A Department of Homeland Security Center of Excellence led by the University of Maryland

A consortium of researchers dedicated to improving the understanding of the human causes and consequences of terrorism

Al-Qaeda's Propaganda Decoded: A Psycholinguistic System for Detecting Variations in Terrorism Ideology


Al-Qaeda's Propaganda Decoded: A Psycholinguistic System for Detecting Variations in Terrorism Ideology

Abstract: 

We describe a novel hybrid method of content analysis that combines the speed of computerized text analysis with the contextual sensitivity of human raters, and apply it to speeches that were given by major leaders of Al-Qaeda (AQ)—both in its “core” Afghanistan/Pakistan region and its affiliate group in Iraq. The proposed “Ideology Extraction using Linguistic Extremization” (IELEX) categorization method has acceptable levels of inter-rater and test-retest reliabilities. The method uncovered subtle (and potentially non-conscious) differences in the emphases that Usama Bin Laden and Ayman Al-Zawahiri put on the various components of their ideological justification for terrorism. We show how these differences were independently recognized as the crux of the rift in AQ, based on documents that were confiscated in Abbottabad following Usama Bin Laden’s assassination. Additionally, several of the ideological discrepancies that we detected between AQ “core” and its Iraqi affiliate correspond to schisms that presumably led to the splintering of AQ Iraq and the rise of ISIS. We discuss IELEX’s capability to quantify a variety of grievance-based terrorist ideologies, along with its use towards more focused and efficient counter-terrorism and counter-messaging policies.

Publication Information

Full Citation: 

Cohen, Shuki J., Arie Kruglanski, Michele J. Gelfand, David Webber and Rohan Gunaratna. 2016. "Al-Qaeda's Propaganda Decoded: A Psycholinguistic System for Detecting Variations in Terrorism Ideology." Terrorism and Political Violence (May): 1-30. http://www.tandfonline.com/doi/abs/10.1080/09546553.2016.1165214

Publication URL: 
Visit Website