Research

When bar fights get mean, bystanders intervene

Bystanders break up about a third of the fights that occur in bars. Credit: MorgueFile.comAll Rights Reserved.

UNIVERSITY PARK, Pa. -- People are more likely to try to break up a bar fight when they believe the conflict is too violent, or has the potential to become more violent, according to an international team of researchers.

Bystanders break up about a third of the fights that occur in bars and are most likely to intervene in conflicts between males, said Michael Parks, who recently earned his doctorate in sociology at Penn State.

These bystanders used nonaggressive interventions to break up about 65 percent of the fights between two aggressive males. Most bystander interventions were classified as nonviolent interventions, which included verbally stopping the fight, or separating the fighters, according to the researchers.

Parks said the threat of severe violence between males may be a cue for bystanders to intervene.

"Male-to-male aggression between two actors is usually considered by third parties to be the most severe, the type of incident that can lead to severe violence," said Parks.

Fights between males also receive the highest amount of aggressive interventions, when people enter the conflicts themselves as fighters, according to the researchers, who report their findings in the current issue of Aggressive Behavior.

The researchers said that incidents that pitted males against females, which was the most common type of aggression in the study, also received the lowest amount of intervention, with only 17 percent of those incidents broken up by third parties.

Despite the abhorrence that most people have for violence between men and women, the researchers theorized that bystanders may not intervene because they do not believe the encounter will escalate to severe violence, said Parks, who worked with Wayne Osgood and Richard Felson, both professors of criminology and sociology, Penn State; Samantha Wells, assistant professor of epidemiology and biostatistics, University of Ontario, and Kathryn Graham, associate professor of clinical public health, University of Toronto.

"It seems a little upsetting that people didn't intervene in incidents that involved a man harassing a women, but the results showed that this was indeed the case," said Parks. "Our data showed that this type of violence had the lowest level of severity, so one explanation for the lack of intervention in these incidents is that third parties perceived that the events won't escalate into higher levels of violence, something that does not have the potential to be dangerous or an emergency."

Third parties were also more likely to intervene when the fighters appeared to be more intoxicated, said Parks.

Social psychologists in the 1960s theorized that bystanders avoided intervention for a number of reasons, such as a person not acting because he or she believes someone else with intervene, according to Parks. The phenomenon is usually referred to as the bystander effect, he said.

"These results support recent research showing that the bystander effect may not be as dramatic or as strong as once thought in emergencies or dangerous situations," said Parks.

The researchers examined data from 860 relevant incidents from the 1,057 incidents of violence recorded in a 2004 survey of bar violence in Toronto. For the survey, 148 trained observers collected data over 1,334 nights at 118 Toronto bars with a capacity of more than 300 people. Each observer was screened for observational skills and had 25 hours of training to classify violence that included practice sessions in bars.

The National Institute of Alcohol Abuse and Alcoholism supported this work.

Last Updated July 28, 2017

Contacts