Homicide Rates
In early modern Amsterdam a committee consisting of a professor of medicine and two or three surgeons had the task of inspecting the bodies of all those suspected of having died of unnatural causes.
The committee would write, for example, that they had observed three stab wounds in the chest and belly of an adult man and considered these as absolutely lethal. These medical examiners refrained from any legal conclusions, but often a court clerk added information about the circumstances in which a body had been found. Due to the committee's meticulous description of injuries and the additional information, ambiguous cases are few. This allows the researcher of today to assign the overwhelming majority to one of three categories: murder, suicide or accident. Starting from here, further quantification is possible. We can compute a reliable suicide rate for the period covered by the series, for example. Making statements about the number of accidents, on the other hand, would be less advisable, as it is unlikely that all were systematically investigated. Here we are concerned with murder. Because each inspection report specifies whether the body is that of a baby or an older person, we can subdivide this category into homicide and infanticide. The latter possibility, once more, is not the subject of this chapter. In the case of homicide, the crucial advantage is that we get information about victims regardless of whether or not a perpetrator was caught or even identified by name. The only missing number is that of murdered persons whose bodies were successfully hidden, and it is generally assumed that this number is negligible.Amsterdam is just one example of the post-mortem record keeping which became ubiquitous in western Europe during the Middle Ages. In England a special official, the coroner, had investigated suspect deaths since the thirteenth century.
Unlike the Amsterdam examiners, coroners always presented a legal conclusion, deciding among other things whether death was caused by homicide or suicide. The registration of suicide was less common in other parts of Europe, but that of death due to homicide, again regardless of the identification of a killer, occurred nearly everywhere. The documentation varies from elaborate investigations, as in England, to simple lists of annual numbers of victims. Often, the records in question have been preserved with chronological gaps, and for some towns or regions they still await study. Taken together, however, they provide us with a clear picture of the incidence of homicide in Europe from the late Middle Ages to the present. Figures from all over Europe have been systematically collected and analysed by several scholars, most notably Manuel Eisner.[515] As with every type of crime, absolute numbers have little meaning; we count homicides relative to the total population of the town, region or country in question. An imaginary example serves to illustrate the soundness of this procedure. When five murders per year take place in a town of 10,000 inhabitants, it means that within a generation nearly everyone has direct experience of killing: in her street, in his family, by knowing people who witnessed the act and in still other ways. When a town of i million has thirty per year, it means that the overwhelming majority has only the indirect experience of killing.The count relative to the total population is called the homicide rate. Unless specified otherwise, homicide rates refer to the annual number of killings per 100,000 inhabitants of a geographic entity. In the examples of the previous paragraph, the rates become 50 - typical for many medieval towns - and 3 - not uncommon for a modern European city - respectively. Note that the population of 10,000 in the first example is about (there is no strict rule) the minimum size of the geographic entity.
Otherwise the calculated homicide rate would depend too much on chance. This is equally the case when we only have a figure for one year or so. Thus, fifteen Jacobins were murdered in Aubagne, a town of 7-8,000 inhabitants near Marseille, in 1795,2 and so the calculated homicide rate for Aubagne in 1795 is 200. This figures tells us only that political violence was high in the town; it is not an indication of overall levels of homicide in southern France in the 1790s. Finally, meaningful figures must be based on systematic and trustworthy records: claims by contemporary commentators are to be treated with suspicion, especially if they complain that times are bad.As suggested above, the foremost and basic result of the systematic study of homicide rates in European history is a picture of massive decline from the end of the Middle Ages to the 1950s and 1960s. Medieval rates vary widely per area investigated, at between 20 and 80 or so, but on average they are the highest found over time. Since the Middle Ages falls outside the scope of this chapter, there is no need for further discussion. Homicide rates have increased slightly since about 1970, but this likewise falls outside the present discussion. As it happens, the bulk of the decline had already taken place by the year 1800. Thus, the downward trend in European murder was especially characteristic for the period examined in this chapter. It should be emphasised that this decline of killing refers exclusively to interpersonal violence within the boundaries of developing states. One note of caution: the death count from violence during periods of civil conflict such as the French Wars of Religion (1562-98), the Thirty Years War (1618-48) or the British Civil Wars (1638-51) was much higher than records suggest. This was not just a result of military deaths; during periods of civil conflict all forms of homicide are likely
2 D. M. G. Sutherland Murder in Aubagne: Lynching, Law and Justice during the French Revolution (Cambridge: Cambridge University Press, 2009).
to increase. The greatest fall in the European homicide rate came after 1650 and particularly after 1700.Moreover, the downward trend in murder did not take place at the same pace throughout Europe. Italy, for example, always had much higher homicide rates than England, but in both countries they went down during the early modern period.[516] Even though the Italian homicide rate fell, in accordance with the general pattern, it continued to remain as much as six times higher than that of England in the eighteenth century.[517] Moreover, in the south, in particular, the homicide rate did not approach those rates in the core area of Europe until the early twentieth century. Within the early modern decline, finally, we observe recurrent upsurges of a local and sometimes national character, not all of which are easy to explain. Thus, English homicide rates rose to 5 or 6 in the 1590s, having been 3 or 4 between 1560 and 1590.[518] In Bologna and its contado the rate peaked to no less than 44.5 by the mid seventeenth century, following a severe plague in 1630.[519] Amsterdam witnessed a temporary peak of nearly 10 between 1690 and 1720. The evidence from Aubagne shows how the Revolutionary decade of the 1790s led to an upsurge of personal vendettas in much of southern France, but it is difficult to place this in a long-term pattern since we have few reliable homicide rates for early modern France.[520] To conclude, the early modern downward trend certainly was not unilinear, but it was unmistakably a trend.
Admittedly, in the past a number of victims of violence died who, under modern circumstances, might have been saved. This makes the observed decline all the more significant: by far the largest share of the decline in the rates - by a factor from five to fifteen depending on the area under examination - took place during the early modern period. This was a period without any significant improvement in medical care; expertise and technology did not start to make a difference to homicide rates until the late nineteenth century or even later.
The prevalence of one type of weapon used as opposed to another did affect injured victims' chances of survival. During the upsurge of homicide in Amsterdam, when knife fighting was rampant, about a quarter of homicide victims might have been saved if modern medical treatment had been available.[521] On average, however, the contribution of injury treatment to survival from attack was the same in 1500 as it was in 1800. The steep decline in European homicide rates therefore reflects a real decrease in the incidence of serious interpersonal violence. Moreover, in nearly every society with a high incidence of homicide, this high level is due, in an overwhelming part, to killings in which men are the victims as well as the perpetrators. Hence, the decline of European homicide was primarily a decline of male-on- male fighting.Nevertheless, early modern Europeans were well aware that victims of assault often died as a consequence not of extreme violence but because of the inadequacy of medical care. Thus, an Amsterdam body inspection report of 1678 concludes that poor Jan Valk had not succumbed to his injuries but rather to the bad treatment that these had received. Among other things, those nursing him had failed to renew his bandages regularly, which had caused dirt to slip into his wounds.[522] When the Amsterdam court arrested an offender for assault and his victim died a few weeks later, the magistrates routinely consulted medical opinion on who was to blame. When the doctors were unable to tell whether death was due to the initial wounding or to something else, the offender received the benefit of the doubt and the symbolic punishment of a sword being swung over his head was used instead of decapitation.
Whereas decline in the incidence of homicide forms the basis of our knowledge of interpersonal violence in Europe, the reverse applies to the rest of the world between 1500 and 1800. Here we lack reliable quantitative figures.
An exception is the daybook kept by the alcalde mayor of the Mexican region of Yanhuitlan for twenty-six months in 1606-8. It lists all offences that came to his attention, but records not a single homicide. One offence consisted of inciting a son to commit murder. We should not draw any conclusions from this small sample.[523] For the rest, we are largely ignorant about the quantitative dimension of lethal interpersonal violence in the nonWestern world. In some cases, this may be due to a lack of documentation. In other regions, research is hampered by the very quantity of the documentation. China, for example, has records similar to those left by English coroners for at least the Ming and Qing periods. Apart from the pioneering work of Thomas Buoye, these records have hardly been studied.11 Did China witness a decline in homicide similar to that seen in Europe, or was serious interpersonal violence, for reasons that await explanation, on the rise under the Ming and Qing? Chinese scholars are beginning to work on the data, and their results will transform our understanding of the problem. Comparison between the homicide rates in China and Europe promises to be a fruitful area for future research and collaboration.