© Copyright, Princeton University Press.

No part of this book may be

distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

Chapter 1

Normative Structures in International Politics

O ver the past decade the structure of world politics has undergone a
number of substantial changes. For the past half-century we have lived in
an international system dominated by the bipolar rivalry between the
United States and the Soviet Union. Although the Cold War system has
dissolved as quickly as it emerged, it is still not clear what kind of interna-
tional structure will take its place. For example, scholars have debated
whether the world is moving toward multipolarity with the rise of Eu-
rope, Japan, and perhaps China, or whether American dominance is still
sufficient to define the international system as unipolar.1 But while these
analysts have been charting trends in relative power to predict the emer-
gence of new rivalries, others have begun to consider the construction of
an entirely different kind of systemic structure. Departing from prevailing
conceptions of the international system, these scholars have suggested
that state leaders’ understandings of their international environment may
constitute an international structure that is as influential as the measures
of relative power that are traditionally viewed as giving shape to world
politics.

Can Our History Explain Our Theories?

Perhaps it should not be surprising that changes in the structure of the
international system have led us to reevaluate many of the theories we
use to understand international relations. The bipolar system of the past
half-century emerged from a long and bloody war that was almost imme-
diately replaced by a tense military standoff between the victorious par-
ties. Predictably, these events led scholars to argue that the international
system is fundamentally shaped by the unbridled competition for power
among the dominant states in the system. Since its inception in the
1950s, this “realist” vision of world politics has exploded into a whole
family of theoretical approaches. This field has become so broad that
some might question whether the label “realist” continues to have mean-

1
John Mearsheimer (1990, 2001) typifies this kind of analysis in his discussion of the
history of Great Power politics as well as international security following the end of the
Cold War.

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

2 CHAPTER 1

ing.2 At its core, however, the “realist” approach to world politics asserts
that two central realities permeate every aspect of world politics.3 First,
the constant environment in which all of international politics takes place
is one of anarchy. That is, no power stands above that of the nation-state.
Second, the only variable that can shape this environment significantly is
the relative distribution of power among the states in the system. Specifi-
cally, the prevalence of conflict and the ease of achieving cooperation
may vary depending on whether one, two, or three or more states domi-
nate the system. This approach tends to see conflict as a pervasive aspect
of world politics, and argues that the only possible constraint on the
outbreak of violence is a countervailing threat of violence. Through the
maintenance of this so-called balance of power, violence can often be
avoided, but these intermissions between disputes are not what one
would normally consider a “peace.” Finally, realists assert that interna-
tional agreement, institutions, and the like are of no consequence in re-
ducing international conflict. What counts are the interests of states and
their ability to achieve and maintain them by force.
The realist vision of the international system as rooted exclusively in
power and national interests has had a profound influence on the devel-
opment of theories of foreign policy behavior over the past several de-
cades. Perhaps the single most prominent and popular theoretical ap-
proach to foreign policy to emerge during this period has been the
rational theory of deterrence.4 Although scholars have produced many
specific variants of deterrence theory, all of these arguments share several
presumptions. Once again, deterrence theorists view conflict as a perva-
sive aspect of international politics. In addition, they argue that the best
way for a state to prevent its enemies from starting a conflict or seizing its
valued assets is to threaten to inflict such damage on them that they
prefer complying with their opponent’s wishes to risking a conflict. Spe-
cifically, theories of deterrence present two necessary and sufficient condi-
tions for preventing enemies from initiating a conflict: (1) demonstrate a
capability to inflict severe damage on one’s opponent; and (2) demon-
strate a willingness to use this capability if its opponent does not comply.

2
For a critique of the realist paradigm as either theoretically incoherent or manifestly
false, see Legro and Moravcsik (1999).
3
The list of major realist works is a very long one, but some of the most prominent and
representative of this approach are Carr (1946) and Morgenthau (1985). Perhaps the cen-
tral realist work in this regard is Waltz (1979). A more recent statement of the realist
perspective on world politics can be found in Mearsheimer (2001).
4
Again, the list of prominent realist works on deterrence is extremely long, but some of
the more significant ones are Schelling (1960, 1966); Russett (1967); Mearsheimer (1983);
and Huth (1988).

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

N O R M AT I V E S T R U C T U R E S 3

These two foreign policy prescriptions epitomize the realist-inspired vi-

sion of international politics.
But while the bipolar Cold War system was created and perpetuated
through violence and the threat of violence, the system collapsed in a
very different way. No great military conflict caused the disintegration of
the USSR or its abandonment of its Eastern European empire. Rather
than violence or a dramatic shift in power resources, the bipolar Cold
War system ended because of changes in the foreign policy ideas held by
important policy makers.5 Gorbachev’s rise to power in 1985, for exam-
ple, eventually sparked Soviet “new thinking” about their security policy.6
This new thinking included a reconceptualization both of legitimate So-
viet security needs and of the most effective ways to meet those needs.
Furthermore, the European velvet revolutions of 1989 were able to take
place because of changes in leaders’ beliefs—both in the USSR and in
Eastern Europe—concerning the legitimacy of using force to impose
pro-Soviet communist rule. There is no military or economic reason why
the USSR and its Warsaw Pact allies could not have responded to the
uprisings of 1989 just as they did to uprisings in Hungary in 1956 and
Czechoslovakia in 1968. Similarly, there is no good military or economic
reason that the USSR could not have remained unified and therefore
remained a superpower. The best explanation of the change in Soviet
behavior both at home and abroad is that Soviet elites changed their
understanding of Soviet interests and how best to achieve them.
Perhaps it was the ideological collapse of the Cold War that has led
theorists to explore conceptions of a new international structure that
might be based on the influence of ideas as much as on the distribution
of power. The most prominent previous example of such an attempt was
made by President Woodrow Wilson following World War I. More re-
cently, however, scholars have begun to look at reinvigorating the United
Nations or some other international institution into a global security
community which could give a new and more peaceful structure to world
politics.7 In the mid-1990s, Charles W. Kegley made a powerful call for
5
For a discussion of the end of the Cold War as rooted in changes in the norms held by
decision makers, see Evangelista (1995); Koslowski and Kratochwil (1994); Mendelson
(1993); and Checkel (1993). For complementary discussions of the role of ideas in struc-
tural change and the end of the Cold War, see Stein (1994); Risse-Kapen (1994); and
Lebow (1994).
6
For discussions of Soviet “new thinking” as it began to diffuse the US-Soviet rivalry, see
Gromyko and Hellman (1988); Meyer (1988); and Snyder (1987).
7
A number of volumes were published in the wake of the Cold War concerning norma-
tive international structures in general as well as the role of specific institutions such as the
United Nations (UN) in creating such a structure. See, for example, Rochester (1993);
Weiss (1993); Ruggie (1993); and Falk, Johansen, and Kim (1993).

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

4 CHAPTER 1

the reexamination of Wilsonian ideas concerning the rule of law and in-
ternational security. Citing a wide range of events occurring during the
early 1990s that he believed to be incompatible with traditional realism,
Kegley suggests that “the time has arrived to revise and reconstruct clas-
sical realism, and build a refashioned paradigm inspired by Wilsonian
idealism.” (Kegley 1993, 143) Over the past several years, a number of
scholars have begun to answer this challenge. In response to the ideolog-
ical collapse of the Cold War, scholars such as Alex Wendt, Martha Fin-
nemore, Kathryn Sikkink, and many other have begun to theorize about
an international system in which ideas are as influential as material capa-
bilities (Wendt 1992, 1999; Klotz 1995; Katzenstein 1996; Finnemore
1996; Keck and Sikkink 1998).

What Do You Mean, “Ideas Matter”?

We must be careful of what we mean, however, when we speak of “ideas”
influencing systemic structures or foreign policy behavior. Even the most
orthodox of realist approaches concedes the influence of ideas in the
sense that national policy makers are human, and they have ideas about
how they should behave and why they should behave that way. But there
are at least two important differences between the role that realists allot
to ideas in international relations and the role being considered by these
new revisionists. First, realists argue that the ideas that influence policy
are not complex. That is, they generally concern the link between means
and ends, and they can be summed up in fairly brief “if . . . then” state-
ments.
Second, and perhaps more important, realists insist that the ideas that
policy makers hold are directly determined by other variables, such as
relative power and national interests. Thus in the realist view, ideas do
not have an independent causal effect on behavior, but are only cognitive
intermediaries between environmental stimuli and policy makers’ re-
sponses. A theoretical approach to international relations that includes
the influence of norms and institutions, however, allows for a more com-
plex vision of the role of ideas in foreign policy. First, making normative
assessments as to whether behaviors are “legitimate” or not involves
more complex ideas than can be accounted for by realist theory. Concep-
tions of “right” and “wrong” are more abstract and complex than simple
“if . . . then” beliefs about the nature of the world.8
Moreover, according to this approach, normative standards are not ex-
clusively determined by power and interests and may have an indepen-
8
For a discussion of various ways in which ideas might “matter” in terms of explaining
international behavior, see Goldstein and Keohane (1993).

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

N O R M AT I V E S T R U C T U R E S 5

dent causal impact of their own.9 These revisionists note that a variety of
different normative structures may operate within any given power struc-
ture.10 Thus norms must develop in a manner that is in some ways inde-
pendent of power and interests. If norms are not entirely determined by
power and interests, then these ideas may have an independent influence
on state behavior.
We must also recognize that the construction of a normative frame-
work for world politics should not eliminate international conflict any
more than the passage of a law should eliminate crime. Nor would such a
framework imply that military power will become irrelevant to state be-
havior. Rather we should expect the development of commonly under-
stood norms to help international policy makers take incremental steps
toward reducing the incidence of violent conflict and stabilizing peaceful
relationships. Conflicts should become less frequent as states’ expecta-
tions about their relations converge, and the conflicts that do emerge
should be more easily resolved in compliance with the accepted norma-
tive standards. Specifically, as I will discuss in chapter 3, we should expect
the accession to behavioral norms to ease the enforcement of these stan-
dards for at least two reasons. First, norms define which acts are illegiti-
mate and demand punishment by other states. Second, punishments
should be more likely to result in compliance because the trespassers rec-
ognize the standards of behavior that they have violated as legitimate. In
this manner, norms can help states to reestablish cooperative relations
and minimize the long-term effects of disputes.

Legitimate Agreements: The Building Blocks

of Normative Structure
Theories of an international structure based on ideas leads us inevitably
to the consideration of theories of foreign policy behavior that incorpo-
rate the influence of norms. If norms and ideas are to have any impact at
the systemic level, that influence must be exerted on individual decisions
and actions by nation-states. Most of the evidence collected thus far con-
cerning the influence of norms on foreign policy has been done in the
9
For a discussion of the distinction between norms as independent causal variables and
spurious correlates of cooperative interests, see Krasner, “Structural Causes and Regime
Consequences,” in Krasner (1983); and Krasner, “Regimes and the Limits of Realism: Re-
gimes as Autonomous Variables,” in Krasner (1983).
10
Alexander Wendt’s (1992, 1999) work is perhaps the central statement of this perspec-
tive. His contention is aptly summarized by the title of his 1992 article, “Anarchy Is What
States Make of It.” That is, Wendt argues that international anarchy and the material distri-
bution of power do not—by themselves—predict anything about state behavior. Instead,
Wendt argues that states must construct shared meanings for these material facts. These
constructed understandings, in turn, influence behavior.

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

6 CHAPTER 1

area of international political economy.11 Work in this area is important

and obviously represents a significant step toward establishing the inde-
pendent role of ideas in international relations. But if norms are to be-
come the foundation of more general approaches to international rela-
tions, then we must establish that these ideas can have an independent
influence on military conflicts as well. One obvious reason for emphasiz-
ing the importance of military conflicts is that most structural changes in
world politics have been a result of military conflict of one kind or an-
other, and so militarized disputes would represent one of the most signif-
icant challenges to a normative international structure. Moreover, while
realism represents a general approach to the study of world politics, its
proponents have always held that its explanatory power is greatest when
it comes to understanding security-related disputes. Consequently, if a
normative approach to international relations is to subsume or supplant
realism, it must be able to demonstrate the impact of norms on military
conflicts.
The central aim of this work is to test the proposition that the con-
struction of normative standards of behavior can influence state actions in
security-related disputes. I hope to demonstrate that international secu-
rity norms can and have helped to stabilize peace and cooperation even
between states that represent substantial security risks to one another.
Looking at the historical record, we find relatively few examples of formal
regional—much less global—security organizations. There have been
many fewer international security organizations, for example, than orga-
nizations that have existed to facilitate international trade. Nonetheless,
the fact that security norms have not been as formalized as their eco-
nomic counterparts does not imply that they cannot or have not affected
behavior. I will examine the construction of security norms of a more
decentralized nature. Specifically, I will examine the bilateral establish-
ment of norms as settlements for security-related disputes and the way in
which such bilateral settlements affect subsequent interactions over the
same disputed issue. The finding that such bilateral norms affect security
behavior would have several significant implications. First, such a finding
would be important information for current national policy makers be-
cause it would provide them with another foreign policy tool. If security
norms can stabilize cooperation, then leaders can move beyond deter-
rence toward a more constructive strategy for maintaining their security.

11
The most prominent works in this area concern the construction of what are known as
international regimes. See, for example, Krasner (1983) and Keohane (1984). I will elabo-
rate further on my discussion of international regimes in chapter 2. Other more recent work
has focused more explicitly on the development of international norms. See, for example,
Keck and Sikkink (1998) and Klotz (1995). In general this work also has not focused on
security affairs. Some exceptions are Price and Tannenwald (1996); Price (1995); Legro
(1995); and Tannenwald (1999). A partial exception is Finnemore (1996).

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

N O R M AT I V E S T R U C T U R E S 7

Second, empirical support for the role of security norms would remind
international relations theorists that we must not restrict our attention
to conventional realist hypotheses concerning power, interests, and re-
sources. Moreover, it would specifically suggest that we must give more
careful consideration to the independent role that ideas may play in the
conduct of world politics. Finally, the finding that norms can affect the
resolution of security conflict would provide some encouragement for
those who hope to build more elaborate security structures based on
common normative standards of behavior.
I will test the proposition that norms affect the resolution of security-
related disputes against the central competing realist paradigm for under-
standing military conflict: rational coercion theory.12 This is perhaps the
most appropriate competing explanation to consider in my analysis be-
cause of its powerful influence as a vision of foreign policy making
throughout the Cold War. Moreover, it exemplifies the realist, power-
based notions of international relations that deny any independent role
for ideas in shaping foreign policy. Briefly, my findings indicate that these
settlements do shape state leaders’ responses to crisis-bargaining behavior
in ways that cannot be explained by a purely realist framework. These
results indicate that international security norms can and have helped to
stabilize peace and cooperation, even between states that represent sub-
stantial security risks to one another. At the same time, however, tradi-
tionally realist variables—such as coercion and reputation for toughness—
are central to my explanation of crisis outcomes. Thus my argument
represents an important step toward integrating normative approaches
with the prevailing—yet incomplete—realist approach to crisis bargain-
ing.
My analysis proceeds as follows. In chapter 2 I will review the norma-
tive and realist literatures that relate most directly to my investigation of
norms in security conflicts. In doing so, I hope to place my research in its
proper perspective and define more precisely its contribution to our un-
derstanding of security policy. I will also outline some of the various
theoretical and methodological pitfalls involved in the study of norms
that I hope to circumvent in this work. In chapter 3 I will carefully out-
line normative and realist theoretical frameworks. Following the discus-
sion of each theoretical model, I will develop specific testable hypotheses
from each approach concerning militarized conflict behavior. In chapter
4 I will turn my attention to constructing a research design for testing
12
As I will discuss and demonstrate in chapter 3, rational theories of deterrence and
coercion are, in fact, identical in every significant respect. Their only difference is in the
specific task to which deterrent and coercive threats are put. Deterrence concerns the pre-
vention of an action before it begins, while coercion concerns the reversal of an action that
has already been taken. I choose to phrase my discussion in terms of a theory of coercion
because it is more appropriate to the bargaining behavior that I will examine.

For general queries, contact [email protected]

© Copyright, Princeton University Press. No part of this book may be
distributed, posted, or reproduced in any form by digital or mechanical
means without prior written permission of the publisher.

8 CHAPTER 1

these hypotheses. Specifically, I will identify the proper population of

cases for my analysis, discuss the data set that I have created, and present
the operational measures for the conceptual variables that I have outlined
in chapter 3.
In chapter 5 I begin presenting my empirical analyses. These analyses
sequentially address three related but distinct questions about interna-
tional norms. I begin by asking whether norms influence behavior in secu-
rity-related crises. Second, I investigate how they exert this influence. And
finally I turn to the question of when such international norms will be
effective. In chapter 5 I will present the results of my statistical tests of
the realist and institutionalist models of crisis bargaining. As I described
earlier, I find strong support for the hypothesis that states can and do
construct normative standards that guide their behavior in militarized
disputes.
In chapter 6 I shift my attention from an aggregate statistical analysis
of dispute behavior to the execution of more detailed case studies from
my data set. While the aggregate analysis in chapter 5 addresses the ques-
tion of whether international norms matter, these case studies are better
suited to illustrating how normative standards of behavior influence inter-
national conflict. Specifically, I examine two crises that illustrate the sta-
tistical findings in chapter 5, and demonstrate that the causal arguments
presented in chapter 3 continue to make sense when we examine individ-
ual crises in detail.
In chapter 7 I turn my attention to the question of when international
norms will shape security crises. While the analysis in chapter 5 demon-
strates that norms do have an impact on military conflict, it also indicates
that norms are not always effective. In this chapter I show that my theory
about the impact of norms can also help us understand when norms will
be influential. Finally, in chapter 8 I conclude by discussing the implica-
tions of my research for broader theorizing about international relations.
In addition, I will draw attention to issues and topics that require further
research and to the policy-making implications of my work.
I am keenly aware that this work will not represent the final word in
our understanding of international norms. I begin my analysis with a
broad discussion of the conceptualization of international norms, and
proceed to examine these broad concepts though the lens of several ex-
tremely specific—and in some sense narrow—empirical tests. However, I
am convinced that we can best advance our understanding of broad theo-
retical concepts through testing them in carefully crafted and concrete
circumstances. Taken in this spirit, I believe that my results advance our
understanding of the impact of international norms. At the same time,
however, I recognize that the findings of my work may raise more new
research questions than they answer.

For general queries, contact [email protected]