Abstract: The recidivism rate for ordinary criminals is extremely high, and since over 200 convicted terrorists have been released in the United States and many more will be in the near future, a natural fear has been that they pose a high risk of recidivism. Using nearly 30 years of data, this articles shows that while not zero, the recidivism rate of those involved in jihadi terror plots targeting the United States is much lower than that of common criminals. Unlike most criminals, prison may deter jihadis from future involvement in violent extremism.
Will those convicted of jihadi-related terror offenses pose a danger once they are released from prison?a This article explores that question by looking at conflicting findings from research examining what to expect from terrorists who have served their sentences. Next, it presents quantitative data on those involved with jihadi plots in the United States over the past three decades. Given the small numbers of jihadi re-offenders with a link to terrorist plotting in the United States, the article then gives a qualitative description of each. Lastly, it discusses possible lessons that may be gleaned from the documented cases of jihadi plot recidivism in the United States.
Why Study Jihadi Recidivism Rates?
In the United States alone, there have been over 500 prosecutions of those with ties to international terrorism post 9/11.1 Although the rate of terrorism-related arrests and prosecutions in the United States has slowed since they peaked in 2015-2016, 191 have been charged in plots related to the Islamic State since 2014 alone.2 Well over 200 convicted terrorists in the United States have already completed their sentences and been released.3 Over 50 who are currently incarcerated in the United States on terrorism charges are scheduled to be released in the next five years.4 Years of studies show criminals in the United States re-offend at rates between 25 and 83 percent.b Similar high recidivism rates (45-55 percent) have been reported in the United Kingdom, Germany, Canada, and the Netherlands.5
There has long been concern about jihadi recidivism. A 2012 report by the U.S. Department of Homeland Security stated that 27 percent of prisoners released from the Guantanamo Bay detention center had returned to the fight.6 The terrorism analyst Dennis Pluchinsky noted that “there is an apparent tendency for global jihadists to become recidivists”7 and that “the propensity for reform is less likely for global jihadists than secular terrorists.”8 Most recently, researchers Mary Beth Altier, Emma Leonard Boyle, and John Horgan studied the autobiographies of individuals involved in terrorist activities that were affiliated with known perpetrator groups and came to the conclusion that “terrorist reengagement and recidivism rates are relatively high”9 and are even, “slightly higher than criminal recidivism rates.”10
If convicted jihadis are indeed more dangerous than secular terrorists and recidivism rates among them approach those of common criminals, then there is a significant problem looming on the horizon. The potential problem may be even worse in Europe where foreign fighters who joined the Islamic State have returned in large numbers, many to countries where criminal sentences of all types tend to be much shorter than in the United States.11 Even if convicted, many will be back on the streets within a few short years.
Yet, there are some reasons to hope that those convicted of terrorism-related offenses might be less prone to repeat offense than more common criminals. For instance, two studies on those involved in militant groups on both sides of “The Troubles” in Northern Ireland found recidivism rates to be much lower than the general criminal population. While 11 percent of those convicted were later re-arrested, only 3-3.6 percent (depending on the study) of these were convicted of paramilitary-related crimes.12 This does not necessarily mean that a very high proportion of these former convicts are “reformed” in the sense that they have given up their underlying ideological commitment to violent manifestations of either the Republican or Unionist cause,c only that they are no longer engaged in the terrorism-related illegal behaviors that led to their initial criminal convictions. As John Horgan finds, disengagement is more common and often just as important as deradicalization.13
Whether or not convicted jihadis represent an increased risk of danger should be open to empirical observation, yet few studies have actually tested the premise. To date, only two studies have looked at terrorist recidivism in the United States.14
Most recently, a report from the University of Maryland’s National Consortium for the Study of Terrorism and Responses to Terrorism (START) looked into disengagement from ideological extremism.15 The eye-catching headline in the report’s description reads “New data shows risk of recidivism is high among extremists.” At first glance, the numbers in the report are alarming. Of the 300 extremists examined in the START sample, 49 percent re-offended after their first known instance of ideologically motivated crime.16 If taken to mean that nearly half of convicted terrorists will return to terrorism, this would be a true cause for concern.
But a closer look shows that this is not what the numbers imply. First, the report does not have recidivism as its primary research question. Its main concern is with why some leave extremism and the barriers to exit they encounter. Second, the report looked at “re-offending,” which is conceived as a much broader category than recidivism. In other words, the report is not necessarily talking about a convicted terrorist completing his sentence, being released, and then returning to terrorism-related crimes. Lastly, according to the primary author of the report, very few jihadis were included in the sample of 300 and the vast majority of those “re-offending” were right-wing extremists.17 In other words, this report should not be taken to mean that a high danger exists from convicted Islamist extremists.
The most thorough study so far on terror-related recidivism in the United States is from Omi Hodwitz’s Terrorism Recidivism Study (TRS), which examined 561 individuals convicted of terrorism-related offenses in the United States after 9/11.18 The underlying data is not at this time publicly available, leaving several unanswered questions.19 For instance, the author does not make it clear if it includes all terror offenses, even those prosecuted as non-ideological due to the prevalence of plea bargaining in the United States. The study also excludes arrests that did not proceed to conviction, which increases the probability that unprosecuted cooperating witnesses who later were involved in terrorist plots would not be seen as recidivists. Moreover, by focusing exclusively on post-prison release convictions, the study may have overlooked individuals who had prior terror-related convictions, were involved in later plots, but who were never prosecuted for a variety of reasons.
Even with these caveats, the study is important as the first to systematically examine terrorist recidivism rates in the United States. Of 297 ideologically motivated extremists released from prison, only nine were charged with crimes post-conviction, yielding a recidivism rate of 1.6 percent. This figure is far below that of non-ideologically motivated crimes, however measured. A closer look indicates an even lower number may be more accurate. Five of the nine were charged while still in prison, mostly of crimes unrelated to terrorism. Only four individuals were charged with crimes post-release, none of them for terrorism-related offenses.
To reemphasize, the TRS study found no individuals in the United States who were convicted of terrorism, released from prison, and then were later convicted of a terrorism-related crime.
Measuring Recidivism Among Those Linked to Jihadi Attack Plots in the United States
So, are would-be jihadis in the United States committed life-long ideologues who are likely to return to their former ways upon release? Or are convicted jihadis much more likely to become deradicalized or disengaged during or after incarceration than once feared?
Some answers can be found by looking at the most dangerous category of jihadi offenders: those individuals linked to jihadi terror plotting.
This study examines recidivism rates among jihadi plotters by using data collected in the author’s ongoing “Jihadi Plots in the United States” (JPUS) dataset.d The JPUS dataset attempts to capture all known plots by would-be jihadis against specific targets in the United States in which at least one of the plotters was physically located within the United States.20 The dataset includes all known plots that were executed, which were executed but failed, or which were in the planning stages but disrupted before being fully executed.21 It is, therefore, more inclusive than other datasets, which focus solely on successfully executed plots.e It excludes plots with connections to the United States but in which the targets were overseas.f
Within the dataset, the author sought to identify what he terms “jihadi plotter recidivists.” For the purpose of this study, the author defines jihadi plotter recidivists as either:
Individuals who were previously convicted of a crime in a case related to a jihadi terror plot involving a specific plan to commit an act of violence on U.S. soil who were then subsequently convicted or are awaiting trial in relation to any jihadi terror activity or who died in the commission of a jihadi attack.
Individuals who were previously convicted of a crime in a case related to any jihadi terror activity who were then subsequently convicted or are awaiting trial in a case related to a jihadi terror plot involving a specific plan to commit an act of violence on U.S. soil or died in the commission of an attack.g
In other words, the author counts as a jihadi plot recidivist as those individuals in the United States who re-engage in criminal jihadi activity after a conviction related to a jihadi terror plot or who become criminally implicated in relation to a terror plot after previously being convicted in relation to jihadi activity. It excludes individuals who took part in jihadi activity not linked to specific attack plotting against targets in the United States.
A former terrorist re-incarcerated for simple parole violations, such as drug or alcohol abuse, is therefore not counted as a jihadi plot recidivist. The JPUS dataset includes all known jihadi plots in the United States from January 1990 through the end of May 2019. Most studies begin post 9/11, and there is a good case to be made that this is the appropriate starting point when studying terrorism in the United States since there was a fundamental change in intelligence and law enforcement orientation after the event. However, terrorism is not a new phenomenon, and a similar overhaul of terrorism-related laws also occurred after the 1995 Oklahoma City Bombing.h The longer time frame allows for some comparisons of those convicted prior to and after 9/11.
Neither is jihadi terrorism in the United States an exclusively post-9/11 phenomenon. The first recorded event in the JPUS dataset is El Sayyid Nosair’s 1990 assassination of Rabbi Meir Kahane in New York City, but the phenomenon of jihadi terror in the United States goes back to at least 1983 and probably much earlier.i The further one goes back, the more difficult it becomes to identify plots as jihadism was not widely recognized as a distinct strand of terrorism and has often been entangled with nationalist struggles. The dataset does capture major jihadi plots in the United States prior to 9/11. For instance, two major plots in 1993. That is the year the World Trade Center was first bombed, killing six and injuring over 1,000, and in which the planned follow-up attacks by an overlapping cell against New York City landmarks were thwarted. The New York City landmark plotters received sentences ranging from 25 years to life in a plot that never was executed.22 Three of the plotters have served their sentences and two more are scheduled for release this year.
From the JPUS dataset, 189 total individuals were identified as being involved in jihadi plots against targets in the United States between January 1990 and the end of May 2019. Of these, 17 were convicted prior to 9/11. Only 31 individuals involved in these plots have been identified as being released from prison, three of those were involved in pre-9/11 plots.j
Only four of the 31 (13 percent) released plotters have been identified as having any criminal involvement in any post-incarceration crime. While much lower than traditional criminal recidivism rates, which range between 25-83 percent, this is much higher than the Terrorism Recidivism Study.
However, one of these, Burson Augustin, was involved in a clearly non-ideological crime.k A second, Abdelghani Meskini, was accused of terror involvement after his initial release, but a closer look at the facts suggests his is not a case of jihadi plot recidivism. Both of these cases will be discussed in further detail below.
In fact, only two individuals—Elton Simpson and Ali Muhammad Brown—can be categorized as being jihadi plot recidivists, yielding a 6.5 percent recidivism rate in the United States for those linked to jihadi attack plots. This figure is far below recidivism estimates for common criminals, however measured. This data, although just looking at those involved in jihadi terror plotting in the United States, suggests that convicted jihadis are less likely to return to terrorism-related crimes than some have feared. Unlike the TRS sample, which captured no ideological crimes committed post-incarceration, this data shows that at least a small number of jihadis remain committed enough to the cause that they attempt to commit acts of terror after their release.
The numbers are small enough that a deeper look at each individual may be illustrative of potential future trends.
The first of the two cases of jihadi plot recidivism presented in this data is Elton Simpson.l In Simpson’s case, a judgment call had to be made whether or not to include him as a jihadi plot recidivist because the charge on which he was originally convicted was not technically jihadi related. In the author’s judgment, the totality of the evidence presented below and of his later actions was enough to include him as a jihadi plot recidivist.
Simpson’s social media presence and his real-world connection with Hassan Abujihaad in his hometown of Phoenix, Arizona, had put Simpson on the FBI’s radar as early as 2006.23 In 2009, Abujihaad, born Paul Hall, was convicted of disclosing classified information that he had acquired during his time in the U.S. Navy to an online publication that supported the Taliban.24 In 2010, Simpson was arrested and charged with lying to the FBI about his intentions of traveling abroad to join al-Shabaab.25 However, the judge in the bench trial did not believe the prosecution had presented a strong enough case that Simpson’s lies were directly tied to a foreign terrorist group,26 a charge that would carry a prison sentence. In 2011, he was convicted of a lesser charge of lying to the FBI and was given the minimum sentence, three years of federal probation.27
Five years later, Simpson’s online activities landed him back on the FBI’s radar, and he was once again placed under surveillance. On May 3, 2015, Simpson and his co-conspirator opened fire at an anti-Islam event in Garland, Texas.m Both of the attackers were killed, and one security guard was injured. The undercover FBI agent who had been in communication with the pair arrived too late.28 The Islamic State later claimed responsibility for the attack as, moments before executing their plan, the pair pledged allegiance to the group on Twitter.29
In hindsight, it seems that Simpson’s commitment to violent jihad was both deep and long lasting. His earlier encounter with law enforcement and the justice system did not deter him from later involvement in a terrorist plot.
The second case is that of Ali Muhammad Brown and also involved a judgment call. Although the state charges for which he was initially convicted were not directly related to a jihadi crime, the evidence presented below as well as Brown’s later actions were enough in the author’s judgment to include him as a jihadi plot recidivist.
Brown was part of a group of men involved in a string of criminal activities based around the Seattle barbershop of Ruben Shumpert.n In 2002, the FBI began to investigate the group after they received tips that jihadi videos were being shown to customers. Over a dozen men associated with the barbershop, including Brown, were arrested in 2004 by police and charged with various crimes, including bank fraud, for which Brown was convicted.30
Shumpert was the main focus of the investigation, but federal prosecutors believed they lacked the evidence necessary to charge the suspected ringleader and the others with terrorism-related crimes. Released on bail awaiting a state trial, Shumpert fled to Somalia and is believed to have died there fighting for al-Shabaab.31
In 2006, an FBI agent involved in the initial investigation claimed that “although this investigation did not lead to terrorism charges … it nipped this one in the bud before it could become more dangerous.”32 The agent could not have known that in 2014, one of the men arrested would kill four people across two states.
Two of Brown’s victims seemed to have been chosen at random, but two more were killed outside a Seattle gay nightclub in what appears to be a symbolic act.33 Brown claimed that the murders were justified as retaliation for the killings of Muslims abroad. He also claimed the killings were in furtherance of the “re-creation of the caliphate, so that Muslims could have peace.”34 However, unlike Elton Simpson, Brown never publicly pledged allegiance to any specific terrorist group nor has any group claimed him as one of their own.
He received life sentences in both Washington and New Jersey and in the latter case was prosecuted under a little used state-terrorism charge. His is the only case of jihadi recidivism in which someone, other than the perpetrator, was killed.
The former head of the FBI’s Joint Terrorism Task Force in Seattle believed that Brown would be better classified as a ‘serial killer’ as Brown’s primary motivation may have been the kind of blood lust more typical of the category.35 But as academics Emily Corner and Paul Gill have persuasively argued, there need not be a conflict between mental illness and terrorism.36 One can be both mentally ill and a terrorist. That is to say, terrorists need not be solely motivated by ideological commitment. Given Brown’s own self-identification as a jihadi, it is not necessary to look further than his own admission to include him as one.
In both the cases of Simpson and Brown, the details outlined above make clear their link to “jihadi terror plot involving a specific plan to commit an act of violence on U.S. soil” came in their later rather than their original offense. It is noteworthy that in the author’s dataset, there is not a single individual in the United States who was jailed in a case related to jihadi plotting, was released, and then became involved in jihadi attack plotting again.
It is useful to explain why Meskini and Augustin were not categorized by the author as jihadi plot recidivists.
Abdelghani Meskini was a con-man involved only tangentially in the 1999 al-Qa`ida-linked LAX Millennial bomb plot.37 Meskini had known criminal ties prior to his terror-related conviction. As a cooperating witness against Ahmed Ressam, the ringleader of the al-Qa`ida-linked cell tasked with carrying out the plot, Meskini pled guilty to material support and document fraud. His part in the plot was in delivering forged documents and ill-gotten money once Ressam had crossed the border into the United States from Canada.38 In return for his cooperation, Meskini was given a light sentence and released in 2005.
In 2010, he was accused of violating the terms of his parole by allegedly buying an AK-47 in Georgia. Whether or not this is a case of jihadi recidivism hinges on both if and why he bought the rifle. Analyst Todd Bensman believes Meskini’s re-conviction is evidence of jihadi recidivism and of a looming problem on the horizon.39 As evidence, he cites testimony that Meskini had conducted internet research on Anwar al-Awlaki and the November 2009 Fort Hood attack.40 Prosecutors allege that Meskini became disillusioned after he lost his job and “was ready to snap.”41
Prosecutors in the second case against Meskini also allege that after his release, “he became a willing participant in drug dealing, prostitution and bank fraud.”42 In other words, he returned to his previous criminal life. The two witnesses against Meskini were a prostitute and a drug dealer, both of whom testified in return for immunity or lighter sentences. The AK-47 at the heart of the accusation that Meskini was on a path toward violent jihad was never found. The judge in the bench trial rejected four of the more serious allegations against Meskini. He was convicted of lying to the FBI and to his parole officer. The lies revolved around his involvement in the drug and prostitution trade at the crime-infested apartment complex he managed and about the handgun he owned, which he claimed was for self-defense.43 He is therefore not classified as a jihadi plot recidivist.
The second excluded individual is Burson Augustin of the 2006 “Liberty City Seven” plot. Augustin served his time, was released, and then was convicted for distribution of cocaine in 2013.44 Because his later offense did not involve any link to jihadism, he is not categorized by the author as a jihadi plot recidivist.
Of all plots against targets in the United States in the author’s dataset, the Liberty City Seven case had the weakest ties to jihadism.
The seven reportedly considered themselves followers of the Moorish Science Temple, a religious movement “blending together elements of Judaism, Christianity, and Islam.”45 The Florida cell, who met in a rented warehouse in the Miami neighborhood of Liberty City, had a membership that was largely drawn from the down and out and those previously involved in crime. The accused group claimed it had pledged allegiance to Usama bin Ladin as part of a scam to get money from al-Qa`ida.46 It took three trials (including two mistrials) to acquit two of the accused and convict the five others, most of whom were given light sentences.47
Prior to joining the group, Augustin had been a low-level hustler and drug dealer. After prison, he returned to that life.48
This article only looks at the most dangerous category of jihadi re-offenders (those linked at one point to attack plotting) rather than all jihadi re-offenders and therefore can only make tentative conclusions about the larger prison population of convicted jihadis. But if the low recidivism rates in this data are representative of jihadi recidivism as a whole, then jihadi offenders in the United States tend to come out of prison deradicalized or disengaged. While the recidivism rate for those linked to jihadi plots is not zero, it is far below that of common criminals.
Something has changed these would-be jihadis in prison, and this change cannot be attributed to any specific nationally coordinated CVE or deradicalization program. The fact is that the United States has no such program in place, so any change of heart or will to commit further crimes must be the result of something else. It suggests that time spent in prison alone may dampen enthusiasm for jihadi re-offending. This runs counter to years of data showing that prison tends to increase criminality over time.49
It is worth pointing out that Elton Simpson, the clearest example of a jihadi plot recidivist in the United States over the past 30 years, did not go to prison. His earlier conviction resulted in probation only. He never made it to prison for the second offense because he was killed in the process of carrying out an act of terrorism.
The most important question left unanswered here is the extent to which the findings represent general trends? This article looks only at the most dangerous category of jihadi offenders in a single country.
Could there be a kind of American exceptionalism when it comes to jihadis? Perhaps other countries will face a larger problem from the dual threat of dangerous religiously based terrorism and high recidivism rates? One possible explanation outlined here is that prison itself may have a deradicalizing effect among some jihadi plotters in the United States. An alternative explanation could be that lower-than-expected recidivism rates might be caused by longer prison sentences in the United States,50 depressing enthusiasm among released inmates for jihadi re-offending because they are older and wearier. Another might be that there is a deterrent effect because they do not want to spend another long period in prison. Further research into jihadi recidivism in different parts of the world is clearly necessary before the issue can be put to rest. CTC
Dr. Christopher Wright is an Associate Professor of Criminal Justice at Austin Peay State University, where he teaches in the Homeland Security Concentration. His research mainly focuses on jihadis in the United States and the ways in which they use technology. He has trained new FBI recruits about online extremism on behalf of the Combating Terrorism Center at West Point. Follow @wrightapsu
[a] Islamist terror-related prosecutions in the United States represent a broad range of offenses from attempted mass murder to lying to a federal agent. The most common terrorism-related offense in the United States has been “conspiracy to provide material support” to foreign terrorist organizations (FTOs)—that is, an organization designated by the U.S. State Department for which any aiding or abetting is considered illegal. However, even such conspiracy prosecutions include a wide range of illegal actions ranging from giving small sums of money to someone believed to be a member of a terrorist organization (often an informant) to those who attempted to travel overseas to join the Islamic State.
[b] These variations are partially explained by how one defines recidivism. For instance, 25 percent of federal inmates are re-incarcerated within eight years of their release. See “Recidivism Among Federal Offenders: A Comprehensive Overview,” United Stated Sentencing Commission, March 2016. Whereas 83 percent of state prisoners are re-arrested within nine years of their release. See “2018 Update on Prisoner Recidivism: a 9-Year Follow-up Period (2005 – 2014),” U.S. Department of Justice, Bureau of Justice Statistics, May 2018.
[c] In the context of Northern Ireland, “Republicans” are those who generally support secession from the United Kingdom and a unified and independent Ireland on the whole island, while “Unionists” are those who support the continuation of Northern Ireland as part of the United Kingdom.
[d] Since the dataset only looks at those who overtly adhere to the salafi jihadi ideology, it excludes right-wing and other ideological strains of terrorism and is therefore of more limited scope in predicting broader terrorist behaviors.
[e] For instance, the START Global Terrorism Database only includes executed plots.
[f] For instance, an American citizen who joined the Islamic State in Syria and was later captured and sent home to the United States for prosecution would be excluded.
[g] In some cases, a judgment call had to be made whether or not the inclusion criteria had been met. Details of some of these cases are discussed later.
[h] The Antiterrorism and Effective Death Penalty Act of 1996, which banned support for designated foreign terrorist organizations, was passed in the wake of the Oklahoma City attack.
[i] The reference is to the 1983 attack against the Hotel Rajneesh in Portland, Oregon, by Jama`at al-Fuqara’ member Stephen Paster. See Elizabeth Shogren, “Fuqra: A Name For Muslim Terrorism,” Seattle Times, July 6, 1993.
[j] These low numbers are partially the result of convicts with long sentences dying in prison before their release. However, it should be noted that six more jihadis either have been or are scheduled to be released in 2019. Another four are scheduled for release in 2020, so that by the end of next year, the number will have jumped by nearly one-third in only two years. JPUS Dataset (maintained by the author).
[k] After his release, Burson Augustin was convicted of low-level drug dealing. See “Former Member Of Liberty City Seven Charged In Federal Court For Drug Distribution,” U.S. Department of Justice, August 15, 2013.
[l] Simpson’s is also the only case from the JPUS dataset of someone who initially wanted to travel overseas to fight, was prevented from doing so, and then was involved in a plot against the homeland. See C.J. Wright, “Sometimes They Come Back: Responding to American Foreign Fighter Returnee and Other Elusive Threats,” Behavioral Sciences of Terrorism and Political Aggression (April 2018).
[m] It is because of Simpson’s involvement in the Garland attack that he was initially placed in the JPUS dataset as a plotter. Only upon closer review of the details of his life and previous encounters with law enforcement does it become clear that he was a repeat offender of jihadi-related crimes.
[n] Shumpert’s case also shows another limitation to the data presented here as those who traveled abroad are not included in the dataset, and yet he clearly showed he was committed to jihadism after his initial arrest.
 “DOJ, DHS Report: Three Out of Four Individuals Convicted of International Terrorism and Terrorism-Related Offenses were Foreign-Born,” U.S. Department of Justice Office of Public Affairs, January, 16, 2018.
 Mary Beth Altier, John Horgan, and Christian Thoroughgood, “Returning to the Fight: What the Literature on Criminal Recidivism Can Contribute to our Understanding of Terrorist Recidivism,” U.S. Department of Homeland Security, 2012, p. 8. The figures here rely on various studies, which all use “reconviction” as the standard measure.
 Ibid., p. 187.
 Mary Beth Altier, Emma Leonard Boyle, and John G. Horgan, “Returning to the Fight: An Empirical Analysis of Terrorist Reengagement and Recidivism,” Terrorism and Political Violence, published online November 18, 2019, p. 15.
 Ibid., p. 2.
 For a discussion of potential problems associated with light sentences for terror-related crimes in Europe, see Paul Cruickshank, “A View From the CT Foxhole: Michèle Coninsx, Executive Director of the U.N. Counter-Terrorism Committee Executive Directorate (CTED),” CTC Sentinel 11:9 (2018). For a more general discussion of sentencing differences between Europe and the United States, see Marc Mauer, “Long-Term Sentences: Time to Reconsider the Scale of Punishment,” Sentencing Project, November 5, 2018.
 Altier, Horgan, and Thoroughgood (pp. 8-9) cites the Independent Monitoring Commission’s Fifth Report (2005) for the three-percent rate. They also cite Clare Dwyer, “Risk, Politics and the ‘Scientification’ of Political Judgement: Prisoner Release and Conflict Transformation in Northern Ireland,” British Journal of Criminology 47:5 (2007): pp. 779-797 for the 3.6-percent rate.
 John Horgan, Walking Away From Terrorism: Accounts of Disengagement From Radical and Extremist Movements (London: Routledge, 2009).
 “Profiles of Individual Radicalization in the United States—Desistance, Disengagement, and Deradicalization (PIRUS-D3),” National Consortium for the Study of Terrorism and Responses to Terrorism (START), July 25, 2019; Hodwitz, pp. 54-64.
 For inclusion criteria for START’s PIRUS dataset, see pp. 3-5 of “Profiles of Individual Radicalization in the United States,” public release version, January 2018 update. However, the codebook only specifies inclusion criteria for the START dataset as a whole and not the sample of 300 used in the cited report.
 Author communication, lead author of the study, Michael Jensen, August 2019.
 The author indicates that the data will be forthcoming.
 For a more thorough explanation of the JPUS, see C.J. Wright, “Sometimes They Come Back: Responding to American Foreign Fighter Returnee and Other Elusive Threats,” Behavioral Sciences of Terrorism and Political Aggression (April 2018): pp. 1-16 and C.J. Wright, “How Dangerous Are Domestic Terror Plotters with Foreign Fighter Experience? The Case of Homegrown Jihadis in the US,” Perspectives on Terrorism 10:1 (2016): pp. 32-40.
 For a copy of the dataset, please contact the author directly.
 Chris Ingalls, “Seattle murder suspect tied to former terrorism ring,” King 4 News, August 20, 2014.
 “Terror plot cooperator in trouble again in NYC,” Associated Press, October 18, 2010.
 Patricia Hurtado, “Ex-Terrorist Violated Release Terms, Judge Rules,” Bloomberg, October 28, 2010.
 Carol J. Williams, “Jury acquits 1 terrorism suspect, deadlocks on 6,” Los Angeles Times, December 14, 2007; “The Released: Trial and Terror (Part 1): In the Liberty City Seven terror case, a group of street hustlers working a scam fell victim to an FBI informant posing as an agent of Al Qaeda,” Intercept, April 20, 2017.
 Trevor Aaronson, “The Released: Trial and Terror (Part 1): In the Liberty City Seven terror case, a group of street hustlers working a scam fell victim to an FBI informant posing as an agent of Al Qaeda,” Intercept, April 20, 2017.