The gunman, Adam Lanza, 20, has been described as a loner who was intelligent and socially awkward. And while no official diagnosis has been made public, armchair diagnosticians have been quick to assert that keeping guns from getting into the hands of people with mental illness would help solve the problem of gun homicides.
Arguing against stricter gun-control measures, Representative Mike Rogers, Republican of Michigan and a former F.B.I. agent, said, "What the more realistic discussion is, 'How do we target people with mental illness who use firearms?' "
Robert A. Levy, chairman of the Cato Institute, told The New York Times: "To reduce the risk of multivictim violence, we would be better advised to focus on early detection and treatment of mental illness."
But there is overwhelming epidemiological evidence that the vast majority of people with psychiatric disorders do not commit violent acts. Only about 4 percent of violence in the United States can be attributed to people with mental illness.
This does not mean that mental illness is not a risk factor for violence. It is, but the risk is actually small. Only certain serious psychiatric illnesses are linked to an increased risk of violence.
One of the largest studies, the National Institute of Mental Health's Epidemiologic Catchment Area study, which followed nearly 18,000 subjects, found that the lifetime prevalence of violence among people with serious mental illness — like schizophrenia and bipolar disorder — was 16 percent, compared with 7 percent among people without any mental disorder. Anxiety disorders, in contrast, do not seem to increase the risk at all.
Alcohol and drug abuse are far more likely to result in violent behavior than mental illness by itself. In the National Institute of Mental Health's E.C.A. study, for example, people with no mental disorder who abused alcohol or drugs were nearly seven times as likely as those without substance abuse to commit violent acts.
It's possible that preventing people with schizophrenia, bipolar disorder and other serious mental illnesses from getting guns might decrease the risk of mass killings. Even the Supreme Court, which in 2008 strongly affirmed a broad right to bear arms, at the same time endorsed prohibitions on gun ownership "by felons and the mentally ill."
But mass killings are very rare events, and because people with mentally illness contribute so little to overall violence, these measures would have little impact on everyday firearm-related killings. Consider that between 2001 and 2010, there were nearly 120,000 gun-related homicides, according to the National Center for Health Statistics. Few were perpetrated by people with mental illness.
Perhaps more significant, we are not very good at predicting who is likely to be dangerous in the future. According to Dr. Michael Stone, professor of clinical psychiatry at Columbia and an expert on mass murderers, "Most of these killers are young men who are not floridlypsychotic. They tend to be paranoid loners who hold a grudge and are full of rage."
Even though we know from large-scale epidemiologic studies like the E.C.A. study that a young psychotic male who is intoxicated with alcohol and has a history of involuntary commitment is at a high risk of violence, most individuals who fit this profile are harmless.
Jeffery Swanson, a professor of psychiatry at Duke University and a leading expert in the epidemiology of violence, said in an e-mail, "Can we reliably predict violence? 'No' is the short answer. Psychiatrists, using clinical judgment, are not much better than chance at predicting which individual patients will do something violent and which will not."
It would be even harder to predict a mass shooting, Dr. Swanson said, "You can profile the perpetrators after the fact and you'll get a description of troubled young men, which also matches the description of thousands of other troubled young men who would never do something like this."
Even if clinicians could predict violence perfectly, keeping guns from people with mental illness is easier said than done. Nearly five years after Congress enacted the National Instant Criminal Background Check System, only about half of the states have submitted more than a tiny proportion of their mental health records.
How effective are laws that prohibit people with mental illness from obtaining guns? According to Dr. Swanson's recent research, these measures may prevent some violent crime. But, he added, "there are a lot of people who are undeterred by these laws."
Adam Lanza was prohibited from purchasing a gun, because he was too young. Yet he managed to get his hands on guns — his mother's — anyway. If we really want to stop young men like him from becoming mass murderers, and prevent the small amount of violence attributable to mental illness, we should invest our resources in better screening for, and treatment of, psychiatric illness in young people.
All the focus on the small number of people with mental illness who are violent serves to make us feel safer by displacing and limiting the threat of violence to a small, well-defined group. But the sad and frightening truth is that the vast majority of homicides are carried out by outwardly normal people in the grip of all too ordinary human aggression to whom we provide nearly unfettered access to deadly force.