Mental illness is a character weakness or lack of willpower rather than a medical condition.

In the mid-nineteenth century, American and European physicians debated whether madness was a disease of the brain or a failure of the will. The dominant social answer was the latter. Asylum superintendents wrote at length about how moral instruction, structured routines, religious observance, meaningful labor, could restore reason to minds that had gone wrong through weakness, vice, or overindulgence. The patients in state asylums were the poor, the eccentric, the immigrant, the alcoholic, the grieving woman whose grief had lasted too long. The explanatory framework mixed compassion with condemnation: these people suffered, and they suffered partly because of who they were.

The twentieth century inherited this framework without most of its religious scaffolding. By the 1940s, mental illness in American culture was most commonly explained as a failure of character, an inability to cope with the demands of ordinary life that reflected underlying weakness of personality. Shell-shocked veterans of World War I had been accused of cowardice or malingering; psychiatrists who knew better struggled for years to establish that the condition was real and involuntary. The most common treatments for psychiatric conditions through the 1930s, insulin shock therapy, lobotomy, and institutionalization, reflected an era in which the underlying causes were poorly understood and stigma suffused even professional approaches.

The American Psychiatric Association published the first Diagnostic and Statistical Manual of Mental Disorders in 1952, the year after the Korean War began producing another generation of men with psychiatric injuries. The DSM-I was not a biomedical document in the modern sense; it was shaped primarily by psychodynamic theory, treating mental conditions as reactions of the personality to accumulated stresses. But it was, significantly, a medical document, an attempt to classify psychiatric conditions systematically, to give them diagnostic names that implied they were the domain of medicine and not merely of character.

The shift toward a biomedical understanding accelerated in the 1960s and 1970s. The discovery that lithium could stabilize bipolar disorder, that chlorpromazine could reduce the positive symptoms of schizophrenia, that imipramine could lift depression, these observations pointed toward brain chemistry as a mechanism. If a drug could reliably change the course of a condition, the condition was probably not primarily a matter of willpower. Brain imaging technologies developed in the 1980s allowed researchers to see structural and functional differences in the brains of people with schizophrenia, depression, and obsessive-compulsive disorder.

The cultural lag behind the scientific understanding was enormous. Surveys conducted through the 1990s consistently found that majorities of Americans attributed depression to personal weakness and believed that people could recover by simply deciding to feel better. Insurance companies covered psychiatric conditions at lower rates than physical illnesses on the explicit theory that mental conditions were less real. Employers fired workers for psychiatric hospitalizations while routinely protecting employees who took medical leave for physical illness.

The evidence base for neurobiological contributions to mental illness, studies of heritability, neuroimaging, pharmacological response, and genetic risk, was well established by the mid-1990s. That evidence changed medical practice more than it changed public stigma. Attitudes toward mental illness as a medical condition rather than a moral failure have shifted, but surveys into the twenty-first century continued to document substantial percentages of people who believe the mentally ill should be able to recover through will and effort alone.