Why do people and groups embrace extremism? Decades of research have failed to answer this question definitively, or more accurately, decades of research have carefully eliminated many proposed explanations.
Although many assumptions about drivers of radicalization have been disproven, policymakers and politicians cling to them nonetheless, as do people working to fight extremism on the front lines of communities where it thrives.
One of the most repeated assertions is that extremism results from structural development factors, such as poverty, high unemployment, or lack of educational opportunity. This assertion is intuitively attractive to some policymakers, in part because they have experience trying to solve structural problems and in part because such explanations soothe their anxieties about the human condition.1
But the role of structural factors has been repeatedly discredited. A study of terrorist attacks from 1986 to 2002 found no correlation between low gross domestic product and incidence of terrorism, a finding that has been replicated again and again across different measures and time frames.2
Two 2016 studies based on Islamic State foreign fighter data deeply undercut structural explanations for the group’s success. One found that countries with higher economic prosperity and lower inequality were more likely than countries without these conditions to see residents travel to Syria as foreign fighters and that unemployment was “not highly correlated” to overall foreign fighter activity.3 The other reported similar findings, noting that correlations with structural factors did appear in some places but not others. Taken as a whole across all geographic regions, the data did not support generalizations about structural development factors as drivers of extremism.4
Regarding education, a correlation exists but not the one you might think. A study of more than four thousand jihadist radicals found that their average education level was considerably higher than the general population.5 In a study of Palestinian terrorism, researchers found that higher levels of both education and economic achievement positively correlated with membership in Hamas and Palestinian Islamic Jihad.6 Despite these findings, no one advocates for reducing educational opportunities as a way to counter violent extremism.
Correlations to lower education or standard of living can be found inconsistently in very constrained social contexts and in geographically limited settings. One large-scale study found a correlation between high unemployment and high foreign fighter flows from within the Muslim world but found the opposite correlation for foreign fighters from non-Muslim countries. It should be noted that the “Muslim world” dataset included failed states and countries beset by civil wars and insurgencies. There are obvious risks in arguing for single-issue causation in settings where multiple variables may apply.7
Generally speaking, any given argument for structural causes looks better when the sample size is smaller. For instance, unemployment correlates for some cities and countries in smaller-scale studies, especially at the level of troubled neighborhoods like Belgium’s Molenbeek8 or Somali communities in Minneapolis. But neighborhoods that have an unemployment problem may not be more prone to produce violent extremists. They may simply produce violent extremists with an unemployment problem.
Billions of people around the world face problems in their personal lives and in their communities—injustice, oppression, discrimination, poverty, unemployment, crime, and more. But only a small fraction of people with problems take up extremism.
Another popular and corrosive assumption is that extremism is primarily caused by religion in general or by one religion specifically. Chapters 1 and 2 of this book document the fact that extremism is not simply a product of religious belief, nor is it confined to any one religion. Religion matters to religious extremists in the same way that race matters to racial extremists—as their particular in-group identity.
The tenets of any religion can be bent to the service of radicalization and used to fill in the contents of an ideology, even when key passages argue against violence. Christian extremists have flourished throughout history despite Jesus’s famous instruction to his disciples to turn the other cheek when they were attacked. Buddhists—renowned for nonviolence in the popular imagination—have also been involved in extremism, both historically, as when sixteenth-century Mongol rulers violently imposed Buddhism as the state religion,9 and in contemporary conflicts in Sri Lanka and in Myanmar.10
That fact that religion is not a proximate cause of extremism is not a reason to avoid studying how religion informs extremism. For religious extremists, scriptures and beliefs are sources of information that are used to define in-group and out-group identities. Understanding the details of extremist religious belief can help us understand how extremist in-groups seek to recruit from eligible in-groups, and it can help us predict specific actions that a religious extremist movement might take. But we cannot attribute extremism to religious belief without excluding or misconstruing some of the most dramatic and destructive examples in history.
The desire for simple explanations keeps many of these incorrect assumptions alive. But to understand why people become extremists and how to combat extremist violence, we must move past the old clichés and find something better.
Discussions about extremism sometimes gloss over the differences between individual radicalization and group radicalization. Although they are closely related, these are distinct processes. With the exception of wholly original ideologues—the 0.0001 percent of the 0.01 percent of people who become violent extremists—group radicalization precedes individual radicalization.
Only in extraordinarily rare cases, such as the Unabomber, can individual extremist action be separated from a clear association with a preexisting movement. Even in that case, Ted Kaczynski drew on previous works and ideas, although his concatenation was fairly unique. Kaczynski also influenced other extremists, including Norwegian mass-murderer Anders Breivik, who plagiarized the Unabomber’s manifesto in his own.11
Although radicalization almost always involves the adoption of a specific ideology, the process of adoption is more instructive than the contents of the ideology. When the study of radicalization as a process becomes fixated on contents, the results are a bewildering mess that distorts analytical efforts. A content-driven approach produces a menagerie of competing theories that make sense only in the context of a single movement at a single point in time, such as arguments that extremism is caused by colonialism or religious fundamentalism.12 When these theories are applied to other movements—even very similar movements—they often fail.
The models discussed below are therefore presented at a very high level of abstraction. But at the same time, they describe concrete criteria that can be applied to specific cases as well as to strategies for countering extremism.
The group radicalization framework presented here was derived from the study of propaganda and ideological texts produced by the white nationalist movement Christian Identity, al Qaeda, and Islamic State.13 The individual radicalization framework is derived from the study of online recruitment and behaviors for white nationalists and jihadists.14
It should be noted that the development and deployment of extremist ideologies are abstracted here, but they are not entirely abstract. Individual ideologues and extremist leaders deploy these frameworks for a variety of reasons, and they possess intangible qualities, such as charisma, that may help advance their arguments.15 For some, this may be a cynical exercise in power-seeking. Within Islamic State, for instance, some leaders were true believers, and others were simply exploiting social dynamics.16 Extremism is frequently a minority enterprise, even if leaders believe it will bring them power.
Osama bin Laden threw away his wealth and status to live in hiding and fight a hopeless jihad. It is difficult to imagine that he acted out of any conventional power-seeking motive. For others, such as Adolf Hitler, it is unclear where the desire for power ended and true belief in Nazi ideals began, but the truth almost certainly lay somewhere between those poles.17 In contrast, the demonization and subsequent extermination of the Cathars by a series of Roman Catholic popes in the eleventh century was driven by a number of purely secular and political considerations, although some element of belief came into play.18
The process of becoming an ideologue or believer and the process of leading or joining a group of extremists are ultimately individual journeys, with almost as many variations as there are participants. But that journey almost always falls within a unifying social framework of ideas and texts that we can understand and apply. As Hannah Arendt writes, citing Plato, political entities “do not spring from oak and rock,” but neither do they “spring from within our particular and individual selves.”19
The preceding chapters define the elements of group radicalization as a series of postulates that we can place in a rough sequence. Not every extremist movement will pass through these stages in exactly the same order. But in order to qualify as extremist using the definition in this book, a movement must take the following steps:
These four conditions place a movement on the extremist spectrum. Once on the spectrum, a movement can radicalize (by adopting increasingly negative views about the out-group and endorsing increasingly hostile action), or it can moderate (by mitigating its views of the out-group or shifting to a lower scale of hostile action, such as abandoning an ideological commitment to genocide in favor of segregation). As a noun, the word moderate implies a nonextremist orientation, but as a verb, it denotes movement to a less extreme orientation. Extremist groups can “moderate” without becoming “moderates.”
After these four conditions are in place, an extremist movement must flesh out each category of description by linking it to sources of information, including historical information, scriptures and prophecy, news (real and fake), analysis, and conspiracy theories. As these details are filled in, a movement’s identity construction becomes more robust. As more negative information is collected about the out-group, the scale of the crisis becomes more extreme, and the prescribed solution becomes more violent.
As the process of radicalization proceeds, the identity construct is paired with the crisis-solution construct to create the extremist value proposition:
The study of extremist recruitment practices suggests that individual radicalization is similar to any process of political mobilization but with an emphasis on identity and on crisis-solution constructs.20 An individual who becomes radicalized will usually pass through the following stages, not necessarily in this order:
Some people take years to go through the process of radicalization, from its initial stages to the decision to act; others speed through it in months or even weeks. Some individuals will leapfrog past certain steps. So-called lone wolf terrorists may skip some of these steps entirely. However, if people carry out lone-wolf-style attacks without having significantly engaged with this radicalization process, we should question whether they should be properly understood as extremist adherents or as pathological mass killers. Members of the latter category may loosely invoke an extremist movement they have not meaningfully engaged with.
One element that runs through both the ideological narratives of extremism and the expressed beliefs of individuals who have been radicalized is the idea of grievance. Grievances are a common element in extremist arguments and rationalizations, to such an extent that some models of radicalization include the adoption of grievances as a necessary element of extremism.21 Others argue that specific legitimate grievances cited by extremists—such as colonialism or oppression—are the proximate causes of extremism and terrorism.22
But as noted at the beginning of this chapter, grievances (legitimate or not) do not always lead to extremism. All extremists have grievances, but not all people with grievances become extremists. When grievances appear in extremist ideologies, they are often generalized and always framed in the context of a crisis (chapter 4).
Specific grievances do appear to play a role in mobilizing people toward violence—in this radicalization model, the stages of self-critique and decision to act. A study of 115 mass murders found that the only significant predictor of violence was “the presence of a grievance, specifically a grievance against a person or entity, as opposed to a grievance against a category of people or a grievance against an idea, movement or religion.”23 Grievance is defined here as “the cause of the offender’s distress or resentment, a perception of having been treated unfairly or inappropriately.” The study was not limited to ideological or extremist murders, although specific grievances against individuals or entities often come into play in such cases.
Identity and crisis-solution construction lay out a broad framework within which individual grievances can be contextualized as part of a sweeping identity-based conflict. This dynamic can be seen in many cases of extremist violence. For instance, Oklahoma City bomber Timothy McVeigh was exposed to The Turner Diaries (chapter 4) during the late 1980s and subsequently developed significant antigovernment views. But he was mobilized to violence by a specific event—the FBI’s botched raid on a cult compound in Waco, Texas, that left seventy-six people dead.24
Many recent jihadists were similarly mobilized by the Syrian regime’s brutal attacks on its own people, which they obsessively analyzed in specific detail through a barrage of brutal videos disseminated over social media.25
Key to the definition of grievance is the idea that it “is more than a momentary feeling of discontent or a short-lived, even explosive, expression of anger or frustration; rather, it is a conclusion reached about the reason for the offender’s suffering (or the suffering of others about whom the offender cares)”: it is “a function of concreteness—a tangible, identifiable object to which causation of suffering can be ascribed.”26
Extremist ideologies promote the legitimacy of the in-group as a collective, expanding the circle of people that adherents care about by affirming in-group identity as so robust and important that the suffering of one member must be contextualized as part of the suffering of all members. Secondarily, they argue that the out-group is the concrete cause of the in-group’s suffering, making this argument tangible through a lengthy descriptive process. Finally, extremist ideologies and propaganda point to out-group targets for violence or other hostile action, with varying levels of specificity, stipulating that the hostile action will (ultimately) ease the suffering of the in-group.
In essence, extremist ideologies weld grievances to a system of meaning in which they become both universal and personal, while insisting on hostile action to resolve the conflict. This toxic worldview can then be applied to help mobilize hostile action against specific individuals or entities associated with an out-group. In this sense, grievances and radicalization should probably understood as co-dependent or enabling, rather than a linear causation. More research on this dynamic would almost certainly produce fruitful insights.
Each of the models above shows how people and groups can escalate into extremism and then into violent extremism, but they don’t illustrate why. As the opening of this chapter suggests, the question of why people turn to extremism has been studied at great length, but many of the explanations proffered are flawed. In some cases, the evidence directly contradicts the proposed theories of why radicalization happens. In other cases, the explanations make sense only when the study is limited by ideology, chronology, geography, or all of the above.
Most of us have experienced the daunting challenge of explaining how any one human being makes decisions and develops beliefs. Sometimes we can’t even explain how we arrived at our own beliefs, so we should not expect that we can make absolute conclusions about the behavior and beliefs of other individuals and groups.
Although we can’t consistently predict behavior, we can identify tendencies that reflect various social and psychological factors that inform outcomes. Sometimes we can identify the tendency more definitively than we can explain it.
The complexity of this topic precludes a comprehensive treatment here, but two potential drivers of radicalization are supported by research and are applicable across ideological boundaries: the effects of categorization and learning bias, and the effects of disruptions to the status quo.
As is discussed in chapter 3, categorization takes place when an individual adopts a collective identity. A number of experimental and research studies have demonstrated that simply understanding oneself to be part of an in-group correlates with a tendency toward discrimination or hostility against out-groups.
This does not necessarily apply in every context. For instance, groups that were placed in competitive settings developed negative attitudes about out-groups much more strongly than groups that were placed in cooperative settings. But groups that were placed in a neutral context also developed negative attitudes toward out-groups, suggesting that the tendency to develop negative attitudes about out-groups is—to some extent—hardwired, either in the minds of individuals or as a necessary byproduct of many contexts in which we interact socially.27
The systematic construction of an identity narrative, which is seen in some mainstream groups and is very pronounced in extremist groups, exploits this tendency to develop prejudice against out-groups. An extremist ideology’s detailed descriptions of in-groups and out-groups creates prototypes for each group, which then become stereotypes when generalized. The qualities of prototypes—“attitudes, beliefs, values and behaviors”28—closely track with the identity definitions described in this book, which were derived from extremist ideological texts. As social psychologist Michael A. Hogg writes:
When we categorize someone as a member of a group we assign the group’s prototypical attributes to that person, and view them through the lens of the prototype; seeing them not as unique individuals but as more or less prototypical group members—a process called depersonalization.29
The accumulation of negative data points is enhanced by learning bias. When people have negative perceptions about something they encounter (as with out-groups), they may tend to avoid further contact with it, which inhibits them from encountering new information that might correct their negative perception.30 For extremist movements, barriers to corrective information are institutionalized when the ideology radicalizes to the point that it urges in-group members to avoid contact with the out-group.
The second major factor that helps tilt people toward extremism is uncertainty. People have a tendency to prefer the status quo. Psychology scholars John T. Jost and Roderick M. Kramer argue that
people tend to use ideas about groups and individuals to justify the way things are, so that existing social arrangements are perceived as fair and legitimate, perhaps even natural and inevitable.31
People have a general cognitive bias toward seeing the status quo as just and fair.32 Biases in favor of the status quo may arise for any number of reasons, including a desire to rationalize one’s current experience of life, systemic rewards for accepting the status quo (such as social and economic success), and the reduction of uncertainty and anxiety. Studies suggest the tendency to prefer the status quo is strong enough to overcome biases against out-groups, even when the status quo disadvantages one’s own group and provides advantages to a competing group.33 Thus, a stable status quo serves as a bulwark against radicalization.
When the status quo is overturned, people become uncertain—about their safety, their livelihoods, and their places in the world and society. The status quo can be disrupted in endless ways—including economic, technological, political, military, and social change. War, for instance, can cause great uncertainty, especially civil wars in which former friends and neighbors may suddenly become enemy combatants. Wars can cause people to relocate away from an area dominated by their in-group and into an area dominated by a different identity collective that considers them an out-group.
Revolutions in communications technology may create new opportunities for intergroup social contact or new arenas for intergroup conflict. A wave of immigration can change the demographics—and even the definition—of a national in-group, affecting the definition of preexisting overlapping group identities, such as race or religion.
On an individual level, deep personal trauma, such as job loss or the death of a loved one, can also cause uncertainty, and this is reflected in the biographies of many lone wolf extremists and other kinds of mass killers.34 Another clue can be found in religious terrorism, particularly jihadism, where religious converts are often disproportionately represented. In some cases, people who adopt religious extremism are drawn to exploring alternative religious identities due to fundamental questions about themselves and their place in the world.35
Seismic shifts erode the boundaries and definitions of in-groups and out-groups, especially when people are being physically, economically, or socially displaced. Because people tend to see the status quo as just, these dramatic changes may be perceived as unjust, overturning norms that are perceived mostly as fair and right.36
Uncertainty is uncomfortable and produces anxiety until it can be resolved. Most people deal with uncertainty in healthy ways, but many do not—enough to make a statistical difference. Although there are a number of ways people act to reduce uncertainty in their lives, one very effective strategy is to adopt a group identity that is “distinctive and clearly defined,” according to Michael A. Hogg, head of the social psychology program at Claremont Graduate University and the creator of uncertainty-identity theory, which seeks to describe the connection between uncertainty and extremism.37
At low levels, uncertainty can spur positive excitement and productive behavior, as when people are presented with a problem or challenge they think they can solve. But when people are presented with problems or threats that they think they cannot solve—because the problem is too big or because they lack the necessary resources to address it—they can adopt negative behaviors.38
In addition to its strong empirical support, the uncertainty frame has the benefit of possibly redeeming some of the disproven structural arguments highlighted at the start of this chapter. For example, poverty or unemployment may not be linked to terrorism, but sudden disruptive changes in an economic environment might create uncertainty that exacerbates in-group and out-group tensions.
People may experience uncertainty when faced with changing demographics or the introduction of new technologies. Weak states, insurgencies, and civil war create high levels of uncertainty because they divide in-groups, create new out-groups, and often defy clean resolutions. Additional historical study may shed light on these questions and help illuminate the conditions under which extremism thrives.
The elements of extremism outlined in chapters 3 and 4 are derived from the close study of extremist ideological texts. These elements intersect with the psychological needs identified by uncertainty-identity theory.
Extremist ideologies meet the need for certainty by providing a quality known as entitativity, defined as “the property of a group, resting on clear boundaries, internal homogeneity, social interaction, clear internal structure, common goals, and common fate.”39 All of these characteristics contribute to making groups feel authentic and real, and most of them are explicitly addressed by extremist ideologies. Entitativity is cultivated by extremist ideologues using the following tools:
The structure of an extremist ideology is designed to be filled with content providing a high level of entitativity, and extremist narratives tend to become more complex over time, drawing on an ever-wider variety of information sources. The more content is generated, the more entitativity accrues to the movement.
The crisis-solution construct also aims to address uncertainty. As discussed in chapter 4, extremist ideologies argue that the in-group has reached a watershed moment that requires dramatic, decisive action, which the ideology also clearly delineates.
The most common extremist crisis narratives—conspiratorial, dystopian, and apocalyptic—further address uncertainty. As noted in chapter 4, conspiracy theories are a powerful tool for “making sense” of the world. They contextualize problems that afflict an in-group by attributing them to deliberate, comprehensible actions by an out-group. Although the theories themselves are intricate, their conclusions are straightforward: the in-group’s problems stem from the out-group’s actions. Countering the out-group solves the problems.
Dystopian and apocalyptic crises are similarly clarifying and proactively explanatory.42 They describe crises so sweeping and severe that extreme solutions are a foregone conclusion, cauterizing the need for deliberation. The nature of dystopian and apocalyptic crises minimizes demand for a strategy to “win the peace.” In a dystopia, tearing down the corrupt system is a victory in itself. Because what replaces it cannot be worse, the details can be worked out at a later date.
For apocalyptic and millenarian adherents, the solution can be simpler still. Sometimes they only have to show up. Believers are mandated to take part in a cataclysmic event whose conclusion and aftermath are divinely ordained. The leaders of some apocalyptic groups may still make strategic choices, as Islamic State did in 2016 when it backed away from a prophesied battle it had previously promoted.43 But at the level of most adherents, participation is the primary requirement.
Human beings are complex, and singular explanations for their actions are often elusive. There are several avenues for additional research to help clarify these questions. But after spending some years fruitlessly debating evidence-free views about the “root causes of violent extremism” with government officials and academics (the former more than the latter), I believe the uncertainty frame holds significant promise.
Uncertainty encompasses a number of social and political situations in which extremism rises up, including some scenarios that have led policymakers to embrace wrong explanations. Uncertainty also helps explain the actions of individuals who take up terrorism and extremism, particularly so-called lone wolves, who often experience traumatic disruptions to the status quo of their lives (such as the loss of a job or a parent) prior to taking part in violence. And there is extensive empirical experimental evidence that supports the uncertainty-identity premise, in addition to the complementary findings from my study of extremist texts.
Certainly, more work remains to be done, but uncertainty-identity theory and related research by social psychologists provide a more functional and better supported explanation of extremism and its causes than any of the current alternatives.