How can uncovering patterns in extremist messaging help with identifying and countering extremist content?

Roshonara Choudhry, a London-based student arrested for her attack on a Labour politician in 2010, is said to have drawn inspiration from a series of YouTube-hosted sermons by the Islamist preacher Anwar al-Awlaki. The Boston bomber Tamerlan Tsarnaev, far-right Norwegian extremist Anders Breivik, and Arid Uka, who carried out the shooting of a serviceman in Frankfurt, Germany, are all thought to have been influenced to some degree by online extremist material. What is it about the content of such communications that makes them so appealing to particular individuals? How do authors convince people of the efficacy of their cause?

Messaging strategies and tactics

My research has sought to establish patterns in the language of extremist groups in order to ascertain a common set of strategies used by authors in their attempts to persuade others. These strategies include argument-focused strategies, such as applying pressure directly to the audience in the form of commands; group-focused strategies such as the use of moral comparisons between in-groups and out-groups or heavy reliance on social norms; and author-focused strategies that include attempts to establish likeability with the audience or inspire them.

Influence tactics vary from group to group and from individual to individual. For example, Osama Bin Laden’s messages were primarily characterised by moral arguments, whereby authors justify the use of violence as a means of redressing perceived immoral actions against their people (this includes, for example, any Western government actions that can be perceived as evidencing double standards or ill-treatment). The messages of Ayman al-Zawahiri, on the other hand, were characterised by employing a wide range of persuasive levers, such as direct propositions and demands combined with morality and authority-based arguments.

Other tactics can include images, for example, those posted by Islamic State affiliates on Twitter. Images such as those showing members posing jovially in everyday settings can help to support written messages, by cementing an appealing sense of camaraderie and belonging between fighters.

These kinds of messages are good at mixing mutually reinforcing strategies. For example, using identity-based arguments referring to one’s duty as a Muslim alongside rational choice-based arguments referring to the perceived quality of life one will gain from engagement in extremist activity.

Engagement with extremist messages

More recently, research in this area has moved beyond message content to investigate individuals’ engagement with extremist messages. A recent study by Vergani and Bliuc highlighted the link between the psychological traits of message recipients and the appeal of particular forms of persuasive rhetoric.

In the study, the language of Islamic State’s 'Dabiq' magazine and al-Qaeda in the Arabian Peninsula’s 'Inspire' magazine was compared using a computerised text analysis programme. 'Dabiq' content was rated higher in religious and authoritarian values.

The researchers then presented participants with extracts of extremist messages from 'Dabiq' and 'Inspire', disguised as science fiction narratives. The participants were asked to express their attitude toward these extracts. Those with higher scores on the character traits of religiosity and authoritarianism were more likely to express positive attitudes towards the extracts from 'Dabiq', thus potentially explaining the group’s comparative popularity with certain kinds of followers.

While research has shown how communications may elicit positive attitudes amongst audiences, do extremist messages persuade individuals to carry out acts of extremism, or do they merely reinforce previously held worldviews? Vergani and Bliuc’s study actually found that consumption of extremist material increased pro-social behaviour in individuals with low or medium trait aggression.

The benefits of establishing patterns of extremist messaging

Of course, not all target-audience members have low aggressive tendencies, and the divisive nature of extremist messages does pose a threat to societal stability. While the influence tactics employed may not be successful in convincing all of their target audience, those the messages do convince can go on to threaten the security and safety of others. Therefore, an understanding of how such messages influence particular individuals remains of great importance.

Establishing reliable patterns of influence in extremist content allows for the development of automated tools to assist investigators in locating and assessing potentially problematic content. Certain tactics might be found to unite a particular type of extremist message, such as the activating of audience members’ commitments in Islamic extremist messages, or the othering tactic frequently used in far-right messaging. However, we should be mindful that each group is driven by its own motivations and causes. Therefore, any attempt to establish reliable patterns in influence tactics must consider within, as well as across group perspectives.

The identification of influence tactics featured in extremist messages may also be useful in the creation of counter messages as an alternative strategy to takedowns. However, a key consideration raised here is the extent to which one can utilise the influence tactics derived from extremist messages to create an effective set of counter-persuasion strategies. Here the focus should shift towards a more enhanced understanding of how consumers respond to particular influence tactics and from whom, with consideration given to individual differences, and favourably received influence tactics informing counter-terrorism responses.

Issues for the road ahead

Progress has been made towards understanding how the influence of extremist messages is affected by external factors, such as psychological traits and credibility. However, there needs to be a continued and broader consideration of the interaction between the influence tactics expressed within a message and both the immediate and external context of their reception.

Such work will help provide much-needed answers to questions such as how the medium and modality of a message and its co-occurring material affect its level of influence, and which messages are of particular cultural importance, thereby affecting their ability to influence within a particular community.

Read more
  • Sheryl Prentice, Paul J. Taylor, Paul Rayson, Andrew Hoskins & Ben O’Loughlin. 2011. Analyzing the semantic content and persuasive composition of extremist media: A case study of texts produced during the Gaza conflict. Information Systems Frontiers131: 61-73. Available at: https://goo.gl/iENnsy
  • Matteo Vergani & Ana-Maria Bliuc. 2017. The Language of New Terrorism: Differences in Psychological Dimensions of Communication in Dabiq and Inspire. Journal of Language and Social Psychology, 0261927X17751011. Available at: https://goo.gl/dCN2hM