THE IMPACT OF RELIGION ON DOMESTIC RIGHT WING EXTREMIST GROUPS AND ORGANIZATIONAL VIOLENCE By Sarah Fitzgerald A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Criminal Justice 2011 ABSTRACT THE IMPACT OF RELIGION ON DOMESTIC RIGHT WING EXTREMIST GROUPS AND ORGANIZATIONAL VIOLENCE By Sarah Fitzgerald The purpose of this study is to investigate the relationship between religion and violent activity among domestic right wing extremist groups. Both religion as well as its intensity will be measured to determine if those groups who are religious are more prone to be involved in acts of violence than those who are not. The study will utilize an existing database of domestic right wing groups. While religion has been a justification for violence for centuries, few studies have examined it in an empirical matter as it pertains to domestic right wing extremist groups. This study will do so and determine the impact it has on organizational level violence. DEDICATION I dedicate this thesis to my boyfriend Matthew St. George who has been there for me and supported me in this process. iii ACKNOWLEDGEMENTS I would like to thank the members of my committee for their patience, assistance and support with this project. First, I am extremely grateful to Dr. Steven Chermak who has been accommodating and encouraging throughout the entire process and helpful to me to further develop my writing as well as knowledge and presentation of the subject matter. Dr. Thomas Holt has also helped guide me throughout my time in the Master’s program and aide me through the methodological aspects of this paper. Furthermore, I would like to thank Dr. David Carter for coming on board with the project and offering his guidance and extensive knowledge of the subject. Finally, I would like to thank my parents for their support and encouragement over the years. iv TABLE OF CONTENTS LIST OF TABLES…………….. ………………. ………………. ………………. ………………. ………………. vi CHAPTER 1 INTRODUCTION………………. ………………. ………………. ………………. ………………. ………………. 1 CHAPTER 2 LITERATURE REVIEW……….. ………………. ………………. ………………. ………………. ………………. 11 Prominent Extremist Groups……………………………. ………………. ………………. ………………. 11 Religoin: An Influencing Factor for Violence………………….. ………………. ………………. 15 Religious Violence and Terrorirsm………………………….…………. ………………. 15 Christian Identity……………………………………….. ………………. ………………. 17 Religious Research…………. ………………. ………………. ………………. ………………. ………………. 20 Other Influencing Factors for Violence ………………. ………………. ………………. 22 CHAPTER 3 METHODOLOGY……………… ………………. ………………. ………………. ………………. ………………. 27 Data…………………….. ………………. ………………. ………………. ………………. ………………. ………………. 27 Measures………………. ………………. ………………. ………………. ………………. ………………. ………………. 29 Analaysis…………… ………………. ………………. ………………. ………………. ………………. ………………. 36 CHAPTER 4 RESULTS………………………. ………………. ………………. ………………. ………………. ………………. ………………. 37 Univariate Results…………… ………………. ………………. ………………. ………………. ………………. 37 Bivariate Results……………… ………………. ………………. ………………. ………………. ………………. 42 Multivariate Results…………….. ………………. ………………. ………………. ………………. ………………. 44 Race Results…………………………………. ………………. ………………. ………………. ………………. 49 Year Founded Results…………………….. ………………. ………………. ………………. ………………. 52 CHAPTER 5 DISCUSSION/ CONCLUSION……………….. ………………. ………………. ………………. ………………. 56 BIBLIOGRAPHY……………… ………………. ………………. ………………. ………………. ………………. ………………. 63 v LIST OF TABLES Table 1- Univariate Results……………………………… …………………….. …………………….. …………………….. 37-38 Table 2- Bivariate Results………………………………………… …………………….. …………………….. …………………….. 43 Table 3- Ideological Violence Regression……………………… …………………….. …………………….. 45 Table 4- Any Violence Regression……………………………. …………………….. …………………….. 47 Table 5- Advocates Violence Regression………………………… …………………….. …………………….. 48 Table 6- Race Bivariate Results……………………… …………………….. …………………….. …………………….. 51 Table 7- Year Founded Bivariate Results…………………………. …………………….. …………………….. 53 vi Chapter 1 Introduction The purpose of this thesis is to examine the role of religious involvement and intensity in domestic right wing groups on participation in violent acts. Religion may be a key predictor in determining if a group is likely to become violent or commit a terrorist act. The analysis utilizes an existing database of domestic right wing groups, and will shed light on a topic that few have examined before. There is an increasing concern about the potential for extremists to commit violent or terrorist acts; religion may be a key influence predicting violent activities as it is often used as a justification for terrorist actions. Religious beliefs in groups may cause prejudice and intergroup conflicts within different segments of society (Jackson and Hunsberger, 1999). This study is unique in that it will examine the role of religion in two different ways. First, a group’s religious involvement will be measured by what particular religion or ideology that they affiliate themselves with. Second, intensity of this involvement will be measured on a scale of high, medium, low or none. Other variables will also be examined to determine whether religion matters and to identify other relevant predictors. In addition, this study explores three dependent variables including any violence, ideological violence, and whether a group advocates violence. Religious involvement is expected to have an impact on group violence as well as a group’s statements of advocating of violent actions. The study of the impact of religion on white supremacist groups is important for several reasons. First, concern about understanding the threat of terrorism increased dramatically following the September 11th attacks (New York Magazine, n.d.; Chermak and Freilich, 2010). The attacks impacted many sectors of society, including policymakers, law enforcement, health care and the general public. It is estimated that 422,000 New Yorkers suffered from post 1 traumatic stress disorder as a result of the incident, and 1.4 million Americans changed holiday travel plans involving flying to instead traveling by car or train (New York Magazine, n.d.). Americans living outside New York also faced stress related to the incident for months afterward (Stein et al, 2004). In addition, the attacks increased the fear that further terrorist attacks might occur on American soil. Indeed, Americans have rated terrorism as a main social problem impacting America, and in 2009, 75% of Americans indicated that they thought terrorism was a very or extremely important issue (Gallup, 2010). Second, law enforcement officials have identified terrorism as a primary concern and have adapted policies and procedures to help combat it. Showcasing this concern, the FBI currently lists its number one priority as protecting the United States from a terrorist attack (FBI, 2010). It is not only a challenge for law enforcement but also for intelligence agencies, security and anyone involved or associated with the safety and security of the public (Pynchon and Borum, 1999). Police officials are concerned not only about lone wolf activities but about group level violence. The types of criminal or violent activities may vary a great deal as well as vary dependent upon region (Chermak et al, 2010). Law enforcement at all levels should be concerned about the threat posed by domestic extremist groups. Local law enforcement will often be the first responders and be able to provide intelligence about their particular community and activities. Information provided on a local level is necessary to assemble a more complex picture of terror networks and prevent attacks. Third, academia has also made the study and understanding of terrorist actions a priority. Since 9/11, the amount of funding for terrorism research has grown and there has been a concomitant increase in the number of studies published that help us better understand issues 2 such as understanding lone wolf terrorism, international threats such as Al Qaeda and characteristics related to terrorism (Kingshott, 2003, Sedwick, 2004, Bellia, 2005). The focus of much of the published research post 9/11 has been on better understanding the threat of international terrorist groups, like Al Qaeda or Hezbollah. Although these clearly are important groups to examine, the threat of violence by domestic groups, especially by far right extremists, is significant (Quarles, 2004). Another way the importance of understanding domestic extremists can be showcased is through the growing body of empirical research that has examined domestic terrorism. There has been a significant increase in the number of studies examining terrorism since 9/11 (Silke, 2008). For example, there were 150 books published on terrorism in 2000. This number increased to 1108 books in 2001 and 1767 books in 2002. It is estimated that a book on terrorism is published every six hours (Silke, 2008). Prior to September 11th, there was limited funding or interest in research related to terrorism. However, this has drastically changed and there is a desire for a greater understanding of issues related to both domestic and international terrorism. Fourth, the threat of homegrown terrorist groups may be increasing as local law enforcement in that law enforcement officials have rated white supremacist groups as a top threat (Chermak et al, 2010). For example, one study that surveyed state agencies on their perceptions of extremist activities found that the vast majority of state respondents are concerned both with lone wolf activities as well as group level violence (Chermak et al, 2010). An example showcasing this threat at the federal level can be seen in the FBI’s release of Project Megiddo, a report outlining the threat of domestic extremist groups, and more specifically right wing extremist groups (FBI, 1999). The report predicts the potential for an increase in violent activities committed by domestic extremist groups. This threat was predicted to be enhanced by apocalyptic religious 3 groups that may see end times and become more violent. This pre-Y2K violence may be related to violence that may occur prior to 2012, surrounding other end-times hysteria. Furthermore, a significant proportion of terror incidents on U. S. soil were perpetrated by far right extremists. For example, Timothy McVeigh and Terry Nichols were found responsible for bombing the Alfred P. Murrah Federal Building in Oklahoma City in April of 1995 (SPLC, 2009). McVeigh believed in conspiracy theories and was linked to being an affiliate of the far right extremist movement. This attack killed 168 people, 19 of which were children located at a daycare center in the building (SPLC, 2009). In 1996, another far right extremist, Eric Robert Rudolph, set off a bomb at the Olympic Games in Atlanta that killed one person and injured 100 more (SPLC, 2009). He ascribed to the Christian Identity ideology and was found to be responsible for other attacks including the bombing of a gay bar as well as abortion clinics in Atlanta. In fact, between 1955 and 1998 there were 2,700 terrorist incidents in the United States, with the vast majority of terrorist killings being caused by Americans (Hewitt, 2000). This threat from the far right is shown to be a significant concern from a variety of extremist groups including survivalists, militias, Christian Identity groups and hate groups. A fifth reason for the importance of this study is that there are a growing number of identifiable hate groups within the United States, and these hate groups are becoming more violent (DHS, 2009). In 2009, there were 1,000 hate groups identified by the Southern Poverty Law Center, which is an increase of 136 groups from the previous year (Potok, 2010). Between 2000 and 2008 the number of groups had risen by 54% (Potok, 2010). Currently two driving forces in this increase in hate groups are the election of the first African American president as well as the economic downturn that are promoting radicalization and recruitment in extremist 4 groups (DHS, 2009). Another reason related to this rise can be related to the current issues surrounding immigration. Concurrent with the increase in the number of right wing extremist groups has been an increase in violent activities committed by members of these groups. This increase in violence today can be easily associated with mirroring the political climate of the 1990s, which also saw an increase in extremist involvement and violent crimes (DHS, 2009). In the 1990s, the increase was caused by criticism of outsourcing jobs, a perceived threat of government power, gun rights and an economic recession. Groups perpetrated many violent acts related to these issues and against symbols of government. Such targets included attacks against law enforcement officers, government facilities, infrastructure divisions and banks. This growth began to drop after government reaction to the Oklahoma City Bombing in 1995 and improvements within the economy. The United States government increased its efforts in handling domestic extremists resulting in a decline in the number of groups and violent actions committed by members and affiliates of these groups. Currently many trends exist within society that the government sees as potential threats that may cause right wing extremist group recruitment as well as radicalization toward violent actions (DHS, 2009). As military veterans return from overseas in Iraq and Afghanistan, there is a prediction that they will be ample targets for right wing group recruitment since they have military training. Many right wing extremists are also opposed to gun control legislation and restrictions on firearms that is leading to increased weapons and ammunition stockpiling as well as paramilitary training. Anti-government conspiracy theories and “end time” prophecies may also cause extremists to stockpile weapons, ammunition and food. Groups known to do this stockpiling in the past are those within the militia movement as well as violent Christian Identity 5 groups. Other issues that may impact the rise of extremist involvement include perceived job loss, free trade agreements, and home foreclosures (DHS, 2009). Many extremists attribute the loss of jobs and the increasing number of home foreclosures to be a direct result of a Jewish conspiracy. These factors can greatly contribute to the rise in this activity as well as the radicalization of groups that are becoming more violent. These issues illustrate the need to understand what may cause variations in violent behavior and better predict attacks. With these gaps in current research, a first step that can be taken is to look at variables that have been associated with the promotion of violence in the past, such as religion. Religion may be a driving factor that may lead to this violence. Furthermore, there are many gaps within the research literature in terms of examining group level behavior and radicalization of groups towards violence. A lot of research is focused on it as it pertains to the radicalization of an individual. A variety of elements may contribute to this indoctrination and a group’s decision to move toward violence, where religion may be a key predictor. Research not only in terms of violent group level behavior but also in terms of religious violence pertaining to domestic extremist groups is severely lacking. A portion of the studies that examine religion in terms of domestic extremist groups were published in the 1980s and 1990s but there are few recent studies (Barkun, 1989; Bolce and De Maio, 1999; Gallagher, 1997; Harper and Beau, 1993; Juergensmeyer, 1998; Juergensmeyer, 1991; Rapoport, 1998, Sprinzak, 1995). Since this threat is growing in the United States and groups certainly change over time a more recent study is necessary in order to provide a better understanding of the current atmosphere. There are few up to date studies on the topic, as well as few that take a more quantitative look at the violence committed by white supremacist groups in the United States. 6 Examining religion as a factor in violent attacks or terror incidents indicates that it may be a justification for action. If religion within a group is more likely to promote violence, it is important to understand that. Some studies have argued that religion matters (Rapoport, 2004; Sedgwick, 2004), while other studies find religion not to be an important factor in terrorism (Masters, 2008; Seegmiller, 2007). Because of these mixed results there is a significant need to more closely examine the linkage between religious ideology and violence in extremist right wing groups to be able to answer the question of whether or not religion matters. If it does not matter, it is also important to study other characteristics that may be the true predictors of violent behavior. There has been shown to be a link between religious involvement and effects on criminality in adults, showing the heavier involvement an adult has in a religious organization the less likely they are to commit crime (Evans et al, 1995). However, there is the potential for a heavy involvement in a group to be the real influencing factor. Therefore if a group is promoting violent actions, and are religious as well individuals may be more likely to commit a violent act due to their attachment to that organization or religious group. The variable of religion may be especially important to study due to its close associations with terrorism. Whether religious groups may be more violent than non-religious groups is debatable. For example the Christian Crusades can be seen as a famous example of how religion, and more specifically Christianity, was used to justify violence. The Crusades consisted of at least nine invasions of the Islamic East, with the first one occurring in 1095 (Setton, 1989). The purpose of the Crusades was to capture the Holy Lands, and these violent attacks were sanctioned by the Church and deemed to be in accordance with God’s will. 7 Another clear and more modern example of religious terrorism is the use of suicide bombers by extremist groups. While suicide bombers operate throughout the world, they clearly showcase the use of religion in justifying violent actions. Some suicide bombers believe they are on a mission from God and are motivated by a promise of being rewarded in the afterlife (Maiese, 2005). Many believe they are involved in a Holy War and are there to act on behalf of God. Martyrdom is seen as a status symbol and often bombers are viewed as heroes within their communities. Those who were involved with the September 11th plot were suicide bombers, willing to fly planes into buildings because of their religious beliefs. This violence justified through religious beliefs has become highly associated with violent actions committed by single issue groups. Single issue groups are typically those groups who are focused on one primary issue, which may be abortion, taxes, guns, etc. For example, abortion clinic violence is often tied to adherents to a far right ideology and extremist religious views. Between the years 1973 and 2003 abortion providers in the United States were targeted through more than 300 violent attacks (Jacobson, 2009). These attacks include bombings, murder and arson. In the 1980s and 1990s there was a drastic decline in the number of abortions as well as the number of abortion providers—which may be a direct result of the violence toward providers (Jacobson, 2009). Violence committed at abortion clinics is considered to be one of the most common forms of domestic terrorism within the United States (Jacobson, 2009). Religion plays a vital role among anti-abortion groups but may be a background predictor among other single issue groups. These examples demonstrate how religion has been used over time to justify violent terrorist actions, and are now being used by far right extremists. There are many reasons that 8 could easily illustrate why religion may facilitate this violence at a group level. Religion plays an important role in creating unity within a group and can play a crucial role in a group’s identity (Dobratz, 2001). To some religion may be a group matter or an individual matter, but often the traditions and shared beliefs will bring individuals together and create more group cohesiveness. A lot of far right groups have an emphasis on racist views, but a diverse number of radical groups can create room for a lot of individuals to find groups that promote similar ideals that they hold themselves. For some, the role religion plays may be a key component in that individual or group’s hope of a “white nationalist dream” (Kaplan, 1998). It is important to understand what factors may influence a group and make some groups behave one way such as through violence, while other groups may promote their ideology through nonviolent means (Rapoport, 1988). Clearly, terrorism is an extremely important issue to study for numerous reasons. Religion may play a key role in not only recruitment of individuals into an extremist group but also into a group using religion to promote violent actions. Currently, there is no research that helps us to completely understand the impact of religion on the criminal behaviors of right wing extremist groups. It would be helpful to have a better understanding of this and the differences that may exist between those that have a strong religious ideology, those that have a moderate one, those that have a low one and those that are not associated with any religious beliefs. There are five chapters in this thesis. The second chapter will cover a thorough literature review of issues of concern related to religion and domestic extremism. The third chapter will discuss the methodology of this study and go into greater detail about the specific variables that will be used in the analysis. The fourth section will supply the results of the study. Finally, the 9 paper will wrap up with a discussion of the implications of these results as well as conclusions of what this means and where future research should go from this point. 10 Chapter 2 Literature Review This portion of the thesis will seek to provide the background information needed to better understand religious extremist groups within the United States as well as discuss the potential influence of religion on extremist violence. Since there is very limited research assessing the influence of religion and other organizational characteristics on violence, it is necessary to consider a variety of broader topics that tie into this particular study. First, domestic right wing groups will be briefly touched on in order to paint a picture of the types of groups associated with the far right extremist movement in the United States and their uses of religion. Second, religion will be more specifically examined under the scope of far right extremists and Christian Identity. Thirdly, the literature review will conclude with a discussion of what we know about the impact of religion on violence. Finally, other potential influencing characteristics will be examined. Prominent Extremist Groups In order to best develop an understanding of the role religion plays within domestic extremist groups it is necessary to understand the basics of the far right movement within the United States and the different kind of groups that exist within it. Hundreds of right wing extremist groups exist within the United States today (SPLC, 2009). For the purpose of this study, all groups will be examined to look at variables related to their religious beliefs and involvement. While groups will not be specifically looked at in strict categories as Klan, Neo Nazi, etc during the analysis stage it is still important to understand how such groups may differ 11 in their structure and religious beliefs. The variety of extremist groups within the United States varies in terms of their religious beliefs and intensity as well as their use of violence or terrorism (Kaplan, 1995). In order to understand how religion is used and its role in groups committing organizational violence one must first see the characteristics of some of the prominent extremist groups and their religious beliefs. The Ku Klux Klan is the typical group that comes to mind when most people think of a white supremacist group in the United States. Ku Klux Klan groups are among the extremists that get the most public attention and hatred toward them, which brings out lot of counterdemonstrators when they hold events (Kaplan, 1995). Klan groups are known to build upon nationalism and patriotism in promoting their racist beliefs (Adams and Roscigno, 2005). In terms of religion, the Klan is known to use Christianity as an ideological foundation in the development of its identity (Adams and Roscigno, 2005). The element of religion has changed over time within the Klan, and is typically influenced by the leadership. While the Klan may use religion as a characteristic to help define the group, Neo Nazis use it as a tool to help create unity. Neo Nazis are another type of group that is prominent within America’s radical right. These types of groups are known for the activism within the political arena, relatively small sizes and use of leaderless resistance (Kaplan, 1995). Unlike Klan groups, the focus of Neo Nazi organizations is primarily on separatism or the creating of a new all-white nation (Adams & Roscigno, 2005). They are also known for using religion to help promote Aryan unity, particularly through websites that display symbols and pagan iconography to showcase history as well as the Aryan present. The use of religion in Neo Nazi groups has 12 changed over time, sometimes being a heavy influencing factors and other times being shadowed by other ideological goals. Another group that uses religion in a similar manner to both the Klan and the Neo Nazis as a driving ideological force is the Aryan Nations, which started by Richard Butler who had a forty acre compound in Hayden Lake Idaho, (which was later lost through a lawsuit) (Aho, 1995). This group is very anti-Semitic and holds Neo-Nazi characteristics. This compound that the group had was a gathering place for many within the radical right, as well as a place for survivalist training (Kaplan, 1995). Religion also plays an important role within this group with a main belief that they promote is that Christ was an Aryan and not a Jew and that white AngloSaxons are the chosen people with the United States being the promise land. Just like Aryan Nations, a group known as The Order began in Idaho (Aho, 1995) and when founded in 1983 was a splinter group of the Aryan Nations that is often viewed as one of the most violent extremist groups in the United States (Al-Khattar, 2003). Members of Aryan Nations left to form the Order because they thought that they should be involved in more extreme tactics in order to achieve the goals of the group. The group was based in northern Colorado and came into the spotlight after it was part of investigations surrounding the murder of a talk show host in Denver (Kaplan, 1995). However, the group is now extinct. Odinism is a less known religion that gets associated with far right rhetoric and is different than the aforementioned groups in that it is a religion followed by many extremist groups. Odinism is sometimes referred to as Asatru or Wotaninism and is a neopaganism religion (Dobrtaz, 2001). It is focused on those values from ancient European and those of religions prior to Christianity. Odinism is argued to be a link between Wiccan witchcraft and neo paganism 13 beliefs and the radical right (Kaplan, 1995). One particular group within America’s extremist right is the Church of the Creator. This group was founded in 1973 by Ben Klassen and is dedicated to promoting the white race and has a survivalist mentality (Dobrtaz, 2001). The main belief within this group is that Christian Identity is misguided along with all other religions (Kaplan, 1995). They believe that Christianity is built upon a lie and therefore all religions are false. They also reject the belief in any kind of a god or otherwise supernatural being and instead belief in what they call “creativity”. This belief involves a form of rewritten Christianity as well as racism and health faddism. Other groups that are prominent in the extreme right as well as have deep ties to religion are The Covenant, the Sword and the arm of the Lord (Rapoport, 1988), Christian Defense League (Quarles, 2004), Committee of the States and the White Knights of the Ku Klux Klan. Besides these specific groups that have a variety of goals and agendas, there are some groups that fall into the far right extremist category that are single issue constituencies. Such issues that these groups may be concerned with might be tax protesting or abortion among a variety of others (Kaplan, 1995). Militias may also fall into this particular category. Many hate groups have existed for several decades but militias are newer and are more paramilitary orientated with an emphasis on guerilla training, and survivalism (Hoffman, 2006). Taken as a whole, there are various groups categorized within the extremist right in the United States and the role religion plays within them. These groups are only briefly touched on not to provide a history or thorough analysis of each of them but rather to paint a broader picture of the types of groups in the far right as well as how they see religion. Some of these groups will be further discussed during the analysis phase to better describe how these large individual 14 groups use religion to justify violent actions. This discussion is necessary in order to develop a comprehension of the current composition of far right extremists. Looking at the breakup of the extremist right in the United States helps to provide a basis for the types of groups that the study will be examining. These groups vary vastly in terms of their characteristics, and more specifically their religious affiliation and religious intensity. Clearly, religion plays a key role in some of these groups and may act as an influencing factor for violence. Religion: An Influencing Factor for Violence Some studies have argued that religion matters and is an important variable to be studied when looking at group level behavior, such as violence (Rapoport, 2004; Sedgwick, 2004). Religion may play an important role in creating an identity for a group (Dobratz, 2001), as well as making a group more cohesive by bringing them together (Kaplan, 1998). To best examine how religion may have importance for this study, first religion will be discussed as it pertains to its current associations with terrorism. Second, the historical context of religious violence will be examined through Christian Identity groups since they are the most prominent and most often associated religious groups with the far right movement in the United States. Finally, research finding that religion may be a key predictor of violence will be examined. Religious Violence and Terrorism There may be a new form of terrorism that is highly associated with religious motivation and mass destruction (Masters, 2008). Religious terrorism is much different than secular terrorism in many fashions (Hoffman, 1998). Recently there has been an exponential increase in the number of religious terrorist groups throughout the world and a decline in the previously 15 predominant secular groups. Acts of religious terrorism are often seen as being more violent than their non religious counterpart and have quite varying characteristics. For example, in 1995, religious international terrorist groups were responsible for 25% of all terrorist incidents, yet were responsible for 58% of the fatalities related to terrorism (Hoffman, 1998). It is important to understand these differences as well as recognize the types and prominent incidents of religious terrorism that have not only occurred domestically but also internationally. Secular terrorists are seen as rarely acting in a situation to perpetrate mass killings because it will not be consistent with their political objectives (Hoffman, 1998). They are also viewed as wanting to be liked by the general public while they promote their beliefs. On the other hand, religious terrorists are viewed as not having constituency and are viewed as wanting to target anyone who is not in their religion or their group. Religious terrorists often want to fight to change an entire order, while secular terrorists are seen as fighting to change a flaw in the system. As discussed previously, religious terrorism has occurred for centuries. The 1990s showcased a recent surge in religious terrorist groups as well as actions committed by these groups (Hoffman, 1998). In 1993, one of the World Trade Center towers in New York City was targeted by religious extremists. While this attack got a lot of attention it was certainly shadowed in the attention received after September 11th, 2001. The 1993 attack was brought on by Islamic extremists who wanted to topple one of the towers and tried to release a deadly gas into the air (Hoffman, 1998). An airplane owned by Air France was hijacked in 1994, by Islamic extremists who attempted to blow the plane and its passengers up while it was over Paris, however, this 16 attack failed. In 1995, a religious cult in Japan set of sarin nerve gas on the Tokyo subway that ended up killing twelve people and injuring nearly 4,000 people (Hoffman, 1998). Numerous other attacks have occurred since these including the Oklahoma City Bombing; however, the most notorious act of religious extremism within the current day United States would be the attacks of September 11th. These attacks brought a great deal of focus on Islamic extremism as well as the concept of suicide bombers. Suicide bombing has clear ties to religious terrorism, with the group practicing it the most being the Tamil Tigers out of Sri Lanka (Pape, 2003). Suicide bombings have been greatly increasing. Contrary to how the media may portray it, the majority of attacks are not done by individual fanatics but are tied to groups which want to perpetrate the attack to further their political or ideological goals. While, suicide bombings is often an issue of international terrorism and is often not associated with domestic extremist groups it does showcase a clear example of how religion can impact individuals or groups to promote violent behavior or jihad. Religious violence may be representative of a new wave of violence and terrorism (Rapoport, 2004). David Rapoport proposes that four waves of terrorism exist, and that the first three are over. The fourth wave is predicted to be what society has entered into now, a wave of religious fundamentalism and violence. Religion may be a vastly important variable to modern day extremist organizations impacting their use of violence. Christian Identity Understanding the Christian Identity movement is critical to having a better understanding of religious violence that occurs in the extremist right in the United States by 17 providing a historical context. Throughout history violence can be associated with religious groups, including Christianity (Juergensmeyer, 1998). Groups such as Al Qaeda are known to use their religion to justify their violent actions. By understanding Jihad and the beliefs, it can help to create a better understanding of why a group like Al Qaeda may commit certain actions. This idea also holds true in terms of right wing extremist groups and their religions. One of these most predominant religions in the extremist right is Christian Identity The origins of the Christian Identity Movement may be ambiguous but most historians believe that the movement originated during the mid 19th Century in Britain (Hoffman, 2006). Its beginning can be traced back to British Israelism. Beliefs of this ideology are that the ten lost tribes of Israel were not Jews but rather white Anglo-Saxons. This way of thinking appealed to those considered elite in Britain, and was greatly inspired by the writings of Richard Brothers (Kaplan, 1995). Many other beliefs in British Israelism differentiated it from other beliefs systems and religions at the time as it morphed into the Christian Identity movement we know today in the United States. The notion of Christian Identity is the bringing together of various ideas, such as British Israelism, anti-government sentiment based on the fear of a threat toward the white race, white supremacy, polygenist views, and fundamentalist Christianity (Sharpe, 2000). The view perpetuates the idea that Jews are seeds of Satan and not the true Israelites while the white race is (Kaplan, 1995). This idea is promoted through a two seeds doctrine, which argues that the true Israelites are the white race and that Jews are separate and were brought about due to a union between Satan and Eve. They also believe that Cain was non-white while Abel was white 18 (Quarles, 2004). This belief system, while started in Britain, soon made its way to the United States. British Israelism came to the United States early on during the 20th century through writings of William Cameron and the teachings of Gerald L. K. Smith (Juergensmeyer, 2000), with the work of Canadian author W. H. Poole also having a great influence (Kaplan, 1995). While in Britain the ideas of British Israelism were more associated with those individuals considered to be a part of the upper class the ideology shifted into more of a political philosophy when it came to the United States (Juergensmeyer, 2000). It was in the 1940s that a group of right wing extremists developed the doctrine of Christian Identity in the United States (Sharpe, 2000). After WWII, conservative protestant churches began to feel alienated and began to take on a more anti-Semitic stance (Kaplan, 1995). Studies conducted during this time found that those who considered themselves fundamentalist Christians had more anti Catholic as well as anti Semitic views and were generally more intolerant (Bolce and De Maio, 1999). The Christian Identity movement has changed as it has developed in the United States. There are four basic tenants that most of those within the current movement believe (Hoffman, 2006). First, the belief is that Jesus Christ is an Aryan and not a Jew. Second, also believe that white Anglo-Saxon people are the chosen people. Third, they believe that whites are from the Lost Tribes of Israel and not Jews. Fourth, is that the United States is actually the chosen promise land. This is where some of their more racist views come in and African Americans are viewed as being talking apes (Aho, 1990), and Jews are believed to be a result of a union between the Devil and Eve (Aho, 1990; White, 1989). The creationist origin exists for whites while non-whites are seen as an evolutionist origin and being less than human (Sharpe, 2000). 19 Today the Christian Identity movement is highly decentralized. Both groups and individuals involved in the movement have a varying amount of attachment and promotion of these beliefs. Some groups are based solely around this concept, while others adopt it later on or are more focused on other issues. Complexity, disorganization and variation are often used to describe the movement (Quarles, 2004). Those who become involved may be so due to the ability of groups to provide fellowship, spirituality and religion (Kaplan, 1995). Younger individuals are often attracted to the movement because of captivating leaders as well as the nature of fundamentalist groups (Sharpe, 2000). This disorganization and variance among Christian Identity groups is one difficulty in attempting to measure the importance of religion. However, one way to help showcase these differences is to not only look at religious intensity and involvement between groups but also within groups. Fifty years after Christian Identity came to the United States it became a prominent voice within the radical right during the 1990s (Juergensmeyer, 2000). Many different radical groups follow the Christian Identity doctrine, yet these groups differ from each other in numerous ways. However, one way many of these groups are alike is in their known use of violent actions that they often justify through their religious beliefs. Religious Research Research has found a link between religion and violence within groups (Rapoport, 2004; Sedgwick, 2004). This link may be discussed in how it pertains to groups using religion to promote discrimination, legitimize beliefs, promote of justify violence and create unity within a group. Religion may be used as a tool that allows for or creates violence among far right extremist groups within the United States. First, religion has been found to be a key predictor for 20 non religious discrimination (Fox, 2004). This showcases that the variable religion may perpetuate discrimination against minorities that may lead to acts of violence. Furthermore, the vast majority of conflicts occur between religious groups rather than within them. Second, religion may be used to legitimize beliefs within America’s radical right (Gallagher, 1997). Religion is used as a means to legitimize the groups’ beliefs like antiSemitism, anti-government, racism and justifying the use of violent acts in order to meet the group’s goals. The idea is that the ends justify the means and that the violence is necessary in order for the group’s goals to be met. Third, religion has been found by many to be used as a justification for violence (Al Khattar, 2003; Appleby, 2000; Barkun, 1997; Fox, 1999). One way this is viewed is through a common belief in the radical right of the immense of a race war within the United States (Blazark, 2001; Kimmel & Ferber, 2000; Stern, 1999). With groups believing that the race war is inevitable, they may be more likely to use violence believing that end days are near (Rapoport, 1988). This belief may cause many Christian survivalists to prepare by stocking up food and weapons (Quarles, 2004), and they may also practice war games in preparation (Sharpe, 2000). This creates a belief among many followers that the war will lead to an Aryan victory that will reinstate the power of white men within the world and thereby restore order. More specifically, violence may be justified through language within the Bible. Groups may use the words within the Bible, take them literately or twist them in order to rationalize the use of violence to further the group’s ideology and goals (Al-Khattar, 2003; Juergensmeyer, 1991). They may reinterpret the Bible in a way that coincides with their goals (Hoffman, 1997). Religious imperatives can certainly be a motivator in justifying violence (Crenshaw, 2000). 21 Individuals involved in abortion clinic bombings are known to cite scripture as a means of justifying their actions (Rapoport, 1988). Fourth, violence may be further facilitated within these groups through the group identity that religion helps to create (Silberman et al, 2005). Individuals within the group may feel that their identity is being threatened by outside sources and feel a need to defend it. Rules or standards of a particular extremist group may also provoke violent occurrences. Myths within groups may be powerful enough and offer collective meaning that creates a sense of togetherness and otherness with those not in the group. Desecration may also call for violence if members of a group feel like those things they hold sacred are being violated or destroyed and feel it requires for immediate action to be taken. Clearly, religious violence is not a new phenomenon but is certainly a growing one. For centuries religion has been used as a justification by groups and individuals to commit violent acts. Religion and extreme fundamentalism may be a way for groups to express their political viewpoints (Poland, 1998). This section has provided a clearer picture of religion in the far right, and violence and its justifications, while the next section will discuss other potential predictors within groups that may cause violence. Other Influencing Factors for Violence Many have argued and some studies have found that religion may be closely associated with acts of violence within a group (Rapoport, 2004; Sedgwick, 2004). However, religion may not be an important characteristic driving violence or terrorist actions (Masters, 2008; Seegmiller, 2007). When looking at involvement in crime, opposite results are found meaning 22 that adults with heavy involvement in religious organizations are actually less likely to commit violent crime (Evans et al, 1995). Religion may also matter for some groups such as those that are international in scope, or associated with such actions as suicide bombings, though these issues may not matter on a domestic right wing level (Maise, 2005). It is vital to address this issue, and determine if religion does not matter what other characteristics may have an impact on determining a groups’ likelihood to be violence prone. One such factor may be a group’s structure and organizational level characteristics. Understanding the norms of a group could help understand not only the group’s attitudes but their decision making process as well (Pynchon and Borum, 1999). An understanding of the structure of the group would also help provide a picture of the internal operations that take place within a group and help govern it. The cohesiveness of a group would showcase how much the members of the group have a “we” or “us versus them” mentality and may therefore justify a collective kind of violence. Finally, one very key factor in understanding the risk a group may pose is to know what current situations are impacting the group and may increase their likelihood of responding in a violent manner. One of these situations may be an increased sense of urgency which may cause irrational reactions or be triggered by a specific catalyst event. Groups may see a nonviolent response as not being a viable option, this sense of urgency to take action may kick in which could easily be provided by religion. A group’s motivation is also a key element to help identity the risk for violent actions that a group may possess. Groups seek to see themselves and their actions in a positive light and may believe themselves to be better than other people not in the group and even see outsiders as being associated with more negative traits (Fiske and Taylor, 1991). Another important factor is an 23 understanding of the level of individual responsibility within a group. If members see responsibility of an action as spread throughout the group then they may have a limited sense of personal responsibility, which may help them justify committing a violent action by having the group behind them. This group cohesion may be developed out of the structure of an organization, since this may impact the likelihood a group may have to become involved in violent activities (Cordes et al, 1985). A strong charismatic leader may be able to recruit more individuals into the movement, as well as create an atmosphere more conducive to violent behavior. For example, a strong leader may help push forward groupthink. Groupthink was a word first coined by Irving Janis, where members of a group want to create unanimity and may begin to think like the group in order to avoid inner group conflict (Janis, 1972). The more cohesive a group becomes the more pressure it has on its members to comply with the group and not disagree with the beliefs. A strong leader in an extremist organization may preach political and religious beliefs to the group that will thereby go unquestioned and becomes the norm within a group, and may be used to justify violent actions. Those attitudes help by the group are often more extreme than the individual attitudes held by the members (Levine and Moreland, 1988). Relating to organizational leadership and having a strong charismatic leader, recruitment practices or strategies may have an influence on violence. Recruitment within right wing extremist groups has been described to be very similar to that of recruitment into gangs (Blazark, 2001). The Internet has had a great influence on this with the creation of many more hate oriented websites and allowed for individuals to be able to spread their literature more easily as well as reach other individuals like them with extremist views (Adams and Roscigno, 2005). 24 Groups recruiting online may also use the internet to learn new violent techniques (DHS, 2009).Those individuals being recruited may be more prone toward violence, especially those who are younger and may be more easily influenced by the previous factor discussed, a charismatic leader (Sharpe, 2000). While recruitment may be one influencing factor, another one may be political involvement within a group (Rummel, 1997). Groups that are heavily influenced in the political process may deter from committing acts of violence due to being involved in a conventional avenue to further their goals. However, a failure to become involved in such an avenue may lead a group to become violent (Sprinzak, 1995). The ideology a group possesses may also make a group more prone to violence (Cordes et al, 1985). Racism in a group may serve as this ideological driving force, where some groups even believe that a race war is imminent in the United States and that they need to prepare for with violent action (Blazark, 2001). Violent action may occur through a process of split deligimization, where a conflict exists with the group and a certain community that they deem to be inferior (Sprinzak, 1995). Groups that are extremely racist may be different than those that have a different ideological driving force, where the ideology may have a great influence on the likelihood those members of the group or the group itself will commit acts of violence. While, the ideology of a group may be an important group characteristic influencing the potential for violence, a main influencing force may be the element of time and depend on the characteristics of the time in which the group exists in (DHS, 2009). As discussed in the introduction, groups that existed in the early 1990s may be different than those groups existing in the late 1990s. There is an argument that those groups that exist now may be more similar to 25 groups existing within the early 1990s than they are to groups existing in the late 1990s in terms of the political and economic climate. Related to the notion of time, the amount of time a group is in existence for may also have an impact on the likelihood that members of a group will commit acts of violence. It has been argued that those groups that are in short existence may be more likely to commit acts of violence (Rapoport, 2001). A group may form for the sole purpose of committing an act of violence and then disband, or be forced to disband or change to operating under leaderless resistance (Cronin, 2006). Certainly, for the purpose of this study, religion is the main variable to be examined to look at how its presence or absence in a group may make its members more prone to violence. However, religion may not be an important factor as some have argued and other factors may be the real influencing characteristics in a group that lead to violence. Therefore other characteristics discussed will also be included in the analysis to determine if religion matters and if not what else matters. The following section will better describe the data used for this study and the variables that are included. 26 Chapter 3 Methodology This study seeks to identify if there is a relationship between religion and violence and whether or not religion has an impact on organizational level violence. Other variables will also be examined to determine if religion is not a good predictor of violence what may be. This will be done through looking at religious affiliation of domestic extremist groups as well as level of religiosity. Violence will be measured in four different manners to best capture the nature of violence. Data The data utilized in this study comes from an existing database of white supremacist groups. This database was compiled by using Southern Poverty Law Center’s yearly bulletins since 1990 to identify hate groups existing within the United States within each year. Lists of groups were gathered for the years between and including 1990 through 2008. This accounted for more than 6,000 groups. In order to parse out the number of groups only those that were around for at least three consecutive years were chosen and the rest were excluded. This was necessary because it was not manageable to code the original 6,000 groups. By including those with at least three years in existence, the groups had some degree of longevity and the potential for more information about their operations and organization. This resulted in roughly 650 groups, which were further parsed by collapsing chapters of large organizations and making them into umbrella organizations. This was done to obtain a more feasible number of groups as well as to prevent duplication of information from large umbrella groups that have chapters throughout the United 27 States. The final product was a list of roughly 250 groups. Several reasons exist for parsing out the original list of groups. Clearly it is not manageable to code the original list of 6,000. Another reason is that it’s helpful to choose those who have some form of longevity, therefore providing a greater time frame and basis for information to exist on that group. The process of parsing out groups resulted in a final total of 278 groups. The groups were then searched in various open sources of both primary and secondary engines. The search engines that were utilized are Lexis-Nexis, Proquest, Yahoo, Google, Copernic, News Library, Westlaw Campus, Google Scholar, Amazon, Google U.S. Government, Federation of American Scientists, Google Video, Center for the Study of Intelligence, Surf Wax, Dogpile, Mamma, Librarians’ Internet Index, Scirus, All the Web, Google News, Google Blog and the Homeland Security Digital Library. Initially the groups were searched by their name. As more information was found on the groups, follow up searches were conducted using key terms that were found. This may have included such searches as member names, locations, years, or specific incidents. This information was compiled into a Microsoft Word document that was given to a coder who read through it in order to properly code it. At this stage, further follow up searches were conducted on the groups utilizing targeted searching. The information was further checked by a second coder to check for consistency in coding and accuracy of information. While multiple searches were conducted, the amount of information found on some of the groups was minimal. A lot of groups may be operating off of the radar, making it difficult to obtain exact information on their specific characteristics. This is the primary limitation of this study, resulting in only a percentage of groups being included within the analysis. The resulting database consisted of 278 groups and hundreds of variables showcasing the information that was found on each group. 28 Measures Many studies have focused on examining particular terrorist events, while few have quantitatively examined terrorist groups (Cordes et al, 1985). This study seeks to examine several characteristics related to extremist groups that have been used by other researchers in examining international terrorist groups, domestic groups and other types of groups. The dependent variable for this study is violence and will be measured in three different ways. Religion will be captured in two different ways including affiliation and religiosity. Several other variables will also be examined along with religion to see if they have an impact on a group’s level of violence. These variables include group scope, size, political process, command structure, monetary support, recruitment, area, activity, prison operations and whether or not a group operates an illegal business. All variables are coded binary with a one or a zero. Dummy variables will be used for any of the other variables that are not dichotomous. This allows for the comparison within variables to a reference group, which is always coded as zero (Lizotte and Zatz, 1986). The use of dummy variables is important when using data that is categorical in nature. Violence Violence is the dependent variable. It is very common in other studies for severity scales to be utilized in measuring violence; however, this was not practical for this data. Most groups that are violent are involved in more than one act of violence; thereby there would have been limited variation if a severity scale was used (Cunradi et al, 2006). Therefore, the dependent variable of violence is measured in three different ways. First, violence is measured by looking at whether or not the group is involved in any type of violent activity. An act of violence is defined as if a 29 group committed any of the following: armed attack, arson, assassination, hostage taking, bombing, hijacking, kidnapping, suicide attacks, CBRNs, WMDs or robbery. T his is simply coded in a dichotomous fashion. Second, a variable will dichotomously measure whether or not a group is involved in any kind of ideological violence. Ideological violence would be any of the previous measures of violence that are committed by the group or its members that was done so in order to further the goals and ideologies of the group. Third, a variable will dichotomously look at whether or not a group advocates violence. Whether or not a group advocates violence is simply whether a group promotes it through its ideology. This will provide a different measure of violence, and cover the aspect of determining if groups that discuss or promote violence are different than those groups that are actually out there committing acts of violence. It is important to look at violence in a variety of fashions in order to see if different types of ways of measuring it may yield different results. The independent variables may showcase factors that may attribute to a group being more violent in one of these manners or make them more likely to promote violence. These three ways of measuring violence are important since violence could be measured in a variety of fashions. Measuring it in different ways may produce different results as well as implications for violence. Religion Religion has been a difficult variable for researchers to determine a proper way to measure it and has been cited as one of the hardest variables to measure while it is frequently ignored in quantitative research (Fox, 2003). It is often put it into similar affiliation related categories (Evans et al, 1995, Cordes et al, 1985, Vernon, 1962). Here, religion, the primary independent 30 variable of concern, will be measured in two separate ways. First, religion will be looked at in terms of the group’s religious affiliation. This was initially captured in the database as coding a group as Jewish, Catholic, Christian/Protestant, Christian Identity, Odinism/Wicca/Norse, Cosmotheism, other cult, World Church of the Creator, other racist faith, Islamic, Atheist/Agnostic, or other. This resulted with the majority being within the categories of Protestant/Christian or Christian Identity or having no religious affiliation. Only 10.9% of groups fell into a category other than those three, and were therefore collapsed into a category labeled other. The final coding of the variable resulted in there being four categories including Protestant/Christian, Christian Identity, other and none. The other category was condensed due to small numbers in most categories. Religious intensity was operationalized into four categories: high, medium, low religiosity or no religion. Most criminal justice studies that attempt to measure religiosity do so by measuring church attendance (Evans et al, 1995; Rohrbaugh & Jessor, 1975), while in sociology religiosity is often a multi faceted variable that is measured in a variety of ways (Evans et al, 1995; King & Hunt, 1972). For the purpose of this study the only feasible way to measure religiosity is to measure it by judging the groups’ religious literature, propaganda, beliefs, services to make a judgment in regards to the intensity perpetuated by the group since this is the only information that is easily ascertainable through open source searching. Groups that are coded as being high in religiosity are those who are guided primarily by religion, have extensive mention of it in literature, church attendance, and heavy religious overtones. For example, the Army of God is coded as having 31 high religiosity because it has a religious name; releases religious literature has religious discussions on its website and promotes religion in their ideology. Those groups that are coded as medium in religiosity are characterized by groups that showcase a religious element but another ideological factor being the primary influence. Aryan Nations is an example of a group with medium religiosity. The group abides by Christian Identity, has church services and has a religious element to it but is primarily focused on promoting a racist agenda. Groups that are coded as low religiosity are those groups who show some mention of a religious affiliation or ideology but do not take it any further and show any heavy religious elements such as literature mentions, church attendance, or religious propaganda. A specific example of a group coded as having low religiosity would be the European American Issues Forum. This group identifies as being a Christian group, but shows no other signs of religiosity. It has a primary focus of individual liberties and promoting white European rights where religion is just one background factor to the group. Finally, a no-religiosity category exists for those groups who are not affiliated with any religion and show no level of religious beliefs. By looking at religion within these two ways it will be able to showcase if the type of religious association has an impact on violence and will furthermore be able to determine if the intensity of religious beliefs or involvement may have an impact of violence. Political Process A group’s level of involvement in the political process is measured through a nominal level variable. This variable is examined in order to see if a group’s involvement in politics may impact its tendency toward violent actions. This variable is coded dichotomously as one involved 32 or zero not involved. Another study examined politics in extremist groups by looking at whether the group had infiltrated any political activist group (Cordes et al, 1985). However, because of limitations that exist with data, a broader perspective is taken by looking as if a group has tried to become involved in the political process by running for office, lobbying or other political related activities. Racism All of the groups that are included within the study have already been identified as hate groups and showcase some level of racism. For the purpose of this study, racism will be looked at as whether or not it is the primary issue of concern for the group. It is examined in this dichotomous fashion since all groups are hate groups, it makes it difficult to judge what makes one group more racist than another. However, through this way results can be seen to examine those that are primarily racist versus those that have another issue that leads them. This is determined by examining a group’s issue founding, current issue, and ideology. For example, a group that is coded as having race as their primary issue is Center Lane Skins. This group has group is against minorities living in the United States and believe in white superiority. On the other hand, the group Power of Prophecy, while still racist, is primarily anti-federal government and believes in promoting many conspiracy theories so they would be coded as not having race as their primary issue. The variable is coded dichotomously with one indicating it is the primary issue for the group and zero indicating that the group has some other issue as their primary issue of concern. 33 Time The dimension of time is an important factor to examine within this study. Time is examined in two ways. Time will be examined in a variable that will showcase the duration of the groups existence. The database captured 1990-2008. Previously discussed in the data section, the minimum existence of a group in this time frame is three years. Time was coded into three dummy variables. The first dummy variable shows groups that have short term existence which is viewed as being in existence for three to six years. The second category is coded as those groups with a medium level of existence being from seven to ten years. Finally, these are compared to those with a long existence of eleven or more years. Another way that time was examined was in terms of era. It will be divided into three different categories. The first category is groups that were founded pre 1995, then groups between 1995 and 2000, and finally groups between 2001 and 2008. The first era was chosen because the heyday of many white supremacist groups has been said to have occurred in the late 1980s and early 1990s. By looking at this time frame it can then be compared to the nature of the groups coming right after it as well as the nature of more recent groups. The second category chosen were those groups that occurred between 1995 and 2000. If they heyday was in the early 1990s, looking at groups that occurred in the late 1990s may show differences based on that time frame. The final category examines groups that were founded between 2001 and 2008. This was chosen as a time frame in order to showcase the groups that formed in and around September 11th as well as those formed after to see if they are significantly different than those groups from 34 the before 1995 or the groups formed in the late 1990s. Command Structure Command structure of a group is another dichotomous variable. It is measured in groups being hierarchical with a single leader or other. For measurement purposes hierarchical with a single leader are coded as one while any other leadership structure was coded as zero. Another study that examined terrorist organizations argues that leadership and group structure is an important variable and examines it in a similar fashion by looking at the type of structure and whether or not it is hierarchical (Cordes et al, 1985). Nearly 80% are controlled in this manner, suggesting hierarchical organizations are the most common structure found in white supremacist groups. Recruitment Other studies that have looked at recruitment among extremist or terrorist organizations typically measure it by examining sources and methods (Cordes et al, 1985). This study captures that in a similar manner by looking at the sources, or ways in which groups recruit. This variable was created by computing a new variable that added thirteen different variables related to recruitment in order to measure the strength of a group’s recruiting methods. Capturing recruitment through an array of variables can better get at the idea of whether or not a group is a heavy recruiter and try to bring in an arrangement of methods that may occur within the process of recruitment. This variable includes if a group recruits in prison, school, newsletters, family/friends, church, internet, personal visits, protests, mail, military, and concerts and if the group has a charismatic leader. While examining a group with a charismatic leader may not on face value appear to be directly related to recruitment strategies it is. A group with a charismatic 35 leader may be better at recruiting and drawing in people to join or affiliate themselves with the group. A factor analysis was run to verify that this was a proper way to scale the variable of recruitment. The new recruitment variable was then defined into those that did not recruit those that did some recruiting (one or two) and those that were heavy recruiters. This was then coded into a dummy variable with the reference category being those groups that do not recruit. Analysis Religion and its potential to influence violence within domestic groups will be measured through logistic regression. Logistic regression is the obvious choice with the measurements for violence, which are all dichotomous in nature. A logistic regression is often the multivariate approach taken when having a nominal dichotomous dependent variable (Press and Wilson, 1978). Running a logistic regression is also highly suggested for this data in order to see if any other variables are having an impact on a group’s tendency toward violence other than religion, and to see if religion matters when taking into account these variables. 36 Chapter 4 Results Results will first be presented by looking at univariate statistics. This will simply describe the distribution of data and the makeup of each of the dependent and independent variables in terms of percentages. Second, bivariate statistics will be examined to look at the relationship between each of the independent variables with each of the dependent variables. Finally, the results section will end with providing outcomes of the binary logistic regressions. Univariate Results Basic descriptive statistics for each variable and how they compared to the three dependent variables are presented in Table 1. First, all of the variables were examined to see how they compared in terms of percentages as well as to the three dependent variables of any violence, ideological violence and whether a group advocates violence. These percentages help to showcase the breakdown of variables that are included within the analysis and can be found in Table 1. Table 1- Univariate Results N Dependent Measures Ideological Violence Yes No Any Violence Yes No Advocates Violence Yes No Percentages 57 221 20.50% 79.50% 63 215 22.70% 77.30% 48 230 Any Violence 17.30% 82.70% 37 Ideological Violence Advocates Violence Table 1- Univariate Results Continued N Independent Measures Religious Affiliation Christian Identity Christianity Other *None Religiosity High Medium Low *None Political Process Yes No Structure Hierarchical- single leader Other Recruitment Heavy Some *None Race Primary Issue Not Primary Issue Years in Existence Short Existence Medium Existence *Long Existence Year Founded pre 1995 1995-2000 *2001-2008 *reference category Percentages Any Violence Ideological Violence Advocates Violence 44 30 49 15 31.90% 21.70% 10.90% 35.50% 40.00% 16.00% 12.00% 32.00% 40.00% 17.80% 11.10% 31.10% 47.70% 13.60% 9.20% 29.50% 49 24 16 49 35.50% 17.40% 11.60% 35.50% 24.00% 28.00% 16.00% 32.00% 24.40% 26.70% 17.80% 31.10% 25.00% 29.50% 15.90% 29.60% 39 89 30.50% 69.50% 32.70% 67.30% 36.40% 63.60% 72.70% 27.30% 102 29 77.90% 22.10% 72.00% 28.00% 71.10% 28.90% 72.70% 27.30% 39 67 172 14.00% 24.10% 61.90% 36.50% 27.00% 36.00% 26.80% 28.10% 45.10% 39.60% 35.40% 25.00% 93 48 66.00% 34.00% 68.60% 31.40% 66.70% 33.30% 71.70% 28.30% 97 56 125 34.90% 20.10% 55.00% 12.70% 15.90% 71.40% 10.50% 14.00% 75.50% 14.60% 14.60% 70.80% 134 79 65 51.80% 28.00% 21.20% 73.00% 12.70% 14.30% 75.40% 10.50% 14.10% 72.90% 12.50% 14.60% Groups that were found to have been involved in at least one act of violence within this study constitute 22.7 percent of all of the groups, with the remainder 77.3 percent appearing to be non-violent. Types of violence typically appeared in two forms, some being group level and 38 some being at the individual level. For example, members of the Imperial Klans of America beat up a young man they believed to be an illegal immigrant in 2006 at a county fair (SPLC, 2011). Members attacked and brutally beat him because of their misguided belief that he was a Latino. The percentages get somewhat smaller when further looking at those with ideological violence and those groups who advocate violence. The majority of groups that were initially found to have any act of violence were also found to have committed ideological violence, constituting for 20.5 percent of all groups. Furthermore, 17.3 percent of groups advocated violence. The first independent variable presented in Table 1 is a group’s religious affiliation. Almost 32 percent of the groups could be defined as Christian Identity groups, 22 percent were Christian/Protestant, and 11 percent were affiliated with some other religion. It is actually interesting that the highest percentage of groups had no clear religious affiliation. There does not appear to be much variation in group religious affiliation when examining the different dependent variables. For example, 40 percent of the groups that committed any act of violence were Christian Identity, 16 percent were Christian/Protestant, 12 percent affiliated with some other religion, and 32 percent were not religious. The results from groups that commit ideological violence are similar. Forty percent of the groups that committed ideological violence were Christian Identity, 18 percent were Christian/Protestant, 11 percent were affiliated with other religions, and 31 percent had no religion affiliation. A somewhat higher percentage of groups that advocated violence were Christian Identity. Specifically, 48 percent of the groups that committed ideological violence were Christian Identity. Fourteen percent were Christian/Protestant, 9 percent were affiliated with other religions, and 30 percent were not affiliation with a religion. 39 The second variable showcased in Table 1 is religiosity. Slightly more than one third of groups are highly religious, the other slightly or somewhat religious (low or medium religiosity), with the last third being not religious at all. Measures of religiosity showcase 35.5 percent of groups as being highly religious, 17.4 percent having medium religiosity, 11.6 percent having low religiosity and 35.5 percent not being religious at all. Groups with no religiosity were those that were more likely to commit violent acts with 32 percent committing any act of violence, 31.1 percent being involved in ideological violence and 29.6 percent advocating violence. The next greatest instance of violence comes in those that are labeled as having medium religiosity where 28 percent commit any violence, 26.7 percent commit ideological violence and 29.5 percent advocate violence. Of the groups with high religiosity, 24 percent were coded as committing any violence, 24 percent committing ideological violence and 25 percent advocating violence. Finally, those groups have low religiosity are the least violent where 16 percent commit any violence, 17.8 percent commit ideological violence and 15.9 percent advocate violence. Roughly one in every three groups examined (30.5%), is involved in or attempts to be involved in the political process. The groups that are involved commit less violence than those that are not involved but are more likely to advocate violence, with 72.7 percent of those groups politically active advocating violence. Structure of a group was examined by looking at groups who are hierarchical with a single leader versus those that are not. Most groups (77.9%) are hierarchical with a single leader and are more likely to be involved in violence as well as to advocate it. Of these groups, 72 percent have committed any act of violence, 71.1 percent have committed ideological violence and 72.2 percent advocate violence. 40 As show in Table 1, recruitment was measured by those with no recruitment strategy, some recruitment and heavy recruitment. No recruitment was the norm within 61.9 percent of groups, whereas 24.1 percent had some recruitment and 14 percent were involved with heavy recruitment. The groups with no recruitment were more likely to be violent but less likely to advocate violent, with 36 percent being involved in any act of violence, 45.1 percent being involved in ideological violence and 25 percent advocating violence. Since all groups included within this study are hate groups by nature, race was examined strictly by whether or not it is the primary issue or driving force within a group. It was found to be the primary issue in 66 percent of the groups, while 34 percent had another issue as their primary guiding force. Of groups involved in any violence 68.6 percent were guided primarily by race, similar to the 66.7 percent in ideological violence. Slightly more, 71.7 percent of groups advocating violence had race as their primary issue. Time was looked at through two different variables. Fifty-five percent of the groups had a long existence (11 + years), 20.1 had a medium existence (7-10 years), and 34.9 percent had a short existence (3-6 years). Groups having a long existence comprised 71.4 percent of the any act of violence category, 75.5 percent of ideological violence and 70.8 percent of groups advocating violence. Finally, time was also measured in eras by looking at those groups forming before 1995, those forming between 1995 and 2000 and those forming after 2001. Results indicate that 51.8 percent were formed before 1995, 28 percent were formed between 1995 and 2000 and 21.2 percent were formed between 2001 and 2008. Those groups that were formed before 1995 comprised 73 percent of the any violence category, 75.4 percent of ideological violence and 72.9 percent of groups advocating violence. 41 Bivariate Results Religious affiliation and religiosity are the first two independent variables in Table 2 in Appendix B showcasing the bivariate results. Groups affiliated with Christian Identity were found to be more likely to advocate for violence as compared to other religious groups as well as those groups that have no religious affiliation. Christianity, other religions and no religion were not found to be statistically significant for any violence, ideological violence or whether a group advocates violence. Groups that have high or medium religiosity were found to be more likely to commit any act of violence, ideological violence or advocate violence. This result makes the indication that those groups that have higher religiosity (medium or high) are more likely to be involved in all three measures of violence and to be more violent groups overall than those that have low religiosity or no religiosity. Low religiosity was not found to be significant, yet no religiosity was found to be significant in terms of any violence and of whether a group advocates violence. Heavy recruitment was found to be correlated with all three independent variables and to be significant at the .001 level, indicating that groups are more likely to advocate for or commit violent acts if they are heavily involved in recruiting. A similar relationship was found for groups that do no recruiting, with significance found within all three dependent variables. Some recruiting was found to correlate with groups who advocate violence. Results indicate that groups on both extremes, those who recruit heavily as well as those who do not recruit at all are more likely to be violent that those who participate in some or little recruiting. 42 Table 2-Bivariate Results Any Violence Ideological Violence Advocates Violence 0.121 -0.067 0.076 0.52 0.233** -0.134 0.059 -0.085 -0.161* 0.17** 0.134 0.092 0.093 -0.118 -0.15* 0.219*** 0.092 .113* -0.05 -0.088 0.334*** 0.047 -0.280*** 0.01 0.336*** 0.121** -0.347*** 0.085 -0.26*** -0.077 0.311*** -0.195*** -0.063 0.238*** 0.277*** -0.201*** -0.112* 0.226*** -0.161*** -0.095 Independent Measures Religious Affiliation Christian Identity 0.131 Christianity -0.105 Other 0.099 *None -0.055 Religiosity High -0.181** Medium 0.211** Low 0.104 *None 0.11* Political Process 0.037 Structure -0.111 Recruitment Heavy 0.35*** Some 0.036 *None -0.283*** Race 0.042 Years in Existence Short Existence -0.252*** Medium Existence -0.058 *Long Existence 0.288*** Year Founded pre 1995 0.269*** 1995-2000 -0.189*** *2001-2008 -0.116* *correlation significant at the .10 level **correlation significant at the .05 level ***correlation significant at the .001 Time and violence also appear to be related. A significant relationship was found between the three measures of violence and groups that are around for a short existence as well as groups around for a long existence that can be seen in Table 2. Groups that exist for six years 43 or less as well as groups that exist eleven years or more are more likely to commit any act of violence, ideological violence or to advocate for violence than those groups that exist for a medium amount of time of between seven and ten years. Both categories of groups existing prior to 2001, those prior to 1995 and those between 1995 and 2000, were found to be significantly more likely to be involved in violence than those formed in 2001 or later. Groups existing before 1995 as well as groups existing between 1995 and 2000 had significant correlations to being involved in any violence, ideological violence and advocating violence. However, some significance was found with groups existing in or following 2001 for any violence or ideological violence, but not as strong as for the prior years. There was no significant relationship between several independent measures and the dependent variables. No support was found for a correlation between group structure, political process, or race between any of the three independent variables. Multivariate Results Three logistic regression equations were run with these data, one for each of the dependent variables. Tables 3, 4 and 5 presents the results for the any violence, ideological violence, and advocates violence variables. Results for involvement in ideological violence are provided within Table 3.These results show that most of the variables had no influence on ideological violence. Specifically, religious affiliation, religiosity, race, recruiting, political process, and structure were not found to be significant. However, a group having a short existence was found to influence ideological violence as well as a group that was founded between 1995 and 2000. 44 Table 3- Ideological Violence Regression B S.E. Exp(B) 0.668*** 0.194 0.513 Constant Religious Affiliation Christian Identity Christianity None Other 0.932 0.872 0.123 0.912 -0.445 0.896 (reference category) 2.539 1.131 0.641 Religiosity High Medium Low None -1.151 0.053 0.739 0.867 0.316 1.055 (reference category) Primary Issue Not Primary Issue 0.577 0.514 (reference category) 1.781 Heavy Some None 0.931 0.667 -0.319 0.63 (reference category) 2.537 0.727 Involved Not Involved 0.692 0.539 (reference category) 1.998 Heirarcical with single leader Other -0.691 0.589 (reference category) 0.501 Short** Medium Long -3.28** 1.209 -0.863 0.766 (reference category) 0.038 0.422 Pre-1995 -1.264 0.939 Between 1995-2000** -2.398** 1.178 After 2001 (reference category) Log Likelihood= 116.101, Cox & Snell R Square= .257 *sig <.10 **sig<.05 0.283 0.091 Race Recruiting Political Process Structure Existence Founded 45 Involvement in any act of violence is portrayed in Table 4. Results are relatively similar to those for ideological violence indicating that most of the variables have no influence. No relationship was found with religious affiliation, religiosity, race, political process or structure. However, some significance was found within recruiting and group existence. Those groups that recruit heavily are more likely to commit an act of violence as compared to those who recruit some and those who do not recruit at all. Furthermore, groups who have a short existence were also found to be more likely to be involved in any act of violence. Table 5 represents the results for groups that advocate violence. Like with the previous measurements of violence, a lot of variables were not found to be significant, including race, political process, existence of when a group was founded. However, groups that are affiliated with Christian Identity are more likely to advocate violence than are other religious affiliations or no religious affiliation. Those groups that have high religiosity are more likely than those with medium, low or no religiosity to advocate for violence. Heavy recruiting was also found to be significant, indicating that those groups that are more heavily involved in recruiting are more likely to advocate for violence than those that are involved in some recruiting or no recruiting. Further, groups that are hierarchical and have a single leader were found to be more likely to advocate for acts of violence than those groups who fall under another category of organizational structure. 46 Table 4- Any Violence Regression B 0.484** Constant Religious Affiliation Christian Identity Christianity None Other S.E. Exp(B) 0.19 0.735 0.809 -0.259 0.847 -0.272 0.848 (reference category) 0.616 2.085 0.772 0.762 Religiosity High Medium Low None -1.068 0.501 0.722 0.835 0.344 1.65 (reference category) Primary Issue Not Primary Issue 0.534 0.487 (reference category) 1.706 Heavy Some None 1.055* 0.631 -0.333 0.57 (reference category) 2.871 0.717 Involved Not Involved 0.252 0.505 (reference category) 1.286 Hierarchical with single leader Other -0.566 0.556 (reference category) 0.568 Short Medium Long 1.973** 0.942 -0.656 0.694 (reference category) Race Recruiting Political Process Structure Existence 0.139 0.519 Founded Pre-1995 -0.635 0.851 Between 1995-2000 -1.012 0.94 After 2001 (reference category) Log Likelihood= 127.852, Cox & Snell R Square= .218 *sig <.10 **sig<.05 47 0.53 0.363 Table 5- Advocates Violence Regression B S.E. Exp(B) 0.593** 0.192 0.553 Constant Religious Affiliation Christian Identity Christianity None Other 1.977** 0.865 0.289 0.897 0.304 0.852 (reference category) 7.224 1.335 1.355 Religiosity High* Medium Low None -1.458 0.285 0.778 0.853 0.233 1.33 (reference category) Primary Issue Not Primary Issue 0.404 0.499 (reference category) 1.498 Heavy Some None 1.095* 0.645 0.015 0.565 (reference category) 2.99 1.016 Involved Not Involved -0.427 0.527 (reference category) 0.652 Heirarcical with single leader Other -0.993* 0.563 (reference category) 0.371 Short Medium Long -0.826 0.88 -0.072 0.662 (reference category) 0.438 0.93 0.393 0.831 0.429 0.872 (reference category) 1.481 1.536 Race Recruiting Political Process Structure Existence Founded Pre-1995 Between 1995-2000 After 2001 Log Likelihood= 125.713, Cox & Snell R Square= .211 *sig <.10 **sig<.05 48 Race Results All of the groups included in the database have been designated as hate groups and have at least some element of racism within them. It was important, however, to examine whether groups whose primarily ideological focus was race-based were different from those where this issue was secondary. These results are presented in Table 6. Only bivariate results are presented because the multivariate model had a small number of cases and tests indicated problems with multicollinearity. Bivariate results show many differences between those groups that have race as their primary issue versus those that have a different primary issue. While looking at religious affiliation, no significant relationships were found between any religious affiliation and that of having another primary issue other than race. However, groups that affiliated with Christian Identity or with no religion at all were found to be more likely to advocate violence as well as more likely to commit ideological violence or any act of violence. Groups with another non-race issue driving them and high religiosity and medium religiosity were found to be statistically more likely to be involved in any act of violence while no relationship was found with the race driven groups that have high religiosity. Groups that have medium religiosity and that are primarily driven by race were found to be more likely to advocate for violence. Further, those groups that have no religiosity and are driven by race are found to be more likely to commit ideological violence or any act of violence. Examining the recruitment variable, no relationships were found with those groups that are primarily driven by race. However, the story is much different when examining the groups that are driven by another factor. Those groups that are heavy recruiters were found to be statistically more likely to advocate for violence and to commit ideological violence or any act of violence. 49 A relationship was found within the two variables measuring the element of time for both types of groups. Groups having a short existence were found to be more likely to commit ideological violence or any act of violence if they are not driven by race. On the other hand, those groups driven by race are more likely to commit ideological violence or any act of violence if they have been around for a medium or a long existence. Further, when examining the year a group was founded groups driven by race that were founded prior to 1995 were found to be more likely to commit ideological violence or any violence. However, those groups not driven by race are also likely to commit ideological violence if founded prior to 1995. Finally, groups that are primarily racist and founded between 1995 and 2000 are more likely to commit any act of violence but those groups that are not primarily racist and founded in this time are more likely to commit ideological violence. 50 Table 6- Race Bivariate Results Race as Primary Issue Any Ideological Advocates Violence Violence Violence Any Violence Other Issue Ideological Violence Advocates Violence Independent Measures Religious Affiliation Christian Identity .012** .005** .000*** .362 .551 .450 Christianity .850 1.00 .117 .385 .655 .510 Other .745 .576 .156 .872 .584 .755 *None .027** .050** .050** .895 .693 .793 Religiosity High .570 1.00 1.00 .071* .239 .136 Medium .333 .257 .034** .052* .204 .108 Low .248 .176 .616 .590 .397 .490 *None .031** .049** .113 .748 .880 .871 Political Process .846 .921 .220 .462 .194 .985 Structure .262 .181 .181 .983 .858 .440 Recruitment Heavy .121 .275 .885 .001*** .000*** .002** Some .102 .162 .708 .154 .313 .127 *None .827 .676 .581 .135 .026** .224 Race Years in Existence Short Existence .444 .527 .723 .088* .014** .341 Medium Existence .029** .037** .320 .937 .540 .694 *Long Existence .018** .031** .286 .165 .015** .284 Year Founded pre 1995 .037** .063* .161 .174 .044** .288 1995-2000 .082* .108 .179 .136 .033** .452 *2001-2008 .444 .527 .723 .874 .752 .563 *correlation significant at the .10 level, **correlation significant at the .05 level, ***correlation significant at the .001 51 Year Founded The time frame a group was formed was also examined to see how it could relate to members’ violent tendencies. This was done by using the variable that was created on when a group was founded. Bivariate correlations were ran with each of the three dependent variables after splitting the file into groups founded prior to 1995, then groups founded between 1995 and 2000 and finally groups founded 2001 or after. An attempt was made to run a binary logistic regression for all of them, however, was not possible due to issues with multicollinearity. When examining the variable of religious affiliation, Christian Identity groups founded prior to 1995 as well as those that were founded 2001 or later were both found to be more likely to advocate for violence than those groups founded between 1995 and 2000. Groups that are labeled as being Christian/Protestant and were formed prior to 1995 were also found to be more likely to advocate for violence than those groups founded later. Yet, between 1995 and 2000 Christian/Protestant groups were more likely to commit acts of ideological violence. This disappeared with Christian/Protestant groups founded 2001 or later with no relationships being found. Finally, those groups categorized as having an “other” religion that were founded prior to 1995 were found to be more likely to commit any act of violence compared to those founded later. 52 Table 7- Year Founded Bivariate Results Founded Prior to 1995 Founded Between 1995-2000 Founded Between 2001-2005 Any Ideological Advocates Any Ideological Advocates Any Ideological Advocates Violence Violence Violence Violence Violence Violence Violence Violence Violence Independent Measures Religious Affiliation Christian Identity .409 .474 .030** .397 .566 .332 .239 .546 Christianity .215 .309 .080* .323 .020** .550 .207 .253 Other .077* .210 .774 .444 .514 .285 *None .622 .825 .775 .650 .254 .973 .784 .859 Religiosity High .112 .088* .034** .545 .758 .3113 .081* .116 Medium .100* .210 .029** .526 .144 .332 .048** .211 Low .292 .210 .334 .566 .697 .512 *None .008** .001*** .010** .346 .312 .157 .049** .012** Political Process .597 .403 .777 .849 .520 .630 .935 .865 Structure .126 .234 .126 .223 .416 .783 .149 .071* Recruitment Heavy .000*** .001*** .003** .318 .182 .001*** .002** .001*** Some .580 .864 .353 .826 .918 .203 .548 .813 *None .008** .010** .001*** .729 .413 .003** .012** .055* Race .846 .856 .661 .437 .931 .289 .515 .749 Years in Existence Short Existence .067* .089* .535 .687 .579 .160 .007** .002** Medium Existence .909 .370 .827 .391 .762 .762 .162 .100* *Long Existence .231 .057* .791 .093* .252 .021** Year Founded pre 1995 1995-2000 *2001-2008 *correlation significant at the .10 level, **correlation significant at the .05 level, ***correlation significant at the .001 53 .048** .306 .256 .160 .005** .556 .562 .865 .000*** .933 .023** .749 .151 .847 - For those groups characterized as having high religiosity that were founded prior to 1995 they were found to be significantly more likely to commit acts of ideological violence or advocate for violence. These findings disappear when examining high religiosity among groups founded between 1995 and 2000, however, groups with high religiosity founded 2001 or onward are then more likely to commit any act of violence compared to their religious or non religious counterparts. Groups having a medium level of religiosity and that are founded prior to 1995 are more likely to advocate for violence, this disappears with groups founded between 1995 and 2000, but then reappears with those founded 2001 or onward and then includes a relationship with any act of violence. Those groups with no religiosity are significantly more likely to advocate for violence, commit any act of violence or any act of ideological violence if founded prior to 1995. Again, like with other measures of religiosity the significance disappears with those groups founded between 1995 and 2000. However, it then again reappears with significance being found with those groups that have no religiosity being more likely to commit ideological violence or any act of violence if founded in 2001 or later. Groups that are heavy recruiters are significantly more likely to advocate for violence regardless of the year they are formed. Those formed prior to 1995 or those formed 2001 or later are also more likely to commit ideological violence or any act of violence. No significance, regardless of year, was found with groups that are involved in some recruiting any a form of violence. The exact same results were found with those groups that are involved in no recruiting practices, with groups advocating violence being significant over any year category they were founded, and those founded pre 1995 and those founded 2001 or later being more likely to commit any act of violence or ideological violence. 54 Those groups that are categorized as having a short existence were found to be significantly more likely to commit ideological violence or any act of violence if they were founded prior to 1995 or if they were founded 2001 or later. The only significant relationship found for medium existence was found that those founded after 2001 were more likely to commit ideological violence. Finally, looking at groups with long existence significance was found with those that were founded between 1995 and 2000 to be more likely to advocate for violence. Groups that have been founded between 2001 and 2005 were not measured for long existence, since they could not fit into that category. Long existence would include those existing 11 years or more. 55 Chapter 5 Discussion/Conclusion Initially, this study sought to examine the influence of religion on domestic extremist groups in whether or not they are committing acts of violence. Previously, there has been disagreement about whether religion matters and if it does, to what degree. The results indicate that religion is not a primary driving force making domestic extremist groups more violent. This was important to discover since previously, there has been suggestion that it may be a key predictor. Other findings, such as the influence of time and recruitment can provide a great basis for future studies as well as bring forth elements that may really be better predictors of violence within these groups. Several variables seemed to impact a group’s involvement in violence. Such variables include those related to the existence and time frame of a group, structure, and religious variables. The results from the analysis initially indicate through bivariate statistics that some aspects of religion may matter, as well as recruitment, years in existence and the year the group was founded. Similar findings can be viewed through the multivariate analysis of logistic regression; however, findings differ based on the way that violence is being measured. Overall, findings indicate that some factors exist that may be able to predict what factors or characteristics may make an extremist group violent or advocate violence. Minimal significance was found in terms of religion, however, some was found. Christian Identity groups were found to be more likely to advocate for violence. There are many reasons that can help explain why this might be the case. Religion may be a key influence or motivating factor behind acts of violence or terrorism (Aho, 1999). It may be used to legitimize violence, or promote hate and discrimination (Fox, 2004; Gallagher, 1997) or by reinterpreting the bible to fit 56 their goals or promotion of violence (Hoffman, 1997). One example of how it may be used to justify violence would be to view minorities as not being human or view them as deserving violent treatment (Al-Khattar, 2003). It may also provide a group an identity and enforce an “us vs. them” mentality (Silberman et al, 2005). Christian Identity was at the forefront of the radical right during the 1990s (Juergensmeyer. 2000). While Christian Identity groups were not found to be more likely to be involved in violence or ideological violence, they are more likely to advocate violence, indicating that these groups are often more talk than they are action. On the bivariate level both medium and high religiosity was found to be related to all three measurements of violence. However, through multivariate analysis significance was only found between high religiosity and groups advocating violence, indicating that those groups that are characterized as being highly religious are more likely to advocate for violence than those that are medium or low in religiosity or have no religiosity. Previously, religiosity has been found to be inversely related to acts of violence (Benda and Toombs, 2000). However, advocating violence may be dependent upon factors such as the situational threat. Groups may be more likely to advocate violence toward interracial marriage but less likely so against job competition posed by other races (Glaser et al, 2002). It is highly dependent upon the climate and circumstances surrounding a group. Related to the previous section of Christian Identity, the more stringent or higher religiosity within a group may make justifying violent actions easier. Race was not found to be significant in the initial results. However, significance was found when running bivariate statistics of comparing the characteristics of groups that use race as a primary issue and those who do not to the three dependent measures of violence. Race is often the ideological driving force for hate groups in the United States, with many groups believing that someday there will be a race war in America (Blazark, 2001). The ideology or driving force 57 is an important factor in examining the characteristics of groups (Cordes et al, 1985). While all of the groups examined are hate groups, differences were found between those that had racism as their primary issue. Looking at the aspect of religion, Christian Identity groups were significantly more likely to commit ideological violence, commit any violence or to advocate for violence. The groups with race as a primary issue and that had a medium or long existence were also significantly more likely to commit ideological violence or any violence. The belief in a race war and ideological values of those groups that are driven by race may be keeping groups together for longer periods of time and driving the violence factor. Other groups may form for a shorter period of time to achieve a specific goal and then disband. Heavy recruitment was found to be significant at the bivariate and multivariate levels. Those groups that recruit heavily are more likely to advocate violence as well as to be involved in any act of violence. Recruitment in hate groups has changed dramatically over the past few decades and is often compared to recruitment into gangs (Blazak, 2001). The Internet has impacted this change with the creation of many more hate oriented websites and has created an avenue to spread literature and material to those it may not have been able to reach before (Adams and Roscigno, 2005). Ideological motivations are also related to recruitment, and individuals that are being recruited may be more violent prone or intense in their beliefs, especially young people who may be easily influenced (Sharpe, 2000). The factor of time was examined in two different ways. Groups having a short existence or those groups founded between 1995 and 2000 were found to be more likely to commit ideological violence or any violence. Many reasons exist as to why a group may have a short existence including being involved in counterproductive strategies, imploding, or making a bad decision such as committing a violent action (Cronin, 2006). Modern day terrorist groups often 58 have a short existence, where 90% of them disappear within a year of formation and of the rest more than half of those are gone within the following decade (Rapoport, 2001). Right wing groups are predicted to exist within a cyclical life span, where groups have violence or hostility toward an enemy, often defined by a race, ethnicity, religion, sexual preference, and will ebb and flow over time (Sprinzak, 1999). Many reasons exist as to why a terrorist group, or other hate group for that matter, may cease to exist including the failure of its implementation into a new generation, the death or arrest of a leader, a move into legitimate political channels, a failure to achieve goals, repression or a failure to gain support (Cronin, 2006). Violence may also cause instability within a group. Groups may go underground or change their name or structure after an act of violence has occurred. After the Oklahoma City Bombing in 1995, when more scrutiny was put on right wing extremists, many groups began to operate under the façade of leaderless resistance (Cronin, 2006). This can be more easily understood the bivariate year results. Time was also examined by splitting the data and examining the differences between groups that were founded prior to 1995, those founded between 1995 and 2000 and those founded 2001 or later. These results may help explain the previous discussion of current society mirroring the times of the early 1990s. Many of the correlations found in groups founded prior to 1995 were similar to those that were founded 2001 or later, where few relationships were found with those founded between 1995 and 2000. However, despite groups going underground during this time other societal characteristics existed that may have promoted acts of violence by right wing extremists between 1995 and 2000. One such reason for increased violence among groups founded during this time may be related to ends time hysteria caused by Y2K, with groups justifying the use of violence since they thought the end of the world was coming near (FBI, 1999). The 1990s political 59 climate, outsourcing of jobs, gun rights, economic downturn and perceived threat of government power are also viewed as being factors that may have influenced the likelihood of violence among domestic extremists (DHS, 2009). The results provide important findings that help to illuminate right wing domestic extremist groups in the United States between 1990 and 2008. However, the study is not without its limitations. First, one major drawback of the study comes in terms of missing data. The study resulted in roughly 280 groups, however; roughly half of them were eliminated in data analysis due to missing data. This shortcoming may be the greatest of the study, yet the data can still be extremely useful in examining a topic area that has been neglected in literature. Second, data was further limited by only being able to use information that was found through open source searching. Other methods of searching may yield more information and as a direct result yield less missing data. Third, this study looks at only group based behavior but has to ignore radicalization and lone wolf behavior which may be an influencing part of the far right movement in the United States. Behavior included may be a group action but also include individual members of a group taking an action on behalf of the group. Lone wolf behavior that is not included would be behavior done by an individual that has a far right ideology but that is not associated with any extremist group. However, such behavior is beyond the scope of this study that strictly seeks to illuminate group level behavior. Fourth, a further limitation of this study would be in determining its relevance to present society. Some may argue that there are drastic differences between the groups that existed in the 1980s or 1990s and those groups that are being formed more recently or exist today and the 60 overall changing nature of the groups (Barkun 1989). However, there may be a resurgence of groups, with many of the political and economic elements of the 1990s mirroring the current day. Fifth, the dimension of time may have limited this study in the results that are supplied. Groups that existed less than three years were eliminated as well as those that fall out of the time frame between 1990 and 2008. Different characteristics may exist between those groups that existed for less than three years than those that were included within this study. Despite this, it was not practical, for the purposes of this study, to be able to include all groups that were initially identified due to the grand scope and potential for limited data on many of these small fringe groups. A final limitation for this study may be the use of the .1 significance level. Most social science research uses the .05 level; however, due to limitations of the data the .10 level is also included here but less strength is given to findings at this level. While the purpose of the study sought to identify the relationship between and violence, it did find this relationship as well as other important results. Religion is not the issue, but rather religious extremism that leads to violence (Pratt, 2010). Antidotal evidence exists that religion has been used as an excuse and justification for violence. This is showcased through the logistic regression results looking at groups that advocate violence. This study provides information on a subject that has previously received little examination and illuminates the topic of domestic extremism, which has recently been overshadowed by the international threat and lone wolf terrorism (Ashby and Brinsfield, 2004). It also is able to provide information pertaining to groups, where as previously the focus has been on incidents or actions committed by groups. It provides a basis for future studies to examine characteristics related to and influencing extremist organizations. 61 Future studies need to occur that are able to more thoroughly examine the dynamics of domestic right wing extremist groups. These studies could greatly contribute to the literature on extremist groups with attempting to eliminate limitations of this current study. Such studies are important as times change and these hate groups may be taking on different characteristics with the different political and economic climates that exist within the United States. 62 Bibliography 63 Bibliography Adams, J., & Roscigno, V. (2005). White supremacists, oppositional culture and the world wide web. Social Forces, 84(2), 759-778. Aho, J. A. (1990). The politics of righteousness: Idaho Christian patriotism. Seattle: University of Washington Press. Appleby, R. S. (2000). The ambivalence of the sacred: religion, violence, and reconciliation. Lanham, Md: Rowman & Littlefield Publishers. Barkun, M. (1989). Millenarian aspects of white supremacist movements. Terrorism and Political Violence, 1, 409-434. Bellia, P. (2005). The "lone wolf" amendment and the future of foreign intelligence surveillance law. Villanova Law Review, 50, 425-478. of Homeland Security. (2009). Rightwing extremism: current economic and political climate fueling resurgence in radicalization and recruitment. DHS Assessment, 1, 1-10. Benda, B., & Toombs, N. (2000). Religiosity and violence are they related after considering the strongest predictors?. Journal of Criminal Justice, 28, 483-496. Blazak, R. (2001). White boys to terrorist men: Target recruitment of Nazi skinheads. American Behavioral Scientist, 44, 982-1000. Bolce, L., & Maio, G. D. (1999). Religious outlook, culture war politics, and antipathy toward Christian fundamentalists. The Public Opinion Quarterly, 63(1), 29-61. Chermak, S., & Frelich, J. (2010). Surveying American state police agencies about lone wolves, far-right criminality, and far-right and Islamic Jihadist criminal collaboration. Studies in Conflict & Terrorism, 33(11), 1019-1041. Cordes, B., Jenkins, B., & Kellen, K. (1985). A conceptual framework for analyzing terrorist groups. Rand Publication Series, 1, 1-131. Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379-399. Crenshaw, M. (2000). The psychology of terrorism: An agenda for the 21st century. Political Psychology, 21(2), 405-420. Cronin, A. (2006). How al-Qaida ends. International Security, 31(1), 7-48. Dobratz, B. (2001). The role of religion in the collective identity of the white racialist movement. Journal 64 for the Scientific Study of Religion, 40(2), 287-301. Evans, D., Cullen, F., Dunaway, G., & Burton, V. (1995). Religion and crime reexamined: The impact of religion, secular controls and social ecology on adult criminality. Criminology, 33, 195-224. FBI Quick Facts. (n.d.). FBI Homepage. Retrieved March 21, 2011, from http://www.fbi.gov/about-us/quick-facts Fiske, S., & Taylor, S. (1991). Social Cognition. New York: McGraw-Hill. (Original work published 1984) Fox, J. (1999). The influence of religious legitimacy on grievance formation by ethno-religious minorities. Journal of Peace Research, 36(3), 289-307. Fox, J. (2003). Religion as an overlooked element of international relations. International Studies Review, 3(3), 53-73. Fox, J. (2004). Counting the causes and dynamics of ethnoreligious violence. Totalitarian Movements and Political Religions, 4(3), 119-144. Gallagher, E. (1997). God and country: Revolution as a religious imperative on the radical right. Terrorism and Political Violence, 9(3), 63-79. Glaser, J., Dixit, J., & Green, D. (2002). Studying hate crime with the internet: What makes racists advocate racial violence?. Journal of Social Issue, 58(1), 177-193. Harper, C., & Beau, B. L. (1993). The social adaptation of marginal religious movements in America. Sociology of Religion, 54(2), 171-192. Hewitt, C. (2000). Patterns of American terrorism 1955-1998: An historical perspective on terrorism-related fatalities. Terrorism and Political Violence, 12(1), 1-14. Hewitt, C. (2003). Understanding terrorism in America: from the Klan to al Qaeda. London: Rutledge. Hoffman, B. (1988). Recent trends and future prospects of terrorism in the United States. RAND Corp, 1, 1-77. Hoffman, B. (1997). The confluence of international and domestic trends in terrorism. Terrorism and Political Violence, 9(2), 1-15. Hoffman, B. (2006). Inside terrorism (Revised and expanded ed.). New York, NY: Columbia University Press. 65 Imperial Klans of America | Southern Poverty Law Center. (n.d.). Southern Poverty Law Center. Retrieved March 22, 2011, from http://www.splcenter.org/get-informed/intelligencefiles/groups/imperial-klans-of-america Jacobson, M., & Royer, H. (2009). Aftershocks: The impact of clinic violence on abortion services. NBER Working Paper, 1, 1-61. Janis, I. L. (1972). Victims of groupthink; a psychological study of foreign-policy decisions and fiascoes. Boston: Houghton, Mifflin. Juergensmeyer, M. (1991). Special issue on violence and the sacred in the modern world. Terrorism and Political Violence, 3(3), 101-117. Juergensmeyer, M. (1998). Christian violence in America. Annals of the American Academy of Political and Social Science, 558, 88-100. Juergensmeyer, M. (2000). Terror in the mind of God: the global rise of religious violence. Berkeley: University of California Press. Kaplan, J. (1995). Right wing violence in North America. Terrorism and Political Violence, 7(1), 44-95. Kaplan, J. (1997). Leaderless resistance. Terrorism and Political Violence, 9(3), 80-95. Kaplan, J., & Bjørgo, T. (1998). Nation and race: the developing Euro-American racist subculture. Boston: Northeastern University Press. Khattar, A. M. (2003). Religion and terrorism an interfaith perspective. Westport, Conn.: Praeger. Kimmel, M., & Ferber, A. (2000). "White men are this nation:" Right-wing militias and the restoration of rural American masculinity. Rural Sociology, 65(4), 582-604. King, M., & Hunt, R. (1972). Measuring the religious variable: Replication. Journal for the Scientific Study of Religion, 11(3), 240-251. Kingshott, B. (2003). Terrorism: The "new" religious war. The Justice Professional, 16(1), 1527. Laqueur, W. (1999). The new terrorism: fanaticism and the arms of mass destruction. New York: Oxford University Press. Law Enforcement Resources | Southern Poverty Law Center. (n.d.). Southern Poverty Law Center. Retrieved March 22, 2011, from http://www.splcenter.org/what-we-do/hate-andextremism/law-enforcement 66 Levine, J., & Moreland, R. (1990). Progress in small group research. Annual Review of Psychology, 41, 585-634. Lizotte, A., & Zatz, M. (1986). The use and abuse of sentence enhancement for firearms offenses in California. Law and Contemporary Problems, 49(1), 199-221. Maiese, M. (n.d.). CRInfo: The Conflict Resolution Information Source - Suicide Bombers. CRInfo: The Conflict Resolution Information Source - . Retrieved March 21, 2011, from http://crinfo.beyondintractability.org/essay/suicide_bombers/?nid=5206 Masters, D. (2008). The origin of terrorist threats; Religious, separatist, or something else?. Terrorism and Political Violence, 20(3), 396-414. Moghaddam, F. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161-169. Newport, F. (n.d.). Nine Years After 9/11, Few See Terrorism as Top U.S. Problem. Gallup.Com - Daily News, Polls, Public Opinion on Government, Politics, Economics, Management. Retrieved December 18, 2010, from http://www.gallup.com/poll/142961/nine-years-few-terrorism-top-problem.aspx Pape, R. (2003). The strategic logic of suicide terrorism. American Political Science Review, 97(3), 343-357. Poland, J. M. (1988). Understanding terrorism: groups, strategies, and responses. Englewood Cliffs, N.J.: Prentice Hall. Potok, M. (n.d.). Rage on the Right | Southern Poverty Law Center. Southern Poverty Law Center. Retrieved March 22, 2011, from http://www.splcenter.org/getinformed/intelligence-report/browse-all-issues/2010/spring/rage-on-the-right Pratt, D. (2010). Religion and terrorism: Christian fundamentalism and extremism. Terrorism and Political Violence, 22(3), 438-456. Press, J., & Wilson, S. (1978). Choosing between logistic regression and discriminate analysis. Journal of the American Statistical Association, 73(364), 699-705. Pynchon, M., & Borum, R. (1999). Assessing threats of targeted group violence: Contributions from social psychology. Behavioral Sciences and the Law, 17, 339-355. Quarles, C. L. (2004). Christian Identity: the Aryan American bloodline religion. Jefferson, N.C.: McFarland. Rapoport, D. (1988). Messianic sanctions for terror. Comparative Politics, 20(2), 195-213. 67 Rapoport, D. (2004). The four waves of modern terrorism. Attacking terrorism: elements of a grand strategy (pp. 46-73). Washington, D.C.: Georgetown University Press. Rohrbaugh, J., & Jessor, R. (1975). Religiosity in youth: A personal control against deviant behavior. Journal of Personality, 43(1), 136-155. Rummel, R. (1997). Is collective violence correlated with social pluralism?. Journal of Peace Research, 34(2), 163-175. Schmid, A. (2004). Frameworks for conceptualizing terrorism. Terrorism and Political Violence, 16(2), 197-221. Schwartz, S., Dunkel, C., & Waterman, A. (2009). Terrorism: An identity theory perspective. Studies in Conflict & Terrorism, 32(6), 537-559. Sedgwick, M. (2004). Al-Qaeda and the nature of religious terrorism. Terrorism and Political Violence, 16(4), 795-814. Seegmiller, B. (2007). Radicalized margins: Eric Rudolph and religious violence. Terrorism and Political Violence, 19, 511-528. September 11 by Numbers. (n.d.). New York Magazine -- NYC Guide to Restaurants, Fashion, Nightlife, Shopping, Politics, Movies. Retrieved March 22, 2011, from http://nymag.com/news/articles/wtc/1year/numbers.htm Setton, K. M. (1989). A history of the Crusades ([2nd ed.). Madison: University of Wisconsin Press. Sharpe, T. (2000). The identity Christian movement: Ideology of domestic terrorism. Journal of Black Studies, 30(4), 604-623. Silberman, I., Higgins, E. T., & Dweck, C. (2005). Religion and world change: Violence and terrorism versus peace. Journal of Social Issues, 61(4), 761-784. Silke, A. (2008). Research on terrorism: A review of the impact of 9/11and the global war on terrorism . Integrated Series in Information Systems, 18(1), 27-50. Sprinzak, E. H. (1995). Right-wing terrorism in a comparative perspective: The case of split delegitimization. Terrorism and Political Violence, 7(1), 17-43. Stein, B., Elliott, M., Jaycox, L., Collins, R., Berry, S., Klein, D., et al. (2004). A national longitudinal study `of the psychological consequences of the September 11, 2001 terrorist attacks: reactions, impairment and help-seeking. Psychiatry, 67(2), 105-117. 68 Stern, J. (1999). The prospect of domestic bioterrorism. Emerging Infectious Disaeses, 5(4), 517522. Stern, J. (2003). The protean enemy. foreign affairs, 82(4), 27-40. Suttmoeller, M. , Chermak, S. M. and Freilich, J. D. , 2010-11-17 "An Examination of Law Enforcement Officer Homicides committed by Right-Wing Extremists" Paper presented at the annual meeting of the ASC Annual Meeting, San Francisco Marriott, San Francisco, California . 2011-03-15 from http://www.allacademic.com/meta/p431535_index.html Vernon, G. (1962). Measuring religion: Two methods compared. Review of Religious Research, 3(4), 159-165. Waugh, W. (1983). The values in violence: Organizational and political objectives of terrorist groups. Conflict Quarterly, 3(4), 5-19. White, R. (1989). From peaceful protest to guerilla war: Micromobilization of the provisional Irish Republican Army. The American Journal of Sociology, 94(6), 1277-1302. 69