ArticlePDF Available

Long-Term Historical Trends in Violent Crime


Abstract and Figures

Research on the history of crime from the thirteenth century until the end of the twentieth has burgeoned and has greatly increased understanding of historical trends in crime and crime control. Serious interpersonal violence decreased remarkably in Europe between the mid-sixteenth and the early twentieth centuries. Different long-term trajectories in the decline of homicide can be distinguished between various European regions. Age and sex patterns in serious violent offending, however, have changed very little over several centuries. The long-term decline in homicide rates seems to go along with a disproportionate decline in elite homicide and a drop in male-to-male conflicts in public space. A range of theoretical explanations for the long-term decline have been offered, including the effects of the civilizing process, strengthening state powers, the Protestant Reformation, and modem individualism, but most theorizing has been post hoe.
Content may be subject to copyright.
Manuel Eisner
Long-Term Historical
Trends in Violent Crime
Research on the history of crime from the thirteenth century until
the end of the twentieth has burgeoned and has greatly increased
understanding of historical trends in crime and crime control. Serious
interpersonal violence decreased remarkably in Europe between the
mid-sixteenth and the early twentieth centuries. Different long-term
trajectories in the decline of homicide can be distinguished between
various European regions. Age and sex patterns in serious violent
offending, however, have changed very little over several centuries.
The long-term decline in homicide rates seems to go along with a
disproportionate decline in elite homicide and a drop in male-to-male
conflicts in public space. A range of theoretical explanations for the long-
term decline have been offered, including the effects of the civilizing
process, strengthening state powers, the Protestant Reformation, and
modern individualism, but most theorizing has been post hoc.
‘‘Symonet Spinelli, Agnes his mistress and Geoffrey Bereman were to-
gether in Geoffrey’s house when a quarrel broke out among them; Sy-
monet left the house and returned later the same day with Richard
Russel his Servant to the house of Godfrey le Gorger, where he found
Geoffrey; a quarrel arose and Richard and Symonet killed Geoffrey’’
(Weinbaum 1976, p. 219). This is an entry in the plea roll of the eyre
court held in London in 1278. The eyre was a panel of royal justices
empowered to judge all felonies and required to inquire into all homi-
cides that had occurred since the last eyre (Given 1977). The story is
Manuel Eisner is reader in sociological criminology at the Institute of Criminology,
University of Cambridge, England. The author thanks Michael Tonry, Kevin Reitz, and
the anonymous reviewers for their helpful comments.
2003 by The University of Chicago. All rights reserved.
0192-3234/2003 /0030-0003$10.00
84 Manuel Eisner
typical of the situational structure of lethal violence in thirteenth-
century London—a disagreement, a quarrel leading to a fight, and a
fight resulting in a death. It could arise in different situations, often
after drinking or over women, sometimes during a feast, but rarely pre-
meditatedly. In two of the 145 instances of homicide recorded in the
London eyre court rolls of 1278, the quarrel broke out after a game of
For quantitatively minded historians, the completeness of the eyre
court records, meticulously drawn up by the clerks of the justices, is a
temptation to count. James Buchanan Given (1977) did this a quarter
century ago, retrieving information on over 3,000 homicides recorded
in twenty eyres in seven counties of thirteenth-century England. Be-
yond simple counts the data also include detailed information about
the sex of offenders and victims, the personal relations between them,
the number of cooffenders, the situations in which the events occurred,
and the decisions taken by the judiciary.
The counts invite attempts at estimating a homicide rate. The 145
cases recorded in the 1278 London eyre court, for example, represent
an average of about six cases per year, as twenty-five years had elapsed
since the last convening of the court. And if the London population of
the time was around 40,000 inhabitants, then the homicide rate—
based on the cases reported to the eyre courtwas around fifteen per
Scientific enticements sometimes come in bunches, and Ted Robert
Gurr (1981) took the issue one step further in an article in this series
entitled ‘‘Historical Trends in Violent Crime: A Critical Review of the
Evidence.’’ Besides the cluster of twenty homicide rates provided by
Given (1977), he reviewed two studies that offered estimates for a few
counties in Elizabethan England (Samaha 1974; Cockburn 1977) and
a series of homicide indictments in Surrey for the period 1663–1802
(Beattie 1974). Gurr plotted the some thirty estimates between about
1200 and 1800 on a graph, added the London homicide rates for the
modern period, and fitted an elegant S-shaped trend curve to the data
points (see fig. 1).
The curve suggested that typical rates may have been about twenty
homicides per 100,000 population in the High and Late Middle Ages,
dropping to ten around 1600, and ending after an extended downswing
at about one per 100,000 in the twentieth century. Gurr interpreted
this secular trend as ‘‘a manifestation of cultural change in Western
society, especially the growing sensitization to violence and the devel-
Long-Term Historical Trends in Violent Crime 85
Fig. 1.—Indicators of homicides per 100,000 population in England, thirteenth to
twentieth centuries. Note: Each dot represents the estimated homicide rate for a city or
county for periods ranging from several years to several decades. Source: Gurr 1981,
p. 313.
opment of increased internal and external control on aggressive behav-
ior’’ (1981, p. 295; see also Gurr 1989). Gurr’s essay easily qualifies as
one of the most influential studies in the field of history of crime re-
search, and the suggestive figure has been frequently reproduced (e.g.,
Daly and Wilson 1988, p. 276; Ylikangas 1998b, p. 10; Monkkonen
86 Manuel Eisner
2001). But it also has raised a large number of questions, most of which
fall into three broad categories.
For one, many historians have questioned whether counting homi-
cides documented in premodern records and comparing respective
rates over eight centuries is scientifically sound. This problem has a
methodological side, including issues of how complete premodern rec-
ords are, what the historical ‘‘dark figure’’ of homicide might look like,
what the relationships among homicide and other forms of interper-
sonal violence are, what impact changing medical technologies have,
and how accurate population estimates are. But some historians of
crime also raise a more substantive problem since the comparison of
numbers over time assumes some comparability of the underlying sub-
stantive phenomenon, namely, interpersonal physical violence. Con-
troversy about this issue divides scholars of more cultural science incli-
nations from those with more behaviorist backgrounds. The former
argue that violent acts are embedded in historically specific structures
of meanings, values, and expectations and claim that considering the
cultural context is essential to understanding historical manifestations
of violence. From this viewpoint, the comparison of homicide rates is
futile and misleading. Behaviorists, by contrast, argue that aggressive
interpersonal behavior is a human universal and that counting the fre-
quencies in manslaughter or murder gives some information about vio-
lence in everyday interactions.
The second set of questions concerns the degree to which Gurr’s
findings can be generalized and specified. Although extrapolation from
a few counties in England to the Western world was probably inevita-
ble at the time when the essay was written, it required a daring as-
sumption. More recently, however, several historians of crime have put
the hypothesis of a long-term decline in interpersonal violence to the
test. Cockburn (1991) contributed an uninterrupted series of indicted
homicide in Kent from 1560 to 1985. Possibly to his surprise, the data
showed a more or less continuous tenfold fall from around three-to-
six offenses per 100,000 to a rate of 0.3–0.7 over more than 400 years.
During the 1970s, Ylikangas (1976) began to examine the long-term
history of violence in Finland. His original findings were increasingly
corroborated by a series of studies on Norway and Sweden, which sug-
gested a coherent pattern in Scandinavia with a massive long-term
decline of homicide rates during the early modern age (Naess 1982;
¨sterberg and Lindstro
¨m 1988; O
¨sterberg 1996). Finally, Spierenburg
(1996) contributed evidence about the Netherlands, primarily focus-
Long-Term Historical Trends in Violent Crime 87
sing on Amsterdam and covering some 700 years. The findings again
coincided with the long-term decline anticipated by Gurr, showing a
drop from about fifty per 100,000 population in the fifteenth century
to about one per 100,000 in the nineteenth century. Thus, at least for
those captivated by transhistorical numbers, there remains little doubt
about the empirical cogency of the broad picture Gurr painted. Sup-
ported by a growing flow of empirical findings, therefore, many have
now started to ask new and more detailed questions (see, e.g., Karonen
2001; Monkkonen 2001; Roth 2001; Eisner 2002b). How can the qual-
ity of the data be ascertained? Can the beginning of the downturn be
more narrowly identified? Does the timing and pace of the decline dif-
fer between large geographic areas? Can processes of pacification be
attributed to specific social groups? Are there sustained periods of in-
creasing levels of homicide rates, and how do they interact with the
declining trend?
The empirical examination of these issues is intricately connected
with the third group of questions: Why? Answering requires some
macrolevel theory of social, cultural, and political developments that is
not among the usual stock of criminological theorizing. Many crimi-
nologists, when theorizing about the effects of long-term social
change, are primarily equipped to explain why urbanization and indus-
trialization should lead to more crime (Shelley 1981). But they are at
a loss when asked to explain declining trends. Historians of crime,
however, found that their empirical observations fit surprisingly well
with the work of the late German sociologist Norbert Elias. In his ma-
jor work, The Civilizing Process (1978), Elias assumed that an interplay
between the expansion of the state’s monopoly of power and increasing
economic interdependence would lead to the growth of pacified social
spaces and restraint from violence through foresight or reflection
(Elias 1978, p. 236). In an attempt to bridge sociological macrotheory
and psychological insight, he suggested that the average level of self-
control would increase to the degree that state institutions stabilize the
flow of everyday interactions. Since these expectations match so well
what crime historians have been finding, Elias has become the major
theoretical reference for scholars who are working in the field and in-
terested in theorizing about the long-term trend. But it is open to de-
bate and the refinement of empirical findings precisely how far the the-
ory of the civilizing process goes toward providing a causal framework
for explaining the long-term decline in lethal violence.
Departing from these groups of questions, this essay is organized in
88 Manuel Eisner
three sections. Section I presents a reanalysis of quantitative estimates
of homicide rates in Europe from the Middle Ages to the present day.
It is based on a much larger set of historical studies than Gurr was
able to examine and discusses the increasing degree to which we can
distinguish different long-term trajectories in the decline of homicide
in various European regions. Section II reviews evidence on various
contextual factors. I examine historical evidence on sex, age, and class
of violent offenders as well as studies that have examined historical pat-
terns of the sex of homicide victims and their relationship with the
offender. Section III explores theoretical approaches used in recent
scholarship for explaining these secular trends, examining how theoret-
ical arguments match the available data and in what ways future re-
search might help to decide between alternative approaches.
I. The Secular Trend in Lethal Violence
What makes any assessment of our knowledge about the long-term
trend in homicide rates relatively difficult is that relevant research has
been published in many different languages, sometimes in difficult-to-
find specialized historical journals, and with widely varying research
questions forming the background of scholarly work. Therefore, this
section builds on a systematic meta-analysis of more than ninety publi-
cations on premodern homicide rates in Europe as well as on a com-
prehensive collection of modern homicide time series in ten countries,
based on national statistics and stretching over periods of more than
120 years. Taken together, these data first confirm the Europe-wide
massive drop—roughly by a factor of 10 : 1 to 50:1 over the period
from the fifteenth to the twentieth century—in lethal interpersonal vi-
olence first observed by Gurr on the basis of English data (1981). Sec-
ond, the transition to declining homicide rates appears to have started
earliest in the northwestern parts of Europe and then to have gradually
diffused to the more peripheral regions of the continent. By the nine-
teenth century, therefore, homicide rates were lowest in the modern-
ized, affluent, and urban regions of Europe, which were surrounded by
a belt of high homicide rates in the periphery. By around 1950, most
European countries experienced their lowest historically known levels
of homicide rates. Since then, an increasing trend has prevailed.
A. Sources: History of Homicide Database
To examine long-term trajectories of lethal violence, I have assem-
bled an extensive database of serial data on homicide in Europe. The
Long-Term Historical Trends in Violent Crime 89
resulting ‘‘History of Homicide Database’’ is an attempt at a compre-
hensive collection of available quantitative information on homicide
over several centuries. The database incorporates two very different
types of sources. National vital statistics providing annual counts of
homicide victims probably constitute the most reliable source. In most
European countries, data series start during the second half of the
nineteenth century, although Swedish national death statistics were in-
troduced in the middle of the eighteenth century (Verkko 1951). The
second main source are statistics on homicides known to the police or
persons accused of murder or manslaughter. Partly based on earlier re-
search (Eisner 1995), the database for modern national homicide statis-
tics includes ten European countries with series of annual data stretch-
ing over more than 100 years.
Prior to the introduction of national statistics, however, statistical
data on homicides accrue from the painstaking archival work of histo-
rians who scrutinize large numbers of judicial sources produced for
widely varying purposes and not originally intended for statistical anal-
ysis. Because of the fragmented judicial structure of premodern Eu-
rope, limitations on the amount of time researchers can spend in ar-
chives, and large gaps in surviving sources, we are left with a patchwork
of local studies. Some of them aim at establishing long-term trends in
serious violence. Many, however, are not primarily interested in esti-
mating the frequency of homicide but aim at gaining insight, through
judicial records, into the administration of justice, the mentalities of
historical epochs, and the lives of ordinary people. However, departing
from the elegant curve boldly drawn by Gurr through some thirty esti-
mates of homicide rates, it is worthwhile to reassess the issue of long-
term trends in violent crime by using the wealth of new research on
the history of crime. Therefore, I systematically collected the results
published in articles and monographs in several languages that pre-
sent data on premodern homicide rates. A number of recent research
reviews facilitated access to this literature ( Johnson and Monkkonen
1996; Schu
¨ssler 1996; Rousseaux 1999a; Blauert and Schwerhoff 2000).
The History of Homicide Database at the time of writing includes
approximately 390 estimates of premodern homicide rates based on
more than ninety publications containing relevant data (Eisner 2001).
Coded variables include information about the geographical area, the
period, the counting units (offenders, victims, and offenses), the type
of sources, the absolute number of homicide cases, the population esti-
mates, and assessments of the quality of the data in the primary publi-
90 Manuel Eisner
cation. When available, I also coded the percentages of female offend-
ers and victims and the percentage of infanticides included in the data.
Although the inclusion of infanticide is not wholly satisfying, it aims
at improving comparability, since the majority of publications do not
allow for separating infanticides and other killings. Three rules guided
the coding process.
First, a threshold decision had to be made about whether to include
a study. I excluded studies that are explicitly based on highly selective
sources such as minor courts, that quote only approximate estimates
without specifying the source, or that are based on extremely small
samples. I included all studies, however, that at least present informa-
tion on the respective territory covered, the type of source, and the
time period covered. If it was sufficiently clear how the information
had been gathered, I ignored occasional warnings by the original au-
thor against using homicide counts for computing rates. Schuster
(2000), for example, extensively explains why he objects to computing
homicide rates on the basis of medieval records. However, since he de-
scribes the origin of his data ( judicial proceedings) and the territorial
unit to which they refer, I decided to include his data on fifteenth-
century Constance in the database.
Second, some publications present time series of counts for each sin-
gle year or for short subperiods. This required a rule about how to
aggregate these counts into larger units in order to reduce random
variation. Generally, I summed up counts for single years and short
subperiods and grouped them into ten-year periods. However, some
flexibility was required to take varying sample sizes into account. Thus,
series for small geographic units, providing low annual numbers, were
aggregated to twenty-year intervals.
Third, whenever possible I used the population estimates quoted in
a publication for computing the homicide rates. If a range of popu-
lation estimates was given, the lower and upper bounds were coded
separately, and the mean was used. Some publications cite counts of
homicides without giving population estimates. In these cases I used
demographic sourcebooks like Bairoch, Batou, and Che
`vre (1988) and
de Vries (1984) for population data.
Most data come from scholarship using one of three types of
sources. A first type encompasses lists of coroners’ inquests or body
inspections of persons purportedly killed irrespective of whether the
suspect was identified. Spierenburg (1996), for example, has used this
kind of source in his analysis of homicide in Amsterdam. The second
Long-Term Historical Trends in Violent Crime 91
type of source is records on the offenders indicted and tried by a court
and constitutes the basis for the vast majority of the data included here.
A third type comprises records of suspected or proscribed homicide
offenders registered by local authorities. An example here are the so-
called medieval Achtbu
¨cher in German urban jurisdictions, which list
persons banished from a city because of homicide, often after they have
fled from the city (see, e.g., Simon-Muscheid 1991; Schu
¨ssler 1998).
Geographically, the data cluster in five large areas. Among those,
England remains exceptional in respect of the wealth of sources that
cover significant territorial units and the number of excellent studies
(for syntheses, see Sharpe 1984, 1988; Emsley 1996), yielding 137 esti-
mates. Historical estimates of homicide rates start in thirteenth-
century England with the impressive analysis by Given (1977) on the
coroners’ rolls submitted to the eyre courts. Hanawalt (1979) then ex-
amined some 16,000 crimes recorded in the jail delivery rolls during
the first half of the fourteenth century. These documents, in which the
key information (e.g., the name and residence of the victim, the type
of crime committed, and the jury’s verdict) was recorded, include al-
most 3,000 homicide cases. From the mid-sixteenth century onward,
comprehensive studies by Beattie (1974, 1986), Cockburn (1977,
1991), and Sharpe (1983, 1984) have traced the development of homi-
cides indicted at the assize courts of several counties over extended pe-
riods of time.
A second area with a wealth of data and a rich tradition of criminal
history research is the Netherlands and Belgium. Beginning with work
by Berents (1976) on crime in fourteenth-century Utrecht, several
studies now cover medieval cities such as Antwerp (Heyden 1983) and
Amsterdam (Boomgaard 1992). Studies by Rousseaux (1986) and
Spierenburg (1996) provide evidence for the early modern period until
the beginning of the nineteenth century.
Since the 1980s, a large body of scholarship has emerged in Scandi-
navia, exploring the very homogenous judicial sources produced by a
centralized and uniform judicial system in existence since the fifteenth
century. O
¨sterberg and Lindstro
¨m (1988) have examined judicial rec-
ords in Stockholm and some smaller cities from 1450 to the mid-sev-
enteenth century. Recent studies by Karonen (1995, 1999), Ylikangas
(1998a, 1998b), and Ylikangas, Karonen, and Lehti (2001) have added
impressive series of estimates for various regions in both Sweden and
Finland, some of which cover more than 200 years.
Fourth, a significant number of studies provide information on long-
92 Manuel Eisner
term trends in Germany and Switzerland, although both the sources
and the judicial structures are extremely complex. Some evidence
comes from an old tradition of local studies on medieval crime and
criminal justice (Buff 1877; Frauensta
¨dt 1881; Cue
´nod 1891). Much
recent scholarship has focused on the Middle Ages with detailed stud-
ies of Cologne (Schwerhoff 1991), Constance (Schuster 1995, 2000),
Nuremberg (Schu
¨ssler 1991), Olmu
¨tz (Schu
¨ssler 1994), Krako
¨ssler 1998), Basel (Hagemann 1981; Simon-Muscheid 1991), and
Zurich (Burghartz 1990). The early modern period had received less
attention, but there are now growing numbers of studies on various
areas in the seventeenth and eighteenth centuries (Schormann 1974;
Henry 1984; Du
¨lmen 1985; Behringer 1990; Lacour 2000).
Although somewhat less thoroughly covered than the other areas,
Italy is the fifth region with a series of studies that permit empirically
based extrapolations. Studies of medieval and renaissance cities include
Bologna (Blanshei 1981, 1982), Florence (Becker 1976), and Venice
(Ruggiero 1978, 1980). Romani (1980) has examined court records in
late sixteenth-century Mantova, and a fascinating study by Blastenbrei
(1995) analyzes wounding reports by medical professionals and judicial
records in late sixteenth-century Rome. Finally, a series of studies on
Padova (Zorzi 1989), Citra (Panico 1991), Sardinia (Doneddu 1991),
Tuscany (Sardi 1991), and Rome (Boschi 1998) yields another cluster
of estimates for the late eighteenth and early nineteenth centuries be-
fore the onset of national statistics.
B. How Reliable Are Estimates of Premodern Homicide Rates?
Historians who analyze extensive criminal justice records usually re-
sort to statistical counts in presenting their findings. But they disagree
whether retrieved historical data should be used to compute homicide
rates, whether such numbers provide any useful information about the
real incidence of killings, and whether long-term historical comparison
is scientifically justifiable (for the recent debate, see Spierenburg 2001;
Schwerhoff 2002). There are five major issues with regard to reliability
and validity.
The first is whether murder or manslaughter cases in premodern
sources qualify as homicides in a modern legal sense, or whether the
data are inflated by cases that would nowadays be regarded as negli-
gent manslaughters or accidents (see, e.g., Aubusson de Cavarlay 2001,
p. 27). Legally, homicide represented a felony throughout Europe
since the High Middle Ages, and definitions invariably included some
Long-Term Historical Trends in Violent Crime 93
notion of intentional aggression. Philippe de Beaumanoir, officer of
the French Crown in the thirteenth century, defined homicide in the
following way: ‘‘Homicide: When one kills another in the heat of a
fight, in which tension turns to insult and insult to fighting, by which
one often dies’’ (cited in Rousseaux 1999a, p. 145). A reading of the
case descriptions in historical sources suggests that, some exceptions
notwithstanding, most cases would qualify as criminal acts. In this vein,
DeWindt and DeWindt (1981, p. 54) conclude—based on a close ex-
amination of 111 presentments of homicide in the Huntingdonshire
Eyre of 1286—that 90 percent were definite acts of aggression or vio-
Another important issue concerns the quality of the immensely var-
ied judicial sources used in historical analyses. Spierenburg (1996), for
example, has argued that homicide estimates based on court records
may yield considerable underestimates because only a fraction of of-
fenders were captured and brought to trial. His comparison between
body inspection records and the judicial sources in late medieval and
early modern Amsterdam suggests that possibly as few as 10 percent
of all homicides may have resulted in a suspect being brought to court
(see also Boomgaard 1992; Spierenburg 1996, p. 80). Even if this is an
extreme estimate, it suggests that early court records constitute selec-
tive evidence. In order to examine this issue empirically, Monkkonen
(2001) recently proposed the use of capture-recapture methods, which
yield estimates of the size of the unknown underlying population of
offenders based on comparisons between different types of sources
(e.g., court records, coroners’ inquests, proscriptions, reports in dia-
ries, or printed sources). It remains to be seen whether such a strategy
can clarify the issue of historical dark figures. Yet it is uncontroversial
that the progressive shift toward more efficient prosecution has the ef-
fect of underestimating the long-term decline in lethal violence (Stone
1983, p. 23).
A related issue is whether homicide rates constitute a leading indica-
tor of overall levels of violence through long historical periods (see,
e.g., Schuster 2000). For present-day societies, homicide appears quite
adequately to reflect variation in overall violence. In the United States
and Great Britain, for example, trends in assault, as measured by the
National Crime Victimization Survey, are highly correlated with fluc-
tuations in homicide rates (Langan and Farrington 1998). Moreover,
cross-national homicide rates are also significantly correlated with lev-
els of robbery, assault, and sexual violence as measured by the Interna-
94 Manuel Eisner
tional Crime Victimization Survey (Eisner 2002a). Yet until recently
there seemed to be no way directly to address this question histori-
cally since alternative data for measuring historical levels of violence
were not available. But recent research for several sixteenth- and
seventeenth-century Swedish cities now suggests that time series of re-
corded assault and homicide are surprisingly parallel in both trends
and fluctuations (Karonen 2001). However, since historical trends for
other types of violence that differ from those for homicide cannot be
ruled out, the subsequent analysis is based on the assumption that ho-
micide may be cautiously construed as an indicator only of serious in-
terpersonal violence.
A fourth, somewhat overemphasized, issue concerns the variability
of homicide rate estimates because of the small sizes of geographic
units, the small number of cases used for computing respective rates,
or both (Aubusson de Cavarlay 2001). With only a handful of estimates
based on a few killings each, this would be a serious issue. But with
several hundred estimates, many based on large numbers of homicides,
covering both urban and rural areas, and converging to coherent pat-
terns despite heterogeneous sources, one may safely assume that the
data are not random noise. Likewise, the low precision of population
estimates, although important, should probably not be regarded as an
insurmountable issue. Better population data are important for more
accurate estimates. Fortunately, however, my interest is not to com-
pare differences in the magnitude of 50 or 60 percent over time, but a
ten- to possibly fifty-fold decline. Therefore, quite considerable inac-
curacy in population estimates—especially if it is randomly distributed
between the different studies—can be accepted.
Probably the most important ‘‘distorting’’ factor in comparisons of
homicide across long periods is the interplay between changes in the
technology of violence and growth in medical knowledge. The lives of
a large proportion of those who died from the immediate or secondary
consequences (e.g., internal bleeding, infections) of a wound in any so-
ciety before the twentieth century could have been saved with modern
medical technology. But until recently it seemed wholly impossible to
estimate the size of this effect. Monkkonen (2001), however, has pro-
posed to use information about the elapsed time from injury to death
as a rough indicator of the potential impact of modern medical tech-
nology. He argues that most deaths occurring within the first one to
two hours after the injury are probably not preventable even with
modern medicine, while the vast majority of those occurring after
Long-Term Historical Trends in Violent Crime 95
twenty-four hours could be prevented by modern technology. A series
of studies yields quite consistent results in regard to the typical time
from assault to death before the twentieth century. In mid-nineteenth-
century New York, about one-fourth of victims died immediately and
another fourth within the first twenty-four hours (Monkkonen 2001).
In seventeenth-century Castile, about 37 percent of the victims died
immediately and another third within the first twenty-four hours
(Chaulet 1997, p. 22); Spierenburg (1996) estimates that somewhat less
than half of victims in seventeenth-century Amsterdam died immedi-
ately. Even if these estimates are far from precise, they give a rough
idea about the order of magnitude, by which the lethal consequences
of violence might have declined with late twentieth-century technol-
ogy. Most authors agree, however, that changes in medical technology
are unlikely to have had any major impact on the chances of surviving
a wounding before the late nineteenth century.
C. Results
Figures 2–7 graphically display the estimates collected for the five
areas. Figure 2 includes all local premodern estimates for the whole of
Fig. 2.—Overall trend in homicide rates, all premodern local estimates and four
national series. Note: All 398 local estimates from the History of Homicide Database;
national series for Sweden, England and Wales, Switzerland, and Italy.
96 Manuel Eisner
Fig. 3.—England: local estimates and national series. Source: History of Homicide
Database; see text for details.
Europe along with national series for four countries. Figures 3–7 show
trends for the five geographic areas over longer periods of time. In
each figure, dots represent single local estimates based on the mean
year of the investigated period and the mean homicide rate, if upper
and lower bonds for the respective population were given. For the pre-
modern period, lines show selected continuous series of estimates for
one geographic subunit. During the modern period, lines show na-
Fig. 4.—Netherlands and Belgium: local estimates and national series. Source: His-
tory of Homicide Database; see text for details.
Long-Term Historical Trends in Violent Crime 97
Fig. 5.—Scandinavia: local estimates and national series for Sweden. Source: History
of Homicide Database; see text for details.
tional homicide rates based on vital statistics or police statistics. For
two reasons, the graphs use a logarithmic scale for the vertical axis.
First, estimated homicide rates range between over 100 and 0.3 per
100,000 population over the centuries. Hence, variation at lower abso-
lute levels would become invisible with a linear scale. Second, a loga-
rithmic scale has the advantage of making relative differences compara-
ble across the whole range of absolute levels. In addition, table 1 shows
Fig. 6.—Italy: local estimates and national series. Source: History of Homicide Data-
base; see text for details.
98 Manuel Eisner
Fig. 7.—Germany and Switzerland: local estimates and national series. Source: His-
tory of Homicide Database; see text for details.
average estimates of homicide rates for specified subperiods. These es-
timates are based on the unweighted averages. It could be argued,
though, that averages should be weighted for differences of population
size and the length of the period on which the estimate is based. How-
ever, since the larger urban areas are more thoroughly covered anyway,
it did not seem appropriate to increase their contribution to the overall
mean even more. The results can be summarized in the following
1. Overall Secular Decline. The data displayed in figure 2 suggest a
common trend in homicide rates across western Europe. Three main
conclusions can be drawn. First, the total of all estimates is located in
a band in which upper and lower limits gradually move toward lower
levels from around 1500 until the mid-twentieth century. Taken to-
gether, the empirical evidence suggests a continent-wide gradual de-
cline of serious interpersonal violence. Computing averages per cen-
tury and including all estimates yields the series displayed in table 2.
Comparing these estimates with the original curve plotted by Gurr
shows impressive consistency. Adding new data, it appears, has little
impact on the overall pattern.
Second, for each century, the estimates show a large degree of dis-
Homicide Rates in Five European Regions
Netherlands Germany
and and
Period England Belgium Scandinavia Switzerland Italy
centuries 23 47 ⋅⋅⋅ 37 (56 )
Fifteenth century ⋅⋅⋅ 45 46 16 (73 )
Sixteenth century 7 25 21 11 47
Seventeenth century:
First half 6 (6 ) 24 11 (32 )
Second half 4 9 12 (3 ) ⋅⋅⋅
Eighteenth century:
First half 2 7 3 (7 ) (12 )
Second half 1 4 .7 (8 ) 9
1800–1824 2 2 1.0 318
1825–49 1.7 ⋅⋅⋅ 1.4 415
1850–74 1.6 .9 1.2 212
1875–99 1.3 1.5 .9 2.2 5.5
1900–1924 .8 1.7 .8 2.0 3.9
1925–49 .8 1.3 .6 1.4 2.6
1950–74 .7 .6 .6 .9 1.3
1975–94 1.2 1.2 1.2 1.2 1.7
Source.History of Homicide Database.
Note.Data are arithmetic means of all available estimates for a given period and
region. Estimates based on local data are rounded to the next integer. Figures in paren-
theses are particularly unreliable because they are based on fewer than five estimates.
Figures in italics are based on national statistics.
Overall Homicide Rates in Europe, Thirteenth
to Twentieth Centuries
Period Homicide Rate Number of Estimates
Thirteenth–fourteenth centuries 32 76
Fifteenth century 41 25
Sixteenth century 19 76
Seventeenth century 11 107
Eighteenth century 3.2 65
(excluding national series)
Nineteenth century 2.6 Mostly national series
Twentieth century 1.4 National series
Source.History of Homicide Database.
100 Manuel Eisner
persion with a ratio between the lowest and the highest bundle of esti-
mates typically being around 1:10. This variation may arise from a
number of different sources. There may be measurement errors (sys-
tematic or random) influencing each estimate for all kinds of reasons
(e.g., gaps in the sources, unrecorded homicides, or faulty population
estimates). Variability may be the result of historically contingent con-
ditions, such as food crises, local warfare, or banditry, that influence
the local level of serious violence. Based on our knowledge of the large
local variability of homicide rates in present societies, we should not
expect anything else when working with historical data. Finally, varia-
tion in each period may reflect large-scale systematic differences be-
tween areas of the European continent. As I argue below, the evidence
suggests that a large-scale pattern of geographic variation emerges
from the sixteenth century onward and is due to different trajectories
in the secular transition from high to low levels of lethal violence.
Third, there appears to be a significant process of convergence be-
tween the mid-nineteenth and mid-twentieth centuries, when very lit-
tle variation remains between various countries of western Europe.
2. Relative Homogeneity in the Middle Ages. Before 1500, the data-
base includes about 100 different estimates of homicide rates. They
come from a widely dispersed sample of areas, primarily larger cities
(i.e., more than 5,000 inhabitants) but also some small towns and rural
territories, and are based on a staggering variety of sources. However,
the evidence suggests a startlingly homogeneous pattern throughout
Europe. Evidence based on coroners’ rolls in fourteenth-century Ox-
ford and London result in estimates in the order of twenty-five to 110
homicides per 100,000 (Hanawalt 1976; Hammer 1978), while esti-
mates for other areas of England typically vary between eight and
twenty-five homicides per 100,000. In the south of Europe, data from
judicial archives in Florence (Becker 1976; Cohn 1980), Venice (Rug-
giero 1980), Bologna (Blanshei 1982), and Valencia (Garcia 1991) yield
estimates between a low of ten and a high of 150 homicides per
100,000. And studies on an extensive sample of urban jurisdictions in
what are now Belgium, the Netherlands, Germany, Switzerland, and
northern France again result in estimates between a low of six and a
high of about 100 homicides per 100,000 of the population. Overall,
there is considerable haphazard variation between individual estimates,
which may result from peculiarities of the surviving sources or reflect
local economic and social conditions, political conflict, or the intensity
of law enforcement. However, and more important, there appears to
Long-Term Historical Trends in Violent Crime 101
be little systematic difference during this period when larger areas of
Europe are compared.
3. Increasing Geographic Differences from the Late Sixteenth Century
Onward. By the late sixteenth century, however, significant large-
scale differences begin to emerge. They suggest that the secular trajec-
tory from high to low levels of lethal violence may have had different
shapes in different areas. More particularly, homicide rates, as esti-
mated on the basis of indictments brought before the assize courts in
Elizabethan and early Stuart England, typically range between three
and ten per 100,000 (see fig. 3). Experts in the field seem to agree that
these estimates indicate a real decline compared with the late Middle
Ages (Sharpe 1996, p. 22). Yet because of the lack of records between
the late fourteenth and the mid-sixteenth centuries, the precise period
when the secular downturn started cannot be identified.
In the Low Countries, too, evidence indicates a marked shift from
the high homicide pattern during the sixteenth century (see fig. 4).
Studies by Boomgaard (1992) and Spierenburg (1996, pp. 80 ff.) based
on body inspection reports suggest that homicide rates in Amsterdam
may have declined from about forty to twenty per 100,000 during the
sixteenth century. For Brussels, Vanhemelryck (1981) suggests that the
rate of homicides recorded by the judiciary may have declined from
about twenty per 100,000 in the fifteenth century to about ten per
100,000 in the sixteenth century. The pattern becomes even clearer by
the end of the seventeenth century, when Spierenburg (1996, p. 86)
calculates a homicide rate of about four per 100,000 for Amsterdam.
The estimate for Brussels, based on the whole century, is four to five
per 100,000 (depending on the population estimate). And another cen-
tury later, a few scattered figures suggest that the homicide rate in late
eighteenth-century Belgium or the Netherlands typically ranged be-
tween 0.7 and about three per 100,000.
The Scandinavian countries show similar trends but differences in
timing. Figure 5 shows that homicide rates remained at very high lev-
els until the first decades of the seventeenth century. Estimates based
on the thorough work by Karonen (2001) yield homicide rates of thirty
to sixty per 100,000 in Turku, Arboga, and Stockholm around the turn
of that century. These rates, considerably higher than anything found
in England or the Netherlands at this time, may have been the result
of an upsurge from the mid-sixteenth century, when estimates tend to
be considerably lower. From about 1620 onward, however, Scandina-
vian scholars observe a staggering decline in homicide rates. By the
102 Manuel Eisner
second half of the seventeenth century, rates had dropped to around
eight to ten homicides per 100,000, while estimates for early eigh-
teenth-century Sweden were in the region of about four per 100,000.
By 1754, when national death statistics were initiated, the Swedish ho-
micide rate had dwindled to a mere 1.3 per 100,000.
Although lethal interpersonal violence had declined to both histori-
cally and cross-culturally remarkably low levels by the late eighteenth
century throughout northern Europe, a very different trend is found
in southern Europe (see fig. 6). Admittedly, the data from Italy have
large gaps, and we lack long-term continuous series similar to those
available in England, the Netherlands, and Sweden. However, the con-
trast is so stark that there is no reason to doubt its main characteristics.
Departing from the handful of estimates for late medieval and early
Renaissance cities, studies by Blastenbrei (1995) on Rome and by Ro-
mani (1980) on the Duchy of Mantova give some idea of typical homi-
cide rates around 1600. Blastenbrei shows that medical professionals in
late sixteenth-century Rome were registering some twenty-five to
thirty-five killings per year, which yields an estimated homicide rate of
thirty to seventy per 100,000. In a similar vein, the criminal justice
records in Mantova include some ten to fifteen cases of murder or
manslaughter each year. Romani (1980) estimates the population at
30,000–40,000, which in turn suggests a homicide rate of between
twenty-five and fifty-five per 100,000.
Another two centuries later a sample of figures suggests some de-
cline. In this period, the rate of convicted homicide offenders in Tus-
cany or Padova can be estimated at between four and ten per 100,000
(Zorzi 1989; Sardi 1991). In the south of Italy as well as in Sardinia,
however, late eighteenth-century homicide rates were still well above
twenty per 100,000 (Doneddu 1991), and Boschi’s figures for Rome
around 1840 put the homicide rate at over ten per 100,000 (Boschi
It is hard to say whether Germany and Switzerland followed the
northern European pattern of a sustained decline or whether the long-
term trajectory resembles the Italian pattern of high homicide rates
well into the beginning of the industrial revolution. Because of the het-
erogeneity of the sources and the political fragmentation of territories,
but possibly also because of a lack of scholarly interest in examining
quantitative long-term trends, the existing data make solid conclusions
As figure 7 shows, the data first suggest a dramatic drop in homicide
Long-Term Historical Trends in Violent Crime 103
rates at the beginning of the fifteenth century. However, this apparent
trend probably reflects a shift in the sources used for historical research
rather than any real change. Most estimates for the fourteenth century
are based on banishment records. These documents include a large
proportion of suspects, who had fled an urban jurisdiction after a crime
and may never have been put to trial (Schu
¨ssler 1994). In the early fif-
teenth century, the practice of banishment without formal trial fell out
of use, and studies for this period are mostly based on proceedings of
the local judiciary. Yet many scholars argue that early modern judicia-
ries may have been able to deal with only a small fraction of actual
crimes, including homicide. By around 1600, estimates for the cities of
Cologne and Frankfurt range between six and sixteen homicides per
100,000, a figure similar to those in England or the Netherlands in this
period (Du
¨lmen 1985, p. 187). However, a series of estimates for sev-
eral areas in southern Germany and Switzerland and primarily based
on offenders tried by the judicial authorities typically hover between
two and ten during the late eighteenth and early nineteenth centuries.
These are consistently higher estimates than are found during this pe-
riod in northern Europe, and they suggest that the frequency of serious
violence in Switzerland and southern Germany may have been some-
where between the low rates found in the north and the high rates
found in the south of Europe.
4. Center-Periphery Structures in the Nineteenth Century. By the be-
ginning of the nineteenth century, the impact of different long-term
trajectories in the evolution of serious interpersonal violence since the
Middle Ages had created a pattern of large-scale regional differences
within Europe. Since the late sixteenth century, England and the
Netherlands had moved a long way in the transition from a high vio-
lence society to one characterized by a much more pacified mode of
everyday behavior. In Sweden, the same process seems to have started
later, occurred faster, and produced a similar result. In Italy, however,
homicide rates appear to have moved little from their late medieval and
early modern levels, especially so in the south and on the islands. The
development in Germany and Switzerland is hard to track, but by the
early nineteenth century a north-south divide may have come into ex-
istence, with higher levels characteristic of many areas in Switzerland
and southern Germany.
Against this background, it seems worthwhile to summarize the
large-scale ecology of lethal violence in Europe around 1900, origi-
nally described by Ferri (1925) and Durkheim (1973) but discernible
104 Manuel Eisner
Fig. 8.—Homicide rates around 1880. Sources: History of Homicide Database;
Verkko 1951; Chesnais 1981.
in much greater detail now thanks to additional recent work (see, e.g.,
Chesnais 1981; Johnson 1995; Eisner 1997; Thome 2001). On the
level of nations, the pattern resembles a trough with low homicide
rates across the highly industrialized countries of northern Europe, in-
cluding Germany and France. A rim of high-homicide countries sur-
rounds the trough and includes Portugal, Spain, Italy, and Greece in
the south, and all eastern European countries and Finland (see fig. 8).
By the end of the twentieth century, this large-scale geographic pattern
changed. While homicide levels in eastern Europe remained high,
rates in southern European countries have converged to levels typically
found in northern and western Europe (see fig. 9).
Within countries, nineteenth-century regional differences appear to
have followed a distinctly similar pattern. In Italy, homicide rates were
higher in the rural south with its low literacy rates than in the more
industrialized north. In 1880–84, for example, the homicide rate varied
Long-Term Historical Trends in Violent Crime 105
Fig. 9.—Homicide rates around the end of the twentieth century. Note: Data re-
fer to completed homicides known to the police, 1998–2000. Source: Interpol at
http:/ / /Public/Statistics/ICS/downloadList.asp (last accessed February
2, 2003).
from a high of 45.1 in the district of Palermo to a low of 3.6 in the
district of Milan (Chesnais 1981). French maps suggest higher levels
of homicide in southern France than in the prosperous and urbanized
north (Durkheim 1973). Within Germany, homicide and assault rates
were generally higher in areas characterized by low urbanization, low
proportions of professionals and public servants, and a high overall
death rate ( Johnson 1995). In late nineteenth-century Poland, homi-
cide and assault were more common in the countryside than in cities;
in Switzerland, too, homicide rates were negatively correlated with
levels of urbanization and industrialization (Kaczynska 1995; Eisner
All in all, it seems, a center-periphery dimension characterized the
geographic distribution of lethal violence across late nineteenth-
106 Manuel Eisner
century Europe. Homicide was low in the centers of modernization
characterized by high urbanization, industrialization, literacy, and edu-
cation. Elevated levels of violence, in turn, were found throughout the
peripheral areas with high birth rates, high illiteracy rates, and a pre-
dominantly rural population. This pattern, I tentatively conjecture, was
the result of differential long-term pathways, some of which can be
traced back to the beginning of the early modern period.
5. The Past 120 Years: The U-Shaped Pattern. From about the
1880s onward, death statistics and police statistics cover the majority
of western European countries. These data permit us quite accurately
to trace main trends in homicide rates over the past 120 years. The
main message can be summarized in three points. First, a comparison
of the 1950s with the 1880s suggests that the frequency of lethal vio-
lence fell by at least another 50 percent even in northern European
countries, and considerably more in the south. Indeed, as Gurr, Gra-
bosky, and Hula (1977; see also Gurr 1989) had shown, declining
trends were the predominating pattern for other types of violence (e.g.,
serious assault, robbery) as well as for property crime in many Western
societies (Gatrell 1980). In a sense, therefore, homicide rates around
1950 may serve as a benchmark for the lowest level of interpersonal
lethal violence as yet attained in any known Western society. It stands
at about 0.4–0.6 deaths per year per 100,000 inhabitants. Second, the
data demonstrate a rapid convergence of homicide rates between the
late nineteenth century and the 1960s. By then, cross-national differ-
ences within western Europe had become inconsequential and have re-
mained small since. Third, the data from 1950 until the early 1990s
point to an upsurge of homicide rates throughout most of Europe ac-
companied by a much sharper rise in recorded levels of assault and
These increases occurred despite advances in medical technology
throughout the twentieth century, which are likely significantly to have
dampened this latest increase. The main trend over the past 150 years,
therefore, corresponds to the U-shaped pattern identified earlier by
Gurr and his collaborators (Gurr, Grabosky, and Hula 1977).
6. Countertrends. The well-documented increase in criminal vio-
lence between the 1950s and the early 1990s may well be just one of
several periods in which violence rates increased over several decades.
For obvious reasons, we know very little about earlier medium-term
periods of increasing interpersonal violence. It might be interesting to
know, for example, whether the upswing in lethal violence documented
Long-Term Historical Trends in Violent Crime 107
for Sweden between the 1790s and the 1840s also occurred in other
European areas. But there is too little evidence to address this question
even tentatively. Recently, however, Roth (2001) offered a fascinating
observation on trends between 1550 and 1800. Comparing time series
for England, Scandinavia, and France, he found evidence of a similar
trend of sharply increasing homicide rates between the 1580s and the
1610s, followed by a continuous drop thereafter. The coincidence be-
tween areas far apart from each other is remarkable. As for England,
Roth suggests that demographic pressure, economic depression, crop
failure, the militarization of culture, and military demobilization to-
gether may have caused homicide rates to soar in the late sixteenth and
early seventeenth centuries. Yet the relative importance of those fac-
tors remains to be explored. Furthermore, one might wonder whether
similar factors played a role in other areas during that time or if the
parallel trends are coincidental.
7. Some Evidence for Other Areas. Research in recent years has ex-
plored long-term trends for geographic areas that I have not discussed.
For example, there is now good evidence for developments in Ireland
from the beginning of the eighteenth century until 1914 (Garnham
1996; Finnane 1997). The data from indictments in two Irish counties
through 1801 suggest homicide rates of around four to seven per
100,000 population in the 1740s and 1750s. These are considerably
higher rates than those found in any of the English counties investi-
gated by Beattie (1986) and Cockburn (1991). By around 1900, na-
tional homicide rates were down to below two per 100,000, and Fin-
nane (1997) concludes that there is strong evidence of a declining
trend in interpersonal violence during the nineteenth century.
Furthermore, Roth (2001) has recently presented data on European-
American adult homicides in Connecticut, Massachusetts, Maine, and
New Hampshire from 1630 to 1800. His work is particularly notable
because it uses sophisticated capture-recapture methods in order more
accurately to estimate homicide rates for European colonists. Before
1637, during the era of frontier violence, he finds that the homicide
rate in colonial New England stood at over 100 per 100,000 adults. It
then dropped to about seven to nine for the next four decades,
which—assuming some underreporting in the English assize court
data—may have been quite similar to the rate that probably prevailed
in southeastern England. It fell again at the beginning of the eigh-
teenth century and reached a low of about one per 100,000 adults at
the end of the century. Examining the causes of this massive decline,
108 Manuel Eisner
Roth argues that the sudden decline correlated with increased feelings
of Protestant and racial solidarity among the colonists (2001, p. 55).
II. Contextual Trajectories
To this point I have primarily traced the long-term trajectory of over-
all levels of lethal interpersonal violence. But historians of crime have
long underscored that these trends need to be embedded in an analysis
of contextual change, including, for example, the cultural meaning of
violence, the typical situations giving rise to conflict and aggression,
the characteristics of offenders and victims, and the framework of legal
and judicial reactions (Rousseaux 1999a). Similarly, criminologists in-
creasingly have become interested in disaggregating violence trends
by, for example, offender-victim configurations (Wikstro
¨m 1992), of-
fender age groups (Blumstein 2000), or weapons used in the offense
(Wintemute 2000). And they found such distinctions highly valuable
for understanding the determinants of change in overall crime levels.
As the consensus about the overall decline grows, therefore, estab-
lishing the long-term variation, and stability, of contextual characteris-
tics of violent crime will become increasingly important. A better un-
derstanding of contextual dimensions may provide the decisive cues
for more refined interpretations of the transition from high to low
homicide levels during the process of modernization. The following
analyses explore some relevant dimensions. They primarily stick to a
statistical framework, presenting numerical evidence on factors that
criminologists find relevant when describing the basic characteristics
of violent crime. The evidence partly derives from the publications
comprised in the History of Homicide Database. I also include data
from a series of studies that have examined historic patterns of robbery
or assault.
The available evidence suggests impressive historical stability in
some respects. Most particularly, both the sex distribution and the age
distribution of serious violent offenders appear to have remained
within very narrow limits over several centuries. However, changes
are apparent along other core dimensions. First, the overall decline in
homicide rates regularly appears to coincide with a decline in the pro-
portion of male-to-male killings. In a similar vein, the drop appears to
be inversely related to a (relative) increase in family homicides. Finally,
evidence suggests that the overall drop in homicide rates may have
been accompanied by a gradual withdrawal of elites from interpersonal
Long-Term Historical Trends in Violent Crime 109
A. Sex of Offenders
Among contextual characteristics of violent crime, the sex distribu-
tion of offenders is the most obvious starting point. This information
should be ascertainable from any historical source that provides of-
fenders’ first names. Unfortunately, however, many historical studies
on crime do not present individual-level data, and only recently have
historians of crime become interested in variability in gender ratios
among offenders. In this respect, research by Feeley and Little (1991)
and Feeley (1994) has exposed fascinating observations on historical
variation in overall female participation rates. Feeley and Little (1991)
first examined Old Bailey Sessions Papers from 1687 to 1912. They
found that women constituted well over one-third of the caseload dur-
ing the eighteenth century, after which the proportion steadily de-
clined to about 10 percent around 1900. Feeley (1994) reviewed a
number of studies that had examined female criminality in early mod-
ern Europe. He found a pattern that should be surprising for those
who believe, with Gottfredson and Hirschi (1990, p. 45), that gender
differences are invariant over time and space. Research on urban areas
in the eighteenth-century Netherlands shows that women accounted
for up to 75 percent of the criminal cases. And even after discounting
the various types of ‘‘moral offenses,’’ the figures remain high and
striking. In Amsterdam, women comprise 50 percent of the persons
accused of property offenses, and similar proportions were found in
other northern European cities (van de Pol 1987; Diederiks 1990).
Feeley (1994, p. 263) argues that the exceptionally high involvement
of women in property crime—found in many eighteenth-century ur-
ban areas throughout northern Europe—reflects their high participa-
tion in the preindustrial mercantile economy. As production shifted
away from the family to the factory, however, women were again rele-
gated to the home, which in turn may explain their gradual retreat
from property crime throughout the nineteenth century.
A series of estimates for the percentage of female offenders from
1200 to 2000 show that female involvement in violent crime has been
much less susceptible to social change. Records across Europe over 800
years consistently show that the proportion of women committing
homicide (excluding infanticide), assault, or robbery was hardly ever
above 15 percent and typically ranged between 5 and 12 percent. Table
3 summarizes the major findings. Exceptions most probably result
from problems in classifications (e.g., inclusion of verbal insult in as-
sault) rather than real differences.
Female Offenders as a Percentage of All Offenders in Various Historical Studies
Region Assault Robbery Homicide Crime Source
Fourteenth–sixteenth centuries:
England, 1202–76 ⋅⋅⋅ ⋅⋅⋅ 8.6 ⋅⋅⋅ Given (1977, p. 48)
England, various counties, 1300–1348 ⋅⋅⋅ 5.1 7.0 9.8 Hanawalt (1979, p. 118)
Cracau, 1361–1405 1.0 .0 1.0* 10.0 Schu
¨ssler (1998, p. 313)
Zurich, 1376–85 1.4 ⋅⋅⋅ ⋅⋅⋅ 4.0 Burghartz (1990, p. 80)
Avignon, 1372 21.0† ⋅⋅⋅ ⋅⋅⋅ 23.7 Chiffoleau (1984, p. 250)
Arras, 1400–1436 13.7‡ ⋅⋅⋅ ⋅⋅⋅ Muchembled (1992, pp. 34, 89)
Constance, 1430–60 4.7 ⋅⋅⋅ .0* 17.2 Schuster (2000, p. 71)
Douai, 1496–1519 ⋅⋅⋅ ⋅⋅⋅ 1.0 ⋅⋅⋅ Fouret (1984)
Amsterdam, 1490–1552 14.0 ⋅⋅⋅ 3.0 15.0 Boomgaard (1992)
Arras, 1528–49 5.0 ⋅⋅⋅ ⋅⋅⋅ 20.0 Muchembled (1992, pp. 34, 89)
Brussels, 1500–1600 8.2 ⋅⋅⋅ ⋅⋅⋅ 7.4 Vanhemelryck (1981, p. 314)
Cologne, 1568–1612 4.4 4.7 5.7* 22.9 Schwerhoff (1995, p. 91)
Seventeenth and eighteenth centuries:
Rural areas near Trier, late sixteenth to early
eighteenth centuries 12.3 4.5 3.7* 5.2 Lacour (2000, p. 535)
Castile, 1623–99 ⋅⋅⋅ ⋅⋅⋅ 1.4* ⋅⋅⋅ Chaulet (1997, p. 17)
Bavaria, 1600–1649 4.5 ⋅⋅⋅ 2.9* 12.4 Behringer (1995, p. 65)
Bavaria, 1685–89 5.0 .0 4.5* 13.2 Behringer (1995, p. 67)
Surrey, 1663–1802 18.2 7.9 13.0 23.9 Beattie (1975, p. 81)
Leiden, 1678–1794 7.8 41.05.5* 47.3 Kloek (1990, p. 8)
Gent, 1700–1789 10.2 ⋅⋅⋅ ⋅⋅⋅ 24.6 Roets (1982)
Stockholm, 1708–18 41.0 ⋅⋅⋅ 43.0 67.0 Andersson (1995)
¸on, northern France, 1715–45 20.0 ⋅⋅⋅ ⋅⋅⋅ 33.5 Champin (1972)
Armagh, Ireland, 1736–95 6.5 ⋅⋅⋅ 7.6 9.7 Garnham (1996)
Neuchatel, 1707–1806 6.2 ⋅⋅⋅ ⋅⋅⋅ 14.7 Henry (1984, p. 660)
Nice, 1736–92 ⋅⋅⋅ ⋅⋅⋅ 3.0 ⋅⋅⋅ Eleuche-Santini (1979)
Rural northern Germany, 1680–1795 3.5 ⋅⋅⋅ ⋅⋅⋅ 18.5 Frank (1995, p. 235)
Late nineteenth-century Germany 8.0 ⋅⋅⋅ 16.0 ⋅⋅⋅ von Mayr (1917, p. 754)
Late twentieth century:
England and Wales, 1997 14.7 8.5 11.9* 23.0 Home Office (1998)
Italy, 1998 16.0 6.9 5.2 15.2 Istituto Nazionale di Statistica (2000)
Germany, 1997 12.0 7.5 10.0* 23.0 Bundeskriminalamt (1998)
United States, 1997 15.0 8.0 10.0 32.0 Federal Bureau of Investigation (1998)
* Infanticide explicitly excluded.
† Includes insult.
‡ Minor assault only.
§ Serious assault only.
Includes pickpocketing.
112 Manuel Eisner
This is not the place to discuss the causes of that apparent long-term
stability. But accepting these data as reasonably valid estimates of
involvement in violent crime probably means that sex is not a relevant
variable in explaining the decline in overall levels of serious violence.
Neither increasing economic prosperity, historical variation in female
participation in the labor market, nor changing cultural models of the
family and gender roles appear to have had a significant impact on
male predominance in serious violent crime.
There is one major exception to this pattern. In early eighteenth-
century Stockholm, women not only accounted for more than 60 per-
cent of property crime offenders but also 45 percent of murder and
manslaughter offenders and 41 percent of assault offenders (Andersson
1995). These are probably the highest female participation rates in se-
rious violent crime found anywhere in the world. Scholars examining
this phenomenon emphasize a combination of factors including—be-
sides demographic imbalance—a highly specific cultural configuration,
which embraced some kind of otherworldly calculus. More particu-
larly, for fear of eternal punishment in hell, suicidal women appear of-
ten to have chosen to kill somebody else, usually their offspring, and
then suffer the death penalty imposed on them by the judiciary ( Jans-
son 1998). Homicide would bring them to purgatory for a limited pe-
riod of time, after which they would enter heaven for eternity, which
was definitely to be preferred to consignment to eternal hell because
of suicide.
B. Age of Offenders
If sex differences have remained more or less constant over 800
years, variability in age patterns should attract scholarly curiosity.
Hirschi and Gottfredson’s seminal 1983 article precipitated a heated
debate among criminologists (see, e.g., Baldwin 1985; Greenberg
1985, 1994; Steffensmeier and Streifel 1991). Hirschi and Gottfredson
argued that the age curve of criminal offending is basically invariant
across time and place, demographic groups, and social and cultural
conditions. Various researchers have produced evidence with the in-
tent of showing the contrary. The debate may be said to have resulted
in a stalemate. Studies convincingly suggest that police-recorded of-
fenders in the past two decades tend to be somewhat younger than, for
example, in the 1950s (see, e.g., Steffensmeier et al. 1989; Junger-Tas
1991). However, the variability of the age-crime curve appears to re-
main within relatively narrow limits, and the overall shape does not
Long-Term Historical Trends in Violent Crime 113
appear fundamentally to differ between different subperiods of mo-
It therefore is useful to explore age patterns in violent crime before
the onset of criminal statistics. However, important limitations of such
an effort should first be noted. Above all, age was not generally re-
corded before the seventeenth century, and most early sources offer
no information whatsoever about offenders’ ages. Second, historians of
crime have not been particularly interested in the age variable (with
the notable exception of King 2000, pp. 169 ff.). Hence, very few stud-
ies on crime in the seventeenth and the eighteenth centuries include
relevant analyses, although the information is probably available in
many primary sources. Third, even when historians have gathered data
on age, the age structure of the underlying population is almost always
unknown, thus making estimation of rates per population impossible.
These limitations notwithstanding, the existing evidence offers some
necessarily crude but nevertheless noteworthy insights. The earliest
evidence that I could find refers to early sixteenth-century Douai, a
city located in the northeast of modern France close to the Belgian
border (Fouret 1987). The archives include information on the ages of
some 100 out of 623 indicted violent offenders and their victims. The
average age of violent offenders in this sample was 26.6 years, while
victims had a mean age of 29.6 years. If these figures represent the
overall age structure of violent offenders in sixteenth-century Douai,
their similarity with modern data is astonishing. In the United States
in 1999, the average age of homicide offenders was 28.6 years, and the
age of victims was 32.3 years. A study by Wikstro
¨m (1985) on violent
crime in Stockholm in the 1980s found a mean age of 31 years for of-
fenders and 34 years for the victims. Thus, not only is the age dif-
ference between offenders and victims almost identical (three years),
but there may also be little difference in the effective mean age, since
we can safely assume that the average population was considerably
younger in the sixteenth century.
This is corroborated by other evidence. I found four studies that in-
clude data on the distribution of violent offenders in the period before
the onset of statistics. The earliest details the age of more than 85 per-
cent of the 1,500 offenders delivered to jail in the Duchy of Mantova
in northern Italy at the end of the sixteenth century (Romani 1980).
These data include all offenders, but violent offenses constitute 40 per-
cent of the total. A second age distribution is based on a small sample
of eighty-three people publicly punished for wounding and attacking
114 Manuel Eisner
in the city of Amsterdam between 1651 and 1749 (Spierenburg 1984,
p. 321). The third series, based on data presented in Ruff (1984, p. 90),
concerns persons convicted for physical violence in two se
in southwestern France near Bordeaux in the period 1696–1789. The
fourth age distribution comes from a study by Champin (1972) of 230
violent offenders indicted in the rural community of Alenc
¸on in Bri-
tanny between 1715 and 1745.
Some of these studies use detailed age brackets. For comparative
reasons, however, I recalculated all distributions for ten-year intervals.
Figure 10 shows the average percentage of offenders per year of age
in the respective age group. Overwhelmingly, the data show a very
similar pattern, with roughly 35–45 percent of the offenders in the
twenty-to-twenty-nine-year age group and a steady decline thereafter.
Certainly, one should bear in mind that these data may be imprecise
in themselves and that no correction for the age distribution of the
population could be made. The extent to which these data support the
notion of an ‘‘invariant’’ age curve of violent offending is open to de-
bate. Future research may come up with more detailed data allowing
for a more elaborate assessment of the age-violence relationship in
Fig. 10.—Age distribution of violent offenders across time and space. Note: Persons
convicted of assault in 1908 in Germany added for comparative reasons. Sources: Man-
tova: Romani 1980; Amsterdam: Spierenburg 1984, p. 321; southern France: Ruff 1984,
p. 90; Alenc
¸on: Champin 1972, p. 55; Germany: von Mayr 1917, p. 766.
Long-Term Historical Trends in Violent Crime 115
early modern Europe. However, at present, one may cautiously con-
clude that evidence from six different areas in Europe and extending
over a period of some 400 years shows a strikingly similar overall pat-
tern. If this finding can be generalized, we may conclude that historical
variation in overall levels in serious violence does not covary with dif-
ferences in the age distribution of violent offenders. That would imply
that changing cultural definitions of youth and young adulthood,
changing marriage patterns, or varying economic prospects for young
men did not result in major changes of the age distribution of serious
violent offenders.
C. Social Status of Offenders
Class is the third primary variable used to describe demographic
characteristics of violent offenders. Research in contemporary society
consistently shows that serious violent offenders are heavily overrepre-
sented among socially disadvantaged groups. Historical studies on the
nineteenth century tell a very similar story, even if the official statistics
of the time are likely to have a stronger class bias. About 50 percent
of a sample of assailants indicted in Bedfordshire between 1750 and
1840 were recorded as laborers or servants (Emsley 1996, p. 45). Simi-
larly, in late nineteenth-century German crime statistics, offenders
from a working-class background were more strongly overrepresented
for aggravated assault and homicide than for any other crime ( Johnson
1995, p. 208).
Many contemporary historians of crime have been strongly inter-
ested in retrieving information about the social background of offend-
ers recorded in the written sources. Examination of these studies yields
a surprisingly consistent pattern. During the Middle Ages, interper-
sonal physical violence was not at all a class-specific phenomenon.
Only to the degree that overall levels of violence fell throughout the
early modern age did violence become correlated with class.
Ruggiero (1980) has done probably the most thorough analysis of
the social status of premodern violent offenders. In a detailed study of
violence in Venice between 1324 and 1406, he was able to identify the
social standing of more than 1,600 offenders dealt with by the secular
judicial authorities. He distinguishes four groups in Venetian society,
for which he also provides estimates of their approximate share of the
total population. The nobility, a group demarcated by its access to po-
litical power, accounted for about 4 percent of the population. Below
it came a group of ‘‘important people,’’ which included merchants,
116 Manuel Eisner
Social Status of Violent Offenders in Early Renaissance Venice,
Type of Crime
Speech Assault Rape Murder Total Population
(%) (%) (%) (%) (%) (%)*
Nobles 35 22 20 4 18 4
Important people 8 11 8 9 9 10
Workers 52 61 65 70 63 75
Marginal people 5 5 7 16 9 8
Number of cases 223 566 416 424 1,629
Source.Data based on Ruggiero 1980.
* Clerics, who may have constituted 3 percent of the population, are not included in
these figures. Therefore, population total is less than 100 percent. Clerics were not re-
ferred to the secular courts.
professionals, and civil officials, and which constituted some 10 percent
of the population. Below them came the large group of workers and
artisans, such as laborers in the textile industry, butchers, bakers, and
marine workers, who may have totaled 75 percent of the population.
At the bottom end came the marginal people, vagabonds and beggars,
who may have been about 8 percent of the population. Ruggiero’s data
show the relative shares of these groups among the cases of recorded
violence, which may be roughly compared with their respective share
in the total population (see table 4). The data suggest that people of
lower standing were not overrepresented among violent offenders and
that nobles had a highly overproportionate share in all but homicide
cases. As Ruggiero points out, these data may be considerably skewed
since much violence among the lower classes may have gone unnoticed
or have been handled with summary justice without leaving traces in
the records (Ruggiero 1980, p. 96). Nonetheless, disregarding the evi-
dence from nonlethal violence and assuming some unnoticed lower
class murders still leaves the impression that the higher ranks of four-
teenth-century Venetian society engaged in their fair share of violent
behavior. In addition, upper-class people seemingly victimized people
of lower standing more often than vice versa, which again contrasts
strikingly with modern patterns. Nobles, it appears, did not scruple to
assault, rape, or kill people of lower standing.
Although probably unparalleled in their detail, these figures do not
Long-Term Historical Trends in Violent Crime 117
seem to be unusual. Several historians of medieval crime have found
similar patterns. In thirteenth-century Bologna, 10 percent of 521 ban-
ishment cases for major crimes were urban magistrates and nobles,
who were aptly labeled by the popolo government as rapacious wolves
(Blanshei 1982, p. 123). In fourteenth-century Lyon, Gonthier (1993)
observes a recurrent involvement of nobles in violent behavior, includ-
ing the organization of gangs to revenge failures to comply with their
interests. Also, tax returns of offenders in fifteenth-century Constance
reveal that wealthy groups were at least as likely to engage in violent
offending as the poor (Schuster 2000, p. 137). Hanawalt (1979, p. 131)
found that members of the oligarchy in a small fourteenth-century En-
glish rural area committed about one-third of all homicides recorded
in the Gaol Delivery Rolls (Hanawalt 1979, p. 131). She concluded
that members of the higher status groups committed at least as much
violence as lower classes since they were likely to become involved in
conflicts over rights and goods, which, in the absence of reliable state
control, often escalated into violent conflicts.
No one has yet attempted to provide comparative data on the social
status of offenders across longer periods. Yet some evidence suggests
that upper classes in northern Europe may have become more pacified
and less prone to physical aggression from the sixteenth century on-
ward. Spierenburg (1998), for example, argues that homicide rates de-
clined in Amsterdam after 1620 because wealthy, churchgoing citizens
renounced violence, while lower-class violence in the form of knife
fighting remained undiminished. Similarly, Sharpe (1984, p. 95) as-
sumes a gradual decline in the involvement of the upper class in crimi-
nal violence in the century after 1550.
In the south of Europe the retreat of the nobility from aggressive
behavior appears to have occurred later. In the French Auvergne, a
mountainous and very poor area, the decisive shift occurred between
the beginning and the end of the eighteenth century, when the upper
classes increasingly withdrew from violent behavior. This transforma-
tion was paralleled by the increasing acceptance, among the nobility,
of merit and competence as core social values, to the detriment of
honor and the military ethic (Cameron 1981, p. 202). Further south,
late eighteenth-century Sardinia offers striking evidence (Doneddu
1991). Ridden by chronic banditry as well as the vendetta, Sardinia had
an overall homicide rate of thirty-five to forty per 100,000 in the years
1767–89. During that period, members of the nobility were recorded
for committing fifty-one homicides. Since the island’s nobility counted
118 Manuel Eisner
some 6,000 members at that time, this puts their homicide rate at
thirty-seven per 100,000, the same as the approximate rate in the total
If this interpretation of upper-class involvement in violent behavior,
based on a few scattered studies, withstands further scrutiny, it may
lead to an important generalization. The transition to lower overall
levels of interpersonal criminal violence, one might hypothesize, was
accompanied by an overproportional withdrawal of the elite from the
use of physical aggression to seize and defend their interests.
D. Sex of Victims
Few studies on premodern homicide tell us anything about the sex
distribution of the victims, although most judicial and nonjudicial
sources presumably include relevant information. Although female
criminality has increasingly become a topic of historical scholarship,
no study has as yet systematically examined female victimization. The
premodern homicide database only includes some thirty estimates of
the proportion of female victims (see table 5).
Most of these estimates are based on work by Given (1977), Hana-
walt (1979), Schu
¨ssler (1991, 1998), and Spierenburg (1996). The pe-
riod up to the sixteenth century is covered by a comprehensive sample
of data, including various counties in England and cities scattered
throughout Europe north of the Alps. The pattern revealed is concor-
dant in suggesting that male victims considerably outnumbered female
victims. The average proportion of female homicide victims during the
period between the thirteenth and the sixteenth centuries is 7 percent
Average Estimates of Gender-Specific Victimization Rates before the
Nineteenth Century
Victims Male/Female Approximate
(percent) Ratio Homicide Rate
Thirteenth–sixteenth centuries 7 12.5:1 30 per 100,000
Seventeenth century 13 6.7:1 8 per 100,000
Eighteenth century 27 2.7 : 1 3 per 100,000
Source.—History of Homicide Database.
Note.—All estimates refer to various regions in England, the Netherlands, Germany,
and France.
Long-Term Historical Trends in Violent Crime 119
and has a range of between 0 and 20 percent. From a criminological
perspective, this figure conveys an air of inevitability. At the beginning
of the twentieth century, Verkko (1951, p. 52) examined the propor-
tions of female victims in countries with high homicide rates (Finland,
Serbia, Bulgaria, Italy, and Chile). The average proportion of female
victims in these countries was 7 percent, the same as the medieval pat-
Only a few relevant observations are available for the seventeenth
and eighteenth centuries. Those suggest that the proportion of female
homicide victims increased as the overall level of lethal violence de-
clined throughout northern Europe. National death statistics for vari-
ous countries corroborate these findings. Trends in overall homicide
rates appear to concur with shifts in the male-to-female victim ratio,
thus confirming the ‘‘dynamic law’’ put forth by Verkko (1951, p. 52),
which holds that fluctuations in overall homicide rates primarily result
from variation in male victimization rates. Generally, the shift toward
lower homicide rates appears to have been primarily—but not exclu-
sively—a drop in male-to-male violent encounters.
E. Personal Relationship between Offender and Victim
This finding can be further substantiated by examining another
variable, the relationship between offender and victim. A number of
studies have examined the proportions of homicides involving family
members (spouses, offspring, and parents). Table 6 shows a series of
estimates from the thirteenth through the twentieth centuries that can
be fleshed out with qualitative evidence on circumstances likely to re-
sult in a (recorded) killing.
The data suggest that the proportion of family homicides was very
low throughout the Middle Ages. Typically, the killing of family mem-
bers made up less than 10 percent in medieval societies. In contrast,
a large proportion of cases occurred in situations of conflict between
(primarily male) acquaintances, with the offender and the victim often
sharing a similar social background or being neighbors in a rural com-
munity. Not only in fourteenth-century Oxford, did ‘‘quick tempers,
strong drink, and the ready availability of weapons’’ contribute to the
great frequency of homicides among men (Hammer 1978, p. 20). In
many urban areas, the tavern was the place where violence occurred.
In sixteenth-century Arras, 45 percent of ninety recorded homicides
were committed in or just outside taverns (Muchembled 1992, p. 94).
Likewise, about half of all violent crimes in sixteenth-century Douai,
Proportion of Homicides against Members of the Family in Various Historical Periods
Family Homicide
Period (percent) Homicide Rate Source
Thirteenth–sixteenth centuries:
England, various counties Thirteenth century 5 22.0 Given (1977, p. 144)
England, various counties 1300–1348 2–8 35.0 Hanawalt (1979, p. 159)
Germany, Nurnberg 1285–1400 9 14.0 Schu
¨ssler (1991, p. 174)
England, Huntingdonshire 1286 5 20.0 DeWindt and DeWindt
(1981, p. 54)
Seventeenth century:
Essex 1620–80 15 6.0 Sharpe (1983, p. 126)
Amsterdam 1651 –1700 11 3.9 Spierenburg (1994, p. 710)
Kent Seventeenth century 26 4.5 Cockburn (1991)
Castile 1623–99 12 35.0 Chaulet (1997, p. 20)
Eighteenth century:
Amsterdam 1701–50 14 10.0 Spierenburg (1994, p. 710)
Amsterdam 1751 –1810 48 6.0 Spierenburg (1994, p. 710)
Kent Eighteenth century 28 2.0 Cockburn (1991)
Surrey 1678–1774 36 4.0 Beattie (1986, p. 105)
Nineteenth century, England 1850s–1860s 55 1.0 Emsley (1996, p. 43)
Twentieth century:
England and Wales 1998 39 1.4 Home Office (2000)
Germany 1996 30 1.4 Bundeskriminalamt (1997)
Long-Term Historical Trends in Violent Crime 121
and probably an even greater proportion in Cologne, occurred in the
context of alcohol drinking (Schwerhoff 1991). In a similar vein Sharpe
argues, summarizing the English evidence, that ‘‘Stuart homicides
were characteristically unplanned acts of violence arising spontane-
ously from quarrels, being simple assaults that went too far in most
cases’’ (Sharpe 1983, p. 131).
However, many historians point out that what seems to have been
impulsive and spontaneous violence often was more culturally guided
than might first be suspected. Male honor seems to have played an im-
portant role here. Thus Liliequist (1999, p. 197) finds that boxing ears,
issuing challenges, fighting and combat interrupted by temporary rec-
onciliation, and drinking rituals constituted the pattern of a culture of
fighting, which was the backdrop of the vast majority of homicide cases
in early modern Scandinavia. Hence, insults constituted a serious af-
front and a large class of crimes dealt with by any court of medieval
and early modern society; throughout Europe knife fighting appears to
have been the appropriate reaction if efforts for reconciliation failed.
During the transition to lower overall homicide rates, however, the
relative share of family killings appears to have increased continuously,
which in turn suggests that overt public fights between men resulting
in serious injury became progressively less frequent. Knife fighting, for
example, became restricted to the lower classes in late seventeenth-
century Amsterdam and all but disappeared as a distinct culture of vio-
lence by the late eighteenth century (Spierenburg 1998). In a similar
vein, the decline of homicide in late nineteenth-century Italy, to a large
extent, probably resulted from the disappearance of public fights be-
tween men over issues of honor. The decline of private revenge and
the vendetta—an almost exclusively male prerogative, too—also ap-
pear to be associated with the overall drop in homicide rates. In coun-
tries such as England or Sweden, as a result, family homicide ac-
counted for more than half of the killings by the end of the nineteenth
century, when the overall level of homicide rates was at most one-tenth
of that before the sixteenth century.
The patterns with respect to homicide victims’ sex and the relation
between offenders and victims suggest another far-reaching generaliza-
tion. Declines in homicide rates primarily resulted from some degree
of pacification of encounters in public space, a reluctance to engage in
physical confrontation over conflicts, and the waning of honor as a cul-
tural code regulating everyday behavior.
122 Manuel Eisner
III. Theoretical Approaches
Gurr’s (1981) original study, though innovative in its pathbreaking
synthesis of empirical developments, has little to offer by way of
theoretical interpretation. Stating that the long-term decline can be
explained by ‘‘the rise of nonaggressive modes of behavior’’ (Gurr,
p. 304) and a sensitization to violence amounts to little more than re-
phrasing and describing the empirical pattern. Since then many histo-
rians of crime, nourished by the flow of empirical findings, have devel-
oped and debated theories that might explain the long-term trends.
The following discussion is based on two prior decisions: first, any
theoretical discussion of the trend in violent crime must assume that it
describes real changes in behavior rather than methodological artifacts
or consequences of the operations of criminal justice systems. There is
still debate about this, but many historians of crime accept the basic
inferences drawn here from existing data. Second, there is debate over
whether the observed patterns require local and specific interpretations
of manifestations of violence or whether we may profitably attempt to
develop general theories. A large part of the craft of historical research
consists in meticulous analysis of specific historical sources resulting in
rich and thick descriptions of some historical reality, sometimes at the
detriment of theoretical generalization. One might argue, for example,
that the declining trend in seventeenth-century Sweden requires a
wholly different explanation from the drop observed in nineteenth-
century Italy. The following discussion, however, starts from the
premise that the existing evidence asks for a generalizing theory, which
takes into account commonalities across large geographic spaces.
These general patterns, partly resting on admittedly shaky empirical
evidence, may be summarized in five points. First, in the long run,
there appears to have been little change in the sex and age structure of
serious violent offenders. Second, serious interpersonal criminal vio-
lence has declined considerably over the past six centuries throughout
Europe. The decline probably started as early as the fifteenth century,
but it is well documented for the long period between the early seven-
teenth and the mid-twentieth centuries. Third, areas in Europe appear
to differ in respect of the timing and pace of the drop in serious vio-
lence. The process may have started earliest in the Netherlands and
England; in Sweden, the main transition may have occurred between
the early seventeenth and mid-eighteenth centuries; and in Italy, homi-
cide rates dropped dramatically only from the mid-nineteenth century
onward. Fourth, historically, high overall levels of violence appear to
Long-Term Historical Trends in Violent Crime 123
be associated with high levels of elite involvement in physical violence.
Drops in lethal violence were disproportionately related to a decline in
elite violence. Fifth, in any high-homicide society, the majority of cases
are male-to-male encounters, often between people of similar social
status, arising out of situational conflicts involving clashes over honor,
property, or other entitlements. Sustained declines in homicide rates,
in turn, are accompanied by some degree of pacification of interactions
in public space.
A. The Theory of the Civilizing Process
Theories may be suspected of being good theories if they predict
something that is corroborated by ensuing empirical work. For this
reason, the work of Elias (1976, 1983) provides the most prominent
theoretical framework discussed by historians of crime who are inter-
ested in explaining the decline in homicide rates. Elias’s theory of the
civilizing process, developed in the 1930s and primarily introduced
into the history of crime and punishment research by Spierenburg
(1984, 1995), embraces long-term social dynamics at a macrolevel as
well as changes in typical psychological traits and developments in
characteristic modes of behavior at a psychological microlevel.
At the microlevel, the theory of the civilizing process holds that,
over a period of several centuries, personality structures have become
transformed in a distinct cumulative direction. The change is charac-
terized by an increasing affect control, a greater emphasis on long-
term planning, a rationalized manner of living, a higher reflexive sen-
sitivity to inner psychological states and processes, and a decreasing
impulsivity—in brief, higher levels of self-control. Higher levels of
self-control imply, in turn, the gradual pacification of everyday inter-
actions, which becomes manifest in lower levels of violent behavior.
The idea that, on the personality level, criminal violence is the result
of low self-control should be an attractive starting point to (many)
criminologists. Many empirical studies now convincingly show that, in
contemporary society, violent and serious offending strongly correlates
with a tendency to seek immediate gratification, a tendency toward
risk-seeking behavior, a high level of impulsivity, and an indifference
to the needs of others (Gottfredson and Hirschi 1990, p. 90; Farring-
ton 1998). For the historical past, a direct measurement of personality
structures is obviously impossible. But assuming causal mechanisms at
a microlevel that do not contradict current criminological knowledge
124 Manuel Eisner
certainly constitutes an advantage for theorizing about long-term
macrolevel dynamics.
The overlap between the theory of the civilizing process and current
criminological thinking ends, however, with the question of why and
how levels of self-control may differ. Criminology has as yet offered
pitifully little on this subject. Elias, by contrast, proposes a coherent
sociological theory. This is partly because of a difference in focus. Elias
was not interested in individual-level variation in self-control, found in
every society, but, rather, in explaining historical variation in popula-
tion averages. On the most general level, he argues that these changes
result from the internalization of outer social control, which, in turn,
results from the increasing interdependency between social actors.
Higher interdependency in complex and extended chains of interac-
tion—buttressed by stable social institutions—promotes self-control,
since it creates advantages for those able to dampen affect and ratio-
nally plan their behavior (Elias 1978, p. 322).
Two interrelated macrolevel dynamics promote this long-term
change since the Middle Ages: the expansion of the state with its mo-
nopoly on violence and the extension of the market economy resulting
in increasing functional interdependency. In respect of the first factor,
Elias argues that the elites of the knightly warrior societies of the Mid-
dle Ages gradually became transformed in the sixteenth and seven-
teenth centuries into relatively pacified court societies, where violence
came to be monopolized by central authorities. The decisive factor was
the rise of monarchic absolutism, in which the state monopoly of
power over a large territorial unit was accomplished to a high degree
(Elias 1976, p. 353). The nobility lost its bellicose functions, which in
turn facilitated the rise of complex economic and social chains of inter-
dependency. As a result, courtly manners became increasingly differen-
tiated, refined, and civilized. This culture of the nobility then gradually
diffused from its very center to other social groups and strata. In re-
gard to the effects of functional interdependency, Elias basically relied
on classical Enlightenment ideas. The view that increasing commer-
cialized exchanges of goods and services creates incentives for restraint
from violence was commonplace among liberal thinkers. Adam Smith,
for example, assumed that ‘‘commerce and manufactures gradually in-
troduced order and good government, and with them, the liberty and
security of individuals, among the inhabitants of the country, who had
before lived in a continual state of war with their neighbors, and of
Long-Term Historical Trends in Violent Crime 125
servile dependency upon their superiors’’ (quoted in Beattie 1986,
p. 137).
Some historians of crime, such as Spierenburg (1996), accept Elias’s
wide-ranging theoretical model of the rise of European modernity.
Others refute the model as insufficient (Schuster 2000). Many, how-
ever, view the theory of the civilizing process as a fruitful point of de-
parture (O
¨sterberg 1996; Sharpe 1996). Thus, if nothing else, most
historians of crime would probably agree that the long-term trajectory
in homicide rates is an indicator of a wider dynamic that encompasses
some sort of pacification of interaction in public space. Beattie, for ex-
ample, when commenting on the decline of homicide in England be-
tween 1660 and 1800, notes that
men and women would seem to have become more controlled, less
likely to strike out when annoyed or challenged, less likely to settle
an argument or assert their will by recourse to a knife or their
fists, a pistol, or a sword. The court record suggests that other
ways of resolving conflicts became increasingly favored and that
men became more prepared to negotiate and to talk out their
differences. This supposes a developing civility, expressed perhaps
in a more highly developed politeness of manner and a concern
not to offend or to take offense, and an enlarged sensitivity toward
some forms of cruelty and pain. (1986, p. 112)
But the problem that divides scholars is the identification of the causal
factors that have brought about sensitization to violence.
B. Social Control
Exploring the notion that there may be a link, however indirect and
complex, between the rise of bureaucratic state structures and the de-
cline of violence, several historians of crime have become interested in
changing patterns of judicial and social control. Two strands of inquiry
can be distinguished. First, scholars have paid progressively more at-
tention to immediate patterns of official attitudes to homicide and vio-
lence, including prosecution and punishment. Second, changes in the
wider context of social control over everyday behavior may constitute
an important element for understanding the secular change in violent
interpersonal behavior.
In regard to official attitudes toward homicide, a decisive shift occurs
126 Manuel Eisner
during the sixteenth century (Rousseaux 1999a). During the late Mid-
dle Ages, although official authorities had become increasingly in-
volved in the regulation of lethal interpersonal violence, homicide was
regarded with lenience if it was perceived as the result of passion or
occurred in defense of honor. Only the most premeditated cases of
murder invariably required the death penalty (see, e.g., Blanshei 1982,
p. 125). However, when the peace between two families was broken
because of a mortal aggression, retaliation by means of private ven-
geance was still regarded, in popular perception, as a legitimate path-
way to reestablish order. Increasingly, however, the parties would be
likely to resort to the courts, where peace treaties comprising a wergild
payable to the victim’s family could be accomplished. In England, be-
cause of early unification under the Normans, jurisdiction of homicide
was the exclusive prerogative of the crown within the jurisdiction of
the royal courts. This led to the normative distinction of three catego-
ries of homicide, namely, culpable homicide punishable by the death
penalty; excusable homicide, which could be pardoned by the crown
with a letter of pardon; and justifiable homicide, which was liable to
be acquitted by a jury.
Between the sixteenth and the seventeenth centuries in continental
Europe, settlement fell out of the hands of families and into the hands
of judges and sovereigns whose aim was to deliver punishment rather
than to reconcile factious families. Only then, Rousseaux (1999a,
p. 154) argues, did homicide invariably become seen as a crime and the
offender as a criminal; its perceived character shifted from an unfortu-
nate accident to a rigorously repressed heinous crime. Manifestations
of this change can been found throughout Europe. In Zurich, for ex-
ample, the concept of honorable manslaughter, punishable by a penalty
only, became the object of intensive judicial conflict and political nego-
tiation between 1480 and 1530, when the primacy of ‘‘urban peace’’
finally won out over the notion of legitimate defense of honor (Pohl
1999). For the small city of Nivelles, the origins of a new model of
social control can be traced to the period between 1520 and 1530,
when ‘‘the aim is no more to re-establish peace between the citizens
but to subordinate the subjects to the social order determined by the
prince’’ (Rousseaux 1999b, p. 266). In Germany, too, evidence suggests
that private reconciliation had become an unusual way of settling ho-
micide by the end of the sixteenth century (Schwerhoff 1991, p. 280).
Rousseaux argues that more rigorous repression may have played a role
in the decline of homicide in early modern Europe: ‘‘Mortal aggression
Long-Term Historical Trends in Violent Crime 127
became the object of a campaign of ‘moralization’ and ‘civilization’
around the sixteenth and seventeenth centuries, between the religious
wars and the Thirty Years War. This undertaking was visible mainly
in the development of criminal law and in the growing sophistication
of legal definitions as well as in the emergence of homicide as a matter
for the gallows. This undertaking was relatively successful if we take
into account the drop in the number of homicides and the virtual dis-
appearance of private dispute settlements’’ (Rousseaux 1999a, p. 157).
However, the replacement of the primacy of private reconciliation
by the dominance of state repression was embedded in a much wider
pattern of increasing social control. From the mid-sixteenth century
onward, social historians find a wave of intensified magisterial social
control spreading throughout Europe that restructured the relation-
ship between the state and its citizenry. It included the creation of
more centralized administrative and judicial organizations, the greater
continuity of bureaucratic intrusions into everyday life, and the con-
struction and expansion of professional armies (Tilly 1992). Particu-
larly, this period saw a flood of ordinances regarding feasts, child
rearing, appropriate clothing, consumption of alcohol, and church
attendance (Oestreich 1968, 1982). Together, these activities resulted
in an acceleration of social disciplining, a process that can be seen as
the result of complex interactions among different social, political, and
economic forces (Du
¨lmen 1993, 1996). The consolidation of state
power is only one of them. Yet factors such as the increased religious
zeal following the Reformation movements, the expansion of literacy
and schooling, and early capitalist organization of work constitute in-
dependent sources of the disciplining process in the early modern age.
Their similar effects on the structures of the self were both to enforce
self-control rigidly and to provide the cultural and social resources
needed for a more orderly conduct of life.
C. Limitations of the ‘‘State Control’’ Model
Strangely one-sided in respect to the role of the state as an internally
pacifying institution, Elias almost exclusively emphasizes the state’s co-
ercive potential exercised through the subordination of other power
holders and bureaucratic control. Echoing the old Hobbesian theme,
the decline in interpersonal violence should thus develop out of in-
creased state control. Although the long-term expansion of the state
and the decline of lethal violence appear to correlate nicely on the sur-
face, a closer look reveals several inconsistencies. Muchembled (1996),
128 Manuel Eisner
for example, points out that the decline of homicide rates in early mod-
ern Europe does not appear to correspond with the rise of the absolut-
ist state. Rather, he argues, the example of the Low Countries shows
that homicide rates declined in polities where centralized power struc-
tures never emerged and the political system much more resembled a
loose association of largely independent units. Neither does intensified
policing nor the harsh regime of public corporal punishment, both
probably the most immediate manifestations of state power in any
premodern society, seem to aid understanding of the trajectories
into lower levels of homicide rates. Police forces in medieval and
early modern Italian cities were surprisingly large—Schwerhoff (1991,
p. 61) cites per capita figures of between 1 : 145 and 1:800—but they
did not effectively suppress everyday violence. Furthermore, no histo-
rian seems to believe that the popularity of the scaffold and the garrote
among sixteenth- and seventeenth-century European rulers decisively
reduced crime.
Rather, the Italian case exemplifies a more general problem. For
whatever the deficiencies of early modern Italian states may have been,
they were certainly not characterized by a lesser overall level of state
bureaucracy and judicial control than, for example, states in England
or Sweden during the same period (see, e.g., Brackett 1992). England
was not centralized in bureaucratic terms, and the physical means of
coercion, in terms of armed forces, were slight (Sharpe 1996, p. 67).
The mere rise of more bureaucratic and centralized state structures
thus hardly seems to account for the increasingly divergent develop-
ment of homicide rates in northern and southern Europe. Examining
Rome, Blastenbrei (1995, p. 284) argues that the divergence may,
rather, be related to the evolution of different models of the relation-
ship between the state and civil society. While northern European so-
cieties were increasingly characterized by a gradually increasing legiti-
macy for the state as an overarching institution, the South was marked
by a deep rupture between the population and the state authorities. In
respect to state control, Roth emphasizes a similar point when examin-
ing the massive drop in homicide rates in New England from 1630 to
1800: ‘‘The sudden decline in homicide did not correlate with im-
proved economic circumstances, stronger courts, or better policing. It
did, however, correlate with the rise of intense feelings of Protestant
and racial solidarity among the colonists, as two wars and a revolution
united the formerly divided colonists against New England’s native in-
Long-Term Historical Trends in Violent Crime 129
habitants, against the French, and against their own Catholic Monarch,
James II’’ (2001, p. 55).
Both Roth and Blastenbrei emphasize, from different angles, a so-
ciological dimension whose importance for understanding the long-
term decline in serious violence has not yet been systematically ex-
plored, namely, mutual trust and the legitimacy of the state as founda-
tions for the rise of civil society. Both are, of course, clearly to be dis-
tinguished from the coercive potential of the state—strong states in
terms of coercion can be illegitimate, while seemingly weak states may
enjoy high legitimacy. And on the level of macro-transhistorical com-
parison, the decline of homicide rates appears to correspond more with
integration based on trust than with control based on coercion.
Intertwined with the rise of legitimate state structures and political
integration, honor probably is an important concept to consider sys-
tematically. Much research emphasizes the crucial role of insults in
triggering situational conflicts in medieval or early modern societies.
Indeed, insult constituted a major class of criminal offenses, frequently
brought to court and often resulting in severe fines to be paid to the
victim. This is in accordance with a society in which ‘‘honor’’ consti-
tuted a highly important symbolic, and therefore also economic, re-
source to be legally protected and publicly regulated (Muchembled
1984; Burghartz 1990; Schwerhoff 1991; Schuster 1995). It required
retributive violence as a potential and culturally accepted means for
maintaining one’s honor (Schmidt 1994). In late fourteenth-century
Zurich, for example, the butcher Welti Oechen stabbed another
butcher in a quarrel (Pohl 1999). The judges decided that the case had
been an ‘‘honorable manslaughter,’’ because the victim had insulted
Oechen by alleging that the Oechen family were villains. The offender
had to pay a fine to the victim’s family and is known to have continued
to live a respectable life thereafter. The example is similar to many
others found in late medieval records.
Homicide here seems to originate in the necessity to react personally
to any challenge to one’s reputation or honor, which is persistently
found in any high-violence society; the judicial reaction is based on ac-
cepting the legitimacy—not necessarily the legality—of the course of
action taken by the offender. The long-term decline in violence, in
turn, appears to have been consistently paralleled by the loss of the
cultural significance of honor. From about the mid-seventeenth cen-
tury in northern Europe, verbal violence—blasphemy, slander, and in-
130 Manuel Eisner
sult—began to cause much less alarm, and rashly spoken statements,
formerly regarded as unwitting but inevitable revelations of nefarious
purposes, lost their awful significance (Soman 1980). The gradual
withdrawal of honor from constituting a symbolic resource to be de-
fended, if necessary, by physical force may be related to the expansion
of reliable state structures. But cultural change may also have played
an important role.
D. Culture
Culture, it is true, is an elusive concept, and explaining the decline
in violence by an increasing sensitization to violence is not likely to be
very helpful. However, systems of values and ideas, when embedded in
social institutions, do have the potential of changing everyday routines
and interaction patterns. But if cultural explanations of the long-term
decline in serious violence are to be kept from being tautological, they
must start with possible causes outside the more narrow subject matter
of attitudes toward violence. This might include, for example, cultur-
ally transmitted and widely shared views of the role of the individual
in society or assumptions about adequate patterns of child raising in
the family. These, in turn, would then have to be shown to impinge
directly on variables that can plausibly be assumed to correlate with
At least two broad cultural streams in Western society may have
been associated with the decline in interpersonal violence, namely,
Protestantism and modern individualism. Max Weber (1922) inter-
preted the Protestant ethic primarily as a gigantic disciplining project
that emphasized fulfillment of one’s duty, sobriety and frugality, and a
methodic conduct of life. Also, inner-directedness and a conscientious
life were among the principal commands of early Protestantism, mak-
ing relentless introspection and the cultivation of shame and guilt per-
vasive cultural goals, especially among the Puritan and Pietist strands
of the Reformation. Furthermore, both Reformation and Counter Ref-
ormation brought about an encompassing wave of church religiosity,
legitimating the intrusion of clerics into the private sphere but also
serving as a backbone of increasing literacy and education.
The rise of modern individualism from the sixteenth century onward
is interrelated with Protestantism but clearly distinguished from it
¨lmen 1997). It embraces the cultural diffusion of a specifically
modern ideal of the self, which is characterized by ‘‘disengagement’’
and ‘‘inwardness’’ as its preeminent qualities (Taylor 1989). It implies
Long-Term Historical Trends in Violent Crime 131
a methodological reflexive distance from the immediate outer and in-
ner world and an orientation toward guiding ideals such as autonomy,
self-responsibility, and authenticity. This development, while follow-
ing its own cultural and philosophical logic, is at the same time linked
to mutually reinforcing religious, political, economic, and artistic prac-
tices (Taylor 1989, p. 206). Examples include the permanent self-scru-
tiny of the religious reformation movements; the sharper delineation
of an independent, private sphere; the rise of a market based on con-
tractual guarantees; and the production of art aimed toward individual
Emile Durkheim—forty years earlier than Elias—explicitly assumed
that the rise of modern individualism may constitute a crucial variable
for explaining the long-term decline in lethal violence. He argued that
individual violence should always be interpreted as the ‘‘product of a
specific moral culture,’’ which regulates the relationship between the
individual and society (Durkheim 1991). Hence, he interpreted the de-
cline of homicide rates primarily as resulting from the liberation of the
individual from collective bonds. High levels of lethal violence mirror
the intensity of ‘‘collective emotions,’’ which bind the individual to
‘‘groups of things that symbolically represent these groups’’ (my trans-
lation, see Durkheim 1991, p. 161). He explicitly refers to the tradition
of the vendetta as an example. Violence thus declines to the degree
that the person becomes liberated from his or her sacred obligation to
the group, and individualism brings about both subjective reflexivity
and emotional indifference in conflict situations (Thome 1995, 2001).
Many specialized historians, when interpreting the contextual cir-
cumstances of declining levels of violence, find that culture change
may have been at least as important as state control or the extension
of economic networks. Commenting on the downturn of interpersonal
violence in Swedish cities after about 1630, Jarrick and So
(1993) emphasize that there is no concomitant increase in state inter-
vention that could explain the shift. Rather, the decline appears to have
coincided with an increased concern, disseminated by the Lutheran
church, about the expiation of sin and an intensified attention to issues
of human dignity and empathy for the weak. Likewise, the decline in
serious violence in seventeenth- and eighteenth-century England ap-
pears to have been embedded in a distinct cultural climate where prin-
ciples of Protestantism combined with notions of individual responsi-
bility (Gaskill 2000, pp. 203 ff.). A pervasive culture of Protestantism,
disseminated through cheap print, embedded violence in a dense rhet-
132 Manuel Eisner
oric of providence, sin, and repentance. Pamphlets and ballads told
an interested audience about how murderers—‘‘troubled in con-
science’’—felt remorse for their acts, how dying victims piously for-
gave their assailants, and how the justice of heaven and earth would
combine to punish the evildoer.
To criminologists, the rise of moral individualism should not be an
implausible candidate for explaining the fall in criminal violence.
Rather, a large number of recent survey studies find that violence is
correlated with low autonomy, unstable self-esteem, a high depen-
dence on recognition by others, and limited competence in coping
with conflict, which together may well be interpreted as subdimensions
of low moral individualism (Agnew 1994; Baron and Richardson 1994;
Heitmeyer 1995). To this we might add the hypothesis that the secular
decline of lethal violence occurred when institutional structures and
educational practices supported the stabilization of that type of indi-
vidualized identity that is shaped to meet the challenges of modern life.
E. Conclusions
Considering the vast field—temporally, geographically, and theoret-
ically—covered in this essay, it may be wise not to attempt an even
more condensing conclusion. Rather, I am tempted to speculate about
elements of further research that may help to clarify some of the issues
pertaining to the long-term development of serious interpersonal vio-
lence in Western society. These suggestions are premised on the idea
that more sophisticated theories and comparison of theories only make
scientific sense to the degree that we dispose of detailed empirical data,
which permit the appraisal of alternative explanations. It therefore
seems obvious to ask for more and better data. There are several di-
mensions to this aim. First, we can improve our understanding of the
accuracy and comparability of historical estimates of homicide rates.
Monkkonen (2001) and Roth (2001) have proposed promising strate-
gies, and it remains to be seen how far capture-recapture methods, bet-
ter population estimates, or more broadly based information on the
effects of improved medical technology can improve estimates of
homicide rates. Second, as our knowledge of overall levels of lethal
violence increases, it may become more important to examine devel-
opments in subtypes such as family homicide, infanticide, or robbery-
related killings. Third, existing research has not yet fully explored
historical variation in contextual variables. Qualitative dimensions are
obviously important here. However, examining to a fuller extent quan-
Long-Term Historical Trends in Violent Crime 133
tifiable information about offenders, victims, and situations—possibly
using some degree of standardized tools across studies—may also sig-
nificantly contribute to our knowledge. Finally, it would be useful to
fill in some of the blank spots on the geo-historical map of homicide
in Europe. France and Spain are conspicuously missing, and more in-
formation about trends in Italy and different areas in the German-
speaking parts of Europe would enrich comparative analyses.
Further empirical research may particularly profit from a more co-
herent set of theoretically based questions. Thus far, attempts at expla-
nation were primarily post hoc interpretations in the light of cultural,
social, and political covariates of the secular trend in homicide rates.
But it might be fruitful to adopt systematically comparative perspec-
tives in future research. Findings from social history research may pro-
vide, for example, indicators of historical and geographic variation in
patterns of formal social control, levels of literacy, political conflict,
and the commercialization of the economy. By comparing regions that
systematically differ in these respects, we might be able to learn more
about what variables contribute to changing levels of homicide.
Agnew, Robert. 1994. ‘‘The Techniques of Neutralization and Violence.’’
Criminology 32:555–80.
Andersson, Hans. 1995. ‘‘Genus och Ra
¨ttskultur: Kvinnlig Brottslighet i
stormaktstidens Stockholm.’’ Historisk Tidskrift (Sweden) 2:129–59.
Aubusson de Cavarlay, Bruno. 2001. ‘‘Les limites intrinse
`ques du calcul de
taux d’homicide: A
`propos des nouveaux standards propose
´s par Eric Monk-
konen.’’ Crime, Histoire et Socie
´—Crime, History and Society 5(2):27–32.
Bairoch, Paul, Jean Batou, and Pierre Che
`vre. 1988. La population des villes eu-
´ennes: Banque de donne
´es et analyse sommaire des re
´sultats, 800–1850. Ge-
neva: Droz.
Baldwin, John. 1985. ‘‘Thrill and Adventure Seeking and the Age Distribution
of Crime: Comment on Hirschi and Gottfredson.’’ American Journal of Soci-
ology 90(6):1326–29.
Baron, Robert A., and Deborah R. Richardson. 1994. Human Aggression. 2d
ed. New York: Plenum.
Beattie, John M. 1974. ‘‘The Pattern of Crime in England, 1660 1800.’’ Past
and Present 62:47–95.
———. 1975. ‘‘The Criminality of Women in Eighteenth-Century England.’’
Journal of Social History 8:80–116.
134 Manuel Eisner
———. 1986. Crime and the Courts in England, 1660– 1800. Oxford: Claren-
Becker, Marvin. 1976. ‘‘Changing Patterns of Violence and Justice in Four-
teenth- and Fifteenth-Century Florence.’’ Comparative Studies in Society and
History 18:281–96.
Behringer, Wolfgang. 1990. ‘‘Mo
¨rder, Diebe, Ehebrecher: Verbrechen und
Strafen in Kurbayern vom 16. bis 18. Jahrhundert.’’ In Verbrechen, Strafen
und soziale Kontrolle, edited by Richard van Du
¨lmen. Frankfurt am Main:
———. 1995. ‘‘Weibliche Kriminalita
¨t in Kurbayern in der Fru
¨hen Neuzeit.’’
In Von Huren und Rabenmu
¨ttern: Weibliche Kriminalita
¨t in der fru
¨hen Neuzeit,
edited by Otto Ulbricht. Cologne: Bo
Berents, Dirk Arend. 1976. Misdaad in de Middeleeuwen: Een onderzoek naar de
criminaliteit in het Laat-middeleeuwse Utrecht. Utrecht: Stichtse historische
Blanshei, Sarah R. 1981. ‘‘Criminal Law and Politics in Medieval Bologna.’’
Criminal Justice History 2:1–30.
———. 1982. ‘‘Crime and Law Enforcement in Medieval Bologna.’’ Journal
of Social History 16:121–38.
Blastenbrei, Peter. 1995. Kriminalita
¨t in Rom, 1560–1585. Tu
¨bingen: Nie-
Blauert, Andreas, and Gerd Schwerhoff, eds. 2000. Kriminalita
¨tsgeschichte: Bei-
¨ge zur Sozial- und Kulturgeschichte der Vormoderne. Constance: Universi-
¨tsverlag Konstanz.
Blumstein, Alfred. 2000. ‘‘Disaggregating the Violence Trends.’’ In The Crime
Drop in America, edited by Alfred Blumstein and Joel Wallman. Cambridge:
Cambridge University Press.
Boomgaard, Johannes Everardus Antonius. 1992. Misdaad en straf in Amster-
dam: Een onderzoek naar de strafrechtspleging van de Amsterdams schepenbank,
1490–1552. Zwolle: Waanders.
Boschi, Daniele. 1998. ‘‘Homicide and Knife Fighting in Rome, 1845–1914.’’
In Men and Violence: Gender, Honor, and Rituals in Modern Europe and
America, edited by Pieter Spierenburg. Columbus: Ohio State University
Brackett, John K. 1992. Criminal Justice and Crime in Late Renaissance Florence,
1537–1609. Cambridge: Cambridge University Press.
Buff, Otto. 1877. ‘‘Verbrechen und Verbrecher zu Augsburg in der zweiten
¨lfte des 14. Jahrhunderts.’’ Zeitschrift des Historischen Vereins fu
¨r Schwaben
und Neuburg 4:160–231.
Bundeskriminalamt. 1997. Polizeiliche Kriminalstatistik Bundesrepublik Deutsch-
land. Wiesbaden: Bundeskriminalamt.
———. 1998. Polizeiliche Kriminalstatistik Bundesrepublik Deutschland. Wies-
baden: Bundeskriminalamt.
Burghartz, Susanna. 1990. Leib, Ehre und Gut: Delinquenz in Zu
¨rich Ende des
14. Jahrhunderts. Zurich: Chronos.
Cameron, Iain A. 1981. Crime and Repression in the Auvergne and the Guyenne,
1720–1790. Cambridge: Cambridge University Press.
Long-Term Historical Trends in Violent Crime 135
Champin, Marie-Madeleine. 1972. ‘‘Un cas typique de justice baillage
`re: La
´dans le baillage d’Alenc
¸on de 1715 a
`1745.’’ Annales de Norman-
die 22(1):47–84.
Chaulet, Rudy. 1997. ‘‘La violence en Castille au XVIIe sie
`cle a
`travers les In-
dultos de Viernes Santo (1623–1699).’’ Crime, Histoire et Socie
History and Society 1(2):5–27.
Chesnais, Jean-Claude. 1981. Histoire de la violence en Occident de 1800 a
jours. Paris: Laffont.
Chiffoleau, Jacques. 1984. Les justices du pape: De
´linquance et criminalite
´dans la
´gion d’Avignon au XIVe sie
`cle. Paris: Publications de la Sorbonne.
Cockburn, J. S. 1977. ‘‘The Nature and Incidence of Crime in England, 1559–
1625: A Preliminary Survey.’’ In Crime in England, 1550 1800, edited by
J. S. Cockburn. London: Methuen.
———. 1991. ‘‘Patterns of Violence in English Society: Homicide in Kent,
1560–1985.’’ Past and Present 130:70–106.
Cohn, Samuel. 1980. ‘‘Criminality and the State in Renaissance Florence,
1344–1466.’’ Journal of Social History 14(2):211– 33.
´nod, John. 1891. La criminalite
´ve au XIX Sie
`cle. Geneva: Aubert-
Daly, Martin, and Margo Wilson. 1988. Homicide. New York: de Gruyter.
de Vries, Jan. 1984. European Urbanization, 1500–1800. London: Methuen.
DeWindt, Anne R., and Edwin B. DeWindt, eds. and trans. 1981. Royal Justice
and the Medieval English Countryside: The Huntingdonshire Eyre of 1286, the
Ramsey Abbey Banlieu Court of 1287, and the Assizes of 1287–88. Toronto:
Pontifical Institute of Mediaeval Studies.
Diederiks, Herman A. 1990. ‘‘Quality and Quantity in Historical Research in
Criminality and Criminal Justice: The Case of Leiden in the 17th and 18th
Centuries.’’ Historical Social Research—Historische Sozialforschung 15(4):57–
Doneddu, Giuseppe. 1991. ‘‘Criminalita
`e societa
`nella sardegna del secondo
settecento.’’ In Criminalita
`e societa
`in eta
`moderna, edited by Luigi Ber-
linguer and Floriana Colao. Milan: Guiffre
¨lmen, Richard van. 1985. Theater des Schreckens: Gerichtspraxis und Strafri-
tuale in der fru
¨hen Neuzeit. Munich: Beck’sche Verlagsbuchhandlung.
———. 1993. Gesellschaft der fru
¨hen Neuzeit: Kulturelles Handeln und sozialer
Prozess: Beitra
¨ge zur historischen Kulturforschung. Vienna: Bo
———. 1996. ‘‘Norbert Elias und der Prozess der Zivilisation: Die Zivilisa-
tionstheorie im Lichte der historischen Forschung.’’ In Norbert Elias und die
Menschenwissenschaften: Studien zur Entstehung und Wirkungsgeschichte seines
Werkes, edited by Karl-Siegbert Rehberg. Frankfurt am Main: Suhrkamp.
———. 1997. Die Entdeckung des Individuums, 1500– 1800. Frankfurt am Main:
Durkheim, Emile. 1973. Le suicide: E
´tude de sociologie. Paris: Presses Universi-
taires de France.
———. 1991. Physik der Sitten und des Rechts: Vorlesungen zur Soziologie der
Moral. Frankfurt am Main: Suhrkamp.
Eisner, Manuel. 1995. ‘‘The Effects of Economic Structures and Phases of De-
136 Manuel Eisner
velopment on Crime.’’ In Crime and Economy. Vol. 32, Criminological Re-
search, edited by Council of Europe. Strasbourg: Council of Europe.
———. 1997. Das Ende der zivilisierten Stadt? Die Auswirkungen von Indi-
vidualisierung und urbaner Krise auf Gewaltdelinquenz. Frankfurt am Main:
———. 2001. ‘‘Modernization, Self-Control and Lethal Violence: The Long-
Term Dynamics of European Homicide Rates in Theoretical Perspective.’’
British Journal of Criminology 41:618–38.
———. 2002a. ‘‘Crime, Problem Drinking, and Drug Use: Patterns of Prob-
lem-Behavior in Cross-National Perspective.’’ Annals of the American Acad-
emy of Political and Social Science 580:201–25.
———. 2002b. ‘‘Long-Term Trends in Violence: Empirical Research and
Theoretical Approaches.’’ In Handbook of Violence Research, edited by John
Hagan and Wilhelm Heitmeyer. New York: Westview (forthcoming).
Eleuche-Santini, V. 1979. ‘‘Violence dans le compte
´de Nice au XIIIe sie
Provence Historique 29:362–78.
Elias, Norbert. 1976. U
¨ber den Prozess der Zivilisation: Soziogenetische und psycho-
genetische Untersuchungen. Frankfurt am Main: Suhrkamp.
———. 1978. The Civilizing Process. Vols. 1–2. Oxford: Oxford University
———. 1983. Die ho
¨fische Gesellschaft: Untersuchungen zu einer Soziologie des
¨nigtums und der ho
¨fischen Aristokratie. Frankfurt am Main: Suhrkamp.
Emsley, Clive. 1996. Crime and Society in England, 1750–1900. 2d ed. London:
Farrington, David P. 1998. ‘‘Predictors, Causes, and Correlates of Male Youth
Violence.’’ In Youth Violence, edited by Michael Tonry and Mark H. Moore.
Vol. 24 of Crime and Justice: A Review of Research, edited by Michael Tonry.
Chicago: University of Chicago Press.
Federal Bureau of Investigation. 1998. Uniform Crime Reports: Crime in the
United States—1997. Washington, D.C.: U.S. Government Publishing
Feeley, Malcolm. 1994. ‘‘The Decline of Women in the Criminal Process: A
Comparative History.’’ Criminal Justice History 15:235–74.
Feeley, Malcolm, and Deborah L. Little. 1991. ‘‘The Vanishing Female: The
Decline of Women in the Criminal Process, 16871912.’’ Law and Society
Review 25:719–57.
Ferri, Enrico. 1925. L’omicida nella psicologia e nella psicopatologia criminale. Tu-
rin: Unione tipografico-editrice torinese.
Finnane, Mark. 1997. ‘‘A Decline in Violence in Ireland? Crime, Policing and
Social Relations, 1860–1914.’’ Crime, Histoire et Socie
´—Crime, History and
Society 1(1):51–70.
Fouret, Claude. 1984. L’amour, la violence et le pouvoir: La criminalite
de 1496 a
`1520. History thesis, Universite
´de Lille III.
———. 1987. ‘‘Douai au XVIe sie
`cle: Une sociabilite
´de l’aggression.’’ Revue
d’Histoire Moderne et Contemporaine 34:3–30.
Frank, Michael. 1995. Do
¨rfliche Gesellschaft und Kriminalita
¨t: Das Fallbeispiel
Lippe, 1650–1800. Paderborn: Scho
Long-Term Historical Trends in Violent Crime 137
¨dt, Paul. 1881. Blutrache und Totschlagsu
¨hne im deutschen Mittelalter.
Leipzig: Duncker & Humblot.
Garcia, Pablo Pe
´rez. 1991. La comparsa de los malhechores, Valencia, 1479– 1518.
Valencia: Diputacio de Vale
Garnham, Neal. 1996. The Courts, Crime and the Criminal Law in Ireland,
1692–1760. Dublin: Irish Academic Press.
Gaskill, Malkolm. 2000. Crime and Mentalities in Early Modern England. Cam-
bridge: Cambridge University Press.
Gatrell, V. A. C. 1980. ‘‘The Decline of Theft and Violence in Victorian and
Edwardian England.’’ In Crime and the Law: The Social History of Crime in
Western Europe since 1500, edited by V. A. C. Gatrell, Bruce Lenman, and
Geoffrey Parker. London: Europa.
Given, James B. 1977. Society and Homicide in Thirteenth-Century England.
Stanford, Calif.: Stanford University Press.
Gonthier, Nicole. 1993. De
´linquance, justice et socie
´dans le Lyonnais me
de la fin du XIIIe sie
`cle au de
´but du XVIe sie
`cle. Paris: Arguments.
Gottfredson, Michael R., and Travis Hirschi. 1990. A General Theory of Crime.
Stanford, Calif.: Stanford University Press.
Greenberg, David F. 1985. ‘‘Age, Crime, and Social Explanation.’’ American
Journal of Sociology 91(1):1–20.
———. 1994. ‘‘The Historical Variability of the Age-Crime Relationship.’’
Journal of Quantitative Criminology 10(4):361–73.
Gurr, Ted Robert. 1981. ‘‘Historical Trends in Violent Crime: A Critical Re-
view of the Evidence.’’ In Crime and Justice: An Annual Review of Research,
vol. 3, edited by Michael Tonry and Norval Morris. Chicago: University of
Chicago Press.
———. 1989. ‘‘Historical Trends in Violent Crime: Europe and the United
States.’’ In The History of Crime. Vol. 1, Violence in America, edited by Ted
Robert Gurr. Newbury Park, Calif.: Sage.
Gurr, Ted Robert, Peter N. Grabosky, and Richard C. Hula. 1977. The Politics
of Crime and Conflict: A Comparative History of Four Cities. Beverly Hills,
Calif.: Sage.
Hagemann, Hans-Rudolf. 1981. Basler Rechtsleben im Mittelalter. Basel: Hel-
bling & Lichtenhahn.
Hammer, Carl I, Jr. 1978. ‘‘Patterns of Homicide in a Medieval University
Town: Fourteenth-Century Oxford.’’ Past and Present 78:3–23.
Hanawalt, Barbara A. 1976. ‘‘Violent Death in Fourteenth- and Early Fif-
teenth-Century England.’’ Comparative Studies in Society and History 18:297
———. 1979. Crime and Conflict in English Communities, 1300–1348. Cam-
bridge, Mass.: Harvard University Press.
Heitmeyer, Wilhelm. 1995. Gewalt: Schattenseiten der Individualisierung bei Ju-
gendlichen aus unterschiedlichen Milieus. Weinheim: Juventa.
Henry, Philippe. 1984. Crime, Justice et Socie
´dans la principaute
´de Neucha
au XVIIIe sie
`cle (1707–1806). Neucha
ˆtel: Editions de la Baconnie
Heyden, J. 1983. ‘‘Misdrijf in Antwerpen: Een onderzoek naar de kriminaliteit
in de periode 1404–1429.’’ Acta Falconis 83(3):223–40.
138 Manuel Eisner
Hirschi, Travis, and Michael Gottfredson. 1983. ‘‘Age and the Explanation of
Crime.’’ American Journal of Sociology 89(3):552–84.
Home Office. 1998. Criminal Statistics, England and Wales. London: Home
———. 2000. Criminal Statistics, England and Wales. London: Home Office.
Istituto Nazionale di Statistica. 2000. Statistiche giudiziarie penali, Anno 1998.
Rome: ISTAT.
Jansson, Arne. 1998. From Swords to Sorrow: Homicide and Suicide in Early Mod-
ern Stockholm. Stockholm: Almqvist & Wiksell.
Jarrick, Arne, and Johan So
¨derberg. 1993. ‘‘Spontaneous Processes of Civilisa-
tion.’’ Ethnologia Europea 23:5–26.
Johnson, Eric A. 1995. Urbanization and Crime: Germany, 1871–1914. Cam-
bridge: Cambridge University Press.
Johnson, Eric A., and Eric H. Monkkonen, eds. 1996. The Civilization of Crime:
Violence in Town and Country since the Middle Ages. Urbana: University of Illi-
nois Press.
Junger-Tas, Josine. 1991. ‘‘Nature and Evolution of the Criminality of Young
Adults.’’ In Tenth Criminological Colloquium on Young Adult Offenders and
Crime Policy, edited by Council of Europe. Strasbourg: Council of Europe.
Kaczynska, Elzbieta. 1995. ‘‘Town and Countryside in Penal Judicature and
Criminality: Kingdom of Poland, 1815–1914.’’ Acta Poloniae Historica 71:
Karonen, Petri. 1995. ‘‘Trygg eller livsfarlig? Va
˚ldsbrottsligheten i Finnlands
¨der, 1540–1660.’’ Historisk Tidskrift fo
¨r Finland 80(1):1–11.
———. 1999. ‘‘In Search of Peace and Harmony? Capital Crimes in Late Me-
diaeval and Early Modern Swedish Realm (ca. 1450–1700).’’ In Crime and
Control in Europe from the Past to the Present, edited by Mirkka Lappalainen
and Pekka Hirvonen. Helsinki: Hakapaino.
———. 2001. ‘‘A Life for a Life versus Christian Reconciliation: Violence and
the Process of Civilization in the Kingdom of Sweden, 1540–1700.’’ In Five
Centuries of Violence in Finland and the Baltic Area, by Heikki Ylikangas, Petri
Karonen, and Martti Lehti. Columbus: Ohio State University Press.
King, Peter. 2000. Crime, Justice, and Discretion in England, 1740–1820. Ox-
ford: Oxford University Press.
Kloek, Els. 1990. ‘‘Criminality and Gender in Leiden’s Confessieboeken,
1678–1794.’’ Criminal Justice History 11:1–29.
Lacour, Eva. 2000. ‘‘Kriminalita
¨t in den Grafschaften Manderscheid-Blanken-
heim und Manderscheid-Gerolstein.’’ Zeitschrift der Savigny-Stiftung fu
¨r Re-
chtsgeschichte, Germanistische Abteilung 117:518–49.
Langan, Patrick A., and David P. Farrington. 1998. Crime and Justice in the
United States and in England and Wales, 1981–1996. Washington, D.C.: U.S.
Department of Justice, Office of Justice Programs, Bureau of Justice Statis-
Liliequist, Jonas. 1999. ‘‘Violence, Honour and Manliness in Early Modern
Northern Sweden.’’ In Crime and Control in Europe from the Past to the Pres-
ent, edited by Mirkka Lappalainen and Pekka Hirvonen. Helsinki: Haka-
Long-Term Historical Trends in Violent Crime 139
Mayr, Georg von. 1917. Statistik und Gesellschaftslehre, Dritter Band: Moral-
statistik mit Einschluss der Kriminalstatistik. Tu
¨bingen: Mohr.
Monkkonen, Eric. 2001. ‘‘New Standards for Historical Homicide Research.’’
Crime, Histoire et Socie
´—Crime, History and Society 5(2):7–26.
Muchembled, Robert. 1984. ‘‘Crime et socie
´urbaine: Arras au temps de
Charles Quint (1528–1549).’’ In La France d’Ancien Re
´gime: Etudes re
´unies en
l’honneur de Pierre Goubert, edited by Pierre Goubert. Toulouse: Socie
´mographie historique.
———. 1992. Le temps des supplices; de l’obe
´issance sous les rois absolus. XVe
XVIIIe sie
´cle. Paris: Colin.
———. 1996. ‘‘Elias und die neuere historische Forschung in Frankreich.’’
In Norbert Elias und die Menschenwissenschaften, edited by Karl-Siegvert
Rehbert. Frankurt am Main: Suhrkamp.
Naess, Hans Eyvind. 1982. Trolldomsprosessene i Norge pa
˚1500–1600–tallet: En
retts- og sosialhistorisk underso
¨kelse. Oslo: Universiteits Forlaget.
Oestreich, Gerhard. 1968. ‘‘Strukturprobleme des europa
¨ischen Absolutis-
mus.’’ Vierteljahreszeitschrift fu
¨r Sozial- und Wirtschaftsgeschichte 55:329
———. 1982. Neostoicism and the Early Modern State. Cambridge: Cambridge
University Press.
¨sterberg, Eva. 1996. ‘‘Criminality, Social Control, and the Early Modern
State: Evidence and Interpretations in Scandinavian Historiography.’’ In
The Civilization of Crime: Violence in Town and Country since the Middle Ages,
edited by Eric A. Johnson and Eric H. Monkkonen. Urbana: University of
Illinois Press.
¨sterberg, Eva, and Dag Lindstro
¨m. 1988. Crime and Social Control in Medieval
and Early Modern Swedish Towns. Uppsala: Academia Upsaliensis.
Panico, G. 1991. ‘‘Criminali e peccatori in Principato Citra alla fine del Set-
tecento (1770–1780).’’ In Criminalita
`e societa
`in eta
`moderna, edited by Luigi
Berlinguer and Floriana Colao. Milan: Guiffre
Pohl, Susanne. 1999. ‘‘ ‘Ehrlicher Totschlag’—‘Rache’—‘Notwehr.’ Zwischen
¨nnlichem Ehrcode und dem Primat des Stadtfriedens (Zu
¨rich 1376–
1600).’’ In Kulturelle Reformation: Sinnformationen im Umbruch 1400–1600,
edited by Bernhard Jussen and Craig Koslofsky. Go
¨ttingen: Vandenhoeck &
Roets, Anne-Marie. 1982. ‘‘Vrouwen en criminaliteit in Gent in de achttiende
eeuw.’’ Tijdschrift voor Geschiedenis 92:363–78.
Romani, Mario A. 1980. ‘‘Criminalita
`e Giustizia nel Ducato di Mantova alla
fine del Cinquecento.’’ Rivista Storica Italiana 92:679–706.
Roth, Randolph. 2001. ‘‘Homicide in Early Modern England, 1549–1800:
The Need for a Quantitative Synthesis.’’ Crime, Histoire et Socie
History and Society 5(2):33–68.
Rousseaux, Xavier. 1986. ‘‘Ordre et violence: Criminalite
´et re
´pression dans
une ville brabanc
¸onne. Nivelles (1646–1695).’’ Revue de Droit Pe
´nal et de
Criminologie 66(7):649–92.
———. 1999a. ‘‘From Case to Crime: Homicide Regulation in Medieval and
Modern Europe.’’ In Die Entstehung des o
¨ffentlichen Strafrechts: Bestandsauf-
140 Manuel Eisner
nahme eines europa
¨ischen Forschungsproblems, edited by Dietmar Willoweit.
Cologne: Bo
———. 1999b. ‘‘Sozialdisziplinierung, Civilisation des moeurs et monopo-
lisation du pouvoir: Elements pour und histoire du contro
ˆle social dans les
Pays-Bas me
´ridionaux, 1500–1815.’’ In Institutionen, Instrumente und Akteure
sozialer Kontrolle und Disziplinierung im fru
¨hneuzeitlichen Europa, edited by
Heinz Schilling. Frankfurt am Main: Klostermann.
Ruff, Julius R. 1984. Crime, Justice and Public Order in Old Regime France: The
´es of Libourne and Bazas, 1696–1789. London: Croom Helm.
Ruggiero, Guido. 1978. ‘‘Law and Punishment in Early Renaissance Venice.’’
Journal of Criminal Law and Criminology 69(2):243–56.
———. 1980. Violence in Early Renaissance Venice. New Brunswick, N.J.:
Rutgers University Press.
Samaha, Joel. 1974. Law and Order in Historical Perspective: The Case of Eliza-
bethan Essex. New York: Academic Press.
Sardi, Laura Carli. 1991. ‘‘Analisi statistica sulla criminalita nel 1700 (reati e
pene) con riguardo allo Stato senese.’’ In Criminalita
`e societa
`in eta
edited by Luigi Berlinguer and Floriana Colao. Milan: Guiffre
Schmidt, Axel. 1994. ‘‘ ‘Wo die Ma
¨nner sind, gibt es Streit’: Ehre und Ehrgef-
¨hl im la
¨ndlichen Sardinien.’’ In Ehre: Archaische Momente in der Moderne,
edited by Ludgera Vogt and Arnold Zingerle. Frankfurt am Main: Suhr-
Schormann, Gerhard. 1974. ‘‘Strafrechtspflege in Braunschweig-Wolfenbu
tel.’’ Braunschweigisches Jahrbuch 55:90–112.
¨ssler, Martin. 1991. ‘‘Statistische Untersuchung des Verbrechens in
¨rnberg im Zeitraum von 1285 bis 1400.’’ Zeitschrift der Savigny-Siftung
¨r Rechtsgeschichte, Germanistische Abteilung 108:117–91.
———. 1994. ‘‘Verbrechen im spa
¨tmittelalterlichen Olmu
¨tz: Statistische Un-
tersuchung der Kriminalita
¨t im Osten des Heiligen Ro
¨mischen Reiches.’’
Zeitschrift der Savigny-Siftung fu
¨r Rechtsgeschichte, Germanistische Abteilung
———. 1996. ‘‘Quantifizierung, Impressionismus und Rechtstheorie: Ein Be-
richt zur Geschichte und zum heutigen Stand der Forschung u
¨ber Krimi-
¨t im Europa des Spa
¨tmittelalters und der fru
¨hen Neuzeit.’’ Zeitschrift
der Savigny-Siftung fu
¨r Rechtsgeschichte, Germanistische Abteilung 113:247– 78.
———. 1998. ‘‘Verbrechen in Krakau (1361– 1405) und seiner Beistadt Kasi-
mir (1370–1402).’’ Zeitschrift der Savigny-Stiftung fu
¨r Rechtsgeschichte, Ger-
manistische Abteilung 115:198–338.
Schuster, Peter. 1995. Der gelobte Frieden: Ta
¨ter, Opfer und Herrschaft im spa
mittelalterlichen Konstanz. Constance: Universita
¨tsverlag Konstanz.
———. 2000. Eine Stadt vor Gericht: Recht und Alltag im spa
Konstanz. Paderborn: Scho
Schwerhoff, Gerd. 1991. Ko
¨ln im Kreuzverho
¨r: Kriminalita
¨t, Herrschaft und Ge-
sellschaft in einer fru
¨hneuzeitlichen Stadt. Bonn: Bouvier.
———. 1995. ‘‘Geschlechtsspezifische Kriminalita
¨t im fru
¨ln.’’ In Von Huren und Rabenmu
¨ttern, Weibliche Kriminalita
¨t in der fru
Neuzeit, edited by Otto Ulbricht. Cologne: Bo
Long-Term Historical Trends in Violent Crime 141
———. 2002. ‘‘Criminalized Violence and the Civilizing Process—a Re-
appraisal.’’ Crime, Histoire et Socie
´s—Crime, History and Societies (forth-
Sharpe, James A. 1983. Crime in Seventeenth-Century England: A County Study.
Cambridge: Cambridge University Press.
———. 1984. Crime in Early Modern England, 1550– 1750. London: Longman.
———. 1988. ‘‘The History of Crime in England c. 1300–1914.’’ British Jour-
nal of Criminology 28:254–67.
———. 1996. ‘‘Crime in England: Long-Term Trends and the Problem of
Modernization.’’ In The Civilization of Crime: Violence in Town and Country
since the Middle Ages, edited by Eric A. Johnson and Eric H. Monkkonen.
Urbana: University of Illinois Press.
Shelley, Louise I. 1981. Crime and Modernization: The Impact of Industrialization
and Urbanization on Crime. Carbondale: Southern Illinois University Press.
Simon-Muscheid, Katharina. 1991. ‘‘Gewalt und Ehre im spa
Handwerk am Beispiel Basels.’’ Zeitschrift fu
¨r historische Forschung 18(1):1– 31.
Soman, Alfred. 1980. ‘‘Deviance and Criminal Justice in Western Europe
(1300–1800): An Essay in Structure.’’ Criminal Justice History 1:3–28.
Spierenburg, Pieter. 1984. The Spectacle of Suffering: Executions and the Evolution
of Repression: From a Preindustrial Metropolis to the European Experience. Cam-
bridge: Cambridge University Press.
———. 1994. ‘‘Faces of Violence: Homicide Trends and Cultural Meanings,
Amsterdam, 1431–1816.’’ Journal of Social History 27:701– 16.
———. 1995. ‘‘Elias and the History of Crime and Criminal Justice: A Brief
Evaluation.’’ International Association for the History of Crime and Criminal
Justice Bulletin 20:17–30.
———. 1996. ‘‘Long-Term Trends in Homicide: Theoretical Reflections and
Dutch Evidence, Fifteenth to Twentieth Centuries.’’ In The Civilization of
Crime: Violence in Town and Country since the Middle Ages, edited by Eric A.
Johnson and Eric H. Monkkonen. Urbana: University of Illinois Press.
———. 2001. ‘‘Violence and the Civilizing Process: Does It Work?’’ Crime,
Histoire et Socie
´—Crime, History and Society 5(2):87–106.
Spierenburg, Pieter, ed. 1998. Men and Violence: Gender, Honor, and Rituals in
Modern Europe and America. Columbus: Ohio State University Press.
Steffensmeier, Darrell J., Emilie Andersen Allan, Miles D. Harer, and Cathy
Streifel. 1989. ‘‘Age and the Distribution of Crime.’’ American Journal of So-
ciology 94(4):803–31.
Steffensmeier, Darrell, and Cathy Streifel. 1991. ‘‘Age, Gender, and Crime
across Three Historical Periods.’’ Social Forces 69(3):869–94.
Stone, Lawrence. 1983. ‘‘Interpersonal Violence in English Society, 1300
1980.’’ Past and Present 101:22–33.
Taylor, Charles. 1989. Sources of the Self: The Making of the Modern Identity.
Cambridge, Mass.: Harvard University Press.
Thome, Helmut. 1995. ‘‘Modernization and Crime: What Is the Explana-
tion?’’ IAHCCJ Bulletin 20:31–47.
———. 2001. ‘‘Explaining Long Term Trends in Violent Crime.’’ Crime, His-
toire et Socie
´—Crime, History and Society 5(2):69–87.
142 Manuel Eisner
Tilly, Charles. 1992. Coercion, Capital, and European States,
990–1992. Rev.
ed. Cambridge, Mass.: Blackwell.
van de Pol, Lotte C. 1987. ‘‘Vrouwencriminaliteit in Amsterdam in te tweede
helft van de 17e eeuw.’’ Tijdschrift voor Criminologie 5:148–55.
Vanhemelryck, Fernand. 1981. De criminaliteit in de ammanie van Brussel van
de Late Middeleeuwen, tot het einde het Ancien Re
´gime (1404–1789). Bru
Paleis der Academie
Verkko, Veli. 1951. Homicides and Suicides in Finland and Their Dependence on
National Character. Vol. 3, Scandinavian Studies in Sociology. Copenhagen:
Weber, Max. 1922. ‘‘Die protestantische Ethik und der ‘Geist’ des Kapitalis-
mus.’’ In Gesammelte Aufsa
¨tze zur Religionssoziologie, vol. 1, edited by Max
Weber. Tu
¨bingen: Mohr.
Weinbaum, Martin, ed. 1976. The London Eyre of 1276. London: London Rec-
ord Society.
¨m, Per-Olof H. 1985. Everyday Violence in Contemporary Sweden: Situa-
tional and Ecological Aspects. Stockholm: National Council for Crime Preven-
tion, Sweden, Research Division.
———. 1992. ‘‘Context-Specific Trends for Criminal Homicide in Stockholm,
1951–1987.’’ Studies on Crime and Crime Prevention 1:88–105.
Wintemute, Garen. 2000. ‘‘Guns and Gun Violence.’’ In The Crime Drop in
America, edited by Alfred Blumstein and Joel Wallman. Cambridge: Cam-
bridge University Press.
Ylikangas, Heikki. 1976. ‘‘Major Fluctuations in Crimes of Violence in Fin-
land: A Historical Analysis.’’ Scandinavian Journal of History 1:81–103.
———. 1998a. ‘‘Die langfristigen Entwicklungstrends der Verbrechen wider
das Leben in Finland.’’ In About Violence, edited by Heikki Ylikangas. Hel-
sinki: University of Helsinki.
———. 1998b. What Happened to Violence? An Analysis of the Development of Vio-
lence from Medieval Times to the Early Modern Era Based on Finnish Source Ma-
terial. Helsinki: Suomen Akatemia.
Ylikangas, Heikki, Petri Karonen, and Martti Lehti. 2001. Five Centuries of Vi-
olence in Finland and the Baltic Area. Columbus: Ohio State University Press.
Zorzi, Domenico. 1989. ‘‘Sull’ammistrazione della Giustizia penale nell’eta
delle riforme: Il reato di omicidio nella Padova di fine settecento.’’ In Cri-
mine, Giustizia e societa
`veneta in eta
`moderna, edited by L. Berlinguer and F.
Colao. Milan: Giuffre
... Analyses of time trends in crime and violence have long traditions both in Nordic and international criminological research (see e.g., Baumer, Vélez and Rosenfeld, 2018;Eisner, 2003;Estrada, 2001Estrada, , 2019. The overall impression from existing research is that violence levels among adolescents increased from the mid-1900s into the 1990s and then declined substantially in the first two decades of the 2000s. ...
... Manuel Eisner (2003) made a very important contribution to our understanding of the evolution of violent interpersonal behavior among humans by Content courtesy of Springer Nature, terms of use apply. Rights reserved. ...
Full-text available
This paper is a summary of the research I did on the development, prevention and treatment of antisocial behavior over the past half century. It is also a reflection on the intellectual and cultural influences that guided my research, from the classic philosophers to my 21st century peers…including David Farrington.
... This means that violence in the region has certain spatial distribution to be accounted in our narrative about the conflict. The end of the Mexican revolution was the beginning of an era in Mexico that I call the Secular Decline of violence, using the term used by Turchin andNefedov (2009) andEisner (2003). Secular decline means a long period of time of a new demographic tendency, in this case, a decline in violence. ...
Since the beginning of the most repressive phase of the Mexican war on drugs, the region of Tierra Caliente, in southern Mexico, has been one of the epicentres of the increase of homicides, the surge of extremely violent drug trafficking organisations. Drawing form historical information, process tracing of the conflict, and in-depth interviews with key informants, this chapter aims to explain the logics of violence as part of the historical political economy of the region. The main argument is that explanatory approaches to violence in Mexico are historically congruent rather than excluding in the Tierra Caliente.
... On one hand, many researchers argued that anonymity among urban dwellers in cities, high population density and less probability of arrest permit the criminals to commit their crime without any fear of recognition (Braithwaite, 1975;Glaeser and Sacerdote, 1999). On the other hand, others believed that at low levels of urbanization, high crime rates may be observed due to the sparsely located residents in an area; meaning that the increase of the urbanization or the size of the city would lead to the reduction of crime level as the result of closer proximity of the residents (Johnson 1992;Spierenburg;1996, Gaviria andPages, 2002;Eisner, 2003). Therefore, it can be argued that urbanization may have both negative and positive effects on crime rates depending on the status of the different urban settings. ...
Full-text available
However, urbanization can have positive effects on the labor force and economic growth in urban areas the unplanned and rapid pace of urbanization process can ultimately lead to unpleasant consequences including marginalization, unemployment, poverty, the emergence of slums and shanty towns and the prevalence of crime in urban areas. Therefore, the objective of this study was to investigate the spatial spillover effects of urbanization on crime in 30 Iranian provinces during the period from 2006 to 2015 using dynamic spatial panel data models. The results of the study revealed that urbanization had a positive and statistically significant impacts on crime per capita in Iranian provinces indicating an increase in urbanization enhances the crime level. In addition, unemployment rate, and human capital had significant and negative effects on crime in Iranian provinces, respectively. Finally, the crime per capita in a previous period exerted a significant positive effect on the current crime which stablished the dynamic nature of this phenomenon. Moreover, the results of the study demonstrated that increasing crime per capita in neighboring provinces could elevate the crime in a given province.
Full-text available
An equation based on network entropy predicts a national economic growth rate from the rate at which national average IQs increase.
In many American Indian nations the security situation is dire. While scholars have studied how institutions shape economic development in American Indian and Alaska Native (AIAN) nations, the role of AIAN institutions for security and violent crime has received much less attention—despite the extensive literature highlighting the important role of effective and legitimate institutions in the long‐term decline of violence. We analyze how varying types of American Indian polities and judiciary institutions fare in tackling violent crime using data across 146 American Indian polities. Our findings indicate that more autonomous American Indian criminal justice institutions with specialized court systems are associated with lower violent crime. However, customary justice institutions do not appear to be effective in reducing violent crime, highlighting the problem of cultural mismatch between traditional and formal justice systems. We argue that analyzing AIAN nations provides important insights into how institutional legitimacy can shape violent crime.
According to the United Nations Office on Drugs and Crime (Global study on homicide. Booklet 1: executive summary. Accessed 5 June 2022, 2019; Victims of intentional homicide. Accessed 5 June 2022, 2020), Italy has the second lowest homicide rate among all European countries, despite being home to mafia organizations such as the Cosa Nostra, ‘Ndrangheta, and Camorra. This chapter provides an up-to-date overview of homicide in the Italian context, describing its legal, historical, and epidemiological aspects. Femicide, infanticide, and filicide are also reviewed as specific classes of homicide, with a focus on their psychological, psychopathological, and sociocultural risk factors. Furthermore, the homicide-suicide phenomenon – which, in Italy, tends to be intrafamilial, involving an older man killing a younger woman (typically his current or former partner) – is discussed in depth.
Full-text available
Violence rising in Latin America since the early 1990s has puzzled media, policymakers and academia. Characterising high scales of violence in non-political confrontations has been one of the main challenges. The main argument of this essay is that the hybrid criminal nature of violence in Latin America by non-state organisations has pushed the discussion to several misinterpretations and conceptual stretching that produces fog rather than clarity. Instead, this essay proposes a concept of criminal war that can capture the complex nature of violence in Latin America by drawing convergences and divergences from diverse fields of literature and confronting usual mischaracterisations in current Latin American research.
Instrumental studies of gender and the accomplishment of violent robbery in the United States (Miller, 1998) United Kingdom (Brookman, Mullins, Bennett, & Wright, 2007) demonstrated that whilst men's and women's motivations for committing street robbery were similar – to fund addiction, hedonism or to supplement meagre incomes – male and female robbers commissioned the offences in markedly different ways. By offering a historical counterpart to modern criminological enquiries, this article examines gender and the enactment of street robbery in Liverpool between 1850 and 1870. By drawing on a database of 260 cases, I argue, in line with the patterns of offending by men and women uncovered in contemporary UK and US criminological studies, that men and women committed street robberies in ways that were shaped by the gender-stratified nature of street life and structured by broader gender and class inequalities. In doing so, the article reflects on how historical criminological approaches can develop our understandings of gender and violent crime across time and place and hopes to encourage future interdisciplinary work in historical criminology, gender and violent crime.
Full-text available
Sumário: Introdução; 1. Retratos da criminalidade; 2. Portugal, um país inquieto?; 3. Situando a insegurança em coordenadas sociais; Conclusão.