Understanding and combating extremism online

Friday June 21, 09.00 - 10.30

Session 4, Auditorium 6, Eilert Sundt building

Chair: Petter Nesser

Paper presenters:

  • Weeda Mehran: Branding Jihad:  Comparative Analysis of Jihadi Extremists’ Media Strategies
  • Chelsea Daymon: ISIS Culture on Telegram
  • Too Good to be True? Evaluating the Accuracy of Automated Data Collection - Brian Wingenroth, Erin Miller, Michael Jensen, Omi Hodwitz, Kieran Quinlan, Michael Distler,
  • The semiotic construction of identity for radicalisation: A linguistic analysis of terrorist online recruitment materials - Aqsa Isa


Branding Jihad:  Comparative Analysis of Jihadi Extremists’ Media Strategies

Weeda Mehran, Maura Conway, Tony Lemieux

Jihadi extremist groups, from ISIS and Al Qaeda to the Taliban have used various means of communication to get their messages across, attract potential recruits and propagate their ideologies.  Jihadi extremists communicate in three ways:   (1) the act of violence itself or propaganda of the deed, (2) the narrative to justify violence, and (3) the message outlining how the group intends to achieve its goals (Baines and O’Shaughnessy 2014).  It is the last two, the narratives and the messages that are the focus of this study. 

Narratives play an indispensable role for jihadi extremists as these groups use narratives to elicit support and challenge perceptions. In this paper, we conduct a comparative analysis of media strategies of ISIS, Al Qaeda, Taliban, Tahrik-e Taliban Pakistan and Lashkar-e Taiba (LeT).  We discuss the similarities and differences in the narratives of these groups and analyse their branding strategies—establishing and positioning an identity—and investigate marketing of these brands to potential audiences.  The analysis is primarily based on the online textual materials of the groups complemented by an extensive review of the relevant literature.

ISIS Culture on Telegram

Chelsea Daymon, School of Public Affairs at American University in Washington, D.C

Culture is an instrumental element in understanding the ecosystem and functions of a group. Culture provides unity, identity, and belonging, while helping to define shared patterns among members. Definitions of culture vary; however, it can be agreed that culture helps shape the customs, beliefs, and social behaviors of individuals, making it a powerful component for understanding groups.   

Despite its loss of territory, supporters of the Islamic State of Iraq and al-Sham (ISIS) are still active on the encrypted platform Telegram. Official and semi-official channels continue to disseminate breaking news, videos, photo reports, and informational products, while pro-ISIS chats expand on these by offering a hybrid version of jihadi culture found in the online environment. Ramsay argues, that the online world offers a jihadi subculture different from the offline world. While there are similarities between the two, there are also differences. Understanding jihadi culture in the online environment, through a case study of ISIS, is of importance since groups continue to utilize platform like Telegram. One could argue that with its loss of territory, online jihadi culture, helps sustain ISIS in the long-term ensuring a continuation of ideas, support, and mobilization, which future groups will learn from.

Too Good to be True? Evaluating the Accuracy of Automated Data Collection

Brian Wingenroth, University of Maryland, Erin Miller, University of Maryland, Michael Jensen, University of Maryland, Omi Hodwitz, University of Idaho, Kieran Quinlan, Booz Allen Hamilton, Michael Distler, University of Maryland

Recent years have seen multiple initiatives that promise real-time, automated event data on social behavior and political conflict. Analysts enticed by the appeal of low-cost, timely data use these datasets with the assumption that developers implement techniques to ensure their validity, but the extent to which the data represent an accurate reflection of reality remains an open question. For example, we observe that relying on these fully automated datasets might lead one to conclude that President Obama was successfully assassinated more than 900 times. No matter how sophisticated one’s analytical methodology, the uncritical use of data can lead to problematic inferences and policy recommendations. Systematic evaluation is essential.

We compare terrorist attack datasets collected using varying degrees of automated and human workflows. We develop a standardized scope and coding scheme to compare event inclusion, date, and location, case by case.  We leverage statistics commonly used in information retrieval to quantify key threats to the accuracy of the data—undercounting, overcounting, and date/location errors. We introduce a strategy for evaluating improvements to automated data collection and conclude by discussing underutilized computational techniques that offer promising solutions in pursuit of data that maximizes cost-effectiveness and validity.

The semiotic construction of identity for radicalisation: A linguistic analysis of terrorist online recruitment materials

Aqsa Isa, Lancaster University

In this research, I examined the ways the terrorist group under study talk about how they work and how people should live and see the world through their online videos and magazines. The analytical framework involves integrating several different approaches from critical discourse studies (CDS) and social semiotics, specifically Wodak’s (2016) discourse historical approach (DHA) and Kress and van Leuween’s (2006) visual grammar. Additionally, I attempted to summarise reasons for radicalisation based on past scholarly work which include finding solutions to personal, social and economic crises (see King & Taylor, 2011; el-Said and Barrett, 2017). A detailed qualitative analysis of the data leads to a taxonomy of semiotic elements essential for highlighting the meaning-making possibilities ­– they are actors, symbols, music and texts that I juxtapose with the existing radicalisation models. Relevant findings include the group positioning themselves as the victim of war as the result of coalition attacks, and what they expect their audiences to do (e.g. “terrorise” the enemies). Some of the arguments used to justify their actions include the use of intertexts (e.g. holy texts) and descriptions of real world events. Ultimately, this research aims to reduce the risk of radicalisation among vulnerable groups and inform public policy.

Published Apr. 23, 2019 1:21 PM - Last modified June 19, 2019 3:01 PM