Violent Extremism in America
Apr 1, 2021
Photo by Jessica Rinaldi/Reuters
Hate, violence, and their co-occurrence—violent extremism—represent increasing threats to society. After decades of combating global jihadism, the United States increasingly is confronting domestic extremism, much of it from those identifying with far-right political movements. The prevalence and nature of this threat have prompted a focus on new approaches and frameworks that go beyond the counterterrorism approach that has dominated the battle against global jihadism.
One approach that has gained increasing attention is applying a public health model to understand and counter violent extremism and its downstream effects (Snair, Nicholson, and Giammaria, 2017). This approach seeks to understand the demographic, community, and psychological drivers of violent extremism to help drive prevention and intervention efforts. Many researchers have called for such an approach (e.g., Alcalá, Sharif, and Samari, 2017; Bhui et al., 2012; Weine and Kansal, 2019).
In our 2021 research study, Violent Extremism in America: Interviews with Former Extremists and Their Families on Radicalization and Deradicalization (Brown et al., 2021), we were struck by how many of the former extremists with whom we spoke felt drawn back to radical ideological thoughts and longed for reengagement with the movements that they left. This is despite their knowing that such thoughts and behaviors are harmful to themselves and others, and despite their wanting to separate themselves from their former activities and social attachments with radical extremist groups.
We were not the first to make this observation. In 2017, Simi and colleagues analyzed a data set of 89 former U.S. white supremacists and observed "lingering" white supremacist identity and ideology that persisted long after disengagement from extremist groups (Simi et al., 2017). The researchers observed that this persistent identity and associated ideology can be described as an addiction and can be manifested in unwanted and situationally induced extremist thoughts, emotions, and physical reactions and, for some, can include relapse to extremist behavior.
As researchers who spent our graduate training studying addiction to substances and other risky behaviors (and who have since shifted our focus to studying extremism), we find merit in this hypothesis and see an uncanny parallel between hate and addiction. In our view, the parallels go beyond the return of unwanted thoughts, feelings, and behaviors. In this Perspective, we review evidence from psychology, neuroscience, sociology, and public health that suggests some similarities between extremism and addiction to substances.
Let us be clear: Our goal is not to suggest that addiction to substances and violent extremism are one and the same or that one causes the other. But by assessing some of the parallels, we hope to identify new ways to prevent hate and radicalization to violent extremism and to improve interventions that can facilitate disengagement and deradicalization. This Perspective concludes by reviewing these implications.
A growing body of evidence suggests that those participating in violent extremism have some experiences that are similar to those with substance use disorder (Simi et al., 2017) and that these experiential similarities likely are supported by common neurobiological pathways (Kimmel and Rowe, 2020; Stahl, 2015). This overlap is supported by seven main lines of evidence: (1) the role of conditioned cues, (2) the neurobiology of vengeful retaliation, (3) the role of stress in compulsive behaviors, (4) features of chronic disease, (5) psychiatric comorbidity, (6) the role of social relationships, and (7) patterns in geographical determinants.
Conditioned cues are environmental triggers that create a craving or strong drive to engage in a behavior that was associated, at one time, with a reward (Crombag et al., 2008; Pavlov, 1927). The importance of conditioned cues in substance use and substance use disorder is well understood from experimental animal models. Such studies often train rodents to press a lever in exchange for an infusion of psychoactive substances, such as cocaine or opiates, and then extinguish that behavior by ceasing to offer the drug reward. These studies show that animals reinstate drug-seeking behavior (e.g., lever-pressing) when they are exposed to "different types of drug cues," such as a light that was previously paired with substance self-administration (see Crombag et al., 2008, for review). The same phenomenon is found in studies of people with substance addiction. These studies show that individuals who are treated for substance use disorder experience a heightened sense of drug craving when they are exposed to visual cues that are associated with former drug use (Hyman et al., 2007). Other studies indicate that substance-seeking behaviors triggered by such cues lead to relapse (Powell et al., 2010).
Evidence from studies of individuals who have been involved in violent extremist movements suggests that conditioned cues play a large role in radical extremism. Simi et al. (2017) described a story about one of their research participants, named Teddy, who would see images of a Nazi flag while watching a movie and get "goose bumps" and re-experience "previous feelings and beliefs related to white supremacy" (p. 1177). Such experiences do not involve just physiological arousal, which, admittedly, could lead to either heightened attraction or aversion toward extremism. Participants in our study of deradicalized extremists described feeling pulled back toward the causes in which they were previously involved, for example, while watching the riots and street violence related to the killing of George Floyd in Minneapolis in May 2020 (Brown et al., 2021). It is not hard to imagine that such cues could cause some individuals to relapse into old patterns of behavior or even rejoin extremist groups.
Recent neurobiological models of perceived grievances indicate that engaging in vengeful retaliation shares properties with addiction to substances, as well as other behavioral addictions, such as sex and gambling addictions (Stahl, 2015). As with addiction to substances (McLellan et al., 2000), the neural architecture underlying the drive for vengeful retribution includes the nucleus acumbens and dorsal striatum and dopaminergic pathways in these and other parts of the brain that respond to reward (Kimmel and Rowe, 2020).
As with cravings that are linked to substance use disorders, vengeful retaliation can become an all-consuming drive that is only sated (temporarily) by taking an action to create harm in the perceived source of that grievance. The process inherent in this cycle of craving and desire for satiation can gradually move from goal-directed reward processes that are focused on inflicting harm to an enemy to compulsive behaviors that are linked with the stimulus (grievance) itself. As a result, engagement in radical extremist thoughts and activities quickly can become habit-forming and self-reinforcing (Stahl, 2015). It is important to note here that we are not claiming that addiction to substances leads to vengeful retaliation, but rather that these processes share some behavioral and neural pathways.
Addiction studies using nonhuman animal models have shown that stress leads to activation of craving and consumption. For example, rodents have been shown to consistently reinstate previously extinguished drug-seeking behavior when exposed to stress. Stress also has been shown to trigger drug cravings and relapse in humans (Hyman et al., 2007; see Sinha, 2007, for review).
Simi et al. (2017) highlighted the possible role of stress in mediating relapse to violent extremism. For example, the authors detailed an "irritating but relatively mundane situation" in which a former white supremacist got in an argument with a cashier of Latino descent at a fast-food restaurant (Simi et al., 2017, p. 12). The former white supremacist felt that the cashier was making unfounded accusations and ignoring the former extremist's request for a new meal. The individual got so upset that she started cursing the cashier, yelling "white power," and giving the cashier the Nazi salute. This individual regretted the behavior later and subjectively felt that she was "out of control" of her behavior at the time, an experience that similarly has been reported by recovering addicts. At the time, however, returning to old behaviors and (temporarily) to an old ideology allowed the former extremist to return to a time when she felt powerful. In this case, the acute environmental trigger for extremism relapse involved feeling ignored or powerless, a cue that seems important for radicalization in general (Hales and Williams, 2018). Our study and others have noted that feeling socially marginalized is an important part of the radicalization process (Brown et al., 2021).
Current conceptualizations of addiction to substances as a chronic disease are rooted in evidence that key features of addiction are similar to those of other chronic conditions, such as diabetes, asthma, and hypertension—for instance, all are long-term, progressive conditions that persist throughout life (McLellan et al., 2000). Similar to other chronic diseases, addiction has a heritable component. Symptoms of addiction can be managed over an individual's lifespan; similarly, individuals, including those interviewed in our research, have deradicalized and left extremist groups and given up extremist ideologies (Brown et al., 2011). And, like individuals who become addicted to substances, individuals who have deradicalized report feeling pulled or drawn back to violent extremism years—or even decades—beyond their involvement with extremist groups.
Individuals with substance addiction have high rates of other psychiatric conditions. In a foundational study, Regier et al. (1990) analyzed data from the Epidemiologic Catchment Area study, which assessed recent and lifetime mental health diagnoses in over 20,000 individuals residing in the United States. Among those with an alcohol use disorder, 36.6 percent qualified for any mental health disorder, including schizophrenia, affective and anxiety disorders, or anti-social personality disorder. For those with nonalcohol drug disorders, 53 percent qualified for a lifetime diagnosis of any mental health disorder. This compares with 22.5 percent of those surveyed who qualified for having any lifetime mental health disorder other than alcohol use. Numerous other studies track similarly high rates of mental health diagnoses in people with drug and/or alcohol addictions (Grant et al., 2004; Kessler et al., 1997; Rounsaville et al., 1991).
Our study of predominantly white nationalist extremists documented that 17 of the 32 individuals reported past mental health challenges that presented obstacles throughout the individual's life. Some identified symptoms, such as overwhelming anger, as drivers of their joining extremist organizations. Trauma or posttraumatic stress disorder, substance use, and physical health issues were also mentioned, although less frequently (Brown et al., 2021).
Other studies have noted this pattern as well. Harris-Hogan, Dawson, and Amarasingam (2020) noted that there was a surprisingly high rate of psychiatric conditions in their study of domestic jihadists in Australia and Canada, especially among those who acted on their own. In addition, Bubolz and Simi (2019) conducted in-depth interviews with 44 white supremacists and found that 57 percent of their sample reported having had mental health problems either before or during their engagement in extremist activities, and 62 percent reported having previously attempted or seriously considered suicide. The researchers also found that 73 percent of those surveyed self-reported a history of having problems with alcohol or other substances, and 59 percent reported a family history of mental illness (Bubolz and Simi, 2019). By comparison, according to the National Institute of Mental Health, as of 2019, around 16 percent of the population in the United States suffers from some form of mental illness (National Institute on Drug Abuse, 2018). Bubolz and Simi (2019) concluded their study by noting that
Individuals with mental health problems may be attracted to extremist causes because of the ideological similarities to certain types of mental health symptoms such as paranoia, elevated levels of anger, and a sense of persecution (p. 1).
Social relationships play critical roles in both addiction to substances and extremism. Social relationships can affect addiction to substances in at least two ways. First, social connections with individuals who are addicted to substances play critical roles in motivating individuals to initiate and continue use (Guise et al., 2017; Mundt, 2011). Second, as an individual becomes addicted, social relationships can evolve to sustain this use and relationships with those who are not addicted to substances weaken (see Henneberger, Mushonga, and Preston, 2021, for a systematic review on this topic).
Similar factors play out in those engaging in hate and violent extremism. Many studies have characterized how radicalization into extremist groups is an inherently social act and influenced heavily by connections with others who have already radicalized or who are coradicalizing (Bastug, Douai, and Akca, 2020; Helmus, 2009; Sageman, 2004). Likewise, as individuals' extremist views and behaviors deepen, they replace social network members who might challenge their involvement in violent extremism with others who help support the cognitive justification for experiencing grievances and seeking and enacting vengeful retaliation (Bélanger, 2021). The social stigma attached to those with hateful views further drives away those who might exert positive influence.
Substance use disorders cluster geographically; for example, while national rates of mortality from alcohol use disorders decreased between 1980 and 2014, in two-thirds of U.S. counties, mortality from alcohol use disorders increased (Dwyer-Lindgren et al., 2018). An emerging field of research examines the social determinants of health and seeks to describe the characteristics of communities in which substance use disorders cluster. Independent of personal attributes, neighborhood characteristics, such as the median income of the neighborhood in which a person lives and the distribution of incomes within a neighborhood, correlate with substance use (e.g., Galea, Alegria, and Chen, 2007) and substance use disorders (Molina, Alegría, and Chen, 2012). Recovery capital is a term that refers to a person's individual and social resources that can help them initiate and maintain recovery from substance dependence (Cloud and Granfield, 2008). Neighborhood attributes, including perceptions of safety, are part of recovery capital and affect treatment outcomes (Evans et al., 2014).
Research has identified geographic concentrations of hate groups in the United States (Goetz, Rupasingha, and Loveridge, 2012; Jefferson and Pryor, 1999; Medina et al., 2018). Medina and colleagues (2018) identify "regions of hate" in the United States on both the West Coast and East Coast and in the Central United States, and they posit that "less diversity, more poverty, less population change, and less education correlate with more hate groups" (p. 1015). In our team's related research, we learned how the lack of exposure to diversity and perceptions of marginalization contributed to individual decisions to join extremist groups (Brown et al., 2021). These are as much place-based attributes as they can be considered individual vulnerabilities. A rich history of research has described the importance of residential segregation in producing race disparities (Williams and Collins, 2001) and constraining socioeconomic mobility (Pais, 2017). Segregation is intimately linked with diversity exposure; in fact, one of the primary indices used to measure segregation across American communities is an exposure index that operationalizes the degree of potential contact between minority and majority group members within neighborhoods (Massey and Denton, 1988).
Photo by Brian Snyder/Reuters
Substance use disorders and violent extremism manifest themselves in unique ways, but both share several factors that increase individual vulnerability and perpetuate patterns of harmful behaviors. To the extent that such similarities exist, there might be unique opportunities to apply lessons from addiction research and treatment to efforts to counter hate and violent extremism.
There is a long history of research that seeks to prevent addiction to substances by acting on what appear to be fundamental causal drivers of such behavior. Three themes from this research are relevant to violent extremism and could assist with primary prevention efforts that focus on shrinking the pool of individuals who potentially are vulnerable to extremist recruitment. First, there is convincing evidence that intervening early has long-term effects on substance use. One of the best examples is the Good Behavior Game (GBG), a classroom intervention for first graders that can reduce substance abuse and dependence in young adulthood. The GBG has also been shown to reduce the risk of committing violent crime from those who engage in the game (Kellam et al., 2008). Coupling GBG-type interventions with efforts that expose children to peers of other races, ethnicities, religions, or socioeconomic status could yield similar promising results in preventing violent extremism.
Second, among adolescents, externalizing mental health symptoms (e.g., disruptive behaviors and hyperactivity) predicts future potentially problematic use of substances (Fergusson, Horwood, and Ridder, 2005; Goodman, 2010), and thus treating externalizing symptoms might be effective at preventing addiction. Identifying and addressing mental health problems early could prevent violence and extremism, but it is unclear whether evidence-based mental health treatment for those in need will independently alter or change their beliefs in hateful ideologies. Finally, identifying structural characteristics of neighborhoods that contribute to substance addiction and extremism and addressing them directly might also be a critical prevention strategy, but additional research is needed to examine the effects of geographically targeted initiatives on both sets of outcomes (e.g., Siegfried and Parry, 2019).
Prevention of substance use disorders also includes secondary prevention efforts that focus on early detection. Because addiction is a chronic condition, health care providers are encouraged to routinely screen for it as they do for other chronic conditions (McLellan et al., 2000). For addiction, this often is done in health care settings but also by family and friends. Our research also suggests that more efforts are needed to help at-risk parents and families recognize and react to signs of extremist radicalization.
Much of substance addiction mitigation focuses on tertiary prevention; that is, intervening once the chronic condition has already taken hold. Efforts to help individuals disengage from extremist social networks and ultimately deradicalize are new but could benefit from integrating approaches that have been successful for treating substance use disorders. Among those with substance use disorders, a person's "readiness to change" is instrumental for treatment to be effective (Henderson, Saules, and Galen, 2004). Motivational interviewing is a low-touch, evidence-based intervention that might increase individuals' readiness to change (Smedslund et al., 2011); it is a nonconfrontational approach that could be helpful for individuals in extremist groups or with associated ideologies to bridge to more-intensive deradicalization efforts. In addition, conceptualizing addiction as a chronic condition demands that it should be covered similarly to other chronic conditions: Evidence-based substance use disorder treatment should include continuous care regimes with no limits or restrictions, especially when individuals relapse. The struggles that people face after deradicalizing are as significant as they are for those in substance addiction recovery: Continuous supports, which might include economic and mental health supports, could be needed to discourage return to participation in extremism. Finally, those seeking treatment for addiction routinely are asked to leave their existing social networks behind and expand relationships with family and friends who are not addicted to substances (National Institute on Drug Abuse, 2018). The same advice goes for those leaving violent extremism; our research indicates that staying away from hate and extremism also requires a supportive network of like-minded individuals (Brown et al., 2021).
Although research on the neural, psychological, and social processes that underlie violent extremism is not as well developed as the study of substance use disorders and how they are treated, current evidence suggests that these behaviors might share some similar underlying causes and possibilities for mitigation. While violent extremism is less prevalent than substance use disorders, political polarization in the United States appears to be on the rise along with perceptions of grievances associated with that polarization. Sadly, this might give us more opportunities to study the draw toward vengeful retaliation and violent extremism in the future.
In the meantime, it would be wise to look back at attempts to fight terrorism and attempts to counter substance use disorders. In both cases, approaches that further stigmatize or marginalize these individuals often seem to backfire, causing the problem to worsen (Cherney, 2017; Werb, et al. 2008). And, in both cases, new approaches that incorporate community-centeredness, harm reduction, and radical forgiveness show promise at addressing what have been persistent, recalcitrant problems. Of course, there must be meaningful and proportionate consequences for behaviors that harm others, but it seems that too much punishment (or not enough support) might feed the cycle of vengeful retribution that is critical in driving domestic extremism.
This Perspective was inspired by the findings of a prior study on radical extremism that uncovered the hints of an addictive pattern in the behavior of former violent extremists. Subsequent conversations and literature review convinced us that it was important to document the parallels between behaviors that are associated with substance dependence and those that are associated with violent extremism. This Perspective is designed for researchers, practitioners, and policymakers who are involved in prevention, intervention, or policy research and practice related to violent extremism in the United States.
We appreciate the generous support for this work from the Ellen Hancock Impact Award for Social and Economic Well-Being, established by longtime RAND Social and Economic Policy Advisory Board member Ellen Hancock. We drew from Ellen's generous gift to ensure that the work reaches a variety of stakeholders and to expand the portfolio of research on this topic. The original research was funded by the National Institute of Justice. We thank our two peer reviewers, Miriam Matthews and Amarnath Amarasingam, as well as RAND quality assurance and publications staff Laurie Martin, Babitha Balan, and Lance Tan for working with us on a tight timeline.
This publication is part of the RAND Corporation Perspective series. RAND Perspectives present expert insights on timely policy issues. All RAND Perspectives undergo peer review to ensure high standards for quality and objectivity.
Our mission to help improve policy and decisionmaking through research and analysis is enabled through our core values of quality and objectivity and our unwavering commitment to the highest level of integrity and ethical behavior. To help ensure our research and analysis are rigorous, objective, and nonpartisan, we subject our research publications to a robust and exacting quality-assurance process; avoid both the appearance and reality of financial and other conflicts of interest through staff training, project screening, and a policy of mandatory disclosure; and pursue transparency in our research engagements through our commitment to the open publication of our research findings and recommendations, disclosure of the source of funding of published research, and policies to ensure intellectual independence. For more information, visit www.rand.org/about/research-integrity.
This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit www.rand.org/pubs/permissions.
The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.