Facebook pages in Europe: a tool for the Jihad in Syria

Team Members

Researchers: Rebecca Cachia, Susan Clandillon, Inte Gloerich, Cristel Kolopaking, Indre Lauciute, Rose Rowson

Gephi Designers: Federica Bardelli, Gabriele Colombo, Carlo de Gaetano, Michele Mauri

Introduction

Since the start of the Arab Spring in 2011, Syria has become a major area of conflict in the Middle East. The current Syrian uprising is the result of armed conflict between forces loyal to President Bashar al-Assad’s Ba’athist government and those seeking to topple the regime. The insurgent group of Jihadists called the Islamic State of Iraq and the Levant (Syria), abbreviated as ISIS, has emerged in opposition to both the regime and other moderate rebel fighters, with the aim of enforcing Islamic rule in Syria (Reuter). In addition to Syrian fighters, ISIS has also been recruiting Jihadists from overseas, with as many as 1800 European citizens joining the Jihadist movement in Syria in 2013 (Lowles and Mulhall 5). This indicates that, as a result, pathways and networks have started to emerge across Europe that are leading European extremists to the battlefields in Syria. The largest radical Islamic network in Europe is run by Anjem Choudary’s group called Al-Muhajiroun, which is the biggest radical Islamic group in England. Among other countries, Al-Muhajiroun is currently active in Belgium, the Netherlands, France, and Germany. The group has sent around 200 to 300 people to Syria and it is considered to be the widest recruiting network in Europe (Lowles and Mulhall 5).

In recent years, extremist groups across Europe have expanded their online presence by taking to social media in order to reach a wider audience than was previously possible. Through the Internet, Jihadist groups are now creating new forms of recruitment within these media that are more efficient and faster in spreading their ideology than previous methods were (Chan 2). In fact, as Finch (32) claims, “[s]ocial media usage by subversive extremist groups is no longer the exception to the rule, it is the standard”. In this regard, social media are especially efficient as a communication tool and in building online communities. As a result, online social networks allow violent extremists to connect quickly, to remain anonymous and to avoid legal sanctions.

This report examines the use of social media by sympathisers of the Jihadist movement in Syria through researching the social media platform Facebook. The aim of the study is twofold: firstly, to establish an understanding of the kind of content being shared through Jihadist pages on Facebook, focusing on English-, French-, German- and Dutch-language pages; and, secondly, to establish connections through shared content and users across the Facebook pages analysed as a portrayal of the European pathways of extremism on Facebook. Therefore, this paper investigates the following research questions.

Research Questions

• How have Facebook pages in Europe been used as a tool for the Jihadist cause in Syria?
• To what extent can Jihadist pages on Facebook be considered a recruitment tool for ISIS and Jihadist networks in Europe?

Literature Review

In his article “Studying Facebook via Data Extraction: The Netvizz Application”, Rieder (346) states that the intimate linkages between data and the technical and functional structure of the medium studied provides rich contextualisation missing from traditional methods of research such as interviews or experiments. Furthermore, the development of tools for the extraction of data in recent years, particularly around big data, has strongly supported the use of the Internet and Internet platforms as sources of information about society. As a case in point, through the technical structure of its interface, the social networking platform Facebook enables, directs and delineates the opportunities for the movement of users across the platform (Rieder 347), in addition to their interaction and engagement with each other and with the content they produce and share. As a result, researchers across academic disciplines have increasingly acknowledged the usefulness of Facebook as a tool for the observation of research subjects in a naturalistic setting (Wilson et al.; Rieder).

In their article “A Review of Facebook Research in the Social Sciences”, Wilson et al. lay out five areas of social research that have so far been facilitated through Facebook. Facebook is used, firstly, to establish a descriptive understanding of the platform’s users; secondly, to establish why people use Facebook; thirdly, to analyse how people present themselves on the platform; fourthly, to understand how Facebook affects the relationships between and among individuals and groups of individuals; and, lastly, to examine why people continue to disclose personal information in spite of potential risks (Wilson et al. 206). While data can be collected in each of these areas by examining output deliberately provided by the users, such as their profile information, the interesting aspect of using Facebook as a tool for social research is the amount of information that the platform collects as part of its very nature of operation. In other words, in addition to the content shared by users (such as images, statuses and links), the ways in which they click, like and navigate through the platform constitute a pool of data representative of behavioural elements of cultural and social significance (Rieder 347). The development of the Netvizz application, originally developed by Rieder in 2009, was motivated by the attraction of extracting data from a user’s personal network of friends, groups and pages for research purposes by working within the opportunities afforded by the Facebook API. In other words, as evidenced by Rieder’s analysis of the “Islam is Dangerous” Facebook group (351-52), by means of Netvizz researchers can collect and compile large amounts of data on, for example, friendship connections between members of a group and establishing potential user “locales” through their interface language. This report builds on Rieder’s claims and previous research by using the Netvizz application to access and repurpose data collected by Facebook to examine the content and connections between Facebook pages linked to the Jihadist movement in Syria at the time of writing.

In their article “Pathways to Violent Extremism in the Digital Era”, Edwards and Gribbon identify a gap in the research relating to the study of extremism online. They suggest that most academic literature focuses on the terrorists themselves and the means by which they attempt to attract support online, rather than the ways in which individuals become radicalized by this content. In this paper they conduct a series of 15 interviews with subjects who have been convicted in relation to terrorism offences, including the possession and intended use of homemade explosives and homemade suicide vests. In doing this, they attempt to gather “primary data” (41) about the individuals and their use of digital tools for extremist purposes. In this way we can draw parallels between their practice of gathering primary data relating to online interaction with extremist content through interviews and our own research using digital methods, which aims to make use of data scraped from Facebook using the Netvizz tool to investigate user interaction with digital content - in this case European Jihadist Facebook pages. Edwards and Gribbon signal the important role that the internet plays with regard to extremist content, ‘Terrorists and extremists use [the internet] for a myriad of purposes: from disseminating propaganda and information to fundraising and operational planning” (40). It is with these functions in mind that we examine the recruitment strategies used by extremist factions on Jihadist Facebook pages. All the while considering Edwards and Gibbons comments on “the rise of the home-grown terrorist” (44). Significantly, we look to one of the key interviewees, A3, and his comments on the importance of Youtube content and image content as having replaced the need for local or national organization. A3 states, “A person who is being radicalized will believe as they watch Al-Awlaki, that he is talking directly to them [...] A picture is worth a thousand words, a video is worth a thousand pictures” (44). With this in mind, this report examines the type of content visible on the Jihadist Facebook pages, to gauge to what extent they are effective elements of the Jihadist recruitment strategy across the European Facebook network.

The work of British NGO Hope not Hate, and specifically the work of Nick Lowles and Joe Mulhall in their report entitled “Gateway to Terror”, outlines the European context relating to Jihadist and extremist groups operating in Europe, especially with regard to the Salafist “Sharia4” organisations. This report has a heavy focus on the Al-Muhajiroun terrorist group operating in the United Kingdom and more widely in Europe, with coverage of the extremist leader Anjem Choudary and his links to the Al-Muhajiroun terrorist network and other Jihadist organisations across Europe. The linking of British terrorist organisations to a wider European network of Jihadist organisations resonates strongly with Bruno Latour’s understanding of the ‘network’ as enabling us to think in terms of “a global entity - a highly connected one - which nevertheless remains continuously local” rather than having to regard the local and the global as distinct from each other (372). In this way, the actors in a network emerge organically, regardless of - or inspite of - their strength of connection to other elements or actors in the network or of their position on a local or global scale. Facebook represents new research opportunities in this regard, and the research data presented in the next part of this report supports a claim for further analysis of the online network of European Jihadist individuals and cross-national terrorist communities.

Method

For the purposes of this study and based on the foreign language skills of the researchers, we focused on the French, English, Dutch, and German language spheres within Facebook in order to gain a representative sample of the spread of Jihadist content across four of the main European languages. New research Facebook accounts were created in order to develop profiles more fitting to the pages and groups being searched for, and to eliminate any kind of influence caused by the personal Facebook accounts of the researchers. The research was conducted on 15 January 2014 and aimed at locating key pages and groups relating to Syrian Jihadist content within the language webs considered.

Finding the first page in each of the language webs proved to be the most difficult aspect in the search for pages espousing the Jihadist movement in Syria. For each of the languages an outside source was used as a gateway into the network of Jihadist pages on Facebook. The German Facebook page ShamCenter is mentioned in an article in Der Spiegel (Spiegel Online International). Further German pages were then discovered by following the Facebook medium through the paths of ‘likes’ from one page to another. A Vice article provided a similar entryway for the English language network, as one of the images used in the article mentions the name of a relevant Facebook page (Roussinos). The report that the British Hope not Hate organisation published on the Al-Muhajiroun network that is active across Europe led to the mostly French language Tawheed4Belgium page (Lowles and Mulhall). The Dutch Ministry of Security and Justice government officials that introduced their work during the Digital Methods Winter School of 2014 mentioned useful starting points for accessing the Dutch Jihadist Facebook network. For each of these languages, care was taken to ensure that the pages were not merely radical, extremist or just promoting Islam, but were actively promoting the Jihadist fight in Syria. In order to determine whether the nature of the page was Jihadist or not, attention was directed to the specific kind of video, photo, and textual content the pages contained. For example, an image with a smiling martyr of which the accompanying text praises his efforts would be an indication of the page being Jihadist. In addition, references made to ‘Sham’ as the Arabic term for ‘Syria’ and to ‘ISIS’ in particular were examined as potential signs of Jihadist content. A total of 17 pages were discovered:

• English (8 pages):

  • The Victorious Party In the Land of Ash-Sham;

  • Sharia4Sham;

  • Bilad al Shaam;

  • Islam is Only One Solution [sic];

  • Support our beloved Mujahideen of ISIS;

  • The Islamic State of Iraq and the Ash Sham - ISIS;

  • Khilafah Shall Rise Again In Sha Allah;

  • Rayat al Tawheed III

• Dutch (4 pages):

  • Ahlus-Sunnah Publicaties;

  • Al Djannah;

  • De Ware Religie Pagina;

  • Shaam al-Ghareeba

• German (3 pages):

  • Generation Mujahida III;

  • ShamCenter;

  • Die Siegreiche Truppe 2

• French (2 pages):

  • La Sharia est meilleur que la république;

  • Tawheed4Belgium

In spite of having found Jihadist groups on Facebook for each of the language webs, they were not taken into consideration during our research as they were closed groups and our requests to join those groups were not granted within the limited timespan.

The pages were then processed using Netvizz, a Facebook research application that extracts data from the pages a user has liked and the groups that a user is a member of. We used Netvizz in order to output data on the pages and the pages’ users discovered in each of the language webs. We chose to analyse the 50 most recent posts as a sample of activity on each page. The spreadsheet created through Netvizz enabled us to analyse each page’s posts by row in the spreadsheet, making it simpler to collect and determine the type of content in each post (image, video, status or link) in addition to the amount of likes each post received and who posted and commented on each post. In order to perform an in-depth content analysis within the timespan given, we focused on the images shared which represented 41% of the total content shared across the Facebook pages. The images were collected and categorised using an emergent coding system on the basis of Stemler’s definition of emergent coding in which the categories emerge organically through the content of each image rather than being imposed on the image (Stemler). The six categories were: 1. death/ violence/destruction; 2. children/children with guns/dead children; 3. Jihadist symbols/martyrs/ political; 4. Qu’ran/religious; 5. enemies of the Jihadists/anti-West/anti-Israel; 6. guns/weaponry/ training. Standardized spreadsheets were created for each language to ensure that the same data would be collected by each researcher for each page in the language group. Data was recorded in terms of the frequency of the category, ‘likes’ on the photo, and the given category. The spreadsheets were then compared to determine the frequency of images per category for each language group (see Appendix 1). The categorised images were visualised using Adobe Illustrator to allow for an overview (see Appendix 2). To get a sense of the visual language used in the images and the overall aesthetic conveyed, we further aggregated the images into one visualisation using Adobe Photoshop (see Appendix 3). In addition, the data collected through Netvizz was visualised using Gephi in order to establish a clear understanding of the network of pages and users (see Appendix 4).

Findings

An analysis of the Jihadist Facebook pages discovered shows that at the time of research 6 out of the 8 English pages were less than 3 months old; 2 out of the 3 German pages were less than a week old; and, 1 of the 2 French pages was less than a week old. Nevertheless, most of the pages across the language webs have accumulated a significant amount of likes over their lifetime. The Dutch web has survived the longest (see Table 1).

	Date of creation	No. of Likes/ Followers	No. of posts analysed	Images	Videos	Statuses	Links
ENGLISH
The Victorious Party In the Land of Ash-Sham	7-1-2014	2720	50	31	10	1	8
Sharia4Sham	10-7-2013	401	23	6	5	11	1
Bilad al Shaam	18-12-2013	3809	50	16	9	22	3
Islam is Only One Solution [sic]	18-12-2011	313	50	20	8	12	10
Support our beloved Mujahideen of ISIS	8-1-2014	1699	50	25	3	14	8
The Islamic State of Iraq and the Ash Sham - ISIS	5-11-2013	1449	50	24	10	14	2
Khilafah Shall Rise Again In Sha Allah	2-8-2013	2818	50	28	0	11	1
Rayat al Tawheed III	25-12-2013	910	50	36	4	7	3
DUTCH
Ahlus-Sunnah Publicaties	15-6-2013	1437	50	6	6	21	17
Al Djannah	4-9-2013	879	50	14	11	21	4
De Ware Religie Pagina	2013*	1711	50	2	0	1	47
Shaam al-Ghareeba	23-8-2013	2720	50	22	16	11	1
GERMAN
Generation Mujahida III	14-1-2014	224	16	11	0	3	2
ShamCenter	7-2013*	2712	50	24	9	7	10
Die Siegreiche Truppe 2	14-1-2014	277	16	5	1	10	0
FRENCH
La Sharia est meilleur que la république	21-6-2013	253	50	26	4	16	4
Tawheed4Belgium	8-1-2014	85	19	10	2	3	4
* Limited information available

Table 1: An analysis of the most recent 50 posts on 17 Jihadist Facebook pages across 4 language webs (English, Dutch, German and French). In some cases limited information about the page was made available to the public.

However, in spite of being comparatively older than the other language webs considered and in spite of gathering a number of likes in that time, the Dutch web has expanded at a much slower rate of ‘like accumulation’ than its counterparts in other languages. Based on further analysis, we determined that this was a result of the type of content shared across the language webs.

Figure 1: The type of content shared across the 17 Jihadist Facebook pages analysed in the English, Dutch, German and French webs.

The most recent 50 posts on each page were analysed for the type of content (or, in the case of pages that had less than 50 posts, all the posts were analysed): images, videos, statuses, or links (see Figure 2). In the case of the English language web, images were by far the most shared type of content. In spite of having fewer posts to analyse due to the age of the pages in addition to the smaller number of pages found, the German and French webs also show that the sharing of images is significantly more popular than any other kind of content. The Dutch web, however, shows a tendency to opt for the sharing of links and statuses rather than images or videos. In other words, the Dutch web is overall more reliant on textual rather than visual content.

German Language Findings:

ShamCenter is a well-known page, and has had an online presence outside of Facebook since July 2013. What is remarkable about the page is the quality of the images that are shared, which constitute more than half of the page’s most recent 50 posts. These images are almost exclusively shot outside of Germany, most likely in Syria. The composition and lighting of the images is of such quality that they have most likely been taken by someone with professional equipment and with an advanced knowledge of photography. Furthermore, some of the images show a video camera. This indicates that the ShamCenter orchestrates its online presence with an understanding of marketing and style. Rather than showing martyrs and fighters, they show their equipment and the landscape. Lastly, they are very careful when showing faces, and often blur them.

Die Siegreiche Truppe 2 is one of the pages that was only created in the week leading up to January 15th. It did, however, gather 277 likes in this short period. In fact, checking the site on the 17th January shows 461 likes, showing that the page is accumulating followers at a rapid pace. The number in the name of the page indicates that there was a previous version of the page, which might explain why people know where to find it. The content of the page mostly consists of statuses that praise martyrs and the Jihad in general. A large part of the images portray Abu Wahib, an important figure in the ISIS organisation in Syria (TRAC). He is shown posing with guns and other military gear. Other images show, for example, dead children and martyrs. All are accompanied by texts praising their bravery and the wish that they are rewarded in the afterlife. The other new page that was found is Generation Mujahida III. In the short period of its existence it has managed to gather over 200 likes. It seems to have been created and managed by a woman. Many of the posts on the page focus on the female experience of the Jihad, showing drawings of female fighters and women in Syria dressed in niqabs. The images of children posted by the page actually glorify their fathers and brothers that are fighting in the Jihad, portraying martyrs and photos of the battlefields.

French language findings:

Through our reading of Lowles and Mulhall’s Hope not Hate report from November 2013, we learnt of the existence of the Sharia4 Facebook groups and pages for France and Belgium, and their connections with key terrorist figureheads such as British leader Anjem Choudary and the terrorist group Al-Muhajiroun; Belgian Sharia4 leader Abu Imran; and, French leaders Jama’at Al Tawhid and Fursan Al-’Izza. However, while the Sharia4 Facebook pages have previously been removed, we were able to find the La Sharia est meilleur que la république through the pages suggested by Facebook. The page contains links to videos depicting French Jihadist fighters in Syria as well as many images with anti-democracy slogans, one depicting French President François Hollande as a clown in addition to rousing war songs set against the background of Jihadist fighters, the Jihadist flag, and verses from the Qu'ran. Furthermore, we discovered that the Sharia4Belgium Facebook page had reformed under a new name Tawheed4Belgium (La Libre). The majority of the content on this page was written in English, Dutch, French, and Arabic, written in the Latin alphabet. However, the page was classified as a French language page due to the links with Sharia4France outlined by Lowles and Mulhall, and the fact that ⅔ of Belgian Muslims live in majority French-speaking Wallonia and in Brussels (Hertogen). Tawheed4Belgium also promotes the Jihadist cause in Syria, and features image content relating to verses from the Qu’ran, the Jihadist flag and references to the Jihadist symbol of the Lion. The page also chides Muslims who do not support Jihad, labelling them as hypocrites. The page also expresses hate sentiment in the form of an image of key governmental leaders from around the globe including Barack Obama and Vladimir Putin with the caption ‘So fight against the allies of Satan’. By these methods the page is thus attempting to glorify the Jihadist cause, in order to attract users to fight for the cause.

English language findings:

All the English language pages found contain a wide array of images depicting imagery associated with the Jihad, including martyred soldiers, weaponry, and the Qu’ran. However, in most cases the title of the page itself is a strong indication of support for the Jihadist movement in Syria: Sharia4Sham (Sham is an Arabic word referring to Syria, and more specifically to the region around Damascus); Support our beloved Mujahideen of ISIS; and, The Islamic State of Iraq and the Ash Sham - ISIS. The English page Bilad al Shaam contains the most likes, with 3809, in spite of it being quite a new page having only been created on 18th December 2013. Based on the video, textual and image content, this page’s unofficial purpose seems to be as a site for the broadcast of news from the battlefield with content associated with the success, death and spread of the mujahideen in Syria. While some of the pages appear to simply be pages set up by Facebook users in support of the cause of Jihad and ISIS, the pages called The Victorious Party In the Land of Ash-Sham and Rayat al Tawheed III appear quite official: the former uses a logo depicting the name of the page on all of its images; the latter uses the same logo depicting Arabic text on all of its images. In addition, the profile and cover pictures used on those two pages do not simply portray images related to the Jihad, but rather contain the logo and/or title of the page itself thereby broadcasting the page as a pseudo-official source of information relating to the Jihad in Syria. What is also interesting here is that images containing the logo of The Victorious Party in the Land of Ash-Sham can also be found shared on other pages such as Khilafah Shall Rise Again In Sha Allah. Therefore, in addition to the kind of content shared, what is interesting in the English web is the amount of shared content (mostly images) creating links between the pages themselves.

Dutch language findings:

The four Dutch pages found do not all focus on the Syrian Jihad, but they support the practices by offering encouraging content or justifying extremist interpretations of the Qu’ran. The page Ahlus Sunna Publicaties is an example of such a page that provides its active community with textual interpretations and supporting videos or images. Most of the content, however, links back to the official website. The same is true for the page De Ware Religie, which also spreads radical ideas of Islamic interpretations and practices. Both pages have around 1500 likes and can be considered quite active. As the Al Djanna page only has 879 likes, it can be considered the least active. This page also does not have much shocking imagery, whereas the first two pages post more graphic images that may arouse an emotional response from supporters. However, the fourth page, Shaam el-Gareeba, is the most closely related to the Jihad, as it offers daily updates on and from the Syrian battlefield. At the time of our data scrape it had 2720 likes and is in active use. Since this page is not linked to an external website, the point of focus is within the Facebook page itself. The majority of the posts consist of images, with fewer posts containing links or videos. It is interesting to see the differences in the character of the pages by comparing the images being posted. By classifying the images of the Dutch pages into categories and comparing the results of the four pages it becomes clear that the following categories are most present: ‘death/violence’; ‘children’; ‘martyrs’; ‘religious texts’; ‘enemies’; and, ‘weaponry’. These kinds of images seem to be used to stimulate the emotions of the page’s followers. The most liked picture in two pages was that of former Israeli president Ariel Sharon who passed away on the 11th January 2014. After that, the most liked pictures portray Jihad fighters as true heroes, dead or alive, since dying for Allah in the Jihad is praised. Next to that children or adults posing with guns is a common way of approving these practices.

Network Findings:

Figure 2: Gephi visualisations of the three language webs (Nodes represent users; edges represent the connection between the users and the 3 language webs - the majority red is the English web; majority turquoise is Dutch; majority blue is German and majority yellow is French. Red nodes represent English language interfaces; Turqoise represent Dutch; Blue represent German; Green represent Arabic; Yellow represent French). The first graph shows an overview of user engagement by interface langauge; the second graph gives a clearer indication of the language webs; the third graph gives a clearer indication of the language webs by increasing the node size of the most active users.

In order to better understand the networks of users and interrelationships between pages and language webs based on the most recent 50 posts of each page, we used gephi (.gdf format) files output through Netvizz for each chosen Facebook page and visualised them with the Gephi software application, thereby establishing the connections shared between the four European language groups (Dutch, French, English, German). The Gephi software also allowed us to further look at language usage within the outlined language webs in terms of the language interface used by each user (see Appendix 4).

The German language web has the greatest plurality of languages, with a majority of German language and Arabic language user interfaces present, as well as French, Danish and Turkish language user interfaces. It should be noted that languages used in the German web were not just European languages. In fact the Arabic language interface users were the most concentrated in the German web, represented as the largest green nodes in the Gephi visualization (see Appendix 4). The German web is particularly well linked to the English language group with a number of users common to both groups. While this is significant, we also must bear in mind that 8 Jihadist Facebook pages were encountered for the English web, compared with 3 for the German web. Thus, as the number of groups increased, so too did the probability of finding common users between the language webs. The German web also shares users with the Dutch web, though not as many as with the English. Notably, the German web has no connections to the French web - although this may be related to the size of the sample - as here the probability of finding common users is lower. As mentioned, the French language group consists of a small sample size of two Jihadist pages. Here, French is the majority language of the users with Dutch and English in the minority. The Dutch interface languages can be explained by one of the pages primarily targeting on a Belgian audience. Some users in the French language group are also common to the English and Dutch language groups. Thus Jihadist Facebook groups across the French language web do not appear to be as multilingual as the German equivalent.

The majority of users within each language group do not deviate from the main language interface connecting each web. This is especially true for the English language web where the majority of users use Facebook in English. The English language web is extremely well connected to the Dutch and German language webs, and it shares the majority of its common users with the Dutch web. While the majority language of the English web is English, there still remains a plurality of global and European languages within the network of users. This suggests that the English language web is reasonably multilingual, although not as multilingual as Germany. The majority of users in the Dutch language web use Jihadist Facebook pages in Dutch, although there are also English, Arabic and French user interfaces present. The Dutch language web is well-linked to the English language web, with a plurality of users common to both webs. The Dutch language web also links to the German language web and to the French language web, though these connections are less pronounced than those connections with the English language web. These strong connections may also be influenced by the larger sample size introduced for the English language web, when compared with the sample for the Dutch language web of four Jihadist pages.

Discussion

General analysis:

Since the Jihad in Syria has been present since 2011, the newness of the majority of the Facebook pages found is evidently not a reflection of the age of the crisis in Syria. A case can be made for the claim that these Facebook pages themselves are new but are not necessarily unprecedented. Most of the pages across the language webs considered have accumulated a rapidly increasing number of likes and followers in a short timespan, indicating that there is a large sector of Facebook users interested in the content of these pages; users who then must be part of a wider network of users who are also interested in the content of these Jihadist pages. Furthermore, there is evidence that some pages have previously been removed and/or are setting up ‘back-up pages’ just in case anyone reports the page and requests its removal. The Victorious Party In the Land of Ash-Sham is linked to more pages of the same name with an added number on the end associated with the number of that page, e.g. The Victorious Party In the Land of Ash-Sham 8 and The Victorious Party in the Land of Ash-Sham 9. The Bilad al Shaam page even included the following in its description: “Bilad al Shaam.. 26.. The true voice of the Freedom Struggle of Vilad al Shaam! Blocked by Facebook thugs for the 25th time. Here we Rise Again.” In fact, Facebook is here a quasi online ‘battlefield’, on which opposing pro-/anti- Assad and Jihadist pages are campaigning to get each other blocked:

Figure 3: An image lifted from the Bilad al Shaam Facebook page showing that the page successfully reported and blocked another Facebook page that was supporting the Assad regime.

The fact that the Dutch pages have survived longer than most other pages analysed is a reflection of the type of content shared: the Dutch pages are overall more reliant on textual rather than visual content. The sharing of images is likely to drive an increasing number of users to a profile or page in addition to motivating a higher rate of interaction and engagement with the content of that page - in this case, Jihadist content - than pages that do not share a high amount of visual content (Jernstedt). Therefore, while the number of people drawn to the Dutch pages accumulates at a slower rate, this enables the Dutch pages to survive for a longer amount of time than the other pages which quickly accumulate followers in addition to the attention of users who do not support the Jihadist movement in Syria (and who then campaign to block or remove the Jihadist Facebook pages). In fact, as mentioned briefly in the Findings section, in spite of the links shown between the Al-Muhajiroun network and the Sharia4 pages in the Hope Not Hate report “Gateway to Terror”, the Sharia4 pages such as the Sharia4Belgium page are likely to be blocked and removed from Facebook. However, blocking this kind of content is not as efficient as a counter-Jihadist strategy as it may seem; the administrators and audiences of those pages do not disappear along with the pages, and tend to reappear and grow under a different ‘alias’, such as the Sharia4Belgium page reappearing as Tawheed4Belgium.

Image sharing:

During our research across the 17 Jihadist Facebook pages analysed in our chosen languages and as discussed above, images were the most commonly shared type of content. This corresponds with an interview with a radicalised extremist conducted by Edwards and Gribbons, where the interviewee puts emphasis on the importance of creation of media online to facilitate both proselytism and radicalisation (44); thus we deemed it important to analyse the images shared on these pages in terms of both aesthetic qualities and through categorisation of content. Motivational factors for media sharing can be numerous: to maintain social relationships, to remind of individual and collective experiences, for self-presentation, self-expression, or task performance (Goh 199). For our purposes in this analysis, we primarily focus on the images as a means to remind of individual and collective experience, and for the improvement of task performance. The individual or collective experience in this case is the role of the Muslim community as a whole, and how that community relates to Jihad and Sharia in relation to Syria via these pages; we think about Sharia and Jihad in the context of social media as tasks to be performed.

Figure 4: A collage of all the images gathered from the Jihadist Facebook pages considered during our research.

To analyse the images as relating to collective experience, we created a cross-page and cross-language collage of all the images shared in the last 50 posts as scraped by Netvizz on 15 January 2014. In total across all languages, there were 339 images in our data set. This process was inspired by Lev Manovich’s work for the Software Studies Initiative, specifically as documented in the article “How To Compare One Million Images?” by Manovich, Jeremy Douglass and Tara Zepel. Limited by time, resources and working with a smaller group of images than typical for Manovich’s research, we manually constructed a collage based on aesthetics and content (See Appendix 3). This collage shows how images appear multiple times across the pages studied, and also how common colours, compositions and themes reappear across the shared images. By constructing this collage of images, we can see that the majority show people: adults and children, both living and dead. Many of the images also employ text, either against a block colour background or added to a photograph to reiterate meaning. Other more incongruous and innocent images appear, for example a photograph of a watermelon with “Ramadan Mubarak” (Happy Ramadan) carved into it. When viewed alone, this image shows the celebratory and tradition oriented nature of Islam; however, when viewed simultaneously with other images depicting death and destruction, this image acts as an important part of the assemblage of images contributing to the potential radicalisation of users. By showing content depicting violence alongside non-extremist Islamic traditions such as Ramadan, these pages normalise extremism and Jihad for Muslim users of social media networks.

This normalisation is further demonstrated through the use of Instagram or Instagram style aesthetics placed on images of martyrs and other scenes of unrest in Syria. These images are typically square framed, and make use of digital filters. When applied to images of martyrs, these filters create an otherworldly, ethereal glow; this method of image presentation used by these groups use easily reproducible, generic software settings to imply divine intervention. This demonstrates an interesting combination of religion and technology: while Islam is based on faith, - which cannot be expressed through a man made image - the administrators and users of these pages utilise images as a form of ‘proof’ for the existence of the afterlife, and to show that these martyrs have performed their duty through the task of Jihad, and have ascended to a higher plane of existence. In terms of our research question, we can consider this distribution of digitally faith confirming content as a potentially driving force for recruitment of radicalised Muslims across different European languages to journey to Syria to assist the Jihadist movement.

Media shared as a motivational factor of task performance “have a functional value and are meant to assist the sender and/or recipient to complete a task” (Goh 201). We can consider these pages as attempting to perform a task: that is, the support of and the recruitment to the Jihadist movement in Syria. As such, the media - and, for our purpose of analysis, specifically images - shared can be seen as a means to an end in the task of encouraging the move from more passive, online involvement, to physical, real life involvement in Syria. Considering the images thusly, they must form specific categories to aid in the completion of this task, demonstrating political, emotional, practical and spiritual incentives to join the resistance movement. These categories can be seen through the groupings of images in Appendix 3, demonstrating the overriding aesthetic approach of these pages; however, to adequately analyse the content of the images for the purposes of our study, we cannot only rely on this kind of visualisation. It is necessary to further visualise our categories to provide a metrified rather than only aesthetically organised presentation of our visual data.

Distribution of image categories:

Figure 5: Categorisation of images analysed.

After putting the images that were used in the most recent 50 posts of each page into categories, we were able to interpret the strategies that are used in these pages. Appendix 1 lists the categorisation and the distribution for each page, each language, and the overall distribution in all European pages we have used in this project. While the graphs that we created with Gephi illustrate the shared users and similarities across languages, the graph in Appendix 2 visualises the differences between languages in the type of images that are used on their pages. Important in reading the graph is that the icons do not represent individual images, but rather percentages of the total number of images used within that language. The most radical differences between languages occur between the French and German pages. As was shown in the Gephi graphs, no users were shared between these languages. As we can see in this visualisation, on top of that their strategies are quite different from each other. The German language pages use a majority of images of guns and training sites. While these images do portray the Jihad, most of them are rather easy to look at or even innocent. The category in green that lists Jihadist and political images on the other hand can be quite graphic and bloody. This category is overwhelmingly used in the French languages pages. This indicates that the French language pages are rather militant in their message. The Dutch and English use images out of each category rather evenly. As the Gephi graphs showed, a lot of used cross the language barriers between these two languages, adding to their interlinkages and similarities. The English language pages do have a slightly larger percentage of images in the Jihadist category, again indicating a more aggressive approach.

Network Analysis (see Appendix 4):

A network of Facebook pages that are not restricted by national boundaries but rather embrace the opportunities for the dissemination of information through a distributed online network allows online extremists and potential Jihadists to maintain a strong and coordinated presence. In this way Facebook is a ready structure to be used as an organisational tool among diffuse groups and individuals (Lewis 112) regardless of their role, presence or hierarchy in organised terrorist networks ‘on the ground’. The strong interlinkages between the language webs indicate a plurality of users common to at least one other Jihadist Facebook page, with some users being common to two other pages. This is initial evidence of a pan-European Facebook network of Jihadist groups and individuals. Furthermore, the multilingual nature of the user interfaces across the network is particularly interesting and is especially evident in the German language web, with the most active language users being Arabic language interface users. In addition, users of the German language web are more strongly linked to the other language webs (except the French web) rather than the other way around. This multilingual aspect further indicates the potential for the recruitment of distinct language groups within the population of users. The online network of Jihadist Facebook pages and users represents a community of like-minded individuals (Hale) that have taken the opportunity presented by Facebook to advance extremist narratives and interact with foreign content and information to an extent that raises the network above any sort of national boundaries within Europe.

Conclusion

Since 2011, Syria has become a region of conflict in which Jihadist forces retaliate against the regime of President Bashar al-Assad. The Jihadists go under the name of ‘Islamic State of Iraq and Syria’ (ISIS) and their aim is to enforce Islamic rule in Syria. In addition to the Syrian fighters are European citizens joining the Jihadist movement united through an online overarching network. It is interesting to take a closer look at this large radical Islamic network and examine its content, cross-European connectedness and the strength of its recruitment function. The possibilities of social media fit the purposes of spreading ideas and connecting like-minded people. As social media facilitates violent extremists by remaining anonymous and avoiding legal sanctions it has become a standard for subversive extremist groups to recruit and operate through social media platforms. Therefore, this research looked into the online network of Jihadist pages through four different languages in Facebook’s platform. The four language webs studied were: English, German, French and Dutch. Next to that, these languages serve as a good research sample for the spread of Jihadist content among Europe. The aim of the research was to understand the kind of content being shared through the European Facebook pages, and to establish connections through shared content and users across Facebook pages, as a portrayal of online European pathways of extremism to Syria. The following research questions were examined throughout this research:

1) How have Facebook pages in Europe been used as a voice for the Jihadist cause in Syria?

2) To what extent can Jihadist pages on Facebook be considered a recruitment tool for ISIS and Jihadist networks in Europe?

We decided to focus on Facebook because it is useful for observing research subjects in naturalistic settings and its data forms a steady base for researching networked relationships or movements (Wilson et al.; Rieder). This research focuses mostly on the network constituted by (inter)national communities which thrive in Facebook’s format. This research relates mostly to Edwards and Gribbon, who gathered ‘primary data’ relating to online interaction with extremist content through interviews. Through our method, we made use of data scraped from Facebook using the Netvizz tool to investigate user interaction with digital content, in this case European Jihadist Facebook pages. We found 17 pages among the four languages, with 8 pages for English, 4 pages for Dutch, 3 pages for German and 2 pages for French. Although some of the pages are quite young, the accumulation of users’ likes shows there is quite a big public for Jihadist content, and the pages have mostly active communities.

As the Netvizz data showed that a majority of the content were images, we took these as the central point of our research. We found that the German language pages mostly use imagery of guns and training sites. On the contrary, France is more militant by portraying mostly political and Jihadist images that have a more shocking nature. The English pages show more pictures from the Jihadist category as well and can therefore be claimed to have a more aggressive approach. The Dutch pages used imagery from all categories evenly, and focused on differentiating between martyrs and enemies clearly. The style of the pictures shows they are taken mostly with smartphones, some even include the use of Instagram filters or other applications like it. Furthermore, a lot of the pictures display updates straight from the Syrian battlefield. Our collage of all 339 images show that many images are shared among different groups, even across language barriers. The strong emotional effect of the picture contributes to the potential radicalisation of users. The pictures function as a reminder of the purposes of the pages: to recruit men to fight for the Jihad and die for Allah.

Using Gephi, we created network graphs of the user groups of each language web. These show that German web has the greatest plurality of language interfaces, with a highly concentrated group of Arabic user language interfaces next to the majority of German language, as well as French, Turkish and Danish. The German language web is linked to the English web with a number of common users. It is linked to the Dutch web to a smaller extent, but is not at all linked to the French web. The English web shares users with the German and Dutch webs and can be considered multilingual, although not to the same degree as the German. The Dutch web has a variety of English, Arabic and French user interfaces besides the Dutch language. The French web turns out to be the least connected, but this is also due to the sample sizes of the pages for each language. Altogether this proves there are strong inter-linkages within the European Jihadist network on Facebook, as many users are shared between one or even two other Jihadist groups. Therefore, we can speak of a pan-European Jihadist Facebook network online with a multilingual nature of user interfaces, especially for the German language web. This indicates the potential recruitment of distinct language groups among Europe for the purpose of the Jihad in Syria.

Appendix 1: Collection of Netvizz Data

Appendix 2: Categorisation of Images

Appendix 3: Collage of Images

Appendix 4: Network Visualisation

Bibliography

Burke, Daniel. “Terrorist attacks and deaths hit record high, report shows.” CNN. 28.10.2013. Accessed 15.01.2014. <http://religion.blogs.cnn.com/2013/10/28/terrorist-attacks-and-deaths-hit-record-high-report-shows/>

Chen, Hsinchun. “Cyber Extremism in Web 2.0: And Exploratory Study of the International Jihadist Groups.” Intelligence and Security Informatics. (2008): 98-103.

Douglass, Jeremy, Lev Manovich, Tara Zepel. “How To Compare One Million Images?”. Software Studies Initiative. 2011. <http://softwarestudies.com/cultural_analytics/2011.How_To_Compare_One_Million_Images.pdf>

Edwards, Charlie and Gribbon, Luke. “Pathways to Violent Extremism in the Digital Era.” The RUSI Journal 158.5 (2013): 40-47.

Global Web Index. 2013. Accessed 17.01.2014. <https://www.globalwebindex.net/products/report>

Goh, Dion Hoe-Lian. “Why We Share: A Study of Motivations for Mobile Media Sharing.” Lecture notes in Computer Science Volume 5820. (2009): 195-206.

Finch, Rob, and Kory Flowers, "Violent Domestic Extremism and the Role of Social Media within Law Enforcement," The Police Chief 80 (June 2013): 32–34. <http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=2952&issue_id=62013>

Hale, W. Chris. “Extremism on the World Wide Web: a research review.” Criminal Justice Studies: A Critical Journal of Crime, Law and Society 25. 4 (2012): 343-356.

Hertogen, Jan. Non-Profit Data. 12.09.2008. Accessed 15.01.14. <http://www.npdata.be/BuG/100/>

Jaiatt, Alexia. “Sharia4Belgium anuncia su ¿disolución?” Minuto Digital. 09.12.2013. Accessed 14.01.2014. <http://www.minutodigital.com/2012/10/09/sharia4belgium-anuncia-su-disolucion/>

Jernstedt, G. Christian. Using Findings from Brain Research to Improve Financial Education Outcomes. Using Social Media, Social Marketing and Brain Research to Improve the Design and Delivery of Financial Education, June 2010. Washington, DC: FINRA Investor Education Foundation, 2011.

Jihad Watch. Robert Spencer. 2014. Accessed 15.01.2014. <http://www.jihadwatch.org>

Kern, Soeren. 'The Islamization of Belgium and the Netherlands in 2013'. Gatestone Institute International Policy Council. 13-01-2014. Accessed on 19-01-2014. <http://www.gatestoneinstitute.org/4129/islamization-belgium-netherlands>.

Latour, Bruno. “On actor-network theory: A few clarifications.” Soziale Welt 47 (1996): 369-81.

Lewis, James A. “Internet and Terrorism.” The American Society of International Law Proceedings 99. (2005): 112-116.

Lowles, Nick. Counter Jihadism: Introduction. London: Hope Not Hate, 2012. <http://www.hopenothate.org.uk/counter-jihad/>

Lowles, Nick and Joe Mulhall. Gateway to Terror. Anjem Choudary and the al-Muhajiroun Network. London: Hope not Hate Educational Ltd., 2013.

Netvizz Facebook app. Accessed 15.01.2014. <https://tools.digitalmethods.net/netvizz/facebook/netvizz/>

Reuter, Christoph. “Masked Army: Jihadist Group Expands Rapidly in Syria.” Spiegel Online International. 18.12.2013. Accessed on 15.01.2014. <http://www.spiegel.de/international/world/isis-shadowy-jihadist-group-expands-rapidly-in-syria-a-939561-2.html>

Rieder, Bernhard. “Studying Facebook via Data Extraction: The Netvizz Application.” University of Amsterdam. (2013): 1-10.

Rogers, Richard. The End of the Virtual. Amsterdam, The Netherlands: Amsterdam University Press, 2009.

Rogers, Richard. “Right-wing Formations in Europe and Their Counter-Measures: An Online Mapping.” CounterExtremism.org. 2013. Accessed 15.01.2014. <https://www.counterextremism.org/resources/details/id/338/right-wing-formations-in-europe-and-their-counter-measures-an-online-mapping>

Roussinos, Aris. “Jihad Selfies: These British Extremists in Syria Love Social Media”. Vice. 2013. Accessed 15.01.2014. <http://www.vice.com/en_uk/read/syrian-jihadist-selfies-tell-us-a-lot-about-their-war>

“Sharia4Belgium toujours présent sur le net.” La Libre. 17.10.2013. Accessed 15.01.14. <http://www.lalibre.be/actu/belgique/sharia4belgium-toujours-present-sur-le-net-525fb2d2357026746e8ab96>

Soeren, Kern. “The Islamization of Belgium and the Netherlands in 2013.” Gatestone Institute: International Policy Council. 13.01.2014. Accessed 15.01.2014. <http://www.gatestoneinstitute.org/4129/islamization-belgium-netherlands>

Spiegel Online International. “The ‘German Camp’: Jihadists from Germany Set up Base in Syria.” Spiegel Online International. 21.10.2013. Accessed on 14.01.2014. <http://www.spiegel.de/international/germany/german-camp-collection-point-for-german-jihadists-in-syria-a-929026.html>

Stemler, Steve. “An overview of content analysis.” Practical Assessment, Research & Evaluation 7.17 (2001). Accessed 15.01.2014 <http://PAREonline.net/getvn.asp?v=7&n=17>

Tarrero Alonso, Óscar. “Monografía: Islamismo Radical en España” Escuela Superior de las Fuerzas Armadas. Departamento de Inteligencia. 09.05.2010. Web. Accessed 14.01.2014. < http://www.defensa.gob.es/ceseden/Galerias/esfas/destacados/en_portada/ISLAMISMOx20RADICALx20ENx20ESPANA.pdf >

Who is Shaker Wahib al-Fahdawi aka Abu Waheebaka aka Nusayri Hunter aka teacher of the Nausayris? 2014. Terrorism Research & Analysis Consortium TRAC. 2014. Accessed 15.01.2014. <http://www.trackingterrorism.org/article/rising-star-isis-sunni-terrorist-abu-wahib-shakir-al-fahdawi-leader-ussud-al-anbar-brigade>

Wilson, Robert E., Samuel D. Gosling and Lindsay T. Graham. “A Review of Facebook Research in the Social Sciences.” Perspectives on Psychological Science 7.3 (2012): 203-20.