The share of violence attributable to drinking

Abstract

While analysts have differed on whether alcohol causes crime, the differences primarily reflect varying definitions of causation. At a population level, more drinking tends to lead to more violence, and less drinking to less. Estimates of the fraction of violence attributable to alcohol have been based both on individual-level and on population-level data. At the individual level, there seem to be cultural differences in the proportion of violent episodes involving drinking, although it is difficult to determine what proportion of these episodes should be attributed to alcohol. At the aggregate level, there are clear variations between different drinking cultures in the fraction of violence attributable to drinking, with the proportion higher in northern and eastern than in southern Europe. These cultural differences underline that there is no single invariant attributable fraction for alcohol's role in violence. The advantages and disadvantages of further individual- and aggregate-level studies are discussed. Suggestions are made for future lines of research on the connection between drinking and violence.