Public drinking and violence: not just an alcohol problem

Homel R, Tomsen S, Thommeny J.
Journal of Drug Issues


OBJECTIVE:<br/>The aim of this study by Homel et al. was to examine the possible links between various aspects of the public drinking environment and the occurrence of violent incidents and behavior.<br/><br/>METHODOLOGY: <br/>The authors employed a non-experimental ethnographic design, using three hundred hours of unstructured observations in licensed drinking establishments in Sydney, Australia in the late 1980s. Semi-structured interviews were also conducted with police, court officials and security officers. In order to compare situational and management variables in establishments known for their violence and in those without any record of violent activities, the authors examined four high-risk and two low-risk premises, as determined by police reports and exploratory visits. Pairs of observers visited each establishment at least five times, for between two and six hours. Sixteen extra sites were visited at least once each, providing a total of 55 visits to 23 sites in 17 premises (some premises had more than one site for drinking). Those establishments that were intensively studied were all in suburban locations, and included licensed clubs or pubs. Observations were noted after leaving each venue, with separate narrative accounts subsequently being written up by each observer and later coded in group meetings.<br/><br/>FINDINGS/DISCUSSION:<br/>The observers noted a total of 32 assaults that involved some degree of physical violence, representing a rate of 11 per one hundred hours of observation. The authors' analysis of the data suggested that the violence that was observed was not due to any inherent risk in public drinking or the patrons of the establishments. Rather, a number of key variables were found to play a role in the development of violent activities: type of patron, social atmosphere, drinking patterns and behavior of doormen. For the violent premises, the typical patron was a young, working class male. However, it was not the age or the class in particular that was related to the violence. Rather, the authors suggested that the critical factors were the ratio of men to women in the venue and the social links between the males. Proportion of males and the presence of groups of males, especially groups that were unknown to each other, seemed to lead to group rivalry and fights. Less violent, although certainly not excluded from violence, were men by themselves or with women, and people in groups with both genders present. The two most important variables in relation to social atmosphere were found to be comfort and boredom. Plenty of room and ventilation, moderate noise levels and entertainment in the form of quality live acts, television, videos and card machines were able to reduce hostility and levels of drinking, which resulted in less aggression and fewer acts of violence. Crowded facilities and boredom led to higher alcohol intake, more aggression and therefore more violence. Drinking patterns were also related to levels of violence. Violence was found to be highest in those establishments that had charged a high price to enter, but then served greatly discounted drinks. Also lacking in the violent premises was the availability of food to help lower levels of intoxication. The behavior of bar staff did not affect levels of violence, although that of the doormen had a significant influence. Aggressive bouncers were not only indifferent to violence between patrons, but they also were often observed to initiate fights and to encourage continuing violence. Many of the doormen seemed to have been poorly trained and overly concerned with their own masculinity, relating badly to groups of male strangers. The authors concluded that violence in public drinking establishments could not be attributed to any single variable. Rather, it was the combination of the presence of groups of male strangers, low comfort, high boredom, high levels of intoxication and aggressive staff that led to the greatest incidence of violent b