By David Zitner and Jeffrey F. Collins

Recent alarming headlines provoke Canadians to take firm, and often conflicting, positions on issues of crime and punishment. Citizens ask how they can protect themselves while behaving in an ethical way toward those who have committed violent crimes.

The story of a Manitoba beheading is well-known and the subsequent events increasingly controversial.

Vince Li, 39, was travelling on a Greyhound bus, and sat next to 22-year-old Tim McLean. For no apparent reason, Mr. Li attacked, killed and decapitated Mr. McLean. The killer reported that voices told him to commit the murder. The court, listening to psychiatric testimony, found him not criminally responsible and sent him to a mental institution.

Three years later, many people were appalled when a panel agreed to his occasional release with supervision.

Now psychiatrists are stoking more controversy by recommending that Mr. Li transfer to a group home and be allowed unescorted outings.

In Nova Scotia, there is the case of Andre Noel Denny, who is accused of second-degree murder in the death of Raymond Taavel, a popular gay activist. Mr. Denny had been found not criminally responsible for an assault causing bodily harm when he was given unescorted leave from the hospital. He will be tried in the fall in the death of Mr. Taavel, who was killed in an altercation outside a bar on Gottingen Street in Halifax.

Those who applaud the decision to release Mr. Li must believe that psychiatric experts are clairvoyant, and therefore able to predict whether someone who has once been violent will not be violent again.

Others are skeptical, believing that violent crime and the limited ability of psychiatrists to predict future dangerous behaviour means that people who have been violent should receive supervision in a secure environment — either a hospital or a jail.

Dangerous and chaotic behaviour is too unpredictable to allow re-entry into society.

Mental health workers are divided in the same way as the rest of us. A decision to release or detain violent offenders depends as much on the composition of the panel making the decision as it does on the facts of the case.

Unfortunately, the current process for deciding fitness to stand trial and the duration of incarceration for violent offenders is capricious and expensive. There is no universal agreement in psychiatry.

Some psychiatrists may attribute bizarre behaviour to one or another form of mental illness. Other psychiatrists, examining the same facts, often reach a different conclusion. According to Dr. F.R. Farnham, publishing in the prestigious medical journal Lancet, the forecasting of dangerousness “remains like that of the weather — accurate over a few days, but impotent to state longer-term outcomes with any certainty …. psychiatric estimations of dangerousness perform less well than chance.”

Clinical intuition is neither adequate nor sensitive enough to determine who is likely to become violent or why. Consequently, the courts are consistently plagued with conflicting testimony from psychiatrists claiming to be experts relying on subjective opinion, not scientific evidence.

Conversely, the key elements for predicting violent behaviour are well-known and readily available to everyone.

Peer-reviewed work published in the journal Schizophrenia Research suggests “readily available … information allowed the prediction of assault … in a sample of general psychiatric patients with psychosis …. The information used in the predictive model was: age, sex, having committed an assault in the last 2 years (self-report) and having used any drug in the last year (self-report).”

Why rely on expensive experts for subjective opinions if objective, scientific information that lawyers, judges and juries can collect may be used to predict dangerousness? In practice, courts resolve psychiatric conflict by asking judges and juries to choose between contradictory opinions.

A quasi-judicial panel with a presiding judge would advance the science behind predictions of danger by insisting that mental health decisions, and expert opinion, must be based on peer-reviewed science and not on the conflicting subjective opinions of people whose sole claim to expertise is that they spent many years at school.

Mercurial medical professionals should not alone make the final determination on whether to detain or release a violent offender.

Someone with expertise in evidentiary judgment should be part of the process. Let medical professionals opine whether detention should be in a hospital or jail.

This article originally appeared in the Chronicle Herald.

David Zitner is a family doctor, past director of Medical Informatics at Dalhousie Medical School and the senior fellow in health-care policy for the Atlantic Institute for Market Studies ( Jeffrey F. Collins holds a law degree and is completing a PhD in political science. He is a research associate with AIMS.