HARVARD GAZETTE ARCHIVES

Felton Earls: 'If we can stop a violent incident, not only have we stopped a victim from being injured, but we have diminished the probability that a whole network of witnessing peers will perpetrate violence.' (Staff photo Kris Snibbe/Harvard News Office)

A self-described child of the 1960s, Earls has spent much of the past 15 years trying to understand violence in society. In a Nov. 17 lecture, he posed the question: "Is violence a medical problem?" but he didn't answer it. "Researchers ask questions," he said. "Only after time, research, and experience should you be in the position to answer the question. I can say that violence is a health problem, but I'm cautious about answering the question of whether it's a medical problem."

Violence is commonly understood to be a public health problem, which means that it's a problem that can be studied among populations and the structures, institutions, and regulations that affect those populations. Medical problems are focused on individuals. Clearly, the effects of violence are often medical problems, but violence itself?

A study, whose authors include Earls and School of Public Health research fellow Jeffrey Bingenheimer, which appeared in the May 27, 2005, issue of Science, suggests that violence acts as a social infectious disease. The study looked at adolescents who witnessed or have been victims of life-threatening gun violence. "We were interested in what happens when young people witness close-up violent episodes in their families or in their neighborhoods," said Earls. "And what's the likelihood that witnessing such an event will mobilize them in such a way as to have them become perpetrators of violence." What they found was that even after controlling for a huge number of variables - including aggression, drug use, gender, and ethnicity - that those who witnessed firearm violence were three times more likely to perpetrate firearm violence within two years than their peers who didn't witness the violence. Thus, violence can be viewed as a disease that "infects" its witnesses.

The ramifications for violence prevention are profound: "If we can stop a violent incident, not only have we stopped a victim from being injured," said Earls, "but we have diminished the probability that a whole network of witnessing peers will perpetrate violence. You can actually break down the cycle in which exposure leads to violent incidence."

Community efficacy

This study, which comes out of a 10-year examination of violence in Chicago neighborhoods, is only the latest study to suggest that people's sense of control over their environment is correlated with violence. Where people feel that they have educational and economic opportunities, when they can participate in family and societal life, then violence rates diminish. Where people feel that they can't escape violence, that there are no jobs, that kids can't be raised in safety, then violence rates are higher. Earls and his colleagues have termed this sense of control "collective (or community) efficacy."

"Community efficacy is about control," said Earls. "What we found is that as we got close to a neighborhood and understood how the residents functioned in terms of the degree to which people watched out for each other, trusted each other, and monitored each other's kids ... if we measured those aspects of neighborhoods, we could draw much sharper correlations and maybe even causal connections with rates of violence."

This was revolutionary in terms of understanding the causes of violence. Previously, it was thought that socioeconomic status and race were primary predictors of neighborhood violence. Community efficacy showed that there were healthy poor, black neighborhoods with lower rates of violence than unhealthy white, middle-class neighborhoods.

"It's a property of neighborhoods that allows one to have public-health leverage, because if community policing, if urban gardens, if public libraries can function in a way to bring people together ... that becomes a point of intervention to make the place safe," said Earls. It moved the equation away from race composition and class composition as causes of violence and said that the cause of violence is the way the community functions. "If you can improve the way the community functions, you can reduce violence."

Healthy communities mean healthier people

A surprise to Earls and his colleagues was that the community efficacy effect was not limited to its beneficial effects on violence. They also found benefits for birth weight, for reduced rates of asthma, and for general mental health among children. "This sense that you live in a place where you feel secure and in control translates into a general health benefit that cuts across a wide range of problems," said Earls.

In studies of adult health in the Chicago neighborhoods, researchers are looking at the general health of adults aged 25-55 in terms of sickness and days missed at work. The findings thus far seem to indicate that general medical well-being is reduced if one resides in a neighborhood that functions poorly, with low levels of community efficacy.

Earls says that he and his colleagues are gratified to have produced research that not only resonates with policymakers, including police departments and urban planners, but also with residents who might be involved in small actions to improve their neighborhoods.

One aspect of violence that continues to trouble Earls is the influence of the media. "I didn't believe the data on TV viewing as a significant exposure to violence for a long time, but now it's really mounted to a point where one can't ignore it. You don't even have to count the number of violent scenes because violent scenes are so common on television, all you have to know is the number of hours that someone watches TV. And as those hours stack up, the probability for violent behavior rises."

Earls, who started his work with violence because it seemed to be one of the most pressing problems of the time, is now taking the community efficacy model to Tanzania, where he hopes to be able to make a difference with the current burden of HIV/AIDS.

bob_brustman@harvard.edu