Ideology and Threat Assessment: Law Enforcement Evaluation of Muslim and Right-Wing Extremism

Abstract

Does ideology affect assessment of the threat of violent extremism? A survey of law enforcement agencies in the United States in 2014 offers a comparison suggesting a small but statistically significant effect: Political attitudes were correlated with assessment of threats posed by Muslim extremists, and threat assessment was not correlated with the number of Muslim Americans who had engaged in violent extremism within the agency’s jurisdiction. By contrast, the perceived threat of right-wing terrorism was correlated with the number of incidents of right-wing violence and not with political attitudes. These findings reflect the context of growing polarization of attitudes toward Muslims in the United States as well as the challenge of bringing counterterrorism policies into proportion with the actual scale of violent extremism.

Keywords

terrorism law enforcement Islam Muslims United States

The study of social problems has long suggested that ideological motivations may fuel “symbolic crusades” (Gusfield 1963) and “moral panics” (Cohen 1972; Critcher et al. 2013; Goode and Ben-Yehuda 1994), generating mass responses out of proportion to the underlying problem they aim to address. A similar approach characterizes constructivist and critical approaches to criminology (DeKeseredy and Dragiewicz 2014; Garland 2001; Schuilenburg, Garland, and Hall 2016) as well as the emerging field of critical terrorism studies (Jackson, Smyth, and Gunning 2009). In this view, security policies and priorities are not direct reflections of relative threats to public safety but are mediated by ideological considerations that heighten the risks associated with some forms of threat and downplay others. With respect to Muslim extremism, arguably the defining threat of the present era, this perspective argues that certain government officials, policy experts, and sectors of the public have exaggerated the scale of the threat, spending fortunes, instituting mass surveillance, and demonizing suspect populations as a result (Kundnani 2014; Lustick 2006; J. Mueller 2006; J. Mueller and Stewart 2011, 2016; Stampnitzky 2013).

This article offers a statistical test of this hypothesis by comparing threat assessments by law enforcement agencies around the United States as recorded in a survey in the first half of 2014. The article examines the correlation of these assessments with the number of cases of Muslim and right-wing extremism in each jurisdiction over the previous dozen years and the ideological context of each law enforcement agency, both in the jurisdiction at large and among law enforcement officers. The article finds that ideological factors were associated with the assessment of threats posed by Muslim extremists but were not associated with the assessment of threats posed by right-wing extremists.

This contrast suggests that concern over Muslim extremism was ideologically motivated years before Donald Trump made this issue a central component of his presidential campaign.

Terrorist Threats in the United States Since 9/11

The attacks of September 11, 2001, transformed the assessment of terrorist threats in the United States. The killing of approximately 3,000 people in those attacks seemed to mark a new era in which relatively small groups of militants, located anywhere in the world, could cause massive destruction in the United States. Terrorism, which occupied a relatively minor place in the 2000 National Security Strategy (Clinton 2000), became the top priority of the 2002 strategy (Bush 2002). A new federal agency was created with counterterrorism responsibilities (the Transportation Security Administration), others were combined into a new Department of Homeland Security whose primary mission was to prevent terrorism in the United States (U.S. Congress 2002), and the Federal Bureau of Investigation (FBI) made terrorism its top priority (R. S. Mueller 2002)—in its previous strategic plan, terrorism was one of nine top priorities (FBI 1998).

The scale of the threat was considered massive. Seventeen months after 9/11, Robert Mueller (2003), director of the FBI, testified to Congress that “FBI investigations have revealed militant Islamics [sic] in the US. We strongly suspect that several hundred of these extremists are linked to al-Qaeda.” Estimates such as this justified expenditures of $50 billion more per year on homeland security than before 9/11 (J. Mueller and Stewart 2011).

A threat on this scale did not materialize. Between 2003 and 2012, the U.S. government averaged 40 indictments per year for offenses related to international terrorism, of which 8 per year had an al-Qaeda “affiliation or inspiration” (Department of Justice 2016). Most involved attempts to travel overseas, not plots targeting the United States; none of these cases involved coordinated, trained operatives comparable to the 9/11 plotters (Kurzman 2012, 2013, 2014a, 2014b, 2015, 2016a, 2016b, 2017; Kurzman, Schanzer, and Moosa 2011).

Beginning in 2006, the director of the FBI acknowledged the reduced threat from al-Qaeda, which “we have made great strides in disabling.” He did not mention or withdraw his earlier estimate of several hundred al-Qaeda–linked extremists in the United States, and he did not lower his assessment of the level of threat from Muslim extremists. Instead, he said that the threat was “evolving” into “smaller, more loosely-defined individuals and cells who are not affiliated with al Qaeda, but who are inspired by a violent jihadist message . . . [and] may prove to be as dangerous as groups like al Qaeda, if not more so” (R. S. Mueller 2006).

This “evolving” threat, “as dangerous as” the threats identified right after 9/11, “if not more so,” became the official assessment of federal law enforcement agencies. In 2012, for example, a year when the FBI indicted only 11 people for offenses related to international terrorism (Department of Justice 2016) and no attacks were carried out by Muslim extremists in the United States (Kurzman 2013), R. S. Mueller (2012) reaffirmed to Congress that “Counterterrorism remains our top priority. . . . The terrorist threat is more diverse than it was 10 years ago,” he explained, and “homegrown violent extremists . . . are increasingly savvy and willing to act alone, which makes them difficult to find and to stop.” James Clapper (2012), Director of National Intelligence, echoed this theme: While al-Qaeda has “weakened significantly,” some homegrown violent extremists (HVE) “have exhibited improved tradecraft and operational security and increased willingness to consider less sophisticated attacks, which suggests the HVE threat may be evolving.” Janet Napolitano (2012), Secretary of the Department of Homeland Security (DHS), made a similar point: “While we have made significant progress, threats from terrorism—including, but not limited to al-Qaeda and al-Qaeda related groups—persist and continually evolve, and the demands on DHS continue to grow.”

Federal counterterrorism efforts focused almost exclusively on Muslim extremists rather than other forms of extremism. The National Strategy for Combating Terrorism (later renamed the National Strategy for Counterterrorism) made no mention of any other form of terrorism (White House 2003, 2006, 2011b). A companion document, a primer on radicalization dynamics produced by the National Counterterrorism Center (NCTC 2012:3), noted explicitly that it “focuses on radicalization inspired by al-Qa’ida and related groups that target US Muslims. . . . There are other types of extremists—such as white supremacists and ecoterrorists—who commit violence in this country and are of concern to law enforcement, but such extremists are not specifically addressed here” (see also White House 2011c:2.)

A Control Case: Right-Wing Extremism

As it happens, one of these other types of extremism—(non-Muslim) right-wing extremism—generated a comparable number of fatalities to Muslim extremism and can serve as a control case to test for ideological influences on threat assessment.

Right-wing extremism encompasses a variety of ideologies, including anti-government, sovereign citizen, and white supremacist movements (Balleck 2014; Caiani, della Porta, and Wagemann 2012; Goldwag 2012; Zeskind 2009). The U.S. government has expressed concern about the violent potential of these movements for many years, especially after the 1995 Oklahoma City bombing, the deadliest attack by domestic terrorists in U.S. history. An FBI assessment in 2007 concluded that “Militias and sovereign citizens demonstrated intent to harm and harass federal and state judges, other officers of the court, and government employees,” although “very few went beyond paper terrorism tactics to commit direct acts of violence or terrorism.” Fewer than 350 of the 2,000 agents in the FBI’s counterterrorism division were assigned to “domestic terrorism” in 2008–2009, the only years for which figures were made public (FBI 2008:6–18; Department of Justice 2010:43). (The FBI categorizes almost all extremism by Muslims in the United States as “international terrorism,” on the presumption that it is inspired overseas.)

In 2009, the Department of Homeland Security warned of a possible upsurge in right-wing extremism (DHS 2009) but retracted the report and shut down the office that produced it after complaints from conservatives who believed the report cast unwarranted suspicion on Americans with conservative beliefs (Johnson 2012). (Similar complaints by Muslim Americans about reports on Muslim extremism did not lead to retractions or offices being shut down.) “The threat is real,” a Department of Justice trainer told local law enforcement officers at a workshop on domestic terrorism in 2014. “Since 2000, 35 law enforcement officers have been killed in 26 incidents in which at least one of the suspects was a right-wing criminal extremist” (Harris 2014). The Department of Justice has offered law enforcement training on right-wing violence since 1996, soon after the Oklahoma City bombing (Chermak, Freilich, and Shemtob 2009).

By the start of 2014, when we carried out our survey of law enforcement agencies, the amount of violence attributed to right-wing extremists was similar to or greater than violence attributed to Muslim extremists.¹ The Global Terrorism Database, maintained by the START Center at the University of Maryland (2016), lists 65 attacks in the United States associated with right-wing ideologies and 24 attacks by Muslims between 9/11 and the end of 2013, although right-wing attacks resulted in fewer fatalities (17 vs. 24).² The International Security Program at the New America Foundation (2016) identified 31 fatalities from “non-jihadist” homegrown extremists and 21 fatalities from “jihadist” extremists through the end of 2013. The Extremist Crime Database, a project tracking extremists who have committed violent crimes (Freilich et al. 2014), identified just over 200 right-wing extremists in the years 2002–2011, as compared with just over 100 extremists associated with al-Qaeda and affiliated movements (Chermak and Gruenewald 2015).³

By another count, Muslim extremists accounted for 67 violent plots aimed at targets in the United States between 9/11 and the end of 2013, including 14 attacks and 43 fatalities (Kurzman 2014b). Right-wing extremists accounted for 66 violent plots over the same period, including 31 attacks and 33 fatalities (Anti-Defamation League 2015). Using a broader definition that includes hate crimes, Arie Perliger (2012) of the Combating Terrorism Center at the U.S. Military Academy counted 3,375 right-wing attacks in the years 2002–2011, leading to 254 fatalities. (On the relationship between hate crimes and far-right ideological violence, see Adamczyk et al. 2014; Dunbar, Blanco, and Crèvecoeur-MacPhail 2016; Mills, Freilich, and Chermak forthcoming.)

Neither Muslim nor right-wing violent extremism accounted for a large proportion of violence in the United States. More than 190,000 Americans were murdered between 9/11 and the end of 2013 (FBI 2016); violent extremism was responsible for less than 1 percent of this total. (After the survey was conducted, the number of fatalities in the United States caused by Muslim extremists more than doubled, to a total of 123 between 9/11 and the end of 2016, due in large part to the mass shooting at a nightclub in Orlando, Florida, which killed 49 people, plus the attacker; Kurzman 2017. The proportion of murders due to violent extremism remained under 1 percent of the overall toll.) While the numbers have been relatively small, the comparison of Muslim and right-wing extremists provides an opportunity to identify ideological influences in the assessment of each form of threat.

Ideological Polarization

The comparison of Muslim and right-wing violent extremism takes place against a backdrop of increasing ideological polarization in the United States (Abramowitz 2010; Campbell 2016). Although most Americans agree on most issues (Fiorina 2011), partisan identities manifest themselves in a number of hot-button issues that have come to symbolize our era. One of these issues is the place of Islam in America. As recently as the fall of 2001, however, that was not the case.

Six days after 9/11, President George W. Bush visited a mosque two miles from the White House and delivered a speech praising Islam and Muslim Americans. “Like the good folks standing with me, the American people were appalled and outraged at last Tuesday’s attacks. And so were Muslims all across the world,” Bush (2011) said. “The face of terror is not the true faith of Islam. That’s not what Islam is all about. Islam is peace.” As a Christian, Bush was not especially qualified to interpret the meaning of Islam. But as a public official who considered Islamic terrorism his number one priority, Bush did not want the war on terrorism to be viewed as a war on Islam. In particular, Bush was concerned that vigilante attacks on Muslim Americans would undermine his administration’s efforts to mobilize the world’s Muslims against al-Qaeda and other revolutionary groups. Bush denounced such attacks as illegal and immoral as well as counterproductive. “Those who feel like they can intimidate our fellow citizens to take out their anger don’t represent the best of America,” Bush stated. “They represent the worst of humankind, and they should be ashamed of that kind of behavior.” Even as the Bush administration placed Muslims in the United States under special scrutiny (Bakalian and Bozorgmehr 2009), officials were careful not to demonize the community whose assistance they sought in preventing terrorism.

A series of surveys suggests that most Americans shared the Bush administration’s reluctance to denounce Muslims in general. One of the most commonly asked items on this subject was some version of the question, “Is your overall opinion of Muslims very favorable, mostly favorable, mostly unfavorable, or very unfavorable?” The question is an imperfect indicator since it forces people to generalize about an entire group in a way that may not reflect more complex judgments, but it has been asked so often in nationally representative samples—24 times since 9/11—that it allows for comparison over time. Unfavorable attitudes toward Muslims—combining somewhat unfavorable and very unfavorable, when both options were offered—rose from a median of 24 percent in the first eight surveys after 9/11 (March 2002–April 2008) to a median of 30 percent in the next 11 surveys (June 2009–August 2012) to a median of 41 percent in the next five surveys (June 2014–June 2016). These surveys are summarized in Figure 1. (An even steeper rise is visible in 33 surveys asking about favorable and unfavorable attitudes toward Islam, where unfavorable responses outnumber favorable responses beginning in 2006. The rise in unfavorable attitudes is not so pronounced in 13 surveys asking about favorable and unfavorable attitudes toward Muslim Americans; Kurzman 2014a.)

Figure 1.

Americans’ unfavorable attitudes toward Muslims since 9/11, by party affiliation.

The overall rise in unfavorable attitudes toward Muslims was accompanied by a striking rise in polarization. In surveys for which partisan breakdowns are available, the difference between Republicans’ and Democrats’ attitudes toward Muslims, also shown in Figure 1, rose from a median of 8 percentage points in 2002–2008 (seven surveys) to 20 percentage points in 2009–2012 (seven surveys) to 29 percentage points in 2014–2016 (four surveys).

This trend toward polarization is not limited to party identification but is also evident with ideological identification: The gap between self-identified conservatives and liberals widened from a median of 11 percentage points in 2002–2008 (six surveys) to 17 percentage points in 2009–2011 (four surveys), although we have found no survey data using these categories after that year.

Other data sources tell a similar story of polarization. The American National Election Studies (ANES) of 2004 and 2012, for example, included a “feeling thermometer” asking respondents to rate their feeling toward Muslims on a scale from 0 (very cold and unfavorable) to 100 (very warm and favorable). The gap in the mean rating of “extremely liberal” and “extremely conservative” respondents widened from 13 points in 2004 to 22 points in 2012 (ANES 2015). Using a different metric, sociologists Clem Brooks and Jeff Manza (2013:68, 72, 76, 160) found a growing gap between Democrats and Republicans over the years 2007–2010 in positive attitudes toward “allowing law enforcement to bring in for questioning people of certain ethnic backgrounds if these groups are thought to be more likely to engage in terrorist activities.”

This polarization coincided with a campaign to cast suspicion on Muslims in the United States as potential terrorists and on American liberals as enablers of terrorism. Funded by conservative foundations and donors, this campaign encouraged media messaging and local mobilizations to block mosque construction, introduce legislation against shari’a law, and demand profiling of Muslims by law enforcement agencies to counteract a supposed plot by Muslims to undermine the Constitution (Ali et al. 2011; Duss et al. 2015; Yazdiha 2014). The campaign also cast liberals’ defense of Muslims’ rights as a front for terrorism and a betrayal of national security (Bawer 2009; D’Souza 2007; Glazov 2009; Horowitz 2004).

Between 2001 and 2008, according to data analyses by sociologist Christopher Bail (2015), fringe anti-Muslim organizations managed to get their message increasingly into mainstream media coverage of Muslims and developed interlocking board memberships with mainstream conservative organizations. Some anti-Muslim activists, posing as experts on Islamic extremism, were hired to train law enforcement officers and military personnel (Sacirbey 2014); according to an internal FBI review, 392 training presentations contained “offensive” stereotypes of Muslims and Islam (Durbin 2012). The polarization of attitudes toward Muslims was vividly displayed in responses to these purported experts. The Obama administration, which entered office in 2009, tried to remove anti-Muslim activists from its pool of trainers in countering violent extremism (CVE), the phrase it adopted in place of counterterrorism (Cole 2012; Department of Defense 2011). By contrast, the Republican Party increasingly embraced anti-Muslim activists, who were named as advisers to all of the leading presidential campaigns in 2012 and 2016 (Kurzman 2016b; Bail 2015:98). The campaigns of Mitt Romney and Donald Trump, for example, both affiliated themselves with an anti-Muslim activist who had expressed the paranoid concern that Muslim extremists “are already in possession of weapons of mass destruction inside the United States” and may have “infiltrated the agencies that protect national security—the FBI, DHS, Defense Department, military, and other institutions” (Phares 2005:287–88).

Unfounded suspicion of American Muslims escalated to the point that many non-Muslims were willing to consider banning the practice of Islam in the United States. In a survey in October 2015, 67 percent of self-identified Democrats said Islam should be legal, 15 percent said it should be illegal, and 18 percent said they were not sure. Among self-identified Independents, the responses were 60 percent legal, 20 percent illegal, and 19 percent not sure. Only two-thirds of these subsamples considered Muslims deserving of freedom of religion, which is commonly hailed as a bedrock principle of Constitutional protection and international human rights (U.S. Commission on International Religious Freedom 2016:9). Among self-identified Republicans, this proportion was less than half: Only 40 percent said Islam should be legal, while 36 percent said it should be illegal, and 25 percent were not sure (Public Policy Polling 2015).

Local Law Enforcement and Violent Extremism

In the aftermath of 9/11, many state and local law enforcement agencies—municipal police departments, sheriffs’ offices, state departments of public safety, and other agencies—increased their emphasis on terrorism prevention. In a survey of 219 state and local law enforcement leaders in early 2004, 86 percent of respondents reported that 9/11 had inspired operational and/or policy changes in their agency (Needle 2005).

The federal government enlisted local law enforcement in this mission. In 2003, the Bush administration began to establish “fusion centers” around the United States to coordinate activities by local law enforcement officials and federal counterterrorism officials. Within a decade, more than 70 such centers were in operation (Monahan and Regan 2012; Regan and Monahan 2013, 2014; U.S. Senate 2012). In 2009, the Obama administration inaugurated the Nationwide Suspicious Activity Reporting Initiative to train local law enforcement in a federal system of information sharing on “suspicious activity” related to violent extremism. In its first three years, more than 200,000 officers received this training (Napolitano 2012). In 2011, the Obama administration announced a new strategy of community partnerships to prevent violent extremism in the United States, relying largely on local agencies because “The Federal Government will often be ill-suited to intervene in the niches of society where radicalization to violence takes place” (White House 2011a:3; see also White House 2011c).

Many local law enforcement agencies embraced the counterterrorism mission, collaborating with researchers to establish best practices for this work (International Association of Chiefs of Police 2012a, 2012b; Miller, Toliver, and Schanzer 2016; Schanzer et al. 2016). Some agencies used this mandate to experiment with community engagement (Schanzer et al. 2016; Weine et al. forthcoming); others, most notably the New York Police Department under Mayor Michael Bloomberg, went in the direction of community surveillance (Apuzzo and Goldman 2013). This difference in approaches suggests that local law enforcement agencies may hold varied ideas about the severity of threats from violent extremism, although this has not been studied directly. This article examines to what extent ideology may be associated with this variation.

Methods

As part of a survey on local law enforcement agencies’ community outreach programs, we asked agencies to rate the level of threat from violent extremism in the United States in the agency’s jurisdiction:

On a scale from 1 to 5 (5 = Severe Threat, 1 = No Threat), please rate how severe your agency believes the threat of violent extremism is within your jurisdiction, for the following forms of extremism (check one box in each column):

Al-Qaeda–Inspired Violent Extremism Other Violent Extremism

1 2 3 4 5 1 2 3 4 5

This survey was conducted in the first half of 2014, before the self-proclaimed Islamic State began to encourage violence in the United States. At the time of the survey, al-Qaeda was seen as the dominant international terrorist network, and al-Qaeda–inspired violent extremism was an umbrella term that avoided the derogatory connotations that were sometimes associated with the phrase Islamic terrorism. The sampling frame was all 480 state, county, and municipal law enforcement agencies with more than 200 sworn officers, plus 63 additional county and municipal agencies with 200 or fewer sworn officers in selected jurisdictions that experienced an incident or prosecution for violent extremism in recent years. The survey yielded responses from 339 of the larger agencies (a 71 percent response rate) and 43 of the smaller agencies (a 68 percent response rate), for a total of 382 law enforcement agencies (a 70 percent response rate), including 35 state agencies, 141 county agencies, and 206 municipal agencies, whose combined jurisdictions covered 86 percent of the U.S. population. The survey was mailed to the commanding officer of the agency, with follow-up e-mails and telephone calls. The commanding officers themselves completed 49, or one-eighth, of the surveys, generally in smaller agencies; the remainder were completed by officers selected by the commanding officer, generally individuals who directed or were involved in divisions investigating violent extremism.

To identify the ideological context in which these threat assessments were made, we merged the survey results with city-, county-, and state-level estimates of ideology calculated by political scientists Chris Tausanovitch and Christopher Warshaw (2013, 2014). Replication data are available at americanideologyproject.com (Warshaw and Tausanovitch 2015). This index, ranging from −1 (most liberal) to +1 (most conservative), draws on dozens of policy questions in seven large-scale surveys, aggregated through a multilevel regression and poststratification (MRP) model. This variable was not available for 18 small jurisdictions in the law enforcement survey, lowering the effective sample size from 382 to 364.

To measure the political attitudes of law enforcement agencies, we draw on data developed by Crowdpac.com, a technology company that has generated an index of political attitudes of law enforcement officers living in each state and county, based on publicly available data on campaign donations. The index, ranging from −10 (most liberal) to +10 (most conservative), is calculated through the campaigns and organizations that the recipients of the officers’ campaign donations have themselves contributed to as well as the candidates’ statements and (for incumbent candidates) their voting record (Bonica 2014). Using the same method, Crowdpac has also created an index of the average political attitude of the entire population within each city, county, and state. City-level population attitudes are available at https://www.crowdpac.com/games/lookup/hometown; county-level law enforcement attitudes are available at https://www.crowdpac.com/games/lookup/police; state-level law enforcement and population attitudes are available in Rohrlich (2015). Crowdpac was kind enough to provide additional data on county-level population attitudes for this project. These variables are imperfectly matched with one another: Law enforcement attitudes are available only at the county and state levels, while population attitudes are available at the city, county, and state levels. We have merged county-level law enforcement attitudes with city-level population attitudes, because many city police officers live outside of city boundaries (Silver 2014). From these variables, we generated a measure of the difference between law enforcement officers’ attitudes and population attitudes. Including this variable lowers the effective sample size from 382 to 323.

To specify the number of incidents of violent extremism occurring within each agency’s jurisdiction prior to the survey in early 2014, we draw on the two most extensive databases of Muslim American extremism and right-wing extremism, respectively. The first is Kurzman (2017), which lists the name, location, and demographic characteristics of 239 Muslim Americans involved with plots of violent extremism from 9/11 through the end of 2013, just prior to the law enforcement survey. The second data source is the list of 4,420 right-wing violent extremist incidents between 1990 and 2011 that formed the basis for the analysis in Perliger (2012); we thank Professor Perliger for sharing the location variable of each incident. Both data sets were aggregated to the level of city, county, and state and merged with the survey results.

The two data sources on violent extremism adopt different measurement approaches: The list of Muslim extremists counts both incidents of violence and plots that were disrupted before they could engage in violence, which comprise a large majority of the cases, while Perliger (2012) counts only incidents of violence. Perliger includes hate crimes and ideologically motivated property damage, while the list of Muslim extremists consists primarily of plots involving indiscriminate violence against persons. Despite these different approaches, the two data sets allow us to compare jurisdictions with less or more violent extremism of each sort—that is, jurisdictions with less Muslim extremism as against those with more Muslim extremism and jurisdictions with less right-wing extremism as against those with more right-wing extremism.

Both measures of extremism involve count data that are highly skewed and include many zero values: 73 percent of the jurisdictions whose agencies participated in the survey had experienced no cases of Muslim extremism since 9/11, and 46 percent of the jurisdictions had experienced no cases of right-wing violence. The inverse hyperbolic sine transformation is designed to handle data of this form (Burbidge, Magee, and Robb 1988), although it is typically used for dependent variables; an alternative approach, adding a positive number to the zero value cases and taking the natural log, generated similar findings. (The pattern of significant and nonsignificant coefficients is the same for logged values using additions of .01, .1, and 1.)

We also control for the population (logged) in each agency’s jurisdiction, drawn from the 2010 U.S. census (Social Explorer 2015), because larger populations may offer a greater pool for recruitment by extremists. Models involving Muslim extremism also include the estimated number of Muslims as a percentage of total population in 2010 in each jurisdiction (Grammich et al. 2012). For the 83 jurisdictions whose Muslim population is not available in this data source, we set the figure arbitrarily to .016 percent, which is the smallest value of the non-missing data as well as the value of the 10th percentile.

Ideology and Threat Assessment

State and local law enforcement agencies were less likely to identify Muslim extremism as a severe threat within their jurisdictions than other forms of violent extremism (see Figure 2). Within the “other” category, the primary ideology identified by these agencies was “anti-government” extremism, which was listed as one of the top three threats of violent extremism in 73.8 percent of the jurisdictions; this rate was almost double the next most widely identified threat, Muslim extremism, which was listed as one of the top three in 39.3 percent of the jurisdictions (Table 1).

Figure 2.

Threat assessment of violent extremism by state and local law enforcement agencies, 2014.

Table 1.

State and Local Law Enforcement Assessment of Top Threats of Violent Extremism, 2014.What are the main violent extremist threats that your agency faces? Please check up to three items below. If your agency does not consider any violent extremist threats to be severe or somewhat severe, please mark “NA–Not Applicable.”

Form of Violent Extremism	Percent of Agencies Listing This Form among the Top Three Threats of Violent Extremism
Anti-government	73.8
Al-Qaeda inspired	39.3
Environmental	33.5
Racist	24.3
Anti-capitalist	14.7
Not applicable	13.6
Other	10.5
No response	2.9
Total N	382

This finding replicates threat assessments in two earlier surveys of law enforcement agencies. In 2002, 85 percent of state agencies and 17 percent of local agencies reported that right-wing terrorist groups were located in their jurisdiction; 38 percent and 3 percent, respectively, reported knowledge of “religious groups using violence” in their jurisdiction (Riley et al. 2005:9). In 2013–2014, a survey of 259 law enforcement officers ranked sovereign citizens as the greatest potential threat (mean response of 3.20 on a scale of 1 [strongly disagree] to 4 [strongly agree]; Islamic extremists/jihadists were ranked second with a mean response of 2.89 (Carter et al. 2014:7). One smaller survey of 37 state police agencies in 2006–2007 found the reverse, with Islamic extremists/jihadists rated as a greater threat to state security than right-wing groups (Freilich, Chermak, and Simone 2009:462).

Switching to multivariate regression with threat assessment as the dependent variable (Table 2), we find that threat assessment of al-Qaeda–inspired violent extremism is correlated with the number of Muslim violent extremists in a jurisdiction (Model 1) and that threat assessment of other forms of violent extremism is correlated with the number of incidents of right-wing violent extremism (Model 4). Logged population size is also correlated with threat assessment, as expected, and the number of Muslims as a percent of the total population of the jurisdiction is correlated with threat assessment of al-Qaeda–inspired violent extremism.

Table 2.

State and Local Law Enforcement Threat Assessment, Violent Extremism, and Political Attitudes.

Dependent Variable	Threat Assessment of al-Qaeda–Inspired Violent Extremism, 2014			Threat Assessment of Other Forms of Violent Extremism, 2014
Dependent Variable	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Independent variables
Muslim American violent extremism (individuals, 2001–2013, inverted hyperbolic sine)	.160* (.064)	.124(.071)	.130(.070)
Right-wing violence (incidents, 1990–2011, inverted hyperbolic sine)				.113** (.040)	.148*** (.044)	.157*** (.045)
Political attitude in jurisdiction (–1 liberal to +1 conservative)		−.635*** (.174)	−.503** (.184)		−.230(.184)	−.056(.200)
Difference between law enforcement officers’ attitudes and population attitudes in jurisdiction			.040* (.019)			.028(.020)
Population (2010, logged)	.151*** (.038)	.138** (.048)	.116* (.052)	.231*** (.033)	.238*** (.040)	.227*** (.042)
Muslims (percent of population, 2010)	.184*** (.054)	.155** (.054)	.190*** (.054)
Constant	.385(.470)	.539(.601)	.752(.651)	−.549(.420)	−.189(.521)	−.061(.555)
Adjusted R²	.15	.15	.16	.13	.12	.10
N	382	364	323	382	364	323

Note: Standard errors in parentheses.

p < .05. **p < .01. ***p < .001.

Introducing ideology, however, has different effects on threat assessment. For al-Qaeda–inspired violent extremism (Model 2), political attitudes of the jurisdiction are a strong predictor of threat assessment, even stronger than the number of Muslim violent extremists, whose coefficient falls to marginal statistical significance (p = .07). For other forms of violent extremism (Model 5), political attitudes do not affect threat assessment, which remains strongly correlated with the number of right-wing violent incidents in the jurisdiction. (Another study has found evidence that conservative attitudes in a county are associated with the presence of far-right ideological violence [Adamczyk et al. 2012], although we are unable to replicate this finding with our data.)

The law enforcement agencies in our survey reported lower levels of threat in conservative jurisdictions than in liberal jurisdictions. A shift of one point in the political ideology scale (e.g., from the most liberal to the midpoint of the scale) is associated with a lower threat assessment of two-thirds of a point, on average, on a five-point scale.

This pattern holds in subsets of the sample as well, such as liberal jurisdictions only (political attitudes scale <0), conservative jurisdictions only (political attitudes scale >0), and middle-of-the-road jurisdictions only (political attitudes scale between −0.5 and +0.5): More conservative jurisdictions, even within these subsamples, are associated with lower threat assessments for Muslim extremism, controlling for the actual number of Muslim extremists, and are not significantly correlated with threat assessments for other extremisms, controlling for the actual number of right-wing incidents.

The pattern also holds with the percentage of votes cast for Obama in 2012 and with an indicator of the party in office. Agencies reporting to a Republican mayor, sheriff, or governor (or county board or executive for county agencies with appointed sheriffs) were marginally more likely (p = .06) to rate threats from Muslim extremism lower than agencies led by Democrats and Independents, controlling for the actual number of Muslim extremists. Party affiliation did not matter for the assessment of other forms of violent extremism. (A dummy variable for elected sheriffs was not significant for either form of threat.)

The pattern holds as well with a direct indicator of law enforcement officers’ political attitudes, as calculated by Crowdpac from campaign donation data, although this indicator is not available for all of the jurisdictions surveyed for this project. As shown in Table 2, threat assessment of Muslim extremism was associated with the difference between the political attitudes of law enforcement officers and the jurisdiction at large: Agencies where officers were more conservative than the population they served rated the threat of Muslim extremism higher than agencies where officers were more liberal than the population (Model 3). The difference between law enforcement attitudes and the population’s attitudes was not significantly associated with the assessment of other forms of violent extremism (Model 6).

Threat Assessment and Subsequent Violent Extremism

Law enforcement agencies’ threat assessments may have been based in part on ongoing investigations of plots that had not yet yielded arrests or otherwise come to public attention. These cases would not have appeared in the data sets of violent extremism.

To test whether this was the case, we counted Muslim Americans associated with violent extremism in 2015, the year following the survey (Kurzman 2016a). Cases were more numerous in 2015 than in any year since 9/11, spurred by online recruitment by the self-proclaimed “Islamic State,” although the pace of new cases dropped off sharply in the middle of the year (Kurzman 2016a, 2017). Most of these cases involved arrests of individuals attempting to join militant groups overseas. No comprehensive list of right-wing violent extremism was available for 2015, comparable to the Perliger (2012) data set for earlier years.

The number of Muslim American extremists in 2015 in each jurisdiction was strongly correlated with the number of cases in previous years as well as with logged population (Table 3, Model 1). The size of the Muslim population, net of total population, was not correlated with the number of cases in 2015—it was not correlated with the number of cases in previous years either, although it was positively associated with law enforcement’s threat assessment for al-Qaeda–inspired violent extremism. Law enforcement agencies’ 2014 threat assessment, added in Model 2, is not significantly correlated with the number of cases in 2015. A likelihood-ratio (LR) test indicates that adding this variable does not improve model fit (LR chi-square = .95, p = .33).

Table 3.

State and Local Law Enforcement Threat Assessment and Subsequent Violent Extremism.

Dependent Variable	Muslim Americans Involved with Violent Extremism, 2015 (Inverted Hyperbolic Sine)
Dependent Variable	Model 1	Model 2
Muslim American violent extremism (individuals, 2001–2013, inverted hyperbolic sine)	.359*** (.027)	.356*** (.027)
Population (logged)	.070*** (.016)	.067*** (.016)
Muslims (percent of population, 2010)	.008(.022)	.004(.023)
Threat assessment of al-Qaeda–inspired violent extremism, 2014		.021(.021)
Constant	−.845*** (.195)	−.853*** (.195)
Adjusted R²	.49	.49
N	382	382

Note: Standard errors in parentheses.

p < .05. **p < .01. ***p < .001.

Whatever nonpublic information about al-Qaeda–inspired violent extremism that law enforcement agencies had in 2014, therefore, did not help them to predict cases of Muslim extremism in 2015.

Conclusion

Four observations emerge from these findings. First, we note that ideology mattered in the assessment of threats from Muslim extremism. Law enforcement agencies in liberal jurisdictions rated the threat of Muslim extremism higher than agencies in more conservative areas. Agencies that were more conservative than the public they served rated threats from Muslim extremism more highly than agencies in which officers were no more conservative than the jurisdiction at large. We do not have evidence in these data sets to explain why this might be so, but we do have evidence that this assessment was correlated more strongly with ideological factors than with the number of Muslim extremists in the jurisdiction.

Second, ideology did not matter in assessment of threats from other forms of violent extremism. These assessments were not the inverse of the threat from Muslim extremism: They were not higher in conservative areas than liberal areas. Instead, threat assessments hewed consistently to the number of incidents of right-wing violence in the jurisdiction. This is particularly notable because of the definitional mismatch between the indicator of violence (right-wing violence only) and the threat assessment category of “other violent extremism,” which is not limited to right-wing extremism (although that was the most common form of extremism identified by the law enforcement agencies). Despite the statistical noise that we would expect from this mismatch, threat assessment in this “other” category was robustly correlated with the prevalence of right-wing violence.

Third, the contrast between the two forms of extremism offers a telling comparison of the effect of ideology on threat assessment. Ideology does not matter equally for all forms of threat assessment. In the context of increasing polarization of attitudes toward Muslims in America, ideological factors are associated with the perception of threat posed by Muslim extremists but not with perception of threat posed by other extremists. This ideological context outweighs the effect of evidence of extremism by Muslims, but it does not outweigh evidence of right-wing extremism. We do not offer a theory of threat assessment, but we note that these findings support the suggestion from the field of critical terrorism studies that American responses to Muslim extremism are related more to ideological considerations than to the observable scale of the threat.

Finally, law enforcement’s threat assessment of al-Qaeda–inspired violent extremism did not predict the number of Muslims involved in violent extremism the following year, controlling for the number of cases in previous years. This finding reinforces the impression that threat assessment regarding Muslim extremists may be detached from the actual level of threat, a “moral panic” in sociological terms. “The United States is obsessed with terrorism, to an unhealthy and illogical degree,” a recent analysis concluded. “When politicians attempt to calm those fears, to put terrorism into perspective, they are accused of ignoring danger, of coddling the enemy. Terrorists want attention; our hyper-sensitivity to their violence feeds that need.” This judgment came not from the field of critical terrorism studies but from the former director of analysis at the National Counterterrorism Center and deputy director of the CIA’s Counterterrorism Center (Liepman and Mudd 2016:14).

Footnotes

Acknowledgements

We thank David Schanzer of Duke University; Jessica Toliver and Elizabeth Miller of the Police Executive Research Forum, who conducted the survey; Crowdpac and the American Ideology Project (Chris Tausanovitch and Christopher Warshaw), who provided data on political ideology; and Arie Perliger, who provided data on right-wing violence.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The survey on which this article is based was funded by the National Institute of Justice (grant 2012-ZA-BX-0002).

1

In late 2014 and early 2015, after the survey was completed, there was an uptick in violence attributed to Muslim extremists, although the scale of violence remained low ().

2

After a review of all of the incident descriptions and notes in the data set, right-wing ideologies were identified by the keywords abortion, Ku Klux Klan, Minutemen, Nazi, white, supremacist, antigovernment, pro-life, hate, gun, and anti-IRS and by targets including reproductive, health, mosque, Obama, liberal, Holocaust, and Sikh. Muslim extremism was identified by the keywords Muslim, Arab, Taliban, Islam, Allah, Osama, Nidal, and Palestinan and by targets including Israel and marathon.

3

present totals for the period 1990–2011 as well as demographic characteristics for the entire period, the pre-9/11 period, and the post-9/11 period. We estimated the number of extremists post-9/11 by comparing the percentage male in each period.

Author Biographies

Charles Kurzman is a professor of sociology at the University of North Carolina at Chapel Hill studying political sociology with a focus on the Middle East and Muslim communities.

Ahsan Kamal is a doctoral student at the University of North Carolina at Chapel Hill studying sociology of development, environment, and social movements. He is also a faculty member at the National Institute of Pakistan Studies, Quaid-e-Azam University, Islamabad.

Hajar Yazdiha is a doctoral student at the University of North Carolina at Chapel Hill studying cultural processes in contentious politics. She is joining the faculty at University of Southern California as a postdoctoral fellow in 2017 and an assistant professor in 2018.

References

Abramowitz

Alan

. 2010. The Disappearing Center: Engaged Citizens, Polarization, and American Democracy. New Haven, CT: Yale University Press.

Adamczyk

Amy

Freilich

Joshua D.

Chermak

Steven M.

Parkin

William S.

2012. Examining the Relationship between General Social Survey (GSS) Measures and Far-right Ideological Violence: A County-level Analysis. College Park, MD: START Center, University of Maryland.

Adamczyk

Amy

Gruenewald

Jeff

Chermak

Steven M.

Freilich

Joshua D.

2014. “The Relationship between Hate Groups and Far-right Ideological Violence.” Journal of Contemporary Criminal Justice 30(3):310–22.

Ali

Wajahat

Clifton

Eli

Duss

Matthew

Fang

Lee

Keyes

Scott

Shakir

Faiz

. 2011. Fear, Inc.: The Roots of the Islamophobia Network in America. Washington, DC: Center for American Progress.

American National Election Studies. 2015. Time Series Cumulative Data File, 1948–2012. Stanford, CA: Stanford University, and Ann Arbor: University of Michigan.

Anti-Defamation League. 2015. Terrorist Conspiracies, Plots and Attacks by Right-wing Extremists, 1995–2015. New York: Anti-Defamation League.

Apuzzo

Matt

Goldman

Adam

. 2013. Enemies within: Inside the NYPD’s Secret Spying Unit and Bin Laden’s Final Plot against America. New York: Simon & Schuster.

Bail

Christopher

. 2015. Terrified: How Anti-Muslim Fringe Organizations Became Mainstream. Princeton, NJ: Princeton University Press.

Bakalian

Anny P.

Bozorgmehr

Mehdi

. 2009. Backlash 9/11. Berkeley: University of California Press.

10.

Balleck

Barry

. 2014. Allegiance to Liberty: The Changing Face of Patriots, Militias, and Political Violence in America. Santa Barbara, CA: Praeger.

11.

Bawer

Bruce

. 2009. Surrender: Appeasing Islam, Sacrificing Freedom. New York: Doubleday.

12.

Bonica

Adam

. 2014. “Mapping the Ideological Marketplace.” American Journal of Political Science 58(2):367–87.

13.

Brooks

Clem

Manza

Jeff

. 2013. Whose Rights? Counterterrorism and the Dark Side of American Public Opinion. New York: Russell Sage Foundation.

14.

Burbidge

John B.

Magee

Lonnie

Robb

A. Leslie

. 1988. “Transformations to Handle Extreme Values of the Dependent Variable.” Journal of the American Statistical Association 83(401):123–27.

15.

Bush

George W.

2001. “Islam Is Peace” Says President: Remarks by the President at Islamic Center of Washington, D.C. Washington, DC: The White House.

16.

Bush

George W.

2002. The National Security Strategy for the United States of America. Washington, DC: The White House.

17.

Caiani

Manuela

Porta

Donatella della

Wagemann

Claudius

. 2012. Mobilizing on the Extreme Right: Germany, Italy, and the United States. Oxford, UK: Oxford University Press.

18.

Campbell

James E.

2016. Polarized: Making Sense of a Divided America. Princeton, NJ: Princeton University Press.

19.

Carter

David

Chermak

Steve

Carter

Jeremy

Drew

Jack

. 2014. “Understanding Law Enforcement Intelligence Processes.” Report to the Office of University Programs, Science and Technology Directorate, U.S. Department of Homeland Security. College Park, MD: University of Maryland START Center.

20.

Chermak

Steven M.

Freilich

Joshua D.

Shemtob

Zachary

. 2009. “Law Enforcement Training and the Domestic Far Right.” Criminal Justice and Behavior 36(12):1305–22.

21.

Chermak

Steven M.

Gruenewald

Jeffrey A.

2015. “Laying a Foundation for the Criminological Examination of Right-wing, Left-wing, and Al Qaeda–Inspired Extremism in the United States.” Terrorism and Political Violence 27(1):133–59.

22.

Clapper

James R.

2012. Unclassified Statement for the Record on the Worldwide Threat Assessment of the US Intelligence Community for the Senate Select Committee on Intelligence, January 31, 2012. Washington, DC: Office of the Director of National Intelligence.

23.

Clinton

William J.

2000. A National Security Strategy for a Global Age. Washington, DC: The White House.

24.

Cohen

Stanley

. 1972. Folk Devils and Moral Panics. London: MacGibbon and Kee.

25.

Cole

James M.

2012. Memorandum for Heads of Components and United States Attorneys: Training Guiding Principles. Washington, DC: U.S. Department of Justice, Office of the Deputy Attorney General.

26.

Critcher

Cris

Hughes

Jason

Petley

Julian

Rohloff

Amanda

, eds. 2013. Moral Panics in the Contemporary World. New York: Bloomsbury.

27.

DeKeseredy

Walter S.

Dragiewicz

Molly

, eds. 2014. Critical Criminology, 4 vols. Abingdon, UK: Routledge.

28.

Department of Defense. 2011. “Screening Process for Countering Violent Extremism (CVE) Trainers and Speakers.” Retrieved April 3, 2017 (http://dailycaller.com/wp-content/uploads/2011/11/Joint-Staff-Action-Trainers.pdf).

29.

Department of Homeland Security. 2009. Rightwing Extremism: Current Economic and Political Climate Fueling Resurgence in Radicalization and Recruitment. Washington, DC: Department of Homeland Security, Office of Intelligence and Analysis.

30.

Department of Justice. 2010. Follow-up Audit of the Federal Bureau of Investigation Personnel Resource Management and Casework (Audit Report 10-24). Washington, DC: Department of Justice, Office of the Inspector General, Audit Division.

31.

Department of Justice. 2016. National Security Division Chart of Public/Unsealed International Terrorism and Terrorism-related Convictions from 9/11/01 to 12/31/14. Washington, DC: U.S. Department of Justice.

32.

D’Souza

Dinesh

. 2007. The Enemy at Home: The Cultural Left and Its Responsibility for 9/11. New York: Doubleday.

33.

Dunbar

Edward

Blanco

Amalio

Crèvecoeur-MacPhail

Desirée A.

, eds. 2016. The Psychology of Hate Crimes as Domestic Terrorism. Santa Barbara, CA: ABC-CLIO.

34.

Durbin

Richard J.

2012. “Letter to the Honorable Robert Mueller, Director, Federal Bureau of Investigation.” Retrieved April 3, 2017 (https://www.wired.com/images_blogs/dangerroom/2012/03/FBI-training-letter.pdf).

35.

Duss

Matthew

Taeb

Yasmine

Gude

Ken

Sofer

Ken

. 2015. Fear, Inc. 2.0: The Islamophobia Network’s Efforts to Manufacture Hate in America. Washington, DC: Center for American Progress.

36.

Federal Bureau of Investigation. 1998. Strategic Plan: 1998–2003. Washington, DC: Government Printing Office.

37.

Federal Bureau of Investigation. 2007. A Threat Assessment for Domestic Terrorism, 2005–2006. Washington, DC: Federal Bureau of Investigation, Counterterrorism Division.

38.

Federal Bureau of Investigation. 2008. FY 2009 Congressional Budget Submission. Washington, DC: Federal Bureau of Investigation.

39.

Federal Bureau of Investigation. 2016. Crime in the United States, 2015. Washington, DC: Federal Bureau of Investigation.

40.

Fiorina Morris

, with Abrams

Samuel J.

Pope

Jeremy C.

2011. Culture War? The Myth of a Polarized America. 3rd ed. Boston, MA: Longman.

41.

Freilich

Joshua D.

Chermak

Steven

Gruenewald

Jeffrey A.

Parkin

William

Belli

Roberta

. 2014. “Introducing the United States Extremist Crime Database (ECDB).” Terrorism and Political Violence 26(2):372–84.

42.

Freilich

Joshua D.

Chermak

Steven M.

Simone

Joseph

Jr.

2009. “Surveying American State Police Agencies about Terrorism Threats, Terrorism Sources, and Terrorism Definitions.” Terrorism and Political Violence 21(3):450–75.

43.

Garland

David

. 2001. The Culture of Control: Crime and Social Order in Contemporary Society. Chicago, IL: University of Chicago Press.

44.

Glazov

Jamie

. 2009. United in Hate: The Left’s Romance with Tyranny and Terror. Los Angeles, CA: WorldNetDaily.

45.

Goldwag

Arthur

. 2012. The New Hate: A History of Fear and Loathing on the Populist Right. New York: Pantheon Books.

46.

Goode

Erich

Ben-Yehuda

Nachman

. 1994. “Moral Panics: Culture, Politics, and Social Construction.” Annual Review of Sociology 20:149–71.

47.

Grammich

Clifford

Hadaway

Kirk

Houseal

Richard

Jones

Dale E.

Krindatch

Alexei

Stanley

Richie

Taylor

Richard H.

2012. 2010 U.S. Religion Census: Religious Congregations & Membership Study. Lenexa, KS: Association of Statisticians of American Religious Bodies.

48.

Gusfield

Joseph R.

1963. Symbolic Crusade. Status Politics and the American Temperance Movement. Urbana: University of Illinois Press.

49.

Harris

Robert F.

2014. Domestic Terrorism/Sovereign Citizen Overview. Tallahassee, FL: Institute for Intergovernmental Research.

50.

Horowitz

David

. 2004. Unholy Alliance: Radical Islam and the American Left. Washington, DC: Regnery.

51.

International Association of Chiefs of Police. 2012a. A Common Lexicon. Alexandria, VA: IACP Committee on Terrorism, Countering Violent Extremism (CVE) Working Group.

52.

International Association of Chiefs of Police. 2012b. Radicalization 101. Alexandria, VA: IACP Committee on Terrorism.

53.

Jackson

Richard

Smyth

Marie Breen

Gunning

Jeroen

, eds. 2009. Critical Terrorism Studies: A New Research Agenda. New York: Routledge.

54.

Johnson

Daryl

. 2012. Right-wing Resurgence: How a Domestic Terrorist Threat Is Being Ignored. Lanham, MD: Rowman & Littlefield.

55.

Kundnani

Arun

. 2014. The Muslims Are Coming! Islamophobia, Extremism, and the Domestic War on Terror. London: Verso.

56.

Kurzman

Charles

. 2012. Muslim-American Terrorism in the Decade Since 9/11. Durham, NC: Triangle Center on Terrorism and Homeland Security.

57.

Kurzman

Charles

. 2013. Muslim-American Terrorism: Declining Further. Durham, NC: Triangle Center on Terrorism and Homeland Security.

58.

Kurzman

Charles

. 2014a. “Anti-Muslim Sentiment Rising in the U.S: What Is Happening to Religious Tolerance?” IslamiCommentary. Retrieved April 3, 2017 (https://islamicommentary.org/2014/02/anti-muslim-sentiment-rising-in-the-u-s-what-is-happening-to-religious-tolerance/).

59.

Kurzman

Charles

. 2014b. Muslim-American Terrorism in 2013. Durham, NC: Triangle Center on Terrorism and Homeland Security.

60.

Kurzman

Charles

. 2015. Terrorism Cases Involving Muslim-Americans in 2014. Durham, NC: Triangle Center on Terrorism and Homeland Security.

61.

Kurzman

Charles

. 2016a. Muslim-American Involvement with Violent Extremism, 2015. Durham, NC: Triangle Center on Terrorism and Homeland Security.

62.

Kurzman

Charles

. 2016b. “When Republicans Needed Muslim Allies.” Retrieved April 3, 2017 (http://kurzman.unc.edu/when-republicans-needed-muslim-allies).

63.

Kurzman

Charles

. 2017. Muslim-American Involvement with Violent Extremism, 2016. Durham, NC: Triangle Center on Terrorism and Homeland Security.

64.

Kurzman

Charles

Schanzer

David

Moosa

Ebrahim

. 2011. “Muslim-American Terrorism Since 9/11: Why So Rare?” The Muslim World 101(3):464–83.

65.

Liepman

Andrew

Mudd

Philip

. 2016. “Lessons from the Fifteen-year Counterterrorism Campaign.” CTC Sentinel 9(10):12–15.

66.

Lustick

Ian S.

2006. Trapped in the War on Terror. Philadelphia: University of Pennsylvania Press.

67.

Miller

Elizabeth

Toliver

Jessica

Schanzer

David

. 2016. Promising Practices for Using Community Policing to Prevent Violent Extremism. Washington, DC, and Durham, NC: Police Executive Research Forum and Triangle Center on Terrorism and Homeland Security.

68.

Mills

Colleen E.

Freilich

Joshua D.

Chermak

Steven M.

Forthcoming. “Extreme Hatred: Revisiting the Hate Crime and Terrorism Relationship to Determine Whether They Are ‘Close Cousins’ or ‘Distant Relatives.’” Crime and Delinquency.

69.

Monahan

Torin

Regan

Priscilla M.

2012. “Zones of Opacity: Data Fusion in Post-9/11 Security Organizations.” Canadian Journal of Law and Society 27(3):301–17.

70.

Mueller

John

. 2006. Overblown: How Politicians and the Terrorism Industry Inflate National Security Threats, and Why We Believe Them. New York: Free Press.

71.

Mueller

John

Stewart

Mark G.

2011. Terror, Security, and Money: Balancing the Risks, Benefits, and Costs of Homeland Security. New York: Oxford University Press.

72.

Mueller

John

Stewart

Mark G.

2016. Chasing Ghosts: The Policing of Terrorism. New York: Oxford University Press.

73.

Mueller

Robert S. III.

2002. The New Focus: Testimony before the House of Representatives Subcommittee for the Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies, Committee on Appropriations, Washington, DC, June 21, 2002. Washington, DC: Federal Bureau of Investigation.

74.

Mueller

Robert S.

III

2003. War on Terrorism: Testimony Before the Select Committee on Intelligence of the United States Senate, Washington, DC, February 11, 2003. Washington, DC: Federal Bureau of Investigation.

75.

Mueller

Robert S.

III

2006. The Threat of Homegrown Terrorism: The City Club of Cleveland, Cleveland, Ohio, June 23, 2006. Washington, DC: Federal Bureau of Investigation.

76.

Mueller

Robert S.

III

2012. Statement of Robert S. Mueller, III, Director, Federal Bureau of Investigation, before the Committee on the Judiciary, United States Senate, at a Hearing Entitled “Oversight of the Federal Bureau of Investigation,” Presented May 16, 2012. Washington, DC: Federal Bureau of Investigation.

77.

Napolitano

Janet

. 2012. Written Testimony of DHS Secretary Janet Napolitano for a Senate Committee on Homeland Security and Governmental Affairs Hearing on the President’s Fiscal Year 2013 Budget Request for the Department of Homeland Security. Washington, DC: Department of Homeland Security.

78.

National Counterterrorism Center. 2012. Radicalization Dynamics: A Primer. Washington, DC: National Counterterrorism Center.

79.

Needle

Jerry

. 2005. Post 9-11 Policing: The Crime Control-Homeland Security Paradigm: Taking Command of New Realities. Alexandria, VA: International Association of Chiefs of Police.

80.

New America Foundation. 2016. Terrorism in America after 9/11. Washington, DC: New American Foundation, International Security Program.

81.

Perliger

Arie

. 2012. Challengers from the Sidelines: Understanding America’s Violent Far Right. West Point, NY: Combating Terrorism Center, U.S. Military Academy.

82.

Phares

Walid

. 2005. Future Jihad: Terrorist Strategies against the West. New York: St. Martin’s Griffin.

83.

Public Policy Polling. 2015. National Survey Results, October 1–4, 2015. Raleigh, NC: Public Policy Polling.

84.

Regan

Priscilla M.

Monahan

Torin

. 2013. “Beyond Counterterrorism: Data Sharing, Privacy, and Organizational Histories of DHS Fusion Centers.” International Journal of E-Politics 4(3).

85.

Regan

Priscilla M.

Monahan

Torin

. 2014. “Fusion Center Accountability and Intergovernmental Information Sharing.” Publius 45(1):39–59.

86.

Riley

K. Jack

Treverton

Gregory F.

Wilson

Jeremy M.

Davis

Lois M.

2005. State and Local Intelligence in the War on Terrorism. Santa Monica, CA: Rand.

87.

Rohrlich

Justin

. 2015. “Love the Cops? That Might Depend on Where You Live.” Vice News, (July 31). Retrieved April 3, 2017 (https://news.vice.com/article/love-the-cops-that-might-depend-on-where-you-live).

88.

Sacirbey

Omar

. 2014. “Anti-Muslim Speakers Still Popular in Law Enforcement Training.” The Washington Post (March 12).

89.

Schanzer

David

Kurzman

Charles

Toliver

Jessica

Miller

Elizabeth

. 2016. The Challenge and Promise of Using Community Policing Strategies to Prevent Violent Extremism. Durham, NC: Triangle Center on Terrorism and Homeland Security.

90.

Schuilenburg

Marc

Garland

David

Hall

George

. 2016. The Securitization of Society: Crime, Risk, and Social Order. New York: New York University Press.

91.

Silver

Nate

. 2014. “Most Police Don’t Live in the Cities They Serve.” Fivethirtyeight.com (August 20). Retrieved April 3, 2017 (https://fivethirtyeight.com/datalab/most-police-dont-live-in-the-cities-they-serve).

92.

Social Explorer. 2015. Census 2010. New York: Oxford University Press.

93.

Stampnitzky

Lisa

. 2013. Disciplining Terror: How Experts Invented “Terrorism.” Cambridge, UK: Cambridge University Press.

94.

START Center. 2016. Global Terrorism Database. College Park, MD: University of Maryland.

95.

Tausanovitch

Chris

Warshaw

Christopher

. 2013. “Measuring Constituent Policy Preferences in Congress, State Legislatures, and Cities.” The Journal of Politics 75(2):330–42.

96.

Tausanovitch

Chris

Warshaw

Christopher

. 2014. “Representation in Municipal Government.” American Political Science Review 108(3):605–41.

97.

U.S. Commission on International Religious Freedom. 2016. Annual Report of the U.S. Commission on International Religious Freedom. Washington, DC: USCIRF.

98.

U.S. Congress. 2002. Public Law 107–296, an Act to Establish the Department of Homeland Security, and for Other Purposes. Washington, DC: Government Printing Office.

99.

U.S. Senate. 2012. Federal Support for and Involvement in State and Local Fusion Centers: Majority and Minority Staff Report. Washington, DC: United States Senate, Committee on Homeland Security and Governmental Affairs, Permanent Subcommittee on Investigations.

100.

Warshaw

Chris

Tausanovitch

. 2015. Replication Data for: Representation in Municipal Government. Version 1. Cambridge, MA: Harvard Dataverse.

101.

Weine

Stevan

Eisenman

David P.

Kinsler

Janni

Glik

Deborah C.

Polutnik

Chloe

. Forthcoming. “Addressing Violent Extremism as Public Health Policy and Practice.” Behavioral Sciences of Terrorism and Political Aggression.

102.

White House. 2003. National Strategy for Combating Terrorism. Washington, DC: Executive Office of the President of the United States.

103.

White House. 2006. National Strategy for Combating Terrorism. Washington, DC: Executive Office of the President of the United States.

104.

White House. 2011a. Empowering Local Partners to Prevent Violent Extremism in the United States. Washington, DC: Executive Office of the President of the United States.

105.

White House. 2011b. National Strategy for Counterterrorism. Washington, DC: Executive Office of the President of the United States.

106.

White House. 2011c. Strategic Implementation Plan for Empowering Local Partners to Prevent Violent Extremism in the United States. Washington, DC: Executive Office of the President of the United States.

107.

Yazdiha

Hajar

. 2014. “Law as Movement Strategy: How the Islamophobia Movement Institutionalizes Fear through Legislation.” Social Movement Studies 13(2):267–74.

108.

Zeskind

Leonard

. 2009. Blood and Politics: The History of the White Nationalist Movement from the Margins to the Mainstream. New York: Farrar Straus Giroux.