Risk Assessment Decisions for Violent Political Extremism 2009-02

Risk Assessment Decisions for Violent Political Extremism 2009-02 - PDF Version (482 KB)

D. Elaine Pressman

Author's Note:

The author wishes to thank the individuals who were instrumental in bringing this work to fruition. R. Karl Hanson, Ph.D. provided the encouragement and support which ultimately led to this work being undertaken. His pertinent suggestions for improvement enhanced the final report. It was Dr. Hanson's international reputation in the field of risk assessment that led me to his office at Public Safety Canada in Ottawa in 2008. A discussion with him was the catalyst for a conceptual shift in my thinking on the challenging topic of violence prediction.

Thanks are also due to Sean Norton, an expert in radicalization at Defence Research and Development Canada (DRDC), who was an early sounding board for the concept. He provided many constructive comments. Dr. Rogelio Alonso, an expert in violent extremism at Universidad Rey Juan Carlos in Madrid, devoted time during a busy schedule in The Hague to review the items included in the protocol for risk assessment of violent extremists (VERA). His suggestions resulted in some modifications to the tool. The author wishes to acknowledge the contribution of Wayne L. Hanniman of the National Security and Criminal Investigations Division of the RCMP. This present work would not have materialized without his interest in earlier work by the author. His involvement in a related project supported by the RCMP led to this current work.

This project was funded by Public Safety Canada. Points of view, information and the tools included in this document are the sole responsibility of the author and do not necessarily represent the official position, policy or viewpoint of Public Safety Canada or the Government of Canada. The support and involvement of the individuals mentioned above cannot be interpreted as an endorsement of the ideas contained in this document. All errors are the sole responsibility of the author.

Correspondence concerning this article should be addressed to D. Elaine Pressman, Ph.D., Senior Research Fellow, Canadian Centre for Security and Intelligence Studies, Paterson School of International Affairs, 1404 Dunton Tower, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario, Canada K1S 5B6 or by email at epressman@rogers.com

Executive Summary

Risk factors for violent individuals have been used in risk assessment protocols for decades. Such tools and guides have been shown to be a valid and reliable way to assess risk of future violence. The risk assessment protocols currently available, however, have questionable relevance to violent extremists and terrorists because the factors used to assess risk do not relate to the background and motivations of this group of violent actors. The need was identified for a relevant tool for the population of violent ideologically motivated extremists.

Approaches to risk assessment for violence are described in the document. These include unstructured clinical judgment, actuarial approaches and structured professional judgment (SPJ). Unstructured approaches have been criticized for not demonstrating high validity or good inter-rater reliability. Given the low base rate of violent extremists, it is difficult to create empirically based actuarial prediction instruments for violent extremism. The structured professional judgment approach (SPJ) has been used successfully with forensic populations and was considered appropriate for a tool to address risk assessment for the population of violent extremists. As current SPJ guides were found to be inadequate to address the specific historical and contextual features of violent extremists, a new SPJ guide was developed and is described in this document.

The major goals of the project were to identify and compare the specific characteristics and factors of those who perpetrate "general" criminal violence and those who perpetrate ideologically motivated violence, to highlight the salient differences among the historical, contextual, attitudinal and protective risk factors of these types of criminal violence and to construct a new tool to assess the risk of violence in ideologically motivated extremists.

'Political violence', 'radicalization', 'extremism' and 'terrorism' were defined and elaborated in the document. The most significant risk factors relevant to violent extremism and terrorism were extracted from the literature and organized into a structured professional judgment (SPJ) protocol. Five categories were identified as risk factors and relevant items were identified for each category. These categories are modeled after other well-established SPJ tools but are item specific to violent extremism. The categories include attitude factors, contextual factors, historical factors, protective factors and demographic factors. A preliminary model for the assessment of de-radicalization and disengagement efficacy was proposed.

The new SPJ protocol, Violent Extremist Risk Assessment or VERA is designed to be used with persons having a history of extremist violence or having been convicted of such offences. At this stage of development, the VERA is a conceptual "research" tool intended to generate debate and discussion.

Introduction

Background

Political, religious and ideological violence has been growing across the world. Canada has been and continues to be a target. A communiqué from Al-Qaeda named Canada as being among seven countries designated for attack. Canada has also been the target of a significant number of violent extremist attacks over the past century. Homegrown and imported violent extremists have been a component of the Canadian scene for many decades. Extremists in Canada have been motivated by nationalistic goals, political ideology (leftwing and rightwing), religious beliefs, and single causes. These single causes include animal rights, anti-abortion and ecological concerns.

A prominent example of "homegrown" violent extremism in Canada motivated by nationalistic fervor and political ideology was the Front de Liberation du Quebec (FLQ). The FLQ was active from 1963 to 1973 pursuing the traditional goals of Quebec independence and the emancipation of the working class. FLQ militancy included the kidnapping and murder of Pierre Laporte and the kidnapping of British diplomat James Richard Cross in October 1970.

Canada has experienced aboriginal violent extremism motivated by nationalistic and political goals. A 2005 draft of the Canadian Forces counter-insurgency manual identified the Mohawk Warrior Society as an example of a domestic group prepared to use terror tactics to further its political aims. In 1990 this group was responsible for a 78 day violent confrontation in Quebec. This conflict, known as the Oka Crisis, resulted from a land claim dispute and ended with three deaths. This was the one of a number of violent conflicts between indigenous peoples and the Canadian government.

Right wing extremists in Canada placed a bomb in a Canadian Immigration Centre in Vancouver in 1986 to protest Canadian immigration policies. In another case, five white supremacists were convicted of beating a caretaker to death at a Sikh Temple in Surrey in January 1998. They were caught on tape bragging about the killing of Nirmal Singh Gill.

Some violent extremist groups share a complex set of motivations combining political ideology with other concerns. The anarchist "urban guerrilla" group referred to as "Direct Action" or "The Squamish Five" were accused and subsequently convicted of committing "terrorist acts". Their motives were "concern for environment, poverty and the threat of nuclear destruction" (Regina v. Belmas, 1986). The appellants were described in the court documents as members of a group "committed to achieving political and environmental objectives by terrorist means". The offences arose out of bombings to a B.C. Hydro substation, an industrial factory in Ontario which manufactured guidance devices for American cruise missiles and retail outlets. They caused $7.5 million dollars in damages and injured 10 people (Regina v. Belmas, 1986).

Ecological concerns motivated violent incidents in 1997 and 1998. Farmer and commune leader Weibo Ludwig was involved in over 150 incidents of vandalism against oil companies which he believed were polluting the area around his property. Violence motivated by single cause extremism in Canada has included anti-abortionists who have been responsible for stealing and destroying medical equipment, vandalizing premises, and the shootings of three obstetricians who performed abortions.

Violence motivated by religious conviction has existed in Canada from as early as the 1920's. Acts of arson and violence were committed by sects of the Doukhobors, a religious community first located in Manitoba and Saskatchewan and later moved to British Columbia. Groups known as the "Sons of Freedom" (SOF) and the "Freedomites" continued the fire and bomb attacks for decades. In 1962, the RCMP rounded up 59 leaders of the Sons of Freedom in British Columbia for a bombing. In 2001, an 81 year old Freedomite woman, who appeared nude in court, was convicted of setting fire to a college building. The "Sons of Freedom" have been considered responsible for an estimated 40 acts of arson and explosions against municipal buildings, railway tracks, homes, schools and a post office.

Foreign political grievances motivated other "terrorist" attacks in Canada. In 1985 three gunmen, members of the Armenian Secret Army for the Liberation of Armenia (ASALA), occupied the Turkish Embassy in Ottawa for four hours and killed a security guard. In 1982, the Justice Commandos for Armenian Genocide (JCAG) assassinated a Turkish diplomat in Canada. The ASALA sent messages to the RCMP threatening to blow up Toronto and Montreal subways. In 2003, charges were brought against suspects of the Air India bombing which was motivated by Sikh nationalism. Members of the Babbar Khalsa and the International Sikh Youth Federation were implicated in the bombing of Air India Flight 182 from Vancouver to India on June 23, 1985. The Sikh groups were motivated by the desire for a separate Sikh state (Khalistan). The Boeing 747 was destroyed at 9,500 meters with 329 people on board including 280 Canadians. This event remains Canada's deadliest single "terrorist" incident.

Canadian security agencies are concerned about the susceptibility of those showing a willingness to enact terrorist violence from within Canada. There is particular concern about groups of "homegrown" violent extremists inspired by Al Qaeda. Bob Paulson, the RCMP Commissioner in charge of the National Security Criminal Investigations Unit remarked in February 2009 that more homegrown extremists and suspected terrorists are believed to be operating in Canada than ever before. The Canadian Security and Intelligence Service (CSIS) identified the threat from violent political-religious extremists as the most immediate danger to Canada and Canadians (CSIS, 2007).

In 2008, the first cases of "terrorism" were successfully prosecuted under Canada's Anti-Terrorism Act. On September 25, 2008, a 20 year old man arrested in June 2006 as part of the "Toronto 18" conspiracy was found guilty of participating in a terrorist group. The plot involved a plan to detonate bombs in downtown Toronto and storm Parliament Hill in Ottawa. A second conviction was handed down on October 29, 2008. Mohammad Momin Khawaja, a 29 year old Ottawa resident, was convicted on five charges of participating in a "terrorist group" and helping to build an explosive device likely to cause serious bodily harm or death to persons or serious damage to property.

Canadian "terrorism" and violent extremism have included incidents as varied as hoaxes and vandalism to hostage takings and mass murder (Leman-Langlois & Brodeur, 2005). Incidents of violent extremism such as those perpetrated by the FLQ have been classified as conventional terrorism and have been distinguished from "new terrorism", which has the goal of "re-establishing a historical desired situation or mythical past" (Leman-Langlois & Brodeur, 2005). Between 1973 and 2003, Canadian based incidents of violent extremism or terrorism included 6 hijackings; 2 airplane bombings; 9 hostage takings or kidnappings; 4 letter bombs; 170 bombs, firebombs and arson; 59 threats; 35 attacks on individuals; 45 acts of vandalism; 14 plots and foiled attacks; and 32 instances of support (Leman-Langlois & Brodeur, 2005).

Risk Assessment, Criminal Justice and Violent Extremism

This type of violence generates several questions pertaining to the dangerousness of these type of offenders. Importantly, what type, if any, risk assessments for future violent extremist and terrorist acts can be undertaken for convicted individuals? What risk do individuals convicted of violent extremism or terrorist offences pose to society at parole? Is terrorist violence substantially different from other criminal violence and is a terrorist merely a criminal? Are available risk assessment protocols appropriate to criminals convicted of violent extremism and terrorist offences?

Maghan and Kelly (1989) completed a comprehensive review of terrorists and the criminal justice corrections systems. They believe that it would be imprudent to regard the violent extremist/terrorist as merely a criminal. Deeming the terrorist 'a mere criminal' blurs important distinctions among types of offenders (Maghan & Kelly, 1989). If the violence of terrorists is significantly different from that of other violent criminality, then are the factors used to assess the risk of violence for criminals irrelevant to terrorists? The protocols in use today to assess general risk of violence were in large measure developed for violent individuals in psychiatric institutions (Webster, Douglas, Eaves, & Hart, 1997). To apply these protocols to violent extremists is to overvalue their applicability. The authors of the HCR-20, one of the most widely used risk assessment guides for violence, specifically state that the scope of the guide should be "restricted mainly to settings where there is a high proportion of persons with a history of violence and a strong suggestion of mental illness or personality disorder" (Webster et al., 1997). The authors recognized the need for special purpose risk assessment tools. If terrorists and violent extremists do not exhibit mental illnesses or personality disorders, and if major differences in motivation and background factors exist, then specific risk assessment protocols for violent extremists are necessary.

Assessing Risk and Determining Interventions

The assessment of risk for violence has two objectives. The first objective is to evaluate an individual to determine the risk that they will commit acts of violence. The second objective is to develop appropriate interventions to mitigate risk.

Evaluation of risk requires consideration of the type, severity, nature and other features of violence. Risk has been described as the chance of something going wrong or of being a "hazard" (Boer, Hart, Kropp & Webster, 1997). Risk is contextual, and related to the environment in which the individual resides (Hart, 1998, 2000). Ideology is an important component of this context. Family, peer and community support for violence are also relevant factors. As the notion of "risk" is only partially understood according to experts, and is not able to be predicted with certainty (Boer et al., 1997), it is prudent to devise risk assessment tools that are as specific and relevant as possible. The more accurate the identification of the motivations and the character of prior violence, the more accurate will be the prediction of future violence.

In the case of violent extremists, contextual factors, such as political views, religious tenets, ideological doctrine and reactions to geopolitical factors require consideration.

In addition, psychological factors, environmental factors and social factors would be required in a risk assessment protocol for violent extremists (Bakker, 2006; Nesser, 2004; Precht, 2007; Sageman, 2004; Silber & Bhatt, 2007).

The second objective of risk assessment is to develop effective intervention programs. The type and nature of violence determines the nature of the intervention program. Intervention for convicted violent extremists will necessarily differ from programs for other violent criminals. Attention to de-radicalization and disengagement from the extremist ideology will be an essential feature of intervention with violent extremists. Such strategies are irrelevant to non-extremists. Correctional rehabilitation has traditionally been focused on education, employable skill building, and strategies related to mental disorders, psychopathy, anger control, and drug and alcohol abuse. These are not the most relevant features for de-radicalization of violent extremists.

Tools which can assess the effectiveness of intervention programs having the objective of establishing disengagement from violent extremism are a concomitant requirement.

Purpose of this work

The purpose of this paper is as follows: (1) to identify the characteristics and risk factors relevant to "routine" non-ideological criminal violence, terrorist violence, violent extremism, and political violence; (2) to highlight the salient differences between risk factors relevant to criminal violent offenders and violent extremists; (3) to evaluate the applicability of current structured professional assessment (SPJ) guides for risk assessment to violent extremists; (4) to propose a structured professional judgment tool specifically relevant to assessing the risk of violence for this population. An approach to the assessment of disengagement and de-radicalization efficacy will be introduced in the form of a SPJ tool.

Radicalization, Extremism, and Terrorism

Historical Basis

The terms used to describe terrorism have been expanding with the proliferation of the literature examining terrorist acts and motivations. Words such as "radical", "radicalization", "extremist", "terrorist" and "terrorism" are used interchangeably by some and differentiated by others. The abundance of terms is confusing and can be misleading. Terrorism has been difficult to define even for the United Nations. Member states interpret acts of terrorism differently. For some states a specific act is "terrorism" and for other states it is not. Different professions also legitimately define terrorism in accordance with their work contexts (Maghan & Kelly, 1989). Confusion results from nuanced differences in usage in North America and Europe (Pressman, 2008).

Some 20 years before the events of September 11, 2001, the United States law enforcement perspective was that terrorism was "unlawful violence calculated to inspire terror in the general public or a significant segment thereof in order to achieve a power-outcome or to propagandize a particular crime or grievance" (Hoffman, 1998; Laqueur, 1997, 2003). This definition highlights the defining feature of terrorism which is the "intent" of the violence. Terrorism is not only a criminal act of violence, but it is an act intended to inspire fear in a population.

The terms "radicalization", "radicals", "extremists", "extremist violence" and "terrorism" have been used within the context of terrorist related violence. They are considered in this section with the corresponding definitions. The definitions highlight usage of these terms within the context of 21^st century violence. Any assessment of an individual's risk for violent extremism should be sensitive to the meaning of these terms, their differentiation from each other, and the intent of the violence.

Radicalization

The current use of the term "radicalization" refers to the process of adopting an extremist belief system (Precht, 2007; Silber & Bhatt, 2007). The "radical" is the person who has adopted this extremist belief system.

When the term "radicalization" is used in North America, it does not assume the use of violence. Radicalization is a process and not all who begin the process progress through to the end stage of violence. One can, therefore, be radicalized and not be involved with violent action. In a democratic system where freedom of thought, belief, opinion and expression are fundamental freedoms, simply having or expressing radical views is a protected right and not legally problematic. The intensification of extremist beliefs occurs throughout the process of radicalization and is a fundamental aspect of the radicalization period.

"Extremism" is a culturally relative term that is subjective, emotionally laden, and pejorative (Hoffman, 1998; Laqueur, 1997, 2003). Extremist beliefs are dependent on perspective. The person who holds views which are considered to be "extreme" within one cultural context or time may not be considered to hold "extremist" beliefs within another cultural context or time. Norms and values are intricately bound up in the definition of "extremism" and also with the process of radicalization.

Ambiguity and "subjectivity" are resolved when radicalization is considered within the framework of an established system of jurisprudence. When radicalization includes violence as a component, it is illegal. Acts of violent radicalization are criminal acts and the actors are seen to be violent extremists or terrorists. When such acts occur, the security and judicial systems take precedence over philosophical debate as to cultural relativity. Within the norms of the state and the justice system, the violent acts are illegal. Violent extremism in western societies is an unlawful action and an attack on the norms and values of the society.

Radicalization as a process was analyzed by the New York Police Department in a 2007 study. Four phases of the radicalization process were described: (1) a pre-radicalization phase or the point of origin; (2) the self-identification phase, where individuals begin to explore the radical ideology and associate themselves with like-minded individuals; (3) the indoctrination stage where beliefs are intensified and reinforced; and (4) the attack or terrorist phase where members accept that it is their individual duty to participate in a terrorist attack (Silber & Bhatt, 2007). Not all who begin the radicalization process move through all four phases and ultimately engage in violent action.

"Violent radicalization" is defined in the United States as the process of adopting or promoting an extremist belief system for the purpose of facilitating ideologically based violence in order to advance political, religious, or social change (United States of America House of Representatives, 2007). The distinction between non-violent radicalization and violent radicalization was underscored in this bill sent to the House of Representatives and referred to as the Violent Radicalization and Homegrown Terrorism Prevention Act of 2007. Although the bill did not become law, it created a heightened awareness of violent radicalization and a clearer understanding of its definition.

The process of radicalization has been defined by the Dutch Intelligence Service as the increasing willingness to use undemocratic methods, or the willingness to use, support or facilitate violence and fear as a method of effecting changes in society (AIVD, 2004). The Dutch Intelligence Service considers both violent and non-violent radicalization problematic. Non-violent aspects of radicalization are also viewed as a danger to the democratic legal order. Open rejection of Dutch democratic institutions and alienation from mainstream Dutch society by large numbers has been attributed to increasing radicalization (AIVD, 2002, 2004, 2007). Radicalization is considered to be a major contributing factor in the establishment of enclaves of non-integrated parallel communities holding views inconsistent with "Dutch" democratic and pluralistic values and laws. Radicalization is considered responsible for the increasing deep division and mistrust between groups in Dutch society (AIVD, 2004, 2007).

Recent studies concur that predicting who will radicalize to the stage of violence is difficult if not impossible. There is no one pattern or profile of terrorists and analyses of past perpetrators have revealed that they have been rather "unremarkable individuals, who have led unremarkable lives, have held unremarkable jobs" and have had little or no previous criminal record (Silber & Bhatt, 2007).

Although it may not be possible to predict with accuracy who will become a terrorist or violent extremist, it may well be possible to construct a tool which will be able to assess the dangerousness of radical extremists. Radicalization is a dynamic process. If sufficient characteristics of the process are understood, and the motivation and factors that lead to ideological violence known, the risk of future violence may be predictable by determining the presence and degree of these factors and characteristics. It may similarly be possible to assess if de-radicalization or disengagement has occurred.

Extremism

Extremism is defined as any political theory that holds to uncompromising and rigid policies or ideology. An extremist is a person who holds such uncompromising or rigid views or acts in a manner far beyond the norm. "Extremism" can be considered a subjective and value-laden term used with the motivation to condemn (Kegley, 1990). The label of "extremist" is a value judgment. Is it possible to construct reasonable and objective boundaries that are appropriate to a determination of "extremism"?

In the age of global migration and the ease of regular travel, the perceived problem of extremism has been heightened. The mass movement of individuals for economic benefit rather than the search for freedom and safety has resulted in large immigrant populations resisting social and political integration (European Union, 2005). Some immigrants resist participation in the democratic process of host societies for ideological reasons and hold to what some label "extremist" beliefs. Individuals are able to live and work in one country and remain psychologically attached to another country. They may live in one political system and yet adhere to political and ideological perspectives that are incompatible in fundamental ways. The long-term impact of trans-nationality is yet to be fully understood. Nor has the relationship between extremism and trans-nationality been adequately explored.

Trans-nationality has not been problematic when values are compatible, although not equivalent. Is trans-nationality problematic when values and political ideologies are fundamentally incompatible? Telecommunication, print, email, internet, television and other media connections permit daily links to countries and individuals around the globe. It is known that the internet is a known facilitator of "extremism" (Van Duyn, 2006; Weimann, 2004) and that the ability to stay connected to extremist ideology can promote radicalization. Extremism has been observed to be proliferating in some areas of European cities. Some of the areas are referred to as "satellite cities" in reference to the masses of satellite dishes visible on apartment buildings.

What is the impact on social cohesion and how much influence do such connections have on the development of "extremist views" and susceptibility to violent extremism? Trans-national connections, however, are clearly not the sole generators of extremist beliefs and goals. Members of extreme nationalistic or ethnic groups, fanatical religious groups, extreme right-wing and left-wing political groups, and devout single cause adherents can all couple violence to their goals of generating change. Is it possible to define the boundaries of acceptable values and attitudes? The jurisprudence system provides the basis for defining such boundaries.

Terrorism

"Terrorism" defined broadly can be included under the spectrum of acts of violent extremism. Terrorism is a relative term. It is known to have many meanings and is used in many contexts. In one study undertaken by the United States Army, over 100 definitions of terrorism were identified (Record, 2003). The United States Federal Bureau of Investigation (FBI) defines "terrorism" as the "unlawful use of force or violence against persons or property to intimidate or coerce a government, the civilian population or any segment thereof, in furtherance of political or social objectives" (FBI, 2001).

The United Kingdom Terrorism Act of 2000 (c 11) defines terrorism as an act of violence for political, religious or ideological purpose or objective or cause (United Kingdom, 2000). The Canadian Anti-Terrorism Act similarly considers motivation in defining terrorist activity. Schmid and Jongman (1988) identify five elements of terrorism: (1) it involves an act in which violence or force is used or threatened, (2) it is primarily a political act, (3) it is intended to cause fear or terror, (4) it is intended to have a psychological effect, and (5) it is intended to result in anticipated reactions.

The victims of terrorist attacks are usually civilians and not combatants. The perpetrators are not recognized combatants. They do not wear recognizable uniforms or insignia prior to or during attacks. The immediate victims of a terrorist attack are not necessarily the intended targets, but may be a target of convenience intended to create the wider reaction of fear and terror.

Crenshaw (2000) defines terrorism similarly as the deliberate and systematic use or threat of violence to coerce changes in political behaviour. It involves symbolic acts of violence intended to communicate a political message to watching audiences. Laqueur (1999) is more general in his definition of terrorism. He believes that little can be said about terrorism with certainly except that it is the use of violence by a group for political ends, usually directed against a government, but at times also against another ethnic group, class, race, religion or political movement. Political terrorism is goal directed with a purpose greater than the act, is enacted in pursuit of political objectives rather than ordinary criminal objectives, is calculated violence rather than impulsive or reactive violence and is directed at affecting the views and behaviour of specific groups.

The United Nations Security Council considers acts of terrorism as criminal acts despite not being able to define the acts precisely. UN Security Council Resolution 1566 defines terrorism as "criminal acts committed with the intent to cause death or serious bodily injury, or taking of hostages with the purpose to provoke a state of terror in the general public, or a group of persons or particular persons in order to intimidate a population or compel a government or international organization to do or abstain from doing an act" (United Nations, 2004). Many countries, including Canada, legally distinguish acts of terrorism from criminal acts done for other purposes.

Acts of terrorism are differentiated by the origin of the actors. Homegrown terrorism is contrasted with terrorism undertaken by foreign actors. "Homegrown" terrorism is defined as "the use, planned use, or threatened use of force or violence by a group or individual born, raised, or based and operating primarily within the country in which the attack takes place to intimidate or coerce the government, the civilian population or any segment thereof, in furtherance of political or social objectives" (United States of America House of Representatives, 2007). The label "terrorist" or "terrorism" has been used by countries to describe that which is considered by the country to be illegitimate violence. It has been used by countries to legitimize a state's own use of force or violence. The term "terrorist" has also been used to de-legitimize political or foreign opponents.

The total set of elements and their interactions that lead a person to terrorist activity are not known. Some elements are suspected of playing an important role. These include early socialization patterns, participation in a violent environment, identity issues, increasing commitment to an ideology or political objective, a religious conversion experience, and personal connections to extremist groups (Crenshaw, 2000; Pressman, 2008; Sageman, 2004).

In summary, terrorism is premeditated, rather than a momentary rage or impulse, has political motivation and excludes criminal violence motivated by monetary gain or personal vengeance, uses targets who are noncombatants and attacks people who cannot defend themselves, and the perpetrators are clandestine agents or sub-national groups (Pillar, 2001). Terrorism is dynamic, has many forms and is associated with many types of individuals and groups. The violent actors are influenced by personal dispositions such as attitudes, mental constructs, situational and contextual factors.

Criminal Justice and Risk: Characteristics of Violent Political Extremists and Non-Ideological Violent Criminals

Attributes and Characteristics of Violent Extremists

Most experts believe that there is no one profile that holds for all violent extremists or all terrorists. Although a single profile may not exist, there is debate as to whether it is possible to identify attributes and characteristics of violent extremists. Sageman (2004) undertook to investigate the origins of collective violence and to identify the characteristics of individual terrorists by searching for common features to explain why individuals become involved in jihadi terrorism through the analysis of 172 global Salafi jihadists. These individuals were involved in terrorist activities in the 1990's and the early part of the 21^st century .

Seventeen variables divided into three categories were identified. The categories included social background, psychological make-up, and circumstances of joining the jihad. Variables of social background included (1) geographical origins, (2) sociological status, (3) education, (4) faith as youth, (5) occupation, and (6) family circumstances. Variables for psychological make-up included (1) mental illness, and (2) terrorist personality. Variables for circumstances for joining the jihad included (1) age, (2) place of recruitment, (3) faith, (4) employment, (5) relative deprivation, (6) friendship, (7) kinship, (8) discipleship, and (9) worship.

Sageman's (2004) results indicated that many of the social explanations previously assumed with regard to terrorists were incorrect. Terrorists were not found to be poor, angry or fanatically religious. Instead, the terrorists analyzed were found to be largely middle-class, educated men from caring, stable and religious families. They grew up with strong positive values of religion and community concern. Sageman found that the terrorists did not display any psychiatric pathology nor patterns of emotional trauma in their past. No evidence of pathological hatred or paranoia was observed in the sample studied. In terms of past experience or historical factors, terrorists did not suffer long-term relative deprivation nor did they suffer from pathological prejudice.

Scholarly work subsequently undertaken on "jihadi" terrorism in Europe and in North America has supported these conclusions (Bakker, 2006; Nesser, 2004; Precht, 2007; Silber & Bhatt, 2007). There were some differences between the Sageman sample and the European sample of 242 analyzed by Bakker. The population of terrorists and suspected terrorists analyzed by Bakker were more likely to be "homegrown", older by an average of two years (27.3 years), predominantly single, and exhibited some educational and socio-economic differences from the Sageman sample (Bakker, 2006). Bakker's work, however, supported most of Sageman's significant findings. There is consensus today that no common profile exists for terrorists and studies that have sought mental dysfunctions or personality disorders as an explanation of terrorist behaviour have failed (Crenshaw, 2000; Horgan, 2003; Sageman, 2004). Evidence was not found in studies that tried to relate propensity for violence to aggressive impulses. In fact, impulsive aggression is considered to be a hindrance to terrorist activity. Executing acts of violent extremism requires individuals to maintain at least a partial clandestine lifestyle and a restricted and carefully planned use of violence. Sudden outbreaks of aggression or anger would be counter-productive to the violent extremist's goals.

Some patterns have been found. One pattern that reoccurred in mujahedin was that they were socially and spiritually alienated (Sageman, 2004). Most often affiliation and commitment to terrorist groups was accomplished through friendship, kinship, and discipleship (Bakker, 2006; Sageman, 2004), progressive intensification of faith and belief that ultimately led to acceptance of the violent ideology, and finally to acceptance in the terrorist network (Nesser, 2004; Precht, 2007; Sageman, 2004; Silber & Bhatt, 2007). It is believed that social bonds are essential to the process of increasing radicalization and participation in terrorist groups. Social support, emotional support, and the development of a common identity play a role in this process. Although some violent extremists and terrorists may exhibit a few of the psychological characteristics of psychopathy (lack of empathy with victims), they are not considered to exhibit antisocial personality disorders, dissocial personality disorders nor high psychopathy scores in terms of the criteria included in the PCL-R (Hare, 2003).

Jihadi terrorists in Europe studied by Bakker were found to exhibit dogmatism, ideological rigidity, and a simplistic and utopian view of the world (Bakker, 2006). This may make them appear to be morally deficient. In fact, they may have an exaggerated sense of morality which requires them to act (sometimes with violence) against anyone who does not comply with their rigid and uncompromising moral judgment system. Other personality features and attributes of violent extremists have included identity confusion (personal and collective), rejection of the collective identity, strong feelings of injustice, strong grievances, the identification of a person, group or country as the cause of the injustice, dehumanization (of the identified person, group), hostility, hate, a sense of internalized martyrdom (willingness to die for a cause), a need for the strong ideological-religious-political cause, a high level of frustration, rejection of "western" values and a need for revenge or morally just retribution (Bakker, 2006; Borum, 2003; Crenshaw, 1995, 2000; Maghan & Kelly, 1989; Nesser, 2005; Precht, 2007; Pressman, 2008; Sageman, 2004; Silber & Bhatt, 2007).

Some interest persists for exploring the role of psychopathy in violent extremists including the criminogenic risk/need perspective (Simourd & Hoge, 2000). Do violent extremists exhibit a sufficient combination of traits such as callousness, manipulation, deceit, remorselessness, irresponsibility and unrealistic goals to be considered psychopaths? Would violent extremists exhibit a sufficient combination of traits to be considered to have antisocial personality disorder (APD) or dissocial personality disorder (DPD)? Do violent extremists exhibit deficits in empathy and conscience?

Psychopathy falls along a continuum. Hare (1991, 2003), who developed the internationally accepted PCL-R checklist to rate psychopathy, commented that the typical citizen would likely score a 3 or 4 on his checklist (Hare, 2007). The violent extremist is likely, therefore, to exhibit at least some of the traits included on the PCL-R. Research results have indicated that the average score for incarcerated male offenders in North America is approximately 22 to 23 out of 40 (Hare, 2003). A score of 20 would qualify as moderately psychopathic. A score of 30 or above is considered highly psychopathic. Although violent extremists may have been included in forensic population studies, there is no information available on psychopathy scores for violent extremists or terrorists. Additional research is required to determine the relevance of psychopathy, APD, and dissocial personality disorder to violent extremists.

There is confusion among professionals as to the precise meanings of APD, dissocial personality disorder and psychopathy. Hare believes that psychopathy is a syndrome distinct from antisocial personality disorder (Hare, 1991, 2003). Some professionals describe psychopathy as a moral dysfunction. Others consider it a severe personality disorder identified by the PCL-R. There is disagreement as to the relationship between APD and psychopathy. The World Health Organization uses the terms dissocial personality disorder for "psychopathy".

Hare suggests that the percentage of incarcerated criminals that meet the requirements of APD is 80 to 85 percent, but only 20 percent of these would be classified by his scale as psychopaths. He asserts that this differentiates psychopathy from APD. It is this 20 percent that account for 50 percent of all serious crime according to Hare (2003). The American Psychiatric Association has reported that antisocial personality disorder is diagnosed in approximately 3 percent of males and one percent of females.

A comparative analysis of the features of APD, psychopathy (as represented by the PCL-R), dissocial personality disorder, and violent extremists was undertaken. The analysis compares the features of these disorders to the known characteristics of violent extremists. This analysis appears on Table 1.

The results illustrate that very few features are common for psychopaths and violent extremists. The results also reveal that violent extremists exhibit an insufficient number of features to generate the profile and score required for "psychopathy". Additional research will be required to support this conclusion.

Developments in Risk Assessment Methodology for the Criminal Justice System

Risk factors for violent individuals have been identified and used in risk assessment procedures for decades. The last 30 to 40 years have seen an evolution of approaches to the assessment of risk. The specific items used to assess risk have been modified as new information was integrated into the available risk assessment approaches.

The task of any risk assessment is to determine the nature and the degree of risk a given individual may pose. This risk is related to an identified or an unidentified target (Borum, Fein, Vossekuil & Berglund, 1999). Risk assessments are a requirement in some legal jurisdictions. They are used with individuals in mental health facilities, correctional facilities, forensic inpatients and individuals incarcerated in psychiatric institutions. Individuals who are detained and charged with criminal offences may undergo a risk assessment for violence.

The assessment evaluates potential risk to third parties, supports release decisions pending trial, and determines appropriate correctional placement. Risk assessments assist in the determination of appropriate treatment options and parole decisions. The likelihood of violent recidivism is also considered in the process of risk assessment (Borum, 2000, 2006; Borum, Swartz & Swanson, 1996; Quinsey, Harris, Rice & Cormier, 1996).

What set of criteria is used to determine "dangerousness" or the risk of violence? Is it possible to identify a set of features that are able to ascertain the risk of violence of an individual? There was pessimism in the 1960's and 1970's concerning the possibility of making accurate predictions of violence (Webster et al., 1997). Interest in developing a reliable approach to predicting violence was revived subsequent to the publication in 1981 of work on predicting violent behaviour (Monahan, 1981). Monahan's work spurred interest and optimism by pointing the way to improved methods of predicting dangerousness.

Risk assessment methodologies include clinical approaches using unstructured clinical judgment, actuarial approaches, and structured professional judgment methodologies. Review articles on the topic of violence risk prediction have surveyed developments in the field (Bonta, Law & Hanson, 1998; Dolan & Doyle, 2000; Ogloff, 2009). These studies have generally focused on what can be described as "mentally disordered offenders" (Dolan & Doyle, 2000). They report that systematic and structured risk assessment approaches enhance the accuracy of clinical prediction of violent outcomes. Dolan and Doyle (2000) observed that most of the data on the predictive validity of available clinical risk assessment tools are based on North American studies and further validation is required in other populations. Recent work undertaken has involved studies with European populations (Webster et al., 1997).

The first approach to violence risk is based on traditional clinical evaluation and it has been termed the "unaided clinical judgment" approach (Dolan & Doyle, 2000). This approach and the unstructured clinical judgments that are the key elements in the method have been criticized for low inter-rater reliability, low validity, unspecified and non replicable decision-making processes and inferior predictive validity as compared to actuarial prediction (Lidz, Mulvey, & Gardner, 1993; Monahan & Steadman, 1994; Mossman, 1994; Webster et al., 1997). The unaided clinical judgment approach is also referred to as the "traditional" approach or the clinical decision making model. It has been labeled an "informal, in the head impressionistic and subjective human clinical judgment" approach (Douglas, Ogloff & Hart, 2003).

Despite the noted weaknesses of the traditional clinical approach, advocates believe that there is value in the use of clinical judgment in violence risk decision making. Supporters suggest that the flexibility of the method permits a broader impression of information and factors relevant to violence prevention and intervention strategies. The clinical approach enables the clinician to focus on the origins and nature of the violence (Buchanan, 1999; Hart, 1998). Clinical ratings have been shown to be better than chance but actuarial measures have been found to be better still.

Actuarial approaches are methods that use data coded in a pre-established manner to make decisions. This approach uses equations or formulas to arrive at a probability value of risk of violence. These approaches may focus on a smaller number of known risk factors of violence and decisions about risk. The actuarial approach has been widely discussed as an improvement over the clinical approach (Borum, Bartel & Forth, 2006; Monahan, Stedman, Silver et al., 2001; Otto, 2000; Quinsey, Harris, Rice & Cormier, 1996; Webster et al., 1997). This approach is considered a more formal method compared to the clinical decision model. Many of the factors in actuarial systems are "static" factors such as demographic variables. This focus on demographic "static" factors has been a criticism lodged against actuarial approaches (Hart, 1998). The approach has also been criticized for failing to account for individual characteristics, professional judgment and to differentiate among different types of variables (Hart, 1998).

Actuarial risk assessment tools have been developed. These include the early developed Base Expectancy Score (Gottfredson & Bonds, 1961), The Level of Supervision Inventory (Andrews, 1982), The Statistical Information on Recidivism (SIR) Scale (Nuffield, 1989) and a score based on what were identified as "critical features" and called The Salient Factor Score (Hoffman, 1983). These tools have been used over the past decades in the penal systems in Canada and the United States to assist in risk of violence judgments and parole decisions.

A third approach, the structured clinical judgment approach (SPJ), is described as a composite of empirical knowledge and clinical professional judgment. The SPJ approach guides the assessor to arrive at a risk level through consideration of a defined set of risk factors. Different SPJ protocols use different factors but they are usually characterized into sections such as historical factors, contextual factors and personality or individual factors. Some of the protocols include a risk management or protective factors section. The assessor evaluates or rates the relevance of the risk factors included in the SPJ protocol to the subject being evaluated. The assessor also evaluates the presence or absence of risk management factors that may have a mitigating effect on risk.

The result of this integration was the development of tools such as the HCR-20. This SPJ "guide" resulted from an examination of the research findings on violence and violence prediction and discussions with experienced forensic clinicians. Research findings from extensive sources were used to identify specific risk characteristics (Borum et al., 1996; Douglas, Cox, & Webster, 1999; Monahan & Steadman, 1994; Mossman, 1994). SPJ approaches provide a level of accuracy that exceeds unstructured professional judgment (Ogloff & Davis, 2005).

Risk Factor Items for Assessment of Violence

It has only been in the last 15 to 20 years that valid risk assessment instruments have been developed and published. Given the low base rate of violent extremists, it would be difficult to create empirically based actuarial prediction instruments for violent extremism. SPJ has been used successfully with forensic populations, and could be used with violent extremists using appropriate risk item factors.

For violent extremism, the two most relevant of the existing SPJ tools are the HCR-20 and SAVRY. Items for inclusion in the tools resulted from an analyses of patients with the most robust predictors selected for inclusion. These items were then developed into the SPJ format.

The HCR-20 consists of three categories, historical, clinical and risk management. The tool contains 10 questions related to (past) historical factors, 5 questions based on (present) clinical factors, and 5 questions based on (future) risk management factors. The HCR-20 items are identified in Table 2 (Webster et al., 1997). The HCR-20 (Webster et al., 1997) is the most accepted, researched and validated "general" SPJ risk assessment tool for adult violence in use today. Research studies with the HCR-20 have been conducted with civil psychiatric, forensic psychiatric, general inmate populations, mentally disordered inmates and young offenders in Canada, England, Germany, Netherlands, Scotland, Sweden, and the United States. A number of studies found that HCR-20 scores relate to violence (Douglas, Webster, Hart, Eaves, & Ogloff, 2002; Ogloff & Davis, 2005).

The HCR-20 places a heavy emphasis on historical data in its assessment of risk. It allocates the same weighting to historical items as to the combined categories of clinical and risk management items. The heavy weighting on the historical items is justified on the basis that historical factors have been shown to be the most robust predictors of violence in risk assessment protocols. Historical data are the anchor of risk assessments (Webster et al, 1997). This may be the case for general criminals but the robustness of historical data for violent extremism has not been demonstrated. In addition, the historical items applicable to violent extremists will differ from those used for general criminality due to the different historical realities.

The SAVRY is a SPJ assessment tool for youth violence in the tradition of the HCR-20 (Borum, Bartel & Forth, 2006). The SAVRY was designed to be used with 12 to 18 year olds. It is a guide rather than a formal test or scale. The SAVRY assessment protocol is composed of 24 risk items in three sections. These are items are drawn from existing research on violence and aggression in youth. The items on the SAVRY, as in the HCR-20, are divided into three sections. These are historical risk factors, social contextual risk factors and individual-clinical risk factors. Six protective factor items are also included. The historical section includes 10 items, similar to the number of items in this section of the HCR-20. There are 6 items in the social/contextual risk factor section and 8 items in the individual risk factor section. Each risk item is rated as low, moderate or high. Protective factors are rated present or absent. Items are rated on a three level scale, but not assigned a numerical value.

The SAVRY is designed to address the reality that adolescence is a time of significant change intellectually, socially and emotionally. Most youth who are violent in adolescence do not persist in violent behaviour in adulthood and personality traits and behaviours in adolescence are less stable across time, contexts and situations (Cauffman & Steinberg, 2000; Grisso, 1998). As a result, repeated administrations of the SAVRY are recommended over time. The SAVRY items are illustrated on Table 3.

Relevance of SPJ Risk Assessment Protocols to Violent Political, Religious, and Other Ideological Extremists

Rationale for SPJ Protocols

The risk assessment protocols described earlier were born out of the need for a systematic manner to assess risk in individuals in forensic services. The HCR-20 was developed to meet the need for a means by which risk assessments could be performed in a systematic manner in the British Columbia Forensic Services Commission. It was among the first protocols specifically designed for assessing the risk of violence in both forensic inpatients and outpatient applications. The HCR-20 has been found to be appropriate for general population inmates and mentally disordered inmates (Douglas et al., 2002). There is no evidence as to its relevance to the population of violent political-religious extremists.

An examination of the risk factors in HCR-20 suggests that they are specifically relevant for adult populations in civil psychiatric and forensic psychiatric institutions, general inmates and mentally disordered inmates. Face-valid risk items applicable to violent extremists are not included. This is not surprising given that the HCR-20 was not designed for violent extremists, nor was much of the research concerning the attributes of terrorist violence available when the HCR-20 was first conceptualized in 1994 and revised in 1997. The authors also specifically state that the population being considered is a clinical population associated with mental illness or psychopathology (Webster et al., 1997). Therefore, should the HCR-20 be administered to populations without mental illness or disorders, it should be done with caution.

It is now known that radicalized violent criminals differ in significant ways from non-radicalized violent criminals. Many of risk factors included in the HCR-20 are not consistent with the risk factors identified in the literature relating to violent radicals and terrorists (Bakker, 2006; Precht, 2007; Sageman, 2004; Silber & Bhatt, 2007). Four significant areas of difference can be identified.

First, the HCR-20 has been constructed with a heavy weighting on mental illness; 40% of the historical items on the HCR-20 relate to issues of mental disorder. An additional item on mental illness is included in the clinical items section of the HCR-20. This single item represents 20% of the clinical section. No strong relationship between mental illness, psychopathology, or mental disorders, however, has been found to be present in convicted terrorists.

Second, for non-terrorist violence, the probability of future crime has been shown to increase with each prior criminal act. It has been demonstrated that "virtually any measure of past offending can be expected to predict future violence" (Borum et al.,1996; Borum, 2000; Borum et al., 1996; Douglas et al., 1999; Klassen & O'Connor, 1994; Monahan & Stedman, 1994; Monahan et al., 2001; Otto, 2000). In the case of terrorists, no demonstrated correlation has been shown between known violent incidents in youth and subsequent violent criminal activity. This fact has made it difficult for intelligence and security services to predict who will be a terrorist actor. The New York Police Department reports that violent radicals are generally unremarkable in this regard (Silber & Bhatt, 2007).

Third, terrorists have not consistently experienced problems with education, employment, and unstable childhoods. Many have been highly educated and well employed. This was the case in Canada with Momin Khawaja, who was convicted in 2008 of a terrorist related offence. It was the case with Mohammed Bouyeri, who was the assassin in Amsterdam of Theo Van Gogh in 2004. It was the case with attacks in the United Kingdom where the perpetrators were academically successful, including university educated professionals (e.g., engineers, physicians).

Four, terrorists have not been observed to be subject to aggressive and impulsive behaviour prior to their action. The perpetrators of violent acts motivated by political, religious or other ideologies have been judged, in general, to be normal functioning members of society. Clear thinking, precise plans, unimpeded concentration and commitment to the cause have been essential elements for the successful implementation of terrorist acts.

Table 4 identifies the relevance of the risk items on the HCR-20 to the risk characteristics of violent radicals and politically motivated terrorists. Overall, the HCR was found to have limited relevance, with 75 % of the items having little or no relationship to risk factors identified with radical violent extremists.

Risk Assessment for Targeted Violence: Relevance to Violent Extremists

Risk assessment for targeted violence includes the most relevant factors for risk assessment for political-religious-ideological extremism. The United States Secret Service has identified ten key questions for risk of violence based on experience and assassination research (Fein & Vossekuil, 1998). These questions relate in large measure to political extremism. They focus on the following 10 factors:

1. Motive. Is violence considered a legitimate means?
2. Communication. Has the threat been communicated in any way?
3. Targeted violence. Has the subject shown interest in targeted violence, extremist groups, or perpetrators of violence or weapons?
4. History. Is there any violent crime history, patterns of harassment, or willingness to use violence, or blame directed at a target for a grievance?
5. Mental Illness. Is there a history of mental illness, delusions or persecution?
6. Organization.Is the subject capable of carrying out a complex plan?
7. Desperation and Despair. Has the subject experienced a recent loss, major stress or crisis which is related to potential violence?
8. Corroboration. Can the subject's information be corroborated?
9. Behavioural Changes and Concern. Are others afraid or worried about the subject's possible violence?
10. Intervention. Are there any protective factors, intervention or support that may mitigate the risk?

In addition, information on five background areas should be gathered (Borum et al., 1999). These include (1) information that brought the subject to attention of authorities; (2) current and background information about the subject; (3) information about the attack related behaviours; (4) the motive(s) for the attack; and (5) the target selection (Borum et al., 1999; Fein & Vossekuil, 1998; Fein, Vossekuil & Holden, 1995).

These areas are important for both political and non-political violent crime risk assessment situations. Borum et al., (1999) provided details for each of these areas in relation to the type of information that would result. The importance of obtaining a comprehensive picture was stressed.

Area 1 relates to the importance of the circumstances that first brought the individual to official attention. The subject may come to the attention of authorities via an investigator, school activity or informant. The veracity of any informant, information pertaining to the subject's background and information about the subject's current situation and circumstances are essential to collect. This includes identifying information (name, physical description, date of birth and other identification), residences, education, military history, any paramilitary or other training, employment history, history of violence, history of criminal behaviour, substance abuse, mental health history and problems, relationship history, expertise, weapons training and use, history of grievances and history of harassing others.

Area 2 relates to information on the subject's life history. It includes stability of living and employment situations, nature and quality of relationships and personal support, recent losses, pending crises or changes in circumstances, hopelessness, desperation and decay in social, operational or psychological functioning.

Area 3 relates to attack-related behaviour. This section investigates the preparatory behaviors, the target of the attack if there was target, the selection and location of the target, the manner in which a weapon was obtained and any subversion of security measures.

Area 4 relates to the subject's motives and the related target selection. Understanding motive is essential to the judgment of risk in targeted violence. Continuity of motive over time is also a factor to consider (Fein & Vossekuil, 1998; Fein et al., 1995; Borum et al., 1999).

Area 5 relates to the target. Attention is paid to factors such as (1) unusual interest in instances of targeted violence, (2) evidence of ideas or plans to attack a specific target including communication and notes concerning a planned attack or intention, (3) following a target or surveillance of a location for an attack, and (4) approaching the target. A history of related violent attacks is relevant to risk.

Table 5 illustrates those risk items that have been found to be relevant to violent extremists. These items are organized into sections in order to create the structure for a SPJ instrument.

Comparative Item Analysis: Risk Assessment Instruments

Violent Extremism and Criminal Violence Factors

In this section the risk factor items used to assess violence in SPJ protocols is compared and contrasted with the factors deemed relevant to political, religious or other ideological extremism. As the SAVRY and the HCR-20 are the SPJ tools that represent risk assessment for violence with the broadest application, they are the tools used for the comparison. The items on these tools were examined for relevance to those factors and items selected for inclusion on the new tool referred to as VERA for violent extremist risk assessment.

Each risk item listed on the VERA for violent extremism was cross referenced to the HCR-20 and the SAVRY to determine the presence or absence of the item on these respective tools. The results indicated that only 3 items, corresponding to 12 %, on the HCR-20 are representative of the risk items considered relevant to violent extremists. The SAVRY had a slightly better result with 7 items, corresponding to 28% of the total items on the SAVRY, representative of the risk items considered relevant to violent extremism. Most risk items for the assessment of violent extremism were not present on either the SAVRY or the HCR-20.

The risk items for violent extremists which are present on one or both of the HCR-20 or the SAVRY SPJ protocols are noted below according to the factor category. On the attitude/mental perspective factor, "low empathy" is present on the SAVRY. On the contextual section, there are no related items present on the HCR-20 or the SAVRY. On the historical section, the items present include "previous violence", "criminal violence of a family member" and "early exposure to violence in the home". In the protective category two items are present, specifically "co-operation in rehabilitation", and "available family or community support". The limited representation of salient items related to violent extremism on the HCR-20 and the SAVRY supports a need for a specific tool to assess the risk of violence for extremists. Table 6 illustrates the results of this analysis.

VERA: Violent Extremist Risk Assessment and SPJ Instrument

VERA Consultation Version Overview

The VERA is designed to be an assessment guide that can satisfy the criteria for a SPJ protocol and that is focused specifically on assessing the risk of what has been referred to as "violent political extremism". The assessment guide is designed to be systematic, empirically grounded, developmentally informed, treatment oriented, flexible and practical (Borum et al., 2006). The instrument includes factors known to be relevant to the process of radicalization leading to violent extremism as well as committed political terrorists.

The structure of the protocol for Violent Extremism Risk Assessment (VERA) is modeled after other structured professional judgment (SPJ) tools developed to assess the risk of violence in adolescents and adults. The HCR-20, Version 2 (Webster et al., 1997) and the SAVRY (Borum et al., 2006) were valuable guides in developing the VERA. At this stage of development, the VERA is intended for consultative purposes only.

Items on the VERA have been supported by the results of research undertaken in the area of radicalization and terrorism, are based on previous work undertaken in collaboration with RCMP personnel having operational experience with criminal violent extremists, have followed from discussions with professionals in the security and intelligence fields and have used relevant information obtained from interviews and self-report questionnaire data on radicalization.

A copy of the VERA can be found at the end of this document.

De-radicalization and Disengagement

Intervention and Protective Factors

Disengagement and de-radicalization are distinct psychological and social processes. Disengagement refers to a behavioural change such as leaving a group or marginalizing one's role within the group. De-radicalization refers to a cognitive change such as relinquishing aspects of an ideology or belief system (Bjorgo & Horgan, 2008).

De-radicalization is the opposite of radicalization. It is the process of becoming less radical. De-radicalization as a process requires the rejection or moderation of a belief or ideological system. Groups or individuals may renounce a radical ideology. This may occur when the decision is taken that radical or violent actions are no longer relevant to the world view (Demant, Slootman, Buijis, & Tillie, 2008). Disengagement occurs when there has been a voluntary behavioural disconnect from the extremist organization.

Disengagement is not sufficient to guarantee de-radicalization, but it often can precede de-radicalization. Rehabilitation, re-socialization and de-radicalization programs are all designed to support a shift in attitude and ideology. The violent extremist who is committed to an ideology will be difficult to de-radicalize unless he or she has already experienced some doubt (a cognitive opening) and some disengagement.

Complete de-radicalization on a collective level means that the movement has ceased to exist or at a minimum has changed its goals. De-radicalization on an individual level means that an individual has ceased violent activities. De-radicalization programs have been developed in Saudi-Arabia, Egypt, Singapore, Iraq, Libya, Yemen, Jordan, Malaysia, Indonesia, Great Britain, and Norway. Programs are being developed or considered in other countries. The first Canadian de-radicalization program, a community based initiative, was announced in February 2009.

In the Canadian program, Mohammed Shaikh, the director of the Masjid el Noor Mosque in Toronto, is reported to be working with youths as young as 12 years old. Shaikh is a mediator, a former police chaplain, and a community activist. He has worked on youth crime prevention and conflict resolution. The program which he developed is described as including 12 steps and provides instruction in the Koran and its interpretation. The program's goals are to stress peace and good conduct, to consider the positive aspects of Canadian society, to explain what an open society means, and to stress the connections between Islam, Christianity, and Judaism. It is not clear how the efficacy of the program will be determined.

Although the goals and objectives of this and other programs may be sound, their efficacy is unknown. In the United Kingdom, Stockwell Green Community Services is an example of a program for released extremists. The program goals are to address social exclusion, economic disadvantage and radicalization. Program organizers agree that program impact and effectiveness has been difficult to measure. De-radicalization programs are a primary risk management strategy for rehabilitation of terrorists and violent extremists. The central issue in the rehabilitation of "terrorists" is related to the success of getting the individuals to relinquish the violent ideology.

The success rates of de-radicalization programs, though in some cases reported to be high, are yet to be substantiated. Saudi Arabia, as one example, has implemented a very well funded program of 8 to 12 week duration. The Saudi government has reported that 3,000 inmates voluntarily participated in such programs and only 9 were rearrested for returning to "jihad". Are people de-radicalizing because of a changing environment? Do they de-radicalize because they are offered financial incentives? Does the de-radicalization remain when the payments cease? How effective are differing de-radicalization programs?

It is possible to identify factors that relate to de-radicalization and disengagement. A potential list of "de-radicalization" factors appears on Table 7. Such factors could be organized into an SPJ protocol to more systematically assess the process of de-radicalization.

Some common elements have been noted in de-radicalization programs. These include (1) understanding the motivations (economic, psychological, and social drivers) for joining the extremist group, which may be more linked to social relationships and group dynamics than beliefs; (2) providing financial incentives to families through stipends, educational assistance or employment as a disincentive to engage in criminal violence; (3) removing individuals from the neighborhood sources and social networks supporting radical viewpoints; (4) use of family members (spouses, parents, relatives) to support de-radicalization. The Exit programs in Norway have found that partners, spouses and family obligations can be a driving force in disengagement. In some cases the use of former extremists are helpful in de-radicalization. Frustration at the lack of progress achieved through violence, trauma after attacks and disillusionment with leadership have also been identified as important factors in disengagement.

A study recently undertaken in 2008 by the Institute for Migration and Ethnic Studies at the University of Amsterdam (IMES) endeavored to answer some questions related to de-radicalization within an analytical framework (Demant et al., 2008). The collective and individual levels were identified in this framework. Motivations for de-radicalization and disengagement are suggested and the authors investigate the role that various factors have played in the decline of three movements (non-religious) and consider ways in which religious movements may differ from non-religious movements. They supplement the review of the literature on non-religious movements with interviews carried out with strict anonymity. Unfortunately the population that agreed to be interviewed was extremely limited. Former leftwing activists were easiest to identify and contact. The interviews consisted of five Malaccan activists, six individuals who had been in leftwing movements in Amsterdam and associated with the Squatters movement, five interviews with former members of extreme rightwing political parties, and five interviews with former Islamist radicals. The authors cite that it is difficult to comment about factors due to the small population interviewed and the difficulty in interviewing current political-religious extremists.

Some violent extremists may renounce violence after de-radicalization but remain committed to the cause. One example of this phenomenon is the radical "dawa" (call to Islam) movement in The Netherlands. This movement, led by political Salafists, preaches a rigid theological interpretation that resists all forms of modernity (AIVD, 2007). It has an intolerant, isolationist, anti-democratic and anti-western message, which is communicated by charismatic preachers and the internet (AIVD, 2007).

The goal of radical 'dawa' is to organize and build a broad base for radical Islam. Some radical 'dawa' groups have renounced violence and are protected by freedom of religion and expression. AIVD (2007) suggests that this has created a democratic paradox, which is the using of democratic means to bring about the fall of democracy.

Studies in de-radicalization of violent rightwing extremists and the IRA have suggested that where active and constructive political engagement is observed, a lasting disengagement from violence is more likely to result (Aronoff, 1999; Sprinzak, 1999). If disengagement and de-radicalization is disingenuous, the risk remains.

Risk Management Programs and Disengagement and De-radicalization Assessment

Effectiveness Issues

Although risk assessment for violence is important, interventions designed to ameliorate the risk are as important as risk assessment instruments for violence (Webster & Douglas, 2001). Questions are raised frequently pertaining to the risk of future dangerousness of those individuals who have gone through de-radicalization or other interventions programs. Programs may be designed to result in the disengagement of a subject from violent extremism, but they may not have demonstrated efficacy.

There is insufficient information on the effectiveness of de-radicalization programs. Most assessments of de-radicalization programs have been anecdotal, or based on vague information. A 2008 study on de-radicalization supported by the Dutch National Coordinator for Counter-terrorism (NCTb) found that it is difficult to obtain information from violent extremists (Demant et al., 2008) and reach conclusions about the de-radicalization process. The true impact of de-radicalization programs active in Saudi Arabia, Iraq, Egypt, Indonesia and Europe may be described as promising, but are limited in documentation, empirical follow-up and reliable measures.

An objective analysis of the efficacy of de-radicalization programs is essential to determine if such programs are able to modify the attitudes implicated in radical extremism. Without such demonstrated efficacy, it is irresponsible to assume that simple participation in such programs will be effective in changing entrenched values and opinions. Without demonstrated efficacy, programs should not be used as a justification for repatriation, release from incarceration or in lieu of incarceration.

It should be possible to identify a set of factors that are considered relevant to the de-radicalization process (e.g., those illustrated on Table 7) and organize these items into a structured professional judgment (SPJ) format. Such a tool would provide structured feedback on the effect of the de-radicalization programs. Obtained results would be of interest to the judicial and correctional systems and to professionals charged with strategy decisions for de-radicalization programs.

Recommendations

1. Pilot Testing, Reliability and Validity of the VERA

It will be necessary to undertake studies to determine reliability of scoring and the validity of the consultative version of the VERA. This will require access to relevant individuals within the justice system. Pilot testing with the tool will precede the reliability and validity phase. A set of probe questions to elicit the information required will be developed during this stage.

2. Consultation

Experts and practitioners in the field of risk assessment and practitioners having an interest or need for a risk assessment tool for violent extremists should be consulted for input as to the conceptualization and application of the VERA. This can be achieved through focus group sessions, interviews and written feedback from these experts. Suggestions should be incorporated into revisions to the VERA. Feedback from the authors of other risk assessment protocols should also be solicited.

3. Training and Trial Application of the VERA

Clinical trials should be undertaken with the VERA to determine the usability and utility of the instrument. This will require training of professionals from the prison, justice, security and intelligence communities who are involved with violent extremists. The training will involve sections on the background and rationale of the VERA and the use of the rating system, application of the system to real and hypothetical cases, integration of other sources of information in the ratings, analysis, the determination of the risk assessment judgment and recommendations for intervention programs based on the results.

4. SPJ Assessment Protocol for De-radicalization

The development of a structured professional judgment tool to assess de-radicalization is recommended. This will include the development and complete description of an instrument including the item ratings, background literature, background of the SPJ protocol development, user qualifications, criteria for use, directions for use, item ratings and interpretation.

Conclusion

Violent extremism includes acts of terrorism. There is little international consensus about specific acts of terrorism but violent extremism can be defined through the violent actions of the perpetrator(s) motivated by political or religious ideology, single causes or combinations of ideologies. These violent actions are contrary to the legal system. There are common elements in what has been labeled "terrorism" and "violent extremism". These common elements include the nature of the violent action and the motivations which precipitate the action. In terms of action, both terrorism and violent extremism utilize threatened or actual violence and both use violence to inflict property damage and/or human casualties. In terms of motivation, both terrorism and violent extremism are motivated by ideology. The use of violence in both cases is to evoke fear and effect change.

Terrorism and violent extremism differ from ordinary crime. Ordinary crime is motivated by financial gain, aggressive narcissism or other personal motivation. Terrorism and violent extremism are motivated by larger ideological issues. Although terrorist acts and violent extremist acts are violations of criminal law, the underlying motivation renders these acts significantly different from other criminal violence.

Concomitant with the difference in motivation is the difference in the underlying characteristics of the actors and the factors relevant to the acts. There is consensus that terrorists do not act out of mental imbalance, psychopathology or psychopathy. They are not typically unemployed, under-employed, under-educated or poor. They are not generally irresponsible, or impulsive. They may have a well developed moral code but this code may not include empathy for those whom they consider the enemy. Their acts are planned, coordinated, morally justifiable within their value system and high in social conscience. Their background and historical factors, situational factors and attitude factors differ from ordinary criminals.

The development of an appropriate risk assessment protocol relevant to criminal actors motivated by political, religious or other extremism was deemed necessary and a tool was developed. This tool, referred to as the VERA, or Violent Extremist Risk Assessment, is a structured professional judgment guide. At this stage it is a research consultative protocol. It may be considered as a supplemental guide in addition to other established and validated tools to address the specific features of violent extremism.

Terrorism and violent extremism are crimes addressed by law enforcement and the judicial system. Risk assessments for violence are routinely required by the judicial system prior to decisions relating to pre-trial release, corrections management and parole. It is hoped that the present developments will be an addition to the practice of risk assessment.

References

AIVD (2002). Rekrutering in Nederland voor de jihad van incident naar trend [Recruiting for Jihad in The Netherlands: From Incident to Trend]. The Hague: Netherlands General Intelligence and Security Service.

AIVD (2004). From Dawa to Jihad. The Hague: Netherlands General Intelligence and Security Service.

AIVD (2007). Radical Dawa in transition: The rise of Islamic Neoradicalism in the Netherlands. The Hague: Netherlands General Intelligence and Security Service.

Andrews, D. A. (1982). The Level of Supervision Inventory: (LSI). Toronto, ON: Ontario Ministry of Correctional Services.

Aronoff, M. J. (1999). Political violence and extremism. Israel Studies, 4, 237-246.

Bakker, E. (2006). Jihadi terrorists in Europe: Their characteristics and the circumstances in which they joined the jihad- An exploratory study. The Hague: Clingendael Security and Conflict Programme.

Bjorgo, T., & Horgan, J. (2008). Leaving terrorism behind: Individual and collective disengagement. New York: Routledge.

Boer, D. P., Hart, S. D., Kropp, P. R., & Webster, C. D. (1997). Manual for the Sexual Violence Risk-20: Professional guidelines for assessing risk of sexual violence. Vancouver, B. C.: Mental Health Law & Policy Institute, Simon Fraser University.

Bonta, J. (2002). Offender risk assessment: Guidelines for selection and use. Criminal Justice and Behavior, 29, 355-379.

Bonta, J., Law, M., & Hanson, R. K. (1998). The prediction of criminal and violent recidivism among mentally disordered offenders: A meta-analysis. Psychological Bulletin, 123, 123-142.

Borum, R. (2003). Understanding the terrorist mind-set-perspective. FBI Law Enforcement Bulletin, 72 (7), 7-10.

Borum, R. (2000). Assessing violence risk among youth. Journal of Clinical Psychology, 56, 1263-1288.

Borum, R. (2006). Assessing risk for violence among juvenile offenders. In S. Sparta & G. Koocher (Eds.), The forensic assessment of children and adolescents: Issues and applications (pp. 190-202). New York: Oxford University Press.

Borum, R., Bartel, P., & Forth, A. (2006). SAVRY: Professional Manual for Structured Assessment of Violence Risk in Youth. Lutz, FL: Psychological Assessment Resources Inc.

Borum, R., & Reddy, M. (2001). Assessing the violence risk in Tarasoff situations: A fact-based model of inquiry. Behavioral Sciences and the Law, 19, 375-385.

Borum, R., Fein, R. A., Vossekuil, B., & Berglund, J. (1999). Threat assessment: Defining an approach for evaluating risk of targeted violence. Behavioral Sciences and the Law, 17, 323-337.

Borum, R., Swartz, M., & Swanson, J. (1996). Assessing and managing violence risk in clinical practice. Journal of Practical Psychiatry and Behavioral Health, 2, 205-215.

Buchanan, A. (1999). Risk and dangerousness. Psychological Medicine, 29, 465-473.

Cauffman, E. E., & Steinberg, L. (2000). (Im) maturity of judgment in adolescence: Why adolescents may be less culpable than adults. Behavioral Sciences and the Law, 18, 1-21.

Crenshaw, M. (1995). Thoughts on relating terrorism to historical context. In M. Crenshaw (Ed.), Terrorism in context (pp. 3-24). University Park, PA: Pennsylvania State University Press.

Crenshaw, M. (2000). The psychology of terrorism: An agenda for the 21^st century. Political Psychology, 21, 405-420.

CSIS (2007). Canadian Security Intelligence Service Public Report (Report # PS 71). Ottawa, ON: Public Works and Government Services.

Demant, F., Slootman, M., Buijs, F., & Tillie, J. (2008). Decline and disengagement: An analysis of the processes of de-radicalization. Amsterdam, Netherlands: IMES, University of Amsterdam.

Dolan, M., & Doyle, M. (2000). Violence risk prediction: Clinical and actuarial measures and the role of the Psychopathy Checklist. British Journal of Psychiatry, 177, 303-311.

Douglas, K., Cox, D. N., & Webster, C. D. (1999). Violence risk assessment: Science and practice. Legal and Criminological Psychology, 4, 149-184.

Douglas, K., Ogloff, J., & Hart, S. (2003). Evaluation of a model of violence risk assessment among forensic psychiatric patients. Psychiatric Services, 54, 1372-1379.

Douglas, K., Webster, C. D., Hart, S., Eaves, D., & Ogloff, J. (2002). HCR-20: Violence Risk Management companion guide. Vancouver, B.C.: Mental Health Law & Policy Institute, Simon Fraser University.

European Union (2005). Addressing the factors contributing to violent radicalization (Report # COM-2005 313). Brussels, Belgium: European Union Council Concerning Terrorist Recruitment and Commission, European Parliament.

FBI (2001). Code of Federal Regulations (28 C.F.R. Section 0.85) relating to terrorism. Washington, DC: Federal Bureau of Investigation.

Fein, R. A. & Vossekuil, B. (1998). Protective intelligence and threat assessment investigations: A guide for state and local law Enforcement officials (NCJ No. 170612). Washington, DC: National Institute of Justice, Office of Justice Programs.

Fein, R. A., Vossekuil, B., & Holden, G. A. (1995). Threat assessment: An approach to prevent targeted violence (Report # NCJ 155000). Washington, DC: National Institute of Justice, Office of Justice Programs.

Gottfredson, D. M., & Bonds, J. A. (1961). A Manual for intake base expectancy scoring. San Francisco, CA: Research Division, Department of Corrections.

Grisso, T. (1998). Forensic evaluation of juveniles. Sarasota, FL: Professional Resource Press.

Hare, R. D. (1991). Manual for the Hare Psychopathy Check List-Revised. Toronto, ON: Multi-Health Systems.

Hare, R. D. (2003). Manual for the Hare Psychopathy Check-List-Revised (2^ndEd.). Toronto, ON: Multi-Health Systems.

Hare, R. D. (2007). Forty years are not enough: Recollections, random musings and prognostications. In H. Hervé & J. Yuille (Eds.), The psychopath: Theory, research, and practice (pp. 3-29). Mahwah, NJ: Lawrence Erlbaum.

Hart, S. D. (1998). The role of psychopathology in assessing the risk for violence: conceptual and methodological issues. Legal and Criminological Psychology, 3, 123-140.

Hoffman, B. (1998). Inside terrorism. New York: Columbia University Press.

Hoffman, P. B. (1983). Screening for risk: A revised Salient Factor Score (SFS: 81). Journal of Criminal Justice, 1, 539-547.

Horgan, J. (2003). Leaving terrorism behind: An individual's perspective. In A. Silke (Ed.), Terrorists, victims and society. New York: John Wiley and Sons.

The United States of America House of Representatives (2007). Violent radicalization and homegrown terrorism prevention act of 2007. Washington, DC: author.

Klassen, D., & O'Connor, W. A. (1994). Demographic and case history variables in risk assessment. In J. Monahan and H. J. Steadman (Eds.), Violence and mental disorder: Developments in risk assessment (pp. 229-257). Chicago, IL: University of Chicago Press.

Kegley, C. W. (1990). International Terrorism: Characteristics, causes, control. Basingstoke Hants, U.K: Palgrave-Macmillan.

Laqueur, W. (1997). Post modern terrorism. Washington, DC: The United States International Information Program

Laqueur, W. (1999). The new terrorism: Fanaticism and the arms of mass destruction. New York: Oxford University Press.

Laqueur, W. (2003). No end to war: Terrorism in the twenty-first century. New York: Continuum.

Leman-Langlois, S. & Brodeur, J. P. (2005). Terrorism old and new: Counter-terrorism in Canada. Police Practice and Research, 6, 121-140.

Litz, C. W., Mulvey, E. P., & Gardner, W. (1993). The accuracy of predictions of violence to others. Journal of the American Medical Association, 269, 1007-1111.

Maghan, J. & Kelly, R. J. (1989). Terrorism and corrections: The incarcerated radical. In J. R. Buckwalter, (Ed.), International terrorism: The decade ahead (pp.29-53). Chicago, IL: University of Illinois Press, Office of Criminal Justice.

Monahan, J. (1981). Predicting violent behavior: An assessment of clinical techniques. Newbury Park, CA: Sage.

Monahan, J., & Steadman, H. J. (1994). Violence risk assessment: A quarter century of research. In L. E. Frost & R. J. Bonnie (Eds.), The evolution of mental health law (pp.195-211). Washington, DC: American Psychological Association.

Monahan, J., Steadman, H. J., Silver, E., Applebaum, P. S., Robbins, P. C., Mulvey, E. P., Roth, L., Grisso, T., & Banks, S. (2001). Rethinking risk assessment: The MacArthur study of mental disorder and violence. New York: Oxford University Press.

Morgan, M. J. (2004). The origins of the new terrorism. Parameter, 34, 29-43.

Mossman, D. (1994). Assessing predictions of violence: Being accurate about accuracy. Journal of Consulting and Clinical Psychology, 62, 783-792.

Nesser, P. (2004). Jihad in Europe: A survey of the motivations for Sunni Islamist terrorism in post millennium Europe (FFI Rapport 2004/01146). Kjeller, Norway: Norwegian Defence Research Establishment.

Nuffield, J. (1989). The SIR Scale: Some reflections on its applications. Forum on Corrections Research, 1, 19-22.

Ogloff, J., & Davis, M. R. (2005). Assessing risk for violence in the Australian context. In D Chappell and P. Wilson (Eds.), Crime and justice in the new millennium. Sydney, Australia: Lexis Nexis.

Ogloff, J. (2009). The violent client: Advances in violence risk assessment. Melbourne, Victoria: The Australian Psychological Society.

Ogloff, J. & Daffern, M. (2004). The dynamic appraisal of situational aggression (DASA). Melbourne, Victoria: Centre for Behavioral Science, Monash University.

Otto, R. K (2000). Assessing and managing violence risk in outpatient settings. Journal of Clinical Psychology, 56, 1239-1262.

Paulson, B. (2009, February 12). Number of homegrown suspected terrorists higher than ever RCMP. Ottawa Citizen. Retrieved July 27, 2009 from http://www2.canada.com/topics/news/story.html?id=1279132

Pillar, P. R: (2001). Terrorism and U.S. foreign policy. Washington, DC: Brookings Institution Press.

Precht, T. (2007). Home grown terrorism and Islamist radicalization in Europe: From conversion to terrorism. Copenhagen, Denmark: Danish Ministry of Justice.

Pressman, D. E. (2008). Exploring the sources of radicalization and violent radicalization: Transatlantic perspectives. Journal of Security Issues, 2, 1-20.

Quinsey, V. L., Harris, G. T., Rice, M. E., & Cormier, C. A. (1996). Violent Offenders: Appraising and managing risk. Washington, DC: American Psychological Association.

Quinsey, V. L., Rice, M. E. & Harris, G. T.: (1995). Actuarial prediction of sexual recidivism. Journal of Interpersonal Violence, 10, 85-105.

Regina v. Belmas (1986). H. M. v. Juliet Caroline Belmas, Appellant [27 C.C.C] 142, 1966. Vancouver, BC: Carswell.

Record, J. (2003). Bounding the global war on terrorism. Carlisle, PA: US Army War College.

Sageman, M. (2004). Understanding terrorist networks. Philadelphia, PA: University of Pennsylvania Press.

Schmid, A. P & Jongman, A. J. (1988). Political Terrorism (2^nd ed.). Oxford: North-Holland Publishing Company.

Silber, M. D., & Bhatt, A. (2007). Radicalization in the west: The homegrown threat. New York: New York Police Department Intelligence Division.

Simourd, D. J., and Hoge, R. D. (2000). Criminal psychopathy: A risk and need perspective. Criminal Justice and Behavior, 27, 256-272.

Spinzak, E. (1999). Brother against brother: Violence and extremism in Israeli politics from Atalena to the Rabin assassination. New York: The Free Press.

United Kingdom (2000). Terrorism Act: 2000 (c.11). London, UK: Acts of Parliament, UK Government.

United Nations (2004). Security council resolution: 1566, 5053^rd Meeting of the Security Council, 8 October, 2004. New York: United Nations Security Council Resolutions.

Van Duyn, D. (2006). FBI congressional testimony to the House Homeland Security Committee: Subcommittee on intelligence information sharing and terrorism risk. Washington, DC: United States House of Representatives.

Vermaat, E. (2005). De Hofstadgroep [The "Hofstad" Group]. Amsterdam, The Netherlands: Aspekt BV.

Webster, C. D., Douglas, K. S., Eaves, D., & Hart, S. (1997). HCR-20 assessing risk for violence: Version 2. Vancouver, BC: Mental Health Law & Policy Institute, Simon Fraser University.

Webster, C. D., Douglas, K. S., Eaves, D., & Hart, S. (2001). HCR-20 violence risk management companion guide: Vancouver, BC: Mental Health Law & Policy Institute, Simon Fraser University.

Webster, C. D., & Douglas, K. S. (2001). Purpose of the companion guide. In C.D. Webster, K. S. Douglas, D. Eaves, & S. Hart (Eds.), HCR-20 violence risk management companion guide, Vancouver, BC: Mental Health Law & Policy Institute, Simon Fraser University.

Weimann, G, (2004). www.terror.net: How modern terrorism uses the internet (Report # 116). Washington, DC: United States Institute of Peace.

Appendix A

VERA: Violent Extremist Risk Assessment and SPJ Instrument

VERA Consultation Version Overview

The structure of the protocol for Violent Extremism Risk Assessment (VERA) is modeled after other structured professional judgment (SPJ) tools developed to assess the risk of violence in adolescents and adults. The HCR-20, Version 2 (Webster et al., 1997) and the SAVRY (Borum et al., 2006) were valuable guides in developing the VERA. At this stage of development, the VERA is intended for consultative purposes only.

Training is required of assessors using this tool. The training will be carried out in a manner consistent with the type of training required for the administration of other SPJ tools. This training will ensure that the meaning of the items and the coding procedures are clearly understood. Training will need to be regularly repeated over time in order to ensure consistency of administration.

Assessors make two types of judgment when using the VERA. The first judgment, at each of 25 item levels, is the presence of the item in relation to the subject and its level (low, medium, or high). Three additional demographic items are scored on the basis of two levels. The second task is to integrate the responses at each item level (low, medium, or high) into a final judgment relating to risk for violence. A standardized framework including a coding form has been constructed to assist in the rating task . The evaluation process requires that the assessor determine the context for the evaluation. This will differentiate, for example, whether the risk assessment is for institutional or community violence.

The VERA should be used as only one component of a multi-method assessment strategy. This is consistent with other SPJ protocols. Other available information, including facts obtained from interviewing the client and gathered from other sources will provide supplemental information to assist in the rating process. Police reports, reports from prosecutors and other professional reports should be obtained and used.

Items on the VERA have been supported by the results of research undertaken in the area of radicalization and terrorism, are based on previous work undertaken in collaboration with RCMP personnel having operational experience with criminal violent extremists, have followed from discussions with professionals in the security and intelligence fields and have used relevant information obtained from interviews and self-report questionnaire data on radicalization.

The VERA is a move away from a "self-report" model to the SPJ model. The SPJ approach to "risk assessment" by a professional evaluator with a structured guide is considered a more reliable protocol and has been an accepted and validated methodology for risk assessment for criminal violence.

Background of the VERA

The VERA is designed to be an assessment guide that can satisfy the criteria for a SPJ protocol and that is focused specifically on assessing the risk of what has been referred to as "violent political extremism". The assessment guide is designed to be systematic, empirically grounded, developmentally informed, treatment oriented, flexible and practical (Borum et al., 2006). The instrument includes factors known to be relevant to the process of radicalization leading to violent extremism as well as committed political terrorists.

The VERA is composed of 28 items. Some items are drawn from the domains of historical, contextual and individual factors used in other SPJ tools. Demographic items used in actuarial approaches and relevant to violent extremism are included in one additional category. Risk items have been selected based on the known characteristics of individuals who have been involved in or convicted of offences related to violent extremism and terrorism. Risk items were structured into the SPJ format to address the salient background variables, personality characteristics, attitudes and motivations of terrorists and violent extremists.

User Qualifications

The VERA can be used by professionals who have been trained in a variety of disciplines. This is a departure from risk assessment tools that are designed to be used exclusively or primarily by psychologists. The primary qualification for using the VERA is training and experience in conducting professional assessments. Training and a demonstrated understanding of the radicalization process, violent extremism and terrorism is a requirement. Experience with intervention program design related to the risk outcome is desirable.

Use of the VERA

The instrument is intended to be rated on a three level scale similar to that of the SAVRY. Each item is not given a numerical number as the VERA is not designed to be a formal test or scale. This is consistent with the HCR-20 and the SAVRY. It is intended to be a professional guide consistent with other structured professional judgment tools.

The objective in assessing risk for extremist violence is similar to that considered for individuals who use "targeted violence" (Borum et al., 1999; Borum & Reddy, 2001). The instrument is developed to assess whether or not the subject of the risk assessment has an identified or identifiable person, group or country as the target of violence. It will determine the ideological, religious or political motivation for the violence. It will determine if the person is acting as part of a group or as a "lone wolf". Assessment should take the form of an inclusive report including the sections identified in Table 8.

The VERA is to be used with and limited to persons with histories of extremist violence or convictions for terrorist related offences. At this stage of development, the VERA is considered a conceptual tool intended to generate comments from other experts in the field working with individuals convicted of terrorist related offences and discussion with potential users. Any results obtained with the tool should be interpreted with caution as the VERA is at this stage a research tool only. Used in a supplemental manner in addition to established risk assessment tools, it may be able to generate additional pertinent structured and guided information related to violence risk with those having been detained or convicted of terrorist related offences.

For purposes of the VERA, violence is defined as a violent act or the threat of committing a violent act of a severity that criminal charges would result and which is intended to cause death or serious bodily harm to civilians (non-combatants) with the purpose of intimidating a population, person or government to do or abstain from doing an act.

VERA Factors and Ratings

Rating information is provided for this first version of the VERA. The Attitude Factor is the first category on the VERA. This is a departure from both the SAVRY and the HCR-20, which use historical items as the anchor for the protocol. Attitude items are the anchor in the case of extremist violence. The other sections include contextual factors, historical factors, protective factors and demographic factors. As attitudes are fundamental to ideological motivated violence, there are at least twice as many items (N=10) in this category as in the other sections.

A rating sheet for the VERA protocol is illustrated in Figure 1. A detailed description of each item rating is described below.

Item Ratings: Attitudes Factor

Ten items in the Attitudes Factor are rated. These attitude items have been found to be important to the motivation for violent radicalization and terrorist related offences.

(1) Attachment to an ideology justifying violence

If a subject is committed to a political, religious ideology or cause where the use of violence to achieve the objective is morally justified and legitimized, then the subject will obtain a rating of "high" on this item. If there is no adherence to the use of violence to achieve ideological goals, then the subject receives a rating of "low". If there is acceptance of the use of violence in some situations but not in others, then the subject will obtain a rating in the "medium range". The rating on this item is determined by the level of commitment to the ideology and to the use of violence to achieve the desired goals.

(2) Perception of injustice and grievance

A subject's perception of injustice and grievances can also be rated as "low", "mid-range" or "high". If a subject has strong feelings that he or his identified group has been the subject of injustice and he/she has a high level of grievance, then the subject receives a rating of "high" on this item. If there is no perception of injustice or grievance, then the subject gets a "low" rating. If there is some perception of injustice and grievance but it is not judged as excessive, then the subject receives a rating in the "medium range".

(3) Identification of cause of the injustice

If a subject has identified the target that is believed to be the cause of injustice with specificity and certainty, he will be rated "high" on this item. If the subject has not identified a specific target or cause of injustice, then this item will be rated as "low". If there is some identification, but specificity is lacking and the target is vague, then this item is rated in the "medium range".

(4) Dehumanization of identified target

If there is dehumanization of the target with a corresponding specificity of a target (individual, group), then this item is judged to be "high". If there is no dehumanization of the target, the item is rated as "low". If there is some dehumanization, but equivocation exists or some concern is noted for the loss of life, then this item is rated in the "medium range".

(5) Internalized Martyrdom

This item relates to the subject's inner reality and acceptance of the necessity to be ready to die for the cause and that such martyrdom is both desirable and noble. If the subject has internalized and accepted the necessity for martyrdom, the item is rated "high". If there is acceptance of the need to be ready to die but the idea of martyrdom is not firmly embedded in the subject's consciousness, the item is rated as "medium range". If there is acceptance of the need to fight for the cause, but not of martyrdom for the cause, then the subject receives a rating of "low".

(6) Alienation from Society and Rejection of Values

A feeling of significant alienation and or detachment from civil society with the concomitant rejection of the democratic values and norms of a secular multicultural society is sufficient to obtain a "high" rating on this item. If the subject expresses limited alienation from society and its values, the item is rated as "low". If the subject rejects some democratic and pluralistic values of the society and is seen to have significant alienation from society, the item is rated in the "medium" range.

(7) Hate, frustration, persecution

A subject exhibiting a high level of hate, anger, frustration and persecution related to his/her world view is rated "high" on this item. This hate and frustration is related to the ideological stance of the individual. A minimal level of hate, frustration, anger and sense of persecution is rated "low". Some frustration and anger is rated in the "medium range".

(8) Need for Group bonding and Belonging

A subject who has a strong need to be part of and accepted by a group is rated "high" on this item. The relative absence of this need is rated "low". Some desire for acceptance and bonding, but not an excessive need is rated in the "medium range". This item is related to personal identification as group bonding is related to personal identity.

(9) Collective Identity

If a subject exhibits an identity that is hostile to the national collective identity, has no sense of belonging to the society of residence and exhibits confused or negative loyalty, then he/she is rated "high" on this item. If the subject exhibits no collective identity issues, the item is rated as "low". The presence of some collective identity issues as elaborated above results in a rating in the "medium range".

(10) Empathy

Very low empathy for those for those outside one's own group can be judged to be in the low, medium or high range. If the subject has no empathy for those outside his/her group, then the rating for this item is "high". If empathy is present for those outside his/her group, then the rating for this item is "low". If there is some empathy for others outside the subject's group, the rating is in the "medium range".

Item ratings: Contextual Risk Factors

Contextual risk factors consider the influence of friends, family and the environment on the subject. It is expanded here to include other influences on the subject such as the internet, training and social contacts from extremist religious, political, ethnic or other ideological motivated groups. It also includes factors related to geo-political situations and government responses that may provoke violent responses.

Contextual factors are important in determining the risk as most terrorists do not plan attacks alone. Violent intention without other human contacts, expertise, leadership, plans, financial and technical support usually results in inaction. Extremist ideology is supported in "group think". This may take the form of peer support, study groups, community support, and access to materials, finances and leadership. The internet has become an important initiator and support for extremist ideas and has served as a link for extremist groups. In other cases, training is undertaken either within the country or abroad. Contextual Factors are the second category of the VERA.

(1) User of Extremist Websites

If the subject is a regular and committed participant and/or user of extremist websites including those that promote and provide instruction in bomb-making, the rating is "high". If the subject does not use the internet or other means to connect to extremists or extremist websites promoting and justifying the use of violent action to achieve ideological goals, the rating is "low". If the subject knows of extremist websites but is not a regular participant or user of the extremist sites, the rating is "medium range"

(2) Community Support for Violent Action

If the subject lives within a community where there is strong support for violent action to further religious or political goals or has a strong attachment to a peer group or study group with whom he has a strong bond and where there is strong support for violent action to further the religious, political or ideological goals, then the rating is "high" (The "peer support for violent action" assessed in this item differs from the subject's own need for a peer group or peer bonding that was as assessed in the previous section.) If the subject lives within a community where there is no support for violence to further the ideological goals, and is not involved with a group supporting the use of violence, then the rating for this item is "low" If there is strong community support for the ideology but low or inconsistent support for violence, then the rating is "medium range".

(3) Direct Contact with violent extremists

If the subject is in contact with three or more "violent" extremists, or is part of a group which includes radical extremists, the rating is "high". If the subject is not in contact with radical/violent extremists, the rating is "low". If the subject has one or two peer/contacts who are violent extremists, the rating is in the "medium range".

(4) Anger at governmental political decisions, actions or other anger directed at the government

This item relates to the degree of anger felt by the subject feels towards the foreign policy decisions and/or actions of a country when such action is against his/her political/ideological or religious views. If there is a strong level of anger and hostility, the rating is "high". If the subject has strong disagreement with actions by the government without high levels of anger, the rating is "low". If there is moderate anger and hostility but the anger is controlled, then the rating is in the "medium range".

Item Ratings: Historical Risk Factors

Historical Factors are the third category of items on the VERA. In the HCR-20 and SAVRY the history of violence is one of the most robust predictors of future violence. Violence in these other tools refers to an act of battery or physical violence that is sufficiently severe to cause injury to another person or persons such as cuts, bruises, broken bones, or death. It includes sexual violence and other forms of domestic violence.

Extremist violence is not related to domestic anger, sexual desires, loss of control or financial gain. Extremist violence motivated by ideology can range from targeted assassinations to violent acts of an apocalyptic or catastrophic nature designed to cause mass-casualties to civilians. In some cases the violence can be characterized as that where destruction and chaos are ends in themselves. Terrorist violence can be seen as "fanatical violence" where the end result is to bring political change, the defense of identity or to address an issue of dignity. The concept of "righteous killing" should be considered in the risk assessment of violent extremists (Morgan, 2004).

The following are included in the historical risk factor section of the VERA:

(1) Early Exposure to Violence

The first item refers to the subject's early exposure to violence. If the subject was exposed to support for violent action related to political, nationalistic, ethnic, religious conflict whether he lived within a democratic country from birth or not, the item should be scored "high". If the subject has not been exposed to nor encountered support for violence either in the home or outside the home, the item should be rated "low". If the subject has been exposed to considerable domestic violence but not to the support of political violence, the item should be scored in the "medium range".

(2) Family and Friends Support for Violent Action

The second item refers to support which family or close friends have demonstrated for violent action. If one or both parents, brothers, sisters, cousins or uncles or close friends have been involved in fighting or raising funds for violent causes, or are promoters of violent action the item is scored "high". If the family or close friends do not support violent action, the item is scored "low". If there is support for some violent action but there has been no direct involvement by the family or friends, the item is scored in the "medium range".

(3) Prior Criminal Violence

This item is related to the subject's prior involvement in violent action that may or may not be ideologically motivated. If the subject has been arrested for prior criminal violence, the rating is "high". If there are no prior arrests for criminal related violence, the item is scored "low". If there are reports of violent incidents at school, at work, at home or elsewhere but these reports have not resulted in arrest, the item is scored in the "mid-range".

(4) Military or Paramilitary Training

If the subject has had paramilitary training that is non-state sponsored legitimate military training, the obtained rating is "high" If the subject has had no weaponry experience or training, the rating is "low". If the subject has had some training but has not been part of a state-sponsored military or paramilitary training program, the rating is "medium range".

(5) Travel out of the Country for Non-State Sponsored Training and Fighting

If the subject has traveled abroad to participate in a paramilitary non-state sanctioned training or has participated in non-state sanctioned training camps for weaponry and tactical, military, or violent resistance, a "high" rating is allocated. If there has been no training abroad a "low" rating is given. If the subject had the intention to travel abroad for such training but was not able to do so, the rating is in the "medium range".

(6) Glorification of violent action

This item is related to the elevation of violence for a perceived noble cause in response to religious or nationalistic beliefs. If the subject has been and remains committed to the view that violence is legitimate and killing is part of a noble mission that is morally justifiable, then the rating on this item is "high". If there is no glorification or legitimization of violent action and no moral justification, then the rating is "low". If there is some moral justification for violent action but no glorification, then the rating is in the "medium range".

Ratings: Protective Factors

Protective factors are the factors that if present, mitigate risk of future violence. The protective factors are rated in terms of the level or extent they are present. If the identified factors are not present, risk of violence is less likely mitigated. Awareness of the presence or absence of these protective factors are useful for intervention program planning.

(1) Shift in ideology

If there is a shift in values away from the previous extremist ideology, then the rating is "high". The rating is in the "medium range" if some adjustment or shift in ideology has been observed in the subject's views or if changes in values are credible. If there is no change in ideology and values from the earlier extremist ideology, then the rating is "low".

(2) Rejection of violence to achieve goals

If the subject rejects the belief that violence used to achieve goals is morally justifiable and legitimate, then the rating is "high". If the subject has not rejected the use of violence to achieve the goals, the rating given to this item is "low". If there is some self-questioning as to the value of the use of violence, the rating for this item is in the "medium range".

(3) Change of vision and concept of "enemy"

If there is a change in perception of the "enemy" from the (prior) extremist ideology, the rating for the subject on this item is "high". If there has been no revision in ideology and the enemy referent remains a target, then the rating on this item is "low". If there has been some modification in the perception of the "enemy" and increased acceptance of democratic values, then the rating is in the "medium range".

(4) Interest in constructive political involvement

If there is interest in participation in the democratic political process, then the rating is "high". If there is no interest in participation in the democratic political process, the rating is "low". If some interest in the political process is noted and the interest in participation is credible, the rating is in the "medium range".

(5) Significant Community Support

If the subject has support for de-radicalization from his/her family or spouse or community organization, then the rating on this item is "high". If there is no support for de-radicalization and disengagement in the family and community, the rating is "low". If there is limited support, then the rating is in the "medium range".

Ratings: Demographic Items

Demographic items related to sex, age and marital status are included as they have been found to be relevant to violent extremists. These items are given two level ratings rather than a three level rating.

These items are rated and considered in the assessment as follows.

(1) Sex:

If the subject is a male the rating is "high".

If the subject is a female the rating is "low".

(2) Age

If the age of the subject is under 30 years of age, the rating is "high".

If the subject is over 30 years of age, the rating is "low".

(3) Marital Status

If the subject is unmarried or has been married or cohabiting for less than a year, the rating is "high". If the subject is married or cohabiting for more than a year the rating is "low".

Date modified:

2022-07-21