Abstract: The wave of Islamic State-linked terrorism experienced in the West over the past couple of years has rekindled debates surrounding mental disorders and terrorist engagement. A very preliminary survey by the authors found that out of 55 attacks in the West where the 76 individuals involved were possibly influenced by the Islamic State, according to media reports, 27.6% had a history of apparent psychological instability, a percentage comparable to that found in the general population. This figure is driven largely by individuals inspired by the Islamic State, as opposed to those directed by it, however. The percentage is likely overinflated for several noteworthy reasons, including poor reporting, low benchmarks, and the tendency to overuse mental health problems as a ‘silver-bullet’ explanation for terrorist involvement. The relationship is, in fact, far more complex than typically presented.

The Islamic State-linked terrorism experienced in the West recently has reignited debates about the connection between mental disorders in terrorist activity. Indeed, the assistant commissioner for specialist operations in London’s Metropolitan Police Service acknowledges the Islamic State is actively trying to recruit, among others, “those with mental health issues.”1 Similarly, Australia’s national counterterrorism coordinator, Greg Moriarty, has outlined that many cases involve individuals “not necessarily deeply committed to and engaged with the Islamist ideology but are nonetheless, due to a range of reasons, including mental health issues, susceptible to being motivated and lured rapidly down a dangerous path by the terrorist narrative.”2 The debates became particularly salient through the summer of 2016 when lone actors inspired and encouraged by the Islamic State killed 135 civilians in separate attacks in Orlando and Nice. More recently, the fatal shooting of five individuals at Fort Lauderdale airport added complexity to these debates. All perpetrators had reported histories of mental disorder.

This article seeks to answer several questions to provide a more rigorous evidence base for these debates. These questions include: What is the existing evidence base regarding psychopathology and terrorist involvement? How prevalent are mental disorders among Islamic State-directed and -inspired offenders in the West? What is the content of these diagnoses? When present, what relationship did the disorder have with radicalization?

The Evidence Base
Very few scientific analyses have focused upon the relationship between mental health and terrorist involvement. The following outlines the headline results from the sum total of these scientific endeavors.3 Some recent studies analyzed the prevalence rates of different disorders within very different samples of terrorist offenders. Based on police files, 6% of Anton Weenink’s sample of 140 Dutch individuals who either became foreign fighters or sought to travel abroad for the purpose of terrorism had diagnosed disorders.4 An additional 20% displayed signs of undiagnosed mental health problems. The diagnoses included ADD, AD/HD, autism spectrum, narcissistic, schizophrenia, post-traumatic stress, and psychotic disorders. The prevalence of schizophrenia (2%) in the sample was higher than what would be expected in the general population, which is currently estimated at between 0.87% and 1%.5

A study of lone-actor terrorists co-authored by the authors of this article also found higher rates of schizophrenia than in the general population, as well as certain other disorders. Based on open source information, over 40% of Emily Corner, Paul Gill, and Oliver Mason’s study6 of 153 lone-actor, solo-actor, and dyad terroristsa had a diagnosed disorder, much higher than in the general population (25-27%).b The diagnoses included schizophrenia (8.5%), depression (7.2%), unspecified personality disorder (6.5%), bipolar disorder (3.9%), post-traumatic stress disorder (PTSD, 3.3%), autism spectrum disorder (3.3%), delusional disorder (2.0%), unspecified anxiety disorder (1.3%), traumatic brain injury (TBI, 1.3%), obsessive compulsive disorder (OCD, 1.3%), unspecified sleep disorder (0.7%), schizoaffective disorder (0.7%), psychotic disorder (0.7%), drug dependence (0.7%), and dissociative disorder (0.7%).c Three disorders held a substantially higher prevalence than found within general populations (schizophrenia, delusional disorder, and autism spectrum disorders), whereas depression, anxiety, PTSD, and sleep disorders are more commonly found within the general population than the lone-actor sample.7

Michael and Simon Gottschalk8 also tested rates of various measures of personality and psychopathology within terrorist samples against the civilian population. By administering the MMPI-2,d they compared 90 incarcerated Palestinian and Israeli terrorists to control groups of Palestinians and Israeli Jews matched on demographic features. The former group scored higher on subscale measures of psychopathic deviate, paranoid, depressive, schizophrenic, and hypomanic tendencies.

It is worth remembering, however, that the Weenink study and the Corner, Gill, and Mason study are based on small samples of particular terrorist types and not necessarily a depiction of all terrorist populations globally. Evidence suggests, for example, that the rates of mental disorders within lone-actor terrorist populations are significantly higher than group-based terrorists. Corner, Gill, and Mason9 found a negative correlation between the level of co-offending and the rate of mental disorder prevalence. Whereas 40% of the study’s lone-actor terrorists had a mental disorder, the prevalence rates were lower for solo-actor terrorists at around 20%, just over 5% for terrorist dyads, and less than 3% for terrorist group members.e In an earlier study, Corner and Gill found that lone-actor terrorists were 13.5 times more likely to have a mental disorder than group-based terrorists.10 This result was in line with Christopher Hewitt’s analysis of U.S.-based terrorists who espoused a range of ideologies (22% vs. 8.1%),11 and that of Jeff Gruenewald, Steven Chermak, and Joshua Freilich, who examined U.S.-based, far-right offenders (40.4% vs. 7.6%).

Evidence also suggests that some terrorist roles may be more likely to experience particular mental disorders. Ariel Merari and his colleagues carried out psychological tests on samples of suicide bombers and various control groups (e.g. other terrorists and non-political criminals)12 employing a range of techniques including clinical interviews, personality tests, the Thematic Apperception Test, and the House-Tree-Person Drawing test. Suicide bombers obtained significantly more diagnoses of Avoidant-Dependent Personality Disorder (60% vs. 17%), depressive symptoms (53% vs. 8%) and more readily displayed suicidal tendencies (40% vs. 0%). On the other hand, the control group was more likely to contain members with psychopathic tendencies (25% vs. 0%) and impulsive-unstable tendencies (67% vs. 27%). Suicide bomber organizers scored higher in ego-strength, impulsivity, and emotional instability than would-be suicide bombers.

Collectively, these results provide important insight into how we should try to understand Islamic State-linked cases. We should expect to see higher rates of specific mental disorders among lone-actors than the general population, and we should expect variance across the roles and behaviors that these individuals engage in once they have chosen to turn to violence (e.g. suicide attack vs. non-suicide attack). We may also expect to find higher rates of disorders among lone actors inspired and encouraged versus those who are directly trained and dispatched by the Islamic State.

The Circumstantial Evidence Base
Whereas the above studies meticulously collected data and compared the terrorist samples to wider comparison or control groups, no such studies exist on Islamic State-linked plotters in the West. Some insight may come from a couple of studies on al-Qa`ida-inspired or -directed terrorism in the 1990s and 2000s to provide a useful framework for trying to understand the current security problem. Marc Sageman’s seminal study of al-Qa`ida networks downplayed the presence of mental disorder. His first study of 172 individuals highlighted one case of “probable mild mental retardation or borderline intellectual functioning” and two cases of “psychotic disorders.” Sageman notes this is “about the incidence of these disorders in the general population.”13 In Sageman’s updated analysis of around 500 individuals, “there were about four with hints that they experienced beliefs that were not based in reality,” but there was “little evidence for any personality disorder.”14 It should be noted, however, that these figures may be a function of the actors being largely network-based. The current threat is more varied with a higher propensity for lone actors, so we may also expect the presence of mental disorders to differ. It may be better to focus on al-Qa`ida inspired lone-actors for insight. In Paul Gill, John Horgan, and Paige Deckert’s study of 119 lone-actor terrorists, 43% were al-Qa`ida-inspired. Of this sub-sample, 25% had a diagnosed mental disorder, which is essentially the same as the population base rate. This figure, however, was significantly lower than those lone actors inspired by single-issue causes (52%) such as anti-abortion, environmentalism, and animal rights.15

In preparing this article, the authors looked at 55 attacks (totaling 76 individuals) between May 1, 2014, and September 30, 2016, where media reports indicated they were possibly influenced in some way by the Islamic State.f (See appendix). These cases encompass both Islamic State-inspired attacks and Islamic State-directed attacks as well as attacks enabled by communications with Islamic State operatives who provided guidance to attackers without them ever connecting to the group in person.g And they include both confirmed Islamists and troubled individuals with unconfirmed ideological commitment to Islamist causes who may just have been influenced by the group’s calls to violence. A history of psychological instabilityh was noted in 21 (27.6%) of the individuals. This is broadly comparable to the current worldwide average (25-27%). The authors were not able to find examples of psychological instability in the media reporting on those alleged to have taken part in Islamic State directed attacks (for example, the November 2015 Paris attacks and the March 2016 Brussels attacks). Excluding Islamic State-directed plots, the percentage of attackers with reported psychological instability was 34.4%.

It is worth noting that some of those with overt mental health problems may not have been inspired by the Islamic State but were simply professing Allahu akbar because it is well recognized as a phrase used in violent attacks. As these cases were dismissed as not being terrorism, there was barely any (in many cases no) further media reporting as to the individual’s intentions, and therefore, the degree to which they were truly Islamic State inspired is not currently known.

These are extremely preliminary results.i There are several practical constraints to a full-blown study akin to Sageman’s at this moment. On a very basic level, it is difficult to put a true figure on the prevalence rates simply because the vast majority of cases have yet to be examined in a court of law. If one thinks about the reliability of sources on a credibility continuum with trial evidence on the highest end, the vast majority of cases are still reliant on sources from the lower end.

When dealing solely with this lower end, researchers need to be careful of a number of potential pitfalls. Many high-profile, Islamic State-inspired individuals have undoubtedly either shown symptoms of psychological distress in earlier years or have been formally diagnosed with a disorder. However, at a time when rigorous and interdisciplinary insight was needed, many researchers, reporters, and practitioners turned to flippant language and coinage. Such phrases not only failed to explain the relationship (if any) between terrorist engagement and psychopathology but also potentially stigmatized the infinitesimally greater number of individuals suffering from some form of mental disorder with absolutely no hint of radicalization or plans to engage in violence. These include Max Abrahms’ “loon wolf”16 phrase, a term that he applies to any lone terrorist suffering any form of mental disorder or psychological distress. The “loon” connotation also implies crazy or deranged and misses the fact that rational planning often occurs in the presence of a mental disorder.

Corner and Gill17 (these authors) empirically compared a sample of lone-actor terrorists with mental disorders with a sample of lone-actor terrorists without mental disorders. They found that those with a mental disorder were just as (and in some cases more) likely to engage in a range of rational pre-attack behaviors as those who were not. Offenders with mental disorders were more likely to express violent desires, seek legitimization for their intended actions, stockpile weapons, train, carry out a successful attack, kill and injure, discriminate in their targeting, and claim responsibility. Most of these traits are typically viewed as rational behaviors to achieve terrorism goals.

Other popular terms used on Twitter are “Islamopsychotics” and “Mentalhadist,” terms that conflate religious devotion with mental illness, thus simplifying and demonizing both.18 Such conceptually loose and boundary-less terms may have the impact of also inflating perceptions of the prevalence of mental disorders (which have a very strict set of operationalized and bounded criteria).

A second obvious problem with this lower end of the credibility continuum is that the sources are sometimes highly questionable. Again, this may artificially inflate the true prevalence rate. In many cases, anonymous sources allude to the individual’s (usually unspecified) strange and erratic behaviors prior to their radicalization. It is often implied and assumed that this is related to mental health problems, but the reporting is unspecific. For example, on September 23, 2014, Numan Haider stabbed and injured two counterterrorism officers in the parking lot of a police station in Melbourne, Australia. Haider was shot dead at the scene. There is evidence that his family previously encouraged him to seek help from a counselor for erratic behavior.19 However, this was most likely linked to their worries over his potential radicalization rather than mental health issues.20 Other cases reporting “erratic behavior” include Michael Zehaf-Bibeau (the Parliament of Canada shooter) and Zale Thompson (who attacked four New York City police officers with a hatchet). All three examples fall short of a confirmed mental health diagnosis of any sort. All three, however, have been held up as irrefutable evidence of the link between mental health problems and radicalization. The bar for confirming evidence should be higher within research and practitioner communities.

Other cases highlight instances where mental health problems were alluded to by non-experts (often family, friends, neighbors) but latched onto by the wider media as concrete evidence of a disorder’s presence (and presumably direct role in the violent intent/actions). Worryingly, now that concrete empirical evidence has demonstrated the greater propensity for lone-actor terrorists to have a mental health problem, it is often one of the first lines of investigation in media reporting. The below examples, while interesting, show the disturbingly low levels of evidence provided within some reporting that depicts the individual as having a mental disorder. Such reporting has potentially overinflated the true rates.

On October 20, 2014, Martin Couture-Rouleau, a 25-year-old French Canadian who converted to Islam in 2013, deliberately rammed his car into two Canadian Armed Forces soldiers in a shopping center parking lot in Saint-Jean-sur-Richelieu, Quebec, Canada. One soldier was killed. Couture-Rouleau was shot following a chase and altercation with a female police officer. No confirmed diagnoses exist, but neighbors claim he was “depressed,” and one additional source claimed Couture-Rouleau’s father had previously attempted unsuccessfully to admit him to psychiatric care.21

On January 7, 2016, Edward Archer shot and wounded a police officer in Philadelphia. When apprehended, Archer confessed that he committed the attack in the name of Islam and pledged allegiance to the Islamic State. Archer’s mother claimed that he had sustained a traumatic head injury from a moped accident and that he had “some form” of mental illness because “he’s been talking to himself … laughing and mumbling” and “hearing voices in his head.”22

On June 12, 2016, Omar Mateen, a 29-year-old American citizen of Afghan descent, attacked the Pulse nightclub in Orlando, Florida, killing 49 individuals. Mateen was shot and killed by law enforcement following an extended standoff. Mateen had a history of disruptive behavior while at school. When Mateen applied for his security guard’s license in 2007 and its renewal in 2013, no psychological issues were noted at screening. Mateen was a habitual user of steroids, and his first wife claimed that he was “mentally unstable and mentally ill.”23 There is no further evidence to support this claim, however. Confusion remains over Mateen’s exact motives, as he pledged allegiance to the Islamic State just prior to this attack over the phone. Multiple witnesses have also claimed that Mateen was homosexual and struggled to come to terms with his sexuality.24

FBI Mateen
Orlando shooter Omar Mateen (F.B.I.)

Other cases are complicated by continuing questions over the degree of psychological distress and other vulnerabilities. On July 24, 2016, Mohammad Daleel, a 27-year-old Syrian refugee, carried out a suicide bombing in Ansbach, Germany. Only the assailant was killed. Investigators believe he had planned to detonate the device remotely, was planning follow-on attacks, and his death was an accident.25 Daleel, who had been treated by a trauma therapist who warned he might engage in a “spectacular” suicide attempt,26 twice was reported as attempting suicide following failed asylum applications in Germany, after which his deportation was stalled.27 Eleven days before the bombing, Daleel was again served a deportation order. Given the Islamic State’s claim that he was previously part of the group in Syria28 and detailed instructions he was receiving in communication with a suspected Islamic State handler overseas,29 some have pointed to the possibility that he might have exaggerated his mental health problems to escape deportation.30 As it is difficult to determine whether his earlier suicide attempts were used as a function to avoid deportation or due to underlying distress, it is difficult to disentangle these actions from the form of violence he ultimately chose to engage in on behalf of the Islamic State.

In other cases, specific diagnoses were made but are not publicly available. On March 14, 2016, Ayanle Hassan Ali, a 27-year-old Canadian, stabbed two soldiers at a recruitment center in Toronto, Canada. Following the attack, Ali told officers “Allah told me to do this, Allah told me to come here and kill people.” Ali was initially deemed unfit to stand trial due to “psychotic symptoms of a major mental illness,” but following mandatory treatment, Ali was found fit to stand trial for a range of charges.31

On July 14, 2016, Mohamed Lahouaiej-Bouhlel, a 31-year-old Tunisian with French residency, drove a 19-ton cargo truck into a crowd of individuals celebrating Bastille Day on the Promenade des Anglais, Nice. Lahouaiej-Bouhlel was shot and killed by police after he managed to drive over one mile through the crowded streets and pedestrian areas. According to reports, Lahouaiej-Bouhlel had only started attending a mosque two months prior to the attack.32 Lahouaiej-Bouhlel’s father claimed that his son had undergone psychiatric treatment prior to 2005 and had a nervous breakdown and a history of substance misuse and depression.33

On July 26, 2016, Adel Kermiche and Abdel Malik Petitjean took six people captive in the church of Saint-Étienne-du-Rouvray, Normandy, before killing a priest. Kermiche was born in Algeria, and he had twice attempted to travel to Syria in 2015 and was jailed.34 The attackers were shot and killed when police stormed the church. Following the attack, psychological reports carried out at Kermiche’s trial prior to the attack in October 2015 suggested that Kermiche had a history of mental health problems.35 This also highlights an additional problem: when the offenders die at the scene of the attack, psychiatrists cannot access the offenders to assess their mental health. If not for his previous dalliance with the law, Kermiche’s mental health issues might have been missed entirely.

The fact is that confirmed diagnoses in recent cases remain few and far between. In December 2014, Man Haron Monis, a 50-year-old Iranian living in Australia, took hostages at the Lindt Cafe in Sydney. The siege lasted over 12 hours and ended when security officials stormed the cafe and shot Monis. Two others were killed. Monis had claimed allegiance with the Islamic State, and following the siege, the Islamic State praised his actions in its propaganda.36 Monis had an extensive history of mental health issues. Forensic psychologist Kate Barrelle, advisor to the coroner in Monis’ case, concluded that the attacker suffered from “narcissism, paranoia, and antisocial behaviour disorder.” 37 Forensic psychiatrist Jonathan Phillips concurred with Barrelle. He dismissed the possibility that Monis also suffered from comorbid disorders by explaining that earlier non-psychosis-related diagnoses were no different from “what everyone experiences ‘from time to time.’” Phillips explained that his diagnostic decision was based on Monis’ history of sexual assaults and that he “orchestrated the incident in a meticulous and callous manner.”38 Counsel assisting the coroner at the inquest, Jeremy Gormly, SC., surmised Phillips’ conclusions: “[Monis] was suffering from some definable personality disorders, including anti-social, paranoid and narcissistic tendencies, but did not suffer a mental illness and his actions cannot be attributed to mental illness.”39

On January 7, 2015, Amedy Coulibaly, a French extremist claiming to act on behalf of the Islamic State, shot and wounded a jogger in Paris.40 A day later, he is suspected of having killed a French municipal police officer. The following day, he murdered four hostages at a kosher supermarket. A psychiatrist had previously assessed him following earlier criminal behavior and found that Coulibaly had an “immature and psychopathic personality” but stopped short of diagnosing a specific mental disorder.41

In December 2015, Muhaydin Mire attacked commuters at Leytonstone underground station in London with a breadknife, injuring three. Mire reportedly claimed, “This is for Syria, my Muslim brothers.”42 Mire had a history of mental illness and had spent time in psychiatric institutions for delusional and paranoid tendencies with some reports of paranoid schizophrenia.43 Mire’s family had also sought to have him admitted one month prior to the attack.44

In a recent and complicated case, Esteban Santiago-Ruiz killed five and injured another six in a shooting at Fort Lauderdale airport in January 2017. Federal officials confirmed that Santiago-Ruiz had received psychological treatment. His family members claim that his problems stemmed from witnessing two fellow servicemen die in combat in Iraq.45 Two months prior to the attack, Santiago-Ruiz visited an F.B.I. field office and reported the U.S. government controlled his mind and made him watch Islamic State-related videos. The C.I.A. was also forcing him to join the group, according to Santiago-Ruiz. After the attack, Santiago-Ruiz initially made similar claims. However, after Santiago-Ruiz was transferred to an F.B.I. office following the attack, he claimed he carried out the attack on behalf of the Islamic State and had not again mentioned mind control, an F.B.I. agent testified on January 17. Despite Santiago-Ruiz’ later claims, the Islamic State are yet to claim responsibility for the attack.46 Investigators are continuing to disentangle Santiago-Ruiz’ motivation, and whether the attack was motivated by voices in his head telling him to commit acts of violence.47 Here, the link between mental state and the violence appears to be much clearer than in many other cases. However, the depth of his ideological orientation toward the Islamic State is highly questionable at the time of writing. For the authors, he looks far more like other mass murderers such as Myron May (the Florida State University shooter), Aaron Alexis (the Washington Navy Yard shooter and Jiverly Wong (the Binghamton shooter).j All three individuals reported being ‘targeted individuals’ and acted violently in order to get the government to turn off their mind-control machines.

In many other cases, when confirmed diagnoses were present, there was a tendency to try dismiss the possibility of terrorism altogether. For example, on December 21, 2015, an unnamed 40-year-old ran over 11 pedestrians across the city of Dijon, France, while shouting “Allahu akbar,” claiming he was “acting on behalf of the children of Palestine,” and brandishing a knife. Police knew the assailant for previous minor offenses, and he had spent time in psychiatric services. Due to the psychiatric history of the assailant, authorities deemed the attack not to be an act of terrorism.48 A similar attack that injured 10 in December 2014 in Nantes was similarly dismissed as non-terrorism because the individual was “unbalanced.”49 On May 10, 2016, an individual attacked commuters with a knife at Grafing train station near Munich. One victim died from his wounds. The assailant reportedly shouted “Allahu akbar” and continued to espouse similar sentiments such as “all infidels must die.” This was later confirmed by police authorities but dismissed as terrorism because the individual was “mentally disturbed.”50 Moussa Coulibaly’s stabbing of three anti-terrorism police officers in Nice in February 2015 was similarly rejected by authorities as terrorism because he was viewed as an “unbalanced individual.”51

Nowhere was this debate on whether attacks were terrorism or non-politically motivated acts of mentally disturbed individuals more evident than in the Man Haron Monis inquest. Many commentators reasoned it was his history of mental disorder that best explains his actions and that he should not be considered a terrorist. For example, Dr. Rodger Shanahan, a research fellow at the Lowy Institute, argued at the inquest that Monis was “not motivated by political, ideological or religious causes, but rather was someone with mental health issues acting on his own personal grudges.”52 Shanahan also argued that “if he [Monis] was following IS [Islamic State] direction why didn’t he go in there and kill everyone?”53 Shanahan instead believed that Monis chose the Lindt Cafe for personal rather than political reasons, highlighting Monis’ historical grievance with Seven Network’s Sunrise television program, whose studios were across the road from the cafe. Shanahan also drew on Monis’ lack of direct connection to the Islamic State and his wearing of a headband, which has an association with the Shi`a sect of Islam and is considered heretical by the Islamic State.54 Clarke Jones of Australian National University agreed with Shanahan, explaining that due to his mental health problems, Monis was desperate to attach himself to a cause, further explaining that his actions were extremely hard to predict.55 Terrorism scholar Bruce Hoffman held an opposing view. Monis’ history of mental disorder, in Hoffman’s opinion, was “immaterial” as to whether he carried out a terrorist attack.56

The debate rests on the question about whether the presence of a mental health diagnosis is enough to state that it was a driver of the radicalization-linked behavior or whether it was just one ingredient in the individual’s vulnerability profile and grievance structure. Some tentative answers are available in the wider research literature.

Corner and Gill57 utilized a sample of 119 lone-actor terrorists and investigated whether certain behaviors within the chain of events that led to a terrorist attack were more likely to co-occur with certain diagnoses than others. Those diagnosed with schizophrenia and associated disorders were the only diagnostic group to be significantly associated with previous violent behavior and this supports previous research in the general violence literature.58 This could be linked to the lowering of inhibitions against violence or potentially the increased chances the individual had previously engaged in (non-terrorist) violence.k Those with personality disorders and autism were less likely to have a spouse/partner involved in a terrorist movement, which may be indicative of not having a spouse due to the detrimental nature of these disorders. Because mental disorders often share symptoms (and because diagnoses are often not available), further research may also focus on analyzing symptoms of mental illness rather than purely the diagnoses themselves.59 The results of this investigation offer insight into links between specific disorder types, and specific behaviors, that in isolation can be linked to a wide range of activities, not just those found in a terrorist attack. Researchers are yet to study temporal trajectory of behaviors, and how these interact with each other and the wider environment in space and time. We just know some behaviors co-occurred at some point in the individual’s life-course. John Horgan correctly notes that although studies of lone actors often find high preponderance of mental health issues within the sample, “detailed research would be needed to further clarify the precise nature and role (if any) of mental health problems in the development of their violent activity.”60

This debate is on-going within the wider study of crime as well. On the one hand, a strand of research assumes a consistent causal link between psychiatric symptoms (where they are found to be present) and criminal behavior.61 On the other hand, a more nuanced strand of research argues there are “a (small) group of offenders whose symptoms relate directly to crime and a (larger) group whose symptoms and crimes are not directly related.”62 For example, various studies illustrate that the offender (across a range of crimes) experienced his/her psychiatric symptoms at the time of the (often violent) crime between 4% and 18% of the time.63 There is no reason to suggest this should be any different for a terrorist subset of offenders. If anything, one might presume the figures to be lower given the wider ideology and ideologues underpinning it provide a grievance and set of instructions on who to target and how. A complex mixture of personality, situational, and personal drivers (among others) likely drives most general crime. Terrorism is no different but for the addition of an overarching ideology. The presence of this ideology in the motivational mix therefore likely lessens the relative cognitive response to mental health problems.

It is simply too early to come to a definitive answer regarding the role of mental health problems and various forms of Islamic State terrorism. Mental disorders appear more prevalent among those inspired by Islamic State than those directed by it. Beyond that, however, it is difficult to make clear conclusions. The available open-source information is clouded by poor reporting practices, the tendency to treat all mental health disorders equally, and the fetishized way mental health is reported. The answer is likely to differ wildly from case to case depending upon the individual’s diagnosis, prior life experiences, co-existence of other stressors and vulnerabilities, and lack of protective factors. Researchers need to have a mature enough response in practice, research, and public discussion that is comfortable with this complexity; understand that where it is present, it is usually one of several drivers; and do so by not stigmatizing the vast majority that suffer from mental health problems while remaining non-violent, non-radicalized, and in need of care.

What we see from the existing research is that lone-actor terrorism is usually the culmination of a complex mix of personal, political, and social drivers that crystalize at the same time to drive the individual down the path of violent action. This should be no different for those inspired by the Islamic State. Whether the violence comes to fruition is usually a combination of the availability and vulnerability of suitable targets and the individual’s capability to engage in an attack from both a psychological and technical capability standpoint. Many individual cases share a mixture of personal life circumstances coupled with an intensification of beliefs that later developed into the idea to engage in violence. What differs is how these influences were sequenced. Sometimes personal problems led to a susceptibility to ideological influences. Sometimes long-held ideological influences became intensified after the experience of personal problems. This is why we should be wary of mono-causal ‘master narratives’ about how this process unfolds. Mental health problems are undoubtedly important in some cases. Intuitively, we might see how in some cases it can make carrying out violence easier. In other cases, it may make the adoption of the ideology easier because of delusional thinking or fixated behaviors. However, it will only ever be one of many drivers in an individual’s pathway to violence. In many cases, it may be present but completely unrelated. The development of radicalization and attack planning behaviors is usually far more labyrinthine and dynamic than one single factor can explain, be it mental disorders (today’s go-to silver-bullet explanation), online radicalization (another popular silver-bullet explanation), or root causes that encompass socio-demographic characteristics.

We must also bear in mind that the relationship between mental health problems and terrorist engagement is just one part of the story. Given the scale and types of violence being conducted by the Islamic State, many perpetrators will develop mental health problems as a byproduct of involvement as opposed to it being a driver of involvement. There will also be a generation of children who were born within the Islamic State and/or trained as fighters, many of whom will return to their parents’ country of origin in the coming years. The interface of mental health practitioners and the Islamic State will, therefore, not just be limited to assessing the risk of whether someone will become a terrorist but will be extended to safeguarding and treatment.

In conclusion, after many years in the dark, the link between mental health problems and terrorist engagement is now often the “go-to” explanation. This is partially due to the studies, cited above, that showed the relatively high rates within specific terrorist sub-samples. These studies coalesced in time with an uptick in Islamic State lone-actor plots and attacks and were latched onto by media, the public and policymaker communities hungry for intuitively appealing and straightforward answers. Much of the nuance within these studies was lost, however. Just because a factor (such as mental disorder) was present, does not make it causal. Nor does it necessarily make it facilitative. It may be completely irrelevant altogether. Contemporary media reporting may have led to a potentially overinflated sense of how prevalent the link truly is and how closely tied it is to the individual’s pathway into terrorism.     CTC

Emily Corner is a research associate at the University College London’s Department of Security and Crime Science. She is currently working on a European Union-funded project on lone-actor terrorists.

Paul  Gill  is a senior lecturer at the University College London’s Department of Security and Crime Science. He is currently working on projects on decision-making and lone-actor terrorism that are funded by CREST, the European Union, and Minerva. 

Appendix 1

Appendix 2

Islamic State-directed attacks by individuals recruited in person into the group and tasked with launching attacks are shaded in gray. The Brussels Jewish museum shooting by a French Islamic State recruit was not categorized as Islamic State-directed because it is not yet clear whether he was tasked by the group to launch an attack.

All the attacks involved one attacker except the Paris attacks (nine attackers), the Brussels attacks (five attackers), the stabbing attack on a Jewish teacher in Marseille (three attackers), the Sikh temple bombing in Essen (three attackers), the San Bernardino attack (two attackers), the Kvissel murder (two attackers), the Saint-Étienne-du-Rouvray church attack (two attackers), the Rochdale murder (two attackers), and the Saint Julien du Puy attack (two attackers). In total, there were 55 cases involving 76 attackers. There was media reporting indicating psychological instability for 21 of the attackers (27.6%). Excluding Islamic State-directed attacks, the percentage of attackers with reported psychological instability was 34.4%.

Substantive Notes
[a] A lone-actor terrorist is defined by John Horgan, Paul Gill, Noemie Bouhana, James Silver, and Emily Corner as “an individual lacking any ties to a terrorist/violent extremist group … These individuals typically engage in violence in support of a group and/or ideology.” Lone actors within the sample included those inspired by Islamist, right-wing, left-wing, nationalist, and single issue ideologies. John Horgan, Paul Gill, Noemie Bouhana, James Silver, and Emily Corner, “Across the universe? A comparative analysis of violent radicalization across three offender types with implications for criminal justice training and education,” National Institute of Justice, June 2016. A “solo-actor terrorist” is defined as “an individual who either (a) carried out the act of terror alone, but under instruction or (b) carries out the act through his/her own volition, but has had previous contact with terrorist/violent extremist groups and/or radical environments to receive planning and support.”John Horgan et al. A dyad in this context refers to two individuals who conceive, plan, and carry out an attack on their own. One or both individuals may have previously been attached to a terrorist group.

[b] A large-scale worldwide epidemiological investigation based on the International Classification of Diseases (ICD) diagnostic tool highlighted that the average prevalence for a diagnosis of any disorder is 27.43%. Ronald Kessler and Bedirhan Üstün, The WHO world mental health surveys: Global perspectives on the epidemiology of mental disorders (Cambridge: Cambridge University Press, 2008). Whereas a worldwide study utilizing the Diagnostic and Statisticians Manual (DSM) as a diagnostic tool identified a lifetime prevalence of 25.0%. Jordi Alonso et al., “Prevalence of Mental Disorders in Europe: Results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) Project,” Acta Psychiatrica Scandinavica 109 (2004): pp. 21-22.

[c] The cases were based on information found in open source accounts, including court documents. Diagnoses were confirmed through extensive cross-checking of sources. Court decisions were deemed most reliable, but in some cases, mental health status was not argued at trial as the individual had an established history.

[d] Developed from the MMPI by James Neal Butcher, it is utilized by mental health professionals to clinically assess personality traits and mental health problems. James Neal Butcher, “Minnesota Multiphasic Personality Inventory (MMPI), International Encyclopedia of the Social & Behavioral Sciences (1989): pp. 9,891-9,894.

[e] An individual in a group (three or more individuals) involved in or intent on carrying out acts of terrorism. This database consists of 544 actors, with links to 85 designated terror organizations. (Cross checked with Terrorism Research & Analysis Consortium, Global Terrorism Database, and the U.S. Department of State’s list of terrorist designations and state sponsors of terrorism).

[f] Cases were chosen following expansive searches of multiple open source outlets, using the LexisNexis open source archive of academic and media articles as a starting point. Cases were determined to be “possibly influenced” by the Islamic State following cross-referencing of information. The authors did not include cases where media reporting indicated inspiration came from Islamist terrorist groups other than the Islamic State, such as the July 2015 Chattanooga, Tennessee, attack. See Devlin Barret and Arian Campo-Flores, “Investigators See Radicalization in Chattanooga Shooter,” Wall Street Journal, July 21, 2015.

[g] By “Islamic State-directed” attacks, the authors mean attacks involving Islamic State fighters recruited in person into the group and tasked by the group to launch attacks.

[h] The threshold the authors used for determining psychological instability was individuals reported to have had a history of psychiatric or psychological intervention for perceived mental problems or mental health interventions and/or diagnoses after their attacks. Cases in which family members reported depression or erratic behavior or mental health issues that did not result in such interventions were not included. For example, Orlando nightclub shooter Omar Mateen was not classed as having psychological instability because although his former wife told reporters he was “mentally ill,” the authors were not able to find any evidence there was any mental health intervention. Greg Toppo, “Ex-wife: Orlando shooter ‘very short-tempered,’ violent,” USA Today, June 12, 2016.

[i] These results differ from the earlier work of the authors, where more comprehensive cases, particularly regarding mental health status, could be formed. The cases of Corner and Gill (2015), and Corner, Gill, and Mason (2016) were drawn from individuals who had either been convicted of a terrorist act or died in the commission of their offense. The large majority of Islamic State-linked cases referenced in this article have not reached trial stage, and there is therefore less concrete information to draw from and analyze.

[j] Myron May recorded an extensive explanation (1 hr. 47 min. video) of his opinion regarding being a targeted individual shortly before carrying out the shooting at Florida State University. Within the recording, he refers to “other targeted individuals, Aaron Alexis, and Jiverly Wong.” This video was uploaded following his attack and death. “The Apprentice of Honor “Myron May mind control victim,” YouTube, March 27, 2015.

[k] For over 40 years, argument has ensued in the schizophrenia and violence literature as to whether schizophrenia is definitively linked to an increased risk of engaging in violence. There are multiple extraneous risk factors that individuals with schizophrenia may be exposed to (e.g. drug use, homelessness) that are now recognized as increasing the risk of violence across populations, not just among those with schizophrenia.

[1] Robert Windrem, “Ranks of ISIS recruits include mentally ill,” NBC News, January 1, 2016.

[2] Paul Karp, “Counter-terrorism: Turnbull defends plan that may increase access to mental health records,” Guardian, July 22, 2016.

[3] For a full history of the study of psychopathology and terrorist involvement, see Paul Gill and Emily Corner, “There and Back Again: The Study of Mental Disorder and Terrorist Involvement,” American Psychologist, forthcoming April 2017.

[4] Anton Weenink, “Behavioral problems and disorders among radicals in police files,” Perspectives on Terrorism 9:2 (2015): pp. 17-32.

[5] Esben Agerbo, Patrick F. Sullivan, Bjarni J. Vilhjalmsson, Cartsen B. Pedersen, Ole Mors, and Anders D. Borglum, “Polygenic Risk Score, Parental Socioeconomic Status, Family History of Psychiatric Disorders, and the Risk for Schizophrenia: A Danish Population-Based Study and Meta-Analysis,” JAMA Psychiatry (2015): pp. 635–641; Jonna Perälä, Jaana Suvisaari, Samuli L. Saarni,Kimmo Kuoppasalmi, Erkki Isometsä, Sami Pirkola et al., “Lifetime Prevalence of Psychotic and Bipolar I Disorders in a General Population,” Archives of General Psychiatry 64 (2007): pp. 19-28.

[6] Emily Corner, Paul Gill, and Oliver Mason, “Mental health disorders and the terrorist: A research note probing selection effects and disorder prevalence,” Studies in Conflict & Terrorism 39:6 (2016): pp. 560-568.

[7] Corner, Gill, and Mason.

[8] Michel Gottschalk and Simon Gottschalk, “Authoritarianism and pathological hatred: A social psychological profile of the Middle Eastern terrorist,” The American Sociologist 35:2 (2004): pp. 38-59.

[9] Corner, Gill, and Mason.

[10] Emily Corner and Paul Gill, “A False Dichotomy? Mental Illness and Lone-Actor Terrorism, Law & Human Behavior 39:1 (2015): pp. 23-34.

[11] Christopher Hewitt, Understanding terrorism in America (New York: Routledge, 2003).

[12] Ariel Merari, Driven to death: Psychological and social aspects of suicide terrorist, (Oxford: Oxford University Press, 2010); Ariel Merari, Jonathan Fighel, Boaz Ganor, Ephraim Lavie, Yohanan Tzoreff, and Arie Livne, “Making Palestinian ‘martyrdom operations’/‘suicide attacks’: Interviews with would-be perpetrators and organizers,” Terrorism and Political Violence 22:1 (2009): pp. 102-119; Ariel Merari, Ilan Dimant, Arie Bibi, Yoav Broshi, and Gloria Zakin, “Personality characteristics of ‘self martyrs’/‘suicide bombers’ and organizers of suicide attacks,” Terrorism and Political Violence 22:1 (2010): pp. 87-101.

[13] Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), p. 81.

[14] Marc Sageman, Leaderless Jihad: Terror Networks in the Twenty-First Century (Pennsylvania: University of Pennsylvania Press, 2008), p. 84.

[15] Paul Gill, John Horgan, and Paige Deckert, “Bombing Alone: Tracing the Motivations and Antecedent Behaviors of Lone-Actor Terrorists,” Journal of Forensic Sciences 59:2 (2014): pp. 425-435.

[16] Max Abrahms, “I’ve coined the term ‘loon wolf’ terrorist to describe a mentally unstable or deranged terrorist acting independently of a terrorist group,” Twitter, December 16, 2014.

[17] Emily Corner and Paul Gill, “A False Dichotomy? Mental Illness and Lone-Actor Terrorism,” Law & Human Behavior 39:1 (2015): pp. 23-34.

[18] Will Gore, “Mental illness has become a convenient scapegoat for terrorism – but the causes of terror are rarely so simple,” Independent, July 25, 2016; Mubin Shaikh “What we MAY have here in the #FLLshooting is a “Mentalhadist”: a diagnosed mentally ill individual who takes on a Jihadist character role,” Twitter, January 7, 2017.

[19] Cameron Houston, Tammy Mills, John Silvester, and David Wroe, “Terror suspect Numan Haider: heightened alert before AFL grand final weekend,” Age, September 24, 2014.

[20] Chip Le Grand, “Numan Haider Inquest: Teen Tried to Kill Policeman, Inquest Told,” Australian, March 7, 2016.

[21] René Bruemmer, “From Typical Teen to Jihadist: How Martin Couture-Rouleau became radicalized after converting to Islam,” National Post, November 9, 2014.

[22] Mari A Schaefer and Julie Shaw, “Mom: Shooting suspect has been ‘hearing voices,’” Philadelphia Inquirer, January 9, 2016.

[23] James Rothwell and Harriet Alexander, “Orlando shooter Omar Mateen was ‘mentally unstable wife-beating homophobe,’” Daily Telegraph, June 13, 2016.

[24] Frances Robles and Julie Turkewitz, “Was the Orlando Gunman Gay? The Answer Continues to Elude the F.B.I.,” New York Times, June 25, 2016.

[25] Andreas Ulrich, “Germany Attackers Had Contact with Suspected IS Members,” Spiegel, August 5, 2016.

[26] Gabriel Gatehouse, “Germany ‘was warned about Ansbach suicide bomber,’” BBC, August 12, 2016.

[27] Matthew Weaver and Martin Ferrer, “Ansbach bombing: Attacker pledged allegiance to ISIS, says official – as it happened,” Guardian, July 25, 2016.

[28] Anthony Faiola, “Islamic State claims German suicide bomber was former militant fighter, Washington Post, July 27, 2016; Samuel Osbourne, “Ansbach suicide bomber Mohammad Daleel ‘fought for ISIS and al-Qaeda’ before coming to Germany,” Independent, July 27, 2016

[29] Andreas Ulrich, “Germany Attackers Had Contact with Suspected IS Members,” Spiegel, August 5, 2016.

[30] “Syrian suicide bomber in Germany ‘had history of mental illness,’” RTE, July 25, 2016; Paul Cruickshank, “There are lots of questions now including whether Ansbach attacker might have faked/exaggerated mental health problems to stay in Germany,” Twitter, July 27, 2016.

[31] Michelle McQuigge, “Man accused in military stabbing ruled not fit to stand trial soon after RCMP lays charges,” Canadian Press, 2016.

[32] Tom Morgan, David Chazan, and Camilla Turner, “Nice killer Mohamed Lahouaiej Bouhlel ‘only started going to mosque this April,’” Sydney Morning Herald, July 17, 2016.

[33] David Chazan, Tom Morgan, and Camilla Turner, “Bastille Day terrorist was radicalised within months and sent £84,000 to his Tunisian family days before attack,” Telegraph, July 17, 2016; “Nice attacker Lahouaiej-Bouhlel had breakdowns says father,’” BBC, July 15, 2016; “Attack on Nice: Who was Mohamed Lahouaiej Bouhlel?” BBC, August 19, 2016.

[34] Caroline Mortimer and Samuel Osbourne, “France church attack: Normandy attacker identified by authorities as 18-year-old Adel Kermiche,” Independent, July 26, 2016.

[35] Kim Willsher, Elle Hunt, and Olivia Solon, “French priest’s killer was freed from jail despite aiming to join jihadis,” Guardian, July 27, 2016.

[36] Michael Safi, “Sydney siege gunman Man Haron Monis praised in Isis publication,” Guardian, December 30, 2016.

[37] Michael Safi, “Sydney siege inquest: Monis a ‘terrible mix’ of disorders and violent beliefs,” Guardian, April 7, 2016.

[38] “Sydney siege inquest: Man Haron Monis was a ‘psychopathic lone wolf terrorist,’” Guardian, May 2, 2016.

[39] Michael Safi, “Sydney siege inquest: Monis may have been driven in ‘by unknown accomplice,’” Guardian, March 21, 2016.

[40] David Gauthier-Villars, Asa Fitch, and Raja Abdulrahim, “Islamic State releases video calling grocery store gunman its ‘soldier,’” Wall Street Journal, January 11, 2016.

[41] Angelique Chrisafis, “Profiles: Key Suspects in Paris Attacks,” Guardian, January 9, 2015.

[42] Laura Proto, “Dramatic footage of police tasering man accused of Leytonstone Tube station attack played to Jury,” Evening Standard, June 1, 2016.

[43] Vikram Dodd and Esther Addley, “Leytonstone knife attack: Man convicted of attempted murder,” Guardian, June 8, 2016; Lizzie Dearden, “Leytonstone Tube stabbing: Isis-inspired attacker Muhiddin Mire ‘thought Tony Blair was his guardian angel,’” Independent, July 27, 2016.

[44] Tom Morgan, Tom Whitehead, and Camilla Turner, “Family of Tube Terror Accused Called in Police Three Weeks Before Leytonstone Attack,” Telegraph, December 7, 2015.

[45] David Fleshler, Susannah Bryan, Paula McMahon, and Linda Trischitta, “Esteban Santiago: Details emerge of suspect in airport shooting,” South Florida Sun-Sentinel, January 7, 2017.

[46] Boris Sanchez and Kevin Conlon, “Fort Lauderdale shooter says he carried out the attack for ISIS, FBI claims,” CNN, January 17, 2017.

[47] Kyle Clayton, Christopher Brennan, Jessica Schladebeck, and Denis Slattery, “Suspected Fort Lauderdale Airport gunman Esteban Santiago, 26, told FBI that CIA was forcing him to join ISIS,” New York Daily News, January 7, 2017; Lizette Alvarez, Richard Fausset, and Adam Goldman, “Florida Airport Attacker May Have Heard Voices Urging Violence, Officials Say,” New York Times, January 6, 2017.

[48] “France Dijon: Driver targets city pedestrians,” BBC, December 22, 2014.

[49] “France attack: Van driven into shoppers in Nantes,” BBC, December 23, 2014.

[50] Kate Connolly, “One dead, three injured in knife attack at train station near Munich,” Guardian, May 10, 2016; Rory Mulholland, “Munich knife attack: One dead after man shouting ‘Allahu Akbar’ attacks four at train station in Grafing, Germany,” Telegraph, May 11, 2016.

[51] John Lichfield, “Moussa Coulibaly: Three French anti-terror soldiers stabbed on patrol outside Jewish radio station in Nice,” Independent, February 3, 2015.

[52] Michael Safi, “Sydney siege inquest: Media apply for suppression orders to be lifted,” Guardian, August 25, 2015.

[53] “Sydney siege would have been extremely hard to predict, inquest told,” Guardian, August 26, 2015.

[54] Michael Safi, “Sydney siege inquest: Experts disagree over Monis’ motives–as it happened,” Guardian, August 25, 2015.

[55] “Sydney siege would have been extremely hard to predict, inquest told.”

[56] Safi, “Sydney siege inquest: Experts disagree over Monis’ motives – as it happened.”

[57] Corner and Gill.

[58] Menachem Krakowski, Jan Volavka, and David Brizer, “Psychopathology and violence: A review of the literature,” Comprehensive Psychiatry 27:2 (1986): pp. 131-148.

[59] Kevin S. Douglas, Laura S. Guy, and Stephen David Hart, “Psychosis as a risk factor for violence to others: A meta-analysis,” Psychological Bulletin 135:5 (2009): pp. 679-706.

[60] John Horgan, The Psychology of Terrorism (2nd Ed.) (Oxon, England: Routledge, 2014), p. 63.

[61] E. Fuller Torrey, “Deinstitutionlization and the rise of violence,” CNS Spectrums 20:3 (2015): pp. 207-214.

[62] Jillian K. Peterson, Jennifer Skeem, Patrick Kennealy, Beth Bray, and Andrea Zvonkovic, “How often and how consistently do symptoms directly precede criminal behavior among offenders with mental illness?” Law and Human Behavior 38:5 (2014): pp. 439-449.

[63] Peterson et al.

Stay Informed

Sign up to receive updates from CTC.

Sign up