Читать книгу Whether to Kill - Stephanie Dornschneider - Страница 7
ОглавлениеIntroduction
As a child, Najeh Ibrahim loved his president. “We all loved Nasser,” he recalls. “He emphasized our country.” However, Ibrahim’s positive attitude toward the leader of his country changed as he grew older. Observing waves of arrests of Muslim Brothers and other political opponents, he began to resent the state. “We were seeing them come out of prison with marks of torture.” When he was seventeen, Ibrahim founded a small group, which quickly spread all over Egypt and soon posed a serious threat to the state: al-Jamaʿat al-Islamiyya. In 1981, this group changed the history of the country: it participated in the assassination of President Anwar al-Sadat. Ibrahim was among the leaders who decided to kill the president. He says the decision was made to resist state repression, by “young and strong” men who had alternatives: “Of course I had an alternative. I am a doctor. Look at bin Laden: He is a millionaire but lives like a beggar.” Had the state not engaged in repression, Ibrahim believes, “Sadat would not have been dead.”
Ahmad Saif al-Islam Hassan al-Banna did not love his president as a child. His father was the founder of the Muslim Brotherhood and assassinated in 1949. When Saif al-Islam became politically active himself, he had bitter experiences. “I tried to oppose parliament twice,” he recalls his time in the People’s Assembly. “They threatened to kill me …. They also threatened my family.” He says he refused to give in and went to court instead—but nothing happened. In spite of such experiences, he did not lose hope and continued to believe that the Muslim Brothers’ participation in politics could change the state. “It is better to succeed. Someone else replaced me,” he comments on his forced withdrawal from elections. “I left, and I understand it is not only me who is treated like that.” Saif al-Islam says he never considered the use of physical force to confront the state: “I will not use violence. I am a judge, and I studied law. My mind does not accept a violation of the law. If I use violence, I will lose. The state will kill us all. Now the state has no reason to do anything against us.”
The difference in the behavior of Saif al-Islam and Najeh Ibrahim could not be larger, but the two individuals have some basic commonalities. They believe in the same religion; they lived in the same country during the same time; and they resented the government enough to become politically active against it—even though this exposed them and their families to state repression. These similarities make the difference in the behavior of Saif al-Islam and Najeh Ibrahim puzzling and raise the following research question: Why do some individuals (like Najeh Ibrahim) take up arms, while others who live under the same conditions (like Saif al-Islam) conduct nonviolent activities instead?
This book is dedicated to investigating this question. Focusing on the individuals who take up arms, this question explores areas that may have been overlooked by the large body of literature on violent groups. Specifically, focusing on groups cannot explain why certain individuals but not others form or join violent groups, carry out particular attacks, or sometimes break away from their groups.
The central argument in my investigation of this question is that, contrary to widespread assumptions, both violent and nonviolent individuals act in response to the belief that the state is aggressive. Unlike what is widely believed, I also find that violent individuals do not act in response to beliefs in Islam. Instead, I argue that the motivations of violent and nonviolent individuals are surprisingly similar, and that there are no significant differences between Muslims and non-Muslims. Specifically, my analysis identifies ten mechanisms related to decisions to take up arms (five mechanisms) and to engage in nonviolent activism (five mechanisms).
These mechanisms show that the belief that the state is aggressive is so significant that individuals who hold it may decide to take up arms even if they also believe the state is stronger than they are, and that they will suffer severe consequences from engaging in violence. They also show that the belief that the state is aggressive may encourage individuals to decide to engage in nonviolent activism, even though they do not believe their activity is going to have any effect on state aggression. The analysis furthermore investigates when the individuals would not have decided to take up arms or to engage in nonviolent activism. This counterfactual analysis shows that in the absence of beliefs about threatening state behavior, no individuals would have decided to take up arms, and significantly fewer individuals would have decided to engage in nonviolent activism. By contrast, absence of beliefs about Islam would not have changed the individuals’ decisions. The analysis also shows that nonviolent individuals may be motivated by the belief that there is economic deprivation in their direct environment—a motivation that is usually attributed to violent individuals.
These findings have implications for our understanding of violent individuals by showing that rather than being crazy religious fanatics, violent individuals are not that different from nonviolent individuals and engage in very similar reasoning processes to those underlying mainstream political behavior. They also put in perspective existing explanations by suggesting that political violence is not a consequence of Islam, economic deprivation, or access to violent groups.
To develop this argument, this book draws on political psychology literature and adopts a cognitive mapping approach (CMA) (Axelrod 1976). To my knowledge, this is the first study that adopts this approach to study violent individuals. Cognitive mapping explores the belief systems underlying human behavior, such as violent and nonviolent activism. In this way, it bridges the gap between actors and structures and adds to theories that focus on external factors, which cannot by themselves explain why people engage in certain behavior. Moreover, cognitive mapping allows the systematic exploration of various types of factors underlying behavior—modeled as beliefs—as well as the study of the mechanisms by which these factors are connected with behavior—modeled as systems of beliefs. In this way, the CMA synthesizes factors that are usually addressed by different theories, and it goes beyond analyses that focus on direct relations between particular variables and behavior, rather than on the microlevel mechanisms underlying this behavior.
Taking the actors’ own explanations as the starting point of the analysis, the CMA provides rich inside knowledge into behavior, which cannot be obtained from other methods that involve external research categories. Cognitive maps do not a priori consider certain factors at the expense of others; instead, they cover a large range of factors that the actors themselves consider relevant. These inside factors, which offer a rigorous basis for a bottom-up analysis of human behavior, can then be analyzed by the researcher.
Cognitive maps are usually highly complex and difficult to analyze: typically, they consist of dozens of beliefs and connections between beliefs, which are related to decisions for action. Because of this complexity, it is not obvious how to systematically analyze such maps, and most political scientists have abandoned the approach—even though cognitive mapping used to be considered a “valuable tool” that “has been used successfully” (Young 1996: 395).
This book reintroduces the CMA to studies of political science by presenting new possibilities for research with cognitive maps. To cope with the complexity of cognitive maps, I present a computer program I developed with Nick Henderson from the Institute of Mathematical and Computational Engineering at Stanford University. The program is nonstatistical and enables the researcher to systematically study the connections between beliefs and decisions. Specifically, it enables the researcher to (1) systematically identify beliefs connected to decisions to take up arms or engage in nonviolent activism; (2) systematically trace belief chains connected to these decisions; and (3) explore counterfactuals, which show under what conditions individuals would not have decided to take up arms or engage in nonviolent activism. The program is applied to cognitive maps that involve trillions of combinations of beliefs.
To construct cognitive maps, this book applies qualitative methods. Specifically, I conducted ethnographic interviews with formerly violent and nonviolent individuals, and employed James Spradley’s theme analysis (1979) to develop a coding scheme that abstracts the beliefs of different individuals into comparable categories. In this way, this book contributes in-depth knowledge about political violence. It also contributes one of the few studies constructing cognitive maps from ethnographic interviews, which present new information that is very difficult to gather. Based on this information, it provides rich insight that both complements and serves as a check on the large body of literature on political violence that focuses on macrolevel factors without engaging with the actors’ own explanations. As I discuss below, the findings obtained from the analysis put into perspective much of the existing research in political violence.
The research design is a double-paired comparison that includes important control groups that remain absent from most existing studies: violent and nonviolent individuals, as well as Muslims and non-Muslims. To investigate these individuals, this study focuses on two countries, Egypt and Germany. Egypt is an authoritarian state located in the Middle East—a region with a long history of political violence. Over the past decades, Egypt has experienced numerous acts of violence, and Egyptians continue to play an important role in violent groups abroad (for example, the current leader of al-Qaeda, Ayman al-Zawahiri, is an Egyptian). A hub of political activism, Egypt is moreover the home of the Muslim Brotherhood—the largest opposition movement in the Middle East and one of the most influential Muslim movements in the world. Most of the individuals who participated in violent and nonviolent activism in Egypt have been Muslim, which indicates that Islam plays an important role in the country. Egypt has a population with a large Muslim majority and a long history of Islamic thought. It has been the home of numerous influential Muslim thinkers (Ibn Khaldun and Muhammad Rashid Rida, for instance, died there), and it houses the greatest authority in Sunni Islam, al-Azhar University. As the most populous state in its region, Egypt also plays a major role in Arab politics. It houses the Arab League and has frequently led talks between Arab states, or between Israel and its neighbors. Finally, Egypt is a rather poor state. In 2013, it ranked number 110 of 187 countries and territories in the UNDP Human Development Index. It has high rates of unemployment and illiteracy.
By contrast, Germany is a democratic state. Since its system was designed very carefully after the fall of the Nazis, it is known as a model of a modern democracy. As the largest economy in Europe, it has played an important role in regional politics, and its people enjoy one of the highest living standards in the world. Nevertheless, Germany has also experienced political violence, and it has been a major setting of political activism in Europe. Unlike their Egyptian counterparts, the individuals who participated in these activities were not Muslim and did not try to introduce an Islamic state. What they had in common was the goal of expelling their government. As a result of these characteristics, violent and nonviolent individuals from Germany can serve as a control group, checking on the findings obtained for the individuals from Egypt. Specifically, studying violent and nonviolent individuals from Germany ensures that results are not subject to individuals who believe in Islam and live in a state that is authoritarian or suffers from economic hardship.
Most of the individuals I study in this book were living in hiding or persecuted by the government at the time I conducted field research for this study (2009–2010). Specifically, the formerly violent individuals I met in Egypt are members of al-Jamaʿat al-Islamiyya and al-Jihad. These groups are responsible for numerous attacks on the Egyptian state, including the assassination of President Sadat in 1981. After the assassination, most of the members of these groups were imprisoned, but some were released a few years later. Some of them, including Ayman al-Zawahiri, left Egypt to unite with other individuals abroad, where they helped form al-Qaeda. Most of them, however, spent decades in Egyptian prisons, where they were often tortured and isolated from their families for extended periods of time. At the end of the 1990s, al-Jamaʿat al-Islamiyya announced an initiative to end violence, which encouraged the state to begin releasing its members from prison. At the time I conducted research for this book, some of them had just been released, which allowed me to contact them in private and try to set up meetings. Although they were usually highly reluctant to talk with me, some agreed to meet with me and, after many hours together, let me ask questions about why they had decided to take up arms.
The nonviolent individuals I interviewed in Egypt are members of the Muslim Brotherhood. At the time I conducted field research, President Husni Mubarak was still in power and the Muslim Brotherhood was his largest opposition. The movement was outlawed and its members were persecuted, although they were allowed to run in the elections as independents. Most of the nonviolent individuals I met had spent time in prison and experienced torture as well. All of them believed they could be arrested and imprisoned because of their resistance to the state. As a result, many of them were difficult to locate and highly reluctant to speak with me.
In Germany, I interviewed formerly violent individuals from the Red Army Faction and Bewegung 2. Juni. The Red Army Faction was the most violent or ga ni za tion in Germany after World War II and responsible for the killing of dozens of people between 1971 and 1993. Bewegung 2. Juni was active during the same period, but since it killed only two people, it received much less attention. Although they are no longer persecuted by the state, the members of these groups receive a lot of media attention, especially related to attacks that remain unresolved (for example, the assassination of German attorney general Siegfried Buback). These individuals were also very difficult to locate, and highly reluctant to speak with me. Dozens refused to be interviewed for this study, but those who agreed gave me the chance to ask questions about their actions for many hours.
The nonviolent individuals I interviewed in Germany are from the Socialist German Student Union and Kommune 1. These groups developed as part of the worldwide student revolts that broke out at the end of the 1960s, and became the major drivers of these protests in Germany. While many of these individuals were under surveillance during and after the student revolts, they were living ordinary lives at the time I conducted field research. Most were even mentioned in the phone book, which made it easy to locate and establish contact with them. With the exception of one individual, none asked to remain anonymous.
Before elaborating on the analysis of these individuals, the following pages further introduce the subject of this book: they critique the existing literature on political violence, develop several working hypotheses that serve as an analytical framework for this study, and define violent and nonviolent activism.
Studying Beliefs Related to Political Violence
This book contributes to the study of beliefs related to political behavior. Beliefs can be considered “the set of lenses through which information concerning the physical and social environment is received” (Holsti 1962: 245), and have been a powerful tool to understand political behavior. Focusing on the actors’ perspective on the world, beliefs contribute to studies of structural factors, which cannot by themselves explain human behavior (see Young 1996: 395). As Jonathan Renshon observes, “In the context of political decision making, leaders react not to an objective reality but to a subjective reality that is filtered through their belief system” (Renshon 2008: 822, my italics).
The study of beliefs originated in the field of foreign policy analysis (George 1969; Holsti 1962; Leites 1953). Although it has spread to other fields including public opinion or voting behavior (Caplan 2007; Page and Shapiro 1992), there are only a few studies about the beliefs connected to political violence.1 Due to this gap, we have very little knowledge about the subjective reality to which violent individuals react, and about whether the external categories presented by existing studies of political violence actually matter from the actors’ own perspective. This gap not only prevents us from understanding the rich microlevel mechanisms that drive the people who engage in political violence, but also from comparing the motivations underlying political violence as opposed to other types of behavior, such as nonviolent activism.
This book explores the beliefs of violent individuals and compares them with the beliefs of nonviolent individuals. This exploration suggests that, unlike what is widely assumed, the motivations underlying extremist behavior such as violence and mainstream behavior such as nonviolent activism appear to be rather similar. Specifically, I find that both violent and nonviolent individuals act in response to particular beliefs about aggressive state behavior. This implies that rather than being crazy religious fanatics, violent individuals are not that different from nonviolent individuals and engage in very similar reasoning processes.
These findings put into perspective existing explanations of political violence that apply external categories of research. In the following paragraphs, I elaborate on these theories and explain how this book contributes to them by investigating the beliefs of violent individuals. Rather than offering a comprehensive overview of the vast literature on political violence, I focus on theories that continue being examined in the major political science publications and that speak to the belief systems literature by implying that violent individuals hold certain beliefs (although they do not explicitly explore the subjective reality of violent individuals). I proceed by providing brief introductions to each theory, commenting on its limits, and explaining the contribution of this book.
Cultural-Psychological Theories
Cultural-psychological theories are arguably the most famous type of theories about the beliefs of violent individuals. Focusing on religion, they imply that violent individuals are motivated by religious beliefs.
In “The Roots of Muslim Rage,” Bernard Lewis (1990) argues that “Islam, like other religions has also known periods when it inspired in some of its followers a mood of hatred and violence”—an argument Samuel Huntington later developed into the theory of the “clash of civilizations” (1993). In spite of their publicity, cultural-psychological theories have been widely disputed. On the one hand, the claim that religion matters to violence has been supported. As Jessica Stern has pointed out, violent individuals are “spiritually intoxicated” and engage in violence “to ‘cleanse’ the world of ‘impurities’” (2003: 281; cf. Juergensmeyer 2009).
On the other hand, recent findings have questioned the role of religious beliefs. “Large-scale political violence is not disproportionately common or deadly in Muslim lands,” Fish, Jensenius, and Michel (2010: 1342) conclude in a statistical analysis building on Monty Marshall’s Major Episodes of Political Violence dataset. At the same time, the authors note that this finding does not provide information about the motivations of individuals (1342). In a statistical analysis of two distinct datasets and an independent sample of Muslims and Jews, Canetti et al. (2010) have explored these motivations. They find that religion does not directly encourage individuals to take up arms, and that “the relationship between religion and political violence only holds true when mediated by deprivations and psychological resource loss” (575).
While these findings contribute insight, we “still lack much hard evidence on whether a relationship between Islam and political violence really exists” (Fish et al. 2010: 1327). More specifically, we cannot explain why only some people who believe in Islam take up arms, or why the majority of Muslims do not take up arms: As Canetti et al. criticize, “on the individual level, existing empirical accounts are both sparse and conflicting” (575).
This book contributes new empirical insight at the individual level, and suggests that the main assumption of cultural-psychological theories is incorrect: investigating the beliefs of violent individuals who are Muslim, my analysis shows that they are not motivated by religious beliefs, but rather by other beliefs about state aggression. By contrast, it shows that beliefs about Islam may help encourage individuals to engage in nonviolent activism instead. The analysis adds analytical rigor to existing studies in the field of cultural-psychological theories by exploring the beliefs of both violent and nonviolent Muslims, and of violent and nonviolent non-Muslims. Moreover, it explores belief systems, which sheds light on the interrelationships by which different types of factors may motivate actors to engage in violence, building on Canetti et al.’s observation that the motivations underlying violence are complex.
Environmental-Psychological Theories
Environmental-psychological theories focus on the environment in which violence occurs. Much of this literature focuses on economic deprivation, implying that violent individuals are motivated by beliefs that they suffer from environmental strains, such as economic hardships.
Early research in this field suggests that people who take up arms against their states are motivated by feelings of frustration related to perceived deprivation and that “rebellions come to be when people cannot bear the misery of their lot” (Victoroff 2005: 19). Building on John Dollard et al.’s frustration-aggression hypothesis (1939) and Alexis de Tocqueville’s work on people’s dissatisfaction with their living situation ([1835] 2000), Ted Robert Gurr argues that “the proposed relation between perceived deprivation and the frustration concept in frustration-anger-aggression theory … provides a rationale for a more general definition of magnitude of violence and a more precise specification of what it comprises” (2011: 9).
Related recent analyses have presented an ambiguous picture. “Economic conditions and education are largely unrelated to participation in, and support for, terrorism,” according to a statistical analysis of the determinants of participation in Hezbollah activities in Lebanon (Krueger and Maleckova 2002, 2003). The authors conclude that “having a living standard above the poverty line or a secondary school or higher education is positively associated with participation in Hezbollah” and “that Israeli Jewish settlers who attacked Palestinians in the West Bank in the early 1980s were overwhelmingly from high-paying occupations” (2002: abstract). In their 2011 statistical analysis of surveys from Afghanistan, Iraq, and the Philippines, Berman et al. reach a similar conclusion: “The data rule out a positive correlation between unemployment and violence for all three countries: if there is an opportunity-cost effect, it is not dominant in any of them” (2011: 498; cf. Mousseau 2011; Blair et al. 2013). On the other hand, Meierrieks and Gries (2012) have provided new support for environmental-psychological explanations (2012). Based on an analysis of panel data for 160 countries from 1970 to 2007, they show that there is a “causal relationship between terrorism and growth” that is “heterogeneous over time and across space.” They conclude: “Growth unidirectionally Granger-causes terrorism for the Cold War era, while terrorism unilaterally Granger-causes growth for the post-Cold War era” (102, my italics). Similarly, James Piazza’s analysis of data from the Minority at Risk Project concludes that economic deprivation matters: “The empirical results show that countries that permit their minority communities to be afflicted by economic discrimination make themselves more vulnerable to domestic terrorism in a substantive way” (2011: 350).
Although not refuting the frustration-aggression hypothesis, these results suggest that its significance is limited, and that the relationship between environmental strains and violence remains unclear. In the words of Piazza, environmental-psychological explanations remain “inconclusive” (2011: 339). A major limitation is that environmental-psychological theories are rather static, implying that if environmental conditions change, the frequency with which political violence occurs will change as well. Such a linear causal relation can be refuted with reference to observations that the majority of people living in poverty do not take up arms (Victoroff 2005), that individuals as well as groups have embraced or renounced violence when environmental conditions stayed the same (Ashour 2007; Wickham 2002), or that individuals of the same group came from different environments (Ansari 1984; Ibrahim 1980).
This book shows that, as opposed to what is expected from environmental-psychological theories, violent individuals are not motivated by beliefs about environmental conditions, such as economic deprivation. Investigating the beliefs of violent individuals from two opposite environments—an authoritarian state suffering from economic hardship (Egypt) and a wealthy democracy (Germany)—the analysis shows that individuals instead decide to take up arms in response to the belief that the state is aggressive. By contrast, it shows that, surprisingly, individuals who engage in nonviolent activism in the same environments may be motivated by beliefs about economic deprivation.
Group Theories
Much of the recent research has examined the role of violent groups. This research emphasizes that individuals usually engage in violence against the state as part of groups, rather than as lone-wolf actors. This implies that violent individuals are motivated by beliefs about their groups.
Although much of this literature focuses on groups rather than individuals, some analysts in this field have also explored how groups may motivate individuals to turn to violence. For example, they have pointed to sociopsychological mechanisms according to which individuals first get in contact and are then absorbed by violent groups. Based on Marc Sageman’s famous analysis of the biographies of 172 individuals, for instance, it has been suggested that “terrorism is an emergent quality of the social networks formed by alienated young men who become transformed into fanatics yearning for martyrdom and eager to kill” (2004: vii). Albert Bandura has specified the psychological aspect of this theory by pointing to “mechanisms of moral disengagement” in which violent individuals abandon the “moral standards” adopted during their earlier “course of socialization” and begin to “refute” moral arguments, referring to “a higher level of morality, derived from communal concerns” (Bandura 1998: 161, 165).
More recently, Omar McDoom’s statistical analysis of 3,426 residents of a Rwanda community has provided support for the social aspect of this theory by showing that violent individuals “are likely to live either in the same neighbourhood or in the same household as other participants” (2013: 1). “Specifically, as the number of violent to nonviolent individuals in an individual’s neighbourhood or household increases, the likelihood of this individual’s participation also increases.” Three studies by Scott Atran and Jeremy Ginges on Palestinians and Jews have further specified psychological aspects according to which violent individuals “are motivated by moral commitments to collective sacred values” (2009: 115).
The finding that violent groups matter is widely accepted. In spite of the insight gained from the mentioned works, focusing on groups cannot explain the behavior of the individuals participating in those groups. Specifically, group theories leave open the questions of why certain individuals join violent groups as opposed to others who have access to the same groups; why certain but not other members of violent groups carry out particular attacks; or why certain members sometimes break away from their violent groups. Often, group theories also fail to consider that nonviolent groups may be interacting with the same individuals, operating in the same environment, and against the same targets.
As opposed to what is expected from group theories, this book shows that violent individuals are not motivated by beliefs about violent groups. Specifically, I find that the beliefs violent individuals hold about their groups do not matter to their decisions to take up arms. Moreover, I find that, surprisingly, nonviolent individuals may hold the same beliefs about violent groups. In particular, the nonviolent individuals I interviewed also hold beliefs that they interacted with members of violent groups. This indicates that they also had group access, but as opposed to what is expected from group theories, this did not motivate them to take up arms.
Psychopathological Theories
Although psychopathological theories have been widely discarded, they deserve attention because they continue to be quoted by numerous analysts of political violence and by the news. Specifically, psychopathological theories have been dedicated to searching for a “terrorist personality.” They have also argued that violent individuals suffer from mental illness. This implies that violent individuals are motivated by beliefs that do not correspond to reality (mental illnesses) or beliefs related to their personality (“terrorist personality”).
Theodor Adorno et al.’s The Authoritarian Personality, written in the aftermath of World War II, presents one of the first contributions to this field. Focusing on “the potentially fascistic individual,” the book explores the hypothesis “that the political, economic, and social convictions of an individual often form a broad and coherent pattern, as if bound together by a ‘mentality’ or ‘spirit,’ and that this pattern is an expression of deep-lying trends in his personality” (1950: 1). Lloyd Etheredge’s “Hardball Politics: A Model” (1979) builds on Adorno’s theory and has inspired numerous additional works.2 Specifically, Etheredge explores “the syndrome” of “hardball politics” where “tough, ambitious, shrewdly calculating men vie for power and status behind a public veneer of civilization and idealistic concern” (3). Treating political behavior as the result of personality structures, Etheredge presents “hardball politics as a subculture … constructed and sustained by a particular personality type, men with what is known clinically as a narcissistic personality disorder” (1979: 3, my italics).
In spite of their contribution and publicity, psychopathological theories have been widely discarded. As Crenshaw (1981: 390) notes, “the outstanding common characteristic of terrorists is their normality.” Moreover, Post has observed that “research on the psychology of terrorists does not reveal major psychopathology” (2004: 128). And Silke has pointed out that “the findings supporting the pathology model are rare and generally of poor quality” (1998: 51).
This book offers new evidence that violent individuals do not suffer from mental illness and that there appears to be no “terrorist personality.” Specifically, my study of the beliefs of violent individuals does not identify any beliefs that do not correspond to the external world, which would indicate that they suffer from illnesses, such as hallucinations. My study also identifies very few beliefs about the individuals’ personality, and it finds that these do not matter to decisions to take up arms. These findings point to the limits of psychopathological theories, which continue to enjoy much publicity.
Analytical Framework
Analyzing the beliefs of violent individuals provides an inside perspective that identifies the limits of existing theories on political violence that focus on external factors such as poverty, or access to violent groups, or on particular beliefs related to Islam or mental illness. Specifically, it shows that violent individuals are not motivated by the main factors that are assumed to explain violence by these theories. Rather, the analysis shows that violent individuals primarily act in response to beliefs about aggressive state behavior. It also suggests that the motivations of violent and nonviolent individuals are surprisingly similar, and that violent individuals engage in reasoning processes that strongly resemble those underlying mainstream political behavior. These findings indicate that analyzing the beliefs of political actors can contribute indepth knowledge that is not available from analyses of external research categories.
By adopting the CMA, this book promises to synthesize the existing literature. Specifically, belief systems represented by cognitive maps consist of beliefs about various types of factors, which are usually addressed by different theories. For example, one can hold beliefs about religious norms like God forbids the killing of innocent people (cultural-psychological theories); about economic conditions like poverty (environmental-psychological theories); or about interacting with violent groups like meeting members of al-Qaeda (group theories).
Cognitive maps can moreover investigate the mechanisms3 by which these factors encourage individuals to take up arms or to engage in nonviolent activism instead. As I explain in Chapter 1, it can do so because beliefs are by nature connected in a systematic way. Because of this, the CMA promises to provide new knowledge about the complex interrelationships between different types of factors connected to violence, adding to other methods such as statistical analyses that focus more on the direct relationships between particular variables and behavior.
Exploring various types of factors, the CMA can shed light on the relevance of existing theories. The following study is guided by an analytical framework of four hypotheses about the main factors addressed by these theories, formulated as beliefs. To formulate the hypotheses, I revisited the theories discussed above and focused on the main factors addressed by each of them. I then related those factors to the beliefs of violent and nonviolent individuals. Table 1 gives an overview.
These hypotheses serve as an analytical framework to guide the following study. Specifically, they allow me to evaluate the importance of the main types of factors that have been found related to political violence by existing research. This serves to reassess the relevance of existing theories on political violence. More specifically, it shows whether, as assumed by cultural-psychological theories, violent individuals are motivated by Islam; whether, as assumed by environmental-psychological theories, violent individuals are motivated by environmental conditions; whether, as assumed by group theories, violent individuals are motivated by access to violent groups; or whether, as assumed by psychological theories, violent individuals are motivated by mental illness or personality.
Table 1: Analytical Framework
Defining Political Violence and Nonviolent Activism
Political Violence
Defining political violence is subject to various challenges. From an empirical perspective, there is “a vast array of types of violence” (Collins 2008: 1), and from a theoretical perspective violence can be considered “a conceptual minefield” (Kalyvas 2013: 19).4 To gain clarity, I identify four features of political violence agreed on by most analysts.
First, the most significant characteristic is that political violence involves “physical force” (della Porta 1995: 2; Tilly 1978: 176). While Gurr (2001) mentions “the use or threat” of force, the majority of works presented above deal with the application of physical force, connected with the occurrence of physical harm. The harm addressed by the literature is mostly death, but it can in principle also include nonlethal physical injuries or the destruction or damaging of objects.
Based on these considerations, I treat political violence as behavior that applies physical force and that is connected with the occurrence of physical harm. As Tilly observes, such a focus establishes a relatively “narrow” definition of political violence and ensures “at least a chance to sort out the regularities in the appearance” of it (1978: 174). By contrast, a broader definition of political violence that goes beyond the application of physical force “to include all violations of human rights not only requires agreement on the character of those rights, but also expands the phenomenon to such a large range of social relations as to make systematic study of it almost unthinkable.”
Second, much of the literature I have discussed relates physical force to civil perpetrators. This confirms the continuing relevance of the original use of the term political violence, which occurred in the 1960s (Gurr 2001) and only later started to include other perpetrators, such as the state, or no perpetrators, as in “structural violence” (Galtung 1990).
In spite of their common focus on civil perpetrators, the literature discussed above deals with these perpetrators from two different angles—groups and individuals. Given my attempt to differentiate individuals who take up arms from those who engage in nonviolent activism instead, this book emphasizes the role of individuals rather than collectives: focusing on groups does not allow a differentiation of the individuals who form, run, or break away from their groups. This does not mean, however, that I think political violence cannot be considered to belong to “the repertoires of collective actions” (della Porta 1995: 2; cf. Tilly 2003). Rather, my focus adds to existing studies by drawing attention to the individuals who are responsible for the exercise of physical force, although groups or larger collectives plan and claim responsibility for it.
Third, as suggested by their common focus on perpetrators, the majority of the works discussed above share a focus on a particular target: the state, defined as a polity that has a government that holds a monopoly of legitimate use of force over a certain territory (Weber 1992). State targets may be people who directly represent the government, such as the head of state, the prime minister, or other ministers. They may also be people who have especially close ties to the government, such as leading businessmen. They may moreover be people who are represented by governments, that is, citizens of the state, or people whose existence is closely connected to that of the government, such as tourists in states where tourism is a major source of income.5 Finally, state targets can be objects owned by the state, such as government buildings and highways. Some examples of political violence involving state targets are the assassination of President Sadat during a military parade in 1981; the murder of German police officer Franz Sippel when he wanted to check the passports of members of the Red Army Faction (1976); the kidnapping and shooting of the chairman of the German Employers’ Organization, Hanns-Martin Schleyer (1977); the kidnapping of the candidate for the mayoral elections in Berlin, Peter Lorenz (1975); or the attack on the building of the Law School in Berlin (1972).
Regardless of its theoretical background, most of the literature is dedicated to physical force applied against civilians living in a state. In spite of extensive criticism related to the term, this is usually referred to as “terrorism.” For the purpose of this study, it is important to emphasize that much of what is called terrorism goes beyond physical force targeting the state (although physical force targeting the state may be called “terrorism”). Quite frequently, for instance, “terrorism” is used to describe physical force targeting particular religious groups. For example, consider the attack on Copts, a religious minority in Egypt, in “The Saints Church” in Alexandria during the Coptic Christmas mass at the beginning of 2011. While the attack has been called “terrorism,” the religious background of the victims (Coptic), the location of the attack (a church), and its timing (during Christmas mass) suggest that the state was not the primary target. Rather, they suggest that this attack is an act of political violence against the Coptic community. This is the case, although one might add that, by exposing the Egyptian state’s inability to prevent the catastrophe, the attack also hurt the state. On the other hand, the massacre of tourists at the Temple of Hatshepsut in 1997, which has also been called terrorism, may be considered an attack against the state: the targeting of tourists is an attack on the Egyptian tourism industry—a major source of state income. Moreover, the perpetrators had a history of targeting state representatives, during which they stated they were indeed targeting the state (Peters 2006).
It might be objected that focusing on political violence against the targets mentioned appears to be somewhat arbitrary and rather narrow. However, the definitions provided in this section establish the basis for a rigorous comparison of violent and nonviolent activism that happens in the same environment—and not for exploring all forms of political violence. Focusing on a wider range of targets (including nonstate targets, for example) would have made it much more difficult to conduct this comparison, even if both violent and nonviolent activism happened in the same environment.
Fourth, physical force is widely considered a means rather than an end, and there is a consensus that the people who exercise physical force do so for a purpose beyond the mere application of physical force.6 Since goals exist prior to the performance of actions, political violence appears to be a planned rather than spontaneous type of behavior.7 Goals also indicate that political violence may involve deliberation, including alternative means—for example, about forcing the repeal of a new law by planting a bomb inside the Ministry of Justice rather than by demonstrating in front of it.
In spite of a more or less general consensus that political violence includes goals, the literature differs on the nature of these goals: cultural-psychological theories assume the goals are based on religious beliefs, and in particular on Islam. Environmental-psychological theories suggest that the goals are based on economic hardship and express people’s desire to improve their living conditions. Focusing on violent groups that absorb individuals who establish contact with them, group theories seem to equate goals with groups, suggesting that the group itself can be a goal. Finally, psychopathological theories imply that goals are subject to the personality of individuals.
This study supports the general consensus that political violence involves certain goals. Since the cognitive mapping approach does not specify particular factors in advance, I do not formulate particular types of goals at this stage. Rather, my construction of data provides insight into the goals related to political violence (see Chapter 4).
Nonviolent Activism and Borderline Behavior
At first sight, the literature discussed above offers little ground to develop a definition of nonviolent activism. This is due to its more general neglect of the absence of political violence8 as well as to the absence of control groups in existing studies of violence. To cope with this difficulty, I draw on the definition of political violence to define nonviolent activism.
Nonviolent activism is a behavior that involves a means that is not physical force. For examples of such activities, Gene Sharp’s famous article on nonviolence (1959) serves as a valuable source.9 In particular, Sharp mentions boycotts, strikes, or noncooperation movements exercised as passive or peaceful resistance in the context of conflicts with the goal of “achieving or thwarting of social, economic, or political change” (53). More specifically, he cites Mohandas Gandhi’s resistance campaign; the Montgomery, Alabama, Negro bus boycott in 1955–1956; the 1952 “Defy Unjust Laws” campaign in South Africa; and the 1942 Norwegian teachers’ resistance to Nazi use of Norwegian schools for indoctrination, which was among the “most important actions in halting Quisling’s plans for instituting the Corporate State in Norway.”
Further drawing on my definition of political violence, I consider nonviolent activism to involve civilian perpetrators and state targets. Such a common focus provides a rigorous basis for comparison—although it limits a wider exploration of violent and nonviolent activism, which may involve noncivil perpetrators and nonstate targets. As mentioned, it is nevertheless much more difficult to compare violent and nonviolent activism conducted by different perpetrators against different targets. It is also important to note that violent and nonviolent activism that involves civil perpetrators and state targets does occur; and that focusing on exactly these activities provides an especially rigorous basis to systematically differentiate between the two.
Following my definition of political violence, I further consider nonviolent activism to involve goals. By definition, violent and nonviolent activism might involve the same goals. However, nonviolent activism is distinguishable from political violence by its means, which are not physical force. As a result, the goals of nonviolent activism are nevertheless related to the rejection of physical force and the embrace of other means instead. This is expressed by Gandhi: “Non-violence is the greatest force at the disposal of mankind. It is mightier than the mightiest weapon of destruction devised by the ingenuity of man” (Gandhi 2007: 23).
It is also interesting to note that, in this context, it is possible to differentiate nonviolent activism from peaceful activism: goals indicate that an activity is planned and is intended to exercise certain but not other means. By contrast, peaceful activism may be considered to not involve any goals and be unplanned or spontaneous. Peaceful activism may moreover be considered to involve goals without considering physical force as a means. In the following analysis, this aspect is addressed by coding the individuals’ direct speech into different types of categories representing peaceful and nonviolent activities (see Chapter 4).
Although the examples of nonviolent activism mentioned above do not involve physical force, the same cannot be said about other activities, which appear to be very similar: As Sharp observes, demonstrations or strikes, for instance, in principle involve nonviolent means—but their occurrence has often been connected with physical force.10 Some examples are the Egyptian revolution in 2011, the Greek protests against austerity measures in 2011, the strikes of textile workers in Mahalla, Egypt, in 2008, or the student revolts in Germany in 1968. In addition, there are activities whose goals involve physical force but whose application does not occur. Examples are the failure to detonate of a bomb intended to kill the hostages and hostage takers during the RAF siege of the German embassy in Stockholm in 1975, or the failed detonation of bombs placed on German trains in 2006.
Table 2: Examples of Borderline Activities (general)
| Failed application of physical force | Application of both violent and nonviolent means |
| Not pulling the trigger of a gun pointed at a target | Joining a protest aft er planting a bomb |
| Failed detonation of a bomb | Protestors hitting the police during a demonstration |
There are at least two types of borderline behavior in which either both nonviolent and violent means are applied or the application of a particular means fails. An example of the first is the throwing of stones at the police by protestors during a demonstration: the activity of demonstrating indicates an application of a nonviolent means, and the activity of throwing stones at the police indicates an additional activity that involves application of physical force. An example of the second type is a person’s refraining from pulling the trigger of a gun that is pointed at a policeman: in this case, there is no application of physical force, even though, by pointing the gun at the policeman, it can be said that physical force is considered a means. Table 2 gives an overview.
Based on these considerations, nonviolent activism may include a large range of activities. What they all have in common is that they (1) exclude physical force, (2) involve goals related to the rejection of physical force and the embrace of other means instead, (3) are conducted by civil perpetrators, and (4) target the state.
The Venn diagram in Figure 1 gives an overview of the concepts discussed in this section. Specifically, it identifies various types of activities as political violence, nonviolent activism, and borderline behavior. Note that although the second type of borderline behavior does not occur, it has been placed at the intersection of political violence and nonviolent activism, because it is identifiable as borderline behavior by negating the major components of violent and nonviolent activism. Table 3 provides particular examples of activities.
Figure 1. Venn diagram of political violence, borderline behavior, and nonviolent activism.
Table 3: Examples of Political Violence, Borderline Behavior, and Nonviolent Activism
| Political violence | Borderline behavior | Nonviolent activism |
| Hatschepsut Massacre, 1997 Assassination of President Sadat, 1981 Kidnapping and shooting of the chairman of the Confederation of German Employers’ Associations, Hanns-Martin Schleyer, 1977 Killing of a German police officer who wanted to check passports of members of the Red Army Faction, 1976 Attack on the building of the Berlin Law School, 1976 Kidnapping of the candidate for the mayoral elections in Berlin, Peter Lorenz, 1975 | Egyptian Revolution, 2011(violent and nonviolent activism) Greek protests against austerity measures, 2011(violent and nonviolent activism) Strike of Egyptian textile workers in Mahallah, April 2008(violent and nonviolent activism) German student protests, 1960s(violent and nonviolent activism) Failure of detonation of bombs placed on German trains, 2006 (none) Failure of detonation of a bomb placed in the main train station of Bonn, Germany, 2012 (none) | Gandhi’s resistance campaign Montgomery bus boycott in Alabama, 1955–1956 “Defy Unjust Laws” campaign in South Africa, 1952 Norwegian teachers’ resistance to the Nazis’ use of their schools for indoctrination, 1942 |
Outlook: Chapter Overview
Drawing on these definitions, the following chapters investigate the question why some individuals take up arms while others who live under the same conditions engage in nonviolent activism instead.
In Chapter 1, I introduce the CMA and show how it can be used to study political violence. I also introduce a formalization of cognitive maps that copes with the maps’ complexity: it allows the researcher to develop computational models processing the cognitive maps as directed acyclical graphs (DAGs). The formalization is based on Judea Pearl’s theory of causality (2000) and provides new possibilities for the application of the CMA. It also presents a new approach to studying counterfactuals. Specifically, it suggests how intervening on the actors’ beliefs about the world rather than on the world itself can explore their behavior in alternative worlds.
In Chapter 2, I introduce the individuals I interviewed for this study and describe several situations in which I interviewed individuals. I elaborate on the research design—a double-paired comparison of violent and nonviolent individuals from Egypt and Germany—and explain my identification strategy of individuals while conducting fieldwork. I also consider potential biases related to this research. Finally, I draw comparisons with other analyses about violent individuals from Egypt and Germany.
In Chapter 3, I describe the historical context in which the individuals decided to engage in violent or nonviolent activism. Specifically, I introduce their groups: the nonviolent Muslim Brotherhood and the violent al-Jihad and al-Jamaʿat al-Islamiyya in Egypt, and the violent Red Army Faction and Bewegung 2. Juni, as well as the nonviolent Kommune 1 and the Socialist German Student Union in Germany. I discuss the beginning, development, and, in some cases, dissolution of these groups with special emphasis on the 1970s, when most of the individuals I interviewed were active.
In Chapter 4, I describe how I constructed cognitive maps from the interviews and how I identified the three main components of cognitive maps from the direct speech of the individuals—beliefs, belief connections, and decisions. Since they are constructed from direct speech, these cognitive maps are not immediately comparable. Accordingly, I employed Spradley’s theme analysis to abstract the beliefs of different individuals into more general categories. I show how I constructed a coding scheme, which consists of several dozen categories. The scheme identifies five broad sets of factors related to political violence and speaks to the analytical framework presented in this Introduction. It shows that the reasoning processes of violent and nonviolent individuals are surprisingly similar, and that there are no significant differences between Muslims and non-Muslims who engage in violence. Rather, the reasoning processes of all individuals primarily rely on beliefs about the state environment. There are also extraordinarily few beliefs about violent groups and about the individuals’ personality. There is no evidence that violent individuals suffer from mental illnesses or act based on a “terrorist personality.”
Chapter 5 presents the computer program and analysis of the cognitive maps. In more than 100,000 runs, the analysis identifies ten mechanisms related to political violence (five mechanisms) and nonviolent activism (five mechanisms). These mechanisms show that both violent and nonviolent individuals act in self-defense by responding to the belief that the state is aggressive. More specifically, this belief is so significant that individuals who hold it may decide to take up arms even if they believe the state is stronger than they are, and that they will suffer severe consequences from engaging in violence. This belief may also encourage nonviolent individuals to make decisions, even if they do not believe their activity will have any effect on state aggression. Nonviolent individuals may further be motivated by the belief that there is economic deprivation in their direct environment—a motivation usually attributed to violent individuals.
In Chapter 6, I show how the program can be applied to study counterfactuals. In more than 100,000 runs, the analysis investigates alternative worlds in which violent individuals would not have taken up arms. Individuals would not have decided to take up arms had they not believed their state was aggressive, and the majority would not have decided to engage in nonviolent activism in the absence of beliefs about state aggression. By contrast, a few nonviolent individuals would have been motivated to make decisions in alternative worlds by the belief that their state was not religious. Confirming that Islam is not found to explain violence, no violent individual would have continued making a decision based on this belief.
In the Conclusion, I reconsider the results and comment on the utility of applying the cognitive mapping approach in the particular field of political violence and political science more generally. In addition, I share what I as a German woman experienced going to Egypt and conducting interviews with violent and nonviolent individuals. I also give a few personal reactions to returning to Egypt for this research, after having lived in Cairo as a student of political science and Middle East studies from 2004 to 2007. I end by commenting on the changed lives of the individuals in the aftermath of the Arab Spring.
