Al-Qaeda’s propaganda decoded: a psycholinguistic system for detecting variations in terrorism ideology
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 gro...
Authors: | ; ; ; ; |
---|---|
Format: | Electronic Article |
Language: | English |
Published: |
2018
|
In: |
Terrorism and political violence
Year: 2018, Volume: 30, Issue: 1, Pages: 142-171 |
Online Access: |
Volltext (Resolving-System) Volltext (Verlag) |
Journals Online & Print: | |
Check availability: | HBZ Gateway |
Keywords: |
Summary: | 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. |
---|---|
Item Description: | Gesehen am 21.11.2023 Published online: 09 May 2016 |
Physical Description: | Diagramme |
ISSN: | 1556-1836 |
DOI: | 10.1080/09546553.2016.1165214 |