Throughout history the public attitude has been that a strong link exists between mental illness and violence. Such perceptions have driven both public policy and the stigma attached to persons with mental disorders. Advocates for the mentally ill have tended to make blanket denials of a connection. However, in recent years a consensus has developed that a statistical association exists, although it is modest and is vastly overstated by both news and entertainment media. Most people with mental illness are not violent; only 3 percent of the variance in violence is attributable to mental illness. Victims are overwhelmingly family members. In addition, demographics are more important to violence than are mental illnesses. Moreover, drug abuse is associated with violence at double or triple the rate of that for mental illness. The crucial relationship between mental disorders and violence is among those with specific kinds of psychotic symptoms. Clinicians have a slightly better than average ability to predict violence; the symptoms are more important than the diagnosis in making predictions. Drug abuse is also much more important than the diagnosis in making predictions; demographic factors are not important in predictions. Finally, rapid developments are occurring in research on this topic. The speaker is Professor of Psychology and legal Medicine at the University of Virginia School of Law. Questions from the audience, answers from the speaker, and introduction by National Institute of Justice Director Jeremy Travis