As I am fortunate to have a large number of data analysts and those involved in the behaviour sciences among my LinkedIn contacts, I would like to point out this paper is not intended to bring anything new to the study of radicalization or extremist behaviour. I also feel sure that many of my contacts in this field will put forward various other methods which may be used to collect the same datasets mentioned in this paper.

Several years ago, as part of my research into the induction and radicalization process used by AQ affiliates via social media (SM), I spent a considerable amount of time reading academic papers on SM mapping and human behaviour.  This information allowed me to research SM, the web and dark web in order to increase my understanding of the drivers associated with violent extremism (VE) and the mindsets of vulnerable people who may be psychologically manipulated to join the extremist cause. It also allowed me to examine and test new theories put forward by various academics.

The following is a basic introduction to the subject which is based on the research of others and which I have modified for my own research needs. Furthermore, due to the limited scope of this paper I have not included data associated with demography, gender; or the analysis of text and visual images which are to be found in the ‘extremists’ virtual world of their making.

Finally, although I and other members of the Narrative Strategies Team (http://www.narrative-strategies.com/) have a comprehensive understanding of the narrative based drivers associated with VE, I have found the following allows us to examine these drivers working over time and space along with the behavioral changes experienced by some members of the target audience.

Analysing Social Media (SM networks)

Virtual social networks, like those found in the ‘real’ world, consist of relationships and relationship building blocks. An examination of this network reveals a combination of relationships which create identifiable patterns of connected people, groups and organisations.  As explained later, this virtual social network which appears to allow users to remain anonymous provides a false sense of security where members are willing to express their concerns, frustrations and other personal information which they may not be willing to discuss in the real world. This provides an indication of an individual’s vulnerabilities which may leave them open to psychological manipulation.  When one examines the communications between like mined individuals within this network it may first appear to resemble a peer-group support network which by its very nature encourages additional personal information to be shared with ‘like-minded’ people. Accordingly, extremist groomers and recruiters can select suitable individuals who may be radicalized.

Virtual Social Networks

It is easier to regard social networks as consisting of social entities: actors, distinct individuals, groups and organisations. We must also be prepared to follow these entities as they migrate to or simultaneously use other SM platforms.  For instance, Twitter is limited to the maximum use of 140 characters (Tweets) and due to this limitation member who are of interests to extremists are often encourage to join a similar network on another SM platform with less restrictions and/or is considered more secure.  Consequently, it is not uncommon to find the same social entities on various SM platforms.

Relationship ties (Contacts)

Some relationships which are tied to others across the network/s are said to be ‘informal’ because they are not widely known by others entities of the network under examination.  For example, on LinkedIn we often find third degree contacts commenting on updates posted by members from outside their network simply because the commentator is connected to one or more of the writers’ first degree contacts.  Such entities, in this example LinkedIn members, are often referred to as ‘Muktiplexity’ or ‘Multiplex’ because these individuals are actors with ties to other actors connected to you. I plan to cover this concept in greater detail at a later date during my examination of Russian trolls and the information war.

The Two Node Network consists of actors who may not have direct ties with each other but they attend similar events within a community (Mosques, sports clubs etc.) or may regularly visit similar websites. Although there are no virtual or physical connections, this provides an opportunity for prominent actors (Focal Actors) to form a false rapport with members of the Two Node Network and the opportunity to form a ‘weak’ link.  The establishment of ‘strong’ links are only attempted after an individual is thought to be of interest to the extremist cause.

Egocentric, also called personal networks, tie directly with Focal actors (those with influence, i.e., groomers, recruiters, propagandists etc.) in the network.  Hanson and Shneiderman describe this as, “Social Centric or complete network consisting of the relational ties among members of a single bounded community. (Social Network Analysis: Measuring, Mapping and Modelling Collections of Connections, D. Hanson and B Shneiderman, 2010).

The examination of networks also allows us to develop what some academics call ‘name generators’ which is simply the names of social entities, in this case people, who frequently communicate with the focal actors.  Hanson and Shneiderman call those names generated by the focal actor, ‘the actors alters’.

The use of name generators, as advocated by Hanson and Shneidrman, allows for the identification of strong ties across a dense network.  To identify weaker ties in more wide ranging networks, acquaintance name generators can be used.

Another useful tool discussed by Hanson and Shneiderman, is the Positioning Generator. This allows the researcher to identify people who fill a particular ‘valued’ role or position within the network and therefore have access to a range of resources. These resources may include professional knowledge, or work related experience beneficial to an extremist group.

Psychological Manipulation

Apart from the same narrative based drivers being used within the real and virtual world, we also find the same methods used to encourage members of their target audience to express their concerns, frustrations, aspirations and how they see themselves.  This information is used to psychologically manipulate suitable members within the network and tie them to others with similar mindsets. The linking of suitable individuals within a network will often reinforce these concerns and produce suitable conditions for physiological manipulation. A United Nations report describes this as, psychological manipulation, “to undermine an individual’s belief in certain collective social values, or to propagate a sense of heightened anxiety, fear or panic within a population or subset of the population…” (The Use of the Internet for Terrorist Purposes: United Nations Office of Drug and Crime, NY 2012) It is also widely acknowledged that certain cognitive propensities can combine to create a mindset that presents a high risk of being radicalized (see Drivers of Violent Extremism: Hypotheses and Literature Review, RUSI, 16 October 2015) and it is these propensities which extremists seek to identify within members of the network.

Social media has made social connections and networks more visible and open to research. “The internet and its use by terrorist organisations, individual members, supporters and recruits afford new avenues for assessing information about groups and their activities…” (Lorraine Bowman-Grieve, Security Informatics, 2013, 2:9) As Bowman-Grieve says, “individual reasons why people become involved are many and varied, with no single catalyst event that explains involvement.” However, research indicates that involvement is a gradual process that occurs over time and the development of this process, which is driven by narratives and supported by inter-personal bonds that have been created for this purpose, can be examined through social network analysis.

By analysing network activities over a period of time not only do we see the use of narratives as efficient drivers towards extremism, we also see the development of identities being slowly formed. This includes perceived victimization and attempts to convince individuals they are victims and linking this to a common or shared identity and the legitimization of violence to address these perceived injustices. We also see the development of dualist thinking which supports the extremist’s’ view of the world, other cultures, religions and western society.