It seems like there's one thing everyone agrees on after a mass shooting: The shooter must have been mentally ill. President Barack Obama said as much in his reaction to the Umpqua Community College shooting on Thursday: "We don't yet know why this individual did what he did, and it's fair to say that anybody who does this has a sickness in their minds, regardless of what they think their motivations may be."
But what if the assumption is wrong, or at least misses the nuance of the issue?
Jonathan Metzl, a professor of psychiatry, sociology, and medicine, health, and society at Vanderbilt University, argues that mental illness is often a scapegoat that lets policymakers and the public ignore bigger, more complicated contributors to gun violence. Metzl, who reviewed the research on mass shootings and mental illness in a paper for the American Journal of Public Health, points to studies that show people with mental illness are more likely to be victims — not perpetrators — of violence, and that very few violent acts — about 3 to 5 percent — are carried out by the mentally ill. And while mental illness can be a contributor to some violent behaviors, other factors — such as substance abuse, poverty, history of violence, and access to guns — are much stronger predictors of violence and shootings.
For example, some studies have found that people with schizophrenia are more likely to commit violent acts. But other research has found that this propensity toward violence is almost entirely due to substance abuse, a common problem among people with schizophrenia who are self-medicating. And substance abuse — particularly when it involves alcohol — can make someone much more likely to commit violent crime.
But the focus on mental illness after mass shootings lives on. In a 2013 Gallup poll, 48 percent of Americans said failure of the mental health system was "a great deal to blame" for mass shootings, compared with 40 percent who cited easy access to guns, 37 percent who cited drug use, and 29 percent who cited the spread of extremist viewpoints on the internet. This may reflect, Metzl suggested, the inability of many Americans to understand why somebody sane would attack and kill strangers.
To dive deeper into the issue, I spoke with Metzl about the research and his thoughts on mass shootings. This interview has been edited for length and clarity.
People with mental illnesses are more likely to be victims of violence, not perpetrators
German Lopez: What do we know about a connection between mass shootings and mental health?
Jonathan Metzl: In the work that I do, I construct arguments that call on the growing body of gun research in fields such as medicine and public health.
One is that there's no real way of predicting a mass shooting. Mass shootings are very often statistical anomalies. The trouble people get into a lot of times is that they think, "Oh, he must have been crazy or something, and a psychiatrist should have been able to see him before and predict it." That's a complete miss, unfortunately.
As an aggregate group, people with mental illness are less likely than the general population to shoot other people, and they are far more likely [to be] victims of violence rather than perpetrators of violence. So I think this intense psychological profile of mass shooters as a way of preventing other mass shootings is really the wrong path to go down, even though it's understandable.
Point number two is that there is research that shows a certain proportion [up to 60 percent] of mass shooters do have some kind of psychiatric or psychological symptoms. As an aggregate group, mass shooters are very often young, white, paranoid men who've had histories of depression and possibly sometimes psychosis. So at least there are profiles that can allow you to aggregate some of these mass shooters into particular groups.
But there are a lot of other factors that aren't linked to mental illness that are equally predictive if not more predictive: access to firearms, substance use or abuse, and past history of violence or arrests. These are all far more predictive than a diagnosis of mental illness — and they're more preventive in the long run.
So while I do think there's a particular profile that many mass shooters fit into, the problem we get into when we diagnose the individual mass shooter and say mental illness is the cause is twofold. One, mental illness is really not causal for shootings; there's no mental illness that causes people to shoot other people. Two, focusing on mental health takes the shooting out of context. If you look at recent mass shooters, whiteness, for example, is a theme that runs through many, many mass shooters, but we would never say whiteness is a cause of this. It shows we collapse and oversimplify the roles of mental illness in shootings, and mass shootings in particular.
German Lopez: The way I look at it is that there are some serious mental illnesses like psychosis or schizophrenia that might contribute to mass shootings, but the vast majority of people with these conditions are not going to participate in an event like this. Is that correct?
Jonathan Metzl: Yeah.
One of the major symptoms of schizophrenia is what are called negative symptoms: people with schizophrenia, they tend to stay home because they lack energy. So even though people tend to jump on schizophrenia as the cause of the cause of many mass shootings, if you look at people diagnosed with schizophrenia as an aggregate group, people with schizophrenia aren't more likely to shoot other people than are sane people. In other words, we might do better by looking at predictive factors from people we deem sane. I can't say that more clearly.
I understand why people jump to this narrative. It's because mental illness is some sort of contributor. And killing strangers seems so far outside the bounds of civilized society and sanity that looking to insanity is understandable. But at the same time, there's no causal link between the diagnosis of something like schizophrenia and shooting somebody else.
Substance abuse is a big contributor to violence
German Lopez: Are there any particular mental health diagnoses that you think can predict violence?
Jonathan Metzl: I don't think there's any one psychiatric diagnosis — except for substance abuse. A lot of studies have shown that the presence of alcohol in the context of an argument when guns are present increases the chance that there's going to be a shooting by five to seven times. That's probably the most important psychiatric diagnosis to look at.
Antisocial personality disorder has also in the past been associated with other kinds of killings. But again, we're talking about 32,000 gun deaths in a year — so it's really hard to pin down a specific diagnosis.
German Lopez: What do you think of mental illness background checks for guns?
Jonathan Metzl: I think background checks are important tools to help stem gun crimes, and they're often very effective. That's not just because of the diagnosis of the mental illness, but because there are predictive factors in gun violence, like history of gun violence and history of substance abuse, that these checks can pick up.
Attempts to stop gun violence should focus on comprehensive factors, not just mental illness
German Lopez: It seems like you want to focus on a comprehensive set of factors. What kind of other factors are you thinking of?
Jonathan Metzl: Well, we know that most gun deaths occur within social networks [such as family, relatives, friends, enemies, and neighborhoods], so we can look at particularly vulnerable social networks. Do people live in an area where there's a lot of tension or a lot of violence? If that's the case, we can intervene on a sociological level — not try to diagnose everybody with some mental illness.
Access to firearms is huge, particularly in terms of gun suicides. [The research shows places with more guns have more homicides, and many experts believe the correlation is at least partially causal.]
Has somebody attempted suicide in the past? That's another factor to consider.
Another is whether somebody has a history of violence. It's much more predictive of someone's future propensity to commit violence than is a diagnosis of mental illness. So I'd ask someone if they have a history of assault or a history of arrest.
Of course, there's substance abuse. That's another very important factor to consider.
We need to acknowledge that it's complex. I think the question is whether we want to prevent the next mass shooting, which is very hard to do, or do we want to do things to prevent gun deaths across the country, in which case everyday violence is what we want to target.
German Lopez: Is your hope that targeting gun violence broadly, including homicides and suicides, could help prevent mass shootings from occurring?
Jonathan Metzl: Yes. I think that's absolutely right.
German Lopez: What would you say to people and lawmakers talking about mental illness as the big contributor to mass shootings?
Jonathan Metzl: Let me be clear. I think mental illness is a factor, clearly, in many mass shootings. But I think there is no one diagnosis that's linked to mass shootings, so there's no psychiatric test that can prevent a mass shooting.
Mass shootings, as traumatic and horrible as they are, are statistically very rare, so they're very hard to predict. So I think we're far better off as a society trying to prevent everyday shootings — and when you look at everyday shootings, there really is no strong correlation between mental illness and shootings.
So mental illness is important, but it becomes a scapegoat. It becomes the one thing we can all agree on — "Oh, yeah, it must be crazy people!" — but I feel compelled to resist that narrative because it gets us off the hook from looking at society, culture, laws, and other things that hit closer to home for people.