Book Search:  


Google full text of our books:


State Death:
The Politics and Geography of Conquest, Occupation, and Annexation
Tanisha M. Fazal

Book Description | Endorsements | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2007, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format

Chapter 1


Of all the states on the map of the world in 1816, nearly half no longer exist today. The fate of these states is the subject of this book. Both as the primary form of organization in international politics and as specific actors on the global stage, states have come to be the central pillar of international relations theory and politics. And while scholars and practitioners alike have questioned whether the state as an institution will survive,1 the survival of particular states is typically assumed and, further, assumed to be desirable. The conventional wisdom appears to be that states do not—and should not—die.

This perception is belied by both recent and distant history. The collapse of the Soviet Union was a defining moment of the late twentieth century, cementing an end to the Cold War. Less than fifty years earlier, World War II witnessed numerous attempts to “kill” other states and ended with the dismemberment of Germany. As any series of historical maps will reveal, international relations is in many ways captured by the emergence and disappearance of states in the international system.

This book addresses three main questions: (1) What is the historical record of state death? (2) Why do some states die while others survive? and (3) What explains the decline of violent state death after 1945? By “state death,” I mean the formal loss of control over foreign policy making to another state. This definition distinguishes state death from cases of extreme regime change, internal state collapse, or minor territorial exchanges. Typically, state death occurs when one state takes over another, or when a state breaks up into multiple, new states. For example, the partitioning of Germany after World War II constitutes a state death, but the Cuban Revolution of 1959 does not. The British conquest of Punjab in 1846 led to the death of the Punjabi state, but even though internal collapse left the state of Somalia without clear internal authority structures, Somalia continued to exist as a member of the international system in the 1990s. Although the argument advanced here has important implications for domestic political changes, the focus is on the survival of states in the international system.

The argument of this book is that geography is a key determinant of state death and survival—it endangers some states while securing others. Buffer states—states that lie between two other states engaged in rivalry —are particularly likely to die. Indeed, buffer states account for over 40 percent of state deaths. Each rival fears that its opponent will take over the buffer state between them and gain a strategic advantage. While both rivals might prefer to maintain the sovereignty of the buffer state so as to decrease the risk of war, the fear of losing strategic ground, coupled with the rivals’ inability to trust each other not to encroach on the buffer, generally signal the demise of buffer states.

Under certain conditions, however, the dynamics of rivalry can be mitigated such that a buffer state survives. If the rivals’ hands can be metaphorically “tied”—because their resources face serious and simultaneous constraints, because they must jointly meet another threat in a separate theater, or because buffer state sovereignty is guaranteed by even more powerful actors—buffer states can escape the fate of their fellows. But these conditions are likely to be (and historically have been) rare. They may occur with the coming of world wars, with revolution, or with the creation of new world orders. And even when buffers are protected by more powerful actors, the incentives to take them over do not disappear; the form of conquest merely changes. The post-1945 era provides an example of a superpower—the United States—sponsoring a rule against conquest that prevented rivals from taking over the buffers between them. The norm against conquest is an intervening variable that reshapes the way in which states seek to control other states. Thus, the bright-line prohibition against taking over other states has led to a proliferation of interventions to achieve ends formerly sought through conquest.


The issue of state survival lies at the heart of international relations theory and practice. International relations scholars typically take state survival to be a minimum, and primary, goal of states, statesmen, and citizens.2 But international relations (IR) scholarship lacks systematic, rigorous analyses of variation in state death and survival.3

The survival assumption is empirically verifiable. States and their citizens do seek survival.4 From Haile Selassie’s impassioned plea to the League of Nations in 1936 to the French resistance to German occupation during the Second World War, loyalty to state and homeland appears to be an extremely strong motivator, one to which people have often sacrificed their lives.5

This push for survival is justified; state death is more frequent than scholars have believed.6 Since 1816, 66 of 207 states have died. Fifty of these cases (over 75 percent) are instances of violent state death, which constitutes the narrower focus of this book. Violent state death occurs when one state uses military force to deprive another of its formal control over foreign policy making. Typically, violent state death occurs in the form of conquest or occupation. It is not unusual for state death to be a cause or a consequence of war. For example, the Iraqi annexation of Kuwait was the trigger for the Persian Gulf War, just as the takeover of Poland led Britain and France to join the Second World War. Following World War II, Germany was dismembered for almost five decades.

Recent events have highlighted the import of thinking seriously about state death. While the lack of scholarship on this topic may be attributed to the relative absence of state death—or, at least, violent state death— since 1945, the recent US occupation of Iraq may reverse this trend. Understanding the dynamics that may endanger some states while bolstering others has reemerged as a central problem for policy makers.

The Literature

State death is a touchstone issue for international relations theorists. I test my argument alongside three major hypotheses drawn from the international relations literature: a balancing argument drawn from realism; a social constructivist claim about the relationship between international legitimacy and state survival; and a set of arguments that suggest that states whose populations are likely to resist conquerors are most likely to survive.

State death plays a fundamental, yet underappreciated, role in neorealism. A basic premise of realist arguments is that states will balance against power; when faced with external threats, states should build up their militaries and seek alliances. Kenneth Waltz, in his famous explication of neorealism, works from the assumptions that states seek survival and that anarchy is the governing principle of the international system.7 The assumption of rationality is also frequently ascribed to Waltz’s work, but the rationality of states or state actors is not a neorealist assumption. Instead, neorealists avoid making a strict rationality assumption by making a selection argument: states that “do not help themselves, or who do so less effectively than others, will fail to prosper, will lay themselves open to danger, will suffer.”8 At the same time, neorealists assert that the death rate of states is low.9 If irrational behavior is punished by death, and the death rate of states is low, neorealists claim, then states must behave as if they were rational. Even if we take the assumptions as empirically correct, the logic behind the deduction is shaky at best. Not all states face equal threats and, further, power and alliances—which are the preferred form of balancing for neorealists—may attract (or indicate) rather than deter threats to survival.10

In addition, if selection pressures are as acute as neorealism suggests, the rate of state death should be high, not low.11 The fact that the rate of state death is higher than previous scholars have assumed does not, however, necessarily support the neorealist balancing argument. I show that the strong relationship between power, alliances, and state survival suggested by this argument does not exist. This finding does not necessarily lead to a conclusive rejection of the balancing argument. Death in the international system could take many forms, from a decline in power to a regime change to state death—I test the application of this argument only to punishment by state death. But my findings do illustrate that balancing behavior does not govern state death and survival.

Unlike realists, constructivists do not assume or conclude that states behave rationally. According to constructivists, states and state actors socially construct their world such that certain actions are permissible and others are forbidden. For example, David Strang and Alexander Wendt have argued that states confer degrees of international legitimacy on other (aspiring) state actors.12 Some of these actors are designated as being in the “in-group” while others are in the “out-group.” This distinction among actors is accompanied by different rules of behavior. To conquer and occupy an illegitimate state is permissible, while conquest of a “legitimate” state is taboo and will be punished. Using a new measure of international legitimacy, and by adding previously ignored states to standard data sets, I find considerable support for this argument. Indeed, the claim that more legitimate states are more likely to survive could be consistent with the argument that a norm against conquest has prevented violent state death after 1945. Particularly if the standards for international recognition of states have been more liberal in the post-1945 era, the general norm against conquest and the specific norm against conquest of legitimate states may have become congruent.

The issue of state death is also relevant for a third set of arguments from comparative politics and international relations theory—what I will call the “nationalist resistance” argument. Proponents of this argument have suggested that nationalistic societies will generate particularly high costs for states that conquer them.13 The resistance to conquest inspired in these particularly nationalistic states would make conquest untenable. Would-be conquerors, behaving strategically, thus avoid taking over states whose populations would generate high levels of resistance. The logic of this argument, as Peter Liberman suggests, may be incomplete because nationalism often coexists with industrialization.14 The infrastructure that characterizes industrialized states may actually make conquest and resource extraction easier, and resistance more difficult, than nationalist resistance theorists would suggest. Furthermore, signs of nationalism may generate a reaction opposite from what one would hope for, angering would-be conquerors and accelerating the process of state death.


The primary focus of this book is variation in the incidence of violent state death. Violent state death is the most common form of state death; one could argue that it is also the most pernicious form. Typically, violent state death occurs via conquest or annexation. Nonviolent state deaths, on the other hand, are usually attached to voluntary state unifications or dissolutions. While there is some overlap among these categories—Bavaria and Württemberg, for example, voluntarily acceded to the Prussian Confederation, but only after observing the conquest of nearby states like Hanover and Saxony—by and large, the processes of violent and nonviolent state death appear quite distinct. It would be surprising, for example, if the same causal explanation could account for the British takeover of the Indian princely states and the breakup of the Soviet Union.15 Because these types of events are so different, I have chosen to focus on the group of outcomes that is most common among state deaths.

I employ both quantitative and qualitative methods to test hypotheses derived from my argument as well as alternative hypotheses. Statistical analysis permits the uncovering of patterns, or correlations; it allows us to verify or disconfirm initial hunches. It also lends the benefit of systematic testing. At the same time, statistical analysis suffers from at least two weaknesses: first, not all variables are easily quantified and, thus, not all hypotheses can be tested using quantitative analysis; and second, while statistical results can show correlation, they typically do not uncover the causal processes that most intrigue political scientists. Even if I am able to show that buffer states are more likely to die than nonbuffer states, I also need to show that my argument explains these outcomes in particular cases. Mixing methods combines the best of both worlds: rigorous, systematic statistical analysis allows generalization, while case studies permit illustration of causal mechanisms.

Plan of the Book

The remainder of this book is divided into three main sections: “Patterns and Causes,” “Buffer State Death and Survival,” and “The Norm against Conquest and State Death after 1945.” Chapter 2, “Definitions and Patterns,” has two tasks. First, I describe in detail the coding criteria and decisions for variables used in this analysis that have not been used frequently in the international relations literature. I explain how and why I generated a new list of states. I also discuss the definition and coding of “state death,” going so far as to explain why certain borderline cases were excluded. Second, I present an overview of state death, taking on the questions of who, what, where, when, and how.

The “why” of state death is addressed in chapter 3, which is devoted primarily to presenting and explaining the argument that the dynamics of rivalry and geography determine state death and survival. I argue that geography, by which I mean location, is a prime predictor of state death— buffer states are particularly likely to die because rivals face strategic imperatives to take over the buffers that lie between them and are typically unable to make credible commitments to preserve buffers. At times, however, unusual circumstances may intervene that decrease the probability that rivals will take over buffer states. I lay out three such conditions: “tiring hands,” when rivals’ resources are simultaneously constrained such that they cannot take over the buffer; “diverted rivalry,” when the rivals must become temporary allies against a threat in another theater, creating incentives to cease fighting over the buffer so that necessary resources are not sapped; and when a particularly powerful state intervenes to guarantee the sovereignty of the buffer state, creating a situation whereby buffers are secure from conquest because rivals fear the costs of violating dictates issued by more powerful actors in the system. This third condition has characterized much of the post-1945 period. The United States in particular has supported a norm against conquest, altering other states’ incentives to take over their neighbors. I trace the evolution of the norm against conquest both in the United States and abroad, in part to understand to what degree it has been internalized and, further, to what degree it may be reversible. Chapter 3 concludes with a review of additional explanations for state death, particularly the balancing argument, the international legitimacy argument, and the nationalist resistance argument.

The heart of the book’s quantitative analysis is presented in chapter 4, which begins the section on buffer state death and survival. This chapter focuses on general explanations for state death, using duration analysis to test hypotheses about the role of buffer states, balancing behavior, international recognition, and nationalism with respect to state death. Analyses of variation in buffer state survival are also included.

Chapter 5 presents pre-1945 case studies of eighteenth-century Poland, the Dominican Republic in 1916, the Dominican Republic in 1870, and late nineteenth- and early twentieth-century Persia. The two variables used to select these cases were buffer state status and state death/survival. I sought to maximize variation on both the dependent and independent variables for each set of case studies. For each case, I ask several questions: Was the state a buffer state? Why did it die/survive? Did the state balance? And, to what extent were (would-be) conquerors concerned about the prospects of resistance? The results from these cases are extremely consistent, illustrating the import of the dynamics of rivalry to both state death and state survival, and challenging the notion that balancing behavior or nationalist rebellion could or did stave off conquest.

Eighteenth-century Poland represents a classic case of a prolonged, some would say agonizing, buffer state death. This case aptly illustrates the dynamics of rivalry that led to Poland’s demise in 1795. By contrast, Persia in the late nineteenth and early twentieth century is an exception to the rule that buffer states will die. Persia’s survival can be explained by extending the logic of the argument that buffer states are particularly likely to be taken over; rare, extenuating circumstances mitigated the Anglo-Russian rivalry that had repeatedly jeopardized the Persian state. In 1870, the United States considered annexing the Dominican Republic, but ultimately rejected this scheme. These events constitute the third case study, and can be explained by the absence of a strategic imperative to take over the Dominican Republic. The Dominican Republic survived precisely because it was not a buffer. Finally, forty-five years later, the US occupation of the Dominican Republic in 1916 appears, on the surface, to be a case of a nonbuffer state death. On examination, however, it becomes evident that the US occupation was driven by fear of German adventurism in the Caribbean; like Poland, the Dominican Republic was a victim of the strategic imperatives facing rivals.

While the empirical chapters begin to provide some answers regarding the causes of state death, they also raise important questions. Why is it that violent state death declined dramatically after 1945? Why are some states resurrected? The third section of this book explores these questions in the context of the norm against conquest that emerged in the early twentieth century.

Chapter 6 turns to the question of state resurrection, asking, why are only some state deaths reversed? A series of quantitative analyses applies a number of hypotheses on state death to state resurrection. I test to see whether the probability of resurrection increases in tandem with the strengthening of the norm against conquest. This argument is supported by results of the quantitative tests as well as case studies of the 1919 resurrection of Poland and 1924 resurrection of the Dominican Republic. Interestingly, the nationalist resistance argument also appears to explain state resurrection, even though it does not seem to shed light on state death generally.

Chapter 7 explores the norms argument further, first by testing corollary hypotheses meant to distinguish the norms argument from alternative explanations for the post-1945 decline in violent state death. While a variable for the post-1945 period is included in the primary statistical analysis in chapter 4, this variable on its own is a very blunt instrument, and could capture a number of other phenomena—such as the onset of the Cold War or the development of nuclear weapons—that coincide with this time period. One method to distinguish the predictions of these different arguments is to generate and test corollary hypotheses of each. Among others, extensions of the norms argument suggest: (1) a rise in reversals of violent state deaths in the twentieth century; (2) an increase in interventions to replace regimes and leaders when conquest is prohibited; and (3) an increase in the number of state collapses as leaders exploit state resources knowing that they face little or no risk of conquest. Each of these arguments is tested using quantitative analysis. The results of these tests consistently support the norms argument.

I turn next to four post-1945 cases: Poland and Hungary in 1956; the Dominican Republic in 1965; and Kuwait in 1990. The questions asked in this chapter are slightly different: Why did the state survive, suffer intervention, or die? Were (would-be) interveners concerned about violating the norm against conquest or about the resistance they might encounter? Again, both primary and secondary evidence confirms the notion that the norm against conquest, as opposed to a fear of increased nationalist resistance, constrained states from taking over their neighbors. Documents from Soviet-era archives are used to analyze the first set of cases— Poland and Hungary in 1956. Both states were buffer states during the Cold War, but the USSR intervened only in Hungary in 1956. This difference in outcome is explained by the fact that Hungary was “more” of a buffer state in 1956, making credible threats to defect from the Warsaw Pact. I also show that the probability of resistance was essentially equal in Hungary and Poland at the time. For the next case, the US intervention in the Dominican Republic in 1965, archival sources reveal that, as in 1916, the United States was very concerned about encroaching rivals, and that concerns about violating the norm against conquest prevailed in preventing a complete US takeover of the republic. The final case, that of the Iraqi annexation of Kuwait in 1990, initially appears to contradict both the buffer state and norms arguments. While Kuwait was not a buffer state,16 this case is the exception that proves the rule that the norm against conquest prevented violent state death after 1945. Simply put, the intent to enforce the norm was not signaled clearly to the Iraqi regime; the norm was violated because it was not clear that violation would be punished.

The final chapter summarizes the main arguments presented here and considers implications for international relations theory and policy. One of the principal theoretical implications of this book is that state death and survival are primarily determined by geography. Geography is a critical, yet often ignored, variable in global politics. Another contribution of this book, then, is to try to “bring geography back in” to international relations scholarship. Several policy implications also follow from this analysis. First, balancing against power does not appear to be an effective option for survival. This implication is especially important for nonbuffer states with limited resources. Second, international recognition does appear to be related to state survival. States like East Timor have done well to acquire it, while states like Taiwan should continue to pursue recognition. Finally, I argue that the norm against conquest that has prevented rivals from taking over buffers since 1945 is by no means a permanent addition to the international system. Indeed, the state that has served as the key proponent of this norm—the United States—is now behaving in ways that directly contradict this norm. A discussion of the 2003 Iraq War and the ensuing occupation suggests that the United States did not intend a long-term occupation of Iraq; nonetheless, the consequences and precedent of this occupation may be particularly fateful for the norm against conquest. Although violent state death has become an extremely rare event in recent years, it may become increasingly common in the future.


The death of a state is a momentous event in the international system. State death forces citizens and cartographers alike to revise their vision of the world. Yet diplomats and scholars have had a poor understanding of this phenomenon, in part because the conventional wisdom assumes that state death almost never occurs. My first goal for this book is to lay out the historical record of state death, showing that this phenomenon has been more frequent than has previously been thought. The next, critical task is to explain variation in state death. Here I also take on the conventional wisdom by arguing that it is the politics of geography—and not a process of selection—that accounts for state death and survival in the international system. My final goal for this book is to explain the shift away from state death after 1945, and to consider how the causes of this shift may bear on future international relations.

Return to Book Description

File created: 8/16/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
Princeton Legacy Library
Exam/Desk Copy
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share