Violence in Europe: historical and contemporary perspectives

Violence in Europe

Sophie Body-Gendrot

l

Pieter Spierenburg

Editors

Violence in Europe Historical and Contemporary Perspectives

Editors Sophie Body-Gendrot University of Sorbonne, Paris, France

ISBN: 978-0-387-74507-7

Pieter Spierenburg Erasmus University, Rotterdam, The Netherlands

e-ISBN: 978-0-387-74508-4

Library of Congress Control Number: 2007936599 # 2008 Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com

Acknowledgments

This collection is the principal result of a series of seminars, directed by the editors, and devoted to the confrontation between historical and contemporary scholarship dealing with violence. Meetings were held in Ferrara (2003), Berlin (2004), Brussels (2004), Rotterdam (2005) and Paris (2005). The series was sponsored by the Posthumus Institute with funds obtained from the Netherlands Organization for Scientific Research (NWO). The seminars were part of the program of the Groupe Europe´en de recherches sur les Normativite´s (GERN), which co-sponsored them. The editors also are grateful to the late Mario Sbriccoli of the University of Macerata for co-organizing the meeting in Ferrara and providing additional funds to Xavier Rousseaux of the Universite´ Catholique de Louvain (Center for Law & Justice History) and to the CEGES/SOMA of Belgium for doing the same for the meeting in Brussels.

v

Contents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sophie Body-Gendrot

1

Part I: Contested Definitions Violence: Reflections About a Word . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pieter Spierenburg

13

Violence as an Essentially Contested Concept . . . . . . . . . . . . . . . . . . . . . . . Willem de Haan

27

Part II: Long-Term Trends Homicide in Scandinavia: Long-Term Trends and Their Interpretations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Dag Lindstro¨m

43

Violence in France’s Past: An Anthropological Approach . . . . . . . . . . . . . . Franc¸ois Ploux

65

The Fall and Rise of Homicide in Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . Ian O’Donnell

79

Part III: Contemporary Trends Violence in Present-Day France: Data and Sociological Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philippe Robert

95

From Old Threats to Enigmatic Enemies: The Evolution of European Policies from Low Intensity Violence to Homegrown Terrorism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Sophie Body-Gendrot vii

viii

Contents

Part IV: Gendering Violent Practices The Language of Violence: Symbolic Body Parts in Marital Conflicts in Early Modern France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Dorothea Nolde Negotiating Violence and Gender: Security and the Night Time Economy in the UK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 Kate O’Brien, Dick Hobbs and Louise Westmarland Part V: Politics, War and Violence Violence and War: Measuring Homicide in Belgium (1900–1950) . . . . . . . . 177 Xavier Rousseaux, Fre´de´ric Vesentini and Antoon Vrints Democratization and Violence: European and International Perspectives . . 205 Susanne Karstedt Violence and Violence-Control in Long-Term Perspective: ‘Testing’ Elias in Relation to War, Genocide, Crime, Punishment and Sport . . . . . . . . . . . 227 Eric Dunning Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

Contributors

Sophie Body-Gendrot Director of the Center for Urban Studies, University of Sorbonne (Paris IV), 1 rue Victor Cousin 75230 Paris cedex 05, email:[email protected]

Susanne Karstedt Professor of Criminology, School of Criminology, Education, Sociology and Social Work, Keele University, Keele, Staffordshire, ST5 5BG, UK, email:[email protected]

Willem de Haan Professor of Criminology, Department of Criminal Law and Criminology, University of Groningen, Faculty of Law, PO Box 716, 9700 AS Groningen, The Netherlands, email:[email protected]

Dag Lindstro¨m Professor, ISAK (Institutionen fo¨r studier av samha¨llsutveckling och kultur) Enheten fo¨r historia [Department for Studies in Social Change and Culture, History Division], Linko¨pings universitet, 581 83 Linko¨ping, Sweden, email:[email protected]

Eric Dunning Emeritus Professor of Sociology, University of Leicester; Visiting Professor, University of Chester, Department of Sociology, University of Leicester, University Road, Leicester, LE1 7RH Dick Hobbs Professor of Sociology, London School of Economics, Houghton Street, London, WC2A 2AE

Dorothea Nolde Institut fu¨r Geschichtswissenschaft/ FB 08, Universita¨t Bremen, Bibliothekstrasse, D-28359 Bremen, email:[email protected]; Secretariat: Claudia Haase, Tel: þþ49-(0)421-218-2226, email:[email protected]; Current position: JuniorProfessor for Early Modern History

ix

x

Kate O’Brien School of Social Policy, Sociology and Social Research, University of Kent, Cornwallis Building NE, Canterbury, Kent, CT2 7NF, email:[email protected] Ian O’Donnell Professor of Criminology, School of Law, University College Dublin, Belfield, Dublin 4, Ireland, email:[email protected] Franc¸ois Ploux Maıˆ tre de confe´rences en histoire contemporaine.[senior lecturer in contemporary history], CERHIO-FRE CNRS 3004 Universite´ de Bretagne-Sud, Lorient; University address: 4, rue Jean-Zay. BP 92116, 56321 LORIENT CEDEX; Private address: 4, alle´e des Fauvettes. 56530 GESTEL. France Philippe Robert Director of Research at the CNRS, France; CESDIP, Immeuble Edison, 43 boulevard Vauban, 78280 Guyancourt, France, email:[email protected]

Contributors

Xavier Rousseaux Senior Research Associate FNRS, Center for Law and Justice History, Universite´ catholique de Louvain (UCL), Louvain-laNeuve, www.chdj.be, email:[email protected] Pieter Spierenburg Professor of Historical Criminology, Erasmus University Rotterdam, Postbus 1738, 3000 DR Rotterdam, Netherlands, email:[email protected] Fre´de´ric Vesentini Postdoctoral Researcher FNRS, Center for Law and Justice History, Universite´ catholique de Louvain (UCL), Louvain-laNeuve, www.chdj.be, email:[email protected] Antoon Vrints Postdoctoral Researcher BOF, Universiteit Gent (UGent) Blandijnberg, 2 B-9000 Gent, email:[email protected] Louise Westmarland Faculty of Social Sciences, The Open University, Milton Keynes, MK7 6AA

Introduction Sophie Body-Gendrot

How much related are present and past violence? The answers are complex due to the limited knowledge scientists have gathered, even after spending a life-time studying this very enigmatic and most serious social phenomenon called violence. All authors agree that the present level of interpersonal violence cannot be sufficiently understood without taking the earlier long-term decrease into account. Ted Robert Gurr (1981, 1989) was one of these pioneers who undertook a statistical overview of the development of homicides from the Middle Ages to the present, looking at England in particular. On his curve, 20 homicides per 100,000 inhabitants were recorded in the High and Late Middle Ages and one case in the twentieth century put an end to the curb. Gurr interpreted this long-term decrease in interpersonal violence as ‘‘a manifestation of cultural change in Western society, especially the growing sensitization to violence and the development of an increased internal and external control on aggressive behavior’’ (Gurr, 1981:258). Currently, both the present and the past have to be considered in any attempt to answer the following questions: is the higher incidence of violence which still prevails a temporary exception or a new trend related to structural dynamics of modern societies? In which regions of Europe is it more specifically pronounced? More generally, this volume claims that historical knowledge of changes in violent behavior and of violence forms an indispensable contribution to an understanding of the manifestations of violence in contemporary societies. This book is organized in five parts, examining contested definitions, longterm trends, contemporary trends, gendering violent practices and politics, war and violence, all contributing to elaborate historical and contemporary perspectives on violence in Europe. In Part One, two scholars agree that common definitions of violence are needed to work along the same parameters.

Sophie Body-Gendrot Director of the Center for Urban studies, University of Paris-Sorbonne, 1 rue Victor Cousin 75230, Paris cedex 05 [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

1

Introduction Sophie Body-Gendrot

How much related are present and past violence? The answers are complex due to the limited knowledge scientists have gathered, even after spending a life-time studying this very enigmatic and most serious social phenomenon called violence. All authors agree that the present level of interpersonal violence cannot be sufficiently understood without taking the earlier long-term decrease into account. Ted Robert Gurr (1981, 1989) was one of these pioneers who undertook a statistical overview of the development of homicides from the Middle Ages to the present, looking at England in particular. On his curve, 20 homicides per 100,000 inhabitants were recorded in the High and Late Middle Ages and one case in the twentieth century put an end to the curb. Gurr interpreted this long-term decrease in interpersonal violence as ‘‘a manifestation of cultural change in Western society, especially the growing sensitization to violence and the development of an increased internal and external control on aggressive behavior’’ (Gurr, 1981:258). Currently, both the present and the past have to be considered in any attempt to answer the following questions: is the higher incidence of violence which still prevails a temporary exception or a new trend related to structural dynamics of modern societies? In which regions of Europe is it more specifically pronounced? More generally, this volume claims that historical knowledge of changes in violent behavior and of violence forms an indispensable contribution to an understanding of the manifestations of violence in contemporary societies. This book is organized in five parts, examining contested definitions, longterm trends, contemporary trends, gendering violent practices and politics, war and violence, all contributing to elaborate historical and contemporary perspectives on violence in Europe. In Part One, two scholars agree that common definitions of violence are needed to work along the same parameters.

Sophie Body-Gendrot Director of the Center for Urban studies, University of Paris-Sorbonne, 1 rue Victor Cousin 75230, Paris cedex 05 [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

1

2

S. Body-Gendrot

In common-sense parlance, violence essentially refers to physical hurt and attack, Pieter Spierenburg observes (Violence: Reflections About a Word). Among scholars, anthropologists in particular feel the need to stick to this everyday usage, acknowledging that any intercultural comparison becomes problematic otherwise. All efforts of earlier scholars to broaden the scope of the concept of violence derived from a definite agenda. This is obvious in the case of ‘‘structural violence,’’ popular in the 1970s. This concept served to legitimize violent social protest with the argument that all unjust social conditions implied violence. Similar objections are in order against concepts such as psychological or symbolic violence. In conditions of high power inequality, they may lead to an intolerable confusion of sufferers and attackers. These notions are all based on the same implicit ‘logic’: (1) Violence is a serious evil; (2) some other evils in society are equally objectionable; (3) consequently, these other evils should be called violence too. But this logic being unscholarly, Pieter Spierenburg proposes to use violence in this volume as a reference to ‘‘all forms of intentional encroachment upon the physical integrity of the body’’ . This is neither an extended nor a too restricted definition. It includes a broader spectrum of interpersonal violence than that prosecuted under the law and hence it is not legalistic. And, next to interpersonal violence, it includes state violence: police action, execution and war. Hence it is independent from moral judgment and personal views of ‘order’. ‘‘Violence is a multifaceted, socially constructed and highly ambivalent phenomenon,’’ Willem de Haan remarks (Violence as an Essentially Contested Concept). It is multifaceted because there are many forms of violence; it is socially constructed because who and what is considered as violent varies according to socio-cultural and historical conditions; and it is ambivalent in the ways it is socially sanctioned, legitimized and institutionalized, as well as culturally transmitted and experienced. Depending on context and perspective, violent actions may either be condemned and considered immoral, illegal and disruptive or admired and perceived as moral, legal and functional. Controversies occur and recur about both the substance of the concept and the scope of the definition of violence. De Haan’s chapter explores the arguments for and against a restrictive or an expansive definition of violence, making use of Gallie’s notion of the ‘essential contestedness’ of concepts. It means that there are no conclusive reasons for accepting one definition and rejecting all others. For him, a proper definition should not be seen as a starting point for empirical research but as its temporary outcome. It would be more fruitful, he argues, to accept that definitions of violence are contested and that they vary depending on the specific contexts of discovery and contexts of justification. ‘‘Violence has indeed many meanings and not all of them are negative’’. ‘‘Violence is intriguing. It is universally condemned yet to be found everywhere. Most of us are fascinated and horrified by it. It is a fundamental ingredient of how we entertain ourselves. . .and an essential feature of many of our social institutions’’ (Litke, 1992:173). The latin root of violence, vis/violentia, refers to

Introduction

3

strength, power but also to force and violence. The very act of coming to life is a violent act. Violence becomes dysfunctional when it is not controled, channeled, contained by rules and laws and civil norms and when it becomes disruptive for social life in society. Robespierre coined the concept of ‘progressive violence’ in the French revolution explicitly for the pursuit of specific political goals. However, direct physical violence which will be studied here – aimed at harming, injuring or killing other people – indubitably stands at the center of the whole issue of violence, Peter Imbusch observes (2003:23). ‘‘Violence always strikes by surprise. Due to its very nature, it exceeds our expectations, disturbs our modes of living, questions our daily life’’ (Ferenczi, 2000:15). It is superior to all other means of control and coercion and its impact which does not need explanation from the author is immediately grasped and is therefore highly disturbing. ‘‘We had secretly made the decision to ignore violence and unhappiness as elements of History,’’ French philosopher Merleau-Ponty wrote about World War II, ‘‘because we lived in a country too happy and too weak to even think about them’’ (quoted in Ferenczi, 2000:15). But the massive trauma of the 20th century have shown how such hope was fragile. The combined historical and contemporary approach of this collection owes much to long-term trends in research on violence and three contributions constitute Part Two. The Scandinavian case offers an excellent opportunity for exploring what use can be made of the study of long-term trends. The homicide ratios in some late medieval and early modern towns were indeed among the highest ever observed. Then a dramatic decrease in deadly violence occurred, starting in the 17th century, as part of a general European change. Dag Linstro¨m’s chapter however reveals a more complicated picture (Homicide in Scandinavia). It is not altogether obvious that the level of overall violence followed the same secular trends as that of homicide. A more detailed analysis of homicide ratios also indicates a much more discontinuous development, with a number of mid- and short-term peaks. Moreover, several local studies indicate considerable regional differences, which sometimes are even more striking than the chronological changes. Regions in northern Sweden and in Finland reveal homicide ratios close to present-day standards during the 16th century already, and in some of these regions they began to rise, contrary to the general trend, during the seventeenth. Finland also offers a divergent development compared to other parts of Scandinavia from the 18th century onward. Whereas the homicide ratios continued to decrease elsewhere in Scandinavia, Finland experienced a rising level of homicide and it still has among the highest homicide ratios in Europe. Since Scandinavian historians disagree on interpretations and explanations, the merit of Linstro¨m’s chapter is to address complexity, to analyze the different theoretical approaches and to discuss interpretations relative to mid- and short term discontinuities, regional differences and the divergent Finnish case. Court records have been used in a huge quantitative enterprise by historians. Implicitly assuming a perfect equivalence between legal norms and social

4

S. Body-Gendrot

norms, researchers in the 1970s lumped all crimes and misdemeanors together as a single category of deviant behavior. Violence was not considered as a part of culture, as the product of a socially constructed ethic. Moreover, this quantitative history drew only a gradual line between crimes of blood and property crimes, considering theft and violence as mere variations of the same phenomenon. The ensuing confusion explains the neglect of the serial study of homicide in France. To redress this situation, Franc¸ois Ploux embarks on a new approach to judicial archives, which no longer confuses the history of violence with the history of crime (Violence in France’s Past). In a new type of historical research, he attempts to recreate the social and normative environment of criminality. This transition from a quantitative history of deviance to a historical anthropology of violence is helpful to understand the social and cultural logic of homicide. His chapter reveals fluctuations of violence during the 19th and early 20th centuries, in particular an increase in rural violence after the midnineteenth. This temporary increase, reversing a longer-term trend away from physical violence, does not so much reflect a greater incidence of violent acts as a shift in the social threshold of tolerance. Consequently, in conflicts of honor, recourse to extra-legal solutions are less frequent. Making a transition with the contemporary period, Ian O‘Donnell’s chapter observes that, considered historically, the rate of lethal violence (excluding infanticide) in the Republic of Ireland reached a peak during the mid 19th century, remained relatively high until the beginning of the twentieth century and then declined until the 1970s (The Fall and Rise of Homicide in Ireland). During the 1990s, the level of recorded homicide rose sharply. This trend can be related to factors such as the economic boom, changing migration patterns and increases in alcohol and drug consumption. To illustrate these shifts, the chapter examines four time periods, each separated by half a century (1845–54; 1895–1904; 1945–54; 1995–2004). The level of lethal violence during each period is described and related to wider social changes such as those wrought by famine (1845–54), the emergence of new family structures and the search for independence (1895–1904), economic depression and mass emigration (1945–54), and the arrival of the Celtic Tiger economy (1995–2004). These phenomena are related to the civilizing process as it played out at the European periphery. The continuation of this theme is found in Part Three. Contemporary social scientists, without relinquishing statistical analysis, acknowledge however that the in-depth and ethographic study of violent groups and their culture forms an essential source of knowledge. For instance, if in ancient European communities, violence was often accepted and sometimes viewed as desirable, currently this acceptance has diminished. Disrespect adds to the pains felt from unfair treatment in democracies claiming equality for all. Too much humiliation, too much disenfranchisement from the mainstream’s values foster ‘counter-worlds’. Violent cultures are usually condemned by the public at large, because ‘‘violence emotionalizes, creates fear and can be politically exploited’’ (Heitmeyer, Hagan, 2003:8). That the majority expresses a rising sensitivity over violence is

Introduction

5

evidenced, among other things, by widespread public concern over isolated acts of brutal murder, which are probably no more frequent today than in the recent or distant past. But such incidents receive ample attention from the media and are exploited by politicians during electoral campaigns. This is the case in present-day France where, as shown by Philippe Robert, interpersonal violence pervades public debate (Violence in Present-Day France). However, the use of this term is undifferentiated and vague. Scholarly work devoted to violence most frequently lacks the necessary accuracy. This shortcoming is due to the poor quality of the data generally used: police statistics. They overestimate the proportion and the growth of serious violence while underestimating the increase of petty violence which is underreported to police agencies. This inaccuracy can be overcome by linking several different data bases. The primary contribution of his chapter to a better understanding of the phenomenon lies precisely in the mobilization of various measures of violence which point at the current rapid growth of low intensity violence. Secondly, this growth is connected with the deterioration of social relationships within (and in close proximity of) relegated urban neighbourhoods. Contemporary cultures of violence belong indeed to insulated and marginalized spaces, where older notions and codes related to an honorable defense of honor have made a come-back. These codes have not returned unchanged though, as Sophie Body-Gendrot’s contribution makes clear (From Old Threats to Enigmatic Enemies). The combined factors of ethnicity and race loom large in the discussion of violence. European countries have become more differentiated in the last quarter of century due to the influx of immigrants from former colonies and other non-Western countries. More recently, the trauma felt in the Netherlands after the assassination of Theo Van Gogh as well as the political consequences of the terrorist attacks in Madrid and in London question attitudes of tolerance for violent expressions of ethnic differences in European cities. If trust and loyalty allow the cohabitation of plural populations sharing the same public spaces, transportation, work, residence, leisure places in everyday life, after such events, negative stereotypes tend to amalgamate perceptions of urban risks of violence associated with idle male Muslim youths in relegated urban areas and with potential homegrown terrorists sharing similar ethnic profiles. The chapter questions whether it is legitimate for research to link these two issues. Who benefits from such links? Political entrepreneurs? Ideologues? Interest groups? Law enforcement agencies? And is there anything new in these endeavors? Are not victims more heard than before? Three types of European policies addressing new types of risks and violent forms of destruction are examined: policies securizing space and ignoring race and ethnicity; policies profiling ethnicized youth for their violent acts or potentially lethal actions; and policies of inclusiveness, participation and cooperation with Muslim leaders. Forms of multiple marginalization are then to be taken seriously by institutions. Even with low-intensity violence, risks of unintended consequences should be weighed carefully. Despite sophisticated technologies of surveillance and policies of prevention and repression, the core is never fully insulated from the margins.

6

S. Body-Gendrot

Besides the links of ethnicity and violence, gender is also a central issue. Gendering violent practices are the object of Part Four. A vast amount of historical studies led to the conclusion that ancient European communities often accepted violence and viewed it as desirable. It was considered as a defense of personal honor and, simultaneously, as a regulation of social tensions. Gradually and with a timing varying per country, this acceptance has diminished. During the last 15 years, historians have paid ample attention to the context of violence, including its meaning for the parties involved (with honor codes, rituals, and types of insult) and its perception by the public at large. In early modern Europe, violence was a ubiquitous part of sociability and, as such, culturally coded to an extremely high degree. Recent studies present physical violence as an almost exclusively male phenomenon. The interest raised by Dorothea Nolde’s work is precisely to address the question of gender-specific meanings and perceptions of both male and female marital violence through the example of France at the end of the 16th century and the beginning of the 17th century (The Language of Violence). As in other European countries, marriage in early modern France was embedded in a culture characterized by an intense presence of violence. The order of marriage, as one of the fundamental institutions of society, was redefined following the Council of Trent and in the context of nascent absolutism. In this process, the regulation of violence played an essential role. On the one hand, new limits were set on the practice of male violence. On the other hand, female rebellion and even more so, female violence were increasingly stigmatized, and the duty of unconditional obedience became the key element of a redefinition of the role of the wife. In this context, the symbolic as well as the communicative meaning of masculine and feminine violence is of particular importance. From a contemporary view point, three associated contributors, Kate O’Brien, Dick Hobbs and Louise Westmarland pursue the theme of gender and violence and look at how it intersects with the working practices and occupational culture of bouncers within the context of Britain’s commercially driven, violence fuelled night-time economies (Negotiating Violence and Gender). They examine violence work as gendered work by exploring how male and female bouncers differ in the way they perform and control violence within licensed venues. Blending together the findings of two British based ethnographic studies of male and female bouncers, they address the social and political context of women’s increased participation within the sector of the private security industry and show that both men and women are commercially viable as specialists of violence. Control is achieved by the gendered (and sexed) body. Violence is instrumentalized and its potential is performed by men and women working in quasi-liminal environments. The major themes of Part Five are politics, war and violence. From one perspective, figures of homicide can be seen as clear indicators of tensions in a particular society. Thus they offer an often neglected, but nevertheless valuable source for social historical analysis. During both the First

Introduction

7

and the Second World War (and their aftermath), societal tensions in Belgium caused unprecedented levels in homicide rates. The contribution of Xavier Rousseaux, Fre´de´ric Vesentini & Antoon Vrints measures homicide in Belgium during the first half on the 20th century (Violence and War). Belgium has interesting and reliable series of statistics from its founding in 1830 up to now. The chapter offers an evaluation of the figures of homicide. Besides official and published figures, it shows how historians can make use of many additional sources to tackle homicide issues. Like the reports written by coroners in Anglo-Saxon countries, in Belgium, medical reports on the causes of death were drawn up for the public prosecutor’s offices. These data are directly compiled from forensic examinations and the unequal geographical distribution of high homicide rates yields valuable clues. During the First World War indeed, the combined factors of massive impoverishment in the cities, of relative wealth in the countryside and the collapse of state authority gave rise to rural banditry on a large scale reflected in the high homicide rates of rural areas. But during the Second World War reflect, reality is far more complex. Mass impoverishment under a totalitarian occupation regime provoked both banditry and political violence. As shown here, violence resulted from social processes, institutional dysfunctions and social conditions. It was not a mere means to an end. Was the European trajectory of modernization and decreasing violence then a more unique historical development and path-dependent trajectory of modernization? Answers rely on a three-pronged strategy based on the analysis of cities. What processes in urban culture and structures impact on social conflicts and violent behavior? Are there countervailing forces? What factors might be responsible for rising urban violence? The development of tolerance and its relation to insecurity and violent conflicts are to be taken into account in the case of Europe. Susanne Karstedt is also concerned with the long dure´e of declining violent crime in cities (Democratization and Violence). In the 19th century, social scientists like Durkheim, Georg von Mayr and others thought that the structural and moral changes that were most visible in cities were driving causal factors. Such developments were remarkably similar across continents, as the comparative study by Gurr et al. of London, Stockholm, Sydney and Calcutta shows. Developmental theories have taken up this perspective during the second half of the twentieth century, and research has provided cross-national evidence that the decline of homicides seems to be a general pattern during the process of modernization and industrialization. However, the chapter points out that this observation obviously does not apply to cities in less developed countries. The surge of violence in the urban conglomerates of the Third World is observed exactly during the same period when cities in the Western industrialized world, the USA and Europe, became more violent and had to cope with rising crime rates in general, and violent crimes in particular. All evidence points to cities as seedbeds where the processes of social change are most intricately linked to changes of social conflicts and violent behavior, and most visible in terms of

8

S. Body-Gendrot

concentration and impact. Cities as laboratories of modernity and modernization offer themselves to explanations of the recent surge in violence. An author seems to link most of the contributions of this unprecedented volume on European violence: Norbert Elias. His thesis relative to the civilizing process and the decline of interpersonal violence is well-known (Elias, 1976). He perceived violence as a phenomenon that decreases due to the development of modernity and as ‘‘one that is securely encapsulated by the state monopoly of violence. . .This reflects a basic pattern of Western industrialized societies, which can be described as self-deception that circumvents enlightenment and in which insufficient account is taken of the ambivalence of the modern age’’ (Heitmeyer, Hagan, 2003:6). A co-author and long-time friend of N. Elias, Eric Dunning, ends the fifth part of this volume by tackling this issue from another angle (Violence and Violence-Control in Long-Term Perspective). Linking war, genocide, crime, punishment, and sport – all of which involve forms of violence which are primarily, though by no means solely, male phenomena, he discusses the meaning of terms such as ‘violence’ and ‘aggression’ and constructs a typology of violence. He makes use of some of the conceptual distinctions introduced in his typology to shed light on these types of violence seen in long-term perspective. The paper focuses on (i) the implications of such issues for Elias’s theory of civilizing processes; and (ii) with light of Elias’ theory – which is widely held to be refuted by current trends – what can be shed upon them. So far, the value of Elias’s theory has only been generally recognized in historical criminology but, if interpreted correctly, it can serve as what Elias himself called a ‘central theory,’ that is as a unifying focus and guide for historical sociological work in a wide range of fields. As shown by this volume, the centrality of violence in representations permeates the history of societies and the present time. ‘‘Processes of the past continue to operate; the past is transformed, it is not obliterated. Novelty is always a rhetorical move and history ever-present’’ (Beauregard, Haila, 2000:23). Dominant groups project their own phantasies and anxieties, their images of ‘Dangerous Others,’ their latent fears and self-rewarding myths, which then trickle down to masses via interactive chains of communication and transmission (Body-Gendrot, 2002). Due to such constructions and to binary perceptions, common to all social categories, of an environment consisting of ‘insiders’ and ‘outsiders,’ blurred or bright boundaries continuously emerge, marking distance, externalization and differentiations loaded with more or less hostility. Tensions between conflicting groups are reactivated in periods of great uncertainty and of search for bearings, less over ideologies, as it is easily believed, than over contentions and brushes from daily life. It is an illusion to believe that violence may remain spatially and symbolically confined to certain areas and within certain groups. As recent lethal actions of interpersonal violence in Europe demonstrate, violence hits the core of societies like a shock-wave. Its destabilizing impact saps the very principles of democracies’ fragile social contracts.

Introduction

9

References Beauregard, R., Haila, A. (2000). The unavoidable continuities in the city. In P. Marcuse and R. van Kempen (Eds.), Globalizing Cities. A New Spatial Order (pp. 22–36), Oxford: Blackwell. Body-Gendrot, S. (2002). The dangerous others: Changing views on urban risks and violence in France and in the United States. In J. Eade and C. Mele (Eds.), Understanding the City (pp. 82–106), Oxford: Blackwell. Elias, N. (1976). U¨ber den Prozess der Zivilisation. Soziogenetische und pyschogenetische Untersuchungen. Frankfurt a.M: Suhrkamp. Ferenczi, T. (2000). Politique ou barbarie. In T. Ferenczi (Ed.), Faut-il s’accommoder de la violence? (pp. 15–22), Bruxelles: Editions Complexe. Gurr, T. R. (1981). Historical trends in violent crime. A critical review of the evidence. Crime and Justice. An Annual Review of Research, 3, 295–350. Gurr, T. R. (1989). Historical trends in violent crime: Europe and the United States. In T. Gurr (Ed.), Violence in America, vol. 1: History of Crime (pp. 21–54), Newbury Park, California: Sage. Heitmeyer, W., Hagan, J. (2003). Violence: The difficulties of a systematic international review. In W. Heitmeyer and J. Logan (Eds.), International Handbook of Violence Research (pp. 3–11), Dordrecht: Kluwer Academic Publishers. Imbusch, P. (2003). The concept of violence. In W. Heitmeyer and J. Logan (Eds.), International Handbook of Violence Research (pp. 13–40), Dordrecht: Kluwer Academic Publishers. Litke, R. (1992). Violence and power. International Social Science Journal, 132, 173–183.

Part I

Contested Definitions

Violence: Reflections About a Word Pieter Spierenburg

The starting point for my reflections about the word ‘‘violence’’ comes perhaps unexpected. In his book Kindly Inquisitors, the journalist and philosopher Jonathan Rauch outlines a radical defense of free inquiry against all forms of censorship, whether traditionally authoritarian or rooted in the modern notion of ‘‘political correctness.’’ This defense of free inquiry leads him to oppose all possible restrictions on verbal expression. In a chapter entitled ‘‘The Humanitarian Threat’’ he reviews measures, proposed and devised, against ‘‘assaultive speech,’’ quoting a professor who had stated ‘‘To me, racial epithets are not speech. They are bullets.’’ Rauch’s reaction is eloquent and uncompromising: ‘‘you do not have to be Kant to see what comes after ‘offensive words are bullets’: if you hurt me with words, I reply with bullets, and the exchange is even.’’ (Rauch 1993: 131) Earlier in the book, Rauch characterizes the views he rejects as ‘‘a theory which said that images and expressions and words could be, for all practical purposes, a form of hurt or violence.’’ (Rauch 1993: 18). They don’t literally hurt, is the obvious implication. If anything, the quoted passages show that the definition of violence, although it can be treated as a scholarly problem, has philosophical and moral dimensions as well. Rauch’s position is all the more intriguing since De Haan in ‘‘Violence as an Essentially Contested Concept’’, bases his diverging views mainly on philosophical literature. My article focuses on the scholarly aspects. Like Rauch, I will essentially argue that words are words and bullets are bullets, or, from the perspective of an early modern historian, that knives are knives. I am pleading for a definition of violence, which essentially limits this word to the intentional encroachment upon a person’s physical integrity. Obviously, the argument cannot be a simple demonstration based on empirical evidence. I will just have to try to be as convincing as possible. The word ‘‘violence’’ primarily belongs to the repertoire of common speech. It is not a legal term; most systems of criminal law do without it, preferring Pieter Spierenburg Section of Criminology, Law Faculty, Erasmus University Rotterdam, Postbus 1738, 3000 DR Rotterdam, Netherlands [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

13

14

P. Spierenburg

concepts like assault or injury. Nevertheless, the word ‘‘violence’’ is very common in criminological writing, its demarcation often left undiscussed. Historians, too, are usually not bothered by the definition question. Yet, this question is a problem which concerns all disciplines within the social sciences and the humanities. There is good reason to begin my discussion with the discipline of anthropology. Its practitioners routinely face the problem of cross-cultural comparison. Two anthropological collections, therefore, form a convenient point of departure. In both of them, the editors set out by tackling the problem of definition and, although they do not opt for a restrictive one at the outset, their discussions show this option to be inevitable.

Definitions by Anthropologists In The Anthropology of Violence (1986) David Riches has written the preface as well as the introductory essay. Taken together, they read like an exercise in presenting propositions and then disclaiming them again. The preface’s lack of reflection is revealing. Without comment, Riches explains that his collection deals with various forms of violence, ‘‘from simple threats of violence at one extreme through to homicide at the other.’’ Thus, he implicitly adopts a straightforward, day-to-day understanding of violence. The qualifications appear in the introduction. First, Riches notes that ‘‘. . . to the Anglo-Saxon mind, ‘violence’ strongly connotes behaviour that is in some sense illegitimate or unacceptable.’’ On an even stronger note, he claims that Anglo-Saxon culture stresses the ‘‘irrationality and bestiality of violence.’’ Such connotations, still according to Riches, prompted the anthropologist Radcliffe-Brown to characterize physical force exercised by the state as ‘‘political organization and not violence.’’ Apparently disagreeing, Riches concludes one should always pose the question who labels an act as violent and from which social position this is done. By then, it looks as if the author prefers a definition of violence that includes all those connotations. Just a little further, however, under the heading ‘‘the performer’s perspective,’’ Riches warns that, for a cross-cultural comparison, it does not matter whether or not the performer or perpetrator views his deed as an act of violence: ‘‘For convenience, the performer’s actions can be labelled violence, corresponding to the commonsensical meaning of the term as ‘the intentional rendering of physical hurt on another human being’.’’ (Riches 1986: quotations on pp. viii, 1, 2, 3, 4). Here we have come full-circle from the casual remark in the preface. The four ‘‘basic properties’’ of violence which Riches identifies, with a claim of cross-cultural validity, are compatible with the definition as intentional rendering of physical hurt. They are: (1) ‘‘The performance of violence is inherently liable to be contested on the question of legitimacy; (2) The discrepancy in basic understandings amongst those implicated in the performance of a violent act . . . is likely to be minimal: . . . violence is unlikely to be mistaken as

Violence: Reflections About a Word

15

such; (3) The practice of violence is highly visible to the senses; (4) The performance of violence to a moderate degree of effectiveness requires relatively little by way of specialized equipment or esoteric knowledge. The manipulative and strength resources of the human body, and knowledge that these resources are capable of destroying physical objects, are sufficient to enable a minimally successful act of hurt against another human being.’’ (Riches 1986: 11). The elaborately formulated fourth property essentially means that you can be violent by slamming or kicking and, hence, that violence has no need for technology, which makes its use in all cultures possible. There might be debate about the first three, even if one accepts the definition of intentional rendering of physical hurt. Toward the end of his essay Riches discusses recent attempts at broadening the definition of violence, to include acts which result in mental anguish for example, but he rejects these attempts, wishing to view mental anguish in terms of a loss of social prestige. (Riches 1986: 22–3). Riches (1991) elaborated on his views in an article published five years later, in which he called for a relatively close correspondence between academic and lay uses of ‘‘violence’’ and related terms. My search into the anthropological literature about violence was unsystematic, but the fact that the editors of Meanings of Violence: A Cross Cultural Perspective (2000) start by referring to Riches suggest that, within anthropology, nothing much of substance about the definition of violence had been published in the meantime. In the second sentence of the preface, Jon Abbink implicitly agrees with the limitation of violence to physical hurt: ‘‘Violence is a human universal: in no known human society or social formation is interpersonal aggression, physical threat, assault, or homicide and armed conflict completely absent or successfully banned.’’ (Abbink in Aijmer/ Abbink 2000: xi). In the next paragraph he cites both Riches’ collection and his 1991 article. In Abbink’s interpretation, Riches’ discussion comes down to four defining elements of violence: its use is contested; it involves damaging physical force; it entails purposeful humiliation of other humans; it necessitates communication of one’s intention, often domination, to the other. Note that only the first of these defining elements exactly matches the first of Riches’ basic properties. Abbink’s preface is followed by Go¨ran Aijmer’s introduction, which stresses the symbolic elements of violence, paying ample attention to representation and discourse. Of the eight essays that follow, some are rather general, while others deal with specific subjects such as executions in China and bull-fighting in Spain (the latter suggesting that inflicting hurt on animals might be included into the definition of violence). Abbink studied the cattle-herding Chai, some 16,000 people living on the border of Sudan and Ethiopia. Chai men frequently engage in cattle-raiding and robberies, with ambushes and killings, against members of a rival sub-group. These practices are mutually seen as unproblematic acts of revenge. The very name of Chai means something like ’’we pay them back.’’ Apart from revenge among sub-groups, unmarried men often fight duels with wooden poles over two meters long; if someone kills his opponent in such a duel,

16

P. Spierenburg

he has to negotiate with the victim’s family about homicide compensation. All this leads Abbink to play down on the first defining element of violence which he had cited in his preface: violence, he concludes, is not always seen as a contested practice. (Abbink in Aijmer/ Abbink 2000: 77–100). We can conclude preliminarily that, although anthropologists are debating the various properties of violence, there seems to be widespread agreement among them about the restriction of the term to the realm of the intentional infliction of physical hurt. This appears to be little different in a German sociological collection entitled Soziologie der Gewalt (1997). The editor, Trutz von Trotha, emphasizes the connection between violence and the body. Additionally, he remarks that causing pain is not the same as inflicting violence, because the activities of a physician, healer or medicine man may be painful for the patient. Trotha also wants to exclude initiation rituals which cause pain from the realm of violence. (Trotha in Trotha 1997: 9–56). In a follow-up article, Birgitta Nedelmann agrees that Ko¨rperverletzung (bodily injury) is the central element of violence. It is always a question of inflicting and suffering, she continues, but she does not appear to equate this with the actions and fate of the perpetrator and the victim respectively, thus leaving open the possibility that, in one and the same fight, the roles of inflicter and sufferer may alternate. (Nedelmann in Trotha 1997: 59–85). The contribution by Katharina Inhetveen, finally, might be read as another counter-case to the thesis that the use of violence is always contested. She studied hard rock concerts and posits that the violence accompanying them increases the social cohesion of the groups involved and their subculture. This ‘‘violence,’’ however, consists of stage diving and rough dancing inside the crowd, which sometimes leads to nose bleeds and swollen lips and often causes black spots. (Inhetveen in Trotha 1997: 235–60). We may compare these activities to those of the female bouncers (O’Brien et al., ‘‘Negotiating Violence and Gender: Security and the Night Time Economy in the UK’’), who rarely made customers literally bounce. A few Dutch studies illustrate the danger inherent in taking the term violence rather broadly. Anxious to include many things into this concept, Hoogerwerf (1996) often confuses violence with all crime. A collection purportedly focusing on violence in schools deals with such subjects as teasing, scolding, truancy, obstinate lying, arson, consuming drugs and ‘‘completely ignoring another person.’’ Additionally, one of the editors distinguishes a separate category of ‘‘behavior that is not immediately recognized as violence such as obstruction, theft, burglary and destroying someone else’s property.’’ (van der Ploeg in Ploeg/ Mooij 1998: 12). The primary aim of studies such as this one is not to inform but to alarm. The public certainly finds violence bad and worrying, so that is the preferred term. Endlessly broadening the range of behavior that falls within this category enables the researchers to present the results of their investigation as disturbing. Notably, policy-makers should be alarmed, so that they are ready to finance even more research, ensuring continued or new jobs for the investigators and their colleagues.

Violence: Reflections About a Word

17

Intentional Encroachment upon the Physical Integrity of the Body Scholars aiming at a better understanding of society would do wise to return to Riches’ notion of violence as the intentional rendering of physical hurt. My proposal is to modify that definition slightly. I include into the category of violence all forms of intentional encroachment upon the physical integrity of the body. That formula excludes forms of encroachment for medical reasons, analogous to Von Trotha’s argument. The ultimate purpose of medical interventions, although they can be experienced as traumatic, is to preserve the body’s physical integrity. If a nurse sticks a needle into my vein, in order to obtain blood for a check-up, it hurts a little, but it is for my own advantage. Even the amputation of a leg, when necessary to ensure the rest of the body’s survival, would be called violence by few people if any. Of course, there is no absence of any relationship here. As several historians have argued, the relatively intense experience of pain suffered by the inhabitants of preindustrial societies, when compared to our own time, influenced their attitudes to, among other things, violence. A telling example from the diary of the eighteenth-century Amsterdamer Jacob Bicker Raye, who was fond of sensational news and regularly visited public executions, involves a captain whose leg had to be amputated. He had drunk the customary amount of liquor prior to the operation, but he had a hard time when the doctors discovered that the infection extended further than they had thought and it was necessary to saw off another slice. All the while, the captain uttered no scream. The word ‘‘intentional’’ is necessary in the definition, in order to exclude physical harm caused without someone wanting it, as in a traffic accident. Here, as in other cases, social scientists should not be bothered too much by legal categories. The law may hold people criminally responsible for neglect or recklessness in cases of accidental death. By itself, this constitutes an interesting example of criminalization, but it is no reason to subsume the cases in question under the rubric of violence. This is equally implied in the discussion about traffic accidents contaminating Belgian homicide rates (Rousseaux et al., in ‘‘Violence and War: Measuring Homicide in Belgium, (1900–1950)’’). Linguistic peculiarities complicate the problem somewhat. In Dutch, if you say that a person has suffered a violent death, everyone thinks of homicide, never of a car crash. In the latter case, it would always be called death by accident. In English, however, dying in a car crash can be called a violent death. Yet, practically all English-language scholars do not confuse lethal accidents with murders. Not mentioned in any of the studies cited so far is the problem of suicide. I see no reason not to categorize the intentional encroachment upon the physical integrity of one’s own body as violence. Indeed, interesting studies have been conducted, in the social sciences as well as history, about the relationship between homicide and suicide, interpreted as outwardly-directed and inwardly directed violence, respectively. (for example, Jansson 1998).

18

P. Spierenburg

Incidentally, such studies suggest that problems of definition are linked closely to the theoretical position a scholar takes. Many historians and social scientists argue that the long-term development concerning homicide and aggression can best be explained within the framework of Norbert Elias’ theory of civilization, which has also been central to my own work. For this theory, it is a relevant empirical question to what extent people were ready to attack and physically hurt each other; how rashly and frequently they would fight out a conflict by scratching, kicking or stabbing; and whether the propensity to do so changed over time or as we move from one society to another. Even scholars who do not subscribe to Elias’ theory may agree that it is an interesting theme for comparative research to find out how common it was to inflict intentional physical hurt upon others in various societies and that it is important not to confuse this with, for example, pickpocketing, lying, or urinating in a canal. Forms of encroachment upon the physical integrity of the body constitute a meaningful subject for comparative study, regardless of whether historical actors in a particular society view these forms as a unity or present-day scholars choose to categorize them as violence. For diachronic and synchronic comparison, we simply need a minimum amount of analytical concepts. We could of course agree to call forms of intentional encroachment upon the physical integrity of the body just ‘‘x’’ or ‘‘combination i-e-b,’’ but I believe that, given the common understanding of the word, ‘‘violence’’ still is a suitable candidate. It should be emphasized that this use of the term is relatively extended after all, because it includes all encroachments, even relatively minor ones which would not be considered serious enough to be registered by the police and prosecuted. Violence, then, is more than just punishable violence, or, to say it differently, my definition is not a legalistic one which would limit the concept to ‘‘the use of force that has been prohibited by law’’ (compare De Haan, in this volume). For one thing, a non-legalistic definition has the advantage of making violence relatively independent from sensitivities, which may increase or decrease even in a relatively short period. In the Netherlands in the 1990s, for example, the range of punishable violence has been broadened by including into it relatively minor violent incidents which previously were not considered serious enough to merit police attention. (Egelkamp 2002). Both when they got police attention and when they did not get it, they fell within the category of (minor) encroachments upon bodily integrity. Historically, the advantage of a non-legalistic definition of violence is that it includes activities such as officially sanctioned feuds in medieval cities or the vendettas and playful homicides among the Chai.

Remaining Problems No definition can solve every problem that might arise and mine, too, will leave a few border cases for which it is hard to decide whether they are in or out. I am in favor of working definitions and sensitizing concepts that broadly refer to

Violence: Reflections About a Word

19

certain areas of social activity. Moreover, definitions should be compatible with (long-term) historical developments. There will always remain peripheral areas. Thus, I am not quite sure whether inflicting hurt on animals can be called violence, but if we do, it would lead to various methodological problems. For example, is slaughtering a cow in order to eat the meat a violent activity? It is of course true that sensitivity toward the suffering of animals has increased since the eighteenth century, although, given the sheer number of species in the world, very few of them are involved. Even if one does not include hurting animals into the category of violence, it is plain that the increased sensitivity toward the suffering of certain species and the objections against blood sports which developed over the last two and a half century or so, are closely related to the parallel increases in sensitivity toward human suffering and that both can be explained very well within the framework of Elias’ theory of civilization. Some scholars argue that the definition of violence should include the clause ‘‘against the will of the sufferer.’’ No doubt that is often the case. Acceptance of this clause makes it easier to accept terms like structural or psychological violence as well, because its alleged victims suffer from adverse conditions or anguish-inducing actions that were usually not wanted by them. But there are less opportunistic reasons to exclude ‘‘against the will of the sufferer’’ from the definition of violence. First, as already noted, the roles of inflicter and sufferer may alternate, as in a duel with knives. Here, the proponent of an ‘‘involuntary’’ definition might argue that duelists usually attempt to ward off being cut or stabbed, so that every wound they suffer nevertheless is unwanted. A duel, however, is always initiated with mutual consent. Second, we may consider situations such as the hard-rock concerts studied by Inhetveen, involving spectators who suffer injuries because of a voluntary activity. Again a counter-argument is possible, based on my skepticism about calling stage diving and rough dancing violence in the first place. Sports, however, constitute a less ambiguous example. Although it would be absurd simply to equate sport with violence, nevertheless, in the European past, sports often had a violent character. The participants voluntarily put their physical integrity at risk and, if a player did not survive the game, the community accepted it as something that, though regrettable, might happen. (Dunning/ Sheard 1979). Thus, violent clashes can be entered upon voluntarily by all parties involved. The question whether threatening with violence is also violence is not very interesting. A threat, to do a, b or c, is just what it is; next, the threat-maker may either do so indeed or refrain from it. A medieval king, for example, might threaten a town with levying a tax. Whether or not he actually proceeded with it, no one would be surprised to find threats of new taxes as a theme in a historical work about taxation. Similarly, no one would be surprised to find the mere drawing of a knife as a theme in a study of knife fighting. Are all attempts to extend the definition of violence beyond the realm of the intentional encroachment upon the body’s physical integrity to be rejected? I believe they are. In this, I have Riches in my camp, although his arguments are a little different from mine. (Riches 1991: 292–5). Like him, for example,

20

P. Spierenburg

I would be disinclined to rank killing by sorcery as violence, but not because it implies representational rather than situational knowledge, but because the act is impossible, even though the practitioners themselves may believe that it works. In his 1991 article, Riches briefly discusses harmful sorcery and witchcraft, mental violence, symbolic violence as proposed by Bourdieu (see also Nolde, in ‘‘The Language of Violence: Symbolic Body Parts in Martial Conflicts in Early Modern France’’), structural/ institutional violence, ritual violence in games and sports, and violence in which the victim is not fully human. His arguments however, are not always clear and the main criterion appears to be that all these categories are incompatable with the interactionist paradigm that he prefers. One concrete argument, on the other hand, is worth citing. It concerns ‘‘economic violence’’ (which he considers as a subcategory of the symbolic and structural categories): ‘‘. . . the notion that economic constraints constitute violence would imply that there is violence in virtually all human social interaction. To assert something like this may not be unreasonable, but it would be a truism and would render ‘violence’ useless as a concept of anthropological analysis.’’ (Riches 1991: 293). Not only of anthropological analysis, I would add. In this series, structural violence on the one hand and mental or psychological violence on the other, are the most relevant concepts to take issue with. Not only do they extend the category of violence to the point of uselessness for scholarly analysis, they are also decidedly ideological. Their main function is not to describe, but to stigmatize. Authors using these concepts rely on the negative connotations of the word violence, in order to mark certain practices as notoriously bad and hence to legitimize action against these practices. Both concepts are variants of the Kantian logic applied to ‘‘assaultive speech,’’ as discussed by Rauch. Despite this weakness, the term ‘‘structural violence’’ has been used frequently and it would be impossible even to take a glance at the entire literature about it. The Norwegian sociologist Johan Galtung, however, has long been acclaimed as a major theorist in this matter. In a programmatic article, Galtung explicitly subscribes to an ‘‘extended concept of violence.’’ He defines violence as ‘‘the cause of the difference between the potential and the actual, between what could have been and what is.’’ (Galtung 1969: 168. He himself designates this essay as programmatic. Cf. Galtung 1972: 29). Thus, if someone dies of tuberculosis in the eighteenth century, it is just bad luck, but if someone dies of tuberculosis in the twentieth century, when this could have been prevented, it is violence. Galtung considers structural violence as the opposite of ‘‘personal violence.’’ Personal violence occurs when one husband beats his wife, structural violence when one million husbands keep one million wives in ignorance. A ‘‘condition of structural violence’’ is the same as ‘‘social injustice.’’ (Galtung 1969: 171). More than any other element, Galtung’s phrase ‘‘what could have been’’ betrays the ideological roots of his conceptual scheme. Indeed, the concept of ‘‘structural violence’’ enjoyed great popularity during the revolutionary ferment of the late 1960s and 1970s. Opposition groups in the Western world were

Violence: Reflections About a Word

21

divided into those advocating non-violent resistance and those willing to employ physical force against the established order. The latter groups faced the problem of legitimating violent resistance against a ‘‘system’’ that was ostensibly based on democratic elections. The concept of structural violence helped solving that problem. (It should be added that Galtung rejects the argument that ‘‘personal violence’’ is necessary against ‘‘structural violence’’: Galtung 1969: 181). This concept referred to much more than just the police shooting at a bank robber. All the inequality, injustice and hardship brought about by capitalism, imperialism and the state organizations upholding them actually was also a form of violence – so the argument ran. The violence that was inherent to the system justified the counter-violence of revolutionary groups: not only of those who merely put up barricades and threw a few stones at the police, but also the kidnappings and killings by the RAF or the Brigate Rosse. When performing such actions, members of these groups believed that the exchange was even. My point is not that their activities should be condemned – although I suspect that most of us do – but that a concept whose primary function is political justification is worthless for scholarly analysis. Therefore, scholars should reject the notion of structural violence outright. Even some of the political activists of the 1960s and 1970s, it seems, realized that this notion had to be taken symbolically rather than literally. In the introduction to a recent anthology, for example, the editors remind us that ‘‘Dom Helder Camara, the ‘little red archbishop’ of Recife, Brazil, railed fearlessly against the military government’s attacks on ‘violent’ landless peasants by reminding those in power of the ‘violence of hunger’ and the ‘atomic bombs’ of sickness and destitution.’’ (Scheper-Hughes/ Bourgois 2004: 2). The use of the term atomic bombs, in particular, indicates that Camara considered the implicit reference to structural violence as a metaphor. Of course, the opposite notion – that the activities of agents representing the state should never be designated as violent – is equally untenable. The very concept of Gewaltmonopol and its translation as ‘‘monopoly of violence’’ imply that the monopoly is ready to be used. (On monopolization of violence over the very long term: Goudsblom 1998). Its most destructive use, throughout history, has been in wars between states, but the uses also include executioners whipping a criminal or policemen shooting in the leg of a fugitive suspect. The fact that state violence is officially pronounced legitimate and often considered as a necessary evil does not make it non-violence. Its level of toughness, on the other hand, can vary over time. In the modern period, state rulers increasingly had to yield to public pressure to moderate the internal use of their monopoly. On a methodological note, it may be added that the distinction between interpersonal and state violence implies that the two should always be kept quantitatively separate. Some scholars include into the homicide rate police shootings which were prosecuted as illegal activities. (E.g. McKanna 1997). This cannot be accepted. When the police, on duty, are encroaching upon a person’s bodily

22

P. Spierenburg

integrity, this is state violence, regardless of whether it is moderate and within legal bounds or excessive and indictable. The argument against ‘‘psychological violence’’ runs along similar lines as that against its structural counterpart. If every activity that produces mental anguish in another person counts as violence, the Kantian logic leads to all imaginable extremes. Take just one example. There are countries in today’s world where women are obliged always to wear a veil in public. If a woman would appear in the streets with her hair visible, this would produce mental anguish in many passers-by. The logic inherent to the concept of psychological violence implies that the unveiled woman is violent. In Iran, after Khomeiny and his followers took over, religious police patrolled public places and often beat up women for having a few curls visible. Such practices, incidentally, have also occurred in European history. In the 1490s, when the preacher Savonarola was Florence’s de facto ruler, his bands of fanciulli, boys of about sixteen who had sworn not to have themselves sodomized, roamed the city streets to harass prostitutes, lavishly dressed women, gamblers, blasphemers and, especially, suspected sodomites. (Rocke 1996: 210–11). If we combine the notion of psychological violence with Radcliffe-Brown’s claim that physical force exercised by state agents cannot be called violence, we would have to consider the beatings by the religious police and the fanciulli’s harassments as merely being part of ‘‘political organization’’ whereas the half-veiled women and the men interested in their own sex were engaging in a violent activity. Most people would find this absurd and it certainly is the opposite of sound scholarly categorizing. Incidentally, the situation in Iran easily meets Galtung’s one-million-men-keeping-one-million-women-ignorant criterion for structural violence. The notions of structural and psychological violence, then, as well as most other extended definitions of violence, including those discussed in De Haan’s contribution, are similarly flawed. They all basically derive from the following ‘‘logic’’: (1) Violence is a serious evil; (2) some other evils in society are equally bad; (3) consequently, these other evils should also be called violence. If applied to a still wider set of social activities, this flawed logic would leave scholars with few usable words at all. Verbal violence, the final notion left for critical examination, derives not so much from the flawed logic just described, but rather from a misguided attempt to reach the minds of past actors. Verbal violence refers to all forms of insult, scolding, slander, derogatory remarks or even insinuating gestures. This notion is left undiscussed by Riches. In fact, it is largely absent from the writings of contemporary social scientists and for an obvious reason: it is closely associated with the experience of European villagers and townspeople of the fourteenth through eighteenth centuries (and perhaps even earlier but we have few good sources from before c. 1300). Because preindustrial Europeans viewed insults as identical or akin to physical attacks, some historians argue that words of abuse should be categorized as ‘‘verbal violence.’’ Foremost among these historians is Gerd Schwerhoff. According to his radical formula, ‘‘whether the protagonists

Violence: Reflections About a Word

23

fought with sharp words or sharp knives does not make a difference.’’ (Schwerhoff 2002: 119. This article takes issue with Spierenburg 2001. See also my reply: Spierenburg 2002). The word ‘‘does,’’ in the present tense, should not conceal the fact that Schwerhoff is talking about people from the past. It did not make a difference to them whether they were hurt with words or knives, but even that thesis is questionable. Schwerhoff largely bases his argument on the well-known preoccupation with honor, found with men and women of the preindustrial period and even later. Notably for men, honor and violence were closely linked. Schwerhoff rightly stresses that this preoccupation has been demonstrated in ‘‘international research.’’ His depiction of a process of escalation closely matches that as described by Dinges (1994). (See also Spierenburg 1998). According to Schwerhoff, many men experienced an encroachment upon their honor with words or one with a weapon as an equally serious matter. ‘‘In daily life, insulting words often marked the beginning of a spiral of violence which could end with manslaughter.’’ (Schwerhoff 2002: 119). To contemporaries, personal honor was at least as valuable as the integrity of the body. Honor constituted a sort of second skin, that had to be defended against attack just as the biological skin. I have no problem with these observations, which are important in their own right. The world view and values of contemporaries and the way they defined social situations are meaningful subjects for historical research, but they cannot form a basis for analytical categorization. It is even doubtful whether preindustrial people really failed to draw a line between insult and physical attack. It is true that many confrontations involved an escalation beginning with words or gestures and ending with bloodshed, but does this mean that, in each individual case, the contestants and the spectators did not know at what stage in the escalation they had arrived? And were contemporaries unable to differentiate between situations in which their honor was encroached upon by words and situations in which this was done by physical attack? I doubt it. Moreover, to speak of verbal violence because this made sense to early modern Europeans, means that we renounce a useful instrument for diachronic comparison. If the term violence would include verbal abuse in some periods and not in others, the word becomes useless as a category for comparing different societies. Rauch’s philosophical statement that words are words and bullets are bullets does not imply that there are no persons, today or in the past, who think otherwise. He simply maintains that the distinction is analytically important. Once more, I conclude that it is meaningful to compare the frequency, seriousness and character of encroachments upon the physical integrity of the body over time. As a supplementary theme, the seriousness with which verbal abuse was viewed by people in different periods should be studied for what it is. If, on the other hand, a historian who has examined the activities of a court that prosecuted verbal abuse, combines insults into one overall category with assault and homicide, in order to distinguish these from theft or vagrancy, no particular objections apply. That is, as long as the overall category is not simply

24

P. Spierenburg

called violence. Alternatively, some historians refer to assaults and insults as ‘‘offenses against the person.’’ This would seem acceptable at first sight: you have crimes against the person and crimes against property. Upon closer examination, however, the two terms do not match logically. In the second case, the property itself is not the victim of a crime. The owner is; when things are stolen, it is a person, too, who suffers. In addition, property crimes usually include robberies. Criminologists often consider robbery together with assault and homicide as violence and with good reason: robbery involves an encroachment upon the physical integrity of the body, or at least the threat of it. Distinguishing different groups of offenses, therefore, is a matter of convenience, but it should not be confused with theoretical analysis.

Conclusion Current definitions of violence range from narrowly restrictive – the use of force that is prohibited by law – to overly extended ones, which include any number of social evils. Extended definitions all suffer from an inherent weakness: they are so broadly inclusive that they make the concept of violence useless for scholarly analysis. Moreover, they are based on a barely hidden agenda which, if not straightforwardly political, is to stigmatize activities or social groups and alarm the public. Perhaps, today’s extended definitions owe something to the modern idea that everyone should be safeguarded against any kind of risk, misfortune or unhappy experience, also in the private sphere. But unhappiness in the private sphere is the fate of most of us at some point in our lives. Two lovers are united for ever, it seems, but suddenly one of them thinks it is time to move on. The partner that stays behind is very sad, feeling angry and frustrated. Should we say that the other has committed violence? Apart from the fact that the person in question will disagree, it is a meaningless comment on the situation in which both ex-lovers find themselves. By contrast, legalistic definitions of violence leave out too much. They are anachronistic, making no room for such historically accepted practices as feuding. Even in a contemporary context, they fail to acknowledge that some encroachments upon bodily integrity are considered too minor to be punishable and that the line between punishableness and unpunishableness fluctuates. Most seriously, legalistic definitions of violence fail to acknowledge that actions by state agents, usually considered legitimate, can involve intentional physical hurt. My demarcation of the concept of violence lies in between the legalistic and extended poles, but not necessarily in the middle. I hope to have made clear that there is much to say for defining violence as the intentional encroachment upon the physical integrity of the body. Within that definition, it is imperative to distinguish – at least in European and Japanese history since about the sixteenth century and in American history since the beginning of the

Violence: Reflections About a Word

25

nineteenth-between interpersonal, or private violence on the one hand and state violence on the other. Acknowledgment I am grateful for their comments to the other contributors in this volume and to Jonathan Simon.

References Aijmer, Go¨ran, & Abbink, Jon, eds. (2000). Meanings of violence: A cross cultural perspective. Oxford, New York: Berg. Dinges, Martin (1994). Der Maurermeister und der Finanzrichter: Ehre, Geld und soziale Kontrolle im Paris des 18. Jahrhunderts. Go¨ttingen: Vandenhoeck & Ruprecht. Dunning, Eric, & Sheard, Kenneth (1979). Barbarians, gentlemen and players: A sociological study of the development of rugby football. Oxford: Oxford University Press. Egelkamp, Margarethe Maria (2002). Inflation von Gewalt? Strafrechtliche und kriminologische Analysen von Qualifikationsentscheidungen in den Niederlanden und Deutschland. Groningen: University of Groningen. Galtung, Johan (1969). Violence, peace and peace research. Journal of Peace Research vol. 6: 167–91. Galtung, Johan (1972). Eine strukturelle Theorie des Imperialismus. in: Senghaas, Dieter, ed. Imperialismus und strukturelle Gewalt: Analysen u¨ber abha¨ngige Reproduktion. Frankfurt a.M.: Suhrkamp, 29–104. Goudsblom, J[ohan] (1998). De paradox van de pacificatie. Amsterdams Sociologisch Tijdschrift vol. 25,3: 395–406. Hoogerwerf, A (1996). Geweld in Nederland. Assen: Van Gorcum. Jansson, Arne (1998). From swords to sorrow: Homicide and suicide in early modern Stockholm. Stockholm: Almqvist & Wiksell International. McKanna, Clare V., Jr. (1997). Homicide, race and justice in the American West, 1880–1920. Tucson: University of Arizona Press. Ploeg, J.D. van der, & Mooij, T., eds. (1998). Geweld op school: Achtergronden, omvang, oorzaak, preventie en aanpak. Rotterdam: Lemniscaat. Rauch, Jonathan (1993). Kindly inquisitors: The new attacks on free thought. Chicago, London: University of Chicago Press. Riches, David, ed. (1986). The Anthropology of violence. Oxford: Basil Blackwell. Riches, David (1991). Aggression, war, violence: Space/ time and paradigm. Man, New Series vol. 26: 281–98. Rocke, Michael (1996). Forbidden friendships: Homosexuality and male culture in Renaissance Florence. New York, Oxford: Oxford University Press. Scheper-Hughes, Nancy, & Bourgois, Philippe, eds. (2004). Violence in war and peace. Malden, MA: Blackwell. Schwerhoff, Gerd (2002). Criminalized violence and the process of civilisation: A reappraisal. Crime, Histoire & Socie´te´s/ Crime, History & Societies vol. 6,2: 103–26. Spierenburg, Pieter, ed. (1998). Men and violence: Gender, honor and rituals in modern Europe and America. Columbus OH: Ohio State University Press. Spierenburg, Pieter (2001). Violence and the civilizing process: Does it work? Crime, Histoire & Socie´te´s/ Crime, History & Societies vol. 5,2: 87–105. Spierenburg, Pieter (2002). Theorizing in Jurassic Park: A reply to Gerd Schwerhoff. Crime, Histoire & Socie´te´s/ Crime, History & Societies vol. 6,2: 127–8. Trotha, Trutz von, ed. (1997). Soziologie der Gewalt. Cologne: Ko¨lner Zeitschrift fu¨r Soziologie und Sozialpsychologie, Sonderheft 37.

Violence as an Essentially Contested Concept Willem de Haan

Introduction Scientific understanding of human violence is one of the most urgent tasks of our time. Given this sense of urgency, it is shocking how little we seem to know about it (Turpin & Kurz, 1997: 1–2). In fact, even among scholars there is not even the beginning of an agreement concerning a definition of what exactly ‘violence’ is (Stanko, 2003: 3). One of the most common ways of defining violence is to only consider forms of criminal violence and to argue that violence is the use of force that has been prohibited by law (Riedel & Welsh, 2002: 3). However, whilst ‘violence’ might conventionally connote physical attack, the notion of physical violence represents a surprisingly broad spectrum of incidents (Waddington, Badger & Bull, 2004: 149). This is obvious, even within the legal point of view which, typically, defines violence rather technically as ‘‘the actual or threatened, knowing or intentional application of statutory impermissible physical force by one person directly against one or more other persons outside the contexts both of formal institutional or organizational structures and of civil or otherwise collective disorders and movements for the purpose of securing some end against the will or without the consent of the other person or persons’’ (Weiner, 1989: 37–38). Thus, it may appear to be self-evident what ‘violence’ is but, in reality, ‘violence’ is ‘‘a slippery term which covers a huge and frequently changing range of heterogeneous physical and emotional behaviors, situations and victim-offender relationships’’ (Levi & Maguire, 2002: 796). Social scientists attempting to come to grips with the phenomenon find the concept of violence ‘‘either under-, or over-defined, or both.’’ (Bauman, 1995: 139) Even though two overarching assumptions seem to be that: 1. violence is typically assumed to be motivated by hostility and the willful intent to cause harm and 2. violence is – legally, socially, or morally – deviant human activity, it Willem de Haan Professor of Criminology, University of Groningen, The Netherlands; Department of Criminal Law and Criminology, Faculty of Law, PO Box 716, 9700 AS Groningen, The Netherlands [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

27

28

W. de Haan

has been argued that these assumptions are based on ‘conceptual quicksand’ in which important questions about the defining attributes of violence have slipped away. The very social and policy imperatives that have driven sociological analysis of violence have molded and warped our understanding of it and spawned a conception of violence that is biased and morally charged and at the same time as it is clouded and unwieldy (Jackman, 2002: 387). Nearly all inquiries concerning the phenomenon of violence demonstrate that violence not only takes on many forms and possesses very different characteristics, but also that the current range of definitions is considerable and creates ample controversies concerning the question what violence is and how it ought to be defined (Heitmeier & Hagan, 2002: 15). Since there are so many different kinds of violence (Reidel & Welsh, 2002: 1) and since violence is studied from different actor perspectives (i.e. perpetrator, victim, third party, neutral observer), existing literature displays a wide variety of definitions based on different theoretical and, sometimes even incommensurable domain assumptions (e.g. about human nature, social order and history). In short, the concept of ‘violence’ is notoriously difficult to define because as a phenomenon it is multifaceted, socially constructed and highly ambivalent. Violence is multifaceted because there are many different forms of violence, which are exhibited in a wide range of contexts. It may, for example, be distinguished in ‘youth violence’, ‘gang violence’, ‘school violence’, ‘street violence’, ‘teen violence’, ‘dating violence’, ‘intimate violence’, ‘domestic violence’, ‘workplace violence’, ‘suite violence’ (Punch, 2000), ‘urban violence’, ‘interpersonal violence’, ‘random violence’, ‘racist violence’, ‘media violence’, ‘mimetic violence’, ‘systemic violence’, ‘symbolic violence’, ‘structural violence’ or even ‘apocalyptic violence’ (Hamm, 2004). Violence can be physical (‘aggression’, ‘abuse’ or ‘assault’), but it can also be verbal (‘bullying’, ‘humiliation’ or ‘intimidation’). It can be overt but also covert like in language and literacy, abstraction, interpretation and representation, and in the violence of ‘censure’ (Valier, 1997). Violence can be individual or collective, interpersonal or institutional, national or international, symbolic or structural. The context may be private or public and the victims may be family members, acquaintances or strangers. Based on the offender’s motive, violence may be angry, impulsive, hostile, expressive, dispute-related, instrumental, or predatory. ‘‘Some incidents occur, more or less, ‘out of the blue’, whereas others occur within some form of relationship in which conflict escalates. Some incidents are concluded in a few moments, whereas others evolve into long-term conflict relationships. Non-physical attacks can be made against the gender, race or sexuality of the victim, or – as in some of these cases – their professional integrity.’’ (Waddington, Badger & Bull, 2004: 159). Instead of or in addition to physical injury; violence can have mental (‘psychological’), social and/or material consequences and there is seems no simple relationship between the apparent severity of a violent act and the impact it has upon the victim. Violence is socially constructed because who and what is considered as violent varies according to specific socio-cultural and historical conditions.

Violence as an Essentially Contested Concept

29

While legal scholars may require narrow definitions for punishable acts, the phenomenon of violence is invariably more complex in social reality. Not only do views about violence differ, but feelings regarding physical violence also change under the influence of social and cultural developments. The meanings that participants in a violent episode give to their own and other’s actions and experiences vary and can be crucial for deciding what is and what is not considered as violence since there is no simple relationship between the apparent severity of an attack and the impact that it has upon the victim. For example, in some cases, verbal aggression may prove to be more debilitating than physical attack. Violence is also highly ambivalent in the ways it is socially sanctioned, legitimized and institutionalized, as well as how it is culturally transmitted and experienced as graphic illustrations of violence abound in world press photos, on television screens, in the movies and in video games. Depending on the context and perspective, violent actions may either be condemned and considered immoral, illegal and disruptive or admired and considered moral, legal and functional. As a topic, violence is ‘deeply emotive’, ‘‘coloring the political and criminal justice responses not just to violence but to crime in general’’ (Levi & Maguire, 2002: 795). However, precisely because violence is so highly ambivalent an increased sensitivity and reflexivity is required in both theory and research on violence (Heitmeier & Hagan, 2002). Not surprisingly, controversies occur and recur about both the substance of the concept and the scope of the definition of violence. In this chapter, debates about various definitions of violence will be interpreted against the backdrop of what the philosopher Gallie (1956) has called the ‘essentially contestedness’ of concepts, i.e. the claim that debates about concepts can never reach closure as invariably good theoretical, methodological and moral arguments can and will be made in favor of one or another definition. However, as we will see, the arguments against restrictive definitions of violence are mostly of theoretical and moral nature. Theoretical arguments tend to be and focused on the question of what violence is. For example. what is the ‘essence’ of violence and how should violence ‘essentially’ be demarcated from aggression? Moral arguments tend to be focused on the question of what the scope of violence is. For example. should sexual abuse, hate speech or bullying be recognized as serious forms of violence? Instead of discussing the question of what violence essentially consists of, this chapter will discuss the scope of the notion of violence and focus on the question whether the notion of violence should best be defined by limiting what qualifies as ‘violence’ to physical attack or should be extended to a broader definition. This, for example, would be a definition, ‘‘which implicitly claims that a ‘violent act’ can be anywhere along a continuum running from an angry and hostile glare (which in certain circumstances can cause a degree of alarm), through verbal abuse, a verbal threat, threatening gestures, a single blow, an attack causing minor injuries, an attack causing major injuries, to an attack causing death.’’ (Waddington, Badger & Bull, 2004: 145). After exploring arguments for and against either a limited notion of violence to be defined in

30

W. de Haan

a rather restrictive way or an extended notion of violence to be defined in a more inclusive way, it will be concluded that, depending on the contexts of discovery and justification, valid arguments are feasible for either inclusive or restrictive definitions of violence. Any definition of violence, however, should be considered as a temporary outcome of theoretical debate – an outcome, which may or may not prove to be useful in future research.

Defining Violence Restrictively In order to provide a conclusive definition, restrictive definitions of violence are being offered which tend to limit violence to physical attack. I will briefly discuss three examples. In The Anthropology of Violence (1986), anthropologist David Riches defines violence as ‘‘an act of physical hurt deemed legitimate by the performer and illegitimate by (some) witnesses’’ (p. 8). According to Riches this definition is compatible with ‘‘basic properties’’ of violence, which have cross-cultural validity, i.e. that the performance of violence requires relatively little by way of specialized equipment or knowledge, that the practice of violence is highly visible to the senses and that, therefore, violence as such is unlikely to be mistaken. Discrepancies in basic understandings amongst those implicated in the performance of a violent act are likely to be minimal. Nevertheless, violence is inherently liable to be contested on the question of legitimacy. In one of the most ambitious attempts at understanding and preventing violent behavior, the National Research Council of the American Academy of Sciences defined violence simply as ‘‘behaviors by individuals that intentionally threaten, attempt, or inflict physical harm on others’’ (Reiss & Roth, 1994: 2). In this way, the panel limited its considerations to intentional, physical, interpersonal violence, thereby excluding forms of interpersonal violence (such as those inflicted by intimidation and emotional abuse) resulting in non-physical harm as well as forms of violence in the institutional and structural spheres. However, while their definition seems to effectively restrict violence to behaviors that are interpersonal, inflict or threaten physical harm, and are motivated by harmful intent, these restrictive criteria are relaxed on an ad hoc basis when they become inconvenient for the inclusion of features that have come under scrutiny in a specific line of research e.g. psychological injuries in the context of family. In this way, Reiss & Roth’s formulation of a ‘precise definition’ merely highlights the points of emphasis – as well as the inconsistencies – that have marked current violence research. Something similar is the case with Neidhardt (1986) who also argues that a restrictive definition of violence is most appropriate both from a sociological and a historical perspective. It is most appropriated from a sociological perspective because physical violence is a universal media of communication. It is also most appropriate form a historical perspective because the monopoly

Violence as an Essentially Contested Concept

31

of violence always applied to physical rather than mental violence. However, even though Neidhardt seems to consider the problem of the definition of violence solved, he does allow for the possibility that in different contexts valid arguments for other definitions may be feasible (1986: 135). Unlike some (e.g. Nedelmann, 1995: 8; von Trotha, 1997: 20) who seem convinced that the question of how to define violence has been adequately solved, at least for the time being, by defining it as bodily harm by physical use of force, Neidhardt admits that nothing raises so many doubts as the definition of violence. The most substantive arguments against restrictive definitions of violence are not of a theoretical rather than a methodological and empirical nature. In his Aspects of Violence, Schinkel (2005) takes issues with Riches’ definition of violence by arguing that this definition has as its major disadvantages 1. that the restriction to ‘‘physical hurt is arbitrary and cannot be adequately grounded, not even in ordinary language; 2. that the initial focus on ordinary language leads to a neglect of forms of violence that use ordinary language a vehicle of disguise; 3. that problems arise out of Riches’’ incorporation of the concepts of legitimacy and illegitimacy in the definition of violence; and 4. that this definition – like any definition which requires violence to be an ‘act’ – is problematic and untenable. Riches’ incorporation of the concepts of legitimacy and illegitimacy in the definition of violence raises problems because it is unclear how the discrepancy in basic understandings amongst those implicated in the performance of a violent act can be ‘likely to be minimal’ when an act of violence by definition is deemed legitimate by the performer and illegitimate by (some) witnesses? Others have argued that differences between legitimate and illegitimate forms of violence tend to be ideologically or morally constructed and have more to do with variations in the extent of social outrage and in the political denial or awareness of these particular forms of violence than with the seriousness of harms. Indeed, injuries from other kinds of harms are often no less (and sometimes even more) painful and tragic than those from acts that have been defined as ‘illegally violent’ (Barak, 2003: 26). Moreover, the usual definition of violence – the use of force toward another that results in harm, narrowly conceived of as physical pain and suffering, tends to omit several critical elements of harm: ‘‘First, it excludes the emotional and psychological pain that result from domination of some over others. Second, it tends to focus on the visible, intentional, interpersonal harm between individuals, while excluding harm against individuals by institutions or agencies. Third, it ignores the violence of social processes, which produce systematic social injury, such as violence perpetuated through institutionalized racism and sexism. Fourth, it excludes the ‘symbolic violence’ of domination, that ‘‘gentle, invisible form of violence, which is never recognized as such, and it not so much undergone as chosen, the violence of credit, confidence, obligation, personal loyalty, hospitality, gifts, gratitude, piety’’ (Bourdieu, 1977: 192)’’ (Henry, 2000: 2). In order to widen our field of vision to incorporate under violence a diverse array of actions which are integral to social life, we need a systematic,

32

W. de Haan

autonomous, generic conception of violence, freed from the biases and inconsistencies of past research (Jackman, 2002: 387).

Defining Violence Inclusively In an attempt to introduce a more systematic, comprehensive analysis of violence several propositions have been made for a more extended notion of violence to be defined in a more inclusive way. I will briefly discuss three recent examples of such inclusive definitions of violence presented by Barak (2003), Jackman (2002) and Henry (2000) respectively. Barak (2003) has made an attempt to be conceptually inclusive, i.e. to take into account the full range of harms associated with a variety of interpersonal, institutional, and structural relationships and behaviors and to not exclude any forms and expressions of violence, whether they refer to individual acts, institutional arrangements or structural conditions as well as whether or not they are prohibited by law. To this end, he adopts a definition of violence as ‘‘any action or structural arrangement that results in physical or nonphysical harm to one or more persons’’ (Iadicola & Shupe, 1998: 26; quoted in Barak, 2003: 26). Jackman (2002) also proposes a generic definition of violence that focuses unequivocally on the injuriousness of actions, detached from their social, moral or legal standing. This definition provides a consistent, autonomous basis for identifying the full population of injurious social behaviors, purely on the basis of their indigenous behavioral attributes. It includes all actions that directly inflict injury as well as those that either threaten or result in injury. It specifies that injurious actions and outcomes may take many forms, immediate or delayed, certain or probabilistic. More significantly, this definition sets no constraints on the motivations of either the victim or the agent, and is agnostic about whether the behavior is unusual or commonplace and whether it meets with society repudiation, disinterest, acceptance or admiration. It thus provides a stripped-down template to identify all behavior that inflict, threaten, or cause injury in order to pursue new questions about violence in social life. In a similar vein, Henry (2000) suggests that a more inclusive, integrated definition of violence is necessary, which replaces the term ‘force’ with ‘power’ and takes a more comprehensive view of harm. Violence is thus defined as ‘‘the use of power to harm another, whatever form it takes’’ (Henry, 2000: 3). In this case, harm is not only physical pain and suffering. It ‘‘can also occur along many dimensions beyond the physical to include psychological or emotional, material or economic, social or identity, moral or ethical, and so on. Within each dimension, the harm can be of two kinds: ‘harms of reduction’ and ‘harms of repression’ (Henry and Milovanovic, 1996: 103). Harms of reduction remove something from a person’s existing status as a human being. For example, physical harms or reduction produce bodily pain or loss (of blood, organs, limbs, physical functioning). Material harms of reduction remove some of the

Violence as an Essentially Contested Concept

33

person’s economic status (property, wealth, money). Psychological harms of reduction have destructive effects on the human mind and weaken a person’s emotional or mental functioning (such as in posttraumatic stress syndrome). Social and symbolic harms of reduction lower a person’s social status (by violating their human rights, sexuality, social identity). Moral or ethical harms of reduction corrupt standards of concern for the well-being of others (as in hate, pressure to cheat, and the like). In contrast, harms of repression reveal how the exercise of power acts to systematically limit another person’s capability of achieving higher levels accomplishment along any of these dimensions. (ibid.: 116). Violence, then, is the exercise of power over others by some individual, agency, or social process that denies those subject to it their humanity to make a difference, either by reducing them from what they are or by limiting them from becoming what they might be’’ (Henry, 2000: 5). Perhaps not surprisingly, these recent inclusive definitions of violence have been met with similar skepticism, especially in Germany where the burden of history renders researchers particularly sensitive to the moral and political implications of definitions of violence. A case in point is von Trotha (1997) who argues that given the moral and political implications of the study of violence, it is especially urgent to avoid any conceptual and descriptive ambivalence, because as soon as one allows imprecision in questions concerning violence, one has not only lost out scientifically but has also taken a politically and morally suspect direction (von Trotha, 1997: 24). More recently, Waddington, Badger & Bull (2004) also found considerable drawbacks associated with an inclusive definition of violence. Whilst for them there is no doubt that the inclusive definition of ‘violence’ is valuable as a methodological tool allowing respondents to express the subjective meaning of their experiences, they do feat that such a definition may become so broad that almost any situation that any person finds disagreeable would qualify as a form of ‘violence’. Secondly, an inclusive definition of ‘violence’ could obstruct our understanding of violence by cloaking the complexities of the phenomenon in a definitional fiat. And thirdly, privileging subjective meaning may involve a risk of circular reasoning when, for example, one concludes that a person must have acted ‘violently’ because others felt frightened by him. Thus, analytically, so broad a definition of ‘violence’ could confuse connotation and denotation, and leave analysts referring to very different events and experiences whilst using the same conceptual apparatus.

Philosophical Clarification Not only in the social sciences but also in philosophy, there is a tendency to, on the one hand, eagerly extend notions of violence and, on the other hand, exhibit a marked reluctance to view violence beyond what may be regarded as the

34

W. de Haan

indispensable point of reference for all analysis of violence: bodily harm by physical use of force. One of the most frequently quoted and reprinted1 definitions of violence has probably been formulated by the philosopher Garver (1968). According to Garver, one cannot comprehend the violence that surrounds us, if one thinks of it as necessarily physical or as necessary illegal. In his view, a successful account of violence has to ‘‘make it clear that violence (1) is a matter of degree, (2) can be social or institutional as well as personal (3) can be psychological as well a s physical, (4) has moral implications when it is social that are radically different from those that it has when it is personal, (5) can be legal as well as illegal (6) needs, when it is social, to be discussed in conjunction with law and justice, and (7) can in principle be excused, however, personally abhorrent one may find it’’ (Garver, 1972: 39). For Garver defining violence poses ‘a typical philosophical question’ to which his answer is ‘‘that violence in human affairs comes down to violating persons’’(1977: 270). Against the more common philosophical view that violence entails causing injury through the use of vigorous physical force, Garver suggests that we focus on violence not as a matter of physical force but rather as the violation of a person, i.e. the violation of basic human rights which are undeniably, indissolubly, connected with being a person: the right to one’s body and the right to autonomy (270). Garver illustrates how persons can be violated either with respect to their bodies (physical violence) or with respect to their ability to make their own decisions (psychological violence) and he shows that each kind of violence has both personal and institutionalized forms. In addition, he contrasts overt to covert forms of violence, which he calls ‘‘. . . the quiet forms which do not necessarily involve any overt physical assault on anybody’s person or property.’’ (1977: 272) As illustration of such ‘quiet violence’, Garver refers to ‘institutionalized violence’ (p. 273), i.e. violence which is not be personal, but done while acting as a faithful agent of an institution. Reactions of other philosophers to Garver’s propositions are mixed. Litke (1992) applauds Garver’s account for giving ‘‘a useful way of viewing a vast range of very diverse and often spectacular human behavior . . . which enables us to see through the diversity and spectacle to certain essential features’’ (1992: 174). But Platt (1992), in response to Garver’s categorization of ‘quiet violence’, comments, sarcastically ‘‘that one can hardly avoid wondering if the discovery of ‘non-violent violence’ is to be the next step in the process.’’ In fact, Garver’s remark that he rather leaves it to the reader to ponder whether all sex acts are acts of violence, suggests to Platt that, indeed, the transition from ‘quiet violence’ to ‘non-violent violence’ is well underway. (Platt, 1992: 186). Although Platt admits that there is much to be said for the extended meaning, which has been given to the notion of violence and has effectively sensitized 1

For an overview of the reprints, revisions and expansions of Garver’s original article, see Betz (1977: 339, note 1).

Violence as an Essentially Contested Concept

35

large numbers of person to the morally dubious status of many social structures and practices, he also argues that extending the meaning of violence leads us in the wrong direction in the final analysis. In his view, the very plasticity of language which allows us to bend or extend the meaning of terms also makes it possible for us to obscure significant distinctions, thus managing to confuse ourselves and others as well as increasing the likelihood of producing fallacious lines of reasoning. According to Platt, the combination of relatively vague descriptive content, coupled with a negative moral and emotional connotation makes the word ‘violence’ ideal for use in polemic discourse. And it is, therefore, not surprising to him that, generally speaking, the extended use of the term violence tends to occur in works that share an obvious intent to change opinions (i.e. ‘consciousness raising’) (1992: 187). In his effort to discredit Garver’s extended approach to violence, Platt has clearly overlooked Garver’s (1988) more recent suggestion that: ‘‘It is useful to think of violence as an ‘essentially contested concept’ and to make use of the insights of W.B. Gallie when trying to understand its normative dimension . . . as well as the contested nature of its applications.’’ (220). Garver refers to Gallie (1956) as the philosopher who has provided convincing arguments to counter the prejudice easily engendered by an overly simplistic empirical or scientific outlook, that any concept which cannot be clearly and unambiguously applied is bound to be confused. Gallie suggests that some concepts are too complex to be adequately captured by a single definition requiring, instead, a number of competing characterizations differing according to which elements are regarded as most salient or fundamental for determining the concept’s extension. Whereas Garver does not fully develop his suggestion, a former student of his (Reilan, 2001) has recently shown, more specifically, ‘‘that the vagueness that the concept of rape has acquired can best be understood in terms of essential contestability, in something like W.B. Gallie’s sense (1956), and that once we understand rape as an essentially contested concept, we will see that extending the boundaries of the concept is part of the proper use of the term’’ (p. 44) For Gallie, an essentially contested concept is a concept that is used to make an appraisal, a value judgment that attaches to all objects that fall under the concept’s scope and is based on a complex set of characteristics drawn from some original exemplars or paradigms, which everyone agrees fall within the concept’s extension. While everyone agrees that objects must bear a resemblance to the paradigms in order for the concept to apply properly, there is no similar agreement over which characteristics of the paradigms count the most or how various characteristics figure into the appraisal. Instead, there are competing understandings of what should fall under the scope of the concept. Adherents to each rival understanding of the concept are aware of other understandings, and maintain their definitions in the face of these rivals (Gallie, 1956: 168–181). According to Reilan, rape can be understood as an essentially contested concept because there is no question that rape has a strong evaluative character. Moreover, there are clear paradigm cases of rape (i.e. examples which everyone

36

W. de Haan

agrees fall under the scope of the concept) that have a complex array of characteristics, all of which are regarded as significant features of the individual cases. It is also clear that different theorists emphasize different aspects of the paradigms as being significant or essential, such that while they agree that the paradigms are instances of rape, they disagree about what makes them instances of rape. Is it the use of force or coercion, the presence of power differentials, the absence of consent, or the degradation and objectification of the victim? The vagueness of the concept of violence, however, is not due to any epistemological consideration; it is, rather, an essential vagueness due to the evaluative character and the ‘emotive meaning’ of the concept (Burgess-Jackson, 1995: 421). Which definition one accepts has significant normative import because the definitional debate is, in effect, a debate over which borderline cases ought to be subjected to the same sort of negative appraisal as the paradigms. In this debate, the feature that has been historically singled out as most significant need not be taken as essential. By the same token, a feature that seems currently to be the most significant might in hindsight be regarded as less or even completely irrelevant, or as a distraction from what really matters.

Implications Having explored some of the arguments for or against, on the one hand, a limited notion of violence to be defined in a rather restrictive way or, on the other hand, an extended notion of violence to be defined in a more inclusive way, the question remains what are the implications for empirical social and historical research on violence. The continuous debates about the preferable definition of violence have been be interpreted against the backdrop of what the philosopher Gallie (1956) has called the ‘essentially contestedness’ of concepts. However, strictly speaking, Gallie’s notion of the ‘essentially contestedness’ of concepts may not be appropriate here because this chapter has been focused not so much on the substance of the concept as on the scope of the definition. Therefore, it could be more appropriate to assume that definitions of violence will always be ‘radically or fundamentally contested’ if only because every definition of violence bears its own theoretical, methodological and moral implications. Recognizing the radically or fundamentally contestedness of any definition of violence seems to be the most realistic as well as the most fruitful starting point for empirical research. While it may be true that the debate about the preferable definition of violence is about whether violence should be defined from the perpetrator’s or the victim’s point of view, trying to define violence from the point of view of an impartial spectator or third party not only seems impossible but also premature (Bufacchi, 2005: 199).

Violence as an Essentially Contested Concept

37

One of the implications of definitions of violence being ‘radically or fundamentally contested’ is that locating violence empirically is not a neutral undertaking, solely dependent on what is ‘out there’ to be found. Empirically, violence will always be seen with reference to a particular conception of violence. Instead of trying to conclusively define the concept of violence empirical researchers could considerer the costs and benefits involved in holding a particular view on violence. This means that various concepts or definitions of violence are best being evaluated for their suitability for coming to terms with a particular research problem, i.e. their enabling or constraining the investigator to ‘see further and deeper’ into his or her material (Lukes, 1979: 272). In this view, it may seem to be preferable for historians concerned with the history of violence to opt for a restrictive, e.g. legal definition of violence, particularly as historians often – but not always – depend upon archival sources, which are often juridical in nature. Historical inquiry concerned with the increase or decrease of violence in societies over long periods of time, therefore, would benefit from a restrictive, e.g. legal definition. At least from a practical point of view, using a limited notion of violence to be defined restrictively may have the advantage of the results of such research being less contested than when a more inclusive definition would be applied. Although sociologist Zygmunt Bauman strongly believes that even historians will never reach consensus with regard to the question of whether or not violence in modern history has been increasing or decreasing for this reason that it is impossible to ‘objectively’ measure the overall volume of violence (2000: 32; quoted in Imbusch, 2005: 53). In order to answer the question of why societies differ in what phenomena they define specifically in terms of violence to include them in the criminal code, a restrictive, i.e. legal definition needs already to be accompanied by a more inclusive social definition of violence. However, for a criminologist a restrictive, e.g. legal definition implies more serious disadvantages. To begin with, a restrictive, e.g. legal definition is not necessarily a precise definition because even if we focus upon an extremely limited notion of violence, it will immediately become apparent that ‘violence’ – however narrowly defined – represents a surprisingly broad spectrum of incidents. Restricting a priori what qualifies as ‘violence’ would unduly and unhelpfully limit our understanding of how violence is socially constructed. An important benefit of a more inclusive definition of ‘violence’ is also that it allows researchers to penetrate the personal experience and subjective meaning of ‘violence’ for those involved either as victim (or perpetrator). In this respect, a broad inclusive definition of violence is preferable to a more restricted one because a restrictive definition tends to be a ‘etic’ while a broad inclusive definition enables emergent ‘emic’ perspectives to be integrated in the concept of violence. Only by refusing to make a priori assumptions about what qualifies as violence or not, can the full spectrum of behavior remain open to empirical research (Waddington, Badger & Bull, 2004: 149). As an example, A˚kerstro¨m (2002) studied in detail how in actual practice demarcations were made in order to either include or exclude certain acts in

38

W. de Haan

the concept of violence. She interviewed staff in nursing homes for the elderly and found that nursing home staff describing how elderly residents sometimes slapped, pinched, or hit them made an effort to avoid pushing persons outside the boundary of accepted normalcy by labeling these events as ‘not violence’. Studying the ‘talk work’ used to construct an act as an instance of ‘violence’ or as ‘not violence’ allowed her to see how the social construction of violence proceeds by drawing lines of demarcation or through the interactional work of distinguishing ‘a slap’ from ‘violence’. When staff used the term ‘slap’, ‘pinch’, or ‘punch’ instead of ‘violence’, they downplayed the event in contrast to police or court statements in which similar terms may be used to invest actions with a ‘sense of drama’. A˚kerstro¨m notes that apparently ‘violence’ has connotations that cannot simply be equated with those of punches, shoves, slaps,’’ and the like and argues that placing the elderly’s violence outside the boundaries of violence meant that the elderly remained ‘care recipients’, the staff ‘caregivers’, and the nursing home a ‘caring context’. In this respect, nursing home staff clearly differs from occupational groups like the police who have a vested interest in ensuring that ‘violence’ is an issue for which they can claim professional competence and which they can use to preserve the image of their jobs as dramatic and dangerous. In both cases, workers evinced interest in defining actions so that they fall either inside or outside ‘violence’.’’ In conclusion, I would argue that, for a criminologist it is more fruitful. both theoretically and methodologically, to consider definitions of violence to be essentially contested, to accept that, depending on the specific contexts of discovery and contexts of justification, valid arguments are feasible for either inclusive or restrictive definitions of violence, and that a proper definition of ‘violence’ should not a priori be seen as a starting point for empirical research but as a temporary outcome, which may or may not prove to be useful in future research. Exploring a diversity of definitions is fruitful because by means of adjusting concepts scientific progress can made. Acknowledgment I would like to thank the members of the international Cultures of Violence research group as well as those of the RIDE research group of the Department of Philosophy of the University of Groningen for their stimulating comments. Special thanks to Pieter Boele van Hensbroek for his useful suggestions to improve an earlier version of this chapter.

References A˚kerstro¨m, M. (2002). Slaps, punches, pinches but not violence: boundary-work in nursing homes for the elderly. Symbolic Interaction, 25, 4, 515–536. Barak, G. (2003). Violence and Nonviolence. Pathways to Understanding. Thousand Oaks: Sage. Bauman, Z. (1995). Violence, Postmodern. In Z. Bauman (Ed.), Life in Fragments. Essays in Postmodern Morality (pp. 139–162). Oxford: Blackwell. Bauman, Z. (2000). Alte und neue Gewalt. Journal fu¨r Konflikt- und Gewaltforschung, 2, 28–42.

Violence as an Essentially Contested Concept

39

Betz, J. (1977). Violence: Garver’s definition and a Deweyan correction. Ethics, 87, 4, 339–351. Bourdieu, P. (1977). Outline of a Theory of Practice. New York: Cambridge University Press. Bourdieu, P. (1992). The Logic of Practice. Cambridge: Polity. Bufacchi, V. (2005). Two concepts of violence. Political Studies Review, 3, 193–204. Burgess-Jackson, K. (1995). Rape and persuasive definition. Canadian Journal of Philosophy, 25, 3, 415–454. Gallie, W.B. (1956). Essentially contested concepts. Proceedings of the Aristotelian Society, 56, 167–198. Garver, N. (1968). What is violence? The Nation, 209, June 24, 817–822. Garver, N. (1972). How to think about violence (book review), The Humanist, 32, January/ February, 39. Garver, N. (1977). What is violence? In T. A. Mappes & J. S. Zembaty (Eds.), Social Ethics. Morality and Social Policy (pp. 268–282). New York: McGraw-Hill. Garver, N. (1988). Violence and the social order. In Philosophy of Law, Politics, and Society. Proceedings of the 12th International Wittgenstein Symposium August 7th to 14th, 1987, Kirchberg/Wechsel, Austria (pp. 218–223). Vienna: Holder-Pichler-Tempsky. Hamm, M.S. (2004). Apocalyptic violence: The seduction of terrorist subcultures. Theoretical Criminology 8, 3, 323–339. Heitmeier, W. & J. Hagan (2002). Gewalt. Zu den Schwierigkeiten einer systematischen internationalen Bestandsaufnahme. In W. Heitmeier und J. Hagan (Hrsg.), Internationales Handbuch der Gewaltforschung (pp. 15–25). Wiesbaden: Westdeutscher Verlag. Henry, S. (2000). What is school violence? An integrated definition. Annals of the American Academy of Political and Social Science, 567, 16–30. Henry, S. & D. Milovanovic (1996). Constitutive Criminology. Beyond Postmodernism. London: Sage. Iadicola, P. & A. Shupe (1998). Violence, Inequality, and Human Freedom. New York: General Hall. Imbusch, P. (2005). Der Gewaltbegriff. In W. Heitmeier und J. Hagan (Hrsg.), Internationales Handbuch der Gewaltforschung (pp. 26–57). Wiesbaden: Westdeutscher Verlag. Jackman, M. (2002). Violence in social life. Annual Review of Sociology, 28, 387–415. Levi, M. & M. Maguire (2002). Violent crime. In: The Oxford Handbook of Criminology (pp. 795–843). Oxford: Oxford University Press. Litke, R. F. (1992). Violence and power. International Social Science Journal, 44, 2, 173–183. Lukes S. (1979). On the relativity of power. In S.C. Brown (Ed.), Philosophical Disputes in the Social Sciences. Sussex: Harvester. Nedelmann, B. (1995). Schwierigkeiten soziologischer Gewaltanalyse. Mittelweg, 36, 2, 8–17. Neidhardt, F. (1986). Gewalt: Soziale Bedeutungen und sozialwissenschaftliche Bedingungen des Begriffs. In Bundeskriminalamt (Hrsg.), Was ist Gewalt?Auseinandersetzungen mit einem Begriff. Bd. 1: Strafrechtliche und Sozialwissenschaftliche Darlegungen. (pp. 109–147). Wiesbaden: Bundeskriminalamt. Plattt, T. (1992). The concept of violence as descriptive and polemic. International Social Science Journal, 44, 2, 185–191. Punch, M. (2000). Suite violence: Why managers murder and corporations kill. Crime, Law & Social Change, 33, 243–280. Reilan, E. (2001). Rape as an essentially contested concept. Hypathia, 16, 2, 43–66. Reiss, A.J. & J.A. Roth (Eds.) (1994). Understanding and Preventing Violence. Washington, DC: National Academy Press. Riches, D. (Ed.) (1986). The Anthropology of Violence. Oxford: Basil Blackwell. Riedel, M. & W. Welsh (2002). Criminal Violence. Patterns, Causes, and Prevention. Los Angeles: Roxbury. Schinkel, W. (2005). Aspects of Violence. Unpublished doctoral dissertation. Rotterdam: Erasmus University.

40

W. de Haan

Stanko, E.A. (2003). Introduction: Conceptualizing the meaning of violence. In E.A. Stanko (Ed.), The Meanings of Violence (pp. 1–13). London: Routledge. Trotha, T. von (1997). Zur Soziologie der Gewalt. Ko¨lner Zeitschrift fu¨r Soziologie und Sozialpsychologie, 37, 9–56. Turpin, J. & L.R. Kurz (1997). Introduction: violence. The micro/macro link. In J. Turpin & L.R. Kurz (Eds.), The Web of Violence. From Interpersonal to Global (pp. 1–28). Urbana: University of Illinois Press. Valier, C. (1997). On the violence of censure. In: Colin Sumner (Ed.) Culture, Violence and Censure (pp. 7–27). London: Taylor & Francis. Waddington, P.A.J., D. Badger & R. Bull (2004). Appraising the inclusive definition of workplace ‘violence’. British Journal of Criminology, 45, 141–164. Weiner, N. (1989). Violent criminal careers and ‘violent career criminals’. An overview of the research literature. In N.A. Weiner & M.E. Wolfgang (Eds.), Violent Crime, Violent Criminals (pp. 35–138). Newbury Park: CA. Sage.

Part II

Long-Term Trends

Homicide in Scandinavia: Long-Term Trends and Their Interpretations Dag Lindstro¨m

Present day Scandinavia is usually considered to be a rather peaceful and secure part of the world. This is generally confirm in the national crime statistics. The post-war homicide rates in Sweden, Norway and Denmark have been comparatively low. In Finland, on the other hand, homicide rates have been quite high even by international standards. (See Fig. 2.) However, the difference between Finland and the other Scandinavian countries does not represent a long historical continuity. It is actually a comparatively recent phenomenon. There is no evidence indicating higher homicide rates in Finland than in other parts of Scandinavia before the eighteenth century. Historical studies also tell us that medieval and early modern Scandinavia in general was much more violent and dangerous than today. Several studies have revealed a terrifying level of lethal violence, especially in the towns, where the highest estimated homicide rates reach a level of about 80 homicides per 100,000 inhabitants. The history of crime has been an important field of research in Scandinavia since the 1980’s and we now have a fairly clear picture of the general development of interpersonal violence from the later Middle Ages to the present. There is a general consensus among Scandinavian historians about a long-term decline in deadly violence, but there is no corresponding agreement about how to explain this transition. On the contrary, it has been a matter of intense discussions and manifest disagreements. Recent research also indicates a process much more complicated and discontinuous than previously assumed.

The Long-Term Trends – An Overview In the early 1990’s Eva O¨sterberg (1991a, b, 1992, 1996) presented a preliminary synthesis describing the major tendencies in the development of crime in Sweden from the mid-fifteenth century to the 1960’s. From a combination of several Dag Lindstro¨m ISAK(Institutionen for studier av samha¨llsutveckling och kultur), Enheten, fo¨r historia, (Department for Studies in Social Change and Culture, History Division), Linko¨pings universitet, 581 83 Linko¨ping, Sweden [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

43

44

D. Lindstro¨m

local studies and aggregated national statistics O¨sterberg was able to establish the existence of a massive long-term decrease in deadly violence in Sweden. According to O¨sterberg a fundamental change took place between the mid-seventeenth century and the mid-eighteenth century at the latest. By the mid-eighteenth century the Swedish homicide rate was already below 1 per 100,000. The Finnish historian Heikki Ylikangas (1976, 1998a, pp. 12–16) has reached similar long-term results concerning Finland. For example, the homicide rate in the province of Tavastia in central Finland fell from approximately 25 per 100,000 between 1506 and 1510 to only 2.6 per 100,000 between 1751 and 1776. Results and conclusions drawn by Norwegian historians point in the same direction. Jørn Sandnes (1990, pp. 62, 74, 77, 83), Hans Eyvind Næss (1994, pp. 68–70) and Erling Sandmo (1999, pp. 218f) all agree that violence decreased rapidly in Norway during the seventeenth century. There is less statistical evidence from Denmark, but at least some Danish historians, such as Jens Christian V. Johansen and Henrik Stevnsborg (1986), maintain that the level of violence fell rapidly in Denmark during the seventeenth century. A more recent Scandinavian survey including further empirical evidence generally supports these conclusions (Ylikangas, Johansen, Johansson & Næss, 2000, pp. 67–79). A rapid decline in deadly violence during the seventeenth has also been confirmed by the Swedish criminologist Maria Kaspersson in her long-term study of homicides in Stockholm (Kaspersson, 2000). Scandinavian research thus confirms the long-term decline in deadly violence observed in other European studies. In the early 1980’s Ted Robert Gurr (1981) and Lawrence Stone (1983) described this fundamental transition. Their conclusions were based on English evidence. Pieter Spierenburg (1994, 1996) later established a similar trend in the Netherlands and Manuel Eisner’s (2001) recent survey further supports a long-term decline in deadly violence as a general European phenomenon. The typical homicide rates have fallen from 20 to 40 killings per 100,000 during the late Middle Ages to only 0.5 to 1 killing per 100,000 by the mid-twentieth century. According to Eisner the decisive turning point may be narrowed down to the early seventeenth century. Homicide rates have been estimated for a number of late medieval and early modern Swedish1 and Norwegian towns and country districts. Although there are considerable differences, they all reveal a level of deadly violence much higher than today. The urban figures range from 10 to 83 homicides per 100,000 inhabitants and the rural figures range from a few up to 40 killings per 100,000. 1

Finland was a part of the Swedish realm until 1809 and if not otherwise is indicated the terms ‘‘Sweden’’ and ‘‘Swedish’’ include Finland until 1809. ‘‘Finland’’ and ‘‘Finnish’’ refer to the eastern part of the Swedish kingdom within the borders of that time. Between 1809 and 1917 ‘‘Finland’’ and ‘‘Finnish’’ refer to the autonomous grand duchy within the Russian empire. In the seventeenth and eighteenth centuries the Swedish realm also included Baltic and German provinces. Contrary to Finland and the former Danish and Norwegian provinces added to Sweden in 1645 and 1658, the Baltic and German provinces never became integrated as parts of Sweden proper. They kept a different social, political, administrative and judicial structure and are not included when references are made to Sweden and Swedish homicide rates.

Homicide in Scandinavia

45

(O¨sterberg, 1992, p. 77; Ylikangas et al., 2000, pp. 68f) Long-term averages, though, seem to be substantially lower than many of the more spectacular figures estimated for shorter periods. The homicide rate in Finnish towns has for example been estimated at 15 killings per 100,000 between 1540 and 1599 and at 12 killings per 100,000 between 1620 and 1660. (Karonen, 1995; Ylikangas, 1998a, p. 13) There are practically no estimated homicide rates for early modern Denmark to be found. This is partly due to the lack of reliable sources and the difficulties in estimating the Danish population. Furthermore, some Danish historians, such as Hans Henrik Appel, are more skeptical about crime statistics based on seventeenth century records. Appel’s study of the district of Skast in Jutland nevertheless confirms the common conclusions about a general decline in deadly violence during the seventeenth century. A total of about 20 homicides per 100,000 appear in the registers, and there is an obvious concentration to the early decades of the century. (Appel, 1999, pp. 276f, 296f) After 1750 there is a much more solid basis for calculating homicide rates in Sweden (and Finland) due to the development of national statistics. Population statistics and statistics on causes of death start in 1750, and from 1836 the Swedish national statistics also include crime statistics. Some differences can be noted between the causes of death statistics and the homicide statistics, but the discrepancies are rather small and both statistics follow the same trends and variations. (von Hofer, 1985, 2003) The homicide rates from the mid-eighteenth century on (based on causes of death) are considerably lower than the estimated homicide rates from the fifteenth to seventeenth centuries. These statistics thus confirm a substantial decrease in deadly violence. They also support the conclusion that the main transition had already taken place by the mid-eighteenth century. The Swedish statistics also show a small but significant rise in deadly violence during the first half of the nineteenth century. After ca 1850 homicide rates decreased again and then remained on a low level until the post-war era when the homicide rates started to increase once again. These statistics also indicate significant differences between Sweden and Finland as shown in Fig. 1. Homicide rates in Sweden were much lower than in Finland already by the mideighteenth century. Yet, the calculated homicide rates from the sixteenth and seventeenth centuries do not indicate a corresponding difference. In fact, at that time Finnish homicide rates were usually rather low, and all the markedly high homicide rates originate from areas in the western part of the Swedish realm. From the eighteenth century on the aggregated homicide rate in Finland continued to increase up to the 1920’s, when it reached a level of more than 8 killings per 100,000. During the same decade the Swedish homicide rate was less than 0.5. The Finnish rates have gradually decreased since the 1930’s, but they are still higher than in the rest of Scandinavia. The Swedish statistics allow regional studies on homicide rates, but very few have been carried out till now. One important exception is Johan So¨derberg’s study on violence in Sweden 1758–1870, where the homicide rate in Stockholm is documented. (So¨derberg, 1998, p. 19) Some important differences between

46

D. Lindstro¨m

0 -4

2 33 19

6

-3

-1 19

20

3 -1 19

14

4

05

19

90

8 18

99

-1

-9

0 -7 18

79

0 18

-6

Sweden

61

5 46 18

-4 31

18

-3

0

6 09

18

80

2 17

93

-1

-9 75

17

17

54

-6

3

9 8 7 6 5 4 3 2 1 0

Stockholm

Finland

Fig. 1 Homicide rates in Sweden and Finland 1754–1940 Finland was a part of the Swedish kingdom until 1809, but the figures have been separated from 1754 on. The Stockholm figures end in 1870. All figures are based on statistics on causes of death, and they are calculated as mean values per 100,000 population annually. Sources: Verkko, 1946; So¨derberg, 1999. Verkko’s first interval for Finland is 1754–1773. So¨derberg’s intervals for Stockholm are 1750–73, 1774–90, 1791–1810, 1811–30, 1831–50 and 1851–1870.

Stockholm and the national homicide rate can be observed. The Stockholm homicide rate was considerably higher, but, on the other hand, the Stockholm figures do not indicate any increase in lethal violence during the first half of the nineteenth century, and by the mid-nineteenth century the homicide rate in Stockholm no longer differed much from the national average. The variations in homicide rates between different regions in Finland are even more striking than the differences between Sweden and Finland. Ostrobothnia in western Finland is the most extreme case. Between 1790 and 1900 about 2000 people were killed, and this period is commonly referred to as the era of the knife fighters (knivjunkare). The homicide rate in Ostrobothnia (with a total population of about 100,000 in the early nineteenth century) rose to a level of 12 per 100,000 at the most. The homicide level started to increase from the 1790’s. It remained on a high level from the 1830’s until the 1880’s when a downward trend began. In some parishes the situation was extremely bad, and homicide rates as high as 50 to 60 killings per 100,000 inhabitants have been noted. This can be compared with the average homicide rate in Finland, which fluctuated between 2 and 3 per 100,000 during the nineteenth century. Another important observation is that Ostrobothnia has not always been characterized by a high level of homicide. Estimated figures from the sixteenth century, on the contrary, indicate a peaceful area with a homicide rate of approximately 3 killings per 100,000 (Ylikangas, 1976, pp. 91–95, 1998b; Ylikangas et al., 2000, pp. 72–75). National Norwegian statistics do not go as far back as those in Sweden, but some estimated homicide rates from the eighteenth and nineteenth centuries still

Homicide in Scandinavia

47

95

91

19

87

19

83

Norway

19

79

19

75

19

71

Sweden

19

67

19

63

Finland

19

59

19

55

19

19

19

51

3,5 3 2,5 2 1,5 1 0,5 0

Denmark

Fig. 2 Homicide rates in Scandinavia 1951–1995. The figures are calculated as mean values per 100,000 population annually, based on reported offences. Source: von Hofer, 1997.

provide a rather good indication of a development similar to that in Sweden. Studies on one of the major judicial districts (Bergen’s lagting, in south-western Norway) shows a homicide rate of 0.6 per 100,000 for the period 1702–1737 and 0.75 for the period 1860–1865. (Næss, 1994, p.69) This is at least a strong indication of a major transition towards lower homicide levels completed in Norway by the early eighteenth century. After the 1930’s the Finnish homicide rate declined, and during the post-war period it has usually fluctuated between 2 and 3 per 100,000, a rate still much higher than in the other Scandinavian countries. (See Fig. 2.) Sweden has experienced a gradual increase from about 1 per 100,000 in the early 1950’s to approximately 2 in the early 1990’s. Denmark and Norway have also experienced increasing homicide rates. In Denmark the rate has increased from about 0.5 in the 1960’s to over 1 per 100,000 in the early 1990’s and Norway, where the rate was less than 0.5 in the late 1950’s and early 1960’s, reached a level of slightly more than 1 per 100,000 in the early 1990’s. Another important difference between the Scandinavian countries is the fact that the homicide rates have increased since the 1950’s and the 1960’s in all Scandinavian countries but Finland, where no significant long-term change can be observed in post-war homicide rates. Thus, the difference between Finland and the other Scandinavian countries has gradually become less significant during the post-war period.

Medieval and Early Modern Homicide Rates – Uncertain Estimations The conclusions concerning a fundamental long term transition in the frequency of violent deaths rests on solid empirical evidence. But the late medieval and early modern Scandinavian homicide figures often referred to are extremely

48

D. Lindstro¨m

uncertain, and they are not always comparable. Calculations are often based on a small number of cases and a rather limited population. The Swedish provincial towns Arboga and Vadstena illustrate this problem. The Arboga homicide rate range between 10 and 60 killings per 100,000 between 1452 and 1543. The estimation is based on a study of 6 periods of 10 years each. Only 14 homicides were registered during those 60 years. One single killing during a 10-year period yielded a homicide rate of 10 killings per 100,000, and one additional killing would have doubled the homicide rate. On the other hand, it seems reasonable to maintain that Arboga experienced a lot more deadly violence between 1462 and 1482, when 10 killings were registered during a total of 20 years, compared with the period from 1493 to 1543 when only 3 killings were recorded during a total of 30 years. The Vadstena homicide rate of 77 killings per 100,000 between 1602 and 1610 is based on only 7 cases during 9 years and an estimated population of approximately 1,000 inhabitants. (O¨sterberg & Lindstro¨m, 1988, pp. 43, 54) The homicide rates for many of the country districts are based on equally small samples. Two registered killings in the district of Vendel in Uppland (Sweden) between 1615 and 1630 have been used to calculate a homicide rate of 14 homicides per 100,000 people. Between 1631 and 1645 the calculated level, based on 6 registered killings, is 26 homicides per 100,000 people, a substantial increase. These calculations have been based on an estimated population of 1,000 inhabitants 1615–30 and 1,500 inhabitants 1631–1645. The homicide rates for La˚nghundra between 1545 and 1560 (14 killings per 100,000 people), another judicial district close to Vendel, are based on a total of only 5 killings during 15 years and an estimated population of 2,300 people.(O¨sterberg, 1991c, pp. 99–112, 1992, p. 77; Ylikangas et al., 2000, p. 68) The uncertain character of the estimated homicide rates becomes even more evident when population calculations are considered. Both Johan So¨derberg and Eva O¨sterberg specify the homicide rate in Stockholm during the early 1620’s at a level of 32. (O¨sterberg, 1992, p. 77; So¨derberg, 1999, p. 80) Arne Jansson, on the other hand, has suggested an interval between 25 and 43, due to the uncertain population size. (Jansson, 1998, p. 16) After 1750 homicide rates in Sweden can be based on population statistics, but before that date it is necessary to rely on calculations usually based on tax rolls. Thanks to the Swedish bureaucracy, the tax records provide a relatively good basis for estimating populations as early as the sixteenth century, and during the seventeenth century a systematic population registration actually began to evolve in Sweden. Population figures for Denmark and Norway are more rare and more uncertain due to less frequent and less complete tax records. Tax registers are usually based on households and in order to calculate the population the number of registered farms or households are usually multiplied with a factor corresponding to the assumed average household size. For rural areas in Sweden this factor has usually varied between 5 and 7. Eva O¨sterberg used the factor 6 when calculating the homicide rates for Swedish country districts during the sixteenth and seventeenth centuries. (O¨sterberg, 1991c,

Homicide in Scandinavia

49

p. 110) Urban households are often assumed to have been smaller. A factor of approximately 4 has often been used (Lilja, 1996), but bigger households have sometimes been assumed as well. Birgitta Lager, for example, used the factor 6 when estimating Stockholm’s population in 1582. (Lager, 1962, p. 29) Recent research also indicates that the populations of small towns in early modern Sweden may have been greatly underestimated. Claes Westling has combined the usual tax registers with a systematic analysis of other available sources. For the town of Vadstena the result was astonishing. A large number of burghers had been omitted from the tax records, and when the population figures for mid-seventeenth century Vadstena were recalculated the number of inhabitants increased more than 120 percent from approximately 500 to more than 1200. (Westling, 2002, pp. 36–89) A corresponding underestimation during early seventeenth century would reduce the Vadstena homicide rate from 77 to 34 homicides per 100,000. An average household size of 6 instead of 4 would further reduce the homicide rate to 22 instead of 77. This would still be a very high figure by modern standards, but less extreme than suggested in the original calculation. The figures are also difficult to compare as different researchers have used different types of sources and they have applied different principles when counting the cases. Pieter Spierenburg has made an important distinction between records listing (a) convicted killers, (b) suspected offenders and (c) actual killings whether a suspect was identified or not (e.g. body inspections and coroner’s inquests). (Spierenburg, 1996, p. 79) These differences must be taken into consideration whenever homicide rates are compared. Most medieval and early modern Swedish (and Finnish) studies are based on local court records. Normally, court records would include offenders tried by the court and possibly found guilty. The local courts were not separated from administrative and political functions, and consequently older Scandinavian court records (both urban and rural) often contain a mixture of judicial records, administrative records and all sorts of important memorandums. There were no formal standards regulating how court records were to be kept, although central control became more thorough in Sweden after 1614. Most of the registered cases refer to convicted killers, but the records may also include information about people suspected, and sometimes tried, but not found guilty. There may even be notations about dead bodies and suspicious deaths as well as comments about homicides without any known or suspected perpetrator. The district of Konga in Sma˚land, near the Danish border, appears as to have been one of the most violent rural districts in early modern Scandinavia with a calculated homicide rate of 40 killings per 100,000 between 1614 and 1629. (O¨sterberg, 1992, p. 77) This calculation is based on evidence presented by Lars-Olof Larsson. He is not very specific about his data, however. The only information he provides describes just over 50 homicides in the court records between 1614 and 1629, while he estimated that the population of the district could not possibly have exceeded 8,000–9,000 inhabitants. Larsson also refers

50

D. Lindstro¨m

to similar amounts of killings in other judicial districts in southern Sma˚land during the early seventeenth century. Finally, he notes that the number of registered killings in early eighteenth century Konga was only one fifth of the amount registered about a century earlier. (Larsson, 1982, pp. 53f) This unspecific character is unfortunately typical for several Scandinavian studies. Precise information is often lacking about the principles used when counting homicides. Some records, especially the court records from Stockholm include notations about private settlements in homicide cases. A total number of 44 such settlements have been registered between 1474 and 1599. During the early sixteenth century an average of one such settlement has been registered annually while the number falls to 0.4 case annually during the late sixteenth century. Some of these settlements were reached before the municipal court in Stockholm, whereas others had been reached elsewhere but were still entered in the Stockholm court records. Nor had all of these homicides been committed in Stockholm, although they were registered into the Stockholm court records. In many cases it is actually impossible to determine whether the killings had occurred in Stockholm or somewhere else. (Lindstro¨m, 2007, pp. 126–28) Accordingly, the Stockholm figures and the estimated homicide rates clearly include some homicides not committed in Stockholm. It is also important to remember that the Stockholm figures were originally generated for purposes other than the calculation of crime rates. (O¨sterberg & Lindstro¨m, 1988, pp. 76f) The Vadstena figures suffer from similar imperfections. Vadstena appears as to have been one of the most violent places in early modern Scandinavia with a homicide rate of 60 killings per 100,000 during the period 1580–1589 and as high as 77, 1602–1610. (O¨sterberg, 1992, p. 77; Ylikangas et al., 2000, p. 68) Petri Karonen has recalculated the Vadstena homicide rates omitting all homicides registered but not committed in Vadstena, and he arrives at considerably lower homicide rates. When presented as five-year moving averages the rates fluctuate between 20 and 60 during late sixteenth and early seventeenth century. (Karonen, 1998, pp. 163f, 184) Some of the Swedish and Finnish figures from the sixteenth century are based on rolls of fines paid instead of court records. This is the case with most rural areas (Liliequist, 1999; Ylikangas, 1976). In late medieval and sixteenth century Scandinavia homicides commonly resulted in economic compensation. Accordingly, fine payment rolls are to some extent usable as records for the study of deadly violence, at least until the early seventeenth century, but they are not as complete as the court records. The reliability of fine payment rolls could be tested by means of a comparative analysis of fine payments rolls and court records, but this has still not been systematically accomplished. Anyway, this means that any comparison between urban and rural areas during the sixteenth century is of limited value. It also puts restraints on any conclusions about long term trends

Homicide in Scandinavia

51

based on comparative analysis of rural homicide rates between the sixteenth and the seventeenth centuries. Fine payment rolls have also been used in Norwegian studies such as Margit Løyland’s study of crime in the district of Agder during the seventeenth century. (Løyland, 1992, pp. 12f, 55–71) Consequently the estimated homicide rate for Agder is based on a possibly incomplete sample of homicides. Cases ending with capital punishment or a private reconciliation are not registered, nor are all the possible cases where the court could not find a guilty perpetrator. Still, the calculated homicide rates are very high in the parts of Agder studied (between 12 and 29 per 100,000 during the early seventeenth century) The fine payment rolls are obviously insufficient for drawing any definite conclusions about the level of homicides and the possible changes in homicide rates. The significant fall in the calculated homicide rate during the seventeenth century may very well be a result of changing legal practices. The new Norwegian Law Code of 1687 also introduced new and harsher punishments. Another problem with these figures is that the estimated population of seventeenth century Agder is based on rather uncertain assumptions. (Løyland, 1992, pp. 26–28) There are no proper court records from sixteenth century Norway, and some early modern Norwegian homicide rates are based on sources very different from the records used in Swedish and Finnish studies. The course of legal proceedings could produce a number of different documents. In medieval and sixteenth century Norway a killer was expected to formally announce his complicity. If not, he would be regarded as an assassin and would become an outlaw. On the other hand, if the announcement was made properly, the king could issue a letter of peace in order to make a judicial investigation possible. During the judicial procedures attacks on the offender were forbidden. A killing did not necessarily lead to a judicial trial at all. A private settlement was another possible outcome. The killer could offer economic and other compensation, and, if this was accepted, he was free from any further legal measures or acts of vengeance. Several agreements of this kind have been documented in official letters. It was also possible for the king himself to pardon a killer and to issue a letter prohibiting any kind of retaliation. These kinds of letters have been analyzed by Jørn Sandnes. (Sandnes, 1990, pp. 65–72) He found a total number of 38 homicides documented this way between 1563 and 1570. The figure obviously represents a minimum. It is not known how common these letters actually were nor to what extent they may have been preserved. Sandnes himself refers to a fine payment roll from 1560 where five homicides are registered. None of these crimes were ever mentioned in any other judicial documents. The 38 homicides found between 1563 and 1570 obviously comprise a very low minimum figure, and Sandnes himself never based any calculations on them. Later, however, they have been used by Ylikangas and others to calculate a Norwegian sixteenth century homicide rate at a level of 10 to 27 killings per 100,000. (Ylikangas et al., 2000, p. 69)

52

D. Lindstro¨m

The famous Bergen homicide rate (83 killings per 100,000) 1562–1571, on the other hand, is based on Sandnes’ own calculations.2 The main source is a diary kept by the well known Norwegian scholar Absalon Pederssøn Beyer. Between 1562 and 1571 he made notes on some 30 homicides. The exact number is uncertain, because Absalon was not always very specific about the time and place of the killings. According to Sandnes, this gives us a minimum figure of 3 to 4 killings annually in Bergen. The dairy also indicates that Absalon sometimes failed to enter notes about homicides he knew about. In 1567 Absalon listed all homicides for several years, and several of them had not been mentioned before in the diary. Sandnes’ conclusion is that five killings annually would be a probable minimum figure for Bergen, i.e. a total number of 50 killings in 10 years. Finally, he estimated the population at 6,000 inhabitants. (Sandnes, 1990, p. 73) The population figure is both uncertain and rather low. Knut Helle has estimated the late sixteenth century Bergen population to have been approximately 8,500 people. (Helle, 1982, p. 487) Thus, the commonly related Bergen homicide rate is actually based on a number of rather uncertain assumptions. A population of 8,500 inhabitants and 40 homicides in 10 years would give a homicide rate of 47 killings per 100,000 population. This would still have been a very high level, and it would not change the fact that Norway and Scandinavia have experienced a dramatic decrease in deadly violence since the sixteenth century, but it would definitely render sixteenth century Bergen a less distinguished position among the most violent places in history. A few conclusions may be drawn from this. The exact Scandinavian homicide rates often presented are illusory and comparative conclusions based on estimated late medieval or early modern Scandinavian homicide rates must be treated with great caution. On the other hand, there is no reason whatsoever to challenge the conclusion that Scandinavia has experienced a dramatic decline in homicide rates similar to the development noted in many other countries. Another important conclusion that may actually be drawn from these still uncertain homicide rates, is that the regional differences in lethal violence were considerable in early modern Scandinavia.

Discontinuities and Regional Differences. A More Complicated Picture of the Long-Term Trends in Homicides The long-term development of homicides in Scandinavia has often been described as a steady and quite rapid decline during the seventeenth century. Johan So¨derberg, for example, illustrates the development of deadly violence in 2

The homicide rate in Bergen was presented by Sandnes as 1 killing per 1,200 persons, and it has later been recalculated as 83 killings per 100,000 by other Scandinavian researchers. Næss, 1994, p. 69; Ylikangas et al., 2000, pp. 69f.

Homicide in Scandinavia

53

Stockholm with a graph showing a steady downward trend that accelerated during the seventeenth century. (So¨derberg, 1993, p. 90)3 Eva O¨sterberg in her early synthesis emphasized a long-term decrease, but she also noted several countertrends and short term peaks (during the 1460’s and 1470’s and during the years around 1600). O¨sterberg noted that the peaks coincided with periods of political unrest in Sweden. She also noted a short-term peak in the 1840’s. O¨sterberg also called attention to the fact that some areas in Finland (especially southern Ostrobothnia) experienced a dramatic rise in violence during the nineteenth century. (Cf von Hofer, 1995; O¨sterberg, 1992, pp. 81f) Since O¨sterberg published her synthesis, a number of new studies have indicated additional discontinuities. Thus, a more complicated picture of the long-term trends is gradually emerging within Scandinavian research on violence. My own study of crime in Stockholm from 1475 to 1625 indicates several short-term fluctuations. The number of registered homicides actually increased during the late sixteenth and early seventeenth centuries. The general level of registered crimes of violence fell dramatically from the late 1560’s and then increased again during the 1580’s. (O¨sterberg & Lindstro¨m, 1988, pp. 78–96) Petri Karonen has more explicitly challenged the smooth transition. His detailed analysis of the development of violence in four Swedish towns (Stockholm, Arboga, Vadstena and A˚bo/Turku) from 1540 to 1660 using five-year moving averages reveals a rather complicated and discontinuous process. Significant increases can be observed in all four towns towards the end of the period, and homicide rates did not decrease steadily in any of the four towns. On the contrary, homicide rates actually increased several times during the period. In Stockholm the homicide rate was much higher during the 1590’s than it had been about the mid-sixteenth century, and in Vadstena the homicide rate about the mid-seventeenth century well exceeded that of the sixteenth century. Karonen concludes that the 1590’s represent a peak in violent crimes both in Stockholm and in Arboga. He explains this as part of a general economic, political and military crisis of the 1590’s, an explanation similar to O¨sterberg’s. (Karonen, 1998, pp. 160–178) Arne Jansson also emphasizes the many discontinuities in the development of violence in Stockholm during the seventeenth century. From a level of 27–38 homicides per 100,000 during the first decade of the century the homicide rate increased to a level of 50–61 during the second decade. After that the homicide 3

A similar description of a steady decline in violence accelerating during the seventeenth century appears in Jarrick & So¨derberg, 1998, p. 217. So¨derberg’s graph has been somewhat modified in a later version of his study. The downward trend is interrupted during the early seventeenth century and the graph shows a significant short term rise in deadly violence. So¨derberg, 1999:79. So¨derberg finds a homicide rate of 72 per 100,000 inhabitants 1475–1495. The graph, though, indicates a level close to 80 around 1500. After that the rate fell to 54 in early 1620’s. Around 1750 the rate was not more than 3. This is approximately twice the level estimated by O¨sterberg. Her calculations are based on an estimated population of about 10,000 inhabitants. O¨sterberg, 1992, p. 76. Consequently, So¨derberg’s calculations must be based on a much smaller population.

54

D. Lindstro¨m

rate fell rapidly and reached a bottom level of less than 10 killings per 100,000 by the mid-seventeenth century. In the 1680’s the homicide rate slightly increased again to more than 10, and at the end of the century the homicide rate declined again to just over 5 killings per 100,000. (Jansson, 1998, p. 16) Norwegian historians have come to divergent conclusions about the development of early modern crimes of violence. Jørn Sandnes interprets the high level of violence in early modern Norway as a continuation of medieval behavior and even as a late remainder of a Viking mentality. (Sandnes, 1990, p. 84) One major problem with Sandnes’ interpretation is the fact that it is not known how violent medieval and Viking Age Norway actually was. It is difficult, if not impossible, to estimate the level of crimes of violence in medieval Norway. A total of about 250 homicides have been found in different judicial documents between 1300 and 1550, which would amount to an average of about 0.5 homicides per 100,000 inhabitants annually. (Bagge & Mykland, 1987, p. 57) Obviously, the real level must have been much higher, but it does not help to determine whether the sixteenth century level corresponds to the medieval, or if it represents a decrease, or if it actually reflects a peak in violence. Some Norwegian historians have actually argued that the early seventeenth century figures do represent a peak in crimes of violence. (Næss, 1994, p. 68; Sandmo, 1999, pp. 218f) Similar trends have been observed in Denmark as well. (Johansen & Stevnsborg, 1986; Næss, 1994, pp. 71f) It has often been asserted that towns have been more violent than the country districts, and that more remote and economically isolated country districts were more violent than central areas. (E.g. Ylikangas, 1998a, pp. 13f; Ylikangas et al., 2000, pp. 71f) But during the sixteenth and seventeenth centuries the differences between different rural areas are often more obvious than the difference between town and country. In many cases these regional differences are of the same dimension as the differences revealed in the long-term trends. It is, furthermore, difficult to draw any general conclusions about remote rural areas. In fact, some of the lowest homicide rates in early modern Sweden are from northern Sweden, whereas some of the highest levels have been observed in central Sweden. The calculated homicide rates in the districts of Ha¨lsingland and A˚ngermanland in northern Sweden during the sixteenth century never exceeded 4 homicides per 100,000 persons. These levels are closer to modern rates than to many other sixteenth century rural districts. Moreover, in northern Sweden increasing homicide rates and several short term peaks have also been observed during the seventeenth century. (Liliequist, 1999, pp. 74–81) These observations provides further confirmations of the long-term discontinuities and the important regional differences. It has already been noted that the development in Finland from the eighteenth century on took another course than in Sweden. The homicide rates in Finland increased from the mid-eighteenth century and reached a peak during the 1920’s with an average of around 8 homicides per 100,000 annually. According to Yllikangas it may be true that there has been a steady long-term reduction in violence in countries like Great Britain and Sweden, and

Homicide in Scandinavia

55

thus the civilizing theory may apply to those countries. But Finland has not experienced a similar long-term development and accordingly, the civilizing theory does not apply very well to that country. (Ylikangas, 1999)

Explanations and Interpretations The observed changes in homicide over time in Scandinavia have been interpreted and explained in different ways. The observed changes in violence have sometimes been associated with changes in property crimes. A possible transition from violence to theft (De la violence au vol) was suggested in the 1960’s and this hypothesis has been discussed by some Scandinavian historians. (Cf. Johansen & Stevnsborg, 1986) Eva O¨sterberg (1992, pp. 87f) regards this interpretation having some relevance when a long-term transition from late medieval to modern society is described. However, violent crimes started to decrease centuries before the number of thefts began to increase in Sweden. So it is difficult to see how these two processes could have been interconnected. This is a line of argumentation also followed by Jan Sundin. (Sundin, 1992, pp. 456–459) A number of Swedish historians have rather emphasized a general transition from personal conflicts (including violence, slander and defamation) towards impersonal property disputes (e.g. debts, inheritance and property). Available studies indicate that such a transition may have started by the early seventeenth century if not earlier. (A˚gren, 1988; Lindstro¨m, 2003; O¨sterberg, 1992, pp. 85f; Sjo¨berg, 1990, pp. 166–170; So¨derberg, 1990; Sundin, 1992, pp. 412–421) Many Scandinavian historians refer to Norbet Elias and his theory on the civilizing process. (Jansson, 1998; O¨sterberg, 1992, pp. 90–92; Sandmo, 2000; Sundin, 1992, pp. 418–421) Some Finnish scholars as early as the 1940’s argued that people in the sixteenth century had been unable to control their emotions and that a change in mentality could explain the observed reduction in criminal acts. (Ylikangas et al., 2000, pp. 78f) Few Scandinavian historians, however, have applied the civilizing process as the only important interpretation when discussing long-term trends in violence. It has usually been combined with other possible explanations. The civilization theory has also been modified and reinterpreted by the Swedish historians Johan So¨derberg and Arne Jarrick. In the 1990’s they formulated an interpretation of the civilization process quite differently from that of Elias. According to Elias the process started among the upper strata of society with the state and its gradual monopolization of violence as a driving force. Jarrick and So¨derberg reduced the influence of the state as less important and interpreted the civilization process more as a spontaneous change from below. Increasing division of labor together with more developed and diversified social contacts promoted tolerance and peaceful cooperation. Along with the development of cooperative strategies and a more intense use of written documents, it became less important to express personal honor and

56

D. Lindstro¨m

thereby brutality further declined. Jarrick and So¨derberg also emphasized the importance of increasing empathic capacities rather than affective control. The development of mutual social respect is a central theme in their concept of the civilizing process. This led not only to less violent behavior, it also promoted individual autonomy, tolerance and better prospects for individuals to participate in the society on equal terms. According to Jarrick and So¨derberg the civilization process did not necessarily manifest itself in more controlled and restricted behavior. It could also liberate emotions and desires. (Jarrick & So¨derberg, 1994, 1998; So¨derberg, 1990, 1993) The Swedish criminologist Maria Kaspersson also emphasizes the importance of the civilization theory when discussing the long-term decline in deadly violence, but, on the other hand, she also argues that the civilization theory itself is not sufficient as an explanation for the very rapid decline in homicides during the relatively short period of the seventeenth century. (Kaspersson, 2000, pp. 205–212, 223f) Several Scandinavian historians have suggested a close connection between the high level of violence and the social importance of personal honor. When challenged, honor had to be defended. The act of defense could be oral, physical or judicial. If honor was not defended it could be severely damaged or even entirely lost, and without honor it was not possible participate fully in normal social and economic life. Violence was to some extent actually accepted and even prescribed as a proper and necessary measure with which to defend honor. The importance of honor may also explain the well-known fact that court records are full of cases dealing with apparently harmless petty violence like single slaps or shoving. The physical attack was an attack on the persons honor rather than on the body itself, and it required a response with suitable measures. (Karonen, 1998, pp. 141–143; Lindstro¨m, 1994; Sandmo, 1992, pp. 55–93; Sandmo, 1994; So¨derberg, 1990, pp. 246–248) Honour did not necessarily lose its social importance even when violence declined. According to Arne Jansson honor seems to have become even more important in homicide cases by the end of the seventeenth century and in the early eighteenth century than it had been before. (Jansson, 1998, pp. 158f) In addition to the civilizing process and a traditional culture of honor, the early modern state formation is also frequently referred to when changes in criminal activates and judicial procedures in early modern Scandinavia are discussed. No one, however, has argued the fundamental importance of state formation as consequently as Heikki Ylikangas. He rejects the theories about a spontaneous civilization process and claims that a more peaceful early modern society developed as a direct consequence of the state formation process during the seventeenth century. A centralized and more effective system of law enforcement was introduced and the authorities were able to provide protection for their subjects. Briefly, according to Ylikangas, we need to explain the rise of the modern state if we want to explain how a more peaceful society developed. Ylikangas also argues strongly against any harmonious description of social order before the rise of the strong state. Pre-modern society was neither peaceful nor balanced, as it has sometimes been argued. On the contrary, according to

Homicide in Scandinavia

57

Ylikangas, this was a society were violence thrived. Like any society where law is not obeyed, it was a chaotic and violent society ruled by despotism and superior force. Nineteenth century Ostrobothnia is a similar example. The knife fighters ruled, life was insecure and the situation did not change until a strong and effective law enforcement was introduced. People then no longer had to surrender to violence and terror. (Ylikangas, 1994, 1998a, 1999) Young men have always been strongly associated with crimes of violence and young men usually become even more dangerous if they are soldiers. (Karonen, 1998, pp. 169f; Kaspersson, 2000, p. 203; O¨sterberg & Lindstro¨m, 1988, pp. 97f; Sundin, 1992, p. 110) Between the mid-sixteenth century and 1721 Sweden was more or less perpetually involved in warfare and soldiers were often around. The effects of this have been discussed. Did warfare lead to more violence or did it in fact reduce violence as many young men, especially those who were less socially integrated, left the local communities for the army. Jonas Liliequist’s study of violence in northern Sweden indicates an increasing level of violence during the seventeenth century due to the presence of soldiers. Soldiers were not only accustomed to violence and the use of weapons. According to Liliequist’s interpretations, they also combined a strong code of honor and manliness with a rather fragile social position, which made them particularly disposed to the use violence. The soldier’s violence may at first sight appear as impulsive and spontaneous, but these acts were rather deliberate and rational acts guided by a strong culture of honor. (Liliequist, 1999) The transition towards a less violent society is expressed not only in decreasing numbers of homicides. The social attitudes towards the use of violence also changed in a fundamental way. In late medieval and early modern Scandinavia crimes of violence were commonly committed by socially well-established people. (Lindstro¨m & O¨sterberg, 1988, p. 154; Sundin, 1992, pp. 109–113) Even the most respected members of the community involved in serious violence during the sixteenth century. Ylikangas has noted that lay members of the local courts actually committed more homicides than ordinary peasants. According to Ylikangas this is because they controlled important economic resources and therefore might use violence to secure their interests. This is a pattern typical for a society characterized by weak compliance to the law and an extensive use of private settlements. In case of homicide or other crimes these people were well equipped to pay fines or to reach a settlement. Both Sundin and Yllikangas observe a fundamental transition during the seventeenth century. Gradually it became less common for wealthy peasants and other socially well established people to be involved in violence. Crimes of violence instead became more and more associated with soldiers, farmhands and their equals. This development continued during the eighteenth and nineteenth centuries. The elite simply handled their conflicts and disputes without using violence. Ylikangas also suggests that the causes of violent acts simultaneously changed from conflicts about influence and property over to disputes caused by personal insults and abuses. A similar transition took place in urban settings. It was no sensation if a councilor or

58

D. Lindstro¨m

even the mayor was involved in violence during the seventeenth century. Merchants and artisans are also commonly found among those who committed violent crimes. Nineteenth century rolls of fines paid show a totally different picture of the typical violent criminal. By that time they usually belonged to the lower social strata in society. (Sundin, 1992, pp. 457f; Ylikangas, 1998a, pp. 88–94; Ylikangas, 1999, pp. 169f) Cultural change, state formation and institutional changes have been the major explanations used by Scandinavian historians when dealing with the long-term changes in crimes of violence. But there are also some examples of socio-economic explanations. Jan Sundin emphasizes the importance of social and economic change when the long-term trends in crime are to be explained. (Sundin, 1992, p. 473) Lars Magnusson (1988) combines cultural and socioeconomic factors to explain the violent behavior of young artisans in nineteenth century Eskilstuna. Cultural patterns were reactivated and transformed as the social position of apprentices and journeymen was undermined because of the development of manufactures and early industry. Young artisans used violence to emphasize their independence and their unique social position. The need and urge to use violence became even stronger when their social expectations no longer corresponded to their real social position. Similar discrepancies between social expectations and economic realities were important in the early studies by Heikki Ylikangas as well. In 1976 he discussed the causes of violence in a long-term perspective. Unequal possibilities and strong differences in social status were considered to be the main reasons behind criminal behavior. According to Ylikangas periods of increasing violence in Finland coincides with periods characterized by a combination of strong common values about life success and very limited possibilities for the majority to attain these goals. Still, it is not possible to establish a simple and direct link between poverty and violence, because the level of violence is always influenced by the efficiency of social control. Thus, the wave of violence in southern Ostrobothnia during the nineteenth century was explained by Ylikangas as the result of a combination of economic inequality, social polarization and the development of an individualistic competitive mentality in a society where social control was breaking down. (Ylikangas, 1976) Martti Lehti has later developed some of these arguments when analyzing the homicide trends in Finland from the late nineteenth century on. The homicide rates in Finland and in several Eastern European countries have been considerably higher than in most of Western Europe. In the case of Finland, at least, this is a comparatively recent phenomenon. It seems that industrialization had a different effect on violence in Finland than in most Western European countries. Many Western European counties experienced a short term rise in violence at the beginning of industrialization but in Finland the rise was much steeper, and it turned out to be more permanent. This increase can be separated into three different waves (1880–1904, 1905–1913 and 1919–1932). The first wave started with industrialization and Lehti calls attention to the specific character of Finland’s industrialization. Increasing

Homicide in Scandinavia

59

homicide rates mainly occurred in towns where the wood-processing industry developed and in the rural lumber districts. These communities were extremely dominated by young men. Life was dull but the income level was relatively high, and the workers consumed large amounts of alcohol. When industry developed in older and more well established urban settings, industrialization did not promote a corresponding increase in violent behavior. According to Lehti, the political situation in Finland also affected how industrialization influenced the use of violence. The lack of democratic development in the Russian empire and the weak labor movement meant that some of the forces restraining violence in many Western European countries were simply not at hand in Finland (or in other parts of the Russian empire). In addition, when the labor movement developed in Finland it also tended to idealize violence rather than to neutralize it. (Lehti, 2001) The second wave of increasing homicide rates in Finland (1905–13) coincided with similar trends in the whole Russian empire. Lehti connects this with the political development. The revolution of 1905 caused a collapse in the ideological basis for the existing political, and social order and the growth in violence was actually part of a more general crime wave. But 1905 also caused changes in the general attitudes to both violence and the control system. Young working class men began to accept the use of violence as a legitimate way of pursuing their economic and social interests. Simultaneously, the acceptance of the legal and political system collapsed among the working class. The third wave during the 1920’s and the 1930’s brought violence to a new peak in Finland. The homicide level of those decades was the result of several coinciding factors. The working class men stuck to their violent behavior, but the increase in homicides was mainly caused by changes within the rural population. Young men of the land-owning agricultural families got more and more involved in violent behavior. This is also an example of developments in Finland that differ when compared to Western Europe. On the other hand, similar trends are known in other former parts of the Russian Empire, such as Estonia. (Lehti, 2001, pp. 56–61) Contrary to the working class men, the young farmers were well integrated in society. They had won the Civil War of 1918, they had reached political power and they could identify themselves with independent Finland. In addition, agriculture developed favorably and farmers’ incomes increased rapidly. Rural violence actually increased most rapidly in those areas where the modernization of agriculture was most rapid. The agricultural population did not benefit equally from this economic development. There was a deficit of land, especially in southern Finland, and this problem grew even worse because of the high birth rates at the beginning of the century. In this situation of social frustration the attitudes to violence changed among young male farmers. As the working class men, they began to accept violence as a way of promoting their social and political goals. This development also coincided with a widespread possession of handguns as a result of the Civil War and with lax gun laws. A rising number of homicides with handguns accounts for almost the whole

60

D. Lindstro¨m

increase in homicide rates. The widespread use of alcohol also contributed to this development. In 75–80 percent of the homicide cases during the 1920’s at least one of the persons involved was intoxicated. When homicide rates began to decrease by the mid-1930’s there was a short term decrease in alcohol consumption due to the Depression. The working class gradually became more integrated in society and the proportion of young men decreased because of declining birth rates. In addition, a new police force was organized with extensive authority to search for alcohol and illegal weapons. New gun laws were passed as well. The combination of these social, economic and political factors led to a substantial decrease in violence. (Lehti, 2001, pp. 58–67)

Conclusions There is general agreement among Scandinavian historians that a long-term decrease in lethal violence occurred, and that the main transition probably took place during the seventeenth century. As new research has been conducted, the long-term development of lethal violence also appears as a much more complicated process than has sometimes been assumed. Numerous mid- and short-term peaks need to be explained. Furthermore, there are considerable regional differences that sometimes are even more apparent than the chronological changes. A major problem, however, is the lack of long-term studies of specific regions. Conclusions on long-term trends are still to a great extent based on a sample of several short-term studies. The estimated homicide rates for late medieval and early modern Scandinavia, often referred to, are extremely uncertain and should be used only with great caution. Still, this does not put the general transition into question. Several researchers have called attention to the divergent development in Finland. Contrary to the rest of Scandinavia and to Western Europe, Finland has experienced several substantial increases in homicide rates after the seventeenth century. The average Finnish homicide rate was much higher than in Sweden already during the eighteenth century. Deadly violence in Finland rose to an exceptional level during the inter-war period. After the 1930’s the homicide rate has been reduced substantially but it still remains on a rather high level by international standards. Finnish historians have discussed the peculiarities of Finland with reference to specific social, economic and political conditions. These arguments are important as a contribution to the general discussions on long-term trends in violence as well. Generally, Scandinavian historians have used a number of different types of interpretations and explanations concerning the trends in violent crimes. In short they can be characterized as: 1. State building and state control, 2. Institutional change,

Homicide in Scandinavia

61

3. Socio-economic change and socio-economic tension, 4. Psychological and cultural change – including fundamental changes in mentalities and in attitudes towards violence and the use of violence. Very little has still been done in the field of systematic comparative studies. It is true that estimated homicide levels have been compared and that both similarities and differences have been observed, but the number of comparative Scandinavian studies where violence is systematically analyzed and combined with socio-economic, ideological, institutional and cultural aspects is still rather limited. For this reason the explanations put forward have largely the character of hypothetical ad-hoc interpretations of empirical results rather than systematically tested hypotheses. A more systematic theoretical and methodological discussion on these matters is necessary to go beyond mere assumptions about causes and explanations. Martti Lehti has called attention to the necessity in combining different explanations, and he also emphasizes the fact that long-term and short-term changes may have very different causes. It seems reasonable to assume that the longterm trend, more or less common for all of Western Europe, must be the result of a very general transformation in the early modern European society. It also seems reasonable to assume that short-term changes and regional differences must be explained differently than the general long-term development. Such explanations must be based on specific local conditions as in Lehti’s explanation of the divergent development in Finland. I also agree with Liliequist when he claims that we need a lot more of systematic analyses of early modern societies with low levels of violence. Too much one-sided attention has hitherto been paid to the high homicide rates. We also need to analyze and to understand the peaceful societies in history.

References A˚gren, M. (1988). Att lo¨sa ekonomiska tvister – domstolarnas fra¨msta sysselsa¨ttning pa˚ 1700talet? Historisk tidskrift, 108, 481–511. Appel, H. H. (1999). Tinget, magten og æren. Studier i sociale processer og magtrelationer i et jysk bondesamfund i 1600-tallet. Odense: Odense Universitetsforlag. Bagge, S., & Mykland, K. (1987). Norge i dansketiden (Politikens Danmarkshistorie). København: Politikens forlag. 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, 618–638. Gurr, T. R. (1981). Historical Trends in Violent Crime: A Critical Review of the Evidence. Crime and Justice. An Annual Review of Research, 3, 295–353. Helle, K. (1982). Bergen bys historie 1. Kongssete og kjøbstad. Fra opphavet til 1536. Alma Mater: Bergen. von Hofer, H. (1985). Brott och straff i Sverige. Historisk kriminalstatistik 1750–1840. Diagram, tabeller och kommentarer. Stockholm: Statistiska centralbyra˚n (CSB).

62

D. Lindstro¨m

von Hofer, H. (1997). Nordic Criminal Statistics 1950–1995 (Department of Criminology, Stockholm University, Report 1997:2). Stockholm. von Hofer, H. (2003). Crime and Punishment in Sweden: Historical Criminal Justice Statistics 1750–2000. Journal of Scandinavian Studies in Criminology and Crime Prevention, 4, 162–179. Jansson, A. (1998). From Swords to Sorrow. Homicide and Suicide in Early Modern Stockholm (Stockholm Studies in Economic History 30). Stockholm: Almqvist & Wiksell. Jarrick, A., & So¨derberg, J. (Eds.) (1994). Ma¨nniskova¨rdet och makten. Om civiliseringsprocessen i Stockholm 1600–1850. Stockholm: Stockholmia. Jarrick, A., & So¨derberg, J. (1998). Odygd och vana¨ra. Folk och brott I gamla Stockholm. Stockholm: Rabe´n Prisma. Johansen, J. C. V., & Stevnsborg, H. (1986). Hasard ou Myopie. Re´flexions autour de deux the´ories de l’histoire du droit. Annales ESC, 41, 601–624. Karonen, P. (1995). Trygg eller livsfarlig? Va˚ldsbrottsligheten i Finlands sta¨der 1540–1660. Historisk tidskrift fo¨r Finland, 80, 1–11. Karonen, P. (1998). A Life for a Life versus Christian Reconciliation: Violence and the Process of Civilization in the Towns of the Kingdom of Sweden during the Years 1540–1700. In H. Ylikangas, P. Karonen & M. Lehti (Eds.), Five Centuries of Violence in Finland and the Baltic Area. Academy of Finland: Helsinki, 129–195. Kaspersson, M. (2000). Do¨dligt va˚ld i Stockholm pa˚ 1500-, 1700- och 1900-talen. Stockholm: Kriminologiska institutionen. Lager, B. (1962). Stockholms befolkning pa˚ Johan III:s tid. Stockholm: Stadshistoriska institutet. Larsson, L.-O. (1982). Sma˚la¨ndsk historia. Stormaktstiden. Stockholm: Norstedts. Lehti, M. (2001). Homicide Trends in Finland and Estonia in 1880–1940: Consequences of the Demographic, Social and Political Effects of Industrialization. Journal of Scandinavian Studies in Criminology and Crime Prevention, 2, 50–71. Liliequist, J. (1999). Violence, Honour and Manliness in Early Modern Northern Sweden. In M. Lappalainen & P. Hirvonen (Eds.), Crime and Control in Europe from the Past to the Present. Helsinki: Academy of Finland, 174–202. Lilja, S. (1996). Sta¨dernas folkma¨ngd och tillva¨xt. Sverige (med Finland) ca 1570-tal till 1810-tal. (Historisk ta¨tortsstatistik, del 2). Stockholm: Norstedts. Lindstro¨m, D. (1994). Oa¨rliga ma¨stare och kivande makar. Ett och annat om ra¨ttskipning, kriminalitet och normsystem i 1500-talets Norden. Historisk tidskrift, 114, 513–554. Lindstro¨m, D. (2003). Fra˚n lokal konfliktlo¨sare till administrativ stab. Ra˚d och ka¨mna¨rsra¨tt i karlstad under 1600-talet. Scandia, 69, 3–33. Lindstro¨m, D. (2007). Fejd i medeltidens och 1500-talets Sverige. In E. Opsahl (Ed.), Feider og fred i nordisk middelalder. Oslo: Unipub, 107–134. Løyland, M. (1992). Slagsma˚l, leierma˚l og bøtlagte egder 1600–1700. Oslo: Tingboksprosjektet. Magnusson, L. (1988). Den bra˚kiga kulturen. Fo¨rla¨ggare och smideshantverkare i Eskilstuna 1800–1850. Stockholm: Fo¨rfattarfo¨rlaget. Næss, H. E. (1994). Vold. In K. Tønnesson (Ed.) Normer og sosial kontroll i Norden ca. 1550–1850. Domstolene i samspill med lokalsamfunnet. (Det 22. nordiske historikermøte, Oslo 13.– 18. August 1994). Oslo: Universitetsforlaget, 62–80. O¨sterberg, E. (1991a). Brott och social kontroll i Sverige fra˚n medeltid till stormaktstid. Godtycke och grymhet – eller sunt fo¨rnuft och statskontroll. Historisk Tidsskrift (Norway), 70, 150–165. O¨sterberg, E. (1991b). Kontroll och kriminalitet i Sverige fra˚n medeltid till nutid: Tendenser och tolkningar. Scandia, 57, 65–87.

Homicide in Scandinavia

63

O¨sterberg, E. (1991c). Violence among Peasants. Comparative Perspectives on Sixteenth- and Sevententh-Century Sweden. In E. O¨sterberg (Ed.), Mentalities and Other Realities. Essays in Medieval and Early Modern Scandinavian History. Lund: Lund University Press, 89–112. O¨sterberg, E. (1992). Criminality, Social Control, and the Early Modern State: Evidence and Interpretations in Scandinavian Historiography. Social Science History, 16, 67–98. O¨sterberg, E. (1996). Criminality, Social Control, and the Early Modern State: Evidence and Interpretations in Scandinavian Historiography. In E. A. Johnson & E. H. Monkkonen (Eds.) The Civilization of Crime. Violence in Town and Country since the Middle Ages. Chicago: University of Illinois Press, 35–62. O¨sterberg, E. & Lindstro¨m, D. (1988). Crime and Social Control in Medieval and Early Modern Swedish Towns (Studia Historica Upsaliensia 152). Stockholm: Almqvisr & Wiksell. Sandmo, E. (1992). Tingets tenkema˚ter. Kriminalitet og rettssaker i Rendalen 1763–97. Oslo: Tingboksprosjektet. Sandmo, E. (1994). Æren og ærekrenkelsen. In K. Tønnesson (Ed.), Normer og sosial kontroll i Norden ca. 1550–1850. Domstolene i samspill med lokalsamfunnet. (Det 22. nordiske historikermøte, Oslo 13–18. August 1994). Oslo: Universitetsforlaget, 81–86. Sandmo, E. (1999). Voldssamfunnets undergang. Om disiplineringen av Norge pa˚ 1600-tallet. Oslo: Universitetsforlaget. Sandmo, E. (2000). The History of Violence in the Nordic Countries: A Case Study. Scandinavian Journal of History, 25, 53–68. Sandnes, J. (1990). Kniven, ølet og æren. Kriminalitet og samfunn i Norge pa˚ 1500- og 1600-tallet. Oslo: Universitetsforlaget. Sjo¨berg, M. T. (1990). Staten och tinget under 1600-talet. Historisk tidskrift, 110, 161–190. So¨derberg, J. (1990). En fra˚ga om civilisering. Brottma˚l och tvister i Svenska ha¨radsra¨tter 1540–1660. Historisk Tidskrift. 110, 229–258. So¨derberg, J. (1993). Civilisering, marknad och va˚ld i Sverige 1750–1870. En regional analys. Stockholm: Almqvist & Wiksell. So¨derberg, J. (1998). Va˚ld och civilisering i Sverige 1750–1870. Stockholm: Podium. Spierenburg, P. (1994). Faces of Violence: Homicide Trends and Cultural Meanings: Amsterdam 1431–1816. Journal of Social History, 27, 701–716. Spierenburg, P. (1996). Long-Term Trends in Homicide: Theoretical Reflections and Dutch Evidence, Fifteenth to Twentieth Centuries. In E. A. Johnson & E. H. Monkkonen (Eds.), The Civilization of Crime. Violence in Town and Country since the Middle Ages. Chicago: University of Illinois Press, 63–105. Stone, L. (1983). Interpersonal Violence in English Society 1300–1980. Past and Present, 101, 22–33. Sundin, J. (1992). Fo¨r Gud, staten och folket. Brott och ra¨ttskipning i Sverige 1600–1840. Stockholm: Nordiska bokhandeln. Verkko, V. (1946), Va˚ldsbrottlighetens utveckling och lagbundenhet i Sverige och Finland 1750–1940. Statsvetenskaplig tidskrift, 49, 325–340. Westling, C. (2002). Sma˚stadens dynamik. Ska¨nninges och Vadstenas befolkning och kontaktfa¨lt ca 1630–1660. Linko¨ping: Linko¨pings universitet. Ylikangas, H. (1976). Major Fluctuations in Crimes of Violence in Finland: A Historical Analysis. Scandinavian Journal of History, 1, 81–103. Ylikangas, H. (1994). A¨tten och va˚ldet. Va˚ldsbrotten i Norden vid o¨verga˚ngen till nya tiden. Historisk tidskrift fo¨r Finland, 79, 3–25. Ylikangas, H. (1998a). What Happened to Violence. An Analysis of the Development of Violence from Medieval Times to the Early Modern Era Based on Finnish Source Material. In H. Ylikangas, P. Karonen & M. Lehti (Eds.), Five Centuries of Violence in Finland and the Baltic Area, Academy of Finland: Helsinki, 7–128. Ylikangas, H. (1998b). The Knife Fighters. Violent Crime in Southern Ostrobothnia 1790–1825. Helsinki: Suomalainen tiedeakatemia.

64

D. Lindstro¨m

Ylikangas, H. (1999). Reasons for the Reduction of Violence in Finland in the seventeenth Century. In M. Lappalainen & P. Hirvonen (Eds.), Crime and Control in Europe from the Past to the Present. Helsinki: Academy of Finland, 165–173. Ylikangas, H., Johansen, J. C. V., Johansson, K., & Næss, H. E. (2000). Family, State, and Patterns of Criminality: Major Tendencies in the Work of the Courts, 1550–1850. In E. O¨sterberg & S. Sogner (Eds.), People Meet the Law. Control and Conflict-handling in the Courts. The Nordic Countries in the Post-Reformation and Pre-Industrial Period. Oslo: Universitetsforlaget, 57–139.

Violence in France’s Past: An Anthropological Approach Franc¸ois Ploux

In France, the 1960s and 1970s were a major turning point in the field of the history of ordinary violence. Up to that period, researchers concentrated above all on studying collective forms of violence and popular protest, in continuation of the work done by Georges Lefebvre in particular. For the most part, the history of violence, and especially that of rural violence, was a history of crowds and protests (Corbin, 1991). From the 1960s on, a certain number of authors endeavoured to use the archives of the legal system with a view to analysing everyday crime and interpersonal forms of violence. The emergence of this new subject in history, which coincides with French sociology’s rediscovery of the crime issue, occured in an intellectual context denoted by the generalised practice of quantitative history and by the importance the disciples of the Annales School had given to the concept of ‘‘mentalities’’. Actually, during these pivotal years two different approaches to violence studies appeared. Two historiography traditions developed and for the most part have remained, up to present day, impervious one to the other (rare are the researchers who, like Jean-Claude Farcy (1989), strive to reconcile the two approaches and go beyond the divide. The history of violence was first an integral part of a quantitative history of crime which was very trendy in the 1970s. Several historians subjected judicial archives to the methods of serial history, then in its golden age1. We must admit that judicial archives are ideally suited to quantification procedures: we find records of all the offences committed, each year, within the jurisdiction of the various courts (bailiwick courts, police courts, criminal courts, etc.). Specialists of the nineteenth century also have at their disposal, with the Compte Ge´ne´ral de l’Administration de la Justice Criminelle, a series of

Franc¸ois Ploux Franc¸ois Ploux, 4, rue Jean-Zay, BP 92116, 56321 LORIENT CEDEX; 4, alle´e des Fauvettes, 56530 GESTEL, France 1

In particular, we can mention the work begun in Caen by Pierre Chaunu (Boutelet, 1962; Gegot, 1966). Concerning the nineteenth century, we can mention a later article by Gabriel De´sert (De´sert, 1981)

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

65

66

F. Ploux

annual volumes listing, from 1825 on, the various crimes and offences committed in the whole of France, together with a certain amount of information concerning the accused (their age, home address, civil status, etc.). It seems important to highlight some of the characteristics of the serial history of violence which ensued. First of all, it reduces the crime (homicide, fight) to a figure. For the researcher it comes down to classifying and to highlighting the diachronic variations. In a way, the researcher decontextualizes the crime which he strives to explain. Nevertheless, the quantification procedure makes it possible to look for connections between the main variables which could explain the temporal and spatial variations of crime (at the expense of theorizing on motives: this macroscopic history of violence has paid little attention to motives for crimes, to the inner workings of violence.) Finally, the researchers who have worked in this direction have never really dissociated the history of violence from a history of crime which is their actual purpose. In more general terms, these first historians of violence have paid very little attention to the problem which is automatically brought up by the study of a society through the prism of categories used in judicial thought: many analyses were founded notably on the opposition between crimes against people and crimes against objects. The category of ‘‘violence’’ itself Was rarely discussed as if it were a universal characteristic. As if this notion could not cover a whole range of behaviour which is actually very heterogeneous. Putting homicide, pilfering and sexual violence together as products of a single cause (man’s aggressiveness), means neglecting social norms, the schemas of perception and assessment that give such actions a particular meaning. It gives violence a single meaning, separate from the multiple forms that it can take. Here we must quote Philippe Robert (1995): ‘‘Penal rules bring together in a legally homogeneous category acts that are behaviourally heterogeneous and whose motivations run the full range of the factors determining human actions’’. Very early on, from the mid 1970s, some authors voiced serious doubts about the plausibility of this method of exploitation of judicial archives applied to the study of real criminality (and not of feelings or representations for example). In a seminal article, Michelle Perrot (1975), wondering how best to use judicial statistics – and more particularly the much talked-about Compte Ge´ne´ral de l’Administration de la Justice Criminelle – thought that no ‘‘purely positivist’’ use of this source was possible. It would be illusory, she stated, to claim to shed light on ‘‘actual’’ criminality using that series: criminal facts do not exist per se, i.e. independently of a set of principles showing specific types of behaviour as being deviant, and situating them on a scale of transgressions. It is that set of principles, which vary according to the obsessions of the persons drawing them up, that criminal statistics may be able to reveal. Variations in the rate of convictions in the light of the type of offence, or depending on the profile of the delinquent, are above all signs of the penal sensitivity of a society or a social class. With a several-year interval, another approach to violence was developed. This time it is a method of history, contrarily to the previous, which proceeds by

Violence in France’s Past: An Anthropological Approach

67

contextualisation. In fact, the replacement of a serial approach by a qualitative approach is a sign of the shift from a history of deviance to a social and cultural history of violence. In modern history, it was the work of Arlette Farge that first showed this new approach to the archives of crime2. Arlette Farge takes care to show the mechanisms guiding collective life in eighteenth-century Paris, by studying the antagonisms, clashes, disagreements and conflicts that were the daily lot of the city’s population. Very careful attention is paid to the tiniest details mentioned in the files consulted. Elisabeth Claverie’s study of violence in the Gevaudan region is also characteristic of the abrupt change in the ways of dealing with the data (Claverie & Lamaison, 1982). The author has examined in detail all the files concerning cases brought before the Loze`re department criminal court. And yet her work contains no figures whatsoever. There are no graphs, statistics, charts, or averages. On the other hand, several remarkable cases are dealt with in the form of a detailed description. The researcher’s work here does not consist of reducing criminal acts to mere quantifiable values, to integrate them in a series. It is not a question of taking the crime out of its context, isolating it, or stripping it of all the elements that set it apart, to bring it into line with other actions with which it has no direct connections, but which are considered as being of a similar nature. The researcher tends rather to put the acts into their context. He takes pains to reconstruct the social, cultural, and normative environment of the crime. To grasp the meaning of the criminal act, he takes into consideration the social identity of its author (and that of its victim), his intentions, and his likely motivations. He has to examine the reactions stemming from the deed, too. It is also possible, in this ethnographic type of study, to establish connections and make comparisons that enable us to distinguish between various categories (family criminality, violence to avenge honour, etc.) and bring to light regularly recurring aspects. But these classification operations are constructed by the researcher. They do not necessarily reproduce the categories of judicial taxonomy: it is indeed advisable to avoid mixing together in a given statistical category behaviour patterns that are seen as being radically heterogeneous. That is why this type of historical anthropology of violence is characterized by the researcher’s desire to bring to light the normative systems and the thresholds of tolerance that are specific to the society being studied. Elisabeth Claverie has thus shown that in the Gevaudan during the nineteenth century, violence was a way of resolving conflict that was seen as being normal most of the time. Murder was part of the exceptional-normal actions.

2

Among the many works written by Arlette Farge, we can mention (Farge, 1974; Farge, 1979; Farge, 1986; Farge, 1989). We should also mention the research carried out at about the same time by Nicole and Yves Castan on criminals in Languedoc region. (Castan, 1974; Castan, 1980).

68

F. Ploux

Violence and Domestic Reproduction We can attribute the recent progress made in historical research concerning analysis of family violence, to the development of an anthropological approach to crime. And the effects of the progress made in this field reflect the importance taken on by family criminality at the beginning of the contemporary period. A considerable number of homicides were committed within households or among close relations. And domestic violence was not only an everyday occurrence: it was also particularly cruel. In the Quercy region during the early nineteenth century, for example, the brutality of a conflict was inversely proportional to the social distance between the protagonists: confrontations between members of a given household, or disputes between neighbours used to helping one another out, were hardly ritualized at all, and they were often terribly bloody (and yet they were much less worrying for the ruling classes than wars between villages or tax riots, which were more spectacular, it is true, but all in all involved much less bloodshed (Ploux, 2002)). We now find it easier to understand the mechanisms behind family criminality, because we have a much better idea of what a peasant family was like in pre-industrial France. Researchers, most of them trained in social anthropology, have studied in great detail the ways in which families were organized, how the family heritage was passed on from one generation to the next, the power relations in the households, how the peasants saw the family and its relationships, etc. And the reflection concerning domestic violence has consisted mainly of comparing the forms of conflictual relations revealed by judicial archives, with the anthropological structures of the rural family. Domestic crime is in fact at once a product of and a pointer to these structures. There can be no doubt that the most extreme cases of violence – those that led to a murder and resulted in an intervention by the forces of justice – were exceptional. But it was their brutality that set them apart from more ordinary clashes. Most of the murders committed within families were the outcome of conflicts that any household could go through, at one time or another in its history. In rural France during the nineteenth century, the modalities under which children acquired a position of domestic power and were able to deal with the family property as they saw fit, were set out in customary rules, which also varied from one region to another. Furthermore, each member of the household was assigned a position in the organization of the household, based on a fairly strict differentiation of the roles and competences. Although these rules set up a system that created inequality between the children, who were more or less favourably treated according to their sex or their order of birth, the underlying principles were rarely contested. Nonetheless, their application could turn out to be difficult and painful. The ambitions of the various members of the household were difficult to reconcile at times. It was common for siblings, or parents and their children, to be opposed concerning how to manage the farm, or how to pass on the powers and the land. Whether the system involved

Violence in France’s Past: An Anthropological Approach

69

dividing up the family heritage into equal shares, or, on the contrary, handing it down to the firstborn, the prospect of an inheritance aroused greed, and led to an atmosphere of competition between the siblings. Violence was often an expression of spite or disappointment. The question of access to the land, which was especially crucial at the time of transmission between generations, was at the heart of family conflicts (Sylvie Lapalus (2004), the author of a considerable amount of work on murder of fathers, further notes that parricide was only rarely committed by tenant farmers, sharecroppers or journeymen, in other words, by peasants who did not own any land). We can add that in the nineteenth century, the customary system for passing on family property was subjected to a set of constraints, linked to the increasing life expectancy and to the adoption of the French Civil Code (which made it illegal, and theoretically impossible, to hand down all the family property to just one child). This led to a fresh upsurge in conflicts between family members, especially in the south of the country. Most of the work done on family criminality concerns the southern part of France, characterized by the persistence, at the time, of non-egalitarian practices in the field of succession and inheritance: the farmhouse, its outbuildings, the land, the tools, and the livestock, but also power in the household, were passed on to just one child, almost always the eldest son, and his younger siblings had to make do with a modest amount of compensation in cash (Augustins, 1989; Claverie & Lamaison, 1982; Ploux, 2005). The numerous studies carried out in the Judicae¨l archives have shown that the mechanisms for passing on household property from one generation to the next (naming the heir, passing the property and the powers on to the future head of the house, exclusion of his younger siblings) led to conflict and violence. Homicides committed within the family were often premeditated, but they were frequently the result of an impulsive, uncontrolled gesture. This means that the criminal’s temperament is one of the explanatory factors that it is necessary to take into consideration in the analysis. Nonetheless, such criminal acts, though apparently irrational, were always a result of an underlying conflict. Furthermore, the incident that triggered the attack was never insignificant. It was frequent, for example, for a son to quarrel with his father about the keys to the cupboards: having the bunch of keys meant having access to the provisions and the savings, in other words holding sway over the household goods. It is true that it was extremely rare, in the nineteenth century, for people to contest the non-egalitarian nature of a customary norm that, in full contradiction with the letter of the French Civil Code, confers household power and the full inheritance on just one child. Such violence was hardly ever perpetrated by younger siblings refusing to accept the basic principles underlying the system. On the other hand, transfers of property and power always led to tension, which often triggered disputes, physical violence, and at times even murders. The disappointment of a firstborn child, suddenly cut off from his inheritance because of his bad behaviour, in spite of the fact that his ranking as the eldest child put him in line as the future head of the household, could lead to

70

F. Ploux

acts of violence. A negotiated settlement concerning the amount of the dowry granted to the younger siblings in consideration of their renunciation of a share in the inheritance often led to tensions among the siblings. Violence could also be a form of contestation concerning the power of the head of the household: at times, an heir who lived in the same house as his parents, even though he was married, found it difficult to have to suffer the authority of an ageing father. Some of them got rid of their mother or father who were too old to work, but who still had to be fed and clothed. Conflicts between husband and wife were no less common, and their issue was often tragic. In fact, the root causes of the poor relations in the couple were almost always the same. The hypothesis most frequently found was that of the husband or wife reproaching their spouse for endangering the economic balance of the household, and hence its future. The long-term existence of the household over the generations was a real cultural imperative: all who jeopardized the future of the farm, through their negligence, incompetence, laziness or drinking, were the object of hatred on the part of the other family members. Violence was also a way of compensating for a lack of authority linked to the status held in the household. In general, the most ghastly murders were committed by younger brothers who had married an only daughter: she had inherited the family property. But in the household, (where he was said to have ‘‘entered as a son-in-law’’) her husband occupied the rank usually reserved for a daughter. His position as compared with his wife went against the cultural principle of male domination. This led to a feeling of unrest that could be the origin of conflicts of incredible violence. Although in a large part of the south of France (Pyre´ne´es, Provence, Quercy, Gevaudan, etc.), the customary system based on the principle of ensuring the long-term existence of the peasants’ home had been applied for a very long time, in other parts of the country the forms of family organization, the practices for passing on property and the types of domestic conflict found were very different. Sylvie Lapalus (2004) has shown this in her recent work dealing with murders of fathers. In 68% of the cases, it was a question of money that was the root cause of the drama (more rarely, parricide was an act of revolt against a use of paternal power that was seen as being abusive). However, although in the south of France, i.e. within the limits of the area where non-egalitarian inheritance was the norm, parricide was mainly committed by firstborn sons impatient to take over full powers concerning the family property, but coming up against the parents’ refusal to give up their control over their property, in the countryside further north, where equality was the rule regarding inheritance, the reason for such murders by some children was to avoid having to pay life annuities to their parents. Once the mechanisms leading to tension within the family have been brought to light, it is possible to examine the factors that exacerbated or on the contrary soothed such antagonism. This means that we can explain either the increases or the falls in the frequency of homicides in the family. Unlike most of the researchers who have worked on these questions, Sylvie Lapalus has carefully assessed changes in the frequency of parricides over the century; the act as such

Violence in France’s Past: An Anthropological Approach

71

is in fact very rare, with on average only about a dozen cases noted per year. Nonetheless, the author found two peaks, the first one during the economic slump in the middle of the century, and the second one during the Great Depression of the 1880s. The frequency of assault and battery on ascendants, for its part, shows a considerable fall as from the middle of the century. It is highly probable that the first half of the nineteenth century constituted a period of intensification of family tensions, at least in the countryside in the south of France. A series of factors may well have contributed to increase the average frequency of domestic violence. The first one was the adoption of the French Civil Code: by establishing a system of almost absolute equality between heirs, the text completely disrupted the system of passing on property that was based on exclusion of the younger siblings. It is true that the peasants in the south of France were able to develop all sorts of schemes that enabled them to get round the law and preserve the former custom at all costs. But this adaptation was nonetheless a cause of tension (to increase their share, the younger siblings did not hesitate to raise the threat of taking their claims to court). The gradual increase in life expectancy slowed down the rate of the transfer of domestic power between the generations: the period of cohabitation between the heir, now an adult, and his elderly parents, tended to lengthen over the century, and many firstborn children found it hard to bear this forced submission to the authority of ageing parents, whose presence also ate into the household economy. Furthermore, due to the increasing influence of the monetary economy in the countryside, the young peasants became increasingly aware of the market value of their work, and the cost of feeding and clothing their parents. This growing ability to make such calculations explains in turn why the children were less inclined to make sacrifices for their family. On the other hand, the fall in the most brutal forms of domestic violence could be correlated with the exodus of younger brothers to the towns or to America, but also with the improved living conditions in the countryside. The very thorough study that Anne Parella (1992) has made concerning family criminality in the Nord department nonetheless reminds us that we always take risks when we seek to restrict our work to studying rates. The author does not find any major changes in the frequency of murders committed within families, between 1815 and 1904. However, in fact this apparent stability masks a radical transformation in the forms and the meaning of domestic violence. In just a few decades, mercenary crime, a tragic outcome to a conflict concerning questions of a material nature (money, inheritance, etc.) fell back and was replaced by crimes of passion. This type of evolution reflects the redefinition of roles and relations within the family: at the beginning of the century, the household group was above all a production unit. Then, as the differentiation between living space and workplace became more marked, and the wage system took hold, the family gradually lost this economic dimension. On the other hand, the affective links between members of the domestic unit showed growing importance. A new type of crime came to light: crimes of passion, committed by workmen whose wives left them.

72

F. Ploux

A Culture of Honour In the nineteenth century, violence among peasants was one of the main components of the negative image of the countryside (an image that was subsequently eclipsed, as the dominant ideology made the village as such into a sanctuary of virtues, in contrast to the town, with its corrupting, criminal, and anti-establishment trends). Administrative agents, magistrates, criminologists and novelists describe country dwellers as being uncouth, impulsive, with uncivilized manners, and easily swayed by their emotions, especially when drunk. Furthermore, rural criminality is frequently linked to the natural conditions (geology, climate), isolation and a lack of education. It is certain that brawls, murders, ambushes, and arson were common ways of settling a conflict. (We should add that most of these attacks were made between people who knew each other: they concerned peasants living in the same hamlet, the same village or two adjoining villages3). To understand these cases of violence, it is necessary to determine the specific aspects of the ethics and the normative system that underlies them. Indeed, in the France during the early part of the nineteenth century, the legal norm and the customary norm did not overlap exactly4. We can define the customary village norm as a non-formalized, highly elastic system of appreciation – but a coherent one – of what is licit, admissible, and tolerable, and of what constitutes a transgression. In a region such as the upper Quercy, it brings together a specific set of ethics based on honour to highlight modes of conflict resolution involving arrangements. We must admit that it is very difficult to describe in precise terms exactly what went into this peasant notion of honour, which gave rise to so much violence, or grasp all the values and representations that were associated with it. It is clear that the notion of honour varied depending on whether it was applicable to a woman, a man, a family or a group5. The main point here is to underline the fact that physical violence was seen as a legitimate method for avenging injured honour. In the village culture, the use of violence was accepted, tolerated, or even appreciated, provided that such violence was aimed at making up for the offence to honour caused by the slander, challenge

3

In a study of violence in the Loire-Infe´rieure (now Loire-Atlantique) department, Bernard Desmars (2005) provides interesting details concerning this question. The accused and his victim were almost always neighbours, living either in the same hamlet (32.4% of the cases), in the same village (35.1%), or in adjoining villages (22.5% of the cases). 4 The difference, wich was gradually to disappear, varied in extent depending on the cultural area considered. The difference is clearly greater in the south of France, where the frequency of attacks is high, than in the country in the Paris region. 5 The honour of the villages is what gives it its identity: hence the frequency of violence between villages, which were caught up in cycles of challenges and vengeance. On this point: see (Ploux, 2002).

Violence in France’s Past: An Anthropological Approach

73

or insult in question. Furthermore, any quarrel or dispute, whatever the actual reason, was seen as bringing the protagonists’ honour into play. Brawls between young men in the dance halls, at the time of dances, over a girl – ‘‘pure’’ types of violence involving honour – are not the only ones that we can put under this heading. Conflicts between neighbours, for example, were often triggered by a quarrel over adjoining plots of land, contestation concerning a right of way, repeated straying of livestock onto ploughed land, etc. Once the informal system of swapping services, so characteristic of relations between good neighbours, had been abandoned, the conflict took the form of a cycle of challenges and vengeance that often ended in a bloody fight between the two families, with all the family members taking part. Let us now take the example of conflicts between rival households on the matrimonial market. Elisabeth Claverie (Claverie & Lamaison, 1982) has shown that in the Loze`re department, there was a logical link between the violence exercised within the families and that opposing the members of two rival households. The decision as to the future heir led to all sorts of conflicts within the household; rivalry between families stemmed from competition, on a tightly compartmentalized matrimonial market, between families wanting to marry off their eldest son to a girl with a dowry, but who was not an eldest daughter. Once again, it was the imperative need to pass on all the family property that accounted for most of the conflicts between villages or villagers. But this rivalry again took on the characteristics of cycles of challenges and vengeance, as though the family honour were the real element at stake during the confrontation. In the early nineteenth century, rural France, and especially in the southern regions, was indeed a society in which challenges and vengeance took on considerable importance. The brutality of the peasants’ manners took its roots in the importance attached to physical violence as a way of showing assertiveness and settling conflicts involving honour. But this specific set of ethics cannot be dissociated from recourse to procedures for settling quarrels that involved compensation rather than sanctions. Examination of the archives of the criminal courts shows that private arrangements were common; they could take the form of pecuniary compensation or a ritual of reconciliation. In fact, challenges and offences to honour could give rise to three types of reaction: Firstly, a countering action. If the offended person rose to the challenge, the adversaries started on a cycle of vengeances. This cycle, which became increasingly brutal, often led to murder of one of the protagonists. Violence here is seen as an exchange: the vocabulary of vengeance considers such vengeance as a debt (hitting a adversary who has insulted you or challenged you is like paying a debt owed to him (Ploux, 2002)). Obtaining full vengeance through murder is hence a cultural imperative. But the testimony recorded in the case files also shows that it was a psychological necessity: until he had avenged the outrage suffered, the

74

F. Ploux

offended person was shaken by an internal crisis that could only be resolved by a criminal act. Second possibility: an arrangement. It was always possible to stop the vicious circle by coming to an arrangement: the author of the attack paid pecuniary compensation to the victim’s parents. Large-scale recourse to this type of solution in a conflict would seem to have been a collective strategy for preservation of a customary norm (the amount of the compensation depended on the local method used to assess the size of the prejudice suffered). Finally, a peasant who felt that he was not strong enough to rise to the challenge gave up his intention to avenge the insult: such absence of reaction on his part was synonymous with humiliation.

The Internalization of the Judicial Norm and the Decline of Rural Violence This ‘‘anthropological’’ notion of violence, which consists of putting criminal acts in relation with the dynamics of social relations and the configuration of cultural norms, provides a possible interpretation of fluctuations in rural criminality. What do the figures tell us about this? According to a number of specialists in rural civilization, from the 1860s on, there was a gradual fall in levels of violence in the countryside (where over half the population of France still lived). On this subject, Alain Corbin (1991) mentions a ‘‘historical fact that has been pointed out for a long time’’. And he goes on to list the probable causes: lessening and finally disappearance of the wheat crises, gradual integration of the peasants into the Nation State, breakdown of the traditional structures making up the village society, and the process of ‘‘civilization of manners’’. Fre´deric Chauvaud (1991) also mentions a fall in levels of violence in the nineteenth century: it ceased to be a commonplace way of dealing with rural conflicts. Chauvaud links this phenomenon to a change in the thresholds of sensitivity. Jean-Claude Farcy (2001), another major specialist in judicial archives, states in a recent summary that the hypothesis of a fall in levels of violent behaviour during the second half of the century would seem to be undisputed. However, the small amount of work done in the field of quantitative history of crime would seem to contradict this image of a fall in levels of social violence, in the countryside, as from the middle of the century. We can examine the studies carried out by Jean-Claude Chesnais (1981), Howard Zehr (1976) and Gabriel De´sert (1981), although we must point out that they are based on a single source: the Compte ge´ne´ral de l’Administration de la Justice Criminelle. It is true that, as Jean-Claude Chesnais (1981) states, the annual number of murders (together with the corresponding rate) tends to fall between 1820 and 1980. However, this fall in the extreme forms of violence was far from being

Violence in France’s Past: An Anthropological Approach

75

regular. The frequency of homicides shows a downward trend from 1825 to 1860, in spite of a peak around 1845–1850 and considerable short-term fluctuations. On the other hand, the graph shows an upward trend between the 1860s and the 1920s (at that date, we find rates close to those for the early nineteenth century). Between 1920 and 1960, the frequency of murders falls steadily. For the nineteenth century, Gabriel De´sert (1981) obtains roughly similar results: a fall in levels of violence in the early decades of the century, but with an upward trend from about 1855 on. Overall, we find a global increase in the frequency of crimes and violence during the century. The work done by Howard Zehr (1976), which is very rarely mentioned by the French authors, is much more detailed than the two studies referred to above. This is due especially to the fact that it makes a distinction between minor cases of violence (assault and battery) and homicides. In fact, the changes in the frequency of these two forms of violence are exactly opposite. Concerning homicide, the results obtained are somewhat different from those of Chesnais and De´sert. The American historian finds a fall in the rates between 1830 and 1870 (interrupted, it is true, by a strong short-term increase in the second half of the 1840s). On the other hand, the graph shows an upward trend from 1870 to 1910. Nonetheless, in 1910, it does not quite reach the rate found for the 1830s. Concerning minor cases of violence, the increase is continuous from 1830 to 1910 (their frequency increased by 121% between these two dates). And the fluctuations recorded during the century are not the same as those noted for homicides. Roughly speaking, the rates increase continuously from 1830 to 1910, except during the first half of the 1850s. Furthermore, Howard Zehr shows that the urbanization did not lead to any major rise in violent crime. In a study published in 1982, Cohen and Johnson (1982) also confirmed, with figures to back up their statements, the absence of correlation between murders and urban growth6. We must thus admit that the overall increase in violent crime during the century reflects at least partially an increase in social violence in the countryside. Howard Zehr (1976) has also established that in the three rural departments of Loir-et-Cher, Vende´e and Creuse, the frequency of assault and battery as compared with the size of the population showed a constant increase between 1827 and 1904. We could object, it is true, that the elements used by these different authors to attempt to assess the changes in levels of violence are somewhat coarsely defined (it would no doubt be useful to carry out more accurately located research, and distinguish between various sub-groups within the ‘‘homicide’’ and ‘‘assault and battery’’ categories, which are far too vague). However, in the absence of more detailed studies, we must admit the fact that the countryside in France was no less violent just before the First World War than immediately after the fall of Napoleon I. What rural history studies have shown very clearly for several

6

Same conclusion in (Lohdi & Tilly, 1973).

76

F. Ploux

decades now, is the sudden disappearance of collective forms of violence after 1860 (what is more, they never resulted in any large numbers of deaths). Riots stemming from food shortages, troubles linked to the defence of customary rights, tax revolts, and brawls between villages were few and far between from that date on. But the frequency of violence between persons only started to fall much later. Although there can be no doubt that ordinary forms of violence have fallen sharply over two hundred years, historians must carry out further work to account for the short-term increase between 1860 and 1910. Two major factors may have accounted for increases or falls in the rates of homicides in the countryside. The first one is the changes in the economic or demographic conditions that led to the tensions from which most of violence stemmed. In particular, it would be useful to study the extent to which a fall in levels of competition on the land market and the matrimonial market could have affected the levels of violence in social relations. Another factor is the changes in relations with the State and the judicial norms. From the 1860s on, we note a process of internalization of the judicial norm and transformation of the concept of honour. During the first half of the century, most of the time, the protagonists of conflict refused to bring their disputes before the courts, as this would have taken management of the conflict out of their hands. This reluctance can be explained by the discrepancies between normative systems: the official court system dealt with conflicts in the light of its own system of values, as expressed in the scale of penalties. More generally, it was the future of the traditional village community that was at stake. This community continued to exist for a good part of the century, as an entity ensuring social regulation, able to supervise, punish, or pardon, and hence apply and perpetuate a customary moral and system of values. The opposition to the process of monopolization by the State of legitimate punishment stemmed directly from the village communities. Furthermore, if the persons offended refused to come to an arrangement with their aggressors, they then sought vengeance, and not justice. Crimes were rarely seen as acts that could offend society as a whole. Claims before the courts were rarely motivated by a desire to defend universal principles of fairness. Obtaining a ruling against an adversary was only an indirect (and highly appreciated) way of taking revenge. Rural people saw hardly any difference between a sentence and vengeance. As a result, any court sentences were seen as a rebuff that called for vengeance. In a way, judicial violence was integrated in the practice of vengeance; recourse to justice was integrated into a specific normative order. This is why interventions by the courts in a private conflict or a war between villages always led to exacerbation of the violence. The person sentenced saw the punishment as an offence that called for vengeance. If the courts intervened in a conflict, this only made matters worse. Gradually, violence ceased to be seen as a legitimate way of dealing with conflicts of honour. And it was not so much the improved efficiency of penal repression that led to a fall in levels of violent behaviour, but rather the internalization of the judicial norm by the population in the villages. Here, it

Violence in France’s Past: An Anthropological Approach

77

is necessary to distinguish between two aspects: a gradual reduction in the discrepancies between the judicial norm and the social norm, which finally came to overlap almost completely, and the gradual acceptance of the courts as the sole legitimate entity for dealing with conflicts. Much has been written about the vectors for spreading bourgeois values in the countryside under the Third Republic. The examples usually mentioned are the schools, conscription, the press or the municipal by-laws concerning order. Nonetheless, no detailed knowledge is as yet available concerning the concrete mechanisms that led to the spread of the legal norm into the social field. On this subject, it would be advisable to follow the example of Laurent Willemez (2003), who recently brought to light the role of unofficial lawyers, people who held a law degree and were often consulted by peasants whom they met in the places where social encounters were common. We conclude with a hypothesis to be explored. The acculturation of the French peasants during the nineteenth century could be the reason for the apparent increase in levels of rural violence in the second half of the century, which, as we have seen, went against a trend seen over several centuries and showing a fall in levels of physical violence. This short-term rise apparently results not so much from an increase in the frequency of violence, as from a shift in the thresholds of tolerance towards such violence, linked to less systematic recourse to private settlements concerning conflicts of honour.

References Augustins, G. (1989). Comment se perpe´tuer? Devenir des ligne´es et destins des patrimoines dans les paysanneries europe´ennes. Nanterre: Socie´te´ d’ethnologie. Boutelet, B. (1962). E´tude par sondage de la criminalite´ du baillage de Pont-de-l’Arche. Annales de Normandie, 235–262. Castan, N. (1980). Les criminels de Languedoc: les exigences d’ordre et les voies du ressentiment dans la socie´te´ pre´-re´volutionnaire (1750–1790). Toulouse:Association des Publications de l’Universite´ de Toulouse-Le Mirail. Castan, Y. (1974). Honneˆtete´ et relations sociales en Languedoc (1715–1780). Paris: Plon. Chauvaud, F. (1991). De Pierre Rivie`re a` Landru. La violence apprivoise´e au XIXe` sie`cle. Paris: Brepols. Chesnais, J.-C. (1981). Histoire de la violence en Occident de 1800 a` nos jours. Paris: Robert Laffont. Claverie, E. & Lamaison, P. (1982). L’impossible mariage. Violence et parente´ en Ge´vaudan. XVIIe, XVIIIe, XIXe sie`cles. Paris: Hachette. Cohen, D. & Johnson, E. A. (1982). French criminality: Urban-rural differences in the nineteenth century. Journal of Interdisciplinary History, XII:3 (Winter 1982), 477–501. Corbin, A. (1991). L’histoire de la violence dans les campagnes franc¸aises au XIXe sie`cle. Esquisse d’un bilan. Ethnologie franc¸aise, XXI, 3, 224–236. De´sert, G. (1981). Aspects de la criminalite´ en France et en Normandie. Cahiers des Annales de Normandie, 13: 220–316. Desmars, B. (2005). La violence et les roˆles sociaux. Les acteurs de l’affrontement dans la socie´te´ rurale de la premie`re moitie´ du XIXe` sie`cle. In F. Chauvaud & J.-L. Mayaud (Eds.),

78

F. Ploux

Violences rurales au quotidien. Actes du 21e colloque de l’Association des ruralistes franc¸ais, (pp. 153–166). Paris: La Boutique de l’histoire. Farcy, J.-C. (1989) Les paysans beaucerons au XIXe sie`cle. Chartes: Socie´te´ arche´ologique d’Eure-et-Loir. Farcy, J.-C. (2001) L’histoire de la justice franc¸aise de la Re´volution a` nos jours. Paris: Presses universitaires de France. Farge, A. (1974). Le Vol d’aliments a` Paris au XVIIIe sie`cle : de´linquance et criminalite´. Paris: Plon. Farge, A. (1979). Vivre dans la rue a` Paris au XVIIIe sie`cle. Paris: Gallimard/Julliard. Farge, A. (1986). La vie fragile. Violence, pouvoirs et solidarite´s a` Paris au XVIIIe` sie`cle. Pari: Hachette. Farge, A. (1989). Le Gouˆt de l’archive. Paris: Seuil. Gegot, J. (1966). E´tude par sondage de la criminalite´ du baillage de Falaise. Annales de Normandie, t.XVI, 103–149. Lapalus. S. (2004). La mort du vieux. Le parricide au XIXe` sie`cle. Paris: Tallandier. Lohdi, A. Q. & Tilly, C. (1973). Urbanisation, crime and collective violence in nineteenth century France. The American Journal of Sociology, 79: 2 (Sept), 296–318. Parella, A. (1992). Industrialization and murder: Northern France, 1815–1904. Journal of Interdisciplinary History, XXII: 4 (Spring), 627–654. Perrot, M. (1975). De´linquance et syste`me pe´nitentaire en France au XIXe sie`cle. Annales E.S.C., vol. XXX 67–91. Ploux, F. (2002). Guerres paysannes en Quercy. Violences, conciliation et re´pression pe´nale dans les campagnes du Lot (1810–1860). Paris: La Boutique de l’Histoire. Ploux, F. (2005). Violence et reproduction domestique dans les campagnes du Haut-Quercy au XIXe` sie`cle. In F. Chauvaud & J.-L. Mayaud (Eds.), Violences rurales au quotidien. Actes du 21e colloque de l’Association des ruralistes franc¸ais, (pp. 137–152). Paris: La Boutique de l’histoire. Robert, P. (1995). Paradigme ou strate´gie : Pire`s et la conception du crime. De´viance et socie´te´, 19: 3, 267–278. Willemez, L. (2003). Diffuser la raison juridique. Le roˆle des avocats dans le de´veloppement de nouvelles normes sociales dans la France du XIXe` sie`cle. Intervention during the symposium on Normes sociales et processsu cognitifs, Poitiers, 18–20 juin 2003. The article reference is: http://www.u-picardie.fr/labo/curapp/Publications/Cv_Willemezfev05.pdf Zehr, H (1976). Crime and the Development of Modern Society. Patterns of Criminality: Nineteenth Century Germany and France. London: Croom Helm.

The Fall and Rise of Homicide in Ireland Ian O’Donnell

Introduction During the 1990s the level of recorded homicide rose sharply in the Republic of Ireland. This trend can be related to factors such as a booming economy, changing migration patterns and increases in alcohol consumption (O’Donnell, 2005a). Considered historically the rate of lethal violence was at a peak during the mid nineteenth century, remained relatively high until the beginning of the twentieth century and then declined until the 1970s (O’Donnell, 2005b). To illustrate these shifts this chapter concentrates on four time periods, each separated by half a century (1845 to 1854; 1895 to 1904; 1945 to 1954; and 1995 to 2004). The level of lethal violence during each period is described and related to wider social changes such as those wrought by famine (1845 to 1854), the consolidation of new family structures and the search for independence during the end of Queen Victoria’s reign and the early Edwardian era (1895 to 1904), economic depression and mass emigration (1945 to 1954), and the arrival of the so-called Celtic Tiger economy (1995 to 2004). It is hoped that this series of snapshots of a society in transition will set the context for further explorations of trends in violence as they were played out at the European periphery. Constraints of space mean that it is not possible to give more than a cursory introduction to some key historical, social and criminological themes. These wider issues are explored by other contributors to this volume. The emphasis here is to describe the Irish case and point to literature that is not always wellknown outside the country but that may be of interest to comparativists. The focus is deliberately narrow and descriptive but hopefully not too intellectually arid as a consequence. Explaining secular trends in violence, particularly with reference to the idea of a civilising process as propounded by Norbert Elias, has been a growth area of intellectual endeavour in recent years (e.g. Spierenburg, 2001; Thome, 2001). Eisner (2001, 2003) has been to the forefront in this regard Ian O’Donnell Professor of Criminology, School of Law, University College Dublin, Belfield, Dublin 4, Ireland [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

79

80

I. O’Donnell Table 1 Murder and manslaughter in Ireland Average annual rate per million population 1845–1854 24.0 1895–1904 15.8 1945–1954 3.4 1995–2004 13.7 Sources: Return of Outrages Reported to the Constabulary Office; Statistical Returns of the Dublin Metropolitan Police; Judicial Statistics, Ireland; Registrar General’s Reports; Annual Reports of An Garda Sı´ ocha´na

and sets out a framework against which the patterns outlined in the following pages can be contextualised and assessed. It is important at the outset to make one point about the range of inquiry. Infanticides (and murders of children aged under one year) are excluded from the analyses presented below because although a significant problem until the early twentieth century they spring from different circumstances, are much more problematic to count than other unlawful killings, and are difficult to place with confidence in any international context. How distressed mothers dealt with unwanted babies demands attention in its own right (for reviews of the extent of this problem see Rose, 1986; Sauer, 1978). To give a flavour of the scale of violent death during the years in question the levels of lethal violence, based on average annual rates per million population are presented in Table 1. The primary data sources are police statistics and mortality reports. During the first two periods the data are based on the entire island of Ireland, which was then a British colony. For the second two, the focus is limited to the 26 counties that became the Irish Free State and later the Republic of Ireland. The marked nature of the decline over the first three time periods is striking, as is the magnitude of its reversal. The general difficulties with official statistics, even for crimes as apparently unambiguous as homicide, have been set out in detail elsewhere (e.g. Morris, 2001; Taylor, 1998) and to avoid repetition are not recapitulated here. For an outline of available Irish sources and the particular interpretative challenges that they pose see Griffin (2005) on the nineteenth century and O’Donnell, O’Sullivan & Healy (2005) on the twentieth.

Famine (1845 to 1854) Little hard quantitative information is available for eighteenth-century Ireland but the literature portrays a society where casual violence was common. Henry (1994, pp. 77–78) notes that in 1780 there were newspaper reports of 26 murders in Dublin leading The Hibernian Journal to comment that: ‘‘Murder in this city

The Fall and Rise of Homicide in Ireland

81

has become so common, that it has lost all its horrors; every day teems with new instances of the most horrid barbarity.’’ Rape was regularly reported in the press throughout the eighteenth century and it was not unknown for a ferocious attack to leave the victim dead or dying. Children were sometimes singled out due to the belief that a male suffering from venereal disease could cleanse himself of the affliction by engaging in sexual intercourse with a virgin or a girl coming into puberty (Kelly, 1995). The data that have been unearthed thus far do not permit any detailed analysis prior to the early 1840s so it is not possible to say whether the trend observed during the first of our reference periods is a continuation of, or a departure from, what pertained previously. Fig. 1 shows the annual rates of murder and manslaughter between 1845 and 1854. The same scale is used in each of the four charts presented in this chapter so that the differences between the intervals in question can be directly compared. Finnane (1997) found that between 1860 and 1914, the number of homicides in Ireland declined steadily. Over the same period there was a noticeable fall in prosecutions for assault, including assaults against the police. This was offered as evidence of a society that was becoming increasingly peaceable. Wilbanks (1996) extended the analysis to the early 1990s and identified the same broad historical trend. He contrasted the falling homicide rate in Ireland with the reputation for violence that Irish emigrants gained in the United States. While the trajectory may have been downward it was still higher than in Britain. Connolly (1999, p. 205) observed that: ‘‘. . . the common perception of Ireland as a more violent society than other parts of the United Kingdom was firmly based in fact. From the 1860s to the 1890s the murder rate recorded in the Irish judicial statistics was about twice that suggested by their English and Welsh equivalent.’’ Ireland’s status as a quarrelsome colony could not simply be ascribed to its endemic agrarian disturbances and periodic political upheavals, most of which resulted in little loss of life.

40 35

33.3 31.0

30

27.9

26.7

24.0

25 20

21.7

23.2 19.2

17.4

16.2

15 10 5 0 1845

1846

1847

1848

1849

1850

1851

1852

1853

1854

Fig. 1 Famine, 1845–1854 (rate of murder and manslaughter per million population)

82

I. O’Donnell

Previous attempts to probe homicide levels in nineteenth-century Ireland have utilised a variety of data sources. Finnane relied primarily on crimes known to the police as shown in the judicial statistics. Wilbanks used a wider variety of sources but for the early period depended on the Outrage reports collated by the Irish Constabulary. These excluded Dublin. Also there is a break in his data series between 1914 and 1947. Connolly drew on a combination of Outrage reports and judicial statistics. I have incorporated all of these sources, and others, into the analyses reported here (for methodological details see O’Donnell, 2005b, pp. 674–675). The Great Famine that occurred between 1845 and 1850, as a result of potato blight, caused widespread devastation. This was not the only famine to have struck the Irish peasantry – crop failures are a regular hazard for subsistence farmers – but its consequences were the most savage. According to Lyons (1973, p. 44) around one million men, women and children perished from hunger and disease over this period and a similar number emigrated. This was the beginning of a perennial haemorrhaging of young men and women. The outflow of people after the Famine was extraordinary in its scale and duration. Guinnane (1997, p. 104) found that by 1881 nearly 40 per cent of Irish-born people were not living in Ireland. He estimated that more than four million people left Ireland for good between the Famine and the First World War (ibid, p. 101). The wandering Irish had an impact on the crime rates in their host communities, especially in England, where they were over-represented among prosecutions and committals to prison (Swift, 1997). In addition to its obvious impact on the size of the national population the Famine also had major consequences for: Sexual behaviour: Family structure:

The Irish language:

Agriculture:

Mental health:

Patterns of recreation:

the emergence of a culture of lifelong celibacy, fewer and later marriages (e.g. Guinnane, 1997). less subdivision of the land and a shift towards the stem family system where one son is chosen to inherit the entire farm (e.g. McCullagh, 1991). tipped into an arguably terminal decline by the perception that fluency in any tongue other than English was a handicap for those whose destiny was emigration (e.g. Hughes, 2001). greater crop diversity and a switch from labour-intensive tillage to land-intensive pasture (e.g. O’Rourke, 1994). the number of patients in public asylums, which were sometimes viewed as places to deposit ‘surplus’ children, increased dramatically (e.g. Finnane, 1981). single men with few prospects found an outlet by engaging in aggressive team sports, faction fighting and other forms of ‘recreational violence’ (e.g. Conley, 1999).

The Fall and Rise of Homicide in Ireland

83

Most of these developments persisted well into the twentieth century and shaped the social organisation of independent Ireland. There was an obvious spike in lethal violence during the mid-1840s. The highest known rate – 33.3 killings per million population – was recorded in 1847, when the Famine was heading towards its catastrophic peak. It fell to half of this level by 1854. The impact on the crime rate of the distress experienced by large swathes of the population did not go unrecognised at the time. In a paper read before the Section of Economic Science and Statistics of the British Association at Dublin on 29 August 1857 one commentator observed: ‘‘That want and privation are fruitful multipliers of all classes of crime; and that to such a parentage we may safely look first, when we find a general and marked increase in the crime of a country’’ (Wilson, 1857, p. 104).

Consolidation (1895 to 1904) The closing years of the nineteenth century and the opening ones of the twentieth were quiet times in Ireland. The national question appeared to nearing resolution with Home Rule an increasingly viable prospect. There was less agitation over land as tenants had been enabled to buy out their (often absentee) landlords. For example the Wyndham Act of 1903 provided loans to tenants at reduced interest for the purchase of land, and gave bonuses to landlords who sold. This added impetus to previous government-assisted schemes and, by the early twentieth century, most of the national territory was owned by former tenants. The ravages of the Famine had long passed and a Catholic middle class was growing in number and in confidence. (The National University of Ireland was established in 1908 to cater for the young Catholic men who aspired to the professions.) Although appalling slums existed in Dublin, the agricultural labouring class, which had barely subsisted throughout the nineteenth century, was in terminal decline. In Ferriter’s words (2004, p. 64) ‘‘. . . the long-term effect of the Famine was a rural Ireland that was more sparsely populated but better off.’’ The second half of the nineteenth century saw the emergence of the small and comfortable tenant farmer. The social reorganisation of post-Famine society meant the virtual elimination of the young rural men with few attachments, and fewer prospects, who found diversion in fighting (Fitzpatrick, 1980). The decimation of the country’s population through disease, emigration and celibacy made society safer for those who remained. Anyone who craved more excitement could always seek it through voluntary exile. In 1895 the homicide rate was slightly higher than in 1854 but over the following ten years it fell by one third (see Fig. 2). In the late nineteenth century some of the recorded cases of manslaughter involved death by careless driving. It was not possible

84

I. O’Donnell 40 35 30 25 20 15 10

18.6

19.7

19.3

20.4 14.6

16.1 10.8

12.6

13.7

12.6

5 0 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904

Fig. 2 Consolidation, 1895–1904 (rate of murder and manslaughter per million population)

consistently to disaggregate these from other cases of manslaughter so, if anything, the decline shown in lethal interpersonal violence is understated. Finnane (1997) found an inverse relationship between interpersonal violence and violence against the self. In other words as homicide went down, suicide went up. A similar pattern had been identified by Durkheim (1897/1952, p. 351) in his classic study of suicide. Observing the state of affairs in the late 1880s he remarked that: ‘‘Spain, Ireland and Italy are the three countries of Europe where there is least suicide . . . Inversely, nowhere else is murder so common. These are the only countries where the number of murders exceeds that of voluntary deaths.’’ Not only was little killing recorded at this time, but it seemed that the citizenry was generally inclined to be law abiding. According to the report of the Committee of Inquiry into the Dublin Metropolitan Police in 1902: ‘‘In Dublin there is practically no serious crime’’ (cited in O’Brien, 1982, p. 179). In 1899 for example the Dublin Metropolitan Police recorded two murders, two rapes and 23 assaults out of a total of 2,466 indictable crimes. Two years later the respective numbers were two, six, 38 and 2,696. O’Brien (1982, p. 183) noted that further evidence for the low level of crime was to be found in the steep decline in the number of convicts in custody (from over 3,400 doing penal servitude for at least three years in 1855 to 292 in 1901). There was also a fall in the number of minor offenders sent to local prisons (from 48,000 in 1855 to 32,000 in 1900). Nonetheless the rate of recorded crime in Dublin was far higher than elsewhere in the country and exceeded the rates for English cities (with only Manchester and Liverpool – key locations for the Irish diaspora – coming close).

The Fall and Rise of Homicide in Ireland

85

Stasis (1945 to 1954) By 1945 state formation in Ireland was advanced and the political and administrative structures inherited from Britain had been adapted to meet the prevailing national priorities. The stability of the new arrangements – a new Constitution had been accepted by the people in 1937 and the same government was in power between 1932 and 1948 – is sure to have had an impact on crime. It is not simply that the presence of a state reduces violence among those subject to its jurisdiction. Cooney (1997) argued that lethal conflict, whether it takes the form of war, rebellion, homicide or execution, tends to be high when state authority is absent and when it is strong or heavily centralized. Between these extremes it typically declines. The pattern observed in Ireland conforms to this idea. Before 1922 the country’s colonial status was the cause of successive waves of agitation in favour of national self-determination. It was also a time when non-political interpersonal violence was common. The unavailability, or perceived irrelevance, of legal remedies created a context where people took direct action in order to resolve disputes. After the establishment of the Irish Free State in 1922, homicide dwindled away and by the time a Republic was declared in 1949, it had dropped dramatically. A culture of late marriage and permanent celibacy persisted up to the 1950s, by which time lethal violence had fallen to unprecedented levels. As Fig. 3 shows, in 1945 the rate of homicide was less than one third of what it had been in 1904. It fluctuated around a very low baseline for the next ten years and reached an historic low of 1.02 per million in 1954, when only three killings were recorded in the entire country. (In addition there was one case of infanticide and

40 35 30 25 20 15 10 5.7 5

3.7

5.4 3.4

3.0

4.7

3.7 1.7

2.0

1951

1952

1.0

0 1945

1946

1947

1948

1949

1950

1953

1954

Fig. 3 Stasis, 1945–1954 (rate of murder and manslaughter per million population)

86

I. O’Donnell

one murder of a child under one year.) Between 1945 and 1954 the rate of lethal violence was just one seventh of what it had been between 1845 and 1854 (see Table 1). This could be described as a precipitous decline. In the early 1950s, Ireland had a predominantly rural population numbering less than three million, enjoyed a low rate of recorded crime, and all of its prisoners (a total of 449 in 1954) could have been accommodated in a single institution of modest size. The quiet declaration of a Republic in 1949 meant that the residual traces of British sovereignty were swept away with little fuss; the functions previously accorded to the king were given instead to the President of Ireland and the country ceased to be a member of the Commonwealth. The country was admitted to the United Nations in 1955. Roman Catholicism was in the ascendant and formal religious observance was almost universal. University education was the preserve of the privileged few and most people had a close connection with the land. It was a country which E´amon De Valera, Taoiseach [prime minister] for most of the years between 1932 and 1959, famously described as aspiring to be: ‘‘. . . the home of a people who valued material wealth only as the basis of a right living, of a people who were satisfied with frugal comfort and devoted their leisure to the things of the soul.’’ But this was not an entirely harmonious society. Economic growth had stagnated, unemployment was high and the country was haemorrhaging its citizens at an alarming rate through emigration (for perspectives on Ireland in the 1950s, see Keogh, O’Shea & Quinlan, 2004). The average annual number of emigrants was 39,350 between 1951 and 1956, a rate of 13.4 per 1,000 population (Central Statistics Office, 2002, p. 11). It is a striking feature of Irish demography that the national population fell for more than a century after the famine of the 1840s. Archer and Gartner (1976) compared the average homicide rate in countries during the five years before a war with the five years afterwards. They contrasted nations that had been directly involved in conflict with others that had not and concluded that combatant nations experienced substantial post-war increases in their homicide rates, whether victorious or defeated. These increases tended to be most frequent among nations with large numbers of combat deaths. A post-war peak would not have been expected in Ireland given the country’s decision to remain neutral during the Second World War (known locally as ‘The Emergency’). Nevertheless large numbers of Irish men joined the British army and an estimated 4,500 lost their lives (McEwen, 2004). Looking specifically at cases of murder involving victims aged over one year, there were 46 such killings recorded during the seven-year period 1939 to 1945 compared with 54 in the same period before the war, and 33 in the same period afterwards. Clearly the country’s marginal involvement in its nearest neighbour’s military operations did not deflect the homicide rate from its downward slide.

The Fall and Rise of Homicide in Ireland

87

Prosperity (1995 to 2004) Contrast the picture of a depopulated, economically stagnant and inwardlooking agricultural society of the 1950s with the heterogeneous country that was Ireland at the end of the twentieth century. A rapid rise in prosperity, growing secularisation and net inward migration (at an annual average of 25,500 between 1996 and 2002) created an increasingly self-confident society in flux. The old certainties and the deeply rooted deference for traditional forms of authority came under profound and sustained attack. By the end of the century, the venerable institutions of church and state were severely bruised and had lost much of the unbending control and unquestioning obedience that had for so long characterised their operations. The reasons for this dramatic change are complex and beyond the scope of this chapter, but the ‘Celtic Tiger’, the shorthand term utilised to describe the remarkable economic prosperity enjoyed in Ireland since the mid-1990s, was accompanied by profound social and cultural dislocation. In the early 1990s Ireland was derided as the ‘basket case’ of Europe. By the beginning of the twenty-first century, it was one of the most globalised countries in the world and held out as a model for other small open economies (see Fahey, Russell & Whelan, 2007; and Nolan, O’Connell & Whelan, 2000 for a collection of views on Ireland’s distinctive experience of growth and inequality). Contrary to the national stereotype, Ireland had a comparatively low level of alcohol consumption for most of the twentieth century. The Pioneer Total Abstinence Association of the Sacred Heart was founded in 1898 to combat what was perceived as a scandalous national tendency to imbibe to excess and became a major force in social life. At its peak in the 1940s and 1950s, it could attract crowds of up to 90,000 to its rallies in Dublin. The temperance movement was inextricably linked with the growing power of the Roman Catholic church and the Association’s goal was to persuade as many Irish Catholics as possible to take a pledge known as the ‘Heroic Offering’ to abstain from alcohol for life. Those who took the pledge often wore a pin depicting the Sacred Heart of Jesus to which their sacrifice was offered. This was seen as a mark of piety. By the end of the 1950s, the Association claimed a membership of 500,000 registered in 1,900 centres around the country. The low level of alcohol consumption in Ireland for much of the twentieth century may have played a role in suppressing violent crime (more generally see Kilcommins & O’Donnell, 2003). This relative abstemiousness changed suddenly. Between 1989 and 1999 the rise in per capita alcohol consumption in Ireland was 41 per cent, by far the highest in the European Union. Over the same period 10 countries showed a decrease, and three a modest increase. According to World Health Organisation figures, by 2000, there were only 11 countries with higher average consumption than Ireland (Strategic Task Force on Alcohol, 2002, p. 16). The change since 1995 was described by the Strategic Task Force on Alcohol

88

I. O’Donnell

as nothing short of a ‘‘dramatic increase in consumption’’ (ibid, p. 5). The habits of Irish drinkers are also of interest, and binge drinking in public houses is common. This has predictable consequences in terms of street disorder. It is possible that the sudden rises in drinking and homicide are coincidental, or that both have a common cause or are explained by different factors. However good reasons exist for thinking that they are related. In around half of the homicide cases examined by Dooley (1995, 2001) either victim, perpetrator or both were intoxicated. The increase in alcohol consumption also has consequences for public order. For example in 2004, criminal proceedings were taken in 21,000 cases of public intoxication. It is plausible therefore to expect that a rise in alcohol consumption would be accompanied by more violence (both lethal and non-lethal). There is support in the wider literature for the proposition that alcohol consumption should be included as a variable in any study of homicide rates (Parker, 1995). Figure 4 shows the annual rate of murder and manslaughter between 1995 and 2004. The suicide rate rose in parallel with the increase in homicide. In 2003 there were 497 suicides recorded, compared with 334 in 1990. Clearly, the inverse relationship described by Durkheim did not hold good a century after he identified it. Another significant factor in Irish homicide as it was manifested around the end of the twentieth century was the emergence of what the police have come to describe as ‘gangland’ killings. These often revolve around drug disputes and invariably involve either abduction and murder (possibly preceded by torture), a shot to the head in a public place such as a bar, or an execution-style hit at the

40 35 30 25 20 15

14.2

14.5 12.7

13.8

14.8

15.1

15.1 13.3

12.6

11.1

10 5 0 1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Fig. 4 Prosperity, 1995–2004 (rate of murder and manslaughter per million population)

The Fall and Rise of Homicide in Ireland

89

victim’s home. When he examined the twenty-year period 1972 to 1991 Dooley (1995, p. 16) found only two incidents of homicide (out of 610) that could be attributed to gangland feuding. By the mid-1990s this had changed considerably and there were eight such incidents in 1996 alone (Dooley, 2001, p. 17). In 2003 more than one in three recorded homicides (20 out of 52) were attributed to gangland feuds. This proportion remained unchanged in 2005 (‘‘World of crime’’, 2005). Ireland joined the European Community in 1973 and for the first time in modern history the country experienced net inward migration. This was largely the return of Irish migrants with their families to meet specific skills shortages in the economy. In the 1980s there was a return to what had become the established pattern of dramatic increases in unemployment and emigration. In the two-year period 1988 to 1989 approximately two per cent of the entire population left the country. This was dramatically reversed in the mid-1990s. Ireland’s population grew by 440,000 (12 per cent) between 1995 and 2004. This was an unprecedented level of expansion that greatly exceeded the European average. The census for 2002 showed that the national population was higher than at any time since 1871. So too had the homicide rate returned to the levels of a century previously.

Conclusion To appreciate the magnitude of the shifts in lethal violence that took place across the four time periods covered by this chapter consider the following raw numbers: l

l

Between 1845 and 1854 there were more than 1,600 homicides (excluding infanticide and the murder of children aged under one year) recorded on the island of Ireland. Between 1945 and 1954 there were fewer than 160 such killings on the island (including both the Republic of Ireland and Northern Ireland).

Even taking account of the substantial decline in the island’s population (from 8.2 million in 1841 to 4.3 million in 1941) this is a remarkable transformation. The difference is not entirely explained by the devastation and panic caused by the Famine, when the level of unlawful killing was at a peak. As we have seen the Famine had an enduring legacy in terms of population decline and how Irish men and women organised their lives. Those who did not stand to inherit or marry faced the prospect of spending their hard earned and meagre savings on a one-way ticket for the emigrant boat. This was a crucial safety valve. But there were other factors at play. A lengthy temperance campaign began to bear fruit in the early twentieth century, the country finally achieved its independence from Britain, and the structures of a modern state emerged. By

90

I. O’Donnell

the 1950s Ireland was a country almost totally insulated from lethal violence. Striving (and failing) to achieve economic self-sufficiency it was a repressively homogeneous place. Half a century later there had been a sea change and the country was coming to terms with rapid prosperity, full employment, soaring alcohol use and a concomitant increase in violent deaths. Two of the periods under examination were times of major population exodus (1845 to 1954 and 1945 to 1954). The others (1895 to 1904 and 1995 to 2004) were times of growing national self-confidence and economic progress. A common theme across all of the periods in question, with the exception of the most recent, is of Ireland as an exporter of potential criminals. While demographic factors are part of the answer the theoretical significance of the recent upswing in lethal violence needs to be disentangled more fully. Mennell (1990, p. 211) has suggested that the rise of public violence might be seen as an indication of a decivilising tendency, but cautions against rushing to premature conclusions with the advice that, ‘‘. . . evidence over at least three generations is necessary to diagnose a decivilising process.’’ More work is required to tease out the key explanatory factors and to establish how they overlap and interrelate. Acknowledging the power of the civilising thesis, Mares (2004) has begun to formulate a theory that can account for both the long-term decline in violence and the periodic reversals that occur within it. His analysis shows that at certain time periods upward trends occurred simultaneously in a variety of regions, suggesting that these events could have been as systemic in nature as the secular decline they interrupted. His focus is on the combined effect, and relative dominance, of state formation and economic growth cycles. Mares argues that in periods when the growth of state institutions was ascendant there was a reduction in levels of violence. However, violence rose when economic processes were given a higher priority than the development of state institutions. Looking specifically at the Irish situation there is some support for this perspective. It seems that when economic factors were dominant, lethal violence was relatively high. This was true during the Famine when the collapse of subsistence agriculture led to widespread destitution and crime, as well as at the end of the twentieth century when there was a sudden, and pronounced, surge in living standards. When the processes of state formation were more significant violence declined. This was particularly pronounced around the time when the country was declared a republic and took full and final responsibility for its own internal and external affairs. It would be instructive to use the Irish experience as a test case for Mares’ theory about the interplay between state formation and the market economy. It might also be fruitful to examine whether the trends for other types of crime mirror those found for homicide. These areas of exploration would likely yield a rich reward in terms of theory testing and refinement. It is to be hoped that these lines of inquiry will be taken up by scholars in the future.

The Fall and Rise of Homicide in Ireland

91

References Archer, D., & Gartner, R. (1976). Violent acts and violent times: A comparative approach to postwar homicide rates. American Sociological Review, 41, 937–63. Central Statistics Office (2002). Census 2002: Preliminary report. Dublin: Central Statistics Office. Conley, C. (1999). The agreeable recreation of fighting. Journal of Social History, 33, 57–72. Connolly, S.J. (1999). Unnatural death in four nations: Contrasts and comparisons. In S.J. Connolly (Ed.), Kingdoms United? Great Britain and Ireland since 1500: Integration and Diversity (pp. 200–214). Dublin: Four Courts Press. Cooney, M. (1997). From warre to tyranny: Lethal conflict and the state. American Sociological Review, 62, 316–38. Dooley, E. (1995). Homicide in Ireland 1972–1991. Dublin: Stationery Office. Dooley, E. (2001). Homicide in Ireland 1992–1996. Dublin: Stationery Office. Durkheim, E. (1897/1952). Suicide: A study in sociology. London: Routledge & Kegan Paul. Eisner, E. (2001). Modernization, self-control and lethal violence: The long-term dynamics of European homicide rates in theoretical perspective. British Journal of Criminology, 41, 618–38. Eisner, M. (2003). Long-term historical trends in violent crime. In M. Tonry (Ed.), Crime and Justice: A Review of Research (pp. 83–142). Vol. 30. University of Chicago Press. Fahey, T., Russell, H., & Whelan, C.T. (2007). (Eds.) Best of times? The social impact of the celtic tiger. Dublin: Institute of Public Administration. Ferriter, D. (2004). The transformation of Ireland, 1900 – 2000. London: Profile. Finnane, M. (1981). Insanity and the insane in post-famine Ireland. London: Croom Helm. Finnane, M. (1997). A decline in violence in Ireland? Crime, policing and social relations 1860–1914. Crime, History and Society, 1, 51–70. Fitzpatrick, D. (1980). The disappearance of the Irish agricultural labourer, 1841–1912. Irish Economic and Social History, 7, 66–92. Griffin, B. (2005). Sources for the study of crime in Ireland, 1801 – 1921. Dublin: Four Courts Press. Guinnane, T.W. (1997). The vanishing Irish: Households, migration and the rural economy in Ireland, 1850 – 1914. Princeton, NJ: Princeton University Press. Henry, B. (1994). Dublin hanged: Crime, law enforcement and punishment in late eighteenthcentury Dublin. Dublin: Irish Academic Press. Hughes, A.J. (2001). Advancing the language: Irish in the twenty-first century. New Hibernia Review, 5, 101–126. Kelly, J. (1995). ‘‘A most inhuman and barbarous piece of villainy’’: An exploration of the crime of rape in eighteenth-century Ireland. Eighteenth Century Ireland, 10, 78–107. Keogh, D., O’Shea, F., & Quinlan, C. (2004). Ireland in the 1950s: The lost decade. Cork: Mercier Press. Kilcommins, S., & O’Donnell, I. (2003). (Eds.) Alcohol, society and law. Chichester: Barry Rose Law Publishers. Lyons, F.S.L. (1973). Ireland since the famine. London: Fontana. Mares, D.M. (2004). Civilization, economic change and trends in interpersonal violence. (Doctoral dissertation, University of Missouri – St. Louis, 2004). McCullagh, C. (1991). A tie that binds: Family and ideology in Ireland. Economic and Social Review, 22, 199–211. McEwen, Y. (2004). Deaths in Irish regiments 1939–1945 and the extent of Irish volunteering for the British Army. The Irish Sword, 24/95, 81–98. Mennell, S. (1990). Decivilising processes: Theoretical significance and some lines of research. International Sociology, 5, 205–223. Morris, R.M. (2001). ‘‘Lies, damned lies and criminal statistics’’: Reinterpreting the criminal statistics in England and Wales. Crime, History and Society, 5, 111–27.

92

I. O’Donnell

Nolan, B., O’Connell, P.J., & Whelan, C.T. (2000). (Eds.) Bust to boom? The Irish experience of growth and inequality. Dublin: Institute of Public Administration. O’Brien, J.V. (1982). ‘‘Dear, Dirty Dublin’’: A city in distress: 1899–1916. Berkeley: University of California Press. O’Donnell, I. (2005a). Violence and social change in the Republic of Ireland. International Journal of the Sociology of Law, 33, 101–117. O’Donnell, I. (2005b). Lethal violence in Ireland, 1841 to 2003: Famine, celibacy and parental pacification. British Journal of Criminology, 45, 671–695. O’Donnell, I., O’Sullivan, E., & Healy, D. (2005). Crime and punishment in Ireland 1922–2003: A statistical sourcebook. Dublin: Institute of Public Administration. O’Rourke, K. (1994). The economic impact of the Famine in the short and long run. The American Economic Review, 84, 309–313. Parker, R.N. (1995). Bringing ‘‘booze’’ back in: The relationship between alcohol and homicide. Journal of Research in Crime and Delinquency, 32, 3–38. Rose, L. (1986). The massacre of the innocents: Infanticide in Britain 1800–1939. London: Routledge and Kegan Paul. Sauer, R. (1978). Infanticide and abortion in nineteenth-century Britain. Population Studies, 32, 81–93. Spierenburg, P. (2001). Violence and the civilizing process: Does it work? Crime, History and Society, 5, 87–105. Strategic Task Force on Alcohol (2002). Interim report. Dublin: Department of Health and Children. Swift, R. (1997). Heroes or villains?: The Irish, crime and disorder in Victorian England. Albion, 29, 399–421. Taylor, H. (1998). Rationing crime: The political economy of criminal statistics since the 1850s. Economic History Review, LI, 569–90. Thome, H. (2001). Explaining long-term trends in violent crime. Crime, History and Society, 5, 69–86. Wilbanks, W. (1996). Homicide in Ireland. International Journal of Comparative and Applied Criminal Justice, 20, 59–75. Wilson, J.M. (1857). Statistics of crime in Ireland – 1842 to 1856. Journal of the Dublin Statistical Society, Vol. II, Part X, 91–122. World of crime: Signs of worse to come. (2005, December 28). The Irish Times, p. 3.

Part III

Contemporary Trends

Violence in Present-Day France: Data and Sociological Analysis Philippe Robert

Violence – that is, interpersonal violence – now pervades public debate in present-day France. For politicians, essayists and journalists its importance is beyond any doubt. Very few of them have taken the trouble – or are armed – to give a concrete content of that terrifying term, however.

Low Intensity Violence To prove that French society is plagued by violence, a single source is cited, as a rule: that is, statistics published by the Ministry of the Interior on the number of reports transmitted to the public prosecutor by the Police and Gendarmerie pertaining to acts possibly constituting more or less serious offenses. Over the last half-century, interpersonal violence has increased fourfold, rising from less than 15 to over 55 acts per 10,000 inhabitants (see Fig. 1). To put this fact in perspective, we would point out that property offenses (thefts, burglaries and so on) were multiplied by 9, and their magnitude is of a completely different order. They rose from slightly under 45 to close to 400 per ten thousand inhabitants. The difference is that this increase took place between the late 1950s and the mid 1980s, and the figures have now leveled off somewhat, although at a very high level. Interpersonal violence, on the other hand, did not begin to rise significantly until the mid 1980s (see Fig. 2), which probably explains why it is attracting so much attention. But we need to take a closer look at what those violent offenses actually include.

Philippe Robert Director of Research at the CNRS, France; CESDIP, Immeuble Edison, 43 boulevard Vauban, 78280 Guyancourt, France [email protected] Translated by Helen Arnold; revised by Rene´e Zauberman.

S. Body-Gendrot, P. Spierenburg (ed.), Violence in Europe. Ó Springer 2008

95

96

P. Robert rates for 10,000 60

50 40 30 20 10

20 02

19 98

19 94

19 90

19 86

19 82

19 78

19 74

19 70

19 66

19 62

19 58

19 54

19 50

0

Fig. 1 Trend of the rates of interpersonal violent offenses – cases recorded by the police and gendarmerie (1950–2004)

rates for 10,000 500

400

300

200

100

0 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 property crime

violent crime

Fig. 2 Trend of the rates of interpersonal violence and property offenses – cases recorded by the police and gendarmerie (1950–2004) Source: Robert, 2002, updated

Violence in Present-Day France

97

Fatal Violence is at the Low-Water Mark When people talk about violence they usually think: homicide. Police data are difficult to use for observing long-term trends, since they did not start distinguishing between attempted and completed homicide until 1988. However, since the proportion of attempts is seen to be practically constant thereafter, we may assume that the same was true for the earlier period. This produces a curve showing a very low level at the beginning of the 1970s,1 rising until the middle of the following decade, then declining until it returns to its original level at the end of the period (Fig. 3). To interpret these figures, a term of comparison is needed. Conventionally, one then turns to statistics on causes of death, established by the health authorities.2 1 900 1 800 1 700 1 600 1 500 1 400 1 300 1 200 1 100 1 000 72 974 976 978 980 982 984 986 988 990 992 994 996 998 000 002 1 1 1 1 1 1 1 1 1 1 2 2 19 1 1 1

Fig. 3 Completed homicides and fatal assault (Police statistics 1972–2003) Source: Robert, 2002 updated Note: Homicide attempts were not a separate category in police statistics until 1988. Since their ratio is practically constant, we have used it to estimate the number of successful homicides between 1972 and 1987.

1

Police statistics have only been published since 1972 in France; however, there are some (rather cursory) unpublished figures since 1950. 2 They are in existence since 1906. Until 1921 they were established by the Public Assistance and Sanitation services of the Ministry of the Interior, until 1943 by the General Statistics of France, until 1967 by the National Statistics Service, later to become the National Institute of Statistics and Economic Surveys (INSEE), and last, since 1968 by the National Institute of Health and Medical Research (INSERM). It differentiates between homicides and accidental deaths since 1930.

98

P. Robert 7 000

nombre absolu d'homicides

6 000 5 000 4 000 3 000 2 000 1 000

05

00

20

20

95

90

19

85

19

80

19

75

19

70

19

65

19

19

55

60 19

50

19

45

19

35

40

19

19

19

19

30

0

Fig. 4 Medically recorded violent deaths (1930–2001) Source: INSERM

They have an advantage over police statistics in that they do not lump together attempts and completed homicides, and they can be observed over a long period of time (Fig. 4). They show a peak during the second world war and another much lower one around 1960, corresponding to the disorders connected with the termination of the Algerian war. After that there is a gradual, moderate rise between 1963 and 1984, followed by a gradual decline, bringing the end-ofcentury level back to that of the early 1970s. Police and health sources converge as to the trend, then, thus justifying the conclusion that public alarm over violence over the last twenty years is not caused by an increase in homicides. Not only did France not experience the same leap in homicide as the United States during the 70s and 80s, but it is still way below the rates presently observed in that country, despite the drop in homicides there during the last decade.

Fragmented Violence When police statistics show a rise in violent offenses (Fig. 1), what is involved is primarily non-lethal violence. There is another difficulty with this indicator, however. Its main heading is intentional assault and battery, which in itself represents approximately two fifths of all violent offenses. The legislation has been modified repeatedly since the early 1980s – and increasingly obsessively in recent years – to extend the perimeter of this offense

Violence in Present-Day France

99

so as to include cases which were previously excluded because less serious, and which therefore were not counted in police statistics (box 1).3 Until 1981, violence had to have caused an incapacity to work exceeding eight days in order to come under the heading of the offense of intentional assault and battery. In 1981, a legislative modification authorized the inclusion of violence having entailed a lesser incapacity to work provided it had been committed with premeditation or entrapment, use of or threat by a weapon, on a vulnerable person, one’s ascendants, a lawyer, a public officer or public authority, an agent of the police forces or a citizen in charge of a public service mission, or on a witness or a victim to prevent them from cooperating with the criminal justice system. These innovations were not eliminated when the law was modified on June 10, 1983. In 1992, the new criminal code was to add minors under age 15 as well as judges and jury members, civil parties and their spouse to the victim list, and agents holding public authority and acts perpetrated in a group to the offenders list, whereas entrapment was eliminated. An Act dated July 22, 1996 made violence entailing no incapacity to work an offense of intentional assault and battery in the presence of one of the aggravating circumstances provided for in the previous laws, and called for heavier sentences in case several aggravating circumstances exist. The June 17, 1998 act adds acts committed in or in the immediate vicinity of a school. The June 18, 1999 Act specifies the inclusion of police officers and adds employees of a public transportation network as protected victins. The September 9, 2002 Act includes violence committed by an adult with the help of a juvenile. Another Act, dated February 3, 2003 designates violence with a racist or xenophobic connotation. The most recent Act, dated March 18, 2003, further lengthens the long list of protected victims, adding the person’s family and intimates, as well as all individuals in charge of any public service mission, health professionals and victims of homophobic violence, as well as victims of violence perpetrated in public transportation vehicles and stations. Independently of any real increase in violence, these legislative changes would have sufficed to induce the perception of a rise in the number of cases of intentional assault and battery recorded by police statistics. By the same token the main indicator of physical violence produced by the police lost its

3

In France, police statistics do not include petty offenses known as contraventions, but only moderately serious and serious offenses – misdemeanors (de´lits) and felonies (crimes).

100

P. Robert

reliability precisely during the most problematic period. The object it measures is not the same at different times. One necessarily wonders why producers or publishers of these statistics never mention this problem4 all the more so since several scholarly publications have pointed out its existence. Be that as it may, we must turn to another source. Data are available, firstly, from a first national victimisation survey5 conducted by CESDIP (Centre de Recherches Sociologiques sur le Droit et les Institutions Pe´nales) for the years 1984–85, and thereafter from ongoing annual surveys by the INSEE (Institut National de la Statistique et des E´tudes E´conomiques) from the mid-1990s on. Use of these data for the study of the trend of violence in France raises three methodological problems, however. In the last analysis,6 we arrive at the following conclusions (box 2). As long as one counts victims (prevalence), the main difficulty in serialization of data is the relatively small size of the samples used.7 For an event as rare as interpersonal violence – especially serious violence – the outcome is relatively large confidence intervals, so that it is not always easy to detect a trend when the changes are not very marked.8 When one wants to count acts (incidence) as well, another difficulty arises. All of these surveys simply ask the respondent to describe the most recent incident for each kind of victimisation. People are also asked how many times they had been subjected to the same sort of victimisation

4 Not only has the headquarters of the National Police Department not pointed out this extension of the perimeter of index 7 (de´lits and crimes of other intentional assault and battery) in its statistics, but further, this black-out has not been modified by the creation of a Observatoire national de la de´linquance, which goes to show that an agency of the sort is only worthwhile if it is really independent of the different administrative bodies and corporations controling crime, and if it applies truly scientific scholarship to the matter. 5 This first survey (see Zauberman, Robert, Perez-Diaz, Le´vy, 1990; also Zauberman, 1991; Zauberman, Robert, 1995), quite shaky owing to the modest funds on which it was run, nonetheless constitutes a most important marker. It is the only general population survey on the trend in violence in France available at the outset of this crucial period. 6 Following a comparison between the CESDIP survey of the mid-1980s and the first INSEE survey of the mid-1990s (Robert, Zauberman, Pottier, Lagrange, 1999, 2001), a serialization was published in 2004 (Lagrange, Pottier, Zauberman, Robert, 2004) covering the national surveys up to 2001. Serialization of the three most recent INSEE surveys is under way and has not yet been published. 7 The INSEE questions some 6,000 households a year, with one person within the household during the early years, and up to three people thereafter. 8 Consequently, the most urgently needed reform for these national surveys would be to considerably increase the size of the samples covered.

Violence in Present-Day France

101

during the two years preceding the survey (Fig. 5). However, instead of asking for the actual number of prior attacks experienced during the reference period, the INSEE surveys itemize the question in terms of once, twice, three times or more, making it difficult to calculate a multivictimisation indicator. Exceptionally, however, the 1999 survey, pertaining to the years 1997–98, gives the exact number of attacks experienced by each respondent. The ’real’ number thus observed allowed us to validate the hypothesis of 3 or more incidents = 4 and to retain this computation for all the available surveys. The last difficulty arises when trying to distinguish the degree of seriousness of all of these attacks, namely, serious physical violence having entailed an incapacity to work exceeding eight days, simple physical violence (assault, injury without an incapacity to work exceeding eight days) and last, attacks with no physical harm (threats, insults, threatening attitudes, rackets, bag-snatching having caused no injury and so on). There is indeed no way of knowing whether the violent incidents experienced previously by the respondent during the reference period were of the same degree of seriousness as the most recent one, the only one described in detail. If we turn back to the same 1999 survey, in which the respondent was asked to describe each incident, we come to the conclusion that multi-victimisation involving serious physical violence is overestimated by 25%

180 000 160 000 140 000 120 000 100 000 80 000 60 000 40 000 20 000

19

19

84

-8

5

94 19 95 95 19 96 96 19 97 97 19 98 98 19 -9 99 9 -2 0 20 00 00 -0 1

0

prevalence

incidence

Fig. 5 Trends of the prevalence (number of victims) and incidence (number of victimisation incidents) for serious physical violence (1984–2001) based on the CESDIP and INSEE victimisation surveys

102

P. Robert

if we assume that each victim of this type of violence had systematically undergone equally serious attacks. This corrective coefficient may then be applied to every survey.9 The number of victims of serious physical violence – work incapacitation exceeding one week or equivalent physical damage – rose between the mid-80s and the mid-90s, after which it stagnated, whereas the upward curve for victimisation incidents never falters (Fig. 5). Furthermore, victims seem to have filed suit increasingly systematically, a fact which increases the chances of having such violence mentioned in police statistics, all else being equal. For less serious assault, these surveys indicate that victims and victimisations have increased even more (Fig. 6). This is particularly true for violence with no physical contact, such as threats and insults, which have literally soared in surveys for recent years (Fig. 7). The chances of having this reflected in police statistics are low, firstly because these victims are increasingly less prone to report (especially for verbal violence), and secondly because the type of offense would not be serious enough to be counted by the police. According to the surveys, the most salient feature of recent years is undeniably the soaring number of low-intensity violence. 3 000 000

2 500 000

2 000 000

1 500 000

1 000 000

prevalence

incidence

19 94 19 95 95 19 96 96 19 97 97 19 98 98 19 -9 99 9 -2 20 000 00 -0 1

19 84

-8 5

500 000

Fig. 6 Trends of the prevalence (number of victims) and incidence (number of victimisation incidents) for other physical violence (1984–2002) based on the CESDIP and INSEE victimisation surveys

9

A series of still unpublished local surveys questioned respondents in detail on the seriousness of every violent attack experienced during the reference period. Their findings corroborate the conclusions drawn from the national.

Violence in Present-Day France

103

5 000 000 4 500 000 4 000 000 3 500 000 3 000 000 2 500 000 2 000 000 1 500 000 1 000 000

prevalence

incidence

19 94 19 -95 95 19 -96 96 19 -97 9 19 7-9 99 8 -2 20 000 00 20 -01 00 -0 2

19

84

-8

5

500 000

Fig. 7 Trends of the prevalence (number of victims) and incidence (number of victimisation incidents) for verbal violence (1984–2002) based on CESDIP and INSEE victimisation surveys

These incidents should not be underestimated. The inconspicuousness of incidents, taken separately, is more than compensated for by their repetition, which may make life unbearable for the victims. Suppose that someone is threatened or insulted by a group of youths every evening when he comes home. In itself the fact is not dramatic, but if it goes on repeatedly by the end of a month the victim will dread it, and after three months he will frantically try to move to a more peaceful environment. Should we view this finding as indicating an increase in the number of this type of assault, or a greater sensitiveness among those who claim to be victims (but who increasingly tend not to report)? It is difficult to choose between these two interpretations. We would simply note that these attacked people tend to belong to the middle – or lower middle – classes, much more so than victims of more serious assaults. It is these very same categories which begin, toward the end of the 20th century, to express concern with security, an attitude previously confined to the most underprivileged classes of society (more on this in Robert, Pottier, 2004). At any rate, be it because there are more victims or because victims are more sensitive, the multiplication of these low-intensity assaults, shown in surveys, attests to increasing tensions in social relations. A similar conclusion is suggested by the higher police figures for assault and battery. These statistics no longer tend to count serious violence only. They now increasingly measure indignant reactions to the harassment affecting some categories of citizens who have succeeded in getting legislators to translate their distress or bitterness into ever enlarging repressive provisions in the criminal code.

104

P. Robert

There is a last, non-negligible fact: a regional survey (Pottier, Robert, Zauberman, 2002) – the only one to ask that specific question – has uncovered that at least half of the attacks mentioned by respondents are in fact violent thefts, such as snatching of a handbag or cell phone, or attempts thereat. Since then a series of still unpublished local surveys have corroborated this finding. It seems as if the fact that homes, cars and business and leisure premises are better protected leads to a shift, marginal at the least, to grabing the coveted good directly from the owner, personally. All in all, then, we have low-intensity violence, often tinted with predation and increasingly poorly tolerated. The question, now, is how to account for it.

Social Relegation, Provocation, Exasperation (Robert, 1999, 2002) Violence is unhesitatingly assimilated with deprived urban areas.10 Often the image of the ghetto is evoked: wrongly so, since the population there is far from homogeneous, networks of relations cross neighborhood boundaries, and there, like elsewhere, people long for a stable job and the apparently indispensable attributes of modern life – a car, a cell phone and fashionable clothing. The problem is that these goals are hard to reach in these socially and urbanistically relegated neighborhoods with their accumulation of features which, although are not exclusively theirs, are particularly dense there: bluecollars, low-rent housing, a large proportion of youths and large families, sometimes with a single parent, poor schooling, a high percentage of immigrants, a closed-in or isolated location and above all, high rates of unemployment and casual labour. The situation is not improving: the INSEE has shown the situation to have deteriorated in three fourths of these urban areas between the 1990 and 1999 censuses (Le Toqueux, Moreau, 2002). In their monograph on the city of Sochaux (Beaud and Pialoux, 2003) found that when the economy picked up between 1997 and 2001 some unemployed youth were definitely able to find a job. . . but only a precarious one, with little socialising effect, therefore, and above all, reversible as soon as the next slump arrived. For these partially disaffiliated groups – that is, disaffiliated essentially from salaried work, at least of the stable sort – the territory, the place where they live again takes on an importance it has generally lost in modern society. We are witnessing repossession of neighborhoods, essentially by young people. There, at least, they are at home in a world where they feel rejected and looked at suspiciously everywhere else. This attempted repossession also involves a sort of race to privatize the neighborhood. These young people feel that their suburban housing project belongs to them, so that other people, including some representing the official organised society, are intruders in their eyes. The strongest, noisiest and most pugnacious individuals and groups therefore fight to gain 10

These ‘‘deprived urban areas’’ contain about 8% of the population.

Violence in Present-Day France

105

control of the public space in the neighborhood: that way they will at least have something of their own, giving them an identity. The diffuse violence of some youths from deprived suburbs has much to do with a draging of latency and idleness. They failed entrance to a longer educational curriculum that would lead to qualification. The normal alternative for working class youth – getting a job at an early age – is no longer open to many. The industrial sector has far fewer openings for unskilled candidates, and at the same time, there is strong hostility to factory work among these young people. The tertiary sector often demands a modicum of schooling and has standards for appearance they do not meet. On the whole, precarious employment demanding very few skills is proliferating. So that the only path open to these youths is an obstacle course of repeated, precarious jobs, unattractive because of the poor pay and hardly providing any socialisation. In some families, chronic unemployment or under-employment has ended up leaving members so far from the working world as to make it difficult for them to return to salaried employment. Lastly, access to employment is further complicated by the (bad) reputation of their neighborhood and xenophobic prejudice against people – even if well educated – whose parents were immigrants, especially if they came from countries such as Algeria: the history of dramatic upheavals attending decolonisation has left deep marks in attitudes and opinions. All this inevitably produces confrontations, verbal at the least, and sometimes physical. Violence mostly occurs between young people. It serves to conquer prestige items, cell phones, fashion clothing, and it is also a part of strategies of continual harassment, constructing power relations and domination. However, taking over and privatizing one’s neighborhood takes more than simply competing with one’s peers. There are also battles with the local representatives of the authorities. Disturbances generally break out after some alleged, sometimes justified, often amplified but also sometimes entirely imaginary police blunder during a raid, a car check... or some so-called wrong: firemen allegedly delayed their assistance to a youthful victim of an accident. Neither vendetta, strictly speaking, nor civil war – calling them rioting is overemphatic – these disorders are somewhere between mourning rituals – in which the conspicuous destruction of cars prevails – and explosions of utter rage. Sometimes they go beyond the neighborhood, but as a rule they stay inside it, thus contributing to its privatisation. Spectacular but sporadic outbreaks actually count less than the ongoing guerilla warfare that attempts to deny official agencies the right to impose their (public) order on the neighborhood. School violence, even more than the stoning of police cars, is typical of this. Now that the traditional working class possibility of getting a job at an early age is waning, to the point of disappearing, demands and anxieties of every sort all focus on schools. More clearly than ever, schooling is seen as the only path to social qualification. Unsurprisingly, then, it elicits anxiety and resentment in those youths who realize they are falling behind as early as junior high school level, as well as those who have already tried and lost the school option. Those

106

P. Robert

who have won entrance to a school curriculum leading to qualification leave the street culture behind them, whereas the others continue, even beyond their junior high school adolescence. This explains why the atmosphere in poor suburban schools is so explosive. Public opinion has been struck by the violence against local representatives of the public authorities, but less so by the violence directed against social workers and citizens’ group militants. And yet, like the deterioration of crime prevention schemes and youth clubhouses, but even more so, they speak of the determination to take possession of an urban area in what may well become a mafia-type conception. Neighborhood take-overs may indeed reflect business/ dealing strategies as well as strategies aimed at achieving distinction. This may ultimately turn pockets of chronic suburban unemployment into strongholds of an underground economy. When we have, side by side, areas whose residents know they are pretty much excluded from the labor market and a prohibition producing illegal profits, the scene is set, and in the long term those people will be drawn into that informal economy. Vandalism, violence and deteriorations are echoed by a lack of maintenance and neglect and inertia on the part of officials. And actually, who is in charge? Often responsibility is passed on to others: not only do agents of the public services, who no longer live in the area, dream exclusively of getting out of it or spending as little time as possible there, but again, the development of semiprivate, semi-public spaces dilutes their responsibility. Is it up to the landlord, the local community, the local representatives of the administration, or managers of shopping centers and supermarkets? When a profitable activity, business or show is to be protected, the necessity and advantages of cooperation in a co-production of security are readily grasped. Some parties are spurred by their business interests, others haunted by the intolerable prospect of a disturbance of public order. But when the only stake is maintaining a truly public space in a derelict neighborhood, the tendency is to pass the buck. The outcome is that those residents who do not participate in the competition for that space may well find themselves confined to their lodgings. Whatever the goals of those people who try to take over public spaces in derelict areas, the fact remains that they drive those inhabitants who do not take part in the conquests back to their private spaces, their homes. They no longer have access to any peaceful public space, open to all, in their own neighborhood. Derelict neighborhoods are not places where victimisation most commonly takes place, but they are where it resounds most, since it is tied to an environment to which one is relegated for lack of sufficient means to go elsewhere. Often they are viewed as pockets of delinquency, but there is a tendency to forget they are also reservoirs of victims who feel abandoned by the authorities. Violence in these areas is fed by the feeling that the world is pitiless for the poor and the weak, and especially for youth at the low end of the social ladder and of immigrant descent. For anyone with no financial capital or professional or family status on which to rely, all that remains is one’s own person, courage

Violence in Present-Day France

107

(physical), eloquence, posturing and clothing. Behind all this looms a honorbased conception of self. It goes with a degree of verbal and physical violence and the return of physical confrontation. But this provocative play around reputations is both alarming and exasperating. It has become totally foreign to our ways of socialising: its grammar escapes contemporary readings. It remains marginal in a society that believes strongly in a model from which the body is excluded. The diagnosis is not that we are reverting to a society based on honor and physical confrontation – even less, a resurgence of natural human ferocity – rather we are simply facing a ’bricolage’ resorting in gusts, and in the margins of society, to an alternative pattern. For those who find themselves permanently excluded from any form of public recognition, who are convinced that they have no place or utility in society, all that remains is illusory compensation through verbal and physical confrontations. Government authority has been wearing down for a long time, and is now combined with chronic denial of access to stable employment, without anything else replacing this affiliation process. The attending normative allegiances have become fragile, and we discover that social workers, citizens’ groups, the schools and even police or Justice have a hard time achieving socialisation when they are no longer backed by the rough discipline of the workplace. Furthermore, the state centre has failed to subdue (or to win over) the periphery. But this is no divorce over social values, as people feign to believe when they enjoin the residents of these fringes to ’be integrated’, to become civilized: they already are, by far, through their consumer habits, cultural references and aspirations. The counter-culture they occasionally exhibit ostentatiously looks more like a series of token reactions to obstructed social affiliation. The reason for the failure is elsewhere. The governmental centre is suffering a breakdown – not only do its control apparatuses work badly now, but worse, it is unable to get the suburbs to enter the political scene, probably because it doesn’t really try to –, above all, it has nothing to offer those fringes that would be worth their while. In short, groups of people who are convinced they are socially useless are being asked to respect the rules nonetheless. Far from being radically other, the derelict neighborhoods accumulate and personify to the utmost the challenges present elsewhere in a more diffuse form. They represent a periphery, not a distant world.

Policies Run on Fashion, not on the Basis of Evaluation The problem of violence in present-day France, a gnawing question for the last quarter-century, is still affected by the inability of the administration to discover the open sesame which would enable it to control what is, after all, a limited problem.

108

P. Robert

There was a time when security policies relied essentially on social measures concentrated on the derelict areas, after which they simply piled up one scheme on top of another, with little concern for coherence or evaluation of efficiency, within an urban policy which was clearly no longer the priority for dealing with security/safety problems. The outcome is not clear-cut. Urban policies manifest a concern with national solidarity and a refusal of extreme relegation, but they have difficulty in producing clearly significant results. Perhaps they remained too indecisive, too ambiguous. Perhaps they recoiled in the face of the cost of conclusive desegregation. Perhaps they suffered from an overly local conception.11 Or they may have encapsulated themselves in a bureaucratic conception more intent on managing territories than on galvanizing communities, preferring to act for rather than to ‘act with. . .’. Did they consistently place emphasis on building – easier to implement and served by powerful corporations – to the detriment of social work? Or more simply, perhaps they were not equal to the job of reversing the weighty trends they strove to combat. Urban policies throw a lifeline to a person gone overboard, but they do not succeed in pulling him back into the boat, or maybe that is not really what they are trying to do. As for prevention schemes, they cower in essentially defensive attitudes. The remarkable development of the private security market has succeeded in encapsulating shopping centers, department stores, public facilities as well as sports and entertainment events in security bubbles. It has also enabled those private individuals who are affluent enough to reduce the risk of intrusion in their home or of removal of their car. On the other hand, the needy hardly have access to that resource, any more than to insurance. And private security has not really made purely public space safe, although that is where most assaults and thefts take place, in spite of people’s fascination with long-distance surveillance technology. Whence the recourse, since the late 1990s, to all sorts of flexwork youth jobs whose purpose is to act as mediators, to compensate for the fact that the administrations, social aid agencies, public transportation companies and even the town halls persistently keep their distance from these people in difficulty. The survival of these schemes has been considerably compromised by cutbacks in the funding and dropping of the employment procedures on which they depend. But above all, supportive socialization schemes for these youths has tended to be neglected. They had taken the form of crime prevention clubs and teams, remarkably well developed during the 1960s and 70s, but were thrown out of kilter when the labor market became less receptive. Although there have been

11

Although the ills definitely are concentrated in some areas, there is no proof that the remedies are all to be found there. Districts with the greatest number of problem areas are often deprived of the means to combat the problems effectively, precisely because they are too poor. Strikingly, hardly a month after the fall 2005 urban disorders, the fines inflicted on districts which refuse to build their quota of low-rent housing, so as to keep their inhabitants away from the vicinity of the poor, were lowered.

Violence in Present-Day France

109

some experiences here or there, no systematic attempt has been made to renovate the formula. Increasingly, the central government has fallen back on an essentially repressive conception of security. In recent years it has been greatly encouraged to do so by the terrorist panic and by the implicit equation between areas with a high proportion of immigrants and pockets of recruitment for apprentice terrorists. Paradoxically, and despite the appearances, the government is not well armed. Citizens feel unsafe because of a smattering of minute offenses, a great many thefts and rather a lot of petty violence. But the huge police and judicial apparatuses find it difficult to cope with these. Whith respect to violence, the reason is that few victims turn to the authorities. If the attack is sufficiently serious, they do of course file a complaint, which is given careful attention, but such instances are infrequent. Most often each act, taken alone, is insufficiently patent to be prosecutable, and yet, when repetitious the effect is disastrous. As to property offenses, things work out differently: the propensity of victims to file complaints is more substantial, but they usually do not receive any response. . . because the theft victim is generally unable to identify the thief and the police has gradually given up on trying to deal with such hopeless cases. France is not atypical in this respect; the same problem is encountered in other, comparable countries. However, some comparisons of detection rates indicate that its magnitude is much greater there than elsewhere. Improved handling of this anonymous offending would require that the police be well imbedded locally, and very familiar with the social milieus in which they function. It is easier for a local or regional police force (of the kind most of our neighbors have, for public security matters) to be effective than for a highly centralised institution which manages its personnel and their assignments nationally. Over the last years of the previous century a community policing solution was tried, inspired by the achievements of North American policing, especially in Chicago and Montreal. However, this reform – conducted, it is true, very rapidly and very administratively12 – raised such objections among the personnel, faced with the prospect of having to change their habits, and above all such hostility from the police hierarchy, that it has been practically abandoned. More pointedly than ever, efforts have therefore concentrated on a law and order policy exerting strong pressure – there is talk about zero tolerance – on such targets as youths from deprived urban areas or anyone resembling them, and which deals essentially with cases of breaches of the public order, traffic violations, drugs and illegal immigration. 12 Not enough attention was given to teaching skills that the personnel no longer had, or to distributing incentives capable of making this innovation attractive, and above all, to evaluating the results step by step.

110

P. Robert

This phenomenon is all the more salient since the intention is to deal with ’fear of crime’ and not only with crime, or even the safety of people and property. Now this famous feeling has two sides: on the one hand, concrete fears that people may experience for themselves or their dear ones, people who are threatened by theft or violence, especially if they feel vulnerable; on the other hand, the abstract concern with security that may be felt by those who, fearing societal changes and above all the growing precariousness, tensely proclaim their rejection of any disorder, even if they are not really threatened by it personally. Fears are fragmented, local and variable. Concern, on the other hand, is easier to perceive through public opinion surveys. The latter is therefore given more attention. Since people concerned with security are usually adamantly xenophobic, intolerant of disorder and avidly punitive, crime control strategies targeting illegal immigration, drugs and public order have understandably gained considerable importance now. Notwithstanding, it is not at all sure that this will lead to the lasting solution of security problems. The threats feared by people who are frightened are not dissipated by the firm repression of public order offenses which hardly affect them. As for people concerned with crime, they certainly are charmed to learn that the government loves order, but there is the risk that they will turn out to be insatiable, since what they hate, actually, is change. Finally, it is dubious that insecurity may be lastingly solved by a law and order policy. . . . Thanks to which it can always be brought up again, but that’s a dangerous game. Anyone who tries populist one-upmanship generally loses out to an even more populist opponent. The fall 2005 urban disorders provide a very characteristic example of public policies and their effects. They followed a period marked by: – the end of the doctrine of local policing, replaced by a simple ordermaintenance doctrine, with extremely conspicuous hassling of youths from derelict neighborhoods; – a very marked cutback in the funding of social work associations working in those neighborhoods; – the elimination of ’youth job’ schemes which had been used by town halls, social aid agencies and public transportation companies to recruit all sorts of mediators whom they used to reinvest public spaces and to prevent any incipient disorders; – finally, a series of statements by public officials demonstrating their deep contempt for the population of deprived urban areas. In this context, it is not surprising that those urban disorders broke out. One only wonders when they will flare up again. They are actually not unprecedented; there were outbreaks in 1981 and 1990. The 2005 events corroborated some characteristic features: – in spite of a few very unfortunate incidents, blown up out of proportion but in fact rare, attacks were aimed at commodities – vehicles and some public (and occasionally private) buildings – rather than at people;

Violence in Present-Day France

111

– the disorders practically never spilled over into the more affluent neighborhoods; – there was hardly any coordination between disorders in different neighborhoods, even if they were adjacent; – no organization was instrumental, nor was there any religious influence. In spite of it all, these spectacular episodes are less significant than the latent violence.

Conclusion On the one side we have groups of youths who are sure they are being discriminated against, denied access to stable work and to consumer society, and who indulge in all sorts of more or less brutal provocations and predatory acts. On the other, we have whole strata of society panicked by growing precariousness and terrified by any disorder, expressing their exasperation at the behavior of these pariahs (box 3). And hovering over that witch’s brew, some political groups allied with some mass media and special interest lobbies surf on this concern with security, cloaking what is actually fragmented violence in terrifying dress and using fear of crime and xenophobia as major levers in the political arena. Nonetheless, I am not convinced that violence is a problem of disreputable suburban neighborhoods only. Our investigations show several types of scenes where violence occurs. There is the private scene, where domestic violence takes place, the workplace, the street, and last, the derelict neighborhoods. The latter are clearly identified with the ’suburbs’; the first one too, for the most part (there turns out to be a high prevalence of violence among intimates in areas of social relegation). For the other two scenes there is no such clear evidence. Violence does not specially take place in problem suburbs, and we actually have no way of knowing whether the offenders live there. I somewhat distrust such apparently obvious assertions. The equation violence = suburbs is a tempting one as it exonerates the rest of society from any such danger. But it may be jumping to conclusions and overly reassuring.13 The variety of scenes of violence makes it quite improbable that youths from derelict neighborhoods are the only ones responsible for attacks occurring anywhere. I recently discovered instances of racketing for fashionable clothing and thefts of cell phones among junior high school students in a rather posh residential suburb, a long way from any problem neighborhood at any rate. They themselves spoke freely about the importance of ’brand names’ – meaning the ones most fashionable at a given point in time – in clashes among schoolmates, and also about strategies for power, prestige and domination. Of course, the brutality probably remained more diffuse there, and less visible, in 13

This equation was strongly challenged by Rey, in a text dating back to 1996.

112

P. Robert

any case. However, its mechanisms were not fundamentally different from those found in ’problem’ neighborhoods. We cannot automatically eliminate the idea that derelict areas suffer cumulatively, and to the extreme, from those challenges which are found more diffusely elsewhere. One unsettling but not improbable hypothesis would be that their violence is simply a paroxysmal form of a latent societal ill. It is perhaps for that reason that the rise in violent acts elicits such an emotional response in spite of the stagnating rate of homicides and the relative infrequency of definite injuries. We do try to reassure ourselves by claiming that violence is specific to the suburban ’wild kids’ (sauvageons, as some politicians have dubbed them), who are immigrants, to boot, but we wonder, privately, whether the problem isn’t much more widespread, whether it doesn’t threaten our social relations as a whole. After all, the withdrawal of government arbitration in some situations already sanctions a degree of brutality. Employees, even the most privileged ones at the managerial level, are increasing complaining that they are stressed and treated badly at work; public life, politics and the media are full of warlike expressions. There is no bloodshed, of course, and no real blows are dealt, but the distance between these harsher social relations and physical confrontation is not an infinite one. Perhaps it is this prospect that makes people particularly sensitive to any flare-up of violence: more or less consciously, we read our apprehensions into them. Those countries such as France and England which have had very low homicide rates for a long time are characterized by the fact that their national State was constructed very long ago and exerted a very strong control on the pacification of internal social relations. A loss, even partial, of the dynamism and legitimacy of that construct may, under certain conditions, contribute to a new upsurge of violence in social relations. However limited that violence, it frightens a society which is no longer accustomed to it. And the fact is that concern with security is above all a discourse directed at the authorities: the insecure are dubious of their ability to maintain the peace. It is understandable that this concern should be particularly acute in a country in which the State has occupied, for so long, such a prominent place in the construction of the society within its national boundaries.

References Beaud, S., Pialoux, M. (2003). Violences urbaines, violence sociale. Paris: Fayard. Lagrange, H., Pottier, M.L., Zauberman, R., Robert, Ph. (2004). Enqueˆtes de victimation et statistiques de police: les difficulte´s d’une comparaison, Crime victimisation surveys and police statistics: a difficult comparison. De´viance & Socie´te´. 28/3: 285–316. Le Toqueux, J.L., Moreau, J. (2002). Les zones urbaines sensibles. INSEE Premie`re. 835. Pottier, M.L., Robert, Ph., Zauberman, R. (2002). Victimation et inse´curite´ en Ile-de-France. Les re´sultats de la premie`re enqueˆte (2001). Rapport final. Paris-Guyancourt: IAURIFCESDIP. http://www.cesdip.com. Rey, M. (1996). La peur des banlieues. Paris: Presses de Sciences Politiques.

Violence in Present-Day France

113

Robert, Ph. (1999). Le citoyen, le crime et l’e´tat. Gene`ve-Paris: Droz. Robert, Ph. (2002). L’inse´curite´ en France. Paris: La De´couverte. Robert, Ph., Pottier, M.L. (2004). Les pre´occupations se´curitaires: une mutation?. Revue Franc¸aise de Sociologie. 45/2: 211–242. Robert, Ph., Zauberman, R., Pottier, M.L., Lagrange, H. (1999). Mesurer le crime. Entre statistiques de police et enqueˆtes de victimation (1985–1995). Revue Franc¸aise de Sociologie. XL/2: 255–294. Robert, Ph., Zauberman, R., Pottier, M.L., Lagrange, H. (2001). Measuring Crime: Police Statistics and Victimisation Surveys (1985–1995). Revue Franc¸aise de Sociologie. 42 supple´ment (Annual English Selection): 133–174. Zauberman, R. (1991). Victimes en France: des positions, inte´reˆts et strate´gies diverses. De´viance et Socie´te´. 15/1: 27–49. Zauberman, R., Robert, Ph. (1995). Du coˆte´ des victimes; un autre regard sur la de´linquance. Paris: l’Harmattan. Zauberman, R., Robert, Ph., Pe´rez-Diaz, C., Le´vy, R. (1990). Les victimes, comportements et attitudes, enqueˆte nationale de victimation. Paris: CESDIP. 2 volumes.

From Old Threats to Enigmatic Enemies: The Evolution of European Policies from Low Intensity Violence to Homegrown Terrorism Sophie Body-Gendrot

Introduction Different patterns of violence have recently disrupted European cities. In France, 8,700 vehicles were burnt, 255 schools, 233 public buildings and 51 post offices damaged and 140 public-transport vehicles stoned during three weeks of disturbances which took place in three hundred neighborhoods of several cities in November 2005. Although hundreds of editorials were written or pronounced after these events (with a lot of sympathy for the rioters expressed by the jet set), what occurred gave an impression of ‘‘deja vu,’’ as if urban threats from marginalized places and residents had to periodically resurface at the core of society1. After two youths were accidentally electrocuted while purportedly fleeing the police in Clichy-sous-Bois, a problem area in the Parisian region, a minority of male youths revolted emotionally and torched cars to make a statement of distress and anger. Torching cars is not unusual; it is one resource in the limited repertoire these youths have (every day in France, 140 cars are burnt on average, according to the police unions). Youths resort to their cell phones, either to communicate with other youths and perhaps challenge them to take action or to watch the news and check whether by chance their specific deeds have been covered. Reporters remarked that they were left alone as soon as the news started on the major television channels, as the youngsters gathered around their cell phones to watch them. I observed a similar competition in Strasbourg where cars are traditionally burnt on New Year’s Eve. It involves youths from poor areas who are stimulated by the presence of television cameras and by the potential attention they can get. Sophie Body-Gendrot Director of the Center for Urban studies, University of Paris-Sorbonne, 1 rue Victor Cousin 75230 Paris cedex 05 [email protected] 1

Three months after the disorders, 86% of the French and 89% of the 30–49 year olds expected more disorders; 82% of the French and 90% of the cohort under 30 years of age did not believe that solutions had been brought for the problems (CSA-Le Parisien opinion poll, January 25–26, 2006).

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

115

116

S. Body-Gendrot

The pressure the groups exert on an individual youth to mobilize cannot be underestimated. The reason why I do not use the term riot for these forms of violence is because they are not ‘‘a prelude to negociation’’ (Hobsbawn, 1966), they do not lead to further social integration via their transformation into conflicts. What is striking is that these youths who were 16 year-old on average at the time of these events seemed devoid of political consciousness and asked for nothing. They had no political spokesmen or allies and were probably aware that there are no structures and no elaborated social proposals aimed at engaging in a dialogue with them. Consequently, due to the powerlessness of the state (or its cynicism on issues of this type) and to society’s general indifference to what takes place on its margins except during unusual events, urban disorders will erupt again. An observer from Monfermeil, another locality adjacent to Clichy, said that at the beginning of the disorders, it was mostly high school students, unknown to the police, that jubilantly torched the cars, but that soon after, older, more determined and much more violent individuals, eager to avoid the police, vandalized the buildings. She added that ‘‘the motivations of these very mobile groups made of three or four individuals were unclear; they acted as if they did not feel French, as if they were not socially integrated because nothing had been done for them or had led them to love and respect the country’’ (interview with the author, December 2005). In these disorders, two interacting factors contributed to this urban form of violence: first, frustrated teenagers feeling close to those who had ‘‘died for nothing’’ acted out, expressing their emotions and grief; second, the chaos created by violence was instrumentalized by others, for various motivations which had to do either with revenge or a will to destroy the status quo or mental disorder or many other reasons. It is difficult to figure out what this second type of violent actors really wanted as no one was ever interviewed and as they acted in solo. It must also be emphasized once more that place mattered and that the forms the disorders took differed from one locality to another and from one region to another. Paris was not burning, however2, and the torched cars or phone booths which attracted the television crews were frequently limited to one or two streets, as in Toulouse. In no locality, did the disturbances last more than four nights. They would start after night had fallen, that is, early because it was Fall. In a majority of the banlieues, there were no disturbances at all. On the whole, besides the two youths of Clichy-sous-bois, two deaths were registered, one of which seems to have been motivated by a murderer’s personal revenge rather than by the disorders. The mayors expressed restraint as well as most elected officials, policemen, educators, parents and other social actors who are used to deal with ordinary violence. The central state also acted vigorously to

2

It seems that the Minister of Interior was carefully monitoring the youths’ blogs and that when an allusion to bringing the disorders to the Champs Elysees was made, three thousand policemen were immediately deployed.

From Old Threats to Enigmatic Enemies

117

restore order. Not only was a national emergency law passed, but prosecutors were required to be ‘‘tough’’. On November 14th, around 2,800 people had been placed in custody. Three hundred and seventy five were sent to prison. Out of the 498 juveniles sent to a juvenile court judge, 108 were locked up (Arteta, 2005:100). Most high school students were set free after their arrest. In the specific case of the Bobigny court, the epicenter of the events, a study commissioned by the Prime Minister under the auspices of the Center for Strategic Analysis showed that out of the 122 young adults (18–21 year old), one fourth had no stable job and one fourth was in training. Most were French and lived with their parents. In 34.8% of the cases, they were not convicted due to unconvincing files from the police. A lot of these youths had been accused of throwing stones and other objects at the police, then of arson or vandalism. For those convicted, they were sentenced on average to a fourth month incarceration, including two months and a half of suspended sentencing. A similar study (Mucchielli, Delon, 2007) emphasizes the ethnicity of the convicted youths by this court, two third of them being of North African origin, one quarter of African origin. The length of the riots, their relative contagion and the impressive damages revealed the depth of the frustrations felt by the youths from these areas. While most adults did not participate in the disorders, a lot of them said that they ‘‘understood’’ the revolt because, since the beginning of the 1980s, no governmental policy has efficiently alleviated the social and economic deterioration that these areas have experienced, brought them more efficient institutions (the failure of public education is to be emphasized), supported their political empowerment in terms of political representation and dealt with issues of contempt, discrimination and racism that numerous residents experience from institutional agencies. The second pattern of violence disrupting European cities has to do with homegrown terrorism. The recent trauma felt in the Netherlands after the assassination of Theo Van Gogh (Bader, 2005; De Haan, 2005) as well as the political consequences of the March 2004 attacks in Madrid which killed nearly two hundred people and those of London in July 2005 (Modood, 2005), not to mention specific threats formulated against the cities of Brussels, Paris, Rome and Berlin, question social ‘‘models of integration.’’ Attitudes of tolerance relative to violent expressions of rejection and hatred coming from some fundamentalist Muslims in European cities, which were formely perceived as ingrained in multicultural societies, are now being rejected. After these recent attacks and acts of violence which, one way or the other, meant a disruption of social order by perceived ‘‘dangerous classes’’ for scared majorities convinced that ‘‘there will be other attacks’’3, a noticeable change has been observed among European national and local governments that share a 3

According to Ian Blair, the chief of the Metropolitan Police Service in London, the question is not if but when, a statement also formulated in other countries.

118

S. Body-Gendrot

consensus and are now advocating strict policies of identification, surveillance and repression of suspects. These reactions look familiar to those who observed ‘‘wars’’ on crime and drugs launched in the U.S. by the Reagan and Bush Administrations, followed by the war on terror after 9/11 which translated into restricted civil liberties and diminished constitutional rights. One can also observe the reinforcement of policies of repression in France in 2001 and in 2006, before Presidential elections. Vulnerable, ethnicized minorities are externalized and targeted as a risk by dominant majorities and their political representatives. In this chapter, which is based on numerous interviews and field work experiences from the past ten years, I study links between violence and ethnicity, particularly regarding young Muslims settled in Europe, and I test the assumption according to which ‘‘ethnically inspired violence is a particular type of violence with a logic of its own’’ (Wimmer, Schetter, 2003:248). It is well-known that ethnicity is a socio-historical and contingent construction. It is not just inherited but assigned as well. On one hand, it is a dynamic mode of identity affirmation, interpreted and reinterpreted in function of social interactions and power relations at specific times and in specific places by subjects opting to construct their primary identity as ethnic and finding some meaning to their life in this choice; on the other hand, ethnicity is assigned to individuals by others. This is what J.P. Sartre meant when he said that anti-Semitism is constituted by the look of others on one person who only then feels as a Jew. Few studies have documented violence and crime resulting from different forms of segregation and discrimination in European cities. For instance, collective research gathered by M. Tonry (1997) and by E. Huttman (1991) does not discuss this link. But if the personal responsibility of actors such as juvenile delinquents interests criminologists, the role of urban space cannot be neglected and both people and places are also to be studied in regards to the impact of global external forces (Body-Gendrot, 2000). What this essay attempts to show is that two types of violent young actors in France sharing the same sociological background, living in segregated housing and socio-economic environments filled with socio-economic problems, adopt very different postures as regards to ethnicity. The case of France is intriguing because, first, the country has the largest Muslim and Jewish minorities of all European countries. In 2005, out of 63 million French, ‘‘second generations’’ of North African origin (it may include those of third or fourth generations) are estimated to be around one million and a half to two million and Muslims are approximately five million people. Most of these ‘‘second generations’’ are French (Withol de Wenden, 2005). These statistical approximations reveal the second challenge researchers meet when studying ethnicity in France. Explicit inquiry about a person’s race or ethnicity in a survey or for purposes of an official record is simply banned by law. There is no consensus on whether ethnic categories should be constructed or not or if the expression ‘‘old-stock French’’ makes any sense at all (Blum, 2002). The voluntary ignorance of ethnicity in official thinking was recently

From Old Threats to Enigmatic Enemies

119

reasserted by the French Higher Council on Integration: ‘‘Contrary to the promise of the Republican pact (ie. ‘to make men and women of various origins and faiths live together’), some agitate tirelessly emergency solutions such as positive discrimination, ethnic quotas, which all tend to segment society and whose ultimate effect is to compromise national cohesion. We are told that communautarianism functions perfectly in Anglo-Saxon countries, at the very moment when the Dutch are interrogating themselves on the deep causes of Theo Van Gogh’s murder and the United Kingdom is setting out to find ways to avoid ethnic murders afflicting inner cities. It is perfectly legitimate to borrow from our neighbors the methods which have succeeded in defending diversity and to combat discrimination and segregation. One must however not forget how individuals which have so much to earn by opening up to others may also lose themselves if they completely forget their identity and history. And, what united the French in the past, at the time of the worst religious conflicts, it is to have found with secularization, a neutral public space, a mode of operation to bring different groups together. . .and to have constructed the Republican idea, via this approach’’ (HCI, 2006:44). Researchers have to resort to monographs or to other stratagems to find out whether French of immigrant origin receive equal treatment or whether they experience discriminations. The recent controversy opposing the philosopher A. Finkelkraut to other intellectuals relative to whether Arab and Black youths were contesting France in November 2005 in the banlieues revealed that pointing to contenders and to their motivations via their ethnicity was simply not tolerated (Cypel, 2005:14). The major problem for this research comes from the very term Islam. Currently, intolerance for Islam – not be confused with intolerance for Arabs – does not establish distinctions between terms like Muslim, Islamist, Fundamentalist, terrorist, etc. This amalgamation embarrasses moderate, democratic Muslims who make the vast majority. The present study focuses on two types of violent actors related somehow to Islam. On the one hand, there are inner-city youths whose ancestors come from primarily Muslim countries. According to Godard and Tassig (2007), only 5% of the five million Muslims living in France are practicing Friday prayers, in one of the 1,685 prayer-rooms. Most of the youths studied here are not religious, similar to other inner-city youths, and they have been socialized in the Western culture. They resent being negatively assigned an ethnic identity and suspected as potential terrorists or delinquents, especially by the police, an imposition which creates tensions and antagonisms and eventually leads to eruptions of violence. On the other hand, young Islamic fundamentalists who may be the brothers of the former chose to assert their militant ethnicity to mark their Otherness and their rupture with Western (French) values. A tiny minority of them engage in acts of violent destruction characterized as ‘‘homegrown terrorism’’. Their self-labelling as Muslims and the fear they inspire legitimizes the use of force against them by the local, national and global repression apparatus. In both cases, ethnicity serves to externalize Muslims. The reversal of British and Dutch multicultural policies on this topic of externalization is also worth analyzing as a minority of young Muslims may be tempted by

120

S. Body-Gendrot

terrorism. Indeed, as noted by de Haan and Nijboer (2005:75), three particular conditions are necessary for youth violence: institutional mistrust, a neighborhood subculture with strong elements of territoriality and an environment that fosters a continuous threat of violence. As for public opinions, their perceptions are heavily influenced by the media so that one may wonder whether very diverse populations and increasingly heterogeneous individuals can feel safe in crowds – share public spaces, parks, bus shelters, mass transportation, schools, malls, public and private buildings, etc. – if they feel that Dangerous Others are in their midst? Will they not exert more pressures on their political representatives and demand more protection? To what extent can public authorities engage in the provision of public safety without jeopardizing minorities’ constitutional rights? It was admitted that until recently in multicultural societies, primary loyalties could go to religion or culture. According to the poll passed by the Pew Research Center (2006) among Muslims of 15 Western countries, 81% of Muslims in Britain consider themselves as Muslims first, 89% in Spain, 66% in Germany and 46% in France. In all these countries but less so in France, Muslims trust their close circles of proximity and their communities as buffering mechanisms against outer forces. This ‘‘natural’’ preference has been accentuated by globalization. While producing a homogenization of cultures, it has generated dysfunctional fragmentations, as people threatened by change have tended to trust their large families and networks more, in particular distrusting police officers who are not close to them. The ex-pluribus has gained at the expense of the unum.

Why the Global Context Leads to More Violence ‘‘When society suffers,’’ Durkheim observes, ‘‘it needs to find someone to whom it can link its pain, against whom to revenge for its disappointments; and those naturally selected for this part are those who are already somewhat disfavored by public opinion. Parias play the part of scapegoats’’. And he adds that to fight this evil trend, one wishes that ‘‘governments would have the strength to enlighten masses about the error in which they are maintained rather than looking for allies in the party of intolerance’’ (quoted in Beaud and Noiriel, 2004). The macro-mutations that all post-industrial societies are experiencing (globalization, immigration flows, the growth of inequalities, the decrease of job security, a more distended solidarity, the enlargement of the European Union, the transformations of the Welfare-States, etc.) provoke malaises, uncertainties and fears which may translate into rougher social interactions that lead to more expressions of interpersonal violence (Body-Gendrot, 2002; 2006). Issues of loyalty, inclusiveness and trust allowing societies to function holistically are indeed at stake here.

From Old Threats to Enigmatic Enemies

121

Currently, xenophobia targets various groups such as Gypsies, Jews4 but also Muslims settled in Europe as proven by the various barometers of the European Observatory of racist and xenophobic phenomena created after 9/11 (RAXEN network). Not only are Muslims visible newcomers among historical working classes, but they bear the burden of a past rife with conflicts due to religion and in some countries, to colonization. With the Middle East conflicts adding to the tensions, one observer remarked that currently, Islam has never had a worse reputation since the Crusades among Western public opinion. It appears as ‘‘the religion of fanaticism, of ‘ayatollahs’, of intolerance, of an impossible adjustment to modernity, to democracy and most of all to the secular French model’’ (Khosrokhavar, 2004:14). All too often, a democracy of opinion conveyed by the media portrays Muslims and their communities through negative stereotypes. Juxtaposing coverage of Islamist terrorists in the news section of international reporting and that of young offenders from Muslim communities in the inner-cities or the banlieues in the section related to social issues is what creates ‘‘a spiral of sense’’ reinforcing biases in people’s minds. This influence is not without consequences on public employees that reflect the mainstream and on policemen’s mentalities in particular. The media however are not the only manufacturers of such stereotypes. All over Europe, the far right and its racist discourse exploit a wide diversity of incidents and amalgamate young Muslims’ social disenfranchisement with terrorist Muslims, arguing that Jihadism blooms wherever social frustration prevails in immigrant communities. Some local politicians label the city of Venissieux in the suburbs of Lyon, ‘‘an Islamic republic’’ alluding to the fact that three of the seven French who have been detained in Guantanamo were residents of that locality, not to mention Tarik Ramadan and his brother. The elusiveness of the term Muslim allows for this negative amalgamation of stereotypes, as it blurs subtle distinctions of age, gender, class and opinions, just as a while ago, the term Black was similarly instrumentalized by the Ku Klux Klan in the U.S. or by Enoch Powell’s followers in the U.K. The new forms of violence associated with youth from the banlieues can be referred to as a ‘‘delinquency of exclusion’’ (Salas, 1998) which distinguishes itself from former types of juvenile delinquency that are associated with rites of passage taking place during adolescence or from pathological forms of juvenile delinquency. Finding its roots in precarious socio-economic backgrounds, the delinquency of exclusion expresses itself via periodic violent and collective outbursts or ‘‘riots’’ in ghettoized areas characterized by a dehumanized built environment. ‘‘Sensitive neighborhoods’’ – 751 located in 200 metropolitan areas represent 7.9% of the national population (4.4 million) in 2006, 1. Between 1990 and

4

Due to the lack of space, the topic of violence between youths of Muslim and of Jewish background will not be tackled here (see Body-Gendrot, 2008 forthcoming).

122

S. Body-Gendrot

1999, these zones lost 5.7% of their residents (due to both the smaller size of families and to departures). Sixty five percent of residents in the Parisian region either moved out or to another sensitive zone. A city like Venissieux, in the suburbs of Lyon, the first city in France in terms of Muslim prayer rooms and mosques, has lost 15,000 residents (out of 75,000) in 15 years (Loubiere, 2005:59–60; Kessous, 2005:20). As early as 1991, sociologist A. Touraine (1991, 171) observed that ‘‘French society based on classes carried with it conflicts and inequalities. The laissez faire society that France is currently becoming carries ghettos within it. . .As a consequence, a ‘social logic’. . .gives a meaning to new identities based on the collective experience of exclusion’’ In the case of France, a society with a logic of equality combined with individualism and non-recognition of ethnicity, it is complex for governing elites to address the question of discrimination and for the state to negotiate with leaders of groups that it does not officially recognize as its partners. The heart of the matter is that the state does not know what to do with the banlieues. It only knows how to act on long trends, whereas media pressure requires quick responses; its answers are usually technocratic, whereas the banlieues require tailor made, a la carte measures. Where the apparatus of the state excels however is when it anticipates a threat that would hit its core, as will be seen with the threat of terrorism.

Ethnic Forms of Violence in Relegated Neighborhoods Previously, we stated that the ethnicization of social relations had been a growing phenomenon in France. None of the dimensions of ethnicity defined by Max Weber (1980–1921) in reference to language, religion, culture, physical appearance, national origins, a singular history and a more or less accepted identity has any value per se. Its meaning comes from the context and from its users. Ethnicity doesn’t exist alone and the importance of location, age, gender, socio-economic disparities and inequalities in the production of social and political cleavages around it cannot be obliterated. Ethnicity thus cannot be studied in isolation. The analysis must include other broader dimensions that include specific circumstances (Body-Gendrot, 2006) In defining the 751 French problem areas, the administration (DIV, 2004:4) has selected a certain number of criteria (poverty, education, public housing, number of immigrant households, family size. . .). A percentage over 20% of immigrant families in these areas is a criterion of eligibility for territorial affirmative actions. 38% of the 15–24 year old males are indeed unemployed vs. 15% in other urban zones in 2003. For people who don’t have the French nationality, the rate of unemployment is 15 to 17% higher than for others in their neighborhoods. Youths of immigrant origin construct their own identity here and now out of territorial solidarities. They frequently feel closer to their French peers than to

From Old Threats to Enigmatic Enemies

123

their parents’ traditions which are respected only – when they are – for emotional reasons. Their attachment to the neighborhood comes from the fact that they have nowhere else to go (many of those living at the periphery do not come to Paris to which they refer by the names of the highway or subway gates). After being called ‘‘beurs’’ and ‘‘youth from housing projects,’’ the label Muslim has been assigned to them. Perhaps it is because of 9/11 and dominant views on ‘‘the clash of civilizations’’. In the neighborhoods where these young men live, their acts of violence are part of a complex set of conflict dynamics and their self-labelling changes according to situations. The statement: ‘‘I am not French, I am from Marseille’’ may lead to another one a few moments later, in different circumstances, and the same person may say: ‘‘When I walk in the street, I would like to have the same serenity as a Michel, a Francis, etc. I would like to be called Hassan in the same state of mind’’ (Body-Gendrot, 1993:88). In incidents opposing youths and interpreted from the outside, particularly by the media, as opposing Jews and Muslims settling scores over the Middle-East issues for instance, obviously ideologies and values may conflict, but there are alternative dimensions and perspectives intervening in such dynamics of youth conflict. Indeed, field interviews reveal conflicts originating over girls switching between boyfriends from one housing project to another, over turf control related to illicit traffics or over revenge from alleged or real injuries that happened several years before. Researcher Tore Bjorgo (2005, 54) who studied racist violence and violent youth gangs in Norway showed how necessary it was to go beyond the perceived ‘‘obvious’’ identities of nationalist/racist Viking groups fighting antiracist or Pakistani Young Guns gangs in Oslo and to understand that they mostly struggled for dominance over certain turfs, certain clubs, the city centre and over sexual reputations. Another study (De Haan, Nijboer, 2005) also showed that between violent Moroccan immigrant youth and native traditional Dutch youths, the differences in the relationship with the police were less acute than expected although the conditions leading to violence were similar.

Why Violent Clashes Between Male Youths and the Police are Persisting in France The rigidity of the Republican model of social integration in France, that ‘‘opposes the recognition of collective rights for any group defined by a community of origin, language and belief,’’ does not prevent French policemen who share a social frame of reference from assigning an ethnic identity to (French) immigrant youths from these disenfranchised areas and from exerting racial profiling. French identity papers do not solve any problems when youths from problem areas encounter the police, and in that respect, ethnic markers fulfill an important function in police work. According to Zauberman and Levy (2003),

124

S. Body-Gendrot

‘‘The habit of judging individuals on the basis of their supposed ethnic characteristics is acquired on the job, through professional socialization. Police racism is of a reactive character. People do not enter the police because they are racist; rather, they acquire racial prejudice through a process of professional socialization’’. Our own research (Body-Gendrot, Wihtol de Wenden, 2003) has shown that French police academies almost never directly tackle the issues of discrimination and racism in the training sessions and on the whole, police unions are hostile to the development of discussions on these issues which were avoided in France until recently. In other words, there is no French industry on race relations. It would be difficult to convince policemen on the field to refrain from ethnic profiling for another reason. New recruits in the police are frequently under the impression that they are doing ‘‘dirty work’’ for which society does not want to take responsibility and that they are submitted to contradictory orders: they must be efficient (proven by the number of arrests they make) and they must be fair (testified by an equal treatment given to everyone, whatever their profile). But stopping one subway user out of five or twelve, the exact sequence chosen at random, as done recently by the New York Police Department to make sure that people’s constitutional rights were respected is not efficient. It requires considerable means; doubling officers patrolling in the subway costs New York City 1.9 million dollars a week in overtime and in the long term, it is bound to create negative reactions which will eventually hurt the police department. Visibly idle in the public space and easily recognizable via their rappers’ paraphernalia, some of these male youths are perceived as a category of risk to public safety by a number of residents who mistrust them and call the police over incidents in the public space or abusive occupation of hallways. Policemen are required to control the youths’ behaviors and instill social discipline to them. In the field, these rank and file policemen have a large freedom of action, sometimes interpreted as a ‘‘hierarchical inversion’’ (Monjardet, 1996, 88). Their professional culture leads them to see these youths as ‘‘hostile’’ even if the likelihood of their resorting to violence is low. In a mimetic posture, the youths also perceive the police as a rival gang trying to control the public spaces which they have somewhat appropriated and privatized for lack of alternatives. They complain of ethnic profiling and of discrimination, a grievance that, at the individual level, street-level professionals know about, hide or sometimes reveal5. In France, the institution’s denial that ethnic profiling is routine police

5

Similar problems occur in the United States and, moreover, the racial issue is ingrained in the formation of the nation as well as in other European countries. But in multicultural countries, efforts are made to treat minorities as groups that have separate collective identities and are endowed with special rights. Racial profiling is denounced and redressed, whenever possible, by the police and justice institutions, as was seen with the Stephen Lawrence case in UK (MacPherson, 1999).

From Old Threats to Enigmatic Enemies

125

work constitutes one of the reasons for violent antagonisms youths feel towards it. The national police does not serve a community police function. Young people’s frustration is caused by the gap between the principles of equal treatment they are taught in school and the reality that they experience every day. The level of accumulation of grievances on the part of both youths and the police explains why, some neighborhoods will experience outbursts of violence and other similar neighborhoods will not. What is striking here is that the institution controling the youths stabilizes their various identities on a single ethnic dimension. It seems that projecting this very identity on them frees police officers from ambivalence and helps them reinforce boundaries (Volkan, 1988). This imposition of Otherness thus legitimizes the use of force by racist police officers. A French youth of immigrant origin reported on the toll-free number 114 that an officer had told him, ‘‘You are just an Arab, you are not even French’’. These attitudes are a postscript to French colonial history which has never been addressed properly. Relations between ethnic groups are still poisoned by repressed feelings forty years after the end of the Algerian war. On each side, memory is still perpetuated in an antagonistic way, without leaving room for the position of the other. Each side feels bitter: it is likely that numerous policemen, like the majority of the French public opinion, think that colonialism played a positive role and feel betrayed by those immigrants whom, they think, the country has received, housed, educated, subsidized and employed. At the same time, poorer immigrants and their children feel betrayed by the French who disrespect and discriminate against them as if they were second class citizens or worse, dangerous terrorists, depriving immigrant parents of voting rights, even though they have lived for decades in the country. Their children experience a slow social mobility due to a lack of adapted education and job access. As I explained in previous studies (2003; 2005, 18–20), some of the male youths of Muslim background in problem areas resent this assigned ethnicization and their criminalization as a category by the police. The analysis of 113 anonymous calls to a toll-free number recording ethnic grievances between 2000 and 2002 reveals that differential treatment by the police mainly occurred during police controls by hostile officers on the streets, in subways and on buses when there was no one to report the verbal aggression. But sometimes witnesses did call the toll-free number. One of the youths I interviewed in a banlieue of the Parisian Region described what shocked him: ‘‘It is as if I were not like them; they ‘killed’ me with the fact that I am dark-skinned. They wounded my parents’ dignity. They have a tooth for tooth attitude and if they are hurt by some youths, then they must hurt other youths in return’’ (Body-Gendrot, 2005, 20). At a time when Europe is harmonizing the defense of human rights, why is the institution not redressing police misbehavior more vigorously? Internal sanctions from inspection corps are unusual (between 250 and 350 a year out of 120,000 policemen). Judges rarely sentence abusive police officers to pay damages to their victims, and between the victims’ words and those of the police, there is an assymetry of positions (Jobard, 2002). Moreover, an

126

S. Body-Gendrot

increasing trend in the last twenty years is that such policemen are prompt to file suits for contempt or insulting behavior as soon as an incident occurs. Social workers told me that during the three weeks of contagious disorders in the Fall of 2005, gendarmes sometimes protected the youths from infuriated policemen. Such a social context explains why any incident involving youths and the police may degenerate. Unable to rely on legal protection, young people can only count on themselves and on the intimidation they exert on rank-and-file policemen, or even on anti-riot teams (CRS). This assumption echoes that of Willem de Haan and his team (2005, 79–84) in their studies of Amsterdam and Groningen in the Netherlands: ‘‘youth violence against residents and the police may be explained as a form of self-help’’ under specific social conditions (distrust of the police, territorial subculture, threats. . .). In sum, ‘‘the radicalization of ethnic stereotypes and the reinforcement of symbolic hierarchies fuels the expectation and propensity to violence’’ (Wimmer, Schetter, 2003:255). As long as the dialectics between poor, ethnicized youngsters and institutions remains negative, the legitimacy of the latter is contested. Youth will not believe that social change may reach them positively and they hold both fatalistic attitudes and a feeling of injustice. Under such conditions, why should they adopt the norms of those by whom they feel rejected? Why would they resist the temptation of violence?

Are Criminogenic Areas the Breeding Ground of Homegrown Terrorism? Before attempting to answer this issue, the methodological question criminologists need to address is whether it is legitimate for them to establish links between youth delinquency and homegrown terrorism. It should be clear from the start that several factors make the two issues very different: the political ideology mobilizing young Muslim terrorists pursuing long-term goals of violent destruction vs. low or medium intensity violence of immigrant youths, the extent and nature of political and judicial responses to threats at the local, national and global levels, the diversity of implied partners, the terrorists’ search for maximum visibility for their actions whose goal is to kill the greatest number of people which contrasts with delinquents’ goals. Nevertheless, the question needs examination because, first, a minority of young male offenders and potential terrorists share similar capacities to inflict violence on their social environment and the governing elites are relatively powerlessness in putting an end to these cycles. Second, the profile of violent youths and homegrown terrorists also bear similarities: they come from the same communities and neighborhoods and sometimes from the same families. Out of their real or alleged social exclusion, they have constructed a confrontational identity that leads them to act out and/or to use violence (to be defined) as a tool. But the status that these two types of actors give their

From Old Threats to Enigmatic Enemies

127

ethnicity constitutes a major difference. Whereas numerous young delinquents who happen to be Muslims do not practice their religion any more than other French (5% to 10%) and refer as much to their parents’ culture as to their peers’, former delinquents turning to Islamic radicalism emphasize their ethnicity as their first identity and express their hatred of ‘‘those who are not us’’. It is important to point out how erronerous some foreign journalists were in linking the November 2005 disorders to religious motivations. The outbursts were not the first instance of a semi-organized Muslim insurgency in Europe, as Daniel Pipes (quoted by Laurence and Vaisse, 2006) wrote in the New York Sun, but rather the youths’ expression of frustration and despair over their experiences of disrespect and their lack of future prospects. By contrast, young Muslims resorting to terrorism say that they are inspired by religious radicalism and that they accept martyrdom for a cause they have embraced. A leaked report of the French Intelligence Service in June 2005 (which should be taken cautiously) offers evidence that the two types of actors share similar backgrounds (Le Monde, July 13:7). From the cases of 1610 converts, detected because of their proselytism or their delinquency or their relations with radicals, the report establishes that one third of them were already known to police services. Males make up 83 percent of the cohort; they live in the same urban zones as other Muslim households. In 44 percent of the cases, the social environment and their relationships have led them to study salafist Islam (23 percent) or the Tabligh (28 percent). The second cause (27 percent) of conversion to salafism is marriage or a relationship. Proselytism (17 percent), which characterizes the Tabligh and its radical preachers, comes third. In 4 percent of the cases, prison has been the breeding ground of their conversion. Half of the cohort left school without graduating and is officially jobless. One third of them have low skilled occupations (such as airport security jobs –this situation rose a controversy and more surveillance is exerted, with the appointment a Perfect in charge of the airport security in 2007). One can wonder what kind of credit should be given to this report, though. Indeed, it is not unusual for intelligence services to fabricate ad hoc enemies (as the U.S. does, with a list of 325,000 people allegedly connected to terrorism) and to increase the potentiality of threats in the eyes of the public, emphasizing, in the words of the French Prime Minister, that ‘‘there is a real continuity between a fundamentalist discourse and terrorist acts’’ (Smolar, 2004a). It is known that the Renseignements ge´ne´raux (R.G.) which was a political police for the state, focused on banlieues after various scandals had tarnished its reputation in the 1980s. It could well be that the 3 500 members of the French Intelligence Service have found a way to carve their niche in a very competitive professional environment with Islamic extremism, a widespread threat that is alarming governing elites, fascinating the media and fueling a diffuse fear in the population, including 73% of Muslims in France (Knowlton, 2005).

128

S. Body-Gendrot

A current Canadian research project, led by J.P. Brodeur and his team (Brodeur, 2005; see also Lal, 2004), draws a stimulating comparison between the ways organized criminals and terrorists operate. Their training is different. Regular delinquents generally train with peers or via older delinquents, ‘‘at home’’, whereas homegrown terrorists train abroad, in camps in Pakistan or Afghanistan. They latter also benefit from ‘‘sanctuaries’’, as was the case for one of the terrorists from Leeds who headed to Rome after the London attacks. Another major dissimilarity is that terrorist networks can have connections with each other (for instance the Algerian terrorist network joining with Al Qaida), while delinquent youth organizations do not. More research is needed to further this comparison. What these two types of violent offenders also have in common are multiple forms of ruptures in their socialization which tend to culminate in forms of negative heroism. Such ruptures have an impact on their social identity, they corrode their legal socialization, they lead them to distrust the law, which they perceive as illegitimate, and to follow contagious behavioral scripts that gain prominence because there are so few pro-social scripts (cf. Anderson, 1999; Fagan and Wilkinson, 1998). A case in point is that of Mourad Amriou, living at La Courneuve, on the outskirts of Paris. Of Algerian origin, he is a member of the UOIF (the Union of French Islamic Organizations), and is also close to the Muslim Brotherhood’s thousands of unpaid activists. His father moved to France in the 1960s and, after finding a construction job, he brought his wife over and became the father of ten children whom he had little time to control. Mourad completed high school but could not get a job. He was sent to prison twice for drug possession. ‘‘One August night in 1999’’, Ian Johnson reports (2005), ‘‘he was hanging out with friends in a square, drinking beer and smoking marijuana. A man on a bike approached him. He was a member of the Tabligh’’. It is worth mentioning here the names of other members of the Tabligh: Zacarias Moussaoui, the only member of the group who was charged in the 9/11 attacks; Herve´ Djamel Loiseau who died fleeing the 2001 American bombardment of Tora Bora in Afghanistan; Djamel Beghal, an Algerian-born French and confessed member of Al Qaida, living in a banlieue and convicted with five other accomplices for plotting to blow up the U.S. Embassy in Paris in March 2005. ‘‘Mr. Amriou’s friends ran away from the missionary, but Mr. Amriou decided to stay on. The older man sat down and made small talk. . .The two men talked until dawn. They met day after day for a week. By the second week, Mr Amriou was, as he puts it, ‘reIslamicized’. He got rid of his jeans and T-shirts and started wearing the white gown and skullcap that he wears today. He cut his hair but let his beard grow thick and long. He quit smoking, drinking and started praying five times a day. He has become a preacher, moving from one UIOF mosque to the other, teaching his audience how to find safety in the mosque from the dangerous outside world.’’ No element here can allow a researcher to predict the evolution of this young man and whether he will eventually resort to

From Old Threats to Enigmatic Enemies

129

homegrown terrorism. So far, he acknowledges his trajectory of rupture from French mainstream society and justifies other preachers’ antisemitism. But, according to Halima, a senior Tablighi activist (quoted by Smith, 2005), young men wishing to migrate from the Tabligh to more militant forms of Islam have no trouble finding their way. ‘‘Everyone knows which mosques attract Salafists, and if you go and ask, it’s easy to get into the Djihadi network.’’ Of the hundred or so Muslims engaged in fundamentalism, he estimates that half of them have already been recruited by Salafists. The testimony of a street educator in the suburbs of Lyon also points to the same rupture phenomenon. Bertrand (2004) explains how the converts to salafism carefully maintain their distance from mainstream society. They take advantage of the closed-up environment that prevails in the microcosm of public housing projects. It requires from them to keep their distance from French society, to take a different appearance, to remain rebels and to contest others in the neighborhood. The evolution in the use of the term Islam and what has been made of it is striking. In the 1980s, in France as in other European countries, Islam was perceived by authorities as a way of buying social peace. At one point, moderate imams or older brothers were pressured by local authorities to exert their leadership on young offenders and to appease tensions. This move is similar to the Ten Point Program which was enacted in Boston after a moral contract was passed between an alliance of 42 churches and the local police. The Churches would ‘‘clean’’ their drug-infested neighborhoods and the police would drop their colonial attitude (Body-Gendrot, 2000). This ‘‘deal’’ worked for a while in France. But for a small number of male Muslims in deep malaise and in search for their bearings, terrorist networks became appealing due to an anti-democratic, anti-Semitic and inflammatory rhetoric displaying their hatred of the West (Kepel, 2004). As shown above by the case of Mr Amriou, marginalized urban areas which were already hot spots on police maps then became a fishing pond for radical recruiters (‘inspirers’). Then in 1995, Khaled Kelkal, an ordinary young man who had joined Islamist networks carried out terrorist attacks and was subsequently killed by the police. His case fit a similar profile. He had found an identity of substitution in Islam; as he told a journalist from Le Monde, ‘‘I am neither Algerian nor French, I am Muslim’’. Currently, out of the dozens of French Muslims from marginalized urban areas involved in the guerilla in Iraq, twelve have died, a dozen are still in Iraq, dozens are currently incarcerated in French prisons for their connections with Iraqi networks. Since 2002, 548 people have been stopped for their relations with Islamist terrorism and of 140 stopped in 2006, 17 were sent to prison (Smolar, 2007:10; Tourancheau, 2005). Roland Jacquard, an adviser on terrorism to the UN Security Council claims that in France, according to judges investigating cases linked to Al Qaida, only 100 people have ties to radical Muslim groups. For DST, those seeing themselves as Allah’s soldiers average 300. The intelligence service, for its part, estimates the number of Salafists to 5000.

130

S. Body-Gendrot

A French website organizing Islamist forums had tripled its audience in the last six months of 2006. Who is right6? Various sources have confirmed that, out of 1700 prayer rooms, 75 would be subjected to Jihadist pressures (Fuller, 2005; my own interviews). It then appears that the number of invisible members of the second or third generations who are ready to get involved in destructive violence is minuscule. Most of them did not convert because of the war in Iraq but because they felt excluded from the Western society of which they are citizens. ‘‘Born again’’, they are rebels looking for a cause, just as certain types of violent young offenders studied in the past were perceived as rebels (Sicot, 2000). Olivier Roy (2005) remarks that ‘‘the Western-based Islamic terrorists are not the militant vanguard of the Muslim community; they are a lost generation, unmoored from traditional societies and cultures, frustrated by a Western society that does not meet their expectations [. . .] They find their cause in the utopia of a universal ummah, in an abstract and pure Islam, in the same way that the Left radicals in the 1970s – the Baader band or the Red Brigades – opted for terrorism in the name of the ‘world proletariat’ or of the ‘Revolution’ without caring for its consequences’’. The phantasmatic fears that they trigger are consequently the hardest to combat. The danger comes from the trend of punitive populism which is now becoming a component of political choices in Western democracies, from the call to revenge against some ‘‘Others’’ grounded in popular emotions, from a general political rhetoric based on fear and from the dramatization of the threat constructed by the media. Promises made by political leaders are radical, public opinion is indifferent to their efficacy, and penal laws are legitimized by the alleged security need for everyone (Salas, 2005: 57). The scope of the political responses goes beyond the local and the national levels and reaches the global level. While with young offenders, the task of the territorial police is to apprehend individuals or groups of individuals who are deemed to be responsible for their acts, the goal with homegrown terrorism is to target a group of individuals embodying high risks – lethal ones – for democratic societies. This task generally resorts to the Intelligence Service (R.G.) and to the territorial surveillance (DST) accountable to the Minister of Interior, to Defense (DGSE-external surveillance and military intelligence) to Justice

6

We share R. Leiken and Brooke’s concerns when they state that the inaccessibility of data on terrorism for research means that much terrorism analysis merely clusters media reports rather than building and analyzing a data set. ‘‘Most of the researchers (are) not producing substantively new data or knowledge. They (are) primarily reworking old material. . .the majority of researchers’’. . .writings and analyses (are) based entirely on data produced by others. This lack of individual data generation and research (is) a profound concern. The paucity of verifiable data often has indeed limited terrorism studies to informed opinion rather than data, except for Mark Sageman (2004) and a handful of other scholars (see della Porta and Merkl) and in France, besides G. Kepel and O. Roy’s studies (2004), Khosrokhavar’s in-depth interviews of foreign and French islamists in prisons (2006).

From Old Threats to Enigmatic Enemies

131

(antiterrorist judges in Paris) and to Customs accountable to the Minister of Economy and Finance. There is an obvious problem of coordination. All the more because the issue is no longer to stop, indict and judge individuals, but to identify threats. Is the process a police operation? An act of war? Internal and external notions of security, strategies of deterrence and the protection of law and order are blurred.

What Should be Done? The treatment of these new risks of violence relies on three possible and simultaneous strategies which have been adopted by most European countries (Bleich, 2005). The first one tends to confront the threat by neutralizing its ethnic dimension, ignoring the influence of ethnicity. Such is the case with the Vigipirate program in France and with all the recent measures decided by Germany, Italy and the U.K to secure public space. Numerous states quietly take measures and get prepared (Bronner, Ceaux, 2005). A second trend is to harden surveillance, repression and the deportation of Islamists hostile to democracies. The surveillance of violent offenders and of homegrown terrorists basically relies on police work, provided it should be coordinated among services. ‘‘When you search a suspicious building,’’ an officer of the New York Police Department told me in 2004, ‘‘you may as well meet a youth gang, maffiosi or Islamist terrorists’’. After the July 2005 attacks in London, the investigative work led by two thousand policemen in order to deter potential terrorists was based on an excellent knowledge of streets, buildings and spaces where potentially suspect categories live and work, as well as on high technologies of identification and surveillance, from cameras to satellites transmitting documents to countries via Interpol, Europol or other agencies. With policemen walking their beat in the London wards, the London police service is ahead of other services. This prevention work also relies on previous urban experiences with terrorist attacks (the Irish Republican Army in London and Basque nationalists in Madrid, Corsican and Algerian networks in France). An explanation of why, at least so far, there have not been similar attacks to those of Madrid and London in Paris may have to do with the early experiences of terrorism that Paris underwent and with the lessons that were then drawn by the state. As early as 1954–62, during the ‘‘Algerian war’’, domestic terrorism was fought by the agency for international surveillance, DST and by the counter-espionage service, SDECE. In 1986, 14 terrorist acts were committed in France in the name of the Committee of Solidarity with Arabic and Middle East Political Prisonners. A law aiming at fighting terrorism was then passed, blurring notions of external and internal threats and coordinating various agencies in charge of safety from the Ministries of Foreign Affairs, Interior and Justice. Then in 1995 and in 1996, the Algerian radical Islamism, the GIA, launched terrorist attacks. In 1998, 138 people were indicted for supporting the ‘‘Chalabi network’’. As a consequence, antiterrorism became an obsession for

132

S. Body-Gendrot

the officials in charge of territorial security and in particular, for the judges working on the case. Due to these dramatic events, tools were elaborated at the state summit, drawing on the connections which had first been elaborated during Algerian colonization and then during its aftermath. As an example, according to Smolar (2004b), for years the French Intelligence Service (R.G.) has specifically monitored three hundred sensitive urban zones suspected of more or less pronounced Muslim communautarianism. For this branch of the police, eight criteria determine what a sensitive zone is: an important number of immigrant – including polygamous – families; many community organizations; ethnic stores; numerous prayer rooms; non-Western and religious ways of dressing; anti-Semitic and anti-Western graffiti; school classes made of primo-migrant, non-French-speaking children; the exodus of ‘‘old-stock’’ French families from the neighborhood. As can be seen, for a country that does not officially recognize ethnicity, such an empirical definition is paradoxical. The R.G. observes that in three hundred urban zones or so, Muslim preachers are part of the landscape. ‘‘Their fundamental proselytism is fruitful, especially among the young and the children who are supervised by numerous organizations in education and sports’’. ‘‘The preachers eventually generate anti-French racism, a sort of negative identity made out of the parents’ culture, street culture codes and broad references to Islam’’. This perception is not shared by researchers who emphasize the appeasing influence of Islam by contrast (Leveau et al. 2005; Smolar, 2004b). Mosques, bars, restaurants, Koranic bookstores, hallal butchers, long-distance telephone shops, day care centers, clothes stores suspected of harboring extremist Islamists are thus under secret surveillance in sensitive zones. When suspicion is beyond reasonable doubt, under the claim of health and safety checks or of a tax audit, undercover round ups take place. Every year, they are conducted and among suspects taken into custody, some are prosecuted and radical clerics are deported. Important cash flow is seized. Theoretically, field information gathered by the intelligence agency, by other services or by territorial police forces is transmitted to a command unit. For instance in Paris, it is sent to the Police Prefect who until recently also headed the Regional Defense Zone and as such coordinated the Army, the fire brigades, the territorial security branches, the judicial police and all the anti-terrorist branches, in symbiosis with the summits of the state. Compared with the U.S. federal fragmentation of antiterrorist units, UCLAT, the French national coordinating agency gathers and exchanges daily information with officials in charge of territorial safety (DST), the Intelligence Service, the national antiterrorist section (DNAT) or the office of external security (DGSE) and it relates to the state leadership as well. But in practice, such clear line of command is blurred and some specialists advocate a reorganization under the leadership of an inter-ministerial cabinet depending on the President. In 2005, a turning point was reached when the Minister of Interior, Nicolas Sarkozy, chose to use the media to make the French public aware of the terrorist risk. ‘‘The risk of violent action on our land is real, the threat is serious,’’ he said, and no distinction

From Old Threats to Enigmatic Enemies

133

should be made between outside and inside danger (Tourancheau, 2005). This fear-based strategy, which is new and unusual in France, seems directly influenced by the American and British leaders’ strategies; they take advantage of diffuse threats to rally the largest numbers behind them for political purposes (for a comparison, see Schain, 2007, forthcoming). Strikingly, countries like Britain and the Netherlands, which had tolerated expressions of hatred of the West from Muslim preachers, seem currently to opt for the French posture of repression. A series of laws after 9/11 2001 have proscribed Islamic organizations supporting terrorism, troublesome places of worship have been closed and pretrial hearings have been organized in ‘‘secret’’ courts. Several hundred people are stopped and examined each year for suspected involvement with terrorism. As in France, some of them can be detained without trial or deported. Police powers have been extended. But then, the government has also met with serious opposition from people eager not to jeopardize civil liberties and habeas corpus protection, specifically from the Law Lords. After the London attacks in 2005, the Prevention of Terrorism Act extended the powers of the police to detain suspects for up to 28 days. It has been estimated that out of 2000 Islamic networks close to Al Qaida, around 30 are engaged in various lethal plots. Targets aim at the public transportation system, the City, Canary Wahrf, etc, according to the M15 leader, E. Mannigham (2007: 43). Threats of this type allowed the former terrorist Rachid Ramda to be sent back to France and Abu Hamza, the former imam of the Finsbury Park mosque in North London to be heavily convicted for exacerbating hatred. For its part, the French system allows prolonged custody in case of criminal conspiracy in relation with a terrorist enterprise. In the long term, the consequences of anti-terrorist measures will hurt moderate Muslim communities which, at the same time, are being asked to cooperate with the police. It is not easy to do so. When Ahmed Aboutaled, a Muslim city councilor in Amsterdam, asked groups of young Dutch Muslims if they would speak out if they learned that a member of their families was preparing to plant a bomb, unsurprisingly the response was silence and evasion (Cohen, 2005). But had they decided to do so, would intelligence services take the information they received from them seriously7? A third strategy consists in increasing measures of social inclusion, favoring forums and dialogues between Muslim communities’ lay and religious leaders, helping Muslims debate about what consensus they share, an approach frequently followed by the British police and its various advisory boards from the Muslim communities. The problem is that, unlike other religions, Islam has no organized leadership. The number of imams who were trained in Western countries (as opposed to imams who are imported from Muslim countries) is 7

There was a controversy in the Netherlands over the lack of attention the AIVD (intelligence services) gave the information provided by a Moroccan who was close to Bouyeri and who had warned them that he was dangerous and ready to act (Stroobants, 2005: 20).

134

S. Body-Gendrot

growing very slowly, even in prisons. Controversial Tarik Ramadan was asked by the British Prime Minister in 2006 to join a circle of advisors he gathered around him. French officials are beginning to tackle the issue of discrimination with the High Authority to fight discriminations and in favour of Equality (HALDE), created in December 2004. Minister N. Sarkozy also departed from previous postures in creating the French Council of Muslim Cult. He took a calculated risk in incorporating the radical branch of Muslims (UOIF), hoping to co-opt and soften the organization.

Conclusion Two types of violence have been examined with links to ethnicity either voluntarily asserted or assigned. It would be wrong to think that with low intensity violence, democracies have little to fear. Any phenomenon of violence generates a political process of repression which, sooner or later, fosters frustrations and more cycles of violence. Of the three types of responses described above, the first two, based on repression, receive a strong support from politicians, the media and voters. Due to ethnicity, an ill-defined and enigmatic Otherness feeds their fears. In the Fall of 2005, after the Danish cartoons on Islam took European politicians by surprise and provoked hostile demonstrations of some Muslim countries, one wondered about the state of freedom of expression and of mutual respect in multicultural societies. Numerous libels, pamphlets, plays and films (a bomb thrown at the movie theater playing The Temptation of Christ in Paris caused one death) were produced against the Roman Catholic and other religions over the centuries. But blasphemies hurting religious practitioners’ deep convictions were nevertheless tolerated by the elites because they took place within in-groups, among homogeneous peoples agreeing on their differences. The elites’ dominant position allowed them the comfort of distance. But with Muslim minorities and newcomers with emotional reactions as their first defense, the elites and majorities are uncomfortable with new codes, and politicians have to step in to ensure minorities’ constitutional protections. The Fatwa against Salman Rushdie and the imams’ violent reactions in Bradford in England, Theo van Gogh’s death and the death threats which followed in the Netherlands reveal that when the status of minority groups is not secure, violence from radical elements is predictable. Courageous political options would be to enlighten masses on the genesis of their fears, on the dangers of catastrophism, of enclosure, of intolerance and of punitive populism. Security comes from a stronger sense of community and from narrowing distances among groups and individuals. The optimal way to fight violence and fears would indeed be for democracies to be as inclusive of their different components as possible and this is what they frequently succeed in doing in many ways.

From Old Threats to Enigmatic Enemies

135

References Anderson, E. (1999). Code of the Street: Decency, Violence, and the Moral Life of the Inner City. New York: W.W. Norton. Andrew, S. (Ed.) (2004). Research on Terrorism: Trends, Achievements, and Failures, London: Frank Cass. Arteta, S. (2005). Pluie de condamnations. Le Nouvel Observateur, November 17–23. Bader, V. (2005). Dutch Nightmare? A Fairy Tale about the Future of Multiculturalism, Unpublished. Beaud, S., Noiriel, G. (2004). Les nouveaux parias de la Republique. Le Monde, February 20. Bertrand, O. (2004). Les proies du salafisme. Liberation, December 9, 38–39. Bjorgo, T. (2005). Conflict Processes Between Youth Groups in a Norwegian City: Polarisation and Revenge. European Journal of Crime, Criminal Law and Criminal Justice, 13/1, 44–74. Bleich, E. (2005, June 17). Religion, Violence and the State in 21st Century Europe. Paper Presented at a Conference on Migration, Religion and Secularism, Paris, Ecole Normale Supe´rieure. Unpublished. Blum, A. (2002). Resisting Identity Categorization in France. In D. Kertzer, and D. Arel (Eds.), Census and Identity. The Politics of Race, Ethnicity and Language in National Censuses (121–147). Cambridge University Press. Body-Gendrot, S. (1993). Migration and the Racialization of the Post-Modern City in France. In M. Cross, and M. Keith (Eds.), Racism, the City and the State (pp. 61–76). London: Rutledge. Body-Gendrot, S. (2000). The Social Control of Cities? A Comparative Perspective. Oxford: Blackwell. Body-Gendrot, S. (2002). The Dangerous Others: Changing Views on Urban Risks and Violence in France. In J. Eade, and C. Mele (Eds.), Understanding the City: Contemporary and Future Perspectives (pp. 82–106). Oxford: Blackwell. Body-Gendrot, S. (2005). Deconstructing Youth Violence. European Journal on Crime, Criminal Law and Criminal Justice, 13/1, 4–26. Body-Gendrot, S. (2006). France: The Politicization of Youth Justice. In J. Muncie, and B. Goldson (Eds.), Contemporary Youth Justice (48–64). Sage. Body-Gendrot, S. (2008). Racist Victimization in France. In G. Anthonopoulos, and J. Winterdyk (Eds.), Racist Victimization in the World. De Sitter Publications, Forthcoming. Body-Gendered, S., Duperz, D. (2002). The Politics of Prevention and Security in France. In D. Duprez and P. Hebberecht (Eds.), The Politics of Prevention and Security in Europe ( pp . 95–132). Bruxelles: UCV University Press. Body-Gendrot, S., Wihtol de Wenden, C. (2003). Police et discriminations raciales. Le tabou franc¸ais. Paris: Editions de l’Atelier. Brodeur, J.P. (2005). Paper Presented at the European Society of criminology Meeting, Cracow, September, Unpublished. Bronner L., Ceaux P. (2005, August 5). Le gouvernement mise sur la videosurveillance contre le terrorisme. Le Monde, p. 5. Brouard, S., Tiberj, V. (2005). Francais comme les autres? Enquete sur les citoyens d’origine maghrebine, africaine et turque. Paris: Presses de Sciences-po. Brubaker, R., Laitlin, D. (1998). Ethnic and Nationalist Violence. Annual Review of Sociology, 24, 423–452. Chiche, M., Chetrit, M. (2005, July 25). Repenser la lutte anti-terroriste, Le Monde. Cohen, M. (2005, June 17). Integration nationale, laicite, secularisation. Paper Presented at a Conference on Migration, Religion and Secularism, Paris, Ecole Normale Superieure. Unpublished. Cypel, S. (2005, November 24). La voix ‘‘tres deviante’’ d’Alain Finkelkraut au quotidien ‘‘Haaretz’’. Le Monde, p. 14.

136

S. Body-Gendrot

De Haan, W., Nijboer, J. (2005). Youth Violence and Self-Help. European Journal of Crime, Criminal Law and Criminal Justice, 13/1, 75–88. De Haan, W. Paper presented at the World Congress of Criminology, Philadelphia, August. DIV- Delegation interministerielle a` la ville (2004, September). Report, Zones urbaines sensibles: un enjeu territorial de la cohesion sociale, 1–8. Fagan, J., Wilkinson, D.L. (1998). Guns, Youth Violence and Social Identity in Inner Cities. In M. Tonry, M.H. Moore (Eds.), Youth Violence, vol. 24, Crime and Justice: A Review Research (105–188). Chicago: Chicago University Press. Fearon, J. and Laitin D. (1999). Weak States, Rough Terrain, and Large-Scale Ethnic Violence since 1945. Paper Presented at the APSA Annual Meeting in Atlanta. Unpublished. Fuller, T. (2005, September 9). Terror Groups Recruit Young North Africans. International Herald Tribune. Garland, D. (2005). Beyond the Culture of Control. In M. Matravers (Ed.), Managing Modernity: Politics and the Culture of Control. London: Routledge. Godard, B., Tassig, S. (2007) Les musulmans en France. Courants, institutions, communautes: un etat des lieux. Paris: Robert Laffont. Godson, B., Chigwada-Bailey, R. (1999). What Justice for Black Children and Young People? In B. Goldson (Ed.), Youth Justice: Contemporary Policy and Practice (15–74), Aldershot: Ashgate. Gurr, T. (2000). Peoples vs. States: Minorities at Risk in the New Century. Washington, D.C.: Institute of Peace Press. Haut Conseil a` l’Inte´gration (2006). Le bilan de la politique d’integration 2002–2005. Rapport au Premier Ministre. Paris: La documentation francaise. Huttman E., Blauw W. and Saltman J. (Eds.), (1991). Urban Housing Segregation of Minorities in Western Europe and the United States. Durham: Duke University Press. Jobard, F. (2002). Compter les violences policieres, faits bruts et mises en recit. Questions penales, XV, 3, June. Johnson, I. (2005, July 11). Islam and Europe: a Volatile Mix. The Wall Street Journal. Kepel. G. (2004). The War for Muslim Minds: Islam and the West. Cambridge: Harvard University Press. Kessous, M. (2005, January 14). La longue derive de Venissieux. Le Monde, p. 20. Khosrokhavar, F. (2004). L’islam des prisons. Paris: Balland. Knowlton, B. (2005, July 15). Muslims Doubt on Extremism. International Herald Tribune, pp. 1–2. Lal, R. (2004). Terrorists and Organized Crime Join Forces. Rand Corporation. Unpublished. Laurence, J., Vaisse, J. (2006). Integrating Islam: Political and Religious Challenges in Contemporary France. Washington, D.C.: Brookings Institution. Leveau R. and K. Finan, (eds) (2005) Musulmans de France et d’Europe, Paris, CNRS Editions. Logan, J. (2001). Quoted by Ricci C. Sorting Out the Census: We are Still Separate and Unequal. University of Albany, N.Y.: University News Page. Loubiere, A. (2005). Une mobilite residentielle insoupconnee. Urbanisme, 345, 59–60. MacPherson, Sir William of Cluny (1999). The Stephen Lawrence Inquiry. London: HMSO. Mannigham-Buller, E. (2007). Partnership and continuous improvement in countering Twenty-first century terrorism. Policing, volume 1, 1, 43–45. Modood, T. (2005). Multicultural Politics. Racism, Ethnicity and Muslims in Britain. Edinburgh University Press. Monjardet, D. (1996), Ce que fait la police: Sociologie de la force publique. Paris: La De´couverte. Mucchielli, L. (2005). Les ‘‘centres educatifs fermes’’: rupture ou continuite´ dans le traitement des mineurs delinquants? Le temps de l’histoire.

From Old Threats to Enigmatic Enemies

137

Mucchielli, L., Delon A. (2007). Qui etaient les mineurs emeutiers de novembre 2005? Melampoulos. Revue de l’Association francaise des magistrats de la jeunesse et de la famille, 10, 97–104. Muncie, J., Goldson B. London, Thousand Oaks, Ca., sage, (2005). England and Wales: The New Correctionalism. In J. Muncie, and B. Goldson, op. cit., 34–47. Pew Research Center (2006 June). The Great Divide: How Westerners and Muslims View Each Other. Washington, D.C.: Pew Global Attitude Project. RAXEN (2001 and following years). Reactions anti-islamiques dans l’UE a` la suite des actes terroristes commis a` l’encontre des Etats-Unis. Reports. Brussels. Roche, S. (2000). La delinquance des jeunes. Les jeunes de 13–19 ans racontent leurs delits. Paris: Le Seuil. Roy, O. (2004). Globalized Islam: the Search for a New Umma. New York: Columbia University Press. Roy, O. (2005, July 23–24). The Ideology of Terror. International Herald Tribune. Sageman, Mark (2004). Understanding Terror Networks. Philadelphia, PA: University of Pennsylvania Press. Salas, D. (Ed.) (1998). La delinquance des mineurs. Problemes politiques et sociaux, Paris: La documentation franc¸aise (812) November. Salas, D. (2005). La volonte´ de punir. Paris: Hachette. Schain, A. Chebel d’Appollonia and S. Reich eds. but I still do not know their publisher., & Schain, M. (2007). Reactions to Terrorism After September 11: New Rules on Immigration in U.S. and Europe. In A. Chebel d’Appolonia (Ed.), The Illusion of Change and Convergence, Forthcoming. Short, J. (2003). Ethnic Segregation and Violence. In W. Heitmeyer and J. Hagan (Eds.), International Handbook of Violence Research (83–96), Dordrecht: Kluwer Academic Press. Sicot, F. (2000). Enfants d’immigres maghrebins. Rapport au quartier et engagement dans la delinquance. Cahiers de la securite´ interieure, 42, 87–108. Smith, C. (2005, May 24). French Islamic group offers rich soil for militancy. International Herald Tribune. Smolar, P. (2004a, May 26). Les renseignements generaux revisent leur priorite´. Le Monde. Smolar, P. (2004b, July 6). Les RG_ s’alarment d’un ‘repli communautaire’ dans les banlieues’’, Le Monde. Smolar, P. (2007, February 7). Les services de renseignement attentifs a` l’hypothese d’un attentat pendant la campagne electorale. Le Monde, p.10. Stroobants, J.-P. (2005, August 1). Le jeune fils du cineaste Theo Van Gogh brutalise´ par deux jeunes Marocains. Le Monde. Tonry, M. (Ed.) (1997). Ethnicity, Crime, and Immigration: Comparative and Cross-National Perspectives, Crime and Justice: A Review of the Research, 21, Chicago: Chicago University Press. Touraine, A. (1991). Face a` l’exclusion. Citoyennete´ et Urbanite´. Paris: Edition Esprit. Tourancheau, P. (2005, November 18). Une bible pour contrer le ‘djihadisme’. Liberation, pp. 5–16. Volkan (1988). The Need to Have Enemies and Allies: From Clinical Practice to International Relationship. Northvale: Jason Aronson. Weber, M. (1980–1921). Wirtschaft und Gesellschaft. Tubingen: J.C.B. Mohr. Whitman J.G. (2003). Harsh Justice. New York: Oxford University Press. Wihtol de Wenden, C. (2005). Seconde generation: le cas francais. In R. Leveau and K. MohsenFinan (Eds.), Musulmans de France et d’Europe (pp. 7–20). Paris: Editions du CNRS. Wimmer, A., and Schetter C. (2003). Ethnic violence. In W. Heitmeyer and J. Hagan (Eds.), International Handbook of Violence Research (pp. 247–260). Dordrecht: Kluwer Academic Press. Wirth, L. (1938, July). Urbanism as a Way of Life. American Journal of Sociology, 44, July, 1–24. Zauberman, R., Levy R. (2003). Police, Minorities and the French Republican ideal, Criminology, 41(4), 1065–1100.

Part IV

Gendering Violent Practices

The Language of Violence: Symbolic Body Parts in Marital Conflicts in Early Modern France Dorothea Nolde

In Early Modern Europe, physical violence was a ubiquitous part of sociability and, as such, culturally coded to an extremely high degree.1 In this context, recent studies have repeatedly emphasized the importance of gender as one of the principal factors that determined acts of violence and their perception.2 Most research, though, presents physical violence as a quasi exclusively male phenomenon, in particular when analysing violence as a means of establishing and maintaining the gender hierarchy. Even the few studies that explicitly address the question of female violence generally conclude that Early Modern Europe’s culture of violence was a male culture, in which women participated, if

Dorothea Nolde Institut fu¨r Geschichtswissenschaft / FB 08, Universita¨t Bremen, Bibliothekstrasse, D-28359 Bremen [email protected] 1

See for example Muchembled, Robert, La violence au village. (XVe–XVIIe sie`cle), Turnhout 1989; Crouzet, Denis, Les guerriers de Dieu. La violence au temps des troubles de religion (vers 1525 – vers 1610), Paris 1990, 2 vol.; Ruff, Julius Ralph, Violence in Early Modern Europe, Cambridge 2001; Greenshields, Malcolm, An economy of violence in early modern France. Crime and justice in the Haute-Auvergne, 1587–1664, University Park 1994; and the articles in Lindenberger, Thomas; Lu¨dkte, Alf, (eds.), Physische Gewalt. Studien zur Geschichte der Neuzeit, Frankfurt a.M. 1995; Eriksson, Magnus; Krug-Richter, Barbara (eds.), Streitkulturen. Gewalt, Konflikt und Kommunikation in der la¨ndlichen Gesellschaft (16.–19. Jahrhundert), Cologne 2003. 2 Barahona, Renato, Sex Crimes, Honour, and the Law in Early Modern Spain : Vizcaya 1528–1735, Toronto, 2003; Hohkamp, Michaela, ‘‘Macht, Herrschaft und Geschlecht : Ein Pla¨doyer zur Erforschung von Gewaltverha¨ltnissen in der Fru¨hen Neuzeit’’, L’Homme 7/2 (1996, special edition, Gewalt’), pp. 8–17; Hufton, Olwen, ‘‘Women and violence in early modern Europe’’, in: Dieteren, Fia; Kloek, Els (eds.), Writing Women into History, Amsterdam 1990, pp. 75–95; Ju¨tte, Robert, ‘‘Geschlechtsspezifische Kriminalita¨t im Spa¨ten Mittelalter und der Fru¨hen Neuzeit’’, Zeitschrift der Savigny Stiftung fu¨r Rechtsgeschichte, Germanistische Abteilung 108 (1991), pp. 86–116; Spierenburg, Pieter, ‘‘How violent were women? Court cases in Amsterdam, 1650–1810’’ Crime, Histoire & Socie´te´s 1, 1 (1997):9–28; Spierenburg, Pieter (ed.), Men and violence, Columbus 1998.

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

141

142

D. Nolde

at all, only marginally.3 The same is true for most of the studies that, due to the growing interest concerning domestic violence, deal with the question of intra-marital violence. For example, Nicole Castan for Early Modern France, Laura Gowing for England and Heinrich Richard Schmidt for Switzerland focus above all on male violence towards women.4 Elizabeth Foyster’s recent study on marital violence in England from the seventeenth to the nineteenth century, despite claiming to present a ‘‘balanced view’’ of both male and female violence, nevertheless addresses physical violence mainly as male violence endured by women. Female violence is mainly described as ‘‘verbal violence’’, whereas physical violence by women is hardly addressed at all and simply appears as ‘‘violent response’’, or ‘‘resistance’’, to male violence.5 As concerns the reciprocal use of physical violence within marriage, Lyndal Ropers study of Reformation Augsburg is an exception, offering a perspective on the role of male violence on the one hand, and pointing out the contradiction between the norm of female non-violence and the reality of female violence within marital relationships on the other.6 There is no doubt, that marital violence should be analysed within the broader context of marital conflict and the contemporary perception of violence. It nevertheless seems crucial, to distinguish between ‘‘violence’’ as an analytical category and the contemporary use of the term, which constitutes itself an object of analysis. In particular, I agree with Pieter Spierenburg’s statement the ‘‘knives are knives’’,7 and that physical violence and verbal abuse cannot be subsumed under the same analytical category called ‘‘violence’’. The following study is

3

See for example the articles in Spierenburg, Pieter (ed.); ‘‘How violent were women? Court cases in Amsterdam, 1650–1810’’ Crime, Histoire & Socie´te´s 1, 1(1997): 9–28; Men and violence (see note 2); see also Daniels, Christine; Kennedy, Michael V. (eds.), Over the Threshold. Intimate Violence in Early America, New York 1999; Greenshields, Malcolm, ‘‘Women, violence, and criminal justice records in early modern Haute Auvergne (1587–1664)’’, Canadian Journal of History 22 (August 1987), pp. 175–194; Hufton (see note 2); Ju¨tte (see note 2); Spierenburg, ‘‘How violent were women?’’ op. cit. 4 Castan, Nicole, ‘‘Puissance conjugale et violence maritale’’, Pe´ne´lope 6 (1982), pp. 95–103; Gowing, Laura, Domestic Dangers: Women, Words, and Sex in Early Modern London, Oxford 1996, in particular pp. 180–231; Schmidt, Heinrich Richard, ‘‘Ma¨nnergewalt und Staatsgewalt. Fru¨hneuzeitliche Ehekonflikte vor Gericht in vergleichender regionalgeschichtlicher Perspektive’’, L’Homme Z.F.G., 14/1 (2003), p. 35–54 ; see also Hohkamp, Michaela, ‘‘Ha¨usliche Gewalt. Beispiele aus einer la¨ndlichen Region des mittleren Schwarzwaldes im 18. Jahrhundert’’, in: Lindenberger; Lu¨dtke (eds.) (see note 1,) pp. 276–302. 5 Foyster, Elizabeth, Marital Violence: An English Family History, 1660–1875, Cambridge 2005, see in particular pp. 38. Physical violence by perpetrated women is mainly discussed in a chapter with the suggestive title ‘‘Resisting violence’’, and in particular under the headline of ‘‘Violent responses’’, ibid. 102–115 6 Lyndal, Roper, The Holy Household: Women and Morals in Reformation Augsburg, Oxford 1989, in particular pp. 165–251. 7 Pieter Spierenburg, Men and Violence, Columbus 1998.

Symbolic Body Parts in Marital Conflicts in Early Modern France

143

therefore based on a definition of violence, that includes all forms of intentional and coercive encroach on the physical integrity of a person.8 I would like to address the question of gender-specific meanings and perceptions of male and female intra-marital violence through the example of France at the end of the sixteenth and the beginning of the seventeenth century. As in other European countries, marriage in Early Modern France was imbedded in a culture that was highly characterised by violence. Marriage being one of the fundamental institutions of society, its order was redefined in the aftermath of the decisions of the Tridentine council and in the context of nascent absolutism. In this process, the regulation of violence played an essential role. On the one hand, new limits were set to the use of male violence. On the other hand, female rebellion and even more so female violence were increasingly stigmatised, and the duty of unconditional obedience became the key element of a redefinition of the role of the wife.9 In this context, the symbolic as well as the communicative meaning of masculine and feminine violence is of particular importance. In order to perceive the communicative dimension of acts of violence, court records, often the main source of historical research on violence, are generally not sufficient. As Denis Crouzet has pointed out for the French wars of religion, it is the context of perception that determines the meaning of a violent gesture.10 Court records (even though they do by no means only reflect the perspective of the authorities but can also be read as evidence for everyday life) reveal this cultural dimension of violence imperfectly. For a historical understanding of violence in Early Modern society a crossing of perspectives with extra-judicial discourses is necessary. The comparison with narrative literature, as demonstrated by Natalie Zemon Davis for pardon letters in sixteenth century France,11 brings to light those cultural evidences, that, although being at work in judicial trials, are hardly ever explicitly addressed there. The study of marital violence in Early Modern France, that I am presenting here, is therefore based on trial records from the Paris Parlement, on normative texts of juridical, theological and humanist origin, as well as on narrative sources, in particular pamphlet literature (occasionnels).

Different Horizons of Meaning The social meaning of marital violence proves to be highly gendered, both on a normative level and in the sphere of everyday behaviour or when it comes to judging deviant behaviour. In form of the right of punishment, male violence was 8

For a more detailed discussion of this discussion see Nolde, Dorothea, Gattenmord. Macht und Gewalt in der fru¨hneuzeitlichen Ehe, Cologne, Bo¨hlau, 2003, p. 5. See also Spierenburg, Men and Violence, Columbus 1998. 9 For further details see Nolde, Gattenmord. 10 Crouzet (see note 1). 11 Davis, Natalie Zemon, Fiction in the Archives. Pardon Tales and Their Tellers in SixteenthCentury France, Stanford 1987.

144

D. Nolde

fully part of the marital order. The right to physical punishment of subordinates was a characteristic feature of all relationships of dominance, from the king’s rule over his subjects to the parents’ rule over their children,12 and it would not have occurred to anybody that marriage could be exempted from this principle. Even less so, as during the sixteenth century the principle of legal minority of the wife was established, while at the same time the post-tridentine Catholic church postulated and propagated the tutorship of the husband over his wife. Jurists, theologians and humanists agreed that the (physical) punishment of the wife was not only the right, but also the duty of the husband.13 In his treatise The instruction of a christen woman (De institutione feminae christianae), a pedagogical work that was particularly influent all over Europe and was translated into French in 1540, the Spanish humanist Juan Luis Vives declares the physical superiority of the husband to be a necessary condition of marital hierarchy: ‘‘[. . .] if the wife was as strong in body and mind, how would she be submitted to someone equal or inferior to her?’’.14 Although the husband’s right of punishment was not controversial as such, its limits were redefined during the sixteenth century. Thus, the common laws, the coutumes, of the sixteenth and seventeenth century no longer included the very broad interpretation of the husband’s right to chastise his wife, granting exemption from punishment even if it had lead to the death of the wife, that had been part of the coutumes until the end of the Middle Ages.15 Simultaneously, royal edicts and juridical treatises established the principle that any homicide, whatever its origin, was to incur capital punishment.16 Only the king, but not the judges, had the right to pardon.17 Judicial practice in fact implemented this change. Defendants, who tried to justify the killing of their wife as a castigation gone too far, were systematically sentenced to death both 12

See Flandrin, Jean-Louis, Familles. Parente´, maison, sexualite´ dans l’ancienne socie´te´, Paris 1984 (1976), p. 170. 13 Koch, Elisabeth, Maior dignitas est in sexu virili. Das weibliche Geschlecht im Normensystem des 16. Jahrhunderts, Frankfurt a. M. 1991, pp. 138–139; Flandrin (see note 9), pp. 143–144. 14 ‘‘[. . .] si la femme estoit aussi robuste en corps et en esperit, comme[nt] se rendroit elle en la subjection d’ung pareil ou moindre de soy?’’ (Vives Juan Luis, De institutione feminae christianae, Antwerp 1524; quoted from the French edition: Vives, Jehan Loys [= Vives, Juan Luis], Livre de l’institution de la femme tant en son enfance, que mariage et viduite´, aussi de l’Office du mary, Geneva 1970 (1540), p. 313; English edition: Vives, Joannes Ludovicus, A very fruteful and pleasant boke called the Instruction of a christen woman, London, Thomas Powell, 1557. 15 Petot, Pierre; Vandenbossche, Andre´, ‘‘Le statut de la femme dans les pays coutumiers franc¸ais du XIIIe au XVIIe sie`cle’’, Recueils de la socie´te´ Jean Bodin XII, La femme (1962), pp. 245–246. 16 See for example Papon, Jean, Secrets du troisieme et dernier notaire, Lyon 1578, pp. 738–739; Duret, Jean, Traicte´ des peines et des amendes, Lyon 1572, pp. 82–85, Imbert, Jean, La pratique judiciaire tant civile que criminelle, Paris 1609 (1550), p. 791; Coras, Jean de, Resolutions de Droict, Paris 1610, pp. 392–393; cf. Laingui, Andre´; Lebigre, Arlette, Histoire du droit pe´nal, Paris 1979, vol. 1, p. 133. 17 ‘‘Ordonnance royale du 14 novembre 1507’’, in: Isambert, Franc¸ois Andre´ et al. (eds.), Recueil ge´ne´ral des anciennes lois franc¸aises, vol. 11, Paris 1827, p. 514; La Roche Flavin, Bernard de, Treze Livres des Parlemens de France, Bordeaux 1617, p. 845.

Symbolic Body Parts in Marital Conflicts in Early Modern France

145

by the courts of first instance and the Paris Parlement as court of appeal. Parallel to this, wives who went to court seem to have been granted much more often than before a juridical separation (se´paration de corps) in order to protect them from the brutalities of the husband.18 Nevertheless, the violence of the husband as such was not sufficient to obtain this measure. The principal presupposition for such a sentence was that the judges recognised a danger for the life of the petitioner.19 The juridical changes were accompanied by a more general debate on the limits of the right of castigation. The theological discourse on marriage often compared the marital dominance of the husband over his wife to God’s dominance over his church. Although, at first glance, this evokes the idea of omnipotence, it nevertheless implied the requirement of a just exercise of power and was presented by the church as a counterpart of any arbitrary dominance. In this context, numerous theologians condemned any violence that ‘‘exceeds’’, as Jean Benedicti puts it, ‘‘the limits of modesty and reason’’.20 At the same time, modesty and reason were disseminated by humanists as male virtues, that any authority holder and consequently also a husband, should show. This argumentation, condemning any form of excess and even more so any excess of violence, referred to Aristotle, who had postulated that a ruler should distinguish himself by prudentia (prudence).21 While confirming the right of castigation in its principle and redefining at the same time its limits, normative discourses remain essentially silent as to female violence within marriage. The reason is that by definition use of violence by the wife was not part of the marital order, so that is seemed superfluous to negotiate its rules and limits. Female violence against the husband is therefore mostly addressed in narrative and satirical literature, as well as in sources concerning everyday history (Alltagsgeschichte), such as the descriptions of mock processions (chevauche´es de l’aˆne), during which husbands beaten by their wives endured public shame, being led through the city facing backward on an ass. It is true that the theatrical staging of this custom presented satirical features as well, but it concerned not fictional but very real husbands, who were either present in person, explicitly named, or recognisable through indicators revealing their profession and their neighbourhood.22 18

Cf. Lefebvre-Teillard, Anne, Les officialite´s a` la veille du concile de Trente, Paris 1972, pp. 180–204. As concerns the role of violence in separation lawsuits in early modern Venice cf. Hacke, Daniela, Domestic Disputes. Women, Sex and Marriage in Early Modern Venice, Aldershot 2004. 19 Boniface, Hyacinthe de, Arrests notables de la cour de Parlement de Provence, Paris 1670, vol. 1, pp. 382–383. 20 Benedicti, Jean, La somme des Pechez et le Remede d’iceux, Paris 1601 (1584), p. 107. 21 See Maclean, Ian, The Renaissance Notion of Woman. A study in the fortunes of scholasticism and medical science in European intellectual life, Cambridge 1990 (1980), pp. 49–50. 22 Pericaud, Marc Antoine, Recueil des che´vauche´es de l’asne faites a` Lyon en 1566 et 1578. Augmente´ d’une complainte ine´dite du temps sur les maris battus par leurs femmes, Lyon 1862;

146

D. Nolde

To what extent female violence within marriage stood outside of social order is also reflected by the fact that sources depict it almost exclusively in two ways: Either female violence is presented in a highly dramatised way as transgression of all social norms, for example in pamphlet literature, or it is ridiculed through satirical representations. According to Sara Matthews Grieco, pictorial representations of female violence within marriage are characterised by the same thematic dichotomy of satirical presentations on one side and dramatic images of crimes against husband, family and social order in general on the other.23 The dominant wife, and even more so the violent wife, was regarded as incarnation of the world upside-down.24 As a consequence, female violence against the husband was always interpreted as grasping for power and as an attempt to reverse marital hierarchy.

Spousal Murder at Court As concerns court cases, marital violence (beside the above mentioned separation cases) was recorded, above all, in its most extreme form, as spousal murder. This was largely due to the fact that in Early Modern France spousal murder was one of the very few violent crimes that was persecuted directly by the state (poursuite d’office). Unlike other crimes, a private plaintiff (who had to bear the costs of the trial to begin with) was not necessary for a trial to take place. The persecution through a poursuite d’office also proves, that marital violence was by no means considered as a private affair between spouses, but was considered of key social and state interest. In the period between 1580 and 1620, 204 cases of spousal murder were brought before the Paris Parlement, the highest judicial instance of the Ancien Re´gime. Ninety seven of these trials concerned female defendants and 106 male defendants.25 A detailed analysis of the auditions reveals that the judges placed male and female violence within different horizons of meaning. This becomes particularly obvious when one looks at the largely differing evidence that figured in trials concerning male and female defendants. Almost half of the evidence in trials against male defendants was constituted of confessions, former acts of violence against the victim and traces of the act. By contrast, more than half of all pieces of evidence mentioned in trials against women consisted in cf. Davis, Natalie Zemon, ‘‘Die aufsa¨ssige Frau’’, in: id., Humanismus, Narrenherrschaft und die Riten der Gewalt. Gesellschaft und Kultur im fru¨hneuzeitlichen Frankreich, Frankfurt am Main 1987, pp. 136–170. 23 Matthews Grieco, Sara F., ‘‘Femmes insoumises et maris battus au XVIe sie`cle’’, L’Histoire 4 (1978), p. 70; id., Ange ou Diablesse. La repre´sentation de la femme au XVIe sie`cle, Paris 1991, pp. 341–342. 24 Davis (see note 19); see also Chartier, Roger; Dominique, Julia, ‘‘Le monde a` l’envers’’: L’Arc 65 (1976), p. 52. 25 One of the defendants appeared twice, in separate trials, for the same crime.

Symbolic Body Parts in Marital Conflicts in Early Modern France

147

testimonies of amoral conduct, marital dispute and what was considered to be suspect behaviour after the act (for example not having sufficiently mourned the dead husband). The auditions of male defendants focused on pieces of evidence that were supposed to provide hints directly concerning the act. Beside confessions, circumstantial evidence in those trials was dominated by clues that presented an analogy with the act, as for example former acts of violence that correlated with the cause of the victim’s death. To peasant Jehan Billard the judge reproached ‘‘that there is a witness against him that he once took her by the throat and that she has been found with prints on the throat’’.26 The auditions of female defendants, on the other hand, were dominated by questions concerning marital dispute and immoral conduct. Moreover, both types of evidence were often linked to each other, the immoral conduct of the defendant being held responsible for marital dispute. Underlying this argumentation, was the question of whether the defendant presented the characteristics of a rebellious wife. Here, the judges based their logic of proof on extra-judicial discourses. The conviction that a rebellious wife presented a danger for the husband’s life and limb and thus was a potential murderer, was indeed a leitmotiv in contemporary discourse on marriage. This idea was even reinforced by theories about the impulsive and uncontrolled nature of women, capable of any kind of excess if not rigorously constrained. In his 1564 treatise, On marital happiness and unhappiness (De l’heur et malheur de mariage), Jean de Marconville entitles one of the chapters ‘‘Of the correction that one must use against women’’ (De la correction de laquelle l[’]on doibt user envers les femmes), where he describes that a man who does not bridle his wife inevitably puts his life at stake. The chapter begins with a warning, what a man has to expect if he does not follow the summons contained in the title: Several are afraid to put themselves into the ties of marriage out of fear to encounter a bad woman, because usually women are more difficult and more aggravating than men, and even so disorderly and dissolute that they have not contented themselves with violating the honour of marriage, but there are many who have killed, murdered or poisoned their husbands. Nevertheless, the fault of this is not so much due to the malignity of women, but to the imprudence and stupidity of men [. . .].27

In the auditions of female defendants regularly appear questions concerning a possible immoral conduct of the defendant, that is to say, sexual misbehaviour in a very large sense. These reproaches seem to be aimed at proving rebellion. In 26

Archives Nationales Paris (A.N.), X2A 959, audition August 7, 1597. Plusieurs craignent a` se mettre e´s liens de mariage de peur de rencontrer une mauvaise femme pour ce que coustumie`rement les femmes sont plus difficiles & fascheuses que ne sont les hommes, & mesmes si dissolues & de´[p]rave´es qu’elles ne se sont seulement contente´es d’avoir viole´ l’honneur matrimonial, mais beaucoup se sont trouve´es qui ont tue´, meurtry, ou empoisonne´ leurs mariz. Toutesfois ceste faulte ne doibt tant estre impute´e a` la malice des femmes qu’a` l’imprudence & bestise des hommes [. . .]. (Marconville, Jean de, De l’heur et malheur de mariage, Paris 1571 (1564), p. 41 r8.) 27

148

D. Nolde

the context of hierarchical order of marriage the sexual relationship between spouses was conceived as an expression of superiority and inferiority respectively – beginning with the church prohibiting the position where the ‘‘man is lying underneath’’ (quando vir succubat), regarded as contrary to the ‘‘natural’’ order of marriage,28 and the anecdotes in Brantoˆme’s Dames galantes,29 and continued through Michel de Montaigne’s Essais. Montaigne calls marital sexuality ‘‘the fundamental action of marriage, the authority of husbands over their wives’’.30 On this foundation, not only adultery but even the wife’s refusal of sexual intercourse was interpreted as challenging the marital power relationship.31 From this perspective, not only adultery as attested by witnesses but the reproach of having shirked the husband or no longer having slept in the common bed became grave evidence at court. Similarly, it was considered incriminating evidence if, according to witnesses, the husband had forbidden his wife to associate with certain persons, or if he had beaten her out of jealousy.

Spousal Murder in Narrative Literature The different horizon of meaning in which spousal murder committed by men and women is placed, becomes even more evident in pamphlet literature. The so called occasionnels were publications of four to sixteen pages, describing spectacular natural phenomena or social events. In the period between 1570 and 1640, at least 25 of those pamphlets were dedicated to spousal murder. Compared to the large thematic variety of pamphlet literature, this is a remarkably high concentration.32 Sixteen of these texts deal with wives who murdered their husbands, eight with husbands murdering their wives, and one publication presents parallel cases of male and female spousal murder. The quantitative preponderance of female spousal murder is also reflected qualitatively in the way it is described. The murder of the husband by his wife was represented as a far more dramatic crime than the opposite case. Stories of female spousal murder thus cumulate elements that characterize the act as particularly evil, 28

Sanchez, Thomas, De sancto matrimonii sacramento, Lyon, 1621 (1602), lib. XI, Disputatio XVI, p. 214; cf. Berriot-Salvadore, Evelyne, Un corps, un destin. La femme dans la me´decine de la Renaissance, Paris, 1993, p. 84; Flandrin, Jean-Louis (see note 9), p. 188; id. ‘‘La vie sexuelle des gens marie´s dans l’ancienne socie´te´’’, in: Arie`s, Philippe; Be`jin, Andre´ (eds.), Sexualite´s occidentales, Paris 1982, pp. 126–127. 29 Brantoˆme, Pierre de Bourdeille, Siegneur de, Les dames galantes, Paris 1990, passim 30 Montaigne, Michel de: Essais, III, 5, Paris, Garnier, 1962, vol. 2 p. 279; see also Cholie`res, Nicolas de, Les apresdisne´es du seigneur de Cholieres, in: id., Œuvres, Paris 1879 (1585), vol. 2. p. 113. 31 Marconville (see note 24), p. 76 r.; cf. Berriot-Salvadore (see note 25), pp. 81–82. 32 Concerning the proportional distribution of subjects see also Seguin, Jean-Pierre, L’information pe´riodique en France avant le pe´riodique: 517 canards imprime´s entre 1529 et 1631, Paris 1964, pp. 69–79.

Symbolic Body Parts in Marital Conflicts in Early Modern France

149

such as intent, ambush, murder during the sleep, hiring of paid murders or dismembering the corpse. Above all, the female murder of her husband is depicted in pamphlet literature as the incarnation of the rebellious wife. Very often she was represented as a fury, who never left her husband in peace during his lifetime, customarily receiving him with railing, insults or even beatings, as soon as he entered the house. Another version of the rebellious woman was the disobedient wife, who resisted her husband’s will and orders. In these stories, the cruelty of the murderous act is depicted as a result of the malicious character of the culprit, and/or of the cruelty and licentiousness of the female nature in general. By contrast, the murder of a wife by her husband was generally not motivated by the husband’s character, but described as originating from a specific situation, as a result of an escalating dispute or as a spontaneous reaction to the wife’s behaviour. The act appears here as blameable but nevertheless understandable, the excess of an otherwise legitimate marital violence. It was accordingly represented as a rather tragic, sometimes almost unavoidable incident, and called an ‘‘accident’’ or ‘‘disaster’’.33 Without any doubt, the murder of the husband by his wife was thought to be far more perverse than the opposite case. According to contemporary understanding, it not only perturbed the social order but reversed it. A wife usurped by this act the power over her husband’s life and thus raised herself above him in an irreversible manner.

The Symbolic Meaning of the Body The symbolic meaning of specific body parts, that can be found in both narrative literature and court records, is particularly revealing. In the first place, there is the importance of the head as target of violence. In narrative literature, stories of wives murdering their husbands always almost mention strokes directed against the head of the victim. In pamphlet literature, in particular, this aspect is generally explicitly emphasized and depicted with many details. Not only was the head considered, at the end of the sixteenth century, as the centre of a person’s identity and will.34 In contemporary discourse, the rule of head over body was one of the metaphors most frequently used to symbolise the rule of the husband over his wife: the term chef designated 33

See for example Tayssonniere, Guillaume de la Histoire du plus espouventable et admirable cas qui ait jamais este ouy au monde nouvellement advenu au royaume de Naples, par laquelle se voit l’ire de Dieu n’estre encore apaise´e, et nous tous humains, sujets a` son juste jugement, Paris 1574, p. 5 v.; Discours tres-veritable de deux meurtres et massacres merveilleux advenuz puis n’agueres en deux et divers mariages, Langres 1603, without pagination. 34 The head was principal target of violent insults, such as knocking down the hat or slapping a person. They were destined to humiliate the attacked person, and often provoked vehement reactions in form of massive physical counter-attacks or lawsuits. (Cf. Muchembled, see note 1, pp. 37, 167–183.

150

D. Nolde

both the head and the authority holder. This metaphor can be found in narrative literature and marriage treatises as well as in the texts of catholic and protestant theologians alike or in juridical treatises.35 To what point this image was culturally self-evident is obvious in the lawsuit between the lawyer Hector L’Aisne´ and his wife Marie Moyse. The wife had already previously obtained a juridical separation and was now bringing an action for the ongoing violence of the husband. In Mai 1600, the first instance pronounced a sentence of life-long detention against Hector L’Aisne´. In June, the Paris Parlement, to which the defendant had appealed, confirmed the sentence. What is particularly interesting in this lawsuit, are the pleadings of the lawyers of the two spouses, who both advanced – with completely opposite aims – the same metaphor. Thus the lawyer of the husband tried to refute Marie Moyse’s accusation with the argument: ‘‘Is he [the husband] severe, cruel, aggravating and inexorable? It is nevertheless her limb, respectively her head[,] the most excellent of all limbs, as so elegantly said Saint Basil [. . .].’’36 On the other side, the lawyer of the wife resumed his understanding of marriage in his pleading as follows: Marriage is a holy and unbreakable society of two equal persons, who unite so much one with the other, that one holds the place of the spirit and the other of the body: the man is the spirit, the father and the head of the wife, and consequently their society must be like that between the father and the son, and not like that between the master and the slave : this is why he must consider himself companion and not master of his wife.37

As shows this quotation, even a pleading for the equality of spouses, who should be companions, does not invalidate the hierarchical understanding of marriage. Simultaneously, the metaphor of the head ruling over the body served to link the rule of the husband over his wife to the rule of Christ over his church and the rule of the king over his subjects. In 1567 was published, in Bordeaux, the first French translation of the Tridentine or Roman Catechism, elaborated in 1565 35 See for example Navarre, Marguerite de, L’Heptame´ron, Paris 1967, p. 269; Caillet, Paul, Le Tableau du mariage represente´ au naturel, Orange 1635, p. 118; Cholie`res (see note 27), p. 115; Marconville (see note 24), p. 79 v.; Vives (see note 11), p. 156; Du Moulin, Pierre, Premiere decade de sermons, Geneva 1643, p. 207; Coras, Jean de, Resolutions de Droict, contenans cent questions notables de matieres Beneficiales, Civiles et Criminelles, Paris 1610, p. 17; Bodin, Jean, Les Six Livres de la Re´publique, Aalen 19772 (1576), I,3, p. 20. 36 ‘‘Est-il [le mari] severe, cruel, fascheux & implacable? C’est ne´ant-moins son membre, voire son chef le plus excellent de tous les membres, comme disoit elegamment S. Basile [. . .].’’ (Peleus, Julien, Les Questions illustres de Me Iulien Peleus, advocat en Parlement, in: id.: Les Œuvres de Me Julien Peleus, Paris 1638, p. 9.) 37 ‘‘Le mariage est une saincte & indissoluble societe´ de deux personnes egales, qui s’unissent tellement l’un avec l’autre, que l’un tient lieu de l’esprit & l’autre du corps : l’homme est l’esprit, le pere, & le chef de la femme, & par consequent leur societe´ doit estre comme entre le pere & le fils, & non comme entre le maistre & le serf: voila pourquoy il se doit estimer compagnon & non pas maistre de sa femme.’’ (Peleus, see note 33, p. 11.)

Symbolic Body Parts in Marital Conflicts in Early Modern France

151

under the direction of cardinal Borromaeus. It stipulates: ‘‘It is thus certain that the husband is compared to Jesus Christ, & the wife to the Church: & that the husband is the head of the wife, as Jesus Christ is of the Church [. . .].’’38 The theorists of the early absolutist state made use of the same image. In 1527, the king was called husband of his subjects and compared to the head ruling over the body’s limbs by the president of the Parlement, Jean de Selve, in a speech.39 Barthelemy de Chasseneux, in 1529, defined the king’s reign in a very similar way: Princeps dicitur maritus Reipublicæ [. . .]. Ita sicut vir est caput uxoris, uxor vero corpus vir. [. . .] Ita Princeps caput Reipublicæ, & Respublica eius corpus [. . .].40

The metaphor of the head ruling over the body thus linked in a mutual strategy of legitimation three of the basic institutions in Early Modern France: marriage, the church and the king’s reign. From this perspective, any attack against the husband’s head by the wife was necessarily a highly symbolic act. An equally symbolic motive in narrative literature are attacks on the victim’s genitals, which was understood as an attack on his virility and thus at the same time on the husband’s rule. Indeed, virility and all its derived qualities – such as will power, activity, etc. – were considered to be one of the principal foundations of marital rule. Although contemporary understanding of manliness was not reduced to sexual potency but comprised a whole series of physiological and personal qualities, procreative capacity as well as complete and intact male genitals were considered a necessary part of virility.41 Eunuchs, for example (including men with incompletely developed genitals), were not only refused marriage but also access to official functions, that is to say to institutional power of any kind.42 In this context the French term impuissance is to be understood in is twofold meaning of impotency and powerlessness. A husband who was deprived of his virility in this double sense, inevitably also lost his claim to dominance.

38 ‘‘C’est donc chose certaine que le mary est compare´ a` Iesus Christ, & la femme a` l’Eglise: & que le mary est le chef de la femme, comme Iesus Christ de l’Eglise [. . .].’’ (Catechisme compose´ et mis en lumiere suyvant l’Ordonnance & Decret du S. Concile de Trente, Lyon 1580, pp. 566–567.) 39 Quoted by Hanley, Sarah, Le ,Lit de Justice‘ des Rois de France, Paris 1991, p. 73. 40 ‘‘The Prince is called the husband of the republic. [. . .] Just like the husband is the head of the wife, and the wife the limb of the husband [. . .], thus the Prince is the head of the Republic, and the Rpublic his limb.’’ (Chasseneux, Barthe´lemy de, Catalogus gloriae mundi, Frankfurt 1603, p. 230.) Cf. Coquille, Guy, Discours des Etats de France (1588), in: id. Œuvres, Paris 1665, vol. 1, p. 322; Le Caron, Louis, Panegyrique III, Paris 1576, f. B.iii; see also Descimon, Robert, ‘‘Les fonctions de la me´taphore du mariage politique du roi et de la Re´publique. France, XVe – XVIIIe sie`cles’’, Annales ESC 47 (1992), p. 1129. 41 Cf. Maclean (see note 18), pp. 30–35, 49–50; Darmon, Pierre, Le Tribunal de l’impuissance, Paris 1979, pp. 30 – 33. 42 Darmon (see note 38), pp. 31–32.

152

D. Nolde

A particularly telling example, assembling both attacks on the husband’s head and on his genitals, can be found in a pamphlet from 1625, entitled True story of a wife who killed her husband, and who afterwards perpetrated cruelties unheard-of to his body (Histoire veritable d’une femme qui a tue´ son mary laquelle apres exerc¸a des cruautez inouyes sur son corps).43 The protagonist, a woman named Marguerite, first kills her husband by hitting him on the head with a stool. Subsequently, she dismembers the body, beginning with the genitals: She pulls the body from under this pallet, stretches it out in the middle of the room, and without being moved by any compassion starts to execute her revenge on his genitals which she cuts : after that, she takes a hatchet with which she gives him a big stroke, thinking to disguise his sex. She cuts his head, puts it into four pieces, cuts the arm underneath the elbow, and the legs underneath the knees.44

Than, it is again the head that becomes the target of her attacks: That fury takes that head, roles it, calls it some words that she throws out with a hoarse voice : after that she comes to the eyes, which she pricks and pulls out with the tip of a spindle : She takes a pair of tongs with which she tears off his nose and ears : That is not all, there still remains some cruelty for her to carry out : she pulls out his beard, without leaving him a single hair.45

Here the beard that is being pulled out introduces another attribute of masculinity. Thus the works of the author and propagandist of gender equality, Marie de Gournay, contain several mocking allusions to the beard as symbol of male dominance. ‘‘That-one, although telling thirty stupidities, will nevertheless win the price due to his beard, or to the pride of a supposed capacity [. . .]’’, reads for example the Grief des Dames.46 Vice versa the lack of beard was regarded as a sign of unmanliness or effeminacy.47 A contemporary proverb resumes this idea in the phrase: ‘‘Peu de barbe sous blesme couleur, / Monstre homme de peu de valeur’’ (Little beard under pale facial colour, shows a man of little value).48 43 Histoire veritable d’une femme qui a tue´ son mary laquelle apres exerc¸a des cruautez inouyes sur son corps, Lyon 1625 (Reprint Lyon 1877). 44 ‘‘Elle tire ce corps de dessoubs ceste paillasse, l’estend au milieu de la chambre, et sans estre esmue¨ d’aucune compassion, commence a` executer sa rage par les parties honteuses qu’elle luy coupe : apres elle prend une hache de laquelle elle luy donne un grand coup, croyant de desguiser son sexe. Elle luy coupe la teste, la met en quatre quartiers, coupe les bras au dessous du coude, et les jambes au dessous des genouı¨ lz.’’ (Histoire veritable d’une femme qui a tue´ son mary (see note 40), pp. 5–6.) 45 ‘‘Cette Megere prend cette teste, la roule, apostrophe dessus quelques paroles qu’elle jette hors d’une voix enroue´e: apres vint aux yeux qu’elle luy creve et tire avec la pointe d’un fuseau : Elle prend des tenailles avec lesquelles elle luy arrache le nez et les oreilles : Ce n’est pas tout, il luy reste encore quelque cruaute´ a` exe´cuter : elle luy arrache la barbe sans luy en laisser un seul poil.’’ (Histoire veritable d’une femme qui a tue´ son mary (see note 40), pp. 6–7.) 46 Gournay, Marie de, Grief des Dames, (1626) Geneva 1993a, pp. 64–65; cf. id., E´galite´ des hommes et des femmes, (1622)Geneva 1993b, p. 50 47 Darmon (see note 38), pp. 34, 82–83. 48 Meurier, Gabriel, Thresor des Sentences dore´es (1578), zitiert in: Le Roux de Lincy, Antoine (ed.), Le Livre des Proverbes Franc¸ais, Paris 1996, (1842), p. 363.

Symbolic Body Parts in Marital Conflicts in Early Modern France

153

Cutting the nose, as mentioned in the story quoted above, also had a particularly degrading signification. In juridical texts, the physical punishment of nose cutting was still an object of theoretical debate in the sixteenth and seventeenth century, but hardly ever applied any more in judicial practice.49 Nevertheless, there is much evidence that cutting or biting the nose was a form of interpersonal violence that was far from rare, and whose signification was unequivocal and familiar to everyone. Mutilating the nose ranked as highly dishonouring; as a loss of face both physically and morally. The integrity of the entire person was thus damaged: ‘‘[. . .] this offence, that causes the mutilation of a part so essential for completing the face and accomplishing the form and the figure of man, is very abominable [. . .]’’, wrote Hyacinthe Boniface when commenting a trial from 1664.50 If such a defamation took place in the context of marital relationship, it undermined the official hierarchy. One could think that the acts, described in pamphlet literature and other stories, are merely fictional elements, originating only in the imagination of the authors. Very similar descriptions can nevertheless be found in the records of authentic trials, even though they then appear in the rather laconic language of court records. This is what a Parlement judge reproached to Germaine Carpe, who was charged with spousal murder: If in the evening he had beaten her, and after having thrown himself onto the bed he fell asleep; she took a cord that he had in his pocket, strangled him and then, with a sickle, she cut his nose and gave several strikes, and that she had confessed it that way.51

Like in Germaine Carpe’s case, when court records of female defendants charged with spousal murder reveal details of the murderous act, they almost always speak of carefully directed violence. Primary targets of violence were the head or the throat of the victim. With the exception of two cases, these were the only parts of the body where injuries were ever explicitly located. Marie de la Fosse, for example, was charged with having pushed her husband, Leonard Jarricq, into the river, having stepped onto his hands with wooden clogs, and having hit him on the head with the root of a tree until he was dead.52 The fact that in this case the details of the act are enumerated at length is probably linked to its symbolic connotations: As the scene is 49

Boniface for example mentions a regulation of the Lex Julia, punishing adultery women, among other things, by cutting the nose. ( Boniface, see note 16, vol. 2, III, p. 48. 50 ‘‘[. . .] cette injure qui fait la mutilation d’une partie si essentielle pour assortir le visage & accomplir la forme & la fac¸on de l’homme, estant tres atroce [. . .]’’. (Boniface (see note 16), vol. 2, III, p. 48; cf. Groebner, Valentin, ‘‘Das Gesicht wahren: abgeschnittene Nasen, abgeschnittene Ehre in der spa¨tmittelalterlichen Stadt’’, in: Schreiner, Klaus; Schwerhoff, Gerd (eds.), Verletzte Ehre, Cologne 1995, p. 369.) 51 ‘‘Si le soir il l’avoit battue et s’estant jete´ sur le lict il s’endormit[;] elle print une corde qu’il avoit en sa pochette[,] l’estrangla et puis avec une serpe elle luy couppa le nez et donna plusieurs coups et qu’elle l’a ainsy confesse´.’’ (A.N., X2A 957, audition August 1, 1594.) 52 A.N., X2A976, audition May 28, 1614.

154

D. Nolde

described, Marie de la Fosse is standing above her husband, striking out at his head. This interpretation is also supported by the fact that sometimes children and servants, that is persons who also were under the authority of the head of family, were charged with similar acts of violence, but never any other accomplices. In the case of Marguerite Martin, for example, it is said that she and their daughter had killed her husband by beating him with a stick and kicking him in the head (with wooden clogs as well, as the indictment specifies).53 Here too, the image that is depicted shows a scene in which the man is literally in an inferior position. In the trial against Guillemette Cany and her twenty-two-year-old son Claude Soulineau, who were charged with having hung the husband (and father), the son is said to have already previously hit his father on the head with a plug.54 And in the case of Marie Hubert it was her servant who confessed to have killed the victim with two axe-strokes on the head at his mistress’ order.55 To cut somebody’s throat, on the other hand, was often connotated with the idea of revenge. It is striking that the throat, just like the head, is often explicitly named to locate injuries, whereas otherwise the notices are generally limited to much vaguer indications, for example that the victim had been killed with a knife or an axe. Twenty-year-old Jehanne Magneau, for example, was reproached no less than three times during one audition with having cut her husband’s throat after he fell asleep in the field, were they were tending cattle together.56 In the trial against Denise Moulnier the link between the murder and an act of revenge was even explicitly established. She was reproached that, to take revenge of her husband for mistreating her, she had asked several persons to stab or to poison him, before finally executing her intention by cutting his throat herself.57 In trials against male defendants, violence was of course present, but the message was totally different. Of the three cases that mention head-injuries, two present the smashing of the skull as just one of several injuries.58 Thus the weaver Pierre de la Pierre from Beauvais confessed having first hit his wife into the face with a rod, before smashing an earthenware bottle on her head.59 None of the records concerning male defendants mentions single, aimed strokes on the head. Here, violence was mostly directed at the body. In two cases the victim’s body was perforated with a sword or a fork.60 The thirty-year-old wine grower Bertrand David was reproached with having so vehemently slapped and 53

A.N., X2A 959, audition August 13, 1597. A.N., X2A 979, audition December 29, 1616. 55 A.N., X2A 971, audition January 26, 1609. 56 A.N., X2A 977, audition August 4, 1615. 57 A.N., X2A 974, audition September 11, 1612. 58 A.N., X2A 982, audition January 7, 1620. 59 A.N., X2A 206, sentence January 7, 1620. 60 A.N., X2A 950, audition June 9, 1581 (Crespin Hubert); X2A 976, audition December 13, 1613 (Jehan de Fontaine). 54

Symbolic Body Parts in Marital Conflicts in Early Modern France

155

kicked his pregnant wife that it resulted in her death. The autopsy had shown that the womb was destroyed, which pointed to foot-kicks (or other blows) to the belly.61 Particularly often the records mention injuries all over the body, stemming sometimes from different weapons. The trial against the cloth merchant Anthoine Lamoureulx gives the following cause for the victim’s death: ‘‘[. . .] the body is battered in several places and there are even several scratches at the throat’’.62 In nine cases the defendants are reproached with having thrashed their wife to death by kicking, boxing, or beating her with a stick. In three more cases the indictment emphasized the large number of injuries caused by beating or stabbing. According to the accusation, the tailor Pierre Maronnier had given twenty-three sword-strokes to his wife, Claudine Minart – according to his own statement it was ‘‘only’’ fifteen.63 The two other cases account for even more injuries. Such a high number of injuries suggests that in these cases also, violence was directed at the entire body. This kind of violent acts does not testify of a direct attack against a person and his or her status, but of a violent appropriation of the woman’s body and person. All the more so, since several of those acts of violence present structural similarities with the legitimate and habitual castigation of the wife. This also reflects in the fact that a relatively frequent, although unsuccessful, strategy of justification of male defendants consisted in presenting the killing of their wife as the result of a failed but in principle legitimate punishment. The three supplementary cases, in which the husband is accused of having killed or strangled his wife with his own hands, suggest a similar kind of violent grip.

Violence and Social Sense Despite the laconic style of court records, about one fifth of the trials against female defendants and one fourth of the trials against male defendants reveal details of injuries. Certain types of injuries are by no means randomly mentioned, but carry a meaning of their own. Since only part of the auditions mention this kind of details, it is hard to say to what extent those injuries reflect a representative image. It is, on the contrary, rather likely that this image of violence is influenced by a selective perception by the judges, who thought certain elements of the act important enough to be emphasized, and others not. In any case, in the trials that are analysed here, acts of violence are understood, beyond the mere aggression, as symbols speaking a language of their own. The message they deliver is different, depending on the gender of actor and victim. 61

A.N., X2A 976, audition September 23, 1614. ‘‘[. . .] se trouve le corps meurtry en plusieurs endroicts et mesmes plusieurs esgratignures a` la gorge’’. (A.N., X2A 955, audition April 2, 1587). 63 A.N., X2A 964, audition November 16, 1601. 62

156

D. Nolde

The symbolic dimension of violence shows, that something more than just the repression of a homicide was negotiated at court, namely the definition and the limits of the respective gender-roles within marriage. While male violence against the wife principally appears as an excess of an otherwise legitimate marital violence, female violence against the husband represents above all an attack on his authority and thereby at the same time a transgression of the role of the (submissive) wife. Accordingly, the juridical sanction also contains a second level. By sanctioning – beyond the homicide itself – male excesses of violence on the one hand and female rebellion on the other, the judges of Parlement effectively contributed to establishing the new model of marriage outlined above. In the interest of the king’s claim to power, the husband’s right of violence was limited while at the same time the wife’s unconditional obedience was established as foundation of marital hierarchy. Moreover, the gendered and highly symbolic meaning of marital violence shows that the social sense of violent acts is not inherent to those acts. The cultural codification of violence does not only happen through the exercise of violence, but at leas as much through its perception. As show the different frames of meaning in which male and female violence were embedded, it is only through the context of perception that violent acts become meaningful gestures.

References Archives Nationales Paris (A.N.), X2A 950, audition June 9, 1581. Archives Nationales Paris (A.N.), X2A 955, audition April 2, 1587. Archives Nationales Paris (A.N.), X2A 957, audition August 1, 1594. Archives Nationales Paris (A.N.), X2A 959, audition August 7, 1597. Archives Nationales Paris (A.N.), X2A 959, audition August 13, 1597. Archives Nationales Paris (A.N.), X2A 964, audition November 16, 1601. Archives Nationales Paris (A.N.), X2A 971, audition January 26, 1609. Archives Nationales Paris (A.N.), X2A 974, audition September 11, 1612. Archives Nationales Paris (A.N.), X2A 976, audition December 13, 1613. Archives Nationales Paris (A.N.), X2A976, audition May 28, 1614. Archives Nationales Paris (A.N.), X2A 976, audition September 23, 1614. Archives Nationales Paris (A.N.), X2A 977, audition August 4, 1615. Archives Nationales Paris (A.N.), X2A 979, audition December 29, 1616. Archives Nationales Paris (A.N.), X2A206, audition January 7, 1620. Barahona, Renato (2003). Sex Crimes, Honour, and the Law in Early Modern Spain: Vizcaya 1528–1735. Toronto: University of Toronto Press. Benedicti, Jean (1601). La somme des Pechez et le Remede d’iceux, Paris: G. de la Noue (1584: Lyon: par Charles Pesnot, de l’imprimierie de Thibaud Ancelin). Berriot-Salvadore, Evelyne (1993). Un corps, un destin. La femme dans la me´decine de la Renaissance. Paris: Champion. Bodin, Jean (19772). Les Six Livres de la Re´publique. Aalen: Scientia Verlag (1576: Paris: chez I. du Puys). Boniface, Hyacinthe de (1670). Arrests notables de la cour de Parlement de Provence. vol. 1. Paris: J. Guignard.

Symbolic Body Parts in Marital Conflicts in Early Modern France

157

Brantoˆme, Pierre de Bourdeille, Seigneur de (1990). Les dames galantes. Paris: Le livre de poche. Caillet, Paul (1635). Le Tableau du mariage represente´ au naturel. Orenge: E. Voisin. Castan, Nicole (1982). Puissance conjugale et violence maritale. Pe´ne´lope, 6, 95–103. Catechisme compose´ et mis en lumiere suyvant l’Ordonnance & Decret du S. Concile de Trente (1580). Lyon. 566–567. Chartier, Roger, & Dominique, Julia (1976). ‘‘Le monde a` l’envers’’. L’Arc, 65, 52. Chasseneux, Barthe´lemy de (1603). Catalogus gloriae mundi. Frankfurt: Willer. Cholie`res, Nicolas de (1879), Les apresdisne´es du seigneur de Cholieres. In: Nicolas de Cholie`res, Œuvres, vol. 2. Paris: Librairie des bibliophiles (1585: ). Coquille, Guy (1665). Discours des Etats de France. In: Guy Coquille, Œuvres (p. 322). Vol. 1. Paris: J. Guignard (1588: ). Coras, Jean de (1610). Resolutions de Droict, contenans cent questions notables de matieres Beneficiales, Civiles et Criminelles. Paris: J. Houze´. Crouzet, Denis (1990). Les guerriers de Dieu. La violence au temps des troubles de religion (vers 1525 – vers 1610). vol. 2, Seyssel: Champ Vallon. Daniels, Christine & Kennedy, Michael V. (1999). Over the threshold. Intimate violence in early America. New York: Routledge. Darmon, Pierre (1979). Le Tribunal de l’impuissance. Paris: Ed. du Seuil. Davis, Natalie Zemon (1987). ‘‘Die aufsa¨ssige Frau’’. In Natalie Zemon Davis, Humanismus, Narrenherrschaft und die Riten der Gewalt. Gesellschaft und Kultur im fru¨hneuzeitlichen Frankreich. Frankfurt am Main: Fischer Taschenbuch Verlag. Davis, Natalie Zemon (1987b), Fiction in the Archives. Pardon Tales and Their Tellers in Sixteenth-Century France. Stanford: Stanford University Press. Descimon, Robert (1992). ‘‘Les fonctions de la me´taphore du mariage politique du roi et de la Re´publique. France, XVe – XVIIIe sie`cles’’. Annales ESC, 47, 1129. Discours tres-veritable de deux meurtres et massacres merveilleux advenuz puis n’agueres en deux et divers mariages (1603). Langres: J. des Preys, without pagination. Du Moulin, Pierre (1643). Premiere decade de sermons. Geneva: P. Choue¨t. Duret, Jean (1572). Traicte´ des peines et des amendes. Lyon: B. Rigaud. Eriksson, Magnus & Krug-Richter, Barbara (eds.) (2003). Streitkulturen. Gewalt, Konflikt und Kommunikation in der la¨ndlichen Gesellschaft (16.–19. Jahrhundert). Ko¨ln: Bo¨hlau. Flandrin, Jean-Louis (1982). ‘‘La vie sexuelle des gens marie´s dans l’ancienne socie´te´’’. In Philippe Arie`s & Andre´ Be`jin (Eds.), Sexualite´s occidentales (126–127). Paris: Ed. du Seuil. Flandrin, Jean-Louis (1984). Familles. Parente´, maison, sexualite´ dans l’ancienne socie´te´, Paris: Ed. du Seuil (p. 170). (1976: Paris: Hachette). Foyster, Elizabeth (2005). Marital Violence: An English Family History, 1660–1875. Cambridge: Cambridge University Press. Gournay, Marie de (1993a). Grief des Dames, (1626). Geneva: Librairie Droz. (1626: ) Gournay, Marie de (1993b). E´galite´ des hommes et des femmes. Geneva: Librairie Droz. (1622: ) Gowing, Laura (1996). Domestic Dangers: Women, Words, and Sex in Early Modern London. Oxford: Clarendon Press. Greenshields, Malcolm (1987). ‘‘Women, violence, and criminal justice records in early modern Haute Auvergne (1587–1664)’’, Canadian Journal of History 22 (August 1987), 175–194. Greenshields, Malcolm (1994). An economy of violence in early modern France. Crime and justice in the Haute-Auvergne, 1587–1664. University Park, Pa.: Pennsylvania State University Press. Groebner, Valentin (1995). ‘‘Das Gesicht wahren: abgeschnittene Nasen, abgeschnittene Ehre in der spa¨tmittelalterlichen Stadt’’. In Klaus Schreiner & Gerd Schwerhoff (Eds.), Verletzte Ehre (369). Ko¨ln: Bo¨hlau.

158

D. Nolde

Hacke, Daniela (2004). Domestic disputes. Women, sex and marriage in early modern Venice. Aldershot: Ashgate. Hanley, Sarah (1991). Le ,Lit de Justice‘ des Rois de France. Paris: Aubier. Histoire veritable d’une femme qui a tue´ son mary laquelle apres exerc¸a des cruautez inouyes sur son corps, Lyon 1625 (Reprint Lyon 1877). Hohkamp, Michaela (1995), ‘‘Ha¨usliche Gewalt. Beispiele aus einer la¨ndlichen Region des mittleren Schwarzwaldes im 18. Jahrhundert’’. In Thomas Lindenberger, & Alf Lu¨dkte, (Eds.), Physische Gewalt. Studien zur Geschichte der Neuzeit (276–302). Frankfurt a. M.: Suhrkamp. Hohkamp, Michaela (1996). ‘‘Macht, Herrschaft und Geschlecht: Ein Pla¨doyer zur Erforschung von Gewaltverha¨ltnissen in der Fru¨hen Neuzeit’’. L’Homme 7/2, special edition ,Gewalt‘, 8–17. Hufton, Olwen (1990). ‘‘Women and violence in early modern Europe’’. In Fia Dieteren & Els Kloek (Eds.), Writing Women into History (75–95). Amsterdam: Historisch Seminarium van de Universiteit van Amsterdam. Imbert, Jean (1550). La pratique judiciaire tant civile que criminelle. Paris 1609 (1550). Isambert, Franc¸ois Andre´ et al. (Eds.) (1827). Recueil ge´ne´ral des anciennes lois franc¸aises. vol. 11. Paris: Belin-le-Prieur. Ju¨tte, Robert (1991). ‘‘Geschlechtsspezifische Kriminalita¨t im Spa¨ten Mittelalter und der Fru¨hen Neuzeit’’, Zeitschrift der Savigny Stiftung fu¨r Rechtsgeschichte, Germanistische Abteilung. 108, 86–116. Koch, Elisabeth (1991). Maior dignitas est in sexu virili. Das weibliche Geschlecht im Normensystem des 16. Jahrhunderts. Frankfurt a. M.: Klostermann. Laingui, Andre´ & Lebigre, Arlette (1979). Histoire du droit pe´nal. vol. 1. Paris: Cujas. La Roche Flavin, Bernard de (1617). Treze Livres des Parlemens de France. Bordeaux: Simon Millanges. Le Caron, Louis (1576). Panegyrique III. Paris: R. Estienne. Lefebvre-Teillard, Anne (1972). Les officialite´s a` la veille du concile de Trente. Paris: Librairie ge´ne´rale de droit et de jurisprudence. Lindenberger, Thomas & Lu¨dkte, Alf (Eds.) (1995). Physische Gewalt. Studien zur Geschichte der Neuzeit. Frankfurt a. M.: Suhrkamp. Lyndal, Roper (1989). The Holy Household. Women and Morals in Reformation Augsburg. Oxford: Clarendon. Maclean, Ian (1990). The Renaissance Notion of Woman. A study in the fortunes of scholasticism and medical science in European intellectual life. Cambridge: Cambridge University Press (1980: Cambridge: Cambridge University Press). Marconville, Jean de (1571). De l’heur et malheur de mariage. Paris: Jean Dallier (1564: ). Matthews Grieco, Sara F. (1978). ‘‘Femmes insoumises et maris battus au XVIe sie`cle’’. L’Histoire, 4, 70. Matthews Grieco, Sara F. (1991). Ange ou Diablesse. La repre´sentation de la femme au XVIe sie`cle. Paris: Flammarion. Meurier, Gabriel (1842). Thresor des Sentences dore´es (1578). Zitiert in: Antoine Le Roux de Lincy (Ed.), Le Livre des Proverbes Franc¸ais, Paris: Paulin. (1996: Paris: Hachette). Montaigne, Michel de (1962). Essais, III, 5, vol. 2. Paris: Garnier. Muchembled, Robert (1989). La violence au village (XVe–XVIIe sie`cle). Turnhout: Editions Brepols. Navarre, Marguerite de (1967). L’Heptame´ron. Paris: Garnier. Nolde, Dorothea (2003). Gattenmord. Macht und Gewalt in der fru¨hneuzeitlichen Ehe. Ko¨ln: Bo¨hlau. Papon, Jean (1578). Secrets du troisieme et dernier notaire. Lyon: Iean de Tovrnes. Peleus, Julien (1638). Les Questions illustres de Me Iulien Peleus, advocat en Parlement. In: Julien Peleus, Les Œuvres de Me Julien Peleus. Paris: P. Billaine.

Symbolic Body Parts in Marital Conflicts in Early Modern France

159

Pericaud, Marc Antoine (1862). Recueil des che´vauche´es de l’asne faites a` Lyon en 1566 et 1578. Augmente´ d’une complainte ine´dite du temps sur les maris battus par leurs femmes, Lyon: N. Scheuring. Petot, Pierre & Vandenbossche, Andre´ (1962). ‘‘Le statut de la femme dans les pays coutumiers franc¸ais du XIIIe au XVIIe sie`cle’’, Recueils de la socie´te´ Jean Bodin XII, La femme, 245–246. Ruff, Julius Ralph (2001). Violence in early modern Europe. Cambridge: Cambridge University Press. Sanchez, Thomas (1621). De sancto matrimonii sacramento, lib. XI, Disputatio XVI. Lyon: Lugduni: sumptibus Societatis typographorum (1602: ) Schmidt, Heinrich Richard (2003). ‘‘Ma¨nnergewalt und Staatsgewalt. Fru¨hneuzeitliche Ehekonflikte vor Gericht in vergleichender regionalgeschichtlicher Perspektive’’, L’Homme Z.F.G., 14/1, 35–54. Seguin, Jean-Pierre (1964). L’information en France avant le pe´riodique: 517 canards imprime´s entre 1529 et 1631. Paris: Maisonneuve et Larose. Spierenburg, Pieter (1997). ‘‘How violent were women? Court cases in Amsterdam, 1650–1810’’, Crime, Histoire & Socie´te´s, 1. Spierenburg, Pieter (Ed.) (1998). Men and violence. Columbus: Ohio State University Press. Tayssonniere, Guillaume de la (1574). Histoire du plus espouventable et admirable cas qui ait jamais este ouy au monde nouvellement advenu au royaume de Naples, par laquelle se voit l’ire de Dieu n’estre encore apaise´e, et nous tous humains, sujets a` son juste jugement. Paris: J. Ruelle. Vives Juan Luis (1524). De institutione feminae christianae, Antwerp: Hillenius, French edition: Vives, Jehan Loys [= Vives, Juan Luis] (1970). Livre de l’institution de la femme tant en son enfance, que mariage et viduite´, aussi de l’Office du mary, Geneva: Slatkine Reprints. (1540: ). English edition: Vives, Joannes Ludovicus (1557). A very fruteful and pleasant boke called the Instruction of a christen woman, London: Thomas Powell.

Negotiating Violence and Gender: Security and the Night Time Economy in the UK Kate O’Brien, Dick Hobbs and Louise Westmarland

I knew that if bodybuilding was going to take off, we’d have to get women into it (Joe Weider)

This chapter examines how gender intersects with the working practices and occupational culture of door security staff, often referred to as door supervisors or ‘bouncers’, and who represent an arm of Britain’s bourgeoning private security industry. Whilst recognising that night-time security work can often be mundane and non-eventful (Monaghan, 2002b) the main body of this chapter deals with the management of violence and aggression by female bouncers. Within the context of Britain’s commercially driven, alcohol fuelled night-time economies we focus on what we term ‘violence work’ (Hobbs, O’Brien & Westmarland, 2007) by exploring how female bouncers are controlling and preventing aggression and physical violent episodes involving male and female customers within licensed venues. Throughout this chapter, we draw on the findings of a two year ethnographic study which involved fieldwork carried out in several major cities across the UK and interviews with over 50 women working as bouncers in addition to interviews with police and local authority licensing officers, night club operators and security firm directors. Participant observation was conducted in a wide variety of licensed venues including large capacity mainstream nightclubs, gay bars and gay dance venues, ‘style bars’, private members clubs and traditional pubs. Our study of ‘female bouncers’ working in British licensed venues is timely because of the recent increased demand for women to work within this sector of the industry (Hobbs et al., 2007). Yet previous research has identified that ‘bouncing’ is an occupation dominated by men and that hyper masculine characteristics such as toughness, ‘muscle’ and violent potential are intrinsic

Kate O’Brien School of Social Policy, Sociology and Social Research, University of Kent, Cornwallis Building NE, Canterbury, Kent, CT2 7NF [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

161

162

K. O’Brien et al.

to this profession (Hobbs, Hadfield, Lister, & Winlow, 2003; Monaghan, 2002c). Operating within ‘public’ environments that have traditionally been viewed as a heterosexual male preserve, and which are governed by the rules of patriarchal cultures (Hey, 1986), means that the position of female security personnel in bars, pubs and nightclubs appears to be anomalous with normative conceptions of femininity, utilities of female bodies, and perceptions of women’s relationship with physical violence, control (Braithwaite & Daly, 1994; Miller, 1998) and aggression (Campbell, 1993; McCaughey, 1997:91). While there is evidence illustrating that women are active in carving out cultural spaces of control within the night time environment (Kirkby, 1997), and that particular groups of women establish their own cultures in pubs (Hunt & Saterlee, 1987), women who are employed to monitor, govern and control licensed spaces appear to be contradicting ‘traditional’ gender expectations and behavioural rules. In this chapter we draw on interview data and our findings show that while commercial demands for the female bouncer reinforce ‘traditional’ notions of femininity; for example, women are perceived to be adept at conciliatory styles of conflict management, we illustrate that within the cultural milieu of the licensed venue women are reworking and ‘detraditionalising’ gender within this occupational setting (Adkins, 2004; Lovell, 2000) whilst simultaneously reproducing gendered arrangements. Our findings suggest that women draw on a range of strategies, and not simply a diffusing strategy, to achieve control and manage violent and aggressive incidents. When enacted by women, ‘violence work’ encompasses a form of gendered action or practice that is context specific. Before we develop this argument further it is important to describe in some detail the conditions that have encouraged the entry of the ‘gender specialist’ female bouncer.

Gendering Nightlife and the Rise of the Female Bouncer Now it’s all aimed at the weekends, it’s all about getting the young crowds in the pubs and clubs, and the venues are like, ‘‘oh, knock another 10 pence off that and another 10 pence off this because next door’s charging such and such and we’ve gotta get them in here’’. SD2 Female Bouncer

The growth of night-time economies in Britain have been explained as part of the transformative shift from industrial to post-industrial economic development in British cities (Hadfield, 2006; Hobbs et al., 2003), and in light of the state’s impotence to provide adequate or appropriate public policing, these shifts have created a need for a specialised form of commercial social control (Hobbs et al., 2003). To date, the problem of violence within Britain’s night-time town and city centres has been presented as a male phenomenon, which is dealt with by men (Hobbs et al., 2003; Monaghan, 2002a, 2002b; Winlow, 2001).

Negotiating Violence and Gender

163

Mirroring the domination by men within public policing in Britain, the private security industry (see Erickson, Albanese, & Drakulie, 2000; Wakefield, 2003) including the door supervising sector, represent an industry that is overwhelmingly a male preserve, with women comprising only 7 percent of the workforce (Walker, 1999). Male door supervisors are drawn from working class environments that are most likely to embrace violent conduct, rather than a cross section of society (Hobbs et al., 2003; Winlow, 2001). Furthermore, male recruits to the door security profession bring with them cultural capital that reflects physicality, and a predisposition to utilising both violence, and the negotiation of violence as an integral part of everyday social interactions (Robins & Cohen, 1978). The centrality of violence to the bouncers working environment feeds into the occupational culture and in turn, the working methods employed by doormen encourage and reproduce the notion of a vigorous and assertive masculinity (Yllo, 1993). On first glance, it may be viewed as quite remarkable that women have managed to ‘break into’ this hyper masculine terrain. After all, understandings of women’s relationship with violence and aggression have tended to hinge on the violence power nexus in presenting women as ‘victims’, ‘survivors’ and ‘peacekeepers’. In so doing, femininity is scripted in terms of vulnerability, non-assertiveness and pacification (see Walklate, 2004). The female bouncer then, presents an interesting case in that she contradicts normative assumptions about femininity and social control. Although still in the minority, women can now be seen standing alongside men controlling ‘the doors’ of pubs, clubs and other late night entertainment venues across Britain. Female bouncers appear to be transgressing ‘traditional’ forms of femininity in ways not dissimilar to other types of women who are required to exhibit control and power within traditional masculine environments (Portillos, 1999; Westmarland, 2001; Yount, 1991; see Doathea Nold’s chapter in this collection). Unruly and intoxicated young men have for some time captured the popular imagination and the attention of academics who have described the problem of violence in licensed venues as a male phenomenon (Gough & Edwards, 1998; Graham & Wells, 2003; Maguire, Morgan, & Nettleton, 2003; Tomsen, 1997; Winlow & Hall, 2006). Yet the expansion of night time economies in Britain has brought about significant changes in patterns of consumption. For instance, with the commodification of gay and lesbian identities (Binnie & Skeggs, 2004; Skeggs, 1997), gay and lesbian night strips in some cities have become much more visible and their venues are increasingly being incorporated into mainstream, heterosexual drinking circuits (Casey, 2004). Perhaps more significant is the expansion in night-time leisure consumption by heterosexual females, especially those aged between 16–24 years. There a number of reasons for this increase but it would appear that lifestyle changes are playing an important role. Young women are enjoying much higher levels of disposable income than women of previous generations and increasing numbers of women are choosing to delay motherhood. Subsequently, young women are consuming night-time entertainment and leisure spaces in growing numbers.

164

K. O’Brien et al.

According to Chatterton & Hollands ‘‘[t]he impact young women have had on transforming the character and atmosphere of urban nightlife is huge’’ (2003:149). It is not surprising therefore, that the percentage of British women aged between 16–24 years consuming more than 14 units a week (the recommended number of units per week) increased from 15% in 1988 to 25% in 2000 and that the number of women of this age group consuming more than 35 units (frequently used as an indicator of ‘binge’ drinking) has doubled (Walker, Maher, Coulthard, Goddard, & Thomas, 2002). The consumption of alcohol by young women, who have recently come to symbolise the problematic British ’binge drinker’ (Marsh, 2004), is linked to new forms of participation in drinking and drug cultures by young women (O’Brien, forthcoming). Implicit in these representations is a perception that violence and aggression amongst young females is increasing (Batchelor, 2001), and the night-time environment is the context in which violence by young women is most visible. Yet studies continue to confirm that young men are ten times more likely than women to engage alcohol related violent offences and to be assaulted in a licensed venue (Hutchinson, Magennis, Shepherd, & Brown, 1998; Maguire et al., 2003). As we have argued elsewhere (Hobbs et al., 2007), the huge growth in the consumption of night-time leisure by young women has created a demand for the ‘gender specialist’ female bouncer. Steph is of white British origin, in her early 30’s and employed by a security firm that provides security to several large nightclubs in central London. She provides context to an important aspect of gender specific work tasks assumed of female security staff in the following extract: We clean up, you know, we carry out the girls that are puking up, we have to look after the ones that have got too hammered, we need to find their friends, you know, we need to find the missing handbags, we have to do all that kind of stuff. (N9)

This aspect of what many of the female bouncers in our study referred to as ‘girl work’, is illustrated further by Josie, a woman in her 40’s who has worked ‘on the doors’ for eight years and during the fieldwork period was employed by a large capacity nightclub venue in the north of England. Josie, like the majority of the women we observed and interviewed, is the only female member of the door team at the venue where she works weekend nights. Here, she highlights a gendered division of labour that is defined by the head doorman whose expectations of her are to manage the behaviours of unruly female customers: If there’s bother with girls, he’ll send for me if he wants to. If we need to search a woman he will send for me. If there’s arguments and he wants us to do something he will get us to go and do it but if something is going on and I get there first, cos’ I do try and get there first, he’ll let us get on with it. (C16)

As these examples also imply, the demand for female bouncers is also linked to the needs and anxieties of capital namely, a commercial concern with the

Negotiating Violence and Gender

165

avoidance of risk and litigation. The nature of physical interaction with customers is typically disciplinary and controlling, and the potential legal consequences of touching customers inappropriately during routine door work has increasingly being viewed as problematic within the industry (Lister, Hadfield, Hobbs, & Winlow, 2001; Monaghan, 2003: 21). It is evident from our interviews with women that their male colleagues express most apprehensive about being accused of wrongfully touching the female body as Josie goes on the illustrates here: They (doormen) won’t even like hold them or touch them as they’re escorting them out. The lads will come and get me. . . and say, ‘‘you know that lady we think she’s doing something, will you give her a quick search?’’ Or, ‘‘will you re-remove her from the premises ‘cause she’s that ill’’. . .the lads are frightened to get a hold of them basically in case they say they’ve touched them or anything like that. So they’ll always come and get me, and I mean the lads that I work with they are nice lads and they won’t touch them. (C16)

The women in our study, who worked in nightclub venues in particular, reported that they are employed specifically for searching purposes and as described elsewhere, searching female bodies for drugs, alcohol and weapons by female staff has become standard (Hobbs et al., 2003:275). In Britain, the Licensing Act 2003 has now created conditions that encourage the employment of female door staff, specifically for searching purposes. Section 182 of the new Act, stipulates that the employment of women can be a condition of granting a venue’s license because licensing authorities can now insist that only women can search female customers (DCMS, 2004). They can also require that a female member of the door team be on duty during licensed hours. This legislation is having a significant impact upon the demand for female door supervisors across Britain. It is also evident from the marketing techniques employed by some venues, such as drink promotions, music policy and de´cor, that female consumers have become increasingly crucial to the growth of the night time economy. The employment of women as bouncers by mainstream venues is commercially driven and appears to be based on the assumption that female customers will more likely patronage a venue fronted by a woman. As one male security company owners informed us, many city centre licensees ask me for a woman in order to attract groups of female customers as they travel the night strip. (D10) We have discovered that in a similar way to men, women’s bodies are being used by the leisure industry as a marketable resource. Yet unlike their male counterparts, unsurprisingly, the majority of women who work as bouncers are not employed for the intimidating functions that a large muscular body supplies (Hobbs et al., 2003; Monaghan, 2002b, 2002c). Rather, we would argue that the commercial value of the utility of women’s bodies is located with the implied emotional qualities of the female body (Hochschild, 1983; Steinberg & Figart, 1999), and supports Monaghan’s arguments concerning the bouncer’s body and the commodification of certain ’gendered attributes’ (Monaghan, 2002c).

166

K. O’Brien et al.

For example, some women’s bodies, such as those described by Munt (2001) as the ’butch body’, give off particular cues and gendered messages and we have found that these types of female bodies are used selectively by some gay venues and female only spaces. Other types of female bodies may appear to achieve a ‘softening’ effect (Guy & Newman, 2004) and are sought after because they help create and maintain particular milieus, venues or brands as this male ex-nightclub owner and security company director illustrates: Door staff are there but they are not blocking you. I tell them to smile, talk to people, ‘look as if you’re going to have a good time’. Your door staff are there like a steward on a boat, if the boat gets into trouble, ‘we’ll get you out, don’t worry’. But that’s all they’re there for. So nobody is scared. The girl at the door gives it a really soft feel, just to talk to, say ‘hiya’, it calms them down, it says to customers, this is a nice place, it’s not violent, ‘I’m at the door, I’m looking after you’. (D7)

Processes of regulation associated with the British security industry have also brought about an increase in the demand for female door staff. The recently established Security Industry Authority (SIA) was created as a result of the Security Industry Act (2001) to act as the regulatory body for security personnel, including door supervisors (Lister, 2002; Lister et al., 2001). A new emphasis on ‘customer service’ is being advocated in an effort to replace the old style ‘coercive’ approach and the SIA is actively promoting this ‘consumer led’ and more consensual style of door work in their efforts to clean up the image of the industry. Customer safety is a core component of the new standardised national training courses for security staff where the emphasis is on teaching door staff how to help and care for customers. Central to the ‘clean up operation’ is the desire on the part of the SIA to remove the violent and aggressive image of bouncing and replace it with ‘professionalism’ and ‘customer service’ (Sabic, Walker, McNamara, Smith, & Kolvin, 2004). Mirroring the increased demand for ‘feminine skills’ within other labour market settings (Lovell, 2000), the security industry is promoting the recruitment of female security personnel because women are assumed to be more suited to conciliatory styles of conflict resolution that the SIA is promoting. In the absence of empirical evidence however, the assumption is being made that women will act as ‘diffusers’ of violence and the SIA are promoting the employment of female bouncers for this purpose.

Gendering ‘Violence Work’: Situating Diffusion In accordance with the SIA policy, women are perceived to be suited to the less aggressive style of managing violence where diffusion techniques are utilised (Monaghan, 2003:22) and we have discovered that some women do in fact employ this technique to manage particular violent situations. However, as the discussion that follows will reveal, it would be misleading to

Negotiating Violence and Gender

167

assume that this ‘preventative’ strategy and ‘peace keeping’ role is simply a matter of gender difference and a ‘feminine trait’ (Gilligan, 1982). Our findings expose a more complex picture. It has already been documented that male bouncers also employ diffusing techniques to manage the violent behaviours of customers (Hobbs et al., 2003; Monaghan, 2002c), yet our observations of female bouncers develop this analysis further. We have found that the utility of this technique is not only linked to the style of venue but is often dependent on the gender of those being controlled. For instance, many of the women we interviewed consistently reported that they were much more effective at diffusing violence involving male customers and conversely, that unruly female customers responded better to male bouncers. This observation is exemplified by Rima who works in an ‘up-market’, independently owned bar located on a busy night strip popular with gay and lesbian night-time consumers: If there’s a disagreement going on between two men, they just send me in. Men always think twice about saying anything rude or being physical with a female whereas women would just go for you. Usually if there’s women going off, one of the lads will go. (C2)

The reasoning behind this strategy is illustrated in more detail in the following extract which draws on interview data with Anna who has worked in a variety of city centre venues. Anna also reveals here, the contradictions that are at play in respect of how female bouncers actually manage violence and aggression, and negotiate with male colleagues, the ‘gender specialist’ role: More than one woman on the door is entirely surplus to requirements really. . .‘You’re only there to sort the women out’, that’s what they said but it’s not the case at all. . . .The case is, I’ve thrown a lot more men out than women, I’ve resolved a lot more conflicts between men that didn’t involve women at all. . . I think that a woman is often in a far better position to sort something out between men, because men don’t actually want to fight a lot of the time they just want bump chests and show off. So if a man goes in, he’ll just start bumping chests with him yeah? If I go in they’ll kind of do the honourable thing and bow down, you know, they back off, they don’t really want to fight, they want a way to get out of it a lot of the time. (C12)

The ambiguities raised by women employed to manage and control violent behaviours is further developed by Shelley, who, typical of many of the women we interviewed, emphasised the importance of being viewed by male and female colleagues as being capable of ‘getting stuck in’, and managing violent confrontations with assertiveness and physical skill: For me to have the respect I’ve got off the men I’ve got to sort of become one of them, rather than think, ‘I’m a women and I’ve got my rights and I’m gonna’ stand up for my rights and you’ll do what needs to be done to accept me’, that’s not the way I’ve took the role of my job on, I’ve went in there and went right, for me to be able to do this job properly I’ve got to become one of the lads. (C10)

168

K. O’Brien et al.

Shelley is married to a doorman and works with a seventeen strong door team in a mainstream, beach-party themed nightclub popular with locals aged 18–24 years. Like many of the women we interviewed, Shelley attaches status and meaning to being labelled as tough, confident and a competent violence expert. In fact, a consistent message expressed by those women employed as a member of a large door team, is that they have to work hard to earn this label and this usually involves proving to male colleagues ‘on the job’, their suitability in terms of temperament, and capability in terms of physicality. A number of women commented on how they achieved the status of ‘one of the lads’ (Miller, 2001) after a specific violent episode, often involving men, and being recognised by colleagues as being the first ‘man’ on the scene. In a sense, therefore, the hyper masculine environments in which these women occupy nurtures the construction of a range of ‘alternative’ femininities (Joe-Laidler & Hunt, 2001; Messerschmidt, 1997; Miller, 1998, 2001) and these femininities can in some cases encompass ‘gender crossing’ (Lovell 2000:14; Miller, 2002:443). For example, a significant number of the women in this study refer to themselves as ‘doormen’ and place a high value on being accepted as ’one of the lads’ by their male colleagues. In the following extract, Lesley, who works in an Australian themed pub in a provincial northern city in Britain, elaborates on these points further in recounting one of her first nights ‘on the job’: We had a really big fight last weekend. . . There were five, six of them and Theo, one of the door lads. . . he’s about 7ft 7’’, well he got bottled in the back of his head, he was on the floor, there were like four of them on to him. I had got one of them, his hands behind his back (shows me) and I got him up eventually and we had a bit of a shuffle and then one doorman had one side and I had the other side of this lad and then I took him down on my own. So I got one out and then I came back and like everyone was everywhere, there was blood everywhere. But I think it’s because I’d got right into the middle of it and didn’t run away. That’s why I’m there, that’s what I am paid to do in my eyes, I’m not there to run away. . . . It wasn’t until that big fight that they really took me seriously. (N2)

Although the motivations of those ultimately awarding status may vary, violence based rituals of induction are commonly experienced by both male and female rookie bouncers (Hobbs et al., 2003; Winlow, 2001). We have found that although women can ‘handle violence’, and importantly, this also encompasses the successful deployment of diffusing techniques, the style in which women enact and perform ‘violence work’ is highly gendered. In discussions elsewhere (Hobbs et al., 2007) we elaborate on this argument in depth in our examination of the resources some women draw upon and the techniques they employ to perform the role of ‘violence worker’. In that work, we focus on a particular group of women who, like their male counterparts (Hobbs et al., 2003), draw on their working class cultural capital (Bourdieu, 1977) in order to ‘work the doors’. These women have experiential knowledge of violence and violent cultures and are concentrated in venues with ‘violent reputations’ or, in the words many of them, work on ‘aggressive doors’ including in particular,

Negotiating Violence and Gender

169

nightclub venues located in provincial British towns and cities. As the following discussion will highlight, the style in which ‘violence work’ is performed by these women extends beyond the diffusion strategy, towards a more instrumental form of violence management. Gail, who describes her upbringing as, traditional working class, I grew up on a rough housing scheme (C8), works in a 2000 person capacity nightclub that is marketed at a young, low-income, predominantly drug taking (ecstasy) crowd and in the extract that follows, she offers a useful example to illustrate that some of the female bouncers in our study are not simply providing a diffusive function, rather, they are frequently required to engage in violent encounters with men and women and hence, ‘‘transforming and replicating traditional door work with its accompanying violent core’’ (Hobbs et al., 2007): A lot of people are under the impression that a guy is not going to turn around and hit a woman. . .but that’s not the case. If you get in the way of half the guys here (club) then they’ll just smack you back out of the way unless you’re prepared to defend your self in some way. The guys don’t care whether they’re hitting a guy or a girl at the end of the day. . .There was one night. . .there was twenty two door staff on and every one of us walked away with some injury or another. The red code had been called to the dance floor and we all ran in and as I got there a boy was on the floor who’d had part of his ear bitten off so I went to pick him up. The boy was in shock, he didn’t know what was going on. I’ve picked him up and went to pick up the bit of ear in the hope that we could may be do something. . .I turned around and there’s this big guy behind me who worked for us. . .I was like, ‘What you doing Lee?’, ‘Get out of my way’. I hate this guy. Lee is massive, he’s body-building material and this wee guy, I mean I’m not the biggest person in the world but I could still pick him up. Lee went to hit the boy- it was Lee who had bit his ear off. He didn’t mean to hit me that time but he hit me over the back of the head and knocked me out. I must have been out for a couple of seconds and then I came back around. Lee is laying into the boy, I’ve got the boy, I’ve pushed Lee out of the way and was shouting at him ‘What you doing? ‘Get out of my way’, and all this and Lee was like, ‘Well I didn’t mean to hit you that time but you’re in my way this time’, and smacked me in the back of the head again so that was me out cold again. (C8)

It is clear from Gail’s account that this is a type of venue where the management of violence and aggressive masculinity represents a routine task requiring a response-led approach. This means that the style in which she engages in ‘violence work’ is not defined by her status as a woman, rather, as a member of the door team closest in proximity to an aggressive or violent incident. Moreover, for women working in similar venues to Gail, dealing with violent and aggressive customers can lead to assault and injury. A version of ’masculine’ cultural capital is clearly an occupational requisite for women like Gail, who in a similar way to her male counterparts, views violence as an intrinsic part of her occupational identity. Female bouncers like Gail, can be viewed as ‘gender crossing’ (Lovell 2000; Miller 2002) when they engage in violence work because their central role violates conventions of femininity, by occasionally ’replicating male excesses’ (Klein 1993:183). For these women, violence, and perhaps more significantly, knowledge of violence

170

K. O’Brien et al.

as an everyday ingredient of their cultural capital, distinguishes these women from others we have included in this chapter.

Discussion and Conclusions The introduction of women into the hyper masculine environment that organises bouncing simply reworks traditional arrangements and performances of gender. Female bouncers are seen as desirable for various commercial, regulatory and legal reasons. Their bodies, and assumptions about their emotional capacities, provide the industry with ‘a resource’ that is suited to the demands of night time leisure economies in Britain, and in particular, attempts to re-brand violent night-strips as safe. Within the security milieu of the night-time economy gender is clearly a mechanism that is situated social action and contributes to the reproduction of social structure (West & Fenstermaker 1995:21). The spectrum of ‘violence work’ represents a profoundly gendered form of social action. Female bouncers are ’doing gender’ (West & Fenstermaker 1995; West & Zimmerman 1987) when they negotiate ways to diffuse, avoid or engage in violent and aggressive incidents. The traditional masculine environment that female bouncers have partially colonised, demands the construction of a range of ‘alternative’ femininities (Messerschmidt 1997; Miller 1998, 2001). The demands of the night time economy trades on and draws upon the willingness of women to transgress traditional gendered boundaries and at the same time simultaneously refashions them (Adkins, 2004). Female bouncers, like female boxers for example (Halbert, 1997; Menneson, 2000), occupy an exceedingly ambivalent position. On the one hand they would appear to dismantle conventional arrangements of gender, yet via the specific roles that the majority of female bouncers are tasked with, for instance, searching and laying their hands upon other women and performing first aid tasks in venues, they are reinforcing traditional ‘feminine practices’ and are actualizing and reproducing traditional gendered identities and expectations. This is further reinforced with respect to ‘violence work’. Women are sought after by some venues to perform the gender specialist role and deal specifically with violent female customers, yet in practice, some women are also dealing with confrontations and violent episodes involving men. As we have shown in this chapter, women are in fact reinforcing gendered codes that underpin ‘violence work’ when they practice diffusing techniques to deal with violent male customers in licensed venues. For those women who occasionally embrace violent action with enthusiasm as part of a mixed sex door team, traditional arrangements of gender are not being resisted, rather, they are being reworked and are transforming a feminised version of cultural capital into economic capital.

Negotiating Violence and Gender

171

References Adkins, L. (2004). ‘Reflexivity: Freedom or Habit of Gender?’ Sociological Review, 52 (2) 191–210. Batchelor, S. (2001). ’The Myth of Girl Gangs’, Criminal Justice Matters, 43: 26–27. Binnie, J & Skeggs, B. (2004) ‘Cosmopolitan Knowledge and the Production and Consumption of Sexualised Space: Manchester’s Gay Village’, Sociological Review, 52 (1) 39–61. Bourdieu, P. (1977). Outline of a Theory of Practice, Cambridge: Cambridge University Press. Braithwaite, J. & Daly, K. (1994). ‘Masculinities, Violence and Communitarian Control’, in T. Newburn & E.A. Stanko (eds.). Just Boys Doing Business? Men, Masculinities and Crime, London: Routledge. Campbell, A. (1993). Men, Women and Aggression, New York: Basic Books. Casey, M. (2004). ‘De-Dyking Queer Space: Female Heterosexual Access and Visibility within Lesbian and Gay Space’. Sexualities, 7 (4) 446–461. Chatterton, P. & Hollands, R. (2003). Urban Nightscapes: Youth Cultures, Pleasure Spaces and Corporate Power, London: Routledge. DCMS (Department of Culture, Media and Sport) (2004). Guidance Issued Under the Section 182 of the Licensing Act 2003 and Guidance to Police Officers on the Operation of Closure Powers in Part 8 of the Licensing Act 2003, London: DCMS. Erickson, B. H., Albanese, P. & Drakulie. S. (2000). ‘Gender on a Jagged Edge: The Security Industry, Its Clients, and the Reproduction and Revision of Gender’, Work and Occupations, 27 (3) 294–318. Gilligan, C. (1982). In a Different Voice: Psychological Theory and Women’s Development, Cambridge: Harvard University Press. Gough, B. & Edwards, G. (1998). ‘The Beer Talking: Four Lads, a Carry Out and the Reproduction of Masculinities’, Sociological Review, 46 (3) 409–435. Graham. K. & Wells, S. (2003). ‘Somebody’s Gonna Get Their Head Kicked in Tonight: Aggression Among Young Males in Bars – A Question of Values’ British Journal of Criminology 43 (3) 546–566. Guy, M. E. & Newman, M. A. (2004). ‘Women’s Jobs, Men’s Jobs: Sex Segregation and Emotional Labour’ Public Administration Review, 64 (3) 289–298. Hadfield, P. (2006). Bar Wars: Contesting the Night in Contemporary British Cities, Oxford: Oxford University Press. Halbert, C. (1997). ‘Tough Enough and Women Enough: Stereotypes, Discrimination, and Impression Management Among Women Professional Boxers’, Journal of Sport and Social Issues, 21 (1) 7–36. Hey, V. (1986). Patriarchy and Pub Culture, London: Tavistock. Hobbs, D., Hadfield, P., Lister, S., & Winlow, S., (2003). Bouncers: Violence and Governance in the Night-Time Economy. Oxford: Oxford University Press. Hobbs, D., O’Brien, K. & Westmarland, L. (2007). ‘Connecting the Gendered Door: Women, Violence and Doorwork’, British Journal of Sociology, 58 (1) 21–38. Hochschild, A. (1983). The Managed Heart: Commercialization of Human Feeling, Berkeley: UCP. Hunt, G. & Saterlee, S. (1987). ‘Darts, Drink and the Pub: The Culture of Female Drinking’, Sociological Review, 35: 575–601. Hutchinson, I. L., Magennis, P., Shepherd, J. P. & Brown, A. E. (1998). ‘B.A.O.M.S United Kingdom Survey of Facial Injuries, Pt 1: Aetiology and the Association with Alcohol Consumption,’ British Journal of Maxillofacial Surgery, 36: 3–13. Joe-Laidler, K., & Hunt, G. (2001). ‘Accomplishing Femininity Among the Girls in the Gang’, British Journal of Criminology, 41: 656–678. Kirkby, D. (1997). Barmaids: A History of Women’s Work in Pubs, Cambridge: Cambridge University Press.

172

K. O’Brien et al.

Klein, A. (1993). Little Big Men: Body Building. Subculture and Gender Construction, New York: State University Press. Lister, S. (2002). ‘Violence as a Commercial Resource: Situating Bouncers and the Use of Force in Context’, Journal of Forensic Psychiatry, 12/2: 245–249. Lister, S., Hadfield, P., Hobbs, D. & Winlow, S. (2001). ‘Accounting for Bouncers: Occupational Licensing as a Mechanism for Regulation’, Criminal Justice, 1/4: 363–384. Lovell, T. (2000). ‘Thinking Feminism with and Against Bourdieu’, Feminist Theory, 1(1): 11–32. Maguire, M., Morgan, R., & Nettleton, H. (2003). Evaluation of the Targeting of Alcoholrelated Street Crime (TASC). London: HMSO. Marsh, B (2004). ‘The Ladette Takeover’, Daily Mail, 19th January, 2004. McCaughey, M. (1997). Real Knockouts: The Physical Feminism of Women’s Self-defence, New York: New York University Press. Menneson, (2000). ‘Hard Women and Soft Women: The Social Construction of Identities Among Female Boxers’, International Review for the Sociology of Sport, 35 (1): 21–33. Messerschmidt, J. W. (1997). Crime as Structured Action: Gender, Race, Class, and Crime in the Making, London: Sage. Miller, J (1998). ‘Up it Up: Gender and the Accomplishment of Street Robbery’, Criminology 36 (1) 37–66. Miller, J. (2001). One of the Guys: Girls, Gangs and Gender, New York: Oxford University Press. Miller, J. (2002). ‘‘The Strengths and Limits of ‘Doing Gender’ for Understanding Street Crime’’, Theoretical Criminology, 6(4) 433–60. Monaghan, L. (2002a). ‘Embodying Gender, Work and Organisation: Solidarity, Cool Loyalties and Contested Hierarchy in a Masculinist Occupation’, Gender, Work and Organisation, 9 (5) 504–536. Monaghan, L. (2002b). ‘Regulating Unruly Bodies: Work Tasks, Conflict and Violence in Britain’s Night Time Economy’, The British Journal of Sociology, 53 (3) 403–429. Monaghan, L. (2002c). ‘Hard Men, Shop Boys and Others: Embodying Competence in a Masculinist Occupation’, The Sociological Review, 50 (3) 334–355. Monaghan, L. (2003). ‘Danger on the Doors: Bodily Risk in a Demonised Occupation’, Health, Risk and Society, 5 (1) 11–31. Munt, S. (2001). ‘The Butch Body’, in R. Holliday & J. Hassard (eds.). Contested Bodies, London: Routledge. O’Brien, K. (forthcoming). Dealing Tac: Young People, Gender and Neighbourhood Drug Markets, Cullomptan: Willan. Portillos, E. L. (1999). ‘Women, Men and Gangs: The Social Construction of Gender’, in M. Chesney-Lind & J. M. Hagedorn (eds.) Female Gangs in America: Essays on Girls, Gangs and Gender, Chicago: Lakeview Press. Robins, D., & Cohen, P. (1978). Knuckle Sandwich: Growing up in the Working Class City, London: Penguin Books. Sabic, I., Walker, A., McNamara, J., Smith, C. & Kolvin, P. (2004). ‘Door Supervision’, in Licensed Premises: Law and Practice, Haywards Heath: Tottel Publishing. Skeggs, B. (1997). Formations of Class and Gender, London: Sage. Steinberg, R. J. & Figart, D. M. (1999). ‘Emotional Labour Since the Managed Heart’ Annals, 561: 8–26. Tomsen, S. (1997). ‘A Top Night: Social Protest, Masculinity and the Culture of Drinking Violence’, British Journal of Criminology, 37 (1) 90–102. Wakefield, A. (2003). Selling Security. The Private Policing of Public Space, Cullompton: Willan. Walker, A. (1999). The Safer Doors Project, London: Home Office Policing and Reducing Crime Unit. Walker, A., Maher, J., Coulthard, M., Goddard, E. & Thomas, M. (2002). Living in Britain: Results from the 2001/2002 General Household Survey, London: The Stationary Office.

Negotiating Violence and Gender

173

Walklate, S. (2004). Gender, Crime and Criminal Justice, Cullompton: Willan. West, C. & Fenstermaker, S. (1995). ‘Doing Difference’, Gender and Society, 9: 8–37. West, C. & Zimmerman, D. H. (1987). ‘Doing Gender’, Gender and Society, 1 (2) 125–151. Westmarland, L. (2001). Gender and Policing: Sex Power and Police Culture, Cullompton: Willan. Winlow, S. (2001). Badfellas: Crime, Tradition and New Masculinities, Oxford: Berg. Winlow, S., & Hall, S. (2006). Violent Night, Oxford: Berg. Yllo, K. A. (1993). ‘Through a Feminist Lens: Gender, Power and Violence,’ in R. J. Gelles and D. R. Loseke (eds.) Current Controversies on Family Violence, Newbury Par, CA: Sage. Yount, K.R. (1991). Ladies, Flirts, and Tomboys: Strategies for Managing Sexual Harassment in an Underground Coal Mine’, Journal of Contemporary Ethnography, 19 (4) 396–422.

Part V

Politics, War and Violence

Violence and War: Measuring Homicide in Belgium (1900–1950) Xavier Rousseaux, Fre´de´ric Vesentini and Antoon Vrints

Introduction Ever since the appearance of Ted Gurr’s famous graph presenting homicide rates for 100,000 inhabitants of England, from the 13th to the 20th century, many historians and social scientists have been fascinated by the multi-secular trend described by its curve. Initially focused on English data, the debate spread around the Western World (from Finland to Amsterdam, and to the United States). Pieter Spierenburg gave the first synthetic interpretation of this trend, based on Amsterdam archival material (Spierenburg, 1994, pp. 701–716). Manuel Eisner has recently published a complex interpretation of the data and the theories built around it (Eisner, 2003, pp. 83–142). The first results of the research in progress we present here is focused on homicide in Belgium during the first half of the 20th century. Why focus on this half century? For two reasons. Firstly, the existence of a relatively sophisticated statistical tool, reformed in 1897 and giving series of data on the perpetrators of homicide (judicial statistics) as well as the victims of violent death (medical statistics). The second reason is particular to Belgium’s history from 1900 to 1950. The country faced two state of war periods: 1914–1919 and 1940–1945, and two military occupations by Germany (August 1914–November 1918 -52 months-, and May 1940–September 1944 -48 months-), in the context of two World Wars. It would appear essential to examine how everyday violence developed in those war periods, and especially in what specific form(s). In this contribution, we will focus on the question of measuring the frequency of homicide and the perturbation brought on by war. Actually, most studies on homicide scrupulously avoid mixing ‘‘extraordinary’’ violence issuing from military conflicts and the ‘‘ordinary’’ violence taking place in local societies. Founded on false theoretical premises, the distinction between ‘‘civilized’’ societies in peacetime and ‘‘brutalized’’ societies at war has been largely Xavier Rousseaux Senior Research Associate FNRS, Fre´de´ric Vesentini, Postdoctoral Researcher FNRS, Center for Law and Justice History, Universite´ catholique de Louvain (UCL), Louvain-la-Neuve, Antoon Vrints, Postdoctoral Researcher BOF, Universiteit Gent

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

177

178

X. Rousseaux et al.

challenged by the last 20th century European wars, in Bosnia and Yugoslavia, as well as the latest historical accounts of 20th century world wars (Mazower, 1998) (Majerus & Vrints, 2005, pp. 52–74). The occupation context, where the military confrontation remains on the margin or outside the occupied territory, might seem to justify such a separation. On the contrary, the occupation context invites us to evaluate the impact of war on violence, in other words, the impact of the culture of war on the local cultures of violence. This is why we present here a threefold contribution on the quantitative measurement of homicide during the two world wars and the interbellum periods – placed in long-term perspectives.

Homicide in the XXth Century: A Long Decline and Recent Upsurges Let us summarize the long-term debate on lethal violence in Eisner’s terms, in three main sets of questions. –Whether counting homicides documented in pre-modern records and comparing respective rates over eight centuries is scientifically sound (Eisner, 2003, p. 86). –To what extent can the English case be generalized to Western society and linked to specific behaviours, times, areas or social groups? –Why did this multi-secular decline occur? The first set of questions combine methodological issues on the reliability of data and epistemological issues on the socially constructed or substantive character of interpersonal aggression. The second set of questions tries to specify the characteristics of the multi-secular decline. The third set of questions might be summarized around a comparison of the findings of crime historians and Norbert Elias’ theory on ‘‘The Civilizing Process’’ (Eisner, p. 87). What are the main results in Eisner’s view? Let us focus on the first two issues for the moment.

Methodological Issues on the Reliability of Data and Epistemological Issues on the Socially Constructed or Substantive Character of Interpersonal Aggression The evaluation of data reliability is generally positive in terms of homicide rates per 100,000 inhabitants (Eisner, pp. 92–98). On the basis of ninety publications on pre-modern homicide rates in Europe as well as based on a comprehensive collection of modern homicide time series in ten countries (Eisner, p. 88) (Eisner confirms ‘‘the Europe-wide massive drop – roughly by a factor of 10:1–50:1 over the period from the fifteenth to the twentieth century – in lethal interpersonal violence, first observed by Gurr on the basis of English data (1981’’). By the 19th

Violence and War

179

century, therefore, homicide rates were lowest in the modernized, affluent, and urbanized regions of Europe, which were surrounded by a belt of high homicide rates in the periphery. By around 1950, most European countries experienced their lowest historically known levels of homicide rates. Briefly, we observe a relative homogeneity of high homicide rates, 6–100 /100,000 between 1300 and 1500; and more geographical discrepancies since the end of the 16th c. For the Netherlands and Belgium, following Boomgaard, Spierenburg and Vanhemelryck, homicide rates in Amsterdam may have declined from 40 to 20 in Amsterdam during the 16th c. and from 20 to 10 for Brussels between 1400 and 1599. By the end of the 17th century, the pattern becomes even clearer with 4 /100,000 in Amsterdam, 4–5 in Brussels (Eisner, p. 101). The statistical material, developed in many European countries for the beginning of the 19th century, provides three main sources of information on a U-Shape trend extending between 1800–2000. Firstly, between 1880 and 1950, the frequency of lethal violence fell by at least 50% in northern European countries, and considerably further in the southern countries, stabilizing around 0.4–0.5 in 1950. These declining trends were the predominating pattern for other types of violence (serious assault, robberies) as well as for property crime. Secondly, cross-national differences within Western Europe had become inconsequential. Thirdly, the data from 1950 until the early 1990s point to an upsurge in homicide rates throughout most of Europe, accompanied by a much sharper rise in recorded levels of assault and robbery. These three main points raise interesting questions about the possible impact of war violence in Western Europe on the declining trend between 1914 and 1950.

Characteristics of the Multi-Secular Decline The answers to the second set of questions on the context of these declining trajectories are more complex. Historians of crime have pointed out the importance of contextual change including, for example, the cultural meaning of violence, the typical situations giving rise to conflict and aggression, the characteristics of offenders and victims, and the framework of legal and judicial reactions. Similarly, criminologists have become increasingly interested in disaggregating violence trends, as, for example, offender-victim configurations, offender age groups, or weapons used in the offence. . . In his detailed analysis of all the data given on sexratios, age and social status of offenders, sex of the victims and degree of personal relationship between offender and victim, Eisner delivers three majors conclusions: First, the overall decline in homicide rates regularly appears to coincide with a decline in the proportion of male-to-male killings. In a similar vein, the drop appears to be inversely related to a (relative) increase in family homicide. Finally, evidence suggests that the overall drop in homicide rates may have been accompanied by a gradual withdrawal of elites from interpersonal violence (Eisner, p. 108).

180

X. Rousseaux et al.

What about the Belgian case? Eisner situated the Low Countries, from the middle ages to the 20th century, as one of the geographical areas strongly resembling the English model of declining violence. Given the precision of the sources, the density of the population and the precocity of autonomous political structures (with the development of cities from the 13th century on), the longterm decline in homicide rates is firmly supported by very reliable data. As of the early 19th century, Belgium offers another specific development in a particularly rich area of statistical material.

Sources for Measuring Violent Death in Belgium On the one hand, the judicial statistics were developed within a twofold perspective, symbolized by one man, Adolphe Quetelet, scientist and administrator (Tixhon, 2001). As a scientist, he was persuaded that crime could be analyzed like other physical events. As an administrator, jumping on the bandwagon of the young Belgian State in 1830, he was concerned by the fact that a criminal policy could influence variations in crime. These two perspectives guided his activity as statistician, in the development of judicial statistics as precise as possible. On the other hand, as founding member of the Central Statistical Commission, he promoted a scientific apparatus of demographic statistics for the new-born State. From the 1840s on, Belgium developed a demographics founded on decennial population censuses and local population registers, counting births, marriages and deaths, as well as emigrants and immigrants in each municipality of the Kingdom. In parallel, death was registered locally, giving rise to a series of medical statistics on causes of death.

Judicial Statistics: Counting the Perpetrators In 1897, the situation of Belgian judicial statistics might be described as deeply in crisis. Under the impulse of Alphonse Delannoy, a statistician at the Belgian Ministry of Justice, an important reform was launched. From 1899–1901, the new statistics were elaborated in two parts. The first part consisted of a statistics of judicial administration activity (prosecution and jurisdiction), from public prosecution to verdict. The second part is a criminal statistics, focused on contextual variables for a ‘‘major’’ part of those convicted. The Belgian Judicial System, Statistical Material, and Homicide In order to understand the nature of the data, it import to give a summary of the judicial organization established in Belgium in 1830, based on the French revolutionary & Napoleonic system symbolized by the Code pe´nal of 1810.

Violence and War

181

The first question is the role of the system of justice administration in categorizing homicide. Published judicial statistics focused only on perpetrators of lethal violence, while medical statistics focused only on victims of violent death.1 –Judicial activity statistics (in short court statistics), count those accused of lethal violence prosecuted before criminal, or assize, courts (cours d’assises/assisenhoven) for crimes and those prosecuted before misdemeanour courts (tribunaux correctionnels/correctionele rechtkanken) for ‘‘correctionalised’’ crimes or misdemeanours. These statistics never mentioned the nature of the cases remaining unprosecuted at the prosecution level (sans suite/zonder gevolg), or dismissed (non lieu a` poursuivre /buitenvervolgingstelling) at the examining chamber level (Chambre du conseil/raadkamer). –Criminal statistics collect data on people convicted for lethal violence and sentenced to criminal or ‘‘correctional’’ penalties by criminal or misdemeanours courts. Criminal statistics don’t give information about people acquitted but sometimes add a table for the unprosecuted homicide cases, termed ‘‘author unknown’’. These two series of statistics are based on the Criminal code, which in Belgium is the 1867 (Code pe´nal/Strafwetboek), largely inspired by the former Napoleonic Code pe´nal of 1810.

Legal-Institutional Classifications of Homicide The second question addresses the impact of the Criminal code on definitions of homicide in Belgium. Statistical registration of homicide depends not only on criminal definition but also on institutional treatment. –The Articles 393–419 of the Criminal Code (1867) defined lethal violence. The Code distinguishes three levels of gravity: Homicide is every form of violence causing death Murder is voluntarily committed homicide (art.393) Assassination is murder committed with premeditation (art.394). Some forms of murders are specified: parricide (art.395), infanticide (art.396) and poisoning (art.397). These types of murders are punished by the death penalty or perpetuity at forced labour. Nevertheless, since 1867, the death penalty has been 1

Actually, only the local prosecution registers (registres aux notices du parquet/notitiesregisters van het parket) mentioned all the suspected deaths examined by the public prosecutor acting as a sort of coroner. Unfortunately, published prosecution statistics don’t indicate the nature of the cases, except for unknown authors.

182

X. Rousseaux et al.

systematically replaced by perpetuity at forced labour. The only exception is a war case in Belgian controlled territory during World War I. On March 26th 1918, Quartermaster Emile Verfaille died on the guillotine in the prison of Veurne, convicted of having killed his fiance´e. It was to be the last execution according to the ordinary penal code in Belgium (Deneckere, 2000, pp. 75–83). Another form of murder is called voluntary serious assault without intention to inflict death (art.401 x1–2) (coups porte´s volontairement ayant intente´ la mort sans intention de la donner). These ‘‘lethal assaults’’ are punished by a criminal penalty of imprisonment. Two interesting restrictions on these punishments are the following: –Involuntary homicide is punished by a simple prison term (emprisonnement) (art.419). –Homicide responding to a provocation is excusable (art.411), while homicide committed in a state of legitimate defence is neither a crime nor a misdemeanour. –From the institutional point of view, another classification system distinguished the qualified homicide (homicide qualifie´) (assassination and murder) which is a crime punished by a criminal court from the simple homicide (homicide involontaire ou de provocation) which is a misdemeanour punished by a misdemeanour court (tribunal correctionnel). This system of prosecuting and punishing perpetrators of homicide evolved during the 19th century. ‘‘Correctionalisation’’ Laws (1832, 1849), transferred certain homicides from criminal courts to correctional courts (without popular juries).

Table 1 Belgian code and court (non-exhaustive) classification of homicide Penal category

Qualification

Crimes

Assassination Infanticide Parricide Poisoning Murder Murder to favorise theft Voluntary serious assault without intention to deliver death Murder Infanticide Poisoning Voluntary serious assault without intention to deliver death Involuntary Homicide Homicide after provocation

Correctionalized Crimes

Misdemeanours

Law article 394 396 395 397 393 475 401 x 1–2 393 396 397 401 x 1–2 419 411

Violence and War

183

At the beginning of the 20th century, the statistical account of homicide is structured by the criminal code and courts around three categories: crime, correctionalized crime and misdemeanour (Table 1). In summary, court statistics and criminal statistics count the perpetrators of homicide differently: the presumed perpetrators prosecuted by correctional or criminal courts for various forms of homicide on the one hand, and the individuals convicted of crimes and sentenced to ‘‘correctional’’ penalties on the other. In any case, judicial statistics cannot establish the link between recorded homicides and perpetrators of homicide for two main reasons: firstly, because the data from the public prosecutors do not discriminate the nature of the crime; secondly because the justice system never established a statistical relationship between the number of cases and the number of presumed authors.2

Medical Statistics: Counting the Victims The focus of the second category of sources is completely different. Instead of counting the overall number of prosecuted homicides or the alleged perpetrators of homicide, the statistics of causes of death take into account the corpse. Since the second half of the 19th century, Belgian communal authorities have had to register the causes of death of their inhabitants (for the Belgian medical statistics, see (Velle, 1985, pp. 213–242). The data were eventually processed by the central administration in Brussels in national medical statistics. By preference, the cause of death was established by a physician who filed a uniform certificate on the identity of the dead and the precise circumstances of death. In practice, especially in the countryside, registration of causes of death was deficient until the second half of the 20th century. Due to the possible judicial and legal implications, the registration of violent death was of far better quality than the statistics of causes of death overall. Each case of violent death had to be examined by a recognized judiciary physician (coroner) in order to send an official medical report to the Procureur des Konings/Procureur du Roi. Four different descriptions were at the disposal of the physician categorizing cases of violent death. Beyond the thin legal framework, the physician’s personal professional evaluation determined the official categorization of the cause of violent death. The first category concerns homicides, violent death having resulted from direct interpersonal violence with or without the intention to kill. In practice, a wide range of causes of violent death could be labelled as homicides: ranging from the victim of a well prepared murder case to a pub brawl that got out of hand. The second category contains all the different forms of suicide. Accidents (mainly mortal traffic and industrial accidents) were classified in the third category. For those cases of violent death with an unclear 2

Except for certain periods when information has been given on homicides committed by unknown authors.

184

X. Rousseaux et al.

cause of death, the residual category of dubious cases (cas douteux) was created. Out of concern to ensure the long term comparability of the data, specific war linked causes of violent death (deaths of soldiers and civilians due to military operations and bombings, military fusillades) were excluded from the four existing categories. A new category (war linked causes of violent death) was created to include those deaths. Medical statistics are undoubtedly a more adequate tool than judiciary statistics for establishing the incidence of homicide in a particular society and making assessments of the long-term evolution of levels of violence. For various reasons, they offer ‘‘harder’’ evidence than judiciary statistics. The main element in this regard is the fact that they directly reflect the number of people dying a violent death. Of course, like all sources, medical statistics have their shortcomings: one might well imagine the impact of social pressure on the physician’s report. It is, for example, probable that in a number of cases suicide was registered as an accident to prevent blame. Murderers, on the other hand, might successfully camouflage homicides as accidents.

Homicide Rates in Belgium (1901–1950) General Patterns Medical Statistics We begin this general survey with the ‘‘series of homicides’’ contained in the national medical statistics (Fig. 1). As mentioned earlier, it is clear that assigning 100%

80%

60%

40%

20%

Homicides

Dubious cases

Suicides

Accidents

Fig. 1 Distribution of causes of homicide in Belgium (medical statistics)

1949

1947

1945

1943

1941

1939

1937

1935

1933

1931

1929

1927

1925

1923

1921

1919

1917

1915

1913

1911

1909

1907

1905

1903

1901

0%

Violence and War

185

bodies to one of the categorizations defined by the statistics is a complex process, combining scientific expertise, social constraints as well as an element of chance, due to factors such as the conservation of bodies and clues. . . So between homicides camouflaged as suicides or suicides publicly proclaimed as ‘‘accidents’’, the manipulation of categories might appear to be rather subtle. Yet a number of elements are of a reassuring nature as regards the utilization of these categories by forensic medicine. First of all, the presence among the four categories of a series inventorying ‘‘dubious cases’’. To the extent that this classification was used by doctors, we might be led to think that the ‘‘borderline cases’’, on which medical expertise alone was unable to pronounce itself definitively, have not been assigned to one category or the other for reasons of time, persons or environment. A second reassuring element as to the use of these categories is their relative constancy. Over a half century of utilization, implying a succession of doctors, in locations sometimes far removed and in contexts oh so different, the distribution between the proportions seems about the same, with perhaps the exception of the years 1943 and 1944 which, as we shall see, are altogether particular. We note in passing the predominance of accidents, often representing over 60% of violent deaths and the rather high number of suicides, in and around 30%. The remaining 10% being equitably shared between homicides and dubious cases. In absolute numbers, the impression emerging is also one of a general constancy, with, nonetheless, an initial increase right after the wars and, especially, an explosion in the years 1943 and 1944. As for the dubious cases, they are rather constant in the long run, with the exception of 1941, when the beginning of the occupation is certainly the source of an increasing use of this ‘‘catch all’’ category, the doctors turning the determination of death over to the public prosecutor’s offices. And the homicide rate in all of this? (Fig. 2) If we refer to the categories of the medical statistics, we must first of all consider what homicide means. They are observed, violent deaths, whose clues tend to prove that they are neither accidents nor suicides, in other words that they are the fruit of a third party, voluntary intervention. If we refer these figures to the movement of the population, we generally find fewer than 3 homicides for 100,000 inhabitants. If we add the dubious cases, we are still below 5 homicides for 100,000 inhabitants, except of course for the two postwar periods and especially the years 1943 and 1944. If we refer to the ‘‘atlas of crime’’ Chesnais presented in his Histoire de la violence (Chesnais, 1981, p. 56) (we note that these rates place Belgium in a rather central position in relation to other Europe countries of the time. We note, nonetheless, that with a rate clearly oscillating between 2 and 4 homicides for 100,000 inhabitants, Belgium presents characteristics closer to Italy or Portugal than to France or England. Within this schema, the war and postwar periods present singularities (Fig. 3). Thus, after the first world war, we observe a rather pronounced peak. Due to the implosion of the central state administration during the first German occupation, regrettably no homicide data are available for the 1914–1918 period. It is thus impossible to indicate the starting point or size of

186

X. Rousseaux et al. 2500 2000 1500 1000

0

1901 1903 1905 1907 1909 1911 1913 1915 1917 1919 1921 1923 1925 1927 1929 1931 1933 1935 1937 1939 1941 1943 1945 1947 1949

500

Homicides

Dubious cases

Fig. 2 Number of homicides and dubious cases in Belgium (medical statistics) Note: The data for 1914–1918 and 1947 are not available.

30 25 20 15 10 5

Homicide rate

1949

1946

1943

1940

1937

1934

1931

1928

1925

1922

1919

1916

1913

1910

1907

1904

1901

0

Homicide rate (dubious cases included)

Fig. 3 Number of homicides for 100,000 inhabitants in Belgium (medical statistics)

a possible rise in the number of homicides during the occupation years. Relying on 1919 and 1920s high homicide rates, it is nevertheless very probable that such a rise had already begun during the war. Contemporary, qualitative sources suggest very clearly that this must have been the case.3 In all available accounts, banditry on a large scale is mentioned to explain the rise in homicide

3

Cf. Antoon Vrints’s forthcoming contribution on crime and protest in Belgium during the First World War.

Violence and War

187

numbers. The massive re-emergence of Ancien Regime style banditry is due to the specific configuration of war time Belgian society. In the view of the Dutch historian Florike Egmond, two societal conditions have to be taken into consideration in explaining the presence of banditry (Egmond, 1986, p. 86). She establishes a link between massive impoverishment and banditry but, at the same time, she warns against economic reductionism. Massive poverty as such doesn’t necessarily give birth to banditry. In Egmont’s opinion, another condition has to be encountered: a high degree of disintegration of state power and authority. Both of these basic conditions were present in Belgium during the First World War. On the one hand, state authority was dramatically weakened during the occupation years. The central government had left the country and the central administration in Brussels stayed behind – in disorder. The state’s repressive capacity was weakened by the entry of the entire Gendarmerie into the Belgian army behind the Ijzer-front (Geschiedenis van de Rijkswacht, 1980, p. 90; Meyers, 1919; Majerus & Rousseaux, 2004, pp. 50–51). The only operative police during the German occupation were the – often very tiny – communal forces. The state’s capacity to maintain its monopoly on the use of violence was further weakened by a strike of the entire Belgian judiciary system in protest against German policies, as of Spring 1918. In addition, as a result of the presence of different armies on Belgian soil, a vast number of firearms ended up in the hands of the Belgian civilian population. On the other hand, the occupation years were indeed a period of massive impoverishment, especially for the population in the urban and industrial areas (Scholliers, 1988, pp. 151–163; Scholliers & Daelemans, 1988, pp. 139–157; Scholliers, 1994, pp. 39–53). Rising food prices and high unemployment rates undermined purchasing powers and living standards – and dramatically so among urban and industrial populations. The end of the occupation did not alter these conditions at once, as the high homicide rates for the post war years indicate. Material conditions remained harsh for vast segments of the Belgian population during the initial post war years and the Belgian state had great difficulty in reaffirming its monopoly on the use of violence and its overall authority. For the Second World War years, statistics on causes of death are available and they show a dramatic overall rise in homicides in absolute terms, especially for the years 1943 and 1944. France too experienced a similar evolution in the number of homicides in the years 1943 and 1944 (Chesnais, 1981 p. 80). To understand this major peak, however, it is important to differentiate chronologically. During the first two years of the occupation homicide rates were remarkably low. In spite of this apparent calm, an important process of polarization was taking place, one which lay the groundworks for the later escalation in political violence.4 Very early on during the first occupation year a major shift occurred in 4

For a general view on political violence during the Second World War and the reaction of Belgian and German authorities in occupied Belgium, see (Dejonghe, 1984, pp. 5–234; Conway, 1993; De Wever, 1995, pp. 580–594; Gotovitch, 1992, pp. 155–194; Van Meerbeeck, Verhoeyen, Van of Vijver, and Van Doorslaer, 1988; Wouters, 2004).

188

X. Rousseaux et al.

public attitudes towards the German occupier. The original relative benevolence was replaced by an overall sentiment of hostility by Spring 1941 (Ge-rard-Libois & Gotovitch,1971). Britain’s enduring resistance and an especially sharp food crisis during the Winter of 1940–1941 quickly discredited the occupier’s fragile authority in the eyes of large sectors of the Belgian population. This anti-German resentment was expressed by growing hostility towards the collaborating minority. A wide pattern of informal social sanctions was used to intimidate collaborators. Because of the real risk of German repression, it mainly involved anonymous practices like window breaking, threatening letters, threats in the resistance press and ritual trashing of homes. Occasionally, the resentment was expressed publicly during relatively violent manifestations in the cities and charivari-like ritual sanctions in the countryside. Nevertheless, politically motivated violence remained limited in scale and gravity. Except for two murders, homicide was not yet committed as a political means. More fundamental than those isolated cases was the increasing social ostracism of collaborators. As moral transgressors, they were cast out of informal networks of sociability: their businesses were boycotted and they were personally avoided. Non-politically motivated violence did not increase during the first two years of occupation either. The low homicide rate for 1941 is a clear indication that the food crisis of the first half of that year did not result in banditry in the countryside to the same extent as its predecessor had during the First World War. This contrast can be partly explained by the fact that the state was able to maintain its monopoly on the use of violence due to the uninterrupted functioning of the Gendarmerie and the judiciary system, at least during the first half of the occupation. During the second half of the occupation, the incidence of homicide skyrocketed. In 1942, the number of registered homicides was twice as high as the year before. In 1943, nearly 700 homicides were committed and in 1944 the spectacular figure of 1,852 homicides was reached. This peak was caused by a complex interplay of various factors, whose respective roles are hard to distinguish. At least a considerable part of the rise can be attributed to the radicalization of political violence during the second half of the occupation. The occupier’s structural violence was increasingly matched by violent acts of resistance. Spring 1942 brought on an escalating cycle of politically motivated eliminations which was to reach its zenith in the Summer of 1944. An important date in this regard was 22 June 1941, signalling the start of the German invasion of the Soviet Union, which released Belgian communists from the ambiguity of self-imposed neutrality. Haunted by repression, the Communist Party replaced their policy of presence by harsh resistance to the occupier. The communists wanted to open a third front in occupied Belgium by organizing guerrilla warfare activities. Sabotage remained the main activity of the partisan cells, but during 1942 the lives of known collaborators were increasingly targeted as well. This initially involved isolated eliminations of tattlers for security reasons, but from April 1942 onwards a systematic terror campaign against collaborators took off. The purpose was not only neutralizing political enemies by killing them, but also frightening their supporters in order to weaken the base of

Violence and War

189

collaboration as a whole. The communist strategy of tension fundamentally aimed at bolstering the population’s resistance by provoking an escalating cycle of violence. German sources for 1942 indicate 67 political murders of collaborators. Taking this number into account, the rise in homicide during that year is largely due to politically motivated violence. In spite of a gradual increase in the number of violent robberies in certain regions, the impact of banditry on homicide was rates still limited (for Limburg see Bouveroux, 1985, pp. 26–28). The spectacular rise in homicide figures in 1943 and 1944 cannot be explained without taking into account the announcement of the imposition on Belgians of forced labour in Germany in October 1942. This highly unpopular measure touched large sectors of the Belgian population, revitalizing harsh memories of the First World War, and thus strengthening hostility towards the occupier. A considerable proportion of the young men called-up preferred a clandestine life over departure to Germany. This choice provided an enormous reserve of potential recruits for resistance organizations. The partisans especially strengthened their ranks in welcoming men who subsequently enabled them to bolster their guerrilla activities. But the imposition of forced labour gave a boost to common-law banditry as well. Men who could not be hidden by the resistance or families often felt obliged to take refuge in woods and relatively scarcely-populated areas where they tried to survive in wandering bands. In this precarious context it is no surprise that some of them resorted to theft and robbery as a survival strategy. Fundamentally, apart from the results of the imposition of forced labour, the two basic conditions for banditry were present during the second half of the occupation. Already in 1940 food scarcity was leading to massive impoverishment, especially amongst working-class populations in industrial and urban areas (Scholliers, 1993, pp. 175–182; Jacquemyns, 1945). But it was only from the second half of the occupation onwards that authorities were no longer willing or able to enforce the state’s monopoly on the use of violence. Out of fear of being drawn into the occupier’s struggle with the resistance, the Belgian judiciary refused to investigate and prosecute most firearms violence. As a result of rising resistance guerrilla activities and common-law banditry, homicide numbers boomed to heights never reached in contemporary Belgian history. Increasingly, collaborationist militia turned to bloody retaliatory actions. By the Summer of 1944 a stage of civil war was reached in some regions. According to a conservative estimate, 850 collaborationists were killed by the resistance and 150 people were killed by collaborationist militia during the entire occupation (Van Doorslaer, 2003, p. 238). To this number should be added 50–100 politically motivated murders of collaborators during the Liberation days as well (Balace, 1995, pp. 75–132). It is worth noting that after 1944 it took two years before homicide rates came down to pre-war levels. The Belgian state apparently had some difficulty in restoring its monopoly on the use of violence after the homicide peak occurring in the second half of the occupation. Due to lack of sources, it is difficult to explain the relatively high homicide figures for the years 1945 and

190

X. Rousseaux et al.

1946. At first sight there are two obvious explanations. Just as after the First World War, it probably took a while before the state was again able to dismantle all robber bands. On the other hand, we should not rule out the possibility that the high rates are at least partly due to wartime accounts being settled afterwards. Of course, seen from another perspective, a completely different interpretation is possible. It might be argued that after the wartime polarization a relatively smooth transition to ‘‘normality’’ took place. As a matter of fact, homicide figures for 1945, the first year after the Liberation, represent only a small fraction of the figure for 1944, the last year of occupation.

The Figures of Justice In order to consolidate our position and put the figures issuing from the statistics on the causes of mortality into perspective, in Belgium we have an interesting series, within the judicial statistics, at our disposal. This series, allowing us to account for the nature of crimes and misdemeanours, are the statistics of the activity of courts and tribunals, as well as the criminal statistics, which are statistics of convictions. From the outset we must point out that these statistics never envisage the fact of ‘‘homicide’’ as such but rather treat the question from the aspect of a judicial fact, in counting those indicted, convicted and acquitted. Unlike the medical statistics, they therefore no longer count homicides but rather persons involved in acts described as such (homicides) by the legal system. From the critical viewpoint, the difference can appear substantial. Nevertheless, it in no way prevents our exploring the general evolution of these series, aware that the figures for homicide and murderers should, logically, be correlated, in taking into account possible distortions caused by the judicial apparatus and the statistical count. A number of statistical categories include homicides. Modelled on the penal code, they list a multitude of definitions essentially dependent on the circumstances surrounding the acts committed. Relationship between criminal and victim, premeditation, use of lethal substances, etc., are all criteria used in creating the various penal definitions of homicide. Thus, in the Criminal Courts activities, among others, they distinguish: assassinations and premeditated murder (Belgian Penal Code [P.C.], art.394), poisoning (P.C., art.397), murder (P.C., art.393), parricide (P.C., art.395), infanticide (P.C., art.396), voluntary administration of a lethal substance causing death (P.C., art.404), assault and battery causing death with premeditation (P.C., art.401, x2), rape causing death (P.C., art.375–376) or, further, justifiable murders and provoked homicides (P.C., art.411–414). On the contrary, there are fewer categories in correctional jurisdictions. Homicides are represented there as crimes judged in terms of attenuating circumstances. Thus we find murders, infanticides or, again, assault and battery causing death without premeditation. We also find two categories of

Violence and War

191

misdemeanours, much more frequent than the crimes: provoked homicides and, especially, involuntary homicide. We note that homicide cases can be referred to other jurisdictions in Belgium: for example, military courts. The data on these jurisdictions are not always available, except in post war periods, in other words when they are at their busiest. In order to assemble the figures established by the post war judicial apparatus and compare them with medical statistics, it would be interesting to look beyond the natural statistical divisions which separate assize courts, correctional tribunals and military courts. If we were to recombine the statistics according to their penal definitions of crime on the one hand, and misdemeanours on the other, we would place an independent series of movements toward indictment at our disposal, distinguish criminal homicides from ‘‘accidents’’ (involuntary homicides) and thus maybe rediscover a division similar to that in the medical statistics. These judicial statistics trace the paths of cases through the penal justice administrative system. One can thus focus on cases at the stage of prosecution, then indictment, then conviction or acquittal . . . As a general rule, the public prosecutor’s offices’ data are never broken down by type of infraction. Also, in order to be as close as possible to the act and minimize the distortion caused by the penal investigation, we shall concentrate our account on cases of indictment (Fig. 4). The general appearance of this graph would never give us the impression that its data come from radically different sources. The order of the scales tally. Certainly one homicide does not always correspond to one and only one indictment. The act may be collective and bring a number of people before the courts and tribunals or, on the contrary, not result in any prosecution. Nevertheless, if we consider the evolution in the number of those indicted being tried for homicide, we realize that, with the exception of the year 1925, their number is always below or equal to that of homicides observed by doctors; 2500 2000 1500 1000 500

1900 1902 1904 1906 1908 1910 1912 1914 1916 1918 1920 1922 1924 1926 1928 1930 1932 1934 1936 1938 1940 1942 1944 1946 1948 1950

0

Med.Stat. - Homicides

Med.Stat - Dubious cases

Jud.Stat. - Indicted for Homicide as Crime

Jud.Stat. - Homicides known by prosecutor office

Fig. 4 Number of homicides (medical statistics) vs. the number indicted for voluntary homicide (judicial statistics) in Belgium

192

X. Rousseaux et al.

700 600 500 400 300 200

Crimes - Criminal Court

Crimes - Correctional Court

Misdemeanours - Cor. Court

Misdemeanours - Military Court

1948

1945

1942

1939

1936

1933

1930

1927

1924

1921

1918

1915

1912

1909

1906

1903

0

1900

100

Crimes - Military Court

Fig. 5 Inductees for homicide in Belgium, distinguished by courts

which seems to lend at least relevance, if not legitimacy, to this comparison. Moreover, the singularity of 1925 may finally be explained rather simply: as the figure here below shows (Fig. 5), 99 homicides were tried by court-martial that year. That exceptional figure, compared to other post war years, very probably testifies to a desire, seven years after the conflict, to turn the page by settling (if need be by acquittals) all cases in suspense since 1918, to say nothing of the need for prosecuting within the statutes of limitation.5 So, generally speaking, the curves of the medical statistics and the judicial statistics have like tendencies and are sometimes very close: between 100 and 200 cases per year, which referred to a population of 8 million inhabitants gives a rate below 2.5 homicides or indictees for homicide for 100,000 inhabitants. With one noteworthy exception nonetheless: the Second World War. Far from skyrocketing as the statistics on causes of mortality would have it, the curve of indictees for voluntary homicide diminishes rather markedly during the second German occupation. The explanation lies essentially in the functioning of the judicial institution during those years. In fact, if we refer to the very lacunary curve of the cases reaching the public prosecutor’s office, we observe values that practically always exceed the homicides declared by the medical statistics. The unit of account in the two instances being the case, these data are perfectly comparable and one can imagine that some of the cases initially considered ‘‘dubious’’ were registered as homicides by the public prosecutor’s office, thence provoking the opening of a criminal investigation. 5

Another explanation tends to link this peak in the activity of Military Courts with the occupation of the Ruhr and the Rhineland by the Belgian army. Yet the data are not very clear in this respect, as regards knowing whether the figures include acts committed abroad and, if so, whether they include acts committed by German civilians.

Violence and War

193

During the year 1944 alone, cases known to the public prosecutor’s office turn out to be less numerous than the homicides observed with certainty by doctors. Hence fewer investigations than homicides . . . The explanation most probably lies in the treatment of cases by the public prosecutor’s offices. The very high number of dossiers entering the public prosecutor’s office in 1945, and especially in 1946, in comparison to the relatively low number of homicides seems indicate that some of the 1944 cases were held over before being dealt with. That would confirm that the rise in the number of indictees after the war is at least partly due to a catching up with what they were not able/did not wish to deal with during the occupation. As stated earlier, for political reasons the Belgian judiciary was unwilling to prosecute homicide during the occupation. That politically motivated attitude is clearly reflected in the wartime judicial statistics. During the most violent years of Belgian contemporary history the state made no attempt to maintain or restore its monopoly on the use of violence. In addition, the political context of the post war period did not permit an unambiguous a posteriori prosecution of wartime homicides. In the Autumn of 1944 an amnesty measure was proclaimed concerning all crimes committed during the occupation in the context of resistance against the occupier. Indeed the relatively low post war prosecution figures indicate that there was no a posteriori prosecution operation on a massive scale. On the contrary, bloodshed by collaborators did not fall under the amnesty measure. The homicides handled by military courts in the post war years reflect that their share in wartime violence was prosecuted.

A Geographical Pattern of Homicide The geography of homicide is just as interesting. Bolstered by the judicial statistics consolidating the medical data, let’s see how these data are distributed throughout Belgian territory. To do this, the data have been gathered into five periods, relevant as seen through the chronological analysis which has just been carried out: the peacetime periods, that is 1901–1913 and 1922–1939; the post war periods or what one might call the ‘‘return to normalcy’’ or the transitional periods 1919–1922 and 1945–1950 and, finally, the German occupation of 1940–1944. A reminder, the data concerning 1914–1918 are not available. As for inventorying homicides, which are statistically rather scarce infractions, we also had to choose a geographically relevant level of analysis; one allowing us not to lose either the granularity of the results nor the possibility of providing context, while offering significant data on the statistical level. The scale adopted is the province, which in fact has the advantage of being the lowest level of jurisdiction sharing administrative and judicial data6 (Tables 2 and 3). 6

In fact the lower levels (i.e. districts and cantons) differ geographically as to administrative maps and judicial maps.

194

X. Rousseaux et al.

Table 2 Mean number of homicides per year in Belgium 1901–1913 1919–1921 1922–1939

1940–1944

1945–1950

Antwerp 14.5 25.7 17.9 33.2 17.8 Brabant 28.1 46.3 29.9 132.0 31.6 Hainaut 24.2 44.0 30.2 166.0 31.4 West Flanders 15.2 23.7 13.1 33.0 12.4 East Flanders 20.2 45.7 16.6 27.8 15.0 Lie`ge 18.8 28.0 18.0 102.2 22.4 Limburg 4.2 10.7 6.1 46.6 10.8 Luxembourg 4.5 9.0 4.0 29.8 8.2 Namur 8.3 15.7* 7.4 36.2 4.8 * The number of homicides in the administrative district of Namur in 1920 has been corrected, based on a cross-checking with the national data. The figure initially mentioned by the medical statistics was 68, which amount to a flagrant aberration given the evolution in others districts. The figure has been changed to 6 for the given year by cross-checking sources.

Table 3 Mean homicide rate for 100,000 inhabitants in Belgium 1901–1913 1919–1921 1922–1939 1940–1944

1945–1950

Antwerp Brabant Hainaut West Flanders East Flanders Lie`ge Limburg Luxembourg Namur

1.4 1.8 2.6 1.2 1.2 2.3 2.3 3.8 1.3

1.5 1.9 2.0 1.7 1.8 2.1 1.5 1.9 2.3

2.5 3.0 3.6 2.9 4.1 3.2 3.6 4.0 4.5

1.5 1.8 2.4 1.5 1.4 1.8 1.6 1.8 2.1

2.7 7.6 13.3 3.5 2.3 10.6 11.3 13.7 10.2

During peacetimes (1901–1913 and 1922–1939), the number of homicides is globally similar for each province. If we refer these data to the population, we obtain a rate ranging between 1.5 and 2.5 homicides for 100,000 inhabitants. Although the data are very, very uniform, it is nonetheless interesting to observe some regional differences, with a slight propensity to kill in the regions of the Walloon ridge formed by Hainaut, Namur and Lie`ge; three provinces characterized as containing most of the country’s heavy industries (metallurgy and coal mining) (Fig. 6). The end of World War One, in fact, was marked by a generalized increase in homicides over the whole territory, and in proportions ranging from 50% to 100% higher. Henceforth, in relation to the number of inhabitants, homicides became more frequent in the rural provinces of Namur, Luxembourg and Eastern Flanders which, apart from its industrial capital Ghent, was predominately rural as well. Contemporary qualitative sources equally suggest that the rise in homicide rates was more marked in rural than in urban areas. The predominantly rural character of mortal violence was due to the specific configuration of wartime Belgian society. For various reasons, the countryside was

Violence and War

195 1901–1914

1922–1939

Antwerp Eastern Western Flanders Flanders

Limburg Brabant Liège

Hainaut Namur

Luxembourg

<1

<1

< 2 et > 1

< 2 et > 1

< 3 et > 2

< 3 et > 2

>3

>3

Fig. 6 Homicide rates in Belgium by province in peacetime

the preferred workplace for bands of armed robbers whose activities were responsible for most homicides during the post war years. On the one hand, the countryside was particularly vulnerable to banditry because it was far more severely affected by the erosion of state authority than the urban areas. A real power vacuum existed in the countryside during the war. Policing in rural areas was traditionally, largely done by the Gendarmerie, which was absent during the occupation. The tiny local police corps lacked the manpower, skills and equipment to maintain order in the countryside. In the big cities, with their large communal police forces, the repressive capacity was far greater. On the other hand, the scope of rural banditry is a reflection of the social polarization between town and country in the post war years [on social tensions in Belgium during the First World War, see: (De Schaepdrijver, 1997, p. 236)]. The deprivations of occupation were not equally shared by the entire Belgian population. Some social groups were more severely hit by the crisis than others. Due to massive unemployment and rising food prices, the purchasing power of the vast majority of the population in urban and industrial areas collapsed. In the countryside the population was generally far better off: due to the need for manpower in the agricultural sector, unemployment rates were low, gardens offered alternative sources of food and income and rising food prices boosted income and living standards for farmers. Given the specific configuration of occupied Belgium (poverty in the cities, relative prosperity in the rural areas, matched with stronger repressive capacities in the cities than in the countryside) it is not surprising that the bandits stemmed mainly from impoverished urban social groups and that in their raids they especially targeted well-to-do farmers (Bekaert & Lempereur-Guimot, 1974, pp. 36–37). The risks of being caught during an attack on an isolated farmhouse

196

X. Rousseaux et al.

in the largely unprotected countryside were relatively low and the potential benefits relatively high. Moreover, farmers generally kept their money at home instead of banking it. The social profile of bandits and victims might suggest that the banditry contained at least an element of protest. In other words, one might ask whether it is legitimate to describe the banditry in Belgium during the occupation years and their immediate aftermath as a form of ‘‘social banditry’’ (Hobsbawn, 1969; for a major critique of the concept see Blok, 2001, pp. 14–28). Apart from the well-known, inherent objections to the relevance of the concept ‘‘social banditry’’, there’s no evidence that the bandits were motivated by a desire to express their indignation against some form of social injustice. Their primary motive was doubtlessly stealing as much money and as many goods as possible. The activities of the robber bands were closely interconnected with the flourishing black markets in the occupied territory. Amongst the vast majority of the population, there was no sympathy whatsoever for the bandits. It is nevertheless probable that resentment against profiteering farmers, felt amongst lower social groups in the urban and industrial areas which formed the recruiting base of the bands, facilitated the use of extreme violence against the farmers attacked. Non-identification with the farmer population at least partially explains the ferociousness of the armed attacks. As for World War II, it literally divided the country into two blocks with annual homicide rates which cross everywhere, globally, but not at all uniformly. Thus the mean homicide rate during the occupation is over 13 in Hainaut and Luxembourg. Hainaut even reached 47.3 homicides for 100,000 inhabitants in 1944! The provinces of Limburg, Lie`ge and Namur are not at rest either and all record a mean rate above 9. Then comes Brabant, which occupies an intermediate position (7.5). At that time, the northwest of the country (both Flanders and Antwerp) recorded rates that were hardly higher than those observed in peacetime. . . (Fig. 7)

1940–1944

1945–1950

<3

<1

< 6 et > 3

< 2 et > 1

< 9 et > 6

< 3 et > 2

>9

>3

Fig. 7 Homicide rates in Belgium by province during and after World War II

Violence and War

197

How can the striking contrast between the relatively calm Northwest and the relatively violent Southeast be explained? There’s no simple and conclusive answer to this question. In the following part a very premature attempt at an answer is formulated by confronting a number of hypotheses with the scarcely available empirical material. One possible explanation of the geographical distribution of homicide during the occupation years is that it is only a radicalization of a pattern existing before the war. In other words, does the geographical distribution of homicide during the second half of the occupation reflect the inter war period pattern? Comparison of the provincial homicide rates between the inter war period and the occupation years refutes the argument of a possible radicalization of an existing pattern. For the inter war years, there’s no evidence of a meaningful difference between the homicide rates for the Northwest and the Southeast of the country, although the provinces with the highest rates are found in the Walloon South. The differences between the provincial rates are further limited. While the borderline between the relatively calm and violent areas during the Second World War might not correspond entirely with the language border between Dutch and French, it remains a remarkable fact that the area with the lowest homicide rates corresponds to three monolingual Dutch provinces. Taking into account that a major part of wartime homicides were politically motivated, is the contrast between the relatively calm Northwest and the relatively violent Southeast, due to a different political climate in ‘‘Flanders’’ and ‘‘Wallonia’’? A major factor in this regard is the relatively weak position of the resistance in the Dutch speaking North. The step to active resistance against the occupier was relatively less frequently taken in the North than in the South. A number of indicators (geographical distribution of clandestine presses and sabotage actions, numbers of victims of German oppression of the resistance, help to hidden Jews) show the relative weakness of resistance in Flanders (with a few notorious exceptions like Louvain) compared to francophone Belgium. Research is underway to explain (or possibly refute) the alleged relative weakness of the resistance amongst Dutch speaking Belgians. A decisive answer is still to be formulated, but a few key factors can nevertheless be indicated (Gotovitch, 1972, p. 226; Gotovitch, 1976, pp. 295–317; Gotovitch, 1991 Laplasse & Steen, 2005, pp. 227–260; Maerten, 2003a, pp. 302–334; Maerten, 2003b, pp. 303–338; Saerens, 2000, pp. 711–716; Temmerman & Boeckx, 1995). The occupier’s Flamenpolitik is beyond any doubt one such factor. The early release of Flemish POW’s by the Germans – in contrast to their francophone companions who remained in custody for the entire duration of the waraborted a powerful source of anti-German resentment in Flanders. More fundamental was the different political balance of power in Dutch and French speaking Belgium. In Flanders, the prudent accommodation policy of the Catholic church, with its party and social organizations carried more weight due to their dominant position on the Flemish political landscape. The relative weight of the socialist and liberal parties was stronger in francophone Belgium. That difference in political profiles had its consequences because, in contrast to

198

X. Rousseaux et al.

the Catholic party, at least segments of the socialist and liberal parties entered the resistance. On the left, anti-fascism had already had more of a voice during the inter war years. The Communist Party had a real, popular base in the Walloon industrial basin, with its tradition of combative unionism, as well as in Brussels. In Flanders, with its tradition of consensual unionism and Catholic predominance, there was no real feeding ground for the radical left. The natural base for recruitment of armed partisans (who were responsible for a major part of the eliminations of collaborators during occupation) was consequently limited in Flanders. Since patriotism is a prime mover for active resistance to an occupier, it is logical that we pay attention to divergences of feelings of national identification in the North and South during the inter war period. In many ways the experience of the First World War determined national identification in inter war Belgian society. On the one hand, in the entire country, dramatic war experiences strengthened a Belgian nationalism closely linked with strong anti-German resentment. On the other hand, a process of radicalization amongst the Flamingants took place, giving birth to an anti-Belgian Flemish nationalism as an independent political factor. The confrontation of the two conflicting tendencies gave birth to real differences in national identification between the North and the South of the country. In francophone Belgium, a strengthened Belgian nationalism was uncontested and used as a political instrument against Flemish demands – which were identified with the occupier’s hated Flamenpolitik. In Dutch speaking Belgium feelings of attachment to the Belgian state remained dominant, but a growing mainly Catholic minority, primarily identified with Flanders – partly out of frustration with the slow tempo of the legal realization of their demands. Although it was a very ambiguous phenomena, it is very clear that the submergence of a concurrent nationalism weakened the overall force of Belgian nationalism in Flanders – and thus one of the major breeding grounds of the resistance during the Second World War. The rise of anti-German Belgian nationalism as a result of the First World War may have been general, but not equally strong in all parts of the country. Some regions were more severely affected by the invasion and occupation than others. It is only reasonable to believe that anti-German resentment was more deeply rooted in those regions. Anti-German feelings ran especially high in those towns and villages where massive ‘‘atrocities’’ had been commited by the invading German army. Research shows that memories of those events still had an impact during the Second World War. As a result of a geographical coincidence, most victims of the ‘‘atrocities’’ of 1914 lived in the southeastern Walloon provinces and in the district of Louvain (Horne & Kramer, 2001). Another important source of anti-German feelings were the difficult material conditions the Belgian population lived under during the First World War. Not all social groups and regions in occupied Belgium suffered equally under the occupation. Historian Henri Pirenne noticed a strong link between the relative standards of living and the degree of hostility towards the occupier (Pirenne, 1929, p. 119). Anti-German resentment was much more outspoken in the

Violence and War

199

industrial and urban areas (where the vast majority of the population lacked food) than in the rural areas (which generally benefited from the high food prices). Anti-German feelings were probably more radical in the Walloon industrial basin than in the Flemish provinces, which were still predominantly rural in 1914. Taking all of this into account, the possibility of a link between the regional variability of anti-German feelings, linked to divergent World War One experiences, and the geographical pattern of homicide during World War Two should not be excluded. Nor should one overlook the fact that the peak of homicide figures consists not only of politically motivated murders, but also includes victims of apolitical common law banditry. An overview of the figures on armed robbery for the years 1943 and 1944 seems to suggest that the regional pattern of armed robberies corresponds fairly well with the data on homicide (Gotovitch, 1985, pp. 33–34). In the Northwest of the country, armed robberies were relatively rare during the second half of the occupation compared to high numbers for the Southeast. It is well known that two basic conditions have to be met to explain the incidence of banditry: weak state authority and massive impoverishment. Since state authority was enforced in the same way all over the country, the explanation of regional patterns of banditry has to be related to the second factor. The Belgian population’s living standards fell dramatically very early on in the occupation, but not all social groups and regions were hit with the same severity. The working-class especially suffered from the crisis. In industrialized areas, the negative impact on the population’s living standards was felt particularly. In those regions the crisis led to a far stronger degree of social disintegration and banditry came to be used as a survival strategy. From this perspective, it is no coincidence that most robberies took place around the Walloon basin, with the highest concentration of industrial workers in Belgium. Belgian historians have agreed for a few decades now that the hunger of 1940–1941 quickly eroded the occupier’s fragile authority and caused the first expressions of resistance by the workers (Gotovitch & Gerard-Libois, 1971) [for a fresh perspective, see (Luyten, 2005, pp. 149–175)]. It is by no means surprising that reaction was most outspoken in the Walloon basin, since it was relatively more severely hit by food shortages than other regions which were less homogenously industrial. Thus in some sense the same specific social constellation which gave birth to banditry stimulated resistance as well. It should nevertheless be stressed that the fact of a partially common breeding ground does not imply the equation of two, inherently very different phenomena. A similar interplay of social and political factors caused diverging responses to the imposition of forced labour, which in its turn influenced homicide rates (Vrints, 2005, pp. 177–204). Research has shown that persons in hiding who wanted to avoid that measure opted for different strategies in the North and the South of Belgium. Walloons in hiding generally joined the resistance or opted for banditry as a survival strategy, while in Flanders and Brussels most of them found refuge near home. Differences in political and religious climates should

200

X. Rousseaux et al.

be taken into account in explaining this conclusion. In the North, the influential Catholic Church managed to take charge of a large number of hiding persons with the clear intention of stopping them from joining the (communist) resistance. In the predominantly dechristianized Walloon basin the Church could not play a similar canalizing role. In the absence of alternatives, Walloons in hiding were more likely to opt for the emergency strategies of banditry and the resistance (which was far more substantially organized in the South anyway). Differences in socio-economic structure played a major role as well. Workers in the Walloon basin had less room to manoeuvre than people in the rest of the country. It was harder for them to find an alternative source of income once they left their official job. People in rural and semi-rural areas easily found refuge with farmers they knew or were related to. In the Walloon basin, characterized by the juxtaposition of vast industrial and rural areas, the links between the worlds of workers and farmers were weak. In Flanders, such a tenuous juxtaposition was almost nonexistent and the borderline between the two worlds far more fluid. But even in comparison with big cities such as Antwerp, Ghent and Brussels, workers of the Walloon basin had greater difficulty in finding alternative sources of existence. The complex economic structure of a city like Brussels offered far more options (black market, working on the side) for survival in hiding. In an industrialized Walloon province like Hainaut there were far fewer options for finding alternative sources of income outside the factories and mines. In such a context, the resistance and banditry were often the only ways to survive. This might also partly explain why in Limburg, the only Dutch speaking province with a similar dependence on a few industrial employers, homicide rates were far higher than in both Flanders and Antwerp. Homicide is, in some sense, a good indicator of the existence of tensions in a particular society. During both World Wars Belgium might be categorized as a society under great pressure. Regional homicide statistics offer a valuable key to a better understanding of the social tensions in Belgian wartime society. In the specific configuration of the first German occupation of Belgium (massive impoverishment in the cities vs. relative wealth in the countryside and an implosion of state authority), a wave of rural banditry came into being which came to be reflected in regional homicide statistics. In the first post war years the geographical centre of gravity of homicide was very clearly situated in rural areas. The wartime polarization between town and country left its traces in the regional homicide rates. The geographical variability of homicide rates during the second half of the second German occupation provides a clear indication that tensions were far more severe in the southeastern part of the country than in the northwestern regions. Social conditions in the South and East – with its concentration of heavy industries and mines- seem to have formed a more fertile breeding ground for banditry, since the impoverishment was more massive in that region. Perhaps even more fundamental is the difference in degrees of political polarization, for various reasons, between the Northwest and the Southeast of the

Violence and War

201

country. In the Southeast, the process of political polarization during the occupation eventually resulted in a cycle of political violence between a relatively strong resistance and an increasingly socially isolated collaboration, reaching a peak in the Summer of 1944. Even if the same polarizing dynamics can be retraced in the Northwest, the political tensions did not result in a de facto situation of civil war.

Homicide in Belgium in Peace and War: Civilization or Brutalization? At the beginning of the 20th century, Belgium presents the image of one of the world’s most ‘‘advanced’’ countries. This highly industrialized and urbanized country had been experiencing declining levels of homicide for 70 years, as established by both judicial and medical statistics. In Elias’ terminology, earlytwentieth-century Belgian society had been taking part in a continuous ‘‘civilizing process’’ where non-lethal physical violence progressively replaced lethal violence in the statistics. As observed by the French criminologist Henry Joly in 1907, proportionally sentenced murders, parricides, poisonings, and infanticides are more numerous in France than in Belgium (. . .) Nevertheless, if we add violent acts lethal or non lethal to serious assault, they are more numerous among our Flemish neighbours.7 Rates of victimization and homicide prosecution remain particularly close. Judging by 1950, after two dramatic occupations, the ‘‘civilizing process’’ had not been stopped nor reversed. Yet, in terms of State control on lethal violence, the impact of occupation and the post occupation period look different in the cases of the First and the Second World Wars. During the First World War, the territorial control of the State had been weakened. Due to institutional malfunctioning during the war, particularly in rural areas, insecurity grew, provoking a vigorous post- liberation reaction by the authorities against the bandits. The authorities aimed at reaffirming the State’s Monopoly on control of violence, which had been contested by an ancient regime style rural banditry, as described by Florike Egmond. In the last years of the Second World War, banditry and political polarization between partisans and opponents of collaboration with the Nazi regime caused a peak in homicide figures in the southeastern part of the country. The State’s reaction was to reaffirm its monopoly on the control of violence, a monopoly contested by the political factions. The discrepancy between the numbers of homicide victims and judicial homicide provides evidence of 7

‘‘Or, si avec toute la prudence voulue, nous comparons la France et la Belgique, nous voyons que la premie`re a proportionnellement plus de gros attentats que la seconde (. . .) En retour, si on additionne les actes de violence, coups et blessures ayant ou non entraıˆ ne´ la mort, . . . ils sont autrement plus nombreux chez nos voisins des Flandres (Joly, 1907, p. 16).

202

X. Rousseaux et al.

political negotiations between the State and the political groups, aimed at dismissing procedures for homicide cases after the war. In conclusion, on the macro-level of the national State, in Belgium the taming of homicide continued during the first half of the 20th century. Generally speaking, the wartime experiences of two occupations by the German army didn’t stop the civilizing process, at least as measured by homicide figures. Nevertheless, a certain brutalization occurred, especially in 1943 and 1944, but that brutalization remained limited in time and restricted to certain social segments and certain geographical arrondissements. Both the long term trend in homicide figures as well as the particularities linked to the wars invite us to moderate the global homicide rate by further scrutinizing regional models and local histories. By comparing homicide rates registered locally by physicians with statistics on penal prosecution, the particularities of the Belgian political context filter through. Hence these homicide rates are key in understanding the fault lines running through the society and opening new windows on the experiences of occupation and, specifically, on the processes of de-civilization and re-civilization of the national community during those periods.

References Balace, F. (1995). Les hoquets de la liberte´. In F. Balace, (ed.), Jours libe´re´s II (Jours de guerre) (pp. 75–132). Brussels. Bekaert, H., Lempereur-Guilmot, N. (1974). Les fluctuations de la criminalite´ en Belgique, Tome I: 1964–1966. Brussels. Blok, A. (2001). Social banditry reconsidered. In A. Blok, Honour and Violence (pp. 14–28). Cambridge. Bouveroux, J. (1985). Terreur in oorlogstijd. Het Limburgs drama. Antwerpen, Amsterdam. Chesnais, J.-C. (1981). Histoire de la violence en Occident de 1800 a` nos jours. Paris. Conway, M. (1993). Collaboration in Belgium. Leon Degrelle and the Rexist Movement 1940–1944. New Haven, London. Dejonghe, A. (1984). De strijd Himmler-Reeder om de benoeming van een HSSPF te Brussel. Vijfde deel : Salzburg voor en na. Politionele ontwikkeling van september 1943 tot het einde van de bezetting. Bijdragen tot de Geschiedenis van de Tweede Wereldoorlog, 8, 5–234. Deneckere, G. (2000). Off with their heads. The death penalty in Belgium. The Low Countries, 8, 75–83. De Schaepdrijver, S. (1997). De Groote Oorlog. Het koninkrijk Belgie¨ tijdens de Eerste Wereldoorlog. Amsterdam, Antwerp. De Wever, B. (1995). Greep naar de macht. Vlaams-nationalisme en Nieuwe Orde. Het VNV 1933–1945. Tielt. Egmond, F. (1986). Banditisme in de Franse tijd. Profiel van de Grote Nederlandse Bende 1790–1799. Soest. Eisner, M. (2003). Long-term historical trends in violent crime. Crime and Justice. A Review of Research, 30, 83–142. Ge´rard-Libois, J., & Gotovitch, J. (1971). L’An 40. La Belgique Occupe´e, Brussels. Geschiedenis van de Rijkswacht, Brussels, 1980. Gotovitch, J. (1972). Beeld van de klandestiene pers in 1940. Bijdragen tot de geschiedenis van de Tweede Wereldoorlog, 2, 223–267.

Violence and War

203

Gotovitch, J. (1976). Wallons et Flamands: le fosse´ se creuse. . . In H. Hasquin (ed.), La Wallonie, le pays et les hommes (t.II, pp. 295–317). Brussels. Gotovitch, J. (1985). Un fichier et une statistique des attentats sous l’occupation. Bulletin du Centre de Recherches et de l’Etudes Historiques de la Seconde Guerre Mondiale, 15, 33–34 Gotovitch, J. (dir.) (1991). Guide de la presse clandestine de Belgique. Brussels. Gotovitch, J. (1992). Du Rouge au Tricolore. Les Communistes belges de 1939 a` 1944. Un aspect de l’histoire de la Re´sistance en Belgique. Brussels. Gotovitch, J., & Gerard-Libois, J. (1971). L’An 40. La Belgique occupe´e. Brussels. Hobsbawn, E. (1969). Bandits. New York. Horne, J., & Kramer, A. (2001). German Atrocities, 1914. A History of Denial. New Haven, London. Jacquemyns, G. (1945). Privations et espoirs. La socie´te´ belge sous l’occupation allemande. Brussels. Joly, H. (1907). La Belgique criminelle. Paris. Laplasse, J., & Steen, K. (2005). Het verzet gewogen. Een kwantitatieve analyse van politieke aanslagen en sabotages in Belgie¨, 1940–1944. Bijdragen tot de Eigentijdse Geschiedenis, 15, 227–260. Luyten, D. (2005). Stakingen in Belgie¨ en Nederland, 1940–1941, Bijdragen tot de Eigentijdse Geschiedenis, 15, 149–175. Maerten, F. (2003a). Les courants ide´ologiques and the Re´sitance belge. In Les courants politiques et la Re´sistance: continuite´s ou ruptures? International Colloquium, Esch-sur-Alzette, April 2002 (pp. 302–334). Luxembourg. Maerten, F. (2003b). L’impact du souvenir de la Grande Guerre sur la Re´sistance durant le second conflit mondial. In L. van Ypersele (ed.), Imaginaires de guerre. L’histoire entre mythe et re´alite´ (pp. 303–338). Louvain-la-Neuve. Majerus, B., & Rousseaux, X. (2004). The impact of war on policing in Belgium. In C. Fijnaut (ed.), The Impact of World War Two on Policing in North-West Europe (pp. 43–89). Leuven. Majerus, B., & Vrints, A. (2005). De brutalisering voorbij? De Eerste Wereldoorlog en politiek geweld in Europa. In T. Zwaan (ed.), Politiek geweld. Etnisch conflict, oorlog en genocide in de twintigste eeuw (Zestiende jaarboek van het Nederlands Instituut voor Oorlogsdocumentatie) (pp. 52–74), Zutphen. Mazower, M. (1998). Dark Continent: Europe’s 20th Century. London: Allen Lane: The Penguin Press. Meyers, M. (1919). La Magistrature et l’Occupation. Audience Solennelle de Rentre´e du 1er octobre 1919. Lie`ge. Pirenne, H. (1929). La Belgique et la guerre mondiale. Paris, New Haven. Saerens, L. (2000). Vreemdelingen in een wereldstad. Een geschiedenis van Antwerpen en zijn joodse bevolking (1880–1944). Tielt. Scholliers, P. (1988). Belgie¨ dreunt in zijn voegen. Sociale en politieke troebelen, 1914–1921. In F. Vanhemelryck (ed.), Mensen in oorlogstijd (pp. 151–163). Brussels. Scholliers, P. (1993). Arm en rijk aan tafel, Tweehonderd jaar eetcultuur in Belgie¨. Berchem. Scholliers, P. (1994). The policy of survival: food, the state and social relations in Belgium, 1914–1921. In J. Burnett, & D. Oddy (eds.), The Origins and Development of Food Policies in Europe (pp. 39–53). London, New York. Scholliers, P., & Daelemans, F. (1988). Standards of living and standards of health in wartime Belgium. In R. Wall & J. Winter (eds.), The Upheaval of War. Family, Work and Welfare in Europe, 1914–1918 (pp. 139–157). Cambridge. Spierenburg, P. (1994). Faces of Violence: Homicide Trends and Cultural Meanings: Amsterdam, 1431–1816, Journal of Social History, 27, 701–716. Temmerman, P., & Boeckx, B. (1995). Deportatie en verzet. Een eerste globale statistische analyse op basis van de erkenningsdossiers Politieke Gevangenen. Brussels.

204

X. Rousseaux et al.

Tixhon, A. (2001). Le pouvoir des nombres. Une histoire de la production et de l’exploitation des statistiques judiciaires belges (1795–1870) (Doctoral dissertation, Universite´ catholique de Louvain, 2001). Van Doorslaer, R. (2003). Gebruikt verleden. De politieke nalatenschap van de Tweede Wereldoorlog in Belgie¨, 1945–2000. In G. Deneckere & B. Wever (eds.), Geschiedenis maken. Liber Amicorum Herman Balthazar (pp. 227–266). Ghent. Van Meerbeeck, P., Verhoeyen, E., Van of Vijver, H., & Van Doorslaer R. (1988). Lexicon. De tijd der vergelding en het verzet. Brussels. Velle, K. (1985). Statistiek en sociale politiek: de medische statistiek en het gezondheidsbeleid in Belgie¨ in de 19de eeuw. Belgisch Tijdschrift voor Nieuwste Geschiedenis, 16, 213–242. Vrints, A. (2005). Patronen van polarisatie. Homicide in Belgie¨ tijdens de Tweede Wereldoorlog. Bijdragen tot de Eigentijdse Geschiedenis, 15, 177–204. Wouters, N. (2004). Oorlogsburgemeesters 40/44. Lokaal bestuur in collaboratie in Belgie¨. Tielt.

Democratization and Violence: European and International Perspectives Susanne Karstedt

Democratization, Democracy and Violence It is a startling fact that when in the mid-80s a ‘third wave’ of democracy took hold in Latin America and Eastern Europe, both democracy and violence were simultaneously on the rise worldwide. Almost by definition democracies represent an institutionalized framework and a way of life that ensures non-violent means to share power between communities of people with widely differing values and beliefs. As Keane (2004) points out, ‘violence is anathema to [democracy’s] spirit and substance’ (p. 1). Accordingly, the process of democratization was accompanied by expectations that violence would generally decrease, and that these countries would embark on a process of reducing levels of violence as Western European countries had done earlier in the 19th and 20th century. Why then was the movement of transitio1ns from authoritarian regimes to democratic rule in South America, Central and Eastern Europe, and South Africa accompanied by waves of individual violent crime and collective violence in these countries? Had democracy lost its capacity to contain violence and to generate peaceful habits amongst its citizens? Or did the new liberties and free market policies that came with democratization generate violence amongst citizens, and democracy was actually criminogenic? Were democratic societies at a disadvantage compared to autocracies when it came to the control and containment of violent conflicts?

A first version of this paper was prepared for a conference of the Democracy Collaborative at the University of Maryland, 2002, and a second version during a Visiting Fellowship at RegNet at the Australian National University in 2004. I am grateful to Boerge Bakken, Sophie Body-Gendrot, John Braithwaite, Manuel Eisner, Willem de Haan, Gary LaFree, Dag Lindstrom, Steve Messner, Pieter Spierenburg, Tom Tyler, and Peter Waldmann for their comments and ideas, and to the Australian National University for a Visiting Fellowship grant. I thank Lena Kruckenberg, University of Bielefeld/Keele University for her assistance in researching and editing. Susanne Karstedt Professor of Criminology, School of Criminology, Education, Sociology and Social Work, Keele University, Keele, Staffordshire, ST5 5BG, UK [email protected]

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

205

206

S. Karstedt

The deepening scepticism as to the capacities of democracies to restrict individual violent crime and collective violence was further fuelled by the fact that since the 1960s, violent crime in mature Western democracies had considerably increased, and a number of these countries had experienced ‘booms’ of violent crime during this period (Eisner, 2001; LaFree & Drass, 2002). Mature and transitional democracies alike were affected by rising rates of violent crime during the last decades of the 20th century. The ostensible coincidence between democratization and violence thus seemed to be related to built-in characteristics and dynamics of democratic societies that generated violence through the liberties they grant, the inequality they produce and their lack of tight social control in contrast to autocratic regimes. In sum, the institutions and values of late modern democracies could be seen as causal factors in the production of individual and collective violence on a considerable scale. Contemporary democracies had lost the capacity to deliver on democracy’s original promise of non-violence. In particular for transitional democracies the wave of violence was deemed to have detrimental effects on the legitimacy of its emerging and still fragile institutions, when such scepticism took hold amongst the citizenry, who saw themselves confronted with unprecedented risks of becoming victims of violence. In this paper I will explore the mechanisms behind the seeming coincidence of democratization and violence during recent decades. To this purpose, I compare the comparative advantages and disadvantages of democracies and autocracies in containing violence, and argue that rising levels of interpersonal violent crime in transitional democracies have to be attributed as much or even more to the legacy of past autocratic regimes as to the emerging democracy (for a detailed analysis see Karstedt, 2006). First, I explore two trends of violence that have been related to the process of democratization: The increase of violent crime in transitional democracies in Eastern Europe and Latin America, as well as in South Africa, and increasing levels of ethnic rebellion and collective violence, in particular in the form of ethnic cleansing in the southern and later in the northern hemisphere (Gurr, 1993, 2000; Mann, 2005). I will then identify comparative advantages and disadvantages of democracies and autocracies in containing violence by contrasting ‘liberal inclusion’ in democracies with ‘repressive inclusion’ in autocracies, and show how comparative disadvantages cumulate in the process of democratization. Here as well as in the following empirical part of the study I will focus on interpersonal violent crime. Finally, I will analyse violent crime waves during transitions to democracy for European countries and contextualise the European experience within an international perspective.

Deficiency and Vulnerability: Democratization and Violence The process of democratisation is basically one that moves a society from an autocratic regime towards a non-autocratic and democratic regime. European states provide examples for such processes taking centuries, and being interspersed

Democratization and Violence

207

with periods of autocratic regimes. The so-called first wave of democratisation was certainly dominated more by slow and drawn-out processes of democratisation, whilst the second wave included a significant number of states with quite abrupt transitions. Such abrupt transitions are clearly the defining feature of the third wave of democratisation starting in the 1980s in Latin America, and coming to its climax with the end of the Cold War in Central and East Europe. The recent shift towards democracy came with an abrupt and extraordinary rise of crime mainly in the post-communist countries. In particular violent crime was seen as indicative of the profound changes that turned these societies and the life of their citizenry upside down. Experiences of violent crime gave rise to feelings of insecurity amongst citizens, and added to feelings of anxiety already at high levels as the result of massive social, economic and political changes (Ewald, 2000). In some of the transitional democracies the electorate brought those back into power who they deemed as guarantors of the regime of ‘law and order’ of the past (see Pridemore & Kim, 2006, for Russia). Outside of Europe, South Africa was emblematic of how the shift toward democracy was accompanied by a wave of violent crime (Steinberg, 2001). Most of the causes that were made responsible for the wave of violent crime were related to the process of democratization and the introduction of free markets: The experience of new levels of social inequality was not attenuated by democratic practices (Karstedt, 2003); rampant market individualism made the loss of collectivistic solidarity deeply felt (Pridemore & Kim, 2006); shortcomings in the democratic institutions of criminal justice and the rule of law produced ‘disjunctive democracies’ (Caldeira & Holston 1999). Sztompka (1993; also Karstedt 2003) identified a lack of civil society and its social bonds, in sum general ‘civilizational incompetence’ at the roots of the wave of violent and other crime in Eastern and Central European countries. While electoral and political democracies were securely established in many Latin American countries, ‘lawless violence’ pervaded their institutions and indicated a ‘clear abdication of democratic authority’ (Mendez, 1999, p. 19). This became particularly visible in the institutions of criminal justice. The wave of violent crime fuelled a militarization of the police in Latin American countries (see e.g. Huggins, Haritos-Fatouros & Zimbardo, 2002) that led to increasing violations of human rights and non-democratic practices. These analyses share a deficiency perspective of the democratization process. Violent crime is the result of a not yet fully established democracy, and of particular deficiencies and maladjustments that arise during the process of democratization. Not the values and institutions of democracy, but the incomplete development of its institutions and practices are identified as causes of violent crime. However, the deficiency perspective partially identifies genuine deficiencies of democracy when compared with autocratic regimes. The rise of crime and violence are attributed to the liberties democracies provide, the inequality they produce, and the diversity of beliefs and patterns of behaviour that are not as tightly controlled as in more authoritarian regimes (Davis Rodriguez, 2006).

208

S. Karstedt

From this perspective, transitions toward democracy are defined by the final goal, the full establishment of democracy. Consequently, autocracies compare favourably to democracies with regard to the control of crime and violence. The deficiency perspective misses out on the fact that transitions are also transitions from authoritarian and autocratic states, and waves of crime and violence during the transition period can be a legacy of the autocratic past. If the downfall of the autocratic state is taken as the starting point, the transition was preceded by increasing tensions, a loss of legitimacy of its institutions and declining trust among the citizenry (Los, 2003). The legacy of the autocratic state is not only a consequence of its downfall, but one of deeper and built-in anomic tendencies that have long-term consequences for transitional societies (Mendes, O’Donnell & Pinheiro 1999; Waldmann, 2002, 2003). Democratization as a process of moving from an autocratic to a non-autocratic regime might be much more path-dependent and anchored in the autocratic past then shaped by the nascent future democracy. The deficiency perspective it seems severely underrates the effect of de-autocratization in the transition process. Collective violence – in the form of ethnic cleansing and ethnic wars – dramatically increased during the 1990s in the successor states of the former Yugoslavia and the Soviet Union, and was part of the transition to democracies and market societies. European countries had not seen collective violence and ethnic cleansing on such a scale since World War II (Gurr, 1993; 2000; Mann, 2005). Mann (2005, p. 4 ) contends that ‘murderous ethnic cleansing has been moving around the world as it modernized and democratized’, and he sees it as the ‘dark side of democracy’. Though Mann does not argue that democracies themselves engage in ethnic cleansing – this would be the end of democracy by definition – they are nonetheless in danger of sliding into ethnic violence through the perversion of either their liberal or egalitarian-socialist ideals. He concludes from a number of case studies covering the 20th century that this danger arises in particular for democracies since they develop strong tendencies toward ethnic homogenization and inclusion. While autocracies can cope better within multi-ethnic environments through authoritarian practices of ‘divideand–rule’ (Mann, 2005, p. 4), democracies have built-in mechanisms that make them vulnerable to this form of collective violence in such environments. This argument is grounded in a vulnerability perspective contending that democracies, their institutions and practices are fragile and easily destroyed. The vulnerability perspective has mainly been developed in relation to ethnic violence. Both the deficiency and the vulnerability perspective aim at explaining transitional processes and their relationship to violence. However, a number of distinct causal patterns have been made equally responsible for the rise of violent crime in mature democracies. Gary LaFree (1998) identified the loss of legitimacy of vital institutions of democracies like the education system, the system of social welfare and the criminal justice system, as decisive factor in the increase of violent street crime in the United States. In a similar vein, Messner and Rosenfeld (1994) argued that institutional imbalances between free market and welfare institutions – both corner stones of the institutional arrangements

Democratization and Violence

209

of contemporary democracies – caused rising rates of violent crime. Individualism and autonomy as core values of democracy have been made responsible for increasing violent crime in established democracies when coinciding with a loss of self-control (Eisner 2001, 2003; for an overview Karstedt, 2001; 2005). Democracy and democratization is seems have built-in mechanisms and dynamics that make them susceptible to waves of violent crime and outbursts of collective violence.

Democracy and Violence: Tensions of Liberal Inclusion Democracy is an extremely multi-dimensional concept. Definitions cover the whole range from their bare institutional framework (e.g. Przeworski, Alvarez, Cheibub, & Limongi, 2000) to those, which define democracy through political and civil rights, democratic practices and values. Others include institutional arrangements like welfare, education, industrial relations and the legal system (see Inglehart & Welzel, 2005; Inkeles, 1993; Jaggers & Gurr, 1995; Lauth, Pickel & Welzel, 2000; Welzel, 2002). ‘Civil society’ – the associations and networks of citizens – is presently seen as a necessary condition for liberal democracies; according to Gellner (1994) ‘no civil society, no democracy’. The notion that civil society underpins the institutions and practices of democracy extends the concept of democracy beyond an exclusively institutional perspective. Transition towards democracy implies the development of the institutional framework and democratic practices and values, embedded in civic associations, and deficiencies in civil society underpinnings of democracy are seen as a severe threat to the transition process (Caldeira & Holston, 1999; Gibson & Duch, 1994; Muller & Seligson, 1994; Nowotny, 2002; Sztompka, 1993). In fact, democratic institutions are based on and foster democratic practices: ‘strong democratic institutions produce strong democratic practices’ (Jaggers & Gurr, 1995, p. 476). Inglehart and Welzel (2005) emphasise in particular the role of democratic values for the future development of democracy and the process of democratization. Democracies are based on the ‘liberal inclusion’ of a plurality of people, classes, values and practices. However, the inclusionary project of liberal democracies is restricted to the borders of the nation state through notions of citizenship. Democratic societies are confronted with the tensions that arise from their inclusionary values and procedures, and their recognition of plural interests, values, and differences of class and ethnicity. The tensions that arise between recognition and encouragement of individual autonomy and diversity on the one hand, and the simultaneous effort to integrate these in the inclusionary project on the other, produce the contradictory and countervailing forces in democracies that give rise to violence and violent crime. The tensions between the inclusionary mechanisms and procedures that modern democracies established and the autonomy and rights which they granted to

210

S. Karstedt

individuals, were clearly seen by one of the first observers of modern democracy, Tocqueville (2000 (1835, 1840)). In his descriptions and analyses of the first modern democracy, the United States, he pointed to the paradoxes and inconsistencies of democracy that arise from its innate tensions. Democracy simultaneously gives rise to individual independence and high levels of conformity: the rights of individuals are counterbalanced by strong forces towards conformity. Majority rules imply dominance and authority, and rely on submission as well as they provoke resistance. Tocqueville observed that egalitarian values co-exist with envy, status differences and social inequality.1 Democracy offers its citizens wide opportunities for deviant, licentious and dangerous behaviour, and simultaneously reduces their desire to do what democracy allows them to do by establishing common and shared values, and procedures that ensure the exertion of individual rights. Toqueville was convinced that the flaws of democracy could be overcome by more democracy (Elster, 1999; Holmes, 1993), or countervailing forces in its institutional arrangements. Contemporary scholars have confirmed the seminal role of two cornerstones of democracy, the rule of law and fair trials for the acceptance of moral norms and sentences in case of criminal violation (Tyler, 1990, 2003; Tyler & Blader, 2000). The innate inclusionary ‘logic’ of democracies engendered on the level of institutions the widening of participation and empowerment for different strata and groups of nation state societies. They developed flexible institutional frameworks within which a diversity of other institutions, economic relations and cultural values could co-exist. Western democracies were and are capable of ‘accommodating’ high levels of economic inequality and tensions between classes, economic discrimination of minorities, and exclusion, as well as ethnic fractionalization, which all are related to high levels of personal and individual violence as well as to incidents of collective violence. Tocqueville saw the mechanisms that could counteract the innate tensions of democratic societies in the relationships and associations that citizens were capable of and free to develop in their communities. Values of tolerance of different views and lifestyles, balanced by mechanisms of interpersonal trust, individual autonomy balanced by mechanisms of social control, participation and membership in associations and groups combine into the pattern of civil society in which the innate tensions of democratic societies are dealt with and attenuated. Citizens engage in relationships beyond their families, and narrow communities with each other in ‘benevolent disinterest’ as Albert Hirschmann once termed it, and these form the foundations for the inclusionary project of democracy below the institutional level. Cohesion in democracies is based on weak ties, rather low level of consensus, and pragmatic dissent (Mann, 1970; Muller & Seligson, 1994). These social bonds are easily endangered by

1

Toqueville however never commented on the big divide in the democracy of the USA, between its African, then slave population, and white population (Cohen 2001).

Democratization and Violence

211

processes, which fractionalize and divide societies like social inequality, ethnic or religious divisions.2 The production of generalized trust in modern societies is based on high social mobility, individualism and social equality, all mechanisms that promote inclusionary and universal values and behavioural patterns in contrast to particularistic ones (Karstedt 2001, 2006). Democratic institutions equally provide and enforce these bonds.3 What are the advantages and disadvantages of democracy in containing violence? As far as democracies rely on and strengthen through their institutions a particular value pattern – individualistic and egalitarian values, as well as higher levels of trust and tolerance – levels of violence should be in principle lower within democratic societies. Democracies are at a disadvantage however, when its inbuilt tensions reach high levels. This will happen when individualistic values of autonomy and self-assertion are not balanced by inclusionary mechanisms, when structural inequality is in contrast with egalitarian values, when parts of the population are denied fairness and justice, and when exclusionary processes affect parts of the population. When turning to those mechanisms and values that can keep levels of violence low in democracies, individualism and egalitarianism both reduce levels of violence (Messner, 1982; Karstedt, 2001, 2005, 2006). From a Tocquevillean perspective those mechanisms that are capable of restraining the disruptive consequences of individualism will reduce violence in democratic society. Studies that analyse the impact of social capital – measured as generalized trust – on violence generally confirm the violence containing role of these mechanisms. In cross-national studies Ledermann, Loayza & Menedez (2002) found that violence increased significantly (although the impact was small) along with a weakening of social capital. Rosenfeld, Messner & Baumer (2001) found in a study of U.S. counties a complex causal pattern that showed a direct and negative impact of generalized trust on homicide rates, while civic engagement and participation had only indirect negative effects, and were dependent on the level of deprivation. Sampson, Raudenbusch & Earls’ (1997) study on collective efficacy provides further evidence that social bonds within communities as well as their bonds with institutions reduce levels of violence. Democracies are at disadvantage in containing violence, when their built-in tensions reach temporarily or continuously high levels and generate a potential for waves or continuously high levels of violence. Thus, the process of increasing individualization and the dominance of values of self-expression had set democracies on a trajectory of social change that had weakened those forces that restrained individual self-assertion and autonomy (see Eisner 2001; 2003). Detachment from vital social bonds and a decline of social control characterizes

2

Muller and Seligson (1994) found that ethnic fractialization decreased levels of democratic development. 3 Several authors, most prominently Rothstein and Stolle (2002) have argued that institutions create trust (institutional theory of trust).

212

S. Karstedt

this process. Established democracies today seem to be overwhelmed by the forces of individualization that they had set free in the first place: market individualism, autonomy and self-expression values (Klages, 1984; Inglehart, 1997). Hagan, Hefler, Classen, Boehnke & Merkens (1998) found that unrestricted hierarchical self-interest was related to violent behaviour of juveniles. Eisner (1997, 2001) argues that unrestricted values of self-expression were responsible for the violent turn in European democracies since the 1960s; his findings have been indirectly corroborated by findings from the World Values Survey (Inglehart & Welzel, 2005: pp. 154) that the more democracies mature the higher they score on values of self-expression. Economic and social inequality produces tensions in relation to the core democratic value of egalitarianism; they are consequently a source of violent conflicts and crime. Social and economic inequality is a strong predictor of violence in all societies (Messner & Rosenfeld, 1998), but its impact might be particularly detrimental in democracies with highly developed egalitarian values. This is further exacerbated, where liberal inclusion is endangered by ethnic fractionalization and economic discrimination against ethnic groups (Karstedt, 2001, 2006). Further tensions arise from imbalances between the economic sphere and other institutions. Messner & Rosenfeld (1994) argue that market institutions need to be restrainted through welfare regimes, and the process of commodification needs to be counterbalanced by values of the common good and civic attitudes. If such restraints and counterbalances fail in democracies, ‘institutional anomie’ generates high levels of violence. The theory seems to be particularly useful in explaining transitions to democracy that come with a transition to a market economy, as in Eastern and Central Europe (Messner, 2004; Savolainen, 2000). Democratic values and institutions provide democracies with comparative advantages not only to keep violence at low levels, but to cope with those tensions that increase violence. Democratic societies that have continuously high levels of tensions – be it as a result of ethnic conflict, economic discrimination of large groups, high levels of inequality, imbalances between market institutions and those that counterbalance them, or because they cannot control the forces of individualism and self-expression values – are generally disadvantaged in containing levels of violence. LaFree & Tseloni (2006) found for a sample of mature and transitional democracies that democratization increased levels of violence in the second half of the 20th century. Their findings point to consistent and increasing tension in democracies during this period.

Autocracy, Totalitarianism and Violence: ‘Repressive Inclusion’ Modern autocratic states are founded on ‘repressive inclusion’. The Fascist states in Europe during the first half of the 20th century, the communist states, as well as populist authoritarian regimes particularly in Latin America all were ‘inclusionary projects’ as far as they – ideologically and politically – explicitly tried to integrate the lower classes, different ethnicities, and economic interests. In the communist

Democratization and Violence

213

states in Eastern and Central Europe this was achieved through wiping out differences between classes, religions, life styles, beliefs and values and imposing an overarching ideological system. In these societies all those organisations and associations, in which such differences can flourish, were prohibited and substituted by a unified system of state-run organisations and a state economy. These state organisation intruded into all spheres of life and eliminated civil society, which at best retreated into a ‘niche’ of private life that was shielded against the intrusion of the state. This generated the very ‘civilizational incompetence’ in these countries (Sztompka, 1993) that burdens the transition to democracy. Its defining features are high levels of distrust in others and institutions, lack of legitimacy particularly of the justice system, and prominently the lack of mechanisms to deal with a diversity of interests, and the resulting tensions (Nowotny, 2002). Latin American authoritarian states took a more ‘etatist approach’ to the control of the population that intruded less into society but relied more on continuously high levels of violent repression (Huggins, Haritos-Fatouros & Zimbardo et al., 2002; Mendes, O’Donnell & Pinheiro, 1999; Waldmann 2002, 2003). In both instances of ‘repressive inclusion’4 we find a high level of anomic tendencies that undermine not only the social norms embedded in civil society, but in addition legal norms and general compliance with them. The use of repressive constraints and mechanisms of control as well as the achievement of ‘order’ come with a considerable erosion of social and legal norms (Waldmann, 2002). Like democracies, modern autocratic regimes – in particular the Fascist regimes during the 1930s and the Communist countries after 1950 – did not only promote egalitarian values, but their practices and institutional patterns aimed at equality – this was their inclusionary project. As a consequence, social inequality was – at least according to official measures – lower than in many democracies where market societies and economic discrimination of minorities produced higher levels. This applies to the Fascist regimes in the first half of the 20th century in Europe, and mainly to the communist regimes in Central and Eastern Europe, which were extremely homogenous societies. In contrast, Latin American autocracies had highest levels of inequality worldwide. Thus, autocratic regimes are capable of producing a sense of social solidarity, which covers tensions underneath (Kharkhordin, 1999). The official and nearly mythical character that solidarity acquired under communist ideology glossed over the actual tensions, levels of mistrust and repressive social control, and was not the least responsible for the double moral standards reigning in these societies.5

4

South Africa seems to be an example of achieving inclusion by ‘repressive segregation’, which makes it a unique case in this respect. 5 East Germany experienced a wave of violence and hate crimes against immigrants and asylum seekers, often against the same groups the state had officially welcomed under the banner of international solidarity some years ago. It is argued that since such international solidarity was never genuinely adopted by the population, but imposed by the state, it contributed to fuelling xenophobia after the fall of the Wall.

214

S. Karstedt

The repressive forms of control exerted by autocratic regimes and the intruding and far-reaching controls of the population in totalitarian states produce a certain level of order and contain violent conflicts (Mann, 2005). Autocratic regimes therefore often report lower rates of both property and violent crime than democracies. However, crime statistics from these countries are notoriously unreliable, in particular with regard to violence. The amount of violence exerted by the authorities is covered up and does not show in official crime statistics. Pridemore (2003) using mortality data found that Russian homicide victimization rates were as high, or even higher than in the United States for at least the last several decades. Importantly, these indicators of ‘order’ hide sources of anomic tendencies (Waldmann, 2002). Since the framework of security and order is not based on the rule of law but on ‘the rule of power’, it is applied arbitrarily and in a non-predictable way. Thus, feelings of insecurity among the population reach rather high levels. Further, such practices of repressive order de-validate norms, rules and regulations as guidelines for behaviour. The fact that the rule of law is substituted by the rule of ‘power’ on all levels of the state bureaucracy (Hagan and Radoeva, 1998; Karstedt, 1999) will increase the feelings of powerlessness in the population, and decrease the legitimacy of and trust in the authorities. Autocratic and totalitarian regimes usually build up considerable bureaucracies and produce a high amount of rules and regulations, with ‘double standards’ emerging, which tacitly allow for their circumvention. This applies to the highest levels of the government, bureaucracy and economic elites as much as to the lower echelons of bureaucracies. For the population non-compliance with the abundance of rules and regulations is the way to get by in daily life (Rose, 1998). The imposition of programmes of economic and social modernization contributes to feelings of powerlessness and lawlessness among the people. Autocracies thus produce high levels of ‘legal cynicism’ in societies. Legal cynicism, which is defined as a general contemptuous and depreciating attitude toward legal norms, is strongly related to high levels of violence (Sampson & Bartusch, 1998). Anomic tensions are also generated by the specific types of inequality in authoritarian regimes. Repressive inclusion ultimately has to rely on inequality of participation and power, and consequently on social and economic inequality. Notwithstanding their egalitarian ideology and overt political practice the communist regimes in Central and Eastern Europe installed a system of privileges for the ruling elites that ‘trickled’ down to the lowest levels of the bureaucracy (Karstedt, 1999; Los, 1988), and produced considerable levels of covert economic inequality in these countries (Karstedt, 2003). Autocratic regimes also produce economic inequality. Autocratic regimes with capitalist economies seem to exploit labour, and in transitions to autocracies income is frequently redistributed from labour to capital (Przeworski et al. 2000). Karstedt (2006) found for a sample of 26 countries that authoritarian and non-egalitarian value patterns and economic inequality were clustered in autocracies, which consistently have higher levels of violence than democracies.

Democratization and Violence

215

Autocracies and totalitarian states have advantages in containing violence that are mainly resulting from their tight grip of social control. When compared to democracies they have however severe disadvantages, which result partially from the very mechanisms and practices of their regimes, and partially from the anomic tendencies which these produce. If they promote non-egalitarian value patterns they should experience higher levels of violence (Karstedt, 2006). Further, their use of violent means of social control pervades society and increases levels of violence. Their practices of arbitrary treatment of citizens, and the resulting legal cynicism and lack of trust might engender an undercurrent of violent conflicts that emerge when the tight grip of control loosens.

Transitions Toward Democracy: Cumulative Disadvantages and ‘Booms’ of Violence During transitions toward democracy disadvantages of both autocracy and democracy cumulate and can cause exceptionally steep increases and high levels of violent crime, as well as eruptions of collective violence. The anomic tendencies that built up during the preceding autocratic regime and contributed to its downfall should push rates of violent crime up. When during the transitional phase the grip of the authorities loosens, the anomic tendencies will produce a wave of violent crime that typically accompanies this period. In the postcommunist states of Central and Eastern Europe, the transition to democracy came with a shift from homogenous to more heterogeneous societies, and higher levels of overt inequality. Combined with rampant legal cynicism, the feeling of injustice will contribute to the level of violent conflicts. The transition is further seen and experienced as a catastrophic loss of social solidarity in these countries, which from hindsight is glorified as a national and historical tradition (Kharkhordin, 1999). Pridemore & Kim (2006) show with homicide data from Russia, that those regions had high levels of lethal violence, where the share of votes for the successor party to the former communist party, which represented the social solidarity of the past, was particularly high. Democratic values and practices, as well as institutions that are capable of containing violence in democracies have not yet a stronghold in the population. Thus, the tensions produced by rising inequality, by free markets not properly balanced by other institutions, and by the depletion of those social bonds that can control individual autonomy and self-expression hit society in a very vulnerable state. Societies in transition experience mainly the disadvantages of democracy and its shortcomings in containing violence. The legacy of the autocratic state is exacerbated by the lack of strong democratic institutions, and value patterns that can deal with and contain violent conflicts. Consequently, the transition period should typically coincide with ‘booms’ of violent crime.

216

S. Karstedt

LaFree & Drass (2002) analysed ‘booms’ and ‘busts’ for a cross-national sample covering roughly the second half of the 20th century. In the section to follow I use their data to explore whether booms did coincide with the transition period in European and Latin American countries. The evidence presented here is based on a small number of countries, which were classified either as stable democracies or autocracies, or for whom transition periods and regime change could be identified. This rather rough classification of democracies and non-democratic societies was developed by Przeworski et al. (2000), who used the following criteria: general elections, party systems (more than one political party), and actual regime change based on elections. These criteria generated two types of democracies – parliamentarism and presidentialism – and two types of autocracies – dictatorships and party bureaucracies, the latter mostly signifying communist regimes. For the period from 1950 to 1990 they classified each country according to periods of regimes. LaFree & Drass (2002) identified crime trends, booms and busts for homicide rates (WHO homicide victimization rates per 100,000 of the population) covering a nearly identical period (1956 – 1997) for many of the countries included in Przeworski et al.’s study (2000).6 These data will be used to explore how booms were distributed among stable democracies, stable autocracies and countries, which experienced a transition from autocracy/bureaucracy to democracy. Table 1 shows that most stable democracies experienced significant trends of increasing homicide rates, the exception being Japan with a decline of homicide rates. Exceptional increases (booms) were identified for one third of the democracies, among them France, Germany and Italy in Europe. Canada and the United States are among the non-European democracies that went through boom periods of high intensity with an average of at least 10% growth in three consecutive years between 1967 and 1971. However, two thirds of the transition countries experienced booms, which were not always and consistently related to the period of transition. Spain experienced a boom directly after its transition to democracy from 1975 to 1977, Poland two periods of exceptionally rising homicide rates in the decade before the final transition, and during the transition, as did Bulgaria, and Czechoslovakia. It seems that democracies experienced rising trends of lethal violence, whilst booms were clustered in transition countries and during the transition period. Since only two autocracies (Mexico and Singapore) are represented in Table 1, the impact of autocracies will be explored further by a comparison of time series for four European transition countries, Bulgaria, Czechoslovakia, Hungary and Poland. Figure 1 (adapted from LaFree & Drass 2002) shows that Poland and Bulgaria experienced some type of ‘booms’ of homicides before 1989, and all countries with the exception of Hungary experienced such a boom phase during

6

For detailed information on the data set and the definition of crime ‘booms’ and ‘busts’, see LaFree and Drass (2002).

Democratization and Violence

217

Table 1 Development of homicide rates in stable democracies, stable autocracies/bureaucracies and ‘transition countries’a Stable Stable autocracies/ Transition democracies bureaucracies countries Trend: increase 13 (62%) 0 (0%) 2 (22%) Trend: decrease/ 1 (5%) 1 (50%) 1 (11%) none Booms 7 (33%) 0 (0%) 6 (67%) Busts 0 (0%) 1 (50%) 0 (0%) Total 21 2 9 a Countries included: Stable Democracies: Australia, Austria, Belgium, Canada, Costa Rica, Denmark, England & Wales, Finland, France, Germany (FRG, West), Iceland, Ireland, Italy, Japan, Netherlands, New Zealand, Norway, Sweden, Switzerland, Trinidad and Tobago, United States Stable Autocracies/Bureaucracies: Mexico, Singpore Transition Countries: Bulgaria, Chile, Czechoslovakia, Greece, Hungary, Poland, Portugal, Spain, Venezuela. Sources: LaFree, Drass (2002); Przeworski et al. (2000).

the transition between 1989 and 1994.7 Steep increases of lethal violence are observed in all of the four East and Central European countries shortly before and during the transition. The unique experience of such ‘booms’ at the beginning of the transition phase has to be contextualised within the long-term trends of lethal violence in these countries as compared to European levels in general. The time series for Bulgaria, Czechoslovakia, Hungary and Poland cover the period from the mid-1950s to the mid-1990s. The ‘booms’ in Czechoslovakia, Bulgaria and Poland and the more steady increase in Hungary, which finally amounted to the marked increase during the transition phase, are clearly visible, but it is obvious that the increase during the transition started from different levels in each country. These ranged from a comparably low rate in Czechoslovakia to an exceptionally high rate in Bulgaria according to European standards. The transition ‘booms’ reflect the different base levels of the preceding period. Throughout the period covered– of which the transition comprises at about 10–15% – Bulgaria had the highest mean rate (2.94), and Hungary, with 2.55, had a considerably higher level than both Poland (1.51) and Czechoslovakia (1.23). The pre-transition time series in Fig. 1 show that by European standards, these rates are by no means exceptionally low, but are located more at the higher end of the European scale. For the same time period the established democratic societies in Western Europe like Denmark (0.87), Sweden (1.05), Netherlands (0.73), England and Wales (0.92), France (0.99) and Germany (1.16) are situated at the lower end of the scale, though all of them experienced a trend of increasing homicide rates. Austria (1.25) and Italy (1.47)

7

The analyses of LaFree and Drass (2002) indicate a trend rather than a boom for Hungary, and show that in contrast to the three other countries, Hungary did not experience a ‘high intensity’ boom either before or during the transition period.

218

S. Karstedt

Bulgaria 5.5 5.0 4.5 4.0 Homicides 3.5 /100.000 3.0 2.5 2.0 1.5 1.0 0.5 1955

1965

1975

1985

1995

1985

1995

Year Czechoslovakia 2.6 2.4 2.2

Czechoslovakia Czech Republic Slovakia

2.0 Homicides /100.000

1.8 1.6 1.4 1.2 1.0 0.8 1955

1965

1975 Year

Fig. 1 Homicide Victimization Rates for Four Transition Countries, 1956–1998 Adapted from LaFree and Drass (2002)

are close to Poland and Czechoslovakia, while Finland (2.77) is the only country among the European democracies of that time that comes close to the rates of Hungary and Bulgaria (LaFree & Drass, 2002).8

8

Computations from my own data set of WHO homicide rates from 1978 to 1990 produce slightly higher levels for Western European and Scandinavian countries, but for the same period, the time series in Fig. 1 equally show increasing levels for Bulgaria, Hungary and Poland as well, but not for Czechoslovakia.

Democratization and Violence

219

Hungary 4.5 4.0 3.5 Homicides /100.000

3.0 2.5 2.0 1.5 1955

1965

1975 Year

1985

1995

Poland 3.0

2.5

Homicides /100.000

2.0

1.5

1.0

0.5 1955

1965

1975

1985

1995

Year

Fig. 1 (continued)

For these countries, the trends and booms of homicide rates indicate that already prior to the transition Eastern European countries had homicide levels above those of most Western European democracies. Taking into account that they had installed tight controls restricting the various forms of potentially lethal violence, the relatively high homicide rates in these communist countries indicate considerable levels of tensions and those anomic

220

S. Karstedt

tendencies that have been identified for autocracies. If we further assume that these rates underestimate the ‘real’ homicide rate by neglecting lethal violence by control agents, the secret police, and in prisons, and if these deaths would be added to the rates of the Eastern European countries, it is more than likely that the ‘real’ level of lethal violence was higher under communist regimes than these data indicate (see also Pridemore, 2003). Within the European context of generally low levels of lethal violence, the contrast between homicide rates for the Western European democracies and the non-democratic countries of Central and East Europe seems to confirm that autocratic and authoritarian states have higher levels of lethal violence. Eisner’s homicide data for 19th and 20th century Europe further corroborate higher rates for countries under communist regimes in the second half of the 20th century. Three countries – Bulgaria, Hungary and Poland – experienced a trend of increasing rates starting in the 1970s, and Hungary and Poland both registered a sharp increase in homicide rates during the 1980s. Though in all four countries homicide rates peaked during the early transition years, the upward trend in the latter two countries began well before the transition, indicating an antecedent increase of anomic tendencies and social problems connected to homicide rates (see Walberg et al., 1998, for Russia). For Poland, the increase in lethal violent crime coincides with the civil strife and unrest, and the imposition of martial law in the 1980s. Table 2 shows how booms are distributed across stable democracies, autocracies, and transition periods. More ‘boom’ years are found in the transition period than during the preceding autocratic regime. However, booms under autocratic regimes did not reach the level of stable democracies. Stable democracies experienced the longest periods of booms, which mainly started during the 1960s and continued until the 1990s. Anomic tendencies of autocratic regimes can be identified for particular cases, where they are related to intense social and political conflicts, particularly in Poland and Czechoslovakia, and these finally erupted in booms during the transition period. Hungary however Table 2 Intense boom periodsa in stable democracies and transition countries Number of Number Average years boom periods of years per country Stable Democracies 12 36 7.2 Transition Countries during 3 11 5.5 Autocracy/Bureaucracy Transition Countries during 8 27 4.5 Democracy/ Transition Period Total 23 74 – a At least ten percent homicide growth in three consecutive years. Countries included: Stable Democracies: Canada, Costa Rica, Italy, Trinidad and Tobago, United States Transition Countries: Bulgaria, Czechoslovakia, Greece, Poland, Spain, Venezuela. Sources: LaFree, Drass (2001); Przeworski et al. (2000).

Democratization and Violence

221

with a slow build-up of its transition during the 1980s and step-wise introduction of free market policies, suffered from a considerable increase in violent crime, which nonetheless did not amount to a boom. During this period, stable democracies experienced trends of increasing lethal violence, sometimes as ‘booms’ albeit from considerably low levels. Autocratic regimes in Europe had and retained comparably high levels of lethal violence, and experienced most of their booms during the transition period.

Advantages of Democracy? The arguments and evidence presented can be taken only as a first step toward a systematic exploration of the links between democracy, democratisation and violence. During the period of the second half of the 20th century, democracies experienced increasing trends of lethal violent crimes, and ‘booms’ were mostly observed for democracies (see also LaFree & Tseloni, 2006). From the perspective of the explanatory framework of ‘liberal inclusion’, these indicate rising tensions within their cultural, institutional and structural patterns. A number of causal factors impacted on mature democracies: the loss of legitimacy and civil strife during the 1960s (LaFree, 1998), imbalances between market and other institutions since the 1980s, rising social inequality and, in particular in the democracies in Europe, the experience of new forms of economic discrimination and cultural conflict through the influx of large numbers of immigrants (Karstedt, 2001, 2005). Though more rarely, waves and booms of lethal violent crime also hit the autocracies in Europe, and seem to be related to rising anomic tendencies, as well as to social and political conflicts as in Poland. However, evidence for comparative advantages of democracies in containing violence is provided by the fact that the Eastern European countries had consistently higher levels of violence than the mature democracies of Western Europe. This characteristic feature of the European landscape becomes obvious in the 19th century and continues throughout the 20th century. With the exception of Russia, which had a high level of violence by European standards throughout the period, the Central and Eastern European states were subjected to different kinds of colonial rule during the 19th century, and shared more features with repressive than liberal inclusive regimes. Steep increases and booms of lethal violence are a defining feature of the transition from autocracy to democracy. Both, the legacy of the past autocratic regime and the role of nascent democratic institutions and practices respectively their deficiencies are emphasised by the fact that such booms occur independently of a simultaneous transition to a market economy; both types of transitions are equally represented in the selection of countries (see Tables 1 and 2), and the examples of Spain and South Africa are pertinent. The case studies of four Eastern and Central European countries show that transitional booms of

222

S. Karstedt

violence are as much a legacy of the autocratic state as they are caused by the new democratic regimes, with the balance of responsibility shifting towards the fallen autocratic regime. Transition countries often tend to turn to the instruments and traditions of ‘repressive inclusion’ in their efforts to deal with the wave of violence. Latin American countries are emblematic for the ‘abdication of democratic authority’ (Mendez, 1999, p. 19), and the development of ‘disjunctive democracy’ (Caldeira & Holston, 1999). However, the potential dynamics of institutional, cultural and structural inclusion that restrain violence in democracies, as well as their comparative advantages point into the opposite direction: Not less, but more democracy seems to be the solution to booms and waves of violence during the transition period. Democracies seem to have more capacities to contain interpersonal violence than autocracies in the long run.

References Caldeira, T. P. R. & Holston, J. (1999). Democracy and Violence in Brazil. Comparative Studies in Society and History, 41, 691–729. Cohen, D. (2001). Chasing The Red, The White And The Blue: A Journey In Tocqueville’s Footsteps Through Contemporary America. New York: Picador. Davis Rodriguez, C. (2006). Civil Democracy, Perceived Risk and Insecurity in Brazil: an Extension of the Systemic Social Control Model. In S. Karstedt & G. LaFree (Eds.), Democracy, Crime and Justice. The Annals of the American Academy of Political and Social Sciences, 605, 242–263. Eisner, M. (1997). Das Ende der zivilisierten Stadt. Die Auswirkungen von Modernisierung und urbaner Krise auf Gewaltdelinquenz. Frankfurt: Campus. 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, 618–638. Eisner, M. (2003). Long-Term Historical Trends in Violent Crime. In M. Tonry (Ed.), Crime And Justice. An Annual Review of Research, 30, 83–142. Elster, J. (Ed.) (1999). Deliberative Democracy. Cambridge: Cambridge University Press. Ewald, U. (2000). Criminal Victimization and Social Adaptation in Modernity: Fear of Crime and Risk Perception in the New Germany. In T. Hope & R. Sparks (Eds.), Crime, Risk And Insecurity: Law And Order In Everyday Life And Political Discourse (pp. 197–219). London and New York: Routledge. Gellner, E. (1994). Conditions Of Liberty: Civil Society and its Rivals. New York: Penguin. Gibson, J. L. & Duch, R. M. (1994). Postmaterialism and the Emerging Soviet Democracy. Political Research Quarterly, 47, 5–39. Gurr, T. R. (1993). Minorities at risk. A global view of ethnopolitical conflicts. Washington D.C.: United States Institute of Peace. Gurr, T. R. (2000). People versus states. Minorities at risk in the new century. Washington D.C.: United States Institute of Peace. Hagan, J. & Radoeva, D. (1998). Both too Much and too Little: From Elite to Street Crime in the Transformation of the Czech Republic. Crime, Law & Social Change, 28, 195–211. Hagan, J., Hefler, G., Classen, G., Boehnke, K. & Merkens, H. (1998). Subterrenean Sources of Subcultural Delinquency Beyond The American Dream. Criminology, 36, 309–342. Holmes, S. (1993). The Anatomy Of Antiliberalism. Cambridge, Mass: Harvard University Press.

Democratization and Violence

223

Huggins, M., Haritos-Fatouros, M. & Zimbardo, P. G. (2002). Violence Workers: Police Torturers And Murderers Reconstruct Brazilian Atrocities. Berkeley: University of California Press. Inglehart, R. (1997). Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies. Princeton: Princeton University Press. Inglehart, R. & Welzel, C. (2005). Modernization, Cultural Change and Democracy. Cambridge: Cambridge University Press. Inkeles, A. (1993). On Measuring Democracy. Its Consequences And Concomitants. New Brunswick and London: Transaction Publishers. Jaggers, K. & Gurr, T. (1995). Tracking Democracy’s Third Wave with the Polity III Data. Journal of Peace Research, 32, 469–482. Karstedt, S. (1999). Social Transformation and Crime: A Crisis of Deregulation. In M. Krygier & A. Czarnota (Eds.), The Rule of Law After Communism. Problems and Prospects in East-Central Europe (pp. 308–326). Aldershot: Ashgate. Karstedt, S. (2001). Die moralische Staerke schwacher Bindungen: Individualismus und Gewalt im Kulturvergleich, Monatsschrift fuer Kriminologie und Strafrechtsreform, 84, 226–243. Karstedt, S. (2003). Legacies of a Culture of Inequality: The Janus Face of Crime in PostCommunist Countries. Crime, Law and Social Change, 40, 295–320. Karstedt, S, (2005). Individualisme et Violence: Modernisation Extreˆme ou ReTraditionalisation de la Socie´te´? Une Comparaison Interculturelle. De´viance et Socie´te´, 29, 273–284. Karstedt, S. (2006). Democracy, values and violence: Paradoxes, tensions and comparative advantages of liberal inclusion. In S. Karstedt and G. LaFree (Eds.), Democracy, Crime and Justice. The Annals of the American Academy of Political and Social Sciences, 605, 50–81. Keane, J. (2004). Violence and Democracy. Cambridge: Cambridge University Press. Kharkhordin, O. (1999). The collective and the individual in Russia: A study of practice. Berkeley, CA: University of California Press. Klages, H. (1984). Wertorientierungen im Wandel. Ru¨ckblick, Gegenwartsanalyse, Prognosen. Frankfurt: Campus. LaFree, G. (1998). Losing Legitimacy. Street Crime and the Decline Of Social Institutions in America. Boulder, Col: Westview Press. LaFree, G. & Drass, K. A. (2002). Counting Crime Booms Among Nations: Evidence for Homicide Victimization Rates 1956 – 1998. Criminology, 40, 769 – 800. LaFree, G. & Tseloni, A. (2006). Democracy and Crime: A Multilevel Analysis of Homicide in 44 Countries. In S. Karstedt & G. LaFree (Eds.), Democracy, Crime and Justice. The Annals of the American Academy of Political and Social Sciences, 605, 26–49. Lauth, H.-J., Pickel, G. & Welzel, C. (2000). Demokratiemessung. Konzepte und Befunde im internationalen Vergleich. Wiesbaden: Westdeutscher Verlag. Lederman, D., Loayza, N. & Menedez, A. (2002). Violent Crime: Does Social Capital Matter. Economic Development and Cultural Change, 50, 509–539. Los, M. (1988). Communist Ideology, Law and Crime. London: Macmillan Press; New York: St. Martin’s Press. Los, M. (2003). Post-communist Fear of Crime and the Commercialization of Security. Theoretical Criminology, 6, 165–88. Mann, M. (1970). The Social Cohesion Of Liberal Democracies. American Sociological Review, 35, 423–439. Mann, M. (2005). The Dark Side of Democracy. Explaining Ethnical Cleansing. Cambridge: Cambridge University Press. Mendez, J. E. (1999). Problems of Lawless Violence: Introduction. In J. E. Mendez, G. O’Donnell & P.S. Pinheiro (Eds.), The (un)rule of law and the underprivileged in Latin America (pp. 1–25). Notre Dame: University of Notre Dame Press.

224

S. Karstedt

Mendes, J. E., O’Donnell, G. & Pinheiro, P. S. (1999). The (un)rule of law and the underprivileged in Latin America. Notre Dame: University of Notre Dame Press. Messner, S. F. (1982). Societal development, social inequality and homicide. A cross-national test of a Durkheimian model. Social Forces, 61, 225–240. Messner, S. F. (2004). An Institutional-anomie theory Of Crime: Continuities And Elaborations in the Study Of Social Structure And Anomie. In D. Oberwittler & S. Karstedt (Eds.), Soziologie der Kriminalita¨t (pp. 93–109). Special volume 43 of the Ko¨lner Zeitschrift fu¨r Soziologie und Sozialpsychologie. Wiesbaden: Westdeutscher Verlag. Messner, S. F. & Rosenfeld, R. (1994). Crime and the American Dream. Belmont, CA: Wadsworth Messner, S. F. & Rosenfeld, R. (1998). Social Structure And Homicide. In M. D. Smith & M. A. Zahn (Eds.), Homicide: A Sourcebook Of Social Research (pp. 27 – 41). New York: Sage. Muller, E. N. & Seligson, M. A. (1994). Civic Culture And Democracy: The Question of Causal Relationships. American Political Science Review, 88, 635–652. Nowotny, T. (2002). Markets, Democracy and Social Capital. O¨sterreichische Zeitschrift fu¨r Politikwissenschaft, 31, 217–229. Pridemore, W. A. (2003). Measuring Homicide in Russia: A Comparison of Estimates from the Crime and Vital Statistics Reporting Systems. Social Science & Medicine, 57, 1343–1354. Pridemore, W. A. & Kim, S.-W. (2006). Democratization and Political Change as Threats to Collective Sentiments: Testing Durkheim in Russia. In S. Karstedt & G. LaFree (Eds.), Democracy, Crime and Justice. The Annals of the American Academy of Political and Social Sciences, 605, 82–103 Przeworski, A., Alvarez, M. E., Cheibub, J. A. & Limongi, F. (2000). Democracy and Development. Political Institutions and Well-Being in the World, 1950–1990. Cambridge: Cambridge University Press. Rose, R. (1998). Getting things Done In An Anti-Modern Society: Social Capital Networks In Russia. Social Capital Initiative Working Paper, 6. Washington: World Bank. Rosenfeld, R., Messner, S. F. & Baumer, E. P. (2001). Social Capital and Homicide. Social Forces, 80, 283–309. Rothstein, B. & Stolle, D. (2002). How Political Institutions Create and Destroy Social Capital. An Institutional Theory of Generalized Trust Project on Honesty and Trust. Workshop 2, Collegium Budapest. Budapest Sampson, R. & Bartusch, D. J. (1998). Legal Cynicism and (Subcultural?) Tolerance of Deviance: The Neighborhood Context of Racial Differences. Law & Society Review, 32, 777–804. Sampson, R. J., Raudenbusch, S. W. & Earls, F. (1997). Neighborhoods And Violent Crime: A Multilevel Study Of Collective Efficacy. Science, 277, 918–924. Savolainen, J. (2000). Inequality, Welfare State, and Homicide: Further Support for the Institutional Anomie Theory. Criminology, 38, 1021–1042. Steinberg, J. (2001). Crime Wave. The South African Underworld and Its Foes. Witwatersrand: Witwatersrand University Press. Sztompka, P. (1993). Civilizational Incompetence: The Trap of Post-Communist Societies. Zeitschrift fuer Soziologie, 22, 85–95. Tocqueville, A. (2000 [1835 and 1840]) Democracy in America. Chicago: Chicago University Press. Tyler, T. R. (1990). Why People Obey the Law. New Haven, Conn: Yale University Press. Tyler, T. R. (2003). Procedural Justice, Legitimacy, and the Effective Rule of Law. In M. Tonry (Ed.), Crime and Justice: A Review of Research (pp. 283–357) Chicago: University of Chicago Press. Tyler, T. R. & Blader, S. (2000). Cooperation in groups: Procedural justice, social identity, and behavioral engagement. Philadelphia: Psychology Press.

Democratization and Violence

225

Walberg, P., McKee, M., Shkolnikov, V., Chenet, L. & Leon, D. A. (1998). Economic Change, Crime, and Mortality Crisis in Russia: Regional Analysis. British Medical Journal, 317, 312–318. Waldmann, P. (2002). Der anomische Staat. U¨ber Recht, o¨ffentliche Sicherheit und Alltag in Lateinamerika. Opladen: Leske und Budrich. Waldmann, P. (2003). Diktatur, Demokratisierung und soziale Anomie. Mu¨nchen: Vo¨gel. Welzel, C. (2002). Fluchtpunkt Humanentwicklung. U¨ber die Grundlagen der Demokratie und die Ursachen ihrer Ausbreitung. Wiesbaden: Westdeutscher Verlag.

Violence and Violence-Control in Long-Term Perspective: ‘Testing’ Elias in Relation to War, Genocide, Crime, Punishment and Sport Eric Dunning

Introduction I shall start this chapter by exploring some conceptual issues related to the thorny problem of defining ‘violence’ and related terms such as ‘aggression’. I shall then become more concrete and discuss various manifestations of violence using a long-term perspective. More particularly, I shall discuss aspects of war, genocide, crime, punishment, and violence in and in relation to leisure and sport, basing my arguments mainly on secondary sources.1 It is necessary to insert two caveats at the outset. The first is that, whilst war, genocide, violent crime and sports-related violence have historically been mainly male activities, females have sometimes been centrally involved as well. However, lack of space precludes an exploration of the gender dimensions of violence in this context.2 The second caveat is that this list of activities and atrocities may seem like a rag-bag of unrelated or only partly related issues which differ from each other markedly in terms of their degrees of seriousness and significance. I want to suggest, though, that apart from some obvious differences, there are similarities and connections among them which have not always been recognized but which become clearer if: (i) the issues involved are viewed from an historical or long-term perspective; and (ii) the occurrence of pleasurable and playful forms of violence at the levels of both fact and fantasy is acknowledged more explicitly as sometimes existing correlatively with the more ‘serious’ forms than is usually done by contributors to the overall field of violence studies. Throughout the chapter, I shall concern myself with the implications of my discussions for Elias’s (1939; 2000) theory of ‘civilizing processes’. That theory, Eric Dunning Emeritus Professor of Sociology, University of Leicester; Visiting Professor, University of Chester, Department of Sociology, University of Leicester, University Road, Leicester, LE1 7RH 1

An obvious omission from this collection of topics is the various forms of terrorism. Just to mention one horrifically infamous example, some 150 women were among the guards at the Auschwitz death camp. One of them, Ilse Greser, was particularly noted for her sadism and was hanged after the war.

2

S. Body-Gendrot, P. Spierenburg (eds.), Violence in Europe. Ó Springer 2008

227

228

E. Dunning

the basic premisses on which it is founded and related constructs will inform most of what I have to say. In fact, this chapter can be seen in a non-positivistic sense as a ‘test’ of key parts of Elias’s theory, those involving trends in violence and violence-control.

A Deeply Rooted Contemporary Belief It is widely believed that we are living today in one of history’s most violent periods. Indeed, it is probably fair to say that, in Western societies at least, the fear that we may be currently undergoing a process of ‘de-civilization’, above all of mounting violence, is deeply imprinted in the contemporary Zeitgeist, one of the dominant beliefs of our times. In the opening chapter of their recently published International Handbook of Violence Research (2003:5), for example, the editors Heitmeyer and Hagan wrote that: ‘In Western societies, the dream of a non-violent modern age clashes with a reality that is massively overshadowed, if not totally plunged into darkness by overt acts of violence and the potential for destruction . . .’ Consistently with this, the following was reported in The Observer, a British newspaper, on Sunday 13 November 2005: A culture of violence in Britain is to blame for an epidemic of school bullying that is devastating the lives of millions of children, according to . . . one of the country’s leading experts on young people. In his first major interview as the new Children’s Commissioner for England, Al Aynsley Green . . . said: ‘I have no doubt that children are being brought up in a society where violence is the norm in many ways. I include in this the violence on television, in the workplace and in the home.’

A moral panic rooted in concern over increasing gun-use by black male adolescents was added to the worry regarding bullying in 2007. It is important though to recognize that similar concerns have been around since at least the 1970s. For example, the psychologists Eysenck and Nias wrote in 1978 of ‘a number of acknowledged facts’ which, they claimed, ‘have helped to persuade many people that the civilization in which we live may be in danger of being submerged under a deluge of crime and violence’ (Eysenck and Nias, 1978:17). From a figurational standpoint, of course, ‘civilization’ is always faced potentially with the danger of collapse. ‘Civilizing controls’ are never more than a relatively thin shell. That is why figurational sociologists lay stress on increasing the understanding, not only of ‘civilizing processes’, but of ‘de-civilizing’ and ‘dys-civilizing’ processes as well.3 3

The concept of a ‘de-civilizing process’ refers to a civilizing process that goes into reverse. That of a ‘dys-civilizing process’ refers to a ‘‘civilized’’ regime which creates and maintains compartments of destruction and barbarism, in meticulous isolation, almost invisible and well-nigh unmentionable. It is as if the civilizing process continues with the same means, but with a different turn: in one word, it has become a dyscivilizing process.’ (de Swaan, 2003:141). This concept is designed to recognize the fact that state-formation does not necessarily produce civilizing consequences and that state monopolies of violence can be used in the pursuit of ‘barbaric’, ‘uncivilized’ ends.

Violence and Violence-Control in Long-Term Perspective

229

Arguing from a psychological perspective different from that of Eysenck and Nias, Peter Marsh contended, also in the late 1970s, that then-recent social developments in Britain had led to a decline in opportunities for ‘socially constructive ritual violence’ — what he called ‘aggro’ — with the consequence that uncontrolled and destructive violence had increased. Using a variation on Erich Fromm’s distinction between ‘benign’ and ‘malignant’ aggression (Fromm, 1973), Marsh argued that there had taken place a ‘drift from ‘‘good’’ violence into ‘‘bad’’ violence’. People, he said, are ‘about as aggressive as they always were but aggression, as its expression becomes less orderly, has more blood as its consequence’ (Marsh, 1978b:142). How it is possible to reach the conclusion that aggressiveness is an historically fixed quantum without historical study is something that Marsh fails to tell us. Clearly he descended with at least this aspect of his case from science to ideology. A similar dose of scare-mongering ideology recurred in a judgement in a book about sport made by Australian journalist Don Atyeo in 1979. Atyeo detected strong parallels between trends in modern sport and trends in their counterparts in Ancient Rome. He argued that a self-destructive trend towards escalating violence is occurring in modern sport, principally as a result of the demands of sensation-seeking spectators. What he wrote remains one of the clearest expressions of a still commonly held belief and is, on that account, worth citing. Atyeo expressed his apocalyptic vision by suggesting that ‘the future of violent sports seems assured’. They will, he argued, grow harder and bloodier in order ‘to feed the rising appetite of an audience which will grow increasingly more jaded and satiated with violence, and increasingly more violent itself’. It will then all come crashing down. This time, though, it won’t be ‘the barbarian hordes banging on the gates outside which will destroy the Coliseum. . . . the violence will be of sport’s own making and will come from within the walls of the Coliseum itself’. (Atyeo, 1979:377). It will be one of my contentions in this chapter that, while a worldwide trend towards increasing violence is undoubtedly currently occurring both in sport and elsewhere, diagnoses such as those of Eysenck, Nias, Marsh, Atyeo and Aynsley involve greater or lesser elements of sensationalism and moral panic and exaggerate the extent and seriousness of the trend to a greater or lesser degree. A discussion of some conceptual and theoretical issues will hopefully help to prepare the way for showing why that is the case.

Towards a Typology of Human Violence The types of violence engaged in by humans are diverse and complex. In their otherwise excellent and thought-provoking chapters in this volume, Willem de Haan and Pieter Spierenburg have argued for approaches to violence that see it, respectively, as ‘an essentially contested concept’ (De Haan), but one which should be restricted solely to ‘physical attacks’ and which is probably not

230

E. Dunning

applicable to non-human animals (Spierenburg). I agree with much of what both authors say. Neither are the cases that they argue for as mutually contradictory as may at first appear. However, that residual elements of metaphysical ‘mindbody’ dualism enter into both of their approaches can arguably be seen from the fact that De Haan distinguishes in his conclusion between ‘physical’ and ‘nonphysical’ violence’, whilst Spierenburg tells us that he is ‘pleading for a restrictive definition of violence, which essentially limits it to the encroachment upon a person’s physical integrity’. Although both authors would probably seek to deny it, arguably inherent in their approaches and much modern thinking in this area is the idea that phenomena such as ‘verbal violence’, ‘symbolic violence’ and ‘mental cruelty’ are somehow ‘non-physical’. This raises the question of where they are supposed to reside? In a ‘realm of spirit’? In ‘heaven’ as in Plato’s theory of forms? In the ether? Viewed scientifically, of course, our so-called ‘minds’ and ‘emotions’ are functions of our physical bodies. Moreover, physical processes are involved in the production and hearing of sounds, and the ‘rational’ and ‘emotional’ experiences of individuals are ‘embodied’ and ‘physical’, too. Although Elias used the concepts of violence and Gewalt in a general and undifferentiated way, I want to suggest that a thoroughgoing application of his process-sociological or figurational approach which is concerned with dynamic balances, gradations and degrees and does not fall foul of ‘mind-body dualism’ can help us to gain a greater purchase on the issues involved than is possible through searching for a single, universal definition of violence or a narrow range of contested definitions. More particularly, a move towards a more comprehensive understanding of violence will arguably be achievable if distinctions are drawn among separable forms and dimensions of violence such as the following: (i) the means employed; (ii) the actors’ motives and values; (iii) the actors’ psychological states, especially their levels of psycho-physical emotional arousal and the degrees of intentionality, rationality and control involved in their actions; (iv) the normative/value parameters within which the violence takes place; and (v) the size, character and degrees of organisation of the social fields within which it occurs. Seen in these terms, at least ten distinctions can be provisionally made among the forms of human violence, namely: 1. Whether the violence is actual/’real’ or ‘symbolic’, that is whether it takes the form of a direct physical assault or simply takes a verbal or gestural/ facial posturing form. It is worth noting in this connection that physical assault is often accompanied by verbal violence, screaming and/or shouting and contorted facial expressions. It can, of course, also be accompanied by the facial and other physical/physiological signs of pleasure;4 4

Recurrently engaged-in facial expressions can become embodied at the level of habitus as Elias shows in the case of medieval warriors (Elias, 2000:161ff). Of course, in cases such as terrorist bombing, suicide bombing and forms of violent crime such as bank robberies, the perpetrators usually have to exercise a high degree of self-control and seek to mask their violent intentions until the moment comes to strike.

Violence and Violence-Control in Long-Term Perspective

231

2. Whether the violence takes a ‘play’ or ‘mock’ form or whether it is ‘serious’ or ‘real’. This dimension clearly overlaps with dimension (1). It might also be captured by means of the distinction between ‘ritual’ and ‘non-ritual’ violence, though it has to be noted that, pace Marsh and his colleagues (1978b), ritual and play can both be seriously violent in intent and content. The most seriously committed football hooligans, for example, invariably seek to hurt their opponents and not infrequently succeed; 3. Whether or not a weapon or weapons is/are used and where they are used, the level of technological knowledge and expertise that was involved in their manufacture, e.g. swords and bows and arrows in contrast to rocket-propelled nuclear weapons; 4. Where weapons are used, whether or not the combatants come directly into contact. Related to this dimension is the degree of distance between combatants who are not in direct contact, especially whether they are able to see and/or hear each other or not; 5. Whether animals are involved in the violence either as targets/victims of the human perpetrators or as trained helpers of the latter. In foxhunting, for example, foxes are the targets/victims while hounds are the killers, trained by their human masters. Similarly in Ancient, Medieval and Early Modern warfare, horses and in some cases elephants played a crucial part; 6. Whether the violence is planned and intentional or the accidental consequence of an action sequence that was not intentionally violent at the outset. For example, if pacifying gestures and language are not swiftly offered, an accidental collision between two people walking along a street or playing in a football match may spark a physically and verbally violent exchange; 7. Whether one is dealing with violence initiated without provocation or with a retaliatory response to an intentionally or unintentionally violent act or perhaps to a real or perceived insult, aggressive facial expression or menacing bodily posture; 8. Whether the violence is legitimate in the sense of being in accordance with a set of socially prescribed rules, norms and values, or whether it is nonnormative or illegitimate in the sense of involving the contravention of accepted social standards. A subcategory of this dimension involves the illegitimate use of violence by agents who have been officially licensed by a state, e.g. the use of violence by police or members of the military in a democratic state to obtain a confession. In totalitarian states such as Nazi Germany, of course, such violence is legitimate; 9. Whether the violence takes a more ‘rational’ or a more ‘affective’ form, that is whether it is rationally chosen as a means for helping to secure the achievement of a given end or goal, or engaged in as an emotionally arousing, pleasurable and satisfying ‘end in itself’. Another way of conceptualizing this dimension would be to distinguish between violence in its more ‘instrumental’ and more ‘expressive’ forms;

232

E. Dunning

10. Whilst violence can be inflicted by persons on themselves whether they are alone or in company, most violence is directed at others and takes place in the context of a social field characterized by chains or networks of interdependence that vary in terms of their length, density, degrees of organization and the balance within them between centripetal and centrifugal pressures5 Following Max Weber, some sociologists would call these distinctions ‘ideal types’ but it is preferable from a figurational standpoint to conceptualize them in terms of interconnected polarities and balances. Let me briefly discuss the concepts of violence and aggression.

A Figurational ‘Take’ on the Concepts of Violence and Aggression ‘Aggression’ is a term which comes out of psychology and it is possible to use it in a constructive sense, whereas ‘violence’ is a popular term which is invariably used with negative connotations. Examples – both of which I have mentioned already – are Fromm’s distinction between ‘benign’ and ‘malignant’ aggression and Marsh’s derivation of ‘aggro’ from ‘aggression’ and/or ‘aggravation’ to refer to violence, e.g. that of football hooligans, that he regards as ritualized and socially constructive. I am less concerned here about terminological niceties, however. I want, rather, briefly to discuss the idea of writers such as Sigmund Freud (1932) and Konrad Lorenz (1967) that humans have an ‘aggressive drive’ or ‘instinct’. Norbert Elias (1988) wrote persuasively on this issue from a sociological and accordingly more reality-congruent standpoint. The idea that humans have an innate aggressive drive which structurally resembles the sex and hunger drives, Elias said, is a false way of posing the problem. What we do have is an ‘innate potential to shift (our) whole physical apparatus to a different gear if (we) feel endangered’. This is the so-called ‘fight-flight mechanism’ through which the human body reacts to danger ‘by an automatic adjustment which prepares the way for intensive movement of the skeletal muscles, as in combat or flight’. According to Elias, however, human ‘drives’ such as the hunger or sex drive are released physiologically, ‘relatively independently of the actual situation’ in which people find themselves. By contrast, the shifting of the body economy ‘to combat- or flight- readiness is conditioned to a far greater extent by a specific situation whether present or remembered’. Such situations can be ‘natural’, for example being attacked by a wild animal, or social, 5

Even when people are alone they are usually parts of a social field. They may, for example, live on their own but have neighbours; or they may spend time on their own in their bedroom. The only partial exceptions are provided by people such as hermits who successfully manage completely to eschew the company of others. Degrees of individualization/aloneness such as are involved in having one’s own bedroom are, in fact, among the variable properties of all social fields.

Violence and Violence-Control in Long-Term Perspective

233

especially conflict.6 However, in conscious opposition to Freud, Lorenz and others who ascribe an aggression drive to people on the model of the sex drive, it is not, said Elias, ‘aggressiveness’ that triggers conflicts but conflicts that ‘trigger aggressiveness’ (Elias, in Keane [ed] 1988:177–8). Of course, there was a degree of rhetorical exaggeration in this. Elias would not have denied that some conflicts result from the disruptiveness of aggressive individuals. He also stressed the interdependence of different human drives (Elias, 1939; 2000:161). In other words, he was aware of the interconnectedness of the sex, hunger and thirst drives and aggression, particularly if the former are frustrated. What he wanted above all to do was to counter the crude psychological reductionism involved in the notion of an ‘aggressive instinct’. Summing up, it was Elias’s contention that whether the ‘fight-flight mechanism’ is directed into fighting or fleeing is fundamentally a question of culture and social learning. More particularly, it is a question of the degree to which the values of the society or group into which one is born or of which one later becomes a member, e.g. through immigration, lay stress on the violent as opposed to the peaceful end of the continuum between extremely violent and wholly peaceful means of handling tensions and conflicts. Applying some of the sorts of conceptual distinctions I have been discussing, let me now turn to the issue of war seen in long-term perspective.

Some Sociological Reflections on Warfare Viewed in Long-term Perspective I want to examine two main things regarding war in this context: the old argument which has resurfaced in some sociological, anthropological, archaeological and social psychological circles in recent years that it is misleading to talk, as Elias does, of ‘civilizing processes’ as involving ‘pacification’ or the imposition of stricter controls on the use of violence because the opposite is true. According to the supporters of this view, far from being conducive to pacification, civilizations and states are in fact among the major causes of violence and war, if not the major causes. Elias, it is implied, took a ‘Hobbesian’ stance, whereas a ‘Rousseauian’ standpoint is more ‘object-adequate’. According to subscribers to the latter view, in societies that present-day Westerners label as ‘primitive’, ‘savage’ or ‘uncivilized’, that is, the structurally simpler, ‘pre-literate’, ‘non-literate’, ‘tribal’ or ‘pre-state’ societies of yesterday and today, war is or was a non-existent, rare or trivial occurrence, a ritualized and quasi-sport-like affair in which few people were hurt or lost their lives. 6

It is, of course ‘natural’ for humans to be ‘social’. That is to say, we evolved biologically as a species that is dependent on social artefacts such as language and with built-in aids to sociability such as smiling and laughing. Adults can, of course, learn to use smiling, laughing and language in what we call ‘anti-social’ ways but these, too, are a variety of ‘social facts’.

234

E. Dunning

American sociologist, William Graham Sumner, appears to have been among the first to adopt such a ‘Rousseauian’ stance. Thus he wrote in 1911 that ‘primitive man ‘‘might be described as a peaceful animal’’ who ‘‘dreads’’ war’ (cited in Keeley, 1996:8). More recently, such a view received powerful support from English social psychologist, Peter Marsh, in the book Aggro (1978a) which I cited earlier. It is a book in which he tells us about what he thinks is the character of warfare among a typical warrior people, the Dani of New Guinea. Marsh’s exposition of the ‘Rousseauian’ case has seldom been bettered and is hence worth discussing at length. The fact that the Dani think of themselves as warriors, Marsh implies, adds strength to his case. Wars among the Dani are frequent, he tell us, and they start in the early morning with the warriors of one tribe issuing a challenge to the warriors of another. Such challenges are usually taken up but they will not be if the warriors think it is likely to rain or if there is no time to eat the ritual sweet potatoes which are central to the Dani warrior tradition. When they do accept a challenge, the Dani warriors paint themselves and put on their war head-dresses. Then, in a state of mounting excitement they advance to the pre-arranged battleground. When they reach it, the two groups stop at a distance from each other of some 500 yards. Then each advances in a stop-go fashion until they are at a distance from each other of some 50 or so paces. At this point, they throw their spears at each other and retreat. Arrows are fired as well but they have no flights, a fact which Marsh construes as resulting not from technological ignorance but as a primitive form of arms control. Marsh also likens the Dani warriors to present-day English football hooligans.7 According to Marsh, the death toll from a year of such fighting is usually between 10 and 20 warriors on each side. ‘In tribes where fighting provides a major focus of activities’, he writes, . . . ‘where battles are essential to the maintenance of their entire social fabric more people die from the common cold than from the spears and arrows of rival warriors’ (Marsh, 1978a:49). ‘Evidence’ of this kind leads Marsh to reach the conclusion that it is misleading to talk of ‘warfare’ among the New Guinea plains warriors because these battles ‘are not warfare in the sense we know it today. They are more like ‘‘skirmishes’’ or ‘‘raids’’.’ Violence of this kind does not, according to Marsh, have ‘the terrifying rationality associated with the struggle for survival in a world of scarcity and need. It arises as a solution to the problem of aggression’. As a functionalist sociologist might express it, war performs the ‘latent function’ of binding together a tribe because it is directed outwards. ‘By channelling the competitive hostility outwards towards the tribe on the other side of the hill, social bonds within one’s own group are reaffirmed and maintained’. Warriors who show character, solidarity and courage on the battlefield are rewarded with prestige and leading positions within the dominance hierarchy of their tribe. Such a framework may seem illogical, says Marsh, but it works. ‘And it copes 7

See Eric Dunning, Sport Matters (1999) for a systematic discussion of this issue.

Violence and Violence-Control in Long-Term Perspective

235

more effectively with the aggression of men than the seemingly more civilized social structures we find in our own society. Aggression is recognised and managed – and that is a rational approach’. (Marsha et al, 1978a:49, 50). I am not totally opposed to the argument Marsh puts forward here. The pattern of ‘in-group/out-group’, ‘we-group/they-group’ or ‘us/them’ bonding (Elias, 1978) that he describes is a common pattern in societies at all levels of social development (in the sense primarily of levels of structural complexity, scientific knowledge and technical ‘know-how’). What I do find problematic is how Marsh explicitly compares Dani warfare to modern football hooliganism, arguing that both are socially constructive outlets for male aggression. The supposedly ‘primitive’ Dani, Marsh implies, are wiser than we are in the modern West, because they recognize and manage aggression. Westerners allegedly deny it and try to stamp it out, partly, Marsh argues, as a result of feminist misunderstandings aimed at subduing the innate aggressiveness of males. What Marsh misses in this connection is arguably twofold: (i) the fact that ‘primitive’ peoples – or better, the members of structurally simpler societies – such as the Dani account for their warfare in magical-mythical terms, i.e. in terms of gods and spirits, not rationally as Marsh implies; and (ii) he fails to appreciate that it is not football hooliganism but sports such as football per se which, if the rules are followed and enforced, take on the character of a mock-fight and which have developed in ‘modern’ societies as major means of channelling aggression.8 Although it can contain elements of ritual in Marsh’s sense, football hooliganism is a potentially dangerous and destructive superimposition on what really do involve forms of what one might term ‘ritualized aggression’, namely highly regulated ‘modern’ sports per se. To my knowledge, one of the first to attempt to counter the ‘Rousseauian’ view adhered to by Marsh was the anthropologist, Bronislaw Malinowski, who wrote in 1941 that ‘anthropology has done more harm than good in confusing the issue by . . . depicting human ancestry as living in a golden age of perpetual peace’ (Cited in Keeley, 1996:8). The data-based arguments of Wrangham and Peterson (1997) and Keeley (1996), however, are considerably more damaging to Marsh’s case than this or my own earlier suggestions. Wrangham and Peterson, for example, cite a ‘global assessment’ of the ethnographies of thirty-one hunter-gatherer societies which found that ‘64% of them engaged in warfare once every two years, 26% fought wars less often, and only 10% were considered to fight wars rarely or never’ (Ember, 1978, cited in Wrangham and Peterson [1997:75]). Wrangham and Peterson, citing Chagnon (1988:986) and Eibl-Eibesfeldt (1989:417), further inform us that, ‘anthropologists have gathered statistics on warfare among ‘‘independent’’ peoples who were not ‘‘inhibited’’ either by governments or more powerful tribes’. Given the relative independence of these peoples and their relative lack of exposure to external 8

According to Elias, in more complex societies, magical-mythical ideas tend to remain as a substratum in human thinking and to resurface in seriously critical situations.

236

E. Dunning

constraints, it is reasonable to assume that the figures cited by Wrangham and Peterson cite are artefacts that provide relatively reliable and reality-congruent representations of this aspect of the lives of the tribal groups concerned. That is, they are not a consequence of preconceived ideas imposed on these groups by the anthropologists who studied them. Summarizing, Wrangham and Peterson conclude that: ‘Violence accounted for the deaths of about 19.5 per cent of adult men among the Huli of highland New Guinea; for the Mae Enga and the Dugum Dani, also of highland New Guinea, warfare produced adult male mortality rates of 25 per cent and 28.5 per cent respectively. For the Murngin of Australia, the figure was 28 per cent’ (Wrangham and Peterson, 1997:77). Keeley’s arguments are also seriously damaging to Marsh’s case. He (Keeley) firstly argues that, while the loss of 20 people per year in war may seem a very small number to, for example, the British who have a population of some 60 million, it is a substantial number—some 7–10% of the total—in a tribal group of some two to three hundred. It is also necessary, according to Keeley, to take note of the fact that large-scale formal battles of the kind which play a part in the wars of ‘civilized’ peoples, tend to take place less frequently in ‘primitive’ war than small raids and ambushes. Typically, says Keeley, these involve just small groups of males sneaking into ‘enemy territory’ to kill as many people, including women and children, as they can. According to Keeley: ‘One common raiding technique (favoured by groups as diverse as the Bering Straits Eskimo and the Mae Enga of New Guinea) consisted of quietly surrounding enemy houses just before dawn and killing the occupants by thrusting spears through the flimsy walls, shooting arrows through doorways and smoke holes, or firing as the victims emerged after the structure had been set afire’ (Keeley, 1996:65). Of course, the simple demonstration that primitive warfare can be more openly ferocious and, in a quantitative sense, more deadly than subscribers to the Rousseauian view hold it to be, does not prove anything regarding the contention that it is their contact with Western civilization that has made some so-called ‘primitive’ peoples warlike and violent. ‘Barbarization’ as a result of contact with the West has in all probability happened not infrequently. However, despite this probable fact, not only have Chagnon, Eibl-Eibesfeldt and Wrangham and Peterson cast doubt on it as shown in my earlier discussion but Keeley has, in my opinion, decisively refuted it. He has done so because, as an archaeologist, he has succeeded in proving conclusively that pre-state tribal peoples from all over the world engaged in violent warfare before both the modern era and contact with the West. He also shows conclusively that, whilst it may not be/have been totally universal, violent warfare has frequently occurred amongst members of our species since the earliest times. In the context of this chapter, it must be enough to illustrate this by summarising Keeley’s conclusion. The facts recovered by ethnographers and archaeologists, Keeley writes, show unequivocally that the warfare of prehistoric and tribal peoples was/is ‘just as terrible and effective as the historical and civilized version. War is hell whether it is fought with wooden spears or napalm’. (Keeley, 1996:174). In fact,

Violence and Violence-Control in Long-Term Perspective

237

pre-state societies were/are rarely peaceful and their adult males saw/see combat repeatedly in a lifetime. They prefer(red) deadly raids, ambushes, and surprise attacks on settlements to the more complex and—leaving aside bombing raids such as the nuclear attacks on Hiroshima and Nagasaki and the non-nuclear raids on the cities of all combatants during World War 2—less deadly battles that are so important as features of the warfare of peoples who believe themselves to be ‘civilized’. Indeed, one could say that ‘primitive’ warfare was/is more deadly than that of ‘civilized’ states because it is more frequent and because of the more merciless way it was/is conducted. The tactics of primitive war and its ‘cousin’, guerrilla warfare, are/were also superior to their ‘civilized’ variants. ‘It is civilized warfare that is stylized, ritualized and relatively less dangerous’, says Keeley. ‘When soldiers clash with warriors (or guerrillas), it is precisely these ‘‘decorative’’ civilized tactics and paraphernalia that must be abandoned if they are to defeat the latter’ (Keeley, 1996:175). This reference to the need to abandon ‘civilized norms’ and ‘civilized’ forms of warfare in order to combat guerrilla fighters ties in with the American experience in Vietnam, and more recently, with that of the Americans and British in Iraq. But how, if ‘primitive’ and guerrilla warfare really are as deadly and effective as Keeley maintains, have peoples who consider themselves to be ‘civilized’ managed regularly to conquer and colonize the pre-state and tribal peoples whom they perceive as ‘barbaric’ and ‘primitive’? According to Keeley, the answer is that, although they may lose particular battles, the economic surpluses, population sizes, transportation technologies, and the planning and logistical capabilities of ‘civilized’ peoples enable them to win the war. In a word, ‘civilized’ peoples may become less good as fighters but ‘civilization’ gives them the men, material and overall wherewithal to fight long and sustained campaigns. Let me now offer some figurational reflections on modern war.

Some Figurational Reflections on Modern War The first thing worthy of note in this connection is that Norbert Elias shared the general left-wing hatred of and horror over war. However, he was also critical of those on the left who glorify ‘revolution’ because, he insisted, ‘revolutions’ are a kind of war. That is, they, too, tend to involve people killing each other and lead to multiple unintended and often violent consequences. However, Elias was not by any means a pacifist. Violence, he argued, is necessary when, for example, a ruling group or ‘establishment’ and an ascendant group such as a class, ethnic group or hitherto subordinate gender become trapped in a ‘frozen clinch’ because the ruling group refuses to make concessions (Elias, 1969). There is also at least one sense in which the belief that the twentieth century witnessed a trend towards growing violence is based on solid foundations. More particularly, as an accompaniment of the increasing pace and scope of global social change, the 20th century was the first in which wars that were literally

238

E. Dunning

‘world wars’ took place. It was also a century in which the violence and effectiveness of the technology of mass destruction increased to hitherto unprecedented levels, a fact evidenced above all in nuclear weapons and the weaponry of chemical and germ warfare. The capacity to produce and deliver such weapons is, of course, a consequence of the application of advancing scientific knowledge and that, in its turn, is arguably in part a consequence of a ‘civilizing process’. More particularly, one aspect of a ‘civilizing process’ involves an advance in people’s capacity for self-distanciation (Elias, 1987) and this enables them to move beyond highly involved, magical-mythical views of the world towards perspectives that are more detached, more ‘reality-congruent’ and facilitate greater control. In order to appreciate this, of course, a relatively detached approach is required of the reader. The ability to exercise a high level of self-control, another consequence of a ‘civilizing process’, is also necessary in order to use many complex modern weapons. Further to this, a lot of modern warfare involves killing at a distance through the firing of rockets and the dropping of bombs on people whom one cannot see or hear. People who consider themselves to be ‘civilized’, do not, for the most part, like the blood and gore of face-to-face battles and most of them have to undergo training that amounts to a process of ‘decivilizing’ in order to be prepared for battle. Such individuals and those who are ‘de-civilized’ by the direct experience of war, often represent a threat to their societies of some magnitude when peace returns. The problems caused by Korean and, to a greater extent, Vietnam veterans in the USA are examples (Joas, 2003:111ff). It is also important to note that there have been numerous violent and destructive wars since 1945. However, they have been, for the most part, local in scope, restricted with the recent exceptions of the wars associated with the break-up of former Yugoslavia and the former USSR, mainly to Third World, formerly colonial countries. Without for one moment wishing to detract from the tragic reality of these wars for the people most directly involved, to say nothing of the often horrific aftermath of, for example, unexploded landmines, it is nevertheless arguably relevant to point out that many areas of the world, particularly in the West, have enjoyed hitherto unprecedented levels of peace and prosperity since the end of World War Two. Such peace and prosperity cannot, of course, be understood independently of the forms of neo-colonial domination and exploitation that derive from the power differentials that are inherent in our rapidly globalizing world. Writing in the late 1980s, just before the end of the ‘Cold War’, Norbert Elias commented on this peace, though not the prosperity, by suggesting that the question is often posed in discussions of physical violence of how men and women can become ‘terrorists’ and kill and maim others? It would be more productive, he suggested, if the question were re-phrased thus: ‘How is it possible that so many people can normally live together peacefully without fear of being struck or killed by stronger parties’ as is generally the case in the urban-industrial nation-states of today? It is often forgotten, he suggests, that never before in the history of humanity have so many people lived together so peacefully as is the case ‘in the large states and

Violence and Violence-Control in Long-Term Perspective

239

cities of our time. This becomes evident when one realizes how much more violent and how much higher in risk of physical attacks were earlier epochs of human development’ (Elias, in Keane [ed] 1988:178). Elias wrote this years before the so called ‘war on terror’ declared by the second President Bush. The male or female terrorists he (Elias) primarily had in mind were groups in Germany such as the ‘Red Brigade’ and the ‘Baader-Meinhof’ gang. They saw themselves as anti-fascist and cannot be understood independently of the culture, politics and tensions of post-Nazi Germany and the ‘Cold War’ era. Bush’s war on terror is post-Cold War, more international and partly rooted, firstly in the ‘First World/Third World’ division9 and the First World’s dependency on and greed for oil, and secondly, in a latter-day flaring up of the Christianity-Islam conflict that started in the Middle Ages. That is, the combatants on both sides, and more particularly some of their leaders, are adherents to monotheistic forms of magical-mythical thinking, though the Christian side is generally more powerful, somewhat more secular and more technically advanced than its Moslem counterpart. Bush’s ‘war on terror’ is also arguably based more on a politically and massmedia driven mis-reading and highly emotional perceptual magnification of the levels of danger involved than it is on a massive escalation of the levels of violence and danger per se. Of course, this might well change if the terrorists and religious fundamentalists, whether Christian or Moslem, from the West or the East, were to succeed in getting their hands on weapons of mass destruction. That is because the present indications are that they would not hesitate to use them.10 I am thinking of Bush, Blair and their cronies in this connection and not simply of Moslem terrorists or their more direct counterparts in the West such as the Baader-Meinhof group, the IRA or the Protestant paramilitaries in Northern Ireland. A world state with a legitimacy comparable to that of the most successful Western nation-states and based, like them, on interdependent monopolies of violence and taxation has not yet emerged and perhaps never will. That is because nuclear and germ weapons, with their immense destructive power, presumably spell an end to processes of state-formation involving violent elimination struggles of the kind which Elias (2000) showed were involved in the ‘civilizing’ of Western Europe. If such weapons were to be used all-out, humanity would destroy itself or at least propel us into some kind of postnuclear ‘dark ages’. If that happened, ‘post-modernity’ in a literal sense would really well and truly have arrived! In a word, the formation of a world state, if it occurs, is likely to be both more long-term and more peaceful in the sense of not 9 The terms ‘First World’ and ‘Third World’ have arguably become illogical with the relative demise of the ‘Second World’, namely the former USSR and its ‘satellites’. 10 President George W. Bush and some of his associates, perhaps including Tony Blair, might be described as ‘religious fundamentalists’. It follows that, if what I have written in this chapter has any substance, they would probably be less hesitant about using nuclear weapons than many of their predecessors were because of their firm belief that ‘God’ is on their side.

240

E. Dunning

involving direct recourse to the use of the most destructive weapons except as a threat. If these weapons do not remain ‘confined to barracks’, it may spell the end of humankind. Assuming that such a version of ‘Armageddon’ does not occur, perhaps what is happening in Europe at the moment with the development of the EU may be a precedent for what may happen on a world level in the future? To express it metaphorically and metaphysically, ‘only time will tell’. Let me turn now to the subject of genocide.

Genocide in the Process of Civilization It is arguably fair to say that a widespread attitude in the twentieth century towards the unspeakably horrific fact of ‘genocide’ – the attempt to eradicate entirely a whole group of people who are socially defined as a ‘race’ or thought of in what amount to ‘racial’ terms – involves the idea that it is specifically and entirely modern as a social phenomenon. A more sophisticated version of the same belief is the idea that the Nazi attempt to exterminate the Jews was specifically modern because it was dependent on such features of ‘modernity’ as bureaucratic organization and industrial technology, an idea most recently propounded by Zygmunt Bauman (1989).11 From a figurational standpoint, however, although ‘modernity’ has probably helped to increase the scale and efficiency of genocides, it is arguably not so much genocide per se that is specifically modern as the growth of revulsion against it. That, at any rate, was Elias’s view. Writing not only of the Holocaust but also of the more general violence of the Germans towards conquered groups during the Second World War, especially towards Slavs and others in the East, he suggested that, in the twentieth century, revulsion was aroused almost world-wide by the mass slaughter of conquered groups by the Nazis and their East European cronies. The shock was all the greater, he contended, because the illusion had grown up in 20th century Europe that such barbarities were, for the peoples of Europe, a thing of the past. More particularly, many people in the West tacitly assumed that they had grown more ‘civilized’, and ‘morally better’ as part of their ‘nature’. They had taken what was, in effect, a kind of racist or ‘Social Darwinist’ pride in being less ‘savage’ than their forefathers and than other people outside the West. However, they had not seriously reflected on, let alone investigated how and why, their behaviour and their feelings had become ‘a little more civilized’ than they had been in former times. According to Elias, ‘(T)he Nazi episode served as a kind of warning’. It was a reminder that ‘restraints against violence are not symptoms of the superiority of the nature of ‘‘civilized’’nations, not eternal characteristics 11 From an ‘Eliasian’ standpoint, the concept of ‘modernity’ is a blanket term which obscures many of the differences regarding content, trajectories and timing (‘phaseology’) that are detectable in the emergence of urban-industrial-nation-states.

Violence and Violence-Control in Long-Term Perspective

241

of their racial or ethnic make-up, but aspects of a specific type of social development which has resulted in more differentiated and stable social control of the means of violence and . . . a corresponding type of conscienceformation. (Elias, in Dunning [ed] 1971:106–107; Elias and Dunning, 1986:143). Nazism and the Holocaust provided stark evidence inside and outside Germany of the fact that this kind of development at the interrelated levels of social structure and habitus/personality can be reversed. According to Elias, though, it is important to recognize that of all the main societies of Western Europe, Germany was the one that was structurally most liable to experience a ‘‘breakdown’’ of civilization. That is because it did not undergo a process of national unification until the late 19th century and because the German variant of this process occurred under the aegis of the Prussian monarchy and ruling caste who were more militaristic than their counterparts in England and France. They also orientated their lives around a harsh honorific ethos, that of ‘Satisfaktionsfa¨higkeit’, in which duelling with swords or pistols played a central part. That was in an overall Western European context in which the incidence of duelling had been declining for a century or more. In addition to this, according to Elias, a ‘breakdown of civilization’ occurred in Germany in the 1920s and early 1930s because the state was no longer able to maintain its monopoly on the use of violence in the face of the rise of the Freikorps (private militias), secret terrorist organizations such as Konsul, and the Nazi ‘brownshirts’, the Sturmabteilungen, (SA) and their Communist ‘streetfighter’ equivalents. While the rise to power of the genocidal Nazis was not inevitable, according to Elias, such a process was demonstrably more likely to have occurred in the Germany of that time than in then-contemporary Britain or France. Elias went on to contend that an almost universal feeling of repugnance against genocide emerged in the 20th century, especially in the West. This provides one more piece of evidence, he suggested, for the fact that Western societies have undergone a ‘civilizing process’, ‘however limited in scope and however unstable its results’. Although it is scanty, further evidence pointing in the same direction is provided by comparison with the Ancient world. What this evidence suggests is twofold: firstly that cases of mass slaughter recurred with fair regularity throughout the history of Greece and Rome; and secondly that, while wholesale massacres sometimes aroused pity in that context, they were not the cause of widespread revulsion and condemnation. Massacres were sometimes a consequence, said Elias, of ‘the battle fury of a long-threatened or frustrated army’. At other times, genocide w as a ‘calculated act aimed at destroying the military power of a rival state’. The wholesale slaughter by the Spartans of all the men of Argos, a potential rival, is an example. A similar act was the massacre of the male population of Melos by order of the Athenian Assembly of Citizens in 415 BC. It resulted, said Elias, providing what was at the time of writing an up-to-date illustration, from a ‘configuration very similar to that which led to the Russian occupation of Czechoslolakia in 1968’. In the Athenian case, however, the invading army

242

E. Dunning

‘killed the men, sold the women and children into slavery, and settled the island with Athenian colonists’. Writers and priests might speak of ‘the gods’ disapproving of massacres of this kind, noted Elias. However, ‘the level of ‘‘moral repugnance’’ against what we now call ‘‘genocide’’ and, more generally, the level of internalized inhibitions against physical violence, were decidedly lower, the feelings of guilt or shame associated with such inhibitions decidedly weaker than they are in the relatively developed industrial nation-states of the twentieth century’ (Elias, in Dunning [ed] 1971:107–108; Elias and Dunning, 1986:144–145). It would not, I think, be too much of an exaggeration to suggest that the historical record, not only of the Ancient world but also before and since, is littered with examples of genocides. In fact, the killing of people perceived as enemies, whether on realistic grounds or not, could even be said to involve a kind and degree of ‘rationality’ as far as peoples living at or near a subsistence level and without a money economy are concerned. Elias suggested as much when he wrote of the ruling class in early feudal Europe that outbursts of cruelty did not exclude them from social life. Killing and torturing others was a ‘socially permitted pleasure’. The social structure even pushed its members in this direction to some extent, making barbarous behaviour seem ‘rational’ under certain circumstances. To keep enemy prisoners, for example, meant to feed them. To return them healthy and unharmed meant to enhance the wealth and fighting power of one’s enemy. ‘For subjects (i.e. working, serving and fighting hands)’, said Elias, were part of the wealth of the ruling class . . . So prisoners were killed or sent back so mutilated that they were unfitted for war service and work’ (Elias, 2000:163). What David E. Stannard (1992) called ‘the American Holocaust’, that is the large-scale wiping out of ‘Native Americans’ in the course of the European conquest of the ‘New World’ is a more recent example of genocide. Another was the massacre of one and a half million Armenians by the Turks during the First World War (Brannigan and Cassis, in Levinson [ed] Vol 1, 2002:769–76). Two more were proposed by Alison Palmer in 1994, namely the killing off of Aborigines in Queensland, Australia by people of British descent, and the massacre by German soldiers during the First World War of the Hereroes of South West Africa (See also Evans 2003:12). She calls them ‘colonial genocides’. Although this was not noted by Palmer, her two examples illustrate the socially acceptable, to a degree even licensed, ‘hunting of humans’ (Leyton, 1989) or what one might call ‘genocide as play’. If I understand him, Slavoj Zizek came close to using such a concept in his critique of Daniel Goldhagen’s Hitler’s Willing Executioners: Ordinary Germans and the Holocaust (1996). According to Zizek, Goldhagen’s argument that most of the executioners felt no shame about their murderous activities is misplaced. What Goldhagen meant, says Zizek, is that such an absence of shame shows that the torturing and killing of Jews was ideologically understood by the executioners as ‘totally acceptable’. They had learned, as a result of exposure to the constant reiteration of anti-Semitic propaganda during the

Violence and Violence-Control in Long-Term Perspective

243

years of Nazi rule, to believe that they were engaged in the ‘noble’ task of ridding the ‘Aryan race’ of its principal enemies in the world-historical struggle for ‘racial’/national survival. However, Zizek argues persuasively that a close reading of Goldhagen’s text reveals how, in fact, ‘the executioners experienced their deeds as a kind of ‘‘transgressive’’ activity, . . . a kind of pseudo-Bakhtinian ‘‘carnivalesque’’ activity in which the constraints of ‘‘normal’’ everyday life were momentarily suspended’. This accounts, says Zizek, for the ‘surplus enjoyment’ they got from excessively torturing their victims. (Zizek, 1997:57). Although it misses the fact that this ‘carnivalesque’ behaviour was condemned as ‘un-German’ by such leading Nazis as Heinrich Himmler and Auschwitz Kommandant Rudolf Hoess (1959), who both wanted the genocide to be puritanically carried out as a duty to the Vaterland, Zizek’s point is well taken. That is, the Nazi Holocaust was far from just being the rational-bureaucratic genocide recently depicted by Zygmunt Bauman (1989), although that was part of it. Indeed, as Elias (1996) has shown, also involved was a ‘breakdown of civilisation’ on several levels. It is sometimes argued that the mass slaughter of (mainly) Jews in gas chambers and gas vans from 1942 onwards was decided upon and planned at the Wannsee Conference and that it replaced shooting and other forms of killing because it was less uncomfortable for the killers (see, e.g. Goldhagen, 1996:157). This is implausible. The Wannsee Conference was more concerned with facilitating and systematizing a planned process that was already well under way. The Einsatzgruppen/Einsatzkommandos and the Ordnungspolizei had already shot some 2 million by the time the Wannsee Conference took place. It is thus more reasonable to assume that gassing was introduced as a more efficient supplement. However, the shootings never stopped, to say nothing of the ‘death marches’ from the concentration camps outside Germany which took place as the armies of the West and its Soviet allies closed in (Goldhagen, 1996). Moreover, while there is some evidence that some of the killers (Goldhagen, loc cit) had breakdowns, they remained comparatively rare whether the killers were Germans, Poles, Lithuanians or whatever. And the idea that the gas chamber technology was somehow more impersonal than shooting is also wrong. Some victims resisted entry or fought vainly to have their children left outside (Hoess, 1959:149–50). In any case, the killers had to look through the windows at the end of an ‘action’ to see whether or not the victims were dead. The screams and death cries of the latter were also clearly audible to those outside the execution chambers, at Auschwitz even in some of the neighbouring blocks. Hans Joas is critical of Bauman on a related issue, more particularly for not paying sufficient attention to ‘the role of spontaneous and individual violence in the everyday life of the concentration camps’ (Joas, 2003:166). Joas also shares with Elias the idea that there is a need, contra Bauman, to stress that the Holocaust took place during the war (Elias, 1996:311; Joas, 2003:168). Let me now turn to the issues of crime, punishment and sport. I shall deal with them together and fairly briefly.

244

E. Dunning

Aspects of Crime, Punishment and Sport in the ‘Civilising’ of the West It may seem strange to some people that I should think it relevant to include in this discussion of ‘serious’ subjects such as war, genocide, crime and punishment, a consideration of the apparently ‘trivial’ subject of sport. However, just as it is arguably the case that forms of sport and fun can be involved in genocidal acts, war, punishment and crime, so it is the case that serious violence and crime of varying degrees and forms can be accompaniments of sport and play. Sport has also visibly risen in the hierarchies of socio-cultural, socio-economic and socio-political significance in recent years, and similar ‘civilising/de-civilising’ trends are arguably evident in all these fields as well. Let me try to demonstrate this with regard to crime and punishment. Although it does not seem to have filtered through into mainstream criminology and the sociology of crime, both of which tend to remain predominantly present-centred, there is a growing consensus among historians of the subject that rates of violence fell in Western countries more or less continuously from the end of the Middle Ages until the 1960s/1970s, at which point they started slowly rising once again. Writing in 2002, Irish criminologist, Ian O’Donnell, cogently summarized the trend by describing how ‘the evolution of modern societies is characterized by ever-decreasing levels of lethal violence between individuals’. More particularly, homicide rates declined steadily across the West from the early 19th century before beginning an upward surge in the 1960s. The 19th century pattern of a decline has been found wherever data have been collected. ‘Deviations from the downward trend were temporary and accounted for by the aftermath of war, demographic changes, economic prosperity or decline, and the early stages of rapid industrialization and urbanization. The increase since the 1960s was common though not universal’ (O’Donnell, 2002:56). Looking at the problem from an even longer-term perspective, Gurr (1981:296) described fourteenth century England as ‘a society in which men (but rarely women) were easily provoked to violent anger and were unrestrained in the brutality with which they attacked their opponents’. Commenting on murder rates since the 13th century, Lawrence Stone concluded that: ‘It looks as if the homicide rates in 13th century England were about twice as high as those in the 16th and 17th centuries and that those of the 16th and 17th centuries were some five to ten times higher than those of today’ (Stone, 1983:25). In a recent review of relevant research in England, Belgium, the Netherlands, Germany, Switzerland, Italy and the Scandinavian countries, Manuel Eisner concluded that the apparent general trend is real and neither a methodological artefact nor a consequence of inadequate data. He wrote: ‘The notorious imprecision of population data, deficiencies in the sources, shifts in the legal definition of homicide, changes in the age structure as well as improved medical possibilities, surely have to be accounted for. But the evidence is so consistent, the secular

Violence and Violence-Control in Long-Term Perspective

245

decline so regular, and the differences in levels so large, that it seems difficult to refuse the conclusion of a real and notable decline’ (Eisner, 2001:628). In order to explain this ‘real and notable decline’, Eisner follows Gurr (1989) in invoking Elias’s theory of civilising processes. In fact, in arguing thus these authors are part of an emerging consensus that Elias’s theory provides a better explanatory purchase on the problem than either ‘modernization’ theories or the conventional explanations the roots of which are traceable to Toennies, Durkheim and the Chicago School. In their The Civilization of Crime (1996), for example, Eric A. Johnson and the late Eric H. Monkkonen point to what they take to be four implications of Elias’s work that run counter to what used to be hegemonic assumptions among historians and sociologists in the 1950s and 1960s and still are far from having lost their hold today. These implications are that: 1. Emergent royal courts played a central part in the control of violent behaviour in developing Western nation-states in their early stages; 2. Behaviour tends to be more ‘civilized’ in urban than in rural areas; 3. Levels of impulsive violence tend to be higher among people in areas where state systems have not yet penetrated; and 4. In Western societies since the Middle Ages overall levels of violence have declined. Johnson and Monkkonen suggest that such implications of Elias’s work may have impeded its acceptance because they run counter to a dominant sociological ideology which derives from interpretations of the work of classical sociologists such as Toennies, Durkheim, Park and Burgess According to this dominant ideology, violent crime has increased as a consequence of the breakdown of family and community (Gemeinschaft) and the rise of mass society (Gesellschaft), especially as a consequence of urbanization, industrialization and the ‘class alignment’ of capitalist societies. Such an ideology made sense, Johnson and Monkkonen argue, because crime did increase somewhat in America following the Second World War and dramatically after 1968. Given that, they suggest, Elias’s discussions of the control of violent impulses in ‘civilized’ societies seemed out of touch. As a result, for historians of crime, ‘the slowly growing conviction that crime has decreased, not increased, over the centuries; that the countryside used to be dangerous, not safe; that, as Barbara Hanawalt put it, ‘‘fur-collar crime’’ was a major threat – all of this changed the status of Elias from curiosity to prescient thinker’ (Johnson and Monkkonen [eds] 1996:4–5). A similar picture holds regarding the punishment of crime. As Spierenburg (1984), Pratt (2002), and less unambiguously Garland (1990) have shown, pace Foucault (1978) the ‘civilising’ of punishment in Western Europe was less a relatively swift process associated with the decline of monarchy than a longerterm process connected with state-formation and changing sensibilities at the level of the social habitus of people. A study we carried out in Leicester in the 1980s points in the same overall direction (Dunning, et al, 1992; Dunning, et al, in Dunning and Mennell [eds] 2003). Using Leicester’s local newspaper, The Leicester Mercury, as our main source of data, we documented the fact that the

246

E. Dunning

reported incidence of violence in Britain over the years 1900–1975 followed a mainly downward trend. More particularly, the reported incidence was relatively high before the First World War, fell between the wars, and increased after the Second World War, especially after the 1960s but without coming anywhere near to approximating the levels of the years before 1914. One of the innovative features of this study consisted of the fact that we distinguished between violence in four ‘spheres’: the ‘political sphere’; the ‘industrial sphere’; the ‘community sphere’; and the ‘sphere of sport and leisure’. Criminal violence per se was not part of our focus. What we found was: a generally downward trend in the reported incidence of political violence in twentieth century Britain; a trend regarding industrial violence that was broadly similar, except that, in this case, the level of reported violence remained high up to the General Strike of 1926; and a steady fall in the incidence of community violence. The only sphere of British social life in which the reported incidence of violence rose more or less steadily in the course of the twentieth century was sport and leisure. What seems in effect to have happened is that a tradition of working class street and pub fighting which had earlier been firmly established, began to decline as a result, for example, of residential relocation, growing affluence, and the increasing incorporation of working class people into dominant values. As this occurred, the comparatively small number of males who continued to cling to the street and pub fighting tradition came increasingly to use sport and leisure, and after the mid-1960s, especially soccer, as a context for expressing it. This invasion of a national sport by working class fighting gangs in a country which, up until then, had prided itself and been widely acclaimed internationally for its peace-loving sports spectators, was amplified by media reporting out of proportion to what was actually occurring, contributing to the impression of a society where law and order were on the verge of breaking down. However, if one takes the whole picture into account, it is clear that, for most of the twentieth century, the reported — and probably also the factual — incidence of violence in Britain had been falling and that the reported rise in and after the 1960s was relatively slight. Even the increase of hooliganism in conjunction with football can be largely accounted for as a kind of transfer of violent traditions into sport and leisure from the community, pub and street. In a word, our findings are broadly consistent with those of other researchers into long-term trends.

Conclusion: Trends in Violence and Elias’s Thesis By way of conclusion, I want to suggest that, — and this slight criticism is in no way intended to detract from the originality and overall significance of their work — contra Johnson and Monkkonen [eds] 1996), Elias saw ‘civilization’ as a ‘beginningless process’ and did not claim that, ‘over time, violence would decline’. In fact, he saw the future as more open-ended and suggested that, at the present level of knowledge, all we can do is investigate why particular past

Violence and Violence-Control in Long-Term Perspective

247

sequences rather than plausible and possible others have occurred (Elias, 1978:158ff). Ted Gurr (1981) was one of the first scholars to recognise that the reported rise in violent crime in many Western countries since the 1960s/1970s is best construed as supporting rather than refuting Elias. This increase, he argues, ‘appears to be a minor perturbation, proportionately no greater than upward swings in homicide rates in Elizabethan times and during the Napoleonic wars — swings which proved to be temporary’. Each major ‘upsurge of violent crime’, he suggests, has been a consequence of ‘fundamental social dislocation’. Among other things, this has separated whole strata from ‘the civilizing institutions which instil and reinforce the basic Western injunctions against interpersonal violence’. These dislocating forces ‘may be migrants, demobilized veterans, a growing population of disillusioned young people for whom there is no social or economic niche, or badly educated young black men locked in the decaying ghettos of an affluent society’. Gurr stresses in this connection that ‘the most devastating episodes of public disorder’ appear to coincide with breakdowns of civilization, short-term ‘changes in values which legitimate violence’, often transmitted through popular culture. For this reason, the instability of lower working class families and the destruction of the compassion institutionalised through the post-war welfare state may have long-term deleterious repercussions (Gurr, 1981:338–340, 343, 346; see also Leyton, 1995:251). This is a cogent argument. In the present context, it is only necessary to add that the currently accelerating processes of globalization which—because they are both ‘causally’ and consequentially associated with the current dominance of ‘New Right’ doctrines — are fundamentally fuelling both the erosion of welfare states and the trend towards increasing violence may make this de-civilizing trend last longer than was the case in Elizabethan and Napoleonic times. Whether this proves to be the case or not, if the analyses and arguments advanced in this chapter have any substance, an interpretation of Elias based on a close and accurate reading of his central texts rather than on an ideological or other form of mis-construal of them as a simple ‘progress’ theory, an ethnocentric celebration of ‘the West is best’, arguably provides a basis on which it will be possible to increase understanding of the structure and dynamics of ‘civilizing’, ’de-civilizing’ and ‘dys-civilizing’ processes beyond the level reached so far. Hopefully, the typology and other data and arguments outlined in this chapter will prove to be fruitful in this connection.

References Atyeo, Don (1979), Blood and Guts: Violence in Sports, London, Paddington. Bauman, Zygmunt (1989), Modernity and the Holocaust, Cambridge, Polity. Brannigan, Augustine and Cassis, Viola R. (2002), ‘Genocide’ in David Levinson (ed), Encyclopedia of Crime and Punishment, London, Sage, Vol 2, pp 769–776. Dunning, Eric (1971), The Sociology of Sport: a Selection of Readings, London, Frank Cass.

248

E. Dunning

Dunning, Eric (1999), Sport Matters: Sociological Essays on Sport, Violence and Civilization, London, Routledge. Dunning, E. et al, (1992), ‘Violent Disorders in Twentieth Century Great Britain’ in Gaskell, G. and Benewick, R. (eds), The Crowd in Contemporary Britain, London, Sage. Dunning, Eric and Mennell, Stephen (eds), (2003), Norbert Elias: Sage Modern Masters, 4 vols, (2003), London, Sage. Eibl-Eibesfeldt, I. (1989), Human Ethology, New York, Aldine de Gruyter. Eisner, Manuel (2001), ‘Modernization, Self-Control and Lethal Violence: the Long-Term Dynamics of European Homicide Rates in Theoretical Perspective’, British Journal of Criminology, 41 (cited in O’Donnell [2002]: 58). Elias, Norbert (1939; 2000), The Civilizing Process, Oxford, Blackwell. Elias, Norbert (1969), The Court Society, Oxford, Blackwell. Elias, Norbert (1978), What is Sociology?, London, Hutchinson, pp158ff. Elias, Norbert (1987), Involvement and Detachment, Oxford, Blackwell. Elias, Norbert (1996), The Germans: Power Struggles and the Development of Habitus in the Nineteenth and Twentieth Centuries (translated from the German by Eric Dunning and Stephen Mennell), Cambridge, Polity. Elias, Norbert (1971), ‘The Genesis of Sport as a Sociological Problem’ in Dunning, Eric (ed), The Sociology of Sport: a Selection of Readings: 88–115; reprinted in Elias and Dunning (1986): 126–149. Elias, Norbert (1988), ‘Violence and Civilization: the State Monopoly of Physical Violence and its Infringement’ in John Keane (ed), Civil Society and the State: New European Perspectives, London, Verso: 177–198. Elias, Norbert and Dunning, Eric (1986), Quest for Excitement: Sport and Leisure in the Civilizing Process, Oxford, Blackwell. Ember, C.R. (1978), ‘Myths About Hunter- Gatherers’, Ethnology, 3,27:239–248 (cited in Wrangham and Peterson, 1997). Evans, Richard J. (2003), The Coming of the Third Reich, London, Allen Lane. Eysenck, Hans and Nias, David (1978), Sex, Violence and the Media, London, Maurice Temple Smith. Foucault, M. (1978), Discipline and Punish, London, Allen Lane. Freud, Sigmund (1932), Civilization and its Discontents, London, Hogarth. Fromm, Erich (1973), The Anatomy of Human Destructiveness, London, Penguin. Garland, David (1990), Punishment and Modern Society: A Study in Social Theory, Oxford, Clarendon. Goldhagen, Daniel (1996), Hitler’s Willing Executioners: Ordinary Germans and the Holocaust, New York, Vintage. Gurr, Ted R. (1981), ‘Historical Trends in Violent Crime’, in Crime and Justice, 3: 295–353. Gurr, Ted R. (1989), (ed), Violence in Americ. Vol 1, the History of Crime, London, Sage. Heitmeyer, Wilhelm and Hagan, John (2003), International Handbook of Violence Research, Dordrecht, Kluwer. Hoess, Rudolf (1959), Commandant of Auschwitz, London, Weidenfeld and Nicolson; reprinted in 2000 by the Phoenix Press, with an Introduction by Primo Levi. Joas, Hans (2003), War and Modernity, Cambridge, Polity. Johnson, Eric A. and Monkkonen, Eric H. (1996), (eds), The Civilization of Crime, Urbana and Chicago, University of Illinois Press. Keane, John (1988), (ed), Civil Society and the State, London, Verso. Keeley, Lawrence H. (1996), War Before Civilization: the Myth of the Peaceful Savage, Oxford, Oxford University Press. Leyton, Elliot (1995), Men of Blood, London, Constable. Leyton, Elliot (1989), Hunting Humans: the Rise of the Modern Multiple Murderers, Harmondsworth, Penguin. Lorenz, Konrad (1967), On Aggression, London, Methuen.

Violence and Violence-Control in Long-Term Perspective

249

Marsh, Peter (1978a), Aggro the Illusion of Violence, London, Dent. Marsh, Peter, Rosser, Elizabeth and Harre´, Rom (1978b), The Rules of Disorder, London, Routledge and Kegan Paul. O’Donnell, Ian (2002), ‘Unlawful Killing Past and Present’, The Irish Jurist, New Series 2002, Vol XXXVII: 56–90. Palmer, Alison (1994), Colonial Genocides: Aborigines in Queensland, 1840–1896, and Hereros in South-West Africa, 1884–1906, Unpublished PhD thesis, University of London. Pratt, John (2002), Punishment and Civilization, London, Sage. Spierenburg, Pieter (1984), The Spectacle of Suffering, Cambridge, Cambridge University Press. Stannard, David E. (1992), The American Holocaust, Oxford, Oxford University Press. Stone, Lawrence (1983), ‘Interpersonal Violence in English Society, 1300–1980’, Past and Present, 102: 22–33. Dunning and Mennell (eds), Norbert Elias, Sage, London pp 137–148. Wrangham, R. and Peterson, D. (1997), Demonic Males: Apes and the Origins of Human Violence, London, Bloomsbury. Zizek, Slavoj (1997), The Plague of Fantasies, London, Verso.

Index

Abbink, J., 15–16 Adkins, L., 162, 170 A˚gren, M., 55 Aijmer, G., 15–16 A˚kerstro¨m, M., 37–38 Albanese, P., 163 Alvarez, M. E., 209 Anderson, E., 128 Appel, H. H., 45 Archer, D., 86 Arteta, S., 117 Atyeo, D., 229 Augustins, G., 69 Bader, V., 117 Badger, D., 27–29, 33, 37 Bagge, S., 54 Balace, F., 189 Barahona, R., 141 Barak, G., 31–32 Bartusch, D. J., 214 Batchelor, S., 164 Bauman, Z., 27, 37, 240, 243 Baumer, E. P., 211 Beaud, S., 104, 120 Beauregard, R., 8 Bekaert, H., 195 Benedicti, J., 145 Berriot-Salvadore, E., 148 Bertrand, O., 129 Betz, J., 34 Binnie, J., 163 Bjorgo, T., 123 Blader, S., 210 Bleich, E., 131 Blok, A., 196 Blum, A., 118 Bodin, J., 144, 150

Body-Gendrot, S., 1, 5, 8, 115, 118, 120–125, 129, 205 Boeckx, B., 197 Boehnke, K., 212 Bo¨hlau., 143 Boniface, H de, 145, 153 Bourdieu, P., 20, 31, 168 Bourgois, P., 21 Boutelet, B., 65 Bouveroux, J., 189 Braithwaite, J., 162, 205 Brannigan, A., 242 Brodeur, J.P., 128 Bronner, L., 131 Brown, A. E., 164 Bufacchi, V., 36 Bull, R., 27–29, 33, 37 Burgess-Jackson, K., 36, 245 Caillet, P., 150 Caldeira, T. P. R., 207, 209, 222 Campbell, A., 162 Casey, M., 163 Cassis, V. R., 242 Castan, N., 67, 142 Castan, Y., 67 Ceaux, P., 131 Chartier, R., 146 Chasseneux, B. de, 151 Chatterton, P., 164 Chauvaud, F., 74 Cheibub, J. A., 209 Chesnais, J.-C., 74–75, 185, 187 Cholie`res, N. de, 148, 150 Classen, G., 212 Claverie, E., 67, 69, 73 Cohen, D., 75, 210 Cohen, M., 133

251

252 Cohen, P., 163 Conley, C., 82 Connolly, S.J., 81–82 Conway, M., 187 Coquille, G., 151 Coras, J. de., 144, 150 Corbin, A., 65, 74 Coulthard, M., 164 Crouzet, D., 141, 143 Cypel, S., 119 Daelemans, F., 187 Daly, K., 162 Daniels, C., 142 Darmon, P., 151–152 Davis Rodriguez, C., 207 Davis, N Z., 143, 146 De Haan, W., 2, 13, 18, 22, 27, 117, 120, 123, 126, 205, 229–230 De Schaepdrijver, S., 195 De Wever, B., 187 Dejonghe, A., 187 Delon A., 117 Deneckere, G., 182 Descimon, R., 151 De´sert, G., 65, 74–75 Desmars, B., 72 Dinges, M., 23 Dominique, J., 146 Dooley, E., 88–89 Drakulie. S., 163 Drass, K. A., 206, 216–218, 220 Du Moulin, P., 150 Duch, R. M., 209 Dunning, E., 8, 19, 227, 234, 241–242, 245 Duret, J., 144 Durkheim, E., 7, 84, 88, 120, 245 Earls, F., 211 Edwards, G., 163 Egelkamp, M. M., 18 Egmond, F., 187, 201 Eibl-Eibesfeldt, I., 235–236 Eisner, E., 79, 206, 209, 211–212, 245 Eisner, M., 44, 79, 177, 178, 179, 180, 205, 206, 209, 211, 212, 220, 244, 245 Elias, N., 8, 18–19, 55, 79, 178, 201, 227–228, 230, 232–233, 235, 237–243, 245–247 Elster, J., 210 Ember, C.R., 235 Erickson, B. H., 163

Index Eriksson, M., 141 Evans, R. J., 242 Ewald, U., 207 Eysenck, H., 228–229 Fagan, J., 128 Fahey, T., 87 Farcy, J.-C., 65, 74 Farge, A., 67 Fenstermaker, S., 170 Ferenczi, T., 3 Ferriter, D., 83 Figart, D. M., 165 Finnane, M., 81–82, 84 Fitzpatrick, D., 83 Flandrin, J.-L., 144, 148 Flandrin, J-L., 144, 148 Foucault, M., 245 Foyster, E., 142 Franc¸ois A., 144 Freud, S., 232–233 Fromm, E., 229, 232 Fuller, T., 130 Gallie, W.B., 2, 29, 35–36 Galtung, J., 20–22 Garland, D., 245 Gartner, R., 86 Garver, N., 34–35 Gegot, J., 65 Gellner, E., 209 Gerard-Libois, J., 199 Gibson, J. L., 209 Gilligan, C., 167 Godard, B., 119 Goddard, E., 164 Goldhagen, D., 242–243 Gotovitch, J., 187–188, 197, 199 Goudsblom, J., 21 Gough, B., 163 Gournay, M de., 152 Gowing, L., 142 Graham. K., 163 Greenshields, M., 141–142 Griffin, B., 80 Groebner, V., 153 Guinnane, T.W., 82 Gurr, T. R., 1, 7, 44, 177, 206, 208, 209, 244, 245, 247 Guy, M. E., 166 Hacke, D., 145 Hadfield, P., 162, 165

Index Hagan, J., 4, 8, 28–29, 212, 214, 228 Haila, A., 8 Halbert, C., 170 Hall, S., 163 Hamm, M.S., 28 Hanley, S., 151 Haritos-Fatouros, M., 207, 213 Healy, D., 80 Hefler, G., 212 Heitmeier, W., 28–29 Heitmeyer, W., 4, 8, 228 Helle, K., 52 Henry, B., 80 Henry, S., 31–32 Hey, V., 162 Hobbs, D., 6, 161–165, 167–169 Hobsbawn, E., 116, 196 Hochschild, A., 165 Hoess, R., 243 Hohkamp, M., 141–142 Hollands, R., 164 Holmes, S., 210 Holston, J., 207, 209, 222 Hoogerwerf, A., 16 Horne, J., 198 Hufton, O., 141–142 Huggins, M., 207, 213 Hughes, A.J., 21, 82 Hunt, G., 162, 168 Hutchinson, I. L., 164 Huttman E., 118 Iadicola, P., 32 Imbert, J., 144 Imbusch, P., 3, 37 Inglehart, R., 209, 212 Inkeles, A., 209 Isambert, F., 144 Jackman, M., 28, 32 Jacquemyns, G., 189 Jaggers, K., 209 Jansson, A., 17, 48, 53–56 Jarrick, A., 53, 55–56 Joas, H., 238, 243 Jobard, F., 125 Joe-Laidler, K., 168 Johansen, J. C. V., 44, 54–55 Johansson, K., 44 John K., 205, 233, 239 Johnson, E. A., 75, 245–246 Johnson, I., 128 Joly, H., 201

253 Julien P., 150 Ju¨tte, R., 141 Karonen, P., 45, 50, 53, 56–57 Karstedt, S., 7, 205–207, 209, 211–212, 214–215, 221 Kaspersson, M., 44, 56–57 Keane, J., 205, 233, 239 Keeley, L. H., 234–237 Kelly, J., 81 Kennedy, M. V., 142 Keogh, D., 86 Kepel. G., 129–130 Kessous, M., 122 Kharkhordin, O., 213, 215 Khosrokhavar, F., 121, 130 Kilcommins, S., 87 Kim, S.-W., 207, 215 Kirkby, D., 162 Klages, H., 212 Klein, A., 169 Knowlton, B., 127 Koch, E., 144 Kolvin, P., 166 Kramer, A., 198 Krug-Richter, B., 141 Kurz, L.R., 27 L’Homme Z.F.G., 141–142 LaFree, G., 205–206, 208, 212, 216–218, 220–221 Lager, B., 49 Lagrange, H., 100 Laingui, A., 144 Lal, R., 128 Lapalus. S., 69–70 Laplasse, J., 197 Larsson, L.-O., 49–50 Laurence, J., 127 Lauth, H.-J., 209 Le Caron, L., 151 Le Toqueux, J.L., 104 Lebigre, A., 144 Lederman, D., 211 Lefebvre-Teillard, A., 145 Lehti, M., 58–61 Leveau, R., 132 Levi, M., 27, 29 Levy, R., 123 Leyton, E., 242, 247 Liliequist, J., 50, 54, 57, 61 Lilja, S., 49 Limongi, F., 209

254 Lindenberger, T., 141–142 Lindstrom, D., 205 Lister, S., 162, 165–166 Litke, R. F., 34 Litke, R., 2 Loayza, N., 211 Lohdi, A. Q., 75 Lorenz, K., 232–233 Los, M., 208, 214 Loubiere, A., 122 Lovell, T., 162, 166, 168–169 Løyland, M., 51 Lu¨dkte, A., 141 Lukes S., 37 Luyten, D., 199 Lyndal, R., 142 Lyons, F.S.L., 82 McCaughey, M., 162 Maclean, I., 145, 151 McCullagh, C., 82 McEwen, Y., 86 McKanna, C. V., Jr., 21 McNamara, J., 166 Maerten, F., 197 Magennis, P., 164 Magnusson, L., 58 Maguire, M., 27, 29, 163–164 Maher, J., 164 Majerus, B., 178, 187 Mann, M., 206, 208, 210, 214 Marconville, J. de, 147–148, 150 Mares, D.M., 90 Marsh, B., 164 Marsh, P., 229, 231, 232, 234, 235, 236 Matthews Grieco, S. F., 146 Mazower, M., 178 Mendes, J. E., 208, 213 Menedez, A., 211 Mennell, S., 90, 245 Menneson, 170 Merkens, H., 212 Messerschmidt, J. W., 168, 170 Messner, S. F., 205, 208, 211–212 Meurier, G., 152 Meyers, M., 187 Miller, J., 162, 168–170 Milovanovic, D., 32 Monaghan, L., 161–162, 165–167 Monjardet, D., 124 Monkkonen, E. H., 245–246 Mooij, T., 16

Index Moreau, J., 104 Morris, R.M., 80 Mucchielli, L., 117 Muchembled, R., 141, 149 Muller, E. N., 209–211 Munt, S., 166 Mykland, K., 54 Næss, H. E., 44, 47, 52, 54 Navarre, M. de, 150 Nedelmann, B., 16, 31 Neidhardt, F., 30–31 Newman, M. A., 166 Nias, D., 228–229 Nijboer, J., 120, 123 Noiriel, G., 120 Nolan, B., 87 Nolde, D., 6, 20, 141, 143 Nowotny, T., 209, 213 O’Brien, J.V., 16, 84 O’Brien, K., 6, 161, 164 O’Connell, P.J., 87 O’Donnell, G., 208, 213 O’Donnell, I., 79, 80, 82, 87, 244 O’Rourke, K., 82 O’Shea, F., 86 O’Sullivan, E., 80 O¨sterberg, E., 43–45, 48–50, 53, 55, 57 Palmer, A., 242 Papon, J., 144 Parella, A., 71 Parker, R.N., 88 Peleus, J., 150 Pe´rez-Diaz, C., 100 Pericaud, M. A., 145 Perrot, M., 66 Peterson, D., 235–236 Petot, P., 144 Pialoux, M., 104 Pickel, G., 209 Pinheiro, P. S., 208, 213 Pirenne, H., 198 Ploeg, J.D. van der, 16 Ploux, F., 4, 65, 68–69, 72–73 Portillos, E. L., 163 Pottier, M.L., 100, 103–104 Pratt, J., 245 Pridemore, W. A., 207, 214–215, 220 Przeworski, A., 209, 214, 216–217, 220 Punch, M., 28

Index Quinlan, C., 86 Radoeva, D., 214 Rauch, J., 13, 20, 23 Raudenbusch, S. W., 211 Reilan, E., 35 Reiss, A.J., 30 Rey, M., 111 Riches, D., 124–15, 17, 19–20, 22, 30–31 Riedel, M., 27 Robert, P., 5, 66, 95 Robert, Ph., 96, 97, 100, 103, 104 Robins, D., 163 Roche, S., 144 Rocke, M., 22 Rose, L., 80 Rose, R., 214 Rosenfeld, R., 208, 211–212 Roth, J.A., 30 Rothstein, B., 211 Rousseaux, X., 7, 17, 177, 187 Roy, O., 130 Ruff, J. R., 141 Russell, H., 87 Sabic, I., 166 Saerens, L., 197 Salas, D., 121, 130 Sampson, R. J., 211 Sampson, R., 214 Sanchez, T., 148 Sandmo, E., 44, 54–56 Sandnes, J., 44, 51–52, 54 Saterlee, S., 162 Sauer, R., 80 Savolainen, J., 212 Schain, M., 133 Scheper-Hughes, N., 21 Schetter C., 118, 126 Schinkel, W., 31 Schmidt, H. R.., 142 Scholliers, P., 187, 189 Schwerhoff, G., 22–23, 153 Seguin, J-P., 148 Seligson, M. A., 209–211 Sheard, K., 19 Shepherd, J. P., 164 Shupe, A., 32 Skeggs, B., 163 Smith, C., 129 Smolar, P., 127, 129, 132 So¨derberg, J., 45–46, 48, 52–53, 55–56

255 Spierenburg, P., 2, 13, 23, 44, 49, 79, 141–143, 177, 179, 205, 229–230, 245 Stanko, E.A., 27 Stannard, D. E., 242 Steen, K., 197 Steinberg, J., 207 Steinberg, R. J., 165 Stevnsborg, H., 44, 54–55 Stolle, D., 211 Stone, L., 44, 244 Stroobants, J.-P., 133 Sundin, J., 55, 57–58 Swift, R., 82 Sztompka, P., 207, 209, 213 Tassig, S., 119 Taylor, H., 80 Tayssonniere, G. de la., 149 Temmerman, P., 197 Thomas, M., 164 Thome, H., 79 Tilly, C., 75 Tixhon, A., 180 Tomsen, S., 163 Tonry, M., 118 Touraine, A., 122 Tourancheau, P., 129, 133 Trotha, T. von, 16, 17, 31, 33 Tseloni, A., 212, 221 Turpin, J., 27 Tyler, T. R., 205, 210 Vaisse, J., 127 Valier, C., 28 Van Doorslaer, R., 187, 189 Van Meerbeeck, P., 187 Van of Vijver, H., 187 Vandenbossche, A., 144 Vanhemelryck, F., 179 Velle, K., 183 Verhoeyen, E., 187 Verkko, V., 46 Vives Juan Luis, 144 Vives, J. L., 144 von Hofer, H., 45, 47, 53 Vrints, A., 7, 177–178, 186, 199 Waddington, P.A.J., 27–29, 33, 37 Wakefield, A., 163 Walberg, P., 220 Waldmann, P., 205, 208, 213–214 Walker, A., 163–164, 166 Walklate, S., 163

256 Weber, M., 122, 232 Weiner, N.A., 27 Wells, S., 163 Welsh, W., 27–28, 81 Welzel, C., 209, 212 West, C., 170 Westling, C., 49 Westmarland, L., 6, 161, 163 Wever, B., 187 Whelan, C.T., 87 Wihtol de Wenden, C., 124 Wilbanks, W., 81–82 Wilkinson, D.L., 128 Willemez, L., 77 Wilson, J.M., 83

Index Wimmer, A., 118, 126 Winlow, S., 162–163, 165, 168 Wouters, N., 187 Wrangham, R., 235–236 Ylikangas, H., 44–46, 48, 50–52, 54–58 Yllo, K. A., 163 Yount, K.R., 163 Zauberman, R., 95, 100, 104, 123 Zehr, H., 74–75 Zimbardo, P. G., 207, 213 Zimmerman, D. H., 170 Zizek, S., 242–243