In the historiography of violence there are two prominent approaches: the statistical, concentrating on long-term trends in homicide rates, and the cultural, focusing on themes as ritual and honor. These approaches, largely separate until now, can be integrated through a diachronic analysis of contextual evidence relating to violence and, in particular, homicide. Theories by Elias about the 'civilizing process' and the Stearns' about trends in anger serve as guides. It is hypothesized that, in the long run, (1) homicidal violence became less impulsive, on the average, and more rational; (2) ritual aspects declined in favor of instrumental aspects; (3) killings of intimates gradually made up a larger share of the total number of killings. Evidence from Amsterdam court records, 1650-1810, lends support to these hypotheses. The actual homicide rates in Amsterdam were calculated from body inspection reports, a source which ought to be preferred by historians everywhere. The rates underwent a steep decline from the middle of the fifteenth century until the beginning of the nineteenth.
", language="", issn="0022-4529", doi="", url="http://dx.doi.org/" }