Political Violence, Crises and Revolutions (Routledge Revivals)

Theories and Research

Ekkart Zimmermann

• Publisher: Routledge
• Year: 2013
• ISBN:
  9781136599743
• Edition: 1
• Topic: Politique et relations internationales
• Subtopic: Politique

First published in 1983, this extraordinary study provides a comprehensive systematic evaluation of cross-national theorizing and quantitative empirical evidence on four interrelated phenomena:

• 
  
  

• Political violence

• 
  
  

• Crises

• 
  
  

• Military Coups D' État

• 
  
  

• Revolutions.

Findings from social-psychological research on aggression are integrated in this outstanding study, as well as results reported in social-historical studies of revolution. The focus of the book is always on analytical perspectives and corresponding empirical evidence. The author continually highlights the sociostructural and political conditions of political violence, crises and revolutions.

This exceptionally detailed and systematic inventory of theories and research on a classic triad of political science (political violence, crises and revolutions) also includes a remarkable bibliography encompassing over 3000 items.

Chapter 1

Introduction

 

 

 

“Some 68 million human beings, according to the most careful estimates, perished from all forms of deliberate human violence—murder, riot, revolution, and war-in the 150 years from 1820 to 1970” (PROSTERMAN 1976: 339). If the victims of state violence suffering from pogroms, brutal repression, and other forms of state coercion were added, the figure would probably be more than twice as high. Obviously, political violence, meaning (for the most part antigovernment or antisystem) violence occurring within a political system and having political consequences, is an issue that deserves the close attention of the social scientist. When carrying out analyses, he or she should bear in mind that, after all, “violence is an issue about which most people have strong opinions, perhaps even violent ones” (MOORE 1968:1).

While there have been several systematic analyses of political violence before 1960, the best known being the classic studies by WRIGHT (1942) and RICHARDSON (1960) on the determinants of war and, more in line with the topics of this study, by SOROKIN (1937) on internal (and external) disturbances, political violence as a distinct field of study developed only recently in the 1960s. The late but extremely rapid development of this branch of study may be attributed to a number of factors:

1. the development of computers of increasing capacity and the con-comitant availability of computer-based programs for data analyses which are relatively easy to handle and produce an enormous output of statistical results;

2. the growing interest in cross-national analyses, leading to the development of a number of data banks described in such publications as Cross Polity Time Series Data (BANKS 1971), Black Africa: A Comparative Hand-book (MORRISON et al. 1972), The Wages of War (SINGER/SMALL 1972), World Handbook of Political and Social Indicators (2d ed., TAYLOR/HUDSON 1972) as well as to numerous other data collections referred to later in the text; and, providing the final and perhaps strongest impetus,

3. the riots among American blacks during the second half of the 1960s. This led to an enormous research program, perhaps second only to the study of voting behavior and to the other dominant issue in social science analyses during the 1960s, student protests in the United States and elsewhere in the world.

In 1965 ECKSTEIN opened up a number of analytical perspectives in a now classic article on cross-national analyses of political violence, the perspective in which we are primarily interested. Other pioneering works are the factor analyses of various forms of conflicts by RUMMEL (from 1963 to the present), the first broad comparative analysis of political violence published (FEIERABEND/FEIERABEND 1966), and GURR's study of civil strife (GURR 1968a), a first high point in theoretical and methodological sophistication. The end of the 1960s and first half of the 1970s have witnessed an enormous and continuing outpouring of studies on political violence, both comparative (cross-national) analyses and case studies. Most of these cross-national analyses study phenomena of political violence taking place after the Second World War. This focus is considerably broadened in the analyses of revolutions which are dealt with in later chapters of this study. Most of those analyses are comparative only in a very limited sense. Our aim is to evaluate these studies and to provide some guidelines as to the possibilities of cross-national analyses of revolutions. In the chapters on military coups d'état, one of the dominant patterns of political violence in many parts of the world, and on crises and political violence, reference is again made to a number of studies generally covering periods after the Second World War, but occasionally, longer periods are treated, and in one instance a period of approximately one hundred fifty years is considered. As might be expected in cross-national analyses of this sort, variations in time and space are considerable, rendering the development of universally valid theoretical statements a herculean task. One purpose of the present study is to bring together various strands of research and to show or suggest where and how they might be fruitfully related to each other. In addition, our book is intended as a critique of existent research.

A few words are in order about what is excluded from this study. First of all, the focus here is on sociostructural and political conditions of political violence, crises, and revolutions and not on the variety of ideological factors (or the structure of idea systems) and their impact on the phenomena mentioned. This is not to deny the importance of ideas. In the present context, this is a necessary limitation in order to find a way through the many studies already published on these explanatory factors. In addition, together with most social scientists, we share the premise that the impact of sociostructural and political conditions is likely to be stronger than that of idea systems (at least in the short run). Secondly, case studies on particular topics are generally not treated here, although some references to their usefulness have been made. This is true as well for other monographic works, be they historical or biographical in nature. Thirdly, at least one topic of current public interest is not considered here in detail: the explanation of the various forms of international terrorism (as well as the possibilities of combatting these forms of international disorder, a topic beyond the scholar's primary role of analyzing the phenomena objectively). To our knowledge, at this time there is not sufficient data to allow for systematic causal analyses of international terrorism. (When writing this, MICKOLUS 1980 was not yet published). Variability in the forms of international terrorism is still too great. However, throughout this study there is a set of recurrent conditions that probably will prove to be of considerable importance for an understanding of terrorist phenomena including irredentist claims, claims for sociocultural autonomy, and separation of certain cultural, ethnic, racial, or other groups (cf. also chapter 8.4.4 for some preliminary considerations).

Concerning the organization of the present study, only a few guidelines will be given here, with additional ones found in each of the major chapters. Chapter 2 deals with conceptual problems in the analysis of political violence. A discussion of various definitional criteria of violence, of typologies of violence, and of the relationships between violence and other key concepts will be found in this chapter. The following two chapters deal with background theories. Especially in cross-national aggregate data analyses of political violence, references are often made to microexplanations of aggressive behavior and to theories of social comparison processes. Our aim is to be more explicit concerning these theories by giving a brief summary of the findings of experimental studies of aggressive behavior (chapter 3) and of theories of social comparison processes (chapter 4). The weaknesses of these theories are discussed, as well as their possible contributions to better studies of political violence and other forms of political protest behavior. After the more preliminary chapters, chapter 5 deals with cross-national approaches to the study of political violence, the stress being on rather broadly defined dependent variables often representing clusters of phenomena of political violence as they emerged from factor analyses. In the first part of this section (chapter 5.1.1) the pioneering approaches of GURR, the FEIERABEND group, and HIBBS (1973) are described, compared, and evaluated. In the second part (chapter 5.1.2) several key variables for explaining political violence are traced through numerous cross-national analyses. A summary discussion of several methodological aspects concludes the section dealing with cross-national analyses of political violence. A chapter on case studies of political violence follows, focusing mainly on the fascinating analyses of social banditry given by HOBSBAWM (cf. 1959, 1969). The richness in detail of this section is lacking in the statistics encountered in the earlier studies. Chapter 5, dealing with approaches to the study of political violence, ends with a presentation and evaluation of theories and empirical results on the causes of the riots of American blacks during the 1960s. These events have been studied in greater detail than any other phenomena in the field of political violence. However, many important questions remain unanswered. It is interesting to compare the theories for explaining political violence on a cross-national basis with those explaining a more homogeneous phenomenon of political violence. As might be expected, there is considerable overlapping in the theories proposed. In addition, numerous other explanations based on survey data have been developed and tested to account for the blacks' riots.

In cross-national analyses of political violence, quite often it is equated with political instability or crises phenomena, yet no justification for such an equation is given. Consequently, in chapter 6 the present study (1) defines crises phenomena and discusses several general approaches to the study of crises; (2) differentiates analytically between crises and political violence, showing possible interrelationships between the two ; (3) lists and discusses those cross-national analyses bearing on crises phenomena and/or the relationship between crises and political violence ; and (4) develops a causal model of crises and crises outcomes (with persistence of a polity as the final dependent variable) and proposes suggestions as to the development of a cross-national crises science.

The chapter on crises and political violence is also the first in a series of three bearing on the analysis of revolutions, the subject of chapter 8. Prior to the study of revolution, however, a chapter on cross-national analyses of military coups d'état has been included. Crises, military coups d'état, and revolutions are topics that, for analytical reasons, must be considered separately. Empirically, however, there is some evidence that they are intricately linked. The behavior of the military and the police forces is of crucial importance when crises situations escalate to revolutionary proportions. However, the behavior of the military and the police forces is only one of the major variables considered in causal modeling of revolutions.

Compared with other sections in this study, the treatment of theories and empirical results dealing with the causes of military coups is more homogeneous, thus facilitating comparisons of theories and results. In contrast with chapter 5.1 where, within the cross-national analyses of political violence, coups were sometimes incorporated into the particular dependent variables, in this chapter the interest is solely on military coups d'état and not on other phenomena of political violence (at least not as far as the dependent variable is concerned). Thus, in chapter 7 some of the theories referred to in chapter 5.1 will be considered, while at the same time theoretical notions and empirical results bearing on later parts of this study will also be treated. This interlocking of theories can perhaps best be demonstrated by the fact that HUNTINGTON's hypotheses (1968) pervade all of these chapters and are dealt with at various points in our discussion. At the end of chapter 7 a causal model of military coups d'état is developed summarizing theoretical and empirical evidence and to some extent going beyond it by suggesting new theoretical linkages to be considered in future cross-national analyses of military coups d'état.

The chapter on the cross-national study of revolutions attempts critical assessment. Of the myriad studies on revolutions, a sizable portion hardly deals with revolutions at all. This also holds true for some of the cross-national analyses of revolutions, as will be shown in chapter 8.4.6. Consequently, the present study attempts to develop a more precise definition of revolutions, to discuss various typologies found in the literature, and to deal with some of the numerous variables which are suggested to explain the occurrence of revolutions. Some of the more elaborate explanations of revolutions, such as HUNTINGTON's (1968) comparison of Western and Eastern models of revolution, MOORE's (1966) broad sociohistorical analyses of the social origins of dictatorship and democracy, and PAIGE's (1975) study of agrarian revolution, are taken up in greater detail. Having criticized the very few quantitative cross-national studies of revolutions, the chapter focuses on historical analyses of revolutionary protests. The quantitative analyses of collective violence in France, Italy, and Germany from 1830 to 1930 (cf., e.g., TILLY et al. 1975) and the largely qualitative analyses of revolutionary crowds by RUDÉ (1964, 1970) are good examples of the way a comparative study of revolutions might proceed (they do not arrive at a dead end as do several cross-national analyses of revolutions which are actually analyses of military coups, and not even recommendable at that). A causal model of revolutions is developed, bringing together the various strands that have been discussed throughout the rather lengthy chapter on revolutions. Finally, several analytical suggestions are made which might be helpful in setting up more successful designs in the study of revolutions.

In the final chapter the three causal models developed in earlier chapters, concerning (1) crises and the persistence of polities, (2) military coups d'état, and (3) revolutions, are briefly compared, providing a short analytical summary of this study. In the concluding paragraphs, reference is made to some other neglected aspects in the study of political violence, crises, and revolutions. The chapter concludes with some comments on other social phenomena that come into focus once the frame of reference is broadened beyond the subjects considered here. The aim of the present study is one of constructive criticism, criticism concerning much of what has been done in analyses of political violence, crises, and revolutions. Suggestions are proposed as to how some of the errors committed in the past might be avoided in more carefully designed future studies. At various points, this text attempts to build on the critical evaluations given here. The author hopes that other scholars will go on in these areas of study, progressing further than we have done in the present context.

Chapter 2

Conceptual Problems

2.1. Toward a Definition of Political Violence

Violence has been a subject for human reflection for hundreds of years. However, rather than presenting an anthology of the different views on violence as found, for example, in the Sermon on the Mount and the writings of MACHIAVELLI, HOBBES, SOREL, and many others, we wish to discuss the usefulness of various definitional criteria proposed by social scientists. Webster's Third International Dictionary defines violence as “exertion of any physical force so as to injure or abuse.”¹ An important distinction is added in the Oxford English Dictionary which denotes violence as “the exercise of physical force so as to inflict injury on, cause damage to, persons or property.” Differentiating between violence against human beings and violence against nonhuman objects is generally accepted among social scientists.

In the introduction to The History of Violence in America (National Commission on the Causes and Prevention of Violence) violence is defined “narrowly … as behavior designed to inflict physical injury to people or damage to property” (in GRAHAM/GURR 1969:xxxii; cf. the similar definitions of VAN DEN HAAG 1972:54, and SKOLNICK 1969:4). This definition, sometimes with slight variations, is also used in the various works of GURR (cf. GURR/McCLELLAND 1971:17/19; GURR 1973b:360: “deliberate uses of force to injure or destroy physically”). GURR also points to additional variables that will be dealt with in this study: “This definition is independent of agents, objects, or contexts of violence” (GURR 1973b:360).

Having specified that violence denotes certain intentions as well as the use of specific means (cf. also the discussion in COUZENS 1971), criteria to be discussed in detail below, the attribute “political” remains to be defined. Some general criteria for political violence can be given, but the distinction between violence and political violence nevertheless often remains arbitrary (see the references to the labeling approach below). The following criteria are offered in an effort to distinguish violence from political violence:

1. The number of persons involved. Other things being equal, the probability that acts of violence will be thought of as political in nature is greater, the more persons are involved in these...

Table of contents

• Cover
• Half Title
• Full Title
• Copyright
• Contents
• LIST OF FIGURES
• LIST OF TABLES
• ACKNOWLEDGMENTS
• 1 Introduction
• 2 Conceptual Problems
• 3 Experimental Studies of Aggressive Behavior: A Brief Overview
• 4 Theories of Social Comparison Processes
• 5 Approaches to the Study of Political Violence
• 6 Crises and Political Violence
• 7 Military Coups d'État in Cross-national Perspective
• 8 The Cross-national Study of Revolutions: Toward a Critical Assessment
• 9 Political Violence, Crises, and Revolutions: Theories and Research—Some Concluding Remarks
• NOTES
• BIBLIOGRAPHY
• INDEX