Evaluation of an Interactive System Dynamics Model

Understanding the Driving Factors of Terrorism A Major Qualifying Project Submitted to: Dr. James K. Doyle, Associate Professor, Worcester Polytechnic Institute Dr. Elise A. Weaver, Assistant Professor, Worcester Polytechnic Institute In partial fulfillment of the requirements for Degree of Bachelor of Science

______________ Bruce K. Skarin April 26, 2002

Abstract This project examines the problem of terrorism used against the Untied States by Osama bin Laden and the alQaida network. Extensive research was conducted on the main causal factors identified to develop a comprehensive background. Using the system dynamics methodology the information gathered was used to build an explanatory model that simulated the terrorist group and U.S. countermeasures to attacks. The model was then presented in several seminars and evaluated using surveys to discover its value as an education tool. In completion of the study a great deal of insight was revealed about the terrorist problem through the model. In addition, the seminars successfully demonstrated the potential of using a system dynamics for policy development and educational implementation.

ii

Table of Contents 1. Introduction _________________________________________________________ 1 2. Background Introduction ______________________________________________ 4 2.1. The Terrorist Group _____________________________________________________ 4 2.1.1. Defining Terrorism __________________________________________________________ 4 2.1.2. Psychology, Fundamentalism, and Regional History ________________________________ 7 2.1.2.1. Mental Models and Terrorism ______________________________________________ 7 2.1.2.1.1. The Origins of Violence and Violent Groups _______________________________ 8 2.1.2.1.2. The Psychology and Sociology of Terrorists, Groups, and Leaders_____________ 13 2.1.2.1.2.1. The Terrorist Mental Model _______________________________________ 13 2.1.2.1.2.2. Terrorist Groups ________________________________________________ 15 2.1.2.1.2.3. Terrorist Leadership _____________________________________________ 16 2.1.2.2. Fundamentalism ________________________________________________________ 17 2.1.2.2.1. Fundamentalism and Violence _________________________________________ 18 2.1.2.2.2. The Violent Fundamentalism of al-Qaida ________________________________ 20 2.1.2.3. The Culture of Islam, Jihad, and the Middle Eastern World ______________________ 21 2.1.2.3.1. Nationalism and Islam _______________________________________________ 21 2.1.2.3.2. The Making of an “Evil” United States __________________________________ 22 2.1.2.3.3. The Failures of Modernization _________________________________________ 23 2.1.3. Osama bin Laden and the al-Qaida Terrorist Network ______________________________ 25 2.1.3.1. The Terrorist Leader: Osama bin Laden______________________________________ 26 2.1.3.1.1. The Declarations of Osama bin Laden ___________________________________ 28 2.1.3.2. The Terrorist Network: Al-Qaida (The Base) _________________________________ 31 2.1.3.2.1.1. The Objectives of al-Qaida ________________________________________ 31 2.1.3.2.1.2. A Chronology of Significant Events_________________________________ 32 2.1.3.2.1.3. Numerical Data _________________________________________________ 34 2.1.4. Critical Flaws in a Terrorist’s Mental Model______________________________________ 35

2.2. The Opposition to Terrorism: The United States of America___________________ 38 2.2.1. U.S. Legal Definitions and Policies on Terrorism __________________________________ 38 2.2.2. The Driving Factors of U.S. Policy _____________________________________________ 41 2.2.2.1. Defensive Policies ______________________________________________________ 41 2.2.2.2. Offensive Policies ______________________________________________________ 42

2.3. Initial System Dynamics insight into Terrorism _____________________________ 43

3. Methodology Introduction _____________________________________________ 45 3.1. Research, Resources, and Processes _______________________________________ 45 3.1.1. Literature Reviews __________________________________________________________ 45 3.1.2. Modeling _________________________________________________________________ 45 3.1.3. Expert Consultation _________________________________________________________ 45 3.1.4. Forums ___________________________________________________________________ 46 3.1.5. Survey Database____________________________________________________________ 46 3.1.6. Internet Site _______________________________________________________________ 46

3.2. Modeling the Dynamics of Terrorism ______________________________________ 46 3.2.1. The Problem Statement and Reference Mode _____________________________________ 47 3.2.2. Dynamic Hypothesis and Base Structure _________________________________________ 48 3.2.2.1. The Terrorist Group _____________________________________________________ 49 3.2.2.2. Defensive Measures _____________________________________________________ 51 3.2.2.3. Offensive Measures _____________________________________________________ 53 3.2.2.4. Terrorist Anger_________________________________________________________ 55 3.2.3. Calibration and Simulated Data Fit _____________________________________________ 57

iii

3.3. Teaching the Driving Factors of Terrorism _________________________________ 58 3.3.1. Seminars on Terrorism_______________________________________________________ 59 3.3.2. Surveys and the Evaluation of Learning _________________________________________ 59 3.3.3. Analyzing the Survey Results _________________________________________________ 60

4. Analysis and Conclusions Introduction __________________________________ 61 4.1. Insights from the Model of Terrorism _____________________________________ 61 4.1.1. The Limitations of Offensive and Defensive Measures ______________________________ 62 4.1.1.1. Expected Success of the “War on Terrorism” _________________________________ 63 4.1.1.2. Potential Behavior of Unchanged Model _____________________________________ 64 4.1.1.3. Policy Alternatives ______________________________________________________ 66

4.2. System Dynamics as a Tool for Understanding Terrorism _____________________ 69 4.3. Final Conclusions and Recommendations __________________________________ 71 4.3.1. Areas for Additional Study ___________________________________________________ 71

References____________________________________________________________ 73 Appendix A Systems Thinking and System Dynamics________________________ 77 Appendix B Literature Keyword Searches & Databases Used _________________85 Appendix C Analysis of Barry Richmond’s “A Systems Thinking Look at Terrorism”________________________________________________87 Appendix D An Analysis of Ivan Taylor’s “Policy Instruments in the War on Terrorism”________________________________________________96 Appendix E Base Model Equations _______________________________________99 Appendix F Survey Example____________________________________________103

iv

List of Figures Figure 1 Group Violence Reinforcing Loops Causal Diagram ........................................ 12 Figure 2 Group Violence Reinforcing Loops Causal Diagram ........................................ 36 Figure 3 Terrorist Reinforcing Loops Causal Diagram .................................................... 37 Figure 4 Defensive Balancing Loop Causal Diagram ...................................................... 42 Figure 5 Offensive Balancing Loop Causal Diagram....................................................... 43 Figure 6 Terrorist Group, Defensive, and Offensive Causal Loop Diagram.................... 44 Figure 7 Weighted Casualties per Year Graph ................................................................. 48 Figure 8 Simplified Terrorist Group, Offensive, and Defensive Model........................... 49 Figure 9 Terrorist Group Model Sector ............................................................................ 50 Figure 10 Effect of Supporters Graphical Function.......................................................... 51 Figure 11 Defensive Model Sector ................................................................................... 52 Figure 12 Effect of Threat on Defense Graphical Function ............................................. 52 Figure 13 Effect of Defense Graphical Function.............................................................. 53 Figure 14 Offensive Model Sector.................................................................................... 54 Figure 15 Effect of Threat on Offense Graphical Function.............................................. 54 Figure 16 Effect of Offense Graphical Function .............................................................. 55 Figure 17 Anger Model Sector ......................................................................................... 56 Figure 18 Effect of Despair Graphical Function............................................................... 56 Figure 19 Effect of Anger Graphical Function ................................................................. 57 Figure 20 Simulated and Real Weighted Casualties per Year Graph............................... 58 Figure 21 Database Relationship Diagram ....................................................................... 60 Figure 22 Switched Off Anger Model Structure............................................................... 63 Figure 23 No Anger Simulation Results........................................................................... 64 Figure 24 No Changes Simulation Results ....................................................................... 65 Figure 25 No Changes Anger Problem............................................................................. 66 Figure 26 Switched Peace Making Policies Model Structure........................................... 67 Figure 27 Peace Making Policies Simulation Results ...................................................... 67 Figure 28 Eight Thinking Skills (Richmond, 2001) ......................................................... 77 Figure 29 Simple Decision Making Process (Sterman, 2000).......................................... 80 Figure 30 Decision Making Process (Sterman, 2000) ...................................................... 81 Figure 31 Decision Making Process with Virtual World (Sterman, 2000)...................... 82 Figure 32 Causal Loop Diagrams ..................................................................................... 84 Figure 33 Stock and Flow Symbols .................................................................................. 84 Figure 34 (Richmond, 2001)............................................................................................. 87 Figure 35: (Richmond, 2001)............................................................................................ 88 Figure 36 (Richmond, 2001)............................................................................................. 88 Figure 37 (Richmond, 2001)............................................................................................. 89 Figure 38 (Richmond, 2001)............................................................................................. 89 Figure 39 (Richmond, 2001)............................................................................................. 90 Figure 40 (Richmond, 2001)............................................................................................. 91 Figure 41: (Richmond, 2001)............................................................................................ 91 Figure 42: (Richmond, 2001)............................................................................................ 92 Figure 43: (Richmond, 2001)............................................................................................ 93 Figure 44: (Richmond, 2001)............................................................................................ 93 v

Figure 45: (Richmond, 2001)............................................................................................ 94 Figure 46: (Richmond, 2001)............................................................................................ 94 Figure 47: (Richmond, 2001)............................................................................................ 95 Figure 48 (Taylor, 2001)................................................................................................... 96 Figure 49: (Taylor, 2001).................................................................................................. 97 Figure 50 (Taylor, 2001)................................................................................................... 97

List of Tables Table 1 Numerical Data on Significant al-Qaida Events.................................................. 35 Table 2 Survey Analysis: Opinion Changes ..................................................................... 70 Table 3 Survey Analysis: Model Acceptance................................................................... 70

vi

1. Introduction Immediately following the terrorist attacks on September 11, 2001 many individuals around the world were asking the question, “How did it happen?” Within a matter of minutes the airwaves were filled with explanations. Some simply claimed that it was the actions of crazed individuals, while others offered comprehensive explanations regarding U.S. policy and the reaction of the world. Regardless of the differences in descriptions, it soon became obvious that terrorism is a subject that has been studied in great detail. With the shock of the tragedy still tender, many citizens of the United States were outraged to discover that the source of such terror had been known and watched for many years. A single man, named Osama bin Laden, had been claiming safe haven in the country of Afghanistan, breeding a new form of terrorism to unleash on the world. Many people responded with only more questions. Why the United States? Why innocent people? Why wasn’t it prevented? Since then, the research of many scholars has presented the public with tremendous amounts of information and theories regarding the formation of terrorist groups and their actions. Some terrorism experts made startlingly accurate predictions of such terrorist acts many years prior. With such information available, it again raises the question, “How did it happen?” The reason most obvious reason is that the problem of terrorism is extremely complex. Even the very definition of terrorism has no simple explanation. Like many other complex problems facing civilization, effective solutions not only require thorough research, but also a means for encouraging the general public to support the best course of action. No matter how precise the explanation, or justified the resolution, it is up to the actors in the problematic system to make theory a reality. In the case of terrorism, or other global problems such as the environment, the system includes nearly every individual around the globe. Yet, in both of these instances the majority of the people in the world do not promote the destruction of the environment or useless acts of violence. So then why do problems like destruction and violence persist? The answer involves the same reasons that allow people to also provide the safe and comfortable environment that we live in. The progress of technology, education, and communication in modern society has greatly amplified an individual’s ability to either end great suffering or cause great suffering. The solution remaining to be discovered is a more effective way to control this progress in a responsible and constructive manner for all the people of the world. In response to these types of problems, numerous techniques for facilitating the understanding of complex systems have been developed. One that shows growing promise is the field of system dynamics. The principles of this field have a great potential for promoting widespread understanding of many problems facing the world today. A system dynamics approach is no different than any other technique in that it cannot predict the exact behavior of a complex system (non-linear). Regardless, it does have many useful benefits, such as helping to clarify the structure and relationship of causes that lead to problematic behavior.

1

Another important benefit of system dynamics is its ability to trace the effects of particular factors. This approach helps to explain why policies built with well meaning intentions can sometimes have unanticipated reactions and unintended consequences that actually end up making the problem even worse. Very often these solutions only address a symptom or one aspect of a problem. Policies like these may indeed help to curb growth of some problems, but such solutions are far less successful in eliminating the source of problematic behavior. Examining a system dynamics model can help avoid this issue by uncovering the root causes of a problem. As a result, more complete solutions tend to be discovered, with an ability to test numerous policies in the model. System dynamics can also be used to help facilitate consensus formation between different groups with differing strategies, and theories by expressing and testing all of the disputable factors within the model. Yet, as stated before, finding the best solution to a problem is only half the struggle to ending problematic behavior. In the end, it is up to the individuals in the system to ensure the success of any plan. The process of educating large groups has always proved difficult, especially when there are sharp differences in abilities and backgrounds. The use of modeling has been demonstrated to be consistently successful in changing behavior under such circumstances. For example, disagreements are often the result of individuals not understanding one another. In these instances, models can help to explain the significance of one another’s actions, allowing for more unified support of solutions. Models also help to take the focus off disagreements between individuals by providing a common objective of working with the model. In returning to the problem of terrorism, it is clear that system dynamics may have a lot to offer. The response of the United States to the tragedy on September 11, 2001 is undoubtedly a critical step towards the end of global terrorism. Despite any progress, there are still a great number of problems in the struggle against violence. If these problems are left unsolved, the United States and other nations will most assuredly be subjected to further acts of terror. It will require the collective effort of many nations, religions, and cultures to reach new agreements and levels of understanding. In dealing with such diversity, tools like those offered by system dynamics will need to be employed in order to reach successful new policies. This project specifically examines the type of terrorism promoted by Osama bin Laden and the al-Qaida terrorist network, which is defined in context throughout the report. This study includes a detailed summary of the background of this “brand” of terrorism, covering expert analysis of both the psychological, social, and regional characteristics as well as the factors leading to group formation and terrorist attacks. The anti-terrorism policies and strategies used to combat this form of terrorism are also presented along with an analysis regarding the effectiveness of each. The information gathered is then divided into sectors that are represented in a model of the system. Tests for comprehensibility by using group forums are also described. In order to measure the effectiveness of system dynamics as an education tool, the conclusion of this study provides an analysis of what different individuals learn from the model.

2

The problem of terrorism is not new to the world, but its progression has reached a point that indicates that a more knowledgeable public will be required to find effective solutions. With terrorism’s global reach, these strategies must be understandable and acceptable to many groups with different or even conflicting backgrounds. This report is intended to provide some initial insight into the dynamics of terrorism and to demonstrate the usefulness of systems dynamics in informing the public.

3

2. Background Introduction In order to accurately develop a model, an extensive understanding of the problem of terrorism must first be developed. One limitation of any model is the requirement of boundaries. The boundaries of a model simply define what factors are represented. Since it is impossible to model every relevant factor of a complex problem, the boundaries explain what is and what is not being considered and why. While restricting the number of factors does make it easier to build understanding, it also highlights the importance of the assumptions made in creating the boundaries of the system. It is therefore necessary to base all of the assumptions on analytical data or, at the very least, an educated opinion. For this reason, the background provided is intended to “funnel” down the numerous overall factors into more manageable assumptions. This process is broken into the two main sectors of the model: 1) the terrorist group (al-Qaida), and 2) the opposition to terrorism, the United States. Each section begins with a broad approach to the material that is then applied towards the specific assumptions regarding the al-Qaida “brand” of terrorism. Since this project is part of a degree requirement, I assume that the reader is already familiar with system dynamics. I do, however, offer a brief summary of some of the key principles in Appendix A, and suggest further readings for those interested.

2.1. The Terrorist Group This section provides a contextual definition of the type of terrorism used by bin Laden and describes the contributing factors to the al-Qaida terrorist group formation as it is expressed in the model. These factors are drawn from psychology and sociology, religious fundamentalism, and regional history. Information is also presented about Osama bin Laden and the al-Qaida network, which was used to establish the group objectives and historical behavior that will be expressed in the model. To aid in continuity, a brief summary at the end of each main subsection explains the relevance to the aspect of terrorism being modeled. The complete application is provided in the modeling section of the report.

2.1.1. Defining Terrorism Before discussing the causes of terrorism, it is first necessary to establish a working definition in the context that will be used throughout this report. There are many definitions that have been proposed throughout the history of terrorism, but most attempts are within the realms of a social perception of norms of behavior. The majority of the approaches found tend to associate terrorism with the type of act committed and the motivation behind it (Henderson, 2001). Problems arise however when attempting to categorize groups and individuals based on types of acts and motivations. In many of the reports to Congress, terrorism is defined as politically motivated (United States, 2002B). In other words, the terrorist’s actions are aimed at disrupting government activities for political reasons. Other approaches take a look at the manner in which the terrorists’ acts are carried out. The difficulty in this reasoning is in separating acts of war and other political violence from terrorists’ acts. The classification of the justification and purpose (ends) for violence and the type of violence carried out (means), quickly becomes a debate of social and moral values. 4

While these factors may provide indications and further elaborate on the definition of terrorism, they should be not considered as binding. Definitions that begin to tread into the areas of religious and societal moral beliefs tend to be highly sensitive to individual perception and interpretation, and are therefore not particularly useful for research. In regards to motivations, Ezeldin (1991) refers to the common statement, "one man's terrorist is another man's freedom fighter." Because of this, one should use extreme caution in linking terrorist groups with other organizations with similar motivations but different means. By establishing that a terrorist group is fighting for a particular cause, other groups are then mistakenly identified with the terrorist group. For example, associating Osama bin Laden's terrorist organization with Islam has proven to be an unacceptable generalization. This point is demonstrated by President Bush's numerous appeals that the U.S. is fighting a war on terrorism, and not the Muslim religion. It is true that many terrorist groups today claim extreme religious motivations, but this fact represents an association and is not necessarily a contributing cause. What defines a terrorist act is also an area of difficulty. This problem is quite clear when examining the objective of each act instead of the overall justification or ends. If the primary objective of an act is to incite terror in order to disrupt activity, then under many definitions it is a potential terrorist activity (Henderson, 2001). The issue with this assumption, however, is that many acts of war and government law enforcement would then qualify as terrorist acts. An example would be the FBI's use of tear gas, loud music, and other disrupting activities in an attempt to bring an end to the standoff in Waco, TX. Some would counter this example by stating that U.S. actions are sanctioned by the government and target only combatants, whereas terrorists target civilians. In return, others would point to the thousands of civilians killed during the U.S. bombing strikes in Afghanistan (BBC News, 2002). Again, it quickly becomes pointless to argue whether or not violence in any form is justifiable, but rather that different groups use and justify violence within a specific perspective and under certain circumstances. The final problem with defining terrorism by actions and motivations is that a group can only truly be categorized as a terrorist group after they have begun to commit actions that fall within the realm of the definition. In addition, the application of a definition based on the type of violence may only encourage groups to use the identified violence in order to gain attention. One approach to a definition that shows promise is to look at the psychology of individual terrorists and to analyze group actions from the perspective of the social psychology of organizations. This approach would be far more consistent in that it begins to address how the decision for violent actions is made. A large majority of people's reactions to the events of September 11th consisted of surprise and confusion. How someone would ever get the idea of crashing a plane into a building full of people is not something that a psychologically sound mind can comprehend. Because of this, great effort should be exerted to avoid associating terrorist thinking with well grounded rational thought. This is not to say that terrorists do not make calculated and logical decisions, but rather that these decisions are founded on completely inaccurate 5

assumptions. This study will refer to the faulty psychology of the terrorists as having flawed mental models, which will be explained further in the following sections. Despite the benefits of social theories, it is still unlikely that a definitive description of terrorism will ever be reached. For this reason, the definition of terrorism must only be used within the context of a study, and not as a means for identification and description. Because of this, the following sections continually develop a contextual definition of Osama Bin Laden’s and the al-Qaida network’s form of terrorism. A summary of the definition developed is as follows: 1) A terrorist act in this study is determined to be any attack aimed at undermining the United States Government’s capabilities in maintaining international interests and domestic peace. Terrorist acts include violent actions against civilians, military personnel, and any other U.S. resource (transportation, building, etc.) as well as similar acts against any other any country aligned with U.S. goals. 2) A terrorist is any individual using a flawed model of the real world to commit or demonstrate a willingness to commit a terrorist act. The primary flaws in a terrorist’s mental model involve issues in the misperception of an enemy and effectiveness of terrorist violence, as well as tendencies towards mass killing or genocide. In this study, a terrorist also behaves in accordance with the objectives set by Osama bin Laden and other leaders of the organizations in the al-Qaida network. 3) A terrorist group is any assembly of the defined terrorists. 4) A terrorist supporter is any individual or organization (government) that commits funds, resources, people, or any other form of assistance to a terrorist group. 5) A terrorist sympathizer is any individual or organization (government) that knowingly or mistakenly tolerates or permits terrorists and terrorist acts. 6) The main terrorist objectives of Osama bin Laden and al-Qaida leaders specifically include: a. The destabilization of the United States Government and its ability to maintain international interests and domestic peace. b. The undermining of all secular governments in the Middle East, which are to be replaced by a single government ruled by the Caliphs and Shari’ah (Islamic Law). c. To expel all non-Muslims, particularly the U.S. military, from the Islamic Holy Lands. d. The extermination of the “infidels or “unbelievers” and defeat of factionalism. Some of the above definitions may seem quite extensive and inclusive, but it should be noted that such broad descriptions are necessary in order to discover the roots of terrorism in the model. In addition, these definitions are applicable only within the context of this report and are not intended to be applied outside of this study in terms of identifying terrorists, supporters, and sympathizers.

6

2.1.2. Psychology, Fundamentalism, and Regional History The previous definitions were formed within the context of the psychology, religious fundamentalism, and regional history factors that were identified as contributing causes of the terrorism described. It was the condition of these factors that enabled Osama bin Laden and the al-Qaida network to successfully build the required support base and plan of action. The following sections provide an intensive summary of these factors and how they relate to the terrorism being modeled.

2.1.2.1. Mental Models and Terrorism The American Heritage Dictionary (1992) defines a problem as “a question to be considered, solved, or answered.” In regards to terrorism, the first problem most people identify is the use of extreme violence. From there individuals usually have a long list of other problems that are associated prevalence of terrorism. In focusing on the problem of extreme violence, some of the first questions considered are, “what is the motivation, and why was it used?” Addressing these questions requires tracking backward through the many different decisions that lead up to actual attack. How these decisions are made is likely an endless debate, and yet every theory about motivation and decision making processes suggest that individuals have some type of structural way of functioning in the world. One way of describing these “thought” structures is in terms of a mental model. There are many names for mental models. Some examples might be strategies, cognitive maps, decision matrixes, and behavior routines. In general a mental model is any pattern that is learned, taught, and repeated from individual to individual. An important characteristic of mental models is that they are always based on a perception of the real world. Defining a mental model is then just a description of the processes for interpreting the feedback received from interacting with the real world and for determining future courses of action (Johnson-Laird, Byrne, 2002). Yet if individuals only experience the world through perception, the result implies that all mental models must make assumptions and are therefore inherently wrong (Richmond, 2002). In other words, no matter how well a model represents the real world, the limitations of human perception prevent a model from ever being completely inclusive. This is not to say that all models are useless, but rather it is important to note that every model has its limits. Some mental models are biologically predetermined, such as the perception of color in response to different wavelengths of light. A model can also be something as simple as the steps for tying your shoes. Language is yet another mental model used for communicating ideas, feelings, and needs. In terms of groups, it has been demonstrated throughout history that the successfulness of any structured organization to accomplish its goals is primarily dependent on the effectiveness of individual motivation. Put another way, groups that succeed in surviving perpetuate a common mental model for living that enables its required goals and objectives to be accomplished. In the case of this project the interest is in the common mental models involved with the problem of terrorism. These include those of al-Qaida terrorists, supporters, and sympathizers, as well as those of U.S. citizens and policy makers. The following sections will primarily focus on the factors that form a terrorist’s model of the world and the behavior it motivates him or her to exhibit. 7

This first part of the background presents many of the key factors that form an alQaida terrorist’s mental model. This information is then used later in the project to develop the model sector that simulates the group’s behavior. In the conclusion I propose that there are several critical flaws in how terrorists perceive their enemy and the effectiveness of using violence, and in their tendencies towards mass killings or genocide. As a result of these flaws it is not unreasonable to assume that the al-Qaida terrorists will not succeed in surviving in the long-term. The main reason for this is that terrorizing methods of violence have been, as Carr (2002) states it, “one of the most ultimately selfdefeating tactics in all of military history.” Regardless of this fact, the time it takes for the mental model to be globally rejected may lead to painful periods of escalated violence and suffering. In addition, the factors originally contributing to the use of violence to incite terror must also be addressed in order to prevent future outbreaks from completely different groups.

2.1.2.1.1. The Origins of Violence and Violent Groups The difficulty in categorizing different manifestations of violence seems an impractical approach to identifying terrorism. The main reason for this is because of the immediate contradiction between experiencing and witnessing violence. For most people, experiencing violence in any form is undesirable. Yet when witnessed secondhand, the understanding of violence is immediately far more sensitive to individual perspective. This reason alone helps to explain how violence is used and justified individually and in groups, even though it is undesirable for most everyone. In situations of appalling violence, many deem it as incomprehensible evil, or barbaric in nature. But these types of descriptions only tend to build evil to mythic proportions. Doing so then only prevents individuals from building a realistic understanding of the origins of violence and how it occurs. The reason why many people avoid developing an understanding of violence may be, as Staub (1989) notes, “because comprehension might lead to forgiving,” which is something that seems inappropriate in the case of extreme violence. The problem, however, is that without a more comprehensive view of violence the possibility of ensuring that “incomprehensible evil” does not happen again is unrealistic. Through studies of genocides and mass killings, Ervin Staub (1989) discovered that there was a certain pattern that led up to the use extreme violence. His work suggests that there are specific characteristics of a culture and societal structure that create an opportunity for violence when combined with hardships and social problems. The result can lead the society and subgroups within the society to turn against one another or against outside forces. Over time disagreements cause increasing mistreatment and dehumanization of the opponent, and can eventually lead to genocide or mass killing (Staub, 1989). It should be noted that the process can be reciprocal. Again, using perspective may deem one side more “deserving” of violence than the other, but in the process of explaining the origins of violence these desires to justify it must be ignored for the time being. Staub (1989) suggests that the psychological impact of the following factors provide the foundations for individual and group violence: difficult life conditions, needs and goals, the coping and fulfillment of needs and goals, a continuum 8

of destruction, cultural and societal characteristics, the role of bystanders, motivation, leadership, the relation of individuals and the system, and inherent evil potential. The likelihood of violence, and more specifically genocide and mass killing, is therefore primarily dependent on the conditions of these factors. Difficult life conditions often create the necessary background of instability that allows for significant changes in attitude and motivation of groups (Staub, 1989). These include extreme economic conditions, political violence and instability, and rapid changes that have a direct impact on an individual’s livelihood, customs, and way of life. In addition, the rigidity or the adaptability of a society can affect how disturbing these difficult life conditions can be. A result of the presence of difficult life conditions is the formation of demanding needs and goals (Staub, 1989). These demands lead to powerful self-protective motives that lead an individual to perceive the world as threatening to both the physical and psychological self. These perceived threats are usually faultily blamed on a single institution or group, instead of taking into account the complexity of the problem. From the blame, individuals and groups become more enabled to construct goals that encourage violence against the governing force. Since difficult life conditions also tend to diminish an individual’s sense of self, goals of rising to a higher stature are also fairly common. The manner in which the previous goals and needs are fulfilled is also a critical step in the progression of violence (Staub, 1989). When a person is presented with a problem, constructive activities often help to provide comfort to an individual and bring into focus the potential for resolution. These activities may be the expression of concern through writing or protest, or actual steps towards what is believed to be a solution. Unfortunately many difficult life conditions are not easy to address and require changes in circumstances beyond an individual’s or particular group’s perceived control. As a result many individual needs and goals are unmet, which can then lead him or her to coping behaviors. In particular these include: the separation of groups into “us” and “them”; devaluation of “them” with terms like “infidels” or “barbarians;” predetermined justice, which is the tendency for a group to believe that “they” (the identified opposition) deserve suffering as a consequence of their action or existence; and finally scapegoating, which is the blaming of others for personal problems. These factors also work together to create a sense of group identity and an ideological plan of action for reaching the individual or group goals. This psychological coping can then turn into a self-fulfilling prophecy by providing the aforementioned constructive activities that bring comfort. These patterns do help to create the potential for extreme violence, but Staub (1989) suggests that there is also “a progression along a continuum of destruction. People learn by doing, by participation, as a consequence of their own actions.” In other words, to sit and think about destructive activities is simply not enough to make someone violent. As an individual experiences destruction firsthand, he or she is forced to create a personal means for understanding, even if it is simply to mimic another person. For this reason, as people take part in more and more of the previous coping activities, such as the creation and devaluation of an enemy, they also increase in potential for becoming 9

perpetrators of violence. Given enough time under the right conditions, nearly any individual may experience a reversal in values, where killing is viewed as necessary, good, and righteous. Learning by doing even includes bystanders who only show passivity or indifference, since deciding to ignore violence is still a decision and doing so repeatedly makes it only easier. This point highlights the danger of how even small and seemingly harmless activities can lead to progression along a continuum of extreme violence. Other important factors are the cultural and societal characteristics out from which the extreme violence grows. These characteristics help to determine the impact of difficult life conditions, the options for addressing them, as well as the aggressiveness in progression along the continuum of destruction. Influential forces include nationalism, respect for authority (government), the perception of individuality, and spirituality. For example, high levels of nationalism and respect for government in Germany most likely assisted in the progression along the path to the Holocaust. The role of bystanders also helps to direct the course of events leading to extreme violence. It is the general support, opposition, or indifference of the people that determines the future of a progression toward violence (Staub). While leaders certainly provide a direction, a sustainable path to violence requires the cooperation of many individuals. Bystanders can certainly be coerced into agreement with the perpetrators of violence, yet often there is a potential for bringing the objectives of violence into question. This is especially true during the early stages of the continuum, when large groups of bystanders may still be unfamiliar or undecided over the issue. The motivation of the perpetrators of violence is most certainly a key part to understanding how genocide and mass killings arise. In this study, motivation is discussed in terms of a mental model. A benefit of this approach is that it allows for the representation of various motivational theories, such as needs based theory, achievement, power, and affiliation theory, as well as Staub’s personal goal theory. Regardless of this feature, any approach to determining behavior is limited to probabilities (Staub, 1989). While these theories help to indicate the potential of extreme violence, no theory is ever likely to be successful in predicting exact behavior. In establishing a mental model for violence, or a pattern in the evolution of such a model, this study hopes to shed further light on the motivation to commit extreme violence. The role of leadership is also undoubtedly a major point in uncovering the origins of violence. This issue will be addressed more thoroughly later in the report, but in general, personal background, charisma, organizational ability, and similar characteristics contribute to the effectiveness of violent leaders. Another important point Staub (1989) notes is that, “If difficult life conditions persist and the existing leadership and societal institutions do not help people cope at least with the psychological effects, the people are likely to turn to radical leadership.” This view helps to provide a limited ability to predict the type of leadership likely to emerge under certain circumstances, yet numerous other factors are still important in determining if it will be peaceful or violent in form.

10

Another significant factor is how individuals interact with the prevailing system. This issue is related to the societal and cultural characteristics, but specifically identifies how individuals are shaped throughout life. For example, the system of the United States promotes general openness to new ideas and self expression. In contrast, many Middle Eastern cultures require strict obedience and unquestioning faith in principles. Because of this, individuals in each system may be manipulated by others promoting violence differently and undergo changes in perspective at differing rates. If exploited properly, nearly any system can be used for extreme violence, yet certain systems may be more vulnerable in different ways than others. The manner by which a system is manipulated is certainly a large part of a system dynamics approach to understanding violence and will be addressed more throughout the study. The final factor that Staub (1989) noted in the origins of extreme violence is what he re referred to as the inherent potential for evil. Even though “evil” is a widely used term that has no specific meaning, Staub believes that most people generally accept it as a term that refers to the destruction of humanity. Under this description, evil would then be the actions causing the destruction. This is an important distinction since evil is not something that exists in ideas, but rather in the consequences of actions (Staub, 1989). It is true that ideas are the basis for action, yet in many cases, the so called proponents of evil claim to be acting on higher ideals. This reason alone uncovers the potential of any group using violence for doing evil. It is this inherent potential for evil that explains how even well meaning ideals can lead to the extreme violence of genocide and mass killings. To romanticize the idea of evil and use it aimlessly in terms of groups or individuals seems quite unproductive. Instead such attempts return to the endless debate of ideals. This study will attempt to avoid this issue by focusing on actions that lead to the evil destruction of humanity through terrorist violence and not the evilness of terrorists themselves. Another way of examining the structure of group violence identified by Staub (1989) is to arrange the factors in a causal loop diagram. The following figure helps to provide a sense of flow to the growth of extreme violence. The black arrows direct the causation link from on factor to another. The red “+” symbol indicates that the two factors move in the same direction, or in other words when one grows the following factor also grows. Since all factors move in the same direction, the loop is therefore a reinforcing loop, which is denoted with the red “R” in the center of the loop. The green arrows are conditional links that are not completely dependent on the cycle depicted. These independent links are conditional in the sense that the factor will only impact the growth of extreme violence if the circumstances are correct. For example, the culture and environment may not generally support or tolerate the use of violence, in which case the role of bystanders would be less significant and the cycle would likely collapse. The link from “Events Changing the World” is orange to emphasize that changes can either increase or decrease the difficult life conditions and are not necessarily constant throughout the cycles.

11

Events Changing the World

+

± Prevalence of Difficult Life Conditions

+ Level of Inherent Potential for Violence

Amount of Demanding Needs & Goals

Progress on Path of Destruction + +

+

R R Bystanders Willingness to Accept Violence + +

Amount of Leadership for Violence

+ Amount of Motivation for Violence

+ Number of Coping & Fulfilling Needs & Goals Activities +

+

Amount of Support for Violence +

Number of Activities+ Dehumanizing & Making of Sides +

Extent of Exploitation of the People & System Interaction

Figure 1 Group Violence Reinforcing Loops Causal Diagram

One benefit of examining these factors in a causal loop diagram is that it helps to explain how extreme violence grows over time. As time passes, the cycle is repeated, allowing for further progress on the path of destruction leading to mass killings or genocide. Though not directly stated in Staub’s work, the closing link from the “Progress on Path of Destruction” to “Prevalence Difficult Life Conditions” is based on the fact that as destructive acts are committed there is often a response “in kind”, which affects life conditions in the region. Conditions also tend to worsen since individuals often leave more comfortable lives in order to participate in the struggle. This section presented several initial background factors that will be developed through this report. In the process of understanding how Osama bin Laden and the terrorists of al-Qaida commit acts of terrorism, the origins of violence must be considered. Through past studies of instances of mass killings and genocide, Staub (1989) has been able identify a specific pattern and structure to extreme group violence. As later sections will demonstrate, when applied to the formation of the al-Qaida terrorist network, “incomprehensible evil” suddenly becomes easier to understand and predict.

12

2.1.2.1.2. The Psychology and Sociology of Terrorists, Groups, and Leaders The previous section discussed the general origins of violence in groups. This brief overview was directed at all forms of violence including both government sanctioned and extremist group violence. While this aids in providing a broader understanding of how violence is used, this study is more specifically concerned with terrorist violence. This section presents some of the key psychological and sociological factors causing group formation and actions. 2.1.2.1.2.1. The Terrorist Mental Model

In considering the individual and group mindset of terrorists, two questions arise: how is the mindset formed and how does it motivate a terrorist to behave? Throughout this study it has become quite clear that the formation of the terrorist’s mental model can follow many different pathways. This problem underlines the difficulty of profiling terrorist members and identifying potential recruits. Perhaps the main reason for so many diverse routes to terrorism is that individuals in general can reach the exact same conclusions through very different lines of reasoning. What is important to realize however is that all terrorists end up with markedly similar behaviors. For this reason it seems more productive to accept that under the correct circumstances and background nearly any individual could become a terrorist. Regardless, I will present a few of the psychological and sociological formation hypotheses regarding individuals and groups for reference purposes before describing how mental models motivate terrorists. There are many psychological hypotheses that have been proposed to explain how the terrorist mindset is formed. These include frustration-aggression (Margolin, 1977), negative identity (Knutson, 1981), and narcissistic rage (Crayton, 1983). In general, these theories attempt to approach the issue of how people can develop violent attitudes towards others (United States, 2002C). As the names suggest, the use of violence is often the result of the individual or group feeling some kind of oppression or alienation. Many of the theories discussed in a Federal Research Division (2002C) report suggest that terrorists reach for violent forms of expression out of desperation or some other enabling feeling. Where many of the theories diverge is over the issue of intelligence. While some theories accurately explain how highly educated individuals with significant achievements become terrorists, they fail when applied to naïve youth or the underachieving and downtrodden. In the case of highly educated individuals, several theories suggest that frustration over the lack of opportunities and with the contradictions between what was taught and the actual state of the world lead to terrorism. In contrast other theories claim that underachievers turn to terrorism since it champions their plight and gives hope for a better life. These discrepancies only indicate that ultimately there is some type of individual desire that is exploited to provide the foundation for violence. One point that most researchers now agree upon is the rejection of the terrorist as mentally ill. While it would be convenient to say that all terrorists are lunatics, findings indicate that very few have truly severe disorders (Hoffman, 1999). In addition, it is highly unlikely that terrorists would be able to complete the demanding organizational tasks required for planning an attack if they had any type of mental disorder. 13

Despite the difficulties in determining how the terrorist mental model is formed, significant progress has been made in explaining terrorist behavior. If one can accept that terrorists are not born into the world, but achieve the required mental model through some cognitive processing of the environment, then it is also reasonable to conclude that terrorists have formed specific needs and goals. While these needs and goals vary significantly depending on the circumstances of certain factors, in general terrorists are seeking power over an individual or group, using the only means they perceive to be effective in achieving this goal (Hoffman, 1999). Very often the means is violence or the threat of violence aimed at influencing public and government opinion, but it also may be expressed through other destructive attacks. Some common examples include assassinations, mass killings, “cyber attacks”, and destruction of property. The key point to note is that terrorists have an unwavering belief that their actions are the only way to achieve the desired goals and needs. Because of this, if conditions appear unchanged, then terrorists are far more likely to use increasingly severe measures. This may be a more defiant attack, such as assassinating government leaders, or one that is more shocking, like the crashing of a plane into a building. The point that many individuals have a hard time understanding is how any human with at least a shred of respect for life could rationalize the killing of others, especially innocent bystanders. In this respect, psychology has discovered a multitude of ways humans cope with killing. Four techniques of moral disengagement that were described by Albert Bandura (1990) are: envisioning the self as a defender of a group threatened by evil; the displacement of responsibility to leaders (following orders); physical disengagement through bombs or suicide; and dehumanization of the enemy. Most common is the dehumanization of the perceived enemy (Hoffman, 1998). Terms like “infidels,” “barbarians,” “unbelievers”, “dogs,” “evil doers,” “enemy of God,” or “children of Satan”, are used to make the victims of terrorism subhuman, and unworthy of living. Terrorists also justify violence by blaming the enemy for causing similar suffering in their world. Terrorists accuse the enemy of being evil in nature and of having an actual plan or conspiracy that is intentionally causing the suffering. These plans are often called “plagues” or “evil plans” and the enemy promoting them are referred to as corrupt, immoral, or “the great Satan” (Bin Laden, 2001). Under this mindset, they are then able to lay responsibility on any individual or group that is not supporting their cause. For example, bin Laden (2001) claims that the United States is purposefully killing Muslims and attempting to destroy the Islamic faith. Under such a broad accusation, he enables terrorists to include anyone who supports U.S. ideals instead of Islamic law as their enemy. The reason that terrorists are so successful in maintaining such broad accusations can be accredited to their narrow lens or perspective of the world (United States, 2002C). In cognitive psychology, this lens affects how individuals make sense of the world and interpret the feedback of their actions. Terrorists tend to perceive the world based on a

14

subjective interpretation instead of objective reality. In other words, they view the world more as it suits their needs, rather than by identifying the actual cause and effect. Physical disengagement also assists terrorist in coping with violent behavior (United States, 2002C). When terrorists use more distant attacks, such as time bombs or long-range riffles, they avoid witnessing the devastation firsthand. Even media reports providing detailed information about that attack can be discredited as a conspiracy of the enemy. The use of suicide bombings also provides an effective means for disengagement, since the individual perpetrator does not live to react to his or her actions. This also aids the group since they can focus on the martyr by praising and worshiping their own loss instead of reacting to the innocents killed by the terrorist act. Another significant commonality between terrorists is living for the future. As Hoffman (1999) points out, by living “in the future, that distant - yet imperceptibly close - point in time when they will assuredly triumph over their enemies,” terrorists have a greater freedom to disregard setbacks and suppression. Very often terrorist are engaging an enemy of significant power and authority, such as the government. This also helps to create a deep sense of obligation and importance. They are battling “against all odds”, which only furthers their willingness to become increasingly destructive. These reasons help to explain how terrorists continue to gain support, despite the lack of measurable progress. In conclusion, there are indeed many similarities in the mental models of different terrorists groups. While the overall goals and needs depend significantly on other factors, terrorists use similar psychology to justify and cope with their violent behavior. This information again adds to the progression of an understanding of the “brand” of terrorism being modeled. 2.1.2.1.2.2. Terrorist Groups

Group dynamics are another important influence on terrorist behavior. In the formation of groups, terrorists fulfill a number of basic human needs and help provide the necessary support and justification for committing acts of violence. In addition terrorist groups are subject to normal group characteristics, such as groupthink, conformity, and demands for action (United States, 2002C). Any group helps to fulfill an individual’s basic needs for affiliation and importance (United States, 2002C). In general, people need to have some sense of self worth and belonging. If normal society fails to provide this for any individual, the likelihood that he or she will join an outlying group is significantly greater (Shaw, 1986). This is a common phenomenon, and is demonstrated by the long history of gangs, cults, and other rebellious groups. As stated before, terrorists often feel alienated or oppressed in normal society. In discovering a terrorist group, these individuals find a new identity that exploits their desire for a better life and feeling of significance. Since terrorist group goals are revered within the group, individuals also sense a connection to something beyond themselves. This helps to create feelings of privilege and honor, as well as an 15

opportunity for contribution. As stated earlier in the report, individuals require a means for coping with and achieving personal needs and goals. Groups usually provide an environment that encourages coping activities and the formation of plans of action. By taking part in this process, group members satisfy needs and develop constructive tasks towards goals that bring the comfort of accomplishment. In addition to fulfilling basic human needs, groups also contribute to violent behavior through other processes. In groupthink, members develop “illusions of invulnerability leading to excessive optimism and excessive risk taking, presumptions of the group's morality, one-dimensional perceptions of the enemy as evil, and intolerance of challenges by a group member to shared key beliefs” (Untied States, 2002C). Because of groupthink, terrorist groups are not likely to significantly change overall goals, which may enable the increased use of violence or other extreme measures. It is also related to the narrowing of perspective described in the terrorist mindset. Groupthink is in part a reaction to the pressures of conformity. Terrorist groups have been known to kill their own members under the suspicion of insufficient commitment. By creating a mindset of complete obedience, group members become absolved from many personal doubts. For this reason, gaining entrance to a terrorist group often requires some kind of initiation or other task that proves loyalty. As a result, every member then shares a similar experience that further adds to the sense of commitment and belonging. The final characteristic of groups to be addressed is the tendency for demands of action. Individuals join groups to take part in a meaningful experience. If a group is reluctant or incapable of fulfilling member needs, the pressures for any form of action often mount (United States, 2002C). This fact plays an important part in the selection and respect of leadership. In general, the wider the range of activities members can participate in, the stronger the groups sense of purpose. Groups that encourage prudence and moderation typically have internal tensions and develop stresses that weaken the confidence in leadership. Both instances can give way to the further use of violence to either occupy the group or provide an outlet for inactivity. 2.1.2.1.2.3. Terrorist Leadership

The leadership of any group is perhaps the single most important component of group activity. This area has been studied extensively in regards to government, business, religion, and other organized groups. In every case, leaders play a central role in determining overall goals and the means for motivating and accomplishing them. Terrorist groups are no different in this regard, yet the leader selection and replacement processes tend to be far less formal than in other organizations. Terrorist group members tend to perceive leaders as the embodiment of the group ideals (Hoffman, 1998). For this reason, terrorist leaders do not necessarily have to be charismatic and engaging, as many other groups require. While these qualities do enhance the group’s responsiveness to commands, terrorists tend to place more 16

importance on the exercise of leadership. If the leadership of a group does not behave consistently with the group’s goals, then it is likely that their ability to command will be diminished over time (United States, 2002C). For this reason formal titles and positions carry less meaning to terrorist groups. Instead, each member plays his or her role towards the greater good of the group. Respect is then given more in terms of proven diligence, rather than assigned recognition. In addition, some of the most successful leaders demonstrate little personal gain, frequently accrediting others for ideas and actions, despite their strong influence in the matter. This further strengthens the group identity and lends greater freedom for the group members to push the ideology towards more extreme ends. This is not to say that leaders do not have a militant command of the group, but rather that command is earned rather than demanded. Since there is often no formal recognition or selection process, leadership tends to emerge from the group. For this reason individuals that bring some added feature, such as recruiting experience, weapons and combat expertise, or financial support are most likely to be drawn into leadership roles. As discussed in the group section, leaders tend to be individuals that continually provide activities and new directions for the group. Defiance and boldness therefore become added qualities. For example, the statements issued by Osama bin Laden often challenge conventional authority and attitudes, like the preeminence of superpowers and respect for royalty. One danger of anti-terrorist movements recognizing the power of terrorist leadership is the tendency for it to only further add to the mysticism and allure of the leader (United States, 2002C). Often terrorist supporters regard opposition claims that group leaders are “monsters” as a feeble attempt to discredit the ideals they embody. As a result leaders are then raised to higher statures in the group’s social world as being defiant and brave.

2.1.2.2. Fundamentalism Many violent struggles have been closely linked to some type of fundamentalist or extremist principle (Hoffman, 1998). These principles can come in many forms, some are religious in nature and others are based on race or a political ideology, but all have a similar thread. While each group might have very different views of the world, they all share a similar structure in their commitment to individual beliefs and causes. As with other definitions in this report, I will begin with a broad explanation of fundamentalism, and then proceed to focus down to the specific definition that is relevant to the model. As stated before, it is not reasonable to conclude that the commitment to specific doctrines defines fundamentalist groups. There are a great number of individuals committed to the Bible, Koran, Torah, U.S. Constitution, or other forms of doctrine establishing a framework for living that are in no respect fundamentalists. In this case it is important to differentiate between traditionalist and fundamentalist. It is true that fundamentalists are also traditionalists in a qualified sense since both groups base their ideologies on the pursuit of purity based on the interpretation of sacred doctrine. But traditionalists are interested in this pursuit only, which leads to marked dissimilarities to fundamentalists. It is the actual structure of the fundamentalist group and how they 17

incorporate ideologies into their movement that defines them as fundamentalist. While there are resemblances between some fundamental doctrines, it is the shared attitude towards each religion or ideology that forms the commonality between fundamentalist groups (Appleby, 1997). In every case fundamentalists view the world in terms of absolute truths. They divide the world into realms of good and evil, claiming exclusive possession of the divine or “correct” truth, thus making it a moral duty to identify and expose the enemy. Because of this commitment, fundamentalist groups often have a powerful mistrust for opposing doctrines. In the case of religious fundamentalism, it is the secular, man made, or “godless” doctrines that breed the most contempt (Appleby, 1997). As a result, these groups develop militant commitments to their own beliefs as a defense against the encroachment of other ideologies, such as secularism and modernization. The process of forming such militant commitments involves a careful selection and interpretation of the binding doctrine, which are then used to build a common ideology and program of action for the group. The leaders in this selection process often claim divine direction to be the guiding force in choosing supportive passages for the movement. What Appleby suggests, however, is that this behavior is in fact more consistent with a political strategy in the pursuit of power. The power that fundamentalists seek is a control over the outsiders or unbelievers. This important fact enables fundamentalists to freely move between doctrines and the modern world. Contrary to popular belief, powerful fundamentalists tend to be very knowledgeable about the latest technologies and sophisticated political strategies, despite their stake in ancient belief structures (Appleby, 1997). To summarize these points, the definition of fundamentalism describes a basic methodology employed by a group in order to build an ideology and program of action for establishing and maintaining power. This methodology uses carefully selected and politically useful sacred teachings, practices, and traditions in order to build a structure of organization and a program of action. By basing this structure on “fundamental” principles, the group enables itself to be identified with larger groups and lends divine meaning to their movement. In short, Appleby (1997) defines religious fundamentalist groups as a “blending of traditional religion and its politicized, ideological defense.”

2.1.2.2.1. Fundamentalism and Violence The definition of fundamentalism does begin to provide a direction for the “brand” of terrorism being modeled, but not every fundamentalist group is a terrorist group. It is therefore necessary to further narrow our definition in order to identify how fundamentalist groups commit violent acts. Studies of groups that use fundamentalism with violence have been able to identify specific structures that govern when violence is used and what the objective is. Sprinzak (Martin & Appleby eds., 1993) describes three variables that are influential in fundamentalist violence: 18

1) “The ideology and self-perception of the movement, and its attitude toward the powers that be; 2) The nature of the enemy of the movement and its perceived threat; and 3) The nature of the leadership of the movement and its formal organization.” It is therefore the combination of these variables that determines the objective and occasion of fundamentalist violence. It is important to note that leadership and perception of the movement and enemy are the critical factors in determining these variables. The dimensions of these factors are addressed in the psychological section of this report. For the time being, I will simply assume that the perceptions and leadership have already been established in order to examine how they influence the use of violence. Sprinzak (Martin & Appleby eds., 1993) uses three case studies of Jewish fundamentalist groups in Israel to illustrate how variations in the conditions of the above variables resulted in noticeably distinct behaviors. In each, violence is used on very different occasions with differing motivations and objectives. Each structure identified on the basis of differences in the variables suggests that violence serves a specific purpose in each case of the different fundamentalist groups. The frameworks identified include: a defensive and reinforcing form, an offensive yet restrained use, and finally an offensive, unrestrained, and ideologically motivated structural violence. The conditions of the variables in the case of the defensive and reinforcing use of violence most reflects a minority group that perceives its cause as a faithful burden that is in an environment that cannot be changed, but instead only tolerated (Martin & Appleby eds., 1993). This resignation of control is facilitated through the belief that a higher power will someday intervene and reward only the faithful. The enemy therefore is only perceived as a threat when the future of the group is perceived to be at stake. Under these conditions the leadership only directs violence in response to direct encroachments upon the group’s livelihood. In the case of the use of restrained offensive violence, the conditions best describe either a minority or majority group that identifies its motives with a larger collective (Martin & Appleby eds., 1993). The perception of self is then more in line with the powers that be. The group’s view of the enemy considers them to be a threat to the larger group as a whole. In this instance, leaders direct violence with an attitude of greater freedom than the governing powers are perceived capable of. The active group therefore feels immune to punishment in the belief that their actions are more broadly condoned, or even applauded. Violence is restrained, however, because the group still has respect for the powers that be. Attacks are therefore more often responses to conflicts with specific objectives. The final group best exemplifies one that perceives themselves beyond any man made authority with the enemy being any unbeliever who takes opposition to the cause (Martin & Appleby eds., 1993). The structure of the group is perceived as divinely 19

determined, with the leadership claiming to only be carrying out the commands of a higher power. A group with these beliefs therefore envision themselves as holy warriors fighting towards a predetermined and divinely ensured future victory. An important point here is that the group does not perceive this to be politically motivated (as researchers do). Instead they perceive themselves as Armstrong (2000) suggests: “Fundamentalists do not regard this battle as conventional political struggle, but experience it as a cosmic war between the forces of good and evil. They fear annihilation, and try to fortify their beleaguered identity by means of selective retrieval of certain doctrines and practices of the past.” With an unlimited perception of righteousness these groups may use violence without any restraint. Moreover, they view the usage of violence as an absolute necessity, or as a divinely mandated surgical procedure to remove an abnormality from God’s “ideal” world. It should be noted that these structures do not specifically define the severity of violence used. In every case the violence may be as mild as vandalism or as severe as a bombing. While the last of the groups described had the greatest amount of self-imposed freedom to exercise violence, the severity still depends solely on the group’s agenda and the actions of the enemy. The important point about these structures is that they help to define when violence is used, and that under no circumstances is it arbitrary. They also provide information regarding overall group objectives and their role in violence.

2.1.2.2.2. The Violent Fundamentalism of al-Qaida In the case of Osama bin Laden and al-Qaida, the structure for the use of violence most reflects the last group described in the previous subsection. The conditions of each variable in this circumstance are clearly described in bin Laden’s declaration of war against the U.S. in 1996, which is summarized in the section on his declarations. In short, the variable descriptions would then be: 1) Al-Qaida perceives itself as warriors in a struggle (jihad) to uphold Islamic law (Shari’ah). The movement is carefully justified as spiritually mandated, and as a religious obligation. Their attitude towards the powers that be is of complete disregard. They have no respect for any other law, especially the man-made secular laws of the U.S. Therefore any government or country that does business with the U.S. is perceived as corrupt and as a potential enemy, including the governments of primarily Islamic nations like Saudi Arabia and Pakistan. 2) Al-Qaida perceives the U.S. to be an aggressive enemy with evil intentions to destroy their religion, and take over the region. They support this belief with selective observations of both direct and indirect impacts of U.S. influence in the region such as modernization and U.S. led sanctions against Iraq. 3) Bin Laden is perceived as a great spiritual leader. He has consulted and gained favor with many other religious leaders around the world. He uses powerful religious images and passages to embolden his cause. Bin Laden often describes his and others’ “visions” of victory and triumph in the future. His active role in 20

the war against the Soviets further added to his persona, and provided him with vital validation of his righteousness. The most dangerous characteristic of his leadership, however, is his active encouragement of individual self actualization. Bin Laden frequently accredits his followers for various revelations, thus giving them their own freedom to act. It also makes it even more attractive for members to seek the more and more extreme attitudes that gain his attention. The results of these effects help to condition his followers to the point where suicide becomes the highest of honors, and encourage increasingly severe attacks. Under these conditions the group objectives are then: to rid the Holy Lands of the U.S., to expel all other forms of law, and to unite the Arab world under Islamic law. The frequency is unlimited by the ongoing battle that does not end until these objectives are met. In this case, the level of violence may actually be somewhat measurable in that it will most likely increase the longer the struggle continues, and in response to any form of opposition.

2.1.2.3. The Culture of Islam, Jihad, and the Middle Eastern World The history of the Middle Eastern world has also played a crucial part in assisting bin Laden wage holy war. When placed in the context of the regional history, many of bin Laden’s seemingly abstract references carry profound meaning to Arabic Middle Easterners, especially the vast majority of Muslims. Many of these statements are related to the ongoing struggle between nationalism, Islam, and the political and economical development in the region.

2.1.2.3.1. Nationalism and Islam In contrast to the Western world, Muslims “tend to see not a nation subdivided into religious groups but a religion subdivide into nations” (Lewis, 2001). Another distinction of the region is its commitment to history. When combined, these two facts lend credence to many of bin Laden’s references. For example, his references to eighty years of humiliation and disgrace stem from the defeat of the last great Muslim empires, the Ottoman sultanate in 1918. During this period, the Ottoman sultanate was abolished. The reason this was so “humiliating” is that the sovereign of the sultanate was also recognized as the caliph, the ruler of all Sunni Islam. As the last in a long line of rulers dating back to the death of the Prophet Muhammad, the utter disregard of this significance to Islam by the revolutionaries was most certainly felt by the entire Muslim world. What made this even more sensitive was that it had happened as a result of an internal revolution after the Turks liberated their country from occupying British and French Empires (Lewis, 2001). In this regard it might have been easier to bear if the move to abolish the sultanate had been the result of conquest, and not a revolutionary idea supported or at least tolerated by Turkish Muslims. Issues like these have also been used to support the idea that nationhood was primarily constructed by the influence of the modern world. Even the names of many Middle Eastern countries are artificial, like Saudi Arabia, where Arabia has no meaning in the Arabic language (Lewis, 2001). In early history, nationalism was more often 21

perceived in terms of a common descent in the form of tribes. On the occasions that unity was required, it was more often the Muslim religion, not the idea of nationality that galvanized the effort. As a result, in modern terms, Middle Eastern nationalism is a relatively new implant with little history or concrete identity (Dawisha, 2000). It has only been further weakened by its historical inability to provide sustainable governance. One of the earliest failures of nationalism being Napoleon’s conquering of Egypt in 1798. This instance caused only further humiliation when the British were able to free Egypt without the assistance of other nations in the region. One of the few instances where nationalism managed to gain ground was under the leadership of Egypt’s president, Gamal Abdel Nasser. The nationalization of the Suez Canal Company in 1956 was perceived as a defiant act against Western imperialism, and was reinforced a few months following by the repelling of a British and French attack facilitated through Israel (Dawisha, 2000). This then assisted the 1958 formation of the United Arab Republic between Egypt and Syria, which breathed new life into the potential of the Islamic population of the Middle East uniting in order to combat the Western imperialism that had helped to establish the Israeli state. This surge in Arabic nationalism in the Middle East proved to be short lived, however, as it was continually eroded by its contradicting nature with the rest of Islamic society. As a result, regional nationalism was never fully able to absorb and represent the Muslim community. Any attempts to manifest Islam politically “marshaled the brutal and unforgiving power of the state to keep it at bay” (Dawisha, 2000). The final end to a sense of nationalism came from the embarrassing defeat of the Arabs led by Nasser in the 1967 attack on Israel. Those six pivotal days marked the end of many Muslims’ hope to soon regain a presence in the holy lands. This profound mistrust in nationalist ideologies leads many regional Muslims to further resent the notion of secularist governments, believing them “to be a spurious effort to undermine the Islamic umma (community)” (Dawisha, 2000). This attitude was later applied to the role of the U.S. in Middle Eastern affairs. During the Iranian revolution of 1979, the seizure of the U.S. Embassy in Teheran and taking of fifty-two American hostages was not the result of decaying U.S. - Iranian relations, but instead because they had shown signs of improvement (Lewis, 2001). Revolutionaries feared that improved relations would only invite more Western encroachment into Iran.

2.1.2.3.2. The Making of an “Evil” United States The encroachment of Western principles, however, can hardly explain the depth of Arab resentment toward the U.S. There have been many other issues regarding U.S. policies and the resulting treatment of the Arab world. Initially the U.S. offered little support to the British led establishment of Israel. Yet, during the cold war, Russian arms sales to Arabic Middle Eastern countries drew the U.S. closer to Israel (Lewis, 2001). This important moment has led up to U.S. support of Israel’s “right to defend itself” by 22

whatever means necessary. In the taking of sides, U.S. policy has created a “double standard” for acceptable violence, in which the assignation of Palestinian leaders and the shooting of Palestinians armed with rocks may be justified under defense. This issue was further sensitized by the United States’ willingness to jump to the aid of the royal family of Kuwait. To many Arabs it appeared that the U.S. was more than willing to support oppressive regimes that cooperated with U.S. policy, than defenseless supporters of religious rights. The most dramatic example of this was what many Arabs see as the betrayal of Iraqi rebels in 1991. In it, the rebels that had responded to the U.S. call for revolt against Saddam Hussein were gunned down by helicopters permitted under the ceasefire agreement. The reason the U.S. refrained from assisting the respondents to the call was because they had intended for it to incite a more predictable coup d’état, not a genuine revolution (Lewis, 2001). This view also parallels the U.S. abandonment of Afghanistan after the Soviet departure. Another example that aided in the Muslim perception of a double standard occurred in 1982. In the Syrian city of Hama, the radical group the Muslim Brothers led an uprising that was bloodily suppressed by artillery and aircraft. Amnesty International estimated 10,000-25,000 casualties, but the U.S gave little attention to the incident. In fact, the U.S. subsequently invited Syrian President, Hafiz al-Assad who had ordered and supervised the action, into a relationship with the U.S. (Lewis, 2001). Additional resentment also resulted from the massacre of 700-800 Palestinian refugees in Sabra and Shatila in Lebanon that very same year. Immediately following, the commander of Israeli forces in Lebanon, Ariel Sharon (the current Prime Minister of Israel), was forced to resign for not responding to the slaughter (Lewis, 2001). While this helped to recognize the significance of the act to the Arab world, it also confused when the standard of crimes against humanity was applied. This left many to question how ruthless rulers like Assad and Saddam where somehow tolerated when the United States claims to be a defender of human rights.

2.1.2.3.3. The Failures of Modernization The final important factor fueling regional resentment of the Western is the failure of modernity. The Middle Eastern world has been plagued by failures in the modernization of the government, military, economy, and education. Despite the richness in oil in several Arab countries, many nations’ per-capita G.D.P. is only a fraction of Israel’s. This only acts as a sore reminder to many Arabs of their long struggle to compete (Lewis, 2001). Political modernization has also had numerous failures. Any attempts to establish more modern democratic styled governments have been undermined by persistent corruption and tyranny. It appears that the only forms of governments capable of maintaining order are oppressive dictatorships. Even then, no leader since Nasser has been able to unite the Arab world with significant success (Lewis, 2001).

23

Perhaps the most important failure of modern innovation is in the Middle Eastern education system. There are many that would argue that it is the highly developed education systems of the Western world that have led to such remarkable progress in personal opportunities and freedoms and living standards. This fact might also explain why economic and political modernization in the Arab world has had so many setbacks, which in turn further hinders the progress of educational reform. One of the best examples of this “double-edged” sword is the situation in Pakistan. In recent years, Pakistan has had a critical string of events prohibiting the advancement of a modern education system. Since its shaky beginnings in 1947, the predominately Muslim nation has relied heavily upon Islamic religious schools, known as madrasahs. With literacy estimated at only 40 percent, the lack of public schools particularly in rural regions has helped to ensure a place for madrasahs in the country’s education system (Stern, 2000). Originally, the government had limited influence over the madrasahs since they were funded by the zakat, which is the Islamic tithe collected by the state. Yet, in recent years, increasingly more schools are being funded privately. Without a dependency on the state, many of the schools have discarded math and the sciences and turned to preaching only narrow and extreme interpretations of Islam (Stern, 2000). Pakistani officials estimate that 10 to 15 percent of the 40,000 to 50,000 madrasahs teach militant forms of Islam. These schools have been known to identify jihad with guerilla warfare and have been the initiation grounds for many of the mujahideen (fighters in the Afghanistan war against the Soviets) and other Islamic fighters. Many graduates of these schools are directed to complete their “spiritual obligations” by fighting in Kashmir or other militant Islamic fronts like al-Qaida (Stern, 2000). The problems of limited government control have been recognized by Pakistani officials. Stern (2000) quotes Pakistani interior minister, Moinuddin Haider, “The brand of Islam they are teaching is not good for Pakistan…Some, in the garb of religious training, are busy fanning sectarian violence.” Yet many attempts to reform these schools have only been met with protest. An attempt to register the madrasahs with the government led only approximately 4,350 of the 40,000-50,000 schools to comply (Stern, 2000). In addition, recommendations to expand curriculum to include math and science garner the familiar claim that the government is only trying to undermine Islamic faith. These schools not only offer free education, but also provide free food, housing, and clothing. This not only increases the attractiveness of the schools to Pakistan’s impoverished youth, but also helps to create a sense of personal debt to the spiritual cause of the school. While not all the schools have ties to terrorist organizations, many encourage more zealous students to seek further “religious” training at terrorist camps. As a result, for many students, jihad has become a way of life. Stern (2000) quotes a mujahideen fighter of 19 years, “A person addicted to heroin can get off it if he really tries, but a mujahid cannot leave the jihad. I am spiritually addicted to jihad.” 24

Jihad does not only receive the attention of the poor, however, it is also supported by wealthy Islamic individuals and families in Pakistan and around the world. By branding the cause as a spiritual duty, many Muslims who do not necessarily support violence, unknowingly do so under the banner of jihad. While the poor are more likely to donate their children to the cause, the wealthy also help to fuel the fight with their money. There are also other issues in the making of a “jihad way of life.” Each fighter that becomes a martyr further adds to the appeal of supporting the cause. Stern (2000) makes the point, “when a boy becomes a martyr, thousands of people attend his funeral. Poor families become celebrities.” In families with large numbers of children, parents often expect that some will die of disease. Therefore, many parents have fewer reservations about donating a child to support what is perceived to be a holy cause. I would again like to make the point that not all Islamic organizations encourage violence. Regardless, few groups have made any attempt to distinguish themselves from those promoting terrorism. Because of this, many Muslims confuse any attempt to reign in the more extreme groups as a general attack on the faith (Stern, 2000). In fact, most identify terrorist violence as a response to secular encroachment, instead of as a specific Islamic agenda for taking control of the region. While many individuals might balk at the implication that peaceful people were unknowingly supporting violence, the strong mistrust of the Western secularist ideology seen in the Arabic Middle Eastern history helps to explain how such confusion occurs. The history of any region most certainly helps to explain how current events come to be. This brief overview of Arab history and their struggles in the modern world are used to help develop key decision structures within our model. In doing so, we hope to explain how terrorism has gained such a strong grounding in the Middle Eastern world.

2.1.3. Osama bin Laden and the al-Qaida Terrorist Network Thus far the information presented has described the origins of violence and the factors that lead religious fundamentalist groups to use terrorism. The final part of developing the background is then to examine the actual perpetrators of the defined terrorist violence. In this study, it is the terrorist network known as al-Qaida, which has come to the forefront of the world’s attention through the leadership of Osama bin Laden. Some people may wonder why Osama bin Laden has received so much attention when so many of his ideas and principles were adapted from various terrorist leaders in the Middle East. What makes bin Laden stand out as a leader is his role in the organization of the terrorist network al-Qaida. Up until his involvement, the numerous Islamic terrorist organizations of the region had acted fairly independently of one another. Yet, this all changed during the war in Afghanistan. For the first time Islamic leaders the world over were encouraging dedicated fighters to join the cause. Even the United State assisted the anti-Soviet fighters through the Pakistani Inter-Services Intelligence agency by providing a total of $2 billion in weapons and aid (Huang, 2002). 25

After the Soviet withdrawal in 1988, bin Laden’s involvement in the mekhtab al khidemat (MAK) or Afghan mujahadeen services office, led him to the formation of the al-Qaida (Alexander & Swetnam, 2001). From this point, a new course was set for Islamic terrorists and their actions in the world under the leadership of the now infamous Osama bin Laden.

2.1.3.1. The Terrorist Leader: Osama bin Laden Since bin Laden has played such an important role in the terrorism of this study it is necessary to take a brief look at his history. Born in Saudi Arabia in 1957 as the 17th son of 51 children of Muhammad bin Laden, Osama bin Laden was raised as a strict Islamist (Alexander & Swetnam, 2001). He was also born into wealth, receiving an estimated $200 million in inheritance from his father’s extremely successful construction group. At a young age, bin Laden was exposed to dedicated Islamic pilgrims that stayed as guests at his home (Alexander & Swetnam, 2001). Yet, his commitment to the faith reportedly did not begin until a later age when the family construction company took several contracts to rebuild the holy mosques in Mecca and Medina in 1973 (United States, 2002C). From this point, bin Laden began to develop a strong passion for Islam and Islamic law. Some reports indicate that at this early age, he was already preaching the necessity of armed struggle and monotheistic rule. Around 1980 bin Laden is reported to have received a civil engineering or public management degree (conflicting reports) from King Abdul-Aziz University (Alexander & Swetnam, 2001). During his time at the university bin Laden became acquainted with two well-known Islamic scholars: Abdullah Azzam, who later assisted with organizing efforts in the Afghanistan war and Mohammed Quttub. When the Soviet Union invaded Afghanistan in 1979, reports indicate that bin Laden made frequent trips to Afghanistan and Pakistan to meet with scholars and leaders who were raising support for the cause (Alexander & Swetnam, 2001). After graduation, bin Laden joined the mujahadeen (Holy warrior) movement that was fighting the Soviets, bringing large amounts of cash and construction equipment. He is also reported to have brought more fighters from the Islamic Salvation Front (ISF) group that he had formed in Saudi Arabia (United States, 2002C). In 1984, bin Laden took part in building a “guesthouse” in Peshawar, Pakistan, which served as a recruiting station for Islamic fighters. Reports indicate it was from this location that the MAK was founded (Alexander & Swetnam, 2001). Bin Laden had also been gaining a reputation as being a fearless fighter during one of the fiercest battles of the war, the siege of Jalalabad in 1986 (United States, 2002C). Around this same time, bin Laden left the MAK and began to build his own training camps and fighting force. In two years time bin Laden reportedly ran six camps and commanded 10,000-20,000 fighters known as the “Afghan Arabs” (United States,

26

2002C). With the help of ex-military advisers from Syria and Egypt, bin Laden gained valuable experience through several successful battles (Alexander & Swetnam, 2001). It was around 1988 that bin Laden is reported to have started a network to keep track of fighters for the purpose of staying in contact and helping friends and family reach loved ones. It is believed that this network, headquartered in Peshawar, Pakistan, was the beginnings of what is now known as al-Qaida (the base). Bin Laden returned to Saudi Arabia in 1989 after the Soviet withdrawal from Afghanistan. Reports indicate that he disbanded the ISF and returned to construction, receiving numerous government contracts in recognition of his efforts in Afghanistan. During this time bin Laden gave several dramatic speeches, which were recorded and sold in numbers as high as a quarter million cassettes (Untied States, 2002C). These speeches touted the victories in Afghanistan and made claims about failures in the Saudi government. Increasing Iraqi activity gave signs of a possible invasion, which bin Laden felt Saudi Arabia needed to be ready to defend against. Yet, during the invasion of Kuwait, the United States began to offer the Saudi government protection in exchange for cooperation, which infuriated bin Laden. In response, bin Laden, with the assistance of religious leaders, began to recruit rebels in opposition to the government for allowing the U.S. “infidels” into the Muslim Holy Lands (Alexander & Swetnam, 2001). Bin Laden’s activities did not go unnoticed, however, and the Saudi government promptly began to threaten him in an attempt to force him to cease his criticism (United States, 2002C). He then used his connections to flee to Pakistan, from where, after a brief period, he returned to Afghanistan. After several reported attempts on his life for his failed attempts to mediate between Afghan factions in the still warring country, bin Laden moved to Sudan in late 1991 (Alexander & Swetnam, 2001). He proceeded to start several new companies including a construction company, which he used to employ many of his former Afghan fighters. Bin Laden enjoyed great success with his businesses which include: the construction company, Al Hijra; an agriculture company, Themar al Mubaraka; a leather and hide company, Khartoum Tannery; the investment companies, Ladin International and Taba Investments; and a transportation company, Qudarat Transportation Company (Alexander & Swetnam, 2001). It is reported that bin Laden used this success to finance al-Qaida activities and as a front for transporting weapons and fighters. During this time, reports indicate that bin Laden moved al-Qaida’s headquarters from Peshawar to the Riyadh section of Khartoum. In December 1992, bin Laden assisted and claimed responsibility for the bombing of a hotel in Aden, Yemen in an attempt to kill U.S. soldiers en route to Somalia (United States, 2002C). It is believed that he was also involved in subsequent attacks against the World Trade Center and U.S. troops in Mogadishu, Somalia, in 1993. Also while in Sudan, it is believed that bin Laden was partially successful in uniting Shiite Muslims organizations under al-Qaida. Reports suggest that he met with 27

leaders of the terrorist organization Hezbollah, and even sent members of his group to train in their camps in Lebanon (Alexander & Swetnam, 2001). In response to bin Laden’s increasingly extremist attitudes, the Saudi government froze his local assets in 1993 and publicly revoked his citizenship in 1994 (Alexander & Swetnam, 2001). Bin Laden was also publicly denounced by family members around the same time. He then proceeded to form the Advice and Reform Council (ARC), which he and other al-Qaida members used to publish several statements that condemned western styled governments, including the Saudi Arabian government, in favor of Islamic rule. In 1996, under immense international pressure, the Sudanese government forced bin Laden to leave the country. He then relocated to Kandahar, Afghanistan where he was welcomed by the ruling Taliban. In August of that year, bin Laden issued his first of several fatwas calling for all Muslims to wage war against the U.S., claiming it to be a spiritual duty. He managed to set-up several terrorist training camps and is suspected to have financed and assisted numerous other al-Qaida terrorist operations. Bin Laden further strengthened his support from the Taliban by giving his oldest daughter to Mullah Muhammad Omar, the Taliban's leader, in marriage. In addition he married his fourth wife, who is reported to be related to a key Afghan leader (United States, 2002C). Since then, bin Laden has continued his campaign against western ideals in favor of a ruthless form of Islamic rule. The escalation in his extreme view of the world and suspected involvement in the September 11th, 2001 World Trade Towers attack has led the United States and allies to topple the Taliban in an attempt to bring an end to his reign of terror. While bin Laden is unique in many ways, there are still many individuals who have adopted his views and pose a severe threat to world security for years to come. His background provides unique insight into the circumstances that helped to form his extreme ideology. These factors must also be taken into consideration, in order to ensure that similar individuals do not rise to unleash comparable forms of terror.

2.1.3.1.1. The Declarations of Osama bin Laden Over the years, Osama bin Laden has made several highly publicized declarations, which explicitly state many of the objectives of al-Qaida and provide valuable insight into the organization and structure of motivation used. The references and claims he made also help to explain how he has gained support both financially and in recruitment of active members. When taken in the context provided in the section on Arabic Middle Eastern history and culture, many of bin Laden’s seemingly obscure references carry profound meaning to the large numbers of Muslims in the world. The first of these statements, or Bayans, was a twelve-page document in Arabic titled “The Declaration of War”. It was issued from the mountains of Afghanistan on August 26, 1996 as a final warning to the U.S. that if they did not leave the “Land of the two Holy Places,” military action would be taken. Many Muslim religious authorities 28

advised that every Muslim study the Bayan in detail, as it concerned not only those in Saudi “America” (Arabia), but those around the world (Bin Laden, 2001). He begins the Bayan with a traditional Muslim prayer asking Allah (God) for guidance and forgiveness for wrongdoings, and stating that he bears witness that there is no other god and no other message of god than that of the prophet Muhammad. He follows this opening with several passages reminding readers of their obligation to Allah, and His overseeing of all events (Bin Laden, 2001). Bin Laden proceeds to accuse the U.S. of conspiring with the “Zionist Crusaders” to bring suffering to the people of Islam around the world. He refers to the spilling of Muslim blood in Iraq, Palestine, Qana, Lebanon, Tajikistan, Burma, Cashmere, Assam, the Philippines, Fatani, Ogadin, Somalia, Eritrea, Chechnya, and Bosnia Herzegovina, claiming that the U.S., under the guise of the United Nations, prevented the people from gaining arms to defend themselves (Bin Laden, 2001). He next charges that these massacres are proof that U.S. and the Zionist Crusaders’ claims to be acting in the interest of human rights is a false cover for aggression. He states that this aggression is most recently exemplified by the occupation of the land of the two Holy Places. He then suggests that there is a “blessed awakening…sweeping the world” (Bin Laden, 2001), which is uncovering these charges, but that the crusaders have responded by killing and arresting those “truthful” scholars and students coming forward. He then lists several of the terrorist leaders recently arrested or killed (Bin Laden, 2001). Bin Laden then proceeds to advocate that a safe base is now available in the mountains of Afghanistan. He claims that from there, “the largest infidel military force (Russia) of the world was destroyed, and the myth of the super power was withered in front of the Mujahideen cries of Allahu Akbar (God is greater)” (Bin Laden, 2001). This claim is referred to numerous times throughout his statement. An interesting note, however is that not once does he refer to the approximately $3.5 billion provided by the U.S. and Saudi Arabia that made the victory possible (Stern, 2000). Bin Laden next berates the Saudi Arabian government for its support of U.S. interests, and for the allowing of U.S. military installations in Saudi Arabia. He also blames the Saudi government for a supposed deteriorating economy, inflation, increasing debt, and filling of jails with “prisoners”. He claims that these supposed failures are the result of the suspension of Islamic law in exchange for man-made law, as well as an inability to protect the country from U.S. interests (Bin Laden, 2001). He then refers to “the famous letter of Shawwaal” issued to the king in 1991 during the Gulf War and the “glorious Memorandum of Advice” delivered in 1992 (Bin Laden, 2001). He claims that both documents were true to Islam and that their subsequent refusal was proof of the alleged government corruption. From this point, bin Laden begins to develop and elaborate conspiracy between the Saudi government, the United States, and Israel. In it he details a “wicked” and “evil” plan to destroy the Muslim people 29

both economically and socially. Some of the more powerful accusations include: that the Saudi regime is not loyal to Islam alleging that the King wore a cross on his chest, that the regime is dependent on U.S. occupation in order to maintain power, and that the occupation was preplanned. The majority of the rest of this statement uses strong references, examples, and passages to further rally support for Jihad. He advocates for a boycott of American goods, and encourages every Muslim to actively support the cause. Bin Laden asks for financial, weapons, and equipment donations. He advises the current military personnel to remain loyal, as to not arouse suspicion. He asks law enforcement agents to ignore terrorist actions, and cover up evidence. Bin Laden encourages the women of Islam to motivate their sons, brothers, and husbands to join in the fight. He guarantees victory claiming that Allah is on their side, and that Americans are cowards. He supports these claims with references to several U.S. military withdrawals in instances where casualties where inflicted. He also rebukes U.S. claims of terrorists being cowards by asserting that “youths believe in what has been told by Allah and His messenger… about the greatness of the reward for the Mujahideen and Martyrs” (Bin Laden, 2001). Bin Laden uses a stirring tale of how two youths struck down a Pharaoh who allegedly abused Islam in order to exemplify a young terrorist’s commitment. He suggests that Allah will bless these martyrs and that each will be guaranteed a spot in paradise and showered with rewards. He also asserts numerous times that it is the moral duty of all Muslims to fight in the name of his cause. Bin Laden further justifies terrorist actions by blaming the U.S. for the Palestinians massacred in Lebanon in 1982. He also accuses the U.S. of causing the death of more than 600,000 Iraqi children as a result of sanctions. From this basis, he asserts that any treaties with the Arab people are now “null and void.” This statement and the others made by Osama bin Laden most certainly contain completely false accusations and appalling misinterpretations of U.S. actions and of the Muslim religion. Regardless, many of the claims are founded on actual events and passages that carry significant meaning to the Muslim world. Because of this, many supporters are drawn to the cause, especially impressionable youths. While each individual may not subscribe to Bin Laden’s extreme interpretations and methods, it only requires a small amount of identification by those in the region with his claims to prevent them from speaking against him. This silent tolerance of extremism is a problem the world over. Even in the United States, suspected racists, homophobes, and sexists are continually permitted to expound harmful rhetoric. While most simply choose to ignore or ridicule such talk, it only takes a single individual to act upon it. The great benefit in the case of the U.S. is that the media most often clarifies false claims and unfounded accusations. In the case of the Middle East, there is far less clarification of actual U.S. intentions and objectives. In fact, such support of U.S. policy would most likely result in accusations of conspiracy. It is through statements like these that Bin Laden has been able to garner such a substantial support base and receive the required tolerance and sympathy from the Arab 30

world. These statements also help to outline his objectives and goals, as well as his means for motivating supporters and group members. These factors are used later in the report to validate the structure and relationship of particular variables in the model, such as terrorists’ willingness to die and the likely reactions to anti-terrorist operations.

2.1.3.2. The Terrorist Network: Al-Qaida (The Base) Osama bin Laden’s history helps to explain how al-Qaida first came into being, but the network is far more than one man and one organization. Al-Qaida is comprised of both a central command as well as a loose coalition among approximately thirty-one Islamic terrorist or fundamentalist groups with similar goals. The central command structure is comprised of several councils and committees for policy, terrorist operations, business and financial issues, media, travel, and religious matters (Alexander & Swetnam, 2001). The network covers an estimated 55 countries and has hundreds of trained terrorists, “backed by many thousands of supporters” (Dannheisser, 2002). In addition, it is likely that there are hundreds of thousands of Muslims that are knowingly or unknowingly sympathetic to al-Qaida. These sympathizers either support the cause by making donations and purchasing propaganda, such as taped speeches, or tolerate the terrorist operations by not reporting suspicious activities and known terrorists to the authorities. Al-Qaida is supported financially by bin Laden’s personal fortune once estimated at $300 million, as well as suspected millions from other sources (Alexander & Swetnam, 2001). These other sources include profits from the numerous affiliated businesses, operative’s personal savings, and laundered money from donations, robberies, fraud, drug-trade, and other criminal activities. 2.1.3.2.1.1. The Objectives of al-Qaida

Many of the goals and objectives of al-Qaida have already appeared throughout the report. These objectives were clarified over time through the various statements and published documents. In summary the main terrorist objectives of Osama bin Laden and al-Qaida leaders in the approximate chronological establishment are: 1) To promote global jihad for the war in Afghanistan and other Islamic struggles. 2) To organize militant Islamic groups. 3) To undermine all secular governments in the Middle East, and replace them with a single government ruled by the Caliphs (Religious Leaders) and Shari’ah (Islamic Law). 4) To destabilize the United States Government and its ability to maintain international interests and domestic peace by killing American civilians and military personnel everywhere. 5) To expel all non-Muslims, particularly the U.S. military, from the Holy Lands. 6) To exterminate the “infidels or “unbelievers” using as much force as possible to bring about the defeat of factionalism.

31

2.1.3.2.1.2. A Chronology of Significant Events

This section provides a year to year summary of the important events surrounding the formation of the Al-Qaida terrorist network. Some of the information has already been reported, but a complete chronology is needed as a reference for the model. In addition, the key numerical data is further summarized in the next sections. In 1979 Soviet troops invaded Afghanistan creating a surge in the demand for Islamic fighters to join the mujahadeen (Alexander & Swetnam, 2001). By 1984, the MAK was established to aid in the recruitment of fighters. In the meantime the radical Islamic objective of overthrowing governments not in favor of Islamic law was exemplified with the 1981 assassination of Egyptian President Anwar el-Sadat. In 1986 bin Laden established a new training camp (Alexander & Swetnam, 2001). That same year a mujahadeen recruitment office was opened in New York City by Egyptian immigrant, Mustafa Shalabi. Then in 1988, bin Laden, Mohammed Atef, and Abu Ubaidah al Banshiri founded al-Qaida and established the first three objectives defined previously. After the Soviet Union withdrew from Afghanistan in 1989, bin Laden returned to Saudi Arabia and began to protest the Saudi government (Alexander & Swetnam, 2001). Egyptian Sheik Omar Abdel-Rahman is also identified as the religious leader of a group of approximately 10,000 veterans called the Arab-Afghans. In the years 1990 and 1991, Abdel-Rahman was placed under house arrest in Egypt, but then managed to escape to Sudan. He then received a visa from the U.S. embassy under the sponsorship of Shalabi and traveled to New York after a visit in Peshawar. Not long after arriving, one of Rahman’s followers assassinated a leader of the Jewish Defense League in New York City. A short while later Shalabi was murdered in Brooklyn. Also during these years, Iraq invaded Kuwait and the Gulf War began. In 1991 bin Laden relocated to Sudan and began to assist 300-480 Afghan war veterans to relocate there as well over the next three years (Alexander & Swetnam, 2001). The following year, civil war broke out in Afghanistan and al-Qaida called for attacks on the U.S. military in Yemen, Saudi Arabia, and the Horn of Africa (goal 4). Also during 1992, Ramzi Yousef traveled from Peshawar to New York under a false name and Ali Mohamed started to train the top al-Qaida commanders in a camp in Afghanistan. Near the years end a bomb exploded in a hotel in Yemen in an attempt to kill U.S. soldiers on the way to Somalia, but two tourists were killed instead. This is believed to be the first alQaida directed attack aimed at the U.S. Around the same time Atef began making frequent trips to Somalia to plan attacks against U.S. forces. Around 1993, Ali Mohamed began training more operatives in Sudan, and then traveled to Somalia to build camps for Somali tribes opposed to United Nations intervention (Alexander & Swetnam, 2001). Also early in 1993, six were killed and over a thousand injured in the World Trade Center bombing. Then in June, Pakistani Mir Amal Kansi, opened fire outside the CIA in the U.S. and killed two personnel and injured

32

three. The fall of that same year, eighteen U.S. soldiers were killed by al-Qaida trained persons in Somalia. Then in 1994 an al-Qaida operative named Mohamed Sadeek Odeh moved to Mombassa, Kenya and started a fishing business (Alexander & Swetnam, 2001). Around the same time Wadih el Hage, another operative, traveled from Sudan to Nairobi, Kenya and opened several businesses. It was also in this year that bin Laden was denounced by his eldest brother and his Saudi citizenship was revoked in response to his recent activities. In the meantime, the ARC was formed and Algerian Armed Islamic Group members received al-Qaida training. In 1995 eight people were killed by a bomb in the Paris Metro set by Armed Islamic Group (GIA) operatives (Alexander & Swetnam, 2001). Also early that year, Ramzi Yousef was arrested for his involvement in the World Trade Center bombing by Pakistani authorities and extradited to the U.S. to face charges. Then, in June, an assassination attempt was made on Egyptian President Hosni Mubarak while visiting Ethiopia. Later in the year, seven were killed when a bomb exploded outside the American-operated Saudi National Guard training center. Shortly after, four operatives associated with bin Laden were found responsible for the attack and executed. Early in 1996 an al-Qaida cofounder and military leader, Banshiri drowned when his ferry sank in Lake Victoria, Tanzania (Alexander & Swetnam, 2001). Later in that year, having been forced to leave Sudan, bin Laden relocated to Afghanistan and issued his first Bayan, which reiterated goals 1-3 and clearly established goals 4 and 5. Also, in June of that year a car bomb exploded outside of U.S. Air Force housing in Saudi Arabia, which killed 19 and injured over 500. The following year, four American oil workers were shot in Pakistan. Also in 1997, Yousef and Eyad Ismoil were convicted and sentenced to 240 years for planning the World Trade Center bombing. In February 1998, several militant groups supported another statement reiterating goals 1-5 (Alexander & Swetnam, 2001). Later in the year, bin Laden published a declaration that added the sixth goal and held a press conference that restated his intentions again. Seven operatives in possession of explosives were also arrested in Brussels and eight were later arrested in London. Then in August of 1998, two bombs exploded outside U.S. embassies in Kenya and Tanzania, killing 234 and wounding over 5,000. That same day, Odeh was arrested in Pakistan with a false passport. Shortly after he was turned over to FBI and admitted to being a member of al-Qaida, which he claimed to be responsible for the embassy bombings. Mohamed Rashed Daoud Al-Owhali was also arrested and extradited to the U.S. in connection to the bombing in Kenya. In retaliation for the bombings, President Clinton ordered a cruise missile strike against several camps in Afghanistan and a suspected bin Laden chemical weapons facility in Sudan. Then in September, U.S. intelligence prevented the bombing of an embassy in Uganda and 20 suspects were arrested in connection with that plot. Authorities in Britain also arrested seven suspected al-Qaida operatives in London. Finally, the U.S. Attorney General indicted bin Laden and Atef for suspected involvement in the bombings.

33

Then in January 1999, a terrorist attack on the U.S. consulates in Calcutta and Chennai, India was prevented when Syed Abu Nasir was arrested with explosives (Alexander & Swetnam, 2001). A month later three suspected al-Qaida associates were arrested by authorities in Uruguay. Also during that year, President Clinton, and later the UN Security Council, imposed sanctions on the Taliban in Afghanistan for harboring bin Laden. In response the Taliban publicly reiterated that bin Laden was a welcomed guest since the Soviet war. Eighteen additional operatives and associate terrorists were arrested in 1999, including Ahmed Ressam who was arrested in the U.S. with a large amount of explosives in his vehicle. Also in that year the U.S. received copies of a CD-ROM from Jordanian authorities that contained an al-Qaida training manual on how to conduct terrorism. In 2000, 32 al-Qaida operatives and associates were arrested around the world, in connection with suspected terrorist plots and previous acts (Alexander & Swetnam, 2001). In addition, in the courts of several countries 30 convictions were made along with 33 indictments, with 18 of them being in absentia. In October, 17 were killed and 39 were wounded when suicide bombers attack the USS Cole in a Yemen harbor. Shortly after, the UN passed Resolution 1333, which banned all military assistance to the Taliban and demanded that all terrorist training camps be closed and bin Laden turned over to the proper authorities. There were also numerous threats made against the U.S. and other nations combating terrorism by al-Qaida and affiliated groups. In the first half of 2001, 11 more al-Qaida operatives were arrested in connection to the Kenya and Cole bombings (Alexander & Swetnam, 2001). Court testimonies also contributed detailed information about the workings of al-Qaida and its connection to terrorist activities. In August, Zacarias Moussaoui was detained for suspicious inquiries into jetliner flight instruction. Then on September 11th, 19 terrorists hijacked four U.S. jetliners, which were then crashed into the World Trade Center Twin Towers, the Pentagon, and a field in Pennsylvania. In total, 3,365 are believed dead and approximately 6,291 were injured (some injuries happened after the collapse of the Towers). Shortly after the attacks, U.S. forces began bombing strikes and with the assistance of allied forces in Afghanistan and around the world have been able to secure a large part of the country. Hundreds of al-Qaida operatives and associated terrorists have been arrested, and several attacks thwarted. Since additional information is reported daily, this study will assume that a significant amount of activity will continue over the next few years. 2.1.3.2.1.3. Numerical Data

The output of a model is given in rates, quantities, and other numeric values as they change over time. It is therefore necessary to extract similar information from the history of the actual system for purposes of calibration. The following table summarizes some of the important yearly totals regarding the terrorism being modeled.

34

Year 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Goal Level 1 1 1 1 3 3 3 3 4 4 4 4 5 5 6 6 6 6

Terrorist Attacks 0 0 0 0 0 0 0 0 1 3 1 1 1 1 1 0 1 1*

Deaths 0 0 0 0 0 0 0 0 2 26 8 7 19 4 234 0 17 4338*

Injuries 0 0 0 0 0 0 0 0 0 1,003 0 0 500 5,000 0 39 6379*

Arrests 0 0 0 0 0 0 0 0 0 0 0 1 0 0 46 27 32 11*

Indictments 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 33 0*

Convictions 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 87 30 0*

Military Actions 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0*

Table 1 Numerical Data on Significant al-Qaida Events *Note: Year 2001 data reflects events up to and including the September 11th attack (Wikipedia, 2002).

(Alexander & Swetnam, 2001)

2.1.4. Critical Flaws in a Terrorist’s Mental Model The progression of this report was intended to develop a theorized pattern and structure for the use of violence by an extreme-religious-fundamentalist group (in the context of this study). I propose that when this theory is applied given the history, goals, and ideology of Osama bin Laden and the al-Qaida network, it becomes reasonable to conclude that if activities continued along the same course, massive killings or genocide would ensue. The modeling section of this report formally develops the structure of the system and explains the calibration of historical behavior and assumptions. As discussed in the calibration section, the results of the base run simulations were found to be consistent with the theorized continuum of destruction leading to mass killings. In conclusion, I suggest this course is the direct result of the critically flawed mental model of the real world perpetuated among terrorists, supporters, and sympathizers (in context). The primary flaws in the individual’s mental model involve issues in the misperception of an enemy and effectiveness of terrorist violence, as well as tendencies towards mass killing or genocide.

35

Events Changing the World

+

± Prevalence of Difficult Life Conditions

+ Level of Inherent Potential for Violence

Amount of Demanding Needs & Goals

Progress on Path of Destruction + +

+

R R Bystanders Willingness to Accept Violence + +

Amount of Leadership for Violence

+ Amount of Motivation for Violence

+ Number of Coping & Fulfilling Needs & Goals Activities +

+

Amount of Support for Violence +

Number of Activities+ Dehumanizing & Making of Sides +

Extent of Exploitation of the People & System Interaction

Figure 2 Group Violence Reinforcing Loops Causal Diagram

Recall the causal diagram of Staub’s (1989) structure behind extreme group violence shown above. The members of al-Qaida came from various backgrounds of difficult life conditions and subjected themselves to even worse through war and violence. In response, al-Qaida and bin Laden have developed needs and goals for coping with the world. They perceive the United States of America to be a “Great Satan”, or evil force aimed at the destruction of their way of life. As a result, they have followed the dehumanizing path and solidified the “us” versus “them” ideology. As the data indicates, al-Qaida has indeed followed a continuum of destruction for the past ten years. This progress is fairly evident in the increasing casualties and aggressiveness of the group’s goals. In addition there was most certainly a large culture that was sympathetic to many of al-Qaida’s cries of injustice due to the poor U.S. reputation as discussed in the Middle Eastern history section. It is highly likely that during the time of al-Qaida’s growth many bystanders were ignoring suspicious activities and known terrorists. The motivation, leadership, and exploitation of the religious system have encouraged undying faith and unyielding force. And finally, there is what most individuals in the U.S. agonize over; a most painful demonstration of the potential evil of human devastation. With such tragedy it is often hard to cope with the concept of evil, yet as stated before it is actions that are evil. While people hold the potential, only actions lead to destruction. If the mental model is changed, then the potential begins to fade, the actions stop, and a balance of peace can be restored. 36

A more specific look at the growth of al-Qaida is depicted in the following figure. This terrorist growth casual diagram provides a simplified view of the cycle that aided in building the al-Qaida network. The loops are reinforcing (as denoted by the R and directional arrow) since all factors move in the same direction. Also, as before, the link from “Changes in World over Time” is orange in order to emphasize the factor’s independence and inconstant increasing or decreasing effect on life conditions. Severity of Terrorist Act + R Changes in World over Time ±

Terrorist Anger

Difficult Life Conditions +

+

Response to Terrorism +

+

Students In Religious Schools

+

R

Terrorist Supporters

Terrorist Act +

+ +

Terrorists Prepared to Commit Acts

+

Planning Done

+

Figure 3 Terrorist Reinforcing Loops Causal Diagram

These concepts and other ideas will be explored in greater detail later in the report. For the time being however, the key point is that there is a highly flawed mental model being used by terrorists and, in a more limited sense, supporters and sympathizers. Any solutions to the problem of terrorism will also need to address the many factors involved in the model. Of course, there are two sides to this problem of terrorism. It is also necessary to look at the main opposition of the terrorism represented, the United States and allied nations.

37

2.2. The Opposition to Terrorism: The United States of America The terrorist group section described how the United States became the primary target of Osama bin Laden and the al-Qaida network. It is therefore necessary to examine the U.S. policy and actions in regards to bin Laden and the region from which he operated. Terrorism is a problem that the U.S. government has struggled with in many forms. From domestic attacks to international attacks on embassies, from “cyber” attacks to the Weapons of Mass Destruction (WMD) threat, the United States government has taken extensive measures to identify and disable the perpetrators of terrorism. With such a broad spectrum, many existing agencies have been involved in finding solutions, and several more have been added to help coordinate efforts. This section summarizes the proposed solutions and actions that were taken that are most relevant to the bin Laden “brand” of terrorism.

2.2.1. U.S. Legal Definitions and Policies on Terrorism The U.S. government is fully aware of the difficulty in formulating a universal definition of terrorism. Most government reports, however, use the definition provided in Title 22 of the United States Code, Section 2656f(d): •

• •

“The term "terrorism" means premeditated, politically motivated violence perpetrated against noncombatant/*/ targets by subnational groups or clandestine agents, usually intended to influence an audience. The term "international terrorism" means terrorism involving citizens or the territory of more than one country. The term "terrorist group" means any group practicing, or that has significant subgroups that practice, international terrorism. /*/ For purposes of this definition, the term "noncombatant" is interpreted to include, in addition to civilians, military personnel who at the time of the incident are unarmed or not on duty.” (United States, 2002B)

This is the formal definition that has primarily been used by U.S. agencies for determining terrorist incidents in analytical studies. Another useful definition that clarifies international terrorism is provided in Title 18 of the United States Code, Section 2331: 1) “The term “international terrorism” means activities that a. Involve violent acts or acts dangerous to human life that are a violation of the criminal laws of the United States or of any State, or that would be a criminal violation if committed within the jurisdiction of the United States or of any State; b. appear to be intendedi. to intimidate or coerce a civilian population; 38

ii. to influence the policy of the government by intimidation or coercion; or iii. to affect the conduct of the government by assassination or kidnapping; and c. occur primarily outside the territorial jurisdiction of the United States, or transcend national boundaries in terms of the means by which they are accomplished, the persons they appear intended to intimidate or coerce, or the locale in which the perpetrators operate or seek asylum” (Henderson, 2001) Some other relevant U.S. policy definitions are provided in other sections of the U.S. Code. These include the sections on sedition, which defines action or advocacy directed at overthrowing or destroying the government as unlawful. Title 18 of the United States Code, Sections 2384 describes a seditious conspiracy to be two or more persons planning to overthrow or destroy the government of the United States through the use of force or declaration of war (Henderson, 2001). Section 2385 of the same title describes perpetrators of sedition to be anyone that knowingly or willfully teaches or encourages violence aimed at overthrowing the government. Seditious activities include preaching, distribution of media and propaganda, and the organization of groups with the intention of promoting violence. This rather broad description has not generally been enforced against those who simply advocate the overthrow the government. Henderson (2001) notes, “during the 1950s and 1960s, the Supreme Court gradually broadened the protections afforded by the First Amendment in such a way that mere belief, advocacy, or discussion was not illegal.” Yet, recent trends in government anti-terrorism policies appear to suggest that these protections are still limited, especially when concerning the immigrants and aliens in the United States. For example, after the September 11th attacks, more than 5,000 requests for voluntary interviews were sent to aliens in the U.S. that were primarily from Middle Eastern countries. Despite being voluntary, it is likely that any suspicions aroused during the interviews or by those failing to respond could lead to deportation or the revoking of visas (Shepardson, 2002). Current definitions of terrorism as it is applied to aliens and immigrants are found in Section 1182 of the Immigration and Naturalization Act. This section identifies the following terrorist activities as grounds for the United States to block immigrants or deport aliens: I) “The highjacking or sabotage of any conveyance (including an aircraft, vessel, or vehicle). II) The seizing or detaining, and threatening to kill, injure, or continue to detain, another individual in order to compel a third person (including a governmental organization) to do or abstain from doing any act as an explicit or implicit condition for the release of the individual seized or detained. III) A violent attack upon an internationally protected person (as defined in section 1116(b) (4) of title 18) or upon the liberty of such a person. IV) An assassination. 39

V) The use of anya. biological agent, chemical agent, or nuclear weapon or device, or b. explosive or firearm (other than for mere personal monetary gain), with intent to endanger, directly or indirectly, the safety of one or more individuals or to cause substantial damage to property. VI) A threat, attempt, or conspiracy to do any of the foregoing.” (Henderson, 2001) In addition, engaging in terrorism is defined as the commitment of individual capacity in support of a terrorist activity. These activities include: I) “The preparation or planning of a terrorist activity. II) The gathering of information on potential targets for terrorist activity. III) The providing of about any type of material support, including a safe house, transportation, communications, funds, false identification, weapons, explosives, or training, to any individual the actor knows or has reason to believe has committed or plans to commit a terrorist activity. IV) The soliciting of funds or other things of value for terrorist activity or for any terrorist organization. V) The solicitation of any individual for membership in a terrorist organization, terrorist government, or to engage in a terrorist activity.” (Henderson, 2001) Another relevant document is the Antiterrorism and Effective Death Penalty Act of 1996. These revisions and additions to the U.S. code further defined terrorist activity and state sponsorship and also addressed issues for victims and the regulation of dangerous materials, weapons, and explosives. There is also a provision that was added to the 1997 Defense Appropriations Act (10 U.S.C. 382), which permits the U.S. military to respond to weapons of mass destruction attacks. The last relevant United States government document recognized in this study is the 1981 executive order (12333) signed by President Regan, which declared that the U.S. would not engage in assassination. This order does not appear to apply, however, under undefined circumstances such as missile or bombing attacks on Muammar al-Qaddafi, Osama bin Laden, and Saddam Hussein (Henderson, 2001). In addition to United State policies, numerous international treaties ratified by the U.S. are pertinent to the type of terrorism used by al-Qaida. These include protocols and provisions for addressing aviation hijackings, biological and chemical weapons, genocide and human rights, maritime protections, torture, and warfare. In general, these treaties prohibit the use of extreme violence in the interest of human rights. There are also several general anti-terrorism treaties that further prohibit terrorist activities like kidnapping, extortion, and financing terrorism (Henderson, 2001). 40

This long list of policy and treaty definitions and descriptions provides the legal foundation that has been used by the United States in combating Osama bin Laden and the al-Qaida network. Additional provisions have also been added or are in the process of being added that further extend government capabilities. New legislation includes special funds for security and the military, the authorization of using military force to respond to the aggressors of the September 11th attack, as well as a $40 billion supplemental military spending package (Center for Defense Information, 2002).

2.2.2. The Driving Factors of U.S. Policy The previous section described the legal framework in which the United States combats the al-Qaida form of terrorism. These bureaucratic definitions and policies are intended to prevent the abuse of U.S. government authority and to ensure consistent and justifiable actions against the perpetrators of terrorist violence. The primary reasons for such precautions are rooted in the rights provided by the Constitution as well as in the general U.S. ideology, which respects the sovereignty of individuals the world over. Based on this theory, in principle, an individual’s rights are therefore only revoked when he or she has wrongfully infringed upon the very same rights of others. In practice, however, the interpretation of this ideal is often fiercely debated. Yet, despite the problem of interpretation many people appear to acknowledge that the protection of individual sovereignty has been a worthwhile endeavor. In fact, it may not be unreasonable to argue that such a pursuit has significantly aided the tremendous advances in the United States’ standard of living and relative peace. The key point in this study, however, is not to argue whether one ideal is superior to others, but rather that these ideals help to explain the driving factors behind United States’ policies and actions. Just as with a terrorist group, United States’ citizens and leaders are motivated by mental models of the world. While there is likely a much greater diversity in U.S. mental models, most share a similar conviction to the described ideology. As a result, U.S. policies and actions also tend to reflect a common structure that exhibits reasonably predictable behavior. The next two sections describe the model sectors that were developed to represent two different responses to terrorist threats. These are the defensive policies implemented and the offensive actions taken. All responses are reactionary to the terrorist threat felt in casualties and damage. In other words, the United States responds when terrorists attack using both security and force. As expected, these responses are not all the same, which is another reason why models are helpful. For the time being however, only simple cause and effect relationships are identified in each of the sectors.

2.2.2.1. Defensive Policies When terrorists successfully strike a United States target, it is almost always the result of the limitations of the current security measures. As a result, in every instance, security measures are increased. Figure 4 is a simple causal diagram representing the U.S. defensive response to terrorism. The loop depicted has a balancing behavior since the 41

effect of defensive measures changes the “direction” of the loop by causing a decline in productivity. These measures help to reduce the productivity of terrorists by making it more difficult to acquire equipment, visas and passports, and other necessary resources. Security measures also increase the planning required, since it takes more preparation to overcome the security measures in place at the time of the act. For example, United States’ boarder controls were able to intercept Ahmed Ressam with 110 pounds of explosives in 1999 (Alexander & Swetnam, 2001). Other defensive measures include restricting access to dangerous materials and vulnerable targets. Perceived Threat

+

+ Defensive Measures

Terrorist Act B

+

-

Planning Done

Productivity +

Figure 4 Defensive Balancing Loop Causal Diagram

Despite the success of defensive policies, history has tragically proven that there are limitations to even the most well designed strategy. As Taylor (2001) points out, probability simply shows that given “a positive value P, then the chance of the event happening at least once, in N years to the future, is 1 – (1 – P)N which becomes closer and closer to 100% for increasing N, as long as P does not change.” Put another way, as long as terrorists can find a way around security measures or discover a new vulnerable target, there is a potential risk for another attack. While there is a great deal that can be done to improve current security within the United States, this fact also explains why other measures are required in combating terrorism.

2.2.2.2. Offensive Policies The United States uses offensive measures in an attempt to directly suppress or engage the perpetrators and supporters of terrorist violence. Offensive measures include military strikes, punitive sanctions, indictments and arrests, and CIA operations. In the case of al-Qaida, the U.S. has taken only a few significant measures. One is the cruise missile and bombing strikes ordered by President Clinton in 1998 in retaliation for the embassy bombings (Alexander & Swetnam, 2001). The targets were several suspected 42

bases of operation and a chemical plant that was believed to be under bin Laden’s control. Reports were unclear as to whether or not the offense had an impact on the groups’ operations. Some of the other important offensive measures include the embargo against Sudan used to help force bin Laden into the open in 1996 and the punitive sanctions levied against the Taliban in 1999 for harboring bin Laden (Alexander & Swetnam, 2001). Figure 5 is another causal diagram that demonstrates how the offensive measures act as a balancing loop to reduce the terrorist threat. The offensive approach is ideal in that it eliminates the most direct causal factor, yet in the instances of al-Qaida related terrorism, the U.S. has been unable to preemptively strike and prevent a terrorist attack. Instead, U.S. offensive measures have been primarily reactionary to terrorist attacks.

Perceived Threat Offensive Measures

+

+

Terrorist Act + B

Terrorist Supporters

Planning Done +

Terrorists Prepared to Commit Acts

+

Figure 5 Offensive Balancing Loop Causal Diagram

2.3. Initial System Dynamics insight into Terrorism The causal loop diagrams in previous sections helped to provide some initial insight into the dynamics of the terrorist group and United States policies over time, yet the loops so far have been shown in isolation. In the real world, these different loops continue to interact and have an impact on one another. Figure 6, shown below, provides a view of the primary connections between the different causal loops.

43

The diagram helps to clarify how the terrorist group (shown in red) might counteract the balancing loops representing U.S. policy (show in blue). Depending on which loop is dominant, the overall behavior of the system may either be escalating or declining. The relationships depicted are based on the material covered so far in this report as well as some of the ideas drawn from Richmond’s (2001 & 2002) “Story of the Month” exercises and Taylor’s (2001) policy study. Reviews of these short studies are located in Appendix C and Appendix D. The key links between the loops are the total casualties inflicted that create a threat which the U.S. acts upon, and the offensive measures which have undesirable impacts that feed terrorist’s hatred for the U.S. For example, the arrest of religious leaders like Sheik Abdel Rahman in 1995 was met in the Middle East with a degree of skepticism. Despite the evidence linking him to the World Trade Center bombing, statements by bin Laden (2001) have helped to generate anger by claiming that the arrest is only a part of the “crusader’s” conspiracy to oppress the Muslim people. Offensive Measures

+

+

R

Terrorist Anger B

Students In Religious Schools

+

Balancing Offense Loop + Terrorist Supporters

-

+

Perceived Threat +

Reinforcing Severity Loop

Total Casualties + Act Severity

+

Defensive Measures

R Terrorist Act Reinforcing Support Loop

+

Planning Done

+

Terrorists Prepared to Commit Acts

+

+

+

B Balancing Defense Loop -

Productivity

Figure 6 Terrorist Group, Defensive, and Offensive Causal Loop Diagram

While there is certainly a great deal more to the terrorist problem than what is shown, the goal at this stage is to simply begin to identify some of the cause and effect relationships in the system of terrorism. This concludes the background material that was covered in this study of terrorism. Now that a fairly complete understanding of the terrorism identified in this study has been developed, additional complexity can be explored in the modeling process.

44

3. Methodology Introduction There were two primary goals in this project. The first was to develop and test a system dynamics model that simulates how the bin Laden “brand” of terrorism came to become a problem in the world today. To achieve this goal, this study presented a detailed summary of the background of this “brand” of terrorism, covering expert analysis of both the psychological, social, and regional characteristics as well as the factors leading to group formation and terrorist attacks. The anti-terrorism policies and strategies used to combat this form of terrorism were also presented along with an analysis regarding the effectiveness of each. This information gathered was then used to build a detailed model of each sector of the system. The second goal of this study was to provide an analysis of what different individuals learn from the model in order to demonstrate the effectiveness of system dynamics as an education tool. The following sections describe how the research, modeling, and analysis processes were completed.

3.1. Research, Resources, and Processes This study began by doing extensive research on of the problem of terrorism and the related factors. A considerable amount of effort was put into sorting through the wealth of information available in order to provide a comprehensive view of the problem. Processes include literature reviews, modeling, expert consultation, and group forums.

3.1.1. Literature Reviews The keywords listed in Appendix B were used to perform database searches on the several different engines, which are also noted in the appendix. Relevant articles and book abstracts were then reviewed and those most relevant to the “brand” of terrorism being modeled were selected. The information gained from the literature reviews was then used to write the background sections and to establish a framework of the factors attributing to terrorist group formation. The information was also used to explore U.S. strategies for combating terrorism and to determine the effects on the system of terrorism. Additional information was collected to help establish the reference modes of the model’s behavior.

3.1.2. Modeling The modeling process was an iterative task of building, testing, evaluating, and refining. The information from the background was used to build a detailed model of each sector of the system. Each sector was first examined in isolation and then in interaction with the others to check for consistency with the expected behavior. Some of the steps taken include the problem statement, model boundaries, dynamic hypothesis, model building, and calibration. The complete details are provided in the next section.

3.1.3. Expert Consultation Several meetings were arranged for both creating the model and evaluating its robustness. Dr. Jim Lyneis, a very experienced modeler and professor of Worcester 45

Polytechnic Institute, provided much of the valuable feedback during the model development process. In addition, fellow student Varun Suri helped to provide some of the initial insights regarding the model’s structure.

3.1.4. Forums Forums were also used to review and revise the model, as well as to test its ability as an educational tool for teaching others about the problem of terrorism. To accomplish this task a presentation of the model was developed. Groups of individuals familiar with system dynamics were assembled first to provide feedback on the model as well as the technique of presenting it for others to learn from. Four additional forums were then held to evaluate the model’s effectiveness in conveying the dynamics of terrorism to others not familiar with system dynamics. The effects were measured using survey administered before and after the presentation.

3.1.5. Survey Database A small database was created for evaluating the surveys from the seminars. After all of the responses were entered and checked, several queries were run to extract information about the survey results and to conduct comparisons. The analysis and conclusions of the project were then formulated from this information.

3.1.6. Internet Site An Internet site was also developed to help facilitate information exchange. Design features include file sharing, discussion boards, and “chat” capabilities for online forums. This site can be accessed at: http://communities.msn.com/TerrorDynamics Due to the limited time available, this website was not utilized in the study, but instead will be used to share the report, model, and other information at the completion of the study.

3.2. Modeling the Dynamics of Terrorism Models of social problems like terrorism raise a number of important issues. As with any model of a complex system, specific assumptions must be made and justified in order to examine the problem in a more manageable form. The use of computers has greatly improved the level of complexity achievable, yet as always, a model by definition implies the existence of boundaries. In the case of this study, the model boundary has already been limited to the specific type of terrorism used by al-Qaida against the United States. In addition, there are many other simplifications in the model structure. These include the aggregation of different variables, graphical function converters, logic statements, and other key assumptions about the cause and effect structure of the terrorism problem. This section examines the details of the model in five sectors: The Terrorist Group, Defensive Measures, Offensive Measures, and Terrorist Anger. In addition, a few different policy structures were developed for representing possible future behavior. I wish to make it clear that the model is simply my interpretation of the numerous theories and studies regarding terrorism and extreme violence in this report. I have most 46

likely made a number of simplifications and debatable assumptions, despite the intensive research invested. The important point however is not to make strict expectations of the model, but rather to explore structure of causes and effects that are leading to the terrorist problem. The benefit of a model is that questionable assumptions can be changed in order to examine the impact on behavior. In other words, if you disagree with a relationship or value in the model, then a quick change is all that is required to see which arrangement reflects the most plausible and historical behavior. Also note that the second goal of this project was to evaluate the potential benefit of a system dynamics model for public education. For this reason, I have refrained from adding too much detail to the model in order to make it more understandable to those unfamiliar with system dynamics. I do, however, suggest some possible areas for extension of the model and additional studies in the final conclusions section.

3.2.1. The Problem Statement and Reference Mode System dynamics is intended to be used for modeling a complex, non-linear, system of differential equations, or in other words a problem. A modeling problem is simply a behavior that cannot currently be explained, or has many possible means of occurring. Non-linear systems are inherently problematic for this reason, and yet it is still possible to make rules or policies for individual relationships based on observations in the world. With these policies in place, specific behavior is generated, which then helps to provide insight as to how the system is driven. While it is possible to model linear systems, the numerical integration technique used with a system dynamics model’s equations makes other methods more appropriate. The importance of being able to solve a system of non-linear equations is that it makes it possible to examine the interactions of the numerous variables in a complex problem like terrorism. More specifically, the model that was developed is intended to simulate the escalation of terrorist violence against the United States by al-Qaida. This objective requires a behavior mode of reference in order to determine the required structure and calibration values of the model. Put another way, it is the historical observations of the changes in the exhibited problem and the related factors. The following Figure 7 is one of the behavior mode graphs that were used to build the model. The “Weighted Casualties” is calculated from the data in Table 1 in the terrorist group numerical data section.

47

Figure 7 Weighted Casualties per Year Graph

Though there are large variances in the severity of each attack, there appears to be a growth trend to the data. The escalation may also be measured by the more and more threatening statements and goals issued by al-Qaida. Additional reference modes include a growth trend in terrorists and terrorist supporters, and reactionary offensive and defensive measures that are scaled by the perceived threat of a terrorist act. These reference modes were then used to formulate a dynamic hypothesis about the problem and to help determine the model structure.

3.2.2. Dynamic Hypothesis and Base Structure The initial system dynamics insights discussed in the background helped to provide the foundation for developing the dynamic hypothesis. In considering the reference modes, it is likely that the reinforcing loops in Figure 8 have been the dominant cycle over the past ten years of al-Qaida’s growth. This figure is a simplified view of the stock flow structure of the model. Many equations and variables in each sector have been omitted for clarity. The main purpose of Figure 8 is to show the interconnections between the different sectors forming the loops: Defensive Balancing Loop, Offensive Balancing Loop, and Anger Reinforcing Loop. The following subsections review the details of each model sector, but it may be necessary to refer back to this figure to view the relationships. There are also iThink® “Storybook” programs that unfold the model structures discussed in this project available on the project web site. I highly recommend examining the model in that manner since it allows for live simulations and the changing of parameters. Finally, the complete base model equations are available in Appendix E. 48

Figure 8 Simplified Terrorist Group, Offensive, and Defensive Model

3.2.2.1. The Terrorist Group Figure 9 shown below is the model structure that was used to represent the alQaida terrorist group and their activities. First note that the items with question marks are those missing information from connections to other sectors. These links will be discussed later in the sectors that are providing the input. In general, the terrorist group sector tracks the number of terrorists and supporters in the system, as well as the planning that is done and then used for terrorist acts. To go through the sector, there are casualties, which are accumulated. These casualties have a specific severity based on some minimum and normal levels and are the result of the terrorist’s planning being used. The logic of how planning is used is simple. The “Ready” converter compares the planning that is done to what is required, and when the two are equal, the planning stock is emptied. Note that this makes for a discrete representation of terrorist acts. Though continuous equations are often more desirable in system dynamics, I believe in the case of al-Qaida’s behavior that a discrete formulation is more appropriate. What this means is that instead of having a continuous output of casualties being inflicted, there are periods with zero new casualties and then sudden spikes of casualties on a given model “day”. To return to the model, planning is 49

accomplished by people. In this case, an average number of terrorists are required to plan an act, and there is also a normal time to accomplish planning. As discovered in the background, the number of terrorists increased over the years. New members were recruited from religious schools like those in Pakistan and others became supporters who later joined the terrorists. Supporters also increase the productivity of planning by contributing money, equipment, and supplies, such as passports and visas, weapons, and sensitive information about potential targets. Normal Severity Al Qiada Terrorists Pepared to Commit Acts

Religious Students In Schools ? Religious Students Commiting to Al Qaida

Normal Fraction of Students Commiting to Al Qaida Maximum Planning

Al Qaida Supporters

New Al Qaida Supporters

Act Severity Terrorists Act on Planning

Time Required to Plan Planning Done

?

Normal Fraction of Supporters Joining

Normal Net Student Growth

Total Casualties

?

Terrorist Required for Planning Act

Net Growth in Religious Students in School

Casualties Inflicted

Minimum Severity

Planning Used

Planning Accomplished Effect of Al Qaida Supporters

? Al Qaida Terrorists Joining

Ready

?

~

Planning Required

Productivity Maximum Planning Requirement

?

Normal New Al Qaida Supporters

Normal Level of Al Qaida Supporters

Figure 9 Terrorist Group Model Sector

The effect of supporters, as compared to some normal level, changes as the number of supporters rises and falls. To express these changes, a graphical function converter is used. The graph drawn in Figure 10 shows what the function does as the number of supporters shrinks or grows. When there are no supporters the minimum value is 10%, and for large numbers of supporters, the maximum value is 250%. The reason “S” shaped growth seems appropriate is because when there are only a few supporters, increases in support are more likely to provide more significant productivity increases. Yet as support becomes more abundant it seems reasonable to assume that not all guns and dollars will be immediately put to use, and stockpiling (not expressed in the model) is probably occurring.

50

Figure 10 Effect of Supporters Graphical Function

3.2.2.2. Defensive Measures Figure 11 presents the model structure that was used to simulate the implementation of United States’ defensive measures and the subsequent effects on alQaida activity. As before in the causal diagram, casualties lead to a perceived threat in the United States, which lasts for a period of time before people become more comfortable again. In response to the threat perceived, citizens and policy makers become more willing to increase the level of security. These effects allow for the implementation of additional security measures. For example after the September 11th attack, fighter jets patrolled the skies and passengers were more willing to wait patiently in long lines at the airports. Yet again, there is a limit as to how large the security increases can be due to financial constraints and concerns over personal freedoms. Figure 12 shows the relationship between the threat level and some normal amount requiring an increase in defense. When there is no threat at all, then no new measures are introduced, and yet the defensive measures in place will not drop below a minimum, which is just the initial defense in the model. This logic is based on the fact that the U.S. is not likely to reduce security below what was already in place. For example, metal detectors aren’t going to be removed from airports even if the terrorist threat subsides. When there is a threat, the effect moves from 0 to 100% as shown in Figure 12.

51

Figure 11 Defensive Model Sector

Figure 12 Effect of Threat on Defense Graphical Function

With defensive measures in place, the productivity of al-Qaida is affected. Security checkpoints screen for terrorists identified by authorities and intelligence. Access to dangerous materials and weapons is restricted and monitored. These policies then help to reduce the group’s productivity in planning. At the same time the planning required to overcome the security measures is increased. The relationship of these effects 52

is again determined by a graphical function, which is shown in Figure 13. An important part of this assumption is that the maximum effect of defensive measures only goes to 10%. The reason for this is because no matter how tight security is, as Taylor (2001) pointed out, the terrorists are likely to eventually find away around the security measures or simply choose a different and more vulnerable target to attack.

Figure 13 Effect of Defense Graphical Function

3.2.2.3. Offensive Measures In addition to defensive measures, the United States also takes offensive action against individuals, groups, and countries that support terrorism. Actions include arrests, punitive sanctions, funds seizures, military strikes, CIA operations, and other types of activity aimed at reducing the terrorists’ tactical position. The structure used to provide these effects is shown below in Figure 14. As before, offensive measures are taken when a perceived threat is identified. Depending on the threat as compared to a normal level, a fraction of the maximum possible offensive attack is implemented. The relationship used in the model is shown in Figure 15.

53

Time to Implement Offense

Maximum US Offensive Measure

Effect of US Perceived Threat on Offense ~

Offensive Loop

US Offensive Measures Taken

US Offensive Measures in Progress

Normal Threat Level Requiring Offense

Normal US Offensive Measure

Effect of US Offensive Measures US Objectives Accomplised

Al Qaida Terrorists Killed or Captured

~

Normal Number of Terrorists Reduced

Minimum Time to Accomplish Offense ~ Effect of US Perceived Threat on Offense

US Perceived Threat

Al Qaida Terrorists Reduced

Al Qaida Supporters Al Qiada Terrorists Pepared to Commit Acts Al Qaida Supporters Reduced

Normal Number of Supporters Reduced

Figure 14 Offensive Model Sector

Figure 15 Effect of Threat on Offense Graphical Function

In the time that it takes for the objectives to be completed, terrorists and supporters are arrested or killed, depending on the effectiveness of the measure, which is calculated by again comparing it to a normal and using the function depicted in Figure 16. The reason for the initial jump from 0 to .25 represents the fact that offensive actions 54

have a certain minimal impact once implemented. The effect is also approaching a maximum since the U.S. is not likely to launch attacks that yield massive deaths.

Figure 16 Effect of Offense Graphical Function

3.2.2.4. Terrorist Anger The final sector of the model is the terrorist anger. This is perhaps the most influential part of the model since it drives both of the reinforcing loops in the system. The general function of this sector is to track the level of anger directed at the United States and the effects this anger has on other factors. Figure 17 shows how the anger sector was represented in the model. The key assumption is that there is an underlying (normal) dislike for the U.S., which maintains equilibrium when there is nothing disturbing the system. But when offensive measures are taken, the effect results in a change in the normal anger directed at the United States.

55

Normal US Offensive Measure

Normal Anger over US Policy

Anger Loop

Normal Time to Recover

Anger with US

US Offensive Measures in Progress

Recovering

Building Anger over US Policy

~

Effect of Anger with US

~ Effect of US Offensive Measures

Effect of Despair over Anger Level

Normal Level of Anger with US

~ Time for Terrorists to Become Disenchanted

Normal Severity

Minimum Severity

Act Severity Al Qaida Terrorists Disenchanted

Religious Students In Schools

Religious Students Commiting to Al Qaida

Al Qiada Terrorists Pepared to Commit Acts

Normal Fraction of Students Commiting to Al Qaida

Normal New Al Qaida Supporters

Time for Supporters to Become Disenchanted

Normal Fraction of Supporters Joining New Al Qaida Supporters

Al Qaida Terrorists Joining

Al Qaida Supporters

Al Qaida Supporters Disenchanted

Figure 17 Anger Model Sector

Anger does subside after offensive objectives are completed, yet if the level of anger becomes much greater than the normal, then the time it takes to recover can increase because of feelings of despair. Figure 18 presents the graphical function used for determining the effect of despair.

Figure 18 Effect of Despair Graphical Function

56

In addition affecting despair, anger also has an impact on several other factors. These include the act severity, supporters joining, students committing, and the rate of new supporters. These effects increase with the level of anger by the function shown in Figure 19. The last effect of anger is on the disenchantment of terrorists and supporters. Considering the commitment of al-Qaida terrorists it is not likely that members will become disenchanted with the fight unless changes lead to the U.S. no longer being perceived as an enemy. In the model this fact is expressed by using a logic statement that says that the disenchantment rate is zero unless the effect of anger is below a minimal amount. Under this circumstance, the members are not angry enough to be terrorists and the group collapses after a period of time. Note that in the base model case, it is not possible for anger to drop below the normal level because of the regular amount of anger directed at the U.S. This assumption is based on the historical observation of Middle Eastern perception of the U.S. As discussed in the background, the U.S. has had a significant history of sending mixed signals and pushing western ideals onto a society with very different values. Also, as discovered in the structure of group violence, given the right leadership a great deal of anger can be created and directed towards a group that is not strictly responsible for all the problems claimed. In addition, my research discovered very few U.S. policies for addressing the anger directed at the country. This issue highlights one of the key insights of the model, which is discussed in greater detail in the analysis section of the report.

Figure 19 Effect of Anger Graphical Function

3.2.3. Calibration and Simulated Data Fit The calibration of system dynamics models to real world data is an issue that has been debated. Disagreements particularly arise when the model includes non-physical variables, like anger or threat. Some suggest that the differences in individual interpretation of numerical scales of factors like anger make calibration rather ambiguous. This does not mean that the model is useless, but rather that the actual values simulated have no real meaning. The value of system dynamics models is in the “shape” 57

of the behavior (how the numbers change over time). Yet others argue that under the correct circumstances, models can make accurate forecasts. Regardless of these issues, it is clear that most people are more comfortable with values consistent with what they already know. In the case of an explanatory model like the terrorism model in this project, it is especially useful to reflect historical behavior. For this reason, the model building process in this study involved matching the reference mode extracted from the background as closely as possible. During the many base runs, each graphical shape and variable constant were incrementally changed in order to recreate a behavior consistent with the reference mode graphs. Using the parameters shown in Appendix D, the model simulation produces the output show in Figure 20. With a good fit for three of the four attacks, I concluded that the model was sufficiently calibrated for its simple role as an explanatory model.

Figure 20 Simulated and Real Weighted Casualties per Year Graph

3.3. Teaching the Driving Factors of Terrorism The second goal of this project was to evaluate the effectiveness of using a system dynamics model to educate others about the problem of terrorism. As described previously in the methodology, this was accomplished by holding several seminars. The first seminar was held for the purpose of evaluating the model and presentation. The next four seminars were held with volunteers who took a survey, watched a presentation of the model, and then repeated the survey. The survey results were then entered into a database for analyzing. 58

3.3.1. Seminars on Terrorism In order to teach others about the model and this project, a presentation was developed. This included several slides that provided background information as well as two “story” versions of the model that progressively developed the model structure. The first version developed the base run model reflecting the past ten years. The second version provided the policy analysis options for the next ten years. These files are also available on the project website. The presentation developed was then given in five different seminars. The purpose of the first seminar was to allow individuals experienced with system dynamics to critique the presentation and model. Five people attended the seminar and one individual unfamiliar with system dynamics remained in the lab during the presentation. In general, the response indicated that the model appeared to be well constructed for its purposes. The critiques also suggested more information about the different policy options and addressed other minor mistakes in the presentation slides. An interesting note was that the individual who remained in the lab was drawn to the presentation. At the end of the seminar he expressed his interest in what was shown and raised an important point regarding future terrorists and the overwhelming numbers of Arabic speaking youth in the region. He suggested that the peace making policies consider this fact, since most terrorists are fairly young and over the next ten years the large youth population will face the decision of supporting or condemning terrorism. The feedback provided was then used to further refine the model and presentation for the next sessions. The next four sessions were aimed at teaching others not familiar with system dynamics about the problem of terrorism. In order to attract volunteers, an email was sent to the campus and flyers were posted to advertise three forums open to anyone interested. Unfortunately only 5 individuals attended these sessions. An additional session was held with a convenience sample comprised of 33 members of the fraternity to which I belong. While this might not seem like an appropriate sample, I believe it actually provided a more ideal representation of differing opinions. Most of the individuals attending the other sessions were already interested in system dynamics or had spent a great deal of time thinking about the problem of terrorism. In the case of the fraternity, more individuals without such strong interests attended, which allowed the forum to be more educational. Overall every forum generated a very positive response. There were very few questions regarding the model structure and most people seemed quite comfortable with viewing the terrorist problem using the system dynamics approach. The complete findings are discussed in greater detail in the conclusions of this study.

3.3.2. Surveys and the Evaluation of Learning Surveys were used to examine what individuals knew about the problem before and after the model presentation. Eleven of the questions were given both before and after the presentation while eight others gathered additional information about the individual and his or her opinion of the presentation. An example of the survey is provided in Appendix E of the report. 59

Four of the survey questions were taken from a Zogby International (2002) poll to help compare each individual’s opinion to the approximate national average. The remaining questions were aimed at discovering what each person understood about the system of terrorism in this study and at collecting some demographic information. For example, questions were asked that gauged how successful the individual believed U.S. anti-terrorism polices will be in reducing the terrorist threat. The goal of the surveys was to measure the changes in opinion after the presentation in order to determine whether or not the individuals learned any of the insights provided by the model.

3.3.3. Analyzing the Survey Results In order to examine the data generated by the surveys in a more meaningful form, a small database was created for storing and retrieving the results. Figure 21 presents the database structure that was used. Several queries were then made to extract the survey results and information about the changes in opinion. Overall these preliminary results indicated that the model had been successful in generating the expected shift in thinking regarding the terrorist problem. The complete findings are discussed later in this report.

Figure 21 Database Relationship Diagram

In addition to examining the survey results, statistical tests were performed to evaluate the significance of the changes in responses of specific questions using a statistics program. In the case of numerical responses, a sample paired t test was used to check the hypothesis that findings were consistently significant. Questions that involved simple “yes/no” answers were checked using the binomial distribution to test the proportion of changed responses for significance. These findings are also presented in the next section of the report.

60

4. Analysis and Conclusions Introduction Approaching effective solutions to complex problems like terrorism is certainly something that cannot be accomplished in a short period of time. Yet, to be overwhelmed by such complexity only ensures efforts towards discovering better solutions to be a hopeless endeavor. For this reason many researchers focus on one aspect of the terrorist problem such as fundamentalism or the psychology involved. While these studies provide a great deal of useful information, it is often difficult to examine how all these factors work together, and how they develop over time. Moreover, each study is written in the context of a specific field, making it harder for experts from different fields to understand the findings of another. An even more important issue is that the large numbers of individuals in the world either supporting terrorism or combating it often do not have nearly the expertise required to benefit from the findings of such intensive studies. In response, different fields studying complexity have begun to contribute new tools for building and examining the terrorism problem (Smith, 2002). In the process of creating a model, a complex problem is simplified to a more manageable and understandable form. With the assistance of computers, researchers are now able to simulate a variety of complex behaviors. In this study, the tools provided by the field of system dynamics were used both to examine the problem of terrorism and to demonstrate the benefit of using models to teach others about the nature of a complex problem. In reaching the end of this study, I believe that a great deal of insight into the problem of terrorism was achieved and then passed on to others. This section details both what was discovered in the model and from the subsequent seminars.

4.1. Insights from the Model of Terrorism The intensive background study involved with this project does indeed help to describe how each individual factor contributes to or diminishes the terrorist problem. Yet, how the policies and factors identified interact and change in response to actions is not something that cannot be mentally simulated effectively. By building a computer model, careful thought can be put into developing a structure consistent with observed expectations. Given that the model structure contains reasonable assumptions, the running of numerous simulations can then provide a great deal of insight into the overall behavior. In addition, individual factors can be examined in detail in order to see how the system’s behavior is affecting them. The model in this study unveiled a number of different issues that appear to be driving the terrorist problem. These issues include the limitations of offensive and defensive measures, effects of anger, and terrorist inflows. Yet, uncovering the problems with policies should not be the sole aim of any study. In the science of problem solving it seems to be far easier to criticize the weaknesses of proposed solutions, rather than come up with a better alternative. For this reason, the following subsections examine several different policy alternatives and how they might influence the overall behavior of the system. 61

In order to evaluate the impact of policies implemented it was necessary to extend the model time boundaries to include the next ten years. By using a switch function in the model, the policy being tested could be enabled after the simulation ran the past ten years to bring the system to where it is thought to be today. Unfortunately the size of this project and the time available did not allow for more than two different policies. The first reflects how many individuals might interpret current United States policies to address the terrorist problem. The second policy offers some system dynamics insight into the solution requirements, suggesting that the inflows to the terrorist group be addressed using peaceful measures. In addition, an unchanged simulation of the twenty-year span is included for reference.

4.1.1. The Limitations of Offensive and Defensive Measures Before exploring policies in the model, it is necessary to examine what has driven the system in the past. One of the key issues with current anti-terrorist policies is the tendency to address only the symptoms of the terrorist problem, which are the terrorists and the terrorist attacks. While defensive measures help to decrease the likelihood of a successful attack, most experts resign that it is not probable to completely prevent every terrorist action. Offensive measures have a similar weakness in that even if every current terrorist is captured, killed, or discouraged, there still remains the question of why there were terrorists in the first place. This is by no means to say that offensive and defensive measures are not necessary in the struggle against terrorism. In fact, as Carr (2002) points out, the United State’s might have been more successful with such measures preventing an attack had they taken bin Laden’s war declaration more seriously. One reason why the U.S. was unable to respond is that, within the legal aspects of engaging violent aggressors, an individual declaration of war does not constitute a full-scale response. In other words, in dealing with transnational terrorism, the restrictions of criminal law do not allow for engagement without enough evidence. In addition to the legal issues, the offensive force used thus far raises several other potential problems. As Carr (2002) notes there appears to be an “undying American belief in the decisive effect of long-range destruction, particularly bombing campaigns.” While such tactics certainly offer far less risk for the loss of American life, the intended impacts of a strike consistently face great uncertainty. Such uncertainties include: the possibility of misinformation about targets, the potential for civilian or allied casualties (such as the Chinese embassy struck in Belgrade), and negligible impact on terrorist forces. Even, worse is that no matter how carefully targeted, the poor U.S. reputation in the Middle East often cause bombings to appear as bullish and as an unfair advantage, breeding only more contempt. These limitations will play an important role in determining the success of future policies. For this reason, the use of a model such as the one in this study can serve as an effective tool for testing and exploring numerous policies for combating the terrorist problem.

62

4.1.1.1. Expected Success of the “War on Terrorism” In a pole by Zogby (2002), 75.3% of Americans believe that the United States is headed in the right direction with the “war on terrorism.” Figure 22 shown below is the structure that was used to represent how I believe most people might imagine the war on terrorism to be successful. During a simulation, the model uses the base run structure for the past ten years. After then, the time switch essentially turns the effects of anger directed at the U.S. off. The reasoning behind this is based on the observation that many U.S. policy makers do not appear to fully understand how terrorism came to be a problem in the first place. In order for offensive and defensive measures to be successful independently, the reinforcing behavior must not be driving the system. In other words, the balancing loops must be more dominant of the system’s behavior. Running the simulation does indeed show that if the effects are absent, then terrorists are captured, and since there is no effect of anger with the U.S, members and supporters are also disenchanted. This disenchantment could also reflect the belief that current U.S. offensive strikes will dissuade others from becoming or remaining terrorists. Figure 23 shows how the number of terrorists drops off sharply and that no additional terrorist attacks are successful. Normal Anger over US Policy ? ~

Normal Time to Recover

Anger with US

Effect of US Offensive Measures

Recovering

Building Anger over US Policy Time Switch

~ Effect of Despair over Anger Level

Effect of Anger with US ~

Anger Loop Figure 22 Switched Off Anger Model Structure

63

Normal Level of Anger with US

Terrorist Activities 800 8,000 2

People Casualties/Month Plans/Month

600 6,000 1.5

People Casualties/Month Plans/Month

400 4,000 1

People Casualties/Month Plans/Month

200 2,000 0.5

People Casualties/Month Plans/Month

0 0 0

People Casualties/Month Plans/Month 0

24

48

72

96

120 144 Time (Month)

"Al-Qa'ida Terrorists Prepared to Commit Acts" : No Anger Casualties Inflicted : No Anger Planning Acomplished : No Anger

168

192

216

240

People Casualties/Month Plans/Month

Figure 23 No Anger Simulation Results

4.1.1.2. Potential Behavior of Unchanged Model In view of the historical reputation of the U.S. in the Middle East, however, it does not seem likely that past anger and anger generated from current U.S. offensive strikes will have no effect on the system. Running the model without any changes to the structure demonstrates the potential of the terrorist problem for spiraling out of control. The simulation output shown Figure 24 suggests that some new policies might be required in order to avoid even more catastrophic attacks fueled by growing despair and anger turning to hatred.

64

Terrorist Activity 20,000 60,000 4

People Casualties/Month Plans/Month

15,000 45,000 3

People Casualties/Month Plans/Month

10,000 30,000 2

People Casualties/Month Plans/Month

5,000 15,000 1

People Casualties/Month Plans/Month

0 0 0

People Casualties/Month Plans/Month 0

24

48

72

96

120 144 Time (Month)

"Al-Qaida Terrorists Prepared to Commit Acts" : No Changes Casualties Inflicted : No Changes Planning Acomplished : No Changes

168

192

216

240

People Casualties/Month Plans/Month

Figure 24 No Changes Simulation Results

Reviewing more of the output reveals how the problem tips out of control. Note how the Planning Accomplished drops after each attack when new defensive measures are implemented and terrorists are captured. Yet, around month 180, the effect of defensive measures reaches the maximum. At that point there are thousands of terrorists prepared to commit acts and the sheer numbers allow for frequent large-scale attacks. Figure 25 uncovers the factors pushing the escalation of violence. In the simulation, high levels of despair cause the recovery from anger to be driven down with each offensive action. This then causes anger to soar, in reaction to the U.S. operations in Afghanistan. As a result the next act is even more severe, causing the U.S. response to wage a fullscale offense that quickly turns into a sustained campaign. The reason for this is because the full-scale offense feeds back into the anger, which then spurs terrorist recruiting. After five years, the anger is so high that the thousands of terrorists required to overpower defensive measures is achieved.

65

High Anger Level 40 8 0.2 20

Anger Fraction Anger/Month Actions

30 6 0.15 15

Anger Fraction Anger/Month Actions

20 4 0.1 10

Anger Fraction Anger/Month Actions

10 2 0.05 5

Anger Fraction Anger/Month Actions

0 0 0 0

Anger Fraction Anger/Month Actions 0

24

48

72

96

120 144 Time (Month)

"Anger with U.S." : No Changes "Effect of Despair with U.S." : No Changes Recovering : No Changes "U.S. Offensive Measures in Progress" : No Changes

168

192

216

240

Anger Fraction Anger/Month Actions

Figure 25 No Changes Anger Problem

4.1.1.3. Policy Alternatives In the same poll by Zogby (2002) mentioned earlier, 80.4% of the respondents believed that it was somewhat likely to very likely that there would be another attack resulting in the loss of American lives. Perhaps even though, many individuals believe the U.S. is headed in the right direction, they are also skeptical that the right changes may not come about soon enough to prevent another attack. Also in the poll, 61.4% were in favor of changing U.S. foreign policy. In doing so, it may be possible to affect the anger directed at the United States that plays such an important role in the driving factors of the model. Figure 26 presents how the implementation of peace making policies might work to address the problem of anger in the model. Like before, the simulation uses the base model structure for the first ten years, after which, the peace policies are implemented to reduce both the normal anger and time it takes to recover by 75%.

66

Time Switch Normal Anger over US Policy

Peace Making Policies

? ~

Normal Time to Recover

Anger with US

Effect of US Offensive Measures

Recovering

Building Anger over US Policy

~ Effect of Despair over Anger Level

Effect of Anger with US

Normal Level of Anger with US

~

Anger Loop Figure 26 Switched Peace Making Policies Model Structure

Running a simulation of the model produces the output shown below. With the peace making policies in place, the anger with the U.S. subsides, and after approximately five years the terrorists become disenchanted with the fight since the U.S. is no longer perceived as an enemy. In addition, the defensive measures and offensive measures are also able to prevent another severe attack in the time it takes for the group to collapse. Terrorist Activity 2,000 8,000 2 800

People Casualties/Month Plans/Month People

1,500 6,000 1.5 600

People Casualties/Month Plans/Month People

1,000 4,000 1 400

People Casualties/Month Plans/Month People

500 2,000 0.5 200

People Casualties/Month Plans/Month People

0 0 0 0

People Casualties/Month Plans/Month People 0

24

48

72

96

120 Time (Month)

144

168

"Al-Qaida Terrorists Prepared to Commit Acts" : Peace Measures Casualties Inflicted : Peace Measures Planning Acomplished : Peace Measures "Al-Qaida Terrorists Killed or Captured" : Peace Measures

192

216

240

People Casualties/Month Plans/Month People

Figure 27 Peace Making Policies Simulation Results

For some reason discussion of peace making polices seems to carry a negative connotation with the U.S. public. For example as noted by Between the Lines (2002), “In covering a Sept. 29 peace demonstration in Washington D.C., the New York Times chose 67

this deliberately inflammatory and misleading headline: “Protesters Urge Peace With Terrorists.”” There are numerous other instances of individuals receiving death threats, sharp criticism, and accusations of being terrorist sympathizers for encouraging nonviolent measures to be used to help address the terrorist problem. While there are some demonstrators that suggest highly unrealistic peace alternatives, there are also numerous strategies seriously worth considering. To see what policies might be effective, it again helps to examine the terrorist problem from the very beginning. As discovered in the background, the al-Qaida network formed out of the U.S. supported war against the Soviet Union in Afghanistan. When the war was succeeded in pushing the Soviets out, bin Laden had learned what many Americans seem to believe is the only solution to the terrorist problem; violence can be successful in achieving goals. After the war, bin Laden’s focus initially landed on the Saudi government, but with the advent of the Gulf War, he shifted efforts to what he thought was the bigger problem, the United States. While the U.S. recognized the threat, the model suggests that all attempts to force the end of terrorism only inflamed the problem. If the assumptions about anger and despair seem reasonable, then the next logical step is to identify why offensive measures are only marginally successful at reducing the terrorist problem and highly inflammatory to the terrorists and supporters. Throughout the history of the Middle East, a strong mistrust of the United States has developed. As a result, ridiculous rumors, like those claiming that it was Jewish people in a conspiracy with the U.S. who pulled off the September 11th attacks have been able to gain reasonable credibility in the region. In addition, the “you are either with us or against us” attitude touted by the U.S. only risks further polarizing marginal terrorist supporters. This is especially true of the youth population. One of the sources of hatred identified in the background was the extreme religious schools in Pakistan that have been steadily supplying dedicated recruits filled with violent attitudes towards secularism and Americans. This issue underlines a key point for future terrorist problems. Currently approximately 44% of the population of Arabic Middle Eastern countries is under the age of 14 (Mena Report, 2002). This means that in another ten years there will be surge in individuals either in favor or in opposition to terrorism. Part of the reason why terrorism has become such a popular strategy is because it has appeared remarkably successful in countries like Pakistan (Kashmir Information Network, 2002). This has most certainly aided in the spread of the terrorist ideology. While the pressure is now on Pakistan to crack down on terrorism, there is again seems to be little attention regarding the source of the problem. It will require a shift in thinking from violence as the only way to hold power, to non-violent means for working through issues. With the youth in the Middle East carrying such an important role, one possibility might be the funding of student exchange programs. Since many of the problems between the U.S. and the region tend to be misunderstandings of one another’s culture, learning more about one another seems to be a natural way of developing more positive relations. In addition, youth that leave the Middle East will gain the opportunity to witness how 68

problems can be solved without the use of violence. Unfortunately the limits of this study do not allow for further exploration of new policies. Instead the focus of this project has been on establishing a better understanding of the problem, in order to bring about the interest in developing such peace making policies in addition to the required offensive and defensive policies already employed. Some suggestions about further research into this area are made at the end of the report.

4.2. System Dynamics as a Tool for Understanding Terrorism Exploring better policies for combating the problem of terrorism is really only a small part of the total solution. As with any other policy, no matter how well a proposed policy change is formulated, its success still depends on convincing the actors in the system that the change is necessary. In the case of terrorism, the individuals involved with the problem come from incredibly diverse backgrounds and can have strongly opposed perspectives of key issues. As a result, many policy changes are subjected to a strong resistance that either prevents the problem from being solved or actually makes it worse than before. For this reason, problem solving techniques like system dynamics not only have a great potential for providing insight and predictions, but also can be used to help teach others about how various policies influence the driving factors of the system. In this study it was discovered that there appeared to be a discrepancy in the American public’s opinion that the U.S. is on the right track for solving the terrorist problem and the likelihood of another attack. One reason for this may be that many Americans are expecting new polices to be developed. What doesn’t seem to be apparent, however, is the willingness to explore policies providing alternatives to the use of force. For some reason, patriotism and the use of force appear to go hand in hand. This connection is perhaps due to the strong desire to seek justice for those lost in the September, 11th attack. While this issue should certainly be a top priority, as discovered in the model, a complete solution to the terrorist problem will require additional measures to prevent the buildup future terrorists armed with an even more extreme agenda. Since the highly vocal U.S. public plays such an important role in government decision making, it is likely that more comprehensive policies will require some explanation in order to prevent public uproar. For this reason, this study used several group seminars to examine the usefulness of a system dynamics model for educating individuals about the problem of terrorism. As described in the methodology, surveys were administered before and after a presentation and then compared to see if opinions had changed. Despite the difficulties with getting a sample population, there appeared to be a clear shift in perspective for the overall group. Some of the key changes are summarized in Table 2 shown below. In addition, there was not a single question where the average response moved in the direction opposite to what the model suggested about the problem. In other words, the groups generally appear to have accepted every insight offered by the model.

69

Question Overall do you think the U.S. is headed in the right direction? Do you favor changing U.S. foreign policy toward the Middle East as a step against the war on terrorism? Do you favor attacking Iraq and removing Saddam as a step against the war on terrorism? Do you believe that support for similar terrorism against the U.S. will decrease or increase, given current policies? (1-10) If terrorists with an agenda similar to al-Qaida’s were to strike, do think it would be less or more severe than the attack on September 11th? (1-5) Do believe that current and planned U.S. policies for combating terrorism will require no change or need to be completely changed? (1-5)

Pre Post 55.3% 39.5% 73.7% 81.6% 47.4% 21.1% Mean 4.84 Mean 2.87 Mean 3.45

Mean 6.08 Mean 3.84 Mean 3.73

Table 2 Survey Analysis: Opinion Changes

Participants also showed an overwhelming support for using the system dynamics methodology to examine the terrorist problem. As seen in Table 3, the vast majority of the participants appear to have felt that the model provided information that was meaningful to them. Even though the presentation was only 45 minutes long, the surveys suggest that the model was able to provide a view of the terrorist problem that was very rich in information. Question In general, how informative was the material presented to you? SomewhatVery Informative Did you find the model to be a useful way of examining a complicated problem like terrorism? Somewhat-Very Useful Do you feel that you learned something about terrorism that you hadn’t known previously? Learned Some-Learned a Lot

Response 94.7% 97.4% 92.1%

Table 3 Survey Analysis: Model Acceptance

Yet despite the positive indications, after analyzing the results with a statistics program, the majority of the shifts in opinion for yes/no questions did not prove to be proportionally significant. Some of the tests regarding numerically scaled responses did show significance however, including those shown in Table 2 (the ones with mean values). The analysis of significance of the mean proportions of support for terrorism was 97%, 99.9% for the question on act severity, and 95% for the question about policy changes. One reason for why changes in opinion for other questions may not be as significant is that many of the participants were already giving responses consistent with the model insights prior to the presentation. As a result, many had little or no change in response after the seeing the model. It is also important to note that the all but one of the participants were males, with an average age of 20 years. Since Worcester Polytechnic Institute is a primarily scientific community, participants also had a higher level of technical education than most Americans. Regardless, it has been my experience throughout this project that given the right explanation everyone I have encountered has been able to follow at least some of the concepts presented in the model. Given additional time, I believe that further 70

interactive studies would have provided an even stronger and more statistically significant shift in view regarding the terrorist problem. If such a conclusion is reasonable, then it would also be realistic to conclude that system dynamics does indeed have the potential to serve as an educational tool for implementing and changing policies aimed at reducing the terrorist threat.

4.3. Final Conclusions and Recommendations In reaching the end of the study it has become clear that terrorism is most certainly a complex problem. Yet, by using tools like those provided by the field of system dynamics, valuable insight into the problem of terrorism can be achieved. In addition, the use of models allows for a great deal of information from many different fields to be presented in a complete and interconnected structure. For example, the majority of the information and concepts in the first 40 pages of this report were able to be expressed in a model that can be viewed on a single page. By centralizing the information, it is also fairly easy to make modifications to represent different assumptions about the system in order to discover those that appear to most consistent with historical observations. In the campaign against terrorist violence, the United States is faced with a great opportunity to develop more effective means for solving complex problems. As demonstrated time and again throughout history, the use of violence and force to solve problems has come at the cost of many lives. The growth of the terrorist problem over the past ten years suggests that many of these past policies proved to be unsuccessful in addressing the necessary issues. Instead of reverting back to older and more risky policies, the United States must push forward to new ideas and strategies that address true source of terrorist violence.

4.3.1. Areas for Additional Study It is my hope that this study was able to bring attention to the need for additional techniques for understanding complex problems like terrorism. Throughout the history of human civilization, mankind has repeatedly demonstrated a lack in ability to find effective solutions to dynamic issues. With the assistance of computers, fields like system dynamics are now able to develop highly detailed models that are capable of simulating almost any system imaginable. When used properly, models like the one in this study can provide powerful insights into the cause of different problematic behaviors. In the case of terrorism it is likely that additional studies will be required to develop more detailed models as well a number of small explanatory models for public education. The high level of aggregation in the model in this study, prevent little operational insights to solving the terrorist problem. Instead models like this tend to be more useful for developing a better understanding of the general driving factors. Having discovered the limitations of offensive and defensive measures, the effects of anger, and terrorist inflows, additional detail could be added to gain a better understanding of these key factors. Some specific areas that would likely need to be looked into include the evaluation different types of offensive measures, such as punitive sanctions, direct 71

engagement, and bombing campaign might benefit from further analysis. Since anger and misunderstandings also provide an environment for breeding terrorism, more specific policies would also seem appropriate to examine. As this study reached its end, the conflict between Palestinians and Israelis was showing signs similar to the predicted behavior of the unchanged model. When the levels of anger and perceived threat reached extremely high levels, the model moved into a more “tit-for-tat” type of behavior mode. This dangerous cycle appears to be acting out today, as there are reports nearly every week of Palestinian suicide bombings and Israeli military retaliation. I have heard numerous individuals say now that there appears to be no possible solution when both sides have such strong emotions. Yet, on both sides there is still talk of finding peace. Such comments sound like the perfect invitation for exploring new approaches like the one used in this study. By allowing both sides to develop models of the problem, I believe it would be possible to move discussion towards identifying more effective operational solutions. One of the key issues in the conflict is the immense feelings of despair. For many Palestinians, the suicide bombings are viewed as the only means for forcing a solution. With such shortsightedness it is not likely that either side will be able regain enough security to dissolve the anger and fear driving the problem. In addition, since some of the accusations made by bin Laden accused the U.S. conducting a Zionist crusade against Muslims, it would be in the interest of the United States to see the conflict resolved. Assistance in resolving the struggle just may provide the perfect remedy to past hostilities towards the U.S. The problem of terrorism is not new to the world, but its progression has reached a point that indicates that a more knowledgeable public will be required to find effective solutions. With terrorism’s global reach, these strategies must be understandable and acceptable to many groups with different or even conflicting backgrounds. It is my hope that this study helped to provide some initial ideas for how recent advances in complex problem solving can be put to use. In time, with faith, perseverance, and compassion, I believe that problems like terrorist violence will one day become just as obsolete as so many of the past problems facing mankind.

72

References “A Bird's Eye View of the Pakistani Terrorist Machinery.” Kashmir Information Network. 20 Apr. 2002 . "Afghanistan's Civilian Deaths Mount." BBC News 3 Jan. 2002. 4 Mar. 2002 . Alexander, Yonah, and Michael S. Swetnam. Usama bin Laden's al-Qaida: Profile of a Terrorist Network. Ardsley: Transnational Publishers, Inc., 2001. Appleby, R. Scott. Introduction. Spokesmen for the Despised. By Appleby. Chicago: The University of Chicago Press, 1997. 1-15. Armstrong, Karen. The Battle for God. New York: Alfred A. Knopf, 2000. Bandura, Albert. "Mechanisms of Moral Disengagement." Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind. Walter Reich, ed. Cambridge: Cambridge University Press, 1990. 161-91. Bin Laden, Usamah bin Muhammad. “Declaration of War against the Americans Occupying the Land of Two Holy Places”. 23 Aug. 1996. 16 Nov. 2001 . Unknown translation. Carr, Caleb. The Lessons of Terror. New York, 2002. Center for Defense Information. Terrorism Project. 6 Mar. 2002. 7 Mar. 2002 . Crayton, John W. "Terrorism and the Psychology of the Self." Perspectives on Terrorism. Lawrence Zelic Freedman and Yonah Alexander, eds. Wilmington: Scholarly Resources, 1983. 33-41. Crenshaw, Martha. Terrorism in Context. University Park: Pennsylvania State University Press, 1995. Dannheisser, Ralph. FBI Experts Say Hundreds of Trained al-Qaida Terrorists at Large. 18 Dec. 2001. U.S. Department of State. 12 Mar. 2002 . Dawisha, Adeed. "Arab Nationalism and Islamism: Competitive Past, Uncertain Future." International Studies Review 2.3 (2000): 79-90.

73

“Economic Factsheet.” Mena Report. 10 Apr. 2002 . Ezeldin, Ahmed Galal. Global Terrorism: An Overview. Chicago: The University of Illinois, 1991. Harris, Scott. “Media's Role during the Nation's Crisis: Investigative Reporting or Cheerleading?” Between the Lines. 20 Apr, 2002 . Henderson, Harry. Terrorism. New York: Facts on File, Inc., 2001. Hoffman, Bruce. Inside Terrorism. New York: Columbia University Press, 1998. Hoffman, Bruce. "The Mind of the Terrorist." Psychiatric Annals 29.6 (1999): 337-340 Huang, Reyko. Lessons from History: U.S. Policy Toward Afghanistan, 1978-2001. 5 Oct. 2001. Center for Defense Information. 9 Mar. 2002 . Johnson-Laird, Phil, and Ruth Byrne. Mental Models Website. 18 Dec. 2000. 4 Mar. 2002 . Knutson, Jeanne N. "The Terrorists' Dilemmas: Some Implicit Rules of the Game," Terrorism, 4, 1980, 195-222. “Latest Zogby Tracking Report: More Americans Believe Nation Headed in Right Direction.” Zogby International. 22 Mar. 2002. 29 Mar. 2002 . Lewis, Bernard. "The Revolt of Islam." The New Yorker 19 Nov. 2001: 50-63. Margolin, Joseph. "Psychological Perspectives in Terrorism." Terrorism: Interdisciplinary Perspectives. Yonah Alexander and Seymour Maxwell Finger, eds. New York: John Jay, 1977. 273-74. Marty, Martin E., and R. Scott Appleby, eds. Fundamentalisms and the State. Vol. 3. The Fundamentalism Project. Chicago, 1993. 462-490. Miller, John. “Usama Bin Ladin: American Soldiers Are Paper Tigers.” Middle East Forum (Dec. 1998). Online: 22 Oct. 2001. Phinney, David. "Mixing Business with Terror." ABC News (1998). Online: 22 Oct. 2001. 74

"Problem." Def. 1. The American Heritage Dictionary. 3 ed. New York: Dell Publishing, 1992. 659. Richmond, Barry, Steve Peterson, and Chris Soderquist. An Introduction to Systems Thinking. Hanover, 1997. Richmond, Barry. “A Systems Thinking Look at Terrorism.” 20 Nov. 2001 . Richmond, Barry. “Keynote Presentation at 2001 STiA Conference.” 8 Feb. 2002 . “September 11, 2001 Terrorist Attack/Casualties.” Wikipedia. 29 Mar. 2002. 2 Apr. 2002 Shaw, Eric D. "Political Terrorists: Dangers of Diagnosis and an Alternative to the Psychopathology Model," International Journal of Law and Psychiatry, 8, 1986, 359-68. Shepardson, David. “More Than 150 Men Agree to Terror Probe Interviews.” The Detroit News (4 Dec. 2001). Online: 18 Apr. 2002 Smith, Roger. “Counter Terrorism Simulation: A New Breed of Federation.” Mar. 2002. 20 Apr. 2002. < http://www.modelbenders.com/papers/siw_s2002/> Staub, Ervin. The Roots of Evil. Cambridge: Cambridge University Press, 1989. Sterman, John D. Business Dynamics. Boston: Irwin McGraw-Hill, 2000. Stern, Jessica. "Pakistan's Jihad Culture." Foreign Affairs 79.6 (2000): 115-126. Stohl, Michael. The Politics of Terrorism. New York: M. Dekker, 1988. Taylor, Ivan. “Policy Instruments in the War on Terrorism.” 10 Dec. 2001 . United States. Department of State. Patterns of Global Terrorism, 1999. 10 Jan. 2002A United States. Department of State. Patterns of Global Terrorism, 2000. 10 Jan. 2002B .

75

United States. Federal Research Division. The Sociology and Psychology of Terrorism: Who Becomes a Terrorist and Why? By Rex A Hudson. Ed. Marilyn Majeska. Sep. 1999. 5 Mar. 2002C . United States. Congressional Research Service. Terrorism, the Future, and U.S. Policy. 2001.

76

Appendix A Systems Thinking and System Dynamics The field of system dynamics was first established in the early sixties by Jay Forrester while developing control theory for servo feedback devices. It has since grown to encompass nearly any system imaginable. Many of the principles of system dynamics are not new, but rather restated using a single jargon and analytical method of application. Other fields of study such as economics, psychology, business management, computer sciences, and field biology often have similar approaches to complex problems, but the resemblances are often masked by differing professional jargons. Early pioneers of the field have progressively developed a generally accepted methodology for modeling complex systems. These standards allow system dynamics to be applied consistently across any discipline, and provide a common language for discussing complex problems. Systems Thinking Applying system dynamics theory does not always have to be in the form of detailed models. One can often gain valuable insight into complex problems by simply taking a “systems thinking” approach. Systems thinking is a technique that was developed by Barry Richmond using the principles of system dynamics to provide a helpful set of practices for understanding and building models. Since this approach to the methodology tends to be less technical it will be used to provide an introduction to the field. Richmond (2001) summarizes systems thinking as “a set of eight thinking skills that enable people to: construct better mental models, simulate them more reliably, (and) communicate them more effectively.” These skills, or paradigms, he lists in the three subsets shown bellow in Figure 28.

Figure 28 Eight Thinking Skills (Richmond, 2001)

Thinking about the cause of a problem is obviously what individuals try to do when looking for a solution. However, in thinking of a “system as cause”, the focus is on identifying only the factors influencing the problematic behavior. In other words, all of the outside factors providing input into a system receive little attention even though they may be influencing the system. The reason for this is because they are beyond the control 77

of the system, and therefore cannot be changed. For example, the sun is outside the system of the earth. Since the sun cannot be turned off or blocked, only the factors on earth provide the system as cause for different behaviors. Another way of describing this concept is to identify the boundaries to a problem. In the example, the sun is beyond the boundaries of the earth system. Within a system, there can be many factors working together to produce highly diverse behaviors. This is called non-linear or dynamic behavior, since the results can be very different depending on the timing and condition of the interactions that take place. Dynamic complexity refers to changes that happen over time within a given system. A way to improve the process of identifying causes is to use “10,000 Meters Thinking.” Richmond (2001) describes this concept as “creating enough ‘distance’ from the specifics of an issue to see, in general, what’s going on.” This is especially true in instances of high emotions and conflicting backgrounds. The perspective or “vantage point” a problem is viewed from has a great impact on determining causes. Sometimes it is necessary to look at a system with a very focused and detailed view. Other times it is better to look from a “distance” in order to see the overall behavior. For example, you are having car trouble and aren’t sure what is causing the problem. If you were to only focus on troubleshooting each individual component it could take a very long time to discover the problem. However, if you were to look at the overall problematic behavior you might be able to determine if the issue was electrical or mechanical. From that point, though you would not be able to solve the issue. Rather, you have to return to a more detailed view in order to find the actual problem. An advantage of modeling is that it allows for both simultaneously. In modeling, a great amount of detail can be put into the individual components and then be viewed alongside with the overall behavior of the system. Operational thinking is another useful skill in the processes of representing the significant factors and the causal structure of a system. When applied properly, only the activities and items contributing to a behavior are represented. Incidental or correlated factors are excluded to prevent confusion and to get at the “root” of the problem. An example Richmond (1997) uses compares two milk forecasting models. One model, developed with proper statistical techniques, described milk production using macroeconomic variables that were historically correlated to milk sales. Yet, from an operational standpoint, only variables directly influencing milk production would be used, beginning with the number of dairy cows. It would also represent consumers and the consumption of milk. While both models might accurately predict actual milk sales, the operational model has an advantage in that it provides insight as to how milk is actually produced and sold (Richmond, 1997). In identifying the system as cause and operational factors, independence is established. Yet within in the system there are often many interdependencies between factors. In describing such connections, modelers often refer to “closing the loop.” Problems that persist over time indicate that a cycle is present. This is also called the feedback process. Even though there may be a chain of events leading to a particular outcome, as long as the interactions are in place, the process repeats. “Closing the loop” 78

requires connections to be made between factors in the process that influence one another. For example, an individual driving a car is a closed loop process. When the road bends, an individual determines that the wheel must be turned. Once around the bend, the wheel is straightened again. As time passed, the decisions that were made changed in response to the feedback received from the action observed. Put another way, in one cycle, the wheel was turned, and in the next it was straightened. One result of “closed-loop thinking” is nonlinear outcomes. That is, a change in X does not always cause Y to change at the same rate over time. In these cases, the exhibited behavior may change exponentially, oscillate, or follow other non-straight-line patterns. This type of thinking does not come naturally, which is another reason computer simulations are useful. Richmond (2001) suggests that simulation skills also include scientific or analytical thinking. Being able to assign quantities is a necessity for effective modeling. As a result, the processes of isolation, identification, and measurement used within the scientific community play an important role. Analytical thinking is also inherent to many of the other concepts discussed. In general, the use a consistent and reliable approach to problem solving is a requirement to any scientific method. System Dynamics Systems thinking has proven to be a useful technique for improving the understanding of many complex problems. Yet, in order to scientifically confirm the behavior of a system it is necessary to build a model. When it comes to complex problems, the human thought process is limited in its abilities to mentally simulate behavior. With the assistance of computers, the systems of non-linear differential equations involved with models of complex problems can now be approximated. In less technical terms, computers allow the many calculations required to simulate the behavior of a system to be done quickly and efficiently. The following material describes some of the limits to the human thought process and some explanation of how system dynamics models work. The Human Thought Process As a modeler, there are many things that must first be understood about the human thought process and the behavior of complex systems. Fortunately the learning process is in itself a complex system that we will use as an introduction. When thinking about what drives an individual’s behavior people often think of the outside influences. While an important factor, there is actually a continual loop between the person’s mental models and the outside world. This connection between the person’s thoughts and the real world is what we call a feedback loop. Figure 29 shows a simple diagram of the cycle contributing to individual behavior. The connecting lines in Figure 29 help to promote a sense of flow throughout the system. Note that there is no representation of any accumulation of knowledge or other variables. Also notice that the effect from one item to the next is not explained. As the

79

complexity of a model grows, additional facilities are added to capture the structure. For now however, this diagram will suffice for describing the decision making.

Figure 29 Simple Decision Making Process (Sterman, 2000)

The diagram in Figure 29 could be read as follows: The real world provides an individual with information feedback, which develops his or her mental models of real world. These mental models influence information feedback by manner of what is already understood through the models. For example, when an individual flips a light switch, his or her mental model expects a light to come on. If that does not happen then the feedback received (light not coming on) is interpreted differently. The individual then makes decisions about how he or she will behave in the real world, which are based on the information feedback and the strategy, structure, and decision rules developed from the mental models of real world. The cycle is completed and begins again with the person having an affect on the real world (Sterman, 2000). Figure 30 provides some additional detail of the problems that can arise as a person goes through the cycle. It is important to be aware of these factors when both creating and interpreting a model. While some individuals are better than others at thinking about complex systems, no one is capable of understanding more than a few functions simultaneously. The average human’s short-term memory can only contain seven items and can be reduced even further when distracted. Additional difficulties stem from our inability to compute numbers. This is especially true with exponential growth. For example when faced with the question, “If a Lilly pad on a pond doubles in size every day till it completely covers the pond, and it takes 30 days to do so, on what day is the pond half covered?” many respond something close to 15 days. This can be attributed to the tendency for people to think linearly, that is, incremental increases that remain constant. Furthermore when confronted with the calculus principles of integration and differentiation, even highly educated individuals perform poorly. The answer to the Lilly pad question is, on the day before.

80

Figure 30 Decision Making Process (Sterman, 2000)

Introduction to Modeling Having defined some of the difficulties encountered in the human thought process, Figure 31 shows how a virtual world can help in the decision making process. There are many advantages that a computer has over the human mind. Computers are not only better at numerical processing, but also provide an environment limited only by the amount of information entered. Through simulation, a computer can calculate complex interactions and provide valuable information as to how a system will function. Moreover, it can be done an infinite number of times to determine the best combination of actions. Of course, there are limitations to what a simulation can provide, but when used in addition to mental models individuals can significantly increase overall understanding of the system.

81

Figure 31 Decision Making Process with Virtual World (Sterman, 2000)

It is clear the use of a virtual model in addition to normal decision-making has its advantages, but it does not provide a solution to problems in itself. While a model may provide accurate and timely information, it is still up to the users and developers to interpret the behavior and implement the insights as decisions and policies. An important note is that any policy implementation also involves the people interacting within the system. As with any policy or plan, no matter how eloquent it is, without widespread support it will most likely fail or only partially succeed. This tendency is know as policy resistance. There are an infinite number of good plans that failed because they were poorly executed. Conversely there are a great number of horrifying policies that succeeded under precise implementation. Stocks, Flows, and Feedback Loops In knowing some of the limitations of the human mind and the possible benefits of using a virtual world, we can now go into further detail as to how a model works. There are three primary attributes to a system dynamics model: stocks, flows, and information links. A stock is the accumulation of any variable that can be quantified. A flow is the rate by which a stock is modified. Finally, informational links simply provide the state of a stock or a flow. While it may seem unlikely that only three features could possibly describe any complex system, if you begin to examine the state of any variable, 82

it becomes clear. For instance physics tells us that matter and energy are always conserved. Therefore, energy and matter are either stored in some object (a stock) or they being transferred to something else (a flow). If this is so, then why is there even a third attribute? In case of models of the real world, the rules determining behavior change as the system changes. Extending the physics example, in the case of ice, as heat is applied the temperature of the ice remains the same, as it turns into water. Yet after it has completely melted, the temperature begins to rise. While science can explain why this happens, in order to model it, information about the state of the water is required to determine whether or not heat melts frozen water, or warms it. In the case of more complex systems like human social systems, there may be many informational links involved. These links also lead the feedback “closed loop” behavior discussed earlier. To illustrate how these attributes work, there is an analogy frequently used by modelers. Imagine that you would like to take a bath, and you begin by initiating a flow by turning on the faucet. Your tub is acting as a stock, accumulating all of the water. You are disappointed to see that the water is not rising, and after a brief moment you realize you forgot to plug the drain. In that moment the water was running out through another flow, but since you want your tub full, you plug it. What has happened now is that the information you gathered from your stock (tub), has led you to change a flow (drain). This change as a result of information is the feedback loop. Note that while you wish for the tub to be full, there is no way to change its status except for altering the flow. Also notice that there is no physical change that led you to change the flow, but rather you inferred from the state of the stock that something needed to be done. To take the analogy further, and to explain another fundamental principle of system dynamics, imagine that you wish to go for a swim, but your pool is empty. Having learned your lesson you immediately ensure that the drain valve is closed and begin to fill your pool. After hours of waiting, the pool is only a quarter of the way full. Finally after the sun sets you give up and turn off the water. The pool has just created a delay. Delays are just one of the affects that stocks can have. Stocks also decouple flows and create inertia in systems. By having a basic understanding of these attributes and how they interact, one can begin to think about how they help to explain the dynamic behavior of systems. Beginning with Causal Loop Diagrams Often it is difficult to effectively develop a stock and flow model in the initial stages of a project. The use of causal loop diagrams helps to facilitate an understanding of the variables involved and their interactions. These are simple diagrams that use qualitative and operational factors with interlinking causal arrows. The arrows are used to show the flow of the system depicted. A symbol or letter is also used to denote the relationship from one factor to the next. These symbols indicate that the item moves in the same or opposite direction and are often + or S for same, and – or O for opposite. In addition, a “loop-gain” is also denoted in order to describe whether the loop is exhibiting reinforcing behavior, or balancing. Reinforcing behavior causes the system to move in a direction without bound, while balancing loops are self-correcting and move towards some goal or equilibrium. They are also sometimes called either positive or negative 83

feedback loops. For example, the loop on the left in Figure 32 is reinforcing since as more sheep are born, the more adults there are to mate, and thus more sheep are born. The loop on the right is balancing, however, since the more sheep there are the more wolves there are, but the wolves eat the sheep and as the herd is thinned, the fewer the wolves there can be.

Sheep Being Born +

+ Adult Sheep

R

B

Wolves +

Figure 32 Causal Loop Diagrams

It is important to not that causal loops do not keep track of any values and are therefore only used to help facilitate mental simulations of a system. For this reason, the stock and flow diagrams provide a richer view of a system, and the use of computers then allows for actual simulations to be run. Stock and Flow Models and Computer Symbols When system dynamics computer models were first built they were simply a list of differential equations. As the field grew, software developers added graphical features to help facilitate model building and to make it easier to examine how the model function. Figure 33 provides an example of what the symbols generally look like. Rectangular boxes are used to denote a stock, while the arrows with “valves” signify flows. The clouds on the ends of the flows signify that the source or sink of the flow is outside of the model boundaries, and therefore not an important part of the model. In addition there are also informational links that are simply arrows from one model factor to another. Converters and constants are also very common in modes. These are sometimes denoted by circles, but can also be just text. Constants provide some value that is used for calculations in the model, while converters perform intermediary calculations.

Water Flowing

Water in Bathtub

Water Draining

Figure 33 Stock and Flow Symbols

84

Appendix B Literature Keyword Searches & Databases Used I. Terrorism a. Actions b. International c. Models d. Osama (Usama) Bin Ladin e. Economic f. Middle-East g. Al-Qaida h. Groups i. Dynamics j. Political Influences k. Sponsors l. Funding m. Religion n. Drug Trade o. Human Rights p. Case Studies q. Psychology r. Leadership s. Group t. Recruiting II. Anti-Terrorism a. Policy b. Sanctions c. Diplomacy d. Foreign Aid e. Actions f. Security g. Prevention h. Organizations i. Alliances j. Military Strategies k. Human rights issues l. Intelligence m. Moderate Islamic/Muslims III. Psychology a. Religion b. Control c. Islam d. Behavior e. Survival 85

f. Learning g. Prejudice IV. Models a. Terrorism b. Leadership c. Group Dynamic d. Economic Stability e. Drug Trade Databases Used: 1) 2) 3) 4) 5) 6) 7) 8) 9) 10)

WorldCat ECO Social Science Index Newspaper Abstracts PsycFIRST ArticleFirst WilsonSelectPlus PIAS International Wilson Business AltPressIndex

86

Appendix C Analysis of Barry Richmond’s “A Systems Thinking Look at Terrorism” Following the events of September 11th, 2001, Barry Richmond (2001) of High Performance Systems, Inc. and notable developer of system dynamics theory produced a short presentation exploring terrorism and systems thinking. Richmond uses HPS’s software to help illustrate several models in an interactive format. Some insight is offered regarding the potential problems with current mental models. He also examines some of the structure leading up to a terrorist attack, and how it might be prevented. The last part of the presentation looks at how to eliminate the causes of terrorism. Richmond (2001) begins by suggesting that the current mental models behind the “war against terrorism” have a structure similar to Figure 34. This is presented using a stock and flow diagram that follows the definitions provided in Appendix A. The logic offered is as follows: A stock of terrorist acts comes from the inflow of acts, which are linked to the number of terrorists. The model assumes that the more terrorists there are the more acts there will be. The number of terrorist is therefore reduced through a flow of actions such as incarceration, death in conflict, or other means.

Figure 34 (Richmond, 2001)

The output of such a model suggests that by eliminating terrorists, the acts committed will also be prevented. Running the simulation then generates the output shown in Figure 35.

87

Figure 35: (Richmond, 2001)

Despite the appeal of such logic, Richmond (2001) argues that there are three additional assumptions that should be considered. 1) “US-led actions cannot completely eliminate the population of terrorists worldwide. As such, recruiting of new terrorist will continue. 2) Actions taken by the US will stimulate anger in some quarters. 3) Terrorist will capitalize on that anger to spur their recruiting efforts.” With this in mind, in Figure 36 he adds a stock of anger that builds from the total # of actions, which influences the productivity of recruiting efforts. This leads to an increase in the number of terrorists, through the relationship shown in Figure 37.

Figure 36 (Richmond, 2001)

This graphical relationship depicted below demonstrates how anger at US led actions begins by only marginally improving recruiting productivity, but as anger grows the productivity of recruiting increases rapidly. 88

Figure 37 (Richmond, 2001)

If the assumptions stated seem reasonable, then the dynamics lead to “an initial decline in the terrorist population and (the) volume of terrorist acts, (but) both rebound and begin growing exponentially.” This can be seen in the model output depicted in Figure 38, and can be explained “by the anger inadvertently created by US-led actions to reduce the population,” as seen in Figure 39.

Figure 38 (Richmond, 2001)

89

Figure 39 (Richmond, 2001)

It should be noted that the timeline and values generated in the model have not been calibrated to existing data. What is important, however, is that the general behavior is precise given the stated assumptions. This is not to say that in the real world the conditions will not change, but rather that if they do not, then it is not at all unlikely that terrorism will persist in the long-term. The insight to be gained is that perhaps a different perspective might find a more permanent solution to the “war on terrorism.” Richmond (2001) next explores what systems thinking reveals about the problem. Defensive Policies Richmond (2001) first examines how the likelihood of an act might be reduced. It should be noted that this section is not “intended to point any fingers,” as to why the events of September 11th happened, but rather to show that there are many preventable steps leading up to an attack. Figure 40 maps out the structure of the system leading up lives being lost, and offers measures that might have prevented each “flow” or transition from one stage to the next.

90

Figure 40 (Richmond, 2001)

While many of these ideas have already been implemented or put forward, describing an attack in this manner provides better insight as to how these measures are linked together operationally. By thinking of an attack as overcoming a system, it is clearer that the people involved must also be aware of each individual’s significance. This is particularly true the farther up the “chain” one is. Improving security and awareness, however, is only half the problem. Considering the number of potential routes attackers might take, it is important to also look at offensive measures that address where terrorism is coming from. Disagreement as a Cause Most would agree that all conflicts begin with some disagreement over some issue. In order for that to be possible it requires two individuals or “sides” to the issue. Richmond (2001) begins by looking at how opposing positions interact in Figure 41.

Figure 41: (Richmond, 2001)

91

The output of such a model is shown in Figure 42. For those unfamiliar with the modeling process, the variable “my (your) rate of adjustment” determines how quickly the position is adjusted. As long as it is positive, the level of disagreement will eventually go to zero. That is, some compromise will eventually be reached.

Figure 42: (Richmond, 2001)

Changing both adjustment rates to zero would reflect irreconcilable differences and would then have the output shown in Figure 43. As demonstrated, the “Level of Disagreement” never changes, and the issue is never resolved. Richmond (2001) then explores how the rates are influenced and what might make the problem worse or better.

92

Figure 43: (Richmond, 2001)

Figure 44 outlines the structure that can cause disagreements to get even worse. In this model Richmond adds stocks of anger and hatred. If hatred builds and leads to violent actions, additional anger can build leading to the reinforcing “cycle of violence.”

Figure 44: (Richmond, 2001)

The cycle does not persist however as long as there is a willingness to change positions and move towards and agreement, as shown in Figure 42. In this case the output of anger and violent activities would be that in Figure 45.

93

Figure 45: (Richmond, 2001)

While anger does build, the willingness to reach an agreement leads it to eventually dissipate, thus avoiding the escalation of violent actions. If both sides are not willing to find an agreement as shown in Figure 10, then the consequences can be more severe as shown in Figure 46.

Figure 46: (Richmond, 2001)

94

In this situations violence and anger “spirals completely off the charts.” Even for non-zero adjustment of position rates, if not sufficiently high, this behavior persists. Since not every disagreement follows this pattern, the structure is expanded to include the effects of tolerance in Figure 47, shown on the following page.

Figure 47: (Richmond, 2001)

As shown, tolerance can impact how hatred builds. When sufficiently high, the “cycle of violence” can be avoided. While this model oversimplifies some of the interactions, it does help to clarify what some of the dynamics of disagreement are. One concern is that the rate of adjustment in position is not affected by hatred. In many disagreements, a building hatred often reduces willingness to find a common ground. Richmond argues that this may not be necessary if tolerance is sufficiently high, since it prevents the cycle of violence from escalating. This would be a case of “agreeing to disagree.” The trouble with not formulating an interconnection, however, is that issues of irreconcilable differences often have inconsistent behavior that would most likely erode tolerance to the point where violence again spirals out of control. These additional dynamics will be explored in the course of our project.

95

Appendix D An Analysis of Ivan Taylor’s “Policy Instruments in the War on Terrorism” Ivan Taylor (2001) offered some insight in a short research paper that uses system dynamics theory to evaluate some of the current and future policy instruments for combating terrorism. The paper explores both US and Canadian counter-terrorism polices in the categories of “deterrence, prevention, control, containment and restoration.” Taylor uses some causal loop diagrams to demonstrate the dynamics of several polices concluding that “vigilance and reconstruction will win out in the end.” Taylor examines counter-terrorist polices in the logical progression of policy goals, beginning with deterrence. He identifies “two primary policy instruments in line with the deterrence goal: the threat of retaliatory military force; and punishment through law enforcement means.” In discussing the dynamics of retaliatory threat, Taylor refers to the causal diagram shown in Figure 48. He suggests that a reinforcing cycle dominates the policy, which only leads to an escalation of violence. This is supported with reference to the ongoing Palestinian and Israeli conflict, but concedes that massive retaliatory action may have a different impact.

Figure 48 (Taylor, 2001)

Taylor also explores some of the dynamics of criminal law enforcement, identifying some of the problems involved. Given the willingness of terrorists “to die for their cause,” he asserts that “the death penalty for their crime is no deterrent.” He also notes the problems with evidence and privacy rights, as well as the global debate on the humanity of executions. Taylor again uses a causal diagram (Figure 49) to demonstrate how the alternative of incarceration “may only lead to more terrorism.”

96

Figure 49: (Taylor, 2001)

The next logical policy goal identified was prevention. Taylor argues that while preventative measures have worthy objectives, the feasibility of complete prevention is unrealistic. An examination of the cycle demonstrates that as “one potential target is defended, the terrorist would simply move to another vulnerable target.” As a result, the probability of an event happening is not reduced. With this in mind, Taylor notes, “If the chance of an event in any year is a positive value P, then the chance of the event happening at least once, in N years into the future is 1-(1-P)N which becomes closer to 100% for increasing N.” In other words, as long as terrorists are motivated to take action, given enough time the likelihood of a terrorist event occurring is nearly certain. Taylor also uses Figure 50to demonstrate how this problem can be worsened by a balancing “cycle of preventive measures by government.” In this cycle, the time between incidents leads to overconfidence in the effectiveness of the measures taken, creating a false sense of security that prevents measures from being re-evaluated.

Figure 50 (Taylor, 2001)

In situations where deterrence and prevention are unsuccessful, Taylor identifies control and containment to be the next logical policy goals. These policies are aimed at reducing the impact of terrorist incidents and regaining the confidence of the public in government. By regaining control, government can “take the initiative away from the 97

terrorists,” demonstrating the futility of the actions taken. He states that communication, timing, and training are significant factors in regaining control following terrorist incidents. Containing the situation is also important to minimize economic and other long-term impacts. Taylor claims that the final stage in recovering from an incident is restoration. He believes that this process has the greatest potential for positive side effects. He suggests that foreign aid for educational, economic and public institutions may help to eradicate the poverty, ignorance, and hatred that encourages future terrorism. In closing, Taylor (2001) concludes that the dynamics show “the fallacy of ‘swift and effective retribution’ such as military force and punitive laws as the only policy instrument against terrorism.” He warns that “we must not allow ourselves to be lulled back into a state of complacency even if the instruments of military force, law enforcement and preventive intelligence result in a temporary reduction in the frequency of terrorist incidents.” Instead, it should be noted that the long-term solution to terrorism remains in reducing the likelihood of terrorist incidents, not in prevention and deterrence.

98

Appendix E Base Model Equations

99

100

101

102

Appendix F Survey Example Please take a few minutes to answer the following questions, before the presentation: *Note: All information provided is strictly anonymous.

1) About how much time would you say you voluntarily spend following the war on terrorism on television, in print, on the radio, or Internet news? (circle one) 1 Hour or Less

More Than 1 Hour

More Than 2 hours

Do Not Follow

2) Overall, do you think the U.S. is headed in the right direction or are things off on the wrong track? (circle one) Right Direction

Wrong Track

3) What do you think is the likelihood of another terrorist attack within U.S. borders in the near future resulting in loss of American lives? (circle one) Very Likely

Somewhat Likely

Somewhat Unlikely

Very Unlikely

4) Do you favor or oppose using each of the following steps against the war on terrorism: a. - Changing current US foreign policy toward the Middle East (circle one) Favor

Oppose

b. - Use of strategic nuclear weapons (circle one) Favor

Oppose

c. - Attacking Iraq and removing Saddam (circle one) Favor

Oppose

5) Before the attack on September 11, 2001, do you believe that experts had any indication that such an attack was possible? (circle one number) No Indication 1

Some Indication 3

2

4

Strong Indication 5

6) How effective do you believe current and future United States’ security measures will be in preventing future terrorist attacks? (circle one number) Prevent a Few Attacks 1

2

Prevent Most Attacks 3

4

Prevent All Attacks 5

7) How effective do you think current and future United States’ offensive measures (i.e. military actions, CIA operations, or punitive sanctions) will be in preventing future terrorist attacks? (circle one number) Prevent a Few Attacks 1

2

Prevent Most Attacks 3

103

4

Prevent All Attacks 5

8) If terrorists with an agenda similar to Al-Qa’ida’s were to strike again, how severe do you think the attack would be compared to the attack on September 11, 2001? (circle one number) Less Severe 1

Just as Severe 3

2

More Severe 5

4

9) Do you believe that support for similar terrorism against the United States will increase or decrease in the next 10 years? (circle one) Decrease 1

2

3

Remain the Same 4

5

6

Increase Some 7

8

Increase Dramatically 9 10

10) Do you believe current and planned United States’ policies for combating terrorism will require any change in order to be successful? (circle one number) No Change 1

Some Change 3

2

4

Completely Changed 5

11) If you do think that there will need to be some changes, what would you suggest? _____________________________________________________________________________________ _____________________________________________________________________________________ _____________________________________________________________________________________ 12) How familiar are you with the field of System Dynamics? (circle one number) Not at all Familiar 1 2

Not Very Familiar 3 4

Somewhat Familiar 5 6

7

Very Familiar 8

9

13) What is your major or occupation? _______________________________ 14) Please fill in the following: Age: ____

Gender: _____

Citizenship: _____

Thank you for completing the survey. Please return it me before the presentation starts.

104

Expert 10

Please take a few minutes to answer the following questions, after the presentation: *Note: All information provided is strictly anonymous.

1) Overall, do you think the U.S. is headed in the right direction or are things off on the wrong track? (circle one) Right Direction

Wrong Track

2) What do you think is the likelihood of another terrorist attack within U.S. borders in the near future resulting in loss of American lives? (circle one) Very Likely

Somewhat Likely

Somewhat Unlikely

Very Unlikely

3) Do you favor or oppose using each of the following steps against the war on terrorism: a. - Changing current US foreign policy toward the Middle East (circle one) Favor

Oppose

b. - Use of strategic nuclear weapons (circle one) Favor

Oppose

c. - Attacking Iraq and removing Saddam (circle one) Favor

Oppose

4) Before the attack on September 11, 2001, do you believe that experts had any indication that such an attack was possible? (circle one number) No Indication 1

Some Indication 3

2

4

Strong Indication 5

5) How effective do you believe current and future United States’ security measures will be in preventing future terrorist attacks? (circle one number) Prevent a Few Attacks 1

2

Prevent Most Attacks 3

4

Prevent All Attacks 5

6) How effective do you think current and future United States’ offensive measures (i.e. military actions, CIA operations, or punitive sanctions) will be in preventing future terrorist attacks? (circle one number) Prevent a Few Attacks 1

2

Prevent Most Attacks 3

4

Prevent All Attacks 5

7) If terrorists with a similar agenda were to strike again, how severe do you think the attack would be compared to the attack on September 11, 2001? (circle one number) Less Severe 1

2

Just as Severe 3

More Severe 4

105

5

8) Do you believe that support for terrorism against the United States will increase or decrease in the next 10 years? (circle one) Decrease 1

Remain the Same 3 4

2

5

6

Increase Some 7

8

Increase Dramatically 9 10

9) Do you believe current and planned United States’ policies for combating terrorism will require any change in order to be successful? (circle one number) No Change 1

Some Change 3

2

4

Completely Changed 5

10) If you do think that there will need to be some changes, what would you suggest? _____________________________________________________________________________________ _____________________________________________________________________________________ _____________________________________________________________________________________ The next few questions are about the presentation. 11) In general, how informative was the material presented to you? (circle one number) Not very Informative 1

2

Somewhat Informative 3

4

Very Informative 5

12) Did you find the model to be a useful way of examining a complicated problem like terrorism? (circle one number) Not very Useful 1

2

Somewhat Useful 3

4

Very Useful 5

13) Do feel that you learned something about terrorism that you hadn’t known previously? (circle one number) Learned Nothing 1

2

Learned Some 3

4

Learned a Lot 5

If you have any additional comments, please make them here: __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________ __________________________________________________________________________________

Thank you for your participation in this study. If are interested in seeing the results of this project feel free to email me at: [email protected] 106

107