During his oral exams that spring, Perpetrator C mumbled unintelligibly in front of the panel of professors trying to grade him. He started isolating himself, playing violent video games one hundred hours per week. During this time, Grace answered a call from his psychiatrist. “Did you know your son was dropping out of school?” the psychiatrist asked. “And because of this, I can’t provide services anymore.” Grace offered to pay out of pocket, but that was against university policy.
Grace immediately called her son to try to understand what was going on. He was more talkative on this call than usual. She offered to come visit him, and she and Larry booked plane tickets for a visit in six weeks.
The shooting took place three weeks after their phone call. Moments before he opened fire, Perpetrator C used his cell phone to call a local behavioral health unit for help. No one answered.
—
Eighty percent of all mass shooters in our database were in a state of crisis in the minutes, hours, days, or weeks prior to committing their shootings. A crisis overwhelms a person’s usual coping mechanisms. Someone in crisis is like a balloon full to bursting. When you inflate a balloon, you apply stress, or tension, stretching the body over the air; and the bigger the balloon gets, the more likely it is to pop. A fully inflated balloon becomes so fragile that it must be handled carefully or it will break. A crisis is the same.
Struggling to manage his mental health symptoms, Perpetrator C was ready to burst, and failing out of graduate school was the final puff of air into the balloon. Our data show that mass shooters don’t just snap, acting violently out of the blue. There’s a slow build over time, with air being added to the balloon little by little. Small failures and indignations add up, so the crisis is more than the sum of its parts. For some mass shooters, the final blow is a major loss, such as having their wives leave them, or being kicked out of school or the military. For others, it’s something smaller, like failing a class, being rejected by peers or coworkers, or experiencing paranoia that eventually becomes unbearable. For mass shooters, no matter the cause, reaching this crisis point makes them violently angry and hopeless that things will ever change.
Half of all mass shooters in our database had been reprimanded, suspended, or fired from work shortly before committing their crimes. Many were workplace shooters who killed their coworkers and supervisors. One, who in 1986 murdered fourteen people at the Oklahoma post office where he worked, had been rebuked for his tardiness and for misdirecting mail. The shooter who, in 1997, killed four people at the maintenance yard in Orange County where he worked had been fired for stealing scrap metal. And the man who went on a shooting spree in 2019 in the West Texas cities of Midland and Odessa had been fired from his job just hours before he killed eight people.
About one in four mass shooters had a relationship end right before they committed their crime, especially those who murdered people at restaurants and retail establishments. The shooter who killed six people at an Oregon restaurant in 1977 had just attempted suicide after his wife left him. In 2018, the man who killed four people at a car wash in Pennsylvania had been stalking a woman who had just broken up with him.
A number of mass shooters faced financial ruin prior to their crimes, which seemed to push them over the edge. The man who killed eight people during a shooting spree in San Francisco in 1993 had recently filed for bankruptcy. The perpetrator who murdered seven coworkers at a technology company in Massachusetts in 2000 had recently had his paychecks reduced owing to an IRS tax lien, which triggered a psychotic episode. And the perpetrator of the worst mass shooting in U.S. history, who killed sixty people in 2017, was a high-stakes gambler who was reportedly depressed after losing large in the Las Vegas casinos.
A true crisis is communicated through a change in behavior from baseline—something different or unusual from that person’s norm, something noticeable. Parents, spouses, friends, coworkers, neighbors, pastors, mental health professionals, and even law enforcement report observing significant behavioral change in a majority of mass shooters.
For example, the perpetrator who killed four people at a Texas church in 2005 had previously gotten into an argument at church service and was asked to leave. He started screaming obscenities at neighbors and firing a gun while in his yard. He also left rambling, nonsensical notes on parishioners’ cars and would yell at them as they left the service.
The Newtown, Connecticut, shooter, who murdered twenty-seven people, including twenty first graders, in 2012, had become a recluse, writing about violent and graphic things in his journal. He would spend hours dancing maniacally at a video game console in the lobby of a local theater.
The signs that mass shooters were in crisis prior to their shooting
Duration of crisis before shooting
One of the only female mass shooters in our research, who killed seven people at the post office in California where she worked as a carrier, had started shouting to herself not long before her crime. She ordered food at restaurants and then bolted out the door before eating. She would kneel in prayer at the roadside and even undressed in public parking lots prior to the shooting. This behavior was visible to the people who knew her, but there was no intervention; she was never connected to mental health resources.
—
Perpetrator A’s second letter to us was focused on his “Adolescence.” In it, he disclosed the following:
I realize now that I was frequently depressed, although at the time I thought everyone felt like I did. I didn’t know I needed help and probably wouldn’t have admitted it if I had. Only “crazy” people needed mental help, right? I wasn’t crazy and wouldn’t have admitted to being so “weak” that I needed help.
I thought about suicide many times. Even stuck the barrel of a loaded gun in my mouth a couple of times. I’d say fear and lack of commitment were the biggest reasons I didn’t pull the trigger.
He sent us a follow-up letter titled “Adulthood.” He labeled the pages 8–10, so we could connect the letter to his previous seven pages of correspondence. He described feeling completely “overwhelmed” at the time of his shooting:
In addition to my military duties + studying for military courses I helped [my fiancée] study for nursing school sometimes until 1–2 am . . .
I was a functional alcoholic + drank nearly every day. At lunch time, I would go to my room in the barracks + slam a pint of MD 20/20 Orange Jubilee + then eat orange flavored cough drops to cover it up. I normally kept 1/2 pint or pint of Jack Daniels in my trunk + would take a few swallows on the way home to my fiancée’s house. Once home or off duty I would continue drinking mixed drinks or beer, especially if I was going out with friends. There were many times I drank alone, also.
In his next letter, titled “Prior to the crime” (labeled pages 11–15), he described the buildup to his crime:
There were quite a few things going on in my life right before my crime. I had just been informed that I would possibly be transferring to another military base, which I did not wish to do. . . . Fighting to maintain a semi-decent level of sobriety + not doing well with it. Adjusting to my promotion in rank and the role/duties expected of me. Adjusting from being single to engaged with two children. The constant worry of my little brother’s battle with Cystic Fibrosis. Slight concern about finances/bills. I had a good bit on my plate + was doing my best to try + handle it all. With the mental/emotional trauma from my childhood still unresolved I can see now that it was only a matter of time before my juggling act collapsed.
The night before the shooting, Perpetrator A went out drinking with friends. He got so drunk that he slept at the army barracks that night. He didn’t get up to go on a planned beach trip the next morning, and he didn’t show up for his dress uniform inspection, either. Instead, he rented movies on his way home. When he got home, he crashed. His fiancée and kids were gone for the weekend.
I got up around 2 pm, got something to eat + started watching movies (the porn first, then “Sometimes They Come Bac
k,” more porn, etc. . . .)+ drinking. I had a fifth of Wild Turkey 101 + a 12-pack of Coor’s [sic]. Sometime that evening, I made a call to my mom + we talked about my little brother, both of us getting pretty emotional. I love my brother + was worried about him. I got so upset/crying that my mom even offered to drive up from Florida to make sure I was okay.
His mom never made it up to see him, because he committed the shooting that night.
—
Much of our public policy debates have centered on the role of mental illness in mass shootings. “Mental illness and hatred pull the trigger. Not the gun,” President Trump said after the mass shootings in El Paso and Dayton that killed thirty-one during the summer of 2019.1 It’s true that about six in ten mass shooters have a history of mental health diagnoses or treatment, but that fact conceals more than it reveals. The vast majority of people with serious mental illness living in the community are not violent, and people with serious mental illness are more likely to be victims than perpetrators of crime.2
Just because someone has a mental illness, it doesn’t mean their every action is related to that diagnosis or that their mental health symptoms directly cause all their behavior. Symptoms of mental illness wax and wane over time, depending on factors such as treatment and stress.3 In order to understand if mental illness played a direct role in a mass shooting, we need to know more than whether the perpetrator had a history of treatment or a formal diagnosis. We need to know if they were actually experiencing symptoms when planning and committing the shooting and if those symptoms influenced their decision to act.
It is difficult to assess the role of symptoms of certain serious mental illnesses in motivating mass shootings because some symptoms are traits that motivate violence for individuals both with and without serious mental illness. For example, irritability and hopelessness are symptoms of depression, and impulsivity is a symptom of bipolar disorder, all traits that potentially could contribute to mass shootings regardless of one’s mental health diagnosis.
The mental health of mass shooters
It’s easiest to examine the role of mental illness in mass shootings when looking at symptoms of psychosis, namely delusions, and hallucinations. This is because delusions and hallucinations tend to be specific to a serious mental illness, and because they are not based in reality, they are easier to conceptualize as directly motivating violence, even if the vast majority of people experiencing them are never violent. Hallucinations can occur in any of the five senses—touch, sight, hearing, smell, and taste. Visual hallucinations, for example, include seeing shapes, colors, objects, or people who aren’t physically there. Hearing voices is the most common form of auditory hallucination and is commonly associated with certain mental illnesses such as schizophrenia. Delusions, by contrast, revolve around concepts, ideas, and beliefs that are strongly held in the mind, such as paranoia, grandeur, or somatic delusions related to the body and its functions.
Our research finds that in 11 percent of cases, psychosis may have played a minor role in the shooting, meaning the perpetrator had experienced delusions or hallucinations when planning or carrying out the shooting that may have influenced their thinking or decision-making, but these were not the main motivating factor. For these shooters, psychosis may have reduced their ability to cope with other life stressors, contributing to their crisis point—as with the 2019 Midland-Odessa shooter who had been fired from work shortly before his shooting spree: He had been repeatedly hospitalized throughout his life for delusional beliefs about government conspiracies, which likely contributed to his actions.
In 9 percent of mass shootings, psychosis played a moderate role in the crime—meaning the perpetrator experienced psychosis prior to or during the crime, but they also had another motive. For example, the 2012 Oikos University shooter was failing out of school and, on the day of the shooting, planned to request a $6,000 tuition refund for the semester. After a university administrator did not meet with him, the shooter got angry and shot his fellow students at Oikos, a Korean Christian college in Oakland, California. However, the shooter was also diagnosed with paranoid schizophrenia and believed that the staff at the university were conspiring against him, alienating him from classmates, and surveilling him. He had started living in his car when he thought the university was wiretapping his home, then had abandoned the car because he believed the university had put a GPS tracker on it. A psychiatrist who examined the shooter before trial said he had auditory and visual hallucinations as well as delusions.
For another 10 percent of shootings, psychotic symptoms appear to have played a major role in the shooting, meaning the shooter experienced psychosis both prior to and during the crime, was responding to delusions or hallucinations in planning and committing the crime, and had no other discernible motive. For these shooters, a psychotic episode was their crisis point. For example, the perpetrator of the 2018 Waffle House shooting in Nashville, Tennessee, had been experiencing psychotic symptoms for years prior to the shooting. He believed that pop star Taylor Swift was stalking him, that they had met at a Dairy Queen, that she had scaled a building to get away, and that she and the police were hacking into his phone and bank accounts. A year before the shooting, he was arrested while trying to break into the White House to meet with President Trump. After his arrest for the shooting, forensic psychologists diagnosed him with severe untreated schizophrenia and found him incompetent to stand trial.
This all means that 70 percent of the time psychosis played no role in a mass shooting, meaning the shooter had no history of experiencing delusions or hallucinations either before or during the shooting. In other words, the data do not support blaming mass shootings exclusively on serious mental illness, as President Trump did. Doing so not only risks stigmatizing the millions of Americans who are affected by serious mental illness each year; it also misses other explanations and motiving factors.
Such stigma exists, in part, because the term psychotic is used frequently and interchangeably with psychopathic to explain mass violence in popular culture. They are, in fact, two distinct concepts. The latter is a personality disorder distinguished by grandiosity, a lack of empathy, limited emotional range, and manipulation. There is a vast academic literature on the exact nature of psychopathy, but colloquially, psychopaths are often described as narcissists with little conscience, which is a natural description for people who kill instinctively without remorse—people such as the character Dexter or Hannibal Lecter, the “psychopaths” we know and love from TV. Some experts argue that mass shooters must be psychopaths.4 Why would anyone perpetrate a mass shooting if they had the capacity to truly understand its consequences? “Law-and-order” politicians, who favor individualized explanations for aberrant behavior over critiques of social structure, also like the label because it magically explains away mass shootings in their communities—the psychopath is the natural ally of the “monster.”
People get labeled psychopaths largely based on the results of a psychological evaluation tool called the Hare Psychopathy Checklist—Revised, named for Canadian psychologist Robert Hare, who developed it in the 1970s. After years of research, however, expert psychiatrists and psychologists have concluded that the test cannot precisely or accurately predict an individual’s risk for committing serious violence.5 Psychiatrist Hervey M. Cleckley, who originally conceptualized psychopaths in his 1941 book The Mask of Sanity,6 describes psychopaths as charming and glib, without empathy or the ability to experience the full range of human emotion. In his framework, Cleckley describes psychopaths as calm individuals, distinctly not anxious, which makes them immune to suicide. However, our research has found that mass shooters are neither charming nor calm, and they are certainly not resistant to suicide. Take the teenager who repeatedly texted his ex-girlfriend about killing himself the week before he killed four people at his Washington State high school in 2015. For most mass shooters, the psychopath label doesn’t fit.
—
Molly reached out to us after hear
ing about the Violence Project on social media. Her initial email was a bit cryptic—she said she had a story to tell—so we didn’t know what to expect when we met via Zoom. Turns out her story is about an incredible moment of crisis intervention.
Molly is engaging, with a loud, friendly voice and a bright smile. She’s clearly an amazing middle school counselor. “I need to start at the beginning and then go straight through,” she starts out. “That’s the only way I can tell this.”
She explains that she was a counselor overseeing seventh and eighth grades in a rural middle school in the South with around seven hundred students. One morning, a student approached her in the hallway and said she was worried about someone in her class; she had noticed cut marks under his watch band. Molly thanked the student for reporting and called the student in question into her office. “He opened up to me,” she explains, “He seemed to have some delusional thinking; he believed he was part of the online music industry and some sort of pyramid scheme for others to advance in. I learned that his mother had passed away, apparently from a drug overdose.”
The next morning, the same young man sought her out in the common area and asked to see her again. Molly was having breakfast with a colleague and hadn’t started her day yet, so she told the student to go wait for her in her office. When Molly entered her office, she knew something was wrong. The student didn’t make eye contact. He was wearing a gray Dickies jacket zipped all the way up, as high as it would go. He was sweating and fidgeting.
“As we started talking, I felt dizzy, like I was going to faint,” Molly tells us. “Something just didn’t feel right. He asked a number of strange questions, like ‘Are there any drugs here? Where is the police officer’s office in the school? What would you do if I pulled all the phone cords? Do you have any scissors?’ ”
The Violence Project Page 7