Crime, Histoire & Sociétés / Crime, History &
Societies
Vol. 6, n°2 | 2002
Varia
Criminalized violence and the process of
civilisation: a reappraisal
Gerd Schwerhoff
Translator: Lukas Hoffmann
Electronic version
URL: http://journals.openedition.org/chs/418
DOI: 10.4000/chs.418
ISSN: 1663-4837
Publisher
Librairie Droz
Printed version
Date of publication: 1 December 2002
Number of pages: 103-126
ISBN: 2-600-00812-8
ISSN: 1422-0857
Electronic reference
Gerd Schwerhoff, « Criminalized violence and the process of civilisation: a reappraisal », Crime, Histoire
& Sociétés / Crime, History & Societies [Online], Vol. 6, n°2 | 2002, Online since 25 February 2009,
connection on 30 April 2019. URL : http://journals.openedition.org/chs/418 ; DOI : 10.4000/chs.418
This text was automatically generated on 30 April 2019.
© Droz
Criminalized violence and the process of civilisation: a reappraisal
Criminalized violence and the process of
civilisation: a reappraisal1
Gerd Schwerhoff
Translation : Lukas Hoffmann
1
The history of crime deals mainly with deviancy and the violation of norms. Nevertheless,
historians of crime try to study the social standard as well. While looking at deviancy,
they hope to shed light on people’s attitudes, for example, towards property, sexuality
and morality or towards the authorities. However, the most intensively debated topic, it
seems, has been for quite some time the issue of violence. We are confronted with
‘violence’ in the court records in a variety of facets: murder and manslaughter, bodily
harm and robbery, infanticide and rape. The level of violent crime is viewed as a central
indicator of the forms and functions of violence in social life. And with the help of
criminal statistics, criminal historians try to measure the degree of violence in a given
society.
2
For the last few years the homicide rate has been regarded as a quantifiable indicator of
the degree of violence in a particular period. The homicide rate in present-day
criminology is the rate which shows the annual average of deaths caused by physical
violence in a population of 100,000. Twenty years ago, scholars concerned with the
history of crime adopted this measurement. In 1983, Lawrence Stone made the
criminological analysis of T.R. Gurr, which had been published two years earlier, known
to the community of historians. Gurr had summarised the results of fragmented studies
on violent crimes in England and projected them on the time scale. Stone developed
Gurr’s analysis of trends even further and came to the frequently cited conclusion «that
medieval English society was twice as violence-prone as early modern English society,
and early modern English society at least five times more violence-prone than
contemporary English society»2. So both argue that a clear, long-term decline of violence
had taken place. Subsequently, a lively debate unfolded whose course cannot be repeated
(in detail) here. Soon, this debate crossed over the borders of England and uncovered
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
1
Criminalized violence and the process of civilisation: a reappraisal
varied new empirical data and led to an exchange on the methodology and the theory of
criminal history.
3
In 1996 a programmatic anthology entitled The Civilization of Crime took stock of the
developments in historical research on crime3. In the introduction, its editors stated that
historians of criminality had been evading the most interesting challenges for a long
time, namely international comparison and analysis of long-term developments.
Recently, however, a surprising mutual agreement could be seen emerging in the field of
criminalized personal violence which seemed to prove that interpersonal physical
violence – violence exercised by the state was not explicitly mentioned here – had
decreased from its level in the Middle Ages. All relevant authors agreed that «the decline
in the level of interpersonal violence took place sometime between the 17th and the 18th
century – a period marked by the rise of state control over the population, but a time
considerably before the great wave of urban-industrial growth and expansion»4. Finally
they concluded that, in connection with this empirical verification of a general trend,
Norbert Elias’ theory of civilisation, the central theses of which are widely accepted, had
gained new credit among historians.
4
In 1939, Norbert Elias published his book, About the Process of Civilization 5. In this major
work, the German sociologist tried to trace the changes in human behaviour from the 12th
to the 18th century. The analysis of table manners using guidebooks on moral discipline
and handbooks on etiquette (Zucht- und Benimmbücher) as the primary sources is the main
topic of his presentation. In this analysis, he describes how nose-blowing and spitting
were subjected to an increasingly restrictive canon of rules. But also other expressions of
emotion were governed by a process of civilisation, as Elias explains in the last chapters
of the first volume of his opus magnum: Apart from the sexual drive, it is above all with
die Angriffslust (that is aggressiveness – in German, ‘the lust for attacking someone’) that
he is concerned here6. According to his description, this drive was also subject to the
fundamental historical trend from ‘external constraint’ to ‘internal constraint’, a shift of
the governing of emotions from the outside to the inside of the person. Elias does not
perceive this increasing internalisation of outer constraints as increasing individual
autonomy, but conversely as the result of a transformation of social constraint into
personal constraint. Other secular processes are intertwined with this development, for
example the tendency for increased functional specialisation in the working process or
the development of a state-monopoly of power.
5
After having been a long-time outsider to the scientific community, Elias has, since the
1970s, progressively won recognition as he author of an important and inspiring classic
study. The renaissance of Elias’ paradigm in historical research on criminality is due
above all to Pieter Spierenburg. Bearing in mind Elias’ theories and his own research on
early modern Amsterdam, the Dutch historian elaborated his interpretations of homicide
rates and further developed them in a creative manner. Recently has he summarised his
point of view in this journal and answered the rhetorical question: «Violence and the
Civilising Process: Does it work?» clearly with a ‘yes’. As he sees it, his perspective is
backed up by a phalanx of historians from the Anglo-Saxon world, whereas a number of
historians in Continental Europe – among them Martin Dinges and myself – take a stand
against the civilisation-theoretical interpretation7.
6
The description of the main lines is very simple. On the one hand, those sceptical of such
a far-reaching interpretation of homicide rates and the critics of the Elias paradigm are
not confined to Continental Europe. In a review of the anthrology just mentioned, the
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
2
Criminalized violence and the process of civilisation: a reappraisal
Dutch historian Herman Roodenburg remarks with surprise that – in contrast to what the
emphatic introduction suggests – only two authors refer to Norbert Elias in detail
(namely, apart from Eva Österberg, Spierenburg himself!)8. No one who has read Jim
Sharpe’s critical remarks in the debate with Stone about the value of homicide rates9 will
be surprised to hear that Sharpe mentions Elias only in passing in his article. On the other
hand, there are enough «Continental» voices who concede that homicide rates – with or
without reference to Elias – are central to furthering knowledge about criminality in the
past and who distinctly criticise any cultural objections10. Hence, the dispute cuts right
through the international family of historians of crime. According to the protagonists of
this dispute, the «civilisation of violence» theory seems to be favoured by the majority of
historians.
7
The following considerations are aimed at questioning some of the central premises and
conclusions of this approach. By no means is this meant to ignite a feud among the
community of historians of crime, particularly as there is common agreement about
many aspects of it. Nevertheless, Pieter Spierenburg rightly claimed that there should be
not only an analysis of Elias’ original paradigm, but also of its creative developments 11.
This is exactly what this article is all about. It focuses exclusively on the topic of ‘physical
violence’ and not on other issues of Elias’ work12. Likewise excluded from the following
analysis is the criminality of the 19th and 20th centuries whose developments would make
an article on their own. In regard to the empirical data used, I am going to deal with the
findings of German historians since their work tends to be unduly neglected in the
international debate. To integrate this research, heretofore inaccessible to those
unfamiliar with the German language, into the discussions of the international research
community could be a rewarding by-product of this article13.
Homicide rates
8
First, the empirical basis: «Was the long-term decline of violence, from the 14 th to the
middle of the 20th century, real?»14. In his most recent article, Spierenburg himself hardly
discusses the data systematically, but deals with only some aspects of their
interpretation. He summarily refers to Manuel Eisner’s re-analysis of European criminal
history research. Taking 60 research papers, the Swiss sociologist distilled 290 estimated
homicide rates recorded in the pre-statistical period15. Neither a comprehensive list nor
an analysis of these data can be presented here, nor is this necessary16. Nevertheless, we
still have to deal with this source material. In doing so, we have to distinguish the late
medieval period (from the 13th to the 15th century) and the early modern period (from the
16th to the 18th century).
9
Extremely high rates for homicide circulate for the 14th century in particular: spectacular
front-runners are, for example, the English university town of Oxford, in the period from
1342 to 1348 with a homicide rate of 110; or the metropolis of Florence, whose rate of 52
during the years from 1352 to 1355 is exceeded by a value of 68 in the period from 1382 to
1385. Similarly high rates can be found in Central Europe during the same time: Freiburg
i. Br., for instance, had a homicide rate of between 60 and 90 in the second half of the 14 th
century; the Moravian town of Olmütz, a rate of 77; and Krakow one of 6417. In the basis of
the autopsy reports, which are viewed as being especially reliable, Spierenburg worked
out a homicide rate of 47 to 59 for Amsterdam during a few years of the 15th century 18.
Significantly lower are the values for some towns in the south and south-east of Germany:
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
3
Criminalized violence and the process of civilisation: a reappraisal
For Basel, a rate between 25 and 30 was calculated for the period from 1376 to 1456;
similar rates were reached for Regensburg (from 1324 to 1350: 25) and Eger (from 1310 to
1390: 30)19. But even lower figures are easily found, especially for the 15th century. In his
recent study of Constance from 1430 to 1472, Peter Schuster finds 16 assaults and
homicides between 1468 and 1472 and thus calculates an average rate of 6.2. In Cologne,
the homicide rate had a value of 10 during the years from 1468 to 1481, according to the
Totenbuch, where deaths caused by violence were registered by the judge20. For Brussels, a
comparable rate of 12 can be estimated in the 15th century21. Finally, there are relatively
low figures for the Hanseatic cities of Wismar (from 1343 to 1428: 15) and Stralsund
(1310-1399: 9)22.
Table 1: Homicide rates in late medieval Europe (= number of deaths caused by violent acts per
100,000 inhabitants)
Regensburg:
1324-1350
25
Nürnberg:
14th century (individual years) 25 - 60
Oxford:
1342-1348
90 - 120
Freiburg i. Br.: Second half of 14 th century
>50
Olmütz:
1353-1389 (individual years)
77
Antwerpen:
1356/7, 1367/8, 1372-87
23
Florenz:
1352-5
152
1380-368
Krakau:
1361-1400
64
Basel:
1376-1456
25-30
Brüssel:
1404-1499
1
Köln:
1468-1481
10.7
Sources: Krakau, Regensburg: Schüssler (1998, pp. 216ff); Nürnberg: Schüssler (1991) p. 122); Oxford:
Hammer (1978, p. 11); Freiburg: Wettmann-Junblut (1997, p. 93); Olmütz: Schüssler (1994, p. 166);
Antwerpen: Schüssler (1998, p. 218); Florenz: Becker (1976, p. 287); Basel: Simon-Muscheid (1991; pp.
30f); Brüssel: Vanhemelryck (1981, p. 105); Köln: Schwerhoff (1991, p. 282).
10
This gives us some figures to play with. The validity of the listed data is often highly
questionable. First of all, they all rely on highly heterogeneous source materials:
Achtregister (lists of escaped and proscribed suspects) are compared with indictments,
lists of executed delinquents with entries in Wundenbüchern (records of violent acts) or
court bills. It is evident that a different subset of criminalized violence is looked into with
each of these partial data. Secondly, the judicial-normative basis in space and time is
extremely variable. That there are no relevant sources for the late-13th century is only an
indicator of their dependency on the respective norms. Originally, homicide was not a
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
4
Criminalized violence and the process of civilisation: a reappraisal
penalized crime. Until far into the early modern period, it was possible to regulate
manslaughter by monetary compensation payable by the culprit to the victim’s family23.
11
In many cases the sources are as heterogeneous as the categories for the various criminal
acts. Did court officials register only mere (affective) homicides or did they also include
(wilful) murders? Did accidents or deaths in the context of feud violence prior to the ages
of larger conflicts and wars find their way into the statistics? And what about lethal
violence in the context of robbery?24. Another central problem is calculating populations.
Usually the number of inhabitants can be estimated only roughly, short-term
demographic changes cannot be taken into consideration. As difficult as such an
undertaking is for the early modern period, attempts to do the same for the Middle Ages
border on a juggling act: Given, for example, in his classic analysis of 13 th-century
England, multiplies the number of Doomesday Books mentioned in the inventory lists of
households from 1086 by five in order to arrive at the total population figure of that time
and multiplies this again by 2.5 to trace back the assumed growth of the population over
the 150 years in between25.
12
Most proponents of the hypothesis of a civilisation of crime saw these and other problems
with the applied methods of quantification and did not hesitate to acknowledge the
difficulties of any such calculations. However, they argued, one should not exaggerate
historical scepticism. Eisner, for instance, countered the criticism with «the socioscientific expectation that, with a sufficiently large number of analyses, the various
sources of error would neutralise each other, at least to such an extent that the rough
structure of secular trends becomes visible». In his view, there were no systematic errors
in his sample26. Another strategy rests on the fragmentary nature of the existing sources
which – in the view of some scholars – only supports the assumption that the resulting
figures are trustworthy minimal figures. Would «more complete» source material not
increase these figures for obvious reasons? Due to these considerations there is a certain
tendency (explicit or implicit) among scholars to favour the highest figures as ‘those
closest to reality’. Even though Spierenburg himself denies, in his most recent article,
that anybody would «advocate a method of always accepting the highest figures
reported», he argued slightly differently in an earlier article: at least «court cases
involving arrested killers» and «record listing all cases with identified killers (including
fugitive suspects) should be taken into consideration only when they are relatively high»
27
. Martin Schüssler, whose works form a central reference for recent studies, interprets
his material similarly. In contrast to Spierenberg (!), he barely touches on preliminary
methodological considerations. A critical evaluation of the source material, for example,
is completely lacking. Thus, his quantitative analyses are highly questionable from a
methodological standpoint. When he calculates a homicide rate of 25 from his Nuremberg
source material, for example, but finds a value of 65 for the year 1392, then it is the latter
which he views as being «the likelier one»28. Furthermore, years for which there is no
source material are excluded in Schüssler’s interpretation due to the assumption that the
data of these years had been poorly registered. Of course, this may be true. But it is also
possible that there were no sanctions during those years or even no deaths to be
registered at all. And he often applies the same method when reproducing the research
results of fellow scholars29.
13
The projection of homicide rates is only one possible source for a systematic biasof the
source material. Population figures, too, are traditionally over- rather than
underestimated30. Furthermore, there are many other problems with the measurement of
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
5
Criminalized violence and the process of civilisation: a reappraisal
«the basic total figure for the local population». Apart from England, which is an early
exception to the rule, we have figures only for medieval towns. Apparently this is the
seed of a systematic bias. Thus Beattie’s figures for the 17th and 18 th centuries, for
example, hint at a significantly higher homicide rate in the urban context (cf. table 2).
Generally speaking, it is quite difficult to ascertain whether such discrepancies between
town and countryside are rooted in the different behaviour of people living in rural areas
compared to those living in towns or if there are institutional reasons for these
discrepancies. The impact of urban courts went far beyond the town wall especially in the
late medieval period. Neither did the culprit and the victim have to originate from the
same area for which the respective judicial court was responsible, nor did the crime
necessarily have to take place within the boundaries of the local jurisdiction. Only rarely
do we learn something about such details as, for example, in the Nuremberg book of
outlaws from the year 1392: here we read about a journeyman, who had been banished for
the rest of his life, «that he had stabbed a woman, who later died at Bamberg, with a
knife»31. It is into this context that I would like to put my remarks about the pseudoobjectivity of the measurement of «manslaughters per 100,000 inhabitants»32. In contrast
to modern small-sized villages, medieval towns of only 5,000 inhabitants could already be
focal points of violence as well as foci of jurisdiction. It remains debatable whether the
projection of small basic units onto this modern measurement of a homicide rate is – in a
statistical or a mathematical sense – validor not: Two manslaughters in a town of 10,000
people would already make a value of 20.
14
What can these figures for the Middle Ages prove? One could interpret them as a trend
curve which shows a dramatic climax in the 14th century, in comparison to the previous
13th century and the following 15 th century. Such an interpretation would be highly
questionable because of the above problems with the analysis of the source material. But
tailoring these data to fit Elias’ paradigm is an equally shaky proposition. The calculation
of homicide rates from the 13th to the 15 th centuries is fraught with so many problems
that a comparison between these figures or even the determination of a trend on this
basis seems to be methodologically inadmissible33. No doubt, the level of violence was
relatively high in the late Middle Ages. But that is all we can safely say for that period.
15
For the early modern times, things are different indeed. The number of sources is
considerably higher and so is the density of the source material. However this does not
mean that the quality of most of the data improves dramatically. Moreover, the number
of relevant analyses is still limited. As far as France is concerned, Spierenburg rightly
complains about the total absence of comparable data. There are no accountable analyses
of serial sources covering a longer period of time for the German-speaking territories
either. Fragmented figures, as those on 17th-century Frankfurt-am-Main, according to
which the number of murders halved during that period and further declined in the 18 th
century, are worth being mentioned explicitly since they are so much the exception to
the rule34.
16
From the 15th century onwards, though, there does exist a new type of sources in some
regions and towns, namely the registers of official body inspections (Totenbeschau) in
cases of sudden death. Such body inspections, as they had been carried out by the
Coroner in England, have already been mentioned for Amsterdam and Cologne as
examples of this type of source material. For the period prior to the 17th century, there
are unfortunately only summary calculations for the Dutch metropolis. According to
these figures, the homicide rate was approximately 28 for the period from 1524 to 1565,
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
6
Criminalized violence and the process of civilisation: a reappraisal
and between 21.1 and 24.2 for the selected years of 1560, 1570, 1580 and 1590 35. For
Cologne, the data can be interpreted in more detail for the time between 1557 and 1620.
Unfortunately there are no records for the following years and decades (cf. table 2).
Conversely, the practice of body inspections in Amsterdam did not resume until 1667, but
for the ensuing period, long-term serial data can be listed. By far the most valuable set of
data are still those figures presented by Cockburn in 1991 for the English county of Kent
on the basis of coroner inquests which cover a period from 1560 well into early modern
times36.
Table 2: Homicide rates in 16th and early 17th century (=number of deaths caused by violent acts,
per 100,000 inhabitants)
Kent
Köln
Amsterdam
1557-63: 13.5 1560-62: 73
1561-81: 3.5 1574-81: 14.5 1566-67: 82
1571-73: 36
1577-80: 46
1582-01: 5.6 1582-01: 9.1
1585: 31
1602-21: 3.9 1602-20: 7.7
Sources: Cockburn (1991, p. 78); Schwerhoff (1991, p. 283); Blastenbrei (1995, p. 71)
Note: The data from Cologne are not mere random samples, as Schüssler (1998), p. 220, wrote; for
this table, the data has been re-organised and newly calculated.
Table 3: Homicide rates in 17th and 18th century (=number of deaths caused by violent acts, per
100,000 inhabitants)
Kent
Surrey
Urban
Amsterdam
Rural
1561-81: 3.5
1582-01: 5.6
1602-21: 3.9
1622-41: 3.8
1642-61: 3.7
1662-81: 4.3 1660-79: 8.1 4.3
1667-1679: 3.5
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
7
Criminalized violence and the process of civilisation: a reappraisal
1682-01: 3.6 1680-99: 5.0 4.7
1693-1709: 10.2
1702-21: 2.7 1700-19: 3.9 2.9
1710-1726: 8.3
1722-41: 1.6 1720-39: 2.8 2.4
1742-61: 2.0 1740-59: 2.0 1.6
1752-1767: 2.5
1762-81: 1.9 1760-79: 1.7 1.1
1768-1783: 3.8
1782-01: 1.8 1780-02: 0.9 0.9
1784-1799: 2.6
1802-21: 1.4
1800-1816: 1.5
1822-41: 0.9
Sources: Cockburn (1991, p. 78); Beattie (1986, p. 108); Spierenburg (1996, p. 83).
17
Though the data presented here show greater validity than those for the medieval period,
considerable problems with their interpretation remain. Many questions have already
been raised by James Cockburn, who – as Spierenburg rightly notes – tends to minimise
the relevance of his data for an analysis of the development of violence or negates such a
development altogether. Up to now his arguments have not been sufficiently discussed.
Insufficiently reflected, for example, is the question of whether there is a connection
between homicide rates and violent offences in general. Apart from some excursions in
this direction, there is no appropriate basis in source material for quantifing physical
injuries and other violent offences37. In order to compensate this lack, it is the explicit or
implicit premise of the civilisation of violence thesis that homicides are a valid indicator
of all forms of criminalized physical violence – if only because it is the only existing
somewhat «hard» indicator left. In fact, such offences can be seen as an extreme pole of a
continuum of unspectacular violent acts running from mere threats and quarrels to
severe physical injury38. This does not mean, however, that there is a continuing stable
relation between the two factors over a long period of time. Most of the people killed by
violent acts in the early modern period died because of their injuries, due to blood loss or
of infections, which they would have no doubt survived in the 20th century. In a
comparison which aims at encompassing several epochs, categories of modern criminal
statistics such as «attempted murder» and «severe bodily injury» have to be added to the
homicides. One could argue here that medicine had not progressed significantly until the
19th century, but there are other relevant factors that need to be taken into
consideration. The accessibility and the use of weapons, for example, seem to be of
central importance – a point which is primarily responsible for the enormous differences
in the violent death statistics between Europe and the US. Still, these figures would not
make us view the United States as being less civilised. The majority of violent deaths in
the 16th and early 17 th centuries, according to James Cockburn’s analysis for Kent, was
caused by thrust weapons, whereas, from the 17th century onwards, the proportion of
firearms and other tools has grown continually39. Does this mean, then, that potential
thrust weapons have increasingly disappeared from daily life? Or are they just no longer
used to settle disputes? These questions illustrate the necessity of a qualitative analysis of
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
8
Criminalized violence and the process of civilisation: a reappraisal
individual violence in the cultural context, which will be dealt with in the third part of
this article.
18
Despite the methodological reservations, historical research, especially that concerned
with pre-modern times, is always faced with a difficult source basis, and it is legitimate to
make the most of quantifying methods. Looking at both the data and the numerous
scattered hints, the impression is substantiated that the level of violence decreases from
the 16th to the 18 th century. Pieter Spierenburg has the merit of having pointed out this
trend. However I would like to qualify this immediately in two respects: First there is no
evidence that this is a long-term process. All that can be proven somewhat empirically is
merely the fact that there has been a change in the levels of violence within a period of
time that can be limited to 200 or 300 years. By the middle of the 18th century at the
latest, the level of lethal violence seems to have levelled out at a comparatively low level
in England and on the Continent40. This does not apply to all parts of Europe, though, not
for example to the Mediterranean area, where a distinct culture of violence prevailed for
a longer period of time41. And this brings us to my second qualification, which is
concerned with the dramatic difference between the graphs for one individual territory
as well as the discrepancies between the respective geographic areas of research. There is
at least the same need for an explanation for those discrepancies as there is one for the
overall trend. Thus, the county of Kent shows a low homicide rate for the 16th century
which is reached by Amsterdam only in the second half of the 18th century. By
comparison, the figures from Cologne are already significantly higher, but the rates
calculated by Peter Blastenbrei for late 16th century Rom seem to be exorbitant. The
privileged observation of a unidirectional development seen through the glasses of the
theory of civilisation tend to obscure such significant differences. Elias’ paradigm is about
to become Elias’ bias – at least in the context of the debate about violence.
The civilising process
19
Here we have arrived at the second pole of the discussion, the theory of civilisation as
such. Pieter Spierenburg pleads for theoretically oriented historical research. According
to his approach, theory and empirical data should always relate to one another without
the empirical data acting blindly or the theory drifting away into unfounded speculation42
. In the main, I agree on this. However the procedure just outlined must not lead to
insecure empirical data being backed by a problematic theory or vice versa, and central
problems being thus obscured. This is exactly the case for the theory of civilisation, its
fields of research and the discussion about violence in particular.
20
Surely some of the aspects brought forward by Spierenburg in the wake of Elias are
basically not contentious. Hence the connection between the development and frequency
of individual violence, on the one hand, and the development of a monopoly of power by
the state, on the other hand, can hardly be dismissed. By the way, the terminology used
by Elias here is less problematic than Spierenburg suggests. Elias talks about the
«monopoly of power» clearly in the sense of a ‘monopolisation of legitimate physical
power’43. In doing so, he follows explicitly the language of Max Weber, who defined the
state as a «political institution which can make use of the monopoly of legitimate physical
constraint»44. Elias talks uninhibitedly of a «monopoly of force»45. Spierenburg’s attempt
to translate this into ‘military monopoly’ is too narrow, even though military and tax
control indeed represent two important aspects of the monopoly of power, according to
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
9
Criminalized violence and the process of civilisation: a reappraisal
Elias. Another important aspect for the history of violence, which Spierenburg touches on
only in passing in this context, is the development of criminal law by the state, with its
severe punishments. In the course of this development an increase in the exercise of state
power was seen. The frequency of executions and other bodily punishments increased46.
From the standpoint of the theory of civilisation, the monopolisation of power by the
state had two opposite consequences: On one side, there was a decrease in interpersonal
violence, an intensified internal pressure exercised by the state on its subjects, and, on
the other, there was an externalisation of power onto the intergovernmental level in the
form of state-building wars47.
21
Our goal should now be to look in more detail at the connection between the
development of state monopoly of power and the changein individual violence. In
following Norbert Elias’ theory of civilisation, however, Spierenburg takes exactly the
opposite route. He refers to a historic «macro-narrative» which seems to make the
empirical data explainable by a plain theoretical explanation. By fitting it into the
occidental process of civilisation, he thus officially ordains a historical trend which makes
a more detailed analysis seem superfluous. This model is, in the end, a reductionist view
in which the frequency of violence is dependent on the degree of affect control. This
affect control is again connected with a slowly increasing development of state monopoly
of power and social interdependencies: «Thus, the long-term decline in homicide resulted
from a taming of aggressive impulses in daily social intercourse, which in its turn was a
function of the rise and growth in power of states in Europe»48.
22
A number of objections to this simple reductionist model can be made. In my view, Elias’
idea of violence and aggression seems problematic. He conceptualises them as if they
were just there naturally, as a proposition of the human being, which can be socially
integrated and abolished only secondarily. From this viewpoint, the changes in the forms
of violence always represent the dependent variable, whereas the social conditions of
society with their twisted structure and the expanding power of the state appear as
anonymous figures responsible for that process. The state apparatus of repression thus
forms a somewhat stable lid for the social pressure cooker, slowly reducing the inevitable
eruptive outbreaks of passions while trying to prevent them altogether. That Elias viewed
violence as a social habit that is learned like any other human behaviour, as Spierenburg
writes, could be surmised from some of Elias theoretic statements. His empirical analysis
of knightly violence in the Middle Ages tells a completely different story, though. It is in
this naïve and prejudiced interpretation of the sources that the various points of
criticism, which I will not repeat here, originate49.
23
By no means is this supposed to be a simple know-it-all criticism of the work of a classic
whose core cannot be shaken by minor, almost inevitable corrections in detail. Like no
other modern thinker in modern times, Elias related empirical data and theory to each
other. He rejected programmatically Weber’s idea of a construction of ideal-types, of
models which can fathom reality by an abstract terminology. His process of civilisation
aims at being both a theory of social processes and their empirical proof. This has farreaching consequences. In Elias’ work there is an unclear relation between empirical data
and theoretical premises. The mention of evidence by the sources is suggestive and
episodic. Hardly ever does he quote in a methodologically controlled and controllable
way. Possible criticism of his analyses can thus be rejected on various levels again and
again, depending on what is needed: empirical objections can be devalued by referring to
the process of civilisation, criticism of the concept as such can be countered with facts.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
10
Criminalized violence and the process of civilisation: a reappraisal
Thus, Elias’ theory is immune to possible attempts at falsification. Spierenburg chooses a
similar strategy when he says that the theory of civilisation is «based on observed past
trends and has no room for evolution» in order to defy ideologically critical reservations 50
. Yet the theory is used again as a «seeing aid» in order to show that there was a
measurably higher degree of violence in the late Middle Ages and that its gradual decline
in the early modern times was part of a secular trend.In my opinion, the borderline
between a precisely fitting alliance of empirical data and theory, on one side, and a
circular reasoning, on the other, has been unduly crossed here.
24
Spierenburg advances many anti-critical arguments with which the older Norbert Elias
impatiently confronted his critics time and again. He counters all criticism against the
ideological character of his theory of civilisation, against its teleological, evolutionistic
main feature, against the transfiguration of modern times and the distorted, cliched
perception of the Middle Ages with conceptual remarks made by Elias: processes of
civilisation had direction but they were blind, unintentional, unplanned processes; they
had no beginning and no end, they went on irregularly and intermittently; there could be
backdrops, periods of de-civilisation and, yes, these processes were even primarily
reversible. But for judging the value of a theory correctly, some general conceptual
remarks are of little help. Instead, their application in practice must be the main criteria.
As inspiring as singular ideas of Elias may be, his whole model for the discussion about
violence is restricting.
25
The continuous change in empirical data and theory can also be observed in the handling
of the French model. To what extent the French way of civilisation can be viewed as the
model case for a process of civilisation through the royal court or to what extent there
are alternative ways of development remain an open question. Spierenburg advocates the
latter point of view and explains that a refinement of habits at the court in no way
necessarily goes hand in hand with a subduing of violence – after all, the royals were still
wearing weapons at court. Their renunciation of violence was based on external
constraints rather than on automatic self-constraint51. For one thing, Spierenburg is
inconsistent with his teacher here: Elias explicitly emphasised that the «affect household
of a human being» had to be viewed as a whole, that various «utterances of urges», such
as hunger, sexual drive or the drive to attack, were inseparably intertwined. This quite
mechanical connection once more casts a bright light on the questionability of Elias’ drive
theory. But the second part of the argument is much more important because it operates
from a theoretical element which is again and again viewed as being central to the theory
of civilisation: the transition from external constraints to self-constraint. Indeed, this is a
very original and highly interesting figure of thought. But how is this development to be
controlled empirically? In regard to table manners, Elias has put forward a number of
sources which actually hint at an internalisation of certain modes of behaviour. But for
the sphere of violence mere arbitrariness seems to govern any analysis. That there were
mere external constraints curbing the exercise of violence in the Middle Ages seems to
have been evident for the sociologist. By using the roughly outlined Dark Ages as a
starting point, he was much better able to contrast the process of civilisation, which
originated in the court. Violence in the Middle Ages is portrayed as being controlled
externally, the avoidance of violence later on, however, was a successful form of selfconstraint. There are no detectable criteria for this historic operationalization. Instead,
an external constraint can be constructed again, if necessary, as Spierenburg’s
argumentation shows, in order to remove inconsistencies from Elias’ paradigm. At other
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
11
Criminalized violence and the process of civilisation: a reappraisal
points of the controversy, the self-constraint argument is omitted completely. In the
process of the devolution of the Soviet Union and during the formation of non-pacifistic
islands within Western metropolises, there was «a weakening of the state monopoly»
which then, of course, was accompanied with a rise in the homicide rate52. In these cases,
a successful internalisation of drives cannot have taken place, otherwise the weakening of
the state’s monopole of power would hardly suffice to explain the new increase in the HR.
It is generally questionable how the curbing of human affects in daily life, on one side,
can go together with an increase of violence in wars and civil wars reaching as far as
genocide in modern times, on the other– at least, if one represents the «drives» of a
human being as a unit, which was what Elias postulated.
Habits and meanings
26
Apart from the macro-historical approach to interpersonal violence, with the help of
homicide rates and the theory of civilisation, the micro-historical perspective mentioned
at the beginning has been successfully established and has produced numerous studies on
the phenomenology of everyday violence. These analyses revolve around «the twin
themes of ritual and honour», as Spierenburg rightly notes. In his view, these were «the
only serious objection» to an application of the theory of civilisation to violence53. Would
the binding of an offender to an all-encompassing code of honour and the strict
ritualization of most duels not implicitly mean a control of affects and thus the exact
opposite of what violence is all about in the theory of civilisation – namely unrestrained
violence? In his discussion of these topics, Spierenburg comes to the conclusion that the
perspective of the theory of civilisation can be combined with the latest studies on
honour and ritual. And he refers to his own attempts, his integration of cultural and
statistical approaches to the analysis of homicide rates in Amsterdam.
27
Let us take a brief look at these attempts. The nucleus of this ideal-model of analysis is a
system «of two related but distinct axes». The poles of the first axis are formed by
«impulsive violence versus planned or ‘rational’ violence», those of the second by «ritual
or expressive violence versus instrumental violence»54. His analysis of a sample from
Amsterdam court cases (the body inspection records do not provide the contextual
evidence needed for an adequate analysis) shows that impulsive violence was quite
dominant up to around 1750, whereas its share declined sharply from the middle of the 18
th century onwards. With this finding, Spierenburg combines another observation, which
is concerned with the victim–offender relation. Impulsive violence in the form of sudden
outbreaks of aggression between strangers or companions was found primarily in
alehouses and on the street. Since approximately 1720, however, he notes an increase of
intrafamilial violence in comparison to the violence against outsiders, «a shift from the
killing of strangers to the killing of intimates»55. In addition, this intrafamilial violence
was to be seen more and more as an expression of long-term tensions, for example
between spouses, rather than as impulsive panic reactions.
28
Spierenburg’s suggestion is undoubtedly the work of a pioneer. Without such attempts at
categorization, a further-reaching analysis of violence will hardly be possible. What we
need is a set of terminological instruments which give the seemingly amorphous
phenomenon of violence a sharper profile. This two-axis model can hardly be the best
solution though. Among other things, Eva Lacour, a psychologist dealing with history,
criticised the comparison of impulsive and rationally planned violence as inadequate and
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
12
Criminalized violence and the process of civilisation: a reappraisal
proposed a three-dimensional analysis instead: motivation (angry or unplanned), form
(ritual/wilful or uncontrolled) and planning (planned or unplanned)56. Both of these two
works57 fail, however, with respect to a methodological problem: the idea of attributing
single cases to the respective poles of the model can hardly be understood – and where
such a categorization is understandable it seems to me very problematic. Lacour, for
example, works with a highly heterogeneous set of «types of offenders», to whose
configuration the reasons for the conflict and the occasion of the conflict, the place of the
crime, offender and victim are used arbitrarily to characterise the respective type. «Fight
for property» is thus juxtaposed with «conflict of honour», «revenge» with «alehouse
conflicts»58. In doing so, the author herself admits that many conflicts over property
could have a strong aspect of honour, for example, or that many types of conflict could be
understood as aggression out of revenge. Her method, which is to select the dominant
motives in a single case, seems arbitrary and exemplifies the difficulties connected with a
solid categorisation of acts of violence on various levels. To put a question mark behind
some of her findings, which themselves seem to question usual commonplaces of
research, is appropriate. Observations such as that violent conflicts were rarely preceded
by threats, that conflicts happened quickly, directly and were uncontrolled become
questionable when the high rate of insults –18.3% for the county of Virneburg, higher
than all acts of violence taken together (13.4%) – are excluded from the analysis of
violence59. Because it is here that those interactions took place which could have led to
physical violence as well, but which did not!
29
The debate about a sensible categorisation of acts of violence has thus just begun and it is
highly welcome. It is therefore especially unfortunate that Spierenburg reads his
observations as evidence for the theory of civilisation and thus jumps to conclusions
where further questions would be more pertinent. In his point of view, impulsive and
ritual violence are combined with a low level of affect control, whereas instrumental and
rational violence can maintain their position in «civilised» forms of society. Here, too,
violence is being conceptualised as if it was a problem of the human nature which needs
to be socially restrained and abolished. From this viewpoint, the changes in forms of
violence are always seen as dependent variables, whereas intertwined social relations and
the growth of state power figure as anonymous causes. Not only is this way of reasoning
conceptually questionable, but it remains also empirically doubtful, because – as we have
seen – long-term trends are hard to detect. Instead we find a shift in the ‘faces of
violence’, which furthermore is quite precisely localisable in time.
30
But Spierenburg objects to this micro-historical perspective that it «simply provides no
alternative theory»60. That is true, if one accredits a ‘theory’ only when it claims a
universal applicability similar to that of the theory of civilisation. Historical macronarratives and theoretical scripts of modernisation are not experiencing a boom at the
moment; in this respect the theory of civilisation is more or less the last theoretical
dinosaur of its kind. But giving up the idea of a general theory about the course of
historical processes does not mean that we have to renounce a theory-oriented
perspective of analysis. Instead of dealing at length with the sociologists’ debate about
violence I would like to look at this discussion only briefly. In his review of research on
violence since 1997, the sociologist, Trutz von Trotha, complains about his colleagues’ low
productivity on violence. According to him, the mainstream of modern empirical
research was in fact not sociology of violence but a sociology of the reasons for violence.
The sociological theory as such, including the most frequently cited classics as Durkheim,
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
13
Criminalized violence and the process of civilisation: a reappraisal
Weber or Simmel, did not have much to offer in this respect. Elias, too, as one of the
«essential reference points in the current theory of violence», was no exception: «In fact,
the nucleus of Elias’ study is not concerned with violence, but with forms and changes in
the control of violence. […] Violence as such instead is understood mostly as an unbridled
«affect», as an impulse and as natural shamelessness about one’s own body and that of
the other». Trotha’s own research project opts for the phenomenological tradition of
sociology. He demands a «thick description» in the sense of Clifford Geertz, an
antireductionist and process-oriented microscopic analysis of violence. By proceeding in
such a way, the «Why?» would be replaced by questions of «What?» and «How?». Still, he
argues that he would be able to find and name sociological basic terms with a high degree
of universal applicability61.
31
The micro-historical analyses of the recent years have already started to tackle this
programmatically postulated task of perceiving violence as «meaningful behaviour»62.
The consequences of such an approach can only be dealt with only briefly here63. I will
start by reconsidering the definition. This may sound strange at first, because physical
violence can be defined quite clearly as «harm done to another person». Instead I would
plead for widening the understanding of the term «bodily harm» for the purposes of
historical analysis, which means in concrete terms that mere threatening gestures such
as the drawing of knifes and also (or even in particular) «verbal violence» should be taken
into consideration as well. Many researchers might view the abolishment of the clear-cut
boundary of «physical violence» as being problematic, and especially in the Englishspeaking world this term may be regarded as being «confusing from a linguistic
standpoint»64. But it is no coincidence that early modern jurisdiction (which could indeed
separate insultsfrom manslaughter) was familiar with the term iniuria realis – a category
into which extremely brutal acts of violence were subsumed65.
32
This was undoubtedly a jurisdictional reflex to social practice. In daily life, insulting
words often marked the beginning of a spiral of violence which could end with
manslaughter. Insults harm one’s honour, which was at least as important as the integrity
of the body. Or perhaps it was even part of this bodily integrity: metaphorical speaking,
this honour could be understood as something like a «second skin» which had to be
defended against violent attacks as much as the biological skin.
33
The definition of violence is thus closely connected with the question of its outward social
form and function. International research has provided us with a «thick description» of
everyday violent behaviour in late medieval and early modern societies which shows
many variations in detail, but matches in outline. The acts of violence committed mostly
by male offenders followed a highly ritualised dramaturgy and can ideal-typically be
described as a process of escalation, which could start with teasing and insults, progress
to threatening gestures – such as the drawing of knives – and assault, and could
culminate in serious injuries and manslaughter. The continuing escalation was not
inevitable, there were a number of ways to de-escalate a tense situation. The argument is
quite convincing that the drawing of knives marked a kind of deadlock and made an exit
from this conflict possible for both parties without either of them losing face.
Accordingly, the urge to defend one’s own honour in the forum of a virtual public was the
driving force behind the acts of violence. On these occasions honour functioned as a kind
of homogenising code which standardized the heterogeneous interests and motives. In
other words: no matter whether conflicts between neighbours, inheritance disputes or
conflicts with fellow members of the guild were at stake, these conflicts have always been
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
14
Criminalized violence and the process of civilisation: a reappraisal
translated in a language of honour and have thus been made negotiable on some level or
have otherwise been settled violently. Whether the protagonists fought with sharp words
or sharp knives does not make a difference, generally speaking66.
34
Such an approach is clearly different from that of the perspective of the theory of
civilisation. Physical violence is no longer conceptualised as a phenomenon which is
socially unwelcome at any place and at any time; instead it becomes describable and
interpretable in its social meaning. Taking the concept of social control as the analytical
reference point, one could say: physical violence in early modern times was not only the
object of this control, but its medium at the same time. In conflicts argued out violently
there was always the fulfilling of social norms at stake as well; such conflicts can be
understood as attempts to sanction breaks with the norm. Of course, these two
dimensions – social control of violence and social control by violence – entertain a
strained relationship. The exercise of violence was sanctioned negatively both by the
authorities and by society in the Middle Ages as well as in early modern times; to bring
about peace, to keep and to re-establish it if necessary, was one of the most treasured
values of these centuries. Hence many of the measures taken aimed at a reconciliation
between the arguing parties or at a reconciliation of the offender with the human
community and with God. Even the judicially sanctioned act of homicide – at least if it
presented itself as a situational and unplanned manslaughter – followed this logic of
functioning and remained within the private sphere of the compensational system far
into early modern times. Shameful punishments were suspended fairly often by acts of
mercy. Throughout this whole period, the regular punishment for physical violence
which did not have a lethal end remained the fine. Physical violence was not criminalized
consistently in pre-modern times, a fact which can be explained by its aforementioned
double-facetted character.
35
This approach goes together with categorisations only if these are not meant as a mere
confirmation of a given theory. Of course, macro-historical factors can and must be
included in a second step. For any study of the influence of demography, the levels of
urbanization and economic development may not seem to be very fashionable, but they
continue to be the necessary tools for the work of the historian. Particular attention must
be paid to the state as a generator of norms, of criminal procedure and as an agent of
sanction. It is important, however, to find the exact mechanisms at work by which state
power influenced the frequency and the shape of violence. Was it really a systematic and
intensified social control by the authorities that helped suppress violence? This «causal
scheme taken from the theory of civilisation …, which in its shortest forms says: more
disciplining social control, hence more methodical self-control, hence less individual
violence» strikes even followers of a modernisation theoretical perspective as being far
too simplistic67. A slightly different light is thrown on this connection by the paradigms of
«judicial standardisation» and the «use of justice». The term «juridification»
(Verrechtlichung), which was actually coined in the analysis of political protests in early
modern times, emphasizes the systematic character of the extension of the judicial
apparatus in the early modern period68. It did not focus on the suppression of violence,
but on the extension of an attractive spectrum of alternatives to resolve conflicts in the
form of a functioning justice system which could be used by a large number of people. In
order to test this hypothesis, a more detailed analysis of the court records would be
necessary. So far there have been only very few serial analyses for Germany; there is no
evidence for an increase of interpersonal violencebetween men of the same social group
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
15
Criminalized violence and the process of civilisation: a reappraisal
at the lower courts of the Fürstbistum Osnabrück since the late 17th century, but the
number of economic conflicts between the representatives of various social groups – as,
for example, between well-to-do peasants and day labourers – have predominated at this
court ever since then69.
36
Such a model, according to which informal mechanisms of social control were replaced by
formal mechanisms, seems to share the mechanistic character of the theory of
civilisation. After all, such informal mechanisms of control still exist today, only – at least
in western societies – violence no longer has a central place. That is why we need to focus
more closely on the interplay between those two levels, formalised social control of the
law and informal control agencies of society, and ask if and how this interplay has
changed over time. It is here that the concept of «the use of justice» sets in, which looks
at how the historical protagonists dealt with the old and the new institutions70. Studies
show how the options of a potential user of these institutions varied and that the decision
between violent do-it-yourself, extra-judicial settlement or a court settlement remained
dependent on the context for a long time.
37
A different explanatory model, which aims at outlining the historical change in the forms
of violence, has been proposed by Joachim Eibach. In his area of research, the imperial
city of Frankfurt in the 18th century, he detected a gradually growing social consent to
resolve conflicts peacefully. As a possible cause for this development, he sees a process of
modernisation towards a bourgeois society (Verbürgerlichung). The new forms of upperclass social and cultural life had an increasingly shaping influence on the behaviour of the
lower classes of urban society. The new ideal of a distinguished morality and
concentrated emotions had undermined the traditional, physically violent culture of
quarrels71. Undoubtedly these observations are inspired by Elias, where a shaping
influence is attributed to the higher classes of society, too. Spierenburg also talks about a
pacification of the élites as a historical development noticeable throughout the early
modern and modern period and about a trend which left violence at the margins of
society. He illustrates this with a highly significant example, namely a comparison
between the inner city of Amsterdam in 1700 and that of Philadelphia in 200072. No doubt
the contrast between the general acceptance of violence beyond class boundaries in early
modern times and their limitation to certain no-go-areas of big American cities, where
only outsiders and losers live, is striking73. But is it really always the upper classes and the
elites that gradually refrain from violence in conflicts? The demonstration of virility and
the exercise of violence were part of aristocratic behaviour well into the early modern
period74. Also urban patricians did not follow this model. Neither in 15th-century
Konstanz nor in 16th-century Augsburg did they attract attention by an ostensible
restraint from violence – quite the contrary75. This is even more noticeable the higher one
goes up the social scale because members of the elite were usually far more able to
protect themselves from legal sanctions due to their powerful social standing. The
absence of a bourgeois upper class, of merchants and rentiers in the criminal statistics
may indeed be an indicator of the social distance of this group with regard to the culture
of violence of craftsmen (including the established masters!). In Cologne, however,
findings show a quite different pattern already for the 16th century and thus well before
the period of Verbürgerlichung in the 18 th century 76. In a rural area of the 18th century,
though, Frank finds an outstanding degree of violence in the rural upper classes 77.
Further research on this topic is absolutely necessary. For example, the change in social
codes and the position of violence within this context need to be analysed. Promising
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
16
Criminalized violence and the process of civilisation: a reappraisal
beginnings are shown by Yves Castan’s analysis for southern France, in which he notes a
shift from the old code of ‘honneur’ towards the new code of ‘honnêteté’ 78. Equally
important is the answer to the question of the extent to which religious movements and
groups, especially the process of confessionalisation, would have to be included into such
analysis79. Moreover, the decline of the culture of violent disputes among craftsmen and
peasants needs further explanation. If our hypothesis that violence is a medium of social
control is correct, then this social code must have fulfilled important functions which
were not replaceable at the drop of a hat. Did social polarisation and pauperisation in
rural areas lead to a basic change in the way people behaved in conflicts80? Are there
signs of the norms mediated by the guilds, losing their binding power or losing their
connection to honour and violence? Following Émile Durkheim, Michael Eisner has most
recently argued that the process of individualisation was an important factor in the
decline of violence, a process which was simultaneously marked by a liberalisation of the
binding norms of honour and shame that had been central to all groups81. Even if one
disagrees with his modernisation-theoretical perspective, one must concede that he has
formulated a hypothesis which can be a starting point for further historical research.
*
* *
38
The basic argument of this article can – like Spierenburg’s – be summed up in a few
sentences. The highly problematic homicide rate as a measurement is unable to certify a
long-term trend of the internalisation of interpersonal violence from the Middle Ages to
modern times. Perhaps the only provable fact is that there was a decline in that rate in
early modern times, namely in the 17th and 18th centuries. As much as Elias’ paradigm has
provided us with an important impetus for our research, it can hardly be applied to
finding an adequate answer to the question of what the reasons for that change really
were. By taking violence only as a blind affect, which had to be controlled by external and
internal constraints,this theory prevents us from gaining a deeper understanding of
violence. In contrast to Elias’ bias, a historical-anthropological perspective to violence
aims at analysing violence also as a medium of social control, as a productive and
meaningful force, and as a social code, without losing sight of the macro-historical
processes of change.
BIBLIOGRAPHY
Bahlcke, J., Strohmeyer, A. (Eds), Konfessionalisierung in Ostmitteleuropa. Wirkungen des religiösen
Wandels im 16. und 17. Jahrhundert in Staat, Gesellschaft und Kultur, Stuttgart, Steiner, 1999.
Beattie, J. M., Crime and the Courts in England, 1660-1800, Oxford, Clarendon, 1986.
Becker, M. B., Changing patterns of violence and justice in fourteenth-and fifteenth century
Florence, Comparative Studies in Society and History 1976, 18, pp. 281-296.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
17
Criminalized violence and the process of civilisation: a reappraisal
Behringer, W., Mörder, Diebe, Ehebrecher. Verbrechen und Strafen in Kurbayern vom 16. Bis 18.
Jahrhundert, in Van Dülmen, R. (Ed.), Verbrechen, Strafen und soziale Kontrolle, Frankfurt am Main,
Fischer, 1990, pp. 85-132.
Blastenbrei, P , Kriminalität in Rom 1560-1585, Tübingen, Niemeyer, 1995.
Blauert, A., Schwerhoff, G. (Eds.), Kriminalitätsgeschichte. Beiträge zur Sozial- und Kulturgeschichte der
Vormoderne, Konstanz, UVK, 2000.
Cockburn, J. S., Patterns of violence in english society: Homicide in Kent 1560-1985, Past & Present
1991, 130, pp. 70-106.
Dinges, M., Die Ehre als Thema der Stadtgeschichte. Eine Semantik im Übergang vom Acien
Régime zur Moderne, Zeitschrift für historische Forschung 1989, 16, pp. 409-440.
Dinges, M., Der Maurermeister und der Finanzrichter. Ehre, Geld und soziale Kontrolle im Paris des
18. Jahrhunderts, Göttingen, Vandenhoeck & Ruprecht, 1994.
Dinges, M., Formenwandel der Gewalt in der Neuzeit. Zur Kritik der Zivilisationstheorie von
Norbert Elias, in Sieferle, R. P., Breuninger, H. (Eds.), Kulturen der Gewalt. Ritualisierung und
Symbolisierung von Gewalt in der Geschichte, Frankfurt/M., Campus, 1998, pp. 171-194.
Dinges, M., Justiznutzung als soziale Kontrolle in der Frühen Neuzeit, in Blauert, A., Schwerhoff,
G. (Eds.), Kriminalitätsgeschichte. Beiträge zur Sozial- und Kulturgeschichte der Vormoderne, Konstanz,
UVK, 2000, pp. 503-544.
Eibach, J., Städtische Gewaltkriminalität im Ancien Régime. Frankfurt am Main im europäischen
Kontext, Zeitschrift für Historische Forschung 1998, 25, pp. 359-382.
Eisner, M., Individuelle Gewalt und Modernisierung in Europa, 1200-2000, in Albrecht, G., Backes,
O., Kühnel, W. (Eds.), Gewaltkriminalität zwischen Mythos und Realität, Frankfurt/Main, Suhrkamp,
2001a, pp. 71-100.
Eisner, M., Modernization, self-control and lethal violence. The long-term dynamics of european
homicide rates in theoretical perspective, The British Journal of Criminology 2001b, 41, pp. 618-638.
Elias, N., Über den Prozess der Zivilisation. Soziogenetische und psychogenetische Untersuchungen, 2 vol.
(Gesammelte Schriften Bd. 3), Frankfurt/M., Suhrkamp, 1997.
Evans, R. J., Rituals of Retribution. Capital Punishment in Germany, 1600-1987, Oxford, Oxford Univ.
Press, 1996.
Frank, M., Dörfliche Gesellschaft und Kriminalität. Das Fallbeispiel Lippe 1650-1800, Paderborn,
Schöningh, 1995.
Frank, M., Ehre und Gewalt im Dorf der Frühen Neuzeit. Das Beispiel Heiden (Grafschaft Lippe) im
17. u. 18. Jahrhundert, in Schreiner, K., Schwerhoff, G. (Eds.): Verletzte Ehre. Ehrkonflikte in
Gesellschaften des Mittelalters und der Frühen Neuzeit, Köln, Böhlau, 1995a, pp. 320-338.
Fuchs, R.-P., Um die Ehre. Westfälische Beleidigungsprozesse vor dem Reichskammergericht 1525-1805,
Paderborn, Schöningh, 1999.
Given, J. B., Society an Homicide in Thirteenth-Century England, Stanford, Stanford University Press,
1977.
Häberlein, M., Tod auf der Herrentrinkstube. Ehre und Gewalt in der Augsburger
Führungsschicht (1500-1620), in Backmann, S. et al. (Eds.): Ehrkonzepte in der Frühen Neuzeit.
Identitäten und Abgrenzungen, Berlin, Akademie, 1998, pp. 148-169.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
18
Criminalized violence and the process of civilisation: a reappraisal
Häberlein, M. (Ed.), Devianz, Widerstand und Herrschaftspraxis in der Vormoderne. Studien zu Konflikten
im südwestdeutschen Raum (15.-18. Jh.), Konstanz, UVK, 1999.
Hammer Jr., C. I.: Patterns of homicide in a medieval university town: Fourteenth-century
Oxford, Past & Present, 1978, 78, pp. 3-23.
Isenmann, E., Die deutsche Stadt des Spätmittelalters. Stadtgestalt, Recht, Stadtregiment, Kirche,
Gesellschaft, Wirtschaft, Stuttgart, Ulmer,1988.
Johnson, E. A., Monkkonen, E. H. (Eds.), The Civilization of Crime. Violence in Town and Country since
the Middle Ages, Urbana and Chicago, University of Illinois Press, 1996.
Klein, G., Liebsch, K. (Eds.), Zivilisierung des weiblichen Ich, Frankfurt/M., Suhrkamp, 1997.
Kolmer, L., Gewalttätige Öffentlichkeit und öffentliche Gewalt. Zur städtischen Kriminalität im
späten Mittelalter, Zeitschrift für Rechtsgeschichte Germanistische Abteilung, 1997, 114, pp. 261-295.
Kottmann, P., Gogerichte in der Agrargesellschaft des Hochstifts Osnabrück (1500-1800),
Historische Mitteilungen,1998, 11, pp. 1-22.
Lacour, E., Schlägereyen und Unglücksfälle. Zur Historischen Psychologie und Typologie von Gewalt in der
frühneuzeitlichen Eifel, Egelsbach, Hänsel-Hohenhausen, 2000.
Lacour, E., Faces of violence revisited. A typology of violence in early modern rural Germany,
Journal of Social History, 2001, 35, pp. 649-668.
Lindström, D., Crime and control in the capital: Stockholm 1475-1625, in Österberg, E., Lindström,
D., Crime and Social Control in Medieval and Early Modern Swedish Towns, Stockholm, Almqvist &
Wiksell, 1988, pp. 67-140.
Pröve, R., Gewalt und Herrschaft in der Frühen Neuzeit. Formen und Formenwandel der Gewalt,
Zeitschrift für Geschichtswissenschaft, 1999, 47, pp. 792-806.
Rehberg, K.-S. (Ed.), Norbert Elias und die Menschenwissenschaften. Studien zur Entstehung und
Wirkungsgeschichte seines Werkes, Frankfurt/M., Suhrkamp, 1996.
Rousseaux, X., Frome case to crime. Homicide regulation in medieval and modern Europe, in
Willoweit, D. (Ed.), Die Entstehung des öffentlichen Strafrechts. Bestandsaufnahme eines europäischen
Forschungsproblems, Köln, Böhlau, 1999, pp. 143-175.
Rummel, W., Verletzung von Körper, Ehre und Eigentum. Varianten im Umgang mit Gewalt in
Dörfern des 17. Jahrhunderts, in Blauert, A., Schwerhoff, G. (Eds.), Mit den Waffen der Justiz. Zur
Kriminalitätsgeschichte des späten Mittelalters und der Frühen Neuzeit, Frankfurt/M., Fischer, 1993,
pp. 86-114.
Schindler, N., Habitus und Herrschaft. Zum Wandel aristokratischer Herrschaftspraxis im 16.
Jahrhundert, in Schindler, N., Widerspenstige Leute. Studien zur Volkskultur in der Frühen Neuzeit,
Frankfurt/M., Fischer, 1992, pp. 78-120.
Schröter, M., Erfahrungen mit Norbert Elias, Frankfurt/M., Suhrkamp, 1997.
Schultheiss, W. (Ed.), Die Achts,- Verbots- und Fehdebücher Nürnbergs von 1285-1400, Nürnberg,
Selbstverlag,1960.
Schüssler, M., Statistische Untersuchung des Verbrechens in Nürnberg im Zeitraum von 1285 bis
1400, Zeitschrift für Rechtsgeschichte Germanistische Abteilung,1991, 108, pp. 117-193.
Schüssler, M., Verbrechen im spätmittelalterlichen Olmütz. Statistische Untersuchung der
Kriminalität im Osten des Heiligen Römischen Reiches, Zeitschrift für Rechtsgeschichte
Germanistische Abteilung 1994, 111, pp. 148-271.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
19
Criminalized violence and the process of civilisation: a reappraisal
Schüssler, M., Verbrechen in Krakau (1361-1405) und seiner Beistadt Kasimir (1370-1402),
Zeitschrift für Rechtsgeschichte Germanistische Abteilung 1998, 115, pp. 198-338.
Schulze, W. (Ed.), Europäische Bauernrevolten der frühen Neuzeit, Frankfurt/M., Suhrkamp, 1982.
Schulze, W. (Ed.), Aufstände, Revolten und Prozesse. Beiträge zu bäuerlichen Widerstandsbewegungen im
frühneuzeitlichen Europa, Stuttgart, Klett-Cotta, 1983.
Schuster, P., Eine Stadt vor Gericht. Recht und Alltag im spätmittelalterlichen Konstanz, Paderborn,
Schöningh, 2000.
Schuster, P., Richter ihrer selbst? Delinquenz gesellschaftlicher Oberschichten in der
spätmittelalterlichen Stadt, in Blauert, A., Schwerhoff, G. (Eds.), Kriminalitätsgeschichte. Beiträge
zur Sozial- und Kulturgeschichte der Vormoderne, Konstanz, UVK, 2000, pp. 359-378.
Schwerhoff, G., Köln im Kreuzverhör. Kriminalität, Herrschaft und Gesellschaft in einer frühneuzeitlichen
Stadt, Bonn/Berlin, Bouvier, 1991.
Schwerhoff G., Falsches Spiel. Zur kriminalhistorischen Auswertung der spätmittelalterlichen
Nürnberger Achtbücher, Mitteilungen des Vereins für die Geschichte der Stadt Nürnberg,1995, 82, pp.
23-35.
Schwerhoff G., Zivilisationsprozess und Geschichtswissenschaft. Norbert Elias’
Forschungsparadigma in historischer Perspektive, Historische Zeitschrift 1998, 266, pp. 561-605.
Schwerhoff, G., La storia della criminalità nel tardo medioevo e nella prima età moderna. Il
‘ritardo’ di un settore della ricerca tedesca, Annali dell’Istituto storico italo-germanico in Trento 1998b
[erschienen 1999], 24, pp. 573-630.
Schwerhoff, G.: Aktenkundig und gerichtsnotorisch. Einführung in die historische Kriminalitätsforschung
, Tübingen, edition diskord, 1999.
Schwerhoff, G., Insel des Friedens oder Brennpunkt der Gewalt? Die Reichsstadt Köln ca.
1470-1620, in Dinges, M., Sack, F. (Eds.), Unsichere Grossstädte? Vom Mittelalter bis zur Postmoderne,
Konstanz, UVK, 2000, pp. 139-156.
Schwerhoff, G., Social Control of Violence – Violence as Social Control: The Case of Early Modern
Germany, in Spierenburg P. (Ed.), Handbook of Social Control (2003, forthcoming).
Sharpe, J. A., Crime in England: Long-term trends and the problem of modernisation in Johnson,
E.A., Monkkonen, E.H. (Eds), The Civilization of Crime. Violence in Town and Country since the Middle
Ages, Urbana and Chicago, University of Illinois Press, 1996, pp. 17-34.
Simon-Muscheid, K., Gewalt und Ehre im spätmittelalterlichen Handwerk am Beispiel Basels,
Zeitschrift für Historische Forschung, 1991, 18, pp. 1-31.
Spierenburg, P., Faces of violence. Homicide trends and cultural meanings: Amsterdam
1431-1816, Journal of Social History, 1994, 27, pp. 701-716.
Spierenburg, P., Long-term trends in homicide: Theoretical reflections and Dutch evidence,
fifteenth to twentieth centuries, in Johnson, E.A., Monkkonen, E.H. (Eds), The Civilization of Crime.
Violence in Town and Country since the Middle Ages, Urbana and Chicago, University of Illinois Press,
1996, pp. 63-105.
Spierenburg, P., How violent were women? Court cases in Amsterdam, 1650-1810, Crime, Histoire
et Sociétés, 1997, 1, pp. 9-28.
Spierenburg, P., Violence and the civilizing process: Does it work? Crime, Histoire et Sociétés, 2001,
5, 2, pp. 87-105.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
20
Criminalized violence and the process of civilisation: a reappraisal
Stone, L., Interpersonal violence in english society, 1300-1980, Past & Present,1983, 101, pp. 22-33.
Trotha, T. von (Ed.), Soziologie der Gewalt, Opladen, Westdeutscher Verlag, 1997.
Vanhemelryck, F., De criminaliteit in de ammanie van Brussel van de late middeleeuwen tot het einde van
het ancien regime (1404-1789), Bruxelles, Koninklijke Acad. voor Wetenschappen, 1981.
Walz, R., Agonale Kommunikation im Dorf der Frühen Neuzeit, Westfälische Forschungen, 1992, 42,
pp. 215-251.
Weber, M., Wirtschaft und Gesellschaft. Grundriss der verstehenden Soziologie, 5., revidierte Auflage,
ed. Winckelmann, J., Tübingen, Mohr, 1980.
Wettmann-Jungblut, P., Penal law and criminality in Southwestern Germany: Forms, patterns and
developments 1200-1800, in Rousseaux, X., Levy, R. (Eds.): Le pénal dans tous ses États. Justice, États
et Sociétés en Europe (XIIe-XXe siècles), Bruxelles, Facultés universitaires Saint-Louis, 1997, pp. 25-46.
Wettmann-Jungblut, P., Der nächste Weg zum Galgen? Studien zur Eigentumskriminalität in
Südwestdeutschland 1550-1850, Diss. masch. Saarbrücken,1997 (to be published 2002).
NOTES
1. A preliminary version of this paper was read at the conference «Gewalt. Wahrnehmung und
Regulierung von Gewalt in der Vormoderne», organized by Neithard Bulst and Peter Schuster in
Bielefeld at the Zentrum für Interdisziplinäre Forschung, 2 th – 29th November 1998. The author is
currently Professor of Early Modern History at the Technical University of Dresden. Besides
criminal history, his main research interests are the history of early modern religion and the
history of public spheres.
2. Stone (1983, p. 32).
3. Besides, «Violence in Town and Country since the Middle Ages» is the subtitle of this
anthology – the topic of the discussion was not crime per se, but violence. Hence, according to
the title, the confusion between violence and criminality, of which Spierenburg (2002, p. 95),
accuses Claude Gauvard, is also prevalent among the editors of this anthology!
4. Johnson, Monkkonen (1996, p. 6f). At another point, the second half of the 18 th century is said
to have been the decisive period (8).
5. Compare Elias (1997) for an annotated new edition which also includes variations from the
1982 English translation, authorised by Elias himself; on the reception of his works, see the essays
in Rehberg (1996) and Klein, Liebsch (1997).
6. Elias (1997, pp. 365ff.).
7. Dinges (1994, pp. 328ff.; 1998); Schwerhoff (1998, 1999).
8. Crime, Histoire et Société 2000, 4, 1, pp. 135-137.
9. Sharpe (1996, p. 30); cf. his ‘observations’ in Past and Present 108 (1985), pp. 206-215.
10. Cf. Rousseaux (1999, p.145); Eisner (2001a, b); Lacour (2000, 2001).
11. This indeed is the actual focus of my previous analysis of the theory of civilisation.
12. So far, the debate has focused on the topic area of «shame and embarrassment». A discussion
would have to start from Schröter’s (1997) further considerations.
13. Compare Schwerhoff (1998b) or my essay in Blauert, Schwerhoff (2000, pp. 21-67).
14. Spierenburg (2002, p. 91).
15. I am most grateful to Manuel Eisner who kindly provided me with a version of his paper
before it was published (Eisner, 2001a; compare also Eisner, 2001b).
16. All following figures presented without an individual reference are taken either from the list
by Schüssler (1998, pp. 218ff.) or from the tables by Rousseaux (1999, pp. 159ff.) respectively.
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
21
Criminalized violence and the process of civilisation: a reappraisal
17. Wettmann-Jungblut (1997, p. 38); Schüssler (1994, pp. 166f.); Schüssler (1998, pp. 216f.).
18. Spierenburg (1996, p. 80).
19. Schüssler (1998, pp. 222f.); also Simon-Muscheid (1991, pp. 30f.) and Kolmer (1997, p. 276).
20. Schuster (2000b, p. 89f.); Schwerhoff (1991, pp. 282-284).
21. Compare Rousseaux (1999, p. 161) – he adds up methodologically correctly the figures
presented by Vanhemelryck (1981, pp. 90-105), whereas Schüssler (1998, p. 218) wrongly states a
very low value of 41 ‘murders’.
22. Schüssler (1998, p. 219).
23. For the development of homicide from case to crime, see Rousseaux (1999).
24. Lindström (1988, p. 92). Another, banal aspect of the categorization problem is the question
of what or who is counted respectively: the dead, the deeds or the killers, compare SimonMuscheid (1991, p. 30) where 45 persons had been convicted for 30 cases of manslaughter.
25. Given (1977, p. 30f.).
26. Eisner (2001a, p. 74).
27. Spierenburg (2001, p. 91; 1996, p. 79).
28. Schüssler (1991, p. 122). As a critique, compare Schwerhoff (1995).
29. His representation of my findings for Cologne is visually distorting in that he takes individual
years with maximum values only. Hence the homicide rate of 10 which I estimated is doubled,
compare Schüssler (1998, p. 220). Typically enough too Spierenburg (1996, p. 66) presents my
figures as «minimum rates».
30. Isenmann (1988).
31. Schultheiß (1961) no. 1008, 127.
32. Schwerhoff (1991, p. 286).
33. Even Manuel Eisner as a follower of the hypothesis of a long-term decline of violence
concedes that «due to the problems with the data one should not make too far-reaching
conclusions» Eisner (2001a, p. 84).
34. According to Eibach (1998, p. 387) there were 31 convicted offences of homicide between 1562
and 1594, whereas only 15 in the period from 1661 to 1694 and less than 10 of such convictions
during the periods from 1741 to 1773 and from 1774 to 1805 respectively. Wettmann-Jungblut
(1997) notes still 7 cases of murder or manslaughter for the small Benedictine convent of St.
Blasien with approximately 3,000 inhabitants, whereas no such case was found for the 20 years
between 1753 and 1773.
35. Spierenburg (1996, pp. 80f.).
36. For a tabular presentation of the data, it is more advisable to have separate tables for the late
16th and early 17 th century and for the late 17 th and the 18 th century. Additionally, contrasting
figures have been added on the one hand for Rom in the late 16 th century as a rare case of very
dense material and on the other hand – rather as a supplement- the well-known data by Beattie
which is based on «softer» source material, the homicide indictments.
37. Compare, for example, Lacour (2000, pp. 88ff.).
38. Spierenburg (1996, p. 74).
39. Cockburn (1991, pp. 79ff.).
40. Compare similarly Eisner (2001 a, b).
41. Compare Blasenbrei (1995); Eisner tries to distinguish the development of violence in the
various European areas and concludes that the pioneer states of the process of modernization
England and Holland were those in which the violence rates declined first. That this
differentiation is empirically provable seems – apart from the obvious north-south divide,
though – questionable to me due to the data basis.
42. Spierenburg (1996, p. 67).
43. Elias (1997, pp. 151ff., pp. 331ff.). For a general discussion on the terminological specification
of «power» and «violence», see Pröve (1999).
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
22
Criminalized violence and the process of civilisation: a reappraisal
44. Weber (1980, p. 29).
45. Elias (1997, p. 332); compare the English text p. 538.
46. For a comparison of the frequency and development of death penalties in early modern
Germany, for example: Behringer (1990, p. 468); and more recently Evans (1996).
47. Pröve (1999, p. 803); cf. Spierenburg (2001, p. 97).
48. Spierenburg (1994, p. 702).
49. Schwerhoff (1998a, p. 576 ff.).
50. Spierenburg (2001, p. 100).
51. Spierenburg (2001, p. 99).
52. Spierenburg (2001, p. 101f.).
53. Spierenburg (2001, p. 94ff.). It should be aknowledged explicitely in this context that
Spierenburg himself has made important contributions to the phenomenology of violence.
54. Spierenburg (1994, p. 704; 1996, p. 70).
55. Spierenburg (1996, p. 94).
56. Lacour (2000, pp. 182ff.; 2001, pp. 651f.). Incidentally, Lacour notes «only a few major
changes» in the profile of violence for her rural area of research in West Germany between the 16
th
and the 18th century – an increase in violence within the family being on of them.
57. For a detailed comparison, see Schwerhoff (2003).
58. Lacour (2000, pp. 111ff.; 2001, p. 659).
59. Lacour (2000, pp. 89, 171ff.).
60. Spierenburg (2001, p. 96).
61. Cf. Trotha (1997, pp. 15, 20ff.).
62. Rummel (1993, p. 88). For an overview of the respective German studies, see Schwerhoff
(1999, pp. 112ff.). Furthermore, the programme of a «historical anthropology of violence»
postulated by Dinges should be mentioned here once again. Dinges has an understanding of
violence similar to Trotha’s.
63. For a more detailed analysis cf. Schwerhoff 2003).
64. Spierenburg (1997, p. 10). If the criteria of «physical violence» was to be generalized like this,
then the exclusion of certain crimes, above all infanticide, but also rape, would have to be
criticised as well. This exclusion, too, originates only from the fact that certain contexts of
crimes can be distinguished from one another quite clearly. Thus, Spierenburg (1996, p. 73),
notes that infanticide tells us less about violence than about shame and despair.
65. Fuchs (1999, p. 139ff.).
66. Apart from the works of Martin Dinges, the essay by Walz (1992) must be mentioned as a
classic analysis of the «agonal» culture; cf. Frank (1995b). It should be pointed out that the
category ‘honour’ is used here as a general category of analysis and not, as in Lacour’s works, as
one type of conflict among many others.
67. Eisner (2001a, p. 92).
68. Cf. Schulze (1982, 1983). For the connecting points between the research on crime and the
study of protests, see contributions in Häberlein (1999).
69. Kottmann (1998, p. 14).
70. Cf. Dinges (2000).
71. Eibach (1998, p. 381f.).
72. Elias (1997, vol. II., pp. 420 ff.); Spierenburg (2002, p. 101).
73. Cf. Schwerhoff (2000, p. 154f.).
74. Schindler (1992).
75. Schuster (2000a); Häberlein (1998).
76. Schwerhoff (1991, p. 185f.).
77. Frank (1995b, p. 337).
78. Cf. the discussion in Dinges (1989, pp. 425ff.).
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
23
Criminalized violence and the process of civilisation: a reappraisal
79. For a discussion of the paradigm of ‘confessionalization’ cf. Bahlcke, Strohmeyer (1999).
80. For hints of such a change, see Frank (1995a).
81. Eisner (2001b, pp. 91ff.).
ABSTRACTS
«Violence and the Civilising Process: Does it work?» – a question raised by Pieter Spierenburg in
a recent issue of CHS will be the reference point for the following article. In contrast to
Spierenburg my answer to this question is: ‘No, it does not’. First I comment on the validity of
late medieval and early modern ‘homicide rates’ for the development of a general history of
violence. A long-term decline in violence, as it is postulated by Spierenburg, cannot be found in
the sources. I will then deal with the role of violence within the theory of civilisation. Norbert
Elias’ interpretations are based on both a naïve reading of the source material and a simplistic
understanding of violence which regards violence as a human ‘drive’ that needs to be suppressed
by social control. In the third and final part of this article I try to develop an alternative
explanation of violence which views violence as a meaningful and culturally encoded form of
social action. Accordingly, violence should not only be perceived as an object of social control but
also as a means of social control.
«Violence et processus de civilisation: est-ce que ça marche?» – Cette question soulevée par
Pieter Spierenburg dans un numéro récent de CHS sera le point de départ de cet article.
Contrairement à Spierenburg, ma réponse à cette question est: «Non, ça ne marche pas». Dans un
premier temps, je commenterai la validité des «taux d’homicides» pour le Moyen-Âge tardif et le
début de la période moderne, dans la perspective d’une histoire générale de la violence. On ne
peut pas trouver dans les sources le déclin de la violence sur le long terme que postule
Spierenburg. Puis je m’intéresserai au rôle de la violence dans la théorie de la civilisation. Les
interprétations de Norbert Elias sont basées sur une lecture naïve des sources de compréhension
simpliste de la violence, qui voit dans celle-ci une «pulsion» qui doit être réfrénée par le contrôle
social. Dans la troisième partie de mon article, j’essaierai de développer une explication
alternative de la violence, en tant que forme d’action socialement et culturellement codifiée. En
conséquence, la violence ne devrait pas être perçue seulement comme un objet mais aussi comme
un instrument de contrôle social.
AUTHORS
GERD SCHWERHOFF
Technische Universität Dresden, Institut für Geschichte Dresden – Allemagne,
gerd.schwerhoff@mailbox.tu-dresden.de
Crime, Histoire & Sociétés / Crime, History & Societies, Vol. 6, n°2 | 2009
24