding of how
Twitter offers resources to engage in processes of radicalisation in Europe. Examining
what is taking place from a network perspective helps capture how samples are
structured, how people are linked to one another and how conversations flow. To
conduct this analysis, we isolated user information pertaining to followers and
followees, as well as streams of tweets and retweets, then used Gephi to design
connectivity of users and conversations.
Research showed close-knit networks of contributors frequently sharing
information, liking or retweeting each other’s messages. Each country - corresponding
to a given colour (Figure 10) - is well-established as a national community, being
gathered with little to no peripheral accounts/dots, implying homogeneity within the
network and strong connectivity. There are two notable exceptions: Belgium, that is
distributed between two linguistic areas - one French/France, the other Dutch/the
Netherlands - as well as Norway, scattered and far from the network of other
countries, implying that the sample is not well-established. Moreover, all country-
level clusters are well-connected to one another, with several connections from one
cluster to the next, with the UK at the centre (likely due to the linguistic dimension of
PATON ET AL.
38
conversations). High levels of interconnectivity and circulation of content as well as
conversations further establish the existence of a far-right European milieu. This
milieu is not restricted to national considerations, although some countries have more
leverage than others within this far right European milieu. Communicative networks
are simultaneously homogeneous at the national level but also well-connected at the
European level.
Figures 10 & 11. Interconnectivity of the Twitter sample by country
Another approach to our sample confirms homophily and the existence of global
conversations (Figure 11). Below we can see the same samples but organised in terms
of communities of interpretation (Figure 12). Communities of interpretations are
networks of users who share interests in the same topic. We can see very little
dispersion and extremely strong connectivity within samples around a few main
themes. The level of homophily appears to be quite extraordinary, especially given
the size of the sample, as it is organised only around five threads, further confirming
the existence of a global conversation related to an European far-right digital milieu.
The structuration of conversation around interpretative communities also emphasises
the fact that this milieu on Twitter is a European closed-circle, with little outside
connection, and a strong bubble-filter effect.
ESSACHESS vol. 15, no. 2(30) / 2022
39
Figure 12. Network of right-wing extremists on Twitter: our sample for all 7
countries and their followers that they follow
These elements allow us to demonstrate the existence of a far-right European milieu
thriving on Twitter that supports online radicalisation for those who wish to engage
in such ideologies through their media participation.
5. Discussion: Participation vs. Radicalisation
Throughout the paper we have seen the disparities between the two ideological
beliefs. Their singularities notwithstanding, we saw that one phenomenon is a shadow
of the past, whereas the other is related to an active base of members, embedded in
extremism, related to a European milieu. The reasons behind this disparity are difficult
to single out since it is most likely a mix of several convergent phenomena.
We know that Islamist extremism still exists but that ISIS-type activity, for
example, takes place elsewhere. ISIS has encouraged its followers on social media
sites to connect with ISIS coordinators and recruiters on Telegram to discuss sensitive
matters such as travel to ISIS-held territory. ISIS also created public channels on
Telegram to broadcast pro-ISIS news updates and disseminate other propaganda
materials through its Amaq News Agency news outlet. In January 2015, an ISIS-
affiliated channel disseminated a guide of the ‘safest’ platforms to use (Helm et al.,
2020; Coker et al., 2015) (Figure 12).
PATON ET AL.
40
Figure 12. A guide of platforms disseminated by ISIS in 2015
Twitter and Facebook are noticeably absent - indicative of jihadist audiences’
apprehension towards using public platforms. More recently, sites like Telegram
announced that they are ‘clamping down’ on jihadist-affiliated accounts and pledging
to remove ISIS accounts from public channels (Helm et al., 2020). However, owing
to Telegram’s encrypted nature, private jihadist groups are still widespread in many
different languages on the platform.
Moderation is a strong factor shaping online communities and laying the
groundwork for far right milieus as much as the lack of a jihadist milieu on social
media. In recent years, websites have received immense pressure from governments
and associated agencies to moderate extremist and terrorist-type content. Early on, the
Obama Administration and the U.S. Congress formulated legislative proposals. The
European Union in the form of a ‘code of conduct’ incited action. The UK, France
and Germany attempted moderation, mainly threatening to fine big tech platforms
(Cope et al., 2017). Europol's EU Internet Referral Unit (IRU), in partnership with the
Netherlands and Belgium, launched a project
2
to improve the detection, analysis and
referral of online terrorist content. Likewise, a UK-French joint anti-terrorist
campaign was created to add pressure on companies like Twitter to improve the
monitoring of terrorist-related activities (Toor, 2017). In all these cases, top-down
moderation prioritarily aims at Islamist extremism content. Recently, far right
activities have started to be more heavily moderated, but by no standards to the same
degree (Bodo, 2017; Wahlstrom & Tornberg, 2019).
2
https://www.europol.europa.eu/newsroom/news/europol%E2%80%99s-eu-internet-referral-unit-
partners-belgium-france-and-netherlands-to-tackle-online-terrorist-content
ESSACHESS vol. 15, no. 2(30) / 2022
41
Beyond the reasons cited above, another element explaining disparity between the
two is the political scene in Western countries. The far right sample is subjected to the
blurred boundaries of the permissible, at a time when policing hate speech by right-
wing extremism has grown more and more difficult given the support it receives daily
with certain political figures in several Western countries, who openly express violent
extremist viewpoints (Berger, 2018).
Finally, we can postulate that academia contributes to this situation by using the
notions of political participation and radicalisation discriminately. In literature, the
expression ‘radicalisation’ is more often applied to characterise Islamist extremism,
whereas ‘political participation’ is more often employed to study the far right’s
activities (Pisoui & Reem 2016), even though each ideological strand can refer to a
same phenomenon of extremism. This disparity is even more striking given that
radicalisation refers to an extreme behaviour, while political participation tells us
nothing of the nature of the behaviour. By using the terms radicalisation and political
participation discriminately, academia creates legitimacy for one while stigmatising
the other. In this article, we have contributed to the labelling of behaviours, even more
so in considering both samples from the same viewpoint. Our findings show that the
far right is the by-product of internal and external labelling processes, whereas the
labelling of Islamist extremist content as radical is more a byproduct of researchers’
work and labelling from other institutions (States, Twitter, etc.). In the frame of our
own work, we often labelled accounts Islamist extremist even though we were dealing
with forms of Purist Salafism that arguably could be framed as conservative rather
than extremist, and therefore participating in the stigmatisation of one ideology over
the other.
Conclusion
In this paper, we discussed far right and Islamist extremism in Europe, based on media
participation on Twitter between 2018-2019 to understand the role of social media in
the process of engagement in radical causes. We focused on three main areas: online
activity, forms of collective action, and digital resources for engagement in a radical
cause. In a final section, we carried out a discussion on how the discriminate use of
notions of political participation and radicalisation for each ideology supports each
form of media participation.
In the first section, we saw that far right Twitter activity increased over the period
of the study. Accounts were active, presented common characteristics pertaining to a
set of signifiers and people with whom they were interacting. Publications displayed
acceptance of hate speech and a high level of tolerance for extremism. Comparatively,
Islamist extremist accounts’ activity was faint. If accounts predominately supported
Salafi purism, the performed identities of the sample were diverse in composition.
This sample appears to have been incapacitated in expressing controversial ideas.
PATON ET AL.
42
Moving forward, we discussed the main repertoires of action. We showed that the
far right uses Twitter to converse, employing the platform for its democratic
conversational properties, taking advantage of the fact that this digital space provides
a direct access to one of the main arenas of public debate that exists in modern times,
to defend a ‘cause’; whereas the Islamist extremist sample uses the platform as a store-
front to broadcast its ideological beliefs, seeking recognition and publicity, to promote
leaders and faith-based brands, and at times, tore-route users to other platforms where
sales can sometimes be performed.
Next, we examined resources for action available for online users. We saw that
there is little to no matter for discussion of resources for action for Islamist extremism,
apart from out-of-the ordinary ISIS type activity. On the far right, however, we
demonstrated that group identity paves the way to radicalisation, knowing that group
identity is fuelled by world leaders and related to a global conversation. The fact that
conversations were global on the far right leads us to support the idea that the
consolidation of nationalist and conservative positions is nowadays being forged on
an international scale. There is a sense of unity that goes beyond national contexts,
supporting evidence of an online digital far-right milieu at the European level,
knowing some countries have more leverage than others within this far right European
milieu. Network analysis, in particular, offers evidence of a close-knit European
network of contributors, that has little outside connection, and is subjugated to a
strong bubble-filter effect. High levels of interconnectivity, circulation of content as
well as conversations within a far-right European milieu show one of the main
resources for radicalisation online.
In a final section of this paper, we consider other reasons behind the disparities of
media participation for each ideological belief. We show how political participation
is constructed at the crossroads of conflicting dynamics on the web, beyond their
offline existence. Moderation is a strong factor shaping online communities, laying
the groundwork for a far right milieu as much as the lack of a jihadist milieu on social
media. Beyond this, the political scene in Western countries plays a major role. The
far right sample is subjected to the blurred boundaries of the permissible, at a time
when policing hate speech by the far right has become more and more difficult given
the support it receives daily in most Western countries. Comparatively, controversial
forms of expression related to Islamist extremism have been banned on big tech
platforms. The contrast between the two forms of political practices is also related to
the ways in which academia uses the notions of political participation and
radicalisation discriminately. The important element to keep in mind here is that
labelling has a performative effect; the treatment of these two ideological beliefs is
asymmetric, whether this treatment is by academia, big tech or governments. Yet, by
singling out the one, without giving the second the same level of attention, the most
dangerous form of extremism of the two may possibly be left unattended. The
asymmetric treatment of these ideological beliefs raises indeed the question of how
social media is nowadays playing the game of extremism, helping legitimise political
leaders outside of traditional national spheres, paving the way to a European far right
ESSACHESS vol. 15, no. 2(30) / 2022
43
milieu in a new territory foreign to past democratic rules and settings tied to
gatekeepers and legitimate institutions.
Acknowledgement & Funding
This project has received funding from the European Union’s Horizon 2020 research
and innovation programme under grant agreement No 725349. Full country reports of
the work package dedicated to studying online radicalisation can be found online at
the following address: https://www.dare-h2020.org/self-radicalisation-and-digital-
sociability.html
References
AIVD (2018). Rechts Extremisme in Nederland: Een fenomeen in beweging [Right-Wing
Extremism in The Netherlands: A phenomenon in motion], The Hague: Ministry of Interior
and Kingdom Relationships.
Berger, J. & Morgan, J. (2015). The ISIS Twitter Census: Defining and describing the
population, Washington DC: The Brookings Project on U.S. Relations with the Islamic
World.
Berger, M.J. (2018). ‘The alt-right twitter census: Defining and describing the audience for alt-
right content on Twitter’, VOX-Pol Network of Excellence. Available online at;
https://www.voxpol.eu/download/vox-pol_publication/AltRightTwitterCensus.pdf
Berry, J., & Keller, E. (2003). The Influentials: One American in ten tells the other nine how to
vote, where to eat, and what to buy, New York: Free Press.
Cammaerts, B. (2012). ‘Protest Logics and the Mediation Opportunity Structure’, European
Journal of Communication, 27(2): 117-34.
Cha, M., Haddadi, H., Benevenuto, F. & Gummadi, P.K. (2010). Measuring user influence in
Twitter: The million follower fallacy, Proceedings of the Fourth International AAAI
Conference on Weblogs and Social Media, http://twitter.mpi-
sws.org/icwsm2010_fallacy.pdf (last accessed January 20, 2020).
Chadwick, A. (2007). ‘Digital network repertoires and organizational hybridity’, Political
Communication, 24(3), 283-301.
Cheong, P. & Halverson, J.R. (2010). ‘Youths in Violent Extremist Discourse: Mediated
Identifications and Interventions’, Studies in Conflict & Terrorism, 33: 1104–23.
Coker et al. (2015) https://www.wsj.com/articles/islamic-state-teaches-tech-savvy-
1447720824
Della Porta D. & Diani M. (2006). Social Movements: an introduction. Oxford: Blackwell
Publishing.
PATON ET AL.
44
Deuze, M. (2006). ‘Participation, Remediation, Bricolage: Considering Principal Components
of a Digital Culture’, The Information Society, 22(2): 63-75.
Figeac, J., Paton, N., Salord, T., Cabanac, G., Cuelho Bezerra, A., Prévost, H., Ratinaud, P.,
(2021). ‘Brazilian left-wing activists on Facebook: the role of cultural events in political
participation’, Brasiliana: Journal for Brazilian Studies.
https://tidsskrift.dk/bras/article/view/125719
Fiske, J. (1989). Reading the Popular. Boston: Unwin Hyman.
Frazer, N. (2005). Transnationalizing the Public Sphere, March 2005.
Hassan, G., Brouillette-Alerie, S., Seraphin, A., Frau-Meigs, D., Lavoie, L., Fetiu, A., Varela,
W., Borokhovski, E., Venkatesh, V., Rousseau, C. & Stijn, S. (2018). ‘Exposure to
extremist online content could lead to violent radicalization: a systematic review of
empirical evidence’, International Journal of Developmental Science, 12(1-2): 71-88,
https://doi.org/10.3233/DEV-190276
Helm, G., Baky, A., Danielle Soskin, D., Monsell, L., Paton N., (2020). “D5.1/2: United
Kingdom Country level reports on drivers of self-radicalisation and digital sociability
among Europeans”, Final Report H2020 Dare : Self-radicalisation” http://www.dare-
h2020.org/uploads/1/2/1/7/12176018/5.1_united_kingdom.pdf
Hoffman, B. (2009). ‘Radicalization and subversion: Al Qaeda and the 7 July 2005 bombings
and the 2006 airline bombing plot’, Studies in Conflict & Terrorism, 32(12): 1100-16.
Jammet, T. & Guidi, D. (2017). ‘Observer Les Observateurs: Du pluralisme médiatique au
populisme anti-islam, analyse d’un site de “réinformation” suisse et de ses connexions’,
Réseaux, 202-203(2): 241-271.
Jenkins, H. (2006). Convergence Culture. Where Old and New Media Collide. New York: New
York University Press.
Koehler, D. (2014). ‘The radical online: individual radicalization processes and the role of the
internet’, Journal for Deradicalization, 15(1): 116-34.
Livingstone S. (2013). The participation paradigm in audience research. Communication
Review. 16: 1-2, 21-30
Medina-Serrano, J.-C., Shahrazaye, M., Papakyriakopoulos, O. and Hegelich, S. (2018). ‘The
rise of the AfD: a social media analysis’, Proceedings of the International Conference on
Social Media and Society, Toronto, ON, Canada, 19–21 July 2019 (SMSociety ’19).
Meraz S. & Papacharissi Z. (2013). ‘Networked gatekeeping and networked framing on#
Egypt’, The international journal of press/politics, 18(2), 138-166.
Miller-Idriss, C. (2017). The Extreme Gone Mainstream Book Subtitle: Commercialization and
Far Right Youth Culture in German, Princeton University Press.
Neudert, L.-M., Kollanyi, B. & Howard, N.P. (2017). ‘Junk news and bots during the German
parliamentary election: what are German voters sharing over Twitter?’ Data Memo 2017.7.
Oxford: Project on Computational Propaganda, https://comprop.oii.ox.ac.uk/research/junk-
news-and-bots-during-the-german-parliamentary-election-what-are-german-voters-
sharing-over-Twitter/
ESSACHESS vol. 15, no. 2(30) / 2022
45
Norris, P. (2002). Democratic Phoenix: Reinventing Political Activism, New York: Cambridge
University Press.
O’Neill, S. & McGrory, D. (2006). The Suicide Factory. Abu Hamza and the Finsbury Park
Mosque, London: Harper Perennial.
Piazza, J. & Guler, A. (2019). ‘The Online Caliphate: Internet Usage and ISIS Support in the
Arab World’, Terrorism and Political Violence, DOI: 10.1080/09546553.2019.1606801
Picart, C. (2015). ‘Imagined Relations in the Self-Radicalization of Colleen LaRose (Jihad
Jane)’, Societies, 5: 354-83.
Postmes, T., Van Bezouw, T. & Kutlaca, M. (2014). ‘Van collectief ongenoegen tot
ordeverstoringen’ [from collective discontent to public disturbances] , The Hague: WODC.
Praet, Y. J. V. (2018). ‘L'émergence de l'islam 'traditionnel' au sein du cyber espace
francophone musulman’. Recherches sociologiques et anthropologiques, 49(1): 127-156.
Ratinaud, P. (2014). IRaMuTeQ: Interface de R pour les Analyses Multidimensionnelles de
Textes et de Questionnaires (Version 0.7 alpha 2). Available from http://www.iramuteq.org
Reinert, M. (1983). ‘Une méthode de classification descendante hiérarchique : application à
l’analyse lexicale par context’, Les cahiers de l’analyse des données, VIII (2): 187-198.
Reinert, M. (1990). ‘ALCESTE: Une méthodologie d'analyse des données textuelles et une
application: Aurélia de Gérard de Nerval’, Bulletin de Méthodologie Sociologique, 26: 24-
54.
Saleh, H. (2014). ‘Western female jihadis employ “soft power” of ISIS online’, Financial
Times, 18 October. Available at: http://www.ft.com/cms/s/0/2d8b020c-5792-11e4-8493-
00144feab7de.html#axzz3vFs63UPn
Silber M.D. & Bhatt A. (2007). ‘Radicalization in the West: The Homegrown Threat' NYPD:
https://seths.blog/wp-content/uploads/2007/09/NYPD_Report-
Radicalization_in_the_West.pdf
Stevens, T. & Neumann, P. (2009). Countering online radicalisation: A strategy for action,
King’s College London: International Centre for the Study of Radicalisation (ICSR).
Tandoc E.C., Wei Lim, Z. & Ling, R. (2017). ‘Defining “Fake News”’, Digital Journalism,
DOI: 10.1080/21670811.2017.1360143
Tilly, C. (1993). ‘Contentious repertoires in Great Britain, 1758–1834’, Social Science History,
17(2), 253-280.
Torok, R. (2013). ‘Developing an explanatory model for the process of online radicalisation
and terrorism’, Security Informatics, 2 (6): 2-10.
Tufekci, Z. (2017). Twitter and Tear Gas: The Power and Fragility of Networked Protest, Yale
University Press.
Tuffal, W. (2015). ‘Rotherham, Rochdale and the racialised threat of the “Muslim Grooming
Gang”’, International Journal for Crime, Justice and Social Democracy, 4 (3): 30-43.
PATON ET AL.
46
Van Laer J. & Van Aelst P. (2010). ‘Internet and Social Movements Action Repertoires:
Opportunities and Limitations’, Information, Communication & Society 13(8): 1146-71.
Williams, R. (1958). Culture and Society. London: Chatto & Windus.
Wiktorowicz, Q. (2006). ‘Anatomy of the Salafi Movement’, Studies in Conflict & Terrorism,
29(3): 207-239. https://doi.org/10.1080/10576100500497004
Venhaus, J.M. (2010). ‘Why Youth Join al-Qaeda’, United States Institute of Peace Special
Report, Washington DC: United States Institute of Peace.
