[SOLVED] reading and come up with questions and answers – Humanities

reading the book chapter and the article come up with 10 questions and answers in total, 5 for each article. Don’t use simple quote as an answer. no yes/no question. No true/false question. reference the page number after the answerexample:What is the evidence of the inequality of colored and white people during the Drug War?Teens of 12 to 17 who are white are more than a third more likely to have sold drugs than African American teens (99). White students use cocaine 7 times higher than blacks, heroine 7 times higher than them and 8 times more on crack (112-113).Long-Term Trends in Crime: Continuity and Change
Long-Term Trends in Crime: Continuity and Change
Marcelo F. Aebi and Antonia Linde
The Oxford Handbook of the History of Crime and Criminal Justice
Edited by Paul Knepper and Anja Johansen
Print Publication Date: May 2016
Subject: Criminology and Criminal Justice, Criminological Theories
Online Publication Date: Jul 2016 DOI: 10.1093/oxfordhb/9780199352333.013.3
Abstract and Keywords
This chapter introduces the available research on long-term crime trends and shows the
major role that cross-national comparisons have played since the nineteenth century, re­
vealing that the first researchers were fully aware of such comparisons’ pitfalls. Since
then, most research has focused on homicide trends—often used as a proxy for trends in
all violent crime—in industrialized Western nations, because data are not widely available
for other regions of the world and other types of offenses. Problems related to finding ap­
propriate crime measures and valid denominators for the absolute numbers provided by
such measures are also investigated, as are the risks of chronocentrism, parochialism,
and causationism. Finally, it cautions that explaining the decrease in Western industrial­
ized countries’ interpersonal violence since the late Middle Ages through the effect of a
civilizing process requires overlooking the genocides, mass killings, and war crimes com­
mitted in these countries and their former colonies.
Keywords: long-term crime trends, chronocentrism, parochialism, causationism, Durkheimian modernization per­
spective, civilizing process, ecological opportunity perspective, Marxian world system perspective, De la violence
au vol hypothesis
Introduction
ASSESSING the state-of-the-art research on long-term crime trends in Western industrial­
ized countries, Johnson and Monkkonen (1996b) conclude that until the late 1970s, re­
searchers considered it almost impossible to obtain reliable estimates of the levels of
crime in earlier times. As a consequence, they were less interested in crime rates and
trends than in “everything surrounding crime-legal institutions from the penal system to
modes of enforcement, popular and elite attitudes, court systems and crimes coming to
courts” (Johnson & Monkkonen 1996b, p. 2). This state of affairs changed during the fol­
lowing twenty years in such a way that by the mid-1990s it was “assumed that crime—in
particular personal violence—in the Western world has declined since the early Middle
Ages until very recently. This simple statement encompasses the two aspects of crime
Page 1 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
that scholars expected never to approach: international comparison and lengthy
trends” (Johnson & Monkkonen 1996b, p. 4).
The goal of this chapter is to introduce the reader to the available research on long-term
trends in crime, paying special attention to its historical development and the method­
ological challenges it involves, and presenting its main findings and their possible inter­
pretations. The reader will thus discover that Johnson and Monkkonen (1996b) are right
about the general consensus on the long-term decrease of interpersonal violence in West­
ern Europe and the United States, but also that international comparisons have been a
key element of research in criminology since the origins of this social science. We start
the chapter with a short overview of the historical development of comparisons of crime
rates and crime trends. On that basis, we present a classification of the types of research
that have been conducted according to their unit of analysis, (p. 58) time span, and offens­
es analyzed. This leads us to analyze the way in which crime has been defined and mea­
sured by researchers, putting the accent on the use of homicide as a proxy for violent of­
fenses. Then we present the main sources of data for research on crime trends and dis­
cuss their validity and reliability. We also examine the pitfalls involved in the process of
interpretation of the trends shown by such sources. Finally, we provide an overview of the
main theories proposed by researchers to explain the trends observed.
I. Historical Development of Research on Crime
Trends
Science is based on comparisons; therefore, as soon as data were available for more than
one year or more than one country, researchers and journalists started comparing them.
Hence, when the first comprehensive national judiciary statistics were published in
France in 1827, the data included in them—which referred back to the year 1825—were
immediately compared to the figures available for other countries, such as England, for
which data existed but were not as rich in detail as the brand-new French statistics
(Aubusson de Cavarlay 1998). Two years later, when the Netherlands—which at that time
included Belgium—published its first criminal statistics, Adolphe Quételet (1829)
accompanied them with an essay in which he included comparisons with the French sta­
tistics as well as his first observations about the regularity of crime across time. Thus, an­
alyzing the French figures for 1825 and 1826—the only two years available at that time—
he mentioned the “frightening regularity” (effrayante regularité) with which the same
crimes were committed. For example, he pointed out that there had been 250 crimes
against relatives in 1825, and 244 in 1826; 613 forgeries in 1825, and 610 in 1826; and
4,841 thefts in 1825, against 4,489 in 1826 (Quételet 1829, p. 28). Later, he extended his
observations to slightly longer series and observed the same “remarkable
consistency” (constance étonnante) (1833, p. 20) not only in the number of crimes, but al­
so in the characteristics of the accused offenders in terms of age and gender (Quételet
1831; Quételet 1835). It is on the basis of the 1842 English translation of his 1835 Trea­
tise on Man and the Development of His Faculties that this remark, which gave birth to
Page 2 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
the illusion of stable crime trends, a conclusion supported also by Guerry (1833), has
been quoted for one and a half centuries, from Buckle (1864) to Hagan (2013), to name
only two researchers at both ends of that time frame.1
Although Quételet (1829) contrasted the figures from the first Dutch criminal statistics
with the French ones, he believed that they could not be compared to the data from Eng­
land, because, among other reasons, the latter only included indictable offenses. The
comparison with France was possible because the data had been collected in a similar
way and both countries applied the same law. The reason for that similarity was that the
Netherlands, during its time under the rule of Louis Napoleon as the Kingdom (p. 59) of
Holland (1806–1810), had adopted the French civil and penal codes with only minor modi­
fications. At the time of Quételet’s comparison, the main difference between the two sys­
tems was that the jury had been abolished in the Netherlands (1829, p. 26).
This shows that since the beginning, the commentators of criminal statistics have been
aware of many of the problems related to cross-national comparisons, such as the differ­
ences in the definitions of offenses, criminal codes and criminal proceedings codes, and
ways of collecting data. Some of them, like Quételet or Mittermaier (1830), have men­
tioned these problems but considered that some types of comparisons could still be made,
while others, like Alphonse De Candolle (1987 [1830]; 1987 [1832]), have been extremely
reluctant to engage in international comparisons. As we will see throughout this chapter,
these two positions can still be found among contemporary researchers. In addition, we
will see, in the subsection on homicide as a proxy for violent offenses, that also since the
beginning researchers have been complaining about the way in which the press compares
levels of crime across nations.
Back in 1829, the Grand Duchy of Baden also published its conviction statistics, which
were compared by Mittermaier (1829) to the data on convictions available from England
as well as some regions of Switzerland, the United States, and other German states (see
Reinke 1998). Then, in 1835, Ducpetiaux published a book including data from four West­
ern European countries that seems to be the first one completely dedicated to cross-na­
tional comparisons of crime statistics.2 Since then, the interest in comparative criminolo­
gy has never ceased.
Among the most influential works of the second half of the nineteenth century, we can
mention the first German doctoral dissertation based on criminal statistics, in which
Georg von Mayr (1865) compared different states of Germany and established crime
trends since the 1830s (Reinke 1998), as well as the studies conducted by researchers
from the French and the Italian schools of criminology. Hence, in France, Gabriel Tarde
(1886) published the classic book Comparative Criminality (La criminalité comparée), in
which he analyzed recorded crime across countries and regions since the beginning of
the statistical series in the mid-1820s. Tarde paid particular attention to the negative ef­
fects of industrialization on levels of crime in the cities compared to those in rural areas,
which he considered healthier. It took almost a century to find data showing that interper­
sonal violence was indeed higher in the countryside (Johnson & Monkkonen 1996b). At
Page 3 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
the same time, Ferri (1895) produced his Atlas of Homicide spanning nations across Eu­
rope, which showed higher levels of homicide in the southern regions than the northern
ones, and led to a debate on the influence of climate—a factor previously mentioned by
Quételet (1835)—on delinquency. Ferri (1892, p. 256) was also among the first re­
searchers to establish long-term homicide and suicide trends—covering in some cases
several decades of the nineteenth century—for a series of European countries.
During the first half of the twentieth century, Verkko compared homicides across nations,
but he published in Finnish, and despite Ekelund’s (1942) efforts to summarize in English
some of his main ideas, his work remained relatively unknown to the scientific community
until he was translated in the second half of the century (see, e.g., Verkko 1967). Rusche
and Kirchheimer (1939) also studied crime trends for different offenses (p. 60) in several
European countries and during different periods of the nineteenth and early twentieth
centuries, including data from police, conviction, and prison statistics, and giving a pre­
dominant role to the relationship between unemployment and crime.
In the 1960s, two different tendencies can be observed. Sociologists of deviance rediscov­
ered the limits of criminal statistics (Kitsuse & Cicourel 1963) and became more interest­
ed in the social reaction to crime than in crime itself. At the same time, some historians
started looking for data on crime from the years before the publication of the first crimi­
nal statistics. Among the latter, Pierre Chaunu directed in France a series of studies
based on the analysis of ancient criminal proceedings that allowed the construction of
time series and the development of the hypothesis known as “violence to theft”—which
will be discussed later—to explain them (Boutelet & Chaunu 1962). Another group of
French historians who specialized in punishment in the nineteenth century were interest­
ed in the development of the prison system. In 1975, the publication by philosopher
Michel Foucault of Discipline and Punish presented an innovative view of the birth of pris­
ons. Foucault (1975) believed that prisons became the main form of punishment in the
late eighteenth and early nineteenth centuries in the context of the development of a dis­
ciplinary society. They thus were part of a carceral system that included schools, hospi­
tals, psychiatric asylums, factories, and similar institutions whose aim was to discipline
the bodies of the persons within them in order to control them. Nevertheless, Foucault in
his analysis followed a heterodox method that was severely criticized by contemporary
historians of punishment in a special issue of a French specialized journal.3 This group
asked Foucault “essentially, whether his dichotomies between l’age classique and l’age
moderne do not reproduce the ‘traditional-modern’ schematization which has so bedev­
iled the historical genealogy of the industrial world” (Ignatieff 1982). They also criticized
his idea of presenting France as having become a disciplined society by 1850 and were
skeptical about the possibility of a hidden plan developed by the bourgeoisie to establish
the prison system: “Who exactly was responsible for putting the carceral archipelago in
place?” (Ignatieff 1982). Although later Foucault (2009) moderated his position, he is tra­
ditionally quoted according to his ideas of the mid-1970s.
Page 4 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
In the meantime, and for reasons related to an increase in crime since the 1960s that will
be developed more deeply in the next section, researchers became particularly interested
in establishing long-term crime trends. Thus, by the end of the 1970s, there was enough
empirical evidence to affirm that violent crime had been declining in Europe since the
Middle Ages (Gurr 1981). Additionally, researchers started questioning the conclusions of
the nineteenth-century authors who had considered that urbanization was related to an
increase of crime (Lodhi & Tilly 1973). They also casted serious doubts on many classic
ideas of that century, such as those related to the relationship among change, strain, and
disorder, which suddenly seemed to lack empirical support (Tilly 1984, p. 53). The validity
of these findings was enhanced by the fact that similar results were found in different re­
gions of Western Europe, such as the Netherlands (Spierenburg 1994) and the Scandina­
vian countries (Österberg 1992), as well as in the United States (for an overview, see
Johnson & Monkkonen 1996a).
In that context, European data became particularly relevant, as the study of longterm crime trends in the United States has been problematic because of the lack of na­
tional crime statistics until the development of the Federal Bureau of Investigation’s Uni­
form Crime Reports in 1930. Before that, many U.S. states produced their own statistical
(p. 61)
collections, but only prison statistics were collected at the national level. Since the begin­
ning of the twentieth century, Robinson (1910; 1911; 1920) had denounced such a situa­
tion and pleaded for the development of national statistics based on convictions. Never­
theless, court statistics at the national level were never fully developed.4 Indeed, Robin­
son highlighted many of the drawbacks that are still mentioned by contemporary re­
searchers, who usually are forced to establish long-term trends on the basis of data for
specific cities or regions of the United States (Roth 2009). As a consequence, most of the
research currently produced in that country focuses on crime rates after the Second
World War.
II. Varieties of Contemporary Research on
Crime Trends
The recent literature on crime trends admits at least three broad classifications based, re­
spectively, on the unit of analysis, the type of offense studied, and the time span covered
by the analysis. According to their unit of analysis, studies can be divided into those cov­
ering crime trends in a particular country and those adopting a comparative perspective
including two or more countries. According to the type of offenses analyzed, studies that
examine a single country or cover short periods of time typically contain analyses of sev­
eral types of crime. On the contrary, the overwhelming majority of comparative studies,
and especially those covering long periods of time, are dedicated to homicide trends. Oth­
er violent offenses, such as assault and robbery, have received less attention, while prop­
erty offenses and cybercrime are seldom studied. Finally, according to the time span of
the analysis, two types of studies can be distinguished. The first includes research cover­
ing the evolution of crime after the Second World War and has been written mainly by
Page 5 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
criminologists. The second includes research covering crime trends since the late Middle
Ages and has been written by researchers from different fields including history, sociolo­
gy, political science, philosophy, and, of course, criminology. Both approaches search for
causal explanations of crime trends; however, the kind of analysis that can be conducted
for each period is heavily conditioned by the available data. Research based on post-1945
trends can take advantage of the multiplication of macro-level indicators and indexes that
are useful as independent variables. Hence, authors usually proceed to study the correla­
tions between different crime measures and a series of socioeconomic, political, and cul­
tural variables. These analyses have become increasingly sophisticated with the availabil­
ity of statistical packages that allow the application of econometric models, although
quite often there are serious (p. 62) doubts as to whether the data available are accurate
enough for such models. On the opposite end of the spectrum, research covering long pe­
riods of time for which few independent variables are available tends to be more descrip­
tive, as well as speculative, regarding the explanations of the trends observed.
In that perspective, the temporal distribution of studies on crime trends conducted dur­
ing the last fifty years shows two peaks: one at the turn of the 1980s and the other at the
turn of the 2000s, coinciding, respectively, with an upward and a downward trend in
crime in the United States and Western European countries. At the time of the first peak,
criminologists were trying to explain the rise in crime that followed the Second World
War, and some researchers reacted by showing that such an increase was preceded by a
long-term decrease. During that period, the most well-known example of an article writ­
ten by criminologists, based on national data covering the period after the Second World
War, including several offenses, and in which crime trends were tested against a series of
independent variables is the study by Cohen and Felson (1979) that introduced the rou­
tine activities approach. At the same time, the most-quoted example of the work of a polit­
ical scientist, covering several centuries, focused on violence, and including several coun­
tries is a 1981 paper by Ted Robert Gurr. It was published in the context of the debate on
crime trends in Western societies, and it stressed that the increase in the 1960s and
1970s was preceded by a much longer downward trend, in such a way that the evolution
of serious crime depicted a distended U-shaped curve. Gurr (1981) also drew a distinction
between trends in the United States and those in Western Europe, and introduced the
ideas of Elias (2000 [1939]) and the civilizing process to explain the evolution of crime
since the early modern period.
At the turn of the 2000s, research on crime trends peaked again as criminologists started
proposing explanations for the decrease in crime observed in the United States since
1992. Curiously enough, even after similar declines have been observed in other coun­
tries, many of the U.S. explanations are still focusing on local causes for such a decline
(Goldberger & Rosenfeld 2008; Rosenfeld 2014), a situation that Tonry (2014, p. 3) de­
scribes as “at least a little parochial.” In a comparative perspective, nonetheless, re­
searchers placed this new decrease in the context of the well-known long-term downward
trend in interpersonal violence (Eisner 2001; Eisner 2003). Thus, the upward trend ob­
Page 6 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
served from the 1960s to the early 1990s came to be viewed as an exception for which
different explanations have been proposed (Aebi & Linde 2014c).
Hence, after almost 200 years of research, it can be said that the hypothesis of stable
crime trends has been falsified, and the same is true for the correlation between urban­
ization and violent crime. Indeed, “rural societies of the distant past were far more vio­
lent than the predominantly urban societies of both the recent past and the
present” (Johnson & Monkkonen 1996b, p. 7). The present state of knowledge has been
reviewed by Pinker (2011), who produced a general overview of the decline of interper­
sonal violence, gathering information from prehistoric times to the twenty-first century.
He shows that researchers currently tend to agree on the fact that, in the context of a civ­
ilizing process, violence has been decreasing in industrialized Western societies since the
late Middle Ages (i.e., 1300–1500 C.E.), and that such a decrease was particularly (p. 63)
important in the seventeenth and eighteenth centuries. In Western European countries,
this decline has only been interrupted in three historical periods, during which the avail­
able data suggest an upward trend of homicide rates. The first one took place in the last
decades of the sixteenth century, the second one at the turn of the nineteenth century
(i.e., before and just after 1800), and the third one from the 1960s to the early 1990s (Eis­
ner 2001, p. 633). In the United States, where the period of observation is shorter, there
have been “three great surges of violent crime which began ca. 1850, 1900, and
1960” (Gurr 1981, p. 295). Nevertheless, it must be mentioned that a major objection to
the concept of the civilizing process is that in order to accept it, one has to focus on the
periods of peace and leave aside the genocides, mass killings, and homicides that took
place in the same countries where the process was taking place (as exemplified in acts
against members of native nations or ethnic or religious minorities), in their colonies, or
in the territories where they were fighting wars in which many of the victims were civil­
ians. This leads us to concentrate on the types of offenses that have been studied by re­
searchers.
III. Defining Crime
A. Property and Violent Offenses
Gottfredson and Hirschi (1990) define crime as “acts of force or fraud undertaken in pur­
suit of self-interest” (p. 15). This wide-ranging definition, inspired by the classic princi­
ples of Roman law, presents the advantage of easily generating consensus. However, the
main challenge for researchers interested in comparative crime trends is to identify the
operational definitions of crime used across time and space.
From that perspective, property offenses are particularly problematic. Even if the basic
principle remains the same—taking something that belongs to someone else against his
or her will—the concepts of property and its distribution across the population have un­
dergone so many changes that it is practically impossible to follow the evolution of theft
through time and space. For example, large-scale economic crime has been documented
Page 7 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
since the times of the Roman Empire (Huisman et al. 2015, p. 3), but until recent times it
had not been measured in terms of rates that could be used to establish trends and com­
parisons across nations. In the case of petty property offenses, Romantic literature of the
nineteenth century conveys the image of the thief who steals to avoid starving. Its arche­
type is the main character of Victor Hugo’s Les Misérables, who is punished for stealing
bread. At that time, poverty was so extreme that the houses of the poor seldom included
suitable targets for theft.5 Hence, authors like Beccaria (2009 [1764]) seem to have taken
for granted that the rich were the victims of theft, while thieves came mainly from the de­
prived classes.6 Indeed, the link between poverty and property offenses seems profoundly
engraved in the collective consciousness. This could explain why the basic version of
anomie theory—a concept developed by (p. 64) Merton (1938) during the Great Depres­
sion based on the distance between economic goals and the legitimate means to achieve
them—is intuitively perceived as plausible by the public, even if the empirical evidence on
it is still a matter of discussion among researchers.
In highly industrialized Western countries, the situation started to change in the second
half of the twentieth Century with the development of a market economy, the generaliza­
tion of access to consumer goods, and the miniaturization of electronic devices. At that
moment, households started to be filled with suitable targets for theft. At the same time,
the development of alternative crime measures, particularly crime victim surveys, helped
researchers better understand the evolution of property crime in such a context and led
to the development of new theories such as the lifestyle (Hindelang, Gottfredson, & Garo­
falo 1978) and routine activities (Cohen & Felson 1979) theories. The problem is that re­
searchers still have not found reliable indicators of theft for the previous centuries and
therefore have been unable to establish consistent long-term crime trends. As the same is
true for white-collar crime, researchers specializing in long-term crime trends have con­
centrated their efforts on the study of the evolution of violent offenses.
The World Health Organization (WHO) defines violence as “the intentional use of physical
force or power, threatened or actual, against oneself, another person, or against a group
or community, that either results in or has a high likelihood of resulting in injury, death,
psychological harm, maldevelopment or deprivation” (WHO 2002, p. 4). From this defini­
tion, the WHO devised a typology of three broad categories of violent acts: self-directed
violence (which includes self-abuse and suicide); interpersonal violence (which includes
family and intimate partner violence and community violence); and collective violence
(subdivided into social, political, and economic violence).
Durkheim, one of the founding fathers of sociology, dedicated a full book to the study of
suicide (1897 [2007]), but he also covered some forms of interpersonal violence. Probably
because in some countries suicide was considered an offense until quite recently—in Eng­
land and Wales, for example, it was decriminalized only by the Suicide Act of 1961—crimi­
nologists continued to pay attention to it during the twentieth century. However, their
main focus of research has been the evolution of interpersonal violence. Initially they
principally studied community violence, defined as “violence between individuals who are
unrelated, and who may or may not know each other, generally taking place outside the
Page 8 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
home” (WHO 2002, p. 5). However, since the last quarter of the twentieth century, family
and intimate partner violence has also been the object of many studies.
Once more, it is the operationalization of the concept of violence that complicates the pic­
ture. First of all, there is currently a general consensus that the perception of what is a
violent act has changed throughout the centuries. In particular, industrialized Western so­
cieties have become more sensitive to violence. To name but one example, physical pun­
ishment of children by their parents was tolerated without much discussion until the sec­
ond half of the twentieth century. This means that an offense such as assault has under­
gone a lot of modifications in its definition. Rape has also been (p. 65) redefined, and rob­
bery, as we have seen before, is affected by the availability of suitable targets.
As a consequence, in order to operationalize violence, most studies concentrate on homi­
cide, which is usually defined as the intentional killing of a person. With this definition,
this behavior is probably the only one whose interdiction seems stable across time, even
if exceptions can be found under specific circumstances such as war. The question then
becomes: Is homicide a good proxy for violent offenses?
B. Homicide as a Proxy for Violent Offenses
Every year, the publication of crime statistics by Eurostat or the Council of Europe is fol­
lowed by a series of press articles that compare figures across countries. Generally, such
comparisons are performed in an inadequate way and produce misleading conclusions.
That is why serious newspapers usually add comments by experts that warn the reader
about the risks of such comparisons.7 The idea of progress is seriously undermined when
one reads De Candolle (1987 [1832]) and finds out that the same debates were taking
place almost 200 years ago.8 A classic mistake, for example, is to compare the total num­
ber of recorded offenses or convictions imposed, as these are the worst indicators for
such comparisons because they amplify the problems linked to the recording of each par­
ticular type of crime. Already in the 1820s, Quételet (1829) had mentioned the debate
generated by the fact that the number of persons accused of crimes in England was six
times higher than in France, and had explained that the comparison was inadequate be­
cause data for England included not only felonies (crimes) but also misdemeanors (af­
faires correctionnelles). In order to solve this problem, he was probably the first re­
searcher to suggest comparing similar crimes—in particular, murder (Quételet 1829, p.
25).9
A century and a half later, Gurr (1981) took into account the results of the first victimiza­
tion surveys—which showed that the correspondence between police and victimization
data increased with the seriousness of the offense (Skogan 1976)—to highlight the rele­
vance of murder for the study of trends in interpersonal violence. If the more serious
crimes are the ones that are most often reported to the police, then “murder is the most
accurately recorded violent crime” (Gurr 1981, p. 298). Similarly, analyzing the validity of
conviction statistics as measures of homicide, Gurr concluded that “because homicide
usually has been committed by people known to the victim (as evinced by many micros­
Page 9 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
tudies) and in Western societies usually has attracted close official attention, the slippage
between act and court record is probably less for homicide than other crimes against per­
sons” (1981, p. 299).
Hence, Gurr (1981) and later Eisner (2003) and Pinker (2011) used homicide trends as a
proxy for trends in interpersonal violence. The fact that homicide and nonlethal violent of­
fenses followed similar trends in English-speaking countries during the second half of the
twentieth century and into the 2000s (Tonry 2014) reinforced that use. However, in conti­
nental Western Europe, assault and homicide had opposite (p. 66) trends—with assault in­
creasing and homicide decreasing—from the 1990s to the mid-2000s (Aebi & Linde 2010;
Aebi & Linde 2012). Tonry (2014, p. 39) suggests that the difference could be artificial,
reflecting “major shifts in cultural thresholds of tolerance of violence.” According to him,
“nonlethal violence should be probabilistically associated with lethal violence in the same
way that traffic deaths are with vehicle miles driven. Both sets of outcomes are products
of the law of large numbers. … Deaths should be probabilistically associated with the to­
tal number of potentially violent incidents” (Tonry 2014, p. 39).
Nevertheless, there is no definitive empirical evidence about such a relationship in West­
ern continental Europe, and the situation is more complicated when the sample includes
countries from other regions of the world. Thus, comparing fifty-eight countries around
the world using the International Crime Victim Survey (ICVS), the United Nations Office
on Drugs and Crime (UNODC) (2011) found that, in general, many of the countries with
high homicide rates also showed high robbery rates. Taking into account that ICVS rates
are not affected by the validity and reliability problems of official statistics, and that
homicide is the offense most effectively recorded by the latter, the authors concluded that
homicide can be considered “a reasonable proxy for violent crime in general” (UNODC
2011, p. 15). However, there is no association between homicide and other violent offens­
es included in the ICVS such as assault (including threats) and rape. Consequently, the
UNODC also recognizes this significant variation across nations: “There are countries in
which there is an abundance of violent crime that does not result in homicide and others
where homicide appears high in comparison to general levels of non-lethal vio­
lence” (2011, p. 15).
Hence, recent data suggest that the correlation between homicide and other violent of­
fenses varies across countries. The characteristics of that correlation during previous pe­
riods of the history of humankind remain unknown. The reason is that although one can
find indirect indicators of homicide for different periods (Pinker 2011), it is almost impos­
sible to find them for crimes such as assault, rape, or robbery—which, incidentally, is con­
sidered a property offense in civil law countries. In sum, there is a tendency among re­
searchers to consider homicide as a reasonable proxy for violent offenses, even if the em­
pirical evidence for that correlation is not conclusive.
Page 10 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
IV. Measuring Long-Term Crime Trends
A. Sources for Research on Long-Term Crime Trends
Studies covering the current level of crime or its short-term evolution are mostly based
on criminal justice statistics—which correspond to those collected by police, prosecution,
court, and prison authorities and are usually known as “official” statistics—victimization
surveys,10 and, for youth samples, self-reported delinquency studies. The limitations of
these sources have been thoroughly studied, and there is consensus (p. 67) among re­
searchers that while none of them produces an accurate picture of the levels and trends
in crime, their combination can increase our understanding of them. Other statistics and
specific surveys can also be useful. In particular, public health data are often considered
the most reliable source for data on homicide victimization (LaFree 2005) and are part of
the mortality statistics that countries elaborate according to the International Classifica­
tion of Diseases developed by the WHO.
The situation is harder to solve for studies covering long-term trends because most of the
measures of crime mentioned previously are relatively new. Self-reported delinquency
surveys did not become a usual measure of crime until the second half of the twentieth
century, and victimization surveys only reached wide usage in the last quarter of it. In the
case of police statistics for Western nations, “rarely do the records antedate the establish­
ment of modern police forces in the mid to late nineteenth century” (Gurr 1981, p. 298).
For example, in England and Wales, police data—in the form of indictable offenses report­
ed to the police—have only been available since 1857, when the country started publish­
ing comprehensive national criminal statistics (Maguire 2007, p. 244). Prison statistics
have existed since the nineteenth century, but, as they mainly reflect the levels of punish­
ment within a society, their relationship with crime trends is only indirect. For example,
studies about the long-term evolution of prison populations in Belgium (Vanneste 2003)
and Italy (Melossi 2009) have usually found a relationship between trends in homicide
and trends in the prison population, even if they were inspired by the work of Rusche and
Kirchheimer (1939) and were mainly testing the relationship between imprisonment and
unemployment. The exceptions to this scarcity of indicators are Sweden and Finland
(which was a part of Sweden until 1809), which disposed of mortality statistics since the
middle of the eighteenth century.
In sum, in Western European countries, conviction statistics are the main official source
of crime data for research covering trends during the last two centuries. As we men­
tioned in the introduction, conviction statistics started to be published in some European
countries in the late 1820s.11 Indeed, such publications usually combined what are nowa­
days called “prosecution statistics” with conviction statistics. That is why Gurr (1981, p.
299) states that “most studies of nineteenth-century crime, and virtually all earlier ones,
use data on indictments/committals or convictions.”
Page 11 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
For all these reasons, researchers interested in long-term trends keep searching for alter­
native measures of crime, and one can say that they are particularly imaginative. In a re­
view of the literature, Rousseaux (1993) identifies studies conducted on the basis of the
sentences pronounced by the courts of specific jurisdictions, medical sources such as
coroners’ records of homicides and death registration data in England and Wales, and the
minutes taken when a dead body was found in Amsterdam and Normandy. For non-state
societies, the two main sources of data for homicides have been the records of ethnogra­
phers—who registered the deaths that occurred in the societies they were studying—and
“forensic archaeologists, who sift through burial sites or museum collections with an eye
for signs of foul play … such as … a spearhead or arrowhead embedded in a bone …
bashed-in skulls, cut marks from stone tools on skulls or limbs, (p. 68) and parry fractures
on ulnar bones” (Pinker 2011, p. 48). These sources allow the production of rough esti­
mates of the percentage of deaths caused by violence.
It is also interesting to note here that, among its many privileges, the aristocracy also his­
torically benefited from detailed records of its demography. Hence, using data from
Hollingsworth (1965), Clark (2007, p. 122) has estimated the net reproduction rate and
the life expectancy for males at birth and at age twenty for English aristocrats (kings and
dukes), as well as the fractions of death from violence among them, which decreased
from 26 percent in the period 1330–1479 to 4 percent in the period 1780–1829.
Finally, trying to measure the evolution of intolerance to violence, researchers have used
indicators such as the abolition of judicial torture, which took place from the 1750s to the
1850s in most continental countries, although it was first abolished in England in 1640
and in Scotland in 1708 (Hunt 2007; Mannix 1964). Nevertheless, there have been set­
backs in this development. For example, on January 20, 2015, U.S. president Barack Oba­
ma stated in his State of the Union address before Congress: “As Americans, we respect
human dignity, even when we’re threatened, which is why I have prohibited torture.”12
Also in the United States, another indicator of increased sensitivity to violence used by re­
searchers is the decrease in execution rates since the beginning of the seventeenth centu­
ry (Pinker 2011, pp. 151). In Europe, one of the conditions for membership in the Council
of Europe—which currently counts forty-seven member-states—is to have abolished the
death penalty or at least introduced a moratorium on its application. However, once more
we must mention that the idea of a civilizing process, which includes the abolition of tor­
ture, has been seriously damaged by the clearly established fact that torture has been
reintroduced during colonial wars, national liberation wars, dictatorship regimes, and the
contemporary combat against terrorism. And, of course, the validity and reliability of all
of the abovementioned measures of crime and violence remain a matter of endless discus­
sion, as we will briefly see in the next section.
Page 12 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
B. The Validity of the Sources for Research on Long-Term Crime
Trends
None of the available sources for research on long-term crime trends is perfectly valid. In
this section we summarize the problems related to the use of conviction statistics, which
provide the longest historical series of crime data. However, most of the problems related
to the use of conviction statistics are similar to the ones found when using other types of
criminal justice statistics (see, e.g., Aebi 2010).
As we have seen, although some criminologists tend to consider Kitsuse and Cicourel
(1963) the discoverers of both the dark figure and the limitations of criminal statistics
(Ouimet & Tessier-Jasmin 2009), the fact is that both these concepts were identified by
the first researchers, and in particular by Alphonse De Candolle (1987 [1830]; 1987
[1832]), in the early 1830s, immediately after the publication of the French national judi­
ciary statistics. De Candolle (1987 [1830]) mentioned that such statistics did not include
the offenses that had not been discovered by their victims and those that the victims
decided not to report. As Beirne (1993, p. 105) points out, this is probably the first
mention of the dark figure of crime. De Candolle (1987 [1830]) also noted that conviction
(p. 69)
statistics based on the number of persons convicted do not include those cases in which
the author of the crime is unknown of those in which the legal procedure does not lead to
a sentence by a court. He identified some of the methodological problems related to the
study of crime trends, mentioning that sudden changes in them could be due to changes
in prosecution activity (De Candolle 1987 [1830]). Similarly, he rejected the idea pro­
posed by Quételet (1831) of a constant proportion between the total number of offenses
committed and the number of offenses known and tried by a judicial authority, stating
that such a proportion could vary across time, hence affecting crime trends (De Candolle
1987 [1832]). He was also reluctant to engage in cross-national comparisons because the
number of unknown crimes varied from one country to another (De Candolle 1987 [1830])
and because there were important differences across national criminal justice systems
(De Candolle 1987 [1832]). In that context, he paid particular attention to the presence or
absence of a jury, which he considered the main cause of mistakes in comparisons. He al­
so identified varying legal definitions of offenses and procedural systems as major
sources of distortions. Finally, he suggested that if cross-national comparisons were to be
made, they should be based on the number of accused persons instead of on the number
of sentenced persons (De Candolle 1987 [1832]). The reason was that many offenses
known to the prosecution authorities do not lead to a conviction.
As can be seen, De Candolle anticipated by 100 years the so-called Sellin’s dictum, which
states that “the value of a crime rate for index purposes decreases as the distance from
the crime itself in terms of procedure increases” (Sellin 1931, p. 346, emphasis in the
original). If the validity of a crime measure is defined as its capacity to measure efficiently
the phenomenon under study—that is, crime—this means that conviction statistics are
less valid than police statistics because the latter are closer to the offense itself. On the
other hand, if the reliability of a crime measure is defined as its capacity to provide mea­
surements that are intersubjective and reproducible—that is, to reach the same result in­
Page 13 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
dependent of the person who manipulates the instrument—conviction statistics can be
seen as more reliable than police statistics. The reason is that conviction statistics are
based on a decision made by a judge that relies on much more information than was
available to the police officers who recorded the offense. Also, if different judges arrive at
different sentences in similar cases, a superior court will intervene in order to unify the
jurisprudence.13 Thus, many authors consider conviction statistics to be a useful indicator
of crime trends (von Hofer & Lappi-Seppala 2014).
The problems related to the counting unit of conviction statistics identified by De Can­
dolle are still a matter of concern. Thus, when conducting cross-national comparisons, it
is essential to identify whether the statistics of the countries being compared are based
on convictions or on persons convicted, and whether they apply a principal offense rule
implying that an offender convicted for several offenses is counted only once and for the
most serious offense. Another source of differences across countries relates to the point
at which data are collected for the statistics, as some countries (p. 70) record court deci­
sions before any appeal is made, while others record them after the appeals have been
exhausted (Aebi et al. 2010, pp. 165–66).
As far as divergences in the definition of offenses are concerned, intentional homicide
provides a good example. It has been mentioned before that this crime is often used for
international comparisons because it is supposed to be defined in a similar way across na­
tions. However, there are several sources of distortion, including the way in which this
crime is distinguished from negligent homicide and manslaughter, and the more subtle
distinctions that can exist among intentional homicide, murder, assault leading to death,
and nonnegligent manslaughter. These differences can be even more important in a his­
torical perspective. For example, in France during the Ancien Régime—from the fifteenth
century to the start of the French Revolution in 1789—the intention to kill was not taken
into account, and each act capable of causing death was considered a homicide
(Rousseaux 1993). Finally, the way in which attempted homicides are recorded can also
introduce major differences. For example, across European nations, the percentage of at­
tempted homicides recorded by the police can vary from 10 percent to 90 percent of the
total recorded homicides (Aebi 2010). In the case of conviction statistics, the influence of
the ratio between completed and attempted offenses on cross-national comparisons is
less well known because most countries provide an overall total of persons convicted for
murder, which include attempts at murder, but are unable to indicate the percentage of
the total the latter comprises.
In sum, multiple problems can affect the comparability of crime measures, and re­
searchers interested in comparative criminology must be aware that they will spend a
good part of their time trying to find ways of dealing with such problems. Whenever it is
possible, a comparison of different sources for the same phenomenon is a good starting
point to identify potential complications and improve time series. In Sweden, for example,
von Hofer (1990) compared homicides during the nineteenth and twentieth centuries ac­
cording to mortality statistics—available since the 1750s—and according to conviction
statistics—available since the 1830s. He found that even if the statistics were based on
Page 14 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
different counting units—victims of homicide in the case of mortality statistics, and per­
sons sentenced for homicide in the case of conviction statistics—the trends were quite
similar. In similar cases, Monkkonen (2001) recommends comparing both series using the
capture-recapture technique of the Chandra-Sekar-Deming method, which can be a useful
tool for identifying homicides omitted or unrecorded in the original sources.
V. Pitfalls of Theoretical Interpretations of
Crime Trends
Once the data have been found, the researcher faces the challenge of establishing appro­
priate hypotheses and interpretations of the trends that can be established. Some of
these challenges are similar to those usually found in criminological research, but (p. 71)
others are specific or take a particular turn when applied to the study of crime trends.
This section focuses on four of these specific issues: the problem of the denominator,
chronocentrism, parochialism, and causationism.
A. The Problem of the Denominator
It is well known that absolute numbers are useless for comparative research. For exam­
ple, the number of thefts in a given city does not provide any information on the number
of potential targets for such thefts, security measures, population, or geographic charac­
teristics of the city. An additional problem for comparative criminologists is that all these
variables fluctuate across time and space. Homicides are usually presented as rates per
100,000 inhabitants, but the population of cities and countries during remote periods of
time is extremely difficult to estimate. For example, in the case of the rates of murders in
the United States before the first census of 1790, not only is it difficult to establish the
numerator (i.e., the number of murders), but “the denominator, too, is often no better
than an educated guess” (Lane 2011). In small towns, changes in the denominator can
drastically modify rates, while in big cities it is sometimes difficult to establish whether
the number of homicides recorded includes those committed in adjacent regions—for
which authors of such crimes may have been judged in the closest capital—whose esti­
mated population is not taken into account in the computation of the rates. Hence,
Monkkonen (2001) posits that homicide rates in medieval cities, such as those calculated
by Hammer (1978) for Oxford, could be highly overestimated. In order to solve that prob­
lem, he suggests following the example of demographers by using the capture-recapture
method to estimate the population (i.e., the denominator) of each place studied. The
structure of the population must also be taken into account in the computation because,
historically, the rate of homicides has been related to the percentage of young males in
the general population, and this percentage has changed widely throughout history under
the influence of variations in infant mortality, waves of immigration—mostly composed of
young males—and infant mortality. Thus, Monkkonen (2001) has shown how the use of
age-standardized homicide rates can change the picture of homicide in several cities in
the 1880s and early 1890s.
Page 15 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
B. Chronocentrism
The concept of chronocentrism (from the Greek chrono, time, and kentron, center) has
been applied to criminology by Rock (2005), who took it from (Morson 1996) and defined
it as “the unsubstantiated, often uninspected, almost certainly untenable but powerful
doctrine that what is current must somehow be superior to what went before” (Rock
2005, p. 474). Although this author uses the concept to show how the younger genera­
tions of criminologists tend to forget what was written by the previous generations, it is
also possible to apply it in a more general sense. (p. 72) In that perspective, quite often
the current generation considers that they are at the end of some sort of evolutionary
process. For example, in The End of History and the Last Man, Fukuyama (1992)
concludes that liberal democracies are the final form of government and represent the fi­
nal point in the sociocultural evolution of humanity. Even if the author used his title with
some irony, and probably a good sense of marketing, there is no doubt that, twenty years
later, history continues its course and the map of the world continues to change. In the
case of crime trends, researchers working in the 1980s talked about a “distended Ushaped curve” to explain the evolution of violence from the early modern period to the
1970s (Gurr 1981, p. 296) because they could not anticipate the future evolution of it.
Thirty years later, the increase in violence from the 1960s to the 1990s can be seen as an
exception to a long-lasting decreasing trend. Thus, Pinker (2011, p. 106), who applies
Elias’s (2000 [1939]) idea of a civilizing process to explain such trends, talks about “decivilization” in the 1960s and “re-civilization” in the 1990s. When the next upward trend
arrives, researchers will probably start to think about a curvilinear evolution of violence
since the Second World War.
C. Parochialism
The online Oxford British and World English Dictionary defines parochialism as “a limited
or narrow outlook, especially focused on a local area; narrow-mindedness.” In the context
of research on crime trends, parochialism can assume two forms. The first one consists of
analyzing trends in one country and proposing local explanations for them without taking
into account what is going on in similar countries. Thus, the researcher loses the opportu­
nity of testing whether there are common trends that would require general instead of lo­
cal explanations. The second form, which can also be seen as a light form of ethnocen­
trism, consists of extrapolating the trends found in one country to other countries without
studying the latter.
Many of the explanations given to the crime drop observed in the United States from
1992 onward can be seen as parochial—according to the first form of parochialism pre­
sented in the prior paragraph—because they focus on local explanations that could not be
applied to other countries experiencing similar trends in crime (Eisner 2008; Tonry 2014).
A well-known illustration of this pitfall can be seen in the presentation of the first edition
of The Crime Drop in America, a book by Blumstein and Wallman (2000), at the American
Society of Criminology conference of 2000 in San Francisco. The session was summarized
by Young (2004), who mentions that “the first question, from a Canadian woman, was
Page 16 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
something of a revelation. She pointed out, ironically, how Canadians were supposed to
be condemned to culturally lag behind their American cousins, but that they too had had
a drop in violence, despite the fact that they had not experienced such a period of rapid
prison expansion, that zero-tolerance policing was not de rigueur and that Canada had
only a small problem of crack-cocaine” (Young 2004, pp. 27–28). The “Canadian woman”
was the well-known criminologist Rosemary (p. 73) Gartner (Laub 2001), and her remarks
were taken into account for the second edition of the book (Blumstein & Wallman 2006).
The second type of parochialism is often a consequence of the fact that long-term crime
data are currently available for only a small group of highly industrialized Western coun­
tries. Thus, the trends observed in these countries are sometimes extrapolated to other
regions of the world for which data are not available. This approach takes a more subtle
form when the classification of countries into nonindustrialized (developing), in transi­
tion, and industrialized—even if many of the latter have relocated their industries in off­
shore countries—is seen as a sort of ordinal and longitudinal scale. Then it is inferred that
the current situation in nonindustrialized nations corresponds to that experienced by in­
dustrialized countries 100 or 200 years ago, and it is supposed that the former will follow
the same evolution of the latter, not only in terms of industrialization, but also in terms of
crime trends. For example, according to LaFree (2005, p. 192), “an important underlying
assumption of contemporary modernization theorists … is that nations evolve through
similar developmental stages, as lesser-developed nations gradually adopt the character­
istics of more developed nations.” This idea was first developed by Shelley (1981), who
later modified her position in consideration of the fact that industrialization and urbaniza­
tion are not always accompanied by political and social modernization (Howard, New­
man, & Pridemore 2000; Shelley 1986). In practice, this idea remains active, and quite of­
ten it is suggested that the high rates of crime observed in nonindustrialized nations will
tend to decline with modernization. Incidentally, the same suggestion can be made when
crime trends are explained through the theory of the civilizing process (Elias 2000
[1939]).
D. Causationism
Causationism is the “doctrine or theory that every event is the result of a prior and ade­
quate cause.”14 Criminologists, sociologists, economists, and other social scientists regu­
larly show their inability to make accurate predictions of future macro-level trends in
their disciplines. Instead, quite often they act as historians, concentrating on previous
trends and providing different causal explanations for them. Thus, in the same way that
economists did not predict the financial crisis of 2008 but have provided several explana­
tions for it, criminologists did not predict the crime drop in the United States but have
proposed, and continue to propose, different explanations for it. The latter are essentially
causal explanations. To name but a few, the drop in crime in the United States has been
partially attributed to an increase in imprisonment (Levitt 1996), the legalization of abor­
tion (Donohue & Levitt 2001), an increase in the number of police officers (Levitt 2004),
Page 17 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
the end of the crack epidemic (Blumstein & Rosenfeld 1998), and changes in demograph­
ics and economic opportunities (Blumstein & Wallman 2006).
However, even when trying to explain the past, social scientists do not always reach con­
sensus. Thus, while some authors continue to quote the legalization of abortion as one of
the causes of the homicide drop, others consider such an explanation to not be (p. 74) ap­
propriate (Joyce 2004). Both groups show correlations between homicide, abortion, and
other variables, but none can definitely prove that they are completely right and the oth­
ers are completely wrong. That is why the reader usually chooses the explanation that
better fits her or his Weltanschauung, or the one that is presented in the most convincing
way from a rhetorical point of view.
We have reached here the main inner limitation of social sciences: they leave a significant
margin of interpretation that, unfortunately, is sometimes filled with ideology. That is why
there are no real scientific revolutions in the social sciences. Social theories are seldom
falsified and tend to cumulate. Thomas S. Kuhn was fully aware of that fact, and, already
in the preface of the book in which he introduced the concept of scientific revolutions, he
stated that he was “struck by the number and extent of the overt disagreements between
social scientists about the nature of legitimate scientific problems and methods” and “the
controversies over fundamentals that today often seem endemic among, say, psycholo­
gists or sociologists” (Kuhn 1970, p. vii).
In the case of crime trends, this paradox has been well summarized by Hanns von Hofer
(2012), who was intrigued by the number and diversity of explanations proposed for the
evolution of crime in Europe after the Second Word War. In particular, he wondered
whether it was really possible to talk about “crime trends” (and consequently to propose
substantial explanations for such trends), or whether we were facing a “random walk” (in
other words, if the evolution observed is purely accidental). According to him, “it be­
comes clear that yearly crime data from the post-war period (presently about 60 observa­
tions) in many instances cannot form a robust empirical basis to decide whether we ob­
serve causal trends or random patterns in the data. Many criminological time series stud­
ies comprise an even shorter time period” (p. 86).
VI. Theoretical Explanations of Crime Trends
Two main theories have usually been used to explain long-term crime trends: the
Durkheimian modernization perspective and civilization theory. For shorter periods of
time, particularly for crime trends since the Second World War, the ecological opportunity
perspective is also quoted quite frequently. Finally, the Marxian world system perspective
has mainly been used to explain cross-sectional differences across nations or the changes
that take place inside a country during the transition toward a market economy. The four
perspectives will be presented here. The reader must be aware that they were developed
ex post facto, that is, that they were produced to explain trends that were already appar­
ent and not to predict future trends. As a consequence, they tend to fit well the trends
they are supposed to explain. For example, civilization theory was developed in the late
Page 18 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
1930s to explain the changes observed in European societies since the Middle Ages; the
Marxian world-system perspective took form in the (p. 75) 1970s and reflects the conflic­
tive period in which it was produced, characterized by the national liberation wars, the
Vietnam War, the dictatorships supported by some core nations across the world, and the
Cold War; the ecological opportunity perspective was developed at the end of the 1970s
to explain the crime trends shown by victimization surveys; and, finally, the Durkheimian
modernization perspective finds its remote origins in the second half of the nineteenth
century, when several researchers were trying to propose explanations for the changes
they observed related to modernity. Curiously enough, we will see that research conduct­
ed in the second half of the twentieth century found that the crime trends that the
Durkheimian modernization theory tried to explain had been calculated incorrectly—in
particular, finding erroneously that old rural societies were more violent than the new
metropolises—and therefore it could be considered the only theory to have been falsified.
However, and probably to the great despair of Karl Popper, the theory was consequently
modified, and since then continues to be one of the most quoted and tested
perspectives.15 Old theories die hard.
A. The Durkheimian Modernization Perspective and Its Derivatives
Most of the cross-national studies conducted by criminologists since the 1970s apply the
Durkheimian modernization perspective to explain variations in levels of crime, particu­
larly homicide, across nations (Neuman & Berger 1988; Nivette 2011). Basically, this per­
spective “suggests rapid societal and structural change to be the primary source of de­
viance within a society” (Kick & LaFree 1985; LaFree 1999; Nivette 2011). As its name in­
dicates, this perspective finds its inspiration in the work of Emile Durkheim (1893
[1997]), who studied the consequences for social cohesion of the transition from the socalled traditional to modern society in France during the second half of the nineteenth
century. Durkheim (1893 [1997]) first focused his approach on the interdependence
among individuals created by the division of labor, considering that such a division could
pacify the relationships among individuals, but could also adopt pathological forms. It is
in this context that he mentioned the anomic division of labor that could lead to behaviors
going against the social rules. Durkheim borrowed the term anomie from Jean-Marie
Guyau, who had coined it from the Greek a-nomos (absence of rules) in 1885 (Boudon &
Bourricaud 1989). Later, in his work on suicide, Durkheim (1897 [2007]) characterized
anomie as a sentiment of alienation that can emerge in periods of transition, when old
values are replaced by new ones. In that context, individuals cannot control their desires,
which are unlimited, and suffer a sickness of infinity, which may lead to suicide. This con­
cept is quite close to the spleen immortalized by Charles Baudelaire in Les Fleurs du mal
(1993 [1857]).
In the 1930s, during the Great Depression in the United States, the concept of anomie
was redefined by Merton (1938) and became less sophisticated than its original formula­
tion. In this form, it started referring to the distance between institutionalized means and
cultural goals, which can be particularly far away from each other in a period of (p. 76)
economic crisis (Merton 1938). Later, Shelley (1981) adapted this theory for use in com­
Page 19 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
parative criminology. Since then, studies finding higher homicide rates in nonindustrial­
ized countries than in the industrialized ones usually justify their findings by saying that
the former are undergoing a process of modernization, suggesting that such rates will de­
crease as industrialization is fully achieved. Thus, nowadays Durkheim is usually inter­
preted as having predicted that crime would rise in periods of transition, but if that is the
case, then one of the most striking results of research on long-term trends in homicide is
that it has proven that Durkheim was wrong (Killias 1991). Western European societies
underwent their major cultural and socioeconomic change in the past 250 years, includ­
ing the rise of industrialization and big cities, the development of the proletariat, and the
spread of science and education, but did not react to such strain with an increase of inter­
personal violence. As Eisner (2008, p. 302) puts it: “If anything, modernity was associated
with decreasing homicide.”
Indeed, the success of the Durkheimian modernization perspective among criminologists
is inversely proportional to its rejection by historians since the 1990s. Sharpe, for exam­
ple, has said:
I have always been a little unhappy with the concept of “modernization,” and it
was never as widely adopted in Britain as it seems to have been in the United
States a decade or two ago: indeed, the English Marxist historian Edward Thomp­
son once referred to modernization theory as “a pseudo-knowledge that has pres­
tige on a few American campuses.” It seems to me that if we accept “moderniza­
tion” as a useful shorthand term, it is another way of stating the main lines of de­
velopment formulated by classical sociology: Marx’s concept of a transition from
feudalism to capitalism, Weber’s notion of the transition from a “traditional” to a
“rational” society, or that other old friend, the shift from Gemeinschaft to
Gesellschaft traced by Tönnies. All of these thinkers (and, one suspects, a number
of others who were influential in the decades around 1900 but who are now more
or less forgotten) were concerned with exploring the problematic of the great
transition from a stable, rural, preindustrial society to an unstable, urban, and in­
dustrial one. (1996, p. 20)
The most well-know derivative of modernization theory in the field of long-term crime
trends is the violence to theft hypothesis developed by Pierre Chaunu and his colleagues
in France in the early 1960s (Boutelet & Chaunu 1962). Studying a sample of criminal
proceedings in Normandy, they found that most of the criminals sentenced in the seven­
teenth century were violent offenders, malnourished and incapable of controlling them­
selves during the summer harvest. On the contrary, in the eighteenth century, most of the
sentenced criminals were reasoning criminals who tried to steal from and deceive others
(Boutelet & Chaunu 1962). These scholars suggested that the delinquency treated by the
courts had shifted from violent offenses to property offenses as a result of changes in
mental structures, self-control, and lifestyles. This hypothesis was quite popular in the
1960s, a period in which many European intellectuals sympathized (p. 77) with a Marxist
analysis of society that considered property the main cause of crime. Nevertheless, the
vast majority of empirical analyses have rejected this hypothesis. According to Johnson
Page 20 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
and Monkkonen (1996b, pp. 6–7), “the popular de la violence au vol (from violence to
theft) ‘modernization’ thesis does not hold up to empirical verification… . Hence the per­
ceived decline in violence over the long term was in fact real and not merely a spurious
correlate of a growing interest in prosecuting property crimes in bourgeois society.
Whereas violence certainly did decrease over the centuries, there is no solid evidence
that property crimes actually increased.”
B. Civilization Theory
Norbert Elias introduced the concept of a civilizing process in 1939, but his work re­
mained almost confidential until the 1960s before becoming a classic at the turn of the
1980s, when he was rediscovered by researchers trying to explain the decline in homi­
cide since the late modern period. In a nutshell, Elias (2000 [1939]) suggested that since
the Middle Ages, Western European societies have experienced a series of changes relat­
ed to the division of labor and the monopolization of the use of violence by the state,
which also led to the development of the modern state. These changes increased the in­
terdependence among individuals and were mirrored by a change in their habitus and an
increase in self-restraint. The process described by Elias fits perfectly well the long-term
decline in interpersonal violence observed by researchers such as Gurr (1981), Eisner
(2003), and Pinker (2011), who consequently have adopted it as their main theoretical
framework.16
In the mid-1990s, Johnson and Monkkonen (1996b) tried to explain why this idea of Elias
had not commonly been accepted during the past fifty years: “Elias’s work has many sub­
stantive implications that historians would have found unacceptable thirty years ago:
first, that control of violent behavior emanated from courts; second, that urban centers
would have more ‘civilized’ behavior; third, that areas where state systems had not yet
penetrated would be more impulsively violent; and, fourth, that, over time, violence would
decline.” Furthermore, they stated, Elias’ work ran contrary to the persuasive argument
“that, with the breakdown of family and community (Gemeinschaft) and the rise of mass
society (Gesellschaft), especially through urbanization, industrialization, and the class
alignment of capitalist societies, crime has increased” (Johnson & Monkkonen 1996b, p.
5). Indeed, until the turn of the 1980s, explanations of crime trends were mostly based on
modernization theory and its derivatives, among which the so-called violence to theft the­
sis (Boutelet & Chaunu 1962) played a major role. It was not until then that Gurr (1981)
introduced the ideas of Elias to the criminological scientific community as an explanation
of long-term crime trends, while Cohen and Felson (1979) proposed the routine activities
approach to explain short-term crime trends.
(p. 78)
C. The Ecological Opportunity Perspective
The ecological opportunity perspective refers mainly to the routine activities approach
developed by Cohen and Felson (1979) and sometimes called opportunity theory (Cohen,
Kluegel, & Land 1981). When it was first proposed by Cohen and Felson (1979), the rou­
tine activities approach suggested that the socioeconomic improvements that took place
Page 21 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
in the United States after the Second World War had led to an increase in crime because
they provided further opportunities to engage in criminal behaviors. Thus, the increase in
the average income of the population meant that more consumer goods were bought and
could be stolen, while technological improvements meant that smaller portable devices
were produced that could easily be stolen. Opportunities for burglary, for example, in­
creased as single households proliferated, females were fully incorporated into the job
market, and increasing household income allowed people to spend more time outside
their homes. Thus there was a decrease in the number of capable guardians who could
protect the houses, which were empty during longer periods of time. The postwar period
therefore saw an increase of suitable targets and a decrease of capable guardians, and
this led potential offenders to find more opportunities to commit crime (Cohen & Felson
1979; Felson 1987). In their original analysis of crime rates in the United States from
1947 to 1974, Cohen and Felson (1979) included homicide, forcible rape, and aggravated
assault as indicators of violent offenses. They suggested that their rise was due to the
growth of the nonhousehold-nonfamily activities that we have just described, which had
increased the risk of personal victimization. In a similar perspective, the latter was ex­
plained by lifestyle theory—which was proposed almost at the same time—which focused
on the increasing amount of time individuals spent in public places, particularly at night
(Hindelang, Gottfredson, & Garofalo 1978). The routine activities and lifestyle theories
have also been used to explain crime trends in Europe, including homicide victimization
from 1960 to 2010 (Aebi & Linde 2014c). However, as far as we know, the ecological op­
portunity perspective has not been applied to explain long-term crime trends, as defined
in this article. The reason may well be that the main indicator of such trends used by re­
searchers are victimization surveys, which were not available before the 1970s. Eisner
(2008) has argued that ecological opportunity theory cannot explain long-term trends in
homicide because several potential drivers of violence in public space, like alcohol con­
sumption and leisure time, were increasing at the beginning of the twentieth century,
while homicide was decreasing. However, it seems an interesting challenge to try to find
alternative indicators for a theory that fits well the evolution of crime since the 1950s.
D. Marxian World System Perspective
The Marxian world system perspective, sometimes called simply conflict theory (LaFree
2005), can be seen as a “synthesis of Marxian criminology and world system
theory” (Neuman & Berger 1988, p. 284). Contrary to the previous theories, it was not
(p. 79) proposed by a single author, but is the result of the works of several and therefore
is not exempt from contradictions. For this reason, LaFree, Curtis, and McDowall (2015)
refer to it using the plural denomination of “conflict perspectives.”
The premise of the Marxian world system perspective is that the uneven expansion of the
market economy has created three categories of countries: core nations, which are indus­
trialized and take advantage of the natural resources of semi-periphery nations—most of
which are their former colonies, rich in natural resources—and periphery nations, which
Page 22 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
are underdeveloped and are at the mercy of the core and periphery nations (Howard,
Newman, & Pridemore 2000)
From a cross-sectional point of view, some of the premises of this perspective can be test­
ed through a comparison of crime in these three types of nations. Such analyses have
shown, for example, that homicide rates are much higher in the peripheral nations of
Central America and Africa than in the core nations of Western Europe (UNODC 2014).
From a longitudinal point of view, which is relevant to the discussion in this chapter, this
theory has mainly been used to explain the consequences of the development of capital­
ism in precapitalist societies and its negative effects on social relationships. However, the
literature on this topic has to face the lack of reliable indicators for establishing longterm trends. Moreover, it has been produced mainly by political scientists, anthropolo­
gists, and sociologists, and, as a consequence, criminological issues are treated only in a
superficial way (Howard, Newman, & Pridemore 2000). An additional inconvenience
comes from the fact that there seems to be no theoretical consensus on the way crime
should develop in the long run, especially as countries achieve their transition from one
economic category to another. For example, in the 2010s, the former Soviet republics that
joined the European Union are considered transition economies, while some Asian and
American nations have joined the category of newly industrialized countries, but conflict
perspectives do not provide a clear hypothesis about the future evolution of crime in
these countries.
Finally, recent empirical analyses of short-term trends in homicide show that, since the
beginning of the 1990s, wealthy nations have shown decreasing trends, while nonwealthy
nations have shown increases (LaFree, Curtis, & McDowall 2015). This increase has af­
fected in particular several Central and Southern American nations (Baumer & Wolff
2014).
Conclusion
Criminologists have been interested in cross-national comparisons of levels and trends in
crime since the beginning of the publication of criminal statistics in the 1820s. Currently,
most of their studies concentrate on trends after the Second World War and often apply
sophisticated statistical models. However, quite often the data available at the interna­
tional level are not as reliable as expected and do not fulfill the requirements for such
analyses. At the same time, there is a limited number of researchers interested in the his­
tory of (p. 80) crime who have been collecting the different indicators available since the
Middle Ages, and sometimes even before that. These researchers come from different
fields, and the data they collect are usually not appropriate for statistical time series
analyses.
On the basis of the data collected by these researchers, there is currently a major consen­
sus that interpersonal violence has declined in western Europe since the late Middle Ages
and in the United States since the seventeenth century, with a few interruptions to this
downward trend. This decrease has been mainly explained as the effect of a civilizing
Page 23 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
process, characterized by the monopolization of violence by the state and the growth of
self-restraint among individuals, which resulted in an increased sensitization to interper­
sonal violence. However, these conclusions are based on a very small set of nations.
Moreover, in order to fully accept them, one has to ignore the periods of war that oc­
curred during this time, as well as the genocides, mass killings, and homicides of civilians
committed by some of these nations in their own countries, their former colonies, and the
places where they were fighting a war. Finally, the long-term evolution of crime in most
regions of the world, such as Asia, Africa, Oceania, Eastern Europe, and Central and
South America, is relatively unknown.
Researchers interested in long-term crime trends are confronted by a series of method­
ological problems. First of all, the definitions of offenses vary across time and space.
Changes in definitions are related to the perception of delinquent behavior by the given
society. In particular, the decrease in interpersonal violence seems related to an in­
creased sensitivity toward such violence by members of the society, which in turn is re­
flected in the enlargement of the definition of violence by the criminal law and an in­
creased enforcement of this law. In addition, data covering several centuries are not easi­
ly available, and whenever absolute numbers are found, they still have to be put in rela­
tion to the general population to which they refer—in order to express them as figures
per 100,000 inhabitants, for example—which is not an easy task. In particular, official
crime statistics are only available for a period of roughly two centuries in Western Euro­
pean countries, while surveys were developed mainly in the second half of the twentieth
century. As a consequence, researchers have searched for alternative sources of informa­
tion. In that context, those who specialize in long-term crime trends have concentrated
their efforts on the study of the evolution of violent offenses, and particularly homicide,
for which it is possible to find indirect measures, such as medical records or findings
from ethnographers and archaeologists. As a result, and quite often without a solid em­
pirical basis, homicide has been used as a proxy for violent offenses. Whether this is real­
ly the case is still a matter of discussion.
Researchers are constantly involved in discussions about the different crime measures
used for the study of trends, and such discussions will probably continue endlessly. The
replication of similar findings with different measures seems to be the best indicator of a
relatively reliable finding. The interpretation of those findings is another delicate issue,
as researchers can easily fall into the traps of chronocentrism, parochialism, and causa­
tionism. Such problems can also affect the theories developed to explain crime trends,
among which the most well known are the Durkheimian modernization perspective, the
Marxian world system perspective, the ecological opportunity (p. 81) perspective, and the
already-mentioned civilization theory. Indeed, one of the main advantages of long-term
crime trends research is that it can prove that some theories and explanations that seem
valid for relatively short periods of time—such as the transition to modernity or the after­
math of the fall of the Soviet Union—are indeed false.
Page 24 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
References
Aebi, M. F. 2010. “Methodological Issues in the Comparison of Police-Recorded Crime
Rates.” In International Handbook of Criminology, ed. S. G. Shoham, P. Knepper, & M.
Kett, 211–27. Boca Raton, London, & New York: CRC Press, Taylor & Francis Group.
Aebi, M. F., B. Aubusson de Cavarlay, G. Barclay, B. Gruszczyńska, S. Harrendorf, M.
Heiskanen, V. Hysi et al. 2010. European Sourcebook of Crime and Criminal Justice
Statistics—2010, 4th ed. Den Haag: Boom Juridische Uitgevers.
Aebi, M. F., & A. Linde. 2010. “Is There a Crime Drop in Western Europe?” European
Journal on Criminal Policy and Research 16 (4): 251–77.
Aebi, M. F., & A. Linde. 2012. “Conviction Statistics as an Indicator of Crime Trends in
Europe from 1990 to 2006.” European Journal on Criminal Policy and Research 18 (1):
103–44.
Aebi, M. F., & A. Linde. 2014a. “Conviction Statistics as Measures of Crime.” In En­
cyclopedia of Quality of Life and Well-Being Research, ed. A. C. Michalos, 1,276–81. New
(p. 83)
York: Springer.
Aebi, M. F., & A. Linde. 2014b. “National Victimization Surveys.” In Encyclopedia of Crim­
inology and Criminal Justice, ed. G. Bruinsma & D. Weisburd, 3,228–42. New York:
Springer.
Aebi, M. F., & A. Linde. 2014c. “The Persistence of Lifestyles: Rates and Correlates of
Homicide in Western Europe from 1960 to 2010.” European Journal of Criminology 11 (5):
552–77.
Aubusson de Cavarlay, B. 1998. De la statistique criminelle apparente à la statistique judi­
ciaire cachée. Déviance et société 22 (2): 155–80.
Baudelaire, C. 1993 [1857]. The Flowers of Evil. Oxford: Oxford University Press.
Baumer, E. P., & K. T. Wolff. 2014. “Exploring the Breadth and Sources of Contemporary
Cross-National Homicide Trends.” Crime and Justice 43: 231–87.
Beccaria, C. 2009 [1764]. On Crimes and Punishments, 5th ed. Trans. and ed. by Graeme
R. Newman & Pietro Marongiu. New Brunswick & London: Transaction.
Beirne, P. 1993. Inventing Criminology: Essays on the Rise of “Homo Criminalis.” Albany:
SUNY Press.
Blumstein, A., & R. Rosenfeld. 1998. “Explaining Recent Trends in U.S. Homicide Rates.”
Journal of Criminal Law and Criminology 88 (4): 1175–1216.
Blumstein, A., & J. Wallman. 2000. The Crime Drop in America. Cambridge & New York:
Cambridge University Press.
Page 25 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
Blumstein, A., & J. Wallman. 2006. The Crime Drop in America, 2nd ed. Cambridge & New
York: Cambridge University Press.
Boudon, R., & F. Bourricaud. 1989. A Critical Dictionary of Sociology. London: Routledge.
Boutelet, B., & P. Chaunu. 1962. “Etude par sondage de la criminalité dans le bailliage du
Pont-de-l’Arche (XVIIe–XVIIIe siècles).” Annales de Normandie 12 (4): 235–62.
Buckle, H. T. 1864. History of Civilization in England, Vol. 1. New York: D. Appleton and
Company.
Champod, C., D. Girardin, L. Lebart, P. Margot, J. Mathyer, N. Quinche, & E. Sapin, E.
2009. Le théâtre du crime: Rodolphe A. Reiss, 1875–1929. Lausanne: Presses polytech­
niques et universitaires romandes.
Clark, G. 2007. A Farewell to Alms: A Brief Economic History of the World. Princeton, NJ,
& Oxford: Princeton University Press.
Cohen, L. E., & M. Felson. 1979. “Social Change and Crime Rate Trends: A Routine Activ­
ity Approach.” American Sociological Review 44 (4): 588–608.
Cohen, L. E., J. R. Kluegel, & K. C. Land. 1981. “Social Inequality and Predatory Criminal
Victimization: An Exposition and Test of a Formal Theory.” American Sociological Review
46 (5): 505–24.
De Candolle, A. 1987 [1830]. “Considérations sur la statistique des délits.” Déviance et
société 11 (4): 352–55.
De Candolle, A. 1987 [1832]. “De la statistique criminelle.” Déviance et société 11 (4):
356–63.
Donohue, J. J. I., & S. D. Levitt. 2001. “The Impact of Legalized Abortion on Crime.” Quar­
terly Journal of Economics 116 (2): 379–420.
Ducpetiaux, É. 1835. Statistique comparée de la Criminalité en France, en Belgique, en
Angleterre et en Allemagne. Belgium: Haumann.
Durkheim, E. 1893 [1997]. The Division of Labor in Society. New York: Simon and Schus­
ter.
Durkheim, E. 1897 [2007]. Suicide: A Study in Sociology. New York: Free Press.
Eisner, M. 2001. “Modernization, Self-Control and Lethal Violence: The Long-Term
Dynamics of European Homicide Rates in Theoretical Perspective.” British Journal of
Criminology 41 (4): 618–38.
(p. 84)
Eisner, M. 2003. “Long-Term Historical Trends in Violent Crime.” Crime and Justice 30:
83–142.
Page 26 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
Eisner, M. 2008. “Modernity Strikes Back? A Historical Perspective on the Latest In­
crease in Interpersonal Violence (1960–1990).” International Journal of Conflict and Vio­
lence 2 (2): 288–316.
Eisner, M. 2014. “From Swords to Words: Does Macro-Level Change in Self-Control Pre­
dict Long-Term Variation in Levels of Homicide?” Crime and Justice 43: 65–134.
Ekelund, E. 1942. “Criminal Statistics: The Volume of Crime.” Journal of Criminal Law
and Criminology 32 (5): 540–47.
Elias, N. 2000 [1939]. The Civilizing Process: Sociogenetic and Psychogenetic Investiga­
tions. Ofxord: Blackwell.
Felson, M. 1987. “Routine Activities and Crime in the Developing Metropolis.” Criminolo­
gy 25 (4): 911–32.
Ferri, E. 1892. L’omicidio-suicidio: Responsabilità giuridica, 3rd ed. Torino, Italy: Fratelli
Bocca Editori.
Ferri, E. 1895. Atlante antropologico-statistico dell’omicidio. Torino, Italy: Fratelli Bocca
Editori.
Foucault, M. 1975. Surveiller et punir: Naissance de la prison. Paris: Gallimard.
Foucault, M. 2009. Security, Territory, Population: Lectures at the Collège de France
1977–1978. Houndmills, UK: Palgrave Macmillan.
Fukuyama, F. 1992. The End of History and the Last Man. New York: Free Press.
Goldberger, A. S., & R. Rosenfeld, eds. 2008. Understanding Crime Trends: Workshop Re­
port. Washington, D.C.: National Academies Press.
Gottfredson, M. R., & T. Hirschi. 1990. A General Theory of Crime. Stanford, CA: Stanford
University Press.
Guerry, A.-M. 1833. Essai sur la statistique morale de la France. Paris: Chez Crochard.
Gurr, T. R. 1981. “Historical Trends in Violent Crime: A Critical Review of the Evidence.”
Crime and Justice 3: 295–353.
Hagan, F. E. 2013. Introduction to Criminology: Theories, Methods, and Criminal Behav­
ior, 8th ed. New York: SAGE.
Hammer, C. I. J. 1978. “Patterns of Homicide in a Medieval University Town: FourteenthCentury Oxford.” Past and Present 78: 3–23.
Hindelang, M. J., M. R. Gottfredson, & J. Garofalo. 1978. Victims of Personal Crime: An
Empirical Foundation for a Theory of Personal Victimization. Cambridge, MA: Ballinger.
Page 27 of 34
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Subscriber: UC – Los Angeles (UCLA); date: 28 January 2020
Long-Term Trends in Crime: Continuity and Change
Hollingsworth, T. H., ed. 1965. The Demography of the British Peerage. London: Popula­
tion Investigation Committee, London School of Economics.
Howard, G. J., G. Newman, & W. A. Pridemore. 2000. “Theory, Method, and Data in Com­
parative Criminology.” Criminal Justice 4: 139–211.
Huisman, W., J. van Erp, G. V. Walle, & J. B…
Purchase answer to see full
attachment

Why Choose Us

• 100% non-plagiarized Papers
• 24/7 /365 Service Available
• Affordable Prices
• Any Paper, Urgency, and Subject
• Will complete your papers in 6 hours
• On-time Delivery
• Money-back and Privacy guarantees
• Unlimited Amendments upon request
• Satisfaction guarantee

How it Works

• Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
• Fill in your paper’s requirements in the "PAPER DETAILS" section.
• Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
• Click “CREATE ACCOUNT & SIGN IN” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page.
• From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.