14.1 Introduction
Radicalization of individuals to violent action or the support of violent action presents an almost impossibly complex problem for study. It can take place across any medium or set of media, on the scale of a purposeful global propaganda campaign in multiple languages or by accident via unrecorded one-on-one conversations. It can take place over the course of long periods of time, or in an afternoon. The consequences of radicalized individuals and institutions, however, are concrete and drastic. Radicalized individuals have committed attacks on public transit systems, in public and private buildings, and against peaceful gatherings of civilians worldwide. In the period from January 2011 to June 2015, a study found that there were 69 terrorism plots in Europe, North America, and Australia. Overall, 19 of those plots came to fruition. The study goes on to argue that 30 had a declared connection to the Islamic State, and that plots with an IS connection were twice as likely than plots not connected to IS to move from planning to practice [1]. Ingram [2] interprets this information to suggest that it is in part the persuasive efficacy of the IS media campaign that led to this increased likelihood of execution, while Hegghammer and Nesser suggest that undercover operatives themselves may have precipitated the higher likelihood of a plot moving beyond the planning stage [1]. The conventional wisdom put forward by Ingram, that propaganda contributes to radicalization, is pervasive throughout the literature on atrocities [3 ā5]. Understanding how messages that incite radicalization are composed, structured, and function to radicalize individuals such that they are more willing to further political agendas through violent, unidirectional means may help efforts to reduce the effectiveness of this kind of messaging.
In this study, we examine the writing produced by various Islamic State affiliates in English so as to disentangle the relationship between the writing and radicalization. Our examination of written propaganda focuses on their use of metaphors. We do so because as Steuter and Willis argue in their study, At War with Metaphor, āthe metaphors we use reflect and reflexively shape our thinking.ā They go further, indicating the kinds of metaphors typically used to radicalize; ā[t]he massacres and genocides that comprise our most painful historical moments are characterized by a persistent dehumanization of the enemy,ā and ā[l]anguage itself, in the way it invites us to understand both the enemy and ourselves, becomes a potential weaponā [6]. Most significantly, they state, ā[i]t is language, rather than logic, that summons us through its emotional affect to a war we can no longer justify.ā
In tracing the role of metaphor in the propaganda efforts of the Islamic State and its affiliates, our research bridges two bodies of theory. The first, conceptual metaphor theory [7], describes how language functions to communicate ideas by connecting a concept in one abstract domain via references to ideas in another more concrete domain. As an example, one might refer to an idea as ābrilliant,ā a connection transferring a feeling from the concrete domain of light and brightness to the abstract domain of inspiration. To supplement the perspective provided by Lakoff and Johnsonās work, we relied on the operationalization of their work by Gordon et. al. [8], which presents 14 abstract ontological categories covering the range of conceptual and linguistic metaphors and their correspondence to the annotation work of [9]. The second body of theory on which this study draws is the informationāmotivationābehavioral skills (IMB) model [10], which describes how to understand the causal pathways linking messaging and behavior. Developed in the public health domain, it first appeared in studies of peopleās adoption of proactive behavior in relation to the public health crises of smoking and HIV. Fisher and Fisher separated messaging into components of information, such as that condoms prevent the transmission of HIV, the reasons and motivation for why the information is significant, such as that HIV can be a fatal disease, and the behavioral skills needed to implement the information, such as how to properly use a condom.
To develop the connections between these two bodies of theory, we simultaneously apply a computational and qualitative research design that allows for us to assess the relationship between the use of conceptual metaphors and the information, motivation, and skills communicated by the documents in our research corpus. Our method involved manual, parallel, qualitative coding of a subset of documents from our jihadi corpus for both metaphor usage and focus on information, motivation, or behavioral skills. Comparing those two sets of manual annotations using a bag-of-words model [11, 12] showed connections between different types of metaphor and different elements in the IMB model. For example, to convey information, wr...
