Mourners pass by a beam with a message left before the burial of Kendrick Castillo on May 17 in Littleton, Colo. Castillo was killed while trying to stop a gunman in his school last week. (David Zalubowski/AP)
Deborah Doroshow is a physician and historian of medicine at Yale University and the author of "Emotionally Disturbed: A History of Caring For America's Troubled Children."

In the two decades since the massacre at Columbine High School, digging into the psychology of mass shooters has sadly become an all-too-familiar habit — now something we seem to do almost weekly.

After the Virginia Tech shooting in 2007, media coverage pointed to the shooter’s odd behavior as a child and his near-mutism as a college student. After the mass shooting at Sandy Hook Elementary School in Newtown, Conn., in 2012, newspapers described the shooter as “withdrawn and meek” and suggested that he might have had Asperger syndrome. The two people responsible for the shooting at STEM School Highlands Ranch in Colorado on May 7 are already the subjects of forensic investigation of their presumed troubled pasts.

This practice is not just a phenomenon of the post-Columbine era of mass shootings. It has its roots in the early 20th century, and it represents an effort to shift blame and find an area of consensus after massacres that could otherwise force uncomfortable conversations. In the process, this practice fosters stigma against one of the most vulnerable groups of Americans: the mentally ill.

In the late 19th century, reports of mass shootings were typically very brief. But by the turn of the century, coverage grew more detailed, often describing how the shooter had gone “suddenly insane” as a result of financial losses or a romantic mishap.

Starting in the 1930s, newspaper coverage of shootings expanded. Journalists and those affected by the shootings searched for clues in the shooter's past that might explain why the tragedy took place. They used the Freudian language of “complexes” that had become a part of daily conversation and the psychiatric language of diagnostic categories to offer an answer.

The first case dissected in this way was the murder of two professors and the wounding of a third at the Columbia University dental school by technician Victor Koussow in 1935. The New York Herald Tribune and the New York Times immediately speculated on the shooter’s mindset. Koussow was a Russian immigrant who claimed to have served in numerous lofty positions and been awarded medals for military bravery while in Russia. Colleagues described him as suffering from a “Napoleonic complex” and a “persecution complex,” while journalists concluded that his menial work in light of his (real or imagined) past in Russia contributed to a “superiority complex” and led him to kill colleagues who had not properly respected his achievements.

Five years later, Verlin Spencer, a junior high school principal, killed five colleagues in South Pasadena, Calif., soon after he had learned that his contract was in jeopardy. Immediately, speculation abounded that his attack was caused by a “persecution complex,” because Spencer had frequently blamed his colleagues for gossiping about him and trying to get him fired. Journalists and colleagues noted a medical leave a year prior for a “nervous breakdown,” suggesting that the problem was not new. Over the next several days, the Los Angeles Times continued its investigation of Spencer’s mind. He had been overworked and lacked sleep; he had been dismissed from a previous job because of mysterious “morals charges” involving a female student; he was terrified of failure; he was addicted to bromides for constant headaches and to amphetamines for his fatigue.

In both cases, this analysis mingled with tributes to the victims, but reporting did not include discussions about how to prevent future shootings. This began to change in the 1940s and 1950s. After 14-year-old Billy Prevatte shot and killed one teacher and wounded two others at his junior high school outside Washington, D.C., in 1956, citizens wrote letters to The Washington Post arguing for increased attention and resources, not only to treating emotionally disturbed children, but also to preventing childhood mental illness in the first place.

A sea change occurred with the 1964 release of the Warren Commission report, which investigated the assassination of President John F. Kennedy. Among its findings, the commission determined that Lee Harvey Oswald had been an emotionally disturbed child. Mental health experts and journalists seized on Oswald’s story as a means of fomenting fear about a potential epidemic of childhood mental illness sweeping the nation. This conversation fostered results: In 1969, the presidentially commissioned Joint Commission on the Mental Health of Children announced that 1 million emotionally disturbed children in the United States were going without treatment, declaring a crisis in child mental health and recommending a renewed commitment to offering support to combat this national epidemic.

This newfound focus helped start the now-familiar pattern, where we look to the abnormal psychology at the root of a shooter’s actions when trying to come to grips with senseless violence. Many then conclude that the mental health system is broken, proffering solutions that have all too often not come to fruition.

This psychoanalysis serves several purposes. Fundamentally, it’s an attempt to figure out how someone so dangerous could slip through the cracks. Blaming mental illness, which is increasingly understood as a result of abnormal biology, allows us to avoid tough or uncomfortable questions such as why specific people, like parents or teachers, didn’t see it coming and do something to prevent it. Focusing on mending a broken mental health system also redirects blame from individuals to infrastructure. Blaming mental illness also allows people to sidestep the inflamed and often vitriolic battle over gun control that erupts in these moments. Consensus is often easier to come by on mental health issues.

But we must resist this tendency, sensible though it may seem, to make mass shootings a cautionary tale about our broken mental health system. Although the mentally ill are depicted in the popular imagination as dangerous, unpredictable and violent, decades of research have shown that mental illness accounts for only a small proportion of violent crimes. By linking mass shootings to debates about mental health, we are perpetuating the stereotype of the mentally ill as violent and the stigma that this already vulnerable group of people must contend with on a daily basis. And it’s a stigma with consequences: People with mental illness are less likely to seek out help.

Mental health services undoubtedly need and deserve increased funding. But we should take care not to make this the defining lesson of each mass shooting. Doing so stigmatizes the mentally ill and prevents us from having the sorts of hard conversations that we need to have about what really causes mass shootings and how we can prevent them.