Social Trust, Anarchy, and International Conflict

Social Trust, Anarchy, and International Conflict

9780230113176_01_previii.indd i

2/5/2011 10:52:21 AM

9780230113176_01_previii.indd ii

2/5/2011 10:52:21 AM

Social Trust, Anarchy, and International Conflict Michael P. Jasinski

9780230113176_01_previii.indd iii

2/5/2011 10:52:21 AM

SOCIAL TRUST, ANARCHY, AND INTERNATIONAL CONFLICT Copyright © Michael P. Jasinski, 2011. All rights reserved. First published in 2011 by PALGRAVE MACMILLAN® in the United States a division of St. Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Where this book is distributed in the UK, Europe and the rest of the world, this is by Palgrave Macmillan, a division of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN: 978–0–230–11317–6 Library of Congress Cataloging-in-Publication Data Jasinski, Michael P. Social trust, anarchy, and international conflict / Michael P. Jasinski. p. cm. Includes bibliographical references and index. ISBN 978–0–230–11317–6 (alk. paper) 1. War. 2. War—Causes. 3. Trust—Social aspects. 4. Politics and war. 5. State, The. I. Title. JZ6385.J37 2010 303.6’6—dc22 2010040862 A catalogue record of the book is available from the British Library. Design by MPS Limited, A Macmillan Company First edition: April 2011 10 9 8 7 6 5 4 3 2 1 Printed in the United States of America.

9780230113176_01_previii.indd iv

2/5/2011 10:52:22 AM

Contents

List of Tables

vi

Acknowledgments

vii

1 2 3 4 5 6 7 8 9 10

Introduction Democratic Peace and Diversionary War Anarchy, States, and Nations Social Trust and Its Origins Overcoming Particularism The Social Trust Theory of International Conflict Research Design Results The Outbreak of World War I Analysis and Conclusions

1 9 29 47 63 77 91 109 127 159

References

173

Index

195

9780230113176_01_previii.indd v

2/5/2011 10:52:22 AM

List of Tables

7.1 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 9.1

Corruption Perception Index (CPI) aggregate values 1990–2001 Incidence of domestic political conflict Presence of minorities at risk (MARs) as a function of social trust Democratic peace test at state pair (dyadic) level Conflict initiation test with social trust test as a joint attribute Social trust as a state attribute test Conflict initiation test with social trust quartiles as dummy variables Conflict targeting test Conflict targeting test with social trust quartiles as dummy variables Economic growth and social trust levels of major European powers in 1914

9780230113176_01_previii.indd vi

99 109 112 113 114 117 118 120 122 154

2/5/2011 10:52:22 AM

Acknowledgments

This book would not have seen the light of day without the comments, advice, and support of a wide variety of individuals. First and foremost of those is Professor Markus Crepaz who, during my doctoral studies at the University of Georgia, not only pulled the curtains on the world of academic research in a way that all but predetermined my future career choices, but was also responsible for planting the seeds of the thought process which eventually germinated with this book. I also owe a considerable debt of gratitude to Professors Jeff Berejikian and Jaroslav Tir for introducing me into the subfields of political decision making and international conflict. I doubt I could have written this book without the guidance provided by these three individuals, each of whom strongly influenced my thinking on these matters. I also cannot understate the support I received from my dearest companions in life, namely, Cordelia, Konrad, and Zosia, who took turns as sources of encouragement, motivation, and sanity, and from my father-in-law Professor Emeritus Charles Bowlus who initiated numerous discussions on the applicability of the theoretical model developed within this volume to the study of medieval conflicts. Last but not least, recognition must be given to my students in the undergraduate Political Psychology class I taught during the Spring 2010 semester, whose active participation in the discussions of many of the issues addressed within this book was of great help in clarifying my thoughts and improving this manuscript.

9780230113176_01_previii.indd vii

2/5/2011 10:52:22 AM

9780230113176_01_previii.indd viii

2/5/2011 10:52:22 AM

CHAPTER 1

Introduction

I

n the summer of 2005 my family and I were marking that year’s Memorial Day by visiting for the first time the most recent additions to the National Mall, namely, the FDR Memorial and the World War II Memorial. While I was deeply impressed with the simple and poignant FDR Memorial, the World War II Memorial at first left me deeply puzzled. With a deep and long-standing interest in the history and politics of the conflict, and having seen such memorials in other countries, I felt that the design of the memorial was quite unlike any other. For one, none of the big issues over which that war was fought seemed adequately reflected in the choice of a design consisting of individual columns representing U.S. states and territories arranged in two semicircles facing each other across a set of fountains. To put it in different terms, what seemed missing from the World War II Memorial was the war itself. Happily, a U.S. Park Service brochure very helpfully eliminated my confusion by informing me the memorial was intended to commemorate the American unity experienced during the conflict. Upon reading that explanation I couldn’t help being struck by this predominantly inward-looking interpretation of the significance of that tremendously consequential global armed conflict. Given the importance of the issues that were decided on the fields of battle in that conflict, why was America’s unity chosen as the most important aspect of that conflict to commemorate? But even more troubling was the growing realization that the memorial design was squarely at odds with its stated intent, for the impression given by the two sets of individual and separate columns arranged in a broken oval was of both fragmentation and division. I remember myself only half-jokingly remarking that the builders accidentally constructed a U.S. Civil War Memorial, so striking was the internal disunity and conflict embodied in that monument.

9780230113176_02_ch01.indd 1

2/5/2011 10:52:39 AM

2

●

Social Trust, Anarchy, and International Conflict

At the time of our visit I was in advanced stages of my graduate education and was grappling with several problems in the study of international conflict, specifically the democratic peace and diversionary war theories. Almost by accident, our visit to the World War II Memorial provided a crucial insight that lies at the core of this project, namely, the nature of the social effects of international conflict. Monuments, especially national ones, are less a reflection of the era they are intended to commemorate than of the political era in which they are built, and the Memorial’s failure to properly reflect the issues that were viewed as important when the war was actually being fought is consistent with that pattern. The Memorial’s focus on American unity allegedly achieved during World War II was all the more remarkable considering that the actual state of unity and societal cohesion in the United States during the war was in fact far from perfect. To cite but one example, African Americans spent the entire war relegated to a virtual second-class-citizen status, in spite of the fact that so many of them volunteered to serve and fight in the U.S. armed forces. Yet even the military was no refuge from the pervasive discrimination that plagued the United States during that era, for even when African American soldiers were allowed to bear arms instead of serving in a variety of supporting, though often hazardous, duties, they did so in segregated, all-black units. Even worse was the plight of the Japanese Americans, many of whom spent the war in internment camps, also in spite of the fact that many of them served with distinction in the European theater of operations. The stark reality of racial and cultural divisions in the United States in the 1940s suggests that for the Memorial to focus on American unity during that period, the craving for a measure of societal unity and cohesion in the 1990s, when the Memorial was at last constructed, must have been very strong indeed. It was, as someone once only half in jest defined nostalgia, a “yearning for things as they never used to be.” Indeed, I do not believe it is a coincidence that the Memorial’s preoccupation with U.S. unity and societal cohesion and the “Greatest Generation” nostalgia both took place during a time of growing concern about the slowly fraying fabric of the U.S. society, the fraying that was comprehensively described in Robert Putnam’s Bowling Alone. Moreover, the United States is virtually (but not entirely) alone in viewing World War II as an event with positive social consequences.

9780230113176_02_ch01.indd 2

2/5/2011 10:52:40 AM

Introduction

●

3

During my extensive travels around Europe and the postcommunist newly independent states of Eastern Europe and former Soviet Union during the 1990s, I was struck by the fact that there was only one other country sharing the American nostalgia for the world’s bloodiest and most destructive war. That country is the Russian Federation, a country that in the wake of the collapse of communism has experienced a far more acute crisis of national identity, unity, and societal cohesion. This evident linkage between attitudes toward international conflict and the health of society within a state suggested that might be the answer to the problems afflicting both the diversionary war and democratic peace theories, and offered a promise of an overarching theory of international conflict that simultaneously accounted for the phenomena of democratic peace and diversionary war. And the importance of societal cohesion as a determinant of international conflict was precisely what was missing from both the diversionary war theory, which focuses mainly on government or leader unpopularity as a cause of conflict, and the democratic peace theory, which links international belligerence with regime type. For, even though nostalgia in general is not the most constructive of emotions, concealing as it does the ugly aspects of the past while at the time overly glorifying or perhaps even inventing positive ones, nostalgia for war and the societal benefits it allegedly brings strikes me as a dangerous fantasy to engage in. The readers no doubt recall a number of political commentators following the Al Qaeda terrorist attacks against the World Trade Center and the Pentagon on September 11, 2001, expressing hope that the attacks would recreate unity among the American people that has been lacking for some time. However, if one begins to think of war, the most senseless and destructive of human enterprises, as a source of salvation from a malaise afflicting one’s society, then surely such sentiments cannot help but increase the chances that the country experiencing the fraying of its civil society will not become more willing to seriously contemplate provoking international conflicts solely to shore up its crumbling internal unity. That societal cohesion matters enormously in domestic politics has been abundantly clear for some time. Ever since its appearance, Robert Putnam’s Bowling Alone (Putnam 2000) has had a profound impact on our understanding of the importance of social engagement on a variety of domestic social and political phenomena. It demonstrated

9780230113176_02_ch01.indd 3

2/5/2011 10:52:40 AM

4

●

Social Trust, Anarchy, and International Conflict

that healthy and unhealthy societies display radically different patterns of internal behavior, and furthermore showed that regime type alone was not enough to account for certain domestic phenomena. Now time has come to develop a similar analytical framework and apply it to the study of international phenomena, starting with international conflict. In fact, the success of Putnam and other scholars of social capital, social trust, and civil society begs the question whether a similar theoretical approach, one emphasizing social engagement and cohesion over regime type, to both domestic and international conflict might not yield a similarly improved understanding of these international phenomena. That such results should be expected is evident from the current state of the domestic and international violent political conflict subfields. While democracy has been one of the more popular domestic variables applied to conflict behavior of states, empirical findings have been ambiguous or even contradictory. For example, the level of democratization has not been conclusively and unambiguously shown to have a salient and profound effect on states’ propensity to initiate international conflict, with the exception of the consistently substantiated “democratic peace” hypothesis, which, however, has been validated only on state pair, or dyadic, level, and only for actual wars, in other words conflicts with relatively high level of intensity. It is also striking that the considerable body of democratic peace research has thus failed to conclusively confirm the notion that democracies are inherently more pacific than non-democracies. Even though such arguments are often made on the grounds that democratic leaders operate under the constraints of checks and balances, which allegedly ensure decisions to use force cannot be made rashly or hastily, popular sovereignty, which holds leaders responsible for their decisions to initiate international conflicts, and the peaceful norms of political conflict resolution prevailing within democratic societies, and which are supposedly externalized and applied to international interactions of democratic actors. This relative weakness of a variable as important as regime type has sparked a number of research agendas aimed at investigating other variables, for example, the level of economic development, or specific aspects of democratic regimes. So far, however, there has not been a major effort to apply a social capital / social trust–based theoretical framework to the study of conflict. This project aims to fill that gap by following in Robert Putnam’s footsteps and further demonstrating the importance

9780230113176_02_ch01.indd 4

2/5/2011 10:52:40 AM

Introduction

●

5

of a theoretical focus not merely on the means of government, but also on its ends. Similarly, while the diversionary war theory has a strong intuitive appeal, strong and robust empirical support for the theory is yet to materialize, in spite of a long-standing and voluminous research program. After all, who can argue against the proposition that state actors when facing severe domestic political problems and thus facing the prospect of losing power may be tempted to improve their political fortunes in order to divert attention of the public from domestic problems and focus it on the international arena. Moreover, the prospect of an international conflict may benefit an embattled political leader by creating a situation in which any would-be opponent has to be prepared for being perceived and/or accused of either directly or indirectly aiding the external enemy, and by enabling the political leaders to employ means of repression against the opposition that would be inconceivable in peacetime. The weakness of the empirical support for the diversionary theory therefore can be taken as an indication that while the theory may be sound, the conditions under which diversionary incentives exist require a reexamination. Here, too, the application of social trust offers a way out by raising the possibility that diversions occur not in response to governmental unpopularity, but rather under the conditions of weak societal cohesion. This would explain the weakness of empirical findings in the large-scale quantitative studies, because the leaders engaging in diversionary behavior may be doing so not in a bid to improve their personal popularity, but rather to increase the cohesion of their society by forcing them to, in effect, close ranks by presenting them with an external enemy. This, of course, also has the desired effect of forcing the now-cohesive polity, all prior internal conflicts and resentments at least temporarily forgotten, to rally not only around the flag, but also around the individual holding that flag, namely, the leader or leaders of the state in question. Therefore the overall objective of this book is a rather ambitious one, namely, the creation of an overarching theory of international conflict that accounts for the phenomena of both democratic peace and diversionary war, but does so in such a way as to eliminate the above-listed shortcomings of these theories. I believe that the concept of social trust, a concept thus far largely ignored in the study of international phenomena, holds the answer to the problems inherent

9780230113176_02_ch01.indd 5

2/5/2011 10:52:40 AM

6

●

Social Trust, Anarchy, and International Conflict

in both of these theories, can be effectively applied to establish a more comprehensive theory of domestic origins of international conflicts and of linkages between domestic and international political violence. To be sure, this project does not represent a simple application of Putnam’s analytical framework to the study of conflict. Whereas Putnam emphasized the importance of civil society and social capital, which in turn are the product of interpersonal associations, this study’s approach is based on the importance of the state’s role in eliminating domestic anarchy, the resulting creation of generalized social trust, and its impact on the state’s international interactions. My intent is to show that the concept of anarchy, usually treated exclusively as a feature of only the international system (with notable exceptions of ethnic conflict literature and research on state failure, both of which recognize the possibility that anarchy and the attendant phenomenon of domestic security dilemma may exist on domestic, state level of politics), not only may exist on domestic level but, moreover, it has a considerable influence on the conflict behavior of the state so afflicted. As so many states in existence today are well ordered, well governed polities where anarchy is all but nonexistent, it is sometimes forgotten that not all states in the world are this fortunate. There are many states in the contemporary international system whose domestic politics display features of anarchy. And yet, it must be remembered that even as many states fail to reach this ideal, it must be remembered that the elimination of anarchy and the creation of a system of effective governance is the main responsibility of the state and arguably the foundation of its legitimacy. Moreover, as I hope to demonstrate, a state’s success or failure in conquering anarchy within its own borders exercises considerable influence on the nature of that state’s interaction with other international actors. Therefore the influences of anarchy, and of its associated phenomenon of generalized social trust, on conflict behavior both on domestic and international level are the focus of this study. The use of anarchy as a concept also offers a useful alternative to the oft-used autocracydemocracy continuum inasmuch the latter concept s essentially refer to the process of governance; focusing on the effectiveness of governance reflects the outcome of that process. The difficulty in linking democratic form of governance to pacific international policies (with the important exception of “joint democracy,” which is the basis of

9780230113176_02_ch01.indd 6

2/5/2011 10:52:40 AM

Introduction

●

7

the democratic peace theory) than those pursued by their authoritarian peers is likely due to the tremendous variation in the quality of governance among both democratic and authoritarian regimes. The focus on the effectiveness of governance may help eliminate empirical problems associated with that variation, and add to the very short list of international conflict initiation-influencing concepts whose effects are evident on the state level of analysis. I also hope that the emphasis not merely on the process by which a state is governed but rather on the effects of said governance will help refocus attention on the importance of maintaining effective governance even in the era in which cutting government programs through a variety of so-called austerity programs appears to be in vogue at the time this manuscript was being finished. Given the importance of a strong, well governed state not only in maintaining domestic tranquility and order but also international stability, any backsliding in this area is fraught with many dangers. While these dangers are avoidable, it may well be that the influence of effective governance has been so pervasive for so long that it is now being taken for granted. The layout of the project is relatively straightforward. I start by outlining the current state of affairs in the democratic peace and diversionary war research programs, whose critiques increasingly point at the importance of not only regime type, but of the “state of the state.” The subsequent chapters examine the question of international and domestic anarchy, including the nature of the phenomenon and its effects, in greater detail, and point toward the considerable ideational effects (suggested by national identity and social trust research programs) of effective governance. I then argue that the nature of the phenomenon of social trust, in turn, has considerable implications for domestic and international conflict, leading to several hypotheses on domestic and international conflict initiation and targeting. The theory I develop in further chapters of this book posits that it is not the type but the quality of governance that really matters when it comes to state propensity to initiate conflicts and even to their likelihood of becoming victims of aggression. High quality governance that is the property of strong states results in the creation of high levels of generalized social trust, reducing the likelihood other states will be perceived as hostile, and projecting an outward image of a nonbelligerent, trustworthy, and cooperative actor, which in turn reduces the likelihood

9780230113176_02_ch01.indd 7

2/5/2011 10:52:40 AM

8

●

Social Trust, Anarchy, and International Conflict

others will resort to force in their dealings with high-trust states. Conversely, poorly governed weak states with low levels of generalized social trust not only tend to view their surrounding world as inherently hostile, menacing, and belligerent, but the violence of their own domestic politics is perceived by other international actors as evidence of their hostile intent. These hypotheses in turn are tested and analyzed in the final chapters of the project. I opted for an interdisciplinary approach with concomitant liberal borrowing of concepts and placing them into hitherto unfamiliar contexts. The fragmentation of social sciences into disciplines, fields, and subdisciplines has hampered the emergence of a truly comprehensive understanding of these phenomena, in no small part due to the sheer proliferation of concepts as each discipline, field, and subfield has developed its own peculiar language. One is reminded here of the old parable on “seeing the elephant.” As a result of the balkanization of social sciences we may have lost sight of not only the fact that there is but one elephant, but also that the various disciplines may often be examining the same aspects of the elephant’s anatomy. Thus, for example, an examination of political science, sociology, and psychology literature reveals that research on the crucial question of intra- and inter-group dynamics and conflict may be found scattered among bodies of literature that address the concepts of anarchy, group identity, nationalism, ethnic conflict, and social capital / social trust. The purpose of this project is to help bridge these already highly developed bodies of research and attempt to ascertain what the impact of the combined wisdom of several social sciences might be on our understanding of domestic and international violent political conflict. It is also my hope that this project will spark a new round of debates on the nature of diversionary war and democratic peace theories and help take these two very important research programs in new directions. Finally, living as we are (at least those of us in the United States) in the aftermath of the end of the “era of big government,” hopefully this book will make a contribution to the national debate on the proper role of government in society by bringing attention to the importance of preserving an active and strong government as a guarantor of domestic tranquility, and the very real negative consequences that will inevitably follow should government prove unequal to its responsibilities.

9780230113176_02_ch01.indd 8

2/5/2011 10:52:40 AM

CHAPTER 2

Democratic Peace and Diversionary War

D

emocracy is one of the more popular domestic variables used to explain conflict both on domestic and international level. It has been used as an explanatory variable to address such disparate types of conflicts as wars, militarized interstate disputes (MIDs), secessions, ethnic conflict, and political repression. Nevertheless, while the empirical findings on the state pair (dyadic) level of analysis are quite robust, it is nevertheless striking that the findings on the monadic level are far weaker, casting doubt on the efficacy of democracy as an explanatory variable. Possibly even more damaging are the lingering doubts concerning the logical underpinnings of the democratic peace theory. The diversionary war theory, on the other hand, suffers from an opposite problem: while the logic of this very intuitively attractive theory appears to be sound, the mystery that begs to be resolved is why the empirical support is not as robust as one would expect it to be. Moreover, given that these are both “second image,” or domestic politics–based theories of international conflict, one has to raise the question whether they are not both part of a larger phenomenon, whether it might not be feasible to construct an overarching theory of domestic causes of international conflict that incorporates both of these theories while at the same time avoiding their empirical or logical flaws. The Logic of Democratic Peace

The straightforward assertion of the democratic peace theory that democracies are extremely unlikely to go to war against one another essentially rests on two separate logical arguments. The structural or

9780230113176_03_ch02.indd 9

2/5/2011 10:53:16 AM

10

●

Social Trust, Anarchy, and International Conflict

institutional argument attributes the existence of democratic peace to constraints placed on decision makers by democratic institutions, as well as the institutionalized forms of peaceful conflict resolution, in the form of checks and balances or divided government that ensures no single individual or group of individuals has the authority to initiate armed international conflicts without the approval of a broad range of domestic political actors. In addition popular sovereignty, which is held to be a pacifying factor on the grounds that the citizens of a state who will, after all, have to bear the cost of the conflict in both blood and treasure, will object to their leaders’ decisions to initiate conflicts and hold them accountable by voting them out of office (Russett 1993). The corresponding ideational argument in favor of democratic peace, on the other hand, rests on the assumption that the decision makers in democratic states, by virtue of possessing an effective means of peaceful resolution of domestic political conflicts, internalize such methods and apply them when dealing with international actors as well. Moreover, leaders of democratic states are also assumed to view one another as sharing liberal values and democratic norms, with the perception of shared membership in this exclusive club being sufficient to avert the resort to violence in the dealings between such states. While some ideational works on democratic peace focus on elite beliefs and attitudes (Hermann and Kegley 1995; Doyle 2005), others extend it to the masses as well. Owen (1994), for example, argues that the citizens of a democratic state are less easily mobilized for conflict against other democratic states due to the perception that other democracies share the values and norms of the state in question. The importance of two democracies correctly perceiving one another as such is underscored by Starr (1997). A similar, though more detailed, argument is offered by Van Belle (1997), who attributes the mutual perceptions of the two democracies to the mass media. When two democracies come into conflict, the domestic news media on both sides accept each other as legitimate sources of information. As a result, due to the mass perceptions of a would-be adversary as another friendly democracy, democratic leaders have far less influence on how the conflict is reported by the media, which in turn raises the political costs of initiating conflict. The role of the media in presenting information about a would-be adversary in an international conflict, according to Van Belle, therefore accounts for

9780230113176_03_ch02.indd 10

2/5/2011 10:53:17 AM

Democratic Peace and Diversionary War

●

11

one of the possible exceptions to the democratic peace theory (since both the United States and Spain had democratic political regimes at the time of the conflict) by suggesting that the Spanish-American War occurred in spite of both adversaries being democracies simply because Spain, at the time a relatively new and weak democracy that would, after all, succumb to a fascist coup a few decades later and not reemerge as a democracy until the late 1970s, lacked the institution of free press that would serve the function of presenting the image of Spain as a liberal democracy to the United States. Moreover, the interaction between United States and Spanish media was complicated by the near-absence of modern electronic communications, which enormously slowed down the trans-Atlantic exchange of information. This explanation of the crucial role of the media has the added benefit of accounting for the occasional resort to covert action by one democracy against another, because such actions naturally avoid the scrutiny of the media and, moreover, covert actions by definition do not require the mobilization of public opinion (Van Belle 1997). The importance of shared perceptions on both elite and mass levels is also underscored by Weart’s (2001) argument to the effect that criticism of democratic peace theory on the basis of wars fought between ancient Greek democracies fails to acknowledge that arguably Athens did not perceive Syracuse as behaving like a fellow democracy. However, upon closer scrutiny both the ideational and institutional explanations are suspect on both logical and empirical grounds. Checks and balances notwithstanding, the fact remains that the chief executives of even democratic states retain considerable discretion in the international use of force. Very often such powers (for example, the powers of U.S. presidents inherent in their role as the commander in chief of the armed forces) are bestowed on them out of the sheer necessity to maintain the ability to react quickly to crises and to preserve the unity of command the effective application of military power requires and national legislatures cannot provide. Moreover, the argument that the leaders of democratic states are constrained by the need to maintain public support or risk losing the next election due to the public disapproving if the war fails takes into consideration the almost inevitable “rally around the flag” effect that is highly likely to accompany an international conflict. In accordance with the logic of diversionary war theory, even a democratic leader initiating

9780230113176_03_ch02.indd 11

2/5/2011 10:53:17 AM

12

●

Social Trust, Anarchy, and International Conflict

an international conflict would likely receive a considerable boost in popular support, support that would wane only in the event the war proved long and costly. The lack of effective institutional checks on the war powers of U.S. presidents was quite clearly demonstrated in the run-up to the invasion of Iraq in 2003. Unsurprisingly, the notion that democratic leaders pursue more pacific policies because of their need to seek the approval of their (presumably antiwar) publics (a frequently unwarranted assumption, as shown by, for example, the high levels of U.S. public support for the invasion of Iraq in 2003) has not remained unchallenged in the international conflict literature, as a number of studies have shown that autocratic leaders may actually have greater reason for concern about the political consequences of wars (Chiozza and Goemans 2004; Rosato 2005). The ideational argument for democratic peace likewise suffers from a number of problems. Although democratic leaders are sometimes alleged to have more benign perceptions of other international actors (a factor sometimes cited as the reason for the existence of the democratic peace phenomenon), and moreover the transparent nature of democratic politics is reputed to account for benign mutual perceptions by democratic states, the linkage between democracy and perception has been questioned as well (Hagan 1994). Explanations such as offered by Van Belle’s emphasizing the role of perceptions and the process by which they are shaped are problematic for a number of reasons. One of the question-begging aspects of that argument is the implication that the democratic peace phenomenon would not have been observable or, for that matter, existing, prior to the emergence of modern communications, given the importance Van Belle attaches to the ability to transmit large numbers of information across large distances. Moreover, apart from suggesting that leaders of democracies are either unwilling to acknowledge the democratic character of their adversaries or are pursuing conflictual policies in full awareness of the democratic nature of the adversary probably Van Belle exaggerates the independent role played by the media during crisis situations. Here, the democratic peace theory research program appears to suffer from insufficient attention being given to the effects of armed conflict, or even an acute international crisis, on public opinion. As the diversionary war theory well documents, such conflicts have the effect of rallying the public around their state and their leaders, and the stress

9780230113176_03_ch02.indd 12

2/5/2011 10:53:17 AM

Democratic Peace and Diversionary War

●

13

induced by the conflict moreover has the effect of greatly reducing the effectiveness and clarity of decision making (Jervis 1976). The inherent assumption made by Van Belle appears to be that, contrary to the expectation of a rallying effect caused by the conflict, the members of the news media would somehow remain unaffected and wholly objective even as the rest of the country was swept by war fever. Indeed, a number of works have suggested that perceptions of other states may vary quite rapidly irrespective of their actual nature. For example, Oren (1995) has demonstrated how even though the nature of the Wilhelmine Germany has not changed over the late-nineteenth and early-twentieth centuries, the perception of that state and its intentions in United States have undergone a dramatic change over that period of time. A similar transformation occurred in the United Kingdom, which as late as the 1890s viewed Germany as an essentially friendly international actor and a welcome, stabilizing influence on the European continent. Yet by the beginning of the twentieth century Great Britain was joining the entente cordiale with France and the Russian Empire and emphasizing the importance of the German threat in its defense plans. It should also be noted that while Japan was viewed in the United States in essentially friendly terms in the late-nineteenth and early-twentieth centuries, that perception rapidly changed once Japanese and U.S. interests began to clash in China and other parts of Asia. Similarly, in the case of the Spanish-American War, the role of the so-called yellow journalism, which whipped up war fever in the United States, must be acknowledged as a factor facilitating the outbreak of hostilities. Oren therefore quite rightly attributes the perceptions not to the objective nature of the regime in question but rather to subjective interests of key domestic actors in the perceiving state, with the subjective interests driving the perceptions. A more recent example of a dramatic change in perception of a state that was quite divorced from any internal regime change occurred between the United States and Soviet Union during the 1940s. The 1940s-era Soviet Union, a totalitarian state by any known criteria, was widely perceived as a practically democratic ally during World War II with the assistance of the U.S. media and film industry that was enlisted in making the case for the U.S.-Soviet alliance against Nazi Germany, a perception that practically overnight vanished following the end of World War II, to be replaced with a

9780230113176_03_ch02.indd 13

2/5/2011 10:53:17 AM

14

●

Social Trust, Anarchy, and International Conflict

this time correct image of an implacable totalitarian adversary. Similarly, in the wartime USSR, the Soviet official media tended to portray the United States in quite favorable terms, only for this benign image to disappear once the shared interest in defeated Nazi Germany vanished (Oren 1995). To cite another example, consider the case of the contemporary relationship between the United States and France. On the face of it, and consistent with the ideational argument in support of the democratic peace theory, these two states should enjoy uniformly strong positive images of one another. After all, they are both well established, strong democratic states with equally well established traditions of free press and other forms of media. They are both at high levels of economic development, fought on the same side in most wars during the past two centuries, and are in fact both members of the North Atlantic Treaty Organization (NATO). Moreover, in view of the fact that we are living in the era of the Internet and the near-instant transmission of vast amounts of information, and the fact that both of these countries have highly literate and comparatively politically educated publics, there are no obvious obstacles to these two states accurately perceiving one another as democracies and therefore maintaining positive images of one another. Therefore, apart from the rather negative image France enjoys in the United States even in the best of times, the level of virulent antiFrench sentiment that found expression even in the august halls of U.S. Congress, which renamed the “French fries” normally served in its cafeteria as “Freedom fries,” due to the French (well-founded, as it turned out) opposition to the U.S. efforts to secure a UN Security Council resolution authorizing the invasion of Iraq, was quite striking. It was quite evident that what drove the perception of France in the United States were not so much the objective attributes of the French state and society, but rather the subjective interests of the United States, which, however temporarily, were perceived as clashing with those of France. These examples make it quite evident that perceptions of other states are very frequently determined by other factors, including the ability to transmit information across large distances and the interests of the states in question. Oren’s discussion of the factors influencing the perceptions of other states and the examples of misperceptions of other states also raise the possibility of a motivated bias (Jervis 1976) influencing the

9780230113176_03_ch02.indd 14

2/5/2011 10:53:17 AM

Democratic Peace and Diversionary War

●

15

perception of leaders and the masses and operating quite irrespective of the objective nature of the state being perceived. Since individuals operating under the influence of motivated bias (for example, induced by the interests at stake in the conflict against another state) tend to ignore, dismiss, or minimize information that is contrary to their biases, the availability of even large amounts of accurate information is unlikely to make an impact on their perceptions (Jervis 1968). Indeed, most works studying the role of misperception in international relations do not distinguish between democracies and autocracies and their intentions and motives in terms of accuracy of perception of (or by) other states, with said perceptions being rather prone to distortion (Jervis 1988), especially when under the influence of stress that is an inherent component of most international crises (Jervis 1976). Similarly, there is no evidence to suggest that the leaders or the publics in democratic states are any less prone to misperceive their adversaries than leaders and publics in autocratic states, and are no less likely to view others as more hostile and threatening than they really are (Jervis 1968). A recent study (Mercer 1996) of reputational effects finds that whereas situational factors are usually seen as accounting for other actors’ desirable behaviors, dispositional ones account only for undesired behaviors. Since a regime type is essentially a dispositional, rather than a transient situational, factor (Mercer 1996), the notion of state leaders or publics attributing positive characteristics to other states on the basis of regime type is therefore highly questionable. Nor is it clear that democratic regimes tend to produce pacific leaders. One of the works examining the linkage between democracy and aggressiveness brought about by perceptions of other states is Hagan (1994), who questions the premise that democracies are less likely to be governed by aggressive, war-prone leaders than autocracies, and, while remaining convinced that international aggressiveness of leaders has roots in domestic politics, seeks to find these roots elsewhere. Hagan is echoed by Hermann and Kegley (1995) who nevertheless argue that the democratic peace phenomenon is caused by leaders’ perceptions and leadership styles. More sanguine than Hagan (1994) about the linkage between democracy and leaders’ perceptions, Hermann and Kegley (1995) rely on the social identity theory to lay out their explanation of how democratic leaders’ images and beliefs about other state actors are formed.

9780230113176_03_ch02.indd 15

2/5/2011 10:53:17 AM

16

●

Social Trust, Anarchy, and International Conflict

An even greater problem for the logic of democratic peace theory is the empirical evidence that the pacifying effect of democracy enjoys strong support only on state pair (dyadic) level, that is to say, only when both would-be combatants are democracies. The strength of this evidence is such that democratic theory scholars acknowledge the alleged pacifying effect of democracy is not to be observed on monadic level (Kinsella 2005). However, this is highly problematic for the argument that a democratic system of government results in a more pacific international policy because, as Rosato (2005) points out, much of the logic of the theory, including its reliance on democratic norms, institutions, the delayed resort to force due to the need to follow the often cumbersome democratic procedures, and the improved ability to signal resolve on the grounds that democratic states enjoy an advantage when prosecuting wars, concerns the characteristics of individual states and not dyads. Given all of the aforementioned qualities of democratic states, one should not unreasonably expect democratic states to be more pacific in their international interactions than non-democracies, regardless of whether the potential adversary is a democracy or an authoritarian state. And yet, although a number of authors have argued that democracies are more peaceful and less conflict initiation–prone (Rummel 1997; Maoz 1998; Russett and Starr 2000), others have found that democratic states have proven to be every bit as willing to go to war as autocracies (Roy 1993; Chan 1997; Rosato 2005). Moreover, democratic nations have not exhibited a reduced willingness to bear the costs of war (Roy 1993), a finding consistent with the democratic peace notion that democracies can credibly signal resolve precisely due to their reputation for ability to bear such costs. Casting further doubt on the impact of democracy on conflict initiation, some have even argued that under conditions such as regime unpopularity, democratic states may have greater incentives to engage in international conflict than authoritarian ones (Gelpi 1997; Hess and Orphanides 1995, 2001; Russett 1990) due to democratic leaders’ need to win an upcoming election, a necessity not faced by authoritarian ones. Indeed, the notion of democracy exercising pacifying influence on state policies also runs counter to the realist theory of international relations that asserts that states are compelled to act in a similar manner regardless of their domestic political system due to either the

9780230113176_03_ch02.indd 16

2/5/2011 10:53:17 AM

Democratic Peace and Diversionary War

●

17

sameness of their interests (Morgenthau 1948) or to the structural forces exerted by the system itself (Waltz 1979). The near-wars between France, Britain, and the United States during the nineteenth century, or the already discussed Spanish-American War, challenge the notion that shared democratic institutions or values are a restraining factor on either the leaders or the masses (Roy 1993), all the more so since democracies do not reliably externalize their domestic norms of conflict resolution and do not trust or respect one another when there is a clash of interests (Rosato 2003). Some supporters of the theory acknowledge that democracy by itself is no guarantee of peace, even in the case of shared democracy (Doyle 2005), but rather that it ought to be viewed as one of the three pillars of liberal, or Kantian, peace alongside economic interdependence and respect for human rights, an argument that unfortunately also undermines the claims of democracy’s salience as a determinant of international conflict. And, indeed, the armed conflicts between ancient democracies point to the possibility that shared democracy does not necessarily eliminate the possibility of war within the dyad (Weart 2001). The weakness of empirical support for the notion that democracies are inherently more peaceful due to their liberal domestic ideologies and nonviolent domestic conflict resolution norms has even led to concerns that Kant’s notion of a pacific union of democracies resting on three pillars of democracy, common moral bonds established between democracies, and the democracies’ economic cooperation toward mutual advantage may not be inevitable (Sorensen 1992). However, if for peace to prevail all three conditions of Kant’s liberal peace must be satisfied, then the contribution of democracy is at best a relatively minor one, assuming it exists at all. Another strand of research concerns differences between different types of democracy and their influence on conflict behavior. The propensity of ancient democracies to go to war with one another suggests a need to focus on differences between ancient and modern democracies, which in turn indicates the possibility that intrademocracy differences in general might be important (Robinson 2001). Specific regime variations singled out for analysis include institutional structure, specific rules of the democratic process (Clark and Nordstrom 2005), as well as political accountability arrangements (Huth and Allee 2002). Similarly, the variation in the behavior of

9780230113176_03_ch02.indd 17

2/5/2011 10:53:18 AM

18

●

Social Trust, Anarchy, and International Conflict

both democratic and authoritarian states has led to calls for more sophisticated conceptualizations of both regime types (Rioux 1998; Clark and Nordstrom 2005). A number of works have raised the possibility that democratic peace is the product of factors other than democracy itself. While Cederman (2001) points to a steadily growing pacifying effect of state pair democracy, he also notes that other, nondemocratic or mixed dyads experience a similar effect, albeit a weaker one, suggesting systemic factors and maturity effects may be responsible. Others have suggested democratic peace is simply the outcome of regime similarity and that identical pacifying effects may be found among dyads sharing political system similarities (Werner 2000). Chan (1997) finds that both autocratic and democratic dyads are more peaceful than mixed dyads, a result that is supported by Peceny et al. (2002) who discover pacific tendencies among certain types of authoritarian dyads and, on a monadic level, considerable variation in conflict-proneness among different types of authoritarian regimes, indicating that variations within each regime type may be responsible for variations in conflict behavior. State conflict-proneness has also been attributed to policy preference congruence, with no clear linkage between these preferences and the type of regime (Clark 2000; Gartzke 2000). Finally, regime stability itself may be a factor determining conflict behavior as well (Russett and Starr 2000); with relationship between democracy and conflict disappearing once political stability is controlled for (Maoz and Russett 1992). In the absence of a clear-cut empirical support for the proposition that democracies are less conflict-prone, there have been suggestions that perhaps the causal arrow runs in the opposite direction. It has been argued that instead of peace being the natural consequence of democracy, democracy may be the natural outcome of prolonged peace (Chan 1997; Gates, Knutsen, and Moses 1996; Rioux 1998). While Mousseau and Shi (1999) do find backing for the pacifying effect of a democratic regime, they acknowledge that conflict may have the effect of undermining a state’s democratic regime over the long term, a possibility suggested by Wolfson, James, and Stolberg (1991) who argue that individuals in their support of the war aims are willing to abandon domestic freedoms. Similar concerns about war ultimately undermining democratic institutions that require peace to

9780230113176_03_ch02.indd 18

2/5/2011 10:53:18 AM

Democratic Peace and Diversionary War

●

19

flourish are voiced by Barzilai (1999). Furthermore, to the extent that endemic warfare in the international system tends to undermine each of the pillars of Kant’s liberal peace by destroying international trade and undermining international norms and respect for human rights, the argument that the prerequisites both Kant and the proponents of democratic peace list as prerequisites for the absence of warfare has considerable logical soundness. Finally, the absence of evidence supporting the pacifying effects of democracy on monadic level has even led to the questioning of whether democracy can be used as an exogenous variable in its own right, but rather a by-product of other phenomena (Wolfson, James, and Stolberg 1991) as the structure of Kant’s formula for peace indicates a need to identify a component that is theoretically prior to democracy, liberalism, and international organizations, and which provides foundation for these components of Kantian peace (Peet and Simon 2000). Attention has also been drawn to material preconditions for democracy (Snyder 2002). Economic development and capitalism, in particular, have been singled out for attention (Mousseau 2009; Gartzke 2007) as conditions that are required for democracy to emerge in the first place. Thus it has been argued that the pacifying effects of democracies are limited only to those dyads where both states are not only democratic but also have well developed economies (Mousseau 2000) on the grounds that economic prosperity fosters stable governments, which in turn establish an atmosphere of individualism and the rule of law, which leads to a separate peace among democratic states, or only to stable, long-lasting and well consolidated democratic regimes (Bueno de Mesquita et al. 1999; Cederman 2001), with wealth being responsible for pacific behavior attributed to democracy (Gartzke 2000), or possibly a confounding variable (Maoz and Russett 1993). Other domestic variables used to account for the democratic peace phenomenon have included intergenerational differences, the aging of the population, and other similar factors (Garnham 1986). Similarly, Braumoeller (1997) argues, on the basis of his analysis of emerging democratic systems in the post-Soviet newly independent states, that democratic peace may be viable, but is nevertheless vulnerable to national issues, and to concentration of power in small groups. However, Braumoeller does not discuss under what conditions the

9780230113176_03_ch02.indd 19

2/5/2011 10:53:18 AM

20

●

Social Trust, Anarchy, and International Conflict

particular form of conflict-prone non-universalist liberalism is likely to exist. Nevertheless, Braumoeller’s injunction to consider the context within which a liberal political system exists likely points in the right direction, as does his argument that the causes of democratic peace phenomenon likely will not be ascertained by measuring democracy. The theory has also been subjected to criticism on the grounds that it suffers from design failures (Gates, Knutsen, and Moses 1996), that the findings confirming democratic peace tend to be nonrobust and dependent on variable construction, model specification, and choice of estimation procedure (Gartzke 2000). Linkages between democracy and various forms of domestic conflict have been similarly ambiguous. Although democratization has been linked with a decrease in political repression, for example, due to the existence of other avenues for conflict resolution (Davenport 1995, 1996, 1999), or the lack of both opportunities and willingness to engage in repression due to their institutional structure (Poe, Tate, and Keith 1999), the findings have not been entirely unambiguous, as the level of democracy deters the government’s use of repression only at very high indicators of democracy (Davenport and Armstrong 2004). Moreover, there appears to exist a curvilinear relationship between democracy and autocracy, with violent repression being less likely both at high levels of democracy and autocracy (Regan and Henderson 2002). Although the end of the Cold War has produced a wave of democratization across the globe and reduced international conflict, that wave apparently has not led to a noticeable improvement in the overall respect for human rights (Cingranelli and Richards 1999). The Empirics of Diversionary War

The diversionary war theory has precisely the opposite problem from that of democratic peace. On the one hand, the logic of the theory itself is extremely attractive, intuitively correct, and enjoying widespread popularity. The classical formulation of the theory posits that domestic problems create an incentive for the political leadership to behave in a belligerent fashion, up to and including war in order to increase that leadership’s popularity and thus remain in power (Levy 1989). And, indeed, there are a number of reasons for embattled

9780230113176_03_ch02.indd 20

2/5/2011 10:53:18 AM

Democratic Peace and Diversionary War

●

21

political leaders to embroil their countries in wars quite apart from the need to divert attention from political problems and thus diminish the risk of losing power. They include the almost inevitable expansion of the coercive and punitive capacity of the state (sometimes including the institution of martial law) when engaged in war, and putting one’s political opponents in the unenviable position of having to oppose a national leader when the said leader is marshalling the country’s resources for war, therefore risking the perception that the opposition is working in collusion with, or is even actively supported by, the enemy state. The popularity of the theory is reflected by, among other things, the almost inevitable likelihood that U.S. presidents will find themselves at some point during their term in office confronted with suspicions and even direct accusations their commitment of U.S. military force bore the marks of diversionary behavior. However, and remarkably given the logical soundness of the theory and the degree to which it has been internalized by the public (it is difficult to think of an international relations theory that has a more widespread support base), the empirics of the theory are the problem because the body of research supporting the notion that state leaders under the conditions of domestic political or economic problems and therefore facing threats to their authority and popularity will attempt to engage in diversionary violence (for example, Ostrom and Job 1986; Russett 1990; James and Oneal 1991; Morgan and Bickers 1992; Hess and Orphanides 1995; Gelpi 1997; Dassel and Reinhardt 1999; Morgan and Anderson 1999; DeRouen 1995; Enterline and Gleditsch 2000) is offset by another that fails to find the proposed relationship (Meernik 1994; Meernik and Waterman 1996; Leeds and Davis 1997; Gowa 1998; Meernik 2000; Mitchell and Moore 2002; Chiozza and Goemans 2003). Yet in spite of these critiques the phenomenon of diversionary war does appear to exist: qualitative studies of, for example, the Argentina’s invasion of the Falklands in 1982 (Lebow 1985) quite convincingly show that the Argentinean military junta was heavily influenced by classical diversionary motives. Indeed, given the conditions under which Argentina’s invasion of the Falklands unfolded, it is difficult to frame the motivation of the Argentinean junta in any terms other than diversionary. The Argentine decision makers had to have known that the British armed forces were about to embark on a round of painful and well publicized

9780230113176_03_ch02.indd 21

2/5/2011 10:53:18 AM

22

●

Social Trust, Anarchy, and International Conflict

cuts that would have greatly diminished the British amphibious assault and naval air capabilities, both of which proved vitally necessary to the eventual and successful British effort to retake the Falklands, an effort that earned the British carrier and amphibious forces a secure place in the British defense budget. Therefore, had Argentina delayed the invasion by only a few years, United Kingdom might have found itself in a position of having to accept the Argentine fait accompli, having at that point no real capability to mount a retaliatory effort. However, had the Argentine junta delayed the invasion long enough to actually assure its success, it could have well fallen victim to the growing popular discontent driven by the rapidly worsening economic conditions, and it is these conditions that forced the junta’s hand and compelled it to engage in an effort to rally the country around themselves by invading the Falklands (Lebow 1985). So how can one account for the difficulty quantitative, large-n studies have encountered in “capturing” the phenomenon of diversionary war? That a theory so intuitively correct has failed to garner robust empirical support is one of the major mysteries of contemporary conflict studies, all the more so that the diversionary effect appears to be very elusive and sensitive to case and method of study selection in large-n studies. A growing body of research has attempted to identify the causes of the theory’s weak empirical support. One set of critiques has focused on alternatives to diversionary violence. Thus Bennett and Nordstrom’s (2000) assessment of policy alternatives open to leaders of states engaged in an enduring rivalry included the peaceful settlement of the dispute, and efforts to alleviate the domestic problems that are the source of the crisis. Levy (1989) suggests the need for an improved understanding of the relationship between internal and external conflict, and for more precise specification of conditions under which a conflict becomes externalized. Bueno de Mesquita (1985) lays out a very wide variety of options open to beleaguered leaders, including shifting the blame to underlings or stepping up domestic repression, and critiques the assumption inherent in some diversionary theory works that the diversion must have the form of a conflict with another country. In that he is echoed by Russett (1990) who likewise frames the choices facing the beleaguered regime in terms being able to choose between external diversion and internal repression of the opposition, with the latter option

9780230113176_03_ch02.indd 22

2/5/2011 10:53:18 AM

Democratic Peace and Diversionary War

●

23

being of particular attractiveness to authoritarian regimes that are not circumscribed by constitutional protections of individual rights, an argument that is corroborated by Gelpi (1997) who finds that regime type does influence the proneness to engage in diversionary behavior. Yet another set of alternatives is explored by Leeds and Davis (1997) who include the option of manipulating the economy to remove the threat to the leader’s popularity instead of engaging in foreign adventures, an approach that is also suggested by De Rouen (1995). However, several other critiques of the theory have pointed out a number of more deeply seated problems. Richards et al. (1993) attribute the problem to the failure of diversionary war theorists to develop a comprehensive model of the relationship between domestic conditions and diversionary uses of force. In examining the impact of impending elections or economic crisis on international behavior of industrialized democracies Leeds and Davis (1997), for example, no consistent support is found for a relationship between domestic conditions and international behavior, and no evidence is found to support the contention such conditions promote displays or uses of force. Similarly, no link between the economic failure (a commonly used operationalization for government popularity) and the uses of force is found by DeRouen (1995), although he finds an indirect link between economy, politics, and the use of force. Examining the influence of presidential election cycles and the composition of the government on U.S. uses of military force, Gowa (1998) also finds no support for the diversionary war theory, suggesting instead that politics indeed do stop at water’s edge. Gowa suggests that the abstention from seeking diversions is consistent with rational expectation models, since the presidents are likely to know that efforts to use military force to improve their political fortunes will fail. Rather, according to Gowa, the U.S. presidents commit military force only in response to changes in national power and the advent of world wars. In the same vein, Meernik (2000) believes that results showing support for diversionary war theory may be skewed by selection effects, which lead to erroneous results, and that the apparent linkage between economic problems and conflict-proneness of states may be due to the appearance of weakness caused by poor economic performance inducing potential challengers to take advantage of the weakened nature of the sate by mounting a military challenge, thus increasing

9780230113176_03_ch02.indd 23

2/5/2011 10:53:18 AM

24

●

Social Trust, Anarchy, and International Conflict

the number of crises a state in dire economic straits may find itself in. A similar argument is put forth by Meernik and Waterman (1996), which argues that instead of reacting to domestic conditions, U.S. presidents make decisions on the use of force on the basis of U.S. national security concerns and U.S. responsibilities as a global hegemon. The higher importance of international variables instead of domestic ones behind the decisions to use or threaten the use of military force is also argued by Mitchell and Moore (2002). Bueno de Mesquita (1985) criticizes the failure to account for courses of action other than initiating an international conflict that leaders under pressure may resort to in order to boost their political standing, and also finds the theoretical support for scapegoating hypotheses predicting international conflict as a result of domestic unrest to be lacking. Tarar (2006) frames the question of diversion in terms of uncertainty about the competence of their incumbent leader who, when faced with economic problems, may engage in diversionary behavior in the hopes of reestablishing own reputation for competence on the grounds that effective performance in an armed conflict will overshadow the leader’s poor management of the economy and put to rest the doubts about the leader’s competence. However, Tarar includes the caveat that the action will have the desired effect only when the target is a comparatively powerful state, as weak adversaries do not create opportunities to display the leader’s prowess as a national leader. However, such a qualification considerably raises the stakes of conflict, because provoking an international crisis with a powerful adversary implies a significant likelihood that the diversionary ploy will backfire if the diverting leader is forced to back down. Yet another alternative explanation for the rarity of empirical support for the diversionary theory is the difficulty in finding suitable targets. Adopting a rather different tack in regard to the diversionary incentives stemming from economic crises, Smith (1996), Leeds and Davis (1997), and Clark (2003) argue potential targets of diversionary violence are aware such incentives will do their utmost to avoid giving provocation, suggesting that the national leaders are constrained by the need to find a plausible target for diversionary violence. A final set of critiques of the diversionary theory suggests weak empirical support may well be due to the manner of conceptualization of the diversionary incentive. One of the main problems facing

9780230113176_03_ch02.indd 24

2/5/2011 10:53:19 AM

Democratic Peace and Diversionary War

●

25

the researchers of diversionary violence is that the motives of political leaders are objectively unknowable. I would argue that there are no political motives that are more closely held than diversionary ones due to the very nature of the behavior being contemplated, namely, risking a costly and bloody war for the sole purpose of individual political gain. Therefore one certainly cannot expect candid disclosure of own motives, even in secret diaries, when it comes to political behavior as crass, cynical, and self-serving as initiating diversionary conflicts, for such a disclosure would surely doom any hope for a positive legacy of any political leader. To the extent that political leaders are expected by their followers and supporters to serve a common good, however defined, diversionary actions amount to a betrayal of such support. To make matters worse (and here admittedly the research is practically nonexistent), it may well be that diversionary motives operate on the subconscious level, due to the individual need to maintain self-esteem blocking the conscious recognition of the true motives of diversionary actions. Therefore, since political actors engaging in diversionary actions are not about to make a confession, diversionary motives unavoidably have to be operationalized by proxy. Diversionary war studies therefore have tended to employ a variety of measures of government unpopularity, which are usually operationalized in terms of economic performance (Russett 1990; Morgan and Bickers 1992; Meernik 2000; Tarar 2006), the level of domestic political unrest (Levy 1989), or imminent elections (Russett 1990; Morgan and Bickers 1992; Smith 1996). However, a considerable number of diversionary war critiques are pointing in a different direction. It may well be that the inconsistent empirical support is caused by an undue focus on government unpopularity, whereas diversionary violence may actually be a reflection of a far more fundamental problem. State Weakness: The Missing Link?

The question of state weakness and societal cohesion is a recurring theme in many diversionary war critiques that point out the “rally around the flag” effect diversionary action seeks to obtain is far from automatic. It has been suggested that diversionary action might be quite ineffective for certain states (Morgan and Anderson 1999;

9780230113176_03_ch02.indd 25

2/5/2011 10:53:19 AM

26

●

Social Trust, Anarchy, and International Conflict

Bueno de Mesquita 1985). While the theory posits that international diversionary crises ought to be launched in order to unify a fractured society. Morgan and Anderson (1999), however, argue that external diversion is going to be effective only if internal conflict is of moderate intensity and there already exists a certain degree of internal cohesion. Absent such cohesion, an attempt to divert might actually backfire if certain sections of society view the foreign “enemies” as less of a threat than other internal actors. Morgan and Bickers (1992) similarly question the assumption that all inhabitants of a given state automatically view themselves as patriots of that state and thus owe it support of that state, and in many cases allegiance to an ethnic group trumps national identity (van Evera 1994). Likewise pointing in the direction of state weakness, Gelpi (1997) argues democracies are more likely to divert because of their higher level of internal cohesion and willingness to identify with the state, and Simon and Starr (2000) argue that democratic regimes have the ability to deal with challenges to their authority by bolstering own legitimacy by means other than military action, which include the improvement in the economic situation in the country. A like pattern has been emerging in democratic peace theory critiques. Several works exploring alternatives to democracy as a variable influencing conflict behavior have shown the potential promise of pursuing this line of research. Peet and Simon’s (2000) study, for example, tests whether what they define as “social purpose” of the state accounts for that state’s propensity to engage in MIDs, and whether states with an embedded liberal social purpose are more cooperative on security issues. The core of Peet and Simon’s argument is that political authority is granted legitimacy if power is exercised in a manner that corresponds to social expectations between the ruler and the ruled, and that states whose governments use their power in a way that are compatible with the purpose for which this power was granted and thus enjoy a high degree of legitimacy tend to be less conflict-prone. The arguments made by Bueno de Mesquita et al. (1999) and Cederman (2001) that democratic peace benefits accrue only to stable, long-lasting, and well consolidated democratic regimes, and the findings that less stable partially free states are more war-prone than both autocracies and democracies (Chan 1997; Davies 2002; Mansfield and Snyder 2002) similarly raise the importance of

9780230113176_03_ch02.indd 26

2/5/2011 10:53:19 AM

Democratic Peace and Diversionary War

●

27

state strength as a contributing factor to international peace. An even more direct approach to the problem of state weakness as a cause of conflict is represented by Holsti (1995). Holsti argues that whereas strong states—which tend to be well integrated entities with a healthy relationship between state and society and strong government legitimacy that have stood the test of time—tend to be peaceful, weak and failing states have a much higher conflict propensity as their leaders attempt to shore up the crumbling national unity by resorting to international conflicts. Literature on ethnic conflict further suggests that such conflict may be less dependent on the nature of the state’s political system but rather its strength and weakness (Posen 1993; Kaufman 1996; Kaufmann 1996; Roe 1992; Saideman and Aires 2000; Taras and Ganguly, 2002), with ethnic minorities finding themselves in particular danger in situations where state weakness creates the so-called domestic security dilemma (Posen 1993) that renders minority groups vulnerable to attack by others. In a similar vein, Benson and Kugler (1998) argue that domestic peace is a function of efficient governance, irrespective of its institutional format, as new democracies tend to be at heightened risk for ethnic conflict and separatism (Simon and Starr 2000). Taken together, these disparate yet converging strands of scholarship suggest pacific domestic and international policies may be less a function of regime type, but rather the product of some quality peculiar to well established regimes, whether democratic or autocratic, which have successfully eliminated anarchy from within the borders of their states. Indeed, when one considers one of the outstanding exceptions to the democratic peace theory, namely, the Spanish-American War, it is quite clear that both adversaries were, by today’s standards, weak states. The United States, once again in spite of considerable nostalgia attached to the nineteenth century, was far from a Tocquevillian bliss it is often assumed to have been. Instead, it was a country only slightly more than three decades removed from the end of its bloody and protracted Civil War, a country riven by ethnic and social conflict, and a country that experienced no fewer than three successful assassinations of U.S. presidents in the late-nineteenth and early-twentieth centuries. Governmental authority in much of what today are the western U.S. states was at best sparse. Overall, the level of the unrest

9780230113176_03_ch02.indd 27

2/5/2011 10:53:19 AM

28

●

Social Trust, Anarchy, and International Conflict

and violent conflict in the United States was high enough to warrant describing the United States of the 1890s a relatively weak state. A similar description can be attached to Spain, a declining major power that was about to lose the last remnants of its once-vast colonial empire, which experienced a series of coups and upheavals in the nineteenth century and, in a little more than three decades, was to fall victim to its own bloody civil war that would be followed by nearly four decades of a fascist dictatorship. In view of the evident importance of state weakness as a factor that appears to be of even greater importance than regime type, it is a phenomenon that deserves considerable further attention. And, to the extent that weakness or absence of governance implies the presence of anarchy, it is anarchy that is the subject of the next chapter.

9780230113176_03_ch02.indd 28

2/5/2011 10:53:19 AM

CHAPTER 3

Anarchy, States, and Nations Anarchy as a Domestic Phenomenon

Although anarchy is deservedly one of the most heavily studied political phenomena, the overwhelming portion of this research has treated it as an international phenomenon, with relatively little thought given to the possibility that it might be a domestic phenomenon with far-reaching international consequences. This, in my view, is a considerable omission, especially since the expansive body of literature on ethnic conflict and state failure makes clear that anarchy can indeed be found as a feature of domestic politics, not solely international ones. Moreover, when considering the widely attributed effects of international anarchy on the nature of interaction between state actors, one becomes cognizant of the importance of effective governance and the effect such governance has through the elimination of anarchy on the domestic level. International Anarchy

Starting with Thucydides and continuing through contemporary structural realists, the long-standing realist argument has held that the effects of international anarchy on state actors are fixed and resistant to efforts to change (Ahrensdorf 2000). These effects include making conflict an ineradicable part of interaction of international actors. Realists claim that in a self-help system where there is no overarching central authority and where all states possess inherent offensive capabilities and therefore the ability to destroy other states, each state as an instrumentally rational and strategic actor must safeguard itself against the possibility of destruction (Grieco 1988). The absence of a world government and the preservation of

9780230113176_04_ch03.indd 29

2/5/2011 10:53:32 AM

30

●

Social Trust, Anarchy, and International Conflict

the institution of state sovereignty that ensures states retain the monopoly on the legitimate use of force and recognize no higher authority means, according to the structural realists, wars occur because there is nothing in the international system to prevent them (Waltz 1988). Given the absence of any institution with real power to prevent international aggression, anarchy is inevitably going to force states not only to seek their security by enhancing their power relative to other states, but the ensuing competition between states will frequently erupt into conflict and even outright warfare (Grieco 1988). Even though the international system may not be in a constant state of war, it is nevertheless under a constant threat of war that limits cooperation due to the logic of security competition. Since international cooperation will almost inevitably benefit one partner more than the other, the prospect of a potential adversary realizing a relative gain is sufficient, according to the structural realists, to forgo cooperation even if such cooperation offers the possibility of an absolute gain (Grieco 1988). The necessity of self-reliance combined with the threats posed by others leads to an overriding concern with relative power position over other states and makes cooperation at best temporary and difficult to sustain due to power distribution concerns and fear of cheating (caused by imperfect information and lack of enforcement mechanisms) even among allies, all the more so since structural realists do not view alliances as anything other than fleeting arrangements forced upon the allies by the presence of a common threat that requiresthe assistance of others (Mearsheimer 1994). The World War II–era “grand alliance” that did not much outlast the destruction of Nazi Germany is a prime example of the fleeting nature of alliances in an anarchic international system. Given these conditions occasional wars are inevitable due to the security dilemma, competition for scarce resources needed to maintain or improve own position in the international system (Roy 1993), or merely miscalculation caused by imperfect information due to adversaries’ incentives to misrepresent own aims and intent (Mearsheimer 1994). Consequently, the argument goes, state leaders do not have the luxury of indulging in idealism, let alone altruism, as each state’s right to ensure self-preservation overrides all other moral obligations (Jackson 2005).

9780230113176_04_ch03.indd 30

2/5/2011 10:53:32 AM

Anarchy, States, and Nations

●

31

To be sure, the assertions concerning the inevitability of conflict have not remained unchallenged. The realist claim that anarchy all but forces states to behave in a competitive or even conflictual manner has been attracting criticism from an increasing number of scholars who have expressed doubts about whether the condition of anarchy inevitably leads to a situation of relative gains maximization. In more extreme cases, it has even been suggested that the concept of anarchy is irrelevant to international relations, rendering the related problem of relative gains-seeking superfluous (Powell 1993). Others have suggested that international conflict has less to do with any inherent properties of anarchy than with realism becoming a self-fulfilling prophecy, as states that are instructed by the theory to assume the worst of others tend to act accordingly and set off a cycle of conflict (Berejikian and Dryzek 2000). The Cuban Missile Crisis—in which both states, motivated by the desire to prevent the other from obtaining an advantage, and which therefore perceived the actions undertaken by the other state as threatening because they were interpreted as likewise representing an attempt to realize a relative gain—is an example of how actions that are seen as defensive and self-protective in intent by their initiators are likely to be interpreted quite differently by others. Another reason suggesting states are not nearly as afraid of their own destruction as the realist theory suggests, and therefore will not act in the highly competitive manner suggested by the theory, is the evidence that states are actually quite unlike individuals. Whereas individuals are mortal and therefore have genuine grounds to fear others, states on the other hand are both durable and capable of rapid recovery from military defeats (Nye 1997), an argument supported by Organski and Kugler’s (1977) discussion of the so-called phoenix factor, undercutting the argument that anarchy automatically infuses states with mutual fear. Probably the best example of a state’s relative position in the international system remaining unchanged in spite of war loss is Germany’s experience in the twentieth century. In spite of having lost two world wars, the second of which resulted, after all, in a dismemberment of the country and an extended foreign military occupation combined with loss of sovereignty, Germany in 2000 was one of the world’s top five economies, its international status having undergone virtually no change when compared to 1900.

9780230113176_04_ch03.indd 31

2/5/2011 10:53:32 AM

32

●

Social Trust, Anarchy, and International Conflict

Moreover, even though states retain their sovereignty, the expansion of influence of international organizations and other non-state actors also leads to the amelioration of the harshest proposed effects of anarchy (Wendt and Friedheim 1995) as an increasing number of functions of government are being assumed by non-sovereign political entities (Buzan, Jones, and Little 1993) which, in spite of the absence of formal legal authority, exercise influence through their sense of international legitimacy (Hurd 1999). Therefore even in the absence of an overarching governing body the international system may assume some features of hierarchy and order (Barkdull 1995; Berejikian and Dryzek 2000; Lake 1996; Wendt 1992), with the existence of security communities such as NATO indicating that a variety of forms of behavior is possible, even under the condition of de facto structural anarchy, ranging from Kantian idealism to Waltzian neorealism (Sorensen 1992). The existence of NATO and the European Union, organizations whose membership includes a number of major powers and former bitter enemies does represent a significant challenge to the notion that lasting cooperation is impossible under anarchy (Jervis 2002). The fact that NATO continues to exist in spite of the disappearance of the Soviet Union, and the continued and growing cooperation among the European Union represents a strong refutation of predictions made for the post–Cold War era by a number of prominent realist theorists (for example, Mearsheimer 1994). Furthermore, contrary to Waltz’s assumptions on the nature of states as functionally similar, the de facto functional differentiation of states similarly contributes to making the effects of anarchy more nuanced than neorealism allows (Buzan, Jones, and Little 1993). The insights from political economy that posit that, under conditions of trade liberalization, states will focus on those economic activities where their economic factor endowments offer them a source of comparative advantage likewise support the notion of functional differentiation of states. The possibility of anarchy being compatible with a wide range of behaviors, above and beyond the narrow range specified by structural realists, is particularly well captured by Wendt’s (1999) identification of Hobbesian, Lockean, and Kantian cultures of anarchy, each representing different images of other states, the nature of their interaction, beliefs in the appropriateness of use of armed force, and indeed the

9780230113176_04_ch03.indd 32

2/5/2011 10:53:33 AM

Anarchy, States, and Nations

●

33

incidence of conflict. Therefore there are strong reasons to believe that the notion advanced by the realist theories that self-help, or the absolute necessity of states to rely for their own security only on their own devices and resources and not on others, while possible, is not the sole and inevitable outcome, the default mode of behavior, of the anarchic international system (Wendt 1992). By the same token, while the possibility that the actions of others may be interpreted as hostile and malicious and therefore trigger the security dilemma cannot be discounted, such an outcome should be understood as but one of several possible social structures possible under anarchic conditions. Other such structures that are as likely to be realized include “other help” (Mercer 1995), which yields behaviors compatible with the interests of other international actors, behaviors that, contrary to self-help, are characterized by the trust in benign nature of intentions and motives of other states. However, Wendt does not convincingly specify under which conditions any of the three cultures of anarchy he describes will prevail. If “anarchy is what we make of it,” and given the sheer unattractiveness of existence in a self-help Hobbesian culture of anarchy or even the competitive nature of interactions under the Lockean culture of anarchy, clearly there must exist strong barriers preventing the evolution toward a Kantian culture of anarchy, where interactions are characterized by cooperation, other-help, and mutual trust. One possible explanation that Wendt’s framework does not adequately consider is that the international culture of anarchy and its effects are actually structurally determined. That structure, however, is to be found in the level of anarchy prevailing on the domestic level of states that make up the international system. From Domestic to International

There already exists considerable support in favor of a linkage between domestic politics and the international “culture of anarchy.” The linkage between domestic anarchy and international anarchy is all the more plausible due to the long-recognized influences domestic factors have on international behavior of states, international anarchy notwithstanding. At the most basic level, our awareness of international anarchy is only possible because of the existence of states where features of anarchy have been all but eliminated (Buzan, Jones, and Little

9780230113176_04_ch03.indd 33

2/5/2011 10:53:33 AM

34

●

Social Trust, Anarchy, and International Conflict

1993), as it is the contrast between the politics of strong and well governed states and the comparatively disorderly interactions characterizing the international system that make us aware of the distinction. In addition, one should note here that even the proponents of the realist theory have long acknowledged the importance of domestic factors. Morgenthau (1948), for example, recognizes that domestic considerations impinge on international behavior of states, particularly democratic ones, as does Mearsheimer (1994), although the latter laments the fact that domestic considerations frequently prevent state leaders from enacting policies consistent with realist theory prescriptions. The notion of state factors exerting strong influence on the international behavior of states is also acknowledged by Waltz (1979) who furthermore argues that authoritarian leaders are under an even greater popular pressure than democratic ones. Nevertheless, some of the more recent critiques of realism have argued that the role of domestic factors is even larger than realists are willing to allow. These critiques posit that the effects of anarchy on the international level may be modified or mitigated by state-level factors. These can be roughly broken down into two categories. The first emphasizes the material factors and interests of dominant substate actors whose interests are reflected in the national foreign policy on the one hand, while the second approach focuses on the development of ideas and identities. The first set of approaches that emphasizes that state actors, far from being unitary actors, are compelled to engage in two-level games (Putnam 1988), where the winning is defined not only in international terms, but domestic ones as well, represents a strong argument in favor of the neoliberal notion that the anarchic international environment is far from the most important set of influences on national decision makers. Instead, in the two-level game the national decision makers are constrained by both international and domestic structures, and face the security dilemma on both levels since the survival of the regime may be undermined both from within and without. Even though a foreign military invasion may well be a threat to the survival of the regime, such threats can equally easily originate from within the state, in the form of challengers for political power. This forces state actors to pursue policies that reproduce the state in such a way as to preserve both the internal and international balance

9780230113176_04_ch03.indd 34

2/5/2011 10:53:33 AM

Anarchy, States, and Nations

●

35

of power (Buzan, Jones, and Little 1993). Often this takes the form of having to choose between allocating resources to alleviate domestic problems by satisfying key domestic constituencies, and international ones by maintaining alliances and deterring enemies (Powell 1993). Therefore states do not have the luxury of focusing on only one level of the game, but rather must simultaneously look inward toward key interest groups and political groupings, and outward at other states, cognizant that the chances of state survival may be improved or worsened by internal as well as external forces. And, indeed the internal forces may present as much, if not more, of a threat to the political survival of the elite than external ones (Spinoza 2000), especially since international actors are not uniform sources of threats. Just as state survival may be threatened by an internal war, especially when the state fails to justify its existence by providing the public goods expected of it by its citizens (Jackson 2005), foreign aid may actually prove to be a source of strength. Alternately, states may be motivated to go to war not out of concern for the relative position in the system (by, for example, engaging in preventive wars or wars in pursuit of natural resources), but through the process of externalization of an ongoing domestic struggle for power, as evidenced by the factors that drove the German foreign policy after 1897 (Levy 1988). As a result, in actuality state actions appear to be far less frequently motivated predominantly by balance of power or relative position considerations than the realist theory allows, which presumably accounts for the frequency of foreign policy “miscalculations” referred to by Mearsheimer (1994). Therefore it stands to reason (and diversionary theory logic is entirely consistent with this argument) that states whose domestic politics are characterized by a high degree of “selfhelp” and which therefore are more violent are more likely to engage in international violence, thus contributing to the perception that the international system is inherently Hobbesian. At its most extreme, the domestic anarchy of a failed state can have a very direct impact on the international system and its sense of security (Holsti 1995), as demonstrated by Al Qaeda’s ability to utilize the anarchic Afghanistan to launch attacks against the United States. The possibility of such a domestic-international linkage is suggested by some of Kant’s writings, as his solution to the problem of international anarchy entails an institutional arrangement akin to a

9780230113176_04_ch03.indd 35

2/5/2011 10:53:33 AM

36

●

Social Trust, Anarchy, and International Conflict

domestic social contract, a great federation of states where the rule of law prevails. After all, Kant’s argument posits that even in a stillanarchic international system, but one composed of well-ordered, non-anarchic states, one would not experience the level of competition and conflict predicted by structural realists (Jackson 2005). And, indeed, it is striking that nearly a century after Kant, even as the well-ordered, effectively governed nation-states of Europe were at last emerging in the nineteenth century, their respective proponents of nationalist ideology were already exploring the possibility of creating an era of peace at least on a regional level (Renan 1995; Smith 1995; Greenfeld 1995; Gellner 1995), even to the point of creating a regional pan-European confederation that would supersede the nation-states that at that time dominated the international system (Renan 1995). Furthermore, the international system’s reflection of values embedded in the domestic politics of its dominant actors is also evident in the process by which domestic norms originate, propagate, and become widely accepted (Finnemore and Sikkink 1998; Hurd 1999; Cortell and Davis 1996, 2000; Goertz and Diehl 1992). The emergence of democracy and human rights as global norms during the twentieth century owes a great deal to the fact that so many of the dominant international actors, the great powers of the international system, are themselves constitutional democracies exhibiting high levels of respect for these norms in their domestic politics. Therefore, given the rather strong influence of domestic factors not only on the international behavior of states but also on the very nature of the international system, one must consider the possibility that one of these highly influential factors is anarchy within state borders. Domestic Anarchy

While anarchy continues to be predominantly a subject for study on the systemic level, a number of authors have pointed out that anarchy is not only a constant and unchanging phenomenon but also not exclusively an international one. It should not pass unremarked that, in spite of occasional admonitions that Hobbes’ work is misused by scholars who want to use his prescriptions for overcoming the international state of anarchy (Heller 1980), Hobbes’ salient works on the “state of nature” were written on the basis of his experience of

9780230113176_04_ch03.indd 36

2/5/2011 10:53:33 AM

Anarchy, States, and Nations

●

37

English civil wars. Hobbes’ argument essentially amounts to a recognition of the crucial role played by the strong state in maintaining the cohesion of the society, preserving internal tranquility, and the fostering of commerce, functions that are closely related to its sovereignty and that, if reduced below a certain level, will cause the state to perish. Since, according to Hobbes, the desire for security is the most reliable and rational desire of our human nature, a state based on satisfying that desire is fully in harmony with human nature and thus fully capable of solving the problem of anarchy. Hobbes therefore argues that the natural human condition, the “state of nature” can be overcome, at least on the state level, through effective political institutions equipped with the tools necessary to implement their functions (Ahrensdorf 2000). In spite of Hobbes’ vital contribution to our understanding of anarchy and the vital role of a strong, sovereign state as a guarantor of domestic peace and stability, domestic anarchy remains a grossly underutilized, and even barely recognized, concept in the study of international politics. To a certain extent this may be due to the influence of structural realists, most notably Kenneth Waltz (1979) whose writings draw a sharp distinction between international and national politics. Thus, according to the structural realists, domestic politics are said to be characterized by order and hierarchy imposed by the impartial state whose monopoly on the legitimate use of force is used for the benefit of all of its citizens, whereas the international system lacks these features due to the absence of a comparable governing institution. The exclusion of domestic anarchy from consideration as an influence on international politics is also the result of the unwillingness by structural realists to consider domestic factors on the grounds that the logic anarchy forces all states to adopt a self-help stance, and also by the deeply rooted tendency to consider factors such as regime type, material indicators of national power, and the level of economic development, unfortunately to the point of nearexclusion of considerations of state strength. Fortunately, a more nuanced understanding of anarchy has emerged in recent decades. Milner (1991), for example, convincingly argues that reality is considerably more complex than Waltz’s dichotomy, with international relations often exhibiting elements of hierarchy and order and domestic politics not infrequently being characterized by

9780230113176_04_ch03.indd 37

2/5/2011 10:53:33 AM

38

●

Social Trust, Anarchy, and International Conflict

self-help. The legitimate right to the use of force claimed by governments that Waltz cites as evidence of the difference of domestic politics is not always in evidence (Milner 1991). Contrary to Waltz’s assertion that force is unimportant as a means of control in domestic politics and instead is used only to serve justice, authoritarian regimes indeed do rely on the use of force to coerce obedience and maintain themselves in power, just as such the power of government is often, in effect, privatized in weak states and used for the personal benefit of individuals occupying state functions (Milner 1991). For Waltz to assert otherwise suggests the existence of an unspoken assumption that all states in the international system are at an equally high level of governance and fairness, an assumption that unfortunately lacks strong empirical support. Remarkably, Milner’s arguments are not all that dissimilar from Morgenthau’s (1948) recognition that the essence of international relations is the same as that of domestic politics and is dominated by the struggle for power, modified only by different circumstances, and E. H. Carr’s (1964) acknowledgment that international and domestic politics are organized along identical principles. Morgenthau and Carr’s acknowledgments, combined with Milner’s arguments on the nature of domestic and international politics, make a strong case for the inclusion of domestic anarchy into the study of international politics. This is even more so, given that the possibility of anarchy as a substate, domestic-level phenomenon and the attendant security dilemma that it creates among substate actors has long been recognized by segments of ethnic conflict literature that argue that the absence of a strong state capable of providing order and security creates a situation where ethnic minorities gradually (and possibly due to political entrepreneurs exploiting the situation through the manipulation of ethnic identity symbols) come to view each other as potential enemies and act accordingly. Ethnic conflict is thus attributable to a combination of state inability to protect its citizens creating a selfhelp environment in which mobilization along ethnic lines may take place, which in turn leads to a security dilemma and threat of violent conflict (Posen 1993; Kaufman 1996; Kaufmann 1996; Roe 1992; Saideman and Ayres 2000; Taras and Ganguly 2002). In extreme cases of state failure, a condition of “open anarchy” may prevail, making the failed state an open environment fully connected to the

9780230113176_04_ch03.indd 38

2/5/2011 10:53:33 AM

Anarchy, States, and Nations

●

39

international system (Vinci 2008). The concept of anarchy existing on the domestic level has even been utilized in descriptions of political interactions in long-established countries, including the United States (Eidelberg 1970). Consider, for example, the wealth of expressions in the contemporary American English vernacular (“you have to look out for Number One,” “it’s a jungle out there,” “it’s a dog-eat-dog world,” “if you want to have something done right, you have to do it yourself,” “there but for the grace of God go I,” et cetera) that are the reflection of the extent to which for ordinary Americans life in the modern American society is characterized by “self-help.” That such vernacular expressions should exist is not surprising, considering that, even though the United States of the late-twentieth century was a far stronger state than it was in the late nineteenth, one must acknowledge that it still lagged behind most other states at comparable level of development in the provision of public goods and services for its citizens. The State and National Identity

Given that domestic politics may represent various cultures of anarchy, this raises the question what characterizes strong, well governed states where the features of Hobbesian and even Lockean anarchy (corresponding to conflict and competition, respectively) are kept to a minimum. One such readily observable characteristic is the presence of national identity. The nineteenth-century-nationalist ideal was the nation-state, or a state whose boundaries neatly corresponded to the boundaries of group identification (usually defined in the primordial terms of ethnicity, language, and religion) of its inhabitants. The concept was invented as a means of addressing the problem of the relationship of state to society on the basis of assumption that what appeared to be primordial bonds of blood were sufficient to establish the national “in-group,” ignoring the strong evidence that such bonds are actually socially invented constructs and not reflections of genuine ethnic solidarity of the state’s inhabitants (Breuilly 1995). Because there is considerable evidence that the primordial sources of identity, such as race or language, are not unshakeable (Renan 1995), and moreover their salience has been known to widely vary from state to state, person to person, or

9780230113176_04_ch03.indd 39

2/5/2011 10:53:34 AM

40

●

Social Trust, Anarchy, and International Conflict

one era to another (Geertz 1995). The relatively low importance of such primordial traits is underscored by the fact that a strong sense of national identity is possible even in situations where the population of a state shares no common ethnicity, language, or religion (Weber 1995). Furthermore, there is growing consensus that national identity, far from being a naturally occurring phenomenon, is something that must be painstakingly constructed by the state (Tombs 1996), specifically through its efforts to overcome anarchy within its borders by the maintenance of public order, law enforcement, the provision of a wide variety of other public goods, and carrying out other policies to facilitate the peaceful coexistence of individuals within the state (Kedourie 1995). Therefore it is unsurprising that citizens with a strong sense of belonging to a nation can only be found in states with well developed educational systems (Gellner 1995), and strong national identification occurs only where social conditions permit homogeneous and centrally sustained high cultures that embrace not only elites but the masses as well (Gellner 1995), with print languages creating the foundation for common national consciousness (Anderson 1995). The state’s role as framework for collective action is a crucial motivating factor behind the emergence of collective national identity (Hobsbawm 1995), especially in an anarchic environment characterized by competition and conflict. And, indeed, when one examines the historical record, it becomes evident that most major powers in the nineteenth century consistently pursued, with varying degrees of success, policies of assimilating its subjects and eliminating provincial attachments and loyalties that were to be replaced or at least transcended by loyalty to the state itself (Richmond 1995). Despite occasional references in the literature to “old” and “new” nations, purporting that nations such as France have led a discrete existence for several centuries and predate the emergence of the nationalist ideology (Seton-Watson 1995), evidence suggests that genuine national identity capable of supplanting regional loyalties in even the so-called old nations is a much later creation, dating to only the nineteenth century (Connor 1995b). To cite but one example, the French national identity sometimes held (including by SetonWatson 1995) to be an ancient one, is a comparatively recent creation, a product of concerted state action in the nineteenth century (Weber 1976) in the form of expanding public education that created

9780230113176_04_ch03.indd 40

2/5/2011 10:53:34 AM

Anarchy, States, and Nations

●

41

a common language for all inhabitants of the French state to communicate, that led to the construction of national infrastructure that connected the far-flung provinces of the state and allowed travel, and that also resulted in universal military service that likewise brought together the inhabitants of many regions of the French state and redefined their individual notions of identity. Prior to these state-led efforts, which also included standardizing administration, laws, and education (Weber 1976; Hobsbawm 1995), the individual identification with the French state was rather abstract, with individual identity being predominantly tied to one’s village or region (Weber 1976; Connor 1995b). That effective state action should create national identity among its inhabitants is unsurprising, given the importance of the state’s role as a provider of public goods to the daily lives of its citizens. An effective state fulfills crucial economic, sociocultural, and political needs (J. S. Mill in Rawls 1999). Consequently, individuals derive a sense of security, a feeling of belonging, and prestige from their affiliation with the state (Wendt 1999). Of crucial importance here is that national identity, as a result of state efforts to eliminate domestic anarchy, is a psychological construct. Famously described as an “imagined community” (Gellner 1995; Anderson 1991, 1995), it requires a psychological bond to exist among its members, determining whether a given set of individuals’ self identifies itself as members of a single nation, with a common ideology and customs, as well as a sense of homogeneity (Renan 1995; Kohn 1995; Greenfeld 1995). It is also accompanied by a sense of solidarity, and the predisposition to make sacrifices on behalf of other members of the nation (Renan 1995; Kohn 1995). Although a belief in the group’s separate origin and evolution apart from other nations, on the grounds of sharing a specific language and ethnicity not found elsewhere, may be an important ingredient of national psychology, this belief need not be (and rarely is) factually based, and instead is a social construct established by the elites (Connor 1995a). However, the corollary that national identity is the product of state action is that the formation of such identity may be incomplete, or entirely nonexistent, in states incapable of undertaking what de facto amounts to a program of nation-building within its own borders. It should be noted that not nearly all states that have existed historically or exist currently have succeeded in establishing a strong

9780230113176_04_ch03.indd 41

2/5/2011 10:53:34 AM

42

●

Social Trust, Anarchy, and International Conflict

sense of national identity on the basis of allegiance to the state. The use of the term “nationalism” to signify both identification with the state as well as identification with one’s nationality has led to an unwarranted assumption that nationalism, as a force in the service of the state, should operate everywhere with the same strength, even in states that are comparatively recent creations and whose governments remain hardly capable of asserting their own sovereignty. Indeed, the possible weakness of national identity means that substate actors may have several identities simultaneously (Suny 1999). Unsurprisingly, in many situations the state is not the primary source of one’s identity (Connor 1995a). Individuals frequently opt to identify themselves with groups other than their nations (Terhune1965; Druckman 1994), including one’s ethnic identity that frequently represents a stronger component of individual identity than citizenship (Horowitz 1985; van Evera 1994; Morgan and Bickers 1992; Vanhanen 1991, 1999). Moreover, very few states are actually “nation-states,” in the sense that borders neatly coincide with the territorial distribution of the self-identified national group defined in either primordial or civic terms. Nearly 90 percent of states contain important minorities, if not majorities, of peoples considering themselves members of nations other than the one constituting the state (Connor 1995a). Not only are the “nations” of Europe a much later creation, in some cases the creation of nationhood has not yet been achieved (Connor 1995b). Indeed, literature on nationalism draws distinction between tradition and modern states (Gellner 1995; Geertz 1995; Anderson 1995; Smith 1995; Giddens 1995), with the former characterized as having weak tradition of civil politics, absence of government bureaucracies (Geertz 1995; Smith 1995), government’s inability to maintain the monopoly of force within its borders (even where such monopoly is claimed) (Giddens 1995), in short, the ability to alleviate the state of anarchy within its borders. Whereas in the modern states national unity is maintained by appeals to allegiance to the civil state, in traditional states primordial ties predominate, with social fissures running along the lines of kin and language (Geertz 1995). This distinction appears similar to dividing lines drawn even by early proponents of the nationalist ideology, including Herder who distinguished so-called natural states that consisted of a single people, and states that were the product of war and conquest, such as Prussia under Frederick

9780230113176_04_ch03.indd 42

2/5/2011 10:53:34 AM

Anarchy, States, and Nations

●

43

the Great (Breuilly 1995). That such distinctions exist is not merely of academic interest because, as it turns out, whether or not the state in question enjoys a high level of popular identification with the state and a resilient national identity can have a profound impact on the international behavior of the state. Patriotism or Nationalism?

That states’ differing levels of strength translate into differing levels of national identity salience is important due to a perhaps understated and understudied relationship between national identity and international conflict. In spite of the expectation that strong national identity should be accompanied with high level of international conflict on the grounds that strong group identity requires the existence of enemies (Huntington 2006), paradoxically, it is the lack of stable identities that is most likely to result in a violent international conflict, as the existence of fragile, multiple, and shifting identities has been identified as a cause of conflict (Suny 1999). Hobsbawm (1995), for example, draws a contrast between France and Germany in the late nineteenth century, wherein the former’s identity was defined in self-referential terms while the latter’s in juxtaposition to “the other.” Whereas the French national identity was established around the institutions of a comparatively ancient French state, the comparatively recently established unified Germany in which, moreover, many of its formerly independent German principalities chafed under the Prussian hegemony was hard-pressed to find much that united all the Germans, apart from their shared dislike of the French, dislike that was a by-product of the onerous occupation of large swaths of Germany by Napoleonic armies. As will be demonstrated in Chapter 9 of this book, the reliance on conflict with the outside world borne out of the insufficiently strongly established German national identity was a key reason for the outbreak of war in 1914 (Hobstawm 1995). To cite more recent examples of weak national identities increasing the likelihood of conflict, an exclusivist national identity on the part of a dominant ethnic group contributed to a civil war and the destruction of the country of Georgia in the early 1990s. On the other hand, a state like Armenia, with a more coherent national identity, was able to conduct an effective and peaceful

9780230113176_04_ch03.indd 43

2/5/2011 10:53:34 AM

44

●

Social Trust, Anarchy, and International Conflict

transition. Overall, states with weak national identity, regardless of whether they are ethnically homogeneous (for example, Tajikistan) or heterogeneous (Azerbaijan) tend to experience regional divisions and civil wars (Suny 1999). Moreover, states with relatively weak central governments such as states whose institutions provide for regional autonomy suffer from diminished attachments by the ethnic minorities toward the state they reside in (Elkins and Sides 2007). Such a diminished level of attachment may thus contribute to raising the level of domestic political violence. The distinction between strong and weak national identities is already well established in literature and has even earned a distinctive set of terms. Although the term “nationalism” tends to be used to identify all cases of identification with one’s nation or state, the literature sometimes distinguishes two forms of identification with one’s country, namely, patriotism that represents strong attachment and loyalty toward one’s own state without the corresponding hostility toward others, and nationalism where such hostility is de rigueur (Orwell 2000 [1945]; Druckman 1994). While individuals who have strong patriotic feelings may be willing to risk their lives for their country, they are actually not as war-prone as the nationalists, whose attachment toward their nation is established on the basis of a built-in enemy (Druckman 1997) as was, for example, the case with the German Empire of the late-nineteenth and early-twentieth centuries (Hobstawm 1995). What is even more striking, the so-called spirit of internationalism, or the loyalty to the world at large competing with, if not exceeding, the national loyalty, appears to be the strongest not only among individuals who have extremely low sense of identification with their state (preferring, for example, to identify themselves with their families or ethnic groups), but also among individuals with a very high sense of national identification (Terhune 1965). This seemingly paradoxical finding actually confirms the notion that genuinely strong national identity is not built on the basis of hatred toward or exclusion of others, but rather is inherently inclusive in its character. It is striking that so many writings on nationalism suppose that the nation-state is but a waypoint toward larger supranational political entities (Richmond 1995) with no apparent upper limit on the size of such political entities (Gellner 1995) up to the point of a global

9780230113176_04_ch03.indd 44

2/5/2011 10:53:34 AM

Anarchy, States, and Nations

●

45

community of peoples (Hutchinson 1995; Smith 1995; Greenfeld 1995) all sharing a common, shared identity as members of the human race. Already Ernest Renan presupposed that the nations of Europe would eventually be supplanted by a European confederation (Renan 1995), a prediction that likely appeared outlandish when it was originally written in the late-nineteenth century, when the nation-states of Europe were grouped into mutually hostile alliances and that were yet to experience the two world wars, but looks quite prescient a century later when the confederation Renan spoke of appears to be well on its way to fruition. The success of the European Union whose member states are fortunate to have highly capable and powerful governments that have successfully combated anarchy within their own borders, and which at the same time boast of strong nationalisms, has demonstrated that strong national identities are not merely not incompatible with transnational cooperation (Mercer 1995), but perhaps even a prerequisite for it. That strong national identity, itself a creation of a strong, activist state, should have this surprising effect is due to another by-product of the elimination of features of self-help from domestic politics. That by-product is the generalized social trust, yet another concept whose influence has thus far been both underestimated and understudied in the field of international relations.

9780230113176_04_ch03.indd 45

2/5/2011 10:53:34 AM

9780230113176_04_ch03.indd 46

2/5/2011 10:53:34 AM

CHAPTER 4

Social Trust and Its Origins

W

hile the linkage between state strength/weakness and national identity has been recognized for a relatively long time, the awareness that efficient governance is also the source of generalized social trust is considerably more recent. However, the properties of social trust suggest that it is a phenomenon whose influence on international conflict cannot be ignored. Indeed, it may be that social trust is the ideational underpinning of international peace that the democratic peace theorists have been attempting to identify, and the determinant of diversionary behavior that will account for the outbreaks of diversionary violence far more efficiently than the hitherto used approaches relying on governmental popularity. Definitions and Characteristics

The concept of social trust already enjoys a considerable body of literature in its own right. The definition of social trust employed here is Uslaner’s (2002) concept of moralistic, or generalized, social trust, which focuses on the individual propensity (or lack thereof ) to view people who are complete strangers (and who may belong to different social classes, ethnic groups, or even nationalities) as trustworthy, or likely to validate one’s trust being placed in them, even in situations where the individuals in question may have no incentive to validate that trust. Generalized social trust is therefore a faith in others and a willingness to act in accordance with that belief, and an assumption that others will behave in a way consistent with one’s belief in their trustworthiness, even though no immediate benefit is expected (Uslaner 2002). As such, the concept of generalized trust stands in contrast with other conceptions of trust, for example, Hardin’s (2001) encapsulated trust, which posits that individuals view one

9780230113176_05_ch04.indd 47

2/5/2011 10:54:08 AM

48

●

Social Trust, Anarchy, and International Conflict

another as trustworthy only when they know cooperation is in the other individual’s interest, or particularized trust, which obtains in situations where the individuals are both members of a cohesive group, know each other well, or otherwise have the expectation of extended future interaction, and are thus operating under a long shadow of the future (Uslaner 2000). Social trust is also frequently identified as one of the elements of social capital as a facilitator of collective action (Putnam 2000). The heuristic nature of generalized social trust is a key feature distinguishing it from particularized trust (Uslaner 2002), which is more context-based, built on enforceable contracts, expectations of mutual behavior grounded in material self-interest, and a variety of other social constraints. In effect, the concept of particularized trust of encapsulated trust closely resemble one another, as both predict that individuals trust one another only under conditions where one actor believes, on the basis of tangible available evidence, that it is in another actor’s self-interest to behave in a cooperative fashion. Such trust molded by experience of interaction with that particular actor (Hardin 1996) rather than by the natural predisposition of the individual in question to place his or her in others. Thus, whereas strategic, or encapsulated/particularized, trust is built on the foundation of available information and evidence by which to judge other individuals, moralistic trust is based on normative statements and beliefs concerning how people should, and will, behave (Uslaner 2002), even in the absence of concrete evidence that would lead the “truster” to believe that such behavior is to be expected. Generalized social trust literature is unanimous in confirming the importance of a high level of trust for the health of the society and both individual and group interactions therein, in interpersonal, commercial, and governmental settings. To cite Uslaner (2002), trust is the “chicken soup of social life.” Generalized social trust is a belief that others will not take advantage of one’s willingness to trust and cooperate, based in the goodwill of others even where there is a prospect of gain by doing so (Uslaner 2002; Thomas 1996) that exists in the knowledge that some of the benefit may go to bystanders, or free riders (Putnam 2000) who benefit from the behavior of others (in other words, consume trust) without contributing anything in return. Generalized social trust is a societal resource that links citizens to each other and enables them to pursue their common objectives more effectively

9780230113176_05_ch04.indd 48

2/5/2011 10:54:08 AM

Social Trust and Its Origins

●

49

(Hooghe and Stolle 2003), in effect eliminating or at least greatly reducing the endemic collective action problem, and a perception of underlying common values that makes cooperation easier (Uslaner 2002) and leads to a generally more efficient society (Putnam 2000) where individuals are more active and engaged (Uslaner 2000), as well as free to live without the fear of others. Generalized social trust is an embedded notion of mutual faith in others on which so many daily interactions depend (Baier 1986; Lewis and Weigert 1985), and a functional prerequisite for the possibility of a civil society. Generalized social trust also functions as a deep assumption underwriting social order (Lewis and Weigert 1985). Trusting someone is to have an attitude of optimism about his or her goodwill and to have the confident expectation that when the need arises, the trusted individual will act consistently with the expectation being placed on him or her. Social trust therefore is even more basic for the construction of groups than even a sense of moral obligation. Even though it involves risk and doubt, generalized social trust is indispensable in social relationships, where it allows social interactions to proceed with confidence (Lewis and Weigert 1985) where, in the absence of such trust, the interactions would not proceed to the detriment of all concerned. Such interactions are possible because generalized trust implies the existence of a generalized reciprocity norm, or the expectation that one’s deeds will not necessarily be reciprocated at the time they are performed, but rather at some unspecified future (Putnam 2000), and the expectation of future reciprocity therefore creates an interest in the welfare of one’s interaction partners. The concept of the generalized reciprocity norm therefore stands in contrast to ordinary contractual obligations, or interactions governed by the norm of particularized reciprocity where individuals participating in the interaction require a reciprocal benefit to be rendered because, presumably, they do not trust their partner to deliver on their obligations in the future. As a result, since individuals with high levels of generalized social trust exhibit the willingness to interact with others even under conditions where no immediate reciprocal benefit can be obtained, they have an expansive view of the size of the so-called moral community, defined as people with whom they have perceived common interests, perhaps not common values (Uslaner 2000; Crepaz 2007) but certainly a sense of a common fate (Uslaner 2002), leading to a tolerant, inclusive view

9780230113176_05_ch04.indd 49

2/5/2011 10:54:09 AM

50

●

Social Trust, Anarchy, and International Conflict

of society (Uslaner 2002). In effect, generalized social trust represents a rather enlightened and expansive concept of self-interest, in the sense that it implies that individuals with a high level of social trust believe their own well-being and self-interest are closely tied to the well-being of others. To put it in more basic terms, high trusters do not view the world in terms of a zero-sum game. Since such an outlook is radically at odds with that advocated by the realist theory, it raises the question of how an international system would appear if it were populated chiefly by states with populations enjoying high percentages of individuals with high levels of generalized social trust. The idea of such an ideational basis for a civil society enjoys a long and distinguished lineage. The social and political upheavals caused by the industrial revolution led a number of political philosophers of the era to address these problems, with Emile Durkheim identifying the question of an underpinning set of shared ideas and societal cohesion. Although he did not discuss social trust per se, to the extent Durkheim was concerned with the role of ideas as ties that bind a society together (Giddens 1971), he may be seen as a forerunner of the social trust research program. Specifically, Durkheim’s argument against constructing the society on solely contractual basis, per utilitarian recommendations (contractual interactions corresponding to particularized, rather than generalized) on the grounds that a society built on such a foundation would surely disintegrate (Giddens 1971), can be interpreted as an argument in favor of fostering generalized social trust. For, whereas interactions based on the norm of generalized reciprocity are essentially cooperative, with each individual acting with the assumption that the benefits accruing to others as a result of one’s own actions will sooner or later redound to one’s own benefit. In contrast, individuals interacting on the basis of particularized reciprocity norm, with each expecting a concrete payment, in cash or in kind, for goods or services rendered, is fraught with conflict with mistrust because the demand for instant and equivalent payment is redolent of the realist theory’s concern for maintaining relative position and preventing others from realizing a relative gain, concern entirely absent from interactions based on the generalized reciprocity norm. And, indeed, just as the presence of generalized trust is associated with a range of positive phenomena, its absence brings about a host of problems. Distrust, after all, is synonymous with wary suspicion,

9780230113176_05_ch04.indd 50

2/5/2011 10:54:09 AM

Social Trust and Its Origins

●

51

pessimism about the goodwill and competence of other actors, concerns about their malicious intent, as well as an expectation that one’s interests are in danger of being harmed by others in the absence of vigilance (Jones 1996). As a direct outcome of such sentiments, daily life in societies where the level of generalized trust is low is fraught with difficulty due to fears of exploitation by others and beliefs that cooperation with others is dangerous (Uslaner 2002), fears that reduce the frequency of interpersonal interaction in all contexts even under conditions where such interaction would have been mutually beneficial. Banfield’s (1967) research on the Italian village of Montegrano, where the basic bonds of trust have broken down to such a degree that all who are outside of the immediate family are seen and treated as potential enemies is a particularly stark illustration of problems afflicting societies with low generalized social trust. However, it would be a mistake to treat social trust as merely an aspect of interpersonal interactions. It is a concept that has considerable influence on a variety of political phenomena, ranging from economic development to violent crime, trust in government, governmental performance, and, as will be demonstrated later, international conflict behavior. High levels of generalized trust also make possible trusting members of other ethnic groups (Bahry et al. 2005), which in turn presumably (for that question still remains to be explored) serves to reduce the risk of interethnic conflict. Furthermore, individuals with low and high social trust also tend to have different attitudes toward their own country’s government. Interpersonal generalized trust has been shown to be closely linked to both trust in government and trust in particular government officials (Lewis and Weigert 1985; Aberbach and Walker 1970; Hall 1999; Iyengar 1980), and even though no government has likely ever had the trust of all its subjects, even the most dictatorial regimes have sought to increase their credibility and sense of trustworthiness by, for example, coating their policies in a veneer of legality. The need for trust to exist between the rulers and the ruled as a precondition for a civil, nonviolent society has been recognized as early as Plato’s Republic (Baier 1986), as resentment and distrust are the first step toward violent resistance (Aberbach and Walker 1970) and form the preconditions for the eventual downfall of government. Moreover, political factions consisting of individuals who are mistrustful of the government and other factions of the

9780230113176_05_ch04.indd 51

2/5/2011 10:54:09 AM

52

●

Social Trust, Anarchy, and International Conflict

polity tend to demand additional resources so as to increase their own relative power that they view, due to their inherent mistrust of others, as the only guarantee of their own security (once again, illustrating the similarities between the international and domestic levels of politics) and may cause the government to react with repression (Hetherington and Globetti 2002). Level of trust in government also strongly influences what policies and strategies are available to political leaders, since polities consisting of high-trust individuals will be more likely to support projects entailing collective sacrifice, or the improvement in the lot of specific segments of society, and even the very ability to govern, as governments lacking the trust of the governed will have to rely more on coercion (which in itself is a costly diversion of resources) to enact their policies. Even though distrust of government may not always or necessarily have disastrous consequences for the government, it will have consequences in the shape of reduced support for government policies, and thus government efficacy, possibly creating a downward spiral of mutually reinforcing distrust and failed policies ultimately resulting in violence (Hetherington and Globetti 2002) and perhaps even state failure. Moreover, the effects of trust on the attitude toward government appear to be more influential than those of partisanship or ideology (Hetherington and Globetti 2002). The wide variation among democracies in terms of their trust levels suggests that while democratic regimes may enjoy a greater potential for social trust generation for reasons that will be discussed further in this volume, they do not always realize it (Uslaner 2002, 2003). Of particular concern are the low levels of trust in the newly emergent democracies of formerly Communist states, suggesting that trust and democracy are not always closely paired, as evidenced by cases where the process of democratization has been accompanied by decreases in social trust (Letki 2004; Uslaner 2003). Instead, a democratic regime’s ability to create trust among its citizens appears to be confined only to long-existing democracies (Uslaner 2003). Instead, it would appear that while a relationship between social trust and democracy may exist, the causal arrow is likely to run from social trust to democracy, as a high level of social trust appears to be a precondition for the existence of a successful democratic regime. The contemporary literature on social trust supports the notion of using it as a concept that is not only independent of democracy; it

9780230113176_05_ch04.indd 52

2/5/2011 10:54:09 AM

Social Trust and Its Origins

●

53

may actually be a necessary condition for it. To the extent that democracy requires the trust in the goodwill of other individuals whose votes will, after all, have an impact on one’s own well-being, it is difficult to conceive of a well functioning democratic system in the absence of an electorate whose individual members feel a sense of generalized trust toward others. Nor does the level of generalized social trust appear to be the by-product of a specific regime type. Although some links have been discovered between generalized trust and the extent of civil liberties in a country, as well as between authoritarianism and absence thereof, there are nevertheless significant differences among both democratic and authoritarian states in their ability to generate social trust (Hooghe and Stolle 2003). Indeed, the importance of interpersonal trust as a societal factor conducive to democratization has been recognized by Almond and Verba (1963). Polities with high number of trusting individuals tend to have governments that are more efficient and responsive, though with a possibility of a mutually reinforcing feedback loop where high trust leads to more efficient institutions that implement policies that foster social trust (Uslaner 2003). States with high levels of generalized social trust have been found to have experienced longer durations of their democratic regimes, indicating that trust is a key component of prodemocracy attitudes (Inglehart 1997). Democratic norms, a variable that is often cited as a cause for more pacific international conflict resolution behavior by democratic states, tend to be more strongly entrenched in states with social trust (Booth and Richard 1998). The absence of trust, on the other hand, leads to proauthoritarian attitudes (Uslaner 2002; Aberbach and Walker 1970). A similar relationship exists between generalized social trust and wealth. Although some have argued that social trust is the property of wealthy nations (Inglehart 1997), the preponderance of evidence suggests the causal effects run in the opposite direction, namely, that wealthy nations became that way due to high levels of social trust. The level of wealth and economic development a country experiences is the result of social trust that permits economic interaction and cooperation (Putnam 2000; Uslaner 2003; Fukuyama 1995). As noted earlier, since the absence of generalized social trust will limit interactions with strangers, and instead limit interpersonal, including commercial, interactions to the circle of individuals one intimately knows or

9780230113176_05_ch04.indd 53

2/5/2011 10:54:09 AM

54

●

Social Trust, Anarchy, and International Conflict

trusts on the basis of familial ties or other sources of the shadow of the future, it is not surprising that areas where mistrust of others should see relatively little economic activity. In spite of the undoubted role free markets play in the creation of economic well-being, markets cannot operate efficiently (if at all) under the conditions of mutual distrust, where each potential contracting party believes the other is acting in bad faith (Guseva and Rona-Tas 2001). The importance of social trust as a factor behind economic development is borne out by the example of southern Italy, whose economic lag behind the more prosperous northern regions only grew after the reunification of Italy in the nineteenth century due to the south’s persistent absence of generalized social trust that made extensive and impersonal forms of economic cooperation that are necessary for sustained economic growth impossible (A’Hearn 1998). The results of absence of trust are also evident in a recent study of the Russian credit markets, which shows how Russian banks are suffering from having to operate in a market limited by the low number of business partners perceived as trustworthy (Guseva and Rona-Tas 2001). Although material factors do play a role in the formation of social trust (a point that will be expanded upon below), they do so only indirectly. What appears to matter is not the absolute level of wealth, but rather one’s relative position in the society. The presence of income and wealth inequalities, specifically, tends to undermine social trust, as individuals are more likely to trust those whom they view as their equals (Hooghe and Stolle 2003; Uslaner 2003). The Gini index of inequality, in particular, has been found to have a very strong influence on the level of social trust (Uslaner 2003). Cross-national variations in trust depend heavily upon a society’s distribution of income, with declines in social trust being closely linked to growth in economic inequality (Uslaner 2002; Chambers and Kopstein 2001). Therefore, given the demonstrated importance of generalized social trust to the political and economic health of a society, the question of how such trust may be generated is of considerable importance. So far the research in this area has tended to cluster into two schools of thoughts, one claiming that social trust is the by-product of interpersonal interactions, and the other attributing the presence of high levels of generalized social trust to the presence of effective institutions of government.

9780230113176_05_ch04.indd 54

2/5/2011 10:54:09 AM

Social Trust and Its Origins

●

55

The Associational Argument

The notion of social trust being the result of informal associations and involvement in the civil society has been embraced by one of the main motivators of the most recent renewal of interest in the questions of civil society, social capital, and social trust, Robert Putnam (Putnam 1993, 1995, 2000). Putnam has argued that the level of trust is caused by the level of involvement in civic organizations, such as parent-teacher associations, unions, even bowling leagues, showing how such involvement has begun to decline in the 1960s due to a wide variety of factors, including television viewership, urban sprawl, competing pressures for both time and money, and similar related reasons (Putnam 2000). Such community commitments are important, according to Putnam (1993) because of personal interaction generating perceptions of trustworthiness of other actors and creating a “stock” of social capital. A similar argument has been advanced by Letki (2004) who has argued that low levels of political engagement are due to low social capital, anti-democratic norms, and attitudes learned while living under undemocratic regimes. However, the associational aspect of Putnam’s argument has come under considerable criticism, and it is arguably the least tenable portion of his treatment of the topic. Indeed, the question of associational life and generalized social trust appears to be a classical “chicken or egg” problem, since social trust and interpersonal interactions are correlated closely enough to make the cause-and-effect determination rather difficult. However, to help disentangle the conflicting causality arrows, consider the following scenario. Two individuals are pondering whether to spend an afternoon together in a bar imbibing significant quantities of alcoholic beverages. The question that one has to ask oneself is under what conditions is the following scenario likely to come to fruition: are these two individuals who already trust one another and treat their bar-hopping exercise as an affirmation of that trust-based relationship, or is this a case of two individuals who mistrust one another and hope to build mutual trust through this form of association. Alas, the latter scenario, which corresponds to Putnam’s theory of social trust creation, appears to be the less likely one. Considering that ingesting alcohol is likely to put one in a vulnerable position and at risk of at least losing one’s wallet, it would appear

9780230113176_05_ch04.indd 55

2/5/2011 10:54:09 AM

56

●

Social Trust, Anarchy, and International Conflict

that the drinking exercise is unlikely to occur in the absence of a fairly high level of mutual trust between the participants, and is equally unlikely to be contemplated as an exercise in trust-building between two mutually mistrustful individuals. This basic flaw with Putnam’s argument has led to an avalanche of literature critiquing the assumption that trust is the by-product of interpersonal interactions. The specific problem lies in the apparent existence of different forms of association and their dissimilar influence on creating generalized social trust. Some forms of civil society appear to do little to create or strengthen liberal norms of tolerance and acceptance of others (Chambers and Kopstein 2001; Levi 1996) and group membership does not make individuals more trusting (Claibourn and Martin 2000; Knack and Keefer 1997). Whereas communal-level activism appears to be associated positively with generalized interpersonal trust, civil society activism appears to have little or no effect in this regard. Moreover, civil society activity actually appears to be responsible for depressing levels of democratic norms and political involvement, in contrast to active involvement in formal organizations. In other words, the available evidence strongly suggests a possibility that civil society involvement is an indicator of particularized trust among closely knit groups of individuals, rather than a broad attitude of trust toward strangers that generalized trust represents (Booth and Richard 1998). Instead of creating generalized social trust, civic engagement usually takes the form of association with one’s peers, which is a form of activity that does nothing to help one expand the scope of one’s “moral community.” Consistently socializing with the same narrow circle of individuals may even contribute to the destruction of generalized trust (Uslaner 2002). The more one is dependent on one’s close associates and kin, the more one tends to draw the line dividing the world into “us” and “them” precisely along these lines, and, consistent with the argument that participation in the democratic process requires a certain level of generalized social trust, their mistrust of others causes them to withdraw from political activity. Since the main factor distinguishing generalized trust from particularized trust is the size of one’s moral community, any sort of activity that promotes particularized trust is unlikely to have the result of promoting generalized trust. As one might expect, interaction with members of a single and narrowly defined group makes one more likely to have more faith

9780230113176_05_ch04.indd 56

2/5/2011 10:54:10 AM

Social Trust and Its Origins

●

57

in that particular group but not in people generally and is only likely to create particularized trusters who are wary of dealing with people who are different from themselves, and who have a narrow view of their moral community. Indeed, it would appear that the effect of civic organizations on attitudes toward politics very much depends on the type of organization. Whereas membership in inclusive organizations reduces one’s acceptance of political violence, individuals who affiliate solely with members of the same ethnic group tend to have more positive attitudes toward violence (Chapman 2008). The problem of “bad” civil society therefore represents a serious problem for civil society argument (Chambers and Kopstein 2001), especially since, as the study of the Nazi Party’s exploitation of Germany’s networks of civic associations showed, widespread civic associations are no obstacle to a xenophobic totalitarian regime taking power (Berman 1997). That the Nazi Party, an openly and militantly xenophobic political movement that advocated a political worldview built on the notion of mistrust toward “outsiders” (here defined in both national and ethnic terms), were to make significant inroads into the German electorate by exploiting the network of civil society groups whose membership consisted of particularized trusters already predisposed to mistrust others is entirely consistent with the literature dealing with the relationship between civil society and social trust. Indeed, it would appear that the Nazi Party’s xenophobia, far from being an obstacle to attracting members from the various German civil society groups, might have been an actual plus. Putnam himself acknowledges the possibility of existence of a “dark side” of social capital, wherein some kinds of associations may actually discourage the creation of generalized social trust. Whereas the purpose of civic engagement is to enable individuals to form “bridging” associations with complete strangers coming from different societal backgrounds, in actuality some forms of civic engagement result in “bonding” with individuals already very much like oneself (Putnam 2000). The linkage between social group membership and generalized trust has also been assailed directly, with findings that while trust has declined in the United States since the 1960s, associations of various types have not only continued to exist but, in certain cases, even increased their membership rolls (Jackman and Miller 1996; Uslaner

9780230113176_05_ch04.indd 57

2/5/2011 10:54:10 AM

58

●

Social Trust, Anarchy, and International Conflict

2002). Nor is this absence of relationship between civic engagement and trust confined to the United States, as a similar absence of relationship is also observable in other countries, further assailing the alleged connection between civic participation and social trust (Uslaner 2002). For that reason civic engagement alone is insufficient to account for the existence of generalized trust (Uslaner 2000). Social Trust and the State

Putnam’s “bottom up” argument that civic engagement builds social trust also came under attack by proponents of the role of the state as an agent of social trust creation (Stolle 2003; Rothstein and Stolle 2003, 2008). In accordance with this line of argument, trustful attitudes toward other individuals appear to be determined not by the degree of participation in civil society organizations, but rather by the general record of government performance in certain policy areas. Trust therefore is not merely a reflection of basic personality traits or a simple function of general social background factors, but instead springs from the workings of the social or political system, including expectations of treatment by government officials, general feelings of deprivation and well-being, as well as beliefs about the status and acceptability of one’s group in society (Aberbach and Walker 1970). Socioeconomic factors, such as relative deprivation and the threat of downward mobility, inequality, and diminished expectations for the future, are important in understanding why individuals lack generalized trust or engage in exclusivist civil society associations that promote particularized trust (Chambers and Kopstein 2001). Social trust is also associated with more equal incomes, effective public institutions, as well as the quality of education (Knack and Keefer 1997). The willingness to trust others is closely connected with the nature and extent of social integration not merely in specific civil society groups, but in the society at large, or the way in which people relate to others depending on the expectations of others, and understanding of their own role in society (Hall 1999), willingness that is unlikely to materialize under conditions of economic inequality (Putnam 2000; Uslaner 2000). As a result, one is unlikely to reverse the decline in generalized trust and the rise of particularized trust in the United States in recent decades until the trend toward the growth in economic inequality is

9780230113176_05_ch04.indd 58

2/5/2011 10:54:10 AM

Social Trust and Its Origins

●

59

reversed (Uslaner 2000). Government performance in addressing these problems is therefore crucial toward creating social trust (Aberbach and Walker 1970), as greater socioeconomic security has been found to lead to increases in social tolerance, itself an indicator of generalized trust (Persell, Green, and Gurevich 2001; Chambers and Kopstein 2001). The experience of states undergoing democratic transitions that have been accompanied by increases in inequality underscores the linkage between trust and inequality, as the drop in generalized trust in such states has occurred even as they were becoming freer and more democratic (Uslaner 2002). Thus while social trust does not appear to be caused by either economic development or democratization (in fact, a much stronger case can be made for a reverse relationship), research to date strongly suggests that a key precondition for the emergence of generalized social trust is the absence of anarchy on the domestic level (Hooghe and Stolle 2003). In view of the findings that in order to create trust governments must address problems of unequal life chances and other forms of relative deprivation (Chambers and Kopstein 2001), it is unsurprising that states where inequality is relatively low enjoy elevated levels of generalized social trust (Uslaner 2002). States displaying high levels of generalized trust are typically characterized by large government bureaucracies engaging in wealth redistribution, education spending, and playing an active role in the economy (Uslaner 2003). Of particular importance is the degree of inequality in the society as evidenced, for example, by the very high levels of social trust in Scandinavian countries that display low levels of income and gender inequality. Countries with most highly developed institutionalized welfare states also tend to have the highest levels of trust (Hooghe and Stolle 2003), and are moreover characterized by welfare states that are impartial and that do not pit one group of the population against another by a variety of exclusionary policies and institutional design that either facilitates fraud or makes access more difficult (Rothstein and Stolle 2003). The importance of government intervention in the building of social trust is also evidenced by a study testing Putnam’s proposition that declining group associations lead to a decline in trust against the experience of Great Britain that has found that, despite the existence of similar problems of time pressure and television, no similar erosion of social capital has taken place in Great Britain. This discrepancy

9780230113176_05_ch04.indd 59

2/5/2011 10:54:10 AM

60

●

Social Trust, Anarchy, and International Conflict

has been attributed to educational reform, transformation of the class structure, and government policies stressing importance of the distributive dimensions of social capital and the impact that governments can have on it. Moreover, the civic sector itself has been reliant on the government involving it in the delivery of variety of social services and providing substantial resources to its maintenance (Hall 1999). To be sure, Uslaner’s treatment of the role of government policies in the creation of generalized social trust appears to be somewhat inconsistent. While acknowledging the role of state action in reducing inequality (Uslaner 2003), Uslaner has argued that governments are not really capable of influencing trust levels, and that the forcing of citizens to obey the law does not build trust (Uslaner 2002, 2003). At the same time, Uslaner believes that government policies are perfectly capable of destroying generalized trust (Uslaner 2002). Instead, Uslaner (2002, 2003) believes that optimistic individual attitudes are responsible for trust rather than vice versa. In Uslaner’s view, while optimists are likely to believe not only in a better future but also in personal political efficacy and translate these feelings into generalized trust of others, the pessimists’ feelings of low efficacy and fearful anticipation of the future are reflected in generalized mistrust and proauthoritarian attitudes (Uslaner 2002). Similarly, Uslaner argues that trust is responsible for more efficient institutions, not the other way around (Uslaner 2003). Nevertheless, while the linkages between optimism, generalized trust, and efficient attitudes that Uslaner outlines are sound, he does appear to neglect the role the government may play in promoting policies of social equality, which he acknowledges are an important component of social trust, and he also underestimates the possibility of existence of a mutually reinforcing loop of efficient institutions that pursue policies promoting social trust and trusting individuals who support their institutions. Moreover, per Uslaner’s arguments that trust cannot take root in a hierarchy and that rigid social orders and strong class divisions that persist across generations are inimical to the creation of generalized social trust, as different class identification undermines the possibility of individuals viewing each other across the class divide as the same ourselves (Uslaner 2002), to the extent that governments are capable of breaking down these dividing lines they are surely agents of trust creation.

9780230113176_05_ch04.indd 60

2/5/2011 10:54:10 AM

Social Trust and Its Origins

●

61

And, inasmuch government is responsible for shaping decisions on resource allocation, which Uslaner acknowledges is a determinant of trust (Uslaner 2002), government is undoubtedly a factor behind trust formation. Uslaner’s unwillingness to attribute the role of social trust promoter to government institutions may be caused by his belief that trust is acquired early in life and does not change much over time. However, these two propositions are not unnecessarily incompatible. The notion that social trust is formed early in life and does not change rapidly is supported by literature on heuristics and schemata, which indicates that these decision-making shortcuts are resistant to conflicting information. Nevertheless, one must acknowledge the possibility that government actions have shaped the environment in which these fundamental beliefs about the human nature and the character of interpersonal interactions (including social trust) undergo the process of initial formation. When one’s ideas about the nature of the world and the trustworthiness of other individuals are being formed, one can reasonably expect that the outcome of this formative process will be affected if one becomes politically socialized in a corrupt state governed by a weak government (Uslaner 2003). Furthermore, governments facilitate trust generation by enforcing rights, rules, and contracts (Hardin 1996; Levy 1998; Guseva and Rona-Tas 2001), as well as by providing information (Levy 1998; Hooghe and Stolle 2003). Political institutions are an important determinant of trust to the extent they are fair and impartial in the way they administer public policies (Rothstein and Stolle 2003; Mishler and Rose 1997; Aberbach and Walker 1970). These aspects of government performance are closely associated with generalized reciprocity and other aspects of social trust (Knack 2002), though only when such governance is fair and impartial (Sztompka 1999). Indeed, it appears that social trust appears to be missing precisely where there is absence of effective government. For example, the discussions of the low levels of social trust in Italy tend to mention the weakness of the Italian state as a factor, particularly in the southern regions of the country (Putnam 1993; Huysseune 2003). Under the conditions, corruption and low government performance prevent trust-fostering policies from emerging in a vicious, interlocking circle of mistrust, which has proven extremely difficult to break (Putnam 1993). Unsurprisingly, it is the most corrupt states that tend to have

9780230113176_05_ch04.indd 61

2/5/2011 10:54:10 AM

62

●

Social Trust, Anarchy, and International Conflict

the lowest levels of generalized social trust (Uslaner 2003). Indeed, the governments’ failure to provide for the well-being of their citizens, in other words the failure to ameliorate the state of anarchy, is a certain path toward loss of legitimacy and subsequent collapse of the state (Jackson 2005). That effective governance should be responsible for the creation of social trust is unsurprising given the already well established role of the state as a creator of national identity, a concept that itself is unimaginable in the absence of generalized social trust. As discussed earlier, this process, which requires citizens living within the borders of a single state to define their identity in terms of that state, implies the existence of a certain level of social trust among its members due to the inherent characteristic of social trust as a property of groups. All states strive to establish and maintain a certain level of trust as a prerequisite to sustaining a sense of national identity, fostering commerce and economic growth. The oft-quoted description of a nation as an “imagined community” implies the existence of a certain minimum level of social trust consistent with individuals regarding other perfect strangers (who moreover may speak a different language, or be of different ethnicity) who merely happen to be living within the same (potentially arbitrary, and in any event socially constructed) borders as in some way the same as themselves. The nineteenth century nation-building efforts in Europe focused on the creation of a sense of common identity and fate, as well as identification with the state, phenomena that are all consistent with the effects ascribed to high levels of social trust. Moreover, the process of creating a strong national identity required considerably more than the redrawing of borders. It took the form of considerable government-directed in various forms of national infrastructure, including education, transportation systems, and welfare states (Hobsbawm 1995; Weber 1976). Social trust, therefore, appears to be the result of effective nation-building by the state. What remains to be explored is the process by which social trust is created by effective governmental action on the micro level. That is the focus of the next chapter.

9780230113176_05_ch04.indd 62

2/5/2011 10:54:10 AM

CHAPTER 5

Overcoming Particularism

W

hile literature on social trust contains considerable empirical support for the proposition that effective governance fosters generalized social trust, the actual workings of the process by which that transformation takes place remain somewhat mysterious. And yet this is a rather important question to answer because, after all, there is overwhelming weight of evidence particularisms, or attachments to one’s own groups, and corresponding enmity toward others both form the basis for violent political conflict at all levels and are unfortunately deeply entrenched and very difficult to overcome. Some have gone so far as to argue that they are impossible to overcome, on the grounds that strong group identity requires the existence of enemies (Huntington 2006), that the sheer cultural diversity of human communities on the planet, each marked with certain distinctive characteristics, makes particularism inevitable (Miller 1989), and that particularized trust’s narrow application makes it highly unlikely that it can be extended to individuals unlike ourselves (Fukuyama 1995). And yet the existence of states that enjoy high levels of generalized social trust, states that have successfully overcome regional loyalties and other sources of particularism to forge strong civic national identity based on the allegiance to the state, are positive proof that such particularisms are not an unavoidable feature of domestic or international politics. Group Membership as a Source of Conflict

Individual identity as a member of a group has long been recognized as a possible source of conflict (Coser 1956), both domestic and international, particularly since group formation may be inevitable (Mercer 1995), due to the variety of public goods that groups provide

9780230113176_06_ch05.indd 63

2/5/2011 10:55:16 AM

64

●

Social Trust, Anarchy, and International Conflict

(Guetzkow 1957), which include security and safety (Druckman 1994; Wendt 1994), and a sense of positive identity and self-esteem (Hogg et al. 1995; Tajfel 1970, 1978, 1982; Tajfel and Turner 1986 [2004]; Hogg and Abrams 1988; Wendt 1994). As such, groups individuals choose to belong to have a profound impact on individual identity, as well as the individual sense of who one is and how one is to relate to others (Hogg and Abrams 1988; Wendt 1994). Group belongingness, in turn, generates intragroup attraction on the one hand, due to the projection of one’s positive self-esteem onto other members of the same group (Tajfel 1981, 1982; Hogg and Abrams 1988). This, in and of itself, is a positive development, as the creation of intragroup attraction goes hand in hand with the creation of the norm of generalized reciprocity within the group, which, as has been demonstrated in the preceding chapter, is the foundation for the possibility of concerted, collective action by the members of the group and cooperation within the group. Unfortunately, group membership can also carry a dark side, as the fact that group membership can be a source of division between individuals; it is the precondition for intergroup conflict. For the flip side of intragroup attraction is intergroup discrimination, or the propensity to treat nonmembers of whichever group one considers oneself a member of in a manner inferior to that afforded to group members. To make matters worse, only minimal conditions are required to trigger intergroup discrimination. No matter how trivial or arbitrary the criterion for division, intergroup discriminations are likely to emerge (Mercer 1995) even in individuals as young as seven years (Vaughan et al. 1981). The strong group identity thus developed is likely to result in discrimination against out-group members (Mercer 1995), and the sense of loyalty built through group socialization process creates negative feelings toward “others” (Druckman 1994). While bias toward one’s group that comes with loyalty is critical to defining who one is and occurs in both cooperative and competitive situations, it moreover tends to be stronger in situations that have a competitive structure, rather than a cooperative one (Druckman 1994). It does not help that so many aspects of politics revolve around the allocation of scarce resources, be they finite budget funds or natural resources. The aggression against out-groups, in turn, may be caused by displacing repressed aggression toward authority is displaced onto the

9780230113176_06_ch05.indd 64

2/5/2011 10:55:16 AM

Overcoming Particularism

●

65

out-group, suggesting the linkage between in-group loyalty and out-group hostility may be difficult to break (Hogg and Abrams 1988). It has even been suggested that conflict is a necessary component of maintaining group identity (Simmel 1955), since conflict serves a function of preventing the disappearance of intergroup boundaries, which may be of interest to the leaders of the group in question as well as to ordinary group members who derive their self-esteem from group membership, and may be cultivated as a way of preserving a certain status quo. Given the usefulness of conflict to maintaining group identity, it is not surprising that in some situations the object of a conflict or a fight may be the fight itself and its sociological effects on group cohesion, a phenomenon that represents the underlying logic of the diversionary war theory. The remarkable ease with which hostility can be suggested does indicate a deeply embedded social function of conflict (Simmel 1955). Polarization through conflict may be pursued in order to achieve not only group convergence, but also homogeneity, through enforced conformity to an (usually fairly extreme) in-group norm, which in turn results in bolstering the position of the group leadership. Since, in accordance with most group norms, challenging authority figures is inherently illegitimate, a legitimate scapegoat must be found creating a psychological need for the existence of out-group scapegoats. Under such conditions, comparisons with dissimilar groups are made in order to achieve homogenization and convergence, thus stifling in-group instability as well as challenges to the power of that group’s elite (Hogg and Abrams 1988). The social effects of conflict on group identity and behavior suggest that, while national identity may be created, as discussed in the preceding chapters, through state efforts to eliminate domestic anarchy, it clearly is not the only path to its creation. Since conflict can have group-forming properties including on state level, state formation on occasion takes place through conflict that, in effect, serves as a substitute for the provision of public goods by the government that is unable or unwilling to provide such goods. Since the provision of public goods in the form of good governance is expensive and requires considerable tax revenue, whose collection may require overcoming the inherent mistrust of the government and the resistance of a variety of entrenched interests, consequently, given the difficulty

9780230113176_06_ch05.indd 65

2/5/2011 10:55:16 AM

66

●

Social Trust, Anarchy, and International Conflict

in establishing effective governance, states have long taken the comparatively easier path of resorting to various forms of violence to establish and maintain themselves, and have even been described as protection rackets, inasmuch as they tend to perpetuate conflict even to the point of creating or inventing internal or external threats due to the constituting effects of conflict on themselves (Tilly 1985). The temptation to create internal stability and cohesion in this manner on state level may even result in the so-called garrison state, where the entire state revolves around the dominance of specialists in violence, so much so that they become the most powerful substate actor and who normalize the perpetual expectation of violence. In such circumstances, the ruling elite maintains itself in power to a significant extent by invoking (real or imaginary) external foes in order to extract internal compliance. This is accomplished through the manipulation of out-group images in order to convey a heightened sense of permanent threat (Regan 1994). Thusly militarized state may even maintain the outward appearance of democracy even as internal decision making has been monopolized by the national security apparatus. The ensuing militarization creates a society that is organized around preparation for war and socialized to accept the continued security elite dominance, and preeminence of the security sector over the rest of society. Under these conditions, the prevailing state ideology holds that the use of force is a natural, expected, and effective means of international intercourse, with few effective limits on such use (Regan 1994). Consistent with Simmel’s (1955) argument, the society may remain under the dominance of the militarist ideology even long after the threat against which it originally organized itself has vanished due to the active promotion of the virtues of strong military preparation, possibly bolstered by their occasional use (Regan 1994). A number of totalitarian states have resorted to the use of such imagery to bolster their own legitimacy, including the Soviet Union whose ruling party’s legitimacy rested largely on its ability to withstand Nazi Germany’s onslaught during World War II and the subsequent militarized stand-off against the United States and NATO. Similarly, the ideology of the Nazi state rested, for the similar reason of pursuit of legitimization, on the assumption of a permanent race-based war against the surrounding world. Nevertheless, such extreme cases represent exceptions rather than the rule, and the available choices are not limited to a mistrustful

9780230113176_06_ch05.indd 66

2/5/2011 10:55:16 AM

Overcoming Particularism

●

67

polity in a badly governed state, or to a militarized garrison state under a state of permanent self-imposed siege. Aside from the preceding discussion of patriotism and nationalism, if the particularist thesis advocated by Huntington and others were correct, then what is one to make of the democratic peace theory itself (Abizade 2005), which, after all, argues for the possibility of like states perceiving themselves as being members of a shared community and therefore overcoming their particularistic differences? Or, for that matter, how is one to explain the existence of strong national identities in fairly diverse states (for example, the United States and Switzerland, to name but two)? We have already seen that the relationship between group identity and conflict is complicated by findings that strong in-group loyalty does not necessarily lead to strong out-group hostility (Yamagishi and Kiyonari 2000; Orwell 2000 [1945]; Druckman 1997), including on national level. Therefore, if one is to successfully address the problem of particularism as a source of political violence at both domestic and international levels, one must first identify the sources of what Granovetter (1973) initially termed the “strength of weak ties,” or the ability of individuals to trust large numbers of individuals who are not members of their own particular group, and thus determine the conditions under which a strong sense of group does not automatically translate into hostility toward members of other groups. For that, one has to more closely examine the properties of inter- and intra-group dynamics and factors influencing them and the interplay of generalized and particularized trust in group interactions. The common origin of national identity and generalized social trust is no coincidence, as possession of social trust is closely associated with the sense of group identity. Strong group identity implies sharing, cooperation, mutuality of interest with other group members, as well as a willingness to sacrifice personal interest when the welfare of the group demands it (Mercer 1995), all of which seem difficult to imagine in the absence of mutual trust among group members. The concept of “moral community” that is inherent in the concept of social trust, or a group of people toward which the norm of generalized reciprocity will apply, should therefore be viewed as at least identical to, if not necessarily synonymous with, the concept of “in-group.” Group membership, after all, implies a certain degree of trust among

9780230113176_06_ch05.indd 67

2/5/2011 10:55:16 AM

68

●

Social Trust, Anarchy, and International Conflict

members of that group, with interactions between members of the same group being driven by the perception of similarity and homogeneity of the group and its members. Generalized social trust, moreover, implies a belief in the underlying commonality of values (Uslaner 2002). However, generalized social trust surely cannot simply be a property of a group—trust shared by members of a small, cohesive group would almost certainly be of the particularized variety, since the members of such a cohesive group would have intimate knowledge of their fellow group members and moreover would operate under the shadow of the future if there was no expectation the group was about to cease its existence. As the preceding discussion of the interaction of civil society participation and generalized social trust amply illustrates, high levels of interaction within a cohesive, closely knit group appears to be contrary to the notion of constructing generalized trust. Instead, what we are looking for here are the origins of trust that transcends such bonds and that allow one to trust individuals who are not members of the same particularized groups. The answer, as it turns out, lies in the mutually complementary role of two of the major theories of group behavior, the identity theory and the social identity theory. Group Members or Role Players?

At first glance, these theories do not have all that much in common. The social identity theory (Tajfel and Turner 1986 [2004]) holds that a positive self-concept is of great importance to most individuals, and that to maintain such self-concept individuals in question seek to make positive evaluations of groups to which they belong. This positive self-evaluation, however, can come only at the expense of other groups, a process that is at the foundation of intergroup conflict. The low regard consequently felt for other groups outwardly appears to offer support for the arguments that strong group identity is inconceivable without the existence of enemies. Nevertheless, the social identity theory and its proponents also make it clear that such outcomes are not necessarily inevitable, as two individuals coming into contact with one another will not necessarily interact with negative mutual views driven by their own positive selfperceptions associated with their own group identities and concomitant

9780230113176_06_ch05.indd 68

2/5/2011 10:55:17 AM

Overcoming Particularism

●

69

negative perceptions of the other. Instead, interaction will fall at a point on a continuum between two extremes of behavior. Under certain conditions the individuals will interact, indeed, as individuals, bringing to the table, as it were, only their individual characteristics without being affected by their group memberships. Interactions falling closer to then “individual” end of the continuum will therefore be inherently more friendly, in the absence of positive own-group and negative other-group identities. However, the interaction could also occur at the other end of the spectrum, where individuals in effect forget the individuality of their interaction partners and only perceive each other’s, and their own, group membership. Inevitably, such interactions will be conflictual in nature for reasons outlined above (Tajfel and Turner 1986 [2004]). Naturally, these two extremes are ideal conditions rarely approached in practice—while the former extreme may be approached by members of the same family or by lifelong, close friends, or perhaps members of a closely knit, cohesive and salient group of individuals (for example, a highly trained, veteran military unit whose members’ alternative group identities have been submerged in their collective military identity by shared experiences in training, service, and possibly combat), the latter is reflective of opposing armed forces’ members engaged in open warfare against one another. What the identity theory does not specify are the conditions that will influence which extreme will exert a greater “pull” on the nature of the interaction. Indeed, it would appear, even from the few examples cited above, that individuals in question have only a limited ability on how to interact with one another. Instead, such interactions appear to be structurally predetermined. After all, for a soldier fighting a war, attempting to interact with a member of the opposing force as an individual would be highly risky behavior, as such attempts would likely be treated as efforts to fraternize with the enemy or, worse, might be exploited by the enemy to gain an advantage in combat. Nevertheless, what is remarkable about the social identity theory is that it allows for the possibility of interaction between individuals as individuals, in spite of the fact that out-group discrimination and therefore conflict can be easily triggered merely by the awareness of group divisions, even if the groups are not engaged in a competition of any sort (Tajfel and Turner 1986 [2004]) as Tajfel’s own “minimal

9780230113176_06_ch05.indd 69

2/5/2011 10:55:17 AM

70

●

Social Trust, Anarchy, and International Conflict

group” experiments suggest (Tajfel 1970). However, given that interaction of individuals as members of groups appears to be rather undesirable on the grounds that it may lead to conflict, this poses the crucial question of determining conditions in which individuals will react to one another as individuals not only when they are both members of the same salient and cohesive group, but even when they are perfect strangers. In other words, how does one create interactions based on trust between individuals who may not have ever even seen one another before? How does one structure interpersonal relations among members of different groups? As luck would have it, such interactions already exist and they are the norm rather than exception, at least in certain countries. We all interact with one another in that manner on a daily basis. Consider, for example, what happens when we undertake to take a motor vehicle for a drive on a public road where, inevitably, we will interact with other drivers and therefore, considering the dangers inherent in operating motor vehicles, we put our lives in danger. Remarkably, almost as soon as we set out, a specter of death begins to loom above us in the form of other motor vehicles approaching us at high rates of speed and thus threatening death or grievous bodily injury. At that point we are held at risk by actions of perfect strangers. Yet these interactions nearly invariably end with no harm inflicted on either of the participants, because these interactions are fundamentally based on trust. Trust among strangers, no less. What makes that trust possible are the single or double, white or yellow, lines separating the lanes, which give both drivers the ability to signal to one another their trustworthiness that in turn makes it possible for the oncoming vehicles to pass one another with mere inches separating them. One of the remarkable aspects of the interaction is that the communication is entirely (and ideally) nonverbal, as drivers signal their trustworthiness to other drivers merely by staying within their lanes and obeying other traffic rules. This trust, moreover, is not based on simply selfinterest. Although it could be argued that individuals trust one another because it is in nobody’s interest to perish in a fiery head-on collision, in practice self-interest does not guarantee self-preservation. Consider, for example, how traffic frequently proceeds in situations like large and mostly empty parking lots where some individuals may choose not to stay between the “herringbones” designating parking

9780230113176_06_ch05.indd 70

2/5/2011 10:55:17 AM

Overcoming Particularism

●

71

spots and in the lanes designated for traffic, but decide to cut across them. Suddenly traffic slows down to a crawl, due to the fear of collision caused by the mistrust of an individual who suddenly decided to ignore the rules and, in the process, signaled his or her own untrustworthiness as a driver. This simple example is precisely how states create trust among perfect strangers and allow interaction to proceed without being hampered by mutual mistrust that would inevitably set in. Because in the absence of a common, shared set of rules which in effect becomes a language in which interacting parties can signal own trustworthiness (often nonverbally), such interactions would inevitably end in conflict due to the inherent propensity to interpret the actions of others as more threatening than they really are (Jervis 1968). Thus a “fender-bender” caused by misunderstanding or lack of attention would quickly cause one or both parties involved to suspect foul play on the part of the other party. Clearly, self-interest is not enough to generate smooth interactions. A trust-building device, in this instance motor traffic rules, is required. The above example suggests that when two or more interacting partners are interacting in accordance with one universally shared set of rules, they cease being members of opposing groups in that interaction. Rather, they become “role players,” as their positive selfconception becomes the product not predominantly of group identity but rather of role fulfillment in the process of interacting with others in the context of a broader society (Stets and Burke 2000), as proposed by the identity theory that complements the social identity theory by proposing an alternative source of a positive self-conception, termed self-efficacy by the identity theory literature (Stets and Burke 2000). Thus whereas the social identity theory holds that the positive self-concept is achieved by identification with own group through contrasting it against other groups deemed less desirable or attractive, the identity theory proposes that a similar effect can be obtained by fulfilling a role while interacting with others. These distinctions have great import for the study of conflict. Social identity theory’s selfesteem building seems very difficult to achieve without at the same time creating hostility toward out-groups. However, the role-fulfillment scenario described by the identity theory is essentially cooperative, not conflictual, in nature. Whereas in-groups make necessary the creation

9780230113176_06_ch05.indd 71

2/5/2011 10:55:17 AM

72

●

Social Trust, Anarchy, and International Conflict

of out-groups with a mutually hostile relationship, one’s roles merely imply the existence of counter-roles whose mutual relationships (for example, “brother-sister,” “husband-wife”) are not conducive to hostility. Indeed, it would seem that mutually extremely hostile roles and counter-roles are difficult to imagine. Take, for example, the scenario of interaction between a prison inmate (role) and a guard (counter-role). Is it really the case that the high self-efficacy of the inmate requires the existence of the guard, or that the inmate somehow derives selfesteem in playing his/her role well opposite that of the guard? No, instead it would appear such an interaction, the subject of Zimbardo’s Stanford Prison Experiment (Zimbardo 2008), is rather the example of an extreme intergroup conflict in which individual characteristics play minimal role the social identity theory describes so well. The Stanford Prison Experiment, incidentally, perfectly showcases the mutually complementary role of the identity and social identity theories. Whereas the members of each of the two groups (prisoners and guards) interact with one another according to one set of rules and seek to achieve a positive self-conception by attempting to become the “best” guard or inmate they can, they also seek such self-conception through an out-and-out denigration of their respective out-group. Moreover, it should be noted that identity theory’s role fulfillment scenario implies the existence of trust, as individuals playing their specific roles signal (and are signaled in return) their ability and willingness to play their assigned roles and in doing so also signal their trust the “counter-role” players will act in ways their own roles demand, much as drivers of vehicles on a busy public road do. The task and, indeed, the capability of a strong, effective state is to allow all inhabitants to interact not as members of groups (a scenario that would surely lead to increased conflict and violence as individuals fall back on their ethnic, religious, or linguistic group identities) but rather as individuals, using shared sets of rules as a foundation for mutual interpersonal trust of the generalized variety. Because as these conflict-free interactions (conflict-free thanks to the existence of a common set of rules designed to make them so) continue, with both sides benefitting from them, a second-order effect of generalized trust will set in as, consistent with the concepts of bounded rationality and satisficing, individuals will simply begin to assume that the interaction with yet another perfect stranger will take place without any problems.

9780230113176_06_ch05.indd 72

2/5/2011 10:55:17 AM

Overcoming Particularism

●

73

This constitutes the first major contribution of the state to the creation of generalized social trust, namely, the provision of shared sets of rules and scripts that individuals can use to communicate and interact with other individuals with minimum of friction. The second major contribution is the reduction of risk associated with trusting others. For, as Hardin (2001) argues, trust is not a solely dispositional phenomenon but a contextual one as well. People trust one another to do certain things, but not others. Therefore the higher the stakes associated with trusting others, the lower the likelihood of trust actually developing in that context. To the extent that a trust-based relationship can be viewed as a prisoner’s dilemma game on the grounds that for mutual trust to dominate the interaction both actors have to have a reasonable expectation that the other actor will reciprocate their trust and thus “cooperate,” state action has the effect of limiting the extent of the damage one will suffer if one’s counterpart “defects,” and fails to reciprocate one’s trust. The reduction of risk takes many forms. One of them is the provision of a variety of societal benefits ranging from the welfare state to education and health care benefits means that one’s existence no longer teeters at the edge of survival. This quite aptly named social “safety net” thus allows one to trust others by eliminating much of the danger associated with falling when one’s trust is not reciprocated. The state’s monopoly on the legitimate use of force similarly increases incentives for individuals to trust one another by placing at least a normative limit on the extent to which interpersonal conflict can escalate and, in extreme circumstances, by punishing “defectors” who seek to prey on the trust of others. For example, it is difficult to underestimate the importance of institutions like the U.S. Securities and Exchange Commission in creating trust in the U.S. stock market by threatening to punish those who would engage in exploiting the trust of others. Finally, the elimination of inequality, which, as discussed in the previous chapter, is also associated with social trust creation, is important because it reduces the relative capabilities of the interacting actor, thus reducing the risk accepted by the weaker of the two actors when entering into the interaction. Consider, for example, the difference between two seemingly identical groups of individuals: a military unit and an organized crime group. Both are highly salient and cohesive groups, at least in part due to the existence of external threats and to the intimacy and “shadow of

9780230113176_06_ch05.indd 73

2/5/2011 10:55:17 AM

74

●

Social Trust, Anarchy, and International Conflict

the future” that influences intragroup interactions. Moreover, in both cases the group members are heavily armed. And yet whereas the military organization can function under extreme stress, with individuals routinely risking their lives for the benefit of the larger group (it is not uncommon to encounter cases of military personnel risking their lives on behalf of other military personnel or civilians who are not known to them personally) and not for personal benefit, such self-sacrificial behavior is considerably rarer in an organized crime organization, even though one would at the very least expect the presence of particularized trust binding group members together against outsiders. Therefore why do well run military organizations display high levels of both particularized and generalized trust, and organized crime organizations do not? There are several reasons why there is no honor among thieves. First, the consequences of misplacing one’s trust in one’s fellow organized criminal are enormous, ranging from imprisonment to death. Second, with large sums of money changing hands, the incentive for breaching trust can be too high to ignore. But these factors are likely secondary, given that the stakes, and consequences of misplacing one’s trust in combat, will likewise be very dire. The presence or absence of generalized social trust therefore is attributable to something else. What an organized crime group lacks and a military organization possesses is a clearly defined set of individual responsibilities, roles each individual is expected to play in the social context of the group. These obligations are moreover fairly enforced, ensuring that there will be minimal violations. Under such conditions, military organizations are able to foster generalized trust on which their battlefield effectiveness depends. In stark contrast to a military organization, an organized crime group represents a lawless, anarchic, self-help institution where rules, even assumed they exist, tend to be enforced in an arbitrary fashion. Consequently, it is not surprising that generalized social trust is altogether absent in such settings. Generalized Social Trust and Conflict

The relationship between state effectiveness, social trust, and political violence has already been implied in the discussion of the sources of national identity and the propensity of states with weak national

9780230113176_06_ch05.indd 74

2/5/2011 10:55:17 AM

Overcoming Particularism

●

75

identities to be more conflict-prone. However, there also exists more direct evidence that social trust and conflict are closely related. The notion that societies bereft of social trust tend to be more fraught with conflict is implied by extant literature that tends to agree that, in the absence of social trust, society becomes less civil. Individuals are more apt to deny their political foes are part of their moral communities and, in the absence of the generalized reciprocity norm, acrimony increases and the norms of cooperation in governance wane. The unwillingness to compromise leads to a more confrontational style of politics (Uslaner 2000). Whereas high trusters have an inclusive view of society and are tolerant, low trusters (individuals with low generalized trust, though possibly high particularized trust) view society as essentially a cluster of disconnected groups struggling for individual advantage, in what amounts to a state-level struggle for relative gains. Such a society is characterized not by cooperation but by competition (Uslaner 2002). Unsurprisingly, social trust has been shown to have a significant direct effect on homicide rates (Rosenfeld et al. 2001) and other types of violent crime (Putnam 2000), as well as attitudes toward the use of violence (Putnam 2000). Individual trust levels have also been demonstrated to influence the outcomes of game scenarios involving resource consumption in a simulated commons dilemma, with high levels of trust being shown to play a major role in mediating resource consumption. The high trusters’ expectation of reciprocation led to their reduced consumption of a rapidly deteriorating resource (Brann and Foddy 1987). Similar results have also been noted in prisoner’s dilemma games, where high trusters reacted positively to cooperative messages by increasing cooperation, while low trusters reacted by decreasing cooperation (Parks et al. 1996). Generalized trust levels also influence attitudes toward other states, which is an unsurprising finding given that mistrust and fear of other actors have played a crucial role in many explanations of international conflict, including the Cold War (Kydd 2000). Individual world-views incorporate a general sense of the extent to which other international actors may be trusted. Sometimes referred to as international trust, this form of trust influences preferences toward internationalism or isolationism and threat perceptions of other states. Whereas “high trusters” tend to view the international

9780230113176_06_ch05.indd 75

2/5/2011 10:55:18 AM

76

●

Social Trust, Anarchy, and International Conflict

system as populated by friendly and trustworthy states, “low trusters” have a considerably darker view of the intentions and motives of other states. Unsurprisingly, this form of trust was a strong predictor of support toward U.S. initiation of military action against Iraq in 2003 (Brewer et al. 2004). Therefore the logic of generalized social trust, the role of the state suggested by the social identity and identity theories, and the clear linkage between the absence of social trust and certain forms of political attitudes, all point toward the need for creating a comprehensive theory of international conflict based on state strength and social trust.

9780230113176_06_ch05.indd 76

2/5/2011 10:55:18 AM

CHAPTER 6

The Social Trust Theory of International Conflict

S

o far we have accounted for particularized trust obtaining within relatively small and closely knit groups whose members have a direct incentive to reciprocate trust placed in them due to the shadow of the future, personal intimacy, and corresponding reputational costs, the group’s likely ability to sanction “non-trusters” (organized crime organizations come to mind here), or the sheer importance of membership in the group to the individual in question. We have also accounted for generalized trust that is created through effective governance that enables individuals subscribing to the same set of laws to interact in a conflict-free manner due to their ability to signal intended behaviors ahead of time. But what of international interactions where, by definition, the anarchic international system limits if not altogether the possibility of actors being able to signal their peaceful intent to one another? How exactly does the presence of high levels of social trust at state level influence international behavior? For international anarchy surely represents a system in which generalized social trust should have no influence, given the absence of effective governance and its benefits, thus validating the view of structural realists that domestic and international politics operate on diametrically opposite principles. Before jumping to that conclusion, however, one must consider the heuristic nature of generalized social trust. Of Alien Encounters and Heuristics

Wendt’s (1992) “alien encounter” scenario well illustrates the influence on international interactions of generalized social trust created

9780230113176_07_ch06.indd 77

2/5/2011 10:55:28 AM

78

●

Social Trust, Anarchy, and International Conflict

at the domestic level. To recap, the scenario posits that two actors, each of which is interested in own survival and has certain material capabilities, meeting each other for the first time with no history of prior interaction would not necessarily act in ways predicted by the realist theory. The question that remains to be answered is how exactly would they react to each other’s presence? In the absence of any prior interactions and therefore knowledge of each other’s capabilities and intentions, how would these actors (specifically, the human actors involved in the interaction) act when confronted with such an utterly unfamiliar scenario, and what factors would influence how they would react to one another? There are several possibilities. The first one is that, in accordance with the predictions of the social identity theory, the absence of any history of prior interaction would mean the human actors would slide toward the “group member interaction” extreme, with ultimately dire results. The sheer potential for miscommunication and misperception combined with the inherent human propensity to impute malign intent to the actions of others even when such intent is entirely absent (Jervis 1968) likewise points in the direction of the encounter ending badly. All in all, the default outcome appears to be conflictual in nature, and overcoming its strong likelihood seems like a tall order. Happily for the well-being of both the humans and the aliens in the encounter, there nevertheless remains a possibility the encounter would not end badly. As far as the human actors are concerned, at least, they would not have arrived at the encounter with anything resembling a blank slate, but instead their manner of interacting with the alien species would strongly reflect their beliefs about the proper mode for interacting with others. As it happens, humans have a well developed (if far from perfect) apparatus for dealing with uncertainty and unprecedented situations. Considering that human actors operate under conditions of bounded rationality, which limits their ability to make evidence-based decisions due to the absence of information or time to process it, and are moreover considered to be cognitive misers in that they opt for the least labor-intensive way of making a decision in order to preserve mental resources for other tasks demanding their attention, the existence of such an apparatus is a necessity. Thus when confronted with complex or uncertain situations, individuals

9780230113176_07_ch06.indd 78

2/5/2011 10:55:28 AM

The Social Trust Theory of International Conflict

●

79

fall back on a number of decision-making aids, such as heuristics or schemata, which are particularly useful when the actor is faced with the need to evaluate a novel situation (Axelrod 1973), and moreover their decision making is likely to be affected by deeply felt needs and intense emotions, with one’s effect toward the target influencing whether situation or dispositional information is going to be utilized (Stein and Welch 1997) even when perfect information is available. Heuristics are therefore simply decision-making shortcuts that individuals use to test propositions embedded in their schemata on the basis of their availability (the tendency to interpret information that is ambiguous in terms that are most available in their cognitive repertoire), representativeness (the tendency to exaggerate similarity between one event and another one), and anchoring (the magnitude of an event estimated by comparing it with an initial value). The use of such decision-making shortcuts by both elites and masses has been well documented in both national and international politics. The use of cognitive heuristics has been found among most voters, particularly in complex situations (Lau and Redlawsk 2001). Political leaders, even though they can usually avail themselves of far more complete information than the average voter, nevertheless rely on multiple heuristics in screening alternatives on the basis of risk minimization and reward maximization (Mintz and Geva 1997). In international decision making, political leaders seek out the most relevant analogies to analyze a novel set of circumstances (Houghton 1996). Moreover, and this is of crucial importance to the argument advanced in this book, it has long been argued that mutual perceptions by international actors are an important factor in explaining conflict behavior (Kim and Bueno de Mesquita 1995; Jervis 1968, 1988), with international discourse being heavily shaped by projecting domestic experiences onto international actors (Berejikian and Dryzek 2000), even if the rules governing the actual use of such heuristics remain relatively poorly understood (Houghton 1996). Thus when faced with the unprecedented situation of an alien encounter, the human actors would not have evaluated the situation solely using data gathered from the encounter, but would have supplemented that data and evaluated it using a wide variety of heuristics, biases, and schemata to deal with the unknown, and there is every reason to expect that the human actors’ level of generalized social trust would

9780230113176_07_ch06.indd 79

2/5/2011 10:55:28 AM

80

●

Social Trust, Anarchy, and International Conflict

influence whether they viewed this novel encounter as a benign or threatening experience. The basic properties of generalized trust are entirely consistent with literature on schemas and heuristics. The nature of trust and its application to solve decision-making problems under conditions of insufficient information warrant the treating generalized trust as a decision-making heuristic. Whether their level of trust is high or low manifests itself as a bias in their decision making by overriding context-specific information (in situations where it is available) concerning other actor’s motives and intentions. Trust therefore may be viewed as a functional alternative to rational prediction for the reduction of complexity, under conditions of absence of time and other resources to rationally predict and control the effects of contingent outcomes (Lewis and Weigert 1985). Individuals trust holistically, evaluating other actors along a single dimension (Mishler and Rose 1997) and by projecting assumptions onto other actors (Lewis and Weigert 1985; Orbell and Davis 1991). Although there are no indications in the literature that having a high level of trust means that an individual is going to be less likely to gather information, other theoretical approaches suggest the low trusters’ decision making is going to be impaired to a greater extent than that of a high truster, for a number of reasons. Low levels of trust are associated with high levels of stress (after all, the low truster thinks he’s living in a Hobbesian state of nature), and stress is associated with impaired decision making. Also, in accordance with the propositions advanced by the prospect theory (Kahneman and Tversky 1979), having a low-trust level (i.e., expecting everyone around you to attempt to take advantage of you) is also likely to translate into a greater likelihood of considering oneself in a losses frame regardless of the objective circumstances of the situation, and thus in a risk-acceptant state of mind. Such an outcome appears all the more likely because, in spite of the influence on the ultimate decision taken of whether the situation is framed as a prospect of gain or loss, so far the process by which frames are generated remains very inadequately understood. Therefore, if anything, one should expect the low trusters to pay even less attention to available information than high trusters, and be more likely to make decisions influenced by preexisting biases and beliefs.

9780230113176_07_ch06.indd 80

2/5/2011 10:55:29 AM

The Social Trust Theory of International Conflict

●

81

This feature of trust as a heuristic device used to fill in gaps in existing knowledge is also supported by recent research in neuroscience that suggests humans have a relatively advanced ability for predicting the actions of others by predicting their mental states through attempts to simulate mental processes of others in their own brains. This prediction, however, appears to take form of simulation of one’s own actions if put in the situation of the individual one is trying to simulate (Ramnani and Miall 2004), which amounts to, in effect, “putting oneself in someone else’s shoes.” Similarly, research on the concept of “mentalizing” suggests that information required for estimating the mental state of other individuals comes from two sources. While one of the sources is, as one might expect, the observation of individuals and assessment of their actions, the other source is one’s own past experiences that are applied to the current situation and, in effect, projected onto other individuals (Frith and Frith 2003). These two separate strands of research strongly suggest generalized social trust is used as a heuristic device in an effort to predict how other individuals would act in an interaction with oneself, with individuals possessing high levels of social trust unsurprisingly drawing conclusions their interaction counterparts will be trustworthy, and with individuals with low levels of trust concluding, on the basis of having evaluated of how they themselves would have acted if placed in the other person’s position, that their counterpart is untrustworthy. It is not difficult to theorize on the basis of the preceding discussion of the heuristic character of generalized social trust as to how the level of social trust might influence the alien encounter described by Wendt (1992). Consistent with the nature of generalized trust that posits that high-trust individuals act toward others on the basis of an assumption that the others are essentially trustworthy, high-trusting individuals are likely to invoke more benign, less confrontational analogies when confronted with novel situations and thus treat the aliens as essentially benign, whereas low trusters are likely to view novel situations in more confrontational terms. In other words, hightrust individuals will find the encounter resembling some sort of benign scenario they have experienced in their lives, whereas lowtrust individuals will inevitably find analogies with an event that ended in conflict. Consistent with the notion of projecting one’s own

9780230113176_07_ch06.indd 81

2/5/2011 10:55:29 AM

82

●

Social Trust, Anarchy, and International Conflict

beliefs onto other actors, individuals with high social trust levels will be highly likely to assume (in the absence of information to the contrary) that their counterparts are likewise trustworthy. Although the alien encounter scenario is a difficult one to test empirically due to the low availability of aliens to participate in such encounters, one nevertheless cannot help but be struck by how the portrayal of alien species in, for example, U.S. movie productions during the Cold War has tended to mirror the ebb and flow of that conflict. Many of the early Cold War–era movies, for example, the 1953 War of the Worlds, which was made during the early stages of the conflict when the United States still did not comprehend the nature of the threat, portrayed the encounter in strictly conflictual terms, with virtually no possibility of coexistence, let alone cooperation. That depiction stands in stark contrast with that shown in the movies made during the era of the U.S.-Soviet détente, especially the 1977 Close Encounters of the Third Kind, which posits the avoidability of conflict between alien cultures and the possibility of communication and even cooperation based on mutual trust. Here one must stop to consider that the alien encounter scenario is useful not only because it is a vehicle for exploring initial interactions among political actors, but also because encounters in the international system not infrequently resemble such an encounter, when representatives of different states meet for the first time to discuss their political futures. Here, too, social trust strongly influences the nature of the interaction. Social Trust and International Relations

Arguably, something similar occurs in international interactions that, at first, at any rate, may resemble the alien encounter scenario inasmuch each individual interacting for the first time with an individual representing a foreign state will likewise not enter into the encounter with an entirely blank slate. The international decision maker, however inexperienced in the realm of international politics, nevertheless will have had ample experience of interaction with domestic political actors, and the history of these interactions will have created a store of heuristics and schemata for use in continued interactions, including to help deal with unfamiliar situations.

9780230113176_07_ch06.indd 82

2/5/2011 10:55:29 AM

The Social Trust Theory of International Conflict

●

83

Here, too, when evaluating the trustworthiness of one’s interaction partner, the lion’s share of the decision will depend on how the simulation of the other actor’s actions will turn out in the evaluating individual’s mind, depending on the level of social trust the evaluating individual possesses. However, as the discussion of the sources of generalized social trust in the preceding two chapters indicates, the level of social trust an individual possesses is strongly determined by state strength and effectiveness of governance in that state. Thus it matters greatly whether the political socialization of the international decision maker took place in an anarchic, poorly governed state with low levels of generalized trust, where national identity is comparably weak and other sources of identity (ethnicity, regional or familial ties) are more salient, the only trust is of the particularized variety (i.e., among members of the salient identity group), and encounters between members of other identity groups tend to acquire a more confrontational air if only because such encounters will gravitate toward the “members of groups” extreme proposed by the social identity theory. It stands to reason that individuals socialized into such differing domestic political environments will acquire radically different notions on the nature of others and their propensity to find them trustworthy. Since in such a state the only individuals to be trusted are people who share one’s own characteristics (ethnicity, language, religion, other aspects of culture) and perhaps only individuals who are personal acquaintances, foreigners will automatically be regarded with mistrust, giving rise to the possibility of conflict arising. Conversely, individuals socialized into a strong, well governed state with a strong state of national identity and relatively weak other identity markers, who are accustomed to trusting and being trusted by strangers due to the existence of a shared, fair, and enforced set of rules, who thus have a high level of generalized trust will, during their first contact with foreigners, at least entertain the possibility that the foreign actors might be trustworthy. Consequently, individuals from high-trust countries will be less prone to enter into conflictual interactions with representatives of other states than individuals from low-trust states. The level of social trust at the elite and mass levels will therefore be a major determinant of the international behavior of states, consistent with the neoliberal (Moravcsik 1997) argument

9780230113176_07_ch06.indd 83

2/5/2011 10:55:29 AM

84

●

Social Trust, Anarchy, and International Conflict

concerning the primacy of domestic actors and their interests in determining the foreign policy, and we should expect very different behaviors from states with high and low levels of trust. Such individual attitudes resulting from the individual levels of social trust and prevailing both on elite and mass level, will in turn translate into mass behaviors and influence the conflict behavior not merely of isolated individuals but entire states. The aggregate level of social trust prevailing within a state will be an accurate predictor of the level of both internal and external political violence experienced by that state. Specifically, states with low levels of generalized social trust will be very conflict-prone. These are, in accordance with the discussion of nationalism and social trust in preceding chapters, states with low levels of national identity. As a result, these are states whose inhabitants, in absence of a strong state to rally around, seek out alternative sources identity such as ethnicity or religion and, consistent with the postulates of the social identity theory, such states experience deep cleavages tearing asunder their polities. Trust in weak states is likely to be of the particularized variety and, as such, it does not extend far outside one’s family, clan, ethnic group, or another primordial form of identity. Such a state has little or no social cohesion and, since the level of social trust strongly corresponds to the level of political trust, neither the state itself nor its rulers are viewed as legitimate. To make matters worse, the prevailing beliefs regarding human nature are rather negative and violence is seen as a normal, routine means of interpersonal and intergroup discourse. Therefore low-trust states will experience high levels of domestic political violence in the form of insurgencies, political repression, civil wars, and interclan vendettas. Some of these domestic conflicts may be provoked by unscrupulous political leaders attempting to stay in power by exploiting internal divisions. The implications for international conflict are similar. After all, the absence of trust means a fundamental belief that other actors, especially strangers who are not members of one’s own politically salient groups, intend to do one harm in some way, and therefore cooperation with them would be foolish as it would merely open one up to exploitation. After all, to return to the prisoner’s dilemma game, it does not make sense to trust others, or to cooperate, if one

9780230113176_07_ch06.indd 84

2/5/2011 10:55:29 AM

The Social Trust Theory of International Conflict

●

85

enters the interaction convinced the other party will defect, or fail to reciprocate one’s trust placed in them. Moreover, given the low level of trust and the high level of domestic political violence, it is likely that both the elites and the masses are thus socialized into accepting quite permissive sets of norms on the appropriateness and efficacy of force as a means of political conflict resolution, norms that will naturally become externalized. As a result, anarchic states with low level of generalized social trust are more likely to react to international developments in a hostile manner than high-trust states, due to the leaders tending to view the international system in more conflictual terms, and/or due to popular support of the more belligerent response to international developments. This proposition is consistent with a study performed on the interaction of social trust and the support for the U.S. invasion of Iraq (Brewer et al. 2004) suggests that low-trust individuals, when faced with two foreign policy options, are more likely to choose the more belligerent one, whereas the high-trust individuals will tend to opt for the nonbelligerent one. If the majority of the members of mass public possess a world view that views conflict as an inevitable and natural part of human interaction, then any elite faction that advocates cooperation and coexistence with other states will naturally find itself at a disadvantage when faced with political competitors advocating conflict. In a low-trust state such a state of affairs may lead to process of “out-bidding,” with each political faction seeking to outdo others in promoting confrontational international policies. The leaders of low-trust states will moreover have strong incentives to engage in diversionary behavior. Per Hagan’s (1994) typology, this type of a state is most likely to create leaders with militant or radical orientations, and the fragmented nature of the state combined with the leaders’ predisposition toward violence and given that political leaders are heavily motivated by their desire to remain in power (Bueno de Mesquita and Siverson 1995; Bueno de Mesquita et al. 1999, 2003; Goemans 2000; Mintz and Geva 1993; Hagan 1994), the unifying, rallying, or homogenizing effects of conflict will be very attractive for political leaders (Barzilai 1999; Chiozza and Goemans 2004) of low-trust states. As a result, low-trust states are likely to display a high propensity to initiate interstate conflicts for the dual and mutually reinforcing reasons of mass suspicion of outsiders, the

9780230113176_07_ch06.indd 85

2/5/2011 10:55:29 AM

86

●

Social Trust, Anarchy, and International Conflict

leaders’ mistrust of outsiders, and the leaders’ incentive to attempt to solidify their hold on power and legitimacy by attempting to create a rally around the flag effect combined with the inadvisability of pursuing nonbelligerent foreign policy that the mass public, whose members themselves believe conflict is a natural part of politics, and who would likely interpret a nonbelligerent policy as a sign of weakness. Therefore a leader of a weak, partly anarchic, low-trust state would have every reason to provoke international conflicts and few incentives to avoid them. One of such incentives is the possibility that an external diversion might disastrously backfire. After all, whether the efforts to provoke a rally around the flag will succeed is an open question itself. Diversionary war literature critiques suggest that fragmented states are unlikely to rally around their leaders if the divisions are too deep and the logic of generalized social trusts does raise the possibility that a state whose inhabitants do not perceive themselves having a shared destiny with all other inhabitants of that state, and who are unable to bring themselves to sacrifice on behalf of strangers, will not fare well when exposed to the stresses of war. However, it is unlikely that the leaders in question will accurately perceive their societies’ level of fragmentation (indeed, even the diversionary war literature fails to provide reliable metrics as to when the rallying effect will fail to materialize), because the desire to effect an improvement in societal cohesion and own political survival chances through conflict will amount to motivated bias that will render the leader in question oblivious to such considerations. Therefore a leader desperate enough to retain power is unlikely to be deterred by such risks, even though it is highly likely that low-trust states will actually fail in most of the conflicts they provoke due to their low level of societal cohesion making it impossible for the society to shoulder the strains of war, a proposition to be more thoroughly tested in a future study. Moreover, the homogenizing effects of conflict on the group cohesion have authoritarian implications, as conflict tends to lead to an increased state interference in a variety of aspects of political life, legitimizes mandatory mobilization of people and resources and reduces individual liberties (Barzilai 1999). All of these increase the desirability of conflict to embattled political leaders of low-trust states, particularly since the expansion of government powers is likely to be

9780230113176_07_ch06.indd 86

2/5/2011 10:55:29 AM

The Social Trust Theory of International Conflict

●

87

a longer-lasting effect of conflict than the rallying effect. Furthermore, international conflict may be desirable to endangered political leaders due to its ability (however temporary) to give individuals a sense of identity and a mechanism for assimilation (Barzilai 1999). In other words, conflict may serve as a substitute (albeit a very poor one) for the existence of generalized social trust by satisfying the needs for identity and common destiny that are otherwise not being satisfied due to the inadequate governance by the state. Finally, adding to their own propensity to initiate conflicts, low-trust states will also find themselves frequent targets of violence by other state actors. The violent nature of domestic politics and the belligerent international behavior of leaders of such state will necessarily appear as threatening (even if this was not the intent). For example, one remembers the oft-played video footage of the Iraqi dictator Saddam Hussein firing off various firearms into the air during political rallies. It is highly likely that such behavior, intended to impress, overawe, or intimidate domestic political actors in Iraq was interpreted by the U.S. public as evidence of the unhinged, irrational nature of the Iraqi leader and was to a large extent responsible for the magnitude of the U.S. popular support for the invasion of Iraq in 2003. Returning for a second to the “empty parking lot” analogy employed in Chapter 4, such international actors, by acting contrary to the norms of political behavior accepted in observing states, make themselves appear untrustworthy and therefore outside of the “moral community.” When faced with such a blatant violation of accepted norms, even individuals with high levels of social trust are likely to reconsider. The conflict behavior of high-trust states is quite the opposite of the one posited for low-trust states. As discussed earlier, high-trust states are characterized by cohesive societies with comparatively few major internal “in-group vs. out-group” fissures. These are states where the national identity is strong and positive in nature, where the dominant ideology is patriotic rather than nationalistic in character, not requiring an external threat for its maintenance. The high degree of willingness to trust others, even complete strangers, translates into trust toward institutions and leaders and into less belligerent individual attitudes toward others. Therefore there is every reason to expect low levels of domestic political conflict in states with high levels of social trust.

9780230113176_07_ch06.indd 87

2/5/2011 10:55:30 AM

88

●

Social Trust, Anarchy, and International Conflict

The high trust levels have a similar pacifying effect on the international level. As the presence of social trust implies an underlying belief that other actors are trustworthy in the absence of strong evidence to the contrary (for example, Saddam Hussein’s behavior referenced above), i.e., that they are not going to take advantage of the individual in question when engaged in cooperation, trust fosters cooperation by ameliorating distributional concerns. Whereas one would have great concerns if an untrustworthy, and thus presumably a potentially hostile, individual were to gain a material advantage, a similar advantage would be of virtually no concern if realized by someone having our trust. Moreover, if one is trusting of others, it also means one is not likely to view violence as a necessary, or even appropriate, means of interaction with others. Other actors are, after all, seen as fundamentally friendly and not disposed to do one harm, a situation in which the resort of violence would not be appropriate. A well governed, high-trust state is also going to have a general public where jingoistic appeals to nationalistic sentiments by unscrupulous leaders are more likely to fall on deaf ears, and leaders with more restrictive beliefs on the efficacy and appropriateness of the use of force and lacking incentives to engage in diversionary behavior are more likely to enjoy political success than in low-trust states. Therefore high-trust states are characterized by a pacifying combination of leaders who are politically secure and who do not treat violence as a routine means of resolving political conflicts and masses that are less susceptible to nationalistic sentiments due to their own propensity to view others as trustworthy. This reduces the willingness and the ability of leaders to resort to an international conflict as a means of enhancing national cohesion and maintaining internal political stability. The same pattern applies also to high-trust states as targets of international violence. Here, high-trust states are able to signal pacific intent and cooperative nature for precisely the opposite reasons low trust cannot. Their domestic politics of high-trust states, and the behavior of their leaders, are such that they are unlikely to provoke a hostile response from others. These are, after all, states that appear to closely observe international rules and, in doing so, signal to others their own trustworthiness. These are states whose governments repress domestic opposition and whose leaders do not engage

9780230113176_07_ch06.indd 88

2/5/2011 10:55:30 AM

The Social Trust Theory of International Conflict

●

89

in gratuitous displays of military power that, while designed to impress others by sending an image of strength, in actuality likely provoke a hostile response. The grandiose Soviet military parades, no doubt intended to convince others of the military prowess of the Soviet state, also had the probably unintended effect of securing funding for a wide variety of U.S. defense programs to counter the large numbers of advanced weapon systems routinely paraded across Moscow’s Red Square.

9780230113176_07_ch06.indd 89

2/5/2011 10:55:30 AM

9780230113176_07_ch06.indd 90

2/5/2011 10:55:30 AM

CHAPTER 7

Research Design Independent Variable

One of the greatest challenges I encountered during implementing this project was finding a way to operationalize the independent variable for use in quantitative tests, namely, generalized social trust. I opted to use the Corruption Perception Index (CPI), compiled on an annual basis for individual states using survey data by Transparency International (Transparency International 2007). CPI is a well recognized index that is frequently used in quantitative analyses (for example, Goldsmith 1999; Sandholtz and Koetzle 2000; Anderson and Tverdova 2003) of a variety of political phenomena. This choice of operationalization, while relatively unique, is nevertheless strongly supported by literature on corruption (see, for example, Rothstein and Stolle 2008), which suggests close links between corruption and domestic anarchy, as well as between corruption (or, more specifically, absence of corruption) and generalized social trust. In short, the presence of corruption almost certainly is indicative of the government’s inability or unwillingness to provide public goods, and moreover is incompatible with the existence of generalized social trust, to the extent the latter concept demands adherence to the norm of generalized reciprocity. By its very definition, corruption in the sense of trading of favors in exchange for payment or other services and favoring one set of individuals over another suggests particularized, or encapsulated, trust, and strong out-group discrimination, both incompatible with the broad understanding of generalized social trust. Indeed, given the properties of generalized social trust, including the generalized reciprocity norm, it seems impossible to view social trust and corruption as anything but two mutually exclusive phenomena, The logic of both corruption and social trust indicates that the presence

9780230113176_08_ch07.indd 91

2/5/2011 10:56:06 AM

92

●

Social Trust, Anarchy, and International Conflict

of one should be seen as evidence of absence of the other, with each phenomenon being an indicator of the state weakness and of domestic anarchy. The connection between corruption and state weakness and domestic anarchy is as strong as it is between corruption and social trust. Corruption is generally defined as behavior that deviates from formal duties of a public role in favor of private or personal gain (Friedrich 1966; Heidenheimer 1989; Nye 1967), an abuse of public power for private benefit (Lovel 2005; Harris 2003; Girling 1997; Alam 1989), and not the use of public office and official power for their intended purposes, namely, to serve the interests of the polity (Lazar 2005; Friedrich 1972). Most treatments of corruption suggest that it a problem of political character (Alam 1989), essentially a strong indicator of social instability and decay (Lovell 2005) as well as political degradation (Girling 1997). It is a phenomenon that afflicts weak, poorly governed states (Shore and Haller 2005; Lazar 2005), where the polity is fragmented and lacking cohesion (Williams 2003). Harris (2003), for example, defines corruption as an interstitial activity, the interstices being cracks in the machinery of the state, and indicators of an immature, incoherent state organization incapable of fulfilling its basic function of provision of public goods for the population it governs. Per Myrdal’s (1970) formulation, corruption is a feature of weak, relatively powerless states or the so-called soft states. Furthermore, corruption is intimately tied to the quality of government, not necessarily its size. Some of the least corrupt states also have very large public sectors (Lovell 2005). It is caused by the failings of the political system itself, rather than failings of individuals (Girling 1997), which means that just as corruption is caused by the weakness of the state (Sandholtz and Koetzle 2000), it can only be cured by government efficiency (Goldsmith 1999). Moreover, the effects of corruption are essentially the same as the effects of low levels of generalized social trust. The departures from the established norms of modern bureaucracy, such as rationality and universalism (Alam 1989), lead to the failure to provide public goods, such as information, coordination, and enforcement (Shore and Haller 2005; Girling 1997; Alam 1989; Anderson and Tverdova, 2003; Sandholtz and Koetzle 2000), increasing transaction costs

9780230113176_08_ch07.indd 92

2/5/2011 10:56:06 AM

Research Design

●

93

(Seligson 2002) and creating an atmosphere of heightened uncertainty (Shore and Haller 2005). Unsurprisingly, and wholly consistently with the argument concerning the linkages between state strength and domestic anarchy, corruption, and social trust, the consequences of the absence of efficient governance include arbitrary and unequal distribution of resources, undermining the principle of equality, and generally creating a system where private gain supersedes public interest, weakening trust and confidence in public institutions (Newell and Bull 2003). All of the above strongly suggest that states that suffer from high levels of corruption are partly anarchic systems where, in the absence of government-provided public goods, a strong “self-help” ethic predominates. It is therefore unsurprising that corruption is also strongly associated with the absence of generalized social trust. Indeed, in addition to works that link corruption with a decline in trust in political system (Bull and Newell 2003; Girling 1997; Anderson and Tverdova 2003; Seligson 2002; Elliott 1997; Zinn 2005), there is also evidence that corruption is incompatible with interpersonal trust as well (Seligson 2002). Moreover, the exclusionarytype relationships that characterize networks of corruption (Shore and Haller 2005; Elliott 1997; Herzfeld 1992) are strongly reminiscent of particularized trust, or trust among a small group of individuals who know each other well and may have an interest in behaving in trustworthy manner, rather than trust of the generalized variety, where individuals are willing to assume others are trustworthy even in the absence of evidence. Thus instead of striving toward the ideal of public service benefiting the general public through provision of public goods, a corrupt state’s institutions will operate to the benefit of individual officials, their close families and/or private cliques (Nye 1967; Azfar et al. 2001). Indeed, high-corruption states are characterized as being dominated by closed and exclusionary networks of clientelism and patronage (Herzfeld 1992, 1997; Harris 2003; Shore and Haller 2005), where relationships must continuously be maintained through face-to-face contact and interaction in some form (Warburton 2001). Moreover, and contrary to the notion of generalized trust, corruption implies a relationship based on quid pro quo reciprocity (Lovell 2005; Harris 2003), where the general expectation is that nothing can be obtained without first paying a bribe or rendering some other type of favor.

9780230113176_08_ch07.indd 93

2/5/2011 10:56:07 AM

94

●

Social Trust, Anarchy, and International Conflict

One of the unfortunate key features of corruption is its robustness and durability. This characteristic is also consistent with its relationship with generalized social trust and state weakness/domestic anarchy. Putnam (2000), for example, recognizes the existence of so-called virtuous and vicious circles, where societies with high levels of trust are able to maintain that trust through a feedback mechanism, or, conversely, are unable to break the cycle of low trust and poor government performance. The evidence for the existence of a similar feedback loop can also be seen in the case of corruption, to the point that it is frequently seen as an ineradicable cultural characteristic, endemic to certain societies (Shore and Haller 2005). Once corruption passes a certain threshold it becomes extremely difficult to eradicate (Newell and Bull 2003; Harris 2003; Sandholtz and Koetzle 2000). This is due to its selfsustaining mechanism, not unlike the socialization mechanism for the generation of social trust described in the preceding chapter, as individuals who receive their political socialization in a corrupt society learn that corruption is the unavoidable way of life. Therefore while the problem of corruption is not unsolvable, the effects of anticorruption measures (for example, democratization, transparency, better governance) may become apparent only in the very long term, in some cases after as long as over half a century (Seligson 2002), or until a new generation of individuals socialized into a noncorrupt way of life have come of age and entered the political life of the country. Literature on corruption also makes it apparent there is little possibility of endogeneity, inasmuch as it indicates that it is a problem of domestic rather than foreign origin. Instead, there is considerable evidence to the fact in the opposite direction: states that frequently fight wars (and, in particular, states that do poorly in wars) find themselves under pressure to enact domestic reforms in order to improve the quality of governance and thus military performance in the next conflict, consistent with the argument laid out in Chapter 5 that weak, anarchic states suffering from low levels of generalized social trust are unlikely to perform well when subjected to the strains of war. The examples of states motivated by military defeats to improve domestic governance include Prussia in the aftermath of the Napoleonic wars, France following the 1871 defeat by Prussia

9780230113176_08_ch07.indd 94

2/5/2011 10:56:07 AM

Research Design

●

95

(Weber 1976), and Russia after the defeats in the Crimean War and Russo-Japanese War (Hosking 1997). Corruption, moreover, is arguably a better indicator of generalized social trust than World Values Survey (WVS) questions pertaining to interpersonal trust. The key problem with using WVS data to operationalize generalized social trust is that the questions themselves are not worded sufficiently precisely to differentiate generalized and particularized trust. In large part this is due to the relatively recent awareness of the bifurcated nature of the trust phenomenon. Since the WVS questions were formulated before the now-routinely made distinction between generalized and particularized, or bridging and bonding, trust, naturally these questions fail to capture that distinction. Consider, for example, the WVS question V66 that evaluates the perception of trustworthiness of others confronts the respondent with having to choose between two statements: “Most individuals are trustworthy” and “You can’t be too careful.” The problem with that formulation, apart from the binary nature of the question, is that it does not specify the population pool the respondent is supposed to identify as inherently trustworthy or untrustworthy. Since it is not clear whether “most individuals” refers to one’s family, neighbors, ethnic group, region, country, or the entire planet, the respondent in effect is allowed to define that aspect of the question himself/herself, which in turn renders the question an unreliable tool for measuring social trust. Since one cannot infer from the answer what the respondent’s frame of reference is, it is equally impossible to ascertain whether the answer “most individuals are trustworthy” in fact represents generalized or particularized trust. For example, would an individual who’s a member of a closely knit community in a heavily fragmented society (one that is characterized by low levels of generalized trust) and who, as is likely the case, regularly interacts only with other members of that closely knit group, not respond that “most individuals (within the group with which the respondent has the most contact) “can be trusted”? To use another example, would an inhabitant of a poorly governed state that lacks even the most basic infrastructure, who has never had the opportunity to even visit other parts of the state, consider the entire state the proper frame of reference for this question? One is reminded here of Eugen Weber’s (1976) description of the evolution of a genuinely national French identity that was made

9780230113176_08_ch07.indd 95

2/5/2011 10:56:07 AM

96

●

Social Trust, Anarchy, and International Conflict

possible by the extensive governmental efforts to bind France’s provinces to one another and to create not merely opportunities but also incentives for its citizens to leave one’s villages and valleys and see the state of which they were citizens. Thus in the absence of a more precisely worded question, the type of trust the respondent is affirming could not confidently be categorized as generalized. The formulation of that question becomes even more problematic after one considers that the properties of particularized, or encapsulated, trust do not preclude an individual from responding in the affirmative to the WVS question V66 while at the same time projecting untrustworthiness onto strangers, as postulated by the “alien encounter” scenario discussed earlier. The key difference between particularized and generalized trust is the norm of generalized reciprocity, which is absent from the definition of particularized trust. Instead, particularized trust is present in situations where there is an incentive the other individual has to behave in a trustworthy manner, or history of lengthy interaction between the two individuals. However, such history of interaction does not imply trusting individuals outside that small circle of closely interacting individuals; indeed, the “bad civil society” critiques of Putnam’s argument that associations build social trust center focus precisely on that point and suggest that particularized and generalized trust may be mutually exclusive phenomena in the same manner that corruption and generalized social trust are. In other words, individuals who “trust” only in very specific contexts are in actuality low trusters (per generalized trust definition) who will consider others untrustworthy except in the presence of a clear incentive to behave otherwise. However, the vagueness of the WVS question V66 allows for the possibility that the responders are referring to those contextualized interactions that may represent the majority of interpersonal interactions of these individuals, consistent with the critiques of social capital approach asserting that willingness to interact with strangers requires a preexisting high level of generalized social trust. Moreover, there is abundant evidence that in cases of extreme hostility there exists the propensity to “dehumanize” members of the out-group, with the possible consequence that a question that makes a reference to “most people” will be interpreted in such a way as to exclude the members of the out-group from the category of “most people.”

9780230113176_08_ch07.indd 96

2/5/2011 10:56:07 AM

Research Design

●

97

In addition, CPI data is preferable to WVS data for studies at this level of analysis due to aggregation issues and its superior coverage of the developing world. Furthermore, the nature of WVS surveys (telephone interviews) and the variation of social trust from individual to individual indicates the need for a relatively sophisticated method of aggregating individual WVS responses into country-level generalized trust values in order to assure that the sample for each country is representative of the country’s population as a whole or, if not, the surveys are appropriately weighted. Spatial and Temporal Span

The temporal span of the study is 1990–2001, essentially the period between the end of the Cold War and the 9/11 attacks. This time period is inherently attractive due to the absence of superpower competition and absence of system-changing major international events. However, the time span does include the first Gulf War following the Iraqi invasion of Kuwait in 1990 and the U.S. reactions to 9/11, events that would have surely skewed the results had they been excluded from the study. Therefore, in the interests of avoiding selection bias by focusing on eras that were relatively peaceful, the events of these two years are included. The choice of time period for study is further dictated by the availability of data, particularly for the independent variable. Here too the CPI data proved preferable to WVS data, as the cumbersome process of aggregation and the wide temporal spacing of the individual WVS waves would have added considerable complexity and, worse, opacity to the research design, and thus put the ultimate results under question. In contrast, the simplicity and comprehensiveness of CPI data meant it was readily usable in a quantitative model, with little manipulation required. Unit of Analysis

In this respect, the project follows the path charted by Holsti (1995), whose topic and scope are most closely related to the hypotheses outlined above. In his effort to determine whether state strength and/or maturity influence international conflict behavior, Holsti used, as his dependent variable, the number of conflicts per region

9780230113176_08_ch07.indd 97

2/5/2011 10:56:07 AM

98

●

Social Trust, Anarchy, and International Conflict

over the span of 50 years (1945–1995), with the unit of analysis being both region and state, an approach that has been mirrored by Collier and Hoeffler (1998, 2002, 2004) who have used 5-year increments instead of Holsti’s 50-year block. I have opted for a refined version of Holsti’s research design, by using a 12-year block (1990–2001), focusing on individual states within directed dyads, and incorporating an event-count model that allows for the incorporation of a variety of control variables commonly used in international conflict research. Therefore this project uses “state” as the unit of analysis for domestic hypotheses and “directed state pair” for international hypotheses due to the inherently interactive nature of international conflict. The use of “state” and “directed state pair” instead of “state-year” or “directed state pair-year” is also consistent with the nature of the independent variable and the available data. While CPI data is available in countryyear format, in many cases yearly country corruption values are calculated on the basis of a small number of surveys. As a result, year-to-year variations may be more due to the small sample size than to the actual changes in the country’s domestic situation, particularly since corruption appears to be a resilient and difficult to change phenomenon. As a result, the use of multiyear averages may be more accurate than relying on single-year values (Seligson 2002). Studies using multiyear CPI averages include Kuznetsov and Kuznetsova (2003). Table 7.1 shows the range of states that are the subject of study, their aggregate CPI values, and quartiles in which they place. Even a quick examination of the states thus ranked in accordance with their social trust levels, as operationalized in terms of aggregate CPI values, offers initial support for a variety of propositions advanced earlier in this book. For example, even though democracies are overrepresented in the top, fourth quartile, the presence of a single strong, well governed non-democracy in the form of Singapore is striking. Moreover, democracy by itself is of no guarantee of high placement on the social trust scale, as the emerging post-Soviet democracies and certain developing world democracies, including India, fare relatively poorly in this assessment, a finding that is wholly consistent with democratic peace literature suggesting that emerging democracies do not receive the benefits of democratic peace. Similarly, while several major powers place in the fourth quartile, a number of others, including

9780230113176_08_ch07.indd 98

2/5/2011 10:56:07 AM

Table 7.1 Corruption Perception Index (CPI) aggregate values 1990–2001 State name Denmark Finland New Zealand Iceland Sweden Singapore Canada Netherlands Norway Switzerland Australia United Kingdom Luxembourg Germany Ireland United States Austria Israel Chile France Japan Portugal Belgium Botswana Spain Slovenia Estonia Taiwan Namibia Costa Rica South Africa Trinidad and Tobago Malaysia Tunisia Hungary Malawi Greece Jordan Uruguay Mauritius Czech Republic

CPI aggregate value

CPI quartile

9.67 9.58 9.41 9.24 9.24 9.06 9.05 8.84 8.81 8.68 8.63 8.53 8.47 7.86 7.85 7.64 7.59 7.30 7.11 6.70 6.59 6.41 6.18 6.12 6.05 6.58 5.66 5.52 5.42 5.26 5.13 5.10 5.09 5.06 4.97 4.85 4.69 4.62 4.61 4.57 4.55

4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 3 3 3 3 3 3 3 3 3 3 3 3 (Continued )

9780230113176_08_ch07.indd 99

2/5/2011 10:56:07 AM

Table 7.1 Continued State name Poland Italy Lithuania Peru Mongolia South Korea Belarus Morocco Jamaica El Salvador Slovakia Sri Lanka Mali Brazil Turkey Croatia Bulgaria Ghana Argentina Zimbabwe Panama Latvia Mexico Dominican Republic Macedonia Egypt Senegal Thailand Zambia China Colombia Romania Burkina Faso Philippines Guatemala Mozambique India Ivory Coast Nicaragua Venezuela Moldova Kazakhstan

9780230113176_08_ch07.indd 100

CPI aggregate value

CPI quartile

4.52 4.51 4.38 4.30 4.30 4.29 4.05 4.05 3.87 3.72 3.70 3.70 3.68 3.65 3.56 3.53 3.52 3.48 3.41 3.38 3.35 3.32 3.31 3.30 3.30 3.26 3.24 3.12 3.12 3.06 3.01 3.01 3.00 2.96 2.92 2.85 2.77 2.70 2.68 2.60 2.59 2.58

3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1

2/5/2011 10:56:08 AM

Research Design

101

●

Table 7.1 Continued State name Vietnam Bolivia Armenia Uzbekistan Albania Russian Federation Ecuador Georgia Ukraine Tanzania Pakistan Honduras Uganda Haiti Kyrgyzstan Cambodia Yugoslavia Kenya Indonesia Azerbaijan Paraguay Cameroon Angola Madagascar Nigeria Bangladesh

CPI aggregate value

CPI quartile

2.57 2.52 2.50 2.45 2.40 2.39 2.36 2.35 2.28 2.24 2.23 2.22 2.22 2.20 2.20 2.20 2.10 2.10 2.06 1.80 1.73 1.73 1.70 1.70 1.39 0.80

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

India, China, and Russia, are shown to have relatively low levels of social trust and place in lower quartiles. Finally, examining the lower quartiles reveals that these quartiles are populated almost exclusively by fairly poorly governed states that suffer from a variety of internal political problems and in some cases teeter on the brink of state failure. Domestic Conflict Hypothesis Domestic Dependent Variable

The dependent variable, namely, the incidence of domestic violent political conflict, is operationalized in two alternative ways. The first,

9780230113176_08_ch07.indd 101

2/5/2011 10:56:08 AM

102

●

Social Trust, Anarchy, and International Conflict

and the more conventional one, is using the PRIO Uppsala Conflict Data (Sorli et al. 2005), a frequently used measure of domestic conflict (see, for example, Caprioli 2005). In this instance, the variable represents the number of outbreaks of violent domestic conflicts within each state’s borders during the temporal span of the study. The second, and less conventional, operationalization is the number of minorities at risk (MARs) within a given state. Although this variable is often used as a control variable in domestic conflict studies (Caprioli 2005), here it is also used as a dependent variable, as the number of MARs should not necessarily be considered an exogenous variable. The logic of the theory laid out earlier in this project suggests that anarchic states with low levels of generalized social trust are characterized by societal fragmentation. The existence of a large number of MARs within a state should therefore be interpreted as an indicator of latent political conflict, even if it does not necessarily translate into actual violence. The fact that a minority is characterized as being “at risk” suggests that the possibility of violence exists between politically mobilized ethnic factions. Domestic Control Variables

The control variables used here follow the pattern established by domestic conflict literature and include the following: GDP growth rate: This is a variable that is frequently used in literature as a proxy for regime popularity and, as such, ought to be a reliable predictor of internal conflict. The theoretical expectations here are quite straightforward. Given that high growth rates ought to translate into high regime popularity, a state experiencing a period of strong economic growth ought to be characterized by domestic tranquility, with few outbreaks of domestic conflict. Conversely, since economic recessions are likely to contribute to regime downfall, increased frequency of domestic political violence is a likely consequence (Sorli et al. 2005; Collier and Hoeffler 2004; Fearon and Laitin 2002; Caprioli 2005; Hegre and Sambanis 2006; Hegre 2004). Data for this variable is obtained from the World Development Indicators (WDI) (World Bank 2005). Natural logarithm of population size, another frequently used control variable (Sorli et al. 2005; De Rouen and Sobek 2004;

9780230113176_08_ch07.indd 102

2/5/2011 10:56:08 AM

Research Design

●

103

Besancon 2005; Hegre and Sambanis 2005; Hafner-Burton 2005), is used to account for the likelihood that the greater the population of the state, the greater the frequency of outbreaks of domestic conflict. Data for this variable is also provided by WDI (World Bank 2005). Regime type, in the form of Polity score (Democracy score minus Autocracy score) is used to account for two sets of possibilities. The first possibility is that democratic regimes are less likely to experience domestic conflict, as the existence of democratic organizations provides a venue for peaceful conflict resolution and, moreover, democracy is frequently associated with the existence of peaceful conflict resolution norms. In addition, a squared Polity measure is intended to account for a second possibility, namely, that it is not regime type but rather degree of regime consolidation that is responsible for presence or absence of conflict. Whereas consolidated democracies are likely to avoid domestic conflict due to their strong institutions and well entrenched peaceful conflict resolution norms, consolidated autocracies may be able to accomplish the same goal through ruthless and capable security forces capable of deterring the opposition (Sorli et al. 2005; De Rouen and Sobek 2004; Esty et al. 1998; Gurr 2000; Hegre and Sambanis 2005; Hafner-Burton 2005). Source: Polity IIId data, generated by EUGene software (Bennett and Stam 2000). The level of militarization of society (Sambanis 2000; Hegre and Sambanis 2005), operationalized here as the percentage of GDP dedicated to military expenditure, is used to more directly address state capabilities to influence the level of political violence. Here, too, there are two sets of expectations. On the one hand, one must entertain the possibility that the more militarized states are better able to deter political violence from erupting. On the other hand, the very presence of a robust security apparatus may by itself represent a strong incentive to use it. As a result, heavily militarized states may instead display increased levels of domestic violence. The index of capabilities, a measure frequently used in international conflict, is not used here due to the inapplicability of that measure to domestic conflict. Index of capabilities includes such factors as population, geographical area of the state, and iron production, which, while pertinent to interstate conflict, are not as relevant in situations involving domestic conflict. The source used for this variable is likewise WDI (World Bank 2005).

9780230113176_08_ch07.indd 103

2/5/2011 10:56:08 AM

104

●

Social Trust, Anarchy, and International Conflict

Finally, domestic conflict variables also include the number of MARs within a state (Sorli et al. 2005, Caprioli 2005). The presence of MARs implies the existence of latent conflict within the society, with the greater the number of MARs, the greater the expected number of domestic conflict outbreaks. Source: Minorities at Risk Project (2005). International Conflict Hypotheses International Dependent Variable

The dependent variable selected for this hypothesis is the number of occurrences of MID initiation and MID targeting during the temporal span of the study, using Maoz’s (2005) State pair MID Dataset, generated using EUGene software (Bennett and Stam 2000). Here it should be noted that both fatal and nonfatal disputes are used, since the goal is to ascertain state actors’ propensity to view the use of force or even threats of force as an appropriate means of international discourse. Even a nonfatal MID, given the possibility it might escalate, should be seen as a reflection of deep underlying beliefs on the appropriateness of the use of force. Furthermore, the inclusion of such seemingly insignificant MIDs makes for a much more exacting test of the theory by keeping the threshold for inclusion of armed conflicts extremely low. Moreover, the difference between fatal and nonfatal MIDs may simply amount to nothing more than principal-agent problem (for example, accidental exchanges of fire contrary to orders issued by the respective national command authorities) or other factors facilitating the escalation of conflict that are both external to the theory being tested and at the same time difficult to control. Since the theory being advanced in this book focuses on the political aspects of conflict and the motivations for the resort to force and threats thereof, factors governing the escalation of violence once that path is embarked upon are naturally outside of its scope. Moreover, I would like to note that one of the weaknesses of the democratic peace theory is that the findings of virtual absence of wars (conflicts with considerably higher levels of violence than militarized interstate disputes since, per the Correlates of War standard, a conflict only becomes a war once the threshold of 1,000 fatalities is reached) in democratic state pairs, the same does not hold true if the

9780230113176_08_ch07.indd 104

2/5/2011 10:56:08 AM

Research Design

●

105

threshold of violence is lowered to include militarized interstate disputes. Indeed, some critiques of the democratic peace theory have argued that the finding of state pair peace may be more due to the sheer rarity of wars (particularly when using the relatively threshold employed by Correlates of War) rather than to any inherent pacifying effects of democracy. However, if the ideational argument of the democratic peace theory is to hold true, namely, that democracies unerringly perceive each other as being friendly, the pacifying effect of joint democracy ought to be detectible even in low threshold of violence conflicts. International Control Variables

Quantitative models for the international conflict initiation and targeting hypotheses use the standard array of control variables commonly found in international conflict studies. The list of control variables is as follows: Contiguity of states (DeRouen and Goldfinch 2005; Oneal and Tir 2006) is used to account for the inherently greater probability that states that share a common border have more opportunities to engage in conflict and therefore are likely to be more conflict-prone than states not sharing a border. For example, contiguous states are more likely to have outstanding territorial disputes, resource conflicts, and other points of friction that may give rise to armed conflict. This variable is generated using EUGene software (Bennett and Stam 2000), with the highest level of contiguity selected for the time period being studied. Joint democracy (DeRouen and Goldfinch 2005; Pickering and Kisangani 2005; Oneal and Tir 2006) is the standard measure of effects of democratic peace, and therefore it is included to account for the possibility that the effects attributed to anarchy and social trust are instead the result of democratic peace. The democratic peace theory posits that absence of armed conflict between states shared democratic values and/or institutions, as posited by the democratic peace theory. This variable is coded in the standard manner, with the state pair coded as 1 when both states in the state pair have a Polity value of 6 or higher, denoting a mature, well established democracy. Source: EUGene software (Bennett and Stam 2000).

9780230113176_08_ch07.indd 105

2/5/2011 10:56:08 AM

106

●

Social Trust, Anarchy, and International Conflict

Economic growth (Oneal and Tir 2006; Pickering and Kisangani 2005), operationalized in terms of per capita GDP growth, is used for two reasons. The first is that, in the absence of reliable data on regime popularity, which is unavailable for most nondemocratic states, economic growth is frequently used as a proxy for that variable in, for example, diversionary war literature. Consequently, in accordance with the diversionary war theory, states suffering from low or negative economic growth have an incentive to engage in diversionary violence and therefore ought to exhibit an increased propensity to initiate MIDs. Conversely, and in accordance with a number of critiques of the diversionary war theory, negative economic growth reduces a state’s ability to wage wars, and therefore ought to translate into both a reduced propensity to initiate international conflicts and into an increased vulnerability to aggression. Data for this variable is obtained from WDI (World Bank 2005). The capabilities ratio (DeRouen and Goldfinch 2005; Pickering and Kisangani 2005; Oneal and Ti, 2006; Bremer 1992) is operationalized using the composite indicator of national capabilities from the Correlates of War dataset generated by EUGene software (Bennett and Stam 2000). Here, too, the reason for including the variable is manifold. Whereas some expect conflicts to occur when there is a strong preponderance of power, others believe armed conflict is most likely to occur in conditions of parity. The incorporation of this variable allows one to account for both of these scenarios. Major power status (Rasler and Thompson 2006; Oneal and Tir 2006) is another commonly used encountered control variable in international conflict literature, due to the numerous findings suggesting that major powers are likely to be engaged in more international conflicts than other powers. Moreover, given the use of politically relevant directed state pairs as the level of analysis in the international conflict hypotheses, and the assumption that a major power will be in a politically relevant state pair with every other state due to its greater capabilities and force projection abilities, it is necessary to control for this increased scope for interaction. Major Power Status data comes from the Correlates of War dataset, as generated by EUGene software (Bennett and Stam 2000). Alliance portfolio (Oneal and Ti 2006) is operationalized using Correlates of War data unweighted global s-score generated by

9780230113176_08_ch07.indd 106

2/5/2011 10:56:09 AM

Research Design

●

107

EUGene software (Bennett and Stam 2000). This variable is included to test the proposition that peaceful relations between states may not be due to social trust levels, but rather because of close alliance ties within the state pair. Capital-to-capital distance is the final control variable (Stinnett et al. 2002). Even with the level of analysis being politically relevant directed state pair, it is still necessary to account for the fact that while some state pairs may actually share a border, others may be separated by hundreds of miles of ocean and, in the case of state pairs involving major powers, by thousands of miles. This variance is likely to have a considerable effect on conflict initiation, given that the greater the distance separating the states, the lower the level of force a state can bring to bear against its target, likely resulting in the discounting of the military option as either unfeasible or unlikely to be successful. Data for this variable is drawn from Correlates of War, as generated by EUGene software (Bennett and Stam 2000). Analysis Method

As noted earlier, the models used are negative binomial regression and Poisson regression, event-count models that represent an accepted method for investigating conflict processes (see, for example, Maoz 2000 and 2001). The choice of event-count models also logically flows from the theory laid out in this book and its corresponding hypotheses. Since the theory predicts that states with low levels of generalized social trust will experience heightened rates of both domestic and international conflict, an event-count model that evaluates the rates at which events (in this case, outbreaks of domestic conflict on domestic level and militarized interstate disputes on the international level) occur over a period of time is an appropriate method of evaluating these hypotheses. Moreover, in order to improve the accessibility of the results, the models were implemented in such a way as to produce incidence rate ratios (termed “factors” in the results tables in the next chapter), so that a result less than 1.0 indicates an inverse relationship between the independent or control variable and the dependent variable, while a result greater than 1.0 indicates a positive relationship.

9780230113176_08_ch07.indd 107

2/5/2011 10:56:09 AM

9780230113176_08_ch07.indd 108

2/5/2011 10:56:09 AM

CHAPTER 8

Results Domestic Conflict Hypothesis

The results shown in Table 8.1 are consistent with the hypothesis that the greater the level of generalized social trust within the country, as a result of effective amelioration of anarchic, self-help features of the system, the lower the level of domestic conflict.

Table 8.1

Incidence of domestic political conflict

Variable Social trust Military expenditure as share of GDP GDP growth GDP per capita, constant dollars Number of minorities at risk Polity score Polity score squared Natural log of population

Factor (Standard Error) .707* (.115) 1.650** (.256) 1.120 (.115) .999 (.478) 1.243 (.174) 1.070 (.059) .981* .0095735 1.844** (.467)

N ⫽ 114 states; LR chi-squared (7) ⫽ 37.35; *** indicates p ⬍ .001; ** indicates p ⬍ .01;

* indicates p ⬍ .05. Note : Dependent Variable: Number of Domestic Conflict Outbreaks, Using PRIO-Uppsala Conflict Data. Model: Negative Binomial Regression

9780230113176_09_ch08.indd 109

2/5/2011 10:56:33 AM

110

●

Social Trust, Anarchy, and International Conflict

The results are as follows: the incidence rate ratio, or factor, for social trust of .707 suggests that each unit increase in the level of social trust reduces, after controlling for other variables, the likelihood a state will experience an outbreak of domestic violence during the period being studied by a factor of approximately .7, as predicted by the theory. To be sure, the variable only narrowly clears the threshold of statistical significance, but it clears it nevertheless. Remarkably, regime type represented by the polity score appears to be both statistically and substantively insignificant, with a low level of statistical significance accompanying a factor not far removed from 1, indicating a near-total lack of effect on domestic conflict incidence. Regime type represented by polity score squared to reflect not so much regime type but strength does better from the perspective of statistical significance, though its substantive effect remains rather paltry, with a factor of only .981. It is notable, however, that here too state strength has the effect of reducing the incidence of domestic conflict. After all, the more mature and established regime is, the more likely it is to be an effective provider of public goods. Of the control variables, the natural logarithm of population also performs as predicted, with a factor of 1.844 suggesting a strong relationship between population size and the frequency of outbreaks of domestic violence. This variable, moreover, enjoys a relatively high level of statistical significance. A seemingly counterintuitive result is the apparently positive relationship between the level of military spending (as percentage of GDP) and domestic conflict, with a factor of 1.65 indicating that each percentage of national GDP spent on national defense corresponds to an increase in the incidence of outbreaks of domestic violence by a factor of 1.65. However, this finding is not nearly as surprising as it appears at the first glance. First of all, one should keep in mind that military expenditures are, as literature on militarism and militarization suggests, an endogenous, rather than an exogenous, variable. Governments do not spend large sums of money on national security unless there is a perception of threat. Such a perception is more likely to exist in anarchic, self-help, low-trust states. The positive relationship in this instance furthermore indicates that an increased degree of militarization is not an effective method of dealing with domestic conflict. Of the other control variables, the insignificance of the

9780230113176_09_ch08.indd 110

2/5/2011 10:56:33 AM

Results

●

111

number of MARs as an indicator of domestic conflict suggests that the number of MARs itself may not be an exogenous variable, a proposition that will be tested shortly. Finally, it cannot be emphasized strongly enough that every percent point of GDP spent on national security is a percent point of GDP not spent on roads, schools, effective governmental institutions, and other methods of creating high levels of generalized social trust. The negative relationship between defense spending and domestic violence therefore raises the possibility that defense spending may indirectly contribute to the decline in the levels of generalized social trust, a hitherto unexplored possible explanation for the decline of social trust in the United States in recent decades. Finally, the insignificance of the rate of economic growth is likewise unsurprising, given that social trust has been demonstrated as a prerequisite for economic growth (Putnam 2000; Uslaner 2003; Fukuyama 1995). While the per capita GDP factor is, at .999, so close to 1 as to be substantively utterly insignificant, it is worthy of note that the rate of economic growth’s factor of 1.12 points to the possibility of economic growth actually increasing the likelihood of incidence of violent conflicts within the borders of the state. In view of the relationship between social trust and inequality, the growth of inequality in the wake of economic growth in the economically liberalized economies in the 1990s is the likely reason for this result. Finally, although the use of the number of MARs as a control variable implies its exogeneity, the theory elaborated above suggests that even ethnicity, which is sometimes treated as a primordial trait, is likely to be related to the level of anarchy within a state’s borders leading to the formation of communities closely bound by particularized trust. Indeed, the results shown in Table 8.2 bear out this prediction. The more anarchic, self-help the state, the greater the number of MARs within its borders, reflecting the fragmented, inherently conflictual nature of social relationships characterized by strong in-group preference and out-group discrimination. Of the control variables, remarkably and unsurprisingly only the natural log of population shows strong results, with a factor of 1.44 indicating that the larger the population of a state, the greater the number of MARs to be found within its borders. However, the results do not indicate that

9780230113176_09_ch08.indd 111

2/5/2011 10:56:33 AM

112

●

Table 8.2

Social Trust, Anarchy, and International Conflict Presence of minorities at risk (MARs) as a function of social trust

Variable

Factor (Standard Error)

Social trust Military expenditure as share of GDP GDP growth GDP per capita, constant dollars Polity score Polity score squared Natural log of population

.892* (.045) 1.08 (.065) .941 (.031) 1 (4.10e-08.996) 1.04 (.016) .997 (.003) 1.44*** (.082)

N ⫽ 114 states; LR chi-squared (7) ⫽ 56.57; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of MARs. Model: Negative Binomial Regression.

domestic political regime, level of militarization, and economic performance have an influence on the number of MARs. The factors of these variables are all very close to 1, indicating a near-absence of their substantive impact, and moreover their statistical significance is also very low. While these variables may have an impact on the actual number of outbreaks of political domestic violence, they do not influence the fundamental level of society’s fragmentation. However, social trust factor of .892 indicates that states with elevated levels of social trust are less likely to have MARs living within their borders, a factor that moreover enjoys statistical significance, albeit at the minimum level. Therefore, the results suggest that social trust is responsible both for the latent level of conflict within a society and for an increase in the number of domestic acts of violence. International Hypotheses—Conflict Initiation

The first model, presented in Table 8.3, in this section is actually a test of the democratic peace alternative, without the social trust variable, in order to test the viability of the model. If joint democracy

9780230113176_09_ch08.indd 112

2/5/2011 10:56:33 AM

Results

●

113

Table 8.3 Democratic peace test at state pair (dyadic) level Variable Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (initiator) GDP per capita growth (initiator) Capital-to-capital distance

Factor .520*** (.082) .997 (.0018) .744 (1.148) 2.265*** (.402) 1.116** (.036) .999*** (.00006)

N ⫽ 1865 directed dyads; LR chi-squared (6) ⫽ 136.85; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of MID Initiations. Model: Poisson regression

shows the expected pacifying effects on the state pair, this will provide a benchmark against which later results can be compared. The results are wholly consistent with the democratic peace theory, since joint democracy appears to reduce MID initiations within the state pair by nearly half, after controlling for the other variables. The shared democracy variable, which shows that both states within the directed dyad are consolidated democracies, enjoys a factor of .520 and with a high degree of statistical significance in the Poisson model. It should be noted, however, that the results are not very robust as they become statistically insignificant (although only narrowly so) in the negative binomial regression model (results not reported). The results also show that international major powers are 2.2 times more likely to initiate militarized interstate disputes than other states, with that factor similarly enjoying a high level of statistical significance. Capital-to-capital distance, expressed in miles that accounts for the factor being so close to one, since it represents the drop in the rate at which militarized interstate disputes will be initiated within the dyad for every additional mile separating the two capitals, is also strongly significant and lower than 1, confirming the hypothesis that the more distant the two states, the less likely they are to come into conflict.

9780230113176_09_ch08.indd 113

2/5/2011 10:56:33 AM

114

●

Social Trust, Anarchy, and International Conflict

Rather more interestingly, the GDP per capita growth’s fairly significant factor of 1.16 suggests that states enjoying high rates of economic growth are more likely to initiate MIDs than states with lower rates of growth. This finding is contrary to the predictions of the diversionary war theory; however, it is consistent with many of its critiques. Therefore the preliminary model not incorporating social trust as a variable yields results broadly comparable with those in contemporary international conflict literature. The next model, shown in Table 8.4, introduces the social trust variable, at first in state pair form. The joint trust independent variable for this hypothesis was generated by coding all state pairs in which both states belong to the highest social trust category, namely, the fourth quartile, with 1, and all others with a 0. Once the joint trust variable is introduced into the model, the results are quite stark. With a factor of .0811 indicating a very strong substantive impact on the number of conflict initiations and a very high level of statistical significance, social trust shows itself to have considerable potential for explaining the onset of international conflicts. Table 8.4 Conflict initiation test with social trust test as a joint attribute Variable Joint trust Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (initiator) GDP per capita growth (initiator) Capital-to-capital distance

Factor .0811*** (.047) .939 (.224) .994* (.002) .710 (.660) 3.135*** (.902) 1.111* (.050) .999*** (.00008)

N ⫽ 1865 directed dyads; LR chi-squared ⫽ 144.43; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05 Note : Dependent Variable: Number of MID Initiations. Model: Negative Binomial Regression

9780230113176_09_ch08.indd 114

2/5/2011 10:56:34 AM

Results

●

115

This is all the more so since the joint democracy variable, which had both a sizable substantive impact and a high level of statistical significance in the model that did not incorporate a measure of social trust, is rendered both statistically and substantively insignificant in the current model. The insignificance of joint democracy leads to the conclusion that the pacifying effects attributed to shared democracy may instead be attributable to generalized social trust, particularly since the effect of joint trust appears to be quite strong. Whereas in the previous model joint democracy accounted for a 50 percent drop in MID initiations within a state pair, joint trust accounts for a drop of over 90 percent, or an order of magnitude change. Moreover, whereas joint trust retains a high level of statistical significance, joint democracy becomes insignificant. Of control variables, the strongest effects are to be found in major power status with a strongly statistically significant factor of 3.135 that actually represents an increase over the earlier model not incorporating the social trust variable. This factor indicates that major powers are, controlling for other variables, more than three times more likely to engage in military interstate dispute initiation than other states. Capital-to-capital distance likewise has strong statistical significance, with the number of MIDs decreasing with the increase in distance separating the states within a state pair. That result remains unchanged from the previous model. Economic growth per capita, with a weakly significant factor of 1.111, has a slight effect of increasing the likelihood of conflict, a result that, as noted earlier, runs counter to the predictions of diversionary war theory inasmuch as it predicts that states that suffer from bad economic performance should be more likely to initiate conflicts as their leaders attempt to shore up their flagging political fortunes. The capabilities ratio variable, while enjoying a minimal level of statistical significance, nevertheless has a low level of substantive impact on the rate at which MIDs are initiated within the dyad. Finally, alliance portfolio does not appear to have a statistically significant effect on conflict although the factor itself suggests that it reduces the incidence of conflict. The weakness of this variable may be attributable to the possibility that the existence of alliances, like the absence of conflict, is the result of shared high level of social trust that promotes cooperation. This is a question that will be addressed in future research.

9780230113176_09_ch08.indd 115

2/5/2011 10:56:34 AM

116

●

Social Trust, Anarchy, and International Conflict

The increase in the standard error of joint democracy from Table 8.3 to Table 8.4 indicates a possible problem with multicollinearity in the model between trust and regime type. However, the level of correlation between trust and joint democracy is only .3, limiting the likelihood of multicollinearity. It should also be noted that multicollinearity is an issue that violates no regression assumptions and that moreover has no statistical solution (Achen 1982). Moreover, the joint democracy variable also shows a significant drop in its marginal effect from Table 8.3 to Table 8.4, with its factor approaching 1. This suggests that the increase in joint democracy’s standard error may in fact be less due to multicollinearity than to its lack of effect independent of social trust. However, if there is a relationship between democracy and social trust, the causal arrow appears to run less from democracy to trust, but rather from trust to democracy (Uslaner 2002; 2003; Letki 2004; Almond and Verba 1963; Inglehart 1997; Aberbach and Walker 1970). Thus the model presented in Table 8.4 suggests that dyads consisting of states enjoying high levels of social trust will experience dramatically lower incidence of militarized interstate disputes between the states, with the impact of generalized social trust dwarfing that of joint democracy. As intriguing as that result is, it is not an adequate test of the theory advanced in this book, which posits that states with high trust levels will be both less likely to be MID initiators and targets, regardless of the level of social trust of the other state in the dyad. Therefore to further ascertain the detailed effects of social trust as a state, not dyadic, attribute on the incidence of MIDs within the dyad, the social trust independent variable is recoded a categorical variable, corresponding to the quartiles shown in Table 7.1. To briefly recap, the first quartile represents states with the lowest levels of generalized social trust (corresponding to lowest CPI values), and the fourth quartile represents states with the highest social trust levels. The results hold up on state level as well. When examining the potential initiators within the state pair (the first state listed within each directed state pair), it is clear from the results displayed in Table 8.6 that the initiator’s level of social trust has a strong and independent effect on reducing the number of MIDs within the state pair. The factor of .454, with a very high level of statistical significance, indicates that the frequency at which a state will initiate MIDs will

9780230113176_09_ch08.indd 116

2/5/2011 10:56:34 AM

Results

●

117

drop by nearly half as it progresses from a lower quartile to a higher quartile. Other variables perform in the same manner as in the state pair hypothesis, with the exception of the alliance portfolio variable, which, although weak statistically, shows a strong effect at reducing the incidence of conflict. The explanatory power of the social trust variable is even more remarkable when considering that democracy as an attribute of individual states does not yield similarly strong results. To further explore the effects of social trust as a property of individual states and tease out the effects the membership in each individual quartile has on the state’s rate of MID initiations, the model depicted in Table 8.6 is largely similar to that from Table 8.5 except that the categorical social trust variable is now recoded into four dummy variables, each corresponding to a specific quartile. The lowest social trust category (the first quartile) is left out of the model to serve as the baseline against which to compare the other quartiles. Thus the model in Table 8.6 represents the most comprehensive MID initiation test of the social trust theory of international conflict initiation. Table 8.5 Social trust as a state attribute test Variable Trust quartile (initiator) Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (initiator) GDP per capita growth (initiator) Capital-to-capital distance

Factor .454*** (.047) 1.039 (.241) .994* (.0027) .139* (.129) 3.759*** (1.029) 1.184*** (.053) .999*** (.00009)

N ⫽ 1865 directed dyads; LR chi-squared(7) ⫽ 176.86; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of MID Initiations. Model: Negative Binomial Regression

9780230113176_09_ch08.indd 117

2/5/2011 10:56:34 AM

118

●

Social Trust, Anarchy, and International Conflict

Table 8.6 Conflict initiation test with social trust quartiles as dummy variables Variable Trust (initiator) second quartile Trust (initiator) third quartile Trust (initiator) fourth quartile Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (initiator) GDP per capita growth (initiator) Capital-to-capital distance

Factor .452** (.133) .259*** (.091) .0858*** (.0286) 1.045 ( .242) .994* (.0027) .144* (.135) 4.107*** (1.244) 1.182828*** (.0556) .999*** (.00008)

N ⫽ 1865 directed dyads; LR chi-squared (9): 177.54; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of MID Initiations. Model: Negative Binomial Regression

The results show that high-trust states initiate MIDs at a rate over 90 percent lower than low-trust states, or an order shift in magnitude in the rate of conflict initiation, ceteris paribus. The factor associated with the fourth quartile with the value of .0858 and very high level of statistical significance indicates that states, controlling for all other factors, in the fourth quartile initiate militarized interstate disputes at a rate approximately 12 times lower than states in the first quartile. The third quartile factor of .259, also highly significant statistically, indicates that third quartile states initiate MIDs at a rate only a quarter as high as that of the first quartile states. And finally, the third quartile factor of .452, also with a fairly high level of significance, indicates that third quartile states initiate MIDs at only half the rate of the states in the fourth quartile. These very strong results, both in the substantive and statistical sense, stand in contrast with the statistically insignificant factor of 1.045 associated with joint

9780230113176_09_ch08.indd 118

2/5/2011 10:56:34 AM

Results

●

119

democracy, which suggests virtually no substantive impact of regime type on MID initiation after controlling for social trust levels. Thus, overall the results show that every social trust category shows a substantively and statistically significant pacifying effect when compared to the baseline bottom quartile. The strong across-the-board effect generalized social trust has on conflict initiation removes the possibility that the result is due to outliers or other statistical anomalies. One is moreover struck by the large differences between the quartiles when it comes to the effects of social trust. Thus, whereas the secondlowest social trust category initiates MIDs at a rate 55 percent lower than the bottom social trust category, the value for the second-highest category is approximately 25 percent, and for the top category a mere 8 percent. Nevertheless, the degree to which generalized trust affects MID initiation is striking, particularly when one stops to consider that low-trust states, from the rational perspective, should be the least likely states to court the prospect of war. A partially anarchic, fragmented state, with low levels of generalized social trust and plagued by corruption is precisely the type of state that is the least likely to do well in actual conflict. The fragmentation, corruption, and ineffectiveness that is plaguing its government is likely to extend to its armed forces as well, with predictable results. The performance of other models depicted in the Table 8.6 model is also of note. The capital-to-capital distance, with its strongly significant factor of .999 continues to perform as expected. The capabilities ratio, with a weakly significant factor of .991, likewise continues to suggest the relative capabilities of states within the dyad exercise relatively little influence on the decisions to employ military threats of force or actual force. The alliance s-score factor of .114, weakly significant, indicates that alliance ties are associated with a reduction of conflict. However, the major power status factor of 4.107 with a very high level of statistical significance, suggests that major powers are more than four times more likely to initiate militarized interstate disputes than other states, and the GDP per capita growth of a state, with a strongly significant factor of 1.18, similarly shows an association between the rate of economic growth and a heightened propensity to initiate MIDs. Social trust therefore turns out to be an important predictor of international conflict initiator. This leaves only one remaining

9780230113176_09_ch08.indd 119

2/5/2011 10:56:34 AM

120

●

Social Trust, Anarchy, and International Conflict

hypothesis to be tested, namely, whether the level of social trust within a state also influences the rate at which it will become a target of MIDs initiated by other states. International Conflict Targeting

To briefly recap, the international conflict targeting hypothesis tests the contention that high-trust states are less likely to become victims of MID initiation, due to their better ability to avoid antagonizing would-be challengers due to their less confrontational domestic policies and less belligerent modes of interaction by its leaders. The model shown in Table 8.7 is identical to the one used to validate Hypothesis 2, with the exception of the dependent variable. In this case, the dependent variable is the number of MID initiations against the state in question, also from Zeev Maoz’s (2005) Dyadic MID Dataset version 2.0. The negative binomial regression model suggests that capitalto-capital distance and the target’s social trust level influence whether a Table 8.7 Conflict targeting test Variable Trust quartile (target) Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (target) GDP per capita growth (target) Capital-to-capital distance

Factor .687*** (.072) .905 (.214) .999 (.0018) .314 (.279) 1.127 (.3142) 1.057 (.0476) .999*** (.00008)

N ⫽ 1865 directed dyads; LR chi-squared (7) ⫽ 113.19; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of times state is MID target. Model: Negative Binomial Regression

9780230113176_09_ch08.indd 120

2/5/2011 10:56:34 AM

Results

●

121

state is likely to become an MID target. The capital-to-capital distance factor of .999 entirely consistent with the earlier findings that the more distant the state is from potential attackers, the less likely it is to become an MID target. Also as predicted, generalized social trust of the target state has a negative effect on MID targetin.g, with a social trust quartile factor of .687 showing that states in higher quartiles become progressively less likely to become targets of militarized interstate disputes. Remarkably, these are the only two variables to have any statistical significance in the model. Of particular interest is the insignificance of the major power variable. Whereas major power status of the initiator was a robust predictor of MID initiation, that same status in a prospective MID target appears to be of no importance. Similarly, the alliance score, the rate of economic growth, joint democracy, and capabilities score are statistically insignificant when it comes to predicting whether a state is likely to become a target of a militarized interstate dispute. The juxtaposition of these two results leads to a number of conclusions. First, it does appear that the pacifying effect of social trust level of the target state is more likely due to signaling effects rather than any deterrent effect that might be expected from the cohesive nature of a strong, well governed state. The insignificance of such “objective” indicators of state power as capabilities ratio or major power status suggests that potential MID initiators are not dissuaded by the power of the prospective target, consistent with the notion that individuals who have powerful incentives to engage in diversionary incentives are operating under a form of motivated bias that makes them oblivious to the power of their adversary as did, for example, Argentina’s leadership when it made the fateful decision to invade the Falklands (Lebow 1985). Since these objective measures of power are also more easily observable than something as intangible as social trust, it seems unlikely the effect of social trust is due to anything other than nonbelligerent signaling effects. Second, the relative weakness of target variables reinforces the conclusion tentatively drawn in the analysis of results of conflict initiation hypothesis to the effect that domestic characteristics of the MID initiator are of more significance than the characteristics of the target. Such a conclusion is entirely consistent with the diversionary theory of war, as well as with the

9780230113176_09_ch08.indd 121

2/5/2011 10:56:35 AM

122

●

Social Trust, Anarchy, and International Conflict

neoliberal argument that state preferences are the product of domestic political processes. The fact that target social trust has a weaker substantive influence on conflict initiation than initiator social trust also dovetails nicely with the discussion of the heuristic qualities of social trust implying that in the absence of knowledge about the nature and intentions of one’s counterpart, the gap in knowledge will be filled by projecting one’s own attitudes onto the target. When broken down into categorical variables (results shown in Table 8.8), the social trust variable continues to have an effect, although both the substantive and statistical significance appear to be somewhat diminished. In substantive terms, while states in the second-lowest social trust category are the targets of only about half as many MID initiations (ceteris paribus), and those in the uppermost category are targeted at a rate two-thirds lower than the baseline (with a factor of .339), there is no discernible difference between the Table 8.8 Conflict targeting test with social trust quartiles as dummy variables Variable Trust (target) second quartile Trust (target) third quartile Trust (target) fourth quartile Joint democracy Capabilities ratio Global unweighted alliance s-score Major power status (target) GDP per capita growth (target) Capital-to-capital distance

Factor .505* (.155) .337** (.119) .339** (.112) .914 (.216) .999 (.0017) .372 (.334) .983 (.292) 1.071 (.050) .999*** (.00008)

N ⫽ 1865 directed dyads; LR chi-squared (9) ⫽ 115.18; *** indicates p ⬍ .001; ** indicates p ⬍ .01; * indicates p ⬍ .05. Note : Dependent Variable: Number of times state is MID target. Model: Negative Binomial Regression

9780230113176_09_ch08.indd 122

2/5/2011 10:56:35 AM

Results

●

123

two topmost categories, which is a major difference from the results on MID initiation. However, overall the results from the Table 8.7 and Table 8.8 models support the main hypothesis by showing that high-trust states are indeed significantly less likely to be victims of MID initiations. This finding is entirely consistent with findings by Parks et al. (1996) who found that low-trust actors in prisoner’s dilemma games are much more receptive to other actors’ signaling of intent to defect than to cooperate. Moreover, the relative weakness (compared to the strength of findings in support of H2) of these results suggests that attributes of international conflict targets are a rather weak predictor of international conflict, consistent with Moravcsik’s (1997) argument that the sources of state preferences are to be found in domestic politics. What is particularly striking is that while major power status and economic growth are strong predictors of conflict initiation, they have no impact as deterrents against others, a finding that has implications for the deterrence theory. Summary of the Results

Overall, generalized social trust (operationalized as perception of corruption) performed in accordance with theoretical expectations. To be sure, the results concerning domestic conflict are relatively weak, and here additional testing may prove necessary. However, the international conflict hypotheses receive very strong and statistically robust support. The high-trust states appear to be less likely to initiate militarized interstate disputes than low-trust states by an entire order of magnitude, after controlling for other factors. Their high levels of generalized social trust generated by efficient domestic existence appear to exert a strong pacifying effect on their international behavior, as predicted by the theory presented earlier in the book. What is more, their pacific demeanor also contributes to such states being less likely to become the targets of aggression evidently due to their ability to signal nonbelligerent intent. Conversely, the anarchic lowtrust states both initiate conflicts at greatly increased rates due to their inherent mistrust of strangers or peoples unlike themselves and possibly due to the leaders of such states perceiving incentives to engage in diversionary behavior, hoping (possibly in vain, in cases of

9780230113176_09_ch08.indd 123

2/5/2011 10:56:35 AM

124

●

Social Trust, Anarchy, and International Conflict

very weak states with highly fragmented polities) that international conflict will at least temporarily diminish the internal differences and perhaps even lead to the establishment of national identity. The lowtrust states’ appearance of belligerence, caused by the comparatively violent, unstable nature of their domestic politics and mistrustful behavior of their leaders, moreover, appears to be a strong factor in explaining initiations of militarized interstate disputes against such states. Low-trust states therefore appear to be plausible enough threats to present tempting targets for leaders of other states seeking suitable diversion targets by initiating militarized interstate disputes against them, whereas high-trust states lack that quality. The findings also have rather surprising implications for the deterrence theory. Since decisions to initiate MIDs are not influenced by capabilities or major power status of the opposing party, but only by “signaling” of friendly or hostile intent, the results are more consistent with the “slippery slope” critique of deterrence theory. Sending a “strong message” to a low-trust state might easily have the counterproductive effect of communicating hostile intent, validating the recipient’s views of the nature of international system and making conflict more, rather than less, likely. Instead, the results suggest that a conciliatory message would not be interpreted as a sign of weakness but rather as a sign of nonhostile intent, making conflict less likely. The variability in perception of other states also has implications for the power transition theory due to its assumption that major powers automatically view other major powers as likely threats, are sensitive to possibility of losing relative position, and therefore may be tempted to launch preventive military action against a would-be challenger. The differing reactions of various major powers to the growth of potential rivals in the decades prior to the outbreak of World War I resulting in acquiescence or indifference in the case of high-trust states and fear and hostility on the part of low-trust states suggest that the theory has predictive powers only in regard to states with low levels of social trust, which in turn increases their sensitivity to relative gains concerns. However, power transitions involving states with high levels of social trust, and consequently low sensitivity to relative gains, are unlikely to create international tension. However, social trust is clearly not the only variable exerting significant influence on the propensity to be an initiator or a target of

9780230113176_09_ch08.indd 124

2/5/2011 10:56:35 AM

Results

●

125

a militarized interstate dispute. It is complemented as a cause of international conflict by a number of factors reflecting the material capabilities of the state in question. Specifically, both major power status and the rate of economic growth have a significant impact on the state’s propensity to initiate international conflicts. These results should not be necessarily seen as mutually contradictory. In conjunction, the importance of perceptions of others encapsulated by social trust levels and of material capabilities reflects Most and Starr’s (1986) “opportunity and willingness” decision making framework, with generalized trust representing willingness, and major power status and economic growth representing opportunity to initiate militarized interstate disputes. While low level of social trust may lead to national decision makers who believe in the inherent malevolence of other actors (as well as masses who share that worldview) and in the appropriateness and the utility of the use of force, the capabilities of the state in question clearly matter as well. MIDs are most likely to occur when such willingness is accompanied by opportunity in the form of the ability to successfully challenge other states militarily. The combination of results suggests that the most likely MID initiator is an anarchic, low-trust major power experiencing a period of significant economic growth, in other words a state that combines an inherent suspicion and mistrust of others (possibly bordering on xenophobia) with capabilities to confidently give that mistrust a concrete expression in the form of a military challenge. By the same token, the ideal “profile” of an MID target is an anarchic, self-help state with a low level of generalized social trust that is geographically located in a close proximity to (or otherwise forming a politically relevant state pair with) an anarchic, self-help major power experiencing a period of economic growth. Returning to the example of the Spanish-American War that represents an exception to the democratic peace theory’s proposition that democracies, for a variety of institutional and ideational reasons, do not fight wars against one another; the framework just outlined explains the outbreak of that conflict quite well. As noted earlier, the declining Spain was a relatively weak state, its politics characterized by coups d’etat and other forms of internal political violence. What heightened the likelihood of the war against the United States, aside from the U.S. interest in taking over former Spanish colonies as part

9780230113176_09_ch08.indd 125

2/5/2011 10:56:35 AM

126

●

Social Trust, Anarchy, and International Conflict

of the “Manifest Destiny” at the time, were the brutal Spanish repressive measures employed against Cuban insurrectionists fighting against Spanish rule. It is these tactics that made Spain a plausible menace that had to be dealt with by U.S. military force, and the U.S. “yellow journalism” of the era utilized the Spanish atrocities as fodder for its prowar propaganda. However, what made the prowar propaganda effective was that the U.S. public, also living in a relatively weak, partly anarchic self-help state, was quite predisposed to accept the arguments for the war. It is that predisposition of the U.S. public that made the war fever possible, and that resulted in an enthusiastic rush to flesh out a wide variety of volunteer regiments (most notably, the future President Theodore Roosevelt’s “Rough Riders”) that would fight in the war (O’Toole 1984). Moreover, the United States was undergoing a period of very rapid economic expansion that gave U.S. leaders a high level of confidence that their foreign adventures would succeed. A similar chain of events can be discerned in the run-up to the U.S. invasion of Iraq in 2003. Saddam Hussein’s Iraq, a very weak and nearly anarchic state whose domestic politics were characterized by high levels of violence, and which was ruled by a seemingly erratic and irrational ruler (the appearance of irrationality being caused by his efforts to stage various displays of regime strength to overawe opponents), could not help but present an external image of a menacing and threatening state. The U.S. public of the time, however, was also receptive to a prowar policy agenda due to the eroding level of social trust documented by, among others, Putnam (2000). Although the contemporary United States is a far stronger state than it was a century ago, the major state-building effort the U.S. government undertook beginning with President Roosevelt’s New Deal clearly petered out by the early 1970s. The increased receptiveness to the idea of war, manifesting itself by the nostalgia for the unifying effects of World War II, also meant that the antiwar message would hardly gain any traction among the U.S. public, while the prowar case was going to be very easy to make indeed. And, just as in the 1890s, the United States has just completed a major economic expansion driven by the newly emerging computer technologies, expansion that likewise promoted high levels of confidence in the ultimate success of U.S. foreign policy undertakings.

9780230113176_09_ch08.indd 126

2/5/2011 10:56:35 AM

CHAPTER 9

The Outbreak of World War I

A

rguably the best case study to comprehensively illustrate the dynamics of the impact of the combination of the level of generalized social trust, major power status, and economic growth identified as key factors determining the onset of militarized interstate disputes in the quantitative analyses above is the interaction of European major powers leading up to the outbreak of World War I. With the centennial of the outbreak of the First World War rapidly approaching, it is perhaps appropriate to reflect on the causes of that system-altering conflict, all the more so as many of the factors that contributed to its outbreak persist even to this day. Although the quantitative portion of this project dealt with militarized interstate disputes rather than wars, it must be noted that wars are almost invariably preceded by such disputes, which then escalate until they pass the war threshold. Such was the case during the fateful events of 1914. One of these factors is the impact of economic development and interdependence on international conflict. Whereas today, in the era of globalization, there exists considerable optimism that the combination of democracy, universal hospitality, and international institutions (the three components of Kantian peace) have all but banished conflict at least among the states that are well integrated into the international economic system (Jervis 2002), one should not forget that similar predictions were being made prior to 1914. That staggeringly destructive and costly war took place in spite of predictions that such a conflict was extremely unlikely to occur due to the heightened international interdependence caused by the industrial revolution and the expanding international trade fostered by the so-called Pax Britannica, the growing wealth of nations, and the increased costs of major conflict due to a variety of modern technologies (Wolf 2005). To the extent that a similar prescription for peace is being offered today, the failure of

9780230113176_10_ch09.indd 127

2/5/2011 10:57:01 AM

128

●

Social Trust, Anarchy, and International Conflict

optimistic prognoses prior to 1914 should be a cause for concern, not the least due to the painful social disruptions and dislocations that have accompanied both the industrial revolution of the nineteenth and early-twentieth centuries, and the globalization processes of the twentieth and early-twenty-first centuries. Even as the modern states were emerging, their societies were undergoing a painful transition from local, traditional, relatively immobile societies to mass modern ones, with the upshot being considerable political upheaval up to and including regime changes and revolutions. These political upheavals, in the form of revolutionary movements sweeping Europe in the nineteenth century (arguably starting with the French Revolution and ending with the Nazi Machtergreifung of 1933) collectively represented the “growing pains” of the emergence of modern societies and the decay of traditional social structures. It is that fraying of the traditional “conscience collective” that prompted Durkheim and others to begin considering the importance of the moral foundational of society and inventing ways to restore it in the wake of changes wrought by industrialization. We must keep in mind that a similar unraveling is a possible consequence of globalization processes in their current form. A number of experts have already observed that globalization has caused growth in intra- and inter-state inequality, has reduced the states’ ability to provide public goods in a number of important policy areas, and has undermined traditional social structures in many parts of the developing world, likewise creating preconditions for major social upheavals (Scott 2001; Naim 2003). Clearly, in the wake of September 11, 2001, terrorist attacks against the United States whose origins lie in several weak and failing states whose traditional societies have crumbled under the onslaught of globalization, the role of social trust and state strength and capacity as determinants of conflict was as central a century ago as it is today. Furthermore, the outbreak of World War I is an attractive case to study the influence generalized social trust largely because the major powers of Europe have by then been engaged in nearly a centurylong effort at social trust creation within their borders. The “strong” states of Europe were, in 1914, a comparatively recent creation. A mere century earlier, governments of European states could only make a weak claim to exercising full internal sovereignty that was purportedly their due according to the Treaty of Westphalia. The nineteenth

9780230113176_10_ch09.indd 128

2/5/2011 10:57:02 AM

The Outbreak of World War I

●

129

century was to be characterized by a feverish cycle of revolutions and nation-building efforts that resulted in the emergence of modern states that would appear familiar in their shape and function to modern observers. However, while European states in the seventeenth and eighteenth centuries fell far short of the ideal of internal sovereignty or a genuine possession of the monopoly on the internal use of force, they gradually acquired these powers over the course of the nineteenth century. This evolution was driven by both political and technological factors, as only in the late eighteenth and nineteenth centuries did technology permit the rapid transmission of information and goods, allowing states to make good on their claim to internal sovereignty (Schulze 1998). The incentives for such exercises in nation-building as documented, for example, by Eugen Weber in his seminal Peasants into Frenchmen were twofold. The ever-present fear of revolution induced by the disruptive effects of the industrial revolution certainly played a role behind the adoption of policies intended to increase societal cohesion (Lincoln 1983). However, an older and probably even more powerful factor driving these reforms was the growing reforms that states with increased social trust make for formidable opponents in wartime. The wars of the French Revolution and Napoleon’s First Empire confronted the rest of Europe with the vision of a state with a powerful bureaucracy capable of considerable (by the standards of early nineteenth century Europe, at any rate) resource extraction and public goods provision and a population that was less inclined to see their own government as an enemy “out-group” engaged in rapacious and forcible resource extraction, rather than a force that could benefit individual citizens (Broers 1996). It is the evolution of the character of the French state rather than the military innovations of the era that made the French armies such dangerous adversaries during that era (Clausewitz 2004). The implications for military effectiveness of the more extensive social contract of the French Revolution, under which both the individual and the state undertook extensive mutual obligations, were such that no state in Europe could afford to ignore the French example. Karl von Clausewitz’s discussion of the Napoleonic wars included the realization that wars ceased to be the province of sovereigns and their armies but instead they became the business of entire nations (Clausewitz 2004). Consequently, the remainder of

9780230113176_10_ch09.indd 129

2/5/2011 10:57:02 AM

130

●

Social Trust, Anarchy, and International Conflict

the nineteenth century was characterized by a veritable, if largely invisible, arms race among the major powers in which they sought military advantage over one another not only through superior numbers, tactics, and technology, but also by seeking to outdo one another in the effectiveness of their nation-building through increasingly more extensive and elaborate provision of public goods, so as to give the state the capacity to fully mobilize their nation’s human and material resources in a time of war. In this respect, Weber’s (1976) process of turning “peasants into Frenchmen” (or, depending on the country, Germans, Russians, Italians, et cetera) appears to have been a step in the larger process of turning peasants into soldiers. Clearly, a “nation in arms” approach to defense could only be successfully implemented if the bulk of the inhabitants of a state viewed themselves as members of a single nation. The ineffectiveness of ancien regime armies composed of long-term professionals officered by aristocrats when confronted with masses of citizen-soldiers commanded by citizen-officers made the process of nation-building imperative. Moreover, since compulsory universal military service requires the existence of a culture of generalized reciprocity, it is evident that the imposition of such a burden on the population could only be implemented as part of a broader provision of public goods. In contrast to long-term professional soldiers who were paid respectable sums of money for their service (which considerably reduced the size of armies states could field, and rendered them quite casualty-averse, which reduced the likelihood of decisive battlefield victories), draftee/conscript soldiers historically and in virtually all historical milieus were paid what can be termed symbolic sums of money. Therefore widespread acceptance of universal military service can only be explained in terms of generalized reciprocity. There is no direct monetary (or otherwise) reward for service. Instead, one serves the state in the expectation that the state’s existence and security will at some future date be of some as-yet undefined benefit to the individual. Willingness to submit oneself to universal military service also implies willingness to sacrifice on behalf of others (Bessel 1997). Therefore it is no coincidence that the emergence of strong states providing a wide variety of public goods for its inhabitants in a bid to establish a strong national identity and to cement the bonds between the state and society took place just at the same time as the

9780230113176_10_ch09.indd 130

2/5/2011 10:57:02 AM

The Outbreak of World War I

●

131

widespread introduction of the universal military service obligation. Although the concept of the draft, or conscription, was a relatively ancient one, with many states practicing it long before the French Revolution, the draft of that era was far from universal, and moreover included a variety of loopholes, including the ability to, in effect, hire a substitute to perform military service. The objective of the prerevolutionary draft was to create a small corps of long-service soldiers loyal to the regime, cut off from the society at large, and therefore reliable enforcers of domestic order, whereas the late-nineteenthcentury draft systems had as their objective the creation of large forces of reservists after the short (two-three years) term of obligatory service (Rothenberg 2006). If the nineteenth-century literature is any guide, the inhabitants of that era were well aware of the importance of the bond between the state and society, which could only be created by the establishment of an extensive social contract between the leaders and the led, to the state’s military fortunes. To cite but a few examples, while Leo Tolstoy’s epic War and Peace describes the events of Napoleonic era, the book was written in the aftermath of the Russian military defeat in the Crimean War and must be viewed through the lens of the debate that took place following that defeat. Tolstoy’s invocation of the myth of Russian popular resistance, and the warning about the inexorable forces of history remaking the world are a well concealed (given the political implications of Tolstoy’s recommendations, they had to be) call for social reform in Russia, lest it be left behind by other major powers and thus subjected to further defeats. Emile Zola’s great novel of the Franco-Prussian War The Debacle, although sometimes described as a pacifist novel due to its unsparing criticism of the flaws of the French military during that war, likewise represents a call on France to recapture its Napoleonic greatness, with social reforms aimed at exploiting the latent French national spirit, reforms whose implementation is the subject of Weber’s Peasants into Frenchmen (1976), presented as a necessity if France is to prevail militarily the next time it is pitted against the unified Germany. Ernest Renan’s statement that German victories in 1870/71 were to the credit of the “Prussian schoolmaster” is likewise a recognition of the growing importance of the provision of public goods to the effectiveness of national mobilization in time of war.

9780230113176_10_ch09.indd 131

2/5/2011 10:57:02 AM

132

●

Social Trust, Anarchy, and International Conflict

This case is also attractive due to the relatively high unit homogeneity among the five major powers that went to war in 1914. The differences between domestic political regimes of these states were not as pronounced. Whereas during World War II or the Cold War the adversary spectrum ran the gamut from mature and well established democracies to solidly totalitarian states, in 1914 that spectrum was considerably more narrow. Even the democratic states of the era like France and Great Britain still enjoyed only a limited franchise (for example, women at the time were still deprived of the right to vote), while the absolutist Russian Tsar was forced to acquiesce to the creation of State Duma that took over a number of governing functions (Lincoln 1983). The homogeneity also extends to the military aspects of the conflict. Every major power belonged to one of the two major opposing alliance systems whose cumulative capabilities were approximately equal, as evidenced by the prolonged and mostly inconclusive nature of fighting between 1914 and 1918. There was also a universal agreement as to how the future war was going to be fought. The general staff of every major power expected that a future conflict would be short and decided by an all-out offensive that would quickly knock out the adversary, consistent with how European wars of the latenineteenth century, for example, the wars of German unification, were decided. The belief in the ascendancy of the offensive, furthermore, created an incentive to initiate hostilities, rather than to await their initiation by its adversary. It was also a war in which the ultimate outcome was won not so much on the battlefield as on the “home front.” Furthermore, on the eve of war all the major powers’ armies were almost identically equipped and organized, with neither side in the conflict enjoying a pronounced qualitative or quantitative advantage (Haythornthwaite 1992). This, in contrast to World War II, made the hoped-for rapid war of movement followed by a quick and relatively bloodless victory elusive. Consequently, none of the major powers suffered a clear-cut, decisive defeat on the battlefield. Instead the war inevitably turned into one of attrition where success would go to the side that could tolerate the strain the longest. As a result, defeat came only as the combatants gradually succumbed to the stresses of the war, war that tested their national cohesion and the legitimacy of their regimes. Since individual participation in a military

9780230113176_10_ch09.indd 132

2/5/2011 10:57:02 AM

The Outbreak of World War I

●

133

mobilization or a national war effort essentially represents willingness to sacrifice on behalf of others (Bessel 1997), it is evident that the level of generalized social trust that characterizes a society must have had a direct impact on the effectiveness of the war efforts of the combatants. The political and military unit homogeneity therefore allows for controlling for these variables and instead focusing on the role social trust and state capacity played in the war initiation, for here there existed genuine differences among the major European powers. The processes of state- and nation-building did not proceed at the same pace, or even in the same direction, in every European state. The outbreak of World War I represented to a large extent the spillover of the “growing pains” of the emerging modern nation-states of Europe. The tragic irony of 1914 is that the nation-building efforts of the European major powers were sufficiently advanced to turn these states into extremely resilient war machines (to cite but one statistic, each major power lost, on average, approximately 1,000 soldiers killed in action per day, every day, for the duration of that four-year war), yet not sufficiently advanced for their “moral communities” to transcend national borders. Mass warfare was already possible, transnational “in-group” formation has not yet arrived. Thus the Sarajevo assassination of Archduke Ferdinand, the heir to the Austro-Hungarian throne, found the major powers of Europe already on a near-war footing, the result of decades of dedicated preparation. The effectiveness of these preparations would be made evident in the swiftness of mobilization and the rapidity with which armies of hundreds of thousands of troops would be dispatched from their peacetime garrisons into war. However, the events of particular interest are the actions of the major powers in the immediate run-up to the war, specifically their reactions to the assassination. It was probably inevitable that a terrorist act of this magnitude was going to have profound consequences. Yet it was not self-evident at the time that it was going to lead to a general European war. While similar crises and provocations had occurred in the preceding decade (for example, the Agadir Incident, the Balkan Wars, et cetera), all of which led to talk of war, ultimately these crises were defused through diplomatic means (Taylor 1980). The Sarajevo assassination proved to be the exception. This ultimately costly exception was due to the fact that while some of the major powers sought to

9780230113176_10_ch09.indd 133

2/5/2011 10:57:02 AM

134

●

Social Trust, Anarchy, and International Conflict

avoid a war, others appeared determined to seize this opportunity to launch one. The Actors

Before the question of conflict initiation is addressed directly, one must consider which of the major powers fit in the conflict initiator “profile” identified above, namely, a major power with a high level of domestic “self-help” experiencing a period of economic growth. Since all the five states (Germany, Austria-Hungary, France, Great Britain, and Russia) whose actions will be considered here are generally acknowledged to have had major power status, this leaves only economic growth and societal cohesion to be considered. Germany

In terms of satisfying the criterion of a major power undergoing a period of steady economic growth, Germany unquestioningly does so. Thanks to several decades of strong industrial expansion and population growth, both in relative and absolute terms, by 1914 it was arguably the most powerful state on the European continent. While it was inferior to both Russia and the United States in terms of population (Kennedy 1989), in industrial terms it was second only to the United States. Moreover, its growth rate was staggeringly rapid, allowing it to not merely maintain its relative position in the international system but even surpass many of its rivals. Between 1880 and 1913, German per capita industrialization increased by a factor of more than 3, or at a rate comfortably greater than any other major power, and its share of world manufacturing output increased from 8.5 percent to 14.7 percent (Kennedy 1989). In terms of iron and steel production, a major yardstick of military capability for industrial-era armies (and, in fact, still used as a component of composite capability indices), in 1913 Germany was second only to the United States, a considerable improvement over 1890 when it was a distant third to Great Britain. Germany also comfortably maintained its relative position of exceeding Russia’s steel production by a factor of 3 or 4, and significantly increased its relative superiority over France (Kennedy 1989). Whereas Germany was only scarcely ahead of Russia in 1880, in 1913 its industrial production exceeded that

9780230113176_10_ch09.indd 134

2/5/2011 10:57:02 AM

The Outbreak of World War I

●

135

of Russia by a factor of 2 (Kennedy 1989). This factor is of crucial importance, particularly since it was the purported growth of Russian power that German war faction found so troubling (Taylor 1980). Therefore, given the apparent existence of a material capabilitiesbased component of the decision making, the German leadership was in a strong position to confidently contemplate international conflict as a means of improving its international position. Germany’s strength meant that opportunities existed for it to successfully exercise its power. The industrial potential allied to a large and professional military establishment translated into a short-term warmaking capacity second to none in 1914, so much so that in peacetime the German leadership did not even fully exploit its economic potential. Even though the United States would remain unsurpassed as an economic power, it would find it difficult to translate that power into military capability even after it entered the war in 1917 (Haythornthwaite 1992). In contrast to France, for example, Germany’s military leadership did not even perceive the need to call up all eligible recruits. Moreover, Germany’s preeminent military position on the European continent was achieved at a relatively lower cost than that expended by other major powers. In the decade prior to the war, Germany’s total military expenditure represented a proportionally smaller share of its national income than that of most other powers. Whereas Germany allocated only 4.6 percent of its national income to defense, France allocated 4.8 percent, Austria-Hungary 6.1 percent, and Russia 6.3 percent. This indicates that Germany still had considerable spare capacity left to meet the perceived threat of the growing power of Russia, which however was achieved only by accepting a defense burden on its struggling economy nearly 50 percent higher than Germany’s (Taylor 1980). And yet, in spite of the relatively moderate defense spending, the German forces were the best trained and best equipped, both in terms of quality and quantity, in Europe. Even the German reserve forces, alone among all major powers, were equipped and trained nearly as well as the regular forces, so much so that they were deemed capable of taking the field alongside the regular forces, in contrast to other armies who used their reserve divisions and corps for essentially internal security or fortress garrison duties. This extra capability gave the German army a crucial and unexpected advantage in the opening

9780230113176_10_ch09.indd 135

2/5/2011 10:57:03 AM

136

●

Social Trust, Anarchy, and International Conflict

battles of 1914, although clearly not enough to provide the margin of victory (Kennedy 1989). Nevertheless, this rapid growth of German military capability, which, in addition to the world’s most powerful army also included the transformation of the German navy from essentially a coastdefense force that was effectively blockaded by the French during the War of 1870 to a top-rank navy capable of threatening the British naval supremacy, apparently did not breed a sense of security. For all its power, Germany was a remarkably insecure major power, and the growth of its power seemed to correspond with growing feelings of encirclement and menace. Given the growth in Germany’s stature since 1871, one might have expected the emergence of the sense of security based on Germany’s enviable material resources. Yet, in 1914, Germany’s leaders appeared convinced, contrary to objective evidence, that they were losing ground and that time was not on their side (Taylor 1980). The antiwar factions in Germany, after all, viewed Germany’s international position in far more positive (and, arguably, objective) terms, felt that there was no need for war, and did not regard Germany’s encirclement as real, with the German industry winning the battle for the control of continent all by itself (Taylor 1980). The reason for this apparent sense of impending doom that engulfed the German leadership and persuaded it to risk a major power war lay in the still low degree of social trust within the German state. For all its economic success, Germany has not yet succeeded in creating a strong sense of nationhood among its still-diverse population. To be sure, in large part this was due to the historically fragmented nature of peoples that eventually were encompassed by the borders of the nascent German state. Prior to the mid-nineteenth century, close mutual identification of the German nation with the German state was not possible due in large measure to the historical absence of existence of a “German” state. As late as the final decades of the eighteenth century, it was considered self-evident that Germans, on the score of their inherent individualism, were unsuited to living within a single centralized state (Schulze 1998). Partly as a result of these inauspicious beginnings, Germany was not as successful at inventing a national myth as France on the basis of shared national heroes, institutions, and history. In the absence of these devices, and incapable of providing social goods at a sufficiently high level, the German

9780230113176_10_ch09.indd 136

2/5/2011 10:57:03 AM

The Outbreak of World War I

●

137

state had to resort to relying on a secular national enemy against whom German people had to struggle to retain unity, and the concept of conquest or cultural and military supremacy. Therefore, in contrast to France, the German identity was a function of who its enemies were (Hobsbawm 1995), and German nationalism was built chiefly on military unification (White 1996), as Bismarck’s creation of a unified German state was the only national historical experience that all citizens of the new Empire had in common. The Franco-Prussian War, an event of central importance in the founding myth of the German nation, was elevated to mythic status, complete with annual commemorations of the Sedan Day, the date on which Emperor Napoleon III surrendered himself and his encircled army to the combined German forces. The craving for national consensus, the fear of dissolution, was stronger in Germany than in other European countries (Schulze 1998). Thus on the eve of World War I, despite its impressive economic development, Germany remained internally fragmented. Under the veneer of the German unity there persisted strong territorial and denominational differences that were enshrined in the legal system. The continued existence of noble privilege played havoc with the notion of the rule of law (Brose 2001). Germany also lacked the generalized sense of a middle class culture of common sense and generalized practices (Schulze1998). The Imperial German military was the only institution that could claim a truly pan-German character (although here too, as will be shown later, the effects of continued fragmentation of the German polity were keenly felt) and, as a consequence, one of the key functions of the German military was to maintain the existing social and political regime (Deist 1997; Goemans 2000b). This entailed safeguarding the German state against its purported domestic enemies, including Social Democrats, Catholics (fully 40 percent of the German population, in itself a sign of the failure to create a sense of common destiny), Polish nationalists, Jews (who were discriminated against in a number of ways, including the ban on obtaining officer commissions except in cases of converts to Christianity), inhabitants of the newly conquered formerly French provinces of Alsace and Lorraine, and “liberals” (Schulze 1998). Fear of domestic opponents of the ruling regime appears to have been at least as great as the fear of external enemies (Goemans 2000b). Due to

9780230113176_10_ch09.indd 137

2/5/2011 10:57:03 AM

138

●

Social Trust, Anarchy, and International Conflict

the apparent inability to create a strong sense of identity with the German state, the German state was forced to resort to coercive policies, for example, the infamous Kulturkampf, in order to force recalcitrant minorities into line with state policy (Schulze 1998). The experience of German sovereignty over Alsace-Lorraine, in particular, demonstrated the difficulty the Second Reich experienced in establishing an overarching national identity that trumped its regional ones. In the years leading up to the outbreak of World War I and during its early stages, the inhabitants of these two provinces were widely suspected of pro-French sympathies, and throughout the 1871–1914 period the supposedly unified German military engaged in discriminatory practices toward recruits from these provinces (Kramer 1997). Although the experience of the inhabitants of Alsace-Lorraine was arguably the most extreme example of the regional distinctions and peculiarities within the Wilhelmine armed forces, one is struck by the extent to which the German military (with the exception of the navy) retained the corporate identities of pre-Unification German states. The German military’s effort to establish its pan-German identity was severely undermined by the territorial nature of recruitment and other regional distinctions. German army units wore uniforms that retained unique identifying features of the armies of pre-Unification German states. In Bavaria the German army’s regionalism went so far as to permit the maintenance of individually numbered Bavarian army corps. Even the Imperial Guard Corps soldiers, the army’s elite, showpiece and the most certain path to officer advancement, were recruited from Prussian provinces in proximity to Berlin to the exclusion of the rest of Germany. Thus, even though the institution of universal military service offered the possibility of using the military as a veritable “melting pot” from which a genuinely common German national identity might emerge, in practice Prussians served only with Prussians, Bavarians with Bavarians, et cetera, in process actually hampering the development of pan-German national identity (Haythornthwaite 1992). Moreover, the comparatively lower percentage of eligible German draftees actually called up, discussed earlier, was in no small part due to the unwillingness of the German military to provide military training to members of suspect ethnic groups and social classes (Kramer 1997), a clear demonstration of the direct impact of social trust on military effectiveness.

9780230113176_10_ch09.indd 138

2/5/2011 10:57:03 AM

The Outbreak of World War I

●

139

The fragmented nature of the German state came into full view once the hostilities began. The so-called spirit of 1914, as the rally around the flag effect in Germany was described, was not quite as widespread or enthusiastic as it has often been depicted. While antiwar demonstrations were suppressed by government forces, displays of prowar enthusiasm were actively encouraged (Verhey 2000). What prowar enthusiasm there was quickly began to fade once the hardships of the war began to be felt in the form of growing casualty lists and shortages of foodstuffs and other goods. The mobilization effort of the Wilhelmine Germany failed to sustain widespread support for the war. Instead, policies adopted by the government only served to gradually exacerbate the already existing divisions within society. Jews, who already were one of the targets of the German Kulturkampf in the late- nineteenth and early-twentieth century, bore the brunt of the official blame for wartime misfortunes. Jewish scapegoating began as early as 1914 (albeit in an unofficial form, by various nationalist organizations), but in late 1916 the military, smarting from the failure of the Verdun offensive to break the French resistance, joined the effort by launching “investigations” into whether Jews were attempting to evade military service or otherwise undermine the war effort (Hoffman 1997). Therefore the seeds of the “stab in the back” myth exploited by Adolf Hitler and the Nazi Party in the 1920s and the 1930s were sown by the imperial German military during the war. The so-called Burgfrieden, or the German state’s attempt to create a unified political front for the time of war, completely unraveled in the last two years of the war (Chickering 2004; Bessel 1997). In these final years Germany experienced an increase in crime, desertion, draft evasion, black marketeering, food riots, and other indicators of flagging national morale (Bessel 1997). The German military also suffered a gradual decline in morale and, starting in late 1916, the hitherto unheard of problem of desertion began to be a significant issue for the German military leadership (Goemans 2000b). While the German military did not suffer widespread mutinies (with the exception of the naval mutiny in Wilhelmshaven that occurred when the German fleet was ordered to undertake a suicidal sortie against the Royal Navy), by 1918 German army units were experiencing ever-higher rates of desertion and disobedience, with many recorded cases of German units refusing to carry out orders to engage in offensive

9780230113176_10_ch09.indd 139

2/5/2011 10:57:03 AM

140

●

Social Trust, Anarchy, and International Conflict

operations. During the redeployment of German troops from the defeated Russia following the October Revolution and the Treaty of Brest-Litovsk, German units suffered desertion rates of up to 10 percent while in transit to the Western Front (Deist 1997). The unraveling of the German society and military is remarkable in that it actually took place against the backdrop of comparative military success. Unlike their adversaries that were reduced to grimly hanging on to their trench lines and were unable to make significant headway against the strongly fortified German lines prior to the 1918 offensives, the German armed forces could boast of having a number of significant accomplishments to their credit, some of only propagandistic value (for example, the aerial bombardment of London that in the long run decimated the German airship forces and drew resources into the ineffectual heavy bomber force), but others of considerable strategic importance (the eastern front defeats of Serbia, Rumania, and, eventually, the Russian Empire itself ). Moreover, throughout the conflict, the combat operations of the German military took place almost exclusively on the territory of other states, be it allies (Austria-Hungary, the Ottoman Empire) or enemies (France, Belgium, Russia, Serbia, Italy, Rumania). Even at the November 1918 Armistice the trench lines of the ultimately decisive Western Front still ran considerable distance from the pre-1914 German territory (Haythornthwaite 1992). The preceding discussion strongly suggests that the German Empire possessed the combination of confidence-inducing strong economic performance and low levels of social trust that made war a very tempting proposition. Austria-Hungary

In comparison with Germany, Austria-Hungary’s economic performance was considerably less successful. While Austria-Hungary’s population was larger than that of Great Britain and France, its industrial capability remained inferior (Kennedy 1989). Austria-Hungary’s power was eclipsed already in the nineteenth century. Its economic growth can at best be described as mediocre. In the period between 1880 and 1914, Austria-Hungary’s economic performance, in terms of its rate of growth, was inferior to that of Germany, United States, and

9780230113176_10_ch09.indd 140

2/5/2011 10:57:03 AM

The Outbreak of World War I

●

141

Russia, but on a par with that of other major powers. Its share of world manufacturing output actually declined between 1900 and 1913, from 4.7 percent to 4.4 percent (Kennedy 1989). In terms of the creation of cohesive national identity, the domestic affairs of Austria-Hungary were in an even greater disarray. The Austro-Hungarian Empire, a highly diverse collection of nations nursing separatist sentiments of varying intensity, was relatively unsuccessful in producing an overarching national identity, and what loyalty to the state there existed was centered on the person of Emperor Franz Josef whose death in 1916 preceded that of his Empire’s by less than two years (Rothenberg 1976). By 1914 the Austro-Hungarian Empire actually consisted of two states, Austria (Cisleithania) and Hungary (Transleithania), each with a separate administration, and sharing only an army, an emperor, and foreign policy/national security apparatus. Even the three ministries the two halves of the Empire had in common, namely, Foreign Affairs, War, and Finance, had limited ability to manage the affairs within Austria or Hungary. For example, the Ministry of War controlled only the Common Army. Empire’s territorial forces, namely, the Austrian Landwehr and the Hungarian Honved, each had their separate administrations and identity, in spite of organizational and doctrinal similarities. Likewise the Finance ministry presided only over joint expenditures, not the internal finances of either Austria or Hungary (Haythornthwaite 1992). Unsurprisingly, Austria-Hungary suffered from major differences in economic development, governance among its regions, differential treatment of various ethnic groups, and consequently, from the absence of national cultural cohesion (Kennedy 1989). Although Austria became the second state in Europe, after France, to introduce conscription, this did little to cement its unity (Schulze 1998) in large part due to the fact the conscription did not amount to a universal obligation to serve. Furthermore, since the weakness of the AustroHungarian state meant the provision of public education to inculcate at least the knowledge of a common language (in this case, German and/or Hungarian) meant that, like in Germany, Austro-Hungarian conscripts served with other conscripts speaking the same language, a policy that also prevented the army from serving as a melting pot from which a shared, civic national identity might have emerged (Rothenberg 2006).

9780230113176_10_ch09.indd 141

2/5/2011 10:57:03 AM

142

●

Social Trust, Anarchy, and International Conflict

The loyalty of many, mainly Slavic, ethnic groups within the Empire was highly suspect even before the war’s outbreak. The onset of the war swiftly laid bare the underlying fissures of the Austro-Hungarian state, whose military performance as a result was quite mediocre. The poor unit cohesion led to a number of cases of mass surrenders to the Russian Army, so much so that by the end of the war the Russian military formed an entire Czech army corps consisting wholly of former Austro-Hungarian prisoners of war of Czech nationality. The Austro-Hungarian military might have collapsed even faster than it did historically if it were not for the extensive aid (to the point of virtually taking over full control of the Austro-Hungarian military) rendered by Germany (Cornwall 1997). Under these conditions, the Ausgleich, or the Austro-Hungarian attempt to maintain political unity for the duration of the war, proved an even more short-lived failure, with component ethnic groups of the Empire preferring to seek independent statehood (Rothenberg 1976). Therefore, while Austria-Hungary’s economic performance was considerably weaker than that of Germany, its level of national cohesion and social trust was at a desperately low level. Thus while AustriaHungary had every incentive to initiate an international conflict, it lacked the means to prosecute it, and that paucity of means likely gave its leaders a pause. Unfortunately, Germany’s unhesitating backing of Austro-Hungarian military adventurism in the Balkans following the Sarajevo assassination provided the confidence its leaders would have likely otherwise lacked. France

French economic performance during the pre-1914 period was comparatively weak, in fact among the weakest when compared to other major powers. France’s slow economic growth meant that France was slowly losing its relative position, falling further and further behind Germany. In terms of population, it has already fallen behind all other major European powers and its industrial growth was insufficient to prevent it from losing relative position in the global economy, in terms of percentage share of global output. France’s relative share of world manufacturing declined from 7.8 percent to 6.1 percent between 1880 and 1913 (Kennedy 1989).

9780230113176_10_ch09.indd 142

2/5/2011 10:57:04 AM

The Outbreak of World War I

●

143

In spite of its uninspiring economic performance, France was quite successful in establishing a strong sense of national identity. In large measure the French program of nation-building was a direct consequence of the defeat in the Franco-Prussian War, which revealed the weakness of not only the French army but also of its national institutions and its ability to rally the French masses to the defense of the state. The rather low sense of national identity in France prior to 1870 very directly impacted on its ability to wage war against a better mobilized Prussian state, a fact that was recognized by Ernest Renan, Emile Zola, and other political commentators of the era. The French Army’s effectiveness in that conflict was very directly impacted by the need to maintain its reliability as an internal security force. In order to prevent close bonds from forming between the troops and local inhabitants, which might render them unreliable when called upon to quell civil unrest, French regiments were frequently shuffled from one garrison to another. This, however, played havoc with French mobilization plans, as regiments found themselves far from their projected assembly areas. Although conscription existed, in contrast to the Prussian model of universal military service the French state could not impose a similar system on its citizens, and moreover, substitution was permitted (Holmes 1984). In view of how the state-society gap contributed to the French defeat at the hands of the emerging German state, closing that gap became a key state priority in the decades following the war (Weber 1976), an endeavor that was largely successful. French political and social reforms in the Third Republic have done a great deal to close the gap between the state and society that existed under the Second Empire (Porch 1981; Audoin 1997), which consequently influenced the outcome of the battlefield clashes against the German army in 1914. These reforms, which are described in considerable detail by Weber (1976) and which entailed a vast expansion in the scope and scale of public goods provision, included a self-referential national mythology appropriate for a nation whose identity, relying as it did on allegiance to a strong, capable state, did not require a reference to an external “other” (Hobsbawm 1995) as a guarantee of internal cohesion. The outcome of these reforms included a heavily reformed and vastly more capable French military based on a system of universal military service (Porch 1981).

9780230113176_10_ch09.indd 143

2/5/2011 10:57:04 AM

144

●

Social Trust, Anarchy, and International Conflict

However, the best testimony to the effectiveness of the French nation-building project was France’s vastly improved ability to sustain the burden of the war. The French government managed to maintain a high level of popular support for war was maintained even among the more radical elements of the society. Even though at times the fortunes of war seemed to disfavor France, particularly during 1917 that was characterized by bloody and failed offensive operations, a near-mutiny of several French Army divisions that refused to follow senseless orders, and Russia’s de facto surrender to Germany that significantly affected the overall balance of power among the two alliances, ultimately its ability to tolerate the strain of war proved superior to that of Germany (Goemans 2000b). The French Union Sacree fared much better at preserving political and societal unity in the face of war, better than comparable efforts in Germany or Austria-Hungary (Smith 1997; Robert 1997). Indeed, throughout the war, France was not in danger of overthrow from either right or left, and government enjoyed a high degree of legitimacy (Goemans 2000b). French leaders did not operate under the expectation that defeat meant their ouster or even death. Whatever political opposition remained was focused predominantly to opposition to the war itself, rather than opposition to the regime. Similarly, French labor unrest during 1916 and 1917 did not represent a major threat to the French war effort as the strikers’ demands were limited n nature and did not entail far-reaching changes to national goals or strategy. The domestic challenges to the French government caused by the hardships of the war were not of the same magnitude as the challenges to German, Austro-Hungarian, or Russian regimes (Goemans 2000b). The effectiveness of the French nation-building efforts and their ability to create a strong allegiance to the French state are also reflected in their successful efforts to utilize native colonial troops in frontline combat roles. Hailing from places as diverse as Algeria, Morocco, Senegal, and Indochina, these native troops proved a wholly reliable component of the French military machine, and the French army leadership experienced considerably fewer disciplinerelated problems associated with its native colonial troops than, for example, Austro-Hungary experienced when trying to maintain the loyalty of its ethnic minority soldiers (Haythornthwaite 1992).

9780230113176_10_ch09.indd 144

2/5/2011 10:57:04 AM

The Outbreak of World War I

●

145

Great Britain

Great Britain’s economic growth in the prewar period was also unimpressive. Britain, once the world’s premier industrial power, has relinquished that position and slipped to the third place and increasingly found itself under economic pressure from newly ascendant world powers. Its relative decline was particularly precipitous, as its share of world manufacturing output declined from the dominant 22.9 percent in 1880 to mere 13.6 percent in 1913 (Kennedy 1989). However, in Great Britain, as in France, there existed a similar consensus on the ideas of nation, state, and the essential nature of “Britishness” (Schulze 1998). As a result, the war, rather than becoming a challenge to the legitimacy of the state, led to a series of reforms on such issues as suffrage, franchise, and social benefits through which many of the grievances of lower classes were addressed. This occurred even though the war was a major shock for the British society, unaccustomed as it were to waging land campaigns of such a magnitude, and at the cost of hundreds of thousands of casualties (Goemans 2000b). That such measures were possible is indicative of a society that did not view politics as a zero-sum game where any concession was perceived as enhancing the position of one’s political enemies rather than addressing grievances and thus resolving the conflict. Redistribution of political power was not equated with a permanent loss of power. Apart from the Irish question, the British government did not fear domestic enemies, and the political rise of the working class was not viewed in the same apocalyptic terms as it was in Germany, for example (Goemans 2000b). Britain’s ability to sustain its war effort is all the more remarkable in that, like France, it sustained a number of significant military failures along the way, including the disastrous amphibious assault on the Dardanelles, failed offensive operations against Ottoman forces in Mesopotamia, the bloody and inconclusive Battle of the Somme, German aerial bombardment of London, and an increasingly effective German submarine blockade (Horne 1997). Moreover, like France, the British Empire was able to draw on the highly diverse native populations from its many far-flung colonies and rely on them to provide reliable and effective military manpower that bore much of the burden of the fighting on various fronts of

9780230113176_10_ch09.indd 145

2/5/2011 10:57:04 AM

146

●

Social Trust, Anarchy, and International Conflict

the war (Haythornthwaite 1985). Once again, such an ability to construct effective fighting organizations out of highly diverse populations testifies to the strength of the British state and the high level of social trust it was capable of generating. Russia

In economic terms, Russia’s performance in the pre–World War I period was mixed. Its industrial sector grew between 1880 and 1914 at an impressive annual rate of 5 percent–8 percent, although these gains were largely offset by the stagnation of its stagnating agricultural sector that even in 1914 still comprised the lion’s share of Russia’s economic activity. Thus, overall, Russia’s share of global manufacturing increased only slightly between 1880 and 1914, and its real national product grew at a rate of only 1 percent per year, less than other major powers. However, during the decade immediately prior to World War I Russia actually declined in relative terms, its share of global manufacturing output slipping from 8.8 percent to 8.2 percent (Kennedy 1989). Of the five major powers that went to war in 1914, Russia was arguably the most anarchic. In terms of success at creating a common national identity, Russia fared poorly. Here, the vast territories of the Russian Empire that were difficult to reach even during the age of railways, the low population density that implied high costs of creating national infrastructure, and the high level of population diversity meant that even a strong, effectively governed state would have been heavily challenged to produce high levels of social trust under these circumstances. Alas, the slowly decaying Russian Empire of the latenineteenth and early-twentieth centuries was not that state. The low level of social goods provision combined with the arbitrariness and corruption of the imperial rule meant that civic nationalism was not a potent force binding the empire, all the more so, given the diversity of the empire’s population, Russian rulers rightly feared the emergence of ethnic nationalism on the grounds such nationalism would ultimately unravel the empire. Consequently, prevailing sources of identity, particularly in the rural areas, were still local and particular, not general (Hosking 1997; Lincoln 1983). The scale of domestic violence resembled that described by Weber (1976) as prevailing in

9780230113176_10_ch09.indd 146

2/5/2011 10:57:04 AM

The Outbreak of World War I

●

147

France prior to the reforms instituted by the Third Republic in the aftermath of the Franco-Prussian War. Russia was the scene of continued unrest and even outright sporadic warfare. In 1902 alone, there were 365 incidents in which troops had to be used to quell rural uprisings (Kennedy 1989). In 1913, 100,000 Russians were arrested for attacking state authority. In 1909, military was used in over 100,000 incidents (Kennedy 1989). After 1905, the Tsarist regime lived in constant fear of another revolution (Goemans 2000b). Unsurprisingly, the Russian war effort failed to live up to its expected “steamroller” (Taylor 1980) effect. While the Russian army performed well against Austro-Hungarian forces, it suffered an almost unbroken string of defeats by German forces. The internal weakness and division of the Russian state became quickly evident under the strains of war, as evidenced by the inability to carry out an effective industrial mobilization for the war or to sustain public support for the regime and its war aims. As a result, Russia began to falter politically in 1917, first with the abdication of the Tsar in February and, in October, with a revolution that swept into power a Bolshevik regime dedicated to pulling Russia out of the war. It accomplished that goal with the Treaty of Brest-Litovsk. Conflict Initiation

Following the assassination of the Archduke Ferdinand in Sarajevo, it is evident that while Austria-Hungary desired a limited conflict against Serbia that preferably would not involve other major powers, its more powerful ally Germany seized this opportunity to push for a broader war. Austria-Hungary’s fear of impending collapse due to growing forces of Slavic nationalisms, particularly in the Balkans, appears to have created an incentive for conflict in the hopes that a sound defeat of Serbia would extend the life of the empire and stifle the proindependence aspirations of increasingly restive minorities within the empire. Therefore, following the assassination of Archduke Ferdinand in Sarajevo, it was almost inevitable that left to its own devices, Austria-Hungary would attack Serbia. Even though Serbia satisfied nearly all conditions set forth in Austria-Hungary’s ultimatum, these concessions failed to deflect Austria-Hungary from its determined path to war (Taylor 1980). To be sure, even the most prowar

9780230113176_10_ch09.indd 147

2/5/2011 10:57:04 AM

148

●

Social Trust, Anarchy, and International Conflict

Austrian decision makers had misgivings. The chief of the Austrian General Staff, Conrad von Hoetzendorf, while an advocate of a war against Serbia, feared a larger war. While he saw a war against Serbia alone as necessary to Austria-Hungary’s survival, in his estimate a war that involved Russia and/or other major powers would lead to the demise of the Habsburg dynasty and break-up of the empire (Rothenberg 1976). However, Austria-Hungary was not left to its own devices. Here, the low material strength of the Austro-Hungarian state clearly proved a brake on the aggressive designs of its leadership. Instead, it was all but prodded toward war by its main ally Germany even as it became clear that the war, if it came, was highly unlikely to remain a limited regional conflict Austria-Hungary desired. Germany essentially gave Austria-Hungary a free hand in dealing with Serbia and assuaged Austria-Hungary’s concerns of Russian involvement by promising that it would work to defuse the Russian threat to come to Serbia’s defense (Taylor 1980). The Austrian order to launch general mobilization on August 31 was issued at German prompting, before the news of Russian mobilization orders reached Vienna. Similarly, the German declaration of war against Russia occurred even before Russia had an opportunity to declare war against Austria-Hungary. Germany declared war on Russia, thus assuring a wider conflict, with only Russian mobilization orders as provocation (Taylor 1980). The German concerns about the Balkans dated to at least the crisis of 1912, when the German General Staff chief, Helmut von Moltke the Younger, advised Kaiser Wilhelm II to launch a war against Russia in order to discourage it from supporting alleged Serbian aspirations (Haythornthwaite 1992). Indeed, it is difficult to underestimate the extent to which the promise of the backing by the powerful German state, combined with the implicit threats to withdraw that support and undermine the alliance with Germany, influenced Austro-Hungary’s actions following the Sarajevo assassination. Germany’s interaction with France and Belgium in 1914 also contains evidence of Germany being singularly uninterested in preventing a larger conflagration. Rather, the manner and timing of German ultimata to these two countries suggests these diplomatic maneuvers were treated as a mere formality before the inevitable war. The German ultimatum to Belgium was issued even before the Austrian declaration

9780230113176_10_ch09.indd 148

2/5/2011 10:57:04 AM

The Outbreak of World War I

●

149

of war on Serbia (Taylor 1980). Germany’s request for a French promise of neutrality was couched in terms virtually guaranteed to provoke a French rejection. As insurance, Germany requested that France turn over to its military the French border fortresses of Toul and Verdun (the latter the site of a titanic battle in 1916, which demonstrated the importance of the fortress to France’s security) (Taylor 1980). When that request was predictably rejected, Germany declared war on France on basis of false claims of violation of German territory (Taylor 1980), in spite of a French pullback from the German border precisely in order to prevent misunderstandings (Doughty 2005). German efforts to prevent British entry into war were also unconvincing. The German government merely promised not to annex any French territory (though not colonies), nor Belgian territory as long as Belgium did not resist Germany (Taylor 1980). Here, the willingness of the German leadership to contemplate an international conflict combined with the confidence imbued by the vast economic power of the German state, proved a lethal combination, a sufficient condition for the outbreak of the war. In contrast, the Entente powers appeared to be considerably less eager to initiate conflict. The most aggressive of the three appears to have been Russia that declared mobilization in response to AustroHungarian moves against Serbia. However, it had war declared against itself by Germany even before it had the opportunity to do so against Austria-Hungary (Taylor 1980). It should also be noted that a number of Russian leaders warned against involvement in conflict, citing Russia’s continuing internal fragility that was made apparent by Russia’s defeat by Japan in 1905 and subsequent unsuccessful though bloody revolution (Kennedy 1989). France, even after Germany had declared war against its ally Russia, hesitated to join the growing conflict until Germany declared war against France (Taylor 1980), and even took measures that threatened its own ability to deliver a shattering first blow its military doctrine demanded in order to avoid engaging in provocative behavior (Doughty 2005). Finally, Great Britain joined the war almost belatedly; following a prolonged peacemaking effort and only after it was clear that the balance of power in Europe was being threatened (Taylor 1980). In France and Great Britain’s case, the high level of social trust and the modest economic growth meant their leaders had neither

9780230113176_10_ch09.indd 149

2/5/2011 10:57:04 AM

150

●

Social Trust, Anarchy, and International Conflict

the means nor the incentive to initiate major international conflict. In the case of Russia, although that country’s leaders clearly had an incentive, the combination of weak economy and a recent experience of defeat that exposed the weakness of the Russian state also had a moderating influence. However, the deep divisions within the Russian state were clearly a motivating factor behind Russia’s eagerness to provoke a conflict with Japan in 1904, conflict that would end in military defeat and internal revolution (Lincoln 1983). Discussion

The events immediately preceding the outbreak of World War I indicate considerable variance in prowar sentiment among the major power leaderships. Germany appears to have been the most aggressive of the five, as its diplomacy did not leave much room for compromise and in any event its rash declarations of war against Russia and France all but precluded the possibility of a peaceful resolution of the crisis. The pro-war fever ran high in Austria-Hungary as well, an entirely understandable phenomenon given the act of terrorism that was perpetrated against its ruling elite, however, Austria-Hungary did not desire a broader conflict that would involve other major powers in large part due to its economic weakness. Russia, likewise, appears to have been relatively eager to initiate hostilities, although whether it would have declared war on Austria-Hungary must remain a mystery since the German declaration of war took that decision out of the hands of Russian leaders. Great Britain and France, for their part, displayed very little eagerness to precipitate a conflict. The explanation why a state like Germany, which enjoyed a dominant position in Europe, should act as if it were faced by a dire threat to its existence, while states that were gradually losing their previously dominant influence appeared to be satisfied with the state of affairs, lies with the domestic make-up of these states, specifically their domestic “culture of anarchy,” which was projected by their substate actors onto the international system, and informed their beliefs about the motives of international actors. States that were successful in their nation-building efforts operated internationally with restraint, while states whose governments were plagued by issues of legitimacy were tempted to shore up their standing by engaging in international

9780230113176_10_ch09.indd 150

2/5/2011 10:57:05 AM

The Outbreak of World War I

●

151

conflicts. Similarly, states that were experiencing expansion of their capabilities were more confident in their ability to perform well in an armed conflict, and therefore acted in a more provocative, less cautious fashion than their weaker allies and adversaries. Moreover, Germany’s record of apparently unstoppable economic growth was a factor contributing to German leadership’s confidence their policies would ultimately succeed. This range of behavior dovetails quite well with the criteria for determining states most likely to initiate militarized interstate disputes, namely, the combination of low social trust and strong economic growth. Of the five major powers discussed above, two, France and Great Britain satisfy neither of the two criteria. Both powers have experienced relatively low economic growth during the pre-1914 period, which contributed to the waning of their relative standing in the European balance of power, and both have enjoyed relative success in their nation-building policies resulting in strong national cohesion that was displayed in the form of absence of violent domestic conflict and high degree of resilience during World War I. For Austria-Hungary and Russia the indictors are mixed. Both states were highly anarchic polities, with low level of social trust and consequently a low level of individual identification with the state. At the same time, their economic growth was indifferent. While their levels of bellicosity appeared to be higher than those of France and Great Britain, they were tempered by their leadership’s awareness of their internal weakness. The one power that satisfied both criteria, Germany, is also the state that displayed the greatest propensity to pursue a prowar policy. Its policy was confidently bellicose, driven by a worldview that saw Germany as an encircled power struggling for its very existence and supported by a rapidly growing economy that underpinned a military machine, on the surface, second to none. Moreover, the prowar German stance following the Sarajevo assassination was not an exceptional development but rather continuation of a long-standing pattern of international bellicosity that was often counterproductive and only compounded the sense of isolation felt by the German leadership. Although Germany’s power has increased considerably both in absolute and relative terms, its ruling elite nevertheless felt Germany was being denied a “place in the sun” and feared its economy was vulnerable to strangulation by the British naval

9780230113176_10_ch09.indd 151

2/5/2011 10:57:05 AM

152

●

Social Trust, Anarchy, and International Conflict

power (Taylor 1980). However, the most significant consequence of the resulting German naval build-up, combined with revisionist rhetoric targeted at the British naval supremacy, was to compel England to set aside its differences with its hitherto main rivals, France and Russia. The German leadership also chose to ignore France’s signaling its interest in a rapprochement, opting instead (for example, during the Morocco Crisis of 1912) to pursue a policy of threatening France, which only backfired by reinforcing France’s alliance ties with Russia and Britain. That French signaling was ignored by Germany is consistent with the findings by suggesting that low-trust actors in Prisoner’s Dilemma games, for example, ignore other players’ signaling intent to cooperate, rather than defect (Parks et al. 1996). German policies only served to undermine the prorapprochement faction in France (Taylor 1980). The externalization of German internal atmosphere of political hostility and the worldview held by German military leadership were also illustrated by the German conduct at war, specifically the spate of atrocities German army committed in the opening weeks of the war. Some 5,500 civilians were killed in premeditated manner (i.e., by firing squads, after being tried by military tribunals). This violence appears to have been a projection of the domestic norm onto the international arena. With the German military expected to defend the regime against various domestic political factions (as it actually did, for example, in 1848), it tended to view civilians, even German ones, as a potential enemy. This evolved into a set of attitudes toward civilians in occupied countries as well. During the run-up to the war, Germany was opposed to provisions in the Hague Convention that afforded combatant status to a variety of irregular home defense forces, which Germans viewed as illegal (Horne and Kramer 2001). The strong sense of being under siege felt by the German leadership and the consequent German policies are in stark contrast with the relatively benign assessments of their geopolitical environment by France and Great Britain. These declining major powers, which due to their very decline had greater justification to view the international environment with concern, did not see themselves as hopelessly beset by hostile powers. These contrasting patterns of behavior are remarkable as they are inconsistent with the power transition theory. In accordance with that theory, states that find themselves in danger of

9780230113176_10_ch09.indd 152

2/5/2011 10:57:05 AM

The Outbreak of World War I

●

153

being overtaken by others are the most likely to initiate preventive wars in order to maintain their relative position. Yet neither France nor Great Britain exhibited particular interest in containing Germany until German actions began to directly impinge upon their interests. While the British reaction to the German naval build-up (i.e., the joining of the entente cordiale, and the termination of the centurieslong Anglo-French rivalry) superficially resembles behavior predicted by the power transition theory, it should be noted that Germany’s identification as a threat by Great Britain had more to do with German policies than with German capabilities. The simultaneous growth of the U.S. naval power did not, after all, result in a U.S.British hostility or arms race (Taylor 1980). France, likewise, did not appear greatly alarmed by its continuing slippage in international standing, and appears to have resigned itself to the loss of provinces of Alsace and Lorraine, so much so that they no longer were a major factor in French domestic politics, and even toyed with the idea of a rapprochement with its increasingly powerful neighbor (Taylor 1980). From the realist perspective, this behavior is also remarkable because it runs counter to the predictions of both offensive and defensive realists: not only are France and Great Britain not vying for global, or at least regional, hegemony, they are not taking aggressive action to staunch the progressive loss of their relative position. Conversely, Germany, which appears to be the least likely candidate for conflict initiation under the power transition theory, went to war in an apparent belief it was losing (or about to lose) its relative position in the international system. While Germany’s belligerence grew, the policies France pursued became more moderate in nature. While Napoleon III was willing to risk war against Prussia-led German coalition (which fielded an army nearly twice the size of France’s own), France of the III Republic achieved reconciliation with Great Britain and even made moves toward reconciling with Germany. Although one might have expected the issue of Alsace and Lorraine to dominate French politics, in the years immediately prior to the war there existed strong factions of French politicians interested in a reconciliation, and even a financial partnership, with Germany. These ideas failed to carry the day due to the opposition to breaking the alliance with Russia a partnership with Germany would have entailed. Worse, Germany itself was not interested.

9780230113176_10_ch09.indd 153

2/5/2011 10:57:05 AM

154

●

Social Trust, Anarchy, and International Conflict

During the Morocco Crisis of 1912, German leadership opted to pursue a policy of threatening France that only backfired by rekindling the anti-German sentiment in France and discrediting the prorapprochement faction in France (Taylor 1980). Nevertheless, in spite of such snubs the idea of reconciliation with Germany persisted in France into 1914 (Taylor 1980). While, once the war began, the recovery of Alsace and Lorraine promptly became a French war aim (Goemans 2000b), these demands were caused by the outbreak of war, and were not of themselves the reason for war. Prior to the war’s outbreak the issue of Alsace and Lorraine had low political salience in France (Taylor 1980). Summary

The pattern of behavior exhibited by the five European major powers in 1914 illustrated in Table 9.1 is strongly consistent with the findings in the quantitative section of this project. Germany’s combination of high growth and relatively low level of generalized social trust, as evidenced by its fragile national identity that was built on military victories and continued confrontation with its neighbors, especially France. Germany combined high growth with a relatively low level of generalized social trust, as evidenced. Similarly, Austria-Hungary’s low level of social trust translated into belligerent policies though, cognizant of its weakness, AustroHungarian leaders acted only when assured of Germany’s support. Russia appears to have pursued a rather more pacific policy than its low level of trust might indicate, however, one must keep in mind that its belligerence was strongly tempered by an acute sense of its weakness brought home by its defeat by Japan in 1905. And, finally, France and United Kingdom, with their relatively slowly growing economies and stronger, more secure national identities, unsurprisingly prove to be the least belligerent of the major powers. Table 9.1 Economic growth and social trust levels of major European powers in 1914

High Trust Low Trust

9780230113176_10_ch09.indd 154

High Growth

Low Growth

None Germany

France, United Kingdom Austria-Hungary, Russia

2/5/2011 10:57:05 AM

The Outbreak of World War I

●

155

Thus material capabilities in the form of economic performance and major power status, and the level of generalized social trust played crucial roles in the state leaders’ decisions to seek or avoid war. It is also evident that material capabilities alone do not account for heightened willingness to risk war, as relatively weak state such as Austria-Hungary displayed much greater war-proneness than the considerably stronger France. Explanations derived from power transition theory likewise cannot account for the relative war-proneness, since the growing power Germany was considerably more war-prone than the waning powers France and Great Britain. Instead, the decisions on whether to go to war appear to be a combination of both capabilities and general worldview that is a reflection of generalized social trust obtaining within the borders of the state. The discussion of the results of the quantitative models in the preceding chapter posited that the leaders of low-trust states may act to provoke international conflicts for a number of reasons: they view the world in inherently conflictual terms, they see themselves surrounded by enemies who seek to do them harm, and they have an incentive to resort to international conflict in order to increase the society’s cohesion and to preserve their own hold on power. Whereas the quantitative models cannot distinguish among the relative importance of these factors, one of the advantages of a qualitative approach is that such motivational questions can be more convincingly addressed. The question therefore is what was the primary motivation the German leaders felt when embarking on a course toward war, the perception of other international actors as imminent threats that had to be dealt with using military means (in effect, making the war, from the German perspective, a preventive one), or the need to weaken the domestic political opposition by forcing a rally around the flag effect and acquiring additional repressive powers. An examination of German actions and evident attitudes suggests the former motivation must have been by far the weaker of the two, given the sheer level of confidence in their ultimate success and the apparent perception of Germany’s adversaries as essentially weak. Germany, after all, decisively defeated France in the Franco-Prussian War, while Russia suffered a humiliating defeat at the hands of Japan in 1905. Great Britain, while powerful at sea, was very weak on land, a factor that figured in Germany’s calculations in the form of selecting

9780230113176_10_ch09.indd 155

2/5/2011 10:57:05 AM

156

●

Social Trust, Anarchy, and International Conflict

an invasion route into France that ran through Belgium (Taylor 1980). While that invasion route had the advantage of bypassing France’s strongest fortresses, it ran the risk of drawing Great Britain into the war. However, the German leadership’s belief its superior military power would deliver a quick victory, thus preempting a large-scale British mobilization and denying the inevitable naval blockade the time it required to work, made that an acceptable risk. Therefore, while the German leaders might have viewed violence as an acceptable means of political conflict resolution (a belief reflecting the violence of German politics of the nineteenth century, including the 1848 revolution and the Wars of German Unification), there is little evidence to suggest they viewed the war in preventive terms. Instead, it is the domestic political motivation that must have provided the lion’s share of the desire to provoke a large-scale international conflict. The events leading to the outbreak of the war also cast an interesting light on the concepts of signaling and deterrence. The French effort to signal nonbelligerent intent in the form of declared willingness to stay above the fray was nullified by the German desire to win a decisive victory on both fronts. In spite of the claims by the German general staff that the two-front offensive was made necessary by the mobilization plans that allegedly could not be modified to support a one-front deployment against Russia, such statements are belied by the flexibility displayed by the German military high command once the hostilities actually broke out. When the Russian military displayed an unexpectedly rapid mobilization capability as a result of the German underestimation of Russian military capabilities (an underestimation that similarly puts into doubt the alleged German perception of a growing menace originating from the Russian Empire), German general staff planners displayed an impressive ability to depart from their prewar plans and dispatch to the endangered Eastern Front several entire army corps, each comprising tens of thousands of troops plus a large number of horses, ground transport, and heavy weaponry (Haythornthwaite 1985). To be sure, France’s signaling of peaceful intent may have been undermined by the belligerent response by its eastern ally, the Russian Empire, which reacted to the Austro-Hungarian ultimatum toward Serbia by threatening own mobilization (Haythornthwaite 1985). Moreover, Russia’s own international image as a brutal power whose

9780230113176_10_ch09.indd 156

2/5/2011 10:57:05 AM

The Outbreak of World War I

●

157

government was capable of massacring its own inhabitants and whose military power rested on seemingly inexhaustible reserves of manpower (Lincoln 1983) meant that the perceived menace Russia’s mobilization efforts likely became of exaggerated importance in the minds of German planners and provided an additional incentive to their already existing stock of incentives, derived from the context of the German internal politics, to seek an armed conflict. By the same token, the British efforts to mediate the conflict and to provide guarantees to Belgium were successful only to the extent that Germany did not declare war on Great Britain in 1914. This suggests that the German leadership, in spite of the declared fears of a potential British naval blockade that were, after all, used as justification for the ambitious and expensive naval build-up in the late- nineteenth and early-twentieth century, did not believe the British represented a major threat and were not going to enter the conflict. Therefore there are reasons to believe the comparatively higher level of generalized social trust obtaining in Great Britain had a pacifying effect on German policies, although such effect was insufficient to avert conflict that was driven by circumstances outside of British control. Overall, although the mixed nature of the alliances containing both low-trust and high-trust states complicates the analysis of signaling of belligerent or nonbelligerent intent among the adversaries, the available evidence does suggest that low-trust states had some effect on pacifying the more belligerent states. That such effects were too small to avert the war is due to two factors. The first is the already mentioned mixed nature of the alliances. It is quite possible that an alliance consisting solely of high-trust states like France and Great Britain would have been better able to prevent conflict through their signaling of peaceful, nonbelligerent intent. Alas, the entente cordiale was not NATO of the Cold War–era. The presence of low-trust and anarchic Russian Empire within the alliance and the exchange of diplomatic notes between the Austro-Hungarian and Russian governments, which swiftly escalated to the level of threats, therefore crossing the militarized interstate dispute threshold, meant that France and Great Britain were, in effect, drawn against their will into a conflict that originated between several European empires with weak national identities and low trust levels.

9780230113176_10_ch09.indd 157

2/5/2011 10:57:06 AM

158

●

Social Trust, Anarchy, and International Conflict

The second, and more important, reason for the failure of nonbelligerent signaling was that, in accordance with the results presented in the quantitative analysis section of this book, the attributes of the potential target of aggression matter far less than the attributes of the potential attacker. Even though the generalized social trust of the would-be target are effective at averting aggression, it is evident that such effects are, both substantively and statistically, far lower than the influence of the generalized social trust of the target on its decisions on whether or not to initiate militarized interstate disputes. Thus, in the final account, it was the low trust levels of the Austro-Hungarian and German empires that proved decisive, combined with the confidence-inducing economic power of the German state without whose guaranteed backing the Austro-Hungarian leadership would have likely lacked the courage to risk war. While war performance and termination are outside of the scope of this book, but will instead be the subject of a future study, the extent to which state strength and state-society cohesion correspond to the success or failure on the battlefield is remarkable. As noted earlier, the defeated powers were not predominantly defeated by superior weapons, tactics, or even numbers. Even during the 1918 Armistice, the allies were yet to breach the formidable barrier of the Rhine river, which guaranteed Germany’s security. Instead, the defeated powers collapsed because their state-society bonds proved too weak to sustain the burden of war for an extended period of time. Thus the Russian Empire, the Austro-Hungarian Empire, the Ottoman Empire, and, ultimately, the German Empire succumbed to growing domestic unrest and revolution. Although at first glance it appears World War I was a triumph of democracies over nondemocracies, the example of Italy, a relatively weak state that very nearly did suffer a decisive military defeat and whose democratic institutions were easily toppled by Mussolini’s fascist movement in the early 1920s confirms the importance of generalized social trust as a determinant not only of conflict initiation, but also of wartime resilience (Haythornthwaite 1992).

9780230113176_10_ch09.indd 158

2/5/2011 10:57:06 AM

CHAPTER 10

Analysis and Conclusions The Social Trust Theory of International Conflict

The results of both the quantitative and qualitative components of the book strongly suggest that the diversionary war and democratic peace phenomena appear to be related to state strength and the resulting level of generalized social trust obtaining within the borders of the state. Democratic peace appears to be, in actuality, peace among strong states. While most strong states are in fact democracies (a fact that leads to the empirical findings suggesting democracy is responsible for the favorable outcome of peace), this should not obscure the fact that non-democracies can also provide high levels of public goods, or that many of the world’s democracies are quite weak. The weakness of such emerging democracy, while a challenge to the democratic peace theory, is easily accounted for by a theoretical framework that relies not on regime type as a determinant of international conflict initiation but on social trust as a product of state strength and capacity. In a similar vein, the diversionary theory of war’s weak empirical findings appear to be the consequence of an undue focus on regime popularity rather than regime strength and legitimacy. Whereas popularity is a fleeting phenomenon, legitimacy is much harder to address. In consequence, while strong states with high levels of generalized social trust almost never experience diversionary incentives, regardless of the level of popularity of their governments, such incentives exist in weak states on a near-permanent level, and are independent of governmental popularity. Democratic Peace

The democratic peace effect appears to be peace between high-trust states with strong, effective governments, as the ideational effects

9780230113176_11_ch10.indd 159

2/5/2011 10:57:22 AM

160

●

Social Trust, Anarchy, and International Conflict

attributed to democracy show up strongly in the analysis of the effects of domestic anarchy. While high-trust states do tend to be democratic, it should be noted that some states in the high-trust category of states are not democratic (for example, Singapore), and that democracies are represented toward the lower end of the social trust scale as well. This is unsurprising, given that transitional regimes have long been recognized as being somewhat exempt from the benefits of democratic peace. Thus while democratic systems of government, in principle, represent a more reliable path toward effective governance, in practice the quality of governance of both democracies and non-democracies varies very widely. Such a conclusion is consistent with existing literature on democratization, democratic transitions, and emerging states. There is considerable evidence that emerging democracies and fragile states in general are more conflict-prone than their democratic status might otherwise indicate. This is evident from, for example, studies on secession movements (Tir 2005). It therefore appears desirable to establishing a more sophisticated test of consolidated democracy that would include a measure of the quality or effectiveness of governance to accompany the usual measures dealing with the process of selecting a government. Yet, as this study suggests, the ends of government matter as much, if not more, than the means of government. Diversionary War

Aside from the broad proposition that diversionary incentives are a function not of popularity but of legitimacy, the implications for the diversionary war are somewhat more complex and ambiguous. On the one hand, the study is consistent with the diversionary theory fundamental assumption that causes of international conflicts are to be found in domestic politics. However, where the findings depart from the standard diversionary war theory is in the specification of the independent variable. To recapitulate, diversionary theory posits that leaders initiate international conflicts when their political fortunes are threatened by some adverse developments that undermine their popularity. Thus the dependent variable is usually operationalized in terms of economic growth (on the entirely reasonable and likely correct assumption that, in general, bad economic performance translates into

9780230113176_11_ch10.indd 160

2/5/2011 10:57:22 AM

Analysis and Conclusions

●

161

government unpopularity) or, sometimes more directly, using popularity ratings of leaders. However, the findings strongly suggest that it is economic growth, not economic slumps, that are responsible for increased rates in MID initiation, which runs counter to diversionary theory predictions, though not to several diversionary theory critiques proposing that just when states might have an incentive to divert, they might lack the means. Moreover, the strong inverse relationship between social trust and militarized dispute initiation points toward a number of additional possible implications for the diversionary theory. The first, and relatively simple, explanation is that there is no diversion per se. The theory assumes that when seeking a target to be a “scapegoat” for one’s own failings, the decision maker(s) in question initiate international conflict in full knowledge that the target is in fact merely a scapegoat. But if MID initiations flow from an inherent mistrust of others, by definition such MIDs cannot be termed diversionary. After all, it is hardly a diversion if one sincerely believes the MID target genuinely represents a threat that must be dealt with as soon as the opportunity presents itself, and can be dealt with only through use, threat, or demonstration of force. One has to consider the possibility that the diversionary and/or scapegoating behavior may be an expression not of rational calculation with the aim of creating a rally around the flag effect that would relieve the political pressure on the leader but rather an expression of genuine belief that the targeted party is actually to blame for whatever problems are afflicting the country. If one has a low level of generalized social trust, if one’s belief on the inherent nature of other actors is that they are fundamentally untrustworthy and bent on doing one harm, then blaming other actors for any setbacks one suffers is the natural outcome. Bad events simply become evidence of others’ bad faith, confirmation of an already existing belief in their untrustworthiness and malevolence. The second, more complex, explanation (and, given the evidence of at least rudimentary rational utility calculation, probably the more likely one) is that leaders of anarchic, self-help states with low levels of social trust are aware of the masses’ xenophobic sentiments and attempt to channel their latent hostility outward, at external or international targets. The masses, for their part, are likely to support their leaders’ aggressive policies because such policies are consistent with

9780230113176_11_ch10.indd 161

2/5/2011 10:57:22 AM

162

●

Social Trust, Anarchy, and International Conflict

the worldview held by most individuals in a low-trust society. Moreover, conflict creates the impression that the state is providing public goods in the form of protection against an external foe. This explanation therefore implies that a diversion is a function of not so much unpopularity as of illegitimacy of the regime. This explanation also helps explain the puzzling lack of findings in support of the diversionary theory of war. Legitimate regimes, even when unpopular, do not engage in diversionary activities while for illegitimate regimes the diversionary incentives are ever-present, and are not contingent on economic performance or other indicators of popularity. Moreover, the combination of a mass public predisposed to be hostile toward “others” and a leader willing to cynically manipulate these sentiments appears to have some empirical support. It certainly appears possible that Adolf Hitler, for example, did not harbor a genuine hatred of the Jews, but rather discovered that “playing the race card” could be an effective method of building a political movement during his years spent in Vienna, the capital of a fairly anarchic, low-trust Austro-Hungarian Empire. There is likewise little evidence to suggest Hitler expressed strong antisemitic views in the prewar period (Kershaw 1998). At the same timer, Hitler’s view on the usefulness and desirability of war reveal an instrumentalist approach wholly consistent with the diversionary war theory and, moreover, his evident contempt for Germany’s adversaries suggests that Hitler did not genuinely consider other states to represent threats to Germany’s security. Instead, he appears to have viewed war as a tool of social engineering to be used, in conjunction with eugenics and other similar techniques, as a means for creating the “master race” (Kershaw 2000). One must also entertain the possibility that the apparent diversionary behavior represents what amounts to populism on part of one or more factions within the ruling elite. The logic of social trust suggests that a significant degree of division is likely even within the ruling elite under conditions of low generalized trust. However, even though the level of state legitimacy may be low, a depressed level of social trust and concomitant fragmentation of the polity into small in-groups bound by particularized trust implies low likelihood of collective action against the ruling elite. Thus external “diversions” are not necessarily embarked upon in order to shore up the popularity, or even legitimacy, of regime as a whole, but rather to create popular

9780230113176_11_ch10.indd 162

2/5/2011 10:57:23 AM

Analysis and Conclusions

●

163

support for a particular faction. International conflict may be the outcome of an ensuing process of outbidding, with none of the ruling elite factions wanting to appear “weak on defense.” The second and third explanations are particularly compelling given the apparent presence of a rational component in the decisions to initiate MIDs. The third possible alternative explanation is that the state’s failure or refusal to provide public goods may be due to leaders’ mistrust of likely political enemies and/or it may be a deliberate attempt to implement a “divide and conquer” strategy of political demobilization. While in an anarchic, low-trust society domestic actors may view the state as illegitimate, since they view each other as untrustworthy they are unlikely to band together against the state. Such a situation may have arisen in Wilhelmine Germany, where the need to maintain universal military service clashed with the fear of many of the members of the very classes that filled the bulk of the German military’s ranks. What may have distinguished Germany from, for example, France is that the latter’s unprecedentedly swift military defeat in 1870 changed its ruling elite’s view as to whether their most dangerous enemy was foreign or domestic. The far-reaching reforms of the Third Republic may have been due to the apparent fact that threat of Germany has displaced the threat of revolution as the greatest concern to the French state and ruling class. This third possible scenario need not be seen as mutually exclusive with the second one. The preceding chapter’s analysis of the motivations of German leaders in1914 likewise discounts the likelihood the leaders were acting on a genuine belief that other major powers represented clear and present dangers to the German security. Instead, they appeared eager to force a rally around the flag effect to validate their own domination of German politics on the basis of their role as defenders of the realm against foreign enemies, and thus silence the growing domestic opposition demanding political and social reforms. The findings of this project also point toward a new interpretation of the question of conflict externalization. Whereas some studies have argued that states are less likely to initiate international conflicts when beset by domestic strife due to the fragmentation of the polity (Morgan and Bickers 1992) or the need to use repression diverting resources from a possible external use of force (James and Hristoulas 1994; Chiozza and Goemans 2004), while others found that domestic

9780230113176_11_ch10.indd 163

2/5/2011 10:57:23 AM

164

●

Social Trust, Anarchy, and International Conflict

conflict made states more likely to initiate international conflicts (Caprioli and Trumbore 2003) due to diversionary/scapegoating incentives (Heldt 1999; Barzilai 1999) or simple spillover (for example, “hot pursuit” across international borders), (Gleditsch et al. 2007), the findings here suggest that domestic and international conflict may well go hand in hand due to the individual views on the nature of other actors, be they within or without a given state’s borders. To put it in different terms, individuals regard violence as an appropriate or an inappropriate method of resolving conflicts, and that attitude is applied regardless whether the other party, or parties, to the conflict are separated from the decision maker in question by an international border or not. Social Trust and International Relations

The results of this study are quite encouraging for the application of the concept of social trust to the study of not only domestic but also international political phenomena. The findings presented earlier suggest that social trust is a powerful concept that is as-yet underutilized in the studies of international relations. In addition to explaining a variety of domestic phenomena (on which there already exists voluminous literature), it has an equally significant influence on the international behavior of states. The results suggest that the most direct path toward establishing an international system free of violent conflict is to assure such a system is populated by strong states whose effective governance fosters a strong sense of national identity and high levels of generalized social trust, as such states appear to be considerably less belligerent (both domestically and internationally) than their anarchic, low-trust counterparts whose influence on the system is quite destabilizing. While international cooperation is outside the scope of this project and instead remains to be explored in the future, the relationship between generalized social trust and international conflict makes it at least plausible that just as states with high levels of social trust are less likely to become sources of violent international conflict, they are similarly likely to prove to be the states most engaged in active cooperation with others, and most willing to assist others with their problems. In a similar vein, the results are also entirely supportive of the earlier findings to the effect

9780230113176_11_ch10.indd 164

2/5/2011 10:57:23 AM

Analysis and Conclusions

●

165

that strong in-group loyalty does not necessarily imply, let alone necessitate, the existence of equally strong out-group hostility (Terhune 1965; Druckman 1994; Yamagishi and Kiyonari 2000). Indeed, it does appear that the concept of state as a stage toward ever-larger political entities (Richmond 1995; Hutchinson 1995; Gellner 1995) is supported by evidence, inasmuch as absence of conflict between states is a necessary condition for their merger into a larger entity. One of the more promising avenues of future research incorporating social trust is the evident possibility of integrating social trust with the considerable variety of psychological and cognitive approaches toward decision making with the goal of establishing a more unified theory of decision making. Of particular interest here are the common features of the existing psychological and cognitive approaches to decision making, particularly their emphasis on framing. Thus, for example, arguably the most crucial and theoretically least understood element of the prospect theory is whether the individual actor in question views his/her position in terms of losses or gains. This is of major importance since while a losses frame leads to risk-seeking behavior, a gains frame results in risk avoidance. Similarly, research on analogical decision depends heavily on selection of an earlier example to which the current situation is compared. Yet here as well the actual process of selection of analogies remains poorly understood, even though, as in prospect theory, this is arguably the most crucial part of the process. Here, too, the selection of the frame, or analogy, is of far-reaching importance since the analogy invoked is an important predictor of the decision that will be made. And yet studies that examine the process of selection from among a number of alternative courses of action presented by advisors to political decision makers offer few generalizable explanations as to why a certain course of action was chosen over another. Since these framing processes remain poorly understood, cognitive and psychological decision-making approaches are largely reduced to offering post hoc explanations of past events. Unfortunately, while a meticulous study of a political event may provide us with an understanding of a frame, or analogy, in which the individual decision maker was operating while undertaking a certain decision, we are still at a loss when attempting to predict when particular frames or analogies are likely to be invoked.

9780230113176_11_ch10.indd 165

2/5/2011 10:57:23 AM

166

●

Social Trust, Anarchy, and International Conflict

This failure has long been one of the key weaknesses of psychological and cognitive approaches, as compared to the rational choice approach. Arguably the greatest part of the problem in predicting framing effects is that such effects are to a large extent rooted in one’s perception of events, not merely the objective nature of events. The findings of this project may point to a way forward. Indeed, one way to view generalized social trust is as the “ur-frame,” the underlying set of beliefs about actors, relationships, and “how things really are” from which all other analogies and frames spring. Social trust, after all, appears to strongly influence how individual actors are predisposed to perceive other actors, and it could even play a role in the formation of individual operational codes. Therefore, it certainly appears more than plausible that a low-trusting individual, who by default perceives others as inherently untrustworthy and predisposed toward doing one harm in some way when given the opportunity, is rather more likely to place oneself in a losses frame in response to any particular change in the situation (even one that objectively improves his or her position) than a high-trusting individual. It is equally highly plausible that a low-trusting individual is likely to invoke all manner of analogies supporting the already preexisting perception of others as untrustworthy. In both cases, it stands to reason that the conflictual alternatives are likely to be the preferred policy options. A closer, “thick description”–type examination of most, if not all, of the MID initiations that took place within the spatial and temporal span of this study would likely show that the decision makers in question felt themselves to be in a losses frame and/or were viewing their current predicament through the lens of an analogy that favored the resort to use of force. Next to its potential as the “ur-frame,” generalized social trust’s second attractive feature, is that, unlike other psychological and cognitive variables, it is readily operationalizable for use in large-n studies that are currently the currency of the realm in international relations, and thus capable of producing generalizable theoretical predictions. International Anarchy and International Relations Theory

Overall, the findings do support the voluminous literature claiming that international anarchy does not have a powerful and uniform effect on actors operating in that system. The findings furthermore

9780230113176_11_ch10.indd 166

2/5/2011 10:57:23 AM

Analysis and Conclusions

●

167

offer support for the possibility of realizing the concept of Kantian peace. Indeed, it would appear that what Kant’s “universal hospitality” listed as one of the three pillars of “perpetual peace” is synonymous with generalized social trust to the extent that it posits identical treatment of, and respect for, all members of humanity. However, whereas Wendt’s structural constructivist interpretation argues that state behavior is the product of the system-level culture of anarchy, the argument advanced here posits the opposite: since social trust is fundamentally a domestic variable that is constructed through government action through its efforts to eliminate anarchy on the domestic level and forms a major part of the state’s identity, it is the domestic rather than international factors that will play the decisive role of how states view other international actors. Individual views (both elite and mass) of the international system and the appropriate norms of behavior regarding international interactions (as well as actual behavior) are therefore conditioned by the level of social trust existing on the national level. Individuals socialized into a particular domestic culture of anarchy use schemata and heuristics formed in the process of socialization onto international situations. They perceive international anarchy, and actions of other actors within that anarchy, using analogies from their experiences first gained in their respective domestic cultures of anarchy, and act accordingly. Beliefs about the state of anarchy and the inherent nature of other actors in the international system are influenced by the already-acquired beliefs concerning the nature of the domestic actors. For example, an individual socialized into a state with a high social trust level and an absence of anarchy will have had little or no experience of the negative effects of anarchy. When exposed to the international system for the first time, this individual will be likely to perceive this anarchy in more benign terms than someone whose political socialization has taken place in a more anarchic domestic milieu. Consequently, while the anarchic international system is compatible with a considerable range of behaviors, the prevalent behavior appears to be structurally determined. International anarchy is, to use Wendt’s phrase, what we make of it, provided that we recognize that our range of choices as to what to make out of international anarchy is severely constrained by the domestic “culture of anarchy” in which we live, and the state of domestic politics of the major powers in the system.

9780230113176_11_ch10.indd 167

2/5/2011 10:57:23 AM

168

●

Social Trust, Anarchy, and International Conflict

On the whole the findings are strongly supportive of Moravcsik’s (1997) thesis of domestic origins of national preferences, or, at any rate, preferences toward conflict, and are consistent with the neoliberal theory of international relations. Attributes of individual states display a consistently stronger impact on international conflict initiation than attributes of potential opponents or of the state pair. Indeed, in the broadest sense, state preferences toward conflict do appear to depend on that state’s quality of governance, and its ability to minimize the elements of anarchy and self-help on the domestic level. However, domestic anarchy and its companion generalized social trust by no means explain the problem by themselves. The results suggest that while generalized social trust creates a predisposition toward viewing others in hostile terms, state actors only act on these feelings if they believe their capabilities are sufficient to threaten military action with low risk. The combination of these two factors indicates a strong possibility that decision-making processes consist of both rational and cognitive components, with the implication being that neither rational choice nor cognitive/psychological approaches are capable of explaining decision making on their own. The support for neoliberalism notwithstanding, there is definitely a place for the realist theory in the study of international relations. The results of the study suggest that the realist findings concerning the effects of anarchy hold true only for interactions among states characterized by low social trust. The inherently conflictual character of international interaction posited by realism is, however, muted if not necessarily wholly absent in interactions of high-trust states. Similarly, the findings provide an additional insight into the eternal debate on whether and/or when states pursue relative or absolute gains. The logic of the social trust arguments suggests that relative gains would become salient only for low-trust actors. Since such actors tend to ascribe malign intent to others, under such conditions the acquisition of any advantage by another actor would be of concern. For hightrust actors, the dynamic is rather different. Since they are not inherently afraid of others, whether others acquire superior capabilities is a matter of much less concern. Moreover, the structural realist proposition that the international environment shapes state behavior merits additional scrutiny, as international anarchy may have a hitherto underestimated influence on

9780230113176_11_ch10.indd 168

2/5/2011 10:57:23 AM

Analysis and Conclusions

●

169

the domestic politics of states that believe themselves to exist in a particularly hostile environment. Of particular interest here is the apparent feedback loop between international anarchy and domestic politics hinted at in the Chapter 8 discussion of the process of emergence of strong European states. The discussion of the international system in the nineteenth century indicates that states’ efforts to cope with the effects of an anarchic international system included eliminating features of anarchy and self-help within their own borders. Over time, the domestic transformation has begun to impact the perception of international anarchy that drove the nation-building projects in the first place. Looking to the Future

The final set of implications concerns the potential impact of the process of globalization on the international system. In view of the frequency with which 1914 is invoked as an analogy in contemporary globalization literature, often with an eye to demonstrating the differences between the world systems of 1914 and of today, the one analogy that bears examining is the state of the state in 1914 and now. While in 1914 the international system still included many states with relatively high levels of domestic anarchy, one must ask the question whether today domestic anarchy is not making a comeback under the guise of globalization. The literature on globalization often engages in comparisons between the current global situation and one that prevailed prior to 1914 (Frankel 2000; Wolf 2005). Yet one must be cognizant that while prior to 1914 there were voices who confidently claimed that a war among major powers could not break out owing to the great cost of such an endeavor, in 1914 the war came nevertheless, a war whose outbreak moved the British Foreign Secretary to presciently remark “the lamps are going out all over Europe; we shall not see them lit again in out lifetime” (Haythornthwaite 1992). Although the lamps were gradually relit in the decades following World War II, nevertheless one should ask the question whether history is set to repeat itself. To be sure, globalization and expansion of trade likely do contribute to overall global economic growth. However, in return for the promise of economic growth and development, states

9780230113176_11_ch10.indd 169

2/5/2011 10:57:24 AM

170

●

Social Trust, Anarchy, and International Conflict

are encouraged to cede control over flows of goods and capital (Rodrik 2001), reducing the set of policy tools available to them for dealing with domestic problems. This potentially Faustian bargain makes globalization a phenomenon whose influence on the international system is still at best only ambiguous. However, one must also pose the question whether economic growth is not coming at the cost of increasing domestic anarchy. The sacrifice of sovereignty that globalization implies in turn translates into weakened capacity for dealing with negative aspects of globalization, such as drug trade, international terrorism, et cetera (Naim 2003). Equally worrying from the point of view of social trust, which is known to exist mainly in relatively egalitarian societies, is the growth of inequality within and among states that has been one of the externalities of globalization (Scott 2001; Garrett 2004). To be sure, some scholars of globalization argue that the role of the state has not been diminished at all (Waltz 1999; Gilpin 2001), citing the state’s monopoly on the legitimate use of violence and the role of the U.S. military as the ultimate guarantor of international stability (Waltz 1999). This argument, however, only points to an imbalance that may be developing due to globalization. One should note that even as states lose control over various aspects of social and economic policies, even as governance on a variety of economic matters is being transferred to international organizations such as the World Trade Organization, one aspect of state authority remains stubbornly unglobalized: the capacity to wage war. Therefore even as the state’s capacity to build social trust may be reduced, its ability to attempt to shore up its legitimacy through international conflict is relatively unaffected. If unchecked, this development could lead to a situation where military force, rather than being the “ultimate argument” of states, might become their first and only one. While some globalization studies (for example, Wolf 2005) discount the possibility of reemergence of exclusivist and xenophobic ideologies with mass appeal, the breakdown in social trust would likely have precisely that effect. Studies positing that growth of globalization may lead to an increase in conflict (for example, Huntington 1993) might be proven correct in their predictions, although not in the predicted causes of the phenomenon. Rather than the result of a Huntingtonian “clash of civilizations,” there is a danger that globalization will cause increased conflict if it serves to undermine the social

9780230113176_11_ch10.indd 170

2/5/2011 10:57:24 AM

Analysis and Conclusions

●

171

trust-building role of the state. However, if the path the process of globalization is currently following leads to the undermining of the social trust–building role of the state, the likely result of the increase in domestic and international conflict would be a collapse of globalization itself, a rollback in the level of international integration similar, if not greater, in its scope than the retrenchment experienced following World War I. This, in turn, would likely lead to a vicious spiral of economic and political calamity not unlike the one experienced in the 1930s. Therefore, inasmuch as globalization contributes to the weakening of the state and undermining its capacity to build social trust, it may be sowing the seeds for another systemic conflagration. It should be once again pointed out that the “strong” states that exist today, states that effectively eliminate anarchy and self-help on the domestic level, are a comparably recent creation. The positive effects they have, both on the domestic and international arenas, should neither be underestimated nor, worse, be taken for granted. As paradoxical as it may seem, another threat to social trust stems from the end of the Cold War–era superpower rivalry. Nevertheless, it must be recognized that the exercises in social trust creation that states embarked upon during the nineteenth and twentieth centuries, and that led to the relatively pacific high-trust nation-states of today, were to a large extent national security projects. The series of defeats at the hands of the comparably highly efficient Napoleonic France propelled other European powers to emulate the French reforms, in many cases even retaining French modes of administration even after 1815 (Broers 1996). The 1870 defeat was the spark that prompted French nation-building reforms so well described by Eugen Weber (1976). Similarly, Russian modernization efforts in the latter part of nineteenth and early part of the twentieth century were to a large degree driven by the growing awareness of backsliding and military vulnerability (Hosking 1997) brought home by the defeats in the Crimean and Russo-Japanese wars. The U.S. post–World War II investments in science, education, and infrastructure were likewise driven by the perception of Soviet threat. Therefore, with the apparent disappearance of major power antagonism, the question of what factors motivate states to engage in social trust–creating policies acquires a particular urgency.

9780230113176_11_ch10.indd 171

2/5/2011 10:57:24 AM

172

●

Social Trust, Anarchy, and International Conflict

Of course, this worst-case scenario is far from inevitable. For one thing, even though the national security–driven nation-building projects appear to belong to history, one has to acknowledge the possibility that once generalized social trust passes a certain threshold, it becomes a self-sustaining phenomenon as suggested, for example, by Crepaz (2007). However, to date no research has been done to ascertain where that threshold is to be found. All of the above suggests that the role of the nation-state is evolving, and the consequences of this evolution will be as profound as those of its initial emergence. The final remaining question is whether the trend toward building ever-more cohesive states enjoying high levels of legitimacy and social trust will continue, culminating in greater economic and eventually political integration, or whether it will be reversed. At the moment, the pattern of events does not point clearly toward one set of outcomes. On the one hand, growing political integration among European states, including a number of formerly “ancient enemies,” is a cause for optimism. However, the trend toward weakening of social trust in the recent decades (Putnam 2000), the trend that is likely to be exacerbated by the so-called austerity measures adopted by many countries in the wake of the 2008 financial crisis, is worrisome inasmuch as it threatens to undo much of the progress made in the twentieth century. If so, the lessons of the nineteenth and twentieth centuries may yet have to be relearned in the twenty-first.

9780230113176_11_ch10.indd 172

2/5/2011 10:57:24 AM

References

Aberbach, Joel D., and Jack L. Walker. 1970. “Political Trust and Racial Ideology,” American Political Science Review 64 (4): 1199–1219. Achen, Christiopher H. 1982. Interpreting and Using Regression. Sage University Paper 29. Thousand Oaks: Sage Publications. A’Hearn, Brian. 1998. “Institutions, Externalities, and Economic Growth in Southern Italy: Evidence from the Cotton Textile Industry, 1861– 1914,” Economic History Review 51 (4): 734–762. Ahrensdorf, Peter J. 2000. “The Fear of Death and the Longing for Immortality: Hobbes and Thucydides on Human Nature and the Problem of Anarchy,” American Political Science Review 94 (3): 579–593. Alam, M. Shahid. 1989. “Anatomy of Corruption: An Approach to the Political Economy of Underdevelopment,” American Journal of Economics and Sociology 48 (4): 441–456. Almond, Gabriel, and Sidney Verba. 1963. The Civic Culture. Princeton, NJ: Princeton University Press. Anderson, Benedict. 1995. “Imagined Communities.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 89–96. New York: Oxford University Press. Anderson, Christopher J., and Yuliya V. Tverdova. 2003. “Corruption, Political Allegiances, and Attitudes toward Government in Contemporary Democracies,” American Journal of Political Science 47 (1): 91–109. Audoin, Stephane. 1997. “Children and the Primary Schools of France, 1914–1918.” In State, Society, and Mobilization in Europe during the First World War, edited by John Horne, 39–52. Cambridge: Cambridge University Press. Axelrod, Robert. 1973. “Schema Theory: An Information Processing Model of Perception and Cognition,” American Political Science Review 67 (4): 1248–1266. Azfar, Omar, Young Lee, and Anand Swamy. 2001. “The Causes and Consequences of Corruption,” Annals of the American Academy of Political and Social Science 573: 42–56.

9780230113176_12_ref.indd 173

2/5/2011 10:57:34 AM

174

●

References

Bahry, Donna, Mikhail Kosolapov, Polina Kozyreva, and Rick K. Wilson. 2005. “Ethnicity and Trust: Evidence from Russia,” American Political Science Review 99 (4): 521–532. Baier, Annette. 1986. “Trust and Antitrust,” Ethics 96 (2): 231–260. Banfield, Edward C. 1967. Moral Basis of a Backward Society. New York: Free Press. Barkdull, John. 1995. “Waltz, Durkheim, and International Relations: The International System as an Abnormal Form,” American Political Science Review 89 (3): 669–680. Barzilai, Gad. 1999. “War, Democracy, and Internal Conflict: Israel in a Comparative Perspective,” Comparative Politics 31 (3): 317–336. ———. 2000. “EUGene: A Conceptual Manual,” International Interactions 26: 179–204. Accessed July 10, 2007, http://eugenesoftware.org. Bennett, Scott D., and Timothy Nordstrom. 2000. “Foreign Policy Substitutability and Internal Economic Problems in Enduring Rivalries,” Journal of Conflict Resolution 44: 33–61. Benson, Michelle, and Jacek Kugler. 1998. “Power Parity, Democracy, and the Severity of Internal Violence,” Journal of Conflict Resolution 42 (2): 196–209. Berejikian, Jeffrey, and John S. Dryzek. 2000. “Reflexive Action in International Politics,” British Journal of Political Science 30 (2): 193–216. Berghe, Pierre van den. 1995. “A Socio-Biological Perspective.” In Nationalism, edited by John Hutchinson and Anthony D. Smith. New York: Oxford University Press. Berman, Sheri. 1997. “Civil Society and the Collapse of the Weimar Republic,” World Politics 49 (3): 401–429. Besancon, Marie. 2005. “Relative Resources: Inequality in Ethnic Wars, Revolutions, and Genocides,” Journal of Peace Research 42 (4): 393–415. Bessel, Richard. 1997. “Mobilization and Demobilization in Germany.” In State, Society, and Mobilization in Europe during the First World War, edited by John Horne, 212–223. Cambridge: Cambridge University Press. Booth, John A., and Patricia Bayer Richard. 1998. “Civil Society, Political Capital, and Democratization in Central America,” Journal of Politics 60 (3): 780–800. Brann, Peter, and Margaret Foddy. 1987. “Trust and the Consumption of a Deteriorating Common Resource,” Journal of Conflict Resolution 31 (4): 615–630. Braumoeller, Bear. 1997. “Deadly Doves: Liberal Nationalism and the Democratic Peace in the Soviet Successor States,” International Studies Quarterly 41 (3): 375–402.

9780230113176_12_ref.indd 174

2/5/2011 10:57:34 AM

References

●

175

Bremer, Stuart A. 1992. “Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 1816–1965,” Journal of Conflict Resolution 36: 309–341. Breuilly, John. 1995. “The Sources of Nationalist Ideology.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 103–113. New York: Oxford University Press. Brewer, Paul R., Kimberly Gross, Sean Aday, and Lars Willnat. 2004. “International Trust and Public Opinion about World Affairs,” American Journal of Political Science 48 (1): 93–109. Broers, Michael. 1996. Europe under Napoleon, 1799–1815. London: Arnold. Brose, Eric Dorn. 2001. The Kaiser’s Army. New York: Oxford University Press. Bueno de Mesquita, Bruce. 1985. “Toward a Scientific Understanding of International Conflict: A Personal View,” International Studies Quarterly 29: 121–136. Bueno de Mesquita, Bruce, James D. Morrow, Randolph M. Siverson, and Alastair Smith. 1999. “An Institutional Explanation of the Democratic Peace,” American Political Science Review 93 (4): 791–807. ———. 2003. The Logic of Political Survival. Cambridge, MA: MIT Press. Bueno de Mesquita, Bruce, and Randolph M. Siverson. 1995. “War and the Survival of Political Leaders: A Comparative Study of Regime Types and Political Accountability,” American Political Science Review 89: 841–855. Bull, Martin J., and James L. Newell. 2003. “Conclusion: Political Corruption in Contemporary Democracies.” In Corruption in Contemporary Politics, edited by Martin J. Bull and James L. Newell, 233–245. New York: Palgrave Macmillan. Buzan, Barry, Charles Jones, and Richard Little. 1993. The Logic of Anarchy: Neorealism to Structural Realism. New York: Columbia University Press. Caprioli, Mary. 2005. “Primed for Violence: The Role of Gender Inequality in Predicting Internal Conflict,” International Studies Quarterly 49: 161–178. Caprioli, Mary, and Peter F. Trumbore. 2003. “Ethnic Discrimination and Interstate Violence: Testing the International Impact of Domestic Behavior,” Journal of Peace Research 40 (1): 5–23. Carr, Edward Hallett. 1964. The Twenty Years’ Crisis, 1919–1939. New York: Harper & Row. Cederman, Lars-Erik. 2001. “Back to Kant: Reinterpreting the Democratic Peace as a Macrohistorical Learning Process,” American Political Science Review 95 (1): 15–31.

9780230113176_12_ref.indd 175

2/5/2011 10:57:34 AM

176

●

References

Chambers, Simone, and Jeffrey Kopstein. 2001. “Bad Civil Society,” Political Theory 29 (6): 837–865. Chan, Steve. 1997. “In Search of Democratic Peace: Problems and Promise,” Mershon International Studies Review 41 (1): 59–91. Chapman, Terence. 2008. “Unraveling the Ties Between Civic Institutions and Attitudes toward Political Violence,” International Studies Quarterly 52: 515–532. Chickering, Roger. 2004. Imperial Germany and the Great War, 1914–1918. Cambridge: Cambridge University Press. Chiozza, Giacomo, and H. E. Goemans. 2003. “Peace Through Insecurity: Tenure and International Conflict,” Journal of Conflict Resolution 47 (4): 443–467. ———. 2004. “International Conflict and the Tenure of Leaders: Is War Still ‘Ex Post’ Inefficient?” American Journal of Political Science 48 (3): 604– 619. Cingranelli, David L., and David L. Richards. 1999. “Respect for Human Rights after the End of the Cold War,” Journal of Peace Research 36 (5): 511–534. Claibourn, Michele P., and Paul S. Martin. 2000. “Trusting and Joining? An Empirical Test of the Reciprocal Nature of Social Capital,” Political Behavior 22 (4): 267–291. Clark, David H. 2000. “Agreeing to Disagree: Domestic Institutional Congruence and U.S. Dispute Behavior,” Political Research Quarterly 53 (2): 375–400. ———. 2003. “Can Strategic Interaction Divert Diversionary Behavior? A Model of U.S. Conflict Propensity,” Journal of Politics 65 (4): 1013–1039. Clark, David H., and Timothy Nordstrom. 2005. “Democratic Variants and Democratic Variance: How Domestic Constraints Shape Interstate Conflict,” Journal of Politics 67 (1): 250–270. Clausewitz, Carl von. 2004. On War. Whitefish, MT: Kessinger Publishing. Collier, Paul, and Anke Hoeffler. 1998. “On Economic Causes of Civil War,” Oxford Economic Papers 56 (4): 563–573. ———. 2002. “On the Incidence of Civil War in Africa,” Journal of Conflict Resolution 46 (1): 13–28. ———. 2004. “Greed and Grievance in Civil War,” Oxford Economic Papers 56: 563–595. Connor, Walker. 1995a. “A Nation Is a Nation, Is a State, Is an Ethnic Group, Is a . . .” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 36–47. New York: Oxford University Press.

9780230113176_12_ref.indd 176

2/5/2011 10:57:34 AM

References

●

177

———. 1995b. “When Is a Nation?” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 154–160. New York: Oxford University Press. Cornwall, Mark. 1997. “Morale and Patriotism in the Austro-Hungarian Army.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 173–194. Cambridge: Cambridge University Press. Cortell, Andrew P., and James W. Davis. Jr. 1996. “How Do International Institutions Matter? The Domestic Impact of International Rules and Norms,” International Studies Quarterly 40 (4): 451–478. ———. 2000. “Understanding the Domestic Impact of International Norms: A Research Agenda,” International Studies Review 2 (1): 65–87. Coser, Louis A. 1956. The Functions of Social Conflict. New York: Free Press. Crepaz, Markus. 2007. Trust Beyond Borders: Immigration, the Welfare State, and Identity in Modern Politics. Ann Arbor, MI: University of Michigan Press. Dassel, Kurt, and Eric Reinhardt. 1999. “Domestic Strife and the Initiation of Violence at Home and Abroad,” American Journal of Political Science 43 (1): 56–85. Davenport, Christian. 1995. “Multi-Dimensional Threat Perception and State Repression: An Inquiry into Why States Apply Negative Sanctions,” American Journal of Political Science 39 (3): 683–713. ———. 1996. “The Weight of the Past: Exploring Lagged Determinants of Political Repression,” Political Research Quarterly 49 (2): 377–403. ———. 1999. “Human Rights and the Democratic Proposition,” Journal of Conflict Resolution 43 (1): 92–116. Davenport, Christian, and David A. Armstrong II. 2004. “Democracy and the Violation of Human Rights: A Statistical Analysis from 1976 to 1996,” American Journal of Political Science 48 (3): 538–554. Davies, Graeme A. M. 2002. “Domestic Strife and the Initiation of International Conflicts: A Directed Dyad Analysis, 1950–1982,” Journal of Conflict Resolution 46 (5): 672–692. Deist, Wilhelm. 1997. “The German Army, the Authoritarian Nation-State, and Total War.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 160–172. Cambridge: Cambridge University Press. De Rouen, Karl, and David Sobek. 2004. “The Dynamics of Civil War Duration and Outcome,” Journal of Peace Research 41 (3): 303–320. De Rouen, Karl, and Shaun Goldfinch. 2005. “Putting Numbers to Work: Implications for Violence Prevention,” Journal of Peace Research 42 (1): 27–45.

9780230113176_12_ref.indd 177

2/5/2011 10:57:34 AM

178

●

References

De Rouen, Karl R. Jr., 1995. “The Indirect Link: Politics, the Economy, and the Use of Force,” Journal of Conflict Resolution 39: 671–695. Doughty, Robert. 2005. Pyrrhic Victory: French Strategy and Operations in the Great War. Cambridge, MA: Belknap Press. Doyle, Michael W. 2005. “Three Pillars of the Liberal Peace,” American Political Science Review 99 (3): 463–466. Druckman, Daniel. 1994. “Nationalism, Patriotism, and Group Loyalty: A Social Psychological Perspective,” Mershon International Studies Review 38 (1): 43–68. Eidelberg, Paul. 1970. “Intellectual and Moral Anarchy in American Society,” Review of Politics 32 (1): 32–50. Elliott, Kimberly Ann. 1997. “Introduction.” In Corruption and the Global Economy, edited by Kimberly Ann Elliott, 1–12. Washington, D.C.: Institute for International Economics. Elkins, Zachary, and John Sides. 2007. “Can Institutions Build Unity in Multiethnic States?” American Political Science Review 101 (4): 693–708. Enterline, Andrew, and Kristian Gleditsch. 2000. “Threats, Opportunity, and Force: Repression and Diversion of Domestic Pressure, 1948–1982,” International Interactions 26: 1–53. Esty, Daniel C., Jack Goldstone, Ted Robert Gurr, Barbara Harff, Pamela T. Wurko, Alan N. Guner, and Robert S. Chen. 1998. State Failure Task Force Report: Phase II Findings. McLean, VA: Science Applications International Corporation. Fearon, James D., and David D. Laitin. 2003. “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97 (1): 75–90. Finnemore, Martha, and Kathryn Sikkink. 1998. “International Norm Dynamics and Political Change,” International Organization 52 (4): 887–917. Frankel, Jeffrey. 2000. “Globalization of the Economy.” In Governance in a Globalizing World, edited by Joseph S. Nye and John D. Donahue, 45–69. Washington, D.C.: Brookings Institution. Friedrich, Carl. 1966. “Political Pathology,” Political Quarterly 37: 70–85. ———. 1972. The Pathology of Politics: Violence, Betrayal, Corruption, Secrecy, and Propaganda. New York: Harper and Row. Frith, Uta, and Christopher D. Frith. 2003.”Development and Neurophysiology of Mentalizing.” Philosophical Transactions of the Royal Society B: Biological Sciences 358 (1431): 459–473. Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press. Garnham, David. 1986. “War-Proneness, War-Weariness, and Regime Type: 1816–1980,” Journal of Peace Research 23 (3): 279–289.

9780230113176_12_ref.indd 178

2/5/2011 10:57:35 AM

References

●

179

Garrett, Geoffrey. 2004. “Globalization’s Missing Middle,” Foreign Affairs 83 (6): 84–96. Gartzke, Erik. 2000. “Preferences and the Democratic Peace,” International Studies Quarterly 44 (2): 191–212. ———. 2007. “The Capitalist Peace,” American Journal of Political Science 51 (1): 166–191. Gates, Scott, Torbjorn L. Knutsen, and Jonathon W. Moses. 1996. “Democracy and Peace: A More Skeptical View,” Journal of Peace Research 33 (1): 1–10. Geertz, Clifford. 1995. “Primordial and Civic Ties.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 29–34. New York: Oxford University Press. Gellner, Ernest. 1995. “Nationalism and Modernization.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 55–63. New York: Oxford University Press. Gelpi, Christopher. 1997. “Democratic Diversions: Governmental Structure and the Externalization of Domestic Conflict,” Journal of Conflict Resolution 41 (2): 255–282. Giddens, Anthony. 1971. Capitalism and Modern Social Theory. Cambridge: Cambridge University Press. ———. 1995. “The Nation as Power-Container.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 34–36. New York: Oxford University Press. Gilpin, Robert. 2001. International Political Economy. Princeton, NJ: Princeton University Press. Girling, John. 1997. Corruption, Capitalism, and Democracy. New York: Routledge. Gleditsch, Kristian Skrede, Idean Salehyan, and Kenneth Schultz. 2007. “The ‘Civil Peace’: Exploring the Relationship Between Civil and International Conflict,” Paper Presented at International Studies Association 2007 Conference, Chicago. Gleditsch, Nils Petter, and Havard Hegre. 1997. “Peace and Democracy: Three Levels of Analysis,” Journal of Conflict Resolution 41: 283–310. Goemans, H. E. 2000a. “Fighting for Survival: The Fate of Leaders and the Duration of War,” Journal of Conflict Resolution 44 (5): 555–579. ———. 2000b. War and Punishment: The Causes of War Termination and the First World War. Princeton, NJ: Princeton University Press. Goertz, Gary, and Paul F. Diehl. 1992. “Toward a Theory of International Norms: Some Conceptual and Measurement Issues,” Journal of Conflict Resolution 36 (4): 634–664.

9780230113176_12_ref.indd 179

2/5/2011 10:57:35 AM

180

●

References

Goldsmith, Arthur A. 1999. “Slapping the Grasping Hand: Correlates of Political Corruption in Emerging Markets,” American Journal of Economics and Sociology 58 (4): 865–883. Gowa, Joanne. 1998. “Politics at the Water’s Edge: Parties, Voters, and the Use of Force Abroad,” International Organization 52: 307–324. Granovetter, M. S. 1973. “The Strength of Weak Ties,” American Sociological Review 78: 1360–1380. Greenfeld, Liah. 1995. “Types of European Nationalism.” In Nationalism, edited by John Hutchinson and Anthony D. Smith. New York: Oxford University Press. Grieco, Joseph M. 1988. “Anarchy and the Limits of Cooperation: A Realist Critique of the Newest Liberal Institutionalism,” International Organization 42 (3): 485–507. Guetzkow, Harold. 1957. “Isolation and Collaboration: A Partial Theory of Inter-Nation Relations,” Conflict Resolution 1 (1): 48–68. Gurr, Ted Robert. 2000. “Ethnic Warfare on the Wane,” Foreign Affairs 79: 52–64. Guseva, Alya, and Akos Rona-Tas. 2001. “Uncertainty, Risk, and Trust: Russian and American Credit Card Markets Compared,” American Sociological Review 66 (5): 623–646. Hafner-Burton, Emily. 2005. “Right or Robust: The Sensitive Nature of Repression to Globalization,” Journal of Peace Research 42 (6): 679–698. Hagan, Joe D. 1994. “Domestic Political Systems and War Proneness,” Mershon International Studies Review 38 (2): 183–207. Hall, Peter A. 1999. “Social Capital in Britain,” British Journal of Political Science 29 (3): 417–461. Hardin, Russell. 1996. “Trustworthiness,” Ethics 107 (1): 26–42. ———. 2001. “Conceptions and Explanations of Trust.” In Trust in Society, edited by Karen S. Cook, 3–39. New York: Russell Sage Foundation. Harris, Robert. 2003. Political Corruption: In and Beyond the Nation State. New York: Routledge. Haythornthwaite, Philip J. 1992. The World War One Sourcebook. London: Arms and Armour Press. Hegre, Havard. 2004. “Size Asymmetry, Trade, and Militarized Conflict,” Journal of Conflict Resolution 48 (3): 403–429. Hegre, Havard, and Nicholas Sambanis. 2006. “Sensitivity Analysis of Empirical Results on Civil War Onset,” Journal of Conflict Resolution 50 (4): 508–535. Heidenheimer, Arnold. 1989. “Perspectives on the Perception of Corruption.” In Political Corruption: A Handbook, edited by A Heidenheimer, M. Johnston, and V. LeVine, 149–165. New Brunswick: Transaction.

9780230113176_12_ref.indd 180

2/5/2011 10:57:35 AM

References

●

181

Heldt, Birger. 1999. “ Domestic Politics, Absolute Deprivation, and the Use of Armed Force in Interstate Territorial Disputes, 1950–1990,” Journal of Conflict Resolution 43 (4): 451–478. Heller, Mark A. 1980. “The Use and Abuse of Hobbes: The State of Nature in International Relations,” Polity 13 (1): 21–32. Hermann, Margaret G., and Charles W. Kegley, Jr. 1995. “Rethinking Democracy and International Peace: Perspectives from Political Psychology,” International Studies Quarterly 39 (4): 511–533. Herzfeld, M. 1992. The Social Production of Indifference: Exploring the Symbolic Roots of Western Bureaucracy. Chicago: University of Chicago Press. Hess, Gregory D., and Athanasios Orphanides. 1995. “War Politics: An Economic, Rational-Voter Framework,” American Economic Review 85: 828–846. ———. 2001. “War and Democracy,” Journal of Political Economy 109: 776–810. Hetherington, Marc J., and Suzanne Globetti. 2002. “Political Trust and Racial Policy Preferences,” American Journal of Political Science 46 (2): 253–275. Hobsbawm, Eric. 1995. “The Nation as Invented Tradition.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 76–83. New York: Oxford University Press. Hoffman, Christhard. 1997. “The Jewish Community in Germany, 1914–18.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 89–104. Cambridge: Cambridge University Press. Hogg, Michael A., Deborah J. Terry, and Katherine M. White. 1995. “A Tale of Two Theories: A Critical Comparison of Identity Theory with Social Identity Theory,” Social Psychology Quarterly 58 (4): 255–269. Hogg, Michael A., and Dominic Abrams. 1988. Social Identifications. A Social Psychology of Intergroup Relations and Group Processes. New York: Routledge. Holmes, Richard. 1984. The Road to Sedan. London: Swift Printers. Holsti, K. J. 1995. “War, Peace, and the State of the State,” International Political Science Review 16 (4): 319–339. Hooghe, Marc, and Dietlind Stolle. 2003. “Introduction: Generating Social Capital.” In Generating Social Capital, edited by Marc Hooghe and Dietlind Stolle, 1–18. New York: Palgrave Macmillan. Horne, John. 1997. “Remobilizing for Total War.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 1–21. Cambridge: Cambridge University Press.

9780230113176_12_ref.indd 181

2/5/2011 10:57:35 AM

182

●

References

Horne, John, and Alan Kramer. 2001. German Atrocities, 1914: A History of Denial. New Haven, CT: Yale University Press. Horowitz, Donald L. 1985. Ethnic Groups in Conflict. Berkeley: University of California Press. Hosking, Geoffrey. 1997. Russia: People and Empire. Cambridge, MA: Harvard University Press. Houghton, David Patrick. 1996. “The Role of Analogical Reasoning in Novel Foreign-Policy Situations,” British Journal of Political Science 26 (4): 523–552. Huntington, Samuel P. 1993. “The Clash of Civilizations?” Foreign Affairs 72 (3): 22–49. ———. 2006. The Clash of Civilizations: Remaking of World Order. New York: Simon and Schuster. Hurd, Ian. 1999. “Legitimacy and Authority in International Politics,” International Organization 53 (2): 379–408. Hutchinson, John. 1995. “Cultural Nationalism and Moral Regeneration.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 122–131. New York: Oxford University Press. Huth, Paul K., and Todd L. Allee. 2002. “Domestic Political Accountability and the Escalation and Settlement of International Disputes,” Journal of Conflict Resolution 46 (6): 754–790. Huysseune, Michel. 2003. “Institutions and Their Impact on Social Capital and Civic Culture: The Case of Italy.” In Generating Social Capital, edited by Marc Hooghe and Dietlind Stolle, 211–230. New York: Palgrave Macmillan. Inglehart, Ronald. 1997. Modernization and Postmodernization. Princeton, NJ: Princeton University Press. Iyengar, Shanto. 1980. “Trust, Efficacy and Political Reality: A Longitudinal Analysis of Indian High School Students,” Comparative Politics 13 (1): 37–51. Jackman, Robert W., and Ross A. Miller. 1996. “A Renaissance of Political Culture?” American Journal of Political Science 40: 632–659. Jackson, Robert. 2005. Classical and Modern Thought on International Relations: From Anarchy to Cosmopolis. New York: Palgrave Macmillan. James, Patrick, and Athanasios Hristoulas. 1994. “Domestic Politics and Foreign Policy: Evaluating a Model of Crisis Activity for the United States,” Journal of Politics 56 (2): 327–348. James, Patrick, and John R. Oneal. 1991. “Influence of Domestic and International Politics on the President’s Use of Force,” Journal of Conflict Resolution 35 (2): 307–332.

9780230113176_12_ref.indd 182

2/5/2011 10:57:35 AM

References

●

183

Jervis, Robert. 1968. “Hypotheses on Misperception,” World Politics 20: 454–479. ———. 1976. Perception and Misperception in International Politics. Princeton, NJ: Princeton University Press. ———. 1988. “War and Misperception,” Journal of Interdisciplinary History 18 (4): 675–700. ———. 2002. “Theories of War in an Era of Leading-Power Peace,” American Political Science Review 96 (1): 1–14. Jones, Karen. 1996. “Trust as an Affective Attitude,” Ethics 107 (1): 4–25. Kadera, Kelly M., Mark J. C. Crescenzi, and Megan L. Shannon. 2003. “Democratic Survival, Peace, and War in the International System,” American Journal of Political Science 47 (2): 234–247. Kahneman, Daniel, and Amos Tversky. 1979. “Prospect Theory: An Analysis of Decision-Making under Risk,” Econometrica 47 (2): 263–291. Kaufmann, Chaim. 1996. “Possible and Impossible Solutions to Ethnic Civil Wars,” International Security 20 (4): 136–175. Kaufman, Stuart J. 1996. “Spiraling to Ethnic War: Elites, Masses, and Moscow in Moldova’s Civil War,” International Security 21 (2): 108–138. Kedourie, Elie. 1995. “Nationalism and Self-Determination.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 205–209. New York: Oxford University Press. Kennedy, Paul. 1989. The Rise and Fall of the Great Powers. New York: Vintage Books. Kershaw, Ian. 1998. Hitler 1889–1936: Hubris. New York: W. W. Norton & Company. ———. 2000. Hitler 1936–1945: Nemesis. New York: W. W. Norton & Company. Kim, Woosang, and Bruce Bueno de Mesquita. 1995. “How Perceptions Influence the Risk of War,” International Studies Quarterly 39 (1): 51–65. Kinsella, David. 2005. “No Rest for the Democratic Peace,” American Political Science Review 99 (3): 453–457. Knack, Stephen. 2002. “Social Capital and the Quality of Government: Evidence from the States,” American Journal of Political Science 46 (4): 772–785. Knack, Stephen, and Philip Keefer. 1997. “Does Social Capital Have an Economic Payoff ? A Cross-Country Investigation,” Quarterly Journal of Economics 112 (4): 1251–1288.

9780230113176_12_ref.indd 183

2/5/2011 10:57:35 AM

184

●

References

Kohn, Hans. 1995. “Western and Eastern Nationalisms.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 162–165. New York: Oxford University Press. Kramer, Alan. 1997. “‘Wackes’ at War.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 125–144. Cambridge: Cambridge University Press. Kuznetsov, Andrei, and Olga Kuznetsova. 2003. “Institutions, Business and the State in Russia,” Europe-Asia Studies 55 (6): 907–922. Kydd, Andrew. 2000. “Trust, Reassurance, and Cooperation,” International Organization 54 (2): 325–357. Lake, David A. 1996. “Anarchy, Hierarchy, and the Variety of International Relations,” International Organization 50 (1): 1–33. Lau, Richard R., and David P. Redlawsk. 2001. “Advantages and Disadvantages of Cognitive Heuristics in Political Decision Making,” American Journal of Political Science 45 (4): 951–971. Lazar, Sian. 2005. “Citizens Despite the State: Everyday Corruption and Local Politics in El Alto, Bolivia.” In Corruption: Anthropological Perspectives, edited by Dieter Haller and Cris Shore, 212–228. Ann Arbor, MI: Pluto Press. Lebow, Richard Ned. 1985. “Miscalculation in the South Atlantic: The Origins of the Falklands War.” In Psychology and Deterrence, edited by Robert Jervis, Richard Ned Lebow, and Janice Stein, 89–124. Baltimore: Johns Hopkins University Press. Leeds, Brett Ashley, and David R. Davis. 1997. “Domestic Popular Vulnerability and International Disputes,” Journal of Conflict Resolution 41: 814–834. Letki, Natalia. 2004. “Socialization for Participation? Trust, Membership, and Democratization in East-Central Europe,” Political Research Quarterly 57 (4): 665–679. Levi, Margaret. 1996. “Social and Unsocial Capital,” Politics and Society 24: 45–55. Levy, Jack S. 1988. “Domestic Politics and War,” Journal of Interdisciplinary History 18 (4): 653–673. ———. 1989. “The Diversionary Theory of War: A Critique.” In Handbook of War Studies, edited by Manus A. Midlarsky, 259–288. Boston: Unwin Hyman. Lewis, David J., and Andrew Weigert. 1985. “Trust as a Social Reality,” Social Forces 63 (4): 967–985. Lincoln, W. Bruce. 1983. In War’s Dark Shadow. New York: Dial Press. Lovell, David W. 2005. “Corruption as a Transitional Phenomenon: Understanding Endemic Corruption in Post-Communist States.” In Corruption:

9780230113176_12_ref.indd 184

2/5/2011 10:57:35 AM

References

●

185

Anthropological Perspectives, edited by Dieter Haller and Cris Shore, 65–82. Ann Arbor, MI: Pluto Press. Mansfield, Edward D., and Jack Snyder. 2002. “Democratic Transitions, Institutional Strength, and War,” International Organization 56 (2): 297–337. Maoz, Zeev. 1998. “Realist and Cultural Critiques of the Democratic Peace: A Theoretical and Empirical Re-Assessment,” International Interactions 24: 3–89. ———. 2000. “The Street-Gangs of World Politics: The Origins, Management, and Termination of International Alliances.” In What Do We Know about War? Edited by John A. Vasquez, 114–141. New York: Rowman and Littlefield. ———. 2001. “Democratic Networks: Connecting National, Dyadic, and Systemic Levels of Analysis in the Study of Democracy and War”. In War in a Changing World, edited by Zeev Maoz and Azar Gat, 143–182. Ann Arbor, MI: University of Michigan Press. ———. 2005. Dyadic MID Dataset (version 2.0). University of California, Davis. Accessed July 12, 2007, http://psfaculty.ucdavis.edu/zmaoz/ dyadmid.html. Maoz, Zeev, and Bruce Russett. 1992. “Alliance, Contiguity, Wealth, and Political Stability: Is the Lack of Conflict among Democracies a Statistical Artifact?” International Interactions 17: 245–267. ———. 1993. “Normative and Structural Causes of Democratic Peace, 1946–1986,” American Political Science Review 87: 624–638. Mearsheimer, John. 1994. “The False Promise of International Institutions,” International Security 19 (3): 5–49. Meernik, James. 1994. “Presidential Decision-Making and the Political Use of Force,” International Studies Quarterly 38:121–138. ———. 2000. “Modeling International Crises and the Political Use of Military Force by the USA,” Journal of Peace Research 37: 547–562. Meernik, James, and Peter Waterman. 1996. “The Myth of the Diversionary Use of Force by American Presidents,” Political Research Quarterly 49: 573–590. Mercer, Jonathan. 1995. “Anarchy and Identity,” International Organization 49 (2): 229–252. ———. 1996. Reputation and International Politics. Ithaca, NY: Cornell University Press. Miller, David. 1989. “In What Sense Must Socialism Be Communitarian?” Social Philosophy and Policy 6 (Spring): 51–73. Milner, Helen. 1991. “The Assumption of Anarchy in International Relations: A Critique,” Review of International Studies 17 (1): 67–85.

9780230113176_12_ref.indd 185

2/5/2011 10:57:36 AM

186

●

References

Minorities at Risk Project. 2005. College Park, MD: Center for International Development and Conflict Management. http://www.cidcm.umd.edu/ mar/. Accessed October 14, 2007. Mintz, Alex, and Nehemia Geva. 1993. “Why Don’t Democracies Fight Each Other? An Experimental Study,” Journal of Conflict Resolution 37 (3): 484–503. ———. 1997. “The Polyheuristic Theory of Foreign Policy Decisionmaking.” In Decisionmaking on War and Peace: The Cognitive-Rational Debate, edited by Nehemia Geva and Alex Mintz, 81–102. Boulder, CO: Lynne Rienner Publishers. Mishler, William, and Richard Rose. 1997. “Trust, Distrust and Skepticism: Popular Evaluations of Civil and Political Institutions in Post-Communist Societies,” Journal of Politics 59 (2): 418–451. Mitchell, Sara McLaughlin, and Will H. Moore. 2002. “Presidential Uses of Force During the Cold War: Aggregation, Truncation, and Temporal Dynamics,” American Journal of Political Science 46: 438–452. Moravcsik, Andrew. 1997. “Taking Preferences Seriously: A Liberal Theory of International Politics,” International Organization 51 (4): 513–553. Morgan, T. Clifton, and Christopher J. Anderson. 1999. “Domestic Support and Diversionary External Conflict in Great Britain, 1950–1992,” Journal of Politics 61: 799–814. Morgan, T. Clifton, and Kenneth N. Bickers. 1992. “Domestic Discontent and the External Use of Force,” Journal of Conflict Resolution 36: 25–52. Morgenthau, Hans J. 1948. Politics Among Nations: The Struggle for Power and Peace. New York: Knopf. Most, Benjamin A., and Harvey Starr. 1986. Inquiry, Logic, and International Politics. Columbia: University of South Carolina Press. Mousseau, Michael. 2000. “Market Prosperity, Democratic Consolidation, and Democratic Peace,” Journal of Conflict Resolution 44 (4): 472–507. ———. 2009. “The Social Market Roots of Democratic Peace,” International Security 33 (4): 52–86. Mousseau, Michael, and Yuhang Shi. 1999. “A Test for Reverse Causality in the Democratic Peace Relationship,” Journal of Peace Research 36 (6): 639–663. Myrdal, Gunnar. 1970. The Challenge of World Poverty. New York: Random House. Naim, Moses. 2003. “Five Wars of Globalization,” Foreign Policy 134: 29–36. Newell, James L., and Martin J. Bull. 2003. “Introduction.” In Corruption in Contemporary Politics, edited by Martin J. Bull and James L. Newell, 1–17. New York: Palgrave Macmillan.

9780230113176_12_ref.indd 186

2/5/2011 10:57:36 AM

References

●

187

Nye, Joseph. 1967. “Corruption and Political Benefit: A Cost-Benefit Analysis,” American Political Science Review 61 (2): 417–427. ———. 1997. Understanding International Conflicts. New York: Longman. Oneal, John R., and Jaroslav Tir. 2006. “Does the Diversionary Use of Force Threaten the Democratic Peace? Assessing the Effect of Economic Growth on Interstate Conflict, 1921–2001,” International Studies Quarterly 50: 755–779. Orbell, John, and Robyn M. Davis. 1991. “A ‘Cognitive Miser’ Theory of Cooperators’ Advantage,” American Political Science Review 85 (2): 515–528. Oren, Ido. 1995. “The Subjectivity of the ‘Democratic’ Peace: Changing U.S. Perceptions of Imperial Germany,” International Security 20 (2): 147–184. Organski, A. F. K. 1958. World Politics. New York: Knopf. Organski, A. F. K. and Jack Kugler. 1977. “The Costs of Major Wars: The Phoenix Factor,” American Political Science Review 71 (4): 134–166. O’Toole, George. 1984. The Spanish War—An American Epic, 1898. New York: W. W. Norton and Company. Orwell, George. 2000 [1945]. “Notes on Nationalism.” In The Collected Essays, Journalism, and Letters of George Orwell, Vol. 3, edited by Sonia Orwell and Ian Angus. Boston: Nonpareil Books. Ostrom, Charles, and Brian Job. 1986. “The President and the Political Use of Force,” American Political Science Review 80 (2): 541–566. Owen, John M. 1994. “How Liberalism Produces Democratic Peace,” International Security, 19: 87–125. Parks, Craig D., Robert F. Henager, and Shawn D. Scamahorn. 1996. “Trust and Reactions to Messages of Intent in Social Dilemmas,” Journal of Conflict Resolution 40 (1): 134–151. Peet, Curtis, and Marc V. Simon. 2000. “Social Purpose and Militarized Disputes,” Journal of Peace Research 37 (6): 659–677. Peceny, Mark, Caroline C. Beer, and Shannon Sanchez-Terry. 2002. “Dictatorial Peace?” American Political Science Review 96 (1): 15–26. Persell, Caroline Hodges, Adam Green, and Liena Gurevich. 2001. “Civil Society, Economic Distress, and Social Tolerance,” Sociological Forum 16 (2): 203–230. Pickering, Jeffrey, and Emizet F. Kisangani. 2005. “Democracy and Diversionary Military Intervention: Reassessing Regime Type and the Diversionary Hypothesis,” International Studies Quarterly 49: 23–43. Poe, Steven C., C. Neal Tate, and Linda Camp Keith. 1999. “Repression of the Human Right to Personal Integrity Revisited: A Global Cross-National

9780230113176_12_ref.indd 187

2/5/2011 10:57:36 AM

188

●

References

Study Covering the Years 1976–1993,” International Studies Quarterly 43 (2): 291–313. Porch, Douglas. 1981. March to the Marne. Cambridge: Cambridge University Press. Posen, Barry. 1993. “The Security Dilemma and Ethnic Conflict,” Survival 35 (1): 27–47. Powell, Robert. 1993. “Guns, Butter, and Anarchy,” American Political Science Review 87 (1): 115–132. Putnam, Robert D. 1988. “Diplomacy and Domestic Politics: The Logic of Two-Level Games,” International Organization 42: 427–460. ———. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press. ———. 1995. “Tuning In, Tuning Out: The Strange Disappearance of Social Capital in America,” Political Science and Politics 28 (4): 664–683. ———. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Ramnani, Narender, and R. Christopher Miall. 2004. “A System in the Human Brain for Predicting the Actions of Others,” Nature Neuroscience 7 (1): 85–90. Rasler, Karen A., and William R. Thompson. 2006. “Contested Territory, Strategic Rivalries, and Conflict Escalation,” International Studies Quarterly 50 (1): 145–168. Rawls, John. 1999. The Law of Peoples. Cambridge, MA: Harvard University Press. Regan, Patrick M. 1994. Organizing Societies for War. The Process and Consequences of Societal Militarization. Westport: Praeger. Regan, Patrick, and Errol Henderson. 2002. “Democracy, Threats, and Political Repression in Developing Countries: Are Democracies Inherently Less Violent?” Third World Quarterly 23 (1): 119–136. Reiter, Dan, and Allan C. Stam III. 1998. “Democracy, War Initiation, and Victory,” American Political Science Review 92 (2): 377–389. Renan, Ernest. 1995. “Que’est-ce un nation?” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 17–18. New York: Oxford University Press. Richards, Diana, T. Clifton Morgan, Rick K. Wilson, Valerie L. Schwebach, and Garry D. Young. 1993. “Good Times, Bad Times, and the Diversionary Use of Force: A Tale of Some Not-So-Free Agents,” Journal of Conflict Resolution 37: 504–535. Richmond, Anthony H. 1995. “Ethnic Nationalism and Post-Industrialism.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 289–300. New York: Oxford University Press.

9780230113176_12_ref.indd 188

2/5/2011 10:57:36 AM

References

●

189

Rioux, Jean-Sebastien. 1998. “A Crisis-Based Evaluation of the Democratic Peace Proposition,” Canadian Journal of Political Science 31 (2): 263–283. Robert, Jean Louis. 1997. “Mobilizing Labour and Socialist Militants in Paris.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 73–88. Cambridge: Cambridge University Press. Robinson, Eric. 2001. “Reading and Misreading the Ancient Evidence for Democratic Peace,” Journal of Peace Research 38 (5): 593–608. Rodrik, Dani. 2001. “Trading in Illusions,” Foreign Policy 123: 54–62. Roe, Paul. 1992. “The Intrastate Security Dilemma: Ethnic Conflict as a ‘Tragedy’?” Journal of Peace Research 36 (2): 183–202. Rosato, Sebastian. 2003. “The Flawed Logic of Democratic Peace Theory,” American Political Science Review 97 (4): 585–602. ———. 2005. “Explaining the Democratic Peace,” American Political Science Review 99 (3): 467–472. Rosenfeld, Richard, Steven F. Messner, and Eric P. Baumer. 2001. “Social Capital and Homicide,” Social Forces 80 (1): 283–310. Rothenberg, Gunther. 1976. The Army of Francis Joseph. West Lafayette, IN: Purdue University Press. ———. 2006. The Napoleonic Wars. New York: Harper Paperbacks. Rothstein, Bo, and Dietlind Stolle. 2003. Social Capital, Impartiality, and the Welfare State: An Institutional Approach. In Generating Social Capital, edited by Marc Hooghe and Dietlind Stolle, 191–210. New York: Palgrave Macmillan. ———. 2008. “The State and Social Capital: An Institutional Theory of Social Trust,” Comparative Politics 40 (4): 441–446, 456–458. Roy, Denny. 1993. “Neorealism and Kant: No Pacific Union,” Journal of Peace Research 30 (4): 451–453. Rummel, Rudolph J. 1997. Power Kills: Democracy as a Method of Nonviolence. New Brunswick: Transaction. Russett, Bruce. 1990. “Economic Decline, Electoral Pressure, and the Initiation of Interstate Conflict.” In Prisoners of War? Ed. Charles S. Gochman, and Alan Ned Sabrosky, 123–140. Lexington: Lexington Books. ———. 1993. Grasping the Democratic Peace. Princeton, NJ: Princeton University Press. Russett, Bruce M., and Harvey Starr. 2000. “From Democratic Peace to Kantian Peace: Democracy and Conflict in the International System.” In Handbook of War Studies II, edited by Manus I. Midlarsky, 93–128. Ann Arbor, MI: Michigan University Press.

9780230113176_12_ref.indd 189

2/5/2011 10:57:36 AM

190

●

References

Saideman, Stephen M., and R. William Ayres. 2000. “Determining the Causes of Irredentism: Logit Analyses of Minorities at Risk Data from the 1980s and the 1990s,” Journal of Politics 62 (4): 1126–1144. Sambanis, Nicholas. 2000. “Partition as a Solution to Ethnic War. An Empirical Critique of the Theoretical Literature,” World Politics 52: 437–483. Sandholtz, Wayne, and William Koetzle. 2000. “Accounting for Corruption: Economic Structure, Democracy, and Trade,” International Studies Quarterly 44 (1): 31–50. Schulze, Hagen. 1998. States, Nations, and Nationalism. Translated by Wiliam E. Yuill. Oxford: Blackwell. Scott, Bruce R. 2001. “The Great Divide in the Global Village,” Foreign Affairs 80 (1): 160–177. Seligson, Mitchell A. 2002. “The Impact of Corruption on Regime Legitimacy: A Comparative Study of Four Latin American Countries,” Journal of Politics 64 (2): 408–433. Seton-Watson, Hugh. 1995. “Old and New Nations.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 134–137. New York: Oxford University Press. Shore, Cris, and Dieter Haller. 2005. “Introduction—Sharp Practice: Anthropology and the Study of Corruption.” In Corruption: Anthropological Perspectives, edited by Dieter Haller and Cris Shore, 1–29. Ann Arbor, MI: Pluto Press. Simmel, Georg. 1955. Conflict. Glencoe: The Free Press. Simon, Marc V., and Harvey Starr. 2000. “Two-Level Security Management and the Prospects for New Democracies: A Simulation Analysis,” International Studies Quarterly 44 (3): 391–422. Smith, Alastair. 1996. “Diversionary Foreign Policy in Democratic Systems,” International Studies Quarterly 40: 133–153. Smith, Anthony D. 1995. “The Crisis of Dual Legitimation.” In Nationalism, edited by John Hutchinson and Anthony D. Smith. New York: Oxford University Press. Smith, Leonard. 1997. Remobilizing the Citizen Soldier.” In State, Society, and Mobilization in Europe During the First World War, edited by John Horne, 144–159. Cambridge: Cambridge University Press. Snyder, Jack. 2002. “Anarchy and Culture: Insights from the Anthropology of War,” International Organization 56 (1): 7–45. Sorensen, Georg. 1992. “Kant and Processes of Democratization: Consequences for Neorealist Thought,” Journal of Peace Research 29 (4): 397–414. Sorli, Mirjam E., Nils Petter Gleditsch, and Havard Strand. 2005. “Why Is There So Much Conflict in the Middle East?” Journal of Conflict Resolution 49 (1): 141–165.

9780230113176_12_ref.indd 190

2/5/2011 10:57:36 AM

References

●

191

Spinoza, Benedictus de. 2000. Political Treatise. Translated by Samuel Shirley. Indianapolis: Hackett Publications. Starr, Harvey. 1997. “Democracy and Integration: Why Democracies Don’t Fight Each Other,” Journal of Peace Research 34 (2): 153–162. Stein, Janice Gross, and David A. Welch. 1997. “Rational and Psychological Approaches to the Study of International Conflict: Comparative Strengths and Weaknesses.” In Decisionmaking on War and Peace: The CognitiveRational Debate, edited by Nehemia Geva and Alex Mintz, 51–77. Boulder, CO: Lynne Rienner Publishers. Stets, Jan E., and Peter J. Burke. 2000. “Identity Theory and Social Identity Theory,” Social Psychology Quarterly 63 (3): 224–237. Stinnett, Douglas M., Jaroslav Tir, Philip Schafer, Paul F. Diehl, and Charles Gochman. 2002. “The Correlates of War Project Direct Contiguity Data, Version 3,” Conflict Management and Peace Science 19 (2): 58–66. Stolle, Dietlind. 2003. “The Sources of Social Capital.” In Generating Social Capital, edited by Marc Hooghe and Dietlind Stolle, 19–42. New York: Palgrave Macmillan. Suny, Ronald Grigor. 1999. “Provisional Stabilities: The Politics of Identities in Post-Soviet Eurasia,” International Security 24 (3): 139–178. Sztompka, Piotr. 1999. Trust: A Sociological Theory. Cambridge: Cambridge University Press. Tajfel, Henri. 1970. “Experiments in Intergroup Discrimination,” Scientific American 223 (5): 96–102. ———. 1978. Differentiation Between Social Groups: Strides in the Social Psychology of Intergroup Relations. London: Academic Press. ———. 1981. Human Groups and Social Categories. Cambridge: Cambridge University Press. ———. 1982. Social Identity and Group Relations. Cambridge: Cambridge University Press. Tajfel, Henri, and John C. Turner. 1986 [2004]. “The Social Identity Theory of Intergroup Behavior.” In Political Psychology, edited by John Jost and Jim Sidanius, 276–293. New York: Psychology Press. Tarar, Ahmer. 2006. “Diversionary Incentives and the Bargaining Approach to War,” International Studies Quarterly 50: 169–188. Taras, Ray, and Rajat Ganguly. 2002. Understanding Ethnic Conflict. New York: Longman. Taylor, A. J. P. 1980. The Struggle for Mastery in Europe, 1848–1918. 2nd ed. Oxford: Oxford University Press. Terhune, Kenneth W. 1965. “Nationalistic Aspiration, Loyalty, and Internationalism,” Journal of Peace Research 2 (3): 277–287.

9780230113176_12_ref.indd 191

2/5/2011 10:57:37 AM

192

●

References

Thomas, Laurence. 1996. “Becoming an Evil Society: The Self and Strangers,” Political Theory 24 (2): 271–294. Tilly, Charles. 1985. “War Making and State Making as Organized Crime.” In Bringing the State Back In, edited by Peter Evans, Peter B., Dietrich Rueschemeyer, and Theda Skocpol, 169–192. Cambridge: Cambridge University Press. Tir, Jaroslav. 2005. “Keeping the Peace after Secession,” Journal of Conflict Resolution 49 (5): 713–741. Tombs, Robert. 1996. France 1814–1914. New York: Longman. Transparency International. 2007. Transparency International Corruption Perception Index. Accessed March 21, 2007, http://www.transparency.org. Uslaner, Eric M. 2000. “Producing and Consuming Trust,” Political Science Quarterly 115 (4): 569–590. ———. 2002. The Moral Foundations of Trust. Cambridge: Cambridge University Press. ———. 2003. “Trust, Democracy, and Governance: Can Government Policies Influence Generalized Trust?” In Generating Social Capital, edited by Marc Hooghe and Dietlind Stolle, 171–190. New York: Palgrave Macmillan. Van Belle, Douglas A. 1997. “Press Freedom and the Democratic Peace,” Journal of Peace Research 34 (4): 405–414. van Evera, Stephen. 1994. “Hypotheses on Nationalism and War,” International Security 18 (4): 5–39. Vanhanen, Tatu. 1991. The Politics of Ethnic Nepotism. New Delhi: Sterling. ———. 1999. “Domestic Ethnic Conflict and Ethnic Nepotism: A Comparative Analysis,” Journal of Peace Research 36 (1): 55–73. Vaughan, Graham M., Henri Tajfel, and Jennifer Williams. 1981. “Bias in Reward Allocation in an Intergroup and an Interpersonal Context,” Social Psychology Quarterly 44 (1): 37–42. Verhey, Jeffrey. 2000. The Spirit of 1914. Cambridge: Cambridge University Press. Vinci, Anthony. 2008. “Anarchy, Failed States, and Armed Groups: Reconsidering Conventional Analysis,” International Studies Quarterly 52: 295–314. Waltz, Kenneth N. 1979. Theory of International Politics. Reading, MA: Addison-Wesley. ———. 1988. “The Origins of War in Neorealist Theory,” Journal of Interdisciplinary History 18 (4): 615–628. ———. 1999. “Globalization and Governance,” PS: Political Science and Politics 32 (4): 693–700.

9780230113176_12_ref.indd 192

2/5/2011 10:57:37 AM

References

●

193

Warburton, John. 2001. “Corruption as a Social Process: From Dyads to Networks.” In Corruption and Anti-Corruption, edited by Peter Larmour and Nick Wolanin, 221–237. Canberra: Asia Pacific Press. Warren, Mark E. 2004. “What Does Corruption Mean in a Democracy?” American Journal of Political Science 48 (2): 328–343. Weart, Spencer R. 2001. “Remarks on the Ancient Evidence for Democratic Peace,” Journal of Peace Research 38 (5): 609–613. Weber, Eugen. 1976. Peasants into Frenchmen. Stanford: Stanford University Press. Weber, Max. 1995. “The Nation.” In Nationalism, edited by John Hutchinson and Anthony D. Smith, 21–26. New York: Oxford University Press. Wendt, Alexander. 1992. “Anarchy Is What States Make of It: The Social Construction of Power Politics,” International Organization 46 (2): 391–425. ———. 1994. “Collective Identity Formation and the International State,” American Political Science Review 88 (2): 384–396. ———. 1995. “Constructing International Politics,” International Security 20 (1): 71–81. ———. 1999. Social Theory of International Politics. Cambridge: Cambridge University Press. Wendt, Alexander, and Daniel Friedheim. 1995. “Hierarchy under Anarchy: Informal Empire and the East German State,” International Organization 49 (4): 689–724. Werner, Suzanne. 2000. “The Effects of Political Similarity on the Onset of Militarized Disputes, 1816–1985,” Political Research Quarterly 53 (2): 343–374. White, Jonathan R. 1996. The Prussian Army. Lanham: University Press of America. Williams, Robert. 2003. “Political Corruption in the United States.” In Corruption in Contemporary Politics, edited by Martin J. Bull and James L. Newell, 66–78. New York: Palgrave Macmillan. Wolf, Martin. 2005. “Will Globalization Survive?” The Third Whitman Lecture, Institute for International Economics Washington, D.C. Wolfson, Murray, Patrick James, and Eric Stolberg. 1991. “In a World of Cannibals Everyone Votes for War: Democracy and Peace Reconsidered.” In The Political Economy of War and Peace, edited by Murray Wolfson, 155–176. Boston: Kluwer Academic Publishers. World Bank. 2005. World Development Indicators. Accessed March 21, 2007, http://data.worldbank.org/.

9780230113176_12_ref.indd 193

2/5/2011 10:57:37 AM

194

●

References

World Values Study Group. 2000. World Values Survey, 1994–1999. Ann Arbor, MI: Institute for Social Research. Accessed March 18, 2007, http://www.worldvaluessurvey.org/. Yamagishi, Toshio, and Toko Kiyonari. 2000. “The Group as the Container of Generalized Reciprocity,” Social Psychology Quarterly 63 (2): 116–132. Zimbardo, Philip. 2008. The Lucifer Effect. New York: Random House. Zinn, Dorothy Louise. 2005. “Afterword—Anthropology and Corruption: The State of the Art.” In Corruption: Anthropological Perspectives, edited by Dieter Haller and Cris Shore, 229–242. Ann Arbor, MI: Pluto Press.

9780230113176_12_ref.indd 194

2/5/2011 10:57:37 AM

Index

aggression averting 158 international institutions and 30 likelihood of becoming a victim of 8, 123, 158 repressed 64 vulnerability to 106 aggressiveness and links to democracy 15 of state leaders 15 alien encounter 96 and difficulty of testing 82 and heuristics 77 outcome influenced by social trust 81 as an unprecedented situation 79 usefulness 19 alliances 13, 30, 35, 45, 106, 107, 113–115, 117–122, 132, 144, 148, 152, 153, 157 altruism 30 anarchy 8, 28, 29 domestic 6, 7, 27, 29, 35–41, 59, 62, 65, 91–93, 105, 150, 160, 167–171 Hobbesian culture of 32–33, 39 international 6, 7, 29, 35–37, 77, 166–169 and international behavior 30–32, 34, 37 Kantian culture of 32–33 links between international and domestic 33, 35, 168–169

9780230113176_13_ind.indd 195

Lockean culture of 32–33, 39 and national identity 41–42, 45 Argentina 21, 22, 121 Ausgleich 142 Austro-Hungarian Empire conflict behavior 133, 142, 147–149, 154, 156–158 international standing 142 national identity 141–142, 144 social trust level 158, 162 state weakness 141, 144, 162 belligerence and regime type 3 and social trust 124 bounded rationality 72, 78 Burgfrieden 139 capabilities and conflict initiation 135, 151, 155, 156, 168 relative 73, 78, 106, 113–115, 117–122, 124–125, 132, 153 state 103, 106 civil liberties and social trust 53 civil society 4, 6 fraying of 3 negative aspects of 56–58, 96 and social trust 49, 50, 55, 58, 68 clash of civilizations 170 Clausewitz, Karl von 129 cognitive miser 78

2/5/2011 11:01:09 AM

196

●

Index

cohesion group 48, 65, 68–70, 73 military 69, 73, 142 national 2–5, 25–26, 37, 50, 66, 84, 86–88, 92, 121, 129, 132, 134, 139, 141, 143, 151, 155, 158, 172 community, imagined 41, 62 community, moral 49, 56–57, 87 as in-group 67 conflict initiation 104–105, 107, 112–124, 132–134, 146, 153, 158–159, 161, 168 and democracy 16–17 factors influencing 7 conflict-proneness 15–16, 18, 20, 23, 26, 44, 75, 83–84, 105, 155, 160 corruption 61, 91–94, 98 and domestic anarchy 91–93, 119, 146 and mistrust 61, 91, 93 and social trust 93, 96, 123 Corruption Perception Index (CPI) 91, 97–101, 116 democracy ancient 11, 17 and autocracy 6, 20, 103, 132 and conflict 4, 9, 11, 15, 17–18, 20, 23, 26, 103, 125, 127, 158–159 and corruption 98 emerging 11, 27, 52, 98, 159–160 joint/shared 6, 16–17, 19, 105, 112–121 as a norm 36 pacifying effect of 4, 16, 19, 103, 15 and perception 10–12, 15, 66, 105 preconditions for 18–19, 52–53 score 103 and trust 52–53, 98, 116, 160 democratic peace theory 2–5, 7–12, 14–16, 18–20, 26–27, 47, 67, 98, 104–105, 113–114, 126, 159–160 critiques of 26, 105

9780230113176_13_ind.indd 196

diversionary war theory 2–3, 5, 7–9, 11–12, 20–25, 35, 65, 85–86, 114–115, 159–164 and behavior 21–25, 47, 86, 123, 124 critiques 7, 21–25, 86, 106, 114, 161 difficulties in validating 21–22, 86, 163 and incentives 11, 21–26, 85, 106, 121, 160–164 and regime type 26 Durkheim, Emile 50, 129 enduring rivalry 22 entente cordiale 13, 150, 153, 157 European confederation 36, 45 European Union 32, 45 Falklands War, the 21–22, 122 Ferdinand, Archduke 134, 147 framing 165–166 France conflict behavior 17, 134, 142, 144, 149–150, 152–156, 163 corruption level 99 democracy in 132 and Germany 43, 135–137, 148, 155, 163 image in the United States 13–14 international standing 134, 140, 142, 151–155 legitimacy 144, 147, 155 Napoleonic 171 national identity of 40, 94, 96, 131–132, 136–137, 143–145 social trust level 154 game, two-level 34 generalized social trust. See social trust Germany conflict behavior 31, 66, 131, 135–139, 142, 148–150, 153, 155–156

2/5/2011 11:01:09 AM

Index civic associations 57 corruption level 99 image of 13 international standing 31, 134–136, 151, 154 legitimacy 137, 139, 163 national identity of 43, 136–138 Nazi 13–14, 30, 57, 66, 162, social trust level 154 Wilhelmine 13, 163 Gini Index 54 globalization 127–128, 169–171 government as generator of social trust 58–61, 63, 83, 93–94, 98, 101, 112, 122, 139, 146, 159–160, 164 loyalty to 27, 51–52, 62, 142–145, 150 performance 5–8, 19, 26–29, 32, 34, 36, 38, 39, 42, 44–45, 47, 51–52, 54, 58–62, 65–67, 77, 83, 87–88, 91–96, 98, 101, 112, 120, 122, 139, 146, 159–160, 164,167–168 popularity 3, 23, 25, 48, 52, 159, 161 trust in 51–52, 54 world 29, 32, 37, 170 Granovetter, Mark 67 Great Britain conflict behavior 13, 17, 21–22, 145, 149–153, 156–157, 170 democracy in 132 international standing 128, 134, 136, 143, 145, 150–153, 155 legitimacy 145 national identity of 145 social capital in 59 social trust level 146 Hitler, Adolf 139, 162 Holsti, K.J. 27, 97–98 Huntington, Samuel P 67, 170 Hussein, Saddam 87–88, 126

9780230113176_13_ind.indd 197

●

197

identity ethnic 38, 42, 83–84 group 8, 43, 63–65, 67–69, 71 individual 41–42, 63–65 national 3, 7, 26, 39–44, 47, 62–63, 67, 74, 83–84, 95, 130, 138, 141, 143, 146, 154 primordial sources of 39, 84 as product of conflict 62–65, 67–69, 87, 124, 137, 154 as product of state action 40–42, 45, 47, 62, 64–65, 72, 74, 76, 83–84, 87, 130, 138, 141, 143, 146, 164, 167 and social trust 67, 83, 87 transnational 45 identity theory 68–69, 71–72 inequality 54, 58–60, 73, 111, 128, 170 in-group 39, 65, 87, 133, 162 conformity to 65 preference 65, 67, 111, 165 industrial revolution 50, 128–129 Iraq, invasion of, 2003 12, 14, 76, 85, 87, 126 Iraqi invasion of Kuwait, 1990 97 Kant, Immanuel 17, 19, 32, 35–36 Kulturkampf 138–139 liberal peace 17, 19, 127, 167 militarized interstate dispute (MID) 9 initiation 104, 106, 113–120, 125, 161, 163, 166 nonfatal 104 targeting 104, 116, 120–123, 125 minimal group experiment 69 minority at risk (MAR) 102, 104, 109, 111–112 misperception 14–15, 78 mistrust 50–52, 54–57, 60–61, 65–66, 71, 75, 83, 86, 123–125, 161, 163 Morgenthau, Hans 34, 38

2/5/2011 11:01:09 AM

198

●

Index

Moravcsik, Andrew 123, 168 motivated bias 14–15, 86, 121 nationalism 8, 42. Also see identity, national civic 146 and conflict 44, 137 ethnic 146–147 and patriotism 43–44, 67 relative strength of 42, 44–45 and social trust 84 nation-building 41, 62, 126, 129–130, 133, 143–144, 150–151, 169, 171–172 neoliberal theory 34, 83, 122, 168 neorealist theory 32 North Atlantic Treaty Organization (NATO) 14, 32, 66, 157 optimism 49, 60 out-group 67 dehumanization of 96 discrimination 64, 69, 72, 91, 96, 111 hostility against 64–67, 72, 165 perception of government as 129 particularism 63, 67 perception 5, 7–8, 10, 21, 35, 49, 68–69, 86, 166–167 and conflict 15, 21, 35, 111, 124, 136, 155–157, 171 and democratic peace 10–12, 14–15, 67, 105, 124 elite 11–12, 15 mass 11, 15 and misperception 13, 15, 78–79 mutual 11–12 and social trust 7, 49, 54–55, 69, 95, 111, 125, 166–167, 169 variability of 13–14 phoenix factor 31 power transition theory 124, 152–153, 155

9780230113176_13_ind.indd 198

prospect theory 80, 165 public goods 35, 39, 40–41, 63, 65, 91–93, 110, 128–131, 143, 159, 162–163 Putnam, Robert 2–4, 6, 55–59, 94, 96, 127 regime survival 22, 26, 34 regime type, significance of 7, 13, 23, 28, 37–38, 51, 103, 110, 112, 132 and conflict 3–4, 11, 15–19, 23, 26, 112, 119, 159 and social trust 52–53, 55, 116, 159 Renan, Ernest 45, 132, 143 rally around the flag effect 5, 11–13, 22, 86, 139, 155, 161, 163 failure to achieve 25, 86, 143 fleeting nature 87 realist theory 16, 31–34, 36, 50, 78, 154, 168 critiques of 34–37 offensive and defensive 154 structural 29–30, 32, 36–37, 77 repression, political 5, 9, 20, 22, 52, 64, 84, 88, 126, 155, 163 reputational effects 15–16, 24, 77 Roosevelt, Franklin D. 127 Roosevelt, Theodore 126 role fulfillment 68, 71–72, 74 and counter-roles 72 Russian Empire, The 13 conflict behavior 13, 131, 140, 142, 144, 147–151, 156–157 international standing 134–135, 141, 146, 154 national identity 131, 146, 150 political reforms 132, 171 political unrest 147, 149–150, 158 and social trust 146, 154, 158 Russian Federation, the and social trust 54, 101 and World War II nostalgia 3 Russo-Japanese War 95, 149–150, 154–155, 171

2/5/2011 11:01:09 AM

Index scapegoating 24, 65, 161, 164 Jews 139 out-groups 65 schemata 61, 79, 82, 167 security dilemma 30, 33–34 domestic 6, 27, 38 self-esteem and diversionary motives 25 and group membership 64–65 and role fulfillment 68, 71–73 self-help 29, 33, 37–39, 45, 74, 93, 109–111, 125–126, 134, 161, 168–169, 171 shadow of the future 48, 54, 68, 77 signaling 77, 156 peaceful intent 77, 88, 121, 123–124, 152, 156–158 in a prisoner’s dilemma game 123 resolve 16 trustworthiness 70–72, 88, 121 social capital 4, 6, 8, 48, 55, 59–60, 96 “bad” 57 social identity theory 15, 68–69, 71–72, 76, 78, 83–84 social trust 4 associational origin of 55–58 and democracy 52–53, 116 definition of 47–48 and economic development 53–54 as an element of social capital 48 encapsulated 47–48, 91, 96 generalized reciprocity norm 49–51, 64, 67, 75, 91, 96, 130 governance origin of 8, 45, 52, 58–62, 91–93, 146, 151, 159–160, 167–168, 171–172 as a heuristic 48 and international interactions 4–6, 8, 75–77, 82–89, 110, 113–126, 128–129, 133, 136, 138, 142, 149, 151, 157–158, 159–165, 168, 170–172 and national identity 45, 61–63,

9780230113176_13_ind.indd 199

●

199

67, 136, 138, 142, 146, 151, 154–155, 162–165, 171–172 particularized 48–50, 56–58, 63, 67–68, 74–75, 77, 83–84, 91, 93, 95–96, 111, 162 societal benefits of 48–52, 74–75 and state legitimacy 51–52, 63 socialization, political 56, 61, 64, 66, 83, 85, 94, 167 sovereignty, popular 4, 10 Spanish-American War 11, 13, 17, 27, 125–126 Stanford Prison Experiment 72 state failure 6, 27, 29, 35, 38, 52, 101, 128 state-society relationship 143, 158 Tajfel, Henri 69 Tocqueville, Alexis de 27 transaction costs 92 Treaty of Westphalia 128 trustworthiness 7, 47–48, 51, 54, 61, 70–71, 76, 81–83, 87–88, 93, 95–96, 161, 163, 166 Union Sacree 144 United States civic engagement 2, 27–28, 57–58 conflict behavior 11, 13, 17, 35, 66, 82, 125–126, 128 discrimination in 2 image of 14 international interactions of 14 international standing 39, 134–135, 140 level of social trust 39, 58, 99, 111, 126 national identity 67 unrest, political 24–25, 27, 143–144, 147, 158 Uslaner, Eric 47–48, 60–61 Waltz, Kenneth 32, 34, 36–38 war fever 13, 126, 150

2/5/2011 11:01:10 AM

200

●

Index

weak ties 67 Weber, Eugen 95, 129–131, 146, 171 welfare state 59, 62, 67, 73 Wendt, Alexander 32–33, 77, 81, 167

9780230113176_13_ind.indd 200

World Values Survey yellow journalism

95–97

13, 126

Zimbardo, Philip 72 Zola, Emile 131, 143

2/5/2011 11:01:10 AM