Protection States Trust?: Superpower Patronage, Nuclear
Behavior, and Alliance Dynamics∗
Alexander Lanoszka
Princeton University
[email protected]
February 25, 2012
Abstract
Nuclear behavior – ambiguous posturing as well as the pursuit, acquisition, and
detonation of sensitive nuclear materials – creates international instability and potentially invites punishment from superpowers. Why then do some states that enjoy the
protection of a nuclear capable superpower nonetheless engage in nuclear behaviors?
Why do states that do engage in nuclear behavior sometimes fully reverse course?
Explanations that claim such behavior is motivated by self-help do not sufficiently account for the conditions under which states respond to threats with nuclear behavior.
Furthermore, self-help explanations are unable to account for the cases of nuclear reversal or voluntary suspension of nuclear activities. To address these shortcomings,
I advance a new theory in which nuclear behavior is principally informed by alliance
dynamics between secondary states and superpower patrons. I argue that leaders of
secondary states refer to economic and political indicators to determine the reliability
of the superpower’s security commitments. When economic and political indicators
suggest that the superpower’s security commitments are incredible, secondary states
adopt nuclear behavior in order to bargain over the terms of the existing ‘security contract’ with the superpower. The superpowers ability to bargain successfully with the
secondary state depends on the latters economic dependence.
∗
For helpful comments on this draft and previous stages of dissertation ‘hunting’, I thank G. John
Ikenberry, Keren Yarhi-Milo, Thomas J. Christensen, Aaron Friedberg, Rex Douglass, David Hsu, Michael
Hunzeker, Raymond Kuo, Michael McKoy, Thomas Leo Scherer, and Sherry Zaks. All errors are certainly
my own, but I would still blame them on Michael Hunzeker.
Contents
1 Introduction
1
2 Disaggregating Nuclear Behavior
6
2.1
Ambiguous Posturing
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6
2.2
Pursuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7
2.3
Acquisition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
2.4
Detonation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
2.5
Reversal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
2.6
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
3 Literature Review
11
3.1
Nuclear Behavior as the Dependent Variable . . . . . . . . . . . . . . . . . .
11
3.2
Alliance Dynamics as the Independent Variable and Causal Mechanism . .
12
4 Theory of Nuclear Behavior in Alliances
15
4.1
Security-Seeking and Superpower Patronage . . . . . . . . . . . . . . . . . .
15
4.2
The Perceived Unreliability of Extended Deterrence as a Driver of Nuclear
Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
18
4.3
Nuclear Behavior as Insurance and Bargaining . . . . . . . . . . . . . . . .
21
4.4
The Role of Economic Statecraft in Nuclear Bargaining . . . . . . . . . . .
25
4.5
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27
5 Alternative Arguments
28
5.1
Existing Realist Explanations . . . . . . . . . . . . . . . . . . . . . . . . . .
29
5.2
Domestic Politics Explanations . . . . . . . . . . . . . . . . . . . . . . . . .
29
6 Conclusion
33
1
Introduction
Why do some states that enjoy the protection accorded by an alliance with a nuclearcapable superpower engage in nuclear behavior? By nuclear behavior I refer to a
variety of actions that encompass ambiguous posturing as well as the pursuit, acquisition, and detonation of sensitive nuclear materials. That some junior alliance
partners choose these actions and not others in the nuclear era is puzzling for several
reasons. First, because these states are aligned with a superpower, they fall under a
nuclear umbrella and thus theoretically enjoy extended deterrence. Accordingly, we
would expect that these states have fewer incentives for possessing these materials.
Second, the mere acquisition of these technologies is destabilizing in international
politics. The risk for proliferation raises the level of threat perceived by states belonging to an adversary alliance, thus increasing the likelihood for preventive attack.
Put differently, whereas superpower patronage offers protection and stability, a secondary state’s nuclear behavior can renew international tensions and cause dangerous
crises.
Despite these disincentives for pursuing nuclear behavior, numerous instances of
junior alliance partners engaging in such activities can be found in the historical
record. The United Kingdom and France both successfully acquired an independent
nuclear capability. The People’s Republic of China (PRC) chose to acquire nuclear
weapons in spite of its alliance with the Soviet Union. Israel, though not a formal ally of the United States (US), furtively developed a nuclear weapons program.
North Korea and Taiwan each began programs aimed at developing their own nuclear
weapons capabilities in the 1960s and 1970s, respectively. Furthermore, South Korea
surreptitiously obtained heavy water and utilized it to enrich weapons-grade uranium
in 2001 — a move that earned it sharp censure from the international community.
These examples highlight a serious oversight in the current literature. Nuclear
acquisition on the part of a subordinate alliance partner largely undermines the superpower’s command of the region in which the proliferator is located. Having a
nuclear capability levels the playing field between states by offering a countervailing measure against attempts at coercion that more powerful states can successfully
make otherwise. Nuclear proliferation sharply revises balances of power between allied states and enhances foreign policy autonomy in a way that makes the superpower
less able to manage its junior partners and alliance structures. As a consequence,
the superpower has particular incentives for challenging a junior alliance partner’s
nuclear behavior. Given the implications that nuclear proliferation has for a superpower’s leading status, it is unclear why a junior alliance partner would spoil its
relationship with its patron and risk bearing the brunt of coercive dissuasion tactics
1
by the superpower.1 Because many scholars have written on the determinants of
nuclear behavior from a monadic perspective, the literature provides little guidance
on understanding this subject. We are left with some key questions: Why would
subordinate allies pursue actions that could conceivably prompt restraint and punishment efforts from the patron? Under what conditions would allies choose military
policies that contradict the interests of a dominant alliance member or the larger
alliance system itself?
Intuitively, it would seem that the answer to the questions posed here is that selfhelp encourages states to engage in nuclear behavior. These states cannot rely on
others, even their superpower patrons, to meet their security needs. Because states
cannot dial 911 to receive assistance when their survival is under jeopardy, they turn
to nuclear weapons as a primary source of insurance. This simple realist view suggests that there is no puzzle. However, this perspective would be wrong. Consider
the cases of South Korea and Taiwan. Arguably, both were largely poor agrarian
economies before being industrially capable to develop an independent nuclear capability. Thus, Taiwan waited several years to develop its capability following the
PRC’s detonation of its first nuclear weapon in 1964 and South Korea took time to
respond to North Korean efforts at procuring materials for a nuclear weapon. However, a capability-centered explanation does not address why both states ceased their
nuclear activities in the 1970s. Moreover, it is unclear by the merits of the theory
why both of these states dabbled sporadically in nuclear activities beyond the early
1980s and as late as even the 2000s. These East Asian examples are not unique.
Israel developed its own nuclear capability around the same time as China, but it
eschewed any outward demonstration of its newly produced weapon. France did not
begin developing nuclear weapons until almost ten years after Great Britain. Even
West Germany, which hosted large American conventional military deployments and
thus of high interest to American security, engaged in nuclear behavior. Indeed, that
some states reversed on their achievements towards developing nuclear weapons adds
to the general insufficiency of the simple realist argument presented above.2
1
I use the terms ‘superpower’, ‘patron’, ‘leading state’, and ‘superpower patron’ interchangeably for
stylistic purposes to describe those states that have significant power project capabilities. All things equal,
only these states are able to fight a conventional war alone against another superpower on its own territory
(Mearsheimer 2001, 5). During the Cold War, the international system contained two superpowers, the US
and the Soviet Union. Though the United States became the sole superpower with the end of the Cold
War, one contemporary debate concerns whether the US will continue to enjoy this privileged status or
China will soon join its rank (see Brooks and Wolhforth 2008). I broadly define secondary states as all
states that do not share any superpower’s ability to wage conventional warfare. I offer a deeper discussion
on the meaning of ‘secondary states’ in section 4.1.
2
Solingen (2007) provides an excellent description of the indeterminacy and insufficiency of existing
structural realist accounts in explaining key cases of nuclear proliferation. In explaining facts that do
not fit well with the original theory, conventional realist accounts become prone to introducing ad hoc
2
I advance the theory that secondary states will engage in nuclear behavior to obtain forms of insurance against both the abandonment of their patron and the threat
emanating from an aggressor state. More specifically, I argue that secondary states
typically rely on nuclear umbrellas that superpowers provide. Nuclear umbrellas are
a form of extended deterrence in which the superpower prevents an attack on its ally
by threatening nuclear retaliation on an aggressive adversary. One disadvantage of
offering a nuclear umbrella is that the superpower needs to credibly demonstrate its
commitment to come to its ally’s defense in the event of a nuclear exchange rather
than simply sacrifice it. I contend that perceptions of the reliability of the superpower’s security guarantees – that is, the ability and willingness of the superpower
to make good on those commitments – shape nuclear behavior and the bargaining
process that ensues.
A hypothetical example helps clarify the claim put forward here. If a superpower patron wishes to enhance the credibility of extended deterrence, it can choose
to deploy conventional military forces on its ally’s territory. This move enhances
the credibility of the nuclear behavior because it ties a portion of the superpower’s
military assets to the integrity of its ally’s own national security. As much as this solution is effective for its costliness, these same costs might be vulnerable to measures
or policies that the superpower’s leadership could later pursue in order to reduce
military expenditures abroad. In times of economic downturn, for example, the superpower leadership will be tempted to substitute away from expensive conventional
forces to take advantage of the relative cheapness of nuclear weapons and allocate
greater resources to boost domestic economic performance. In such a situation, the
secondary state develops reasons to believe that the superpower patron is reducing
its military commitments abroad. Nuclear programs become an especially attractive
option for the secondary state either because they can be leveraged to pursue an
independent foreign policy or extract greater assurances of support from the patron.
I thus also argue that when secondary states engage in nuclear behavior they also
enter a bargaining process with their patrons over the terms of the security offered
to them. The strategic interaction following the initial decision to embark on nuclear behavior turns on two issues. First, the superpower needs to demonstrate the
credibility of its commitment. Second, the secondary state needs to make commitments that – if it were to reverse on its nuclear activities – it would not engage in
nuclear behavior again. One particular source of leverage for the superpower may
be a favorable asymmetry in its economic relations with the secondary state. The
superpower could use economic statecraft to threaten the client state with punishment. However, punishment does not obviate the resolution of credible commitment
hypotheses that depart critically from the theory’s core assumptions (see Elman and Elman 2002).
3
problems because the superpower does not want the secondary state to be a nuclear
recidivist.
My theory of intra-alliance nuclear behavior is applicable when a key scope condition is met. For the theory to have explanatory power, the secondary state must
be part of a ‘strategic triangle’. I define ‘strategic triangles’ as a situation in which
three states make interdependent decisions regarding alignment and military armaments. During the Cold War many secondary states found themselves in the middle
between two opposing superpowers, aligning with one while fearing the other. With
the demise of the Soviet threat strategic triangles have become less numerous. Yet
they are still in prevalent in some regional contexts. For example, a resurgent Russia
has pushed some Central-Eastern European states to align more closely with the US.
In East Asia, a rising China has led international observers to believe that threatened
states will look to further their security ties with the US.
I test my claims against two competing theories of nuclear behavior. To some
extent, my theoretical framework borrows a number of insights from realist theory
because I emphasize protection from external security threats, fears of abandonment,
and anarchy in the international system. However, I argue that conventional realist
perspectives on nuclear proliferation are inadequate. The body of realist scholarship
is often insufficient and underdetermined in accounting for why secondary states
would engage in nuclear behavior. The assertion that states acquire nuclear weapons
because they have an overwhelming security interest for doing so ignores the number
of nuclear reversals we observe. A meaningful share of these reversals are a product of
the interactions secondary states have with their patron states. In contrast to realist
theories I probe more deeply in these bargaining dynamics. I also dispute research
that emphasizes domestic politics as an explanation for nuclear behavior. Preferences
over a state’s role in the global economy or the level of nationalism exhibited by a
state’s leadership do not explain the broader scope of nuclear behavior.
Several motivations guide this project. The first motivation is that the diffusion of
nuclear technology has major implications for how leading states manage and sustain
an international order of their liking. Oddly, this insight has been overlooked until
very recently by analysts who explore such related issues as international hierarchy
and hegemony (see Deudney 2011). Analyzing the dynamics between superpowers
and their junior partners over nuclear issues can reveal fresh insights on how international hierarchical relations are asserted and preserved.
A second and related motivation is to further our understanding of internal alliance dynamics. Alliances are organizations created by at least two states that have
common security objectives. For alliances to be effective and cohesive member states
have to coordinate on their military policies to some extent. Nuclear behavior is one
4
area of military policy that has dramatic, and often negative, ramifications for the integrity and cohesion of the alliance. Uncertainty over the status of a state’s nuclear
ambitions generates anxiety among non-nuclear alliance partners that they might
find themselves in a regional arms race or be exposed to future attempts at nuclear
blackmail. Contentious military issues shape the dynamics of alliance partners, yet
this phenomenon has not received adequate attention in the existing literature on
alliances.3
A third motivation is rooted in the observation that nuclear behavior will continue to be a salient feature of international politics. It is tempting to believe that
examining alliance dynamics and contentious nuclear behaviors with a focus on empirical examples drawn primarily from the Cold War has little relevance for today.
This view is wrong, however. Gavin (2004/2005, 2010) notes that there is remarkable continuity in how states confront nuclear proliferation throughout the Cold War
and beyond. Revisionist, nuclear-seeking states like North Korea and Iran are not
a recent phenomenon. American and Soviet decision-makers were similarly anxious
about the PRC and its nuclear program in the 1960s. Indeed, a salient concern at
the time was that Chinese proliferation would kindle wider nuclear proliferation in
East Asia – a belief that turned out to have merit. In the present day, one feared
consequence of Iranian and North Korean nuclear arsenals is that American allies in
the Middle East and East Asia, respectively, would be spurred to develop their own
independent nuclear capabilities. Understanding past empirical variation is critical
for addressing urgent policy questions that will confront American decision-makers
in the very near future.
This chapter proceeds as follows. Section II examines the dependent variable
– nuclear behavior – and develops definitions for each action that falls under this
general description. I use this section to refer to historical examples to illustrate
the empirical variation I wish to explain. Section III reviews the existing literature
by assessing both the wider scholarship on nuclear behavior and the relevant work
on intra-alliance dynamics. I show that there are several important shortcomings in
this literature that this project seeks to address. Section IV describes the theory and
offers an account of the strategic calculus secondary states make in their decisions
regarding nuclear programs. Section V discusses three sets of alternative explanations
that also need to be considered in any empirical evaluation of the main argument
presented here. Section VI offers concluding remarks.
3
Pressman (2008) offers a key example of the existing scholarship, but describes only the failed American
alliance restraint efforts in preventing Israel from acquiring the bomb.
5
2
Disaggregating Nuclear Behavior
Since nuclear weapons provoke such intense unease and concern, we must widen the
scope of our analysis to examine the gamut of behaviors capable of inciting these
reactions. The dependent variable on which this project focuses its attention is the
nuclear behavior of secondary states that enjoy the patronage of a superpower. As
noted, I use the term ‘nuclear behavior’ to refer to a variety of different actions that a
state can undertake with respect to sensitive nuclear materials. These actions include
the ambiguous posturing and the pursuit, acquisition, and detonation of sensitive nuclear materials.4 I also consider the many numerous reversals in nuclear behavior
that have occurred historically. In this section I proceed to consider each of these
actions. I use historical examples to both illustrate their definitions and provide the
reader a better grasp of the empirical variation I seek to explain.5 These (positive)
values on the dependent variable are useful for deepening our understanding of the
actions that elicit superpower concern and tracing the evolution of bargaining dynamics between the superpower and the secondary state. Reversal excepted, each of
these stages represent a ratcheting up of nuclear behavior.
2.1
Ambiguous Posturing
Ambiguous posturing denotes a situation in which decision-makers of a non-nuclear
state create uncertainty over their intentions towards the use of sensitive nuclear
materials.6 This uncertainty arises if there exists a lack of clear commitment to
renounce nuclear weapons acquisition, the commission of ‘feasibility studies’ to determine the viability of nuclear weapons development, and an unwillingness to adopt
proper safeguards for existing nuclear facilities. The West German leadership exemplified this approach in the 1960s. During this time, its leaders sought a greater
role in the North Atlantic Treaty Organization’s (NATO) nuclear strategic planning
while resisting calls to make treaty commitments to not acquire nuclear weapons.
My conception of ‘ambiguous posturing’ encompasses a stage of nuclear weapons
development that Singh and Way label as ‘exploration’. This step falls short of actually committing material resources towards the production of enriched uranium and
other components that are needed to build an atomic weapon.
4
I largely borrow the conceptualization of the first three steps from Singh and Way (2004).
See Appendix I for a complete table that lists all historical examples of nuclear behavior amongst
secondary states.
6
To be sure, a state that has already acquired nuclear weapons can adopt a posture that some observers
might call ‘ambiguous’. Some observers have thus described Israel’s concealment of its nuclear arsenal as
an ‘ambiguous nuclear posture’. I restrict this category, however, to instances in which the state has not
yet decided on whether to fully pursue the acquisition of nuclear weapons.
5
6
Regardless of whether the non-nuclear state has intentions to acquire nuclear
weapons, this action still provokes anxiety for a state’s patron. After all, it indicates
either discontent with the superpower’s patronage or a greater willingness to establish
a more independent foreign policy. Such gestures also help leaders of secondary states
to manipulate risk in a way that increases the level of attention that the superpower
affords to them. Of course, there are incentives to take a more subtle undertaking of
exploration so as not to draw harmful attention. Nevertheless, even if this concern has
implications for reliably counting instances of exploration, we are still able to identify
unambiguous cases in the historical record. Taiwan in the 1970s and Australia in
the late 1960s and early 1970s are two such cases.
2.2
Pursuit
Pursuit occurs when a state’s decision-makers actively deploy material resources and
expertise towards the production of nuclear weapons. They have finished deliberating on whether to start an independent nuclear program by making an affirmative
decision. However, pursuit does not necessarily mean that the state effort to build an
independent nuclear capability is either inexorable or efficient. States might dedicate
insufficient resources towards the production of nuclear weapons not only because
resources are limited, but also for strategic reasons because they allow states more
time and flexibility in responding to counter-proliferation efforts.
Pursuit, therefore, does not necessarily lead to acquisition. There are a number of
states that embarked on nuclear weapons programs without ever completing them.
South Korea in the 1970s is a classic example when it chose to cancel its program
following intense American pressure. Furthermore, towards the end of Francisco
Franco’s dictatorship American intelligence learned that Spain had begun its own
nuclear program in earnest. Historical instances of this variety were not restricted
to members of American-led alliances. In the Eastern bloc Romania began its surreptitious pursuit of nuclear weapons sometime during the 1980s.
This type of nuclear behavior produces similar concerns for the superpower patron
as those generated by ambiguous posturing. However, these concerns are amplified
because the secondary state’s leadership is demonstrating its willingness to bear the
material costs associated with pursuit. In other words, committing some level of
resources towards the manufacture of nuclear weapons sends a more credible signal
of the secondary state’s discontent with the superpower’s patronage or willingness
to obtain greater foreign policy autonomy. Furthermore, if the efforts at pursuit
are transparent, states found in the potential proliferator’s region are more likely to
feel threatened. Pursuit is often regionally destabilizing, particularly if it motivates
7
other states to reconsider their own defense policies. In such an event, the superpower
patron would not only have to dissuade the pursuing state from engaging in nuclear
weapons production, but also reassure regional allies of its commitment to their
security.
2.3
Acquisition
Acquisition transpires in one of two ways. First, the state’s own nuclear program was
able to indigenously produce warheads. Acquisition completes the process that began
with the initial exploration of creating an independent nuclear capability. Second,
the state might have successfully obtained nuclear weapons from another source.
This means of procurement is very rare, but the discovery of the A.Q. Khan network
in the mid-2000s alerted Western policy-makers of this possibility. It is conceivable
that the Pakistani nuclear physicist had supplied countries like Iran and North Korea
with sensitive nuclear materials prior to his arrest.
The number of states that actually possess nuclear weapons has varied over time.
Britain and France were the first two non-superpowers to acquire nuclear weapons.
Israel and the People’s Republic of China acquired nuclear weapons in the 1960s,
though the former has adopted an ambiguous status over its proliferation efforts.
South Africa also came into possession of a nuclear arsenal. Since the mid-2000s
North Korea is alleged to have several nuclear weapons in its arsenal.
Acquisition undercuts the superpower’s ability to project power. The possession
of a nuclear weapon empowers the secondary state to block attempts at coercion
and blackmail by its patron. The regional repercussions that follow one state’s
proliferation in the context of broader alliance structures, however, are arguably
more important. For example, a local arms race in the proliferating state’s region
has a higher probability of ensuing. This concern is a salient feature in American
attempts to dissuade Iran from producing nuclear weapons. A nuclear Iran might
provoke American allies in the region like Saudi Arabia to establish their own nuclear
programs, thus undermining the ability of the US to manage security affairs in the
Middle East.
To be sure, the acquisition of a nuclear weapon is not alone sufficient to obtain
a deterrent capability. Though it depends on the nature of the state’s adversary, a
nuclear weapon is less useful without the appropriate warheads and delivery systems.
Indeed, states require a degree of survivability and a second-strike capability so to
prevent an initial nuclear attack from taking place. For my purposes, I consider states
that assemble a military nuclear device to have reached this stage of acquisition.
8
2.4
Detonation
Nuclear behavior does not stop at acquisition. Upon acquiring a nuclear weapon, the
proliferating state can choose to either explicitly demonstrate its newfound capability
or maintain a policy of concealment. Though choosing to conceal a nuclear arsenal
is useful for maintaining strategic ambiguity, detonation has its benefits. According
to Dunn (1991, 20), states that choose not to test nuclear weapons might be unable
to acquire the technology necessary to improve their arsenals. Without adequate
tests states will have a greater difficulty developing “lower weight, more compact,
and efficient fission weapons.” Detonation also reduces uncertainty over whether the
state has a nuclear deterrent.
The most vivid form of demonstration involves an above-ground detonation because it allows the state to signal in no uncertain terms that it possesses such
weapons. Great Britain performed such tests prior to the Partial Test Ban Treaty,
which was an agreement struck in 1963 that banned above-ground detonations.
France, however, continued to engage in these tests until 1974, after which it continued underground testing until as late as 1996. The 1964 detonation by People’s
Republic of China upset the Soviet Union and stirred American policy-makers into
adopting a more vigorous counter-proliferation policy. By contrast, some states that
have acquired nuclear weapons have chosen not to do such exercises. Israel has made
efforts at concealing or downplaying their nuclear arsenals.
States can continue detonations long after the initial acquisition of nuclear weapons.
Such actions are public declarations that serve as a reminder that the perpetrating
state maintains an arsenal of nuclear weapons. Moreover, because above-ground
detonations are banned with the Partial Test Ban Treaty and all forms of detonations with the Comprehensive Test Ban Treaty, the normative injunction against
such actions increases the political costs of carrying out these tests. Performing such
detonations thus also raises questions about the state’s intentions. Indeed, France
incurred international condemnation by exploding nuclear weapons in the South Pacific into the late 1990s.
In the context of alliance relations, any ally that performs detonations is flaunting
its capabilities and autonomy towards its patron while also generating uncertainty
over its motives amongst its neighbors. For these reasons, if a decision needs to be
made between two bad choices, a superpower patron would strictly prefer a nucleararmed ally that does not perform these tests to one that does.
9
2.5
Reversal
It is important to note – contrary to how much of the conventional political science
literature treats the issue – that nuclear behavior is not constrained to exhibiting
a ratcheting effect. History is rife with instances of states that go no further than
ambiguous posturing or pursuit of sensitive nuclear materials. Some historical examples have already been highlighted: West Germany in the early Cold War as well as
South Korea in the 1970s. When states engage in nuclear behavior, there are choices
being made such that acquisition is hardly an inevitability. Even acquisition itself is
not a ‘done deal’. South Africa is an example of a state that chose to renounce and
fully dismantle its nuclear arsenal. Similarly, Argentina and Brazil both relinquished
their nuclear programs in the 1990s. The leadership of these countries judged that
nuclear programs were no longer necessary to enhance their foreign policy autonomy
or improve their security.
Indeed, the choice to ‘deproliferate’ signals greater reliance on a superpower patron to meet some of its security needs. This signal gains credibility if the action
of deproliferation is handled in such a way that makes a renewed effort at proliferation costly in the future. That some states have chosen to renounce their nuclear
weapons also demonstrates that the choice to maintain an arsenal is continually
reproduced. There is no compelling theoretical reason why other existing nuclear
powers would never similarly dismantle their weapons programs. The acquisition of
nuclear weapons is reversible.
2.6
Summary
Nuclear behavior encompasses a variety of different actions that can provoke varying
degrees of anxiety on the part of the superpower. These actions include ambiguous
posturing, pursuit, transmission, acquisition, and detonation of sensitive nuclear
materials. They represent distinct and increasingly higher order actions pertaining
to the use of nuclear weapons. In parsing out the dependent variable this way, it
becomes apparent that there is much richer variation in the nuclear behavior of
states than other analyses often suggest. States do more than simply acquire or
not acquire nuclear weapons. Rather, they have to pass through different stages in
which backsliding is often possible. Each of these stages sends important signals to
the patron regarding alliance cohesion.
10
3
Literature Review
The purpose of this section is to critically examine the international security scholarship that speaks to the issues examined in this project. I divide this section into two
parts. First, I evaluate how the literature has hitherto analyzed nuclear behavior.
Second, I explore the relevant literature on alliances, focusing on those studies that
use alliance dynamics to explain nuclear behavior. I use this latter section to not
only discuss the independent variable, but also the causal mechanisms that, I argue,
have been ignored in the extant literature.
3.1
Nuclear Behavior as the Dependent Variable
A weakness of the existing literature on nuclear behavior is that studies often truncate
the dependent variable. One result of this tendency is to ignore interesting variation
in the intermediate steps between the decision not to pursue the nuclear weapon
at all and the complete acquisition of an arsenal. Goldstein (2000) studies only
positive cases of nuclear weapons acquisition efforts. Hymans (2006) and Horowitz
(2010) examine only final decisions to proliferate or not to proliferate. Jo and Gartzke
(2007) examine cases in which states have an active nuclear weapons program or not.
These scholars are by no means unique. Solingen (2007) traces the process by which
decisions to acquire and forego nuclear weapons. The limited scope of her dependent
variable leaves her unable to properly account for the richer variation we actually
observe in nuclear behavior in her case studies. If a desire to be better integrated in
the global economy explains South Korea’s renunciation of nuclear weapons in the
1970s, then why did it begin the program in the first place?
The examples of South Korea and South Africa highlight another problem in the
extant political science literature on nuclear behavior. Few studies take seriously the
pattern of reversals we observe being committed by a large number of states (Levite
2002/2003; Sagan 2011). In a rare study of deproliferation Mueller and Schmidt
(2010) use cross-tabulations in support of their argument that the Nuclear NonProliferation Treaty (NPT) motivated states to abandon their nuclear programs.
That nuclear reversals have occurred on numerous occasions raises the question of
the motivations behind nuclear behavior. Do states seek nuclear weapons to have a
deterrent capability indefinitely, or are they produced to meet short-term objectives?
Further, when states take steps to build nuclear weapons, do they always intend on
actual production? Many statistical analyses do not provide an explanation for
why these reversals or seemingly half-hearted attempts to acquire nuclear weapons
materialize.
11
Singh and Way (2004) broaden the dependent variable to include exploration
and acquisition, yet they and many others implicitly assume that nuclear behavior
ends with acquisition. France’s actions with respect to nuclear weapons did not end
with acquisition. Instead, its leadership continued to perform above-ground tests
until the mid-1970s. Indeed, a recent debate in the nuclear proliferation scholarship
has emerged over how to measure nuclear acquisition (see Hymans 2010). A central
question concerns whether the proliferator needs to actively demonstrate that it has
access to a nuclear bomb. Regardless of one’s attitudes towards this debate it is
unclear why above-ground detonation should not be considered as significant of a
nuclear behavior as the others. After all, such tests embody the fulfillment of all the
steps undertaken up to and including acquisition.
3.2
Alliance Dynamics as the Independent Variable and Causal Mechanism
Goldstein (2000) offers a first major step in examining the role of intra-alliance
dynamics in nuclear proliferation decisions. He argues that middle powers like the
United Kingdom, France, and the People’s Republic of China cannot fully depend on
the support of their superpower patron in the event of nuclear conflict. In France’s
view, for example, the US would sooner allow Paris to be devastated rather than
allow its own cities bear the brunt of a Soviet attack. Consequently, to hedge against
the perceived incredibility of the patrons security guarantees while balancing the
threat of the opposing superpower, the middle power elects to have its own nuclear
capability that is sufficient to deter attack.
Goldstein’s analysis, however, has several important shortcomings that must be
addressed. First, he focuses his attention on only three positive cases of nuclear
proliferation, thus ignoring a larger number of non-cases in which the allies of either
the US or the Soviet Union did not acquire nuclear weapons. To some extent this
shortcoming is an artifact of focusing specifically on acquisition. As indicated above,
a number of secondary states engaged in a variety of nuclear behaviors during the
Cold War that stopped short of actually obtaining a nuclear weapon. Second, the
independent variables that Goldstein emphasizes are in fact scope conditions for his
theory. He attaches importance to anarchy and bipolarity, yet these two properties
of the international system did not vary in the 1950s and 1960s when the aforementioned middle powers made their proliferation decisions. Finally, the behavior of the
superpower patron remains under-theorized. He does not clarify why the superpower
patron could not make a more credible promise to protect its allies. Nor is clear as
to why the superpower would not apply statecraft to curtail efforts by its allies in
acquiring nuclear weapons.
12
Unfortunately, much of the nuclear weapons literature has neglected Goldstein’s
important insight that alliances and ambiguous patron support generate incentives
that secondary states face in making proliferation decisions. When alliances and nuclear umbrellas are taken into account, the hypotheses pertaining to these variables
are theoretically poorly developed and crudely studied. Singh and Way (2004, 878)
do find that alliances are negatively associated with exploration, pursuit, and acquisition, though the estimated effects vary in magnitude across the outcomes of interest.
Thus, alliances are powerful constraints in exploration and acquisition, yet they have
effects on pursuit that are statistically indistinguishable from zero. Similarly, Jo and
Gartzke (2007, 174) use a dummy variable to indicate a state’s alliance with a nuclear power. Their prediction is that having a nuclear umbrella lowers the probability
of a state seeking a nuclear arsenal. Their findings are mixed. On the one hand,
they note that “the nuclear umbrella provided by nuclear patrons dissuades potential
nuclear contenders from acquiring nuclear weapons.” On the other hand, they find
that “nuclear protégés are no less likely to initiate nuclear programs” (176). This
seeming contradiction raises interesting and important substantive questions about
the role of alliance dynamics, yet the authors do not explore this issue.
Recent scholarship has claimed that policies intended to strengthen alliance relations in fact encourage proliferation. Using data on over two thousand bilateral
civilian nuclear cooperation agreements between 1950 and 2000, Fuhrmann (2009b)
argues that states sign nuclear cooperation agreements with other states when they
seek to strengthen either their allies and alliances or the ‘enemy of their enemies’.
Similarly, Kroenig (2010) claims that states provide others with sensitive nuclear assistance in order to empower the rivals of their adversaries. He qualifies this finding
by arguing that states do not offer such assistance when they already can project
power over the receiving state. These policies help spread nuclear weapons in international relations (Fuhrmann 2009a; Kroenig 2009). However, it is plausible that
states provide civilian nuclear assistance to dissuade their alliance partners from
pursuing their own independent nuclear capabilities. By attaching safeguards and
conditions to provisions of nuclear assistance, states seek to gain a level of control
over the nuclear activities of their alliance partners.
The importance of properly taking into account alliance dynamics is that doing
so helps us gain a wider appreciation of the wider environment in which nuclear
decisions are made. Unfortunately, many studies ignore alliance dynamics and take
an otherwise monadic view of nuclear proliferation. Other studies that focus on
domestic attributes and normative concerns are especially prone to committing this
reductionist error. Hymans (2006) argues that some leaders hold certain world-views
that make them have greater proclivity towards the development of an independent
nuclear capability. In another work, Horowitz (2010) contends that states that can
13
meet the financial costs and possess the organizational structures that can absorb
costly technologies are most likely to acquire and retain nuclear weapons. As different as these works may be, they are similar insofar as their main theories are
pre-environmental. The efficacy of mapping world-views onto state policy requires a
favorable international environment. Likewise, having the organizational and financial capital to integrate nuclear weapons might be meaningless if the state cannot
defeat efforts by other states to prevent its attempt at acquisition.
Simply put, although monadic analyses highlight some important independent
variables, they ignore the larger strategic context in which nuclear decisions are
made. Realist variables like dispute involvement and enduring rivalry certainly provide strategic motives for nuclear proliferation. However, these variables do not capture the possible reaction of other states to any level of nuclear behavior. Even in
the presence of the foregoing conditions a state might not engage in nuclear behavior
if its leadership believes it would be the target of a preventive strike, let alone alternative forms of sanctioning that impose high costs. Indeed, the existing literature
assumes that the superpower is ultimately powerless when it comes to curbing the
nuclear ambitions of other states. Yet superpowers have much to lose from the spread
of nuclear weapons such that they actively engage in counterproliferation policies.
Studies of nuclear proliferation, therefore, should take more seriously the bargaining dynamic that ensues between secondary states and superpower states to better
explain variation in states’ behavior with respect to nuclear weapons programs.
To conclude, three particular weaknesses characterize the current literature relevant to the topic of this project. First, studies of nuclear behavior unnecessarily
limit the scope of their dependent variables. By largely focusing exclusively on proliferation and non-proliferation, they ignore more subtle patterns of behavior that
we observe between those outcomes. Further, the literature is wrong to assume that
nuclear behavior ends with the acquisition of nuclear weapons. Upon acquisition
states make decisions on whether to demonstrate their newfound nuclear capabilities
through the use of above-ground tests. More recent literature has broadened the
range of nuclear behavior by calling attention to states that provide others with sensitive nuclear technology. Second, there is very limited work that properly takes into
account intra-alliance dynamics and nuclear proliferation. This body of scholarship
has yielded inconclusive theories that unsatisfactorily explain why we observe secondary states that enjoy the patronage of a superpower engage in different nuclear
behaviors. Third, a more general problem of the literature is to divorce nuclear decisions from the strategic context in which they are made. The significance of nuclear
behavior is that such activities provoke insecurity amongst neighboring states and
challenge a superpower’s ability to control a region. Existing arguments that focus
on domestic-level variables are especially prone to overlook these important strategic
14
consequences. There remains further work to be done in examining the interactions
between secondary states hat engage in these activities and the superpower patrons
that have incentives to curb them.
4
Theory of Nuclear Behavior in Alliances
In this section I develop a theory of nuclear behavior in the context of alliances with
two goals in mind. The first goal is to identify the conditions under which secondary
states engage in nuclear behaviors when they already have the patronage of a superpower. I begin with the assumption that states are security-seeking. Secondary
states find alliances with superpowers attractive because of the protection allotted
to them from common adversaries. However, these very protections vary in their
perceived strength. In light of this issue, I discuss how states evaluate and respond
to the credibility of alliance commitments, which ultimately reveals the conditions
under which secondary states initiate nuclear behavior. The second goal is to describe the bargaining dynamics that ensue between the superpower and the secondary
state once the latter decides to initiate in nuclear behavior. I first explore in greater
depth the secondary state’s objectives associated with their nuclear decisions. Then
drawing on the previous section, I analyze how information asymmetry and credible
commitment problems affect the bargaining process.
4.1
Security-Seeking and Superpower Patronage
A fundamental observation in the study of international relations is that no overarching sovereign authority exists to adjudicate the affairs of states and enforce contracts
between them. Scholars differ in their views of the implications of the anarchic nature
of the international system for the behavior of states. Rather than wade into this debate, I assume that states are generally security-seeking and thus prefer policies that
encourage stability rather than undermine it. Only under certain conditions, generated by the security dilemma, do states adopt aggressive policies.7 Otherwise, states
are usually biased towards the status quo, eschewing efforts aimed at expansion,
whether through territorial aggrandizement or the active coercion of other states
into compliance. The introduction of nuclear weapons in the international system
has reinforced the status quo bias of states. After all, these weapons favor defensive
postures because they enable states to possess a retaliatory capability that sharply
raises the costs of an attack by an adversary, significantly reducing the incentives
7
In particular, these conditions include the offense-defense balance and whether offensive or defensive
postures can be usefully distinguished (Taliaferro 2000/01, 130).
15
to adopt offensive strategies.8 Even when states do not have their own independent
nuclear capability, the risk of nuclear punishment by a superpower for a conventional
attack reduces the appeal of offensive postures.9
Secondary states require protection in order to prevent the encroachment and use
of coercion tactics by an adversary. Typically, states that face an external threat
are compelled to make choices over whether to arm (internal balancing) or form
alliances (external balancing). The choice is not necessarily mutually exclusive because states often select a mixed bundle of these goods (Morrow 1993). Secondary
states, in particular, lack the material bases of power that can be converted into a
conventional military power that is alone sufficient for ensuring the integrity of their
values. Accordingly, during the formative years of the Cold War, the sheer imbalance in military capabilities within the international system led secondary states to
form coalitions with their preferred superpower. Some security-dependencies, such
as the American-led alliance system in East Asia, survived the end of the Cold War
and the passing of the Soviet threat. Many secondary states elected to enjoy ‘peace
dividends’ while the United States maintained (and expanded) its global military
presence. Indeed, the protections offered by the United States have retained their
appeal for those secondary states (such as South Korea) that continue to face local
threats.10
Since the introduction of nuclear weapons to world politics, one salient form of
superpower patronage comprises a particular variety of nuclear deterrence. Under
deterrence both adversaries recognize each other’s capacity to escalate a nuclear
exchange to the point where the costs of attack sharply outweigh its benefits. Yet
the Cold War was broader in scope than a simple rivalry between two states. It was
a condition of political conflict and military tension between two competing alliance
systems that the superpowers managed. The fate of each superpower’s dominance
was, thus, closely tied to the fate of their junior partners.
In light of this strategic situation, the superpowers pursued the policy of extended
deterrence. This form of deterrence consists of threatening retaliatory action on an
alliance partner’s adversary in order to prevent a military attack on that junior
partner. As Huth (1988, 424) notes, this policy is adopted when “the policymakers
8
See Taliaferro (2000/01, 139), Jervis (1989, Chapter 1), and Lieber (2000, 100). For an extended
discussion on how the presence of nuclear weapons generates disincentives for intended war within the
international system, see Adams (2004, 54-58).
9
For a rejoinder to the view that nuclear weapons have salutary effects on the international system, see
Mueller (1988).
10
In fact, cases exists in which secondary states fostered new security ties with the US following the end
of the Cold War. Many former members of the Warsaw Pact joined NATO and sought close relations with
the US. Even Vietnam, a communist state, has sought to normalize relations with the US in the face of a
rising PRC.
16
of one state (‘defender’) threaten the use of force against another state (‘potential
attacker’) in an attempt to prevent that state from using military force against an
ally – or territory controlled by an ally (‘protégé’) – of the defender.” With the
advent of nuclear weapons, extended deterrence often took the form of a nuclear
umbrella.
This policy has several advantages for both the superpower and the secondary
state. For the superpower, upon achieving deterrence with the primary adversary, the
technical costs of raising a nuclear power to the secondary state are minimal. Second,
the nuclear umbrella strengthens the alliance relationship with the secondary state,
which can provide assets (such as basing rights, territory, or political support) in
return that further the superpower’s own foreign policy goals (Lake 2009; Morrow
1991). As a result of this transaction, the secondary state has what appears, all else
equal, a much lesser need for developing its own nuclear capability. It can free ride
on the nuclear services of the superpower patron while allocating its own resources
towards domestic programs (see Goldstein 1995).
These benefits, however, come with associated disadvantages. The superpower
must demonstrate the credibility of extended deterrence to two different audiences.
First, the superpower needs to persuade the adversary that it will come to the defense of its protégé despite the costs it might incur as a consequence. Second, the
superpower also has to assure the leadership of its protégé that its support is reliable
in the event of nuclear conflict. The adversary and the protégé especially have to be
convinced that the superpower is not willing to simply sacrifice friendly territory.
Another problem with providing extended deterrence is that the superpower is
vulnerable to taking on additional commitments beyond its original goals.11 Hypothetically, a superpower that is able to credibly commit to extended deterrence
generates incentives for the junior partner to engage in risky international behavior
that might provoke an unfavorable military reaction from an adversary. In such a
situation, the superpower may find itself dragged into a conflict it would prefer to
avoid.12 The reasons for coming to the aid of a wayward ally can encompass a variety of factors. The superpower does not want to be seen as willing to abandon its
junior partners nor would it want a situation to develop that would grant its rivals a
geopolitical advantage. Unfortunately, by taking on these additional commitments,
the superpower creates a moral hazard problem amongst other alliance partnerships
by demonstrating its willingness to support a secondary state in spite of its undesired
behavior.
11
Snyder (1984, 467) and Taliaferro (2004, 184) describes this pathology of decision-making as ‘entrapment’ and ‘escalating commitments’, respectively.
12
For more on states being dragged into armed conflict because of the behavior of their allies, see Christensen and Snyder (1990) and Leeds (2000).
17
4.2
The Perceived Unreliability of Extended Deterrence as a Driver of
Nuclear Behavior
The challenges associated with guaranteeing security in the form of extended deterrence imply that the credibility of superpower support varies across space and time.
How secondary states regard the credibility of a superpower’s security commitment,
my independent variable, affects their likelihood of engaging in nuclear behavior.
Doubts over the protection seemingly afforded by the superpower encourage the secondary state leadership to seek greater insurance for the preservation of its values.
When these concerns are allayed, fewer incentives exist to build an independent
nuclear capability.
I argue that states refer to various indicators in order to determine the credibility
of superpower support. These indicators can be distinguished between those that
relate to the superpower’s willingness and those that relate to the superpower’s ability
to protect the secondary state. Willingness captures the convergence of political
values between the superpower and the secondary state. Specifically, overlapping
security interests and shared strategic objectives serve as one basis for bolstering
the credibility of nuclear policy. This point might seem at first redundant because
the formation of an alliance is already a function of some similarity in foreign policy
goals and political preferences (Gartzke and Simon 1996). However, this observation
obscures how foreign policy interests of allies can change, and even diverge from each
other, once the alliance has been established (Ratner 2009; Voeten 2004).
These divergences may develop for several reasons. The superpower might seek
pursue friendly relations with rivals unique to the secondary state or has adopted new
goals that meaningfully contradict with those of its junior partner. These divergences
matter if they potentially endanger the secondary state. When geopolitical goals
strongly converge the secondary state recognizes that its own integrity advances the
foreign policy of the superpower. The credibility of the nuclear umbrella is reinforced
if the fate of both the patron and the secondary state are closely bounded. The fear
that the superpower would not simply abandon its ally rather than incur the punitive
costs of a nuclear exchange is attenuated, if not eliminated. Unfortunately, relying
on strongly convergent security interests alone does little to prevent opportunistic
behavior by the secondary state and the problems of escalating commitments that
might ensue.13
One powerful strategy that the superpower can adopt to demonstrate its willing13
In keeping with the ‘security-seeking assumption’, note that the superpower is a ‘first-mover’ insofar
as it seeks foreign policy objectives at variance with those of its ally. We must first observe the patron’s
actions, and then see if proliferation occurs. Otherwise, we risk ex post reasoning in which we observe
proliferation and then look for policy divergence. I thank Michael McKoy for raising this specific point.
18
ness to support the secondary state is to deploy a sizable contingent of conventional
military forces on the territory of its junior partner. This policy solution addresses
many of the shortcomings of providing extended deterrence by mitigating the moral
hazard problem described above. Fortifying friendly territory in this manner raises
the costs of enemy attack and, with the consent of junior partner, sends a powerful signal of the superpower’s commitment to protecting its ally. After all, the
superpower is willing to risk its own troops and expend war material in providing
protection to the lesser power. The costs of troop deployments overseas, even if they
are intended to protect allied territory rather than fight wars, are expensive to sustain. They require substantial fiscal resources to quarter and train troops, maintain
and replace military hardware such as vehicles and weaponry, and even pay suppliers of required food and other services. Although American military bases have
expanded in number since the end of the Cold War, one estimate pegs the cost of
maintaining overseas bases and troop deployments (excluding those in Afghanistan
and Iraq) at about $102 billion dollars per annum in 2009.14 To put this figure
into perspective, this amount is in fact almost as large as the next largest military
spender’s total expenditures for the same year.
This approach is also advantageous because it also curbs the potential moral hazard problem of offering a nuclear umbrella. With the presence of its conventional
forces, the superpower places de facto limits on the foreign policy options available
to the hosting state. The secondary state might also have to consult even more on
foreign policy and coordinate extensively on military policy. To be sure, the initial
intentions behind the stationing of troops might have had less to do with enhancing
nuclear deterrence than placing these constraints on a state’s foreign policy. For
example, the American occupation of West Germany and Japan was a direct consequence of the Axis Powers’ defeat in the Second World War rather than a desire to
strengthen alliance commitments. Nevertheless, military deployments of even this
type can serve to generate credibility for extended deterrence once these needs arise.
The second set of indicators that secondary states use concerns the superpower’s
ability to protect the secondary state. I have already discussed how significant conventional military deployments constitute a strong signal of the superpower’s commitment to the security of the secondary state by virtue of their costs. However, the
very expenses associated with having military bases abroad offer the superpower’s
leadership a target for cutting costs in times of severe economic constraints. The
temptation to reduce conventional force deployment abroad and retrench militarily
under these conditions is especially attractive in light of the substitution effects that
nuclear weapons provide. US President Dwight Eisenhower recognized this bene14
Statistic quoted in Hugh Gusterson, “Empire of Bases,” Bulletin of Atomic Scientists, March 10, 2009.
Available online: http://www.thebulletin.org/.
19
fit of forward-deployed nuclear forces in his advocacy of the ‘New Look’ policy in
Western Europe. By investing more in a large but relatively cheap nuclear arsenal,
Eisenhower hoped to keep American military spending under control while Western
European allies would bear more of the conventional defense burden (Friedberg 2000;
Gaddis 1982).
Trends in defense expenditures, thus, serve as a useful indicator to gauge the
superpower leadership’s interest in providing firm commitments to allies and other
security partners around the world. Downward trends, especially if they are persistent and large, might reflect a policy of superpower retrenchment. According to
MacDonald and Parent (2005, 11), retrenchment is a “policy of retracting grand
strategic commitments in response to a decline in relative power” and can include
options that “[economize] expenditures, [reduce] risks, and [shift] burdens. Losses
in relative power, however, are not the only drivers for the usage of these policy
options. Public attitudes hostile to high military expenditures or an active global
presence would place pressure on leaders to allocate resources towards the domestic
front and enjoy some semblance of ‘splendid isolation’. Alternatively, leaders in the
superpower might anticipate entanglements abroad that they prefer to avoid so not
to be dragged into future conflicts.15
Indicators that track the superpower’s ability and willingness to provide support
in a nuclear crisis are useful for secondary states in drawing general conclusions
about security guarantees. To simplify, each of these dimensions can be distinguished
dichotomously as either weak or strong. Table 1 crosses the values so to highlight the
level of risk for the secondary state to engage in nuclear behavior associated with each
configuration of willingness and ability. When the superpower demonstrates a strong
ability and willingness to back its security commitments, then the secondary state has
little reason for engaging in nuclear behavior. The risk is, therefore, low. When those
attributes of a security government are both judged to be weak, then the secondary
state is at much higher risk to seek nuclear weapons as insurance. If only one of
ability or willingness is judged weak, the risk for nuclear behavior is greater than
when the secondary state evaluates the superpower’s security commitment as strong
but less severe than when the secondary state perceives the security commitment to
be completely unreliable.
To summarize, I begin with the assumption that secondary states are primarily
security-seekers. I argue that the perceived political unreliability, embodied in foreign
policy divergence, or the perceived economic unreliability, embodied in anticipated
15
To be sure, willingness and ability are not independent of each other. Variations in the willingness and
ability of a superpower to offer strong security commitments are, therefore, correlated to some extent. For
example, it is possible that involvement in a protracted and costly conflict abroad induces both a need for
military retrenchment and a contraction of foreign policy commitments.
20
Superpower Ability
Weak
Strong
Superpower Willingness
Weak
Strong
High Risk
Medium Risk
Medium Risk Low Risk
Table 1: Superpower Security Commitments and Secondary State Riskiness for Nuclear Behavior.
conventional military withdrawals, by the leadership of the secondary state increase
the likelihood that nuclear behavior will be initiated. With a sharp divergence of
foreign policy goals the secondary state might not want to entrust its security to the
superpower, thus offering added motivation for internal balancing through nuclear
means.
When policy proposals that suggest a reduction of conventional military forces
abroad are mooted, the secondary state similarly develops greater reason to doubt the
reliability of the patron’s support. In the view of the secondary state, the potential
withdrawal (or reduction) of conventional forces implies a lack of commitment on the
part of the superpower to come to its defense. The commitment problem in defending
friendly territory during a nuclear attack becomes especially acute. Secondary states
might come to the belief that it would be abandoned or sacrificed more readily if the
superpower has fewer troops on the line. Of course, allied states that do not host the
superpower’s conventional military force would also observe such developments with
unease. Superpower retrenchment still suggests a shying away from commitments
that junior alliance partners would prefer to see preserved. Simply put, the costs and
logistics of extended deterrence reveal both why and when secondary states might
seek to develop their own arsenals.
4.3
Nuclear Behavior as Insurance and Bargaining
As doubts over the superpower’s commitment increase, the secondary state will be
more apt to explore military policies that insure against the effects of patron abandonment. They are more likely to adopt ambiguous nuclear postures or even begin
pursuing their own nuclear weapons program. Having a nuclear weapons arsenal
offers a robust insurance policy for the secondary state. Goldstein (2000) notes that
the secondary state is not required to develop such an extensive and technologically
advanced arsenal as those possessed by the US and the Soviet Union. Rather, it
needs to have a sufficient number of weapons that are capable of second-strike delivery to deter the adversary from launching a direct attack. Indeed, the philosophy
guiding the secondary state’s approach to deterrence is different from that of their patrons. Superpowers rely on the threat of ‘controlled escalation’ in which they proceed
21
through limited but gradually more intense exchanges to communicate their resolve
in inflicting damage. Engaging in controlled escalation requires advanced command
and control systems as well as the ability to absorb nuclear damage. These requirements are especially demanding for smaller states that are less able to meet them.16
Consequently, such states opt for a ‘poison pill’ strategy in which their deterrence
policy rests on the threat of ‘uncontrolled escalation’. The high likelihood of both
parties losing control of a nuclear exchange characterizes this form of confrontation.
For such an exchange to occur there needs to be an element of risk that neither side
could attenuate (Powell 1987, 719). A state’s technological capacity for managing
its nuclear weapons poses such a risk if it is involuntarily underdeveloped and thus
prone to accidents and other organizational failures. These concerns gain significance
when it comes to secondary states. Their national command structures are likely to
be small and more concentrated than is the case for superpowers. In the event of a
nuclear exchange, they face a much higher probability of being thrown into disarray
during the conflict’s initial stages. Nuclear retaliation, therefore, becomes less inhibited and results in the infliction of massive damage on the adversary (Goldstein
2000, 47-51). Backwards inducing from this possibility leads the adversary to refrain
from direct military attack on the secondary state.17
Such are the advantages of acquiring nuclear weapons, but secondary states have
to pass through various stages of nuclear development first. Indeed, there is a paradox underlining nuclear weapons acquisition. As much as having a nuclear arsenal
might engender international stability, the process by which states finally acquire
nuclear weapons generates instability (Sagan and Waltz 1995). Adopting an ambiguous nuclear posture or pursuing a nuclear weapons program provokes alarm amongst
neighboring states, regardless of whether they are allies. Those states might be
unsettled by the uncertainty of the potential proliferator’s intentions and the fear
of being vulnerable to nuclear blackmail in the future. Moreover, though the secondary state acts to hedge against superpower abandonment in their effort to obtain
greater foreign policy autonomy, they also risk punishment from the superpower for
threatening to undermine its alliance structures.
States that engage in nuclear behavior are in effect bargaining over the terms of
16
As Sagan (1995) indicates, the technical demands of ‘controlled escalation’ are high for even superpowers
to meet and, thus, conceivably much higher for less capable states.
17
A nuclear weapons program might seem to some readers as an extreme course of action to take in the
face of incredible security commitments. However, it is important to keep in mind the scope conditions
I defined earlier. These scope conditions pertain to the existence of ‘strategic triangles’ in which a state
finds itself between a patron and an adversary. In the face of a nuclear-armed adversary, nuclear weapons
increase the secondary state’s abilities to deter an attack. Relying exclusively on a conventional military
build-up is problematic if the state wants to be able to retaliate following a nuclear strike.
22
the superpower’s security guarantees. Bargaining is commonplace in international
politics because many resources and other goods (such as security) that states aspire
to possess are scarce. How these goods should be allocated are often subject to
dispute (Reiter 2003, 27). If security partnerships, in the form of basing rights and
alliances, can usefully be likened to contracts (Cooley 2008; Cooley and Spruyt 2009;
Lake 2009), nuclear behavior consists of revising the terms of the original ‘security
contract’ regarding how and to what extent the superpower patron should provide
security.
When nuclear behavior is at issue, two mechanisms – asymmetric information
and credible commitment problems – that prevent states from reaching an agreement
deserve mention. Informational problems generally occur because states have private
information regarding their military capabilities, their resolve in obtaining foreign
policy goals, or their payoffs in prevailing in certain crisis situations. Without a
central authority in the international system that can monitor state behavior, states
also have incentives to misrepresent private information. Commitment problems arise
when states cannot agree to mutually preferable bargains for fear that the other side
would renege on their promises. The inability to credibly commit to an agreement
is a function of anarchy because there is no sovereign authority that can enforce
agreements.18
These two mechanisms are especially salient in bargaining over nuclear technologies and the superpower’s provision of security. Informational concerns emerge
because states might prefer to be discreet about their activities due to the sensitive
nature of nuclear proliferation. Provoking alarm from their regional neighbors could
serve the secondary state’s interests so to startle the patron into granting concessions. However, the manipulation of risk that such behavior entails is dangerous
because it invites harmful sanctions and aggressive action by other states. Information problems are also likely because states might mislead their patrons into granting
concessions that in fact increase the risk of nuclear proliferation. Fuhrmann (2009a,b)
argues that states might offer their allies with civilian nuclear assistance in order to
signal their commitment to their security partnership. Though the sending state
might not desire nuclear proliferation, it inadvertently raises the risks by transmitting such sensitive technologies (see also Kroenig 2010). It is theoretically possible
that states promise to use such nuclear assistance for civilian purposes while secretly
planning to deploy this technology towards military ends. Secondary states, thus,
18
According to Fearon (1995), issue indivisibilities impede a negotiated settlement because a state absolutely will not compromise on an object of disagreement. This third explanation for why states cannot find
bargains is less convincing because states should still find issue linkages and offer side payments. However, I
show below that some alternative arguments regarding nuclear behavior emphasize the importance of issue
indivisibility.
23
have incentives to dissemble their activities by claiming that any nuclear program in
operation is intended solely for civilian purposes.
Commitment problems also are manifest when secondary states engage in nuclear
behavior. I have already discussed how commitment problems create incentives for
states to initiate nuclear behavior. The perceived unwillingness of the superpower
patron to defend the secondary state needs to be satisfactorily addressed so to assuage
the secondary state’s concerns. One simple strategy is for the superpower to rectify
whatever it was that indicated its lack of reliability to the secondary state in the first
place. Towards this end the superpower could, for example, retain or increase its
conventional military presence on the ally’s territory. Alternatively, it could establish
forward deployed nuclear forces in order to signal its commitment to its ally. Once
nuclear behavior is initiated, however, the superpower is not the only state that needs
to demonstrate the credibility of its commitments.
Even if it were to accede to the superpower’s demands, the secondary state also
needs to resolve commitment problems created by its own behavior. It has to demonstrate that if it were to back down it would not renege and thus resume its nuclear
activities in the future. The superpower, after all, does not want to bargain with the
same ally over the extent and form of the security it provides repeatedly. As a result, the superpower would demand a costly action by the secondary state that would
impair its ability to reinitiate nuclear behavior in the future. To lend greater credibility to its nuclear reversal, costly actions include the open disavowal of a nuclear
program, the complete destruction of physical materials integral to the production
of nuclear weapons, and accession to multilateral agreements that proscribe nuclear
behavior. I consider the meaning and costliness of each of these actions in turn.
The open disavowal of a nuclear program is costly because the secondary state
is making a public admission of its nuclear behavior. By doing so the state accepts
certain damages to its reputation because nuclear weapons program are a possible
indication of dishonesty (if maintained covertly), ill-intent, and its propensity for
risk-taking behavior. In other words, a declaratory statement that renounces past
nuclear behavior might make other states suspicious of the state’s foreign policy
objectives. To be sure, an admission might also be a pre-condition for security
cooperation so to eventually regain the trust of other states. However, an admission
communicated publicly to other governments cannot be construed as ‘cheap talk’.
Nuclear recidivism would reveal a level of insincerity and irresponsibility that would
trigger a more aggressive reaction by those states already threatened by the offending
state’s behavior.
The complete destruction of physical materials integral to the production of nuclear weapons is another costly action that a state can undertake to make its nuclear
24
reversal credible. Nuclear weapons are difficult to manufacture even if a state possesses nuclear technology aimed at peaceful materials. States require heavy water
and uranium – resources that have historically been difficult to obtain in the nuclear
era – to produce nuclear weapons. By ridding themselves of such materials the state
can more credibly assert that it would not be able to resume its nuclear behavior
without significant effort. Of course, the specialized knowledge that is required in
building a nuclear weapon cannot simply be forgotten. Without the necessary physical materials, however, the state’s ability of properly resuming nuclear program is
severely handicapped.
Accession to multilateral agreements that proscribe nuclear behavior is costly
because it includes similar effects as open disavowal. As Martin (1993) argues, working through international institutions raises international audience costs that leaders
would incur should they renege. In other words, such multilateral commitments enhance credibility by “[raising] the ex post costs of noncompliance above those that
might be incurred in the absence of the treaty.” Further, these legal commitments
also serve to reduce uncertainty over their intentions with respect to nuclear technologies because states might have to agree to monitoring and enforcement mechanisms
(Axelrod and Keohane 1985; Koremenos, Lipson, and Snidal 2001). If the institution
or agreement effectively generates information and enforces agreements, becoming a
signatory member places major constraints on the state’s freedom of action in the
future.
Getting the secondary state to make such credible commitments has its challenges. Attacking the nuclear facilities of a junior alliance partner is not an option.19
The threat to use force is incredible because the superpower both relies upon the
secondary state for some strategic assets and does not want to develop a reputation
for using military force on its own allies. The superpower thus has to resort to either diplomacy or economic statecraft in order to dissuade the ally from engaging in
nuclear behavior and bring it into compliance.
4.4
The Role of Economic Statecraft in Nuclear Bargaining
I argue that economic statecraft offers the superpower a more promising route to exercising leverage over the secondary state. By economic statecraft I refer to the use
of economic instruments intended to induce changes in the target state’s behavior.
This definition encompasses the actual or threatened use a variety of tools such as
embargoes, boycotts, withdrawal of ‘most-favored-treatment’, tariff increases, freez19
For more on when states attack or consider attack nuclear facilities of non-nuclear weapons states, see
Fuhrmann and Kreps (2010).
25
ing assets, aid suspension, and import and export controls (Baldwin 1985, 39-41).
Superpowers are particularly adept at applying economic statecraft because their
economies are sufficiently diverse and characterized by high levels of aggregate demand such that they influence world prices and counter potential coercers (Drezner
2007).
States that lack diverse or large economies are particularly sensitive to the superpower’s application of economic statecraft. These states might depend on the
superpower for aid packages as well as its market for selling exports. Cutting off
aid flows or imposing export controls could harm the secondary state’s economy and
even undermine the incumbent government’s chances for domestic political survival.
After all, the economic costs created by a superpower’s action might be borne by
members of the secondary state’s population. To avoid the further escalation of these
costs the secondary state might be required to desist from its nuclear activity and
make the commitments desired by the patron.
Many states likely backwards induce the harmful effects of a superpower’s application of negative sanctions, thus leading them to eschew nuclear behavior altogether.
Some states, however, place an overriding importance to their security to such an
extent that they are willing to risk the costs of negative sanctions. These states face
such salient external threats that they feel particularly obliged to pursue an insurance policy in the face of a seemingly unreliable security commitment offered by a
superpower. Indeed, a perceived decline in the credibility of extended deterrence
may even lead some leaders of secondary states to correspondingly believe that the
superpower will be unable to apply significant pressure on them to reverse on their
nuclear behavior.
Aside from having the technological capacity to even create their own nuclear
arsenals, secondary states that wish to ratchet up their nuclear behavior need to,
therefore, satisfy an important requirement. They have to be able to respond effectively to the superpower’s demands to curb their activities and not succumb to
diplomatic or, more importantly, economic pressure. Economic power offers secondary states the means to inoculate themselves from superpower punishment and
coercion tactics.20
To be sure, states lacking in these characteristics can still make strategic gains
in pursuing lower forms of nuclear behavior, even if they realize that they might not
be able to ever achieve full acquisition. A common assumption in the international
security scholarship is that nuclear behavior only generates insurance by granting
20
A number of quantitative studies show that economic power and military power are highly correlated
with nuclear proliferation (Horowitz 2010; Jo and Gartzke 2007; Singh and Way 2004). These same studies,
however, include the United States and the Soviet Union.
26
proliferating states a deterrence capability. I argue that this is not the case. Such
behavior might be intended to attract the superpower’s attention and extract assurances over its security guarantee. Knowing that the superpower would prefer it to
lack a nuclear weapons arsenal, the secondary state can exploit its patron’s fears that
it will break ranks, pursue a nuclear weapons program, and cultivate an independent
foreign policy. Thus, when the patron’s leadership is openly considering scaling back
conventional force deployments, the secondary state could use nuclear behavior so
to remind the superpower’s leadership over the need to assure its allies of its defense
commitments. States that engage in nuclear behaviors towards such ends will only
do so until they either receive a desirable and credible amount of assurances. Yet
they might fail in obtaining these objectives if they experience an intolerable level of
coercion by the superpower.
The discussion so far has focused largely on ambiguous posturing as well as the
pursuit, and acquisition of nuclear weapons to the exclusion of detonation. I argue
that a similar logic guides this behavior as well. Detonation, as explained above, is a
declaratory action that affirms a state’s possession or acquisition of nuclear weapons.
When a secondary state chooses to acquire nuclear weapons but foregoes detonation,
it is certainly using its opacity to keep adversaries second-guessing over the outcome
of attack. More relevant for this discussion, however, choosing not to detonate might
still be a function of the secondary state’s dependence on the superpower for other
security goods. Not explicitly confirming its nuclear weapons program in such a
manner does not merely maintain the fiction of extended deterrence; it is an action
that deliberately eschews making such a strong signal over the secondary state’s
intentions to pursue an independent foreign policy. Because a superpower would
prefer nuclear opacity to nuclear transparency on the part of the secondary state,
non-detonation could be a product of a ‘gentleman’s agreement’ between the two
states. In such an agreement the superpower pays off the secondary state with some
goods (e.g., diplomatic support, economic aid, military hardware) in exchange for
non-detonation.
4.5
Summary
I advance the argument that doubts over the credibility of a superpower patron’s provision of extended deterrence lead secondary states to engage in nuclear behavior.
To draw inferences regarding the reliability of these security guarantees, secondary
states refer to various indicators believed to track the superpower’s willingness and
ability to come to their support. Specifically, secondary states assess the extent to
which their strategic objectives and political values are similar so to gather information on superpower willingness. Secondary states observe military spending trends
27
as well as the economic health of the superpower in gauging the ability of the superpower to uphold its commitments. When these indicators point to a shirking
of superpower commitment, secondary state’s are more willing to initiate nuclear
behavior in order as a form of insurance against superpower abandonment and an
attack by the adversary.
This discussion regarding the onset of nuclear behavior raises several observable
implications and hypotheses. They are enumerated below:
Observable Implication 1 Nuclear behavior by a secondary state is more likely
during a severe economic downturn in the superpower’s economy.
Observable Implication 2 Nuclear behavior by a secondary state is more likely
when there are significant downward trends in the
superpower’s military expenditures.
Hypothesis 1 The probability of a state initiates nuclear behavior increases if
there is growing reason to believe that its superpower patron is scaling back on its conventional military commitments.
I also argue that nuclear behavior constitutes a bargaining process through which
the secondary state negotiates the security it receives from the superpower. Problems
relating to asymmetric information and incredible commitments hamper the ability of
these states to reach an agreement. However, the superpower can practice economic
statecraft in order to punish the secondary state and force it to back down. This
observation leads to a second hypothesis:
Hypothesis 2 The ability of the secondary state to ratchet up its nuclear behavior
is inversely related to its economic dependence on the patron.
5
Alternative Arguments
I call attention to three alternative explanations that might better account for the
observed variation in nuclear behavior. The first set of explanations are drawn from
the realist literature. The second explanation emphasizes the role of domestic politics. Each of these explanations also yield different expectations for how bargaining
processes should unfold.
28
5.1
Existing Realist Explanations
My theory of intra-alliance nuclear behavior includes some realist qualities, particularly in its emphasis on security-seeking behavior and assessments of superpower
support. Nevertheless, it is useful to contrast it with existing realist explanations of
nuclear behavior. Several deductive arguments and their hypotheses can be gleaned.
As noted earlier, the anarchic structure of the international system compels states to
be primarily concerned with their own interests. This self-help environment requires
states to select military policies that reduce their dependence on others for help.
The structural realist view is that the most powerful states in the international
system will acquire nuclear weapons in order to level the capacity of destruction
between them. Alternatively, states that face existential external threats will acquire
nuclear weapons.21 Indeed, some realists argue that states that have a strong desire
for nuclear weapons would spare no effort in getting them (Betts 2000). With respect
to less capable states, structural realism yields less conclusive predictions. On the
one hand, states with sufficient industrial capacity should develop their own nuclear
weapons to obtain sufficient foreign policy autonomy. On the other hand, middling
or weak states also depend on superpower support, which either lowers the incentives
for nuclear weapons or makes them vulnerable to demands to cease nuclear weapons
activities. The clearest hypotheses regarding nuclear behavior we can thus derive
from existing realist explanations concerns material capabilities and the external
threat environment.22
Hypothesis 3 A secondary state facing an intense external threat environment is
more likely to engage in nuclear behavior.
Hypothesis 4 A secondary state is more likely to ratchet up nuclear behavior as
its industrial capacity increases.
5.2
Domestic Politics Explanations
Some theories contend that domestic politics account for patterns of nuclear behavior. Solingen (2007) provides a key work that applies such an approach to studying
21
This claim is similar to Walt’s assertion that states respond to threats rather than capabilities in making
alignment decisions. See Walt (1987).
22
My portrayal of ‘realist’ theory is simplified and obscures the rich theoretical tradition associated with
this paradigm. For competing versions of realist theory, see Rose (1998); Schweller and Priess (1997). I
refer exclusively to threats and capabilities so not to introduce other variables that some would contest as
non-realist. For a discussion of this point, see Legro and Moravcsik (1999).
29
nuclear proliferation. She argues that decisions to acquire nuclear weapons after the
NPT entered into force correlated with governing coalitions’ preferences over their
state’s role in the global economy.23 Liberal internationalist coalitions that seek
greater integration of the global economy eschew the nuclear option, recognizing
that nuclear behavior would risk reducing them to pariah status. Nationalist coalitions are, by their nature, inward-looking and prefer to adopt self-sufficient policies
that detach them from the global economy. Economic self-sufficiency and nationalism maps onto their willingness to embrace nuclear weapon programs. Indeed,
nationalist leaders might seek these weapons as a tactic to rally their populations,
stir nationalist rhetoric, and divert attention away from domestic problems (Snyder
2000). Thus, we can derive two hypotheses:
Hypothesis 5 Economically liberal states are significantly less likely to engage in
nuclear behavior.
Hypothesis 6 Economically nationalist states are more likely to engage in nuclear
behavior.
These explanations are generally silent on how the political and economic interests
of governing coalitions affect bargaining with superpower patrons. Nevertheless,
these theories provide a coherent account of the preferences of the actors as well as
the incentive structures they face. As a result, it is possible to explore how such
actors would bargain with a superpower patron.
Because economically liberal states are not expected to engage in nuclear behavior, we can turn our attention immediately to states ruled by economic nationalists.
These states generally exhibit non-democratic features. The choice to forego engagement with the global economy often reflects a set of political values that center
on national self-sufficiency and tight economic policies that severely constrain the
commercial activities of the state’s population. To justify these restrictions, leaders often appeal to ideologies that use the ‘nation’ as a key organizing principle for
political life. These ideologies provide such leaders as a basis of legitimacy because
they highlight the form that political loyalty and identity members of society should
possess (Greenfeld 1993, 3). Moreover, nationalism offers leaders a heuristic device
that clarifies the boundaries between acceptable and unacceptable political behavior
by denouncing undesired attitudes and activities as harmful to the ‘nation’ and, by
extension, the incumbent regime.
Because nationalism often relies on sharp distinctions between the ‘in-group’ and
the ‘out-group’, it is more prone to include xenophobic tendencies. Leaders can
23
It is not clear why she distinguishes between the pre-NPT world and the NPT world.
30
use nationalist rhetoric in this way to cast blame on alleged foreign depredations
for their country’s misfortunes and rationalize otherwise unpopular security policies
(Christensen 1996). Accordingly, when actually confronted with external pressure,
the leadership would publicize its defiance in the international dispute to further
bolster the regime’s legitimacy (Pape 1997, 33).
This observation regarding non-democratic regimes has unique implications for
how nuclear bargaining should unfold. If the leadership of the secondary state uses
nationalist rhetoric to rationalize its nuclear behavior, then it might make it harder
for the superpower to get the secondary state to back down. Goddard (2006) argues
that how actors legitimate their strategies affects the divisibility of the issue indivisibility under dispute. By issue indivisibility she refers to a ‘constructed phenomenon’
in which actors represent the object of contention in a way that it cannot be divided,
thus reducing the bargaining range to zero. In other words, no division of the good
is acceptable to at least one side of the conflict (37). In studying territorial disputes
in Ireland, Goddard shows that leaders choose rhetorical strategies designed to legitimate their claims. These strategies sometimes have unintended consequences as
they rally a coalitions of different interests together in support of the cause that
might in the future prove decisive in sustaining political authority. Thus, leaders
become dependent on maintaining these coalitions to survive politically, regardless
of whether these legitimation strategies prove to be suboptimal and prevent peaceful
settlement.
Non-democratic allies that rely on nationalism and choose to engage in nuclear
behavior thus may be harder for superpowers to face in nuclear bargaining. This possibility still exists even if the superpower is an ally because the secondary state can
still portray the struggle as ‘David against Goliath’. Leaders of such regimes are reluctant to back down because doing so compromises their stated agenda for advancing
national sovereignty and, by extension, undermines their basis of legitimacy. Claims
over the level of security provided by the superpower and the proliferation status of
the secondary state become incompatible and eliminate the bargaining range altogether. The mechanism preventing a bargain between the secondary state and the
superpower is neither asymmetric information and credible commitment problems,
but rather issue indivisibility.
Hypothesis 7 The ability of a secondary state to ratchet up its nuclear behavior
is positively related to its level of authoritarianism.
31
A related thesis argues that states under political transition, and not just those
that are authoritarian or economic nationalist, exhibit similar pathologies.24 States
that emerge from autocratic systems experiment with democratic institutions including elections, a wider franchise, and an open and competitive media. Regardless
of whether they mature into stable democracies or revert back to authoritarianism,
these states are empirically more likely to engage in risky and aggressive foreign policies, including war-fighting (Mansfield and Snyder 1995). One explanation for this
pattern is that elites left over from the previous regime engage in political competition with new elites that represents a democratic movement. In trying to muster
resources and build popular support, elites take advantage of weak institutional structures to aggrandize their power and control news media to advance their agendas.
The coalition that militant elites produce in these political environments, however,
are unstable. To sustain support from the coalition, these elites “[seek] victories
abroad” and generate prestige, thus leading them to engage in foreign policies that
risk war (33).
Though it is not clear why war should result from these dynamics (presumably,
neighboring states would try their best to prevent these wars from occurring), leaders
of these states might engage in nuclear behavior to build their domestic reputation.
In doing so, they might not be able to succumb to a superpower patron’s demands
for the same reasons described above. For fear of alienating their base of support,
these leaders become unwilling to reverse their nuclear behavior. Indeed, because
these elites shape the policy of a transitioning regime, they have added incentive
to flaunt the authority of the superpower that provided the previous regime with
support (Cooley 2008; Ratner 2009). By demonstrating their ability to defy external
pressure, elites can bolster their nationalist credibility. We can draw two hypotheses
from this discussion:
Hypothesis 8 Transitioning states are more likely to engage in nuclear behavior.
Hypothesis 9 Transitioning states are more likely to ratchet up their nuclear behavior.
Some scholars claim that states seek nuclear weapons to build international prestige. By international prestige, I refer to a coveted status that members of the
international community confer on a state on the basis of its accomplishments or
attributes. Hymans (2006); O’Neill (2006) have advanced hypotheses that explicitly
link prestige concerns and nuclear weapons. More pointedly, Rosecrance (1966, 35)
24
I refrain from describing these states as ‘democratizing’ in light of the assumed teleology found in this
appellation.
32
goes as far to say that “prestige may be the signal operative motivation for the acquisition of nuclear capability.” The problem with this type of explanation is that it is
too difficult to disentangle nationalist motivations from prestige motivations. After
all, acquiring international prestige boosts a leaders’ credentials in raising the stature
of the state’s national community. Because of this theoretical and methodological
issue, I consider any statements regarding the need to acquire international prestige
to imply nationalist motives for weapons development.
6
Conclusion
To summarize, nuclear behavior – ambiguous posturing as well as the pursuit, acquisition, and detonation of sensitive nuclear materials – creates international instability
and potentially invites punishment from superpowers. Why then do some states that
enjoy the protection of a nuclear capable superpower nonetheless engage in nuclear
behaviors? Explanations that claim such behavior is motivated by self-help do not
sufficiently account for the conditions under which states respond to threats with
nuclear behavior. Furthermore, self-help explanations are unable to account for the
cases of nuclear reversal or voluntary suspension of nuclear activities. To address
these shortcomings, I advance a new theory in which nuclear behavior is principally
informed by alliance dynamics between secondary states and superpower patrons.
I argue that leaders of secondary states refer to economic and political indicators
to determine the reliability of the superpower’s security commitments. When economic and political indicators suggest that the superpower’s security commitments
are incredible, secondary states adopt nuclear behavior in order to bargain over the
terms of the existing ‘security contract’ with the superpower. The superpowers ability to bargain successfully with the secondary state depends on the latters economic
dependence.
33
Appendix I
Table 1: States, Alignments, and Nuclear Behavior Years, 1945-2005
State
Alliance Relationship Years of Nuclearization
Algeria
Soviet Friend
1983-2000
Argentina
US Defense Ally
1968-1990
Australia
US Defense Ally
1956-1973
Belarus
Russian Friend
1991-95
Brazil
US Defense Ally
1953-1990
Canada
US Defense Ally
1946
Chile
US Defense Ally
1961-95
People’s Republic of China
Soviet Ally
1955-64
Egypt
Soviet Friend
1955-80
France
US Defense Ally
1946-Present
Germany (West)
US Defense Ally
1966-70
India
Neutral
1954-Present
Indonesia
Neutral
1961-70
Iran
Soviet Friend
1976-Present
Iraq
Neutral
1976-2003
Israel
US Friend
1955-Present
Italy
US Defense Ally
1956-70
Japan
US Defense Ally
1970-71
Korea, North
Soviet Ally
1965-Present
Korea, South
US Defense Ally
1970-1975, 1977
Libya
Soviet Friend
1970-2003
Nigeria
Neutral
1976-95
Norway
US Defense Ally
1951-65
Pakistan
US Friend
1972-Present
Romania
Soviet Defense Ally
1985-93
South Africa
US Friend
1969-93
Spain
US Defense Ally
1971-1974
Sweden
Neutral
1954-69
Switzerland
Neutral State
1946-1969
Taiwan
US Ally/Friend
1967-88
Ukraine
Russian Friend
1991-95
United Kingdom
US Defense Ally
1945-Present
Yugoslavia
Neutral
1954-1988
Table 1: Names of countries and time periods for beginning and end of nuclear weapons activities are drawn
from Mueller and Schmidt (2010, 157), Jo and Gartzke (2007), and Singh and Way (2004). Disagreement
exists over the precise timing of the onset and termination of nuclear behavior (Mueller and Schmidt 2010).
34
References
Adams, Karen Ruth. 2004. “Attack and Conquer? International Anarchy and the
Offense-Defense-Deterrence Balance.” International Security 28(3): 45–83.
Axelrod, Robert, and Robert O. Keohane. 1985. “Achieving Cooperation under
Anarchy: Strategies and Institutions.” World Politics 38(1): 226–254.
Baldwin, David A. 1985. Economic Statecraft. Princeton, NJ: Princeton University
Press.
Betts, Richard K. 2000. “Universal Deterrence or Conceptual Collapse? Liberal
Pessimism and Utopian Realism.” In The Coming Crisis: Nuclear Proliferation,
U.S. Interests, and World Order, ed. Victor A. Utgoff. Cambridge, MA: MIT
Press.
Brooks, Stephen G., and William C. Wolhforth. 2008. World Out of Balance:
International Relations and the Challenge of American Primacy. Princeton, NJ:
Princeton University Press.
Christensen, Thomas J. 1996. Useful Adversaries: Grand Strategy, Domestic
Mobilization, and Sino-American Conflict, 1947-1958. Princeton, NJ: Princeton
University Press.
Christensen, Thomas J., and Jack Snyder. 1990. “Chain Gangs and Passed Bucks:
Predicting Alliance Patterns in Multipolarity.” International Organization 44(2):
137–168.
Cooley, Alexander. 2008. Base Politics: Democratic Change and the U.S. Military
Overseas. Ithaca, NY: Cornell University Press.
Cooley, Alexander, and Hendrik Spruyt. 2009. Contracting States: Sovereign
Transfers in International Relations. Princeton, NJ: Princeton University Press.
Deudney, Daniel. 2011. “Unipolarity and Nuclear Weapons.” In International
Relations Theory and the Consequences of Unipolarity, ed. G. John Ikenberry,
Michael Mastanduno, and William C. Wolhforth. New York, NY and Cambridge,
UK: Cambridge University Press.
Drezner, Daniel W. 2007. All Politics is Global: Explaining International
Regulatory Regimes. Princeton, NJ: Princeton University Press.
Dunn, Lewis A. 1991. Containing Nuclear Proliferation. Adelphi Papers London,
UK: Brassey’s for the International Institute for Strategic Studies.
35
Elman, Colin, and Miriam Fendius Elman. 2002. “How Not to Be Lakatos
Intolgerant: Appraising Progress in IR Research.” International Studies
Quarterly 46(2): 231–262.
Fearon, James D. 1995. “Rationalist Explanations for War.” International
Organization 45(3): 379–414.
Friedberg, Aaron L. 2000. In the Shadow of the Garrison State: America’s
Anti-Statism and Its Cold War Strategy. Princeton, NJ: Princeton University
Press.
Fuhrmann, Matthew. 2009a. “Spreading Temptation: Proliferation and Peaceful
Nuclear Cooperation.” International Security 34(1): 7–41.
Fuhrmann, Matthew. 2009b. “Taking a Walk on the Supply Side: The
Determinants of Civilian Nuclear Cooperation.” Journal of Conflict Resolution
53(2): 181–208.
Fuhrmann, Matthew, and Sarah E. Kreps. 2010. “Targeting Nuclear Programs in
War and Peace: A Quantitative Empirical Analysis, 1941-2000.” Journal of
Conflict Resolution 54(6): 831–859.
Gaddis, John Lewis. 1982. Strategies of Containment: A Critical Appraisal of
Postwar American National Security Policy. Oxford, UK: Oxford University
Press.
Gartzke, Erik, and Michael W. Simon. 1996. “Political System Similarity and the
Choice of Allies: Do Democracies Flock Together or Do Opposites Attract?”
Journal of Conflict Resolution 40(4): 617–635.
Gavin, Francis J. 2004/2005. “Blasts from the Past: Proliferation Lessons from the
1960s.” International Security 29(3): 100–135.
Gavin, Francis J. 2010. “Same As It Ever Was: Nuclear Alarmism, Proliferation,
and the Cold War.” International Security 34(3): 7–37.
Goddard, Stacie E. 2006. “Uncommon Ground: Indivisible Territory and the
Politics of Legitimacy.” International Organization 60(1): 35–68.
Goldstein, Avery. 1995. “Discounting the Free Ride: Alliances and Security in the
Postwar World.” International Organization 49(1): 39–71.
Goldstein, Avery. 2000. Deterrence and Security in the 21st Century: China,
Britain, France, and the Enduring Legacy of the Nuclear Proliferation. Stanford,
CA: Stanford University Press.
36
Greenfeld, Liah. 1993. Nationalism: Five Roads to Modernity. Cambridge, MA:
Harvard University Press.
Horowitz, Michael. 2010. The Diffusion of Military Power: Causes and
Consequences for International Politics. Princeton, NJ: Princeton University
Press.
Huth, Paul K. 1988. “Extended Deterrence and the Outbreak of War.” American
Political Science Review 82(2): 423–443.
Hymans, Jacques E.C. 2006. The Psychology of Nuclear Proliferation: Identity,
Emotions, and Foreign Policy. New York, NY and London, UK: Cambridge
University Press.
Hymans, Jacques E.C. 2010. “The Study of Nuclear Proliferation and
Nonproliferation: Toward a New Consensus?” In Forecasting Nuclear
Proliferation in the 21st Century: The Role of Theory, ed. William C. Potter,
and Gaukhar Mukhatzhanova. Stanford University Press.
Jervis, Robert. 1989. The Meaning of Nuclear Revolution. Ithaca, NY: Cornell
University Press.
Jo, Dong-Joon, and Erik Gartzke. 2007. “Determinants of Nuclear Weapons
Proliferation.” Journal of Conflict Resolution 51(1): 167–194.
Koremenos, Barbara, Charles Lipson, and Duncan Snidal. 2001. “The Rational
Design of International Institutions.” International Organization 55(4): 761–799.
Kroenig, Matthew. 2009. “Importing the Bomb: Sensitive Nuclear Assistance and
Nuclear Proliferation.” Journal of Conflict Resolution 53(2): 161–180.
Kroenig, Matthew. 2010. Exporting the Bomb: Technology Transfer and the Spread
of Nuclear Weapons. Ithaca, NY: Cornell University Press.
Lake, David A. 2009. Hierarchy in International Relations. Ithaca, NY: Cornell
University Press.
Leeds, Brett Ashley. 2000. “Do Alliances Deter Aggression? The Influence of
Military Alliances on the Initiation of Militarized Interstate Disputes.” American
Journal of Political Science 47(3): 427–439.
Legro, Jeffrey W., and Andrew Moravcsik. 1999. “Is Anybody Still a Realist?”
International Security 24(2): 5–55.
37
Levite, Ariel E. 2002/2003. “Never Say Never Again: Nuclear Reversal Revisited.”
International Security 27(3): 59–88.
Lieber, Keir A. 2000. “Grasping the Technological Peace: The Offense-Defense
Balance and International Security.” International Security 25(1): 71–104.
MacDonald, Paul K., and Joseph M. Parent. 2005. “Grateful Decline? The
Surprising Success of Great Power Retrenchment.” International Security 35(4):
7–44.
Mansfield, Edward D., and Jack Snyder. 1995. “Democratization and the Danger of
War.” International Security 20(1): 5–38.
Martin, Lisa L. 1993. “Credibility, Costs, and Institutions: Cooperation on
Economic Sanctions.” World Politics 45(3): 406–432.
Mearsheimer, John J. 2001. The Tragedy of Great Power Politics. New York, NY:
W.W. Norton.
Morrow, James D. 1991. “Alliances and Asymmetry: An Alternative to the
Capability Aggregation Model of Alliance.” American Journal of Political
Science 35(4): 904–933.
Morrow, James D. 1993. “Arms Versus Allies: Trade-Offs in the Search for
Security.” International Organization 47(2): 207–233.
Mueller, Harald, and Andreas Schmidt. 2010. “The Little Known Story of
De-Proliferation: Why States Give Up Nuclear Weapons Activities.” In
Forecasting Nuclear Proliferation in the 21st Century: The Role of Theory, ed.
William C. Potter, and Gaukhar Mukhatzhanova. Stanford, CA: Stanford
University Press.
Mueller, John. 1988. “The Essential Irrelevance of Nuclear Weapons: Stability in
the Postwar World.” International Security 13(2): 55–79.
O’Neill, Barry. 2006. “Nuclear Weapons and National Prestige.”.
Pape, Robert A. 1997. “Why Economic Sanctions Do Not Work.” International
Security 22(2): 90–136.
Powell, Robert. 1987. “Crisis Bargaining, Escalation, and MAD.” American
Political Science Review 81(3): 717–736.
Pressman, Jeremy. 2008. Warring Friends: Alliance Restraint in International
Politics. Ithaca, NY: Cornell University Press.
38
Ratner, Ely. 2009. “Reaping What You Sow: Democratic Transitions and Foreign
Policy Realignment.” Journal of Conflict Resolution 53(3): 390–418.
Reiter, Dan. 2003. “Exploring the Bargaining Model of War.” Perspectives on
Politics 1(1): 27–43.
Rose, Gideon. 1998. “Neoclassical Realism and Theories of Foreign Policy.” World
Politics 51(1): 144–172.
Rosecrance, Richard N. 1966. Problems of Nuclear Proliferation: Technology and
Politics. Berkeley, CA and Los Angeles, CA: University of California Press.
Sagan, Scott D. 1995. The Limts of Safety: Organizations, Accidents, and Nuclear
Weapons. Princeton, NJ: Princeton University Press.
Sagan, Scott D. 2011. “The Causes of Nuclear Weapons Proliferation.” Annual
Review of Political Science 14: 225–244.
Sagan, Scott D., and Kenneth N. Waltz. 1995. The Spread of Nuclear Weapons.
New York, NY: W.W. Norton.
Schweller, Randall L., and David Priess. 1997. “A Tale of Two Realisms:
Expanding the Institutions Debate.” Mershon International Studies Review
41(1): 1–32.
Singh, Sonali, and Christopher R. Way. 2004. “The Correlates of Nuclear
Proliferation.” Journal of Conflict Resolution 48(6): 859–885.
Snyder, Glenn H. 1984. “The Security Dilemma in Alliance Politics.” World
Politics 36(4): 461–495.
Snyder, Jack L. 2000. From Voting to Violence: Democratization and Nationalist
Conflict. New York, NY: W.W. Norton.
Solingen, Etel. 2007. Nuclear Logics: Contrasting Paths in East Asia and the
Middle East. Princeton, NJ: Princeton University Press.
Taliaferro, Jeffrey W. 2000/01. “Security Seeking Under Anarchy: Defensive
Realism Revisited.” International Security 25(3): 128–161.
Taliaferro, Jeffrey W. 2004. “Power Politics and the Balance of Risk: Hypotheses on
Great Power Intervention in the Periphery.” Political Psychology 25(2): 177–211.
Voeten, Erik. 2004. “Resisting the Lonely Superpower: Responses of States in the
United Nations to U.S. Dominance.” The Journal of Politics 66(3): 729–754.
39
Walt, Stephen M. 1987. The Origins of Alliances. Ithaca, NY: Cornell University
Press.
40