The hypothesis that the association between alcohol consumption and risk of alcohol-related violence would be lower during a period when adolescent alcohol consumption was high, compared to a period when adolescent alcohol consumption was relatively low, was supported in this study. The results suggest that with increasing consumption it is possible that heavy drinking to a lesser extent is associated with deviant behavior such as violence. The finding is in line with the notion of the complexity of underlying mechanisms of the alcohol-violence association. It has been well demonstrated that alcohol plays a significant role in violence and other aggressive behavior. In Norway, total alcohol consumption has increased by approximately 30 percent over the past 10 years, and among young people consumption has increased by more than 60 percent. It is therefore assumed that the level of violence, particularly alcohol-related violence, has increased substantially over the same period. However, on the other hand, it is also possible that with increasing consumption among young people, high levels of alcohol intake and alcohol intoxication to a lesser extent are deviant behaviors and to a lesser extent associated with various problem behaviors. In the latter case, it is assumed that the association between alcohol consumption and the risk of violent behaviors would diminish when alcohol consumption and alcohol intoxication become more common and widespread. This hypothesis was tested by analyzing data from annual surveys among 15 to 20 year olds in Norway for the period of 1990 to 2004. Tables, figure, references