ABSTRACT Title of Document: EXPLORING TRENDS IN THE TARGETING OF VIOLENCE IN IRAQ THROUGH THE LENS OF CONFLICT THEORY: MARCH 2003 TO MAY 2006 David Robert Foster, PHD, 2008 Directed By: Dr. Gary LaFree, Department of Criminology and Criminal Justice Elements from Weber?s Conflict theory are combined with a focus on legitimacy in order to predict the effects that major social and political events have on political violence in the Diyala province of Iraq. Using 39 months of time-series data from Diyala in an ARIMA analysis, this research determines the effects of eight major events on the targeting of violence across Diyala as well as against four specific target sectors: law enforcement, government, military and civilian. This research finds mixed support for conflict theory, although a clearer picture emerges when findings are summarized by target sector, hypothesis and by event. This model is particularly effective at explaining trends in political violence directed against civilian targets. Despite the suboptimal data used, this research provides support for conflict ability to offering insight into the causes and course of political violence in a nation undergoing rapid social and political change. EXPLORING TRENDS IN THE TARGETING OF VIOLENCE IN IRAQ THROUGH THE LENS OF CONFLICT THEORY: MARCH 2003 TO MAY 2006 By David Robert Foster Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2008 Advisory Committee: Dr. Gary LaFree, Chair Dr. Laura Dugan Dr. David Weisburd Dr. Jean McGloin Dr. George Quester ? Copyright by David Robert Foster 2008 ii Table of Contents Table of Contents.......................................................................................................... ii List of Tables ...............................................................................................................iii List of Figures............................................................................................................... v Chapter 1: Introduction................................................................................................. 1 Chapter 2: Theoretical Conceptualization .................................................................... 6 Chapter 3: Data and Methods ..................................................................................... 48 Chapter 4: Results....................................................................................................... 77 Chapter 5: Summary and Discussion........................................................................ 104 Appendices................................................................................................................ 130 Appendix One: History and Demographic Background of Iraq ............................... 131 Appendix Two: Analysis of the main groups in Iraq................................................ 142 Appendix Three: Additional descriptive statistics.................................................... 152 Appendix Four: Timeline of events .......................................................................... 156 Bibliography ............................................................................................................. 211 iii List of Tables Table One: Descriptive statistics by week and month for each sector ? page 51 Table Two: Number of time periods with no attacks ? page 52 Table Three: Event Hypotheses and Target Sectors ? page 69 Table Four: Model Specification Results ? page 78 Table Five: ARIMA Results for Province-wide Violence ? page 80 Table Six: Hypothesis results for province-wide violence reduction ? page 83 Table Seven: Hypothesis results for province-wide violence increases ? page 84 Table Eight: ARIMA Results for Law Enforcement Targets ? page 85 Table Nine: ARIMA results for Law Enforcement Targets ? page 88 Table Ten: Hypothesis results for law enforcement violence reduction ? page 89 Table Eleven: ARIMA Results for Military Targets ? page 89 Table Twelve: Hypothesis results for military violence ? page 93 Table Thirteen: ARIMA Results for Government Targets ? page 94 Table Fourteen: hypothesis results for government violence reduction ? page 97 Table Fifteen: Hypothesis results for government violence increase ? page 98 Table Sixteen: ARIMA Results for Civilian Targets ? page 98 Table Seventeen: Hypothesis testing for civilian violence decrease ? page 101 Table Eighteen: Hypothesis testing for civilian violence increase ? page 102 Table Nineteen: Concurrence between expected and actual effects by target sector ? page 105 Table Twenty: Comparison of expected and actual effects by hypothesis ? page 106 iv Table Twenty One: Comparison of expected and actual effects by event ? page 113 Table Twenty Two: Attack Type ? page 152 Table Twenty Three: Fatalities for military, civilian and terrorists ? page 153 Table Twenty Four: Location of Attacks ? page 154 Table Twenty Five: Groups claiming responsibility for attacks ? page 155 v List of Figures Figure One: Target-type databases drawn from the main database ? page 50 Figure Two: Attacks against all targets (monthly) ? page 54 Figure Three: Attacks against all Targets (weekly) ? page 54 Figure Four: Attacks against Government Targets (monthly) ? page 55 Figure Five: Attacks against Government Targets (weekly) ? page 55 Figure Six: Attacks against Law Enforcement Targets (monthly) ? page 56 Figure Seven: Attacks against Law Enforcement Targets (weekly) ? page 56 Figure Eight: Attacks against Military Targets (monthly) ? page 57 Figure Nine: Attacks against Military Targets (weekly) ? page 57 Figure Ten: Attacks against Civilian Targets (monthly) ? page 58 Figure Eleven: Attacks against Civilian Targets (weekly) ? page 58 Figure Twelve: Temporal Spacing of 8 Events ? page 62 Figure Thirteen: Timeline of Attacks against All Targets ? page 81 Figure Fourteen: Timeline of Attacks against Law Enforcement Targets ? page 86 Figure Fifteen: Timeline of Attacks against Military Targets ? page 91 Figure Sixteen: Timeline of Attacks against Government Targets ? page 95 Figure Seventeen: Timeline of Attacks against Civilian Targets ? page 100 1 Chapter 1: Introduction Widespread political violence in Iraq has resulted in thousands of deaths (Iraqi Body Count 2007), frustrated attempts at the installation of a democratic government (Baker & Hamilton 2006) and ripped apart the fabric of Iraqi society (Biddle 2006; Laird 2005). Despite the importance of understanding and ultimately bringing this violence to an end, very little is known about its causes. By examining this violence through the criminological lens of conflict theory, this research increases our understanding of trends in the targeting of violence in Iraq and how these trends are influenced by the changing political and social climate there. Using a time-series analysis of political violence, this research finds mixed support for the conflict theory-based proposition that political events that increase group tension or reduce a group?s legitimacy will increase the frequency of violence targeted against that group. Conflict theory was developed to explain the violence and disorder that often result from power imbalances between differentially stratified groups (Dahrendorf 1959; Weber 1919). Conflict theory posits that the unequal stratification of groups within society produces competition for the limited resources of power, status and wealth (Collins 1975; Vold 1958). Over time the stronger groups will come to dominate, using brute force, the law or shifts in cultural norms to reinforce their position at the expense of other groups (Black 1976; Chambliss & Seidman 1971). Legitimacy, an assessment of whether a group is perceived to have gained or used their power in a lawful or appropriate manner, is central to understanding group interaction and violence as power gained or used in an illegitimate manner is often violently contested by other groups (Collins 1975; Weber 1919). Although conflict theory enjoyed a period of popularity in 2 the United States in the 1960?s and 70?s, it has largely fallen from view in the last two decades (Berry 1997; Wright & Friedrichs 1998). Despite finding only mixed support for conflict theory, this research may provoke further interest in using a similar theoretical approach to understand acts of political violence in similar settings. Conflict theory is a part of the critical tradition, and likewise shares many of the strengths and weaknesses common to that school of thought. Critical theories focus on the inequality that exists within many group relationships, one of the most notable being the Marxist focus on the unequal economic relationship between demographically similar groups (Bonger 1916; Greenberg 1981; Marx 1858). Unlike many of the more specifically focused critical theories, one of the strengths of conflict theory lies in the applicability to a variety of environments of its assumptions about group interaction and the use and abuse of power (Dahrendorf 1959). Specific to this research, the main differences between the groups in Iraq are those of ethnicity, religion and geographical location (Dayaratna 2005; Hafez 2006). The group stratification, societal unrest and weakened legitimacy that serve as primary motivating factors in many conflict theories (Bonger 1916; Collins 1975; Dahrendorf 1959; Marx 1848; Vold 1958; Weber 1919) are immediately apparent within Iraq (Baker & Hamilton 2006; Hafez 2006). The range of groups engaged in violence in Iraq all share the same fundamental purpose in that they are in competition for the scarce resource of political power (Hafez 2006). Iraq is home to a diverse range of groups, including sizable populations of Sunni and Shia Arabs and Kurds, small numbers of foreign insurgents (Kreuger 2006; Neumann 2006), and the coalition forces comprised primarily of the United States and British military. Each group is seeking to influence the political apparatus in Iraq to further their 3 own ends (Hafez 2006). Some groups use voting and political maneuvering to secure resources while others initiate violent attacks against the government or other groups that threaten their claim to power. The strategic use of violence is perceived by these groups to be an effective means to contest legitimacy, enabling them to eliminate rivals, weaken the influence of the coalition and prove to the civilian population that their group ought to play a central role in Iraq?s future (Hafez 2006; Perl 2006). As these groups jockey for political power and influence, the appropriate conditions are met for a modern-day test of the applicability of conflict theory?s unique perspective on changing legitimacy and inter- group tension as the primary drivers of trends in political violence. Building on the central premise of conflict theory that competitive group interaction is the driving force behind crime and violence, in this research I posit that the competing interests of violent groups will be manifested in their choice of targets. This allows for the development of testable hypotheses that link conflict theory?s focus on competitive group interaction and the importance of legitimacy to trends in violence directed against four specific target sectors in Iraq: military, government, law enforcement and civilian. I focus first on a non-target specific province-wide analysis of violence before analyzing attacks directed against these target sectors. While the group-based focus of conflict theory implies that a group-based rather than target-based approach would be simpler and more direct, an empirical evaluation of the violence from a group-based perspective is made more difficult by the limited data identifying which groups are responsible for each attack. The database developed for this research linked a violent attack to a specific group in only 14% of the cases; too low a 4 number for a valid statistical analysis of this violence at the group level. This low level necessitated the specific focus on target sectors taken in this research. In line with conflict theory?s emphasis on the importance of the political sector (Dahrendorf 1959; Weber 1919), a substantial portion of the violence in Iraq today is best referred to as political violence, as attacks are meant to alter the political power balance between competing groups (Ruggiero 2005; Schmid & Jongman 2005). This power struggle supports the position that the factors most relevant to the selection of targets are events that alter the political climate in Iraq such as elections, changes in the governance structure and far-reaching military, religious or social changes that strengthen or weaken the legitimacy of the targeted groups. This research focuses on the relationship between the effects of these political changes and trends in the targeting of political violence in Diyala, one of the 18 provinces of Iraq. Diyala, located immediately to the northeast of Baghdad and sharing a border with Iran, was chosen due to its high rate of violence (Anderson 2007; Biddle 2006; Hafez 2006), its representation of all major ethnic and religious groups, as well as the immediate availability of unclassified data on levels of political violence within the province. A database of political violence from the beginning of the Second Gulf War in March of 2003 to May of 2006 in the Diyala province was developed from open-source material in order to allow for an empirical test of conflict theory as well as to increase our understanding of the changing levels of political violence. The underlying theoretical model introduced in this section may be summarized as follows: trends in the targeting of political violence in Iraq are a function of inter- group competition for scarce political resources and are mediated by the legitimacy of the 5 targets involved. A group's legitimacy, and therefore its likelihood of being targeted by political violence, may be altered by social or political events that alter how it is perceived by other groups involved. The following section further develops the concept of political violence by analyzing how it is defined and by reviewing research that links trends in political violence to government actions. I also provide a review of the critical perspective and conflict theory in particular and its connection and contribution to the study of crime. For those unfamiliar with the history, demographics and course of the war in Iraq, Appendix One at the end of this report provides a brief summary of historical events in Iraq and the course of the current war and Appendix Two covers the history and current status of the main demographic groups in Iraq. This information is placed as appendices at the end of this paper in order to not detract from the development of the theoretical and empirical foundations for this research. 6 Chapter 2: Theoretical Conceptualization In this chapter I review the relevant literature concerning political violence, government actions, critical and conflict theory, legitimacy and the selection of targets. The first section focuses on political violence and its relation to government action and reaction to that violence. The second section delves into the critical perspective, with an analysis of conflict theory, legitimacy and how research on political violence and conflict theory informs our understanding of the variables that could influence target selection. The third section describes the hypotheses linking political violence and legitimacy for each target sector that I derived from this theoretical model. It is important to note that a good deal has been written about political violence, conflict theory and legitimacy from a variety of academic disciplines. I concentrate here only on research that is directly relevant to the model being developed. Conceptual development: Political violence Political violence is violence that is meant to have a political effect or bring about a political change (Turk 2004, pg 273). Political violence covers a broad range of activities, including collective actions such as riots as well as bombings, insurgencies, political kidnappings and assassinations (Collier 2000; LaFree & Drass 1997; Olson 1971; Wilkinson 2000). It is differentiated from other crimes by its motive and not its method, in that it is not the form that the violence takes, but rather the purpose or expected effect of the violence that categorizes it as political. 7 Political violence, including much of the violence in contemporary Iraq, is frequently referred to by the western media as terrorism (Kilcullen 2005) although the exact definition of terrorism and how it connects with political violence is still subject to some debate (Hoffman 1998; LaFree & Dugan 2004; Turk 2004). Definitions of terrorism vary based on the frame of reference of the agency involved. Agencies and organizations define violence as it pertains to their mission; law enforcement agencies emphasize the illegality of the violent act while military organizations contrast their definition with that of regularly defined warfare. Despite these definitional issues, most groups use a definition in line with the following: ?terrorism is politically motivated violence perpetrated against noncombatant targets by subnational groups or clandestine agents? (Perl 2006, pg 31). 1 This definition clearly shows that one of the main distinguishing characteristics of terrorism is that it is politically-motivated violence. This research is focused exclusively on the political violence-terrorism nexus, as this is the form of political violence that is most relevant to the situation in Iraq. Despite its nominally more insidious overtones, it is important to note that terrorism, like all forms of political violence, is usually perceived by those who use it as a legitimate means of bringing about a political change (Cooper 2007; Gibbs 1989). McCauley (2006, pg. 46) refers to political violence as ?politics by other means?, suggesting that violence is one avenue available for creating political change. Understanding the decision to use violent rather than non-violent means to solve a political problem is essential to making sense of the phenomenon. Political violence is a 1 Even a relatively standard definition of terrorism such as this can cause confusion when one considers that much of the ?terrorist? violence in Iraq is directed against uniformed coalition soldiers, a group that many consider not to be a valid target of terrorist violence as presently defined (MIPT 2004). Despite this definitional uncertainty, the controversy over the definition of terrorism should not hinder research (Ranstorp 2006). 8 specific and forceful response to certain conditions of social existence that are deemed to be unacceptable (Parker 2007). Individuals or groups turn to political violence in order to address essentially political issues when conventional avenues of change are perceived as being less effective (Black 2004; Callaway & Harrelson-Stephens 2005; International Crisis Group 2006; LaFree & Dugan 2004). While many avenues for addressing political grievances exist, violence is often selected as the most appropriate method due to its perceived effectiveness (Crenshaw 1990; Hoffman 1998; Wilkinson 2000). Violence can bring about direct changes in a political outcome through assassinations or the closing of elections, or it can act indirectly to alter behavior through the spread of fear, mistrust or government-over reaction (Marighela 1970; Schmid & Jongman 2005). Well-planned political violence enables violent actors to affect an audience larger than just the immediate victims of the attack (Crenshaw 1990; Hoffman 1998), furthering its usefulness as a tactic. Literature Review: The connection between political violence and government actions. Governments rarely refrain from responding when their people are targeted by political violence. There is a good deal of literature about the relationship between government reactions and interventions and the amplification or reduction of political violence (Della Porta 1995; Parker 2007; Tucker 1997). This literature tracks the actions taken by governments to suppress political violence, insurgent movements or terrorist cells, comparing the political, social and military interventions with changes in the rate of violence. Reactions to political violence often take the form of an offensive or aggressive military or police action meant to capture or kill the attackers. A good deal of research 9 focuses on these actions, which often produce spectacular and widespread media coverage, increasing their visibility to academics and the public as a whole. Government reactions to political violence, specifically military reactions, are meant to show the resolve of the government to confront the problem head on (Tucker 1997). The research literature suggests that an excessive military or police response can be seen as repressive or heavy-handed by the general population, shifting popular support toward the violent actors and ultimately increasing levels of violence (Tucker, 1997). There is little empirical research on this topic, but case studies of the aggressive military and police actions taken by governments in dealing with terrorist or subversive movements in Northern Ireland, Libya, Israel, Italy, Quebec and Germany all increased violence rather than having the intended effect of its reduction (Parker 2007; Sandler, Tschirhart & McCauley 1983). Furthermore, while many governments prefer to maintain their monopoly on the use of force by taking aggressive actions against those who use force illegally, democracies walk a fine line in responding to terrorist violence in this fashion as they must avoid instituting policies that needlessly repress civilians or result in unacceptable levels of collateral damage (Parker 2007; Plous & Zimbardo 2004; Wilkinson 2000). Government overreaction is something that violent groups often specifically try to elicit (Marighela 1970; McCauley 2006) as an overreaction may strengthen anti- government support among the population, increasing the perceived legitimacy of the violent group?s cause. Violent groups often specifically base their operational centers in heavily populated areas, both to better blend in with the surroundings as well as to 10 decrease the likelihood that government actions against them will avoid collateral civilian damage (Roberts 2002). The United States has experienced being seen as overly heavy-handed in dealing with domestic terrorism (Wilkinson 2000). Following what was perceived as an excessive use of force to end the siege of the Branch Davidian compound in Waco, TX, anti-government leanings among right-wing radical groups flared. Timothy McVeigh?s bombing of the Federal building in Oklahoma City in 1997 was allegedly meant to strike back at the government for its perceived mishandling of the Waco siege (The Times 1997). How a government chooses to handle violent or potentially violent groups is closely related to the level of violent activity directed against that government. In the case of Northern Ireland, when the British forces were seen as having violated the rights of the citizens in Ireland the number of attacks against the British and the popular support for the struggle there increased (Dayaratna 2005; LaFree, Dugan & Korte 2008). The same effect can also be seen in the French experience in Algeria. As the French forces tried to stifle the growing resentment to their rule, the legitimacy of their presence fell as they were perceived to have overstepped their bounds as they attempted to secure the country, increasing the violence directed against them (Alexander & Keiger 2002). A government?s legitimacy may be reduced by engaging in poorly planned or over-zealous offensive operations against politically-motivated dissidents, and it is this reduced legitimacy that often increases popular support for the very perspective that the government is attempting to suppress. 11 Early research on Northern Ireland by Soule (1979) suggests that British actions increased rather than decreased violence between the British and Irish forces. As previously described, violence is thought to increase under the defiance hypothesis as a result of a popular reaction to overly restrictive limitations on personal freedoms or overly brutal acts of the police (Sherman 1990). While conventional wisdom suggests that increasing military or police presence in an area should decrease violence, more advanced research done by LaFree, Dugan & Korte (2008) that studied the British actions in Northern Ireland showed that each British intervention, with the exception of Operation Motorman, which was a form of troop surge, produced a defiance effect and increased political violence. Francisco (1996) updated the proposed relationship between state coercion and defiance by examining whether forceful state coercion would increase protest, a concept he refers to as backlash. 2 Francisco used data from Germany and Northern Ireland to test this link, and found strong support for the existence of a backlash effect following forceful government interventions. In a similar vein, research done by Callaway & Harrelson-Stephens (2005) found that when governments, such as Britain during its attempt to quell violence in Northern Ireland, instituted policies that were overly restrictive or coercive political violence increased. Campbell & Connolly (2003) suggest that government actions repress the freedoms of the innocent as well as the guilty and in doing so create a groundswell of anti-government feeling that can be manipulated by violent groups into support for their cause (see also Filkins 2002). Research often finds an immediate and short-term 2 See Pridemore & Freilich (2007) for research that finds no connection between perceived government coercion and backlash relating to violence surrounding abortion rights. 12 deterrent effect of military actions, as violent groups must hide from the increased military presence, restricting their actions, but also a long term increase in support for the violent group, ultimately correlated with an increased level of violence (Chalk 1998; Eppright 1997). Government actions that are less repressive do not usually elicit this response. By way of example, the installation of metal detectors in airports or the use of target-hardening strategies tend to have the expected deterrent effect (Collins 2004; Enders & Sandler 1993; Nevin 2003; Tucker 1997 3 ). Violence is not always connected to government actions, but sometimes occurs in reaction to government inaction. Weisburd (1989) offers a unique perspective on the relationship between government inaction and violent actions taken by Israeli settlers. He shows how groups of settlers launched their own vigilante actions against perceived Palestinian encroachment when they felt their government was not taking the appropriate steps to resolve their security concerns. By framing his argument in terms of violence being a collective deviant reaction to external attempts at control, Weisburd demonstrates that violence, in the form of restricted vigilantism is often used as a means of social control by groups who perceive their government as inadequately addressing their social concerns. While the literature on the connection between government actions and political violence is filled with a number of historical case study analyses, it is of only limited applicability to this research as none of the interventions selected for analysis in this research are direct government reactions to political violence. Furthermore, no research that I am aware of has conceptualized the action-reaction interplay between governments 3 Softer reactions to violence may be more successful, but that doesn?t mean they always work. See Tucker (1997) for a review of the ineffectiveness of economic sanctions and the US policy of not negotiating with terrorists. 13 and violent groups as linked explicitly to conflict theory, despite the fact that conflict theory appears to offer a useful explanation for the effects seen. This research is unique in that, unlike much of the work before it that focused on studying the effects of interventions meant to have an effect on levels of violence, this research focuses instead on political events that should not directly affect the levels of violence except through how they alter the legitimacy of the targets involved, or aggravate the tension between groups in competition with one another. 4 This research will contribute to the literature on political violence by testing the unintended effects of policy choices not directly linked to a government reaction to terrorism, an approach that is rare within the current discussion of terrorist violence. This review demonstrates that aggressive military or police tactics often produce a defiant rather than a deterrent effect on levels of political violence. Although this body of literature is not directly relevant to the hypotheses being tested in this research, it does offer support for the idea that violent groups do respond to the actions of the government and that many of these actions, intentional or not, can have a counterproductive effect on reducing levels of violence. Some events that I analyze, such as the bombing of an important Mosque by insurgents, were likely meant to increase the tension between groups and ultimately the violence; however these events are not specifically government-sanctioned interventions. This research will be the first of which I am aware to empirically measure whether political events that are not directly meant to reduce 4 While the interventions studied in this research were not meant to directly affect levels of violence, it is likely that at some level policy makers considered the affect of their actions on levels of violence. For example, holding a democratic election for internal governance is a potential strategy for winning the ?hearts and minds? of the Iraqis. I do not believe that the affect of these policies on political violence was at the forefront of the planner?s minds though, as the original intent of the Coalition was to hold these elections, and the violence that sprung up caught many of the planners by surprise. 14 violence have an effect on levels of violence, hypothesized to operate through their effect on the legitimacy of the target sector involved. Theoretical development: Criminology and political violence While a good deal has been written about political violence over the last two decades, the direct study of political violence, particularly terrorism, received little empirical attention from criminologists until after the events of September 11 th , 2001 (LaFree & Dugan 2004; Lum, Kennedy & Sherley 2006; Ranstorp 2006; Turk 2004). Before 9/11 there were only a small number of academics who devoted themselves exclusively to studying terrorism. Of the articles written on terrorism in major journals in the 1990?s, 83% were done by authors writing on the subject for the first time (Cronin 2006). In the field of psychology, more articles on terrorism were written between 2001 and 2004 than in all of the years before combined (Plous & Zimbardo 2004). Further limiting theoretical development, much of the published work on terrorism and political violence across a range of disciplines tended toward the philosophical or the descriptive rather than the evaluation of theories or the testing of assumptions (Lum, Kennedy & Sherley 2006). The absence of a rigorous and consistent emphasis has left researchers without a clear consensus on the best theoretical approach for analyzing this type of violence. The lack of theoretical attention paid to this phenomenon stems in part from the popular notion that crime and political violence are fundamentally different, restricting theoretical cross-pollination from criminology to the study of terrorism. This distinction is 15 unnecessary, as terrorism and political violence are both forms of illegal violent behavior and therefore de jure and de facto crimes themselves. While a range of academic disciplines are needed to fully understand the cause and course of terrorism, criminologists have substantial expertise in understanding the causation and control of violence and it is essential that they apply their knowledge to the problem of political violence. Additionally, criminology has its roots in the study of political violence, as many of the founders of the discipline were concerned with the political actions of early terrorists or anarchists and with violence that grew from periods of social change (Durkheim 1897; Merton 1968). One of the first criminologists, Cesare Beccaria (1764) argued that sedition was one of the most serious forms of crime, and he described these offenses as acts of ?lese majesty?, stating: ?every crime, even of a private nature, injures society, but it is not every crime that aims at its immediate destruction? (Beccaria 1764, pg 68). Criminology?s most important contribution to the study of political violence came in the form of conflict theory. Conflict theory asserts that social problems such as crime and violence are the result of conflict stemming from a power imbalance between social groups (Collins 1975; Doleschal & Klapmuts 1973; Sellin 1938; Taylor & Walton 1970; Taylor, Walton and Young 1973; Weber 1908). While the conflict perspective can be applied to a wide array of problems, conflict theory itself fits within the larger theoretical framework of critical theory. Before delving into conflict theory specifically, it is necessary to explore the development, growth and more recent decline in popularity of critical perspectives as a whole. Understanding the intricacies of critical criminology is 16 essential to understanding and evaluating the conflict framework that this research utilizes. It is important to note that the terms critical, conflict, radical and Marxist are frequently used interchangeably without appreciation for their subtle differences. In this review I explain the differences between the terms, the core ideas that each represents, and the most influential thinkers in each theoretical tradition. Furthermore, by presenting a fresh look at these theories, this research aims to breathe new life into an often- neglected field of study. This research is structured around the belief that despite its falling out of view in the United States, conflict theory is still a useful theoretical model to explain rates of political violence in countries undergoing rapid social and political change. As this research finds a moderate level of support for the conflict model as an explanation for changes in the trends in violence in Iraq, perhaps theoretical attention will return to conflict theory as a relevant explanation for political violence. Critical Criminology At its core, critical criminology focuses on the unequal interaction between differently stratified groups as the underlying cause of crime, disorder and social unrest (Thomas & O?Maolchatha 1989). Crime and violence result from the need to maintain or challenge the power imbalance that results from this stratification (Berry 1997; Bohm 1982; Dahrendorf 1959). Unlike consensus theories (Hagan 1977; Sutherland & Cressey 1960) that view society as being in agreement about what behaviors are fundamentally right and wrong, critical theory views the definition of crime and deviance as not universally agreed upon and therefore open to interpretation or challenge (Vold 1958). This view of society is 17 referred to as the conflict model, although this distinction can be particularly confusing in that critical theories were based on the conflict model of society, while the conflict model and conflict theory are two separate, though related ideas. Critical theory holds to a conflict model of society, while conflict theory is a sub-field of critical theory. Much of the philosophical work that supported the birth of the critical theories was a result of social changes during and after the industrial revolution (Aron 1970; Walsh & Ellis 2007). The rapid social changes that occurred in the post-industrial revolution world of the 19 th and early 20 th centuries helped raise new awareness of the importance of class, race and gender. Theorists from a variety of disciplines did much to develop the critical perspective during this time period, examining societies that were undergoing rapid social, economic or political change (Walsh & Ellis 2007). Within criminology, the idea that crime was a function of the competitive interaction between societal subgroups became a common theme for theoretical research in the three decades following World War II (Dahrendorf 1959; Taylor, Walton & Young 1973; Vold 1958). It is important to note that many critical theories are referred to as radical theories in that they seek to understand and ultimately address the root causes of social problems. 5 To their credit, critical and radical criminologists have often worked directly on implementing social changes and applying their ideas to the reduction of crime or inequality (Berry 1997). That being said, many critical theorists see a revolution or dramatic restructuring of the social structure as the only effective way to reduce crime. Another common conclusion of critical theory is that inequality, in whatever form it manifests itself is directly linked to crime. Michalowski (1985) argues that an 5 The Latin root of the word radical is radix, meaning ?root?, in this case the root cause, while, interestingly, radix is also the root of the word ?radish? (no pun intended). 18 individual?s actions result from their understanding of the world, which is based on their position within society. As inequality is experienced, not necessarily poverty but also social, ethnic and other forms, levels of involvement in crime and disorder tend to increase. Inequality, particularly when manifested in economic or educational status, has been a frequent topic of study in this field, and a significant amount of research finds support for the link between inequality and criminal behavior (Blau & Blau 1982; Platt 2002). Although the critical perspective encompasses a wide range of topical areas, it may be divided into two conceptual categories depending on whether the focus of the specific approach is on the etiology of conflict or on the application of power to differentially stratified groups. This second approach is commonly referred to as the status characteristics or radical labeling approach. The next section briefly describes the etiological perspectives and the status characteristics or radical labeling perspective. The etiological focus of critical theory is based on the idea that differentially stratified groups are at odds with each other within society. Crime, violence and social unrest find their origin in this conflict, whatever the differences in stratification may be. What is often seen as random crime or meaningless violence to an outsider is actually a form of rebellion against a threatening opponent or regime. Vold wrote of crime: ?On the surface these offenses seem to be ordinary crimes, but on closer examination they are revealed as the acts of good soldiers fighting for a cause against the threat of enemy encroachment? (Vold & Bernard 1986, pg 275). Vold wisely noted that not all crimes fit this model, as some crimes are irrational and do not reflect a class or group power 19 struggle 6 . The etiological approach is the main focus of this research and will be discussed in greater detail in the following section. The status characteristics perspective states that based on their personal or group characteristics some people will be more likely to be defined as criminal by those with more power (Erikson 1966; Triplett 1993; Turk 1975; Turk 1980; Turk 1982; Vold 1958). Those personal characteristics can include race, class, gender or other socially relevant variables. This perspective sees crime generally as a result of the symbolic interaction between an individual, a potentially criminal label and a biased criminal justice system under the influence of a controlling group. The radical labeling approach explains differences in crime rates among social groups by suggesting that while underprivileged groups generally do not commit more crime than more influential groups, they are disproportionately targeted by the criminal justice system which makes their criminality seem greater (Krisberg 1975; or see Hindelang 1978 for evidence to the contrary). Compared with the etiological approach, the status characteristic approach has received the bulk of the empirical evaluation efforts and remains the most? popular and widely discussed aspect of the critical perspective. Critical theory, particularly its etiological predictions, was a main focus within criminology during the 1960?s and 70?s, but in the last two decades has not been applied with nearly as much vigor. This was clearly illustrated in a review of the most frequently cited criminologists from 1986 to 1990 by Cohn & Farrington (1994). Their review found that of the forty seven most cited criminologists, only two were considered critical criminologists. As a result of this relatively low popularity, much of mainstream 6 Quinney (1970) believed that this model could be extended to all crimes, as irrational people and the mentally ill were in fact a recognized segment of society, if not a formal group. 20 criminology is uninformed of the contributions of the critical theorists and know the field only through the work of a few contributors, especially Richard Quinney (Wright & Friedrichs 1998). This research seeks to address this situation by showing that an etiological argument, based in Weberian conflict theory 7 , may provide an explanation for the trends in violence in contemporary Iraq. The next section explores the two main etiological thinkers, Karl Marx and Max Weber, within the conflict tradition, their contributions and specifically the ways in which Weber?s ideas support the hypotheses developed in this research. The Etiology of Conflict: The Marxist and Weberian Perspectives Two of the most important thinkers within the critical perspective are Karl Marx and Max Weber, both of whom founded their own unique branches of the critical school 8 . The approach that Marx founded is aptly referred to as Marxist, and more recently neo- Marxist criminology, and the approach founded by Weber is referred to Weberian conflict theory. This research draws exclusively from the Weberian school, but this review includes an analysis of Marxist thought in order to clarify the distinction between the two as well as draw useful conclusions from that body of literature. Although its intellectual roots can be traced further back in time, most theorists credit Karl Marx (1848) as being the father of critical theory. Marx was one of the first theorists to address the importance of inter-group conflict as a driver of history. 7 Weberian conflict theory refers to the specific conflict theory outlined by Max Weber, as opposed to other theories that utilize similar conflict theory principles. 8 There is a third major philosophical tradition within critical theory, that of Georg Simmel (1908). Simmel is most closely associated with the idea of symbolic interaction, an idea that grew out of his early work. Simmel is largely excluded from most conventional accounts of critical theories, despite the fact that many famous criminologists, including Sellin, Vold, and Quinney can be seen as falling within his philosophical tradition (Wolff 1950). 21 Problems in society, according to Marx, are a function of the unequal relationship between the owners of production and the workers. Marx wrote very little specifically on crime, focusing on social problems related exclusively to the economic sector such as unemployment. Even so, Marx?s ideas are important as they had an incredible impact upon the history of the twentieth century, sparking new theoretical development as well as altering the governments of many nations and leading to a bitter and continuing struggle between capitalist and Marxist groups (Ruggiero 1995). Bonger (1916) was the first theorist to explicitly outline a Marxist criminology. Bonger believed that individuals possessed two moral dimensions, altruism and egoism. The capitalist mindset, based in egoism, encouraged competition, the acquisition of wealth and the virtue of greed and therefore reduced altruism and increased egoism across society. Crime and a variety of other social ills were a result of this increased egoism, as capitalism created unnatural needs that could only be fulfilled among the lower classes by crime. Marxist criminology grew following Bonger?s contributions and eventually was restructured into a neo-Marxist criminology that focused on a wider range of topics than were being included in the traditional Marxist analysis of social problems. Testing and expanding upon Marx?s theory by looking at governmental reactions to crime, Rusche and Kirchheimer (1939) examined the connection between the growth of capitalism and rates of incarceration. The Marxist perspective predicts that as capitalism increases, so should imprisonment as a way of controlling potentially dangerous surplus population. Rusche and Kirchheimer found statistical support for this hypothesis, as rates of incarceration closely correlate with the economic fluctuations of 22 capitalist societies. Following this early finding a school of Marxist penology developed and much of the research on Marxism today falls within this realm (Adamson 1984). Neo-Marxist criminology flourished in the United States during the 1960?s and 70?s, taking Marxist themes of economic stratification and applying them in unconventional ways (Greenberg 1981; Platt 1975). Neo-Marxist criminologists continued Marx?s theme of challenging the status quo and often accused mainstream criminology of supporting an oppressive and unjust criminal justice system. Many critical criminologists believe that mainstream criminology is hopelessly compromised by its relation to the State (MacLean 1997). This abrasiveness led in part to the rejection of neo-Marxist ideas and leanings by much of mainstream criminology (Wright & Friedrichs 1998), as criminological research depends in part upon the funding and access that can only come from a harmonious relationship with government and the criminal justice system. Criticizing the criminal justice system, and by extension the government as tacitly maintaining the capitalist power structure has been a common theme for critical theory. Chambliss (Chambliss 1975; Chambliss & Seidman 1985) further developed this line of reasoning with an analysis of the relationship between law, crime and economic problems. To Chambliss, ruling powers use law and law enforcement to draw attention away from more serious economic problems. The use of the criminal justice system to control the surplus population, the unemployed and the unemployable, has been an idea frequently studied by Marxist criminologists (Spitzer 1975). Finally, Quinney?s (1974) early work focused on the use not of the law, but of other lower class groups as a means of controlling the surplus population. By pitting the lower class against itself, the ruling 23 class could avoid the social unrest or rebellion that Marxism predicted would eliminate the distinction between classes. Marx's ideas were further developed, if under a new guise, by Taylor, Walton & Young (1973) with their creation of a ?new criminology?. Their work was based around the belief that the structure of society, including differences in the means of production, determined how individuals acted within the various facets of their lives. This idea was developed into a range of theories, including links to the economy (Michalowski 1985), social bonding (Colvin & Pauly 1983) and gender differences (Messerschmidt 1986). Hagan (1989) more recently updated neo-Marxist thought with his findings that class, as a function of economic power and authority, was related to crime. All told, Marx?s ideas have had, and continue to have, a great deal of influence on the focus of critical theory and offer a unique angle through which to study the mechanics of a society. Unlike Marx, Max Weber (1919) believed that social class was not determined exclusively by control over the economic sector. Weber held that class was a function of three elements: wealth, status and power. As the founder of conflict theory much of his work, although in the same vein as Marxist thought, contains a uniquely different approach to understanding social interaction. While Marx believed that society could evolve into a relatively disorder-free utopia with the acceptance of socialism, Weber believed that conflict was part of human nature and therefore inescapable (Aron 1970). Conflict, in Weber?s eyes, was socially useful as it helped to define norms, draw boundaries and clarify the social hierarchy. Sellin (1938) was the first criminologist to explicitly take up Weber?s theory. Sellin argued that different cultural groups have different conduct norms and when these 24 cultural groups came into contact, either through immigration or neighborhood mixing, disorder was likely to result. Vold?s (1958) work followed that of Sellin?s but put less emphasis on cultural groups and more on groups that sprung up to maintain specific interests or needs. Although these groups were thought to compete on a number of levels, this interaction was most readily seen in the realm of politics as each group tried to move the legal structure closer to supporting its own group?s interests. Turk (1969) helped to explicate this model further with his ?theory of criminalization? in which more powerful groups sought to criminalize the actions of their competitors. A review of the use of the legal system to control weaker groups is presented later in this section with a discussion of Donald Black's work. Particularly relevant to this research, Weber?s introduction of the concept of power helps to explain the ability of one group to impose its? will over another. Weber defined power as the ability to secure compliance against someone?s will to do otherwise (Weber 1968, pg 53). Groups actively seek power, both for its own sake and for the ability to control other groups that power confers. Building upon Weber?s theme of power, Dahrendorf (1959) argued that each sector within society was divided into imperatively coordinated associations. Each association was broken into groups, those with and those without power. As the goals of the groups diverged, Dahrendorf believed that social order within society was only possible because some groups were able to constrain others, effectively coercing them into following their orders. When opportunities presented themselves there would be direct attempts to alter the power balance by the groups involved, and as a result different groups would gain power at different times. Competition and coercion during these time periods would not only 25 serve to propel individual groups forward, but also to restrict the activities of potential competitors (Turk 1966). The study of the use of power is a common theme within conflict theory. Groups that gain control or influence over the government have the ability to use that power to benefit themselves at the expense of the others, either economically, socially or even by using the government to actively repress competitors (Chambliss & Seidman 1971; Dahrendorf 1959; Vold 1958). Power can take a number of forms, and one of the most often analyzed forms of the usage of power is control over the law. The Weberian tradition included a heavy focus on the behavior of law, the most important theorist from this tradition being Donald Black. Quinney (1970) and Black (1976) both argue that groups in power use the law to exert social control over other groups and to support their self-interest. The more organized a group, the better they are able to gain control of the law and use it to their advantage (see Jamieson & Blowers 1993, for a test of this hypothesis). In support of Black?s ideas, Chambliss & Seidman (1982) tested the idea that the reality of the practice of law is a function of the perspective of those in power. They found that judges often relied on personal values as much as on precedent in determining guilt in legal cases (see Gottfredson & Hindelang 1979, for a counter- argument about the use of law). As the social gap widens between those with and without power, law becomes increasingly important in regulating the actions of the powerless (Chambliss 1975). Research on the use of the law helped connect the etiological arguments of Weber with the status characteristics focus of much of the later work done within critical criminology. 26 Law is only one form of social control, and when enforcement of the law is weaker, other forms of social control, including violence, may be used in its place (Black 1989). The active and violent repression of minority groups by governments has been a source of civil unrest for centuries. If resources for defiance are available however, the coerced group is likely to fight back, contesting the legitimacy of the government with violent action (Etzioni 1961). In line with my earlier discussion of government overreaction, research on crime (Bailey 1998; Sherman 1993) and on terrorist violence supports the position that repression can increase violence as a form of defiance (Bohara, Mitchell & Nepal 2006). Although this body of literature often does not explicitly mention conflict theory, it is in line with its basic ideas and predictions about power, legitimacy and group interaction. Legitimacy and the Weberian Tradition Weber (1919, 1964) was one of the first theorists to explicitly tie legitimacy to group interaction. Legitimacy generally means having followed socially accepted rules or procedures in order to gain power, and once in power utilizing resources in a fashion in accordance with fairness and law (LaFree 1998; Weber 1964). Legitimacy is a function of perspective, and a group may be seen as legitimate in the eyes of the majority while being seen as illegitimate in the eyes of a minority group. Whether a group or a government is seen as legitimate is an important variable in determining how other groups, particularly those with competing interests, will interact with it (Blau 1964). A group?s ability to dominate another group successfully is predicated in part upon the 27 controlling group?s legitimacy (Bendix 1960), as an illegitimate group will lack public support, encouraging defiance against it. Aside from the importance of popular support, legitimacy helps to bind groups together, encouraging the establishment of and adherence to norms and bonds within the groups. Weber?s theory, as described by Collins (1975), links norms and group bonds with group competition: ?The creation of emotional solidarity does not supplant conflict, but is one of the main weapons used in conflict? (Collins 1975, pg 59). The more closely knit a group, the better it is able to define itself and likewise define other groups as being different and perhaps a threat to the original group?s interests. Groups require legitimacy in order to support their own internal solidarity and prevent the development of internal fissures. As long as groups can maintain their legitimacy they can maintain group solidarity and use it to extend their power. While non-violent and norm-based coercion is a useful tool for groups to maintain power, the use of violent coercion is always a potential resource, although one that can threaten a group?s own legitimacy if used incorrectly (Collins 1975). When legitimacy falls, confidence drops and internal and external fractures appear within and between groups, often resulting in an increase in violence (Lipset 1960, Marighela 1970; Parker 2007). Collins (1975) built on Weber?s earlier work to further explore the conflict between different groups for legitimate political power. Collins argued that groups exist within a zero-sum political environment where the loss of legitimacy of an authority is a potential gain in legitimacy and power for a competing group. In this research I assume that events which reduce a group?s legitimacy present an opportunity for rival groups to contest that weakened legitimacy through the use of violence. As a result, I expect that 28 legitimacy-changing events will result in changes in the frequency of political violence. Weber's original argument supports this view as he argued that for groups in conflict with an illegitimate foe, any means of rebellion used would be justifiable (Wilkinson 2000). As an extension of Weberian conflict theory, modeling political violence as a function of perceived legitimacy of the nation-state has been a fruitful theoretical path (Callaway & Harrelson-Stephens 2006). Violent groups threaten a government?s monopoly over the use of force within its own territory and thereby weaken its legitimacy, inviting further attacks or dissension (Weber 1919). In nations such as Afghanistan or Colombia where the central government has lost control over large regions of the country, groups willing to employ violence are major holders of power. States in which the government is too weak to control violence within its territory are known as failed states (Crisis States Research Centre 2007). Iraq currently ranks second on the Crisis States Research Centre?s list of failed states (Fund for Peace 2007); a testament to the weakened central government and high rates of violence the country is experiencing. The loss of legitimacy is something that many nations have suffered through during periods of growth or change. During the 1960?s and 70?s the United States government?s perceived legitimacy fell as a result of the Watergate scandal, the Vietnam war, civil rights protests and economic crises. This declining legitimacy and the public?s failing trust in social institutions was linked to the increase in crime during the later half of the 20 th century (LaFree 1998). As individuals reduce their involvement with or lose faith in the ability of their social system to support them and provide for their needs, they begin to search for other ways, often criminal, to fill the vacuum. When governments or 29 institutions lose legitimacy, citizens and groups are less likely to obey or support those institutions. This research supports my model that the legitimacy of the Iraqi government and the other institutions involved will mediate their popular support and ultimately their being selected as a target. However, governments are not the only group for whom legitimacy is important (McCauley 2006). Groups that use political violence as a tool depend heavily on their perceived legitimacy, as they cannot hope to gain political influence without some form of public support (Cronin 2006). Popular support for a violent group can dissipate for a number of reasons, but if a group acts in an overly violent, callous or unacceptable fashion it will weaken its public support. Legitimacy and public support are so important that terrorist groups often propagandize heavily in order to sway public opinion in their favor (Ashraf & Brigden 2006). Winning the support of the local population is central to having a legitimate claim to power and both violent groups and governments seek to paint their own actions as just and those of their opponents as unreasonable. In support of the conflict perspective, Biddle (2006, pg 5) puts the struggle between insurgents and the coalition-backed government for legitimacy and support in the eyes of the Iraqi public as: ?Using a mix of coercion and inducements, the insurgents and the regime compete for the allegiance of a common pool of citizens, who could, in principle, take either side?In such situations, even the government is typically an instrument of one communal group, and its opponents champion the rights of their subgroup over those of others.? 30 Unlike Marxist theory and its focus on ideology, Biddle sees the conflict in Iraq as being about group survival at its core, as control of the government and the power it bestows is the only reasonably sure way to ensure a group?s continued existence. In sum, legitimacy plays an important role in mediating political violence. While being illegitimate may invite attack, having strong legitimacy ought to help prevent a group from being targeted. As all groups seek to bolster their own legitimacy, targeting a legitimate or popular group in a violent attack will weaken the attacker?s popular support, and as a result I believe that a target sector that enjoys increased legitimacy will be targeted less frequently. This research will put this idea to the test, determining whether legitimacy-weakening events encourage violence against specific target sectors, and legitimacy-strengthening events reduce violence. The evaluation and decline of the critical perspective While Marx, Weber and subsequent critical thinkers developed and refined many unique theoretical ideas, the empirical evaluation of their theories was less rigorous. Most of the evaluation has centered on later theorists, as the earlier etiological work did not lend itself well to empirical evaluation. Evaluation of the critical perspective, and particularly of the etiological arguments can be complex as early theorists were highly critical of positivistic methods of research as a means to support their relatively unconventional ideas (Mann & Dashiell 1975; Marx 1848). Both Marx and Weber sought to understand the present by dissecting the past, with no thought of testing the connections that they might uncover. Marx and Weber preferred to rely on qualitative data, historical events and secondary sources to support their ideas (Lynch 1987) rather 31 than to test them empirically. As a result, many of the subsequent critical theorists who followed in their footsteps preferred case studies to empirical evaluations (Sparks 1980). Critical theories, even those with relatively straightforward ideas, are often incredibly complex as they grew out of a rich sociological and philosophical tradition. For example, Black?s (1976) simple assertion that the law will be used disproportionately against minorities has been subject to a heated debate about the empirical support for and best way to test this proposition (Bernard 2002). Research on this proposition finds (Gottfredson & Hindelang 1979), contrary to the conflict theory proposition that the legal system will be used against weaker or smaller groups, that minorities are the greatest consumers of the law and not primarily unfairly victimized by it. Black has challenged this finding as being a faulty conclusion drawn from a mistaken conceptualization of his ideas, a response that is frequently encountered within critical theories owing to the complexity of many of their core ideas. Critical theories and their notions about power and group conflict have been tested by comparing statistics about the distribution of wealth and membership on corporate boards and ruling bodies (Vold & Bernard 1986). This approach generally finds a relationship between levels of power or social class, crime and disorder. What this research is generally unable to provide proof of is the intent of the stronger groups to dominate the others. There is little research available that shows that stronger groups purposefully use their power exploitatively in order to control or disrupt weaker or less well-organized groups. Further adding to the difficulty of properly assessing the empirical support for critical theory, there is confusion among critical theorists about the proper focus of the 32 field itself, as each sub-theory focuses on a unique manifestation of the conflict believed to be inherent in society (Taylor, Walton & Young 1975). Critical criminology was seen as contradictory in places, given the wide range of ideas that fell under its umbrella (Berry 1997). In addition, critical theorists tend to be critical of other critical approaches, and one of the weaknesses of the perspective is the lack of theoretical consensus about the proper focus and level of analysis of the field (Fine, Kinsey, Lea, Picciotto & Young 1979). Aside from definitional infighting, many critical ideas are rejected out-of-hand by those who hold to a consensus view of society (Hagan 1977; Sutherland & Cressey 1960). One criticism of the conflict perspective is the apparent widespread value consensus within society that stands as proof of the weakness of these conflict model assumptions. Critical theorists respond by stating that what appears to be societal consensus is in fact a result of overt coercion from the criminal justice system and covert coercion in the form of ideological propaganda and persuasion, as value conflict still exists beneath the fa?ade of societal consensus (Walsh & Ellis 2007). Both sides have ample evidence to support their perspectives and the argument about the actual level of conflict versus consensus within society continues. As a macro-level theory, measuring many of the variables described by conflict theory is difficult, and proxies are frequently used (Hagan 1989). Research on critical theories generally focuses on comparing social inequality, either economic, class, race or gender-based and looking at differential rates of criminality or criminal justice processing, ultimately mixing the status characteristics and etiological arguments (Chambliss 1973; Hindelang 1978; Lynch & Groves 1986; Reiman 1995; Thornberry 33 1973). There have been a wealth of cases along these lines, testing for bias within the criminal justice system as a means of testing critical theory (Turk 1982), however these tests are not directly relevant to this research as they focus primarily on the status characteristics perspective. This research poses a different type of test for conflict theory, looking directly at group conflict and political change as a cause of violence, and not at institutionalized bias or labeling effects thought to result from these conflicts. Critical criminology declined in popularity in the United States over the last three decades for a number of reasons. As the intricacy of quantitative methods increased, theories that were more difficult to quantitatively test or that opposed the usage of quantitative measures altogether became less popular. The quantitative testing of theories with large datasets has become standard practice in criminology and theories that purport to be untestable, or which aren?t supported empirically, tended to decline in popularity. Critical criminologists respond to this accusation by pointing out that official statistics, so often used in empirical tests of mainstream theories do not accurately represent the actual crime rate or the characteristics of offenders (Berry 1997). Furthermore, many of the conditions that encouraged the growth and development of critical criminology such as societal change and civil unrest became less frequent in the United States. Critical criminology has also moved its centers of intellectual activity to England from the United States, reducing its visibility within the United States (Berry 1997). Aside from falling from fashion, critical theory also morphed into other theories, including left realism, peacemaking criminology (Pepinksi 1995; Quinney 1991) and deconstructionist theories that looked at the bias inherent in the use of language (Arrigo 1995) as a source of control and repression itself. As a result of these previously 34 mentioned concerns and criticisms, conflict theory as well as the broader range of critical theories, have largely fallen by the wayside, especially in the United States. This research seeks to bring renewed attention to this perspective through a modern-day analysis of political violence and group conflict within Iraq. Criminological research on target selection Before considering target selection from a Weberian conflict perspective, it is important to note that mainstream criminology has not been silent concerning how criminals select their targets. Traditional theories such as routine activities (Clarke & Newman 2006; Cohen & Felson 1979), rational choice (Cornish & Clarke 1985) or a focus on geographic areas with a high frequency of repeat offenses (Weisburd 2005), examine the rational process involved in deciding to commit a crime and selecting a target. For a crime to occur, a motivated offender must meet a suitable target in the absence of a capable guardian, and have the potential benefits of the crime outweigh the potential costs (Cohen & Felson 1979; Cornish & Clarke 1985). This line of reasoning has been very fruitful in terms of crime prevention, but offers little to this research as a result of the limited information available about the motivations of the groups, the suitability of the targets, the capable guardians that might be present, and the ultimate costs and benefits of violent acts. This is not to say that the contributions of this approach should be ignored, for example, modeling guardianship as being a function of the number of coalition or Iraqi security forces in an area would be an interesting study. However, the level of detail available within my dataset on these variables is too poor to test any but the crudest of macro-level rational choice or routine activities models. 35 Conflict theory, legitimacy and violence in Diyala As mentioned in the introduction, this research consists of an initial empirical evaluation of conflict theory?s ability to predict trends in violence directed against all province-wide targets in Diyala. After this initial test, a more detailed target sector- specific evaluation follows that uses conflict theory to more precisely determine the specific expected effects on each target sector of the events selected in this research. Before discussing the specifics of the overlap between conflict theory and sector-specific target selection it is necessary to directly explicate the hypothesis relating non-target- specific violence in Diyala to conflict theory. Weberian conflict theory states that societal subgroups are in competition with each other for political and social power. The perceived legitimacy of the group in power should mediate the opposing group?s choice to engage in political violence to contest that power. To this end, events in Iraq that alter the legitimacy of the more powerful groups, be they the government, the coalition or major religious/ethnic groups, should directly correlate with changes in the frequency of political violence in Diyala. When the more powerful or better organized groups are seen as legitimate, they are less likely to be violently opposed by other groups, in part because the Iraqi people will be less likely to tolerate violence against a group with higher perceived legitimacy. When a strong group holds legitimate power, it reduces inter-group tension as there is little justification to initiate overt or covert conflict. When a group loses legitimacy, tensions increase as the justification for challenging that group increases. From this model the following hypotheses were derived that relate group conflict and legitimacy to violence in Diyala. 36 HYPOTHESIS 1A: Events that increase the legitimacy of controlling groups and reduce inter-group tension will decrease political violence in Diyala. HYPOTHESIS 1B: Events that decrease the legitimacy of controlling groups and increase inter-group tension will increase political violence in Diyala. In this case the language ?controlling groups? refers to the coalition and the Iraqi government, although the influence of each has waxed and waned over the time period in question. The coalition was the key controlling group during the first half of the time period examined, and once control of the country was handed to the elected Iraqi government officials in May of 2006, they became the primary controlling group in Iraq. This changeover and the proper way to interpret the findings connected with the test of this hypothesis will be explained during the discussion section of this research. I admit that this all-targets assessment for Diyala is relatively unsophisticated, as conflict theory provides excellent target-specific hypotheses. I include this initial analysis in order to get a better sense of the baseline effects of the intervention events selected in this research, as well as to provide a comparison group for the later target- specific analyses. For clarity, I refer to the all-targets analysis as province-wide violence, in order to avoid confusing it with the target-specific analyses. Target-specific conflict theory research and hypotheses Conflict theory, ultimately a macro-level theory meant to explain group interaction, helps to shed some light on how specific targeting decisions are made aside from the research already cited. The following section explores the conflict theory research that is specifically focused around the following target sectors: law enforcement, 37 military, government and civilian, and then outlines the hypotheses linking these sectors to political violence in Diyala. As stated, this research posits that groups are in conflict with one another and will therefore select as a target the groups with which they are aggrieved, provided the popular support for that target, as manifested by its legitimacy, is low. My intent was to select target sectors that were mutually exclusive and theoretically distinct, which would have made the theoretical and empirical analysis much simpler. In most nations, the police are closely intertwined with the government, and the government is often closely intertwined with the presence of the military. In Iraq, the coalition military is independent from the government, although the Iraqi government was beholden to the coalition military during the rule of the Coalition Provisional Authority from April, 2003 to June of 2004. Iraq is currently developing its own military for external defense, although the attacks against military targets in this research refer to those targeted against the coalition or allied armed forces. Given the close relationship between the Iraqi government and its police force, an argument can be made that these target sectors should be combined into one. I made the decision not to combine the two sectors based on the different trends revealed in the timelines of the attacks against each sector. While the two sectors map very closely in the beginning of the dataset, in July of 2005 the government sector experiences a large increase in attacks that the law enforcement sector does not. Aggregating these two sectors would mask this difference, and therefore they were left separate. In sum, conclusions about target-specific violence will be interpreted with caution, as factors that affect the selection of one target sector will likely have an effect 38 on targets in another sector. The following section outlines some of this crossover as well as the research supporting the conflict theory propositions believed to be behind the selection of the targets of political violence. Tied into this review is an analysis of the research from outside the field of conflict theory that offers historical examples of changing trends in violence directed against specific target sectors. I also list the target- specific hypotheses for each target sector that will be evaluated in this research. Law enforcement targets Disentangling the selection of law enforcement and government targets can be difficult, as police are a visible and accessible proxy for the government in many cases. Just as groups compete for control of the government, Vold (1958) believes that groups struggle for control over the justice system in order to use that system to curb the interests of competing groups. Vold argues that more powerful groups gain control of the criminal justice system and use the police to coerce their opponents (Chamlin & Cochran 2000; Liska 1992). Control over the government and its police forces can be used to perpetuate overt or covert bias against weaker groups (Krisberg 1975) and as a result of this coercion, minority groups may choose to target the police as the visible and most accessible representation of that government. As mentioned earlier, this form of violent reaction is similar to Sherman?s definition of defiance (Sherman 1990). Weber (1919) was one of the first theorists to link the actions of the State with the illegitimate use of force through the use of the police. When police are seen as taking orders from illegitimate masters, or of being illegitimate themselves, they encourage the defiance discussed by Sherman (Sherman 1993; Turk 1982). If the government is not 39 seen as legitimate due to its exploitation or support of the exploitation of another group in society, then neither are its agents of control: the police and commonly the military. As support for defying a government increases, it is reasonable to expect that attacks against police increases in a similar fashion. If this connection holds true, then I expect to observe an increase in attacks against law enforcement targets following a loss of legitimacy by the Iraqi government, and a decrease in attacks when the legitimacy of the government is increased. Linked to their perceived legitimacy, the efficacy of the Iraqi police force has fluctuated since the beginning of the war (Deflem & Sutphin 2006). The Iraqi police force was created amidst serious internal reorganization, as many former Iraqi police were Ba?athists and the Coalition leadership decided that it was necessary to purge the loyalists from the old regime before the reconstruction of the police forces could commence (Deflem & Sutphin 2006; Hafez 2006). Further adding to the difficulties, the daily responsibilities of the police are made nearly impossible by the high levels of continuing violence in Iraq (Baker & Hamilton 2006). As an example of the difficulty of reconstructing a police force within a turbulent nation, a case study on the reconstruction of the government in Kosovo by Wilson (2006) found that it took a substantial amount of time for a reforming nation to reconstitute an effective police force and legal system. Until that reconstitution is complete and the new government is firmly established, violence against civilians and the police is likely to continue. The legitimacy of the Iraqi police has been called into question as a result of incidents where men dressed in police uniforms have taken part in sectarian killings. For 40 example, in 2007 men dressed in police uniforms rounded up and killed 15 Kurds in a Kurdish village (Komarow 2007). As a result, given that many Kurds are Sunni, many Sunni increasingly see the security forces as a tool of the majority Shia, a situation that is likely to lower the legitimacy and increase violence against the Iraqi police (Flashpoint 2006). Deflem and Sutphin (2006) analyzed attacks against the Iraqi police forces (see page 275 of their article for a chart of attacks against Iraqi police forces over time). They discuss the role of the Iraqi police and the attacks against them, but do not offer an explanation for changes in the violence directed against the Iraqi police. This research will be the first attempt of which I am aware at modeling the variables that influence violent attacks against the Iraqi police. Based on the literature and the conflict model previously described, this research tests the following hypotheses related to the selection of law enforcement targets in Diyala: HYPOTHESIS 2A: Events that increase the legitimacy of the Iraqi government or police will decrease attacks against law enforcement targets. HYPOTHESIS 2B: Events that decrease the legitimacy of the Iraqi government or police will increase attacks against law enforcement targets. Military targets Conflict theory is a particularly useful approach for understanding attacks against the coalition forces in Iraq. The coalition forces depend largely on their perceived legitimacy to gain the support and trust of the Iraqi public which is essential to 41 accomplishing their mission (Baker & Hamilton, 2006). As research previously discussed on the link between aggressive government actions and violence has shown, the higher the perceived legitimacy of the coalition?s presence in Iraq, the less they ought to come under attack. For the coalition forces, legitimacy in Iraq oscillates as events cast their presence as either a liberating or occupying army (Laird 2005; PIPA Jan 31 st 2006; PIPA Feb 28 th 2006). Although many Iraqi political and religious leaders were at first skeptical about the goals of the coalition and the viability of the new government, some have embraced the coalition-backed democratic process as a means to secure power and representation for their people (Finn & Zenilman 2003). Despite the civil infrastructure that the coalition has helped to construct, their presence is still resented by many Iraqis. The following quote from an Iraqi citizen captures this sentiment nicely: ?The Americans, yes, they do good things, but only to enhance their reputation. They are occupiers. We want them to leave? (Filkins 2003). This perception is based in part on the fear in the region that the presence of international military forces, in particular the United States, is meant to secure the Western world?s access to petroleum resources (Clarke & Newman 2006; Hoffman 1998; Korany 2005). Many nations have good reason to fear the presence of the industrialized world in their affairs, as the history of the last three hundred years is littered with exploitive colonial relationships carried out by the militarily strong Western nations at the expense of the developing world (Aran 2003; Wallerstein 1979). There are many cases where nations have sought to gain independence from colonial powers through the systematic use of violence to reduce the political will of the mother country. Israel, Northern Ireland, 42 Algeria, and even the birth of the United States required the use of violent tactics to win independence from a militarily and economically stronger foe. The continued presence of de facto colonial relationships between the first and third world is subject to much debate, and it is important to note that nations do not like to abandon influence over former colonies that have become independent. In the words of the British statesman Benjamin Disraeli: ?Colonies do not cease to be colonies because they are independent? (Disraeli 1863). While Iraq is somewhat different than these colonial examples, the forces that set violence in motion for those nations, from a conflict perspective, are the same as those that are hypothesized to be at work in Iraq. The legitimacy of the coalition presence in the Middle East is dependent upon whether they are viewed as being there for security and humanitarian reasons, or for imperialist or colonial reasons to take control of the region?s resources (Cronin 2002). Events that reduce the coalition?s legitimacy by portraying them as being the tool of an imperialist force, indiscriminate in the targeting of civilians or overly brutal in their tactics ought to reduce their legitimacy and increase the violence against them. Based on the previously described literature, this research tests the following hypotheses related to attacks on military targets in Diyala: HYPOTHESIS 3A: Events that increase the legitimacy of the coalition forces in Iraq will decrease attacks against military targets. HYPOTHESIS 3B: Events that decrease the legitimacy of the coalition forces in Iraq will increase attacks against military targets. 43 Government targets While many critical theories outline the struggle between individual groups and the government, it is important to note that from the perspective of conflict theory, as outlined in this review, the government is merely a resource controlled by a group. When groups rebel against the actions of the state they are rebelling against the actions taken by the group that is using the instrument of state power coercively. This perspective allows for the adaptation of conventional conflict theories to support the model used in this research. Events that reveal the government to be a tool or puppet of one group in the exploitation of other groups will reduce the government?s legitimacy and increase attacks against it. According to Turk (2004, pg 274), political violence against the government is a normal ?response to feelings of indignation and frustration developed in a repressive political environment? and therefore as these feelings oscillate as a result of the changing perceived legitimacy of the government, the violence directed against the government should oscillate as well. One source of frustration or resentment towards the government comes from charges of corruption. In support of this theoretical model, one case study of government corruption in Kosovo found that corruption dramatically reduced the legitimacy of the government, and was believed to have increased the violence against it (Wilson 2006). Although corruption in Iraq doesn?t play a direct role in this research, these findings are in line with the conflict theory model. Other research suggests that States that are undergoing periods of governmental transition experience lower legitimacy and higher levels of conflict and violence (Feierabend, Feierabend & Nexvold 1969), further supporting this model. 44 Recent research on public attitudes in Iraq completed by the Program on International Policy Attitudes (PIPA Jan 31 st 2006) highlights the importance of the political legitimacy of the new government to the Iraqi people. A poll conducted in Iraq in January of 2006 found that while two thirds of the Iraqi people believed that the December 2005 elections were fair and representative, 94% of Sunni?s polled believed that the elections were not fair, and 92% said the new government was not legitimate. While this doesn?t directly address the importance of legitimacy, it does illustrate the divide between the main ethnic groups pertaining to the elections. Weber noted (Bendix 1960) that the electoral process was particularly subject to abuse by powerful groups, and that elections were a particularly sensitive activity, leading me to believe that periods of elections in Iraq will be focal points for violence. The evolving Iraqi government is basically sectarian (Baker & Hamilton, 2006) and the main leaders often act to benefit their sects more than fulfill their position?s duties. The Iraqi government has gone through periods of not producing adequate levels of safety or living conditions for many Iraqis and runs the risk of losing its legitimacy during these downturns as a result (Baker & Hamilton 2006). This variability supports the core conflict ideas cited in this research in that some groups see the government to be legitimate, while others believe that its legitimacy has fallen substantially since it was created. Based on this literature, this research tests the following hypotheses relating to the targeting of government targets in Diyala. HYPOTHESIS 4A: Events that increase the legitimacy of the Iraqi government will decrease attacks against government targets. 45 HYPOTHESIS 4B: Events that decrease the legitimacy of the Iraqi government will increase attacks against government targets. Civilian Targets The fourth category of targets that this research investigates is civilian targets. Attacks against civilian targets include any attacks against a non-governmental, law enforcement or military target. These attacks are often directed against markets, public gatherings or religious sites and processions. As religion is one of the main differentiating factors in Iraq, it is no wonder that many of the attacks against civilians are directed specifically at religious groups. Conflict theory offers a number of angles from which to examine the targeting of civilians. Unless specifically directed at the other three target sectors, the conflict perspective can be directly applied to the targeting of civilians without need for additional adaptation or explanation. In short, as most societal groups are civilian groups and are at odds with other civilian groups, it is to be expected that civilians will be a frequent and direct target of political violence. This can be seen throughout history as civilians have born most of the violence of civil conflicts (Wilkinson, 2000). Furthermore, terrorist attacks generally kill civilians more than non-civilians, although as previously discussed, some definitions define terrorism only as violence targeted against civilians. McGarrell and Castellano (1991) explored the role of status differentiation within the context of Weber?s theory. They posit that people are vastly different from one another in a range of variables including class, race, ethnicity and inequality. Societies 46 with stable or lower levels of differentiation are more stable but as differentiation increases, crime and intergroup conflict should increase. As Iraq is currently undergoing a period of transition where the conventional power balance between the main groups is shifting, this differentiation ought to be changing, supporting our hypotheses that political changes will be linked to changes in violence. Conflict theory has been relatively silent on the actions of religious groups when they act independent of political groups or movements. In Iraq, religion and politics are often interchangeable, as religion is one of the main dividing lines between ethnic groups. As the tensions between the ethnic groups in Iraq increases, civilians are likely to experience an increased number of attacks, as the political and symbolic nature of terrorist attacks often makes use of seemingly random violence against civilians to send political messages (Hoffman 1998). Events that increase tension between the groups will likely result in the specific targeting of religious sites. Attacks on religious sites may also trigger reprisal or retaliatory attacks by the targeted group, suggesting that one predictor of the targeting of religious sites would be previous attacks on other religious sites. Given the limitations of our data, I am not able to tell the ethnicity of the civilians targeted by political violence, and therefore am be unable to more precisely link the motivation to attack civilians to conflict theory directly. Based largely upon the conflict model previously developed, this research tests the following hypotheses relating to the targeting of religious and civilian targets in Diyala: HYPOTHESIS 5A: Events that increase tension between sectarian Groups in Iraq will increase attacks against civilian targets. 47 HYPOTHESIS 5B: Events that decrease tension between sectarian groupsin Iraq will decrease attacks against civilian targets. In summary, there has been a large amount of research on political violence, terrorism, conflict theory, legitimacy and the targeting of violence. Critical theory, particularly based on the conflict theory developed by Max Weber, offers an appropriate theoretical platform from which to study the targeting of violence in Iraq. Weber?s ideas about group interaction, power and the importance of legitimacy allow me to develop specific testable hypotheses that will help to determine how and why major political events affect the trends in the targeting of violence in Iraq. My hypotheses include an initial province-wide analysis of all violence in Diyala and then four subsequent target specific hypotheses that take advantage of the complexity of conflict theory and the detail available in my data. The following chapter describes the data and the operationalization of my variables and lays out the statistical approach that is best suited for testing these hypotheses. 48 Chapter 3: Data and Methods This chapter describes the independent and dependent variables used in this research. I begin with a description of the database of political violence in Diyala that was created for this study. Important choices about the selection of the level of analysis of the data and the selection of target sectors within the data are included herein. Additionally, this chapter includes descriptive statistics and visual timelines for the dependent variable that help to illustrate how the data relate to the hypotheses being tested in terms of the timing and distribution of events. The independent variables are introduced in this chapter as well, and their selection and relationship to my hypotheses are explained in detail. Finally, the analytic procedure for the ARIMA process is explained. I made the decision to place less relevant descriptive statistics about the dataset in Appendix Three. This appendix describes the different types of attacks, their lethality, their location and details about how many groups claimed responsibility for the attacks. These statistics do not directly relate to the hypotheses being tested, and are therefore excluded from the main body of this text in order to improve the flow and clarity of this section. While it is common to include information about the location, in this case Iraq, where the data were collected, the decision was made to place this information in Appendix One at the end of this report. I invite readers who are unfamiliar with Iraq to refer to this section. Placing this section at the end of this report also helps to improve the flow of the narrative, as the primary purpose of this research is to describe a conflict 49 theory test of political violence, and not to provide a historical or geographic summary of Diyala and Iraq. The Data This project has its roots in a database that contains a record of incidents of political violence in the Diyala province of Iraq from June 2003 to May 2006. This database was originally compiled by the Lawrence Livermore National Laboratory (LLNL) from open sources (O?Grady 2006) including newspapers, Internet news sites and public reports of political violence in the province. The inclusion criteria for violence in this database were very broad and included all events that were reported as political violence or terrorism in the sources listed. To ensure the comprehensiveness of the data, the dataset was updated and validated by students at the University of Maryland in the Spring and Summer of 2006. The students checked the incidents in the database against a database compiled by the National Memorial Institute for the Prevention of Terrorism (MIPT) known as the Terrorism Knowledge Base (TKB). Data from MIPT were used to supplement the original database where the LLNL database was missing information about incidents, although less than 10% of the cases in the database are exclusively from the MIPT data. The complete database contains a record of 735 events consisting of 676 violent attacks and 59 failed attacks in Diyala. Failed attacks were so named because they were attacks that were discovered or interrupted before they could be carried out, or that failed in their purpose, i.e. planted bombs not exploding. The 59 failed attacks will be included in this analysis as if they were successful attacks as they were meant to be successful. 50 The database developed for this research is the most complete and comprehensive open source database on violence in the Diyala province of which I am aware, and will be treated as if it contains the population of all attacks over the time period being studied. The database measured all violence in Diyala that was not explicitly defined as criminal violence or conventional warfare. Target Sectors I classified the violent attacks in Diyala into four mutually exclusive categories based on which sector the attacks were directed against: law enforcement (39.9%), military (10.4%), government (26.4%), and civilian targets (23.4%). A visual representation of this break-down is presented below in Figure One. Law enforcement targets included Iraqi security and police forces, police barracks, training centers, recruits and police patrols. Attacks against military targets included attacks on coalition military patrols, bases and soldiers. Government targets included government buildings, embassies, government officials, and polling places. Civilian targets included markets, schools, commercial establishments, infrastructure, public gatherings as well as mosques, temples, holy sites, religious processions and the assassination of religious leaders. Main database of political violence: 735 events Law Enforcement Targets: 292 events Government Targets: 194 events Military Targets: 77 events Civilian Targets: 172 events Figure One: Target-type databases drawn from the main database 51 Level of analysis When working with longitudinal data it is essential to carefully consider how the data will be aggregated, as different levels of aggregation can often mask important trends. For this research data were collected at the incident-level and were aggregated by week and by month. This section describes the choices behind aggregating the data in this fashion, as well as provides visual references through the use of timelines to explore the trends revealed by these different aggregations. While data collection stretched back from the beginning of the US led invasion in March, 2003, the first attack recorded in this database occurred on the 24 th of June, 2003. The database captures violence that occurred on 153 different weeks, although for 41 of the weeks there were no recorded attacks. When the data are aggregated by week, the first week with an attack is the week of the 18 th of June, 2003, and weeks are measured from Sunday to Saturday. The first month recorded was June, 2003, and there were 39 months recorded in the database. Only four months of the 39 did not register an attack. Table One provides descriptive statistics for the target sectors, as well as for the database as a whole, first for the data when aggregated by month and then by week. Table One: Descriptive statistics by week and month for each sector Aggregated by week (N=153) Target Sector Target Sector Target Sector Target Sector All Targets All Targets All Targets All Targets Government Government Government Government Law Enforcement Law Enforcement Law Enforcement Law Enforcement Military Military Military Military Civilian Civilian Civilian Civilian 52 Aggregated by month (N=39) Target Sector Target Sector Target Sector Target Sector All Targets All Targets All Targets All Targets Government Government Government Government Law Enforcement Law Enforcement Law Enforcement Law Enforcement Military Military Military Military Civilian Civilian Civilian Civilian Table Two illustrates an important problem within the dataset. The table shows that the frequencies of attacks are difficult to assess analytically as attacks are not evenly distributed across the time-period in question. There are many weeks without an attack and these attack-free time periods are highly concentrated in the first half of the dataset. The number of time-periods without attacks, broken down by target sector, is presented in Table Two. Table Two: Number of time periods with no attacks Series By Week (N= 153) By Month (N = 39) All Targets 41 4 Law Enforcement 58 8 Government 75 11 Military 105 17 Civilian 71 8 To visually illustrate the distribution of attacks against various target sectors and how the attacks are concentrated over time, the following ten figures show the frequency 53 of attacks aggregated by week and by month for the entire database as well as for each of the target sectors. Figure Two: Attacks against all targets (monthly) Figure Three: Attacks against all Targets (weekly) 0 10 20 30 3/03 3/04 5/0 5 5/06 0 20 40 60 80 03/03 1/04 11/04 9/05 6/06 Figure Four: Attacks against Government Targets (monthly) Figure Five: Attacks against Government Targets (Weekly) 0 5 10 15 20 3/03 3/04 5/05 5/06 0 5 10 15 20 03/03 1/04 11/04 9/05 6/06 56 Figure Six: Attacks against Law Enforcement Targets (monthly) Figure Seven: Attacks against Law Enforcement Targets (weekly) 0 5 10 15 3/03 3/04 5/05 5/06 0 10 20 30 03/03 1/04 11/04 9/05 6/06 57 Figure Eight: Attacks against Military Targets (monthly) Figure Nine: Attacks against Military Targets (weekly) 0 2 4 6 8 10 3/03 3/04 5/05 5/06 0 2 4 6 8 10 03/03 1/04 11/04 9/05 6/06 58 Figure Ten: Attacks against Civilian Targets (monthly) Figure Eleven: Attacks against Civilian Targets (weekly) 0 5 10 15 20 25 3/03 3/04 5/05 5/06 0 5 10 15 03/03 1/04 11/04 9/05 6/06 Choosing between weekly and monthly aggregations One of the most critical decisions made in this research was the time period at which to aggregate the data. Data aggregated at the monthly level produce a smoother overall picture of trends in violence, as the preceding timelines illustrate. However, data aggregated at the monthly level may hide important trends that are apparent when the data are aggregated by week. There were important statistical and methodological concerns that ultimately provided the basis for the decision to run the analysis at the weekly level and these concerns are described below. The general rule of thumb is that time-series methodology requires a certain number of observations, usually set at 50 (Biglan, Ary & Wagenaar 2004; Cook & Campbell 1979) in order to accurately determine the error structure of a series. As stated, I have 39 observations when aggregating by month and 153 observations when aggregating by week. Given that I am often concerned with the effects of multiple interventions on each target sector?s time-series, it was necessary to use the weekly data and not the monthly data for the primary analysis to avoid losing statistical power. The converse of this is that there are fewer attacks in each of the weekly categories, and more weeks than months with zero attacks, increasing the variability of my data. I was able to account for this increased variability by logging the data, as will be discussed shortly, but it is important to note that both aggregations are less than optimal. The main thrust of this analysis focuses on the data aggregated by week and the results of this research will be interpreted carefully, so as not to make generalizations that are not reasonably supported by the data. 60 Independent Variable: Major political events The model developed in this research posits that far-reaching political and social events in Iraq are likely to have an effect upon the legitimacy and standing of the various groups active in Iraq, altering the likelihood of their being a target of violence. In order to determine which events to include in this analysis, a wide range of information on political and social events in Iraq was collected. A comprehensive list of more than 500 events in Iraq from January of 2003 to April of 2006 were compiled (for full list see Appendix 4) and then subjected to two inclusion criteria: magnitude and political impact. This list represented every political development, military operation and milestone of major significance to the coalition, the Iraqi people and other related international events. . During the event selection process, Internet and newspaper accounts were used to provide as much detail as possible about each event to determine the projected effect of the event. In line with other research, I refer to these events as interventions, although these events are unique in that, for the most part, they were not designed to directly have an affect on levels of violence and therefore do not fit the more common usage of the term intervention. Inclusion criteria: magnitude The first criteria used to determine the most relevant events was their magnitude. Since very few major events actually occurred within the province of Diyala itself, selected events must be of sufficient magnitude to have a far-reaching effect that might be more likely to be felt across all of Iraq. Events such as national elections, national 61 changes in the governance structure and the capture of Saddam Hussein are representative of the well-publicized and far-reaching events that are believed to be related to trends in violence. While more localized groups might feel the effects of local events more acutely than large scale events, the lack of information on localized events as well as the locations of the groups involved suggests that the most defensible strategy is to utilize events that should have an effect across all of Iraq. Inclusion criteria: political impact As this research is meant to test the importance of competitive group interaction and legitimacy on the targeting of violence, events were selected from the initial list that had strong direct political or social effects. The expected political and social effects were determined by studying media reports of each event. Within Iraq, the social, political and religious landscapes are closely intertwined, as the primary social and thereby political division is one of religion. Many of the political parties are divided by religious affiliation, and events that altered the political or social balance between these groups were more likely to be selected. As political events received a good deal of media coverage, it is no surprise that four of the eight events were overtly political in nature. The other four events ran the gamut from successful coalition operations such as the capture of Saddam Hussein to particularly damaging violent attacks such as the Feb 22 nd , 2006 bombing of the Shia Mosque of the Golden Dome. It is important to note that the events selected were roughly evenly distributed over the time period in question. This is beneficial as it reduces the chance that the 62 effects of one event would confound the influence of other temporally proximal events. The spacing of the events over the time period can be seen below in Figure Twelve. Figure Twelve: Temporal Spacing of 8 Events Event descriptions Now that the basic framework for the selection of events has been described, the following section outlines the events that were selected for inclusion in the model as well as their expected effect on levels of political violence. In order to place the events in their historical context and to see what events occurred immediately before and after, Appendix 4 lists a chronology of every significant event in Iraq covering the time period measured in this research. Event One: The capture of Saddam Hussein, December 13 th , 2003 Saddam Hussein, the deposed former leader of Iraq, was captured by Coalition forces near his hometown of Tikrit on December 13 th , 2003. Hussein had been in hiding since the war's beginning in March and his capture was a major achievement for the 0 0 5 0 10 0 15 0Timelin Event #1 #2 #3 #4 #5 #6 #7 #8 63 coalition. While he remained in hiding, Saddam Hussein was allegedly able to direct resistance against the coalition and there was uncertainty about whether the Coalition would fail in controlling Iraq resulting in Saddam?s return to power (Baker & Hamilton 2006). A number of former Ba'athists, military officers and party officials remained loyal to Saddam and until his capture strove to return his government to power. His capture served to be both a symbolic and practical victory for the coalition, and the legitimacy of their presence in Iraq increased (Baker & Hamilton 2006). Saddam Hussein was tried by the Iraqi government in October of 2005 for his part in ordering the deaths of a number of Kurds in Iraq in 1992 and was executed on December 31 st , 2006. Expected effect: Hussein's capture should increase the legitimacy of the coalition, reducing attacks against them, and conceivably province-wide violence as well. This event also increases the legitimacy of the new Iraqi government and its police forces, which were under the control of the coalition at that time, reducing attacks specifically against them too. Event Two: Abu Ghraib prison scandal, April 30 th , 2004. On April 30 th , 2004, information came to light that American soldiers had been abusing and humiliating prisoners at the Abu Ghraib Prison in Iraq. This abuse was photographed and the photographs became public on April 30 th on the American television program 60 Minutes II. A number of the photographs showed Iraqi prisoners in American custody in situations that were particularly derogatory to Muslims (CBS 2004). As a result a number of US military prison guards and military officials were tried and ultimately convicted of abusing the prisoners. There was a widespread public outcry 64 against the coalition as a result of this scandal, and American Nicholas Berg was abducted and beheaded by insurgents in May of 2004 in retaliation. Expected Effect: I expected this event to reduce the legitimacy of the coalition forces as it portrayed them as having little regard for the rights or dignity of their captives. The legitimacy of the Iraqi government was also called into question as it was still under the control of the Coalition Provisional Authority at this time, although it is more likely that retaliatory attacks would be directed directly at the coalition forces and not at the Iraqi's working with them. This event should have increased province-wide attacks, given its far reaching nature. Event Three: Government transferred to Interim Iraqi Government, June 28, 2004 Control of the Iraqi government was transferred from the US lead Interim Governing Council to the Interim Iraqi government (IIG) on June 28 th , 2004. This was the first major step in returning control of Iraq to the Iraqi people through a democratically elected government. The IIG was still technically under the authority of the Coalition Provisional Authority, but granted the Iraqi's considerable autonomy in preparing their country for its first democratic election for the Iraqi National Assembly. The new Iraqi Prime minister under this government was Iyad Allawi, and his cabinet was made up of 13 Shiites, 5 Sunnis and 5 Kurds, who were also Sunni, as well as a Turk and an Assyrian. While the new government was specifically structured to be as widely representative as possible, it did not enjoy support from all of Iraq's ethnic groups. Local militias, particularly the Shia Mahdi Army under Muqtada al Sadr, remained openly defiant throughout the rule of the Interim Iraqi Government. 65 Expected Effect: The removal of a portion of the coalition influence should increase the legitimacy of the Iraqi Government and its police forces and reduce attacks against them, as well as province-wide attacks. This effect may be tempered by the fact that the new government was not democratically elected, and may have been seen as a proxy for the coalition forces. Furthermore, violent opposition from local militias, particularly in Baghdad, may have offset or overwhelmed any pacifying affect that this transfer had. Event Four: Election Period 2004, Dec 15 th , 2004-Jan 30 th , 2005 The fourth event selected was the election period for the Iraqi Transitional Government and the Iraqi National Assembly. This event was particularly significant as it was the first multi-party election held in Iraq in 50 years. Turn out was high within the Shiite and Kurdish community, but fearing the defeat of their political candidates, many Sunni's boycotted the election. Shiite and Kurdish political parties ultimately won the election and gained control over the Iraqi National Assembly. Expected Effect: According to Weber (1919), elections are a time of increased tension and potential violence as groups jockey for power and influence, suggesting an increase in violence during this time period for all targets across Diyala. Although there was much hope (Baker & Hamilton, 2006) during this election for a fair and representative government, foreign insurgents, dispossessed Ba'athists and other dissidents had a stake in preventing the elections from occurring. This event should produce increases in province-wide violence as well as against sectors such as the coalition, government and the Iraqi police forces which were protecting the election sites. 66 Violence should also be heavily focused on civilians, as only by the targeting of civilians attempting to vote could the disaffected groups alter the outcome of the elections. Event Five: Transitional Government Installed, March 15 th , 2005-April 6 th , 2005 The Iraqi Transitional Government was installed on March 15 th , 2006, following the election. Shortly thereafter the Iraqi National Assembly convened and selected its new Prime Minister and President. The President, Jalal Talabani was a Sunni Kurd and the Prime Minister, Ibrahim al Jaafari was Shia. One of the main goals for this government was to create a constitution for Iraq and prepare the country for transition to a permanent government. Expected Effect: Despite the earlier boycott of the elections by many Sunnis, the representation of both Shia and Sunni on the new government should increase the legitimacy of the Iraqi Govt and its police forces, reducing attacks against them. This may have served to ease inter-group tensions, reducing attacks against civilians as well as against all targets in Diyala. Even with this representation, Shia and Kurd parties controlled the majority of the seats within the National Parliament leading to a growing sense of disenfranchisement by the Sunni. Event Six: Sunni withdrawal from the government, July 20 th , 2005 Following a political stalemate, Sunni members of the Constitutional committee suspended their participation, effectively removing the Sunni voice from the Iraqi National Assembly. Although the Sunni believed they were marginalized in the earlier election they still held seats within the Assembly and therefore had a voice and a role in 67 determining the constitution and other important governmental decisions. By withdrawing from the governmental process, the legitimacy of the government in the eyes of the Sunni?s was dramatically reduced. Expected Effect: This event increased tension between the Shia and Sunni groups and reduced the legitimacy of the government. This should have resulted in an increase in attacks against government and law enforcement targets and potentially all targets in Diyala. Civilian targets should have also experienced an increase in attacks given that this event increased the tension between the Sunni and Shia. Event Seven, Election Period 2005, November 15 th , 2005-December 15 th , 2005 This election was held to elect the first permanent democratically-elected government in Iraq?s history. Turnout was high across Iraq, believed to be near 70% (Baker & Hamilton 2006), especially among the Sunni community that had boycotted the previous election. The Iraqi Accord Front, a Sunni political group, won 15% of the legislative seats, a huge increase given the lack of Sunni representation and participation in the prior government. Additionally, insurgent groups declared a moratorium on attacks during the elections and many formerly antagonistic groups took part in guarding the election sites. Expected effect: Unlike the first election period in which many Sunni's boycotted, widespread participation in this election should reduce inter-group tension, and reduce attacks against government, law enforcement and civilian targets, and province-wide violence. This event should have a long-term effect on the levels of violence, as the 68 newly elected government should be more representative and therefore more legitimate than the prior governments. Event Eight: Bombing of a major Shia shrine, February 22, 2006 The Al-Askari Mosque, referred to as the Mosque of the Golden Dome was bombed in Samarra on February 22 nd , 2006. This mosque was of major religious significance to the Shia community not just in Iraq, but all across the Muslim world. There were retaliatory attacks against Sunnis as a result and tensions were dramatically increased. Although this was a localized event, the significance of the mosque and the retaliatory attacks caused a ripple of violence across Iraq (Baker & Hamilton 2006). The remains of the mosque were further attacked in 2007 by alleged al-Qaeda operatives, deepening the pain felt by many Shia at the loss of a major holy site. Expected effect: This event increased tension between the Sunni and Shia and should increase attacks against civilian targets as well as province-wide violence. As the Iraqi government was still controlled by a Shia majority, these attacks could reasonably be seen as attacking government and law enforcement targets as well. In order to make explicit which hypotheses I expect to be supported by the theoretical model used for this research for each sector, the following table, Table Two summarizes the events and their expected effects. Boxes are marked with a + to show that I posit, based on the conflict model developed in this research that the specific event will increase violence, or a - to show that I expect the event will decrease violence against the indicated target sector. A blank space indicates there I predicted that the 69 event should have no measurable impact upon this target sector. This group of ten interactions where I expect no effect to occur will serve as a quasi-comparison group for this research, and will be discussed once the results from the ARIMA analysis are presented. Table Three: Event Hypotheses and Target Sectors All Police Military Govt Civilian E1: Hussein Captured - - - - E2: Abu Ghraib Scandal + + E3: Interim Iraqi Govt takes power - - - E4: Election One + + + + + E5: Iraqi Transitional Govt installed - - - - E6: Sunni Withdrawal + + + + E7: Election Two - - - - E8: Shia Shrine Bombed + + + + - means that the event would reduce attacks against that target sector + means that it should increase attacks against that target sector Event coding Once the events had been selected, the next step was to determine the duration of the effect they might have on political violence. The events all have finite start-dates that were clearly reported in the sources that described them, making it easy to determine when the expected effect would begin. The two election periods lasted for multiple weeks, and were therefore left ?on? in the database for their duration. The effects of the 70 all events were measured beginning in the week after the week the event occurred or began in. This is not a perfect coding strategy, as an event that occurred on Sunday will have six days before the effect period was measured. Given the nature of the data and the analytical strategy, this method is the most efficient, and will provide reliable results. Furthermore, I have no reason to believe that the events selected would result in a lagged or delayed effect given their magnitude, social and political importance. However, determining the duration of an expected effect is more complicated than determining its starting point. According to McDowall, McCleary, Meidinger & Hay (1980), interventions can have one of four possible effects based on the length of their effect and the abruptness with which it begins. The possible effects consist of an abrupt permanent effect, a gradual permanent effect, an abrupt and temporary effect or a gradual and temporary effect. They suggest that an a priori understanding of the expected effect of each variable should guide the selection of the expected type of effect in the statistical model. In this case, I believe that the eight intervention variables I selected will have an immediate and not gradually building affect. Some of the interventions may have a long-term effect on the frequency of attacks, while some may only have a temporary effect. While I do have a priori expectations of what the effects ought to be, these analyses will test the effects across three different potential effect durations in order to be thorough in determining their effects. Events were first coded as a pulse function, having an effect only during the week that they occurred. The two elections were coded as ranges, while their effect was measured from the first week that they occurred. Next the interventions were coded as having a one month effect, and then a three month effect. For these functions, events 71 were coded as a ?1? during the time period the intervention was thought to have an effect and a ?0 during the pre-intervention and post-intervention phase. The election periods were coded as a ?1? for each week that the elections were occurring. Analytical procedure Analyzing the effects of a series of intervention events using time-series data poses a number of unique challenges. Time-series analyses have to contend with the effects of serial correlation as well as the existence of trends, seasonal, cyclical and disturbance effects (Box & Taio 1975). This research utilizes an autoregressive integrated moving average (ARIMA) model as ARIMA is an effective statistical technique for fitting a model to time-series data that accounts for these effects (Box & Jenkins 1976; D'Alessio & Stolzenberg 1998). ARIMA models work by determining the underlying structural model for a time-series, of which the observed series is just one possible realization. This underlying model is what drives the series, and the residuals between a well-defined model and the actual series should not be significantly different from a white-noise (stochastic) process, eliminating the effects of serial correlation and cyclical disturbances. ARIMA is preferred to conventional regression because it can estimate parameters more reliably as it takes into account the lag structure between observations as well as cyclical trends within the series (Chamlin & Cochran 1998). Standard OLS regression is inappropriate for analyzing time-series data due to the fact that OLS assumes that the error terms in a series are uncorrelated. With time-series data the errors are often correlated as preceding observations often impact subsequent observations. 72 This results in an inflated t-statistic, which increases the likelihood that any significant effect will be erroneous (Legge 1990). There are three components to a standard ARIMA model, labeled p, d, q. These three components allow for corrections to be made to the model in such a way as to remove the non-stochastic elements, allowing for a more valid test of the effects of an intervention upon the data. The p component refers to the autoregressive structure of the model, meaning the number of ?p? preceding realizations or observations that influence the current realization. The d component refers to the presence of an overall trend in the data. If the data exhibits a rising or a falling trend, it must be differenced by subtracting each observation from its prior observation. The d component specifically refers to the number of times, sometimes more than once, that the data must be differenced in order to eliminate that trend. The q component refers to the moving-average of the model, meaning the portion of the preceding q shocks that determine the level of the current observation. The general form of the ARIMA equation for time-series data t X is: ? ? ? ? ???? p i q i t i it di i LXLL 1 1 )1()1)(1( ??? Where the p term refers to the order of the autoregressive term, i ? refers to the parameter of the autoregressive term within the model, d refers to the differencing required to make the series stationary and q to the order of the moving average term and i ? to the moving average parameter. L refers to the presence of a lagged effect term. Finally, t ? refers to the error term of the model, and errors are assumed to be independent and normally distributed with a zero mean. This assumption is supported by a well-tested pre- 73 whitening process, as after correct specification of an ARIMA model of form (p,d,q) the remaining variation in the data should be purely stochastic, following an essentially random white-noise process. In order to ensure that this is the case, the portmanteau test for white-noise, reported through the Q-statistic, is run to test that the adjusted series is not significantly different from a white-noise process. Estimation of the most efficient ARIMA model for a time-series dataset is an iterative process in which values for the autoregressive, moving average, and differencing components are varied and the efficiency and fit of the model checked. This process is referred to as the pre-whitening process, and involves checking the auto-correlation function (ACF) and the partial auto-correlation function (PACF) for each potential model. These functions show the nature of the auto-regressive, trend, and moving average processes at work, allowing for more efficient model specification, as certain characteristics of the ACF?s and PACF?s indicate improper model specification. In order to assess the effect of an event on the number of attacks, an ARIMA model needs to be estimated for the pre-intervention time-series. There is some debate within the field as to whether the estimation process should be done on the entire series or just on the pre-intervention series only. Some authors (Chamlin & Cochran 1998; Legge 1990; Simpson, Bouffard, Garner & Hickman 2006) prefer to use the entire series to estimate the original model, given that many pre-intervention series contain too few data points to get a valid estimate of the structure of the series, however the definitive texts on the ARIMA process use the pre-intervention time-series only for their model estimation procedures (Cook & Campbell 1979; McDowall, McCleary, Meidinger & Hay 1980). 74 One of the main limitations of the ARIMA method is that it requires a long series of prior observations in order to correctly estimate the structure of the auto-correlations (Biglan, Ary and Wagenaar 2000). Estimating the pre-intervention models using fewer than 50 data points is not recommended as it increases the size of the standard error for the parameter estimates, reducing the power of the statistical analysis (Pridemore, Chamlin & Cochran 2007). This precludes the creation of a statistically reliable ARIMA model looking at the first event, the capture of Saddam Hussein, which occurred on the 25 th week, as I don?t have enough baseline data to validly test the effect of this event on subsequent attacks. The second event, the Abu Ghraib scandal, occurred on the 45 th week of the dataset and will be analyzed, although the findings are somewhat weak for this intervention, as discussed subsequently. An exploratory analysis of the effect of the capture of Saddam Hussein is included in the results section further below in order to be consistent with the theoretical model. Other researchers who have modeled data with multiple interventions using an ARIMA process use only the period before the first intervention as the pre-intervention period as the correct model specification applicable to all later interventions (Kim, Myeong & Kweon 2006). I do not believe that this is an appropriate model to follow, as my data are relatively sparse for the first third of the dataset and estimating a model from an early period to apply to an event that falls well within the latter third of the data would miss much of the underlaying structural relationships that are apparent within the data. The best way to determine the appropriate model for my data was to run model specification analyses for each of the four targets sectors as well as for the entire province-wide dataset with each pre-intervention series being a separate model. 75 As a result of the growing frequency of violent attacks over time, each of my series shows increasing error variability as well. The simplest way to account for the non-stationary error variance is to log-transform the data, which doesn?t alter the model specification process procedures (Cook & Campbell 1979; McDowall, McCleary, Meidinger & Hay 1980). Before this transformation took place, a small amount (.1) was added to each of the weeks, as log-transformations can not be done on null-values and many of the time periods contained no recorded attacks. Log-transformation does not substantively change the outcome of the ARIMA process, but does require a reverse transformation to return the resulting coefficients to their more readily interpretable form. The results of these model specification tests, as well as the results of my ARIMA analyses and an initial interpretation of the findings are presented in the following chapter. Power analysis I ran a post-hoc power analysis in order to check whether or not I had a sufficient number of observations for each target sector to be able to be confident in the ability of my ARIMA model to find an effect where there actually was one (Cohen, 1988). Many statistical tests rely on random samples drawn from a population in order to extrapolate results to the larger whole. In this case, I believe I have come as close as possible to measuring the population of political violence in Diyala over the time period measured, and therefore my results are impacted less by the sample-size restrictions of many statistical analyses. 76 Power analyses determine the likelihood that a test will be able to find an effect based on the size of the actual effect in reality. A small effect size is commonly seen as being near .02, a medium effect size near .15 and a large effect size as being .30 or larger (Cohen, 1988). In this research, many of the effect sizes I did find were between a small and medium size. Only the bombing of the Shi?a shrine, event 8, had a large effect size, as the event was associated with an increase of 17 attacks per week from the baseline level of 40-50 attacks per week. Other effects were more moderate, finding increases or decreases ranging from 1-2 attacks to 7-8 attacks per week with varying baseline levels. The results of my power analysis checking for province-wide violence found that, given the 745 events measured in my database, combined with all eight predicted effects from my events, I had a strong chance of finding even a small effect if there was one. For law enforcement targets, with 292 observations and seven predicted effects, I had a strong chance of finding medium-sized or larger effects, but not of finding small effects. For government targets, with 194 observations and seven predicted events, and civilian targets, with 172 observations and five predicted events, I had a strong chance of finding medium-sized or larger effects, but not small effects. For military targets, which had the lowest number of observations, 77, but also the lowest number of predicted effects, 3, I would also have a strong chance of finding medium-sized or larger effects, but not small effects. In sum, my ARIMA analyses had moderate power, and they should have reliably located the existence of all but the smallest effects. 77 Chapter 4: Results This chapter presents the results of my pre-whitening model specification tests as well as the ARIMA results for the effects of the independent variables on violence directed against the five sets of targets. Results for the province-wide analysis are presented first, followed by the results for the four target-specific subsets. An examination of the support provided for each hypothesis is included after the results for each sector are presented. Chapter Five summarizes the findings by target sector and by hypothesis, and then provides a summary of the cumulative effects of each event for the target sectors in question. These three analyses, sector-specific, hypothesis-specific and event-specific, provide a more complete picture of the violence in Diyala than any one type of analysis would provide. I had to choose between including the first two events in my analysis given their theoretical importance and excluding them based on the weakness of the ARIMA model in detecting effects so early in a time series. I decided to include the first two events, the capture of Saddam Hussein and the Abu Ghraib scandal in the analysis as they were too theoretically important to ignore. The ARIMA model can still process the effect that these events had on the trends in violence, just with less confidence in the validity and reliability of the results. Results are presented for these events alongside the rest of the events, although during the discussion of the findings I explain that they should be interpreted with caution. Ultimately, I did not have enough data points before these events to reliably evaluate them and therefore the findings related to these events are given less weight. 78 After using pre-whitening model specification techniques to correctly fit pre- intervention ARIMA models to my data, I found that the province-wide model and three of the four target-specific models only required one change in model specification across the eight events. This generally fits what most of the research I had consulted finds (Biglan, Ary and Wagenaar 2000; Cook & Campbell 1979; McDowall, McCleary, Meidinger & Hay 1980) as few time-series undergo dramatic enough shifts to warrant multiple changes in their model specification. Only the Law Enforcement target sector required three separate pre-event model specifications to ?pre-whiten? the series for each of the eight interventions. I present the specifications for each model along with their Durbin-Watson and portmanteau Q-statistics, indicated in Table Three. The Durbin-Watson test measures the presence of autocorrelation, and the Durbin-Watson statistic ranges from 0-4, with a value of 2 meaning that autocorrelation has been minimized or eliminated. The Q- statistic measures the presence or absence of a stochastic process. It is important to note that the Q-statistic is not meant to be significant as it is a test that the series is not a white- noise process and therefore the higher the P-value the better. The ARIMA model is correctly specified when auto-correlation has been minimized and the series approximates a white-noise process. I was able to achieve the optimal specification for every model used in this research, and the results are presented below. Table Four: Model Specification Results Sector and specific Events Model p,d,q Q-statistic (P-value) Durbin-Watson 79 Province-wide: Events 1-3 0,0,1 9.38 (p<.99) 1.99 Province-wide: Events 4-8 0,1,1 48.16 (p<.18) 1.97 Government: Events 1-3 0,0,1 23.13 (P<.95) 1.98 Government: Events 4-8 0,1,1 36.72 (P<.62) 1.73 Civilian: Events 1-5 0,0,0 43.67 (P<.32) 1.90 Civilian: Events 6-8 3,1,0 46.65 (P<.22) 2.07 Military: Events 1-3 0,0,0 33.67 (P<.57) 1.52 Military: Events 4-8 0,1,1 .80 (P<.58) B-stat* 1.96 Law Enforcement: Events1-2 0,0,0 27.76 (P<.27) 1.92 Law Enforcement: Event 3 0,0,1 40.59 (P<.27) 1.94 Law Enforcement: Events 4-8 0,1,1 34.01 (P<.73) 1.98 * The B-statistic is another test for white-noise that can be employed when the Q-statistic doesn't approach significance despite adequate model specification as determine by analysis of the ACF and PACF (White, Goldkamp & Campbell 2005). In the following section I present the results of my ARIMA analyses and an interpretation of the findings for each model which show the effect of each of the interventions at the three different potential effect durations of one week (immediate effect), one month and three months. Results are presented in their non-logged form for easier interpretation. Positive coefficients indicate that the event was associated with an average increase of the listed coefficient amount in violent attacks during the time period in question, and negative coefficients with a similar average decrease. Non-significant 80 results are presented in the initial tables showing the effects of each event across the five target sectors in this analysis in order to show the general direction and magnitude of the findings. They are not presented in the subsequent charts in order to make it easier to discern the actual effects that each event had on levels of violence. The p-values for the non-significant findings are placed in parentheses after the coefficients. Results for Province-wide Violence Table Five: ARIMA Results for Province-wide Violence Event Model p,d,q Expected direction of change 1 week 1 month 3 months Saddam?s capture (0,0,1) Reduce -.82* -1.17* -.72 (.21) Abu Ghraib (0,0,1) Increase -1.66* .37 (.79) -1.08 (.22) IIG installed (0,0,1) Reduce 3.95** 2.16 (.80) -.94 (34) Election One (0,1,1) Increase -4.04 (.24) 2.17 (.23) 5.61** Trans Govt installed (0,1,1) Reduce -7.42** -2.87 (.32) -.34 (.84) Sunni withdrawal (0,1,1) Increase -3.88** 1.18 (.62) 1.14 (.55) Election Two (0,1,1) Reduce -.60 (.86) -1.92 (.37) -3.27** Shrine bombing (0,1,1) Increase 17.2** 3.43 (.48) 5.5** * = p<.05, ** = p<.01 Figure Thirteen: Timeline of Attacks against All Targets 0 10 20 30 0 50 100 150 Saddam ->AbuGhraib -> <- IIG installed <- Trans Gov Elect 2 -> Bomb-> Elect 1 -> Sunni out -> Analysis of province-wide violence Saddam?s capture and the Abu Ghraib prison scandal appear to have reduced violence against all targets in Diyala. While these findings are statistically significant, as stated earlier the y occurred too early in the dataset to establish reliable baseline data, and therefore my confidence in their validity is low. The installation of the Iraqi Interim Government increased violence during the week of its installation by almost four attacks from the baseline trend. The first election did not register a change in attack frequency across the province; however the three month average level of attacks following the election period was significantly higher than the baseline. The installation of the Transitional Government following these elections greatly reduced violence, as the level of attacks was seven fewer than the week before. However, there were no long-term effects for the governmental transition. The Sunni withdrawal was associated with an immediate short term reduction in attacks, although only in the week it was announced. The second election period was associated with an overall three month reduction in the number of attacks, while the Shia Shrine bombing increased attacks by the largest margin, nearly 17 in the week following the bombing. Hypothesis testing results for province-wide violence Hypothesis 1A posits that events that increase the legitimacy of controlling groups and reduce inter-group tension will decrease violence in Diyala. The following table shows the ARIMA results for province-wide violence that was specifically expected to reduce violence. 83 Table Six: Hypothesis results for province-wide violence reduction Event 1 week 1 month 3 months Saddam?s capture -.82* -1.17 - IIG installed 3.95** - - Trans Gov installed -7.42** - - Second election period - - -3.27** * = p<.05, ** = p<.01 In line with this hypothesis, the capture of Saddam Hussein reduced attacks in the week of his capture, and in the following month, but only by a small amount. Counter to my hypothesis, the installation of the Interim Iraqi government increased attacks by nearly four per week during the week of the change, but there were no long term effects associated with this event for all targets. The second election period was not associated with an immediate change in attacks, although over the three months following the elections there was an average decrease of 3.27 attacks per week. These results provide mixed, but generally positive support for the conflict theory proposition that increased legitimacy and decreased group tension will reduce attacks. In the case of the installation of the Interim Iraqi government, the increase in attacks might be explained by the fact that this was an unelected and therefore less legitimate government that was essentially put into power by the coalition forces. Hypothesis 1B posits that events that decrease the legitimacy of controlling groups and increase inter-group tension will increase violence in Diyala. The following table shows the province-wide results for events that were expected to increase violence. 84 Table Seven: Hypothesis results for province-wide violence increases Event 1 week 1 month 3 months Abu Ghraib -1.66* - - Election Period One - - 5.61** Sunni Withdrawal -3.88** - - Shrine Bombing 17.2** - 5.5** * = p<.05, ** = p<.01 Counter to my hypothesis, the Abu Ghraib scandal reduced attacks in the week it became public. In line with my hypothesis, the first election period was associated with an increase in violence, but only over the three month period following the start of the election, and not immediately during the election period itself. Also counter to my hypothesis, the Sunni withdrawal from the Constitutional committee was seen to reduce violence by nearly four attacks per week during the week of the withdrawal, with no long term effects. Finally, the strongest effect found in this research, the bombing of the Shia shrine, resulted in a massive increase of 17 attacks in the week of the bombing, as well as a 3 month average increase of 5.5 attacks per week. These findings provide weak support, aside from the shrine bombing, for my hypothesis that increased tension and decreased legitimacy will be associated with an increase in violence. Results for Law Enforcement Targets Results for the ARIMA models focused solely on law enforcement targets are presented below in Table Twenty One. 85 Table Eight: ARIMA Results for Law Enforcement Targets Event Model Expected direction of change 1 week 1 month 3 months Saddam?s capture (0,0,0) Reduce -.19 (.28) -.28 (.08) -.33 (.17) Abu Ghraib (0,0,0) No change -.53 (.24) .26 (.69) -.05 (.92) IIG installed (0,0,1) Reduce 2.14** 1.69 (.10) .01 (.98) Election One (0,1,1) Increase -2.17 (.14) 2.31 (.32) 1.67** Trans Govt installed (0,1,1) Reduce -3.02** -1.28* -.81* Sunni withdrawal (0,1,1) Increase -2.81** -.78 (.22) -.67 (.21) Election Two (0,1,1) Reduce -.45 (.63) -.64 (.49) -1.19 (.67) Shrine bombing (0,1,1) Increase -1.17 (.34) .14 (.89) 1.28 (.13) * = p<.05, ** = p<.01 Figure Fourteen: Timeline of Attacks against Law Enforcement Targets 0 5 10 15 0 50 100 150 Saddam -> AbuGhraib -> <- IIG installed <- Trans Gov Elect 2 -> Bomb-> Elect 1 -> Sunni out -> Analysis of Law Enforcement Targeting Attacks against law enforcement targets were only affected by three of the eight interventions. The transfer to the Iraqi Interim Government increased attacks against law enforcement targets during the week of the transfer. This effect is questionable however, as the Iraqi police force really only came into being once the Iraqi government was constituted. The size and visibility of the Iraqi police force has been growing over time, and as it grew its accessibility as a target should have increased as well, accounting for this effect. This steady increase in the attacks against law enforcement targets can be clearly seen in the timeline listed in Chart Eight. The first election period had no immediate effect on violence against law enforcement targets, but does appear to be associated with a long-term increase in violence directed against law enforcement targets. This long term effect is probably a function of the previously mentioned gradual development of the Iraqi police forces. Contrary to this rising trend, the second governmental transfer after the elections reduced attacks over the short-term and also over the long term, although to a lesser extent. Also contrary to the overall rising trend, the Sunni withdrawal from the government had a short term negative effect on attacks against this sector as well. These findings also fit visually with the overall trends in law enforcement attacks series. Hypothesis testing results for law enforcement targets Hypothesis 2A posits that increase the legitimacy of the government or of the police will decrease attacks against law enforcement targets. These events include the capture of Saddam Hussein (event one), the installation of the Iraqi Interim Government 88 (event three), the installation of the Iraqi Transitional Government (event five) and the second election period (event seven). Table Eight below presents the ARIMA coefficients for these models. Table Nine: ARIMA results for Law Enforcement Targets Event 1 week 1 month 3 months Saddam?s capture - - - IIG installed 2.14** - - Trans Gov installed -3.02** -1.28* -.81* Second election period - - - * = p<.05, ** = p<.01 The capture of Saddam Hussein had no effect on levels of violence against law enforcement targets. Contrary to my hypothesis, the installation of the Interim Iraqi government increased attacks against law enforcement targets during the week the new government was installed. In line with my hypothesis, the installation of the Iraqi Transitional government reduced attacks the week of the transition, as well as in the month and three-months following. This event was also associated with a permanent reduction in the levels of violence. Contrary to my hypothesis, the second election period had no effect on violence against law enforcement targets. I discuss this in greater length in the next chapter, but my initial belief is that due to the moratorium on violence that many of the formerly violent groups imposed for the second election period, violence against this target sector could reasonably be seen as falling. 89 Hypothesis 2B posits that events that decrease the legitimacy of the government or of the police will increase attacks against law enforcement targets. The events that should have decreased the legitimacy of the government/police should have been Election One (event four), where the police were guarding the polling places and the Sunni withdrawal (event six). I present the ARIMA results for this event in Table Ten below. Table Ten: Hypothesis results for province-wide violence reduction Event 1 week 1 month 3 months Election One - - 1.67** Sunni withdrawal -2.81** - - * = p<.05, ** = p<.01 Election One was associated with a three month increase in violence against law enforcement targets, but no immediate or shorter-term effects. The Sunni withdrawal was associated with a one-week reduction in attacks against law enforcement targets, contrary to what my hypothesis predicted. There were no other lasting effects for this event however. The following timeline illustrates the trends in weekly violence and where the five events related to law enforcement hypotheses fell. Results for Violence against Military Targets Table Eleven: ARIMA Results for Military Targets Event Model Expected direction of change 1 week 1 month 3 months Saddam?s capture (0,0,0) Reduce -.02 (.55) -.06 (.35) -.05 (.60) 90 Abu Ghraib (0,0,0) Increase -.07 (.15) -.16 (.11) -.22 (.64) IIG installed (0,0,0) None -.16 (.32) -.26 (.09) -.06 (.67) Election One (0,0,0) Increase 1.54** .56 (.49) .72* Trans Govt installed (0,1,1) None -.30 (.40) .51 (.38) .87* Sunni withdrawal (0,1,1) None -.45 (.61) 1.37* 1.10 (.08) Election Two (0,1,1) None -1.34** .92* -.08 (.78) Shrine bombing (0,1,1) None 2.78** .14 (.89) -.32 (.34) * = p<.05, ** = p<.01 Figure Fifteen: Timeline of Attacks against Military Targets 0 2 4 6 8 10 0 50 100 150 Saddam -> AbuGhraib -> <- IIG installed <- Trans Gov Elect 2 -> Bomb-> Elect 1 -> Sunni out -> Analysis of Violence against Military Targets My research found that attacks against military targets were increased during the first election period, and reduced during the second. This makes sense, as the first election was seen as invalid by the Sunnis, while the second featured more widespread participation and a moratorium on violence by many groups. Coalition forces would have been guarding the election sites, and therefore been caught up in violence targeted at the polls. The first election period, and the installation of the new transitional Iraqi Government were both associated with 3 month average increases in violence against the coalition forces. The Shia shrine bombing was associated with an immediate increase in violence, as well as a more gradual long-term effect. This effect was not expected, as the targeting of a religious site should not have provoked retaliation against the coalition. What does not appear to be captured by the model is the decline in violence against the coalition around the time of the Sunni withdrawal. Visual inspection of the monthly trends reveals this pattern clearly, although what accounts for it is not clear from this research. Hypothesis testing for military targets Hypothesis 3A posits that events that increase the legitimacy of the coalition forces will decrease attacks against military targets. The main event that should have increased the legitimacy of the coalition was the capture of Saddam Hussein (event one). This event was not associated with a significant change in any direction for attacks against military targets. 93 Hypothesis 3B posits that events that decrease the legitimacy of the coalition forces will increase attacks against military targets. The main events that should have decreased the legitimacy of the coalition were the Abu Ghraib (event two) scandal and election one (event four). The election period would not have directly reduced the legitimacy of the coalition, but the military forces were guarding the election sites and could have been targeted by groups seeking to alter the current political situation and prevent the elections from occurring through the use of violence. Table Twelve: Hypothesis results for military violence Event 1 week 1 month 3 months Abu Ghraib - - - Election One 1.54** - .72* Abu Ghraib had no effect on violence targeted against the coalition, which is counter to what my model predicted. The first election period, where the coalition forces were guarding the polling places, was associated with a short term increase in violence and three month average increase as well. This finding supports my conflict-based hypothesis. Results for Government Targets The following table lists the model specification results and the ARIMA model results for attacks directed against government targets. 94 Table Thirteen: ARIMA Results for Government Targets Event Model Expected direction of effect 1 week 1 month 3 months Saddam?s capture (0,0,1) Reduce -.14 (.26) -.21 (.17) -.27 (.07) Abu Ghraib (0,0,1) None -.39 (.07) .33 (.54) -.19 (.54) IIG installed (0,0,1) Reduce 1.82** .41 (.63) -.33 (.40) Election One (0,1,1) Increase -1.91** -.31 (.83) 1.15** Trans Govt installed (0,1,1) Reduce -2.57** -1.24* -.14 (.30) Sunni withdrawal (0,1,1) Increase -.13 (.89) .52 (.75) 1.13 (.55) Election Two (0,1,1) Reduce -1.02 (.31) -1.31 (.12) -1.15 (.29) Shrine bombing (0,1,1) Increase 1.22* -1.38** -.55 (.48) * = p<.05, ** = p<.0 Figure Sixteen: Timeline of Attacks against Government Targets 0 5 10 15 20 0 50 100 150 Saddam -> AbuGhraib -> <- IIG installed <- Trans Gov Elect 2 -> Bomb-> Elect 1 -> Sunni out -> Analysis of Violence against Government Targets For attacks against government targets, four events were associated with changes in the levels of violence. The first government transfer increased attacks against the government sector immediately following the transfer, an effect that fits with my hypotheses that a less legitimate government would inspire violence against it. The first election had a short-term reductive effect on violence against this sector, but appears to have increased violence over the long-term. Figure Sixteen shows the overall trend in violence over the period measured by this data, and this long-term increase may be a function, similar to the case for law enforcement targets, of the increased availability of government targets. Contrary to this perspective though, the installation of the new Iraqi government following the election period was associated with a decrease in violence both in the week and the month following the election. The second election period had a long term reductive effect on the violence, but this effect didn?t show up until after the third month following the elections. Finally, the bombing of the Shia shrine increased attacks in the short-term against this sector, but was associated with a decline in attacks during that month, a finding that is difficult to clearly interpret. Hypothesis testing for government targets Hypothesis 4A posits that events that increase the legitimacy of the government will decrease attacks against government targets. Events that were expected to increase the legitimacy of the Iraqi government included Hussein?s capture (event one), the transfer of the government to the Interim Iraqi Government (event three), the installation 97 of the Transitional Government (event five) and election two (event seven). The results are presented below in Table Fourteen below. Table Fourteen: hypothesis results for government violence reduction Event 1 week 1 month 3 months Saddam?s capture - - - Interim Govt installed 1.82** - - Trans Govt installed -2.57** -1.24* - Second election - - - * = p<.05, ** = p<.01 The capture of Saddam Hussein had no impact on levels of violence against government targets. The transfer of the government to the Interim Iraqi Government increased attacks against government targets in the week of the transfer, contrary to my predictions. There were no longer term effects of this transition though. In line with my predictions, the installation of the Transitional Government decreased attacks both in the week of and in the month following its installation. The second election period saw no significant changes in the levels of violence against government targets immediately following the election, although a six-month and permanent reduction in violence appears to follow this transition. Hypothesis 4B posits that events that decrease the legitimacy of the government will increase attacks against government targets. These events included the Abu Ghraib 98 scandal (event two), Election One (event four) and the Sunni withdrawal (event six). Table Fifteen below presents the results of these tests. Table Fifteen: Hypothesis results for government violence increase Event 1 week 1 month 3 months Abu Ghraib - - - Election One -1.91** - 1.15** Sunni Withdrawal - - - Only the first election period was associated with a change in violence against government targets, and this effect was in the opposite direction predicted by my model. Attacks against government targets fell by nearly 2 attacks during the week the elections were initiated. There was a three-month average increase associated with the election period, although not a one-month change, which is unexpected given that the elections lasted for one month and any effect they may have had should have lasted for the duration of that period. Results for violence against civilian targets Table Sixteen: ARIMA Results for Civilian Targets Event Model Expected direction of effect 1 week 1 month 3 months Saddam?s capture (0,0,0) None -.29 (.47) -1.00 (.08) -.67 (.12) Abu Ghraib (0,0,0) None -1.61** -.04 (.89) .08 (.83) 99 IIG installed (0,0,0) None -.15 (.65) .45 (.07) .36 (.15) Election One (0,0,0) Increase .80** -.37 (.61) .44 (.35) Trans Govt installed (0,0,0) Reduce -1.61** 1.07 (.36) .49 (.67) Sunni withdrawal (3,1,0) Increase -.39 (.51) -.02 (.94) .39 (.22) Election Two (3,1,0) Reduce 1.47 (.72) -3.23 (.27) -3.5 (.30) Shrine bombing (3,1,0) Increase 2.44** 1.50* 2.80 (.42) * = p<.05, ** = P<.01 Figure Seventeen: Timeline of Attacks against Civilian Targets 0 5 10 15 20 25 0 50 100 150 Saddam ->AbuGhraib -> <- IIG installed <- Trans Gov Elect 2 -> Bomb-> Elect 1 -> Sunni out -> 101 Analysis of violence against civilian targets Attacks against civilian targets were reduced following the Abu Ghraib scandal, although the relative scarcity of attacks during this time period suggests caution be used in interpreting these results. Attacks against civilian targets saw a small increase during the first week of the election period, but were then reduced when the Iraqi Transitional Government was installed. As expected, the Shia Shrine bombing was associated with a large short-term and moderate long-term increase in attacks against this target sector. Attacks against this sector increase over the time period studied, and it is unclear whether the events measured in this research fully capture this rising trend. Clearly there was an increase in sectarian hostility and the targeting of civilians over the time period in question. Hypothesis testing for civilian targets Hypothesis 5A posits that events that increase tension between groups will increase attacks against civilian targets. These events included election one (event four), the Sunni withdrawal (event six) and the Shia Shrine bombing (event eight). These results are presented below in Table Seventeen. Table Seventeen: Civilian Targets Event 1 week 1 month 3 months Election One .80** - - Sunni Withdrawal - - - Shrine bombing 2.44** 1.50* - 102 * = p<.05, ** = P<.01 The first election period increased attacks against civilian targets during the first week of the elections. This occurred in line with my predictions, however no longer term effects were seen, which is unexpected as the election period lasted for a month. The Sunni withdrawal was associated with a permanent increase in the average level of attacks against civilians, but not a short-term increase. The Shia shrine bombing was associated with an increase in attacks during the week of the bombing as well as a smaller average increase in the month after, suggesting that the event caused a spike in violence that gradually trailed off. This event was also associated with a permanent increase in the average number of attacks against civilian targets. Hypothesis 5B posits that events that decrease tension between groups will decrease attacks against civilian targets. These events included the installation of the Iraqi transitional government (event five) and the second election period (event seven). These results are presented below in Table Eighteen. Table Eighteen: Civilian Targets Event 1 week 1 month 3 months Trans Govt installed -1.61** - - Second election period - - - * = p<.05, ** = P<.01 103 The second election period was not associated with a change in the level of attacks against civilian targets. In line with my hypotheses, the installation of the Iraqi Transitional Government was associated with a reduction in attacks against civilians. 104 Chapter 5: Summary and Discussion This chapter provides an in-depth discussion of the results from the preceding chapter, a discussion of the methodological and theoretical concerns raised by this research and an assessment of the value of conflict theory in predicting violence in Diyala. This chapter begins with a summary and comparison of the different target sectors and an analysis of the effectiveness of conflict theory in predicting violence, including a summary of the support for my hypotheses. Next I include a review of the specific effects that each event had on political violence in Diyala. This chapter concludes with a review of the methodological and theoretical concerns, directions for future research and policy recommendations that this work has brought to light, as well as a concluding statement summarizing this research. Summary of violence against specific target sectors To further summarize the results by target sector, Table Eighteen presents the concurrence between the expected effects based on the model developed in this research and the actual effects as found in the ARIMA analysis. It is useful to summarize these results by sector in order to compare the efficacy of my conflict model in explaining different sectors. The concurrence score is the fraction of results that concurred with my expected effects, including not finding an effect when one was not predicted. A result that matched the expected effect during any of the three time periods analyzed would show concurrence. 105 Table Nineteen: Concurrence between expected and actual effects by target sector Target Sector Concurrence between expected and observed effects All Targets 3/8 = 37.5% Law Enforcement 4/8 = 50% Military 4/8 = 50% Government 2/8 = 25% Civilian 5/8 = 63% My model appears to predict changes in violence reasonably well against civilian targets, moderately well for military and law enforcement targets, and poorly for government and province-wide targets. My predictions were especially poor in predicting violence against government targets, where only two of the eight events had expected effect. Conflict theory is essentially a theory about how different groups interact, partially explaining why the explanatory power of the theory appears to be higher when focused on civilian targeting, although there is no clear reason why my application of the theory to other sectors should have less predictive power. The fact that the model was successful in explaining trends in the targeting of civilians provides support for the continued use of conflict theory in understanding political violence and civil conflicts. These concurrence scores provide mixed support for the idea that legitimacy changing events are a major driver of trends in political violence. In only four cases where I predicted an increase in violence did one actually occur, and in only four cases where I predicted a reduction in violence did one occur. There were 40 potential effects 106 tested within these data (5 sectors x 8 events), and my hypothesized results fit the actual findings for less than half of the cases. There are a number of reasons why this might be the case, both methodological and theoretical and these are discussed following this section. The next section summarizes the results by hypothesis in order to fully explicate the support for conflict theory that this research provides. Summary of results by hypothesis Table Twenty summarizes the full set of results by hypothesis. Within the table, the concurrence column indicates whether I correctly estimated the effect of the intervention within any of the three time periods analyzed, showing concurrence between my hypotheses and my findings from the data. I include an indicator to state when the actual results were in the opposite direction of my predictions. Concurrence indicates that the data provides support for my model. Table Twenty: Comparison of expected and actual effects by hypothesis Hypothesis Expected effect Actual short-term effect Concurrence Hypothesis 1A Saddam?s capture Decrease Decrease Yes IIG installed Decrease Increase No (opposite) Trans Gov installed Decrease Decrease Yes Second election period Decrease No effect No Hypothesis 1B Abu Ghraib Increase Decrease No Election Period One Increase No effect No Sunni Withdrawal Increase Decrease No Shrine Bombing Increase Increase Yes Hypothesis 2A Saddam?s capture Decrease No effect No 107 IIG installed Decrease Increase No (opposite) Transitional Govt Decrease Decrease Yes Election Two Decrease No effect No Hypothesis 2B Election One Increase Increase Yes Sunni withdrawal Increase Decrease No (opposite) Hypothesis 3A Saddam?s capture Decrease No effect No Hypothesis 3B Abu Ghraib Increase No effect No Election One Increase Increase Yes Hypothesis 4A Saddam?s capture Decrease No effect No IIG installed Decrease Increase No (opposite) Transitional Govt Decrease Decrease Yes Election Two Decrease No effect No Hypothesis 4B Abu Ghraib Increase No effect No Election One Increase Increase Yes Sunni withdrawal Increase No effect No Hypothesis 5A Election One Increase Increase Yes Sunni withdrawal Increase No effect No Shrine bombing Increase Increase Yes Hypothesis 5B Transitional Govt Decrease Decrease Yes Election Two Decrease No effect No In only ten of the twenty nine predicted effects was my model accurate in predicting the effect that the intervention would have on the trends in violence. For no one hypothesis test was I correct in predicting all of the effects of each event, but I was also never completely wrong for tests where there was more than one event. In only four cases did I observe a finding in the opposite direction of that predicted by my hypothesis. 108 Taken as a whole, this is difficult to interpret and suggests that my model is relatively ineffective at explaining the targeting of violence in Iraq. As a whole, the results provide weak support for the conflict model, but as described in the prior section, the conflict model is better at predicting violence against civilian targets and less effective for other sectors. The next section describes the specific effects of each event, summarized for each of the sectors. It is important to note that the findings from Table Nineteen and Twenty do not match up directly. Table Twenty shows the hypotheses where I expected to measure a change while Table Nineteen includes null findings in the expected effects. Put more simply, when testing my hypotheses I was only concerned with events that changed levels of violence. When measuring the effects by target sector, when I expected a null finding, and found no effect, I recorded that as concurrence. Discussion of the effects of each event on trends in political violence This section explores the specific effects that each event had on trends in violence in Diyala. Although not directly linked to the testing of my conflict-theory hypotheses, this approach is useful as it summarizes what was learned from this research about each event, which could be extrapolated to similar events in the future. Event One, the capture of Saddam Hussein, was not associated with a significant change in the frequency of attacks for any target sector aside from a small but significant decrease for province-wide attacks. This goes against my initial idea that Hussein's capture would increase the legitimacy of the Coalition forces, which should have directly reduced attacks against them. I believe that no effect was observed because the capture 109 occurred relatively early in the dataset, during a period when attacks were still relatively infrequent and there were not enough pre-intervention or post-intervention attacks for ARIMA to reliably detect an effect, if there actually was one. A simple visual check at the pre-event and post-event levels of violence by week and by month for this event confirms that there were very few attacks in Diyala during this time. Event Two, the Abu Ghraib scandal, also occurred relatively early in the dataset, potentially confounding my ability to properly identify an effect if there was one. I predicted that this event would increase attacks against military targets, which it did not appear to do. There was a significant (p<. 01) one-week reduction in attacks associated with this event for civilian targets as well as a small reduction seen for province-wide attacks. Attacks against this target sector were reduced by 1.6 attacks for the week following this event. However, attacks had been down for the two weeks before the event, and I believe that, similar to Event One, the relative sparseness of attacks in the data at that time, particularly in this target sector suggest that this finding should be ignored. These findings do not fit visually what appears to be occurring in the data. In the dataset, in the week following the Abu Ghraib scandal there was a noticeable increase in violence. Attacks against law enforcement targets rose by 3, against government targets by 2 and against religious/civilian targets by 1 from the pre-intervention levels. The ARIMA analysis did not detect this one-week lagged increase in violence. This lack of concurrence between the ARIMA findings and the actual short-term trends in the data is a result of the nature of the ARIMA process. The ARIMA process estimates the underlying model, that when combined with a series of random shocks (white noise), 110 produces the observed series reflected within the data. When inadequate data are used, in this case from having too few pre-intervention cases, the model produced may be unreliable. Event Three, the transfer of Iraq from direct Coalition control to the Interim Iraqi Government, was the first event for which there was enough pre-intervention data to be able to have confidence in the findings. Instead of reducing violence, this transition was associated with a distinct increase. The transfer of the government increased attacks by over 2 per week against law enforcement targets, as well as by nearly four per week against province-wide targets. The data show that attacks against law enforcement targets remained elevated for a period of three weeks. This event was also associated with a 1.8 attack per week increase against government targets. I had hypothesized that this event should increase the legitimacy of the Iraqi government and its police forces as it reduced the control by of the Coalition, but it may have merely put into a place a Coalition-selected government, exacerbating the conflict already in existence there. These findings run counter to my hypotheses, and I will offer an alternate interpretation after the results from the other events have been discussed. Event Four, the first election period for the new Iraqi government, saw a 1.5 attack per week increase against military targets for the week following the beginning of the election period. Civilian targets experienced a one week increase in attacks as well. Government targets experienced a one week reduction in attacks, but the 3-month average period was associated with an average increase in attacks following this event. A 3 month average increase was seen for province-wide violence as well. All of these results, when taken as a whole, present a clear picture of the relationship between the 111 election period and the associated attacks that was predicted by my legitimacy model. Elections are essentially a political reshuffling of the cards and groups that want to prevent the elections or alter the results in some way may engage in violence in an attempt to terrorize the civilian population. This is directly supported by my finding of increased attacks against civilians, as well as against military and law enforcement forces that were guarding the election sites. As stated, the Sunni largely boycotted this election, and the resulting violence supports my hypotheses here. Event Five, the installation of the democratically elected Iraqi Transitional government, ought to have reduced attacks against the various target sectors, and it did. Attacks against province-wide targets fell by 7.5 during the week of the transition. Attacks against law enforcement targets fell by 3 per week and this event was associated with a one month and three month reduction in attacks. Attacks against government targets declined by 2.5 for the week following the election, and the installation of the government were associated with both a one month and three month reduction in attacks against this sector. There should be no effect on military targets following this transition, as the transfer did not directly alter the legitimacy or standing of the coalition forces and there was no change in levels of violence immediately following the transfer. This event reduced attacks against civilian targets for the week following the transition by 1.6 attacks. All of these findings squarely fit my hypotheses. Event Six, the Sunni withdrawal from the government and constitutional processes was predicted to increase violence. Counter to my predictions, this event was associated with a decrease in attacks for province-wide violence as well as a specific reduction against law enforcement targets, although only for one week, and was not 112 significantly related to any other immediate changes in attacks. There was an increase in attacks against military targets in the month after the event. One week after the week of the Sunni withdrawal there was a major jump in attacks against religious/civilian targets that is also not immediately apparent due to its lagged nature in my ARIMA results. This lagged increase squarely fits my hypotheses, as the Sunni withdrawal would have dramatically increased sectarian tensions. Event Seven, the second election period, saw an increase in attacks against civilian targets immediately after the event. This increase was of more than 2.2 attacks according to the ARIMA model, and attacks remained elevated for four weeks, the entire duration of the election period. This is counter to what I expected would occur, as this election saw a large Sunni turnout and a supposed moratorium on violence. This moratorium wouldn't have kept foreign insurgent groups such as al-Qaeda from launching attacks though, and this may account for these findings. This event was also associated with a decrease in attacks against military targets for the first week of the event, perhaps capturing the expected effect of the moratorium, although attacks rose slightly thereafter. Finally, Event Eight, the bombing of the Shia Mosque of the Golden Dome was associated with an increase in province-wide attacks, as well as for the specific government, military and civilian target sectors. Attacks against government targets increased in the first week after the bombing but then decreased in the month after. Attacks against civilian targets increased in the month following this attack. Military targets were subjected to an increased level of attacks during this week although this effect was short-lived. As predicted by my model, this event appears to have 113 dramatically increased sectarian tension, resulting in the elevated level of attacks against civilians as well as attacks against the government and Coalition forces. To summarize the mixed support for my hypotheses that these results show, as organized by event, the following table compares the expected effect from the conflict theory model with the actual effects of each event, along with a statement about the concurrence between the expected and the actual effects. Table Twenty One: Comparison of expected and actual effects by event Event Expected effect Actual effect Concurrence 1 (Hussein captured) All decrease Police decrease Military decrease Govt decrease Civilian none All decrease Police none Military none Govt none Civilian none All yes Police no Military no Govt no Civilian yes 2 (Abu Ghraib) All increase Police none Military increase Govt none Civilian none All decrease Police none Military none Govt decrease Civilian decrease All no (opposite) Police yes Military no Govt no Civilian no 3 (Govt Transition One) All decrease Police decrease Military none Govt decrease Civilian none All increase Police increase Military none Govt increase Civilian none All no (opposite) Police no (opposite) Military yes Govt no Civilian yes 4 (Election One) All increase Police increase Military increase Govt increase Civilian increase All none Police increase Military increase Govt decrease Civilian increase All no Police yes Military yes Govt yes Civilian yes 5 (Govt Transition Two) All decrease Police decrease Military none Govt decrease Civilian decrease All decrease Police decrease Military none Govt decrease Civilian decrease All yes Police yes Military yes Govt yes Civilian yes 114 6 (Sunni Withdrawal) All increase Police increase Military none Govt increase Civilian increase All decrease Police decrease Military none Govt none Civilian none All no (opposite) Police no (opposite) Military yes Govt no Civilian no 7 (Election Two) All decrease Police decrease Military none Govt decrease Civilian decrease All none Police none Military decrease Govt none Civilian increase All no Police no Military no Govt no Civilian no 8 (Shrine bombing) All increase Police increase Military none Govt increase Civilian increase All increase Police none Military increase Govt increase Civilian increase All yes Police no Military no Govt yes Civilian yes This table illustrates that my model was very effective in predicting the violence that occurred after the first election, the second governmental transition and the bombing of the Shia Mosque of the Golden Dome. In only five cases did I find an effect in the opposite direction compared to what my model predicted. Three of these opposite findings occurred for my province-wide predictions, which may suggest that including every sector in a common model may gloss over the intricacies of conflict theory, resulting in faulty predictions. The model produced only mixed results for the other events analyzed. While there was no specific reason why the model should have been particularly effective for these events and not others, these findings do provide general support for conflict theory. As stated earlier, the interactions where my model predicted that no to change in trends in violence should occur can serve as a quasi-comparison group. There were ten interactions where no changes were expected to be observed, and in six of these cases no change in violence during any of the measured time periods was observed. This finding 115 supports in the conflict perspective predictions in that in sixty percent of the places where I did not expect to find any effect from an event, none occurred. Another interesting finding is that while both elections increased violence against civilian targets, the installation of both governments either had no effect on violence against this sector or reduced it. This appears to fit a group-competition model, as elections encourage competition, but once the results of the elections are decided and the new parties inaugurated, attacks against civilians decline as they would serve little purpose in changing political outcomes from that point forward. Additionally, the first government transition in which the Coalition Provisional Authority selected the Iraqis who would be governing their country, resulted in increased attacks against the new government and law enforcement targets, suggesting that the new government had little legitimacy. The second government installed during this time period was chosen by the Iraqi people, although the Sunni?s boycotted the election in mass, but this government?s inauguration was associated with a decrease in attacks against government and law enforcement targets, suggesting increased legitimacy as a result of the electoral process. One conclusion that can be drawn from this research is related to the evolution of post-invasion Iraq. Processes that move a nation toward a more stable and democratic society, i.e. elections and the installation of new governments, tend to be associated with increases in short-term violence as opposition groups contest the political and social changes that these events herald. For a nation, such as Iraq in this case, to move toward long-term political and social stability, it appears necessary to undergo a period of growing pains in which violence and disorder are elevated. 116 Methodological concerns Interrupted time-series designs provide a fairly effective quasi-experimental test of whether an event or intervention affects subsequent observations within a longitudinal series of events. This methodology is not without its weaknesses however, and this section will discuss some of the threats to the internal validity of this research. The ARIMA methodology is also less than ideal for the dataset and questions involved. ARIMA analyses require a pre-intervention period of sufficient length in order to get stable estimates of the parameters (Biglan, Ary & Wagenaar 2000). There are statistical techniques for performing time-series analyses on short series, but given the sparseness of the data and the number of weeks in which there were no attacks, these techniques are inappropriate for measuring the effects of early interventions. The levels of violence are so low for these early interventions that a simple visual inspection of the data is about the only option for determining whether there were any related effects occurring. As a result, I did not give these first two events much weight in the discussion of my findings. Additionally, the extreme variability week to week for many target sectors calls into question the ability of any statistical process to correctly estimate the true effects that the eight intervention variables had upon political violence in Diyala. One of the main threats to this research comes from ?history?, in that other events occurring at roughly the same time as the selected interventions may be responsible for the observed changes in the dependent variable (Chamlin & Cochran 1998; White, Goldkamp & Campbell 2005). History is essentially an omitted variable problem, where the observed effect or outcome is a result of an untested variable and not of the 117 independent variable in question. In this research I attempted to control for the effects of other events in temporal proximity to my events by choosing the largest and most prominent events. The impact of these major events should dwarf the effects of smaller contemporaneous events, although this is an imperfect control strategy and results should be interpreted with caution. An additional shortcoming of this research is that it focused on only one location, the province of Diyala. The original research question that prompted the creation of the province-specific database was focused solely on Diyala, and therefore this shortcoming is not directly relevant, although it does suggest a direction for future research. A comparison across provinces within Iraq would yield interesting information that this study does not attempt to address. Another related shortcoming is that while the events chosen should have had effects felt across Iraq, local Diyala-specific events may have had effects on the level of violence that were unmeasured and therefore combined in with our results. As mentioned repeatedly, the data used in this research were the best available data, but were also less than optimal. Their accuracy and reliability are unknown, despite the fact that they were verified against the MIPT dataset for the same time period. With few other databases with which to corroborate the data, I was forced to assume that the data are both a reliable and valid representation of the actual violence occurring in Diyala. Even the relatively obvious finding that the data trends upward over time, showing an increase in violence, may be a data collection artifact. The coalition forces were not expecting the violence that continued long after the war was declared over and the reporting of the violence would have been less systematic earlier in the time-series, as 118 there was less attention being paid to it. Fewer media resources would have been directly focused on systematically reporting on these attacks. This may not have been the case with MIPT?s recording of the violence though, as they were an independent source. People tend to find what they are looking for, and the more emphasis put on tracking the violence in Iraq, the more comprehensive the tracking would have become, with part of the increase in the recorded violence due to increased scrutiny. Violence did fall for attacks on military targets in the latter half of the time period measured, suggesting that this potentially confounding factor may be of less importance. Related to the prior point, the validity of the data used is also subject to debate. Whether criminal events were captured by the database is unknown, and what types of reporting bias might be present are also unknown. The rigor with which the data were collected lead me to believe that this is not a problem however the original reporting of events in Iraq, given the country's changing and unsettled nature, may have been more problematic. The accuracy and thoroughness of the primary reporting of events in Iraq may be less than optimal. Target substitution Another potential confounding factor that this research can not account for is target substitution effects. I assume that there are no changes in the availability of targets within each of the four sectors over the time-period being studied, although this assumption is weaker for the Iraqi government and police, who would have seen their numbers increase as time went on and the country?s government became more stable and functional. Targets may change in their vulnerability, military bases or convoys may 119 become more secure, law enforcement patrols or training centers may become more dispersed. This factor was mentioned earlier in the discussion of the targeting of law enforcement targets; as time passed and the new Iraqi law enforcement forces were recruited and trained there would have been more, and more accessible, law enforcement targets. This research posits that the main factor driving an attack against a sector would be its reduced legitimacy, or the presence of strong inter-group rivalry, and this does not take into account changes in the vulnerability of targets over time. An insurgent group that seeks to challenge both the coalition and the Iraqi government and police forces may shift their attacks away from more heavily armed coalition targets in favor of the less well-trained law enforcement arm of the new government. Theoretical concerns Legitimacy is an elusive concept grounded in a subjective assessment of the fairness of and popular support for an authority. Legitimacy is not overly complex, and I have no reason to believe that the direction of my hypothesized effects is incorrect. Even still, I recognize that I use a relatively crude assessment in that I utilize no direct test of legitimacy in this research, and instead rely upon secondary sources and my judgment to discern the legitimacy-changing effects of each intervention. There are few polls of legitimacy in Iraq, and none that directly capture respondents in Diyala (PIPI 2006). The PIPA polls do focus on the perceived legitimacy of the government and the coalition, but their pan-Iraq focus makes them inappropriate for this research. While Diyala should not be substantially different than other provinces or the nation as a whole, Diyala is unique given the much higher percentage of Sunnis, increasing the importance of sectarian 120 harmony to the reduction of violence. There are few expert assessments of the situation in Iraq, particularly Diyala, that have been made public and the situation there has changed significantly since the beginning of the conflict meaning that any legitimacy measurement would have to be taken repeatedly. Another problem with using legitimacy in this model relates to the fact that legitimacy concerns affect different groups differently. An event may increase one group?s legitimacy, but reduce it in the eyes of another group or groups, and this difference is very difficult to accurately capture. There are a large number of groups involved in the violence in Iraq from a wide range of ethnic and ideological backgrounds. Broad-brush attempts to measure legitimacy without a thorough understanding of all of the groups involved are likely to be inaccurate. At best this research is a cross between an empirical study and an educated guess about which conditions might affect legitimacy in Iraq in what way. This should not detract from the worth of this research though, as the conditions do not permit a more rigorous evaluation of the factors involved. Ground- breaking research rarely finds a well-structured pre-existing environment from which to conduct complex analyses. This research does the best it can with the tools and data that are available and provides useful conclusions as well as advice to those who might come later about the conditions affecting violence in Diyala. With any theoretical test there is the possibility that some other theory provides a better explanation for the changes observed. In this case, there are few other sociological or criminological theories that speak to group violence, social and political reorganization and reform and legitimacy. Sherman?s (1993) defiance theory, although not explicitly tied to conflict theory, would make many of the same predictions about violence directed 121 against forces seen to be illegitimate or oppressive. Theories of rational choice would also make many of the same predictions, as groups and individuals would seek to maximize their benefits and minimize their risks by only attacking targets when the popular support for those targets are low, or the expected payoff from such an attack was elevated. In sum, while this research does not pit one theoretical explanation against another, conflict theory appears to be the most general, most appropriate theoretical explanation for the observed changes in political violence in Diyala. Assessing conflict theory Conflict theory provides a reasonable account of the changing trends in violence in Diyala, particularly for attacks against civilian targets. Ultimately, conflict theory's core notion that violence is a function of intergroup conflict is supported by this research. Specifically, I found that the installation of democratically elected, and thereby more legitimate governments tends to reduce violence while the installation of the government selected by the Coalition Provisional Authority increased violence. Events such as the bombing of the Mosque of the Golden Dome and the Sunni withdrawal increased intergroup tension between the Sunni and Shia and were linked to increases in violence against civilian targets, lending further support to conflict theory. While the conflict theory model was less successful in explaining trends in the violence against government, military and law enforcement targets, it was much more successful with civilian targets. As the focus of the theory is most directly on the role of civilians and the differences that define the various groups, this finding suggests that the model, and the manner in which it was applied to the data on political violence in Iraq, are fundamentally sound. 122 There were contradictory findings, such as the bombing of the Shia shrine, which was associated with increased attacks against military targets. This connection can not be explained by the theoretical perspective developed here, as military targets should not have been attacked following what was a strongly religious and civilian-oriented event. Despite the analytical strength of the ARIMA model, and the rigor used in developing the proper model specifications, the high variance between weeks within the data itself made clear effects more difficult to identify. These contradictory findings do not invalidate the conflict model, they only suggest that a more rigorous and better designed and controlled study would be better suited to reach reliable conclusiosns about the predictive value of the conflict theory perspective. In this research I also correctly predicted that there would be no effect between a target sector and a major event in sixty percent of the cases. This lends moderate support to conflict theory?s ability to not only find effects when there are effects, but also to correctly specify sectors that should not have been effected by certain types of events. One of the strengths of conflict perspectives is that they are very difficult to conclusively disprove. As long as there are differentially stratified groups and a scarcity of resources there will always be some form of competition for those resources. Some groups will win out over others, and the natural response, regardless of time or place is to hoard one?s spoils at the expense of the less fortunate. Karl Marx (1848) was not the first to note that greed and the instinct to protect one?s own group are deeply engrained responses. Absent society-changing revolutions, such as those advocated by Marx, the conditions that give rise to competition and ultimately violence, from the conflict perspective, are likely to continue. Conflict-based analyses done on any social problem 123 are likely to find support, given the underlying conflict-theory forces that drive individual and group behavior. Legitimacy is the lynchpin that holds society together despite the inequality and intergroup competition that conflict theory suggests should drive it apart. Groups, even those that find themselves stuck with an unfavorable outcome will often accept their lot if they perceive the process that got them there to be fair. Violence can be used to challenge a stronger groups claim?s to power, but those who seek the support of the local population must use violence in a fashion that supports their own legitimacy, so that their future claim to power will not be tarnished. As this research hints at, it is when legitimacy is reduced that violence and disorder begin to spring up around the naturally occurring fault-lines of society. Future Research To improve upon this research, two things could be done. First, an extension of this research could be conducted on more recent events in Iraq by extending the database of violence in Diyala to the present. Updating the data in this fashion is not easy, as the original LLNL database collection efforts closed down in May of 2006. While other sources have been compiling this information 9 , they lack the rigor and comprehensiveness of the LLNL data and the two datasets are not directly comparable. Second, a comparable model could be applied to other provinces in Iraq, or to Iraq as a whole. These data have been compiled by other agencies, although again with questions about their comprehensiveness and accuracy, and the comparison awaits being done. 9 MIPT, the other main source of data used for this research was closed down March 2008 due to funding cuts. 124 Both of these directions would help determine the effects of political events on the targeting of violence in Iraq through the lens of conflict theory. Outside of Iraq, conflict theory can be subjected to a similar test by applying the theoretical model developed in this research to historical data. The hypotheses tested in this research do not apply just to Diyala, or even solely to Iraq. Similar hypotheses could be developed for other nations or regions undergoing similar changes. Other nations have gone through comparable political changes, although there are no modern examples of a nation being taken over by a military coalition in order to remove a dictator and install a democratically elected government. Few other nations that have gone through such a shift have experienced such a dramatic level of violence, either internally or as a result of foreign insurgency. Nations such as post-war Germany and Japan went through incredible political and social changes under the guard of a foreign military, but there was little insurgent violence or sectarian strife during these changes, owing to their relative homogeneity as well as other cultural and geopolitical factors. Heterogeneous nations such as Bosnia experienced high levels of civil strife during a period of governmental change (Wilson 2006), suggesting that ethnic tensions are a key predictor of violence during periods of change, although there are always examples to the contrary, further complicating the development of strong theory. Policy Recommendations A number of policy recommendations can be drawn from this research, both specific to the situation in Iraq and generalizable to future situations in which political or social upheaval are linked with inter-group violence. The first recommendation is drawn 125 directly from the fact that this research provides support for the underlying presence of conflict theory principles in Iraq. Whenever a country with differentially stratified groups suffers from political upheaval, regardless of the source of the upheaval, it is reasonable to expect that there will be as increase in violence as groups jockey for power and influence. The first step to addressing a problem is preparing for its existence, and therefore I would recommend that nations and the international community prepare to respond to inter-group conflict during international situations similar to what occurred in Iraq. This research does not directly address the best ways with which to soothe sectarian tensions, but it does offer some advice as to what events to watch for. Building upon this first recommendation, and directly supported by this research, elections are a time of political reordering and as seen following the first election, can be a time of increased violence against a range of targets. Special care should be taken to guard election sites, and to ensure that the elections are fair so that the government selected during the election is seen as legitimate. From our standpoint in an established Western democracy, we might assume that the first election should have brought the Iraqi people together, as it theoretically should have enabled them to follow a fair process to elect a legitimate government. Instead, this research found that the first election was associated with a substantial increase in violence in Diyala, directed against nearly all target sectors. Conflict theory suggests that elections are a time for a reshuffling of the political cards and are therefore an ideal time to attempt to influence or coerce rival groups through the use of violence. The second election was also associated with an increase in violence against civilians, as local and insurgent groups attempted to coerce the populace into not 126 taking part in the Democratic process. Future elections should be monitored very closely and security concerns should be balanced against the need to have an open and transparent process that will be seen as legitimate. The greatest increase in violence that this research detected was following the February 22, 2006 bombing of the Al-Askari Mosque, referred to as the Mosque of the Golden Dome. This was a major Shi?a holy site, and the bombing inflamed sectarian passions and touched off a web of sectarian violence in Diyala. Great pains should be taken to reconcile the Shi?a and Sunni whenever a major attack occurs against a religious target in order to avoid touching off retaliatory attacks against other sects. Al-Qaeda understands that the sectarian divide in Iraq can be inflamed, and many attacks against religious targets have been carried out by foreign insurgents seeking to plunge Iraq into a sectarian civil war. The thread that ties conflict theory to the events in Iraq and the changes in violence is legitimacy. As legitimacy appears to be the moderating factor in target selection, the importance of using fair, transparent and inclusive methods to create a new government or to restructure a society is paramount. When an event changes the perceived legitimacy of a group, the Sunni withdrawal from the constitutional process for instance, the likelihood of violence being directed against that group will increase. If a government is not seen as legitimate it is likely to experience an increase in violence directed against it. If a group withdraws from the political process, as the Sunni?s did in July, 2005, sectarian tensions are likely to increase, resulting in violence directed against civilian targets. Governments should strive to increase legitimacy of institutions in areas subject to social disorder and violence. 127 Another effect that can be used to base a recommendation upon, the installation of the Iraqi Interim Government in June, 2004 could reasonably have been expected to reduce violence, as the new government had representatives from all major ethnic groups, and represented an increase in Iraqi control over their national destiny. Instead, this event was associated with a spike in attacks against that government and its fledgling police forces. The installation of a new government always brings with it a period of uncertainty and change, and this change may be seen as an opportunity to contest to the power of the new government openly through the use of violent or nonviolent means. It is recommended that new governments, and the internal and external actors that support them, provide patient support while the new government develops and solidifies. Conclusion Terrorism and political violence are violent actions meant to have a broader political or social effect beyond the harm done to the immediate victims. This research provides a test of the inverse proposition, that political violence can be influenced by political and social events. The simplicity of this proposition belies the fact that violence and political change are both complex issues that are subject to influence by a wide range of variables. Legitimacy appears to be a critical factor in the determination of whether a government and its law enforcement elements will be targeted by political violence. Events that reduce a group?s legitimacy in some cases appear to increase violence against it. While legitimacy is complex, democratic elections and widespread representation of 128 ethnic and religious groups within a political process appear to be important characteristics of a legitimate government. Intergroup conflict appears to be related to levels of political violence. Events, even remote ones, that increase group tensions along religious or ethnic lines tended to increase violence against civilian targets in Diyala. Events that reduce this tension tended to reduce violence against these and other target sectors, furthering the support for the conflict theory model used in this research. This research is not merely a study of the past, as it can provide lessons or predictions for the future. Given the results of this research, it is reasonable to assume that future elections in Iraq, or in other heterogeneous developing nations, will require the support of all parties involved in order to prevent groups from violently interfering with the electoral process. Governments inaugurated without popular support are likely to have their position challenged with acts of violence, directed at them, their law enforcement forces or civilians. Finally, this research suggests that legitimacy is not a zero sum game, in that it is possible to have cooperation and peace, as evidenced by the reduction of violence following the installation of the new Iraqi government, provided there is broad-based support for the new leadership. The ultimate conclusion of this research is that while Diyala and Iraq are unique within the international community, the underlying factors driving political violence there appear to be no different than the factors that moderate political violence in other times and places. One of conflict theory's strengths lies in its historical basis; the theory was created in a period of strife and social change and it applies uniquely well to other situations characterized by strife and rapidly changing conditions. While this research 129 does not find sweeping support for conflict theory, it does suggest that conflict theory- related forces are at work and that a more refined test of this theory would likely uncover further support for its core premise that group conflict and legitimacy are driving forces in the targeting of political violence. 130 Appendices 131 Appendix One: History and Demographic Background of Iraq 10 The land that comprises present-day Iraq is the site of an important section of human history. Iraq currently encompasses most of what was known as ancient Mesopotamia, the ?land between the two rivers?, the Tigris and the Euphrates. A fertile area amidst relatively arid surroundings, many of mankind?s earliest civilizations including the Babylonian, Sumerian and Assyrian kingdoms as well as the ancient cities of Babylon, Nineveh, Ur, Uruk and Nippur were located in present day Iraq. The pride that the Iraqis have for their role in history can be seen in the opening paragraph of the preamble of the recently completed Iraqi Constitution: We the sons of Mesopotamia, the land of the prophets, resting place of the holy imams, the leaders of civilization and the creators of the alphabet, the cradle of arithmetic: on our land, the first law put in place by mankind was written; in our nation, the most noble era of justice in the politics of nations was laid down; on our soil, the followers of the prophet and the saints prayed, the philosophers and the scientists theorized and the writers and poets created (Iraqi Constitution, United Nations translation 2005). Iraq?s approximately 170,000 square miles are home to 28 million people. Roughly the size of the state of California, Iraq is divided into 18 provinces or governorates, with the national capital at Baghdad. Iraq shares borders with Iran to the 10 Where not specified as from a specific source, information from these appendices is drawn from reliable public sources such as the CIA factbook and online sources on the history of Iraq and is considered to be public knowledge. 132 east, Kuwait to the south, Turkey to the north and Saudi Arabia, Syria and Jordan to the west. Although almost entirely landlocked, Iraq has a small port on the Persian Gulf at Umm Qasr as well as access to a number of major rivers including the Tigris, the Euphrates and the Diyala river. Iraq has significant oil reserves, many of which remain untapped. These oil reserves and the agricultural fertility of the central region have made Iraq a relatively wealthy nation as well as a major player in regional and international politics. The agricultural and petroleum resources of the country are concentrated in the north and south (Korany 2005), while the seat of government is in the center of the country, an area without many natural resources. Iraq also sits at a geographic and cultural crossroads between the Arab, Turk, Kurd and Persian cultures, making its culture a unique mix of ethnicities and religions. Iraq?s demographics The Iraqi people are not ethnically homogenous and one of the main social differences is in the practice of religion. Iraq is primarily comprised of three major ethnic/religious groups, Sunni Arabs, Shia Arabs and Kurds. From a religious standpoint, Iraq is 97% Muslim, with 60-65% of the population being Shia and 32-35% being Sunni (CIA Factbook 2007). The largest group are Shia Arabs, who make up 55% of the population. Sunni Arabs comprise close to 20% of the population, and the Kurds who are mainly Sunni comprise 21% of the population. Other ethnic groups make up the remaining 4% of the population. The central region, including Baghdad and many of the main cities is dubbed the Sunni triangle as it is predominantly Sunni. The southern areas 133 of the country are primarily Shia, while the Kurds are concentrated in the northern third of the country. The Diyala province is to the northeast of Baghdad and is often called ?Little Iraq? (Emery 2007) as Diyala?s demographics generally mirror those of Iraq as a whole, a mix of Sunni, Shia and Kurd. This suggests that the results of this research will be generalizable to Iraq as a whole, as Diyala is not substantially different demographically from the rest of Iraq. Iraq is relatively unique in that its people are primarily of the Shia faith. Worldwide close to 90% of Muslims are Sunni and the remainder are Shia or affiliated with minor sects. Iraq has a history of religious tolerance and pluralism, in part due to the fact that Iraq was ruled by the secular Ba?ath party from 1968 to 2003. The Ba?ath party was a socialist and relatively secular government that encouraged pluralism and religious tolerance, although Saddam and most of the Ba?athist political officials were Sunni. Despite this pluralism, the Shia were long oppressed by the Ba?ath party, and the minority Sunni tended to be the recipients of most of the nation?s wealth and resources (Piazza 2006). To understand the relationship between the Shia and the Sunni it is important to understand the origins of the two main sects within Islam. The Shia / Sunni schism occurred in the 7th century following the death of the Muslim Prophet Mohammed (Tripp 2000). The followers of Mohammed were divided as to who the rightful heir and holder of religious authority should be. The Shia, a shortened form of the Shi?at Ali, or ?Party of Ali?, believed that Mohammed?s cousin and son-in-law Ali was the rightful heir, and that authority traveled down his bloodline. The Sunni believed that authority fell to the Caliphate, or the religious/political governing structure that Mohammed had established. 134 This division is still important, as many Shia and Sunni do not view the social and religious institutions of the opposite sect as legitimate. Military Conflict in Iraq Iraq has a long and bloody history of having been invaded by other nations. Each invading group has altered the social and cultural landscape in the process making Iraq a fairly heterogeneous society. The Persians in the 6th century BC, the Greeks under Alexander the Great in the 3 rd Century BC, the Mongols in the 13th century, the Turks until World War One, and the British have each occupied the land that is Iraq (Polk 2005). In 1916 the British and the French divided up the Middle East into regions of control, a compromise known as the Sykes-Picot Agreement. This agreement made Iraq a British domain until 1932 when Iraq gained nominal independence, with Britain maintaining a military presence in the area. Britain re-invaded Iraq during World War Two to keep valuable oil reserves from falling into Nazi hands. Following the granting of independence in the post-WWII era, Iraq has experienced a series of military coups as groups struggled for power, ultimately resulting in the rise of the Ba?athist party under Saddam Hussein (Slugett 2005). Iraq?s most infamous modern political leader, Saddam Hussein, came to power in 1979 as head of the Ba?ath party. The Ba?ath party supported a secular and moderately socialist political philosophy and maintained its control over the nation through oppressive and often brutal methods (Tripp 2000). Hussein?s Ba?athist party espoused socialist and secular beliefs and was despised by fundamentalist Muslim countries in the region, particularly Iran. Saddam Hussein, a Sunni in a predominantly Shia nation, used 135 his political power to benefit the Sunnis at the expense of the Shia and Kurds (Deflem & Sutphin 2006; Hafez 2006; O?Leary 2002). Hussein did manage to bring some order to Iraq, largely through the use of fear and violence, dissuading many groups from attempting to challenge his power (Clarke & Newman 2006). The Kurds and to some extent the Shia were subjected to ethnic persecution under Saddam Hussein?s rule, including the 1988 gassing of the Kurdish village of Halabja that killed thousands of civilians. Far from an isolated incident, Hussein had a history of violence against his own people, and he was ultimately sentenced to death in November 2006 for ordering the killing of 148 Shia in the town of Dujail in 1982. He was executed as a result of this conviction on December 30 th , 2006. The Iran-Iraq War One of the most significant regional events in the last century was the conflict between Iran and Iraq from 1980 to 1988. The longest running war of the 20 th century, Iran and Iraq began the conflict over a border dispute, although centuries of animosity between the two nations had set the stage for the conflict (Tripp 2000). The United States and Russia both openly supported Iraq, in part to prevent the Iranian theocracy, birthed in 1979 under Ayatollah Khomeni, from gaining too much influence in the region. As a result of continuous war with Iran, Iraq had developed into a well-armed society supporting one of the largest land armies in the world. The war ended in a stalemate and resulted in nearly one million casualties across both sides. Chemical weapons were widely used by the Iraq during this conflict, resulting in serious and lasting illnesses for many Iranians. Following the war the Iraqi government was mired in debt, 136 as it had purchased many weapon systems from Russia, China and other allied nations. Kuwait, Iraq?s prosperous southern neighbor, had loaned Iraq $14 billion to help pay for the war. After the war Iraq demanded that Kuwait forgive these debts, but the Kuwait government stood firm, and the stage was set for another conflict. Operation Desert Shield/Desert Storm Iraq invaded Kuwait in August of 1990, capturing their oil fields, the capital city and access to the Persian Gulf in a matter of hours. Widely seen as an unjustifiable violation of the sovereignty of Kuwait, the United Nations demanded that Iraq withdraw from Kuwait. When the deadline for withdrawal passed, the United Nations authorized an international coalition to use force to free Kuwait. The resulting liberation of Kuwait and the subsequent invasion of Iraq, named Operation Desert Shield/Storm, resulted in the expulsion of Iraqi troops from Kuwait and the near annihilation of the Iraqi military. An undetermined number of Iraqi soldiers and civilians died in the conflict while coalition casualties numbered near 300. The coalition forces came near to, but did not enter Baghdad and Hussein seeing that his cause was hopeless sued for peace. As part of the deal to end the war, Saddam Hussein made a number of concessions to the coalition, including participation in an oil-for-food program and the creation of no-fly zones over the northern and southern parts of Iraq (Polk 2005). There was a limited revolt by many Iraqis in the north and south of the nation, particularly among the Kurd and Shiite populations. The no-fly zones were meant in part to limit Saddam Hussein?s ability to retaliate against these groups, although many Iraqi?s were killed in the purges that followed the peace settlement. Although the first Gulf War was quickly won, the peace 137 that was brokered was not as successful as the coalition would have hoped, as Saddam Hussein kept power and remained a threat to regional stability and the security of his own people. It is relevant to note that following the invasion of Kuwait in 1990 by the Iraqis, Osama Bin Laden, the head of al-Qaeda, contacted Saudi Arabia and proposed using his army of mujahideen from the Russo-Afghanistan war to expel Saddam from Kuwait. It was widely feared that Saddam would invade Saudi Arabia after he took control of Kuwait. Bin Laden, a devout Muslim, feared that Islam?s two holiest cities, Mecca and Medina, both in Saudi Arabia, would fall into Iraqi hands (Polk 2005). The Sauds denied Bin Laden?s request and chose to allow the United States military to set up bases within Saudi Arabia for its protection in preparation for the invasion of Iraq. This event served to deepen the hatred of Bin Laden for the United States. Operation Iraqi Freedom On March 20 th , 2003, a United States led coalition of more than 45 countries entered Iraq in a bid to oust the Ba?athist government and its dictator Saddam Hussein. The coalition was made of close to 350,000 troops from 49 different nations. 90% of the troops were from the United States, with Great Britain supplying the majority of the remainder. The invasion was precipitated by intelligence suggesting that Hussein?s regime had acquired weapons of mass destruction and was an active supporter of international terrorist groups (see the final report of the WMD Commission for a review). Rather than wait until Saddam had made the first move, as he did in his 1990 invasion of neighboring Kuwait, the coalition determined the best way to reduce the threat to the 138 stability of the region posed by Saddam Hussein was to move against him with a pre- emptive strike. The pre-emptive strike, dubbed ?shock and awe? (see Ullman, Wade & Alberts 2004, for a review of the effectiveness of the strike) due to the speed and force with which it occurred, was meant to quickly eliminate the Iraqi command and control structure while leaving the civilian population and infrastructure relatively untouched. Although the pre-war intelligence relating to weapons of mass destruction would later prove to be incorrect, by May 1 st coalition forces had driven Saddam into hiding, gained nominal control of the country and set up the Coalition Provisional Authority to begin the process of establishing a democratically elected government in Iraq. The Coalition Provisional Authority was installed to govern the country from April 21 2003 until it was dissolved on June 28, 2004. The Iraqi Interim Government was handed the reigns to the country and prepared the country for the coming democratically elected government. The Iraqi National Assembly was elected in January of 2005 to help with the transition. The Iraqi Interim government was dissolved on May 3 rd , 2005 when the Iraqi Transitional Government took over. The purpose of this government was to draft a Constitution, which was approved on October 15 th , 2005 in a nationwide referendum. The first permanent government of Iraq was installed on May 20 th , 2006. The governments that have been in charge of Iraq, as well as their leaders and their leader?s ethnic background are as follows: 1. Coalition Provisional Authority (oversaw the Iraqi Governing Council) a. April 21, 2003 to June 28, 2004 b. First Chief Executive: retired U.S. Army Lieutenant General Jay Garner 139 c. Garner was replaced by L. Paul Bremer, who headed the CPA until it ended 2. Iraqi Interim Government a. June 28, 2004 to May 3, 2005 b. President: Ghazi Mashal Ajil al-Yawer (Sunni Islam) c. Prime Minister: Iyad Allawi (Shia Islam) 3. Iraqi Transitional Government a. May 3, 2005 to May 20, 2006 b. President: Jalal Talabani (Kurd, Sunni Islam) c. Prime Minister: Ibrahim al-Jafaari (Shia Islam) 4. Current Government in Iraq a. May 20, 2006 to present b. President: Jalal Talabani (Kurd, Sunni Islam) c. Prime Minister: Nouri al-Maliki (Shia Islam) It is interesting to note that the governments attempted to be as inclusive as possible, with rough representation of the major parties, Sunni, Shia and Kurd. Despite the fact that the current President of Iraq is a Sunni Kurd, most Sunni Arabs don?t view the government as legitimate (PIPA 2006). Iraq after Operation Iraqi Freedom As the international community would come to realize, unseating the former Iraqi regime would be the easy part, as the Provisional, Interim and Transitional governments struggled with uniting a nation that was strongly factionalized along religious and ethnic 140 lines (Ikenberry & Slaughter 2006). Furthermore, coalition troops, the fledging Iraqi government and their police and security forces found themselves under nearly constant attack from foreign insurgents and disaffected Iraqis. Despite the capture of Saddam Hussein on December 13 th , 2003, his execution on December 30 th , 2006, the ratification of a Constitution, sets of elections for both a transitional and a permanent government and the installation of a democratically elected parliament, prime minister, and president, Iraq has not seen a cessation in violence. While the coalition did relatively little damage to the infrastructure and generally avoided attacks on the civilian population, many Iraqis were without basic necessities such as water and power following the change of power. Although much of the nation cheered the deposing of Saddam?s government, the failure of the coalition to return the country to a secure and livable environment caused many Iraqi?s view of the coalition operation to sour, reducing the legitimacy of their presence there (PIPA 2006; Schweid 2006). One of the first steps taken by the Coalition Provisional Authority and the Transitional Government was the ?de-Ba?athification? of the country (Ikenberry & Slaughter 2006). Nearly all political appointees in Saddam?s era were members of the Ba?ath party, and systematically removing them from power drained the nation of some of its most experienced leaders and civil servants (Packer, 2006). Many of these civil servants were supporters of the Ba?ath party by necessity and were happy to renounce the party and maintain their positions, although the stigma of being a former Ba?athists prevented their easy reintegration. There are still a large number of former Iraqi military, intelligence and other Ba?ath party officials who were displaced when Saddam?s regime 141 fell and who have low prospects of finding a place within the new Iraq (Deflem & Sutphin 2006; Ikenberry & Slaughter 2006). While Saddam Hussein and the Ba?ath party were widely despised, the absence of the order that they brought to the ethnically and religiously diverse nation of Iraq has created a power vacuum in which various groups are struggling to gain power or settle old scores (Hafez 2006). A sectarian war has broken out between Sunni and Shia, as each group seeks to control the new political apparatus. Once restricted to attacks on the coalition and the new government, terrorist attacks have begun to target civilian populations, mosques and public gatherings. While the Iraqi government was quickly deposed, one unforeseen consequence of the rapidity of this political change was widespread looting, particularly of abandoned Iraqi military posts (Hoffman 2004). Following the collapse of the Iraqi government and military, a significant amount of armaments were lost as soldiers walked off the job taking their weapons with them and the abandoned posts were looted. Additionally, vast stockpiles of munitions, hidden by Saddam Hussein over the years continually appear in the hands of insurgent groups. These stockpiles supply the terrorist groups in Iraq with a nearly unlimited supply of explosives, weapons and ammunitio 142 Appendix Two: Analysis of the main groups in Iraq 11 Iraq is a heterogeneous nation comprised of a number of different ethnic and religious groups, primarily the Sunni and Shia Arabs and the Kurds. Foreign insurgents and coalition forces add two elements of international influence into the competition between the native Iraqi groups. The following section provides a brief introduction to what is known about the Shia, Sunni, Kurdish and foreign insurgent groups: their motivations, goals and relevant relationships with other groups. Much of the violence in Iraq is centered on the regions where there is a mixture of ethnic groups, providing basic support for my hypothesis that conflict theory will be able to explain trends in the violence there. The four Iraqi provinces that comprise the Sunni heartland, home to many Sunnis as well as a significant proportion of the Shia and Kurds, are believed to account for nearly 85% of the attacks (Biddle 2006). The other provinces, mostly Shia and Kurds living in ethnically homogenous communities, are linked to only 15% of the attacks. Mohammed Hafez (2006) categorized the violent groups in Iraq as being in one of two camps based upon their ultimate view for the future of Iraq. System-reintegration groups wish to create a unified and strong Iraq where they are a major political player. System-collapse groups wish to turn Iraq into a puppet or failed state so that they may conduct their business unmolested by a central government. This distinction helps to clarify the basis of the decision to use violence for many groups. System-reintegration 11 Where not specified as from a specific source, information from these appendices is drawn from reliable public sources such as the CIA factbook and online sources on the history of Iraq and is considered to be public knowledge. 143 groups are often willing to work with the government or other groups for the good of Iraq when it fits their interests, while system-collapse groups have little interest in cooperating with other groups. While conflict theory can be used to explain the specific goals and actions of each type of group, the lack of good data about what type of group is responsible for each attack prevents this research from following too closely along this line of thinking. The Shia The Shia Arabs make up 55% of the population of Iraq, and comprise 40% of the population of Diyala (CIA Factbook, 2007). This research assumes that the main goal of the Shia is to gain control of the political system to prevent a non-Shia from using the system to limit their freedoms. This assumption is based upon the historical situation that the Shia found themselves in under Saddam Hussein, who used the government to repress the Shia and Kurds (Tripp 2000). Given that they are the largest demographic group in Iraq, the Shia should be able to use their majority status within the democratic system in order to gain control over the government. Many Shia have embraced this strategy. During the most recent elections Ayatollah Ali al-Sistani, a moderate Shia cleric who represents the middle and upper- class Shia, gained a significant amount of political power by working with the coalition forces (Finn & Zenilman 2003). Through cooperation with the coalition forces al-Sistani pursued a peaceful strategy to ensure the political representation of his people. The second most powerful political figure, at least in Baghdad, is Moqtada al-Sadr, a younger Shia cleric who represents the lower class Shia and controls the Mahdi Army. The Mahdi 144 Army, a Shia militia based in Baghdad, seeks to provide meaningful security and service needs to the local Shia people in order to ensure their support for their leader, al-Sadr (Finn & Zenilman 2003). Originally using the Mahdi Army to oppose the coalition and the new government, al-Sadr has recently become more amenable to the coalition and new government and has used his militia to provide security. Both al-Sistani and al-Sadr cooperated in the last election to garner the largest number of votes and seats in the new parliament, and to ensure that Shia would gain control of the new government. The victory of Ali al-Sistani and Moqtada al-Sadr?s Shia political parties in the election on December 6 th , 2006, resulted in his supporters renouncing terrorism. Data from a PIPA poll (Jan 31 st , 2006) in Iraq found that 98% of Shia polled believed that ousting Saddam was worth doing, and 84% of Shia believed that the country was heading in the right direction, suggesting that they are supportive of the current political situation and their place within it. The Sunni The Sunni Arabs are 20% of the population of Iraq and comprise 50% of the population of the Diyala province (CIA Factbook 2007). The Sunnis lost power following the fall of the Ba?athist regime of Saddam Hussein, and many Sunnis would be happy with a return to the former status quo (PIPA, Jan 31 st , 2006). While the Shia and the Kurdish populations won a significant number of parliamentary seats in the new government, the Sunnis, fearing that they would not received adequate representation, chose to delegitimize the entire political process by boycotting the development of the government in July of 2005 (Hafez 2006). As a result of not having a satisfactory 145 political voice, many disaffected Sunnis have turned to violence in order to force a reshuffling of the political cards. The Sunnis feel that if they do not have a strong political hand, gained through violence or the political process, that they will be disenfranchised completely within the new Iraq and suffer serious economic and social set-backs as a result. Natural resources within Iraq are not equally distributed geographically, and the Sunni people are both a numerical minority and tend to live in areas that are lacking in natural resources (Polk 2005). One of their stated goals is to be able to secure control over the natural resources, such as Iraq?s oil fields, that fall geographically outside of their traditional domain (Hafez 2006). Conflicts over control of natural resources (Collier 2000) are not uncommon in many countries. Control over these resources brings profits, jobs, and greater bargaining power within the Iraqi nation and the international community. Many Sunnis were members of Saddam?s Ba?athist party and have lost power and prestige following the collapse of that regime. These individuals were often high-ranking politicians, military commanders or members of the Saddam?s intelligence services. Many of these former party members have formed insurgent groups to fight the establishment of a non-Ba?athist government in Baghdad (Hafez, 2006). Following the execution of Hussein in December of 2006, many of these groups have joined with the foreign insurgency in attacks on the Shia, the coalition and the Iraqi government and security forces (Iraqi Insurgency 2007). 146 The Kurds The Kurds make up 21% of the population of Iraq, and comprise 10% of the population of Diyala (CIA Factbook 2007). It is important to note that most Kurds are Sunnis, which tips the balance in Diyala in favor of the Sunni sect. The Kurds occupy a unique place among the world?s ethnic groups, as they are the largest ethnic group that lacks a country of its own (O?Leary 2002). There are an estimated 27-36 million Kurds, with the majority living in Turkey and a significant amount in Iran, Iraq and Syria. The governments of the nations in which the Kurds have resided have often persecuted them, leaving their population without a safe homeland. Attempts at creating an independent and self-governing Kurdish nation over the past four decades have failed. The Kurds are less concerned with gaining control over the central government in Iraq, and more concerned with gaining local autonomy and political control for their people (O?Leary 2002). PIPA polls have found that most Kurds (91%) believed that ousting Saddam was the right thing to do, and 76% said the country was heading in the right direction. Given the nature of the data collected for this research, it will be very difficult to determine Kurdish involvement in political violence. The Kurds are rarely mentioned in the political discussions or violent attacks within Diyala, and while their presence will not be ignored, it is less relevant than the Sunni-Shia interaction in Diyala. An exception to this is the Kurdish group, Ansar al-Islam, a Sunni-linked jihadist group with ties to Al- Qaeda (Neumann 2006) that attempted to kill the former Interim Prime Minister of Iraq, Iyad Allaqi while he was in Germany ( Deutsche Welle 2004). Ansar al-Islam is mentioned as the responsible group in a little over 1% of the attacks in Diyala. 147 Foreign Insurgents While the interplay between Shia, Sunni and Kurd political groups defines most of the political landscape in the new Iraq, the presence of foreign insurgents has had a major impact on the situation (Hafez 2006; Ricks 2006). The presence of foreign insurgents adds an additional level of complexity and violence, as the goals of these groups rarely have the best interests of the Iraqis in mind. Despite being a primarily Muslim nation, Iraq?s secular government and decade-long war with fundamentalist Iran led to an aversion by many Muslims in the Middle East toward Iraq. Al-Qaeda and other foreign insurgent groups with religious motivations do not want to see an Iraqi government that does not reflect their predominantly fundamentalist beliefs (Kilcullen 2005; Masharbash 2005). As a result they have launched attacks against the newly elected officials and the polling places during the elections in an attempt to undermine the government?s power. Neighboring states such as Iran that were traditionally hostile towards Saddam?s regime would like nothing more than to turn the new Iraq into a puppet state by installing a government beholden to their own ideals (Hafez 2006). Insurgent groups such as al-Qaeda are working to see Iraq transform into a failed state so that they can establish training camps to be able to spread their campaign further across the region. Many terrorist groups have a decidedly international bent and their stated grievances relate to what they see as a violation of one nation?s sovereignty by another. Often referred to as a ?clash of civilizations? (Huntington 1993) or cultures, the boundaries where Western industrialized countries meet the developing world has been a 148 hotbed for terrorism, violence and social unrest for decades (Hoffman 1998; Korany 2005). Just as Karl Marx opposed the concentration of the means of production in the hands of a few, many violent groups point out that certain western capitalist nations have been using their political, economic and military might to exploit less powerful nations. Many third-world nations have seen companies and governments enter into exploitative agreements to allow the Western world access to natural resources as well as cheaper labor (Korany 2005). While many of the contemporary violent groups operate under a religious banner, their proclamations make it clear that they view it as their religious duty, as opposed to what Marx thought was a political and economic duty, to protect their lands, people and resources from foreign exploitation (Aran, 2003). The following statement made by Al-Qaeda, shortly after the events of September 11 th , 2001, illustrates how the group?s motivations are framed within a conflict perspective. The Islamic nation has been groaning in pain for more than 80 years under the yoke of the joint Jewish-Crusader aggression. Palestine is living under the yoke of the Jewish occupation and its people groan from this repression and persecution while no-one lifts a finger. The Arabian Peninsula is being defiled by the feet of those who came to occupy these lands, usurp these holy places, and plunder these resources. (Aran, 2003, date of statement: October 10 th , 2001) This statement demonstrates that Al-Qaeda, an international jihadi group that seeks to spread its own interpretation of Islam, couches its arguments in the perceived 149 illegitimacy of the West?s presence in the Middle East, their support for Israel and their perceived intent to steal the area?s resources. By reducing the legitimacy of the presence of the foreign nations, Al-Qaeda seeks to encourage others to attack the ?crusaders? and to see al-Qaeda as the legitimate spokesmen and organizer of the resistance. It is difficult to estimate the number of foreign insurgents in Iraq, and conservative estimates place the figure between 1,000 and 3,000 (Krueger 2006; Neumann 2006; US Department of State 2006). The US military believes that between 4-10% of insurgents are from outside Iraq (Finer 2005). To date, captured foreign insurgents have been from Algeria (20%), Syria (18%), Yemen (17%), the Sudan (15%) and from Egypt (13%). Very few insurgents have been directly tied to Iran, which brings into question the Iranian role in the insurgency. Given the relatively small numbers of foreign insurgents, many insurgent attacks are designed to be more spectacular and thus get more publicity than many low-level attacks by other groups. Al-Qaeda and other groups rely heavily on publicity and propaganda to spread their message (Kepel 2004; Packer 2006) and gain support for their cause. Groups such as al-Qaeda depend upon being able to frame their objectives in a fashion that better fits with the religious and social goals of Iraq to garner legitimacy and support from the population. When foreign insurgents targeted coalition forces in an attempt to free Iraq of coalition interference, the population often supported them (Cronin 2002). When the same insurgents targeted civilians or non-combatants their legitimacy was tarnished and groups that had previously welcomed their aid turned upon them (Parker & al-Kubacy 2006). For example, Sunni insurgents specifically told al-Qaeda members not to target polling places in the December elections or else they would face 150 reprisal attacks (Hashim 2003). The Sunni group involved wished to give its people a chance to win power by embracing democratic methods. During the elections, Al-Qaeda sought to oppose the creation of an Iraqi government backed or influenced by the United States, and waged a campaign of terrorist attacks to unsettle the new government and its attempts at providing security and basic services to the Iraqi people. These attacks have weakened the popular support for al-Qaeda and other foreign insurgent groups, and many Iraqis have become more active in repelling these foreign groups (Anderson 2007; Parker & al-Kubacy 2006). This supports my legitimacy hypothesis in that strong perceived legitimacy among the populace will result in a less hospitable environment for violent action. Furthermore, al-Qaeda is attempting to destabilize the country by attacking the Shia, with hopes that the Shia will retaliate against the Sunni and draw the country downward into a civil war (Hafez 2006). Al-Qaeda is not large enough to wage an effective war against the coalition or the fledgling Iraqi government, but by starting a civil war they aim to use one group against another to weaken the system as a whole. The following quote, believed to be from the head of Al-Qaeda in Iraq, Musab al- Zarqawi, illustrates this strategy: [The Shi?a] in our opinion are the key to change. I mean that targeting them and hitting them in [their] religious, political, and military depth will provoke them to show the Sunnis their rabies???and bare the teeth of the hidden rancor working in their breasts. If we succeed in dragging them into the arena of sectarian war, it will become possible to awaken the inattentive Sunnis as they feel imminent danger and annihilating death at 151 the hands of these Sabeans. Despite their weakness and fragmentation, the Sunnis are the sharpest blades, the most determined, and the most loyal when they meet those [Shi?a], who are a people of treachery and cowardice. (Cited from Hafez 2006, page 7) As a result, the foreign insurgents are often viewed with extreme contempt by many Iraqis (Anderson 2006). The nationalist group, the Anbar Revolutionaries, was formed to kill foreign fighters entering Iraq, and is comprised of Iraqis who had lost a relative to Al-Qaeda attacks (Parker & al-Kubacy 2006). The fact that groups are fighting amongst themselves supports the model of inter-group dynamics and political competition developed in this research. 152 Appendix Three: Additional descriptive statistics Although not directly related to the hypotheses being tested, other descriptive statistics help to paint a clearer picture of the nature of the violence in Diyala. This appendix describes some of the additional information available in my database of political violence in Diyala. Within the database, attacks can be broken into ten mutually exclusive categories based on the method of attack used, and the frequency of these attacks is presented below in Table Twenty Two. Table Twenty Two: Attack Type Attack Type Percent of Attacks Small arms fire attacks 30.8% IED attacks (Improvised Explosive Device) 19.2% Assassinations 16.0% VBIED attacks (Vehicle-born Improvised Explosive Device) 9.0% Ambushes 7.4% Indirect fire attacks 4.2% Suicide Attacks 6.9% Military complex attacks 3.9% Kidnappings 1.9% Attacks with knives or sharp objects .6% 153 Small arms fire attacks account for nearly 1/3 of all incidents and IED attacks and assassinations account for a sizable portion of the remainder. IEDs and VBIEDs have been a major problem for the coalition forces, as these types of attacks are very difficult to detect and often utilize large amounts of explosives, causing significant casualties. Fatalities across the other types of attacks vary widely. Table Twenty Three shows the range and distribution of fatalities for attacks within the dataset. In 76 (10%) of the cases, no data on fatalities was listed and these incidents are excluded from the following chart. Table Twenty Three: Fatalities for military, civilian and terrorists Range # of Fatalities Percentage 0 419 57% 1 124 17% Military fatalities 2-50 114 17% 0 384 58% 1 135 18% Civilian fatalities 2-62 138 18% 0 580 79 1 72 10 Perpetrator fatalities 2-21 14 1 This table shows that many attacks resulted in no actual fatalities, although injuries were common. The perpetrator was not killed in a significant portion of the attacks, and the attacks where the perpetrator was killed were often suicide bombings. 154 The attacks were also geo-coded by location with the city closest to which they occurred. 25 of the attacks, or about 4%, were not geo-coded and were listed as ?Diyala? in the data. Aside from the capital city of Baquba, there is very little public information in English available about the other cities. The following table lists the top five cities and the number and percent of attacks that occurred there. Table Twenty Four: Location of Attacks Location # of Attacks Percent Baquba 483 66% Khalis 43 6% Miqdadiya 36 5% Hawija 30 4% Buhriz 29 4% Other locations 112 15% There is little usable geographic information here, aside from the simple fact that most attacks occurred in the city of Baquba. This is most likely due to the fact that, aside from civilian targets, there are many more government, law enforcement and military targets within the city itself. Much of Diyala consists of small towns, arable or unusable land, and there would be little motivation for a politically-motivated attacker to target these relatively insignificant locations. I have very limited data about which groups were responsible for each attack. The groups that were responsible for the incidents are only listed in 14% of the cases. 155 This is actually relatively uncommon, as across the Global Terrorism Database (LaFree & Dugan 2007) collected by the START Center at the University of Maryland, a group claims responsibility for an attack in roughly half of all cases of terrorism. Data on the groups responsible could only be collected where a group publicly claimed responsibility for an attack, something that is relatively infrequent in Iraq. Table Twenty Five: Groups claiming responsibility for attacks Group Responsible # of Attacks Percent Zarqawi Network/ Al-Qaeda 41 5% Sunni Insurgents 37 5% Ansar Al-Sunnah 10 1% Other groups 10 1% Unlisted/Unknown 635 86% Our understanding of these groups is limited by the availability of data about them. What is known about these groups comes from publicly available sources, and is subject to inaccuracy and incompleteness. Fortunately, many groups publish statements in which they detail their goals and intents. As stated in the literature review and appendices, these statements helped confirm the importance of legitimacy and group- conflict as a potential driver of political violence. These statements are useful, but are generally meant to paint a situation in a certain light, and more can be determined about a group?s position by their actions rather than their words (Hamm 2004). 156 Appendix Four: Timeline of events This appendix lists the timeline of events that were compiled in order to determine the most relevant events to be included in the model. This list of events was compiled by a group of undergraduate students from a variety of open sources, including newspaper articles, official reports and websites tracking events in Iraq. No citations are listed here, for the information or the quotations, as the process that generated was not conducive to citation. I include this list for illustrative purposes only, and can not vouch for its accuracy. 2003 Timeline -January- -Jan. 16, 2003-UN inspectors discover 11 undeclared empty chemical warheads in Iraq. -Jan. 27, 2003-The UN's formal report on Iraqi inspections is highly critical, though not damning, with chief UN weapons inspector Hans Blix stating that "Iraq appears not to have come to a genuine acceptance, not even today, of the disarmament that was demanded of it." -Jan. 28, 2003-In his state of the union address, President Bush announces that he is ready to attack Iraq even without a UN mandate. -February- -Feb. 14, 2003-In a February UN report, chief UN inspector Hans Blix indicated that slight progress had been made in Iraq's cooperation. Both pro- and anti-war nations felt the report supported their point of view. -Feb. 15, 2003-Massive peace demonstrations take place around the world. 157 -Feb. 22, 2003-Hans Blix orders Iraq to destroy its Al Samoud 2 missiles by March 1. The UN inspectors have determined that the missiles have an illegal range limit. Iraq can have missiles that reach neighboring countries, but not ones capable of reaching Israel. -Feb. 24, 2003-The U.S., Britain, and Spain submit a proposed resolution to the UN Security Council that states that "Iraq has failed to take the final opportunity afforded to it in Resolution 1441," and that it is now time to authorize use of military force against the country. -France, Germany, and Russia submit an informal counter-resolution to the UN Security Council that states that inspections should be intensified and extended to ensure that there is "a real chance to the peaceful settlement of this crisis," and that "the military option should only be a last resort." -March- --Mar. 1, 2003-Iraq begins to destroy its Al Samoud missiles. -Mar. 14, 2003- The U.S. and Britain's intense lobbying efforts among the other UN Security Council members yield only four supporters (in addition to the U.S. and Britain, Spain and Bulgaria); nine votes (and no vetoes from the five permanent members) out of fifteen are required for the resolution's passage. The U.S. decides not to call for a vote on the resolution. -Mar. 17, 2003- UK's ambassador to the UN says the diplomatic process on Iraq has ended; arms inspectors evacuate; All diplomatic efforts cease when President Bush delivers an ultimatum to Saddam Hussein to leave the country within 48 hours or else face an attack. -Mar. 19, 2003-President Bush declares war on Iraq. 158 -Mar. 20, 2003-The war against Iraq begins 5:30 AM Baghdad time (9:30 PM EST, March 19), when the U.S. launches Operation Iraqi Freedom. Called a "decapitation attack," the initial air strike of the war attempted to target Saddam Hussein and other Iraqi leaders in Baghdad. -American missiles hit targets in Baghdad, marking the start of a US-led campaign to topple Saddam Hussein. -The U.S. launches a second round of air strikes against Baghdad, and ground troops enter the country for the first time, crossing into southern Iraq from Kuwait. Defense Secretary Donald Rumsfeld claims that the initial phase of the war is mild compared to what it to come: "What will follow will not be a repeat of any other conflict. It will be of a force and a scope and a scale that has been beyond what we have seen before." -Mar. 21, 2003-The major phase of the war begins with heavy aerial attacks on Baghdad and other cities. The campaign, publicized in advance by the Pentagon as an overwhelming barrage meant to instill "shock and awe," is in actuality more restrained. -Mar. 24, 2003-Troops march within 60 miles of Baghdad. They encounter much stronger resistance from Iraqi soldiers and paramilitary fighters along the way, particularly in towns such as Nassiriya and Basra. -Mar. 26, 2003- About 1,000 paratroopers land in Kurdish-controlled Iraq to open a northern front. -Mar. 30, 2003-U.S. Marines and Army troops launch first attack on Iraq's Republican Guard, about 65 miles outside Baghdad. Secretary of Defense Donald Rumsfeld deflects criticism that the U.S. has not deployed enough Army ground troops in Iraq. 159 -March 2003 - Chief weapons inspector Hans Blix reports that Iraq has accelerated its cooperation but says inspectors need more time to verify Iraq's compliance. -In the following days US and British ground troops enter Iraq from the south. -April- -Apr. 2, 2003-Special operations forces rescue Pfc. Jessica Lynch from a hospital in Nasiriya. She was one of 12 members of the 507th Ordnance Maintenance Company captured by Iraqi troops on March 23. -Apr. 5, 2003-U.S. tanks roll into the Iraqi capital and engage in firefights with Iraqi troops. Resistance weaker than anticipated. Heavy Iraqi casualties. -Apr. 7, 2003-British forces take control of Basra, Iraq's second-largest city. -Apr. 9, 2003-The fall of Baghdad: U.S. forces take control the city, but sporadic fighting continues throughout the capital. -US forces advance into central Baghdad. Saddam Hussein's grip on the city is broken. In the following days Kurdish fighters and US forces take control of the northern cities of Kirkuk and Mosul. There is looting in Baghdad and elsewhere. -Apr. 11, 2003-Kirkuk falls to Kurdish fighters. - The Los Angeles Times later reported that the fall was ?stage-managed? by the Army -Apr. 13, 2003-Marines rescue five U.S. soldiers captured by Iraqi troops on March 23 in Nasiriya, and two pilots who had been shot down on March 24 near Karbala. -Apr. 14, 2003-Major fighting in Iraq is declared over by the Pentagon, after U.S. forces take control of Tikrit, Saddam Hussein's birthplace and the last city to exhibit strong Iraqi resistance. Saddam Hussein's whereabouts remain unknown. -Apr. 15, 2003-Gen. Jay Garner, appointed by the United States to run post-war Iraq until 160 a new government is put in place, met with various Iraqi leaders to begin planning the new Iraqi federal government. -April 2003- US lists 55 most-wanted members of former regime in the form of a deck of cards. Former deputy prime minister Tariq Aziz is taken into custody. -May- -May 2003 - UN Security Council backs US-led administration in Iraq and lifts economic sanctions. US administrator abolishes Baath Party and institutions of former regime. -May 1, 2003-The U.S. declares an end to major combat operations. -May 12, 2003-A new civil administrator takes over in Iraq. Paul Bremer, a diplomat and former head of the counter-terrorism department at the State Department, replaces Jay Garner, who was seen as ineffective in stemming the continuing lawlessness and violence taking place throughout Iraq. -May 22, 2003-The UN Security Council approves a resolution lifting the economic sanctions against Iraq and supporting the U.S.-led administration in Iraq. -MAY 29, 2003: The U.S. believes to have found the WMD. -May 30, 2003-In separate speeches, U.S. secretary of state Colin Powell and British prime minister Tony Blair deny that intelligence about Iraq's weapons of mass destruction was distorted or exaggerated to justify an attack on Iraq. Both administrations face mounting questions because no weapons of mass destruction (WMD) have been found. Each had claimed that Iraq's WMD were an imminent threat to world security. -June- - JUNE 6, 2003: The U.S. blames Iraq problems on ?pockets of dead-enders? are trying to reconstitute, Gen. Franks and his team are rooting them out 161 -July- -July 2003 - US-appointed Governing Council meets for first time. Commander of US forces says his troops face low-intensity guerrilla-style war. Saddam's sons Uday and Qusay killed in gun battle in Mosul. -Jul. 7, 2003-Bush administration concedes that evidence that Iraq was pursuing a nuclear weapons program by seeking to buy uranium from Africa, cited in January State of the Union address and elsewhere, was unsubstantiated and should not have been included in speech. Over summer Tony Blair faces even stronger criticism than his American counterpart concerning flawed intelligence. -JULY 11, 2003: Condoleezza Rice that doubts about Iraq intelligence were not communicated to the president -Jul. 13, 2003-Iraq's interim governing council, composed of 25 Iraqis appointed by American and British officials, is inaugurated. The council has power to name ministers and will help draw up a new constitution for the country. The American administrator Paul Bremer, however, retains ultimate authority. -Jun. 15, 2003-Operation Desert Scorpion launched, a military campaign meant to defeat organized Iraqi resistance against American troops. U.S. and British troops face continued attacks; about one American soldier has been killed per day since the end of combat was declared. -Jul. 16, 2003-Gen. John Abizaid, commander of allied forces in Iraq who replaced retiring general Tommy Franks on July 7, calls continued attacks on coalition troops a "guerrilla-type campaign" and says soldiers who will replace current troops may be deployed for year-long tours. 162 -Jul. 17, 2003-U.S. combat deaths in Iraq reach 147, the same number of soldiers who died from hostile fire in the first Gulf War; 32 of those deaths occurred after May 1, the officially declared end of combat. -Jul. 22, 2003-Saddam Hussein's sons, Uday and Qusay Hussein, die in a firefight in a Mosul palace. - July 24, 2005-U.S. releases sons' death photos -August- -August 2003- Deadly bomb attacks on Jordanian embassy and UN HQ in Baghdad. Saddam's cousin Ali Hassan al-Majid, or Chemical Ali, captured. Car bomb in Najaf kills 125 including Shia leader Ayatollah Mohammed Baqr al-Hakim. -August 7, 2003- Bombing at the Jordanian Embassy in Baghdad -Aug. 9, 2003-U.S. combat and noncombat casualties reach 255 at 100-day mark after declared end of combat on May 1; 43 British have died. -Aug. 19, 2003-Suicide bombing destroys UN headquarters in Baghdad, killing 24, including top envoy Sergio Vieira de Mello, and wounding more than 100. -Aug. 29, 2003-A bomb kills one of Iraq's most important Shi'ite leaders, Ayatollah Muhammad Bakr al-Hakim, as well as about 80 others, and wounds 125. -September- -Sep. 7, 2003-Continued violence and slow progress in Iraq lead to President Bush's announcement that $87 billion is needed to cover additional military and reconstruction costs. -October- 163 -Oct. 2, 2003-According to an interim report by David Kay, the lead investigator searching for weapons of mass destruction in Iraq, no WMDs have been found as yet. -Oct. 5, 2003-White House reorganizes its reconstruction efforts in Iraq, placing National Security Adviser Condoleezza Rice in charge and diminishing the role of the Pentagon. -Oct. 16, 2003-The UN Security Council unanimously approves the U.S. and UK resolution on Iraq's reconstruction, which supports an international force in the country under U.S. authority. Several countries originally opposed the resolution unless Washington agreed to a faster timetable for transferring power to the Iraqis, but in the end voted for the resolution without requiring changes. -Oct. 23?24, 2003-The Madrid Conference, an international donors' conference of 80 nations to raise funds for the reconstruction of Iraq, yielded $13 billion in addition to the $20 billion already pledged by the United States. This amount fell short of the overall target of raising $56 billion, the figure the World Bank and the UN estimated that Iraq needs over the next four years. -Oct. 27, 2003-Four coordinated suicide attacks in Baghdad kill 43 and wounded more than 200. Targets included the headquarters of the Red Crescent (Islamic Red Cross) and three police stations. -November- -Nov. 2, 2003-In the single deadliest strike since the Iraq war began, guerrillas shoot down an American helicopter, killing 16 U.S. soldiers and injuring 21 others. Other attacks over the course of the month make it the bloodiest since the war began: at least 75 U.S. soldiers die. 164 -Nov. 14, 2003 -The Bush Administration reverses policy and in a deal with the Iraqi Governing Council, agrees to transfer power to an interim government in early 2004. -December- -Dec. 9, 2003-A directive issued by Paul Wolfowitz, deputy secretary of defense, bars France, Germany, Canada, Mexico, China, and Russia from bidding on lucrative contracts for rebuilding Iraq, creating a diplomatic furor. -Dec. 13, 2003-Iraq's deposed leader Saddam Hussein is captured by American troops. The former dictator was found hiding in a hole near his hometown of Tikrit and surrendered without a fight. 2004 -January- -Jan. 18, 2004- Car bomb in Baghdad leaves 20 dead -Jan. 25, 2004- The U.S. admits that there is no evidence that Iraq stockpiled WMDs -Jan. 27, 2004- Insurgents killed 6 US soldiers -February 2004- -Feb. 1, 2004- At least 67 dead in suicide blasts in Kurdistan -Feb.10,2004- Car bomb kills at least 50; 47 are Iraq army recruits -Feb. 21,2004-Red Cross visits Saddam in Baghdad -March- -Mar. 2, 2004- 182 killed in attacks in Baghdad, Karbala -Mar. 8, 2004- Interim constitution signed -Mar. 9, 2004- Abu Abbas, Achille Lauro hijacker, dead 165 -Mar.28,2004-Protests as U.S. closes Iraqi paper -Mar.31,2004- Iraqis hang U.S. bodies off bridge U.S. officials said they expect more attacks like the one that killed 4 U.S. civilian contractors -April- -Apr. 6, 2004-Heavy fighting in sealed Falluja -Dozen U.S. Marines killed -Apr. 7, 2004-Coalition battled Shiite, Sunni insurgents -Apr. 8, 2004 -Japanese, Koreans taken hostage -Apr. 11,2004- Truce bid calms Falluja fighting -Apr. 16,2004- Missing U.S. soldier on videotape -Apr. 20,2004- Iraqi tribunal in Baghdad to try Saddam Hussein -Apr. 21,2004-Car bombs in Basra kill 68 -Apr. 27,2004-U.S. forces kill 64 insurgents near Najaf -U.S. warplanes pound Fallujah targets -Apr. 29,2004- Former Iraqi generals offer Fallujah plan -Apr. 30,2004-Iraq prison abuse sparks outrage. A coalition military spokesman is 'disgusted' by photographs that apparently show U.S. soldiers abusing detainees. -May- -May.11, 2004-Video shows U.S. captive beheaded Captors say they offered exchange man, identified himself as Nicholas Berg, 26, for Abu Ghraib prisoners -May.14,2004-Interrogation techniques banned in Iraq -May 17,2004-Interim council head killed in car bombing Ezzedine Salim was killed as he was waiting to enter headquarters of the coalition. Several other people also died 166 -May. 20,2004-Chalabi angered by raid on home -May. 24,2004-Bush outlines 5-step Iraq plan -May. 28,2004-Allawi named as future prime minister. Former exile Iyad Allawi has been chosen to head an interim government. -May 31, 2004 - Coalition Provision Authority (CPA) Order number 92 establishes the Independent electoral Commission of Iraq (IECI) as the exclusive electoral authority in Iraq. -June 2004- -Jun. 1,2004-Interim Council in Baghdad dissolved as new president named -Jun. 8,2004-Iraq Kidnapped Italians, Pole freed -Iraq plan gets unanimous backing in US -Jun. 16,2004-Probe rules out Iraq-9/11 links -Jun. 17,2004-Car bomb kills 35 in Baghdad -Jun. 22,2004-South Korean hostage beheaded Insurgents carried out their threat to behead Kim Sun-il they were holding hostage -Jun. 23,2004-'Zarqawi tape' threatens Iraq PM -Jun. 24,2004-Multiple attacks kill 90 -Jun. 27,2004-US hands over sovereignty in Iraq US administrator Paul Bremer gave legal documents to an Iraqi judge two days ahead of schedule. -June 28, 2004 - Sovereignty is transferred from the US-led coalition to the Interim Iraqi Government. -July 2004 - 167 -Jul. 1, 2004-Defiant Saddam in court: Saddam Hussein is hearing preliminary charges against him that included the gassing of Kurds and the invasion of Kuwait. -Jul. 7,2004- Iraq adopts emergency powers law -Jul. 13,2004-Manila begins troop pullout A move is being made to satisfy demands by kidnappers of a Filipino hostage -Jul. 15,2004-Iraq to form new security force in Baghdad. Attacks by insurgents continued as the government is establishing an intelligence service to 'annihilate' terrorist groups -Jul. 23,2004-Militants take Egyptian diplomat hostage -Jul. 24,2004-Militant threat to Australia and Italy -Jul. 28,2004-Day of violence leaves at least 115 dead 68 killed in a Baquba suicide bombing and 42 in fighting in south-central Iraq. Militants claim to kill 2 Pakistanis -August- -Aug. 1,2004-Bomb blasts rock churches -Aug. 6, 2004-Marines: 300 insurgents killed in Najaf -Aug. 7,2004-Iraq amnesty targets insurgents -Aug. 11,2004-U.S. launches offensive in Najaf The military launched a major offensive against militants loyal to renegade Shiite cleric Muqtada al-Sadr. -Aug. 12,2004-Radical cleric wounded in Najaf -Aug. 13,2004-Najaf truce brings calm to city -Aug. 15,2004-Conference on interim assembly -Aug. 17,2004-Al-Sadr rebuffs truce delegates 168 -August 18, 2004 - Iraqi National Conference selects a national assembly to act as a parliament until elections are held. -Aug. 19,2004-Fierce fighting seen in Najaf -Aug. 25,2004-Al-Sistani returns, urges march on Najaf -Blast at mosque kills 74 -Aug. 26,2004-Ayatollah reaches peace deal with al-Sadr Al-Sadr has reached an agreement with al-Sistani that grants his freedom from murder charges and secures peace in Najaf -Aug. 31,2004-Militants: 12 Nepalese 'killed' -September 2004- -Sep. 7,2004-U.S. death toll in Iraq passes 1,000 More than three-quarters have died in combat, and 647 of have been killed since Bush declared an end to major combat -Sep. 14,2004-Car bomb at police station in Baghdad kills 47 Assailants launched two deadly assaults at police targets. -12 police officers killed in a drive-by shooting in Baquba -Sep. 15,2004- Iraq war illegal, says U.N. head Kofi Annan -Sep. 20,2004-U.S. hostage beheaded -Sep. 23,2004-Iraq PM thanks U.S.; Interim Iraqi Prime Minister Ayad Allawi is laying out his government's aspirations to a joint meeting of Congress -Sep. 24,2004- Six Egyptians taken in Baghdad -Sep. 27,2004- Kidnapped Iranian diplomat freed -Sep. 28,2004-Italians freed, Frenchmen to be released -Sep. 30, 2004- BaghdadBombs toll hits 45 169 -October 2004- -Oct. 1,2004-109 insurgents killed in Samarra 3,000 U.S. troops moved into Samarra. One American soldier was killed -Oct. 8,2004-Hostage Bigley killed -Oct. 10,2004-Weapons handover in Sadr City -Oct. 11,2004-Nuclear assets 'vanish' -Oct. 12,2004-Babies found in mass grave -October 12, 2004 - Elections regulations published by IECI. -Oct. 14,2004-Green Zone blasts kill 8 -U.S., Iraqis in Falluja assault -Oct. 15,2004-Blasts at churches in Baghdad -Oct. 16- US bombers strike at Falluja -Oct. 18 -U.S. bombs targets linked to al-Zarqawi -Oct. 19 -Agency halts aid projects after kidnap -Oct. 20 - Abu Ghuraib prison guard pleads guilty -October 21, 2004 - Voter registration materials distributed. -Oct. 22 - Falluja raid 'nets Zarqawi aide' -Oct. 24 - Massacre of 50 Iraqi army recruits The bodies of recruits have been found following an ambush near the Iranian border -Oct. 25 - IAEA: Tons of Iraq explosives missing -Oct. 27 - British troops move toward Baghdad -Oct. 28 - Militants claim killings of 11 Iraqis 170 -Oct. 30 - Eight U.S. Marines killed -Oct. 31 - U.S. forces launch new assaults in Falluja -November 2004- -November 1, 2004 - Voter and candidate registration begins for the election of a transition government. -Nov. 3 - Three Iraqi guards 'beheaded' -November 4, 2004 - Determination is made to allow Iraqis abroad to vote in the elections. -Nov. 5 - US military 'seals off Falluja' -Nov. 6 - Rebel attacks in Samarra kill 33 -Nov. 7 - Allawi declares state of emergency Iraqi commandos seize Falluja hospital -Nov. 8 - Battle for Falluja under way -Soldiers advanced into Falluja -Nov. 9 - U.S., Iraqis enter Falluja's center 10 U.S. troops and 2 Iraqi soldiers had been killed, 22 others wounded. -Nov. 10 - Troops find hostage 'slaughterhouses' -Nov. 11 - At least 19 die in car bombing -U.S. troops push into south Falluja -Nov. 13 - '1,000 Falluja insurgents killed' -Nov. 14 - U.S.: 'Enemy is broken' in Falluja Marines spread through the deserted streets, kicking in doors during a dangerous house-to-house search -Nov. 16 - Margaret Hassan apparently killed The director of CARE International in Iraq 171 was kidnapped on October 19. Videotapes showing her killing -Nov. 18 - Falluja assault 'breaks' rebels -Nov. 21 - Elections set for January 30 -December 2004- -Dec. 1 - U.S. Troop Numbers in Iraq to Hit 150,000 -December 1, 2004 - Recruitment of polling station staff begins. -Dec. 3 - Baghdad Violence Kills at Least 25 Iraqis -Dec. 5 - Insurgents Kill 23 Iraqis Countrywide -Dec. 15 - Shi'ites Bombed as Poll Campaign Begins -December 15, 2004 - Deadline for registering parties and individuals for the coming election. Validation of voter registration list. Election campaign starts. -Dec. 19 - Bombs in Karbala, Najaf kill more than 60 In Najaf, a suicide car bomber plowed his vehicle into a funeral procession just yards from Imam Ali shrine, killing 48 -Dec. 20 - Police arrest 50 after bombings -Dec. 21 - 22 killed in attack on U.S. base Multiple rounds hit a dining hall near Mosul, killing U.S. troops, members of the Iraqi national guard, and Iraqi civilians -Dec. 23 - Rumsfeld arrives for surprise visit 2005 January -January 1, 2005 - Process of distributing ballots begins. -The the militant group led by Abu Musab al-Zarqawi, released a video of five national guardsmen being shot dead in a street as passers-by stopped to watch. 172 -January 2, 2005 - A suicide car bomber struck a bus carrying Iraqi national guardsmen north of Baghdad, killing 26 people and wounding six, as insurgents showed no signs of letting up in their campaign to wreck elections scheduled for the end of the month. -Four Iraqi policemen were killed and one wounded when their vehicle was ambushed near Samarra. -In Diyala province in the east the deputy governor, Ali Haddawi, was shot dead. -January 3, 2005 - A roadside explosion and three separate car bombs in Iraq killed at least 17 people, including three British civilians, as insurgents pressed their campaign to disrupt the general election, and the defence minister said that polling might be postponed -An explosive-laden car struck near Mr Allawi's party headquarters in Baghdad, killing two police officers and one civilian. -Another car bomb in Balad killed four Iraqi national guard soldiers and wounded 14. -Yet another explosion, at a roadside in Saddam Hussein's home town Tikrit, killed at least six guardsmen and injured four others, the police said. -January 4, 2005 - Ali al-Haidri, the governor of the Baghdad province, was assassinated in the al-Hurriya suburb of Baghdad as his car was ambushed in the street in the most high-profile killing for several months. Al-Qaida in Iraq, a militant group led by the Jordanian Abu Musab al-Zarqawi, claimed responsibility for the attack and said that it had killed a "tyrant and American agent". -It also claimed responsibility for a suicide truck bomb attack on an Iraqi commando base near Yarmouk hospital, the second time the base had been hit, killing 10 people, including eight Iraqi commandos, and injuring 60. 173 -In separate attacks, 5 American troops were killed and a further 10 Iraqis died when a suicide truck bomber drove into an Iraqi commando base in Baghdad early in the morning. -January 5, 2005 - The Iraqi prime minister, Ayad Allawi, insists that the general election would go ahead as planned in three weeks' time, despite growing calls for a delay. - Three suicide bombers struck in separate towns, killing more than 20 people. -In Mosul, Omar Mahmoud Abdallah, an official of the Iraqi Islamic party (the mainstream Sunni party) was assassinated. In another incident a prominent Iraqi union leader, Hadi Salih, was shot dead at his Baghdad home. -Many of the bigger Sunni parties have withdrawn from the campaign, leaving the minority community from which the violent insurgency has grown likely to be heavily under-represented in the national assembly. - A suicide car bomber attacked a police graduation ceremony in Hilla, south of Baghdad, killing at least 15 people, including at least 10 policemen. The militant Islamic Army in Iraq claimed responsibility. - In Baquba, a third car bomb killed six at a checkpoint. Earlier, a police colonel and his driver were shot dead in the town. -January 12, 2005 - The U.S. ends its hunt for weapons of mass destruction in Iraq. The two-year search failed to yield the weapon stockpiles that Saddam Hussein's regime was alleged to possess. -January 23, 2005 - Last day for expatriates to register to vote. -JANUARY 27, 2005: 30 Marines, Sailor Die In Copter crash in Iraq, the deadliest single event for U.S. forces since the invasion 174 -January 28, 2005 - Official campaign period ends. -January 30, 2005 - An estimated 60 per cent of Iraqi voters - about 14 million people - turn out to elect a transitional National Assembly in the country's first free polls in 50 years. The elections are won by a Shiite coalition, the United Iraqi Alliance, who win a majority of the assembly seats. Kurdish parties come second -February -February 3, 2005 - Gunmen kill 12 Iraqi soldiers in an ambush near the village of Zab south of Kirkuk. -Insurgents kill five Iraqi policemen and an Iraqi National Guard officer in an ambush on Baghdad?s airport road. -February 6, 2005 - Insurgents kill 17 Iraqi policemen and 5 Iraqi national guardsmen in an attack on a police station in Mahawil, south of Baghdad. Fourteen of the militants are also killed. -Iraqi security forces disclose the December 20 capture in Beiji, 155 miles north of Baghdad, of Khamis Masin Farhan Ugaydi, a former Saddam Hussein-era army general suspected of financing insurgent bombings and plotting attacks on election targets. -February 7, 2005 - A suicide car bomb kills 15 civilians and wounds 17 outside the main police headquarters in Baquba. -February 8, 2005 - Gunmen open fire on the convoy of Mithal al-Alousi, the general secretary of the Iraqi Nation Democratic Party, in western Baghdad, killing his son and bodyguard. -A suicide bomber kills 21 Iraqis and wounds 27 near an Iraqi army recruitment center in central Baghdad. 175 -February 10, 2005 -A suicide car bomber kills four Iraqis in a public square in central Baghdad. -February 11, 2005 - A car bomb kills 12 Iraqis and wounds 17 outside a Shiite mosque in Balad Ruz. -February 12, 2005 - A suicide car bomber kills 17 Iraqis at the entrance to a hospital south of Baghdad. -A masked gunman kills 10 at a bakery in a Shiite area of Baghdad. -Shia holy religious mourning period begins. -February 13, 2005 - Official results of the January 30 elections for seats in the 275- member National Assembly are announced. The United Iraqi Alliance, a slate backed by Shiite Muslim clergy, receives 48% of the vote (140 seats); the Kurdistan Alliance, a coalition of the two main Kurdish factions, earns 26% (75 seats); the Iraqi List, headed by interim Prime Minister Ayad Allawi, gains 14% of the vote (40 seats). Voter turnout is reported at 58%. -Gunmen firing from a car kill two high-ranking policemen in Baghda. -February 14, 2005 - One roadside bomb kills a U.S. soldier and wounds three others, and a second kills three Iraqi national guardsmen, both near Baquba. -A mortar attack kills an Iraqi woman and a 2-year-old girl and wounds seven others in Samarra. February 18, 2005 ? -A suicide bomber kills 15 and wounds 24 as Shiites celebrate the religious festival of Ashura in a procession to al-Khadimain mosque in southern Baghdad. 176 -A suicide bomber kills two and wounds eight outside the al-Bayaa Shiite mosque in western Baghdad. -A suicide bomber kills two and injures five near a religious procession in a Shiite district in northwestern Baghdad. -A rocket fired in the direction of the Husseiniyat al Thaqilan mosque in northwestern Baghdad hits a coffee shop, kills one Iraqi, and wounds three. -In Karbala, police find the bodies of two kidnapped police officers, both the sons of the chief of police of Najaf, Ghalib al-Jazaeni. -A roadside bomb kills a U.S. soldier and wounds another during a patrol in Tal Afar, west of Mosul. -February 22, 2005 ? -Day of Ashura ? the holiest day in Shiite Islam. -Ibrahim al-Jaafari is chosen by the victorious Shiite alliance as its candidate to become Iraq?s new prime minister. -NATO leaders announce in a joint statement that all 26 members of the alliance will help train Iraq?s new security forces. -Six suicide bombers, including one on a bicycle, kill 39 people and wound about 150 in five separate attacks in and around Baghdad. The attacks are meant to disrupt Ashura, the holiest day in Shiite Islam. -February 24, 2005 - A suicide car bomb kills 12 policemen and wounds 29 others at a police station in Tikrit -A roadside bomb kills two Iraqi policemen and wounds three others in Kirkuk. 177 -A roadside bomb kills a U.S. soldier in Qiryat in Diyala province, and an improvised explosive devices kills a U.S. soldier and wound two others near Samarra -U.S. forces detain 29 suspected insurgents and seize weapons caches during operations throughout Anbar province -Iraqi national guard troops capture Mohammed Najem Ibrahim, a suspected terrorist cell leader with links to Jordanian militant Abu Musab al-Zarqawi in Baquba. -February 25, 2005-A roadside bomb kills three U.S. soldiers and wounds eight in Tarmiya. -February 27, 2005 - Iraqi officials announce Syria has captured and handed over a half- brother of Saddam Hussein, Sabawi Ibrahim al-Hassan al-Tikriti, who is accused of playing a leading role in organizing and financing the insurgency in Iraq. - FEBRUARY 28, 2005: Car bombs kill at least 114 Iraqis in Hilla, south of Baghdad. [BBC, 2/28/05] -March -March 2, 2005 - Suicide car bombers kill 13 Iraqi soldiers and wound 30 in two separate attacks in Baghdad. -Gunmen kill a judge and a lawyer with the special tribunal that will try Saddam Hussein and former members of his government outside their family home in Baghdad. The victims, Parwiz Muhammad Mahmoud al-Merani and Aryan Mahmoud al-Merani, were father and son. -March 3, 2005 - Two suicide car bombs kill five Iraqi police officers and wound seven outside the main gate of the Iraqi Interior Ministry in Baghdad. 178 -A car bomb kills an Iraqi civilian and wounds 16 others, including five police officers, outside the main police station in Baquba -March 4, 2005- Gunmen kill Sheik Saad Kamil, a Shiite imam associated with the rebel cleric Muqtada al-Sadr, as he arrived at the Kadmenain Mosque in Baghdad. -March 7, 2005 - In Baquba, three separate militant attacks on Iraqi Army checkpoints and convoys kill 12 Iraqi security officers and wound 25. -In Balad, a suicide bomber drives a truck laden with explosives into the house of an Army officer, kills five civilian bystanders, and wounds 24. -March 8, 2005 - Gunmen assassinate Major General Ghazi Mohammed, a senior Iraqi interior ministry official, in a drive-by shooting in Baghdad. -March 9, 2005 - Twenty bodies, including those of male and female civilians, are found in a remote valley near the Syrian border. -Gunmen fire on the car of Iraq?s interim planning minister, Mahdi al-Hafidh, in Baghdad. Two of the minister?s guards are killed and a third is seriously wounded. Hafidh is not injured. -A suicide bomber detonates a garbage truck full of explosives in a parking lot next to Baghdad?s Sadr Hotel, which houses a number of foreign contractors. One Iraqi policeman is killed; more than 40 are wounded, including 30 U.S. contractors. -March 10, 2005-A suicide bomber kills 53 people at a funeral inside a Shiite mosque in Mosul. -The White House announces President Bush has chosen Zalmay Khalilzad, the U.S. ambassador to Afghanistan, to become the new ambassador in Baghdad. 179 -March 12, 2005- Gunmen kill three Iraqi police officers as they drive to a colleague?s funeral in Mosul. -March 16, 2005 - Iraq?s first freely elected parliament in a half-century meets for the first time in the heavily fortified International Zone in central Baghdad. -A suicide car bomber kills three Iraqi soldiers and wounds 12 more at an army checkpoint in Baquba. -March 19, 2005 - A bomb kills three Iraqi police officers in Kirkuk in a funeral procession for a fellow officer shot by insurgents the day before. -March 20, 2005 - Iraqi insurgents ambush a U.S. military convoy, igniting a battle that leaves 24 of the attackers dead and 7 wounded. Six U.S. soldiers are wounded in the ambush, at Salman Pak -March 22, 2005- Militants ambush a convoy carrying Iraqi security force officials, including a senior police officer, Brigadier General Abu Al-Waled, in front of a mosque in Mosul. Police kill 17 militants and capture 14. None of the security troops are killed. -March 23, 2005-Iraqi and U.S. forces kill 85 insurgents during a raid on a guerrilla training camp 50 miles northwest of Baghdad. Seven Iraqi police officers are killed and six wounded. -March 24, 2005-A suicide car bomber kills 11 Iraqi special police commandos and wounds nine others in Ramadi. -Militants shoot five Iraqi women dead as they leave work at a U.S. military base in southern Baghdad. -March 30, 2005- Militants launch three attacks on Shiite pilgrims on their way to Karbala for a religious festival. Gunmen open fire on a minibus and wound eight near 180 Latifiya, 25 miles south of Baghdad. A second group of gunmen kill one and injure two in a car near the town of Mahaweel, 40 miles south of Baghdad. A suicide bomber on a bicycle rides into a police patrol protecting pilgrims and kills two officers, also near Mahaweel. -March 31, 2005-A suicide car bomber kills two Iraqi army soldiers and two civilians in a crowd of Shiites celebrating a religious holiday in the northern city of Tuz Khurmato -A suicide car bomber rams a U.S. military vehicle, killing one Iraqi civilian and wounding seven in Samarra -April -April 2, 2005 Seven suicide car bombers and 40 to 60 insurgents armed with an array of weapons attack the U.S.-controlled Abu Ghraib prison. Twenty U.S. soldiers and marines are wounded. -A car bomb kills four Iraqi policemen and one civilian at a police station in the town of Khan Bani Saad, 10 miles north of Baghdad. -April 3, 2005 The Iraqi National Assembly appoints Hajim M. al-Hassani, a prominent Sunni Arab and the minister of industry in the interim government, as speaker, and Hussain al- Shahristani, a nuclear physicist and leading Shiite Arab, and Arif Taifour, a Kurd, as his deputies. April 4, 2005- A bomb in an abandoned taxi kills one U.S. soldier and wounds four others in southern Baghdad. -A bomb in a parked car kills an Iraqi civilian and wounds two others in Amiriya. 181 -General Jalal Mohammed Salah, the commander of an Iraqi mechanized armored brigade, is kidnapped. -April 5, 2005-Iraq?s transitional assembly elects Kurdish leader Jalal Talabani to the largely symbolic post of Iraqi president. Shiite Arab Adel Abdel Mahdi, a member of the United Iraqi Alliance who served as interim finance minister, and Sunni Arab Sheikh Ghazi al-Yawar, who served as the interim president, are elected vice presidents. April 6, 2005 - Kurdish politician Jalal Talabani is Iraq's first freely elected president by parliament in 50 years. Shiite Ibrahim al-Jaafari is nominated as prime minister. April 7, 2005 - Iraq?s new president, Jalal Talabani, names the Shiite leader Ibrahim al- Jaafari prime minister. -April 11, 2005- Hundreds of Iraqi troops and commandos backed by U.S. soldiers capture 65 suspected insurgents in central and southern Baghdad in one of the largest raids in the capital since the fall of Saddam Hussein. -Kidnappers abduct a U.S. contractor, Jeffrey Ake, in Baghdad. -A pickup truck packed with explosives kills at least 3 people and wounds 20 near a U.S. military convoy in Samarra. -April 12, 2005 - A suicide car bomber kills four Iraqis and wounds seven near a U.S. military convoy in Mosul. -The Iraqi government announces the capture at a farm northeast of Baghdad of Fadhil Ibrahim Mahmoud al-Mashadani, a former official in Saddam Hussein?s government accused of coordinating and financing a series of insurgent attacks. 182 -April 14, 2005-Two suicide car bombs detonate in quick succession near an Interior Ministry building in central Baghdad and kill 15 Iraqis. More than 30 are injured. A third blast in the same area several hours later kills one person. -In Latifiya, 25 miles south of Baghdad, gunmen assassinate the mayor. -April 15, 2005 - A suicide bomber steers his car into an Iraqi police patrol and kills four officers in Mahaweel, 40 miles south of Baghdad. Another three bomb attacks in Baghdad wound nine people. -April 16, 2005- Three U.S. soldiers die in a mortar attack in Ramadi -A bomb kills 10 Iraqi police officers and three others in a crowded restaurant in Baquba. -April 17, 2005-Insurgents kill Brigadier General Younis Mohammad Sulaiman, the Mosul police public information officer, as he drives to work. -April 18, 2005 - Ten insurgents dressed in Iraqi military uniforms kill Iraqi Lieutenant General Adnan Qaragholi, an adviser to interim Defense Minister Hazem Shaalan, and his son at their home in southern Baghdad. -April 19, 2005 -A suicide car bomber kills two members of the Iraqi army and two recruits and wounds 38 others near the Adhamiya Palace in northern Baghdad. -April 20, 2005-A suicide car bomber kills one and wounds four near the headquarters of Iraqi interim Prime Minister Ayad Allawi?s political party in Baghdad. -Iraqi President Jalal Talabani says the bodies of more than 50 people he believes to be Shiites killed by Sunni extremists were recovered from the Tigris River. -Nineteen Iraqis are lined up against a wall and shot to death in a soccer stadium in the town of Haditha, about 140 miles northwest of Baghdad. 183 -One Iraqi policeman is killed and two seriously wounded when their patrol is hit by a roadside bomb in Mowailha, south of Baghdad. -April 21, 2005 - Eleven people, including six Americans from the security firm Blackwater USA working for the U.S. military, are killed when a commercial helicopter crashes near Baghdad. The Islamic Army of Iraq, a jihadist group, says it brought down the helicopter and executed a survivor. -April 22, 2005 - A car bomb kills nine people and wounds 24 outside a Shiite mosque in Baghdad. -April 23, 2005-U.S. soldiers capture six men suspected of shooting down a civilian helicopter April 21 in Baghdad. -A car bomber wounds three U.S. servicemen and seven Iraqi soldiers on a street that adjoins the road to Baghdad International Airport. -Gunmen assassinate the prison director for Nineveh Province, Colonel Khalid Najim Abdallah, in Mosul. -April 24, 2005 - Two bombs kill 15 Iraqis and wound nearly 60 near the Shiite Ahl al- Beit mosque in Baghdad. -April 27, 2005 - Insurgents assassinate Lamia Abed Khadouri, a newly elected member of the National Assembly, at her home in Baghdad. -April 28, 2005-Iraq?s National Assembly votes overwhelmingly to approve a cabinet, creating the first fully and freely elected government in Iraqi history. The assembly?s political factions continue negotiations on candidates for five ministries--defense, oil, electricity, industry, and human rights. 184 -April 29, 2005-A coordinated series of car bomb attacks targeting Iraqi security forces kills 27 Iraqis, seven of them civilians, and injures more than 100 in and around Baghdad -April 30, 2005-Syria announces it is restoring diplomatic relations with Iraq after a break of more than two decades. -May- - Surge in car bombings, bomb explosions and shootings: Iraqi ministries put the civilian death toll for May at 672, up from 364 in April. -May 3, 2005-Iraq?s first fully and freely elected government takes power as new cabinet members are sworn into office. Several key cabinet positions remain unfilled. -Insurgents kill two Iraqi civilians and one Iraqi soldier at a checkpoint in Ramadi, about 50 miles west of Baghdad. Twelve insurgents are also killed. -May 4, 2005-A suicide bomber kills 60 prospective policemen and wounds more than 200 others inside a police recruitment office in the Kurdish provincial capital of Erbil. -May 6, 2005-A suicide car bomber kills 58 Iraqi civilians and wounds 44 more in Suwayra, about 30 miles south of Baghdad. -A suicide car bomber drives into a police bus and kills 10 Iraqi policemen in Tikrit. -Iraqi police unearth the bodies of 14 Iraqi men in a garbage dump in Baghdad. -May 7, 2005-The Iraqi National Assembly approves new defense, human rights, oil, electricity, and industry ministers. -Suicide bombers attack a U.S. private security convoy and kill two U.S. security contractors and 22 Iraqis. -May 8, 2005-Gunmen kill a senior transportation ministry official, Yasser Khudair Almaaini, and his driver in Baghdad. 185 -Attacks by insurgents kill six U.S. servicemen in separate attacks in the towns of Qaim and Obeidi, both near the Syrian border, about 185 miles west of Baghdad. -May 9, 2005-Hashim al-Shibli, who had been named human rights minister May 7, announces he will not accept the post. -The U.S. military announces it has killed more than 100 insurgents in an offensive in Anbar province in western Iraq. -Iraqi police find the bodies of a senior Iraqi border policeman and five of his relatives in Markab al-Tair, a village near the Syrian border. -May 10, 2005-Iraq?s transitional national assembly names a 55-member committee to draft the country?s permanent constitution. -May 11, 2005-A suicide bomber kills 38 people and wounds more than 80 in Tikrit. -A suicide bomber kills 32 and wounds more than 40 outside a national guard recruitment center in Hawija, about 150 miles north of Baghdad. -May 12, 2005-Four car bombs in Baghdad and one in Kirkuk kill 21 people and wound more than 70. -May 13, 2005-A car bomb kills five Iraqis in Baiji, about 130 miles north of Baghdad. -May 15, 2005-Police find the corpses of 12 Iraqi men in northeastern Baghdad, 13 bodies in eastern Baghdad, and 11 more near Iskandariya, 25 miles south of Baghdad -Militants armed with bombs and mortars kill four Iraqi soldiers and wound four more in an attack on a fire station in Khan Bani Saad, about 15 miles northeast of Baghdad. -Two suicide bombers kill five Iraqis in Baquba, about 40 miles northeast of Baghdad. -May 17, 2005-Insurgents kill 21 Iraqis: six truck drivers delivering supplies to U.S. forces north of Baghdad, three policemen in Mosul; in Baghdad, two government 186 officials, a former member of the Baath Party and his three grown sons, two Iraqi soldiers, a Shiite cleric and his driver, and two Sunni clerics -May 19, 2005-Insurgents kill eight Iraqis in Mosul during an attack on the house of Fawwaz al-Jarba, a Sunni Muslim elected to parliament as a member of a Shiite- dominated coalition. Al-Jarba survives the attack. -In Baghdad, gunmen kill an Oil Ministry official outside his home, and a car bomb kills two people and wounds five near a Shiite mosque. A roadside bomb kills two policemen in Baquba. Another roadside bomb kills a policeman and wounds three more in Iskandariya, 25 miles south of Baghdad. Gunmen kill a local police chief in Anbar Province. A police officer and his father are shot dead in Samarra, 60 miles north of Baghdad. -May 23, 2005-Two suicide car bombs kill 15 Iraqis outside the home of Hasan Bagdash, a Shiite Turkmen politician, in Tal Afar, about 90 miles east of the Syrian border. Bagdash survived the attack. -A suicide truck bomb kills five people and wounds 18 outside the office of the mayor of Tuz Khurmatu. -A car bomb kills 11 Iraqis and wounds more than 110 outside a Baghdad restaurant. -Gunmen kill Wael Rubaie, an official of the Ministry of State for National Security, and his driver, in Baghdad. -A suicide car bomber kills 21 people and wounds 23 outside a Shiite mosque in Mahmoudiya, 25 miles south of Baghdad. -Two suicide bombers, one in a car and the other on foot, kill four Iraqis and wound four U.S. soldiers outside a U.S. base in Samarra. 187 -May 25, 2005-Around 1,000 U.S. and Iraqi troops search homes and seize suspected insurgents in Haditha, about 140 miles northwest of Baghdad. Ten insurgents are killed; two U.S. marines are wounded. -In Mosul, gunmen kill Mikhlif Khalaf, police chief of the northern town of Sharqat. -May 26, 2005 A reconnaissance helicopter is shot down 35 miles north of Baghdad, killing two U.S. soldiers. -Insurgents kill Fakri Abed al-Amri, a member of the Shiite Muslim Dawa Party headed by Prime Minister Ibrahim al-Jaafari, at al-Amri?s brother?s home in Baghdad. -Militants gun down Professor Moussa Salum, a deputy dean at Baghdad?s Mustansiriya University, and three of his bodyguards, in Baghdad. -May 27, 2005-A suicide bomber kills seven people, including three police officers, in Tikrit. -May 29, 2005 Insurgents attack two police stations, an Iraqi army barracks, and a checkpoint in western Baghdad, killing three civilians at Baghdad?s police Major Crime Unit. -Insurgents retaliate against Iraqi forces during a routine sweep of vehicles at checkpoints, killing 27 people, including a British soldier. -Gunmen in Kirkuk kill Major General Ahmed al-Barazanchi, director of internal affairs of Kirkuk province and a former police chief. -Police in Rawah find the corpse of Raja Nawaf Farhan al-Mahalawi, governor of Anbar province. He was abducted May 10 near Qaim, a town near the Syrian border. 188 -May 30, 2005-A single-engine plane crashes near Jalula, 80 miles northeast of Baghdad, killing four Americans and an Iraqi pilot. The aircraft was used by the Iraqi air force for surveillance and personnel transport. -Two suicide bombers in Hillah, 60 miles south of Baghdad, kill 30 people, part of a crowd of 500 police officers outside the mayor?s office protesting a government decision to disband a police special forces unit. -A roadside bomb kills one civilian and injures two others in Adhaim, 30 miles north of Baquba. -Gunmen kill Sunni Muslim tribal leader, Sheik Bishar Abdul Karim al-Dobradani, outside his home in Mosul. -May 31, 2005-Four Italian troops die in a helicopter crash eight miles south of Nasiriyah. -A suicide bomber kills two Iraqi soldiers at an army checkpoint near Buhriz, about 35 miles north of Baghdad. -June- -June 2, 2005-A suicide car bomb targeting a Northern Oil Company facility in Kirkuk kills a four-year-old and wounds 11 others. -A suicide car bomb in Baquba kills four, including a deputy provincial council leader. -A firefight between insurgents and police in Baiji, 130 miles north of Baghdad, kills a Turkish truck driver. Two members of the security forces die in the shootout. -In Siniyah, 150 miles north of Baghdad, a mortar attack kills an Iraqi soldier. -June 3, 2005-A suicide bomb in a remote village north of Baghdad kills 10 Iraqis. -Gunmen kill an Iraqi contractor in charge of renovating a mosque in western Samarra. 189 -A suicide bomber kills 10 Iraqis and wounds 10 more at a home in Saud, a remote village near Balad, about 50 miles north of Baghdad. -In Tuz Khormato, a suicide bomber kills 12 people who had stopped for lunch at the popular highway stop 55 miles south of Kirkuk. -A suicide bomber, trying to attack a convoy of civilian contract workers in Kirkuk, kills a young boy and three other Iraqi bystanders. -Gunmen kill Shiite cleric Ali Abdul Hussein outside his home in Basra. -June 7, 2005-Three explosions in and around Hawija, 40 miles southwest of Kirkuk, kill 34 people; the deadliest kills 10 people at a checkpoint in Dibis, on the outskirts of Hawija. -An attack on a U.S. military base in Tikrit kills two U.S. soldiers. -June 8, 2005: Iraqi President Jalal Talabani visits for talks; meets representatives of Iraqi community in Britain - Massoud Barzani is sworn in as regional president of Iraqi Kurdistan. -A roadside bomb kills a U.S. soldier near Adwar, 10 miles south of Tikrit. -June 9, 2005-Five marines die during combat operations near the town of Haqlaniyah, 90 miles north of Baghdad. -June 10, 2005-A roadside bomb kills two marines near the volatile Anbar province town of Saqlawiya, west of Falluja. -The bodies of 21 Iraqis are found scattered in various locations in an area close to the Syrian border. 190 -June 11, 2005=A former member of the Wolf Brigade, an Iraqi militia composed of roughly 2,000 fighters, kills three people when he blows himself up at the heavily guarded headquarters of an Iraqi police unit in Baghdad. -Gunmen open fire on a minibus in Diyara, killing 11 Iraqi construction workers. -A four-hour bombardment near the Anbar province city of Qaim kills 40 insurgents, according to the marines. -An explosion in a cemetery in Najaf kills two Iraqis. -June 12, 2005-In the northern city of Tuz Khormato, a suicide car bomber, targeting a U.S. convoy near the headquarters of the Patriotic Union of Kurdistan, kills two Iraqis. -In western Baghdad, insurgents attack two police stations, an Iraq army barracks, and a checkpoint; three civilians are killed. -A suicide attacker in a car filled with explosives attacks the Iraqi Oil Ministry in Baghdad, killing two security guards at the heavily fortified complex. -June 14, 2005-A suicide bomber kills 23 people outside a bank in Kirkuk. -Two marines die during combat operations near Falluja. -U.S. and Iraqi forces kill five civilians after they ignored warning signals to stop at a Ramadi checkpoint minutes after a suicide bomber blows himself up, killing an Iraqi soldier. -U.S. forces arrest Mohammed Khalaf, the leader of the Mosul branch of al Qaeda in Iraq. Khalaf is considered a top aide to Abu Musab al-Zarqawi. -June 15, 2005=A militant strapped with more than 100 pounds of explosives kills 26 Iraqi soldiers in a crowded mess hall at an army base in Khalis, 45 miles north of Baghdad. 191 -After seven weeks of captivity, Iraqi and U.S. troops free Australian hostage Douglas Wood. -Insurgents kidnap and kill two senior officials in Kirkuk?s anti-terrorist squad, their driver, and another passenger. -A suicide car bomb kills five Iraqi soldiers at a checkpoint in Kanan, 30 miles north of Baghdad. -Insurgents kill seven people in the northern town of Tal Afar. -June 16, 2005-Six masked gunmen kill a judge and his bodyguard in an eastern Mosul neighborhood. -After weeks of deadlock, Sunni Arabs accept a compromise offer to increase their representation on the Shiite-led parliament committee set up to draft a constitution. -June 17, 2005-A suicide bomber slams his car into a fuel tanker in Baghdad?s eastern suburbs, killing two people and injuring six others. -For the first time, Saddam Hussein?s lawyer is permitted to attend an interrogation session with his client. -June 19, 2005-A suicide bomber detonates his charges in a popular Baghdad restaurant during lunchtime, killing 23 people. The cafe is located just 400 yards from the Green Zone. -June 20, 2005 A suicide bomber kills 15 traffic police officers and wounds 100 people outside the unit?s headquarters in the northern Kurdish city of Irbil. -A suicide bomber kills five soldiers at an army checkpoint in Kirkuk. -Insurgents kill eight officers and an eight-month old baby at a Baghdad police station. 192 -A suicide bomber kills a Kurdish town?s security director and three of his bodyguards while they are traveling in a convoy. -June 22, 2005-Four car bombs, which explode within minutes of each other, kill 23 people and injure 56 others in and around western Baghdad?s Shula neighborhood. Gunmen kill a prominent Sunni law professor, Jassim al-Issawi, and his son in the same neighborhood. -June 23, 2005-Simultaneous car bombs in Karradah, central Baghdad?s shopping district, kill 15 civilians and injure 28 others. -An ambush of a U.S. military convoy in Fallujah kills four Marines-including a female Marine-leaves two other troops missing and presumed dead, and injures 13 other troops. -U.S. and Iraqi troops kill seven insurgents in a home in western Baghdad?s Jamiaa neighborhood. -June 24, 2005-U.S. troops shoot and kill an Iraqi reporter working for a Knight-Ridder, an American news organization, after he did not respond to a shouted signal from a military convoy in Baghdad. -A suicide bomber trailed by five cars loaded with armed insurgents attacks the home of an Iraqi special forces police officer in Samarra, killing nine people. -June 26, 2005-Insurgents kill three people at a barbershop in Baghdad?s Jadida district. Barbershops are the target of some Muslim extremists who believe men?s beards should remain long, authorities said. -A suicide bomber kills five police officers on guard at a teaching hospital in Mosul. -June 27, 2005-A car bomb explodes between a movie house and a mosque in eastern Baghdad, killing seven people. 193 -June 28, 2005-A suicide bomber kills a Shiite member of parliament-87-year-old Dhari Ali al-Fayadh-his son, and two bodyguards in Rashidiya, 20 miles northeast of Baghdad. -Car bombs kill two U.S. soldiers; one in Balad and the other in Tikrit. -A car bomb in Baqouba, 35 miles northeast of Baghdad, kills five people and injures nine others. -Gunmen kill a local city council member, Sabah Hussein, as he drives to work in Mansur. -July- -July 1, 2005-Gunmen kill Kamal Ezz al-Deen al-Ghuraifi, an aide to leading Shiite cleric Grand Ayatollah Ali al-Sistani, and two of his bodyguards in a drive-by shooting outside a Baghdad mosque. -In Baghdad, a suicide bomber detonates a car bomb outside the party offices of Prime Minister Ibrahim al-Jaafari, killing one guard and wounding another. -A mortar attack at a power station causes a water plant to shut down, leaving millions of Baghdad residents with no running water in 100-degree temperatures. -July 2, 2005-In Hillah, 60 miles south of Baghdad, an attacker detonates an explosive belt at a police checkpoint, killing six police officers. Ten minutes later, a second suicide bomber wounds 26 people in the crowd of police and civilians that gathered at the scene. -The head of the Egyptian diplomatic mission to Iraq, Ihab al-Sherif, is pistol-whipped and forced into the trunk of a car by kidnappers when he stops to buy a newspaper in western Baghdad. -July 5, 2005-In separate ambushes, gunmen attack convoys carrying the senior diplomats from Bahrain and Pakistan in Iraq as they drive through the streets of Baghdad. 194 Hassan Malallah al-Ansari, the Bahraini diplomat, is treated at a hospital for a gunshot wound to the hand. The Pakistani ambassador, Muhammad Yunis Khan, is unharmed, though the attack prompts Pakistan to announce the diplomat?s withdrawal from Iraq. -July 7, 2005-Al Qaeda in Iraq announces that it has executed the Egyptian diplomat Ihab al-Sherif, four days after his abduction. The Egyptian government confirms his death. -July 10, 2005 An attacker detonates an explosive vest outside a Baghdad army recruitment center, killing 25 and wounding 47. -A suicide bomber kills four civilians by detonating a car bomb on a highway in Kirkuk, 180 miles north of Baghdad. A bomb in a parked car, apparently targeting security forces responding to the first blast, is discovered and safely detonated by U.S. soldiers. -Eight members of a Shiite family-a mother and seven children ranging in age from 2 to 19 years-old-are shot to death while asleep in their Baghdad home in the early morning. -July 11, 2005=Insurgents firing mortars and automatic weapons assault a police checkpoint in Khalis, 45 minutes north of Baghdad, killing eight Iraqi soldiers. Ninety minutes later, a nearby car bomb hits an Iraqi army patrol, killing two more Iraqi troops. -July 12, 2005 Four members of the local International Organization for Human Rights, including the head of the organization, are killed when gunmen storm their Baghdad office. -July 13, 2005-Twenty-seven people, including 18 children and one soldier, are killed and at least 70 are injured when a suicide car bomber speeds into U.S. troops handing out candy to children in Baghdad. 195 -July 15, 2005-A dozen explosions aimed at U.S. and Iraqi forces sweep across Baghdad throughout the day. Altogether the attacks claim 33 lives and leave 111 wounded. One of the bombs, detonated near President Jalal Talabani?s residence, kills four guards. -July 16, 2005-In one of the deadliest suicide attacks since the start of the U.S. occupation, a man wrapped in explosives detonates himself under an oil tanker at a gas station in a crowded area of Musayyib, about 40 miles south of Baghdad. At least 90 people are killed and 150 are injured. -The Pentagon announces charges against 11 U.S. soldiers suspected of assaulting seven Iraqis they captured during an attack on a power plant in Baghdad. -July 17, 2005-Iraq?s Special Tribunal files its first criminal case against Saddam Hussein and three aides for their roles in the 1982 massacre of dozens of Shiite villagers in the town of Dujail, 50 miles north of Baghdad. If convicted, they could face the death penalty -A bomb explodes in Baghdad while police are inspecting two bodies left on the road, apparently as a trap. Two officers and one civilian die in the blast. An hour later, a suicide car bomber targets a police convoy, killing three officers and four civilian. -July 18, 2005-A series of small ambushes and shootings throughout Baghdad and central Iraq claim the lives of 24 police, soldiers, and government workers. - Death toll rises to 100 in suicide blast in Iraq [Washington Post, 7/18/05] -July 19, 2005-Two Sunni members of Iraq?s constitutional committee are assassinated along with a bodyguard in a drive-by shooting just outside the walls of Baghdad?s Green Zone. -A British research group, Iraq Body Count, releases a report claiming that 25,000 civilians have died in violence in Iraq since the start of the U.S. invasion. 196 -July 20, 2005-Sunni members of the constitutional committee, in response to the assassination of two Sunni delegates, suspend their participation in the committee. -July 21, 2005-Two Algerian diplomats are abducted in Baghdad when gunmen stop their car and drag them from the vehicle. -July 24, 2005-A suicide attacker drives a truck filled with explosives into sand barriers outside a Baghdad police station. At least 39 people die in the massive explosion, which leaves a gaping crater and dozens of burning shops and vehicles in its wake. -July 25, 2005-The Ministry of Interior police command station, housed in one of Saddam?s former palaces, is targeted in a suicide attack. Two people are killed and four are injured. -July 26, 2005-Gunmen in cars assault a pair of busses carrying Industry Ministry employees home from work. Sixteen workers die and 27 are injured. - Sunni delegates resume participation in the constitutional committee, ending a week- long boycott. -July 27, 2005-Al Qaeda in Iraq announces it has killed the two Algerian diplomats who were kidnapped the previous week. -Seven Iraqi soldiers defending a water plant in Tarmiyah, 30 miles north of Baghdad, are killed in an assault by a score of well-armed insurgents. - August ? Draft constitution is endorsed by Shia and Kurdish negotiators, but not by Sunni representatives. -August ? 197 -August 1, 2005-Six U.S. Marines are killed by small-arms fire when their unit is attacked by insurgents in the town of Haditha, 140 miles northwest of Baghdad. In the town of Hit, fifty miles southeast of Haditha, a marine from the same unit is killed by a car bomb. -August 3, 2005-In the deadliest roadside bombing against U.S. troops to date, fourteen marines and their translator are killed when a massive explosion destroys their lightly armored vehicle outside the town of Haditha, 140 miles northwest of Baghdad. -Steven Vincent, an American freelance journalist, and his translator are found shot to death in Basra. They had been abducted at gunpoint the previous evening. -August 5, 2005-Insurgents mount a coordinated assault on several Iraqi army posts in Baghdad. U.S. forces help to repel the onslaught, but not before one Iraqi soldier and six attackers are killed and a dozen insurgents arrested. -August 10, 2005-Gunmen in Baghdad abduct Brig. Gen. Khudayer Abbas, a senior official in the Iraqi Interior Ministry, which oversees the nation?s police and paramilitary operations. -August 12, 2005-An explosion outside a mosque kills four people, including three children, and injures nineteen in the town of Nasaf, seventy miles west of Baghdad. -Police kill three insurgents as they try to break into a polling station to be used in the October 15 constitutional referendum. -August 14, 2005-An Iraqi commando unit discovers thirty bodies in a two-week-old grave in Baghdad. The discovery follows a raid in which thirteen insurgents are arrested, some of whom confess to dumping bodies in the grave. 198 -August 15, 2005 - Date by which a constitution is to be drafted by the transition government. -The deadline to produce a draft of the new constitution expires with Iraq?s Constitutional Committee still unable to reach an agreement. The committee grants itself an additional week to produce the document. -August 17, 2005-In a coordinated attack, two bombs explode at a Baghdad bus station, one detonated by a suicide bomber. Minutes later, another suicide bomb targets the nearby hospital as the wounded begin to arrive. The blasts kill forty-three and injure at least sixty-eight. - A former Iraqi Army Colonel and his son die when they are ambushed by gunmen in Baghdad. Another of the colonel?s sons is kidnapped in the attack. -August 18, 2005-Four U.S. soldiers are killed by a roadside bomb in Samarra -An Iraqi judge and his driver are killed by gunmen as they travel through Baghdad. -August 19, 2005-Three Sunni election workers, kidnapped a day earlier as they hung voting posters, are killed in front of a crowded mosque in Mosul. -August 22, 2005-Minutes before the already-delayed midnight deadline, Iraq?s constitutional committee submits a draft constitution. -August 23, 2005 In an online statement, al-Qaeda in Iraqclaims responsibility for a rocket attack against a U.S. war ship in the Jordanian port of Aqaba, which occurred August 19. -A suicide bomber in Baquba, 35 miles northeast of Baghdad, kills one U.S. soldier, one U.S. contactor, an Iraqi policeman, and four Iraqi civilians while wounding 20. 199 -August 24, 2005-The Iraqi Special Tribunal confirms reports that Saddam Hussein fired all but one lawyer in his 1,500-member defense team. -Dozens of insurgents launch an assault against Iraqi police patrols in Baghdad. At least thirteen people die and forty-three are left wounded. -Cleric Muqtada al-Sadr?s followers try to reopen his office in Najaf, which was closed last year, but are blocked by a rival Shiite militia. Four people are left dead, twenty are injured, and Sadr?s office is set ablaze in the fighting that ensues. -In a failed assassination attempt, gunmen open fire on Deputy Justice Minister Awshoo Ibrahim?s convoy in Baghdad. Four bodyguards are killed and five are wounded in the second attempt on Ibrahim?s life in as many days. -August 25, 2005-Bodies of thirty-six men are discovered in southeastern Baghdad, alongside a road leading to the Iranian border. -August 27, 2005-The U.S. military releases nearly 1,000 prisoners from Abu Ghraib prison. The prisoners, who had not committed serious offenses and had admitted to their crimes, renounce violence and pledge to be good citizens. -August 28, 2005-After a week of negotiations, Iraq?s Constitutional Committee submits a final draft constitution to be voted on in a national referendum October 15. Sunni negotiators, who criticized the document throughout the drafting process, reject the draft. -August 30, 2005 Fighting between pro- and anti-government tribes breaks out near the Syrian border in the city of Qaim, 200 miles west of Baghdad, leaving thirty-five members of the Bumahl and Karabila tribes dead. 200 -U.S. air strikes destroy three houses in the towns of Husaybah and Karabilah, some 200 miles west of the Baghdad. Abu Islam, a ?known terrorist,? and several of his associates are killed in the attacks, according to a U.S. statement. -August 31, 2006 - Up to 1,000 Shiite pilgrims die in a stampede on a bridge in Baghdad. Thousands crossing the bridge to visit a Shiite shrine panic amid speculation that suicide bombers are among the crowd. The incident is reported to be the worst single disaster since the start of U.S. occupation in Iraq in 2003 September ? -September 3, 2005-Gunmen attack a checkpoint in Baquba, thirty-five miles northeast of Baghdad, killing six Iraqi policemen. A second gunfight kills an additional two policemen -September 4, 2005-A series of car bombs aimed at U.S. and Iraqi targets in Hit, eighty- five miles west of Baghdad, kill eight civilians, an Iraqi soldier, and three suicide bombers. U.S. Marines blame al-Qaeda in Iraq for the attacks. -September 6, 2005-U.S. forces leave their base in Najaf, 100 miles south of Baghdad, handing over full control of the city to Iraqi security forces. -U.S. and Iraqi forces raid a rebel stronghold in Karabilah, 185 miles west of Baghdad. -September 7, 2005-U.S. and Iraqi forces arrest 200 suspected insurgents?including at least 150 foreign fighters?in Tal Afar, 250 miles north of Baghdad. -September 8, 2005-Police discover the bodies of fourteen people, dressed in civilian clothing and shot to death, near the town of Mahmoudiya, twenty miles south of Baghdad. 201 -September 10, 2005-Some 5,000 Iraqi troops, backed up by 3,500 U.S. military personnel, descend upon the insurgent stronghold of Tal Afar, 250 miles north of Baghdad. Some 150 insurgents are killed and forty-eight are captured. Five Iraqi soldiers are killed in the fighting. -September 11, 2005-The invasion of Tal Afar concludes after three days of fighting. Though a total of 200 insurgents are killed and over 400 are captured, many escape into the countryside through a network of tunnels. The joint offensive uncovers eighteen weapons caches and a large bomb-making facility. -September 13, 2005-U.S. ground and air forces mount an attack on Haditha, an insurgent stronghold 140 miles northwest of Baghdad. Four militants are killed and one is captured. -Two Sunni clerics are killed by gunmen in Baquba, thirty-five miles northeast of Baghdad. -September 14, 2005-Twelve coordinated explosions tear through Baghdad killing at least 167 people and wounding nearly 600. At least 112 day laborers gathered in search of work in a predominantly Shiite neighborhood are killed. Al Qaeda in Iraq claims responsibility for the attacks, which are in retaliation for the Tal Afar military offensive earlier in the week. -In a predawn raid, gunmen sweep through Taji, a Sunni village ten miles north of the capital, rounding up sixteen Iraqi contractors working for the U.S. military and one policeman. The victims are later found handcuffed, blindfolded, and shot to death. -September 16, 2005-A suicide car bomber attacks worshippers leaving a Shiite shrine in Tuz Khormato, 130 miles north of Baghdad, killing a dozen people and wounding 202 twenty-three. A man wearing an explosive belt is detained while en route to a second mosque in the city. -September 18, 2005-Faris Nasir Hussein, a Kurdish lawmaker, is killed along with his brother and driver when their vehicle is ambushed outside of Dujail, fifty miles north of the capital. Haidar Shanoun, another Kurdish member of parliament, is wounded in the attack. -Police discover four bodies in east Baghdad and pull an additional twenty out of the Tigris River near Balad, fifty miles north of Baghdad. The dead are part of retaliatory attacks between Sunni and Shiite militants. -September 19, 2005-British armored vehicles crash through the walls of a Basra prison, freeing two British soldiers but allowing 150 Iraqi prisoners to escape. The two Britons were arrested earlier in the day for shooting a pair of Iraqi policemen. -The Iraqi Central Criminal Court in Baghdad sentences Ayman Sabawi, Saddam Hussein?s nephew, to life in prison for funding and making bombs for the violent insurgency. It is the first verdict against a family member of the former ruler. -The body of Fakher Haider, a Basra-based Iraqi journalist and photographer for the New York Times, is found beaten and shot to death. He is the second journalist to be killed in Basra in as many months. -September 20, 2005-Insurgents mount an ambush on fifteen lost vehicles carrying supplies for U.S. forces in the Sunni town of Dhuluiyah, killing four security contractors and wounds two additional contractors, as well as a pair of U.S. soldiers. In response, U.S. war planes mount a series of air strikes on the town, though there are no reports of casualties. 203 -The Mosul bureau chief of As-Safeer an independent Iraqi national newspaper is gunned down in the street two days after one of his reporters was killed. -September 23, 2005 A passenger aboard a Baghdad minibus detonates his explosive belt as the vehicle approaches a crowded terminal. At least five people are killed in the blast and eight are wounded. -September 25, 2005-U.S. and Iraqi military personnel, operating on a tip, raid a Baghdad high-rise where they find Abdullah Abu Azzam, a top deputy in the al-Qaeda in Iraq terrorist group. When Azzam refuses to surrender, he is killed. -A convoy of U.S. and Iraqi military personnel takes on small-arms fire as it enters Baghdad?s Sadr city to arrest several members of the al-Mahdi Army, a militia based there. At least five Shiite gunmen are killed in the ensuing ninety-minute shootout. -A suicide car bomber targets an Interior Ministry convoy in Baghdad, killing seven members of the elite commando unit on board along with two civilians. Nineteen people are injured in the attack. -Baghdad police discover the bodies of seven people. Six of the bodies?including that of a policeman?were bound and shot; the seventh body?a young woman?was strangled and tortured. -September 26, 2005-As a Baghdad school is letting out, nine gunmen drag five Shiite teachers and their driver from their van, past a crowd of students, and into an empty classroom, where they line them up against a wall and execute them. 204 -U.S. and Iraqi officials, in an attempt to win support in the coming constitutional referendum, release some 500 detainees from the notorious Abu Ghraib prison near Baghdad to mark the coming Islamic holy month of Ramadan. -Stray mortar shells land in a residential section of Samarra, sixty miles north of Baghdad. One shell hits a house, killing seven members of the family inside. -September 27, 2005-A suicide bomber sets off an explosion amidst a group of Iraqis applying for jobs in Iraq?s Quick Reaction Police Force in Baqouba, thirty miles north of Baghdad. Nine of the applicants are killed and twenty-one are injur -Police in Kut, 100 miles southeast of Baghdad, discover the bodies of twenty-two men in civilian clothes?most of them bound and blindfolded?shot in the head. -September 28, 2005-=Iraq ?s first-known female suicide bomber disguises herself as a man and attacks a line of army recruits in Tal Afar, near the Syrian border, killing six and wounding thirty-five. -September 30, 2005 A car bomb detonates in a food market in Hillah, a mostly Shiite town sixty miles south of Baghdad, killing five people and wounding thirty. -October- -October 1, 2005-U.S. forces launch their anti-insurgent offensive, Operation Iron Fist, in the village of Sadah, eighty miles northwest of Baghdad. Some 1,000 Marines and other troops assault the small village near the border with Syria in an attempt to drive out al- Qaeda linked fighters. 205 -October 2, 2005-Operation Iron Fist continues for a second day as U.S. forces expand their offensive into Karabilah, a town 200 miles west of Baghdad, killing twenty-eight insurgents. -October 3, 2005-A seven-car convoy carrying the Iraqi oil minister is hit by a roadside bomb in Baghdad. The minister survives; three of his escorts do not. -The UN begins circulating millions of copies of Iraq?s draft constitution in anticipation of the October 15 referendum. -October 4, 2005-2,500 U.S. troops accompany Iraqi forces as they sweep into three cities along the Euphrates River in an anti-insurgent offensive dubbed Operation River Gate. -A suicide car bomb strikes at the gates of Baghdad?s heavily fortified Green Zone, killing two Iraqi policemen and wounding one. -October 5, 2005-Prayers to mark the beginning of the Islamic month of Ramadan at a Shiite mosque in Hillah, sixty miles south of Baghdad, are cut short by an explosion killing at least twenty-five people and wounding eighty-seven others. -Faced with UN criticism and Sunni anger, Iraq?s National Assembly backtracks on a last-ditch attempt to alter the rules governing the October 15 constitutional referendum that would have made it more difficult for the draft constitution to be rejected. -October 6, 2005-A combination of suicide bombers and roadside bombs kill twenty-one people in various attacks across Iraq. In Baghdad, a man detonates an explosives-laden belt while traveling in a packed minibus, killing nine and wounding nine. Elsewhere in the capital, a convoy ferrying private security contractors is targeted by a suicide car bomber, who kills three bystanders. 206 -October 7, 2005-Twenty-nine insurgents are killed in Karabilah, 185 miles west of Baghdad, during the continuing Operation Iron Fist offensive, designed to crush al-Qaeda insurgents near the Syrian border. -October 12, 2005 - murderous attack in Baghdad aimed against a delegation of the Arab League. -October 15, 2005-Iraq holds a referendum on its Constitution to create an Islamic federal democracy, which is adopted despite a "No" campaign among Sunni Muslims. More than 61% of Iraqis vote in a historic referendum on a new constitution, which is passed by an overwhelming majority. -October 19,2005 - The trial of former Iraqi dictator Saddam Hussein over crimes committed during his reign begins in Baghdad, before being adjourned until November 28. Hussein faces at least 12 charges on war crimes allegedly committed during his reign. He defiantly pleads not guilty. -October 19. 2005 - Hussein co-defendant's lawyer killed in Baghdad. -OCTOBER 25, 2005- Italy contributes to Iraq Police training reform in wake of events at Basrah. -Draft constitution passes The referendum has passed, according to final provisional results issued by Iraqi election officials. -OCTOBER 26, 2005-UN: Secretary-General commends Iraqis on approval of constitution, hopes historic event will mark milestone on Iraq's path to democracy -November- 207 - NOVEMBER 8, 2005: Powerful new evidence emerged that the United States dropped massive quantities of white phosphorus on the Iraqi city of Fallujah during the attack on the city in November 2004 - Hussein co-defendant's lawyer killed. -November 10, 2005- Condeleeza Rice on surprise visit. -November 11, 2005 - Key fugitive Saddam aide dead -November 15, 2005 - Detainees 'found starving'The government begun an investigation into the abuse of more than 170 detainees held by Iraqi forces. -November 16, 2005 - U.S. admits using white phosphorous. Pentagon acknowledged that U.S. troops used white phosphorous as a weapon against insurgent strongholds -November 18, 2005- Suicide bombers kill dozens in 2 cities -November 19, 2005-Suicide bomber kills 35 at funeral November 28, 2005- Hussein trial adjourned amid rancor -November 29, 2005- Video airs of purported hostages. Al-Jazeera broadcast a video issued by an unknown group showing what it claimed were 4 hostages. Watching developments following Al-Jazeera's screening of video from previously unknown Swords of Righteousness group showing four kidnapped foreigners; foreigners - two Canadians, one Briton, one American - described on video as "spies for the occupation forces"; UK citizen Norman Kember among those kidnapped -2005 November - Suicide bombers target mosques in Khanaqin, killing at least 74 people. -December- -December 2, 2005- 10 Marines killed near Falluja 208 -December 3, 2005-Attackers wielding small arms and detonating roadside bombs struck in the late morning in Adhaim -December 4, 2005-Allawi chased from mosque -December 5, 2005- Witness at Hussein trial describes torture: The witness is describing the torture of man from a Shiite village in 1982: 'They broke all his body parts' -December 6. 2005-Suicide bombers kill 36 -Saddam witness tells of abuse -December 8, 2005- Bomber on bus kills 30. -December 15,2005 ? Iraqi National assembly elections. Iraqis vote for the first, full- term government and parliament since the US-led invasion. A crucial political transition. -Legal term for deployment of Japan's Self-Defense Forces troops in Iraq to expire -Polls open in historic election -Republic of Iraq to France, signed the bilateral agreement between France and Iraq on Iraq's debt to France by providing for 80% of Iraq?s debt to be cancelled between 2005- 2008. It is a substantial contribution to the restoration of the country's financial situation and its reconstruction. The debt canceled by France will amount to about 4 billion euros in the period from 2005 to 2008. -December 18,2005 ? Bush admits that much of the intelligence turned out to be wrong. -Bush hails Iraqi anti-terror role -December 20, 2005-Hussein claims he was beaten in custody -December 23, 2005- More than 825 people have died and another 1, 300 have been wounded this year in the Diyala province. Consists of a diverse mix of Sunni Arabs and Shiites alongside a Kurdish minority. 209 -December 27, 2005-Ukrainian, Bulgarian troops left Iraq -December 29, 2005-International team to review elections - Death threats cut oil flow 2006 -January- -January 4,5, 2006- More than 150 people are killed in suicide bombings and attacks targeting Karbala, Ramadi, Miqdadiya and Baghdad. -January 20,2006- Shia-led United Iraqi Alliance emerges as the winner of December's parliamentary elections, but fails to gain an absolute majority. -January 28,2006- Captors: Last chance for demand -February- -February 2006- A bomb attack on an important Shia shrine in Samarra unleashes a wave of sectarian violence in which hundreds of people are killed. -Feb. 12-Jaafari set to stay as PM -Feb. 14 -Baghdad Saddam in 'hunger strike' protest -Feb. 21 - Blast damages Shia shrine A bomb attack has damaged one of the Muslim world's holiest shrines, sparking demonstrations outside. Thousands have gathered at the al-Askari shrine in Samarra where 2 men blew up the famous golden dome. The shrine is one of Samarra's 2 tombs for revered Shia imams. -Feb. 22- Shrine blast sparks protests -Feb. 23 - Curfew aims to curb violence The government has put Baghdad and 3 210 provinces under curfew for much of Friday -Feb. 26 - Renewed violence amid calls for calm -March- -Mar. 10 - FBI: Body of U.S. hostage found The body of Tom Fox, a Christian peace advocate, has been found -Mar. 14 - More than 80 dead in reprisal killings 86 bodies were found in the capital- during a 30-hour period. Officials announced a vehicle curfew -Mar. 16- Military launches largest air assault 50 aircraft are involved in Operation Swarmer with more than 1,500 Iraqi and U.S. troops near Samarra -Mar. 23 - Peace group hostages freed -April- -Apr. 1 - Pressure mounts in Baghdad on PM to go -Condelizza Rice and Straw arrive -Apr. 4 - Hussein charged with genocide -Apr. 5 - Hussein grins during cross-examination -April 7,2006- 80 people are killed when three suicide bombers attack a Shia mosque in Baghdad. - Bombs kill at least 12 amid deadlock 211 Bibliography Adamson, E. (1984). Toward a marxist penology. Social Problems 31, 435-58. Alexander, M. S, & Kieger, J. F. V. (2002). "France and the Algerian War: Strategy, Operations, and Diplomacy," Journal of Strategic Studies 25(2), 6-7. Allison, P. D. (1995). Survival analysis using the SAS system: a practical guide. Cary, NC: SAS Institute. Anderson, J. W. (March 7 th , 2006). ?Iraqi Tribes Strike Back at Insurgents: In Turbulent Areas, Zarqawi's Fighters Are Target of Leaders and a New Militia.? The Washington Post. A12. Anderson, J. W. (May 28 th , 2007). US forces free al-qaeda in Iraq captives. The Washington Post. A11. Aran, S. (2003). Genesis and the future of suicide terrorism. Science 299, 1534- 1539. Aron, R. (1970). Main currents in sociological thought, volume II. New York, NY: Double Day. Asraf, A. & Brigden, K. (2006). Iraqi insurgency- using the internet to balance the battlefield. Paper presented at workshop ?Cooperating against terrorism: EU-US relations post September 11? at conference ?Future Challenges for Crisis Management in Europe? arranged by the Swedish Emergency Management Agency (SEMA) and the European Crisis Management Academy (ECMA), Stockholm, 3?5 May 2006. Baker, J. A., & Hamilton, L. H. (2006). The Iraq Study Group Report. Available from: http://www.usip.org/isg/iraq_study_group_report/report/1206/ Bailey, W. C. (1998). Deterrence, brutalization and the death penalty. Criminology 36(4) 711-733. Bartollas, C. (2005). Juvenile delinquency, 7 th edition. Boston: Allyn & Bacon. Beccaria, C. (1764). On Crimes and Punishments. Translated by Henry Paolucci (1963 ed), Indianapolis, IA: Bobbs-Merrill. Bernard, T. J. (2002). Donald Black?s influence on criminology. Contemporary sociology, 31(6) 650-652. Berry, B. (1997). The future of critical theory in criminology. Critical criminology newsletter Critical Criminology--Past, Present and Future. Special Issue, Winter, 1997 (Vols 2) 212 Bendix, R. (1960). Max Weber: an intellectual portrait. Garden City, NY: Anchor Books. Biddle, S. (2006). Seeing Baghdad, thinking Saigon. Foreign affairs 85(2), 2-14. Biglan A., Ary D. & Wagenaar. A C. (2000). The Value of Interrupted Time- Series Experiments for Community Intervention Research. Prevention Science 1(1), 31- 49. Black, D. (1976). The behavior of law. New York, NY: Academic press. Black, D. (1989). Sociological Justice. New York, NY: Oxford university press Black, D. (2004). Terrorism as social control. Sociology of crime, law and deviance 5, 9-18. Blau, P. (1964). Exchange and power in social life. New York, NY: John Wiley & Sons. Blau, J. & Blau, P. (1982). The costs of inequality: metropolitan structure and violence crime. American Sociological Review 47, 114-129. Bogen, K. T., & Jones, E. D. (2006). Risks of mortality and morbidity from worldwide terrorism: 1964-2004. Risk analysis 26(1) 45-59. Bohara, A. K., Mitchell, N. J., & Nepal, M. (2006). Opportunity, democracy, and the exchange of political violence. Journal of conflict resolution 50(1), 108-128. Bohm, R. M. (1982). Radical criminology: an explication. Criminology, 19, 565- 589. Bonger, W. (1916). Criminality and economic conditions, Boston, MA: Little. Box, G. E. P. & Tiao, G. C. (1975). Intervention analysis with applications to economic and environmental problems. Journal of the American Statistical Association 70(349), 70-79. Boyns, D. & Ballard, J. (2004). Developing a sociological theory for the empirical understanding of terrorism. The American Sociologist 35(2), 5-25. Callaway, R, & Harrelson-Stephens, J. (2006). Toward a theory of terrorism: human security as a determinant of terrorism. Studies in conflict & terrorism 29: 679- 702. 213 Campbell, C. & Connolly, I. (2003). A model for the war against terrorism? Military intervention in northern Ireland and the Falls Curfew. Journal of law and society 30:341-375. CBS Report. 60 Minutes II, April 28, 2004 Abuse of Iraqi POW's by GI's probed. Available from: http://www.cbsnews.com/stories/2004/04/27/60II/main614063.shtml Chambliss, R. (1973). The saints and the roughnecks. Society 2, 24-31. Chambliss, R. (1975). Toward a political economy of crime. Theory and society, 2, 149-170. Chambliss, W. J. & Seidman, R. (1982). Law, order, and power. Reading, MA: Addison-Wesley. Chamlin, M. B. & Cochran, J. K. (1998). Causality, economic conditions and burglary. Criminology 36(2). 425-440. Chamlin, M. B., & Cochran, J. K. (2000). Race riots and robbery arrests: toward a direct test of the threat hypothesis. Social Pathology 6(2) 83-101. CIA Factbook (2006). Iraq file: Available from: https://www.cia.gov/cia/publications/factbook/geos/iz.html#People Clarke, R. V. & Newman, G. R. (2006). Outsmarting the terrorists. New York, NY: Praeger Publishers. Cohen, L. & Felson, M. (1979). Social change and crime rate trends: a routine activities approach. American Sociological Review 44, 588-608. Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd Edition). Hillsdale, NJ: Lawrence Earlbaum Associates. Cohn, E. G. & Farrington, D. P. (1994). Who are the most cited scholars in major American criminology and criminal justice journals. Journal of criminal justice 22:517- 34. Collier, P. (2000). Rebellion as a quasi-criminal activity. Journal of conflict resolution 44(6), 839-853. Collins, R. (1975) Conflict Sociology: toward an explanatory science. New York, NY: Academic Press. Colvin, M. & Pauly, J. (1983). A critique of criminology. American journal of sociology 89, 513-31. 214 Cook, T. D. & Campbell, D. T. (1979). Quasi-experimentation: design and analysis for field settings. Boston, MA: Houghton Mifflin Cooper, B. (2007). ?Jihadists? and the war on terrorism. The intercollegiate review. 42(1) 27-36. Cornish, D & Clarke, R. (1985). Modeling offender?s decisions: a framework for research and policy. In M. Tonry & N. Morris (eds), crime and justice annual review of research (pp. 147-185). Chicago, IL: University of Chicago Press. Cox, D. R. (1972). Regression models and life-tables. Journal of the royal statistical society, Series B (methodological) 34:187-220. Crenshaw, M. (1990). The logic of terrorism: terrorist behavior as a product of strategic choice. In The origins of terrorism. Ed. Walter Reich. Washington. DC: Woodrow Wilson Center Press. Crenshaw, M. (1992). Current research on terrorism: the academic perspective?. Studies on conflict and terrorism 15, p1. Crisis States Research Centre, (February 2007). Definition of failed states: Available from: http://www.crisisstates.com/download/drc/FailedState.pdf Cronin, A. K. (2002). Behind the curve: globalization and international terrorism. International security. 27(3) 30-58. Cronin, A. K. (2006). How al-Qaeda ends: the decline and demise of terrorist groups. International security 31(1) 7-48. Dahrendorf, R. (1959). Class and class conflict in industrial society. Palo Alto, CA: Stanford University Press. D'Alessio, S. J., & Stolzenberg, L. (1998). Crime, arrests and pretrial jail incarceration: an examination of the deterrence thesis. Criminology 36(4) 735-762. Dayaratna, K. D. (2005). Ireland and Iraq: two sides of the same coin. Berkeley paper, 03/30/05. Deflem, M. & Sutphin, S. (2006). Policing post-war Iraq: insurgency, civilian police, and the process of the reconstruction of society. Sociological focus 39(4), 265- 283. Dejong, C. (1997). Survival analysis and specific deterrence: integrating theoretical and empirical models of recidivism. Criminology 35(4) 561-575. 215 Della Porta, D. (1995). Social movements, political violence and the state, a comparative analysis of Italy and Germany Cambridge, England: Cambridge University Press. Deutsche Welle (2006). Germany examines Allawi assassination plot. 7 December, 2004. Disraeli, B. (1863). Speech in the House of Commons, February 5 th , 1863. Doleschal, E. & Klapmuts, N. (1973). ?Toward a new criminology? Crime and Delinquency 5:607 Dugan, L, LaFree, G. & Piquero, A. R. (2005). Testing a rational choice model of airline hijackings. Criminology 43(4) 1031-1066. Durkheim, E. (1897). Suicide. Published 1952, New York, NY: Macmillan Emery, E. (Jan 2, 2007). Carson unit returns from ?little Iraq?. The Denver Post. Enders, W. & Sandler, T. (1993). The effectiveness of antiterrorism policies: a vector-autoregression-intervention analysis. American political science review 87: 829- 844. Eppright, C. T. (1997). Counter terrorism and conventional military force: the relationship between political effect and utility. Studies in conflict and terrorism 20:1-23 Erikson, K. (1966). Wayward Puritans. New York, NY: Wiley. Etzioni, A. (1961). A comparative analysis of complex organizations. New York, NY: Free Press. Feierabend, I. K., Feierabend, R. L. & Nexvold, B. N. (1969). Social change and political violence: cross-national patterns. pg 476-479. In H.D. Graham and T. R. Gurr, Eds. Violence in America. New York, NY: Bantam. Filkins, D. (27 November, 2003). Attacks on G.I.?s in Mosul rise as good will fades. New York Times. Filkins, D. (21 July, 2002). Dateline: Kabul Afghanistan. New York Times, p A1. Fine, B., Kinsey, R., Lea, J. Picciotto, S. & Young, J. (eds). Capitalism and the rule of law: from deviancy to Marxism. London, UK: Hutchinson. Finer, J. (Nov 17 th , 2005). Among insurgents in Iraq, few foreigners are found. Washington Post Foreign Service, page A01 216 Finn, E. & Zenilman, A. (May 15, 2003). A Guide to Iraq's Shiite Clerics Available from Slate Magazine Online: http://www.slate.com/id/2082980/ Flashpoint (2006). Halting Iraq?s descent into chaos. Newsletter of the armed conflict database,1:1. Available from www.iiss.org/acd. Francisco, R. A. (1996). Coercion and Protest: an empirical test in two democratic states. American Journal of Political Science, 40(4)1179-1204. Fund for Peace (2007). Iraq as a failed state. Available online at: http://www.fundforpeace.org Gibbs, J. P. (1989). Conceptualization of terrorism. American Sociological Review 54(3) 329-340. Gottfredson, M. & Hindelang, M. (1979). Sociological aspects of criminal victimization. Annual review of Sociology, 7: 107-128. Greenberg, D. (1981). Crime and Capitalism: readings in marxist criminology. Palo Alto, CA: Mayfield. Hafez, M.. (2006). Suicide Terrorism in Iraq: a Preliminary Assessment of the Quantitative Data and Documentary Evidence. Studies in conflict and terrorism 29: 591- 619. Hagan, J. (1977). Conflict and consensus in the designation of deviance. Social Forces, 65: 719-744. Hagan, J. (1989). Structural criminology. New Brunswick, NJ: Rutgers University Press. Hamm, M. S. (2004). Apocalyptic violence: the seduction of terrorist subcultures. Theoretical criminology 8(3) 323-339. Hashim, A. S. (2003). The Sunni Insurgency In Iraq. Center for Naval Warfare Studies. Available from: http://www.mideasti.org/articles/doc89.html Hibbs, D. Jr. (1973). Mass political violence: a cross-national causal analysis. New York, NY: Wiley. Hindelang, M. J. (1978). Race and involvement in common law personal crimes. American Sociological Review 43: 93-109. Hoffman, B. (1998). Inside Terrorism. New York, NY: Columbia University Press. 217 Hoffman, B. (2004). Insurgency and counterinsurgency in Iraq. Occasional Paper, June 2004, RAND Corporation Hudson, R. A. (1999). The Sociology And Psychology Of Terrorism: Who Becomes A Terrorist And Why? http://www.loc.gov/rr/frd/pdf- files/Soc_Psych_of_Terrorism.pdf Huntington, S. P. (1993). The Clash of Civilizations? Foreign Affairs, 72(3) 22-49 International crisis group (2006) The arab-israeli conflict: to reach a lasting peace. Middle-East report No. 58. (5 October, 2006). Ikenberry, G. J. & Slaughter, A. (2006). ?Forging a world of liberty under international law?. Final report of the Princeton project on national security. http://www.wws.princeton.edu/ppns/report/finalreport.pdf Iraqi Body Count (February, 2007). http://www.iraqbodycount.org/ Iraqi Constitution (2005). United Nations translation. Available from: http://www.washingtonpost.com/wp- dyn/content/article/2005/10/12/AR2005101201450.html Iraqi Insurgency (February 18 th , 2007). The insurgency in Iraq. Wikipedia entry, available at http://en.wikipedia.org/wiki/Iraqi_insurgency Iraqi Opinion Polls (2006). Available from: http://www.brookings.edu/iraqindex Jamieson, K. & Blowers, A. N. (1993). A structural examination of misdemeanor court disposition patterns. Criminology 31(2) 243-262. Kepel, G. (2004). The war for muslim minds: Islam and the West. Cambridge, MA: Belknap Press of Harvard University Press. Kilcullen, D. J. (2005). Countering global insurgency. The journal of strategic studies 28(4) 497-617. Kim, K. S., Myeong, M. H., & Kweon, Y. (2006). Evaluating the effects of safety policy measures on traffic fatalities in Korea. Transport reviews 26(3) 293-304. Komarow, S. (May 19 th , 2007) Washington Post, May 19 th , 2007. Search for US soldiers in Iraq widens. Korany, B. (2005). The middle east since the cold war: torn between geopolitics and geoeconomics, pp. 59-76 in Louise Fawcett, ed., International relations in the middle east. New York, NY: Oxford Press. 218 Kreuger, A. B. (2006). The national origins of foreign fighter in Iraq. Available online from: www.aeaweb.org/annual_mtg_papers/2007/0105_1430_1601.pdf Krisberg, B. (1975). Crime and privilege. Englewood Cliffs, NJ: Prentice-Hall. LaFree, G. (1998). Losing Legitimacy. Boulder, CO: Westview press. LaFree, G. (2005). Evidence for elite convergence in cross-national homicide victimization trends, 1956-2000. The Sociology Quarterly 46, 191-211. LaFree, G. & Drass, K. (1997). African-American collective action and crime, 1955-91. Social forces 75(3): 835-853. LaFree, G. & Dugan, L. (2004). How does studying terrorism compare to studying crime? In: Terrorism and counterterrorism: criminological perspectives. Sociology of crime, law and deviance 5, 53-74. LaFree, G. & Dugan, L. (2007). "Introducing the Global Terrorism Database." Terrorism and Political Violence 19:181-204. LaFree, G., Dugan, L. & Korte, R. (March 2008). Efficacy of counterterrorism approaches: examining Northern Ireland. College Park, Md: Research Brief , National Consortium for the Study of Terrorism and Responses to Terrorism: www.start.umd.edu Laird, M. (2005). Iraq: learning the lessons of Vietnam. Foreign Affairs 84(6). 22- 43. Laqueur, W. (1977). Interpretations of Terrorism: fact, fiction and political science. Journal of Contemporary History 12(1) 1-42 Legge, J. S. Jr. (1990). Policy alternatives and traffic safety: mandatory seat belts and drunk driving reform in California. The Western Political Quarterly 43(3) 613-630. Lipset, M. S. (1960). Political man: the social basis of politics. New York, NY: Doubleday. Liska, A. (1992). Social threat and social control. Albany, NY: SUNY Press. Lofland, J. (1969). Deviance and identity. Englewood Cliffs: NJ. Prentice-Hall. Lum, C., Kennedy, L., W. & Sherley, A. J. (2006). The effectiveness of counter- terrorism strategies, a Campbell systematic review. START Center publication, www.start.umd.edu 219 Lynch, M. J. & Groves, W. B. (1986). A primer in radical criminology. New York, NY: Harrow & Heston. Lynch, M. J. (1987). Quantatitive analysis and Marxist criminology: some old answers to a dilemna in Marxist criminology. Crime and social justice 29, 110-127. MacLean, B. D. (1997). Thinking critically about crime. Richmond, British Colombia: Collective press. Mann, L. & Dashiell, T. C. (1975). Reactions to the draft lottery: a test of conflict theory. Human relations 28(2) 155-173. Marighela, C. (1970). Minimanual of the urban guerrilla Havana, Cuba: Tricontinental Marx, K. & Engels, F. (1848). The Communist Manifesto. New York, NY: International library Masharbash, Y. (2005). What al-Qaida really wants. Speigel Online, 12 August, 2005. Available from: http://www.spiegel.de/international/0,1518,36944,00.html. McCauley, C. (2006). Jujitsu politics: terrorism and responses to terrorism. In Collateral Damage: the psychological consequences of America?s war on terrorism. Paul R. Kimmel & Chris E. Stout (Eds.). Westport, CT: Praeger. McDowall, D., McCleary, R. Meidinger, E. E., & Hay, R. A. Jr. (1980). Interrupted time series analysis. Beverly Hills, CA: Sage. McGarell, E. & Castellano, T. (1991). An integrative conflict model of the criminal law formulation process. Journal of research in crime and delinquency 28, 174- 96. Memorial Institute for the Prevention of Terrorism (MIPT) (2004). MIPT Knowledge Base: a comprehensive databank of terrorists incidents and organizations. Available at: http://www.tkb.org/home.jsp. Merton, R. K. (1968). Social theory and social structure. Glencoe, IL: The Free Press, Michalowski, R. J. (1985). Order, law and crime: an introduction to criminology. New York, NY: Random House. O'Grady, E. M. (2006). Diyala Event Data Since March 12, 2003. Q Division, Lawrence Livermore National Laboratory, Livermore, CA 94550. 220 Neumann, P. R. (2006). Al Qaeda in Europe: policy priorities after the Iraqi war. Paper presented at ?a transatlantic partnership against terrorism-priorities and challenges ahead?. Swedish emergency management agency, Stockholm 4-5 May, 2006. Nevin, J. A. (2003). Retaliating against terrorists. Behavior and social issues 12:109-128. O'Leary, C. A. (2002) The Kurds of Iraq: recent history, future prospects. Middle East Review of International Affairs 6(4) http://meria.idc.ac.il/journal/2002/issue4/jv6n4a5.html Olson, M. (1971). The Logic of Collective Action. Boston, MA: Harvard University Press. Packer, G. (2006). Knowing the enemy. The New Yorker, Dec 18, 2006. Parker, N. & al-Kubacy, M. (Sept 13, 2006) ?Arm tribes to fight al-Qaeda, say Sunnis? The London Times. Available online from: http://www.timesonline.co.uk/tol/news/world/iraq/article637081.ece Parker, T. (2007). Fighting an Antaean enemy: how democratic states unintentionally sustain the terrorist movements they oppose. Terrorism and political violence 19: 155-179. Pepinsky, H. (1995). "Peacemaking Primer." Peace and Conflict Studies 2:32-53. Perl, R. (2006). ?Terrorism and National Security: Issues and Trends,? CRS Issue Brief for Congress (Updated March 9, 2006), http://www.fas.org/sgp/crs/terror/IB10119.pdf Piazza, J. A. (2006). Rooted in poverty?: terrorism, economic development and social change. Terrorism and political violence 18(1) 159-177. 35 Nation Poll on Iraq. PIPA: Program on international policy attitudes (Feb 28 th , 2006). World Public Says Iraq War Has Increased Global Terrorist Threat. Available online at: http://www.worldpublicopinion.org/ What the Iraqi Public wants. PIPA: Program on international policy attitudes (Jan 31 st , 2006). Majority of Iraqis Endorse Election and Show Optimism. Available online at: http://www.worldpublicopinion.org/ Platt, T. (1975). Prospects for a radical criminology in the USA. In Taylor, Walton & Young (Eds.), Critical criminology (pp. 95-112). Boston, MA: Routledge & Kegan Paul. 221 Plous, S. L. & Zimbardo, P. G. (2004). How social science can reduce terrorism. The chronicle of higher education. Sept 10 th , 2004. B9-B10. Polk, W. R. (2005). Understanding Iraq: The Whole Sweep of Iraqi History from Genghis Khan's Mongols to the Ottoman Turks to the British Mandate to the American Occupation. New York, NY: HarperCollins Inc Pratt, T. C. (2002). Conflict theory, economic conditions and homicide: a time- series analysis. Homicide studies 6(4), 61-83. Preamble to the Iraqi Constitution http://portal.unesco.org/ci/en/files/20704/11332732681iraqi_constitution_en.pdf/iraqi_co nstitution_en.pdf Pridemore, W. A., Chamlin, M. B. & Cochran, J. K. (2007). An interrupted time- series analysis of Durkheim's social deregulation thesis: the case of the russian federation. Justice Quarterly 24(2). 271-290. Pridemore, W. A., & Freilich, J. D. (2007). The impact of state laws protecting abortion clinics and reproductive rights on crimes against abortion providers: deterrence, backlash, or neither? Law and Human Behavior 10:1-17. Quinney, R. (1970). The social reality of crime. Boston, MA: Little, Brown. Quinney, R. (1973). Critique of the social order. Boston, MA: Little, Brown Ranstorp, M. (2006). Mapping terrorism research: state of the art, gaps and future direction. Swedish national defense college, Sweden. Reiman, J. (1995). The rich get richer and the poor get prison. Boston, MA: Allyn & Bacon. Ricks, T. E. (2006). Military plays up the role of Zarqawi. The Washington Post, April 10, p A01. Roberts, A. (2002). Counter-terrorism, armed force and the laws of war. Survival 44:7-32. Robinson, K., Crenshaw, E., & Jenkins, C. (2006) Ideologies of violence: the social origins of islamist and leftist transnational terrorism. Social Forces 84, 4. Ruggiero, V. (2005). Brigate Rosse: Political violence, criminology and social movement theory. Crime, law and social change 43: 289-307. Rummel, R. J. (1976) Understanding conflict and war: Vol 2: the conflict helix. Beverly Hills, CA: Sage Publications 222 Rusche, G. & Kirchheimer, O. (1939). Punishment and social structure. New York, NY: Russell & Russell. Sandler, T., Tschirhart, & Cauley, J. (1983). A theoretical analysis of transnational terrorism. The American Political Science review 77(1), 36-54. Schmid, A. P. & Jongman, A. J. (2005). Political Terrorism. London, U.K: Transaction Publishers. Schweid, B. (Sept 27, 2006). Poll: Iraqis Back Attacks on U.S. Troops. The Associated Press. http://www.washingtonpost.com/wpdyn/content/article/2006/09/27/AR2006092701435.h tml Sellin, T. (1938). Culture conflict and crime. New York, NY: Special sciences research council. Senate Democratic Policy Committee Hearing. (October 12, 2006) An Oversight Hearing on the Planning and Conduct of the War in Iraq: When Will Iraqi Security Forces Be Able to "Stand Up," So American Troops Can Begin to "Stand Down"? http://democrats.senate.gov/dpc/hearings/hearing39/transcript.pdf. Sherman, L. (1993). Defiance, Deterrence, and Irrelevance: A Theory of the Criminal Sanction. Journal of Research in Crime and Delinquency 30(4), 445-473. Silke, A. (2001). The devil you know: continuing problems with research on terrorism. Terrorism and political violence, 13, 1-14. Simmel, G. (1908). Conflict and the web of group affiliations. New York, NY: Free Press. Simpson, S. S., Bouffard, L. A., Garner, J. & Hickman, L. (2006). The influence of legal reform on the probability of arrest in domestic violence cases. Justice Quarterly 23(3) 297-316. Singer, J. D. & Willett, J. B. (1993). Its about time: using discrete-time survival analysis to study duration and timing of events. Journal of educational studies 18(2) 155- 195. Slugett, P. (2005). The cold war in the middle east, pp. 41-58 in Louise Fawcett, ed., International relations in the middle east. New York, NY: Oxford Press. Soule, J. W. (1989). Problems in applying counterterrorism to prevent terrorism: two decades of violence in northern Ireland reconsidered. Terrorism 12:31-46. 223 Sparks, R. F. (1980). A critique of Marxist criminology. In N. Morris and M. Tonry (Eds). Crime and Justice, Vol 2. (pp. 159-208). Chicago, University of Chicago Press. Spitzer, S. (1975). Toward a marxian theory of deviance. Social problems 22: 638-51. Sutherland, E. & Cressey, D. (1960). Criminology, 8 th ed. Philadelphia, PA: J.B. Lippincott. Taylor, I., Walton, P. & Young, J. (1973). The new criminology. London: Routledge & Kegan Paul. Taylor, I, Walton, P., & Young, J. (1975). Critical Criminology. London: Routledge & Kegan Paul. Thomas, J. & O?Maolchatha, A. (1989). Reassessing the critical metaphor: an optimistic revisionist view. Justice Quarterly 2:143-72. Thornberry, T. P. (1973). Race, socioeconomic status and sentencing in the juvenile justice system. Journal of criminal law and criminology 64, 90-98. Times (1997). Oklahoma Bomb was revenge for Waco Cult Deaths. The Times, 8 February 1997. Toby, J. (1979). Societal evolution and criminality: a parsonian view. Social problems 26: 386-391. Triplett, R. (1993). The conflict perspective, symbolic interactionism, and status characteristics hypothesis. Justice Quarterly 10:4, 541-558. Tripp, C. (2000). A history of Iraq. Cambridge, England: Cambridge University Press. Tucker, D. (1998) Responding to terrorism. The Washington Quarterly 21(1) 103- 117. Turk, A. (1966) Conflict and criminality. American Sociological Review 31: 338-52. Turk, A. (1975). Prospects and pitfalls for radical criminology: a critical response to Platt. Crime and Social Justice, 4. 41-42. Turk, A. (1980). Analyzing official deviance: for non-partisan conflict in criminology. In J. Inciardi (Ed). Radical criminology, the coming crisis (pp. 78-91). Beverly Hills, CA: Sage. 224 Turk, A. (1982). Political Criminality: the defiance and defense of authority. Beverly Hills, CA: Sage Publications. Turk, A. (2004). Sociology of terrorism. Annual Review of Sociology 30. 271- 286. Ullman, H. K., Wade, J. P. & Alberts, D. S. (1998). Shock and Awe: Achieving Rapid Dominance. Darby, PA: Diane Publishing Company. United States Counterinsurgency operation in Bacuba, Diyala. Available from http://www.foxnews.com/story/0,2933,123814,00.html United States Department of State. (2006). Country reports on terrorism 2005. Office of the coordinator for counterterrorism, April, p. 131. Vold, G., B. (1958). Theoretical criminology. New York, NY: Oxford University Press. Vold, G. B., & Bernard, T. J. (1986). Theoretical criminology. New York, NY: Oxford University Press. Walsh, A. & Ellis, L. (2007). Criminology: an interdisciplinary approach. Thousand Oaks, CA: Sage. Wallerstein, I. (1979). The capitalist world economy. Cambridge, UK: Cambridge University Press. Weber, M. (1919). Politics as a vocation. Available online: http://www.ne.jp/asahi/moriyuki/abukuma/weber/lecture/politics_vocation.html Weber, M. (1964). The theory of social and economic organization. New York, NY: Free Press. Weisburd, D. (1989). Jewish settler violence: deviance as social reaction. University Park, PA: Pennsylvania State University Press. Weisburd, D. (2005). Hot spots policing and criminal justice research: lessons from the field. The Annals of the American academy of political and social science 559: 220-245. White, M. D., Goldkamp, J. S. & Campbell, S. P. (2005). Beyond mandatory arrest: developing a comprehensive response to domestic violence. Police practice and research 6(3) 261-278. 225 Wilkinson, P. (2000). Terrorism versus democracy: the liberal state response. Portland, OR: Frank Cass. Wilson, J. M. (2006). Law and order in an emerging democracy: lessons from the reconstruction of Kosovo?s police and justice systems. Annals of the American Academy of Political Science, 605: 152-177. WMD Commission Report (March 31 st , 2005). Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction. Available from: http://www.wmd.gov/report/ Wolff, K. (Ed) (1950). The sociology of Georg Simmel. New York, NY: Free Press. Wonnacott, T. H. & Wonnacott, R. J. (1990). Introductory Statistics for business and economics, 4 th edition. New York, NY: Wiley. 'Zarqawi' beheaded US man in Iraq (2004). BBC News, 2004, May 13 th Online at: http://news.bbc.co.uk/1/hi/world/middle_east/3712421.stm