The relationship of mental illness and violence is an issue of long-st
anding clinical and policy importance, and recent research on this ass
ociation has sparked renewed debate. The author formulates six stateme
nts on the association that seem warranted by recent investigations an
d reviews the research evidence. In general, contrary to findings of e
arlier research, an association does appear to exist between mental il
lness and the likelihood of being involved in violent incidents. A dua
l diagnosis of mental illness and substance abuse probably significant
ly increases the risk for violence, and the association between mental
illness and violence is probably significant even when demographic ch
aracteristics are taken into account. Given the considerable limitatio
ns of current research, priorities for future research include attenti
on to the strength of the association for individual subjects, inclusi
on of adequate comparison groups of non-mentally-ill persons and a bro
ad range of variables, and intensive studies of repetitively violent i
ndividuals over time.