Abstract
In her first book Whether to Kill, Stephanie Dornschneider applied the methodology of cognitive mapping analysis (CMA) to study the beliefs that differentiate violent from nonviolent individuals. Fluent in several languages, the author courageously spent 2 years recruiting individuals across violent and nonviolent activist groups in Germany (nonviolent were the Socialist German Student Union and Kommune 1; violent were the Red Army Faction and Bewegung 2. Juni) and Egypt (nonviolent were Muslim brotherhood; violent were al-Jama’at al-Islamiyya and al-Jihad). The author independently interviewed 27 individuals across these seven groups and constructed cognitive maps based on her own textual analysis from these interviews. The author developed a unique programming code in MATLAB to adapt CMA and derived four main assertions: (1) violence cannot be attributed to beliefs about Islam, (2) violent individuals are not motivated by economic deprivation (although nonviolent activists may be), (3) violent individuals are not motivated by beliefs about or access to violent groups (she found no difference between nonviolent individuals’ access to violent groups vs. violent individuals’ access), and (4) violent individuals do not suffer from mental illness such as hallucinations.
Dornschneider reviewed literature from a variety of fields such as international relations, criminology, political, and social psychology, including such theorists Scott Atran and Marc Sageman. However, she omitted research about group dynamics that could have explained how individuals loosely connected to violent groups are more prone to joining them (e.g., work from John Horgan and Clark McCauley). Dornschneider neglected important organizational psychology constructs such as group leadership, structure, norms, and goals in formulating her analysis. It seems the only criterion of mental illness the author addressed in assessing the mental health of participants was the presence of “hallucinations.” She concluded that hallucinations were not present in her subjects since identified State acts of aggression were documented and hence not imagined. However, according to the most recent Diagnostic and Statistical Manual which outlines numerous criteria associated with mental illness, there are many characteristics associated with a variety of personality disorders that could help explain what moves someone toward a path of violence.
Following her literature review, Dornschneider presented the benefits of CMA and then detailed the nature of her interviewee recruitment methods and sample characteristics. This ethnographic research across cultures and time is extraordinarily important. Furthermore, her efforts to gain access to both Muslim and non-Muslims as well as violent and nonviolent individuals fills an important gap in the literature for comparisons. One note of concern from her description of her methods, however, was in the lack of best practices in interview research as outlined in a 2015 issue of Studies of Conflict and Terrorism. Specifically, the author did not use a recorder to capture her interview responses and instead relied upon her own notes. In addition, it is unclear how the interviews were conducted since there was limited information in her method chapter as to the nature of her interview guides. Dornschneider also mentioned becoming close friends with some of her Egyptian participants and their families, which may lead readers to question the objectivity of her interview process, as other ethnographic researchers have cautioned against such relationships when research requires hypothesis testing (cf. Horgan). The remaining chapters of her book detail her meticulous methods in constructing the cognitive maps, which are based on the “actors” own explanations about why they decided to engage or not engage in political activism. Her treatment of the computational analysis she developed using custom MATLAB code is formidable. This book provides excellent methodological resources for anyone wishing to use such computational analysis for examining complex relationships among beliefs, collections of beliefs, and decisions to act.
In conclusion, Whether to Kill contributes to the literature on political violence in several significant ways. First, it provides a framework for CMA both in initial text analysis and in the computational MATLAB approaches. The author’s detailed Chapters 4–5 should serve as an excellent example of how to conduct these analyses. Her inclusion of both violent and nonviolent activists is crucial for the study of political extremism. Dornshneider’s work allows for comparison of individuals who share similar political beliefs but endorse markedly different ways to achieve them (via violence toward the target vs. nonviolent activism). Her description of her sample members and their groups’ histories in Chapters 2–3 could be published as stand-alone case studies, as they are rich with demographic detail. Finally, the most important contribution of this book is her acknowledgment of the multifaceted beliefs that lead individuals toward violence in the name of an ideological cause. She highlighted perceived aggression from the state as being a critical belief shared by both violent and nonviolent individuals in her sample. Her concluding chapter also pointed to conditions of economic deprivation (for nonviolent members), access to violent groups (for both violent and nonviolent), and a host of other microlevel beliefs. Although the author set out to discern differences among violent and nonviolent individuals’ beliefs, she identified more similarities than differences among their cognitive maps. However, this finding is predicated on the methods she used and restricted to her sample of 27 individuals across these comparison groups. Moreover, it is difficult to draw firm conclusions from this work, as the author did not make obvious how she achieved reliability across her coding scheme (i.e., she was the sole interviewer, coder, and interpreter of themes) nor demonstrate consistency across a large enough sample of relevant comparison groups (e.g., Muslims in Germany nor non-Muslims in Egypt were not considered). Thus, the primary contribution of Whether to Kill is the excellent and detailed treatment of CMA both in its textual analysis and in its computation forms. She deserves immense credit for the courage and determination displayed in her research effort.