REGIME SECURITY THEORY: WHY DO STATES WITH NO CLEAR STRATEGIC
SECURITY CONCERNS OBTAIN NUCLEAR WEAPONS?
by
MATTHEW BEASLEY
A THESIS
Presented to the Department ofPolitical Science
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Master of Science
September 2009
11
"Regime Security Theory: Why Do States With No Clear Strategic Security Concerns
Obtain Nuclear Weapons?" a thesis prepared by Matthew Beasley in partial fulfillment of
the requirements for the Master of Science degree in the Deparment ofPoltical Science.
This thesis has been approved and accepted by:
, Chair of the Examining Committee
Committee in Charge:
Accepted by:
Lars Skalnes, Chair
Jane Cramer
Mikhail Myagkov
Dean of the Graduate School
An Abstract of the Thesis of
Matthew C. Beasley for the degree of
in the Department ofPolitical Science to be taken
III
Master of Arts
June 2009
Title: REGIME SECURITY THEORY: WHY DO STATES WITH NO CLEAR
STRATEGIC SECURITY CONCERNS OBTAIN NUCLEAR WEAPONS?
Approved:
Dr. Lars Skalnes
Current realist explanations ofwhy states decide to develop nuclear weapons
cannot account for the behavior of states that lack a clear strategic threat. An alternative
explanation is necessary to explain such behavior. I argue that domestic regimes in states
with no clear strategic threat may develop nuclear weapons in order to ensure their
survival. Such regimes are internationally isolated, under pressure from major powers in
the international system and possess some preexisting nuclear capacity. Under these
conditions, increasing domestic instability causes regimes to pursue a course ofnuclear
development. Nuclear weapons allow a regime to change the preferences of the great
powers that would otherwise prefer to see the regime overthrown. Ifthe regime
possesses nuclear weapons, because of the costs and risks associated with those weapons,
the great power will favor maintenance of the status quo and may even prop up a regime
it intensely dislikes.
IV
CURRICULUM VITAE
NAME OF AUTHOR: Matthew Beasley
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon, Eugene Oregon
DEGREES AWARDED:
Master of Science, Political Science, September 2009, University of Oregon
Bachelor of Science, Political Science, June 2005, University of Oregon
AREAS OF SPECIAL INTEREST:
Intemational Relations
Game Theory
PROFESSIONAL EXPERIENCE:
Graduate Teaching Fellow, University of Oregon, September 2005 to June 2008
vACKNOWLEDGMENTS
I wish to thank Dr. Lars Skalnes for his continued help, feedback and patience
during in the preparation of this manuscript. I would also like to thank Dr. Jane Cramer
and Dr. Mikhail Myagkov for their time, input and assistance in completing this
manuscript. lowe a special thanks to my drafted draft editors, Melinda 0'Sullivan and
Madeline McKee. Their assistance was irreplaceable.
VI
TABLE OF CONTENTS
Chapter Page
I. INTRODUCTION...................... 1
The Question 1
The Existing Literature 3
Domestic Politics - Liberalism 4
Prestige Variables 5
Condition Variables 6
Regime Security Theory 7
Capacity 9
Perception ofIsolation 10
Domestic Instability 10
Competing Explanations 11
Security... 12
International Economics 13
Systemic Prestige Arguments 13
An Introduction to the Methodology 14
Case Choices and Omissions 14
An Overview of the Rest of the Work 17
II. THEORY AND METHODOLOGy........ 18
Regime Security Theory 18
Variables 22
Capacity 22
Isolation 25
Domestic Instability 27
Vll
Chapter
Alternative Theories
Security Variables .
Democracy, International Economics, Domestic and Bureaucratic
Politics .
Prestige .
Methodology .
Page
32
32
38
44
49
III. SOUTH AFRICA 51
A Brief Chronology ofForeign and Domestic Relations 52
A Litany ofDomestic Woes 52
The Deteriorating International Relations of South Africa .. 54
South Africa's Nuclear Policies: A Timeline for Nuclear Decisions 56
Analysis of the South African Case 65
Traditional Threat Arguments 66
The Strategic Nuclear Threat against South Mrica 66
The Overwhelming Conventional Threat against South Africa 69
Regime Security Theory 72
Evidence of South African Domestic Instability 73
A Perception of International Isolation 75
The United Nations, Namibia and Apartheid 76
A United Southern Mrica with a Little Soviet Help 79
A Deteriorating Relationship with the West 85
Checking the Conditions ofIsolation 89
Capacity Concerns 90
Checking the Timing ofNuclear Development 92
Additional Evidence ofRegime Security Theory 95
Discussing Alternative Explanations 96
Timing ofDismantlement 99
Conclusions 101
V1l1
Chapter Page
IV. NORTH KOREA 103
A Historical Outline 104
A Context ofDomestic Decay.............................................................. 105
North Korean International Relations 107
A Timeline for Nuclear Decisions: North Korea and the IAEA 109
Analysis of the North Korean Case 116
Traditional Threat Arguments 117
The Strategic Nuclear Threat against North Korea......................... 117
The Overwhelming Conventional Threat against North Korea 119
Regime Security Theory 120
Evidence ofNorth Korea Domestic Instability.......................... 121
A Perception ofInternational Isolation 124
The United States 125
South Korea 128
China 130
Russia 133
Checking the Conditions of Isolation 135
Checking the Timing ofNuclear Development 136
Discussing Alternative Explanations 139
Conclusions 141
V. CONCLUSION 143
BIBLIOGRAPHY 148
1CHAPTER I
INTRODUCTION
The Question
What causes states to acquire nuclear weapons? Despite the plethora of
theories this question has produced, the dominant explanation for nuclear acquisition
is the intuitive one - security. States seek a deterrent against a nuclear armed
opponent. According to Waltz, nuclear weapons for the U.S. mean nuclear weapons
for Russia. Further, "The Soviet Union having them was a spur for the Chinese
Communists getting them. China's having them was a spur for India, India was a spur
for Pakistan, and so it goes. Now that's not uniformly the pattern but it's a very
important part of [it]."l Great Britain and France fit into the chain ofbalancing in a
similar way: Great Britain and France acquired nuclear weapons in response to the
threat of the Soviet Union coupled with the lack of certainty about American security
guarantees in Europe? This "chain ofbalancing" hypothesis is by no measure
unchallenged. It provides a strong logic for the dominance of strategic security
1 Waltz 1981. There is some disparity in Chinese reasoning; Sagan cites balancing the US followed by
balancing Russia when he lays out the same string that is described by Waltz above. Sagan 1996, 58.
For the general "chain of balancing" logic see also, Shultz 1984.
2 Sagan 1996, 58.
2concerns in the nuclear decision making of states. It does not explain why states seek
to acquire nuclear weapons in situations where there is not a strategic threat. This
thesis seeks to answer the question of why some states with no clear strategic threat
nevertheless obtain nuclear weapons.
The two cases of successful nuclear development which cannot be adequately
explained by traditional threat hypotheses are North Korea and South Africa. Long
before the completion of Pyongyang's nuclear program, shifts in international
alignment and norms meant that the strategic threat against North Korea had
evaporated, and North Korean artillery had established a reliable deterrent against any
conventional threat. South Africa's non-aligned position during the Cold War, along
with its overwhelming advantages in military strength compared to its neighbors,
meant that it was never in real danger from nuclear weapons or conventional assault.
Even if the traditional security explanation sufficiently explained every other
case of nuclear proliferation, a compatible alternate causal theory is necessary to
explain proliferation for these two states in order for threat to be a satisfactory
explanation of proliferation as a whole. While it is unlikely that the causal process is
identical in these cases, it makes sense to compare these two non-standard states in
search of similarities. This has not previously been done. When comparing North
Korea and South Africa one striking similarity emerges. Scholars separately studying
these states cite the security interests of the domestic regime as distinct from those of
the state as a whole. 3
3 Moore 1987; and Reiss 1995a.
3Regime security as a variable appears underdeveloped in the existing literature.
To date, it is addressed most thoroughly by Solingen, who describes the difference
between regime security and state security as "the most important frontier in the study
of nuclear choices." Solingen argues that examining regime security is critical because
treating the state as unitary undermines a successful study of a state's nuclear cost
benefit equation. However, Solingen goes on to test multiple conceptual approaches
against a large number of cases rather than focusing on a regime security centered
1 . 4ana YSIS.
The Existing Literature
Existing nuclear proliferation literature arose in response to the reality of
nuclear weapons and the possibility of additional states joining the existing nuclear
club. The specific task of explaining why states with no clear strategic security
concern develop nuclear weapons evolved from this body of literature. The existing
nuclear proliferation literature is tackled from two angles: why do states get nuclear
weapons, and why do states not get nuclear weapons. This chapter will briefly review
theories and variables which attempt to answer both questions. The following chapter
will focus on the major explanatory theories as to why states do acquire nuclear
weapons. 5
4 Solingen 2007.
5 The two subgroups of non-proliferation explanative theories which do not have an equivalent
proliferation explanative theory are institution constraints and resource constraints, both of which have
4Specific variables which explain the nuclear choices of states will be
categorized under security, domestic politics, prestige, and condition variables. The
variable list borrows explicitly from James Walsh, but modifies the categorization.
First, because this paper deals with variables that affect a state's decisions to acquire
nuclear weapons rather than their decisions to forego them, prestige variables are
included. Second, the variables categorized as condition variables cannot be causes of
proliferation, as they can be with non-proliferation, but they critically affect
proliferation, and so they are listed and dealt with separately from potential
independent variables. Third, because of the attention given to threat theories above,
they will not be reviewed below. However, they will be addressed in detail in the
following chapter.
Domestic Politics - Liberalism
Two variables that affect the chances of proliferation are level of democracy
and level of integration into the international political economy. A high value for
either of these factors is theorized to undermine the chances that a state will acquire
important implications for proliferation theories as condition variables. For a thorough theoretical
division of non-proliferation variables see: Walsh 2001, 19-42. Walsh divides theories as to why states
do not achieve nuclear weapons into 15 "testable hypothesis" They are, summarized; HI: lack of
threat, H2: bipolarity, H3: security guarantees from great powers, H4: superpower pressure, H5: lack of
funds, H6: lack of human resources/knowledge, H7: denial by suppliers, H8: norms against nuclear
weapons, H9: democracy, H10: electoral politics, H11: liberalizing political economy, H12: restrictions
in organizational politics, H13: membership in anti-proliferation regimes, H14: the prisoners dilemma
of nuclear weapons can be solved by regimes, HIS: pressure ofintemationallaw.
5nuclear weapons.6 A third variable, the effect of electoral politics, is theorized as
potentially having either a positive or negative effect on the chances of proliferation.7
The final domestic politics theory falls into the institutionalist school, and
specifically comes from bureaucratic politics theory.8 This theory argues that the
particular interests of important subgroups within the domestic government of the state
affect nuclear decision making. For example, the atomic energy branch ofgovernment
may push for or against nuclear weapons based on whether they think it will garner
them more funding or undermine their authority in favor ofa military program. 9
Prestige Variables
Prestige arguments suggest that states acquire nuclear weapons in order to
enhance their own prestige. The two major subsets of prestige arguments are domestic
level prestige theories and systemic level prestige theories. Domestic prestige
arguments are similar to electoral politics arguments except that they are more general.
These arguments hold that nuclear decisions reflect the goal ofgarnering or
6 Walsh 2001,31-34. Also, for effects of democracy see Chafetz 1993, 133. For effects of economic
liberalization see Solingen 1994, 324-333.
7 Reiss 1988, 117-119,249-251; and Sagan 1996,63-73.
8 Walsh 2001,36-37; Allison 1971; Sagan 1996, 11; and Scheinman 1965.
9 Additional international integration variables of interest to non-proliferation scholars are membership
in international regimes and the effects of international law. However, these variables specifically
address why states may refrain from seeking nuclear weapons and thus have little to say in explaining
why they would seek nuclear weapons. Meyer 1984, 102; Reiss 1988,260-263; Sagan 1996, 80-82; and
Walsh 2001,39-41.
6maintaining domestic support, whether from voters or other powerful interest
groups. 10
International prestige arguments, or systemic prestige arguments, follow two
lines: they argue either that states seek nuclear weapons to increase their rank in power
and position, regionally or globally, or to get attention and force interaction from more
powerful states in the system. 11
Condition Variables
Conditional variables determine whether it is possible for a state to get nuclear
weapons. No value of the condition variable provides an impetus to seek nuclear
weapons, but certain values may be required to weaponize. It is necessary to look at
condition variables for two reasons. First, while these are treated as condition, or
scope, variables by regime security theory, they have been treated as independent
variables when explaining why states do not develop nuclear weapons. Second, these
variables do not always take a zero or one value, and may dissuade a state from
acquiring nuclear weapons by either increasing the cost or decreasing the benefit
without holding a value that is strictly prohibitive of development.
10 For examples of South Mrican domestic prestige arguments see Moore 1987, 132. Sagan classifies
the domestic symbolic power type ofprestige argument as a normative argument; nuclear weapons
produce an "important normative symbol of a states modernity and identity." Sagan 1996, 55.
1l For examples of South Mrican systemic prestige arguments see Moore 1987, 129-132. The fust line
of argument, and the debate as to whether this is differentiable from security logic, is discussed in
Sagan 1996,76.
7Condition variables include the technological and economic capacity to
produce nuclear weapons, as well as access to external resources. 12 In order to have
nuclear weapons capacity it must be true that a state can either produce those weapons
domestically or buy them internationally. This basic requirement means that either
both technological and economic capacity must be sufficient for the task ofbuilding a
bomb, or there must be international access to ready-made bombs available, or both.
Regime Security Theory
South Africa and North Korea both developed indigenous nuclear capacity
without a strategic nuclear threat. 13 Why did these two states develop nuclear
weapons? In the case literature for each individual state a specific observation emerges
as to the state's behavior which deserves comparative attention. That is, the domestic
regime of the state, but not the state itself, was under threat of internal destabilization
exacerbated by the action or inaction of outside powers. The hypothesis proposed in
this thesis is:
States that acquire nuclear weapons and do not face a strategic nuclear threat do
so in order to increase the security of the state's domestic regime. Nuclear
weapons deter intervention targeted at fomenting domestic unrest. Nuclear weapons
also coerce intervention intended to alleviate existent domestic unrest. Rather than
12 For technical and resource capacity see Meyer 1984, 68; and Walsh 200 1. For access to external
resources see Walsh 2001; and Snyder 1985, 183-212.
13 The timing of U. S. nuclear threats against North Korea precludes this explanation. See Chapter IV.
8deterring attack, these weapons make it irrational for other states to support the
overthrow of the regime.
This hypothesis assumes that actors are rational in the most basic sense: they
maximize their utility and value survival. 14 The key element of the regime security
hypothesis, as it differs from traditional security explanations, is that the actor
concerned is the domestic regime and not the state. I suspect that this explanation has
not been well developed within the existing security literature because it tends to treat
states as unitary actors, thus ignoring domestic politics. The regime security
hypothesis applies under the condition that the regime has the capacity to acquire
nuclear weapons. In an environment where a regime has the requisite capacity, the
independent variables which combine to cause a regime to acquire nuclear weapons
are international isolation and domestic instability.
The acquisition of nuclear weapons allows isolated regimes to survive
domestically unstable situations by changing the interests of the great powers in the
international system. If an isolated regime does not have nuclear weapons, it may be in
the interest of the great powers for it to fall; they may even aid in undermining such a
regime. If the same regime has nuclear weapons, the great powers' interest is to
maintain stability and the status quo. The great powers may even aid the isolated
regime because of the high potential cost of an unstable nuclear state. The following
14 This assumption is actually less stringent than the realist assumption that states are rational actors
because the unit of analysis is smaller. The regimes interests are not argued to be analogous to the states
interests.
9sections will provide further explanation of each of the variables of regime security
theory as laid out above.
Capacity
The capacity variable is introduced under resource condition variables above,
and encompasses domestic technological and economic capacity, as well as access to
international resources. In order to achieve the reality of a nuclear arsenal, a state must
be able to either produce weapons domestically or buy them internationally. The
weaknesses of capacity as an explanatory independent variable are dealt with in
Chapter II. In general, the conclusion is that capacity is not a good explanatory
variable because all states could technically come up with the capital to buy a nuclear
weapons program. 15
Despite its lack of strength as an independent variable, capacity is an important
condition variable. It is true that some states that have the capacity to build nuclear
weapons have not done so. Nevertheless, the technology for, or cost of, nuclear
weapons is difficult for smaller developing countries to come up with, especially over
a short period oftime.
15 Walsh 2001,85.
10
Perception of Isolation
The domestic regime must feel extremely isolated. A regime which perceives
itself as isolated will feel threatened by pressure for domestic reform from at least one
major pole and be viewed as an illegitimate regime by the world at large. The regime
must be concerned that one or more ofthe major powers is considering interference
with its domestic affairs. 16 This is a realistic fear when world opinion reduces the
chances of repercussions to actions by the larger power.
Domestic Instability
The domestic regime must fear current or future high levels of domestic
instability. In order to understand the causal chain that leads from isolation and
domestic instability to nuclear weapons it is necessary to layout the interests of the
actors. The regime's interest is to remain in power. It is the interest ofthe great
power(s) and other states which view the regime as illegitimate that the regime be
overthrown. Because domestic instability threatens to destabilize the regime, it worries
that the great power(s) may decide to exacerbate the instability in order to overthrow
the regime. By acquiring nuclear weapons it is possible for a regime to change the
preferences of the great power(s) and other relevant states.
16 While the process of acquiring nuclear weapons temporarily increases the chance of international
interference, the successful acquisition of nuclear weapons should dissuade great powers entirely from
the same interference.
11
Once a state has nuclear weapons it is no longer in the interest of the great
power(s) for the regime to be destabilized. Because survival is the priority of the
regime, a regime in a stable environment will not use nuclear weapons offensively, as
that would ensure retaliation and the end of its existence. Instability in the country
greatly increases the chance that nuclear weapons will be used in two ways. First, in
the chaos of regime change, unstable elements or small groups could get control of the
weapons. Second, the regime itself may launch them out of desperation if it feels it has
nothing left to lose.
Once a regime has nuclear weapons it is in the interest of the great power(s) to
maintain the status quo in order to prevent an unstable environment in which there is
an increased likelihood of nuclear weapons being used on either the domestic
population or a neighbor. Thus, growing isolation and instability causes regimes to
acquire nuclear weapons in order to change the incentives ofother powers in the
system.
Competing Explanations
The theories in the existing literature that are strongly critiqued or informed by
a study of regime security are those based on national security, international trade and
international prestige. However, as my work suggests that regime security is sufficient
to explain the nuclear acquisition of both South Africa and North Korea, that
12
conclusion suggests that competing theories not directly engaged are unnecessary to
explain those cases. They may contain variables that matter to a degree or in some
cases; however, those variables are not determinate in the cases studied.
Security
The two variables which realists have argued cause states to acquire nuclear
weapons are nuclear armed adversaries and overwhelming conventional security
threats. 17 The cases chosen in this study cannot be well explained by the argument that
states develop nuclear weapons in response to a strategic threat. By a detailed review
of the circumstances of each case in the following chapters, it will become clear that
strategic nuclear security cannot explain proliferation in these cases because such a
threat did not exist during the time period that would be necessary to explain
development. Additionally, the case studies will reveal that there is no overwhelming
conventional security threat which can explain nuclear weapons acquisition in either
case.
The study does not imply that strategic nuclear and conventional security
theory cannot explain other cases; only that they do not apply to all cases of nuclear
weapons acquisition. This is still an important implication because threat hypotheses
have been previously generalized to explain all cases of proliferation. The implications
of this study do provide evidence for a critique of structural realism and its assumption
17 Walsh 2001,21-24; Frankel 1993, 46; and Mearsheimer 1998.
13
that it is unnecessary to look inside the black box. Factors other than the position of
units within the system are necessary to explain the South African and North Korean
cases.
International Economics
The isolation variable suggests that the argument that the degree of
international economic integration has an impact on decisions carries weight. The
autarkic nature of the Kim regime minimizes the negative international trade
externalities of nuclear development. In the South African case, nuclear development
did not occur until after a drastic drop in international integration. This outcome is also
consistent with the international economic integration argument. 18
Systemic Prestige Arguments
While rejecting the idea that states acquire nuclear weapons for prestige as an
end in itself, my findings support the argument that states acquire nuclear weapons to
affect the type of attention and treatment they receive from other powers in the system.
However, this is a security argument. The unit of analysis is the regime and not the
state, but the interests of the regime are security and survival. Analysis of the cases
suggests that international prestige arguments correctly identify the efforts of non-
18 Walsh 2001,31-34; and Solingen 1994,324-333.
14
standard proliferators, but misunderstand the logic behind those decisions. The logic is
survival, further impetus is unnecessary.
An Introduction to the Methodology
As noted, South Africa and North Korea are the two cases of nuclear
acquisition which are cannot be well explained by the strategic nuclear security threat,
or chain of balancing, logic. This thesis employs comparative case study methodology
in order to engage in theory generation based on these two cases. Because of the
failure of existing theories to adequately explain the cases concerned, case comparison
is used in order to build and propose a theory which provides an alternate causal
hypothesis in order to explain the behavior of these two states.
Case Choices and Omissions
South Africa and North Korea were chosen as cases because regime security
theory attempts to explain cases where the traditional security explanation fails, cases
which must have a separate explanation. The theory does not argue against the chain
of balancing logic put forward by realists, as that theory does appear to explain the rise
of most nuclear armed powers. South Africa and North Korea are the cases for which
the strongest arguments can be made against the traditional security logic, and thus
they are cases which still require explanation.
15
It is certainly possible to argue that France and Great Britain do not perfectly
fit the chain of balancing logic; however, it is very possible that these states developed
nuclear capacity as a response to weakening American security guarantees in the
continued face of the Soviet threat. 19 It is true that Argentina and Brazil have been
argued to possess alternative motivations for nuclear acquisition; nevertheless, their
behavior falls directly into the predictions of threat based theories. They entered the
nuclear age as dyadic competitors. It is not clear that Israel developed nuclear
weapons in response to overwhelming conventional security threat or in response to
nuclear ambitions from Pakistan, but nor is it clear that the security explanation falls
through.20
In the cases of South Africa and North Korea, conventional security logic fails.
After analyzing the evidence of the individual cases we can say with a high degree of
confidence that the traditional security explanations do not account for the
development of nuclear weapons by either South Africa or North Korea. It is because
of the confidence that can be built in the failure of the traditional threat explanations
that these cases are interesting. These cases indicate that there must be another path
states can take to obtain nuclear weapons. The cases were chosen in order to analyze
similarities between these two non-conformers in an attempt to ascertain the cause or
causes of their nuclear weapons capacity.
19 Sagan 1996, 58.
20 Maoz 2003,46.
16
The weakness of this case selection is that South Africa and North Korea are
weak cases for the traditional security argument and strong cases for the regime
security argument. Thus, discounting a strategic nuclear security explanation does not
strongly undermine that argument and supporting a regime security explanation does
not provide high confidence in the theory's applicability to other cases explained by
the strategic nuclear security argument. Regime security theory does not deal with the
same body of cases as the traditional security argument, rather with those cases that a
strategic nuclear security hypothesis cannot sufficiently explain. As such regime
security theory does not attempt to undermine the traditional security explanation of
nuclear proliferation as it applies to states which faced a continuous nuclear threat
during their period of nuclear weapons development. The opposite is true; this study
begins by accepting as accurate the strategic nuclear threat theory for most cases.
By the same logic, South Africa and North Korea are neither easy nor hard
cases for regime security theory. They are, rather, two cases which did not fit well into
previous models and, as such, were studied in order to generate new ideas. Because
South Africa and North Korea were chosen in order to produce a theory, they are not
legitimate means of testing regime security theory. This thesis does do significant
theory testing, as it applies existing general theories of proliferation to the cases
studied. With regard to the regime security hypothesis presented, this thesis is
explicitly an exercise in theory building and not theory testing. Further testing of
regime security theory would require its application to additional cases.
17
An Overview ofthe Rest of the Work
This work is laid out in three major sections. Chapter II expands on the
discussion of the existing literature on proliferation, provides additional discussion of
the methodology, and lays out regime security theory in greater detail. Chapters III
and IV provide details of cases and analysis. Chapter III focuses on South Africa.
Chapter IV focuses on North Korea. Chapter V provides conclusions and observations,
as well as suggestions for future research and cases to consider.
18
CHAPTER II
THEORY AND METHODOLOGY
Regime Security Theory
Regime security theory attempts to explain why states acquire nuclear weapons
while not facing a strategic nuclear threat. Regime security theory posits an answer to
the question ofwhy decision makers in states not facing a strategic nuclear threat feel
it is in their interest to develop nuclear weapons. The answer put forward is that
decision makers do not consider nuclear weapons to be in the interest ofthe state, but
rather, think that nuclear weapons increase the survivability of the state's domestic
regime.
The logic of regime security theory is that a domestic regime, under threat of
internal destabilization exacerbated by the action or inaction of outside powers, may
use nuclear weapons not to deter attack, but to change the preferences of the great
powers such that it is in the interest of those powers that the regime survives. This
change in preferences is made possible by the risks associated with political instability
and government overthrow in a country that possesses nuclear weapons.
Nuclear weapons add two major risks for a great power if it supports instability
or overthrow of a regime. First, nuclear weapons introduce the loose nukes issue. Non-
19
state actors may take advantage ofthe power vacuum to secure the state's nuclear
arsenal for their own purposes. Stable governments closely and jealously guard their
nuclear arsenal. As stability decreases, and government resources become more and
more strained, the safeguards against theft degrade. Second, as a domestic regime
approaches overthrow there is a chance that the regime will adopt a ''use it or lose it"
mentality. This might mean using its arsenal for nuclear blackmail by threatening to
fire against neighbors unless aid is forthcoming, or launching against rebellious
population centers within the state in an attempt to crush domestic resistance.21
Conversely, because survival is the priority ofthe regime, a regime in a stable
environment will not use nuclear weapons offensively, as that would ensure retaliation
and the end of its existence.
Regime security theory assumes that decision makers within the regimes of
potential nuclear states understand the effect nuclear weapons have on the priorities of
the great powers, and, with that understanding, seek to ensure regime survival by
acquiring a nuclear arsenal. Because of risks introduced above, once a state has
nuclear weapons, it is no longer in the interest of the great power(s) for the regime
within that state to be destabilized. A regime with conventional arms must cope with
domestic instability and the strong possibility of outside interference. A regime with
nuclear arms must deal with domestic instability, but outside interference is no longer
21 There are historical examples of dictators using weapons of mass destruction against their own
population. For example, Saddam Hussein is accused of using chemical weapons against the Kurds
within Iraq. Pelletiere 2003. While there has not yet been a case of a state launching a nuclear weapon
at its population, the logic remains the same - it is possible that dictators may use weapons of mass
destruction against their own population to eliminate domestic dissent in times of crisis.
20
likely. Instead, other powers will condemn the regime verbally while simultaneously
propping them up financially.
The above logic leads to the following hypothesis: States that acquire
nuclear weapons and do not face a strategic nuclear threat do so in order to
increase the security of the state's domestic regime.
The key difference between regime security theory and traditional security
explanations is that the actor concerned is the domestic regime and not the state. I
suspect that a regime security explanation has not been well developed within the
security literature thus far because realist theory typically treats states as unitary
actors, thus ignoring domestic politics, while liberal and institutionalist theories, which
deal with conditions inside the state, tend not to identify international security threats
as explanatory variables. 22 Regime security theory makes the assumption that actors
are rational in the most basic sense: they maximize their utility and value survival.
Regime security theory does not assume the state is a unitary actor. It does, however,
assume that the regime is a unitary actor, acting inside and outside the state to
maximize its own utility.
Regime security theory, at its heart, is an attempt to explain why regimes
develop nuclear weapons in the absence of a threat to state security. The theory
identifies what variables must be present in order for this to occur. The theory must
22 Krasner 1992.
21
assume that states which have the requisite condition and independent variables will
successfully deduce the rationale by which nuclear weapons increase the chances of
regime survival. The argument that regimes, under certain conditions, should seek
nuclear weapons in order to maximize their utility can be made deductively. The
observation that regimes, under such conditions, understand and act upon this cannot.
Solid support of the logic of regime security theory would require direct evidence of
the internal logic of these isolated regimes, evidence which is very difficult to
substantiate. So, the best evidence available to support the theory is that states behave
in the manner the theory predicts they should if they operate under the logic of regime
security theory. Such evidence should not only consist of the development of weapons
and the presence of the requisite conditions, but also the timing of these events. This
means that evidence of regime security theory will almost always be indirect. This
tends to be true of any theory which attempts to explain the internal rationale of
secretive regimes. 23
23 There is some direct evidence of members of the Apartheid government using regime security theory
logic; however, such internal evidence is still highly suspect to bias and revisionism due to the nature of
the regime. See Chapter III.
22
Variables
Regime security theory argues that, in order for it to be in the interest ofa
regime to seek nuclear weapons absent a strategic nuclear threat, several variables
must exist. The regime must possess the capacity to acquire nuclear weapons. The
regime must perceive itselfto be isolated from the international community. The
regime must fear for its survival from within, either through domestic instability or a
significant perceived future increase in domestic instability. The following sections
will explain each of these variables in detail to expand on the concepts laid out in
Chapter I.
Capacity
The capacity variable, introduced in Chapter I as a resource condition variable,
encompasses the domestic technological and economic capacity of a state, as well as
its access to international resources. Capacity has been disproved as an independent
variable. The capacity argument for nuclear acquisition is that nuclear technology is
inevitable, the benefits outweigh the costs, and as states acquire the capacity to
produce nuclear weapons, they will do so. This argument, which comes from
technological inevitablism, was popular in the 1950s and 1960s when few states were
sufficiently advanced to produce nuclear arsenals, and the ones that had the
technology had done so. The arsenals of Great Britain and France provided some
23
support for the argument, as well as the growing nuclear aspirations of other
industrialized countries. However, in the 1970s, as states realized that the original
projections of the economic benefits of nuclear programs had been drastically
overestimated, and states which reached the point of technological capacity chose not
to pursue the bomb because of economic and/or political costs, the variable lost most
of its weight. Today, the simplest, most damning condemnation of capacity as an
explanatory variable for proliferation is the number of western European countries that
long ago exceeded the capacity to produce nuclear weapons and have chosen to refrain
from the bomb.24
While it is an unconvincing independent variable, capacity is an important
condition variable. For many countries, especially small, poor, less developed
countries, the cost of the technology and resources required for nuclear weapons is
prohibitive, especially over a short period of time. For many developing countries the
price tag on a nuclear weapons program means sacrificing the lion's share of the state
budget for a decade or more. Therefore, small states which desire an indigenous
nuclear capacity must be willing to wait ten years for it to materialize unless they can
purchase weapons ready-made internationally.25 Technological capacity operates
similarly. While it is possible for states to develop domestic technologies, it is
24 Walsh 2001,85; and Davis 1993, 106.
25 It is unlikely that an extremely isolated state will be able to purchase nuclear weapons from another
state. This chance is dependent on the degree of regime isolation, the breadth of their isolation, the
number of nuclear actors in the international community at the time and the number of those actors
which have ratified the non-proliferation treaty to the point that they have safeguards in place to prevent
proliferation.
24
extremely costly and time consuming to do SO.26 These factors suggest that while
capacity does not ultimately restrict a state from achieving nuclear capability, certain
values of the capacity variable define the timetable for development.
It is likely that all states, regardless of how poor, can come up with the money
to produce nuclear weapons. However, low domestic capacity can be prohibitive of
nuclear development for two reasons. First, unstable states may worry that putting so
much of the economy into weapons development will undermine their ability to
control an already deteriorating domestic situation for an additional decade. Second,
regimes in states with low economic capacity are particularly susceptible to the
concern that over the long course of development their activities will attract unwanted
attention, spurring the interference they are trying to prevent. Therefore, a state with a
preexisting infrastructure advantage or a stronger economy is able to pursue nuclear
weapons, whereas a regime starting from scratch, with a poor economy, is unlikely to
do the same.
Operationalizing the capacity variable is deceptively complicated. As
described above, categorizing capacity as a 1 or a 0 would be too simplistic. Also,
such a measurement would be useless because every state would score a 1. Instead,
capacity should be measured in two parts in an attempt to determine if it is feasible for
a state to develop nuclear weapons.
First, can the state easily develop nuclear weapons? Universal measurements,
such as percent ofGDP, are not very useful because the cost of nuclear weapons is
26 While the technology to build a simple gun-type device is not particularly difficult, the technological
demands of producing sufficient enriched uranium 235 domestically are much higher.
25
dependent upon the level of technology in the state and in the international system.
Instead, this question must be answered on a case by case basis, contingent upon the
time period and development level of the country. If the answer to this question is yes,
the state possesses the capacity to develop nuclear weapons. Often the answer is no, it
is not easy for states to develop nuclear technology.
Once it has been established that a state cannot easily produce nuclear
weapons, the next question is: does a state have a preexisting nuclear infrastructure or
resource base? If the answer is no, then it is not feasible for the state to develop
nuclear weapons. If the answer is yes then it is feasible for the state to develop nuclear
weapons. While starting from scratch requires an extended period ofvulnerability,
hardship and scrutiny, a state which possesses significant pre-existing nuclear
infrastructure or nuclear natural resources is more capable of developing nuclear
weapons because these variables decrease the costs and timeline for weapons
development.
Isolation
Once the regime of a state possesses the capacity to embark on a path to the
bomb, it is necessary to explain the timing of nuclear development. The independent
variables which explain the timing of weapons development are domestic instability
and a perception of isolation in a capable regime. Defined below are statements which
should be true if the regime perceives itself to be isolated internationally.
26
• The regime is under pressure for domestic reform from at least one
major pole.
• The regime perceives that one or more major powers are considering
interference with its domestic affairs.
• General world opinion views the regime as illegitimate, corroborated
by specific instances of international condemnation, for example, near-
unanimous votes critical of the regime in international organizations.
• The regime does not have reliable support from any pole.
If all the statements above are true, the regime is both significantly isolated and under
threat of international interference enabled by that isolation. There have been few
regimes which met all ofthese criteria. This was especially true during the Cold War.
As the U.S. and the U.S.S.R. jockeyed for influence they were willing to extend their
patronage to many unpalatable governments. The number of regimes which potentially
have met the criteria increased following the Cold War, but not drastically. Most
regimes continued to court and be courted by patron states eager to extend their
influence, and often their control of raw material sources. It is rare to find a state and
regime so undesirable in resources and politics that no powerful state feels it is in their
. h' d 27mterest to act as t elf a vocate.
27 Solingen posits an alternative path to isolation, suggesting it comes from nationalistic economic
policies within the regime rather than political pressure from outside. While such a path would provide
the same reduced costs of nuclear development, because a purposefully autocratic regime would suffer
less from sanctions, it would not necessarily be identified with diplomatic isolation. Solingen 2007.
27
Domestic Instability
Domestic instability can take the form of either growing domestic instability or
predicted future high levels of domestic instability. In order to understand why
domestic instability causes states to acquire the bomb it is necessary to layout the
interests and preferences of the concerned actors - the state's domestic regime and the
great power(s) the regime believes threatens its survival. The interest of the state's
domestic regime is survival. The preferences of such regimes are universal in that their
preference ordering for policies will always maximize their chances of survival. The
interests ofthe great power(s) may be defined by the stability ofthe international
system, influence in the international system, domestic political gain, normative views
of leaders or a number of other things.28 Regardless, the interests of the great power(s)
(and often the rest of the international system), produce a preference ordering which
runs counter to the existence of the domestic regime in question. Therefore, the
preference ofthe great power(s) towards the regime is overthrow, regardless of the
interest which spawned it.
Increased domestic instability, or perceived increases in future domestic
instability, puts the regime in a precarious position. The regime fears, often accurately,
that the great power(s) may decide to take advantage of the instability and support the
overthrow ofthe regime, either by purposely exacerbating the issues causing the
28 I suspect great power interests are most often defined pragmatically, either by systemic concerns or
domestic popularity concerns. However, why the great powers care is an argument in itself, and no part
of regime security theory is contingent on what led them to believe that it is in their interest to oppose
the regime.
28
instability (an example could be trade sanctions), or by funding and/or supporting
domestic resistance within the state. The regime's policy preference shifts to acquiring
the bomb as a result of the regime's fear of outside interference aimed at its
overthrow. This policy is meant to ensure its core interest, regime survival. A
functioning nuclear arsenal serves to ensure regime survival by changing the
preferences of the great power(s) and other relevant states.
The downside of a policy in which a regime develops nuclear weapons to
ensure its security is that an incomplete nuclear program increases the incentives for
the great powers to remove the regime without providing additional security. The
policy remains rational for two reasons. First, as heavily evidenced in both cases
studied, regimes spend a large amount of resources trying to obscure their nuclear
weapons program as long as possible, thus minimizing the window of increased threat
from the great powers. Second, if a regime believes that it is unlikely that it will be
able to hide its nuclear weapons program until fruition, it is often still rational to take
the risk that the cost of increased short term pressure is worth the payoff of long term
regime survival.
Domestic instability is a difficult variable to quantifY. Some states obviously
meet the criteria. Extended periods of open rebellion within the country, as in South
Africa, and millions of starvation deaths, as in North Korea, clearly substantiate
domestic instability, but there are many unclear cases. Do LR.A. activities mean that
29
Great Britain was unstable? There is no easy way to operationalize domestic
instability as 1-0 without overlooking marginal cases.
A state is potentially unstable ifit possesses: a domestic resistance group
powerful enough to have attacked the central government, a failed or failing state
infrastructure (to the point that it cannot support the existing population), or an
attempted or recently successful military coup. There are a number of variables that
affect the probability that a potentially unstable state is or will become so unstable that
the domestic regime is in real danger of collapse. In their 2003 article, "Ethnicity,
Insurgency, and Civil War", Fearon and Laitin describe and test a list of variables that
are theorized to increase the chances of civil war in a state. Many of the variables they
described which correlate well with civil war can be used to further delineate grey area
cases between unstable and not. In states in which one of the above events occurred,
the value of these variables provides additional context for measuring domestic
instability on a case by case basis.29
• Low GDP per capita increases the chances of significant instability in
two ways. First, GDP per capita acts as a proxy for relative strength or
weakness of insurgents compared to the state. Insurgents will have less
strength relative to the state the wealthier the state. Second, recruitment
of young males into guerrilla groups is easier when the economic
alternatives are worse. In states so poor that recruits have a higher
29 Fearon 2003.
30 Ibid., 80.
31 Ibid., 81.
32 Ibid., 81.
30
standard of living than they would otherwise guerilla groups are at a
significant advantage in acquiring support.30
• High disparity in income can produce the second effect of low GDP per
capita even in states with high GDP per capita that have an
accompanying race or class based income disparity. For example, the
Apartheid regime was not poor, however blacks in South Africa had
disproportionately low GDP per capita, and so the second affect of
GDP per capita stilled existed because, within the possible guerrilla
recruitment demographic, GDP per capita was very low.
• A large rural population makes centralization of authority difficult for
the government, as well as increases the low income recruitment pool.3!
• Foreign aid for insurgents significantly increases the chances of their
success. While foreign support of domestic dissent is discussed under
the isolation variable, it is worthwhile noting that direct economic
support of domestic dissension has a compounding effect, increasing
the level of threat produced by an insurgent group dramatically.32
• A number of geographic factors can exacerbate domestic instability.
Difficult terrain, (generally mountains or jungles) providing safe haven
for guerrillas increases the effectiveness of guerrilla groups. Other
types of terrain that make it easier for guerrillas to maintain control
31
over an area or remain hidden from authorities include large, sparsely
populated areas and a physical division between the center of resistance
and the country's center, such as a body of water or another state.33
To summarize, a state can be coded as unstable if it has powerful domestic
resistance groups, a failed state, measured by starvation, or a recent coup. However,
the level of the exacerbating variables helps determine the severity of the instability on
a case by case basis. States which have one of the key events but low values for the
associated variables are strictly unstable but may have a very low degree of domestic
instability. This helps answer the question posed earlier: does the existence and
activities of the IRA mean that Great Britain was unstable? Technically, but the high
GDP per capita of Great Britain at the time greatly decreases the likelihood of
significant instability for multiple reasons and the low values of the exacerbating
variables in general correctly suggest that a strong British government able to contain
the problem and eventually negotiate a peace.
A prediction which follows naturally from regime security theory is that only
illegitimate governments go nuclear without a strategic nuclear security threat.
Illegitimacy is not laid out as a central hypothesis because, especially at the margins, it
is a subjective label, and because the causal process which directly explains nuclear
development is based on instability and isolation in a capable state. This makes regime
illegitimacy at best an ultimate cause, not a proximate one. So, while it is almost
33 Ibid., 81.
32
certainly the case that illegitimate regimes are more likely than others to push for
nuclear weapons outside of the requirements of strategic nuclear security illegitimacy
as a variable is of little use to regime security theory.
Alternative Theories
This section will explain and critique the major existing theories of nuclear
proliferation as defined in Chapter 1. It will attempt to provide support for the
argument that none of the current lines of reasoning satisfactorily explain existing
cases of non~standardproliferation. The following alternative theories of nuclear
proliferation will be addressed: strategic nuclear threat, overwhelming conventional
threat, democracy, international economics, domestic and bureaucratic politics,
international prestige and domestic prestige.
Security Variables
There are two main security explanations. First, arguably the most cited,
supported and intuitive explanation of nuclear acquisition, is strategic nuclear threat.
Second is acquisition as a response to an overwhelming conventional threat. This
section will discuss strategic nuclear threat and then briefly discuss overwhelming
conventional threat and additional relevant security variables.
33
The strategic nuclear threat explanation is that states acquire nuclear arsenals
as a response to an emerging or present strategic nuclear threat from another state. As
Mearsheimer puts it, "no state is likely to attack the homeland or vital interests of a
nuclear-armed state for fear that such a move might trigger a horrific nuclear response.
Not surprisingly, therefore, states are often tempted to acquire nuclear weapons to
enhance their security.,,34 The logic of strategic nuclear threat predicts two processes
by which nuclear proliferation occurs. First, a state may go nuclear in response to a
strategic rival obtaining nuclear capacity. This, in turn, causes the state's other
strategic rivals to go nuclear. This process provides an explanation for most of the
current nuclear powers. The United States got the bomb, which forced the Soviet
Union to get the bomb, which forced Britain, France and China to get the bomb.
Chinese nuclear capability forced India to get the bomb, which forced Pakistan to get
the bomb, which forced Israel had to get the bomb. The second process suggested by
the logic of strategic nuclear threat is the development of dyadic nuclear powers,
Argentina and Brazil for example. 35
Strategic nuclear threat has significant empirical support. Many ofthe cases
above appear clear cut. However, there are objections to the strength of the argument
34 Mearsheimer 1998.
35 Walsh 2001, 21-24; Frankel 1993, 46; and Mearsheimer 1998. Mearsheimer specifically deals with
the implications of realism to the logic of nuclear proliferation in the above New York Times article.
However, those implications are built out of ideas more thoroughly discussed in Mearsheimer 2001.
The idea that states balance strategic threat as opposed to possessing a tendency to balance strategic
threat identifies this interpretation of strategic nuclear threat as coming out of offensive or neo-realism,
rather than defensive realism as explained by Waltz 1979. The opposite end of the this spectrum, as
argued by Paul, is that states prudently calculate the level of threat in their region and will develop
nuclear weapons in response to threat only if the benefits to security outweigh the costs of seeming
more threatening to neighbors. Paul 2000.
34
in a number of the cases it attempts to explain. It has been argued that Great Britain
and France did not need to obtain nuclear weapons as a response to the Soviet threat
because they were protected by the nuclear umbrella of the United States. This
criticism is not particularly damning because during the period of British and French
acquisition, early in the nuclear age, those states doubted the effectiveness of the
United States nuclear umbrella in deterring an attack outside of U.S. territory. The
concern of the British and French was that the Soviets would not be convinced that the
U.S. would escalate all the way to a nuclear exchange in response to a Soviet strike on
Western Europe. The strongest evidence of this uncertainty, stronger than any policy
statement, was the existence of an American tripwire in West Germany. The tripwire
was intended to deter Soviet aggression by signaling that an attack on Western Europe
meant a direct attack on the United States military. This signal would not have been
needed had western governments been convinced that the Soviet's believed that an
attack on Europe would precipitate a full scale American response.
Another hard case for strategic nuclear threat is Israel. If Israel developed
nuclear weapons out of strategic nuclear concerns, it needs to have either been in
response to Pakistan's development or to the progressing development ofIran or Iraq.
While Pakistan has historically been hostile towards Israel, the claim that Israel
developed nuclear weapons as a direct response to Pakistani nuclear ambitions is not
particularly strong. And Iran and Iraq have not, to date, successfully tested any nuclear
weapons, though that may have been affected by Israel's proactive air attack on the
35
Osirak reactor.36 Perhaps the stronger strategic nuclear threat argument concerning
Israel is that it developed nuclear weapons in an intra-regional competition with Iran
and Iraq, only it was successful where its competitors were not. However, the size of
Israel and its history of regional conflict may provide a stronger case for the bomb as a
means of deterring overwhelming conventional force.
The two most difficult cases for strategic nuclear threat are South Africa and
North Korea. South Africa cannot realistically be argued to have developed weapons
in response to a strategic nuclear threat, and, while North Korea did feel pressure from
the U.S., the timeline of plausible U.S. nuclear threats against North Korea does not
come close to matching the timeline for North Korean nuclear developments.
Strategic nuclear threat correctly sits atop the pile of explanations for
proliferation. The theory has a convincing causal mechanism, significant empirical
support in a number of cases, and is theoretically unambiguous and clear cut. The
weakness of strategic nuclear threat as an explanation for proliferation is that it does
not provide a good explanation for all cases. This noticeable weakness provoked the
line of inquiry that lead to alternative security explanations - what causes states to get
the bomb in cases which cannot be explained by strategic nuclear threat?
The second major security explanation, which evolves out of the preceding
question, is that states may develop nuclear weapons in response to an overwhelming
conventional military threat. The logic of the argument is that, if a state has a nuclear
36 Raas 2007.
36
deterrent that it can use in response to attack on its territory, it will be safe from its
rivals regardless of any disparity in conventional forces. There are two reasons why
governments are argued to pursue nuclear weapons along these lines. The first reason
is that when a state is vastly outclassed by its rivals and has significant cause to fear
attack it will resort to nuclear weapons development. The second is that a state may
develop nuclear weapons as an alternative to spending the money on a comparable
conventional force because it is a more cost effective way to achieve a solid deterrent
against attack. This second reason has fallen out of favor since it became clear that
early estimates of the cost of a domestic nuclear program, either peaceful or military,
had been drastically underestimated. The overwhelming conventional threat
explanation does not attempt to explain all cases of nuclear proliferation. Rather, it is
most often addressed within individual cases which are poorly explained by strategic
nuclear threat??
On their own, the strategic nuclear threat hypothesis and the overwhelming
conventional threat hypothesis both over predict the number of instances in which
proliferation will occur. The factors considered by both security hypotheses that
mitigate over prediction are security guarantees from great powers, bipolarity, and
superpower pressure on potential proliferators?8 Great powers dissuade proliferators
by acting as patrons, by providing security guarantees to states under their influence.
37 Overwhelming conventional threat will be addressed specifically in the case study chapters.
38 These variables stem from the structural realist school of thought. They generally arise as
explanations as to why early predictions of rapid proliferation fell short. Walsh 2001, 21-24; Frankel
1993,46; and Mearsheimer 1998.
37
The nuclear arsenal of the power acts as a proxy arsenal for the smaller state once the
great power has provided that state with a guarantee.
Bipolarity increases the number of states affected by great power patronage. In
bipolarity, as the two poles compete for influence, states tend to fall under the
influence of one pole and thus the security umbrella of that pole. In addition, because
of the competition for influence, states concerned about security from an unaligned
neighbor can successfully appeal to either pole for patronage. The problem with
attributing the discrepancy between predicted and actual levels of proliferation to
bipolarity is that doing so also means that the declining influence of the superpowers
following the cold war should have encouraged proliferation. For the most part, this
did not occur. 39 That the decline of bi-polarity failed to result in rapid proliferation is a
problem which is indicative of a weakness in the general realist perspective on
proliferation. Realist theories have consistently over predicted proliferation and then
fallen back to new caveats when the predictions did not pan out.
Superpower pressure on potential proliferators can also dissuade states from
going nuclear through a number of mechanisms including sanctions, refusal of aid,
threats of alienation, or the removal of security guarantees. This is especially effective
when that pressure comes from multiple poles.
As a whole, the tendency for realist scholars to justify over-prediction of
proliferation with paste tense caveats is problematic. However, while these conditions
that supposedly dissuade states from proliferation are not entirely convincing, they do
39 Davis 1993, 81.
38
not significantly impair the broader security reasoning. It seems clear that many states
seek a nuclear arsenal to enhance their security against potential rivals. With
clarification and expansion, the security centric explanations of proliferation provide
the most promising building blocks for a general logic of proliferation.
Democracy, International Economics, Domestic and Bureaucratic Politics
This category includes the major variables addressed by the liberal and
institutional schools of thought. The three variables rooted in liberalism that affect the
propensity of a state to acquire nuclear weapons are its level of democracy, level of
international economic integration and the effect of electoral politics on nuclear
decisions.4o The major institutionalist argument which delineates a potential cause of
nuclear acquisition comes from bureaucratic politics theory. The argument is that
particular interests of important subgroups within the government have determinate
impact over nuclear decision making.
Many other institutionalist arguments exist in the broader proliferation
literature. Generally though, they are concerned with how international institutions
prevent, constrain or impair the spread of nuclear weapons. These arguments are
relevant in the debate on why states seek to acquire nuclear weapons as they affect the
40 Walsh 2001,31-34; for effects of democracy see: Chafetz 1993, 133; and for effects of economic
liberalization see Solingen 1994, 324-325,332-333. For domestic politics see Reiss 1988, 117-119, 249-
251; and Sagan 1996,63-73.
39
cost-benefit analysis of states. However, they are not alternative theories of the causes
ofnuclear proliferation, and as such will not be addressed individually here.
The application of the democracy variable to nuclear proliferation produces
two main claims. Democratic peace theory argues that democracies do not fight each
other because they share a common identity and set of norms. Democratic peace
theory, as applied to nuclear proliferation, suggests that democracies would tend to be
less likely to go nuclear because they do not face strategic nuclear threats from other
democracies, which significantly reduces the number of states in the system they could
be in conflict with.41 A second claim is that democratic transition in a state can reverse
nuclear aspirations by undercutting the domestic position of factions that favored
weapons.42
Currently, the first claim, that democracies are less likely to develop nuclear
weapons, is not directly applicable to or testable by regime security theory. The group
of states which score well on freedom and democracy variables, and thus are likely not
to proliferate, do not tend to overlap with states which are alienated from the
international system and have potentially unstable domestic regimes. The
complementary claim implied by democratic peace theory may apply. It is almost
certainly the case that rogue states are more likely to develop nuclear weapons. Their
lack of investment in the international system means that they pay lower isolation
41 The implications of applying democratic peace logic to the issue of nuclear proliferation are
discussed in both Walsh 200 1, 3 1; and Solingen 2007, 16-17.
42 Walsh 2001,31.
40
costs for obtaining nuclear weapons. While this logic is discussed more extensively
elsewhere it is worth noting here that it is intuitively consistent with democratic peace
ideas.43
The second claim, that democratic transition in a state can reverse nuclear
policy, seems to apply in one of the cases studied, South Africa, though not for the
reasons suggested by proponents. Democratic transition in South Africa appears to
have reversed the state's nuclear course because those in power before the transition
did not trust the incoming democratically elected government with nuclear weapons. 44
The second liberal - institutional variable is the degree ofglobal economic
integration. The argument that the degree of economic integration may affect the
propensity of a state to acquire nuclear weapons has joint components on the systemic
and domestic level. On the systemic level, increased economic integration increases
the costs for states which acquire nuclear weapons because nuclear weapons tend to
produce economic sanctions; these are more damaging the more integrated the state.
On the domestic level, increased economic integration produces influential players on
the domestic political scene (companies, lobbyists and any others dependent on
increased trade) who tend to be against the open development of nuclear weapons as it
runs counter to their interests. 45
43 Ibid.
44 See Chapter 3.
45 Walsh 2001,33; and Solingen 1994, 317.
41
The degree-of-integration argument, like many of the liberal and
institutionalist arguments, is about mechanisms that prevent or impair proliferation.
However, the complementary argument, that a low level of integration reduces the
costs of a nuclear weapons program, is useful. Globally, the level of international
economic integration is such that isolation from the world market is now the exception
rather than the rule. Thus, the rare regimes which are isolated from the world market
can be considered significantly more likely to acquire nuclear weapons because they
do not suffer the accompanied costs of isolation. This prediction both makes sense and
is consistent with regime security theory. Regime security theory argues that political
isolation lowers the cost of the nuclear option through several mechanisms, one of
which is that isolated states have low initial levels of economic integration, with little
cost associated with losing existing trade options to sanctions. The problem with using
international economic integration alone to explain nuclear decision making is that it
identifies a cost which must be taken into a state's decision making process, but not an
impetus to develop nuclear weapons.46
The third argument that comes from liberalism and institutionalism, though it
is often difficult to distinguish from the domestic prestige argument, is the effect of
domestic electoral politics on a state's decision to acquire nuclear weapons. The
46 There are multiple perspectives as to the cause of high levels of international isolation, whether as
international responses to unrelated domestic policies or nationalist policies eschewing the benefits of
integration, as argued by Solingen 2007,4.
42
argument is that politicians will reject or embrace nuclear weapons dependent upon
what is electorally advantageous at the time.47
This argument has particularly unreliable data. The data that would support the
theory would be a high correlation between announcements of nuclear intentions as a
major campaign issue and the modification of nuclear behavior after a change in
government. In such a correlation it is difficult to separate the signal from the noise.
When campaign promises do not produce change it is not an exceptional occurrence.
Politicians often make campaign promises that have little or no effect on their policy
agenda. Because of this it is difficult to deduce meaning, or signal, when campaign
rhetoric happens to align with actions. In short, political speeches in general provide
data of very poor quality.
Additionally, it is difficult for the electoral politics argument to reconcile the
issue of concealed nuclear weapons programs in democratic states. Finally, the
electoral politics argument cannot explain proliferation in states that are not
democracies. As South Africa is a democracy that concealed its nuclear program and
North Korea is not a democracy, the electoral politics argument does not provide a
strong counter to either of the regime security theory test cases.
The fourth argument is institutional. The bureaucratic politics model argues
that factions within a government have their own bureaucratic incentives, and that
those incentives will push factions towards or away from nuclear weapons. For
47 Walsh 2001,32.
43
example, the civilian atomic energy agency in a government may push for or against a
nuclear weapons program dependent on whether the agency believes it will result in
increased influence and funding or undermine the agency's authority in favor of a
program run by the military. The Bureaucratic Politics model effectively argues that
governments are sectioned off into small groups of rational actors which produce
irrational state policies as an unintended consequence of maximizing their individual
utility.
Evidentiary support for the Bureaucratic Politics models tends to consist of
demonstrating that groups within a particular government advocated a policy in their
own interests, other groups advocated against the policy because it ran counter to their
interests and the policy outcome was determined by which group(s) ended up having
critical influence over the decision making process. The shortcoming with this model
is that for any given state it will almost always be true that, a) some sections of the
government want a policy and some do not, and b) one side prevails. To support the
model, it is only necessary to construct a counterfactual such that, had the interests of
actors not aligned in a certain way, another decision would have been reached. In most
cases this means making a subjective determination about which interests and
influences were critical in the policy outcome of the state. This process is unsatisfying
because it is hard to imagine a situation that would disprove such a model save a case
where all parties had interests in one outcome and another was reached.
The bureaucratic politics model could apply to both of the cases studied.
However, both because of the level of opacity of the regimes and the subjective
44
importance which can be assigned to various interest groups in order to support or
undermine the model, the model will not be tested against the North Korean of South
African case.
Generally, the democratic and institutional arguments provide interesting
insights into the process of nuclear proliferation, but they do so by identifying trends
and tendencies rather than laying out causal paths. This is not the case in the
Bureaucratic Politics Model. However, the model is both overly general and difficult
to disprove. Another limitation of the democratic and institutionalist arguments, and
the primary impediment to testing them against the cases in this study, is that they
often rely on states being democracies or having transparent government functions,
conditions which are not generally met with by states which are at risk or have
undertaken non-standard proliferation activities.
Prestige
The prestige argument is that, given certain nuclear norms, developing nuclear
weapons is politically advantageous, either internationally or domestically. The
international prestige argument is that states acquire nuclear weapons in order to
increase or maintain their international prestige, standing among nations, or regional
influence. The domestic prestige argument is that politicians within a state pursue
nuclear weapons in order to enhance their domestic prestige and thereby enhance their
hold on power. This section will discuss the normative nature of the prestige
45
argument, the specifics of the international arguments, and critiques that apply to both
international and domestic prestige arguments.
The prestige argument has been erroneously criticized by those arguing that
"nuclear weapons are only prestigious ifpeople believe so. ,,48 This critique dismisses
the prestige argument on the grounds that the argument is dependent on a norm
(nuclear weapons are prestigious). However, the prestige argument specifically
identifies the norm as the independent variable and argues that changes in the norm
affect the dependent variable. To critique the prestige argument for being dependant
on a norm it would be necessary to argue that the prestige argument made unrealistic
assumptions about the norm, such as assuming that the norm is a constant when it
could realistically change. Simply identifying that an argument is dependent on the
existence of a particular norm is not grounds for dismissal.
Versions of the prestige argument which do not explicitly identify the norm of
nuclear prestige as the independent variable instead make the implicit assumption that
international or domestic nuclear norms which identify nuclear weapons as prestigious
exist. Therefore, a better critique of prestige theory is that the degree to which nuclear
weapons are prestigious is not set and defined, but based on international norms; thus
it is by no means clear that the gains in international prestige garnered from nuclear
weapons outweigh the costs in prestige or diplomatic standing in a given setting. This
becomes increasingly problematic the lower the relative status of a state in the
international community and the more vulnerable the state is to sanction.
48 Davis 1993, 198.
46
The international prestige argument is that states want nuclear weapons to
increase or retain their grandeur and stature in the international community. This
argument is originally derived from the experiences of Great Britain and France, who,
declining in relative world influence following World War II, are argued to have
sought nuclear weapons in order to maintain the level of world status they perceived
themselves to deserve.49 Some versions of the argument claim that states use prestige
to establish themselves as a regional power and make a bid for regional hegemony. An
example of this comes from the South East Asian experience, where the argument is
made that India sought nuclear weapons both to remain a regional power in Asia and
to assert itself as the dominant power in South Asia, and Pakistan sought nuclear
weapons in order to assert itself as a peer to India in South Asia.
There are two significant methodological issues common to prestige
arguments. First, desire for prestige does not vary among states any more than desire
for survival varies among states, as prestige provides additional security and thus
enhances the chances of state survival. Prestige arguments must be based on what
variables cause some prestige seeking states to acquire nuclear weapons and not
others. 50 Second, while the prestige argument is general and should be universally
applicable, it is often applied only to cases not explained by security. This means that,
in a broad sense (though this argument is not made explicit), many authors promoting
49 Davis 1993, 218.
50 A response to the critique that prestige is a constant is that desire for prestige is opportunistic based
on setting. This response is unsatisfying because the independent variable is then setting. Ibid., 107.
47
prestige arguments do not claim that prestige is an alternative causal path to nuclear
weapons, but rather that prestige is always around and explains whatever cases don't
have sufficient evidence to be explained by security concerns. 51
Empirically, it is difficult to provide unambiguous evidence supporting
prestige arguments. Much of the evidence comes from the political rhetoric of state
officials who have a target audience ofcitizens or other states. Such evidence is
unreliable because governments seeking the bomb should be expected to maximize
political gain, regardless oftheir core impetus for development. More convincing
evidence must be from within the internal dialogue ofgovernments. When such
evidence exists, it is difficult to access. This data collection issue is not unique to
prestige, many arguments are forced to rely on indirect or circumstantial evidence and
find a way to compensate for the issue. However, the problem is acute for prestige
arguments because of the degree to which many case analyses in support of prestige
arguments tend to be based on public rhetoric. 52
From a theoretical perspective, the prestige argument has problems with
explanatory power. In cases where governments do not consider it advantageous to
announce their nuclear capacity prestige arguments have a very difficult time
explaining why states retain a hidden nuclear weapons program after it has succeeded.
If states sought nuclear weapons for prestige we should observe them advertising their
accomplishment.
51 Ibid., 198.
52 Ibid., 201.
48
An additional theoretical issue is that prestige arguments have a tendency to
morph into arguments about security. Distinct prestige arguments define prestige as
providing or enhancing the chances of achieving or maintaining a particular place
within the system. Less distinct prestige hypotheses argue that states can affect the
type of attention and treatment they receive from other powers via the development of
a nuclear bomb. However, this line of logic, where the perception of other powers is
changed in a way that makes the state a less appealing target, is not significantly
distinct from security arguments. Where a prestige hypothesis argues that security is
enhanced by the general alteration of the perception of potential adversaries, a security
hypothesis argues that security is enhanced by developing a sufficient deterrent. Both
claim that nuclear weapons enhance security by changing the cost-benefit analysis of
potential rivals. While they may explain the exact mechanism differently, it is not
clear that this distinction is significant; it does not affect the potential policy or path of
the state. Perhaps more importantly, almost no evidence could substantially undermine
one line of reasoning while supporting the other. For these reasons, such indistinct
prestige arguments seem to add little to existing security explanations.
On the whole, prestige appears a dated argument. In the 1950s and 1960s it
provided a viable alternative to security explanations as the western European powers,
clearly concerned with preventing their fall from importance on the international
stage, were among the first to push for nuclear weapons. However, not only have their
not been good cases for prestige since then, but the effectiveness of such a strategy is
49
made questionable by the postwar relative decline in international prestige of nuclear
Britain and France compared to non-nuclear Germany and Japan. 53
Methodology
The spark of regime security theory was a specific question - why did North
Korea develop nuclear weapons? While it is certainly possible to make the traditional
proliferation arguments regarding North Korea, none stands out as a good explanation
of the behavior of the Kim regime. A strong explanation of the North Korean nuclear
program requires looking at similarly puzzling cases and addressing the limitations
imposed by a low number of cases. In order to satisfy these criteria regime security
theory was derived using a modified comparative case study approach. Generating a
research question entailed:
• Determining a general question that reflects the research objective;
• Looking for outlier cases with similar deviations from the norm;
• Asking the general question of each case under study
• Generating a hypothesis consistent with the data from both cases;
• Employing variables that provide policymakers some leverage in order
to enable them to influence outcomes;54 and
• Describing the conditions under which the theory can be disproved.
53lbid.,218.
5411Iis methodology borrows from George and Bennett's structured and focused comparison. George
2005,69.
50
The case which appears most similar to North Korea, and thereby provides a strong
candidate to investigate for the purposes of theory generation, is South Africa. Like
North Korea, South Africa developed a nuclear arsenal without an obvious security
threat. Both South Africa and North Korea had significant preexisting nuclear
infrastructure before they began to work towards weapons. Both countries expressed
fear of interference by great powers. Both countries, during the course of their nuclear
development, were ruled by domestic regimes which faced international isolation and
condemnation. Both countries also underwent a significant period of domestic unrest.
Like North Korea, South Africa is not explained well by any existing theory of nuclear
proliferation. Regime security theory was generated as a potential explanation of the
behavior of both of these anomalies based on a detailed study of available information
on both cases and comparison of that information.
51
CHAPTER III
SOUTH AFRICA
Apartheid was officially ushered in with the election of the National Party in
1948, though the racial laws and constitution which define the regime had already
been in place for some time. 55 The Apartheid regime was one of the most criticized
and isolated regimes the world has ever known. I argue in this chapter that it was this
isolation combined with domestically inspired and internationally exacerbated regime
instability that led the Apartheid government to embrace the nuclear option.
The first section provides a brief chronology of South African domestic and
foreign political issues from the rise to the fall of the Apartheid government -1948 to
1993. The second section of the chapter provides a chronology of nuclear decision
making and development in South Africa. The third section presents the theoretical
arguments discussed in previous chapters and tests of how well those arguments apply
to the South Mrican case. Specifically, the arguments dealt with are strategic nuclear
security, conventional security, regime security, prestige and international economics.
It ends with the conclusion that the South African case provides fairly strong support
for regime security theory. Finally, the fourth section of the chapter compares the
55 Moore 1987, 46.
52
timing of the South African decision to renounce their nuclear weapons capacity with
the proposed theories and logic.
A Brief Chronology of Foreign and Domestic Relations
A Litany of Domestic Woes
The Apartheid government came to power in 1948 with the election of the
Afrikaner-dominated National Party. In 1955 the African National Congress (ANe),
which would quickly become the major domestic political opponent of and resistance
to the Apartheid regime, spelled out the provisions of the Freedom Charter, with
universal suffrage and a unitary, non-racial South African state as its goa1. 56
On March 21, 1960 during a protest of the "pass laws," between 67 and 69
blacks were killed in what became known as the Sharpeville Massacre. This event set
the tone ofthe tense domestic atmosphere of Apartheid South Africa. The pass laws
restricted black movement in the white areas of South Africa. For the white South
African community Sharpeville served to raise the fear of black urban violence, as
well as reminding the white South Africans oftheir minority status. Following
Sharpeville, the Apartheid regime banned both the ANC and the Pan Africanist
Congress of South Africa (PAC) from operating as political parties. The ban
backfired. The ANC, with legitimate political activity ruled out, turned to violence.
56 The incident is also referred to as the Sharpeville Tragedy. Ibid., 46-50.
53
Umkhonto we Sizwe (Spear of the Nation), the military wing of the ANC, began
attacking police stations and sabotaging military targets following its 1961 founding. 57
The South African economy struggled in the 1970s due to the Apartheid
regime's growing international isolation. However, the economic downturn was
mitigated by the high price of gold, of which South Africa is the largest producer in
the world market. Racial tensions in South Africa hit a new high during the Soweto
riots of June 16, 1976, in which between176 and 600 blacks were killed, most of them
schoolchildren. 58
ANC resistance against the Apartheid regime continued into the 1980s, and
broad level anti-Apartheid demonstrations continued in the mid 1980s, with attacks on
police, and even on black South Africans seen as collaborators with the regime.
Domestic instability caused investors to become increasingly worried. Foreign capital
flight ensued, causing the collapse of the rand. Political discontent against Apartheid
hit the South African economy with a 1-2 punch. South Africa was unable to repay its
short term loans because of the collapse of the rand, and pressure from within the
United States caused U.S. banks to refuse to renew those failing loans. This crisis
57 Reiss 1988, 174; and Moore 1987, 50. The PAC was similar to the ANC, but with exclusively black
membership. Over the following years it played a substantially smaller role than the ANC. Much of the
subsequent information in the South Africa timeline comes from the case studies written by Reiss and
Moore. This information is supplemented by several less thorough cases studies providing information
that does not directly overlap with Reiss or Moore, and is also supplemented by an update by Reiss,
written after South Africa renounced its nuclear weapons.
58 Moore 1987,65.
54
prompted Pretoria to suspend payment on foreign loans and impose government
controls on the foreign exchange market. 59
The Deteriorating International Relations of South Africa
There is unanimous consensus among those writing on the situation that the
racial policies of the Apartheid regime were the primary cause of South Africa's
isolation from the rest of the world. J.D.L. Moore goes so far as to describe South
Africa as a "pariah state," defying "a post-war global consensus that finds racism
abhorrent and seeks to promote decolonization worldwide." Moore describes
international reaction to such states, and South Africa specifically, as a "united and
overwhelming attack from non-aligned and socialist countries in international forums
such as the UN.,,60
In 1987, Moore writes, "South Africa faces a combination of internal and
international opposition aimed at her isolation and the eventual overthrow of the
government." The push towards isolation by both widespread public opinion and civil
rights activists supports these claims. There is a wealth of specific evidence of
sanctions against South Africa beginning in the early 1960s. Two examples which
59 Ibid., 52.
60 Snyder 1985, 156; and Moore, 1987, 46-48.
55
show substantial opposition to South African policies are the oil and arms embargos
against South Africa. 61
A microcosm of broader South African isolation, the oil embargo against
South Africa, began in 1973 with the original participants being the Arab states at the
behest of the Organization of African Unity (OAU). Britain and other states followed
suit, and, critically, the embargo was joined in 1979 by Iran following the Islamic
revolution. Before the revolution, South Africa got 90% of its oil from the Shah,
whereas after the revolution South Africa was forced to get oil at increased cost
through spot market purchases, swap arrangements and loopholes. Tellingly, South
African production of oil from coal and stockpiling of oil in abandoned mine shafts
are indications that it expected and prepared for increased isolation during the 1980s.62
Early in the 1980s, the Apartheid government prepared for a siege. Pretoria
was not alone in this. The regime's siege mentality was widely shared by the
Afrikaner electorate. South Africa's cumulative setbacks in international relations
spurred talk of"fortress South Africa" as early as 1980, with South Africa viewing
herself as a "solitary actor against the outside world." The country's fears were not
unfounded. By 1981 South Africa had formal diplomatic ties with only 15 countries.63
61 Moore 1987,50-52.
62 Ibid., 50-65.
63 Reiss 1988, 178. Additional evidence for South African knowledge and fear of isolation can be found
in Reiss 1988, pages 178 and 179; and Moore 1987, 66.
56
South Africa's Nuclear Policies: A Timeline for Nuclear Decisions
This section outlines South Africa's nuclear energy and weapons programs,
and where possible, delineates the range of time in which the series of critical
decisions culminated in the construction of nuclear weapons. Much of the timing of
actions on nuclear issues corresponds to events in other policy areas. Analysis of this
correlation will be kept to a minimum in this section and addressed later.
South Africa's nuclear industry got a very early start. In the late 1940s, South
Africa began extraction of uranium from gold for export to the United States and
Great Britain. At that time it was estimated that South Africa held 25% of the non-
communist world's uranium reserves. 64 The South African Atomic Energy Board
(AEB) was established in 1949, and in the beginning worked closely with its U.S. and
U.K. counterparts.
During the early nuclear period, South Africa participated as a responsible
member of the Western nuclear infrastructure. The first South African uranium
processing facility was opened in 1952, primarily funded by Great Britain and the
United States. In 1957, the United States and South Africa signed a twenty-year
agreement under Eisenhower's Atoms for Peace program. The agreement with the
U.S. allowed South Africa to purchase the U.S. designed Safari I, its first research
reactor, as well as low enriched uranium (LEU) to fuel the reactor. The same
agreement allowed South African nuclear scientists and technicians to train in the
64 Snyder 1985, 148; and Moore 1987, 82. As seen above, the 25% estimate was later revised downward
to about 13% of easily extractable reserves.
57
U.S., gave South Africa the means to begin its own nuclear research, and made the
Apartheid regime a charter member ofthe International Atomic Energy Agency
(IAEA). During this period, South Africa was considered a leading member of the
IAEA. However, the regime was interested exclusively in the technical aspects ofthe
IAEA's work and ignored the ideological and political goals of the other members. 65
As the 1960s progressed, South Africa focused more on the development of an
indigenous nuclear capacity. In 1965, the Safari I research reactor went critical. While
the reactor operated under IAEA safeguards, those safeguards primarily existed to
prohibit the removal of spent fuel and other byproducts. The safeguards did not
prevent Safari I from being used for experiments designed to develop a nuclear
weapons capability.
When the Safari II reactor went critical in 1967, unlike Safari I, it was not
subject to IAEA safeguards. Safari II was a heavy water moderated reactor, so it
possessed the potential to produce plutonium, the removal of which would not be
monitored. However, because ofthe limited capacity of Safari II, both in power output
and plutonium production, the international community considered it not to be a
significant risk from a proliferation viewpoint. The Safari II reactor was online for
three years before being dismantled in 1970. According to the AEB, budget
constraints forced the organization to choose between investing capital in the reactor
or on a pilot uranium enrichment facility.66
65 Snyder 1985,148-149; and Moore 1987, 82-107.
66 Snyder 1985,149; and Moore 1987, 83-84.
58
In July 1970, Prime Minister Vorster announced the indigenous development
ofa unique and economically efficient means ofcommercial uranium enrichment, as
well as plans for the Valindaba pilot uranium enrichment plant.67 At the time ofthe
announcement, Vorster claimed the purpose of the enrichment plant was to make light
enriched uranium (LEU) for energy purposes. During the mid-60s many countries
presented plans for large nuclear power programs before later realizing they had
underestimated the costs and overestimated the profits.68
Suspicion that the Apartheid government was not being completely
forthcoming came with the announcement that the Valindaba plant was not subject to
IEAE safeguards, with the South African explanation being that they were worried
that international monitors would expose their new, secret enrichment process. While
this rationale was consistent with the economic basis for the program which had been
presented to the international community, it was received dubiously because there
were doubts about how original or indigenous the South African enrichment technique
was.
69
67 Snyder 1985, 150; and Moore 1987, 84-89. Also from Moore: The name, Valindaba, translated from
a local African language means "We do not talk about this at all." The original/alternate name for the
Safari IT reactor, Pelindaba, adjacent to the new enrichment plant, means "We don't talk about this
anymore."
68 Moore 1987, 95.
69 In 1975, at the European Nuclear Conference in Paris, it became evident that the South African
process was a modification of the West German Becker jet-nozzle process, with the South African
innovation being a helicon cascade arrangement needed to compensate for the low "cut" used by South
Africa. The South African process produced a smaller quantity of more highly enriched uranium. This
was a suspicious modification because it suggested that South Africa was prioritizing higher enrichment
59
This second phase of South African nuclear development, beginning in the
1970s looked more suspicious and less peaceful. In 1972, the ANC obtained and
released a document published by the AEB assessing the seismic effect of a 10-kiloton
nuclear explosion in the border areas, and what effect the explosion might have on the
white population. Additionally, it was later revealed that the 1972 and 1973 annual
reports of the AEB contained information on the board's investigations into peaceful
nuclear explosions. Unannounced to the international community at the time, in 1974
the AEC informed Prime Minister Vorster that it could build a nuclear device. Vorster
authorized development of the device, at the time still described as a "peaceful nuclear
explosive." Construction began at the Kalahari test site?O
In 1975, the pilot enrichment plant at Valindaba came online, but at low
capacity. The pilot plant didn't hit full capacity until March 1977. Also in 1975,
Abraham Roux, head of the AEB, announced plans to build a large-scale enrichment
plant with the stated purpose ofproducing more LEU for export. This was an attempt
to move South Africa up a step on the production chain. 71
In 1977, South Mrica entered a more confrontational phase in its nuclear
development. In June of that year, increased tensions over the racial policies of the
Apartheid regime resulted in South Africa's removal from the IAEA board of
governors. Two months later, on August 6, 1977 the Soviet Union claimed that South
level, which makes more sense ifyou are trying to achieve REU than ifyou are trying to achieve LEU.
For more on the specifics of the South African process see, Ibid., 85-88,169-171.
70 Moore 1987, 114; and Reiss 1995b, 9.
71 Snyder 1985,150; and Moore 1987, 90.
60
Africa was preparing for a nuclear test at the Kalahari site. Senior U.S. intelligence
officials, having taken their own pictures of the site, were convinced of the Soviet
claim. South Africa failed to offer a convincing alternative explanation for the
Kalahari site. Though unconfirmed at the time, the Kalahari site had been preparing
for a nuclear test and the attention from the big powers "put the entire testing program
on ice."n
In January 1978, the pilot emichment plant produced a small quantity of highly
emiched uranium (BEU), with production continuing throughout the year. That year
South African leaders publicly opted for a smaller plant than the 1975 proposal
outlined by Roux. The factors which affected the decision to scale down the
emichrnent plant were the doubling of the projected cost of the plant to over $2
billion, the Carter administration's 1977 decision to suspend long term contracts to
supply LEU for the two reactors under construction in Cape Town, and the U.S.
Nuclear Non-Proliferation Act, which formalized the ban on selling fuel to states
which were not signatories of the Nuclear Non-Proliferation Treaty (NPT).73
According to the South African account, it was in 1977 that Vorster convened
senior officials to discuss the future of South Africa's nuclear program. This
information carne to light in 1993 as South Africa announced the destruction of its
weapons stockpile. In April 1978, Vorster formally approved his officials'
72 Snyder 1985, 154; Moore 1987, 106-111; Reiss, 1995b.
73 Reiss 1995b, 8; and Snyder 1985, 150.
61
recommendation. Later that year, in October of 1978, the newly elected Prime
Minister P.W. Botha, moving forward with the broader plan endorsed by Vorster,
appointed an action committee to recommend future plans for producing nuclear
devices. The committee delivered its recommendation, proposing seven nuclear
weapons in July of 1979.74
In September of 1979 the "double-flash" over the South Atlantic was theorized
to be an atmospheric nuclear test. There was high level debate about whether it was a
natural event or a nuclear detonation. At the time, South Africa was the primary
suspect, followed by Israel. As of 1995, the unofficial widespread view in the U.S.
government was that it was most likely that Israel tested a low yield warhead. This
view does not appear to have changed since. 75
According to the South African government's 1993 revelations, South Africa's
first workable gun-type device was fitted with not-quite-weapons-grade RED in
November 1979, precluding its culpability for the September 1979 double flash. 76 It
was not until 1981 that South Africa publicly announced its success in producing
"small quantities" ofuranium-235 enriched to 45%. This report was contradicted by
South Africa's 1993 account, as 45% enriched uranium is nowhere near enriched
74 Reiss 1995a, 9.
75 For the specifics of the early debate see Snyder 1985, 148,154-157; and Moore 1987, 116-117. For
later revelations by the South African government see Reiss 1995, 11.
76 After the device was assembled to ensure everything fit properly, it was disassembled into halves, and
two sections were stored separately at high security installations. It is worth noting that this process was
repeated for all South African bombs. A note on the devices themselves: In order to insure effectiveness
without testing, the South African devices were packed with about twice the amount ofHEU typically
used in a gun-type nuclear weapon. Reiss 1995, 8-11.
62
enough for weapons productions, with the standard percent enrichment to qualify at
90%.77
The 1980s saw a decline in public debate over the South African nuclear
situation. By this period, the state had developed a minimal deterrent and was content
to sit on it. Koeberg 1 came online in 1984, producing power for the Cape Town
region. This is significant because the Koeberg project, delayed repeatedly, was the
first nuclear power plant to produce significant power for public consumption in South
Africa. 78
June 1988 saw the last test preparations conducted by the Apartheid regime.
President Botha ordered test preparations again at the Kalahari site. Unlike 1977, these
preparations were not publicized by the Soviet Union or the United States because of
the delicate state of negotiations over the Namibian situation at the time. Undertaken
almost a decade after South Africa achieved the bomb, the test preparations were
perceived by the United States as a transparent signal by the South African regime to
take the Namibia negotiations seriously, and were almost certainly a direct response to
the bellicose language of Castro during the incident.79
In September, 1989 President de Klerk was briefed on the current status of
South Africa's nuclear program, after which he directed an "Experts Committee" to
77 Snyder 1985, 151.
78 Moore 1987, 97.
79 Reiss 1995a, 14.
63
evaluate the pros and cons of the existing program. In February of the following year
de Klerk commissioned a study on the best way to destroy the existing nuclear arsenal.
In July, he settled for a plan of gradual destruction of South Africa's seven nuclear
devices. The weapons dismantlement was completed one year later, and the vast
majority of the technical documents were destroyed. That same month, July of 1991,
South Africa at last signed the NPT. In September 1991 South Africa signed a
safeguards agreement with the IAEA and authorized inspections for "go anywhere,
. ,,80
anytIme.
There are four conclusions that can be drawn from the timetable of the South
African nuclear program. First, even with an active nuclear program, it took South
Africa almost a decade to build an actual nuclear weapon. Plans for the Valindaba
plant, the most likely source of the RED for South African weapons, were announced
in 1970, and yet it was not until the end of 1979 that South Africa was able to produce
a rudimentary nuclear device. The earliest available evidence suggesting that South
Africa was seriously considering developing nuclear explosives came in 1972, seven
years before successfully producing a weapon. It is likely that had the Valindaba plant
not already been in the works this time lag would have been even longer. This timeline
demonstrates that implementation of nuclear decisions is far from immediate. Even
with sufficient nuclear technology it took South Africa almost a decade to produce
enough RED for a bomb.
80 Ibid., 17-19.
64
Second, there is a range of time between 1970 and 1977 during which critical
South African decisions on the building of a nuclear weapon likely occurred. South
Africa began production of the infrastructure that would be required for nuclear
weapons production in 1970. The major indication that 1970 may have been the
decision point is the refusal of South Africa to allow foreign observers into the
Valindaba site with the excuse that they might steal "unique" South African
technologies. However, it seems more probable that Pretoria was maneuvering to keep
its nuclear options open than that it had actually made the decision to develop nuclear
weapons in 1970. The 1977 endpoint to the range exists because 1977 is when
construction of the Kalahari test site began. By August 1977 the Apartheid
government had clearly decided it was in its best interest to construct a nuclear
weapon, and was gearing up to undertake a test as soon as enough HEU was available.
Third, the most likely final decision point to build nuclear weapons was 1977.
The above analysis suggests 1977 as the decision point because it is the same year that
South Africa was removed from the IAEA, the year of Vorster' s conference on the
future of South Africa's nuclear program, and the same year it began construction of
the Kalahari test site. It also occurred during a period of dramatic increase in internal
and external pressure on the Apartheid regime. Clearly, this decision point does not
encompass the entire development process; South Africa had been considering nuclear
weapons for some time beforehand.
The final conclusion that can be pulled from the timing of the South African
case is that the timetable for nuclear weapons development is heavily dependent upon
65
the existing nuclear infrastructure. The likely 1977 decision point shows that, while it
may take a decade to move from generic nuclear technology to a weapon, with an
operating, un-safeguarded uranium enrichment facility, the timetable to build a bomb
is limited only by the productive capacity of that facility, in this case about 2 years.
Analysis of the South African Case
This section will test the arguments explained in Chapter II against the
observations of the South African case. Analysis of this case can be summarized in
three main points. One, the traditional security arguments cannot explain the
acquisition of nuclear weapons by South Africa. Two, detailed analysis of the South
African case strongly supports regime security theory. Three, neither the international
economics argument nor the systemic prestige argument adequately explains the South
African case. Support for these conclusions will be developed in three subsections.
These will:
• Test the strategic nuclear threat argument and the conventional security threat
argument;
• Review the evidence of the South African case as it applies to each part of the
regime security argument; and
• Apply the international economic and international prestige arguments to the
South African case.
66
Traditional Threat Arguments
The Strategic Nuclear Threat against South Africa
South Africa was under substantial pressure from the Soviet Union from the
early sixties to the early eighties. During this period a number of factors may be
argued to have led Pretoria to think: they were under threat of nuclear attack from the
Soviet Union. First, the Soviets had committed resources to southern Africa, and over
the course of this twenty year period, an increasing number of African states were both
friendly to the Eastern bloc and reliant on the U.S.S.R. as a patron. In particular, the
governments of Angola, Mozambique and Zimbabwe had all come to power through
revolutionary violence directly or indirectly supported by the Soviet Union.S! Second,
the Apartheid regime, especially in dealings with the West, emphasized the strategic
importance of South Africa, as it sat on the Cape of Good Hope. Third, the Apartheid
regime was not confident that they could rely on the nuclear umbrella provided by the
Western powers. However, none of these factors stand up to scrutiny when purported
to signify a Soviet nuclear threat.
While the Soviet Union did provide support for communist forces in southern
Africa, this support came primarily as equipment and training. When actual troops
were deployed they came in the form of Cubans, not Soviets. The Soviet Union
81 Moore 1987,52.
67
committed limited resources for communist movements in southern Africa because the
region was never a strategic priority during the Cold War.
Pretoria drastically exaggerated the strategic importance of oil shipments
through the Cape of Good Hope in order to push for support from the West. Pretoria
attempted to use the Cape argument to push for a special NATO extension which
included South Africa. The regime also made overtures on both sides of the Atlantic
for a South Atlantic Treaty Organization, emphasizing the strategic importance of the
Cape route as the country's value added. These attempts were repeatedly denied by all
involved parties, evidencing that the Cape route was not of significant strategic
. 82Importance.
The supposed importance of the Cape route is also undermined by the West
completely ignoring the South African response to the expiration of the Simonstown
agreement, where Pretoria announced that "the West would have to police the
shipping lanes around the Cape by itself." However, the most damning evidence
against the strategic importance of the cape argument is that at the time Simonstown
expired in 1975 Western analysts simply didn't take seriously the idea that the Soviets
would be prepared to go to war by impeding the flow of raw materials around the
Cape during peacetime.83
The above logic speaks to the concern that South Africa could not count on the
nuclear umbrella of the West against the Soviets. It is possible that, if Great Britain
82 Ibid., 63.
83 Ibid., 63-64.
68
and France were worried about the resolve of an extended U.S. nuclear deterrent,
South Africa had cause to be. At the same time, South Africa was a stretch for the
Soviet sphere of influence in the first place, and was significantly less strategically
important than many other non-aligned states. Southern Africa was so unimportant to
the Soviets strategically that their lack of economic incentive to invest resources in the
region is cited as one of the reasons they pulled out in the early eighties and offered
minimal support to the frontline states throughout the Cold War period. 84 At no time
was South Africa close to being worth the global political and military costs of a
nuclear attack from the standpoint of the Soviet Union.
It is possible to argue that the nature of the Apartheid regime caused a unique
level of ire from the Soviet Union. However, a nuclear response would not follow
from such an objection. Using nuclear weapons because of an objection to how a
government treats its citizens would kill hundreds of thousands of citizens. If the
Soviet interest in South Africa was an objection to the Apartheid regime, and not a
strategic interest in the state - an idea to which the chronology of Soviet activities in
the region strongly points, the Soviet nuclear threat was non-existent from the start.
The lack of logical support for the idea that South Africa needed to be concerned with
a nuclear threat from the Soviet Union undermines the strategic nuclear threat theory.
If the strategic nuclear threat against the Apartheid regime did not exist, it does not
make sense for South Africa to have developed nuclear weapons as a response to that
threat.
84 Ibid., 56.
69
The Overwhelming Conventional Threat against South Africa
By the mid seventies white South Mrica feared it would be alone in facing the
conventional Soviet threat. In 1981 foreign observers speculated that Pretoria might
develop nuclear weapons in order to alleviate the country's insecurity and to guarantee
the preservation of its singular racial policies.85 There are three main claims that need
to be addressed in regards to the possible conventional military threat to South Mrica.
One, because South Africa could consider the majority of its neighbors as rivals, it
was therefore in a hostile environment and legitimately concerned with the
conventional military balance. Two, there is evidence of cross-border skirmishes
between the Apartheid regime and its neighbors. Three, there is significant evidence of
Soviet activities in southern Mrica during the Apartheid period.
The first claim is partially correct; South Africa was surrounded by rivals.
However, the conventional military balance between South Africa and its immediate
neighbors was overwhelmingly in favor of South Africa. By 1976 South African
defense spending amounted to half of the defense budget of the rest of the African
states combined, excluding Egypt. From 1976 to 1986 South Africa's defense budget
continued to rise. During the 1980s South Africa's military vastly exceeded the
capacity of regional rivals, especially in armor and air power.86 The power of the
South African economy, coupled with the military weakness of the newly emerging
85 Reiss 1988, 176-179.
86 Additional details on the makeup of the conventional military balance can be found at Moore 1987,
53-55; and Reiss 1988, 175.
70
southern African states, ensured that, without significant great power assistance, South
Africa's neighbors could not be considered a conventional threat.
While there were instances of cross-border skirmishes, those skirmishes were
primarily initiated by South Africa crossing the border into Angola, Zimbabwe or
Mozambique. In the case of Angola, this was to intervene in the Angolan civil war; in
the cases of Zimbabwe and Mozambique it was to target SWAPO and ANC bases in
those countries. While the Apartheid regime did come under limited attack by
numerous non-state actors, most prominently the SWAPO in Namibia and the ANC in
South Africa itself, these were resistance organizations, not national militaries. At no
time did the national governments of South Africa's regional competitors attack South
African territory. During the Apartheid period the closest South Africa came to being
under attack by another state was the threat by Castro that his forces would cross the
border from Angola into Namibia.
Soviet activity in southern Africa provides the strongest argument that South
Africa may have been imperiled by conventional forces. Prime Minister P. W. Botha
outlined the Soviets' objectives in South Africa as follows: "The main object ... under
the guidance of the planners in the Kremlin, is to overthrow this state and to create
chaos in its stead, so that the Kremlin can establish hegemony here." Deputy Defense
Minister Coetzee claimed that East German troops were deployed in Angola and
Mozambique at one point, and the presence of Soviet tanks and support vehicles
71
across the border indicated that South Africa must prepare to face a conventional
threat at some point.87
It is important to contextualize comments of the Apartheid government during
this period. The government had every incentive to overstate the Soviet threat, both
domestically and internationally. Domestically, the nationalist party received
increased popularity for its hard line against the nebulous Soviet threat, and
internationally, the more threatening the Soviets, the more South Africa thought it
could gamer support from the West. Domestically, this strategy of threat inflation
appears to have been convincing. The communist threat was believed by the populous.
In 1982, 80% ofwhite survey respondents disagreed with the statement "the
communist threat against the country is exaggerated by the government. ,,88
The actual record of Soviet actions in southern Africa paints a very different
picture than that put forward by the Apartheid government. Soviet actions appear
limited by the same rationale outlined under strategic threats, above; the region was a
low priority. Historically, Soviet strategy in the region followed the pattern of
supporting domestic resistance groups and supplying arms and training, not overt
military action. Summarizing Soviet involvement, Moore describes it as limited to the
armament and training of anti-South African, non-state actor groupS.89
Considering the conventional balance of military forces, the predominance of
South African aggression in the border conflicts of the period, and the weII established
87 Snyder 1985, 158-159.
88 Ibid., 158-159.
89 Moore 1987, 56.
72
pattern of limited Soviet intervention through funding opposition groups, a
conventional attack against South Africa by any party in the region was extremely
unlikely. The lack of anything close to an "overwhelming conventional threat" in the
South African case suggests that the conventional military threat hypothesis cannot
explain the decision of South Africa to acquire nuclear weapons.
Regime Security Theory
Because we know that South Africa did develop nuclear weapons, in order for
regime security theory to provide an adequate explanation of the South African case,
four things must be true.
• The Apartheid government must be dealing with a period of increasing
domestic instability.
• The Apartheid regime must have a perception of international isolation,
including pressure from at least one major pole for domestic reform and fear
that one or more major powers is considering interfering with its domestic
affairs.
• The Apartheid regime must have the capacity to produce nuclear weapons, and
the time lag between increased pressure and active weapons should reflect that
capacity.
• The timing of the decision to build nuclear weapons should correspond to a
period of increasing domestic instability and international isolation.
73
Direct evidence of the logic that the Apartheid regime wanted weapons in order to
prevent unwanted international interference and to promote international support in
order to maintain the status quo is not requisite for regime security theory, but, if
available in the case, strengthens the theory.
Evidence ofSouth African Domestic Instability
The domestic political history of South Africa is characterized by an
increasingly prevalent pattern of instability beginning in the 1960s and continuing
through the 1980s. The account paints a picture of instability which hits a new high in
1976, and continues to go up from there. There is no doubt that the degree of domestic
instability that existed in South Africa was sufficient to cause the Apartheid regime to
worry about its continued control. In fact, many of the Apartheid regime's policies
were openly made with the goal of preventing regime overthrow.
As early as 1960, the Sharpeville massacre undermined the regime's claims of
domestic stability. For white South Africa, Sharpeville raised fears of black urban
violence. The Apartheid regime responded to Sharpeville by banning black opposition
political parties. The ban backfired. With the removal of legitimate political activity as
an option the ANC actively rebelled, founding its military arm the following year. The
result of Sharpeville was that by 1961, an active, organized and well supported
domestic resistance group was making attacks against and sabotaging Apartheid
targets. The undermined political stability and the subsequent outflow of over half the
74
country's foreign capital led to the first predictions of the Apartheid regime's
imminent collapse originating in the early 1960s.
The Apartheid government quickly rebounded, however, restoring substantial
domestic stability in the mid-1960s. This was accomplished by outlawing black
resistance organizations and leading a series of raids against their hideouts, including
the Rivonia raid, which resulted in the public treason trial of high ranking members of
the ANC, including Nelson Mandela.9o
With low level racial tension and continuing ANC resistance during the
interim, the next major destabilizing event was the Soweto riots of June 16, 1976. The
possibility of increased internal turmoil resulting from the riots is one of the reasons
cited by the Apartheid regime for enacting its "total national strategy" of military
mobilization. The predictions of increased turmoil were accurate. The late seventies
saw a rise in black unemployment and continued black disenfranchisement, which led
to increasing anger at the Apartheid regime and caused an exodus of black youth.
These youth left South Africa proper and joined up with militant resistance
organizations such as Umkhonto we Sizwe, which in turn increased attacks on the
regime. In addition to black anger, in late 1978 Pretoria's authority was undermined in
white South Africa following revelations of high level corruption. 9 ]
During the 1980s, resistance by the ANC consisted primarily of military
sabotage and attacks on symbolic targets. In 1981 the ANC blew up oil storage tanks
90 Reiss 1988, 174; and Moore 1987, 52.
91 Reiss 1988, 176-178.
75
in Sasol. In July 1982 the ANC claimed responsibility for a fire at the not yet
operational Koeberg nuclear power station. In December 1982, the ANC claimed
responsibility for fours bomb explosions at the Koeberg site. Finally, in 1983 the ANC
claimed responsibility for a bomb exploding in an air force headquarters building in
downtown Pretoria.92
In 1984 and 1985 there were broad level anti-Apartheid demonstrations and
attacks on police and black collaborators. In July of 1985 the Apartheid government
declared a state of emergency, enacting additional repressive measures which resulted
in the deaths of over 1000 black South Mricans. Widespread unrest continued into
1986. The domestic instability predictably made investors nervous and precipitated a
major flight of foreign capital which in tum caused the collapse ofthe rand and
increasing economic chaos.93
A Perception ofInternational Isolation
The short history of the Apartheid regime is one of increased international
isolation from the world at large (primarily via the UN), its neighbors and the Soviet
block, and finally from its former Western patrons. This section will present
substantial evidence of increasing regime isolation, and then check for the specific
conditions of the perception of isolation variable. Evidence will be subdivided into
92 Snyder 1985, 163; and Moore 1987, 97. Both acts of sabotage against the Koeberg plant took place
before fuel was put into the reactor, so there was no danger of radioactive release from the sabotage.
93 Moore 1987,51-52; and Reiss 1988, 174-178.
76
three categories: the relationship between South Africa and the United Nations,
focusing on the UN reactions to the Namibia situation and Apartheid itself; the
relationship between South Africa, the rest of Southern Africa and the Soviet Union;
and South Africa's deteriorating relationship with the Western powers. The specific
conditions of the isolation variable that will be checked are, first, that the Apartheid
regime feared domestic interference by one or more of the great powers, and second,
that in addition to being isolated internationally, the Apartheid regime widely
perceived that it was isolated internationally.
The United Nations, Namibia and Apartheid
Apartheid was formally placed on the United Nations agenda as an issue in
September of 1952. However, the focus of the UN at this point was on the emerging
superpowers and areas of direct competition between them. Significant UN attention
did not shift to South Africa until the 1960s. In November of 1960, Ethiopia and
Liberia mounted a legal challenge before the International Court of Justice over South
Africa's right to administer the League of Nations mandate of South West Africa. In
June 1963 a special committee on Apartheid was established and resolutions were
passed condemning the regime. On October 27, 1966 the United Nations General
Assembly passed resolution 2145 (XXI), which declared that South Africa failed to
77
fulfill its obligations under the League ofNations mandate, and that South West
Africa was now "the direct responsibility of the UN." 94
Momentum on the South West Africa (Namibia) issue stalled in July of 1966
when the International Court unexpectedly dismissed the South West Africa case. In
June 1971, the International Court delivered an Advisory Opinion on South West
Africa that reversed the 1966 decision. The court ruled that the Republic of South
Africa's mandate for the territory had been lawfully terminated by the General
Assembly in 1966. In 1973, despite South Africa's continued administration and
military control ofNamibia, the claim ofthe South West African Peoples Organization
(SWAPO) was recognized by the United Nations.95
In 1974 the continued occupation ofNamibia and Apartheid policies led to an
attempt to expel South Africa from the United Nations, a move only blocked by the
veto power of the big Western states - the U.S., France, and Great Britain. However,
South Africa was barred from the XXIXth General Assembly session. In 1976, just
three years after SWAPO was recognized by the General Assembly, it was granted
observer status.96
In 1977 the UN Security Council passed a resolution making the 1963
voluntary arms embargo against Pretoria mandatory. This resolution followed a
pattern of especially publicized civil rights abuses by the Apartheid regime. The 1963
94 Reiss 1988, 175; and Moore 1987,48.
95 SWAPO was a resistance organization operating within Namibia and bordering states to represent the
Namibian people. Reiss 1988, 175-176; and Moore 1987,48.
96 Reiss 1988, 176-177. Taiwan and the Vatican both have observer status at the United Nations.
78
voluntary arms and mixed use embargo followed the Sharpeville massacre; the 1977
mandatory arms embargo followed the detention of Steven Biko, a prominent activist
and black consciousness leader. More general international criticism continued to
match up with Apartheid abuses later in 1977 with increased criticism following the
beating and death of Steven Biko while still in the custody of the Apartheid regime.97
The Namibia dispute continued to draw attention during 1977. Five Western
states - the U.S., Britain, France, West Germany and Canada - created a contact group
to take a more active role in the transition of South West Africa to popular rule. This
contact group led to a UN resolution calling for supervised elections in 1987, adding
to pressure on the Apartheid regime to release its hold on the territory. The regime's
response to this pressure was to set up their own "transitional government," effectively
a stalling tactic, and in 1980 to raid deep into Angola and establish virtually
continuous occupation of southern Angola in an effort to control the movements and
capacity ofSWAPO. 98 It was not until 1988 that Pretoria, under pressure from
neighbors, the West, and the Eastern bloc, granted Namibia its independence under the
condition of Soviet satellite troops being withdrawn from the continent.99
97 Reiss 1988,177; and Moore 1987, 50.
98 Reiss 1988, 177; and Moore 1987,48-49.
99 The key actors in the actual negotiations on the release of Namibia were South Africa, Angola, and
Cuba, whose troops the South Africans wanted removed. Reiss 1995a, 34.
79
A United Southern Africa with a Little Soviet Help
The Organization of Mrican Unity (OAD) was formed in June 1963. One of its
main goals was the independence ofAfrican nations still under minority rule. From
the onset the OAU faced a specific set of challenges. Its membership consisted
primarily of countries with small, dependent economies and younger, less established
governments emerging from the decolonization of the continent. Over time these
problems were exacerbated with the continued weakening ofOAU member economies
and the increasing corruption and instability in their governments. 100 Despite these
challenges, the OAU was able to put significant pressure on the Apartheid regime by
appealing to its more powerful allies for resources and diplomatic support. South
Africa's continuing strategy in regards to her regional opponents during this period
was to use economic and military power to keep them off balance, relying heavily on
the weaknesses of the OAU, and the frontline states (those bordering South Africa and
South West Africa) in particular.
Possibly the most important success of the OAU was that in 1973 it was able
to forge an alliance with Arab nations in order to get them to boycott oil exports to
South Africa. Iran joined the boycott in 1979, significantly increasing its effectiveness.
Arab nations agreed to undermine South Africa because the OAU reciprocally
ostracized Israel. 101 While the oil boycott was a point oflight for the OAU, the
100 Reiss 1988, 174; and Moore 1987,48.
101 Moore 1987,49-50.
80
regional juggernaut that was the South African economy continued to allow the
Apartheid regime to engage in a strategy of international manipulation over the
following two decades.
South Africa's use of financial pressure to manipulate its neighbors goes back
to their decolonization. From the independence of the southern Mrican states until
well into the 1980s, South Africa was able to maintain strong trading links with each
of them despite those countries' objections to the Apartheid government. Pretoria used
her economic dominance to undermine the stability of her neighbors through methods
such as the manipulation of credit flows and utility infrastructure.
In 1966, shortly after the International Court's reversal of its decision on
Namibia, South Africa partially reversed course. Pretoria embarked on a concentrated
campaign ofuitwaardse beweging (outward movement), seeking to use economic
power positively to increase ties to neighbors and bolster their economies in order to
overcome regional inhibitions to dealing with the Apartheid regime, while at the same
time continuing to use economics as a tool of manipulation. 102
The policy of reducing external hostilities to the regime through economic
openness was partially successful, producing positive change in the relationships
between the Apartheid regime and Botswana, Lesotho and Swaziland. This success
was moderated by two factors. First, Lesotho and Swaziland were small, relatively
weak states already deep within South Africa's sphere of influence and economic
102 Reiss 1988, 175.
81
orbit. Second, these states comprised the entirety of the subgroup of southern African
nations which feared communism more than Apartheid. 103
The event that perhaps most undermined South Africa's comfortable
dominance of the region in the early post-colonial period was the overthrow of the
Caetano regime in Portugal in 1974. This precipitated Portuguese divestment of its
interests and colonies in southern Africa. Following the Portuguese pull-out, the
peaceful transition of power in Angola (a former Portuguese colony) broke down in
January 1975 and led to the Angolan civil war. South Africa embarked on an
intervention mission into Angola in October 1975, with the goal of installing a
friendly government in Luanda. The intervention mission failed, and a government
hostile to the Apartheid regime came to power. The primary cause of the failure of the
South African intervention mission was the interference ofCuban troops, supported by
arms and advisors airlifted into Angola by the U.S.S.R. The new Angolan government
itself received substantial arms transfers from the U.S.S.R., and 20,000 Cuban troops
remained in Angola to support the new state.104
The emergence of a hostile Angola was only one of a string ofregional
security setbacks for South Africa during the 1970s. Angola provided a safe haven that
allowed increased exposure of South African forces in Namibia to guerilla raids by
SWAPO. In addition to Angola, a Marxist government came to power in
Mozambique. The late 1970s also saw the fall of the beleaguered white settler
103 Ibid., 175.
104 Reiss 1988, 176; Moore 1987,49; and Snyder 1985, 157.
82
government in Rhodesia. In a blow to South Mrican regional prestige, the Lancaster
House accords of 1979, which paved the way for the peaceful transfer of power in
Rhodesia, were negotiated without South Mrica's participation. 105
The Apartheid government's response to these setbacks in neighboring states
was to formulate a "total national strategy" in 1978 and develop it through the late
1970s and early 1980s. The "total national strategy" is cited as a response to a Soviet
backed "total onslaught," which referred to the shift in the governments of Angola,
Mozambique and Zimbabwe (formerly Rhodesia). All ofwhich came to power
through Soviet-backed revolutionary violence. It is important to note that at the time
Angola, Mozambique and Zimbabwe formed an almost complete cordon around South
Mrica and South West Mrica on its northern borders. Pretoria, not without cause, felt
surrounded by hostile governments. 106
Statements by Deputy Defense Minister H. M. J. Coetzee reinforced the
perception of "total onslaught." In 1979 Coetzee claimed that four to six thousand East
German troops were deployed in Angola and Mozambique and were preparing to
directly interfere in support ofSWAPO. In 1981, Coetzee again promoted the idea of
an overwhelming military threat, claiming that the presence of300 Soviet tanks and
similar numbers of support vehicles meant that Pretoria must prepare to face a
conventional Soviet threat. 107 Regardless of the drastically exaggerated claims of the
105 Moore 1987,48,113; and Reiss 1988, 176-178.
106 Moore 1987, 52. For another perspective on the specifics of this policy see Zolberg 1989.
107 Snyder 1985, 159.
83
Apartheid regime, the Soviets did fund Umkhonto we Sizwe (the military arm of
ANC), providing specialist training and weapons. lOS
Shortly after the election ofPrime Minister Botha in 1978, and concurrent with
the formulation ofa "total national strategy" on the military front, the new South
African Prime Minister pushed for a regional constellation of states in order to
increase economic cooperation between an increasingly regionally isolated South
Africa and other African nations. In response, other African leaders established a
South African Development Coordination Conference (SADCC) in 1979 to promote
intra-regional trading while excluding South Africa. The ultimate goal of SADCC was
to escape South Africa's economic orbit and thus give members the freedom to pursue
active measures against Pretoria.
The Apartheid regime successfully minimized the effects ofthe SADCC by
employing the same economic pressure tactics it had used previously (as well as actual
sabotage) in order to undermine the independence of the SADCC states. These same
economic destabilization tactics were also used to attempt to undermine the
movements and activities of the ANC in neighboring countries during this period. 109
In the early 1980s the Apartheid regime signed non-aggression pacts with its
neighbors Botswana, Lesotho and Mozambique. While this relieved tensions on the
eastern edge of South Africa's border, these were the states least threatening to South
108 Moore 1987,51.
109 Reiss 1988, 178; and Moore 1987, 58-60.
84
African security and most cooperative in past agreements. In fact, the only substantial
benefit of these non-aggression pacts appears to be that they allowed Pretoria to curtail
ANC movements within these border countries. This suggests that the primary
purpose ofthese external arrangements was to enhance domestic stability - security
against the ANC, and not to decrease regional threat - security against other states in
the region.
Despite the non-aggression pacts, the Apartheid regime engaged in a military
operation in Mozambique in which South African armed forces attacked ANC
personnel in a compound outside ofMaputo. Again, this suggests the pacts were
valuable to South Africa only to the point at which they undermined the ANC.
In response to this attack, Mozambique called upon a treaty of friendship with
the Soviet Union, who reluctantly responded. Ten days later two Soviet warships
appeared in Maputo harbor to "show the flag.,,110 With the exception of this flag
waving, the 1980s saw a dramatic reduction in the Soviet presence in southern Africa.
It was practically non-existent between 1981 and 1987 with the exception of the
Cuban troops in Angola. The last push of the Soviet bloc and its allies in the region,
corning in late 1987 and early 1988, was made not by the Soviet Union, but by Castro,
who increased hostilities along the Angolan border in a final effort to increase pressure
for Namibian independence. 111
llO Moore 1987, 56-58; and Snyder 1985, 163.
m Reiss 1995, 14; and Moore 1987,56-60.
85
A Deteriorating Relationship with the West
From the 1950s onward, South Africa had been dependent on the great powers
ofthe West to relieve its growing isolation. The nature of South Africa's relations with
the West was critical throughout the period because the UN veto ofthe United States
and Great Britain was the only barrier between South Africa and significant economic
sanctions. Relations with the West were also the only thing standing between the
Apartheid regime and stronger condemnation and intervention against South African
activities. These included brutal responses to domestic unrest, failure to grant
independence to Namibia, and military raids into neighboring countries. Finally, South
Africa constantly pushed for technological assistance from the West.
Western relations with South Africa degraded as the reign of the Apartheid
regime progressed, starting strong and undergoing constant strain and eventual
deterioration. Protection within the international community waned, and assistance
was increasingly withheld.112
In the 1950s and as late as the early 1960s South Africa fought to ensure its
place as a staunch ally to the Western powers in their battle against communist
expansion. Its primary asset in this bid was as a natural resource exporter. South
Africa supplied a wide range of raw materials, but its huge and accessible uranium
reserve was of special interest to the Western powers. Even up until the 1970s, South
112 Moore 1987,60-66.
86
Africa supplied 13% tol4% of the non-communist world's demand for uranium.
According to a 1983 report by the IAEA, South Mrica contained 13% ofthe world's
total supply ofaccessible uranium, second only to Australia. During the early
development of nuclear programs South Africa participated in the Combined
Development Agency (CDA) with the United States and Great Britain, contracting
with them on uranium mining and export. 113
There were a number of other factors which South Africa used in attempts to
prove itself a valuable ally against communism besides its stores ofuranium. Probably
the most important of these was a more general reciprocally profitable trading
relationship with the West. South Africa exported raw materials and imported Western
goods while remaining a stable venue for reliable, profitable return on investment for
Western companies. 114
Additionally, it had a stated and open devotion to opposing the spread of
Soviet influence. This was reinforced by the Apartheid government dispatching a
fighter squadron to Korea and negotiating the Simonstown naval base agreement of
1955 with Great Britain, which allowed the British a port on South African territory.
As a final incentive for the West, South Africa pushed the idea that its position on the
Cape ofGood Hope was important for the stability ofoil shipments. 115
I13 Snyder 1985,149; Reiss 1988, 173; and Moore 1987,61-71. This infonnation originally comes from
OECD Nuclear Energy Agency and lAEA, Uranium Resources, Production and Demand, Paris:
OECD,1983. .
114 Moore 1987,61.
115 Reiss 1988, 173-174; and Moore 1987,62,63.
87
The 1960s saw the beginning of the deterioration of relations between the
Apartheid regime and the West. Pretoria endured constant condemnations of
Apartheid's racial policies from the U.S. These came in increasingly severe tones as
the decades passed. However, except for the arms and "dual purpose" items embargos,
there was little in the way of substantive sanctions against Pretoria.
Western countries had divergent interests in South Africa. Western companies
wanted continued economic ties, whereas Western publics wanted to condemn and
isolate the Apartheid regime. These conflicting interests created a dilemma for
Western governments when dealing with the region. They tried to maintain the money
flow while simultaneously condemning the Apartheid system. This resulted in a
hands-off approach, and a low, but constant, pressure on South Africa to change.
In April 1961 South Africa was forced to withdraw from the British
Commonwealth. In 1975 the Simonstown naval agreement expired and was not
renewed. In April 1976 the United States rescinded its policy of "selective relaxation"
toward white minority regimes in southern Africa, instead strongly endorsing majority
rule on the continent. 116
These policy changes on the part of the United States and Great Britain
through the 1960s and 1970s caused South Africa to feel abandoned by the West, a
perception which was driven home by the lack of U.S. support in Angola. In 1976
international response to the Soweto Riots was emphatic, with increased worldwide
116 Reiss 1988, 174-177; and Moore 1987, 63. The withdrawal from the British Commonwealth was an
enforced withdrawal and was precipitated by Apartheid policies, directly by the Sharpeville massacre.
88
censure of the Apartheid racial policies. International reactions included a United
Nations mandatory arms embargo against South Africa. 117
The level of international ire against Apartheid rose to the point that Pretoria
informed the South African public that in the event of military conflict it would not be
supported by the West. The state's international status and lack of allies, coupled with
the possibility of increased internal turmoil or aggression by neighbors, led the
Apartheid regime to announce a "total national strategy" for mobilizing the state for
defensey8
The inauguration of the Carter administration, with its emphasis on human
rights, ensured the continued commitment of the United States to the cause ofblack
majority rule. Further, the Carter administration's self-described absence of "an
inordinate fear ofcommunism" signaled to the Apartheid regime that it could no
longer gamer U. S. support by portraying itself as a bastion against communism in
southern Africa. In 1985, responding to increased domestic repression by the
Apartheid regime, Western governments tightened the arms embargo, curtailed
nuclear cooperation and finally imposed their own, limited, economic sanctions. 119
117 Reiss 1988, 177; Moore 1987, 10; and Snyder 1985, 159. All three authors list different casualty
figures.
118 Reiss 1988, 177; and Moore, 1987,49.
119 Reiss 1988, 177; and Moore 1987,52.
89
Checking the Conditions ofIsolation
There is ample evidence that the Apartheid regime feared domestic
interference. South Africa both feared and experienced Soviet interference with her
domestic affairs. While the U.S.S.R. never directly attacked South Africa, it did fund,
train and arm Umkhonto we Sizwe. The Soviets also funneled some of their arms
shipments in southern Africa into the hands of the small South African communist
party. In addition to funding the ANC and South African communists, the Soviets also
aided SWAPO in its fight against Pretoria for control of South West Africa (Namibia)
through Cuban troops with Soviet arms, advisors and airlifts. In sum, the Soviet Union
provided critical funding for each of the organized non-state actors attempting to
undermine South Africa from the inside. 120
It is also clear from the historical record that the Apartheid regime understood
and fought against its growing isolation. Following are quotes and direct policy
explanations made by members and associates of the Apartheid regime in which they
discuss the consequences of their perceived isolation. In 1977, referring to the growing
sense of South African isolation, then Defense Minister Botha told the National Party
Congress that South Africa was "moving more and more in the direction in which the
state ofIsrael has already been since 1948."121 Also in 1977, an excerpt from the pro-
government paper "Beeld" lays out the great powers, both the Soviets and the West, as
120 Moore 1987,49-56.
121 Snyder 1985, 160.
90
bullying South Africa: "The great powers which have nuclear weapons have adopted
an odd attitude. One would have thought that it would have been tactically more
profitable for them to draw closer a potential member of the nuclear club, which South
Africa is. Their bullying attitude could result in making us a maverick bull in the
nuclear herd, and that is surely not a sound situation from their point of view. South
Africa will go its own way and its own interests will be decisive.,,122
Capacity Concerns
South Africa clearly had the economic and scientific capacity to create an
indigenous nuclear weapons program. Two things are interesting about the capacity
variable in the South African case: one, how capacity issues affected the timing of
nuclear development, and two, what decisions were modified by capacity concerns to
the degree that suggests the variable is important as a conditional variable in other
cases.
How the capacity variable affected the timetable for decision making is
discussed in the conclusion of the section on South Africa's nuclear policies. There are
two important points to reiterate. First, beginning with an active nuclear power and
research program but no domestic uranium emichment, it took the Apartheid regime
almost a decade to build a nuclear weapon. Second, the timetable for nuclear weapons
development is heavily dependent upon the existing nuclear infrastructure. The 1977
122 Ibid., 168. Originally from Beeld, August 24t\ 1977.
91
decision point shows that, while it may take a decade to move from generic nuclear
technology to a weapon, an operating, un-safeguarded uranium enrichment facility
allows a state to build a bomb in the time it takes that facility to produce sufficient
quantity and quality ofREU. In the South African case, it took about 2 years.
The South African case provides some well documented examples of how
restricted resources affect the scope and timing of development. The U.S. refusal to
supply enriched uranium for the Safari I reactor greatly impaired research operations,
reducing the scope of programs and forcing the reactor to spend alternate weeks off-
line. Supply restrictions by the United States also affected the scale of the uranium
enrichment facility South Africa was able to build. The Valindaba plant, as announced
in 1970, was scaled down because the costs of a large scale, commercial enrichment
plant increased to the point that a massive amount of foreign capital would have been
required. This price increase was primarily precipitated by the U.S. government
convincing European countries to agree to withhold critical technological components,
including computers. 123
Another significant modification that was made to the South African nuclear
path was the design of their testing facility. When explaining why the entirety of the
Kalahari facility was not concealed underground to avoid satellite detection, lW. de
Villiers, at various times weapon designer, president and chairman of the board of the
directors of the AEC, complained that it was "too damned expensive to go
123 Moore 1987,91-98.
92
underground completely.,,124 This is perhaps the most interesting resource defined
component of the South African plan because it suggests limited resources and
increased costs may prevent states from building the capacity to secretly test a nuclear
device or complete a test facility without the great powers finding out significantly
beforehand.
Lastly, South Africa claimed its nuclear program cost a total of $200 million,
though when taking into account associated costs and rand depreciation this looks
more like $300 to $600 million. Even with this budget and an abundant domestic
uranium supply, it still took South Africa a decade for its nuclear program to proceed
to the point where it could produce a weapon, suggesting that a fast-track plan of
weapons development would be prohibitively expensive for most states. 125 Overall,
evidence suggests that both the available funds of the AEC, and the limited capacity of
the Apartheid regime to obtain sensitive equipment and resources from the outside
world, had a significant effect on the timetable for weapons production.
Checking the Timing ofNuclear Development
Testing the timing of increased instability and international isolation against
the decision to acquire nuclear weapons is problematic because it is difficult to
establish when the decision points for nuclear weapons development occurred with a
124 Reiss 1995a, 10.
125 Ibid., 10.
93
high degree of confidence. The difficulty in establishing decision points is due in large
part to the time lag between policy decision and production implementation, especially
because that time-lag is not set. Nevertheless, the conclusions provided in the nuclear
decision timeline section attempt to outline the most likely decision points and define
the range of possible times for the final decision of South Africa to acquire weapons.
The range of possible dates for nuclear decisions most likely falls between
1970 and 1977. The 1977 endpoint to the range appears the most likely point of final
decision because it is the same year that South Africa was removed from the IAEA,
the same year of Vorster' s conference on the future of South Africa's nuclear
program, and the same year South Africa began construction of the Kalahari test site,
which indicates an active decision to proceed with a nuclear test. As cautioned above,
this decision point does not encompass the entire development process; South Africa
had clearly been considering nuclear weapons for some time beforehand.
The available information on decision points provides us with two questions,
both of which should be able to be answered in the affirmative if the South African
case supports regime security theory. First, if development and discussion of nuclear
weapons occurred over the course of the 1970s, does that coincide with a period of
domestic instability or perceived future domestic instability and a period of increased
international isolation? Second, if 1977 is the most likely final decision point on
nuclear weapons, does that year coincide with a period of heightened domestic
instability and increased international isolation?
94
The answer to the first question is yes. The 1970s was a period of some
domestic unrest in South Africa, and it followed significant upheaval during the 1960s
in response to the Apartheid regime's racial policies. Internationally the Apartheid
regime experienced increasing pressure during the 1970s from the United Nations on
Namibia and from the West on its racial policies. Finally, the 1970s was the period of
heavy Soviet intervention in neighboring countries, including but not limited to the
Angolan civil war during 1975.
As the 1970s progressed the level of domestic unrest certainly increased. The
Soweto riots in 1976, accompanied by the domestic and international response,
increased the perception of vulnerability of the Apartheid regime. This is explained
earlier in this section under the evidence for domestic instability. The answer to the
second question, does 1977 make sense as the likely decision point, is an unqualified
yes. During 1977 the aftermath of Soweto was still playing out both domestically and
in the resulting alienation from Western powers. The year was the beginning of a
period of increased ANC activity as the late seventies saw a jump in ANC recruits. It
was also the year that an attempt was made to expel South Africa from the United
Nations. 126
126 The timing of the South African decision to develop nuclear weapons provides additional evidence
for the importance of domestic threat rather than a conventional military threat from the Soviets. While
there were increased international tensions during the period, they were predominantly (with the
exception of Angola where South Africa was the aggressor) created by the South African response to
significant domestic unrest. The aftermath of the Soweto riots strongly suggests that the final decision
to develop nuclear weapons was a response to the situation at home, and would most likely have
occurred regardless of the intensity of the echoes from abroad.
95
Additional Evidence ofRegime Security Theory
This section will provide additional evidence of regime security logic. It
consists of both an outline of South Africa's stated nuclear strategy and an observation
on the timing of Soviet departure. South Africa's nuclear strategy stated that, in the
event that South Africa was under threat, it would quietly reveal its nuclear arsenal to
the U.S., who would then intervene before it was overrun. The stated reason for U.S.
intervention would be that the U.S. would be disturbed by the potential use of nuclear
weapons in the region and sensitive to the affect on its non-proliferation agenda. 127
While this strategy referred to an amorphous threat to South Africa, it may be implied
that it was in reference to the conventional Soviet threat. However, the strategy would
have been equally ifnot more effective in convincing the U.S. to intervene against a
threat of domestic origin to South Africa. A full Soviet invasion of South Africa would
not have gone unanswered by the West, thus, ensuring U.S. intervention for such a
contingency was unnecessary. However, if South Africa had not possessed nuclear
weapons, the U.S. would have been unlikely to have intervened against domestically
incited instability. U.S. enmity towards the Apartheid regime was such that the need to
ensure nuclear stability would have been the only motivation sufficient for the U.S. to
have intervened in an internal conflict in support of the Apartheid government.
127 Ibid., 15. The full national nuclear strategy included three stages. Stage 1 was nuclear ambiguity as a
deterrent. Stage 2 was revealing nuclear capacity to the U.S. in order to get U.S. intervention. Stage
three consisted of South Africa conducting active nuclear tests and demonstrating its capacity to deliver
warheads to targets, again with the goal of pushing for U.S. intervention.
96
The timeline of Soviet withdrawal from southern Africa provides an ancillary
note in support of the application of regime security theory to the South African case.
As noted above, the Soviets and their allies ceased activities in southern Africa almost
entirely in 1981, with the exception of the Cuban presence in Angola. This partial
withdrawal occurred less than two years after the completion ofthe first rudimentary
nuclear device by South Africa. While it is unclear what level of knowledge the Soviet
government possessed as to the exact developments within the South African program,
it does seem significant that after two decades of low level interventions in southern
Africa (which had produced significant political changes in favor of the Soviets), they
would withdraw so soon after the culmination of the South African weapons program.
This suggests that, not only was the goal of the South African program to undermine
the Soviet will to intervene in Pretoria's affairs, but that the goal was achieved. 128
Discussing Alternative Explanations
Evidence in the South African case can be applied to a number of the
competing theories of nuclear proliferation. The following section will examine these
128 Ifthe South African nuclear weapons program was in part a response to Soviet activity, and the
Soviet's were forced to withdraw upon its success, why did the Soviet Union not withdraw much early
so that South Africa did not develop nuclear weapons in the first place? First, it is unclear that the
Soviet Union was aware their behavior would precipitate nuclear development. To date, the Soviet
experience had been that states developed nuclear weapons in response to nuclear threat. Second, even
with 20/20 hindsight, it is unclear how worried the Soviets were about South Africa developing nuclear
weapons. South African weapons did not pose a threat to the Soviet Union; from the Soviet perspective,
they only precluded further undermining of the Apartheid regime. Southern Africa was a low priority
region for the U.S.S.R. and their interests were to push their goals through proxies where possible and
stop where not. It is very possible that the Soviet's were comfortable with the possibility that the
Apartheid regime would develop nuclear weapons and were simply progressing because it might not
happen, and ifit did it would not be a major setback.
97
relevant alternative proliferation arguments. These arguments include democratic
transition, international economic integration, electoral politics, and prestige.
The argument that democratic transition in a state can reverse nuclear policy
appears to hold in South Africa, though not for the reasons suggested by
proponents. 129 While South Africa had competitive elections during the Apartheid
regime, only a small minority of the population was enfranchised, and thus the
Apartheid regime scored low on the democracy variable. The move to popular rule,
effectively a democratic transition for South Africa, does appear to have reversed the
state's position on nuclear weapons. However, the South African renunciation
decision was made before the transfer of power to a popularly elected government,
rather than after. This timing strongly suggests that a nuclear free South Africa was
not caused by undercutting the domestic position of factions that favored weapons, as
those factions were in power at the decision point, nor any norms or ideals of the new
democracy. Rather, it was caused by the fears of the Apartheid regime, which did not
trust the incoming democratically elected government with nuclear weapons.
At first glance the South African case appears to undermine the argument that
high levels of economic integration lower the chance that a state will develop nuclear
weapons. During the early nuclear period South Africa was a critical trading partner of
the West, exporting uranium and other raw materials to the Western powers, granting
129 Walsh 2001,31.
98
exclusive contracts for its uranium to the U.S. and u.K. 130 Looking at the timing of
South African nuclear development as described above, it is evident that South
Africa's push toward nuclear weapons followed rather than preceded the increased
isolation from the West. This means that during the period of nuclear development the
Apartheid regime had drastically reduced international economic integration as
compared to the early nuclear period. Thus, the South African case does not conflict
with the idea that high levels of integration inhibit nuclear weapons development. The
complementary argument that a low level of integration reduces the costs of a nuclear
weapons program appears to be supported by the South African case.
The argument that politicians will reject or embrace nuclear weapons is
dependent upon what is electorally advantageous at the time is not supported by the
South African case. South Africa's nuclear program remained concealed from the
public until after the decision to dismantle it. Since the program was secret, the
decision to start the nuclear weapons program cannot be based on electoral politics,
nor can the one to end it.
The prestige argument falls flat when applied to South Africa for the same
reason as electoral politics. It is very difficult to explain the case with prestige
arguments because the Apartheid government did not announce its nuclear capacity to
the world until after the program succeeded. If the regime had sought nuclear weapons
for prestige, it would have advertised its success.
130 Snyder 1985,149; Reiss, 173; and Moore 1987,61-71. This information originally comes from
GECD Nuclear Energy Agency and IAEA, Uranium Resources, Production and Demand, Paris:
GECD,1983.
99
Timing of Dismantlement
South African nuclear disarmament began with the election of President de
Klerk on September 14, 1989. Why did South Africa give up the bomb? Two major
factors seem to have impacted the decision of the regime to de-weaponize. First, the
Soviet Union, while not gone, was failing. Second, de Klerk's major policy change
was to move South Africa towards popular rule, ending the Apartheid system. 13l The
following section looks at how these possible reasons for disarmament provide
additional support for regime security theory.
First, both traditional security arguments and regime security theory are partly
supported by the removal of the Soviet Union from the picture. The Soviet threat was
the variable which justified a nuclear route for traditional security hypotheses; it was
also the power which attempted to interfere with South Africa's domestic affairs as
theorized in regime security theory. Any of the three security hypotheses would
predict that removal of the Soviet Union as a threat would remove one of the major
incentives to acquire and hold nuclear weapons. However, it's not clear that removal
of the Soviet Union as a power undermining South African domestic stability would
precipitate disarmament in regime security theory. The perception of isolation would
have remained in effect, and the groups working to undermine domestic stability could
have looked for another arms supplier. Even if another supplier was not forthcoming,
it would have been very difficult for the Apartheid government to determine if that
131 Reiss 1995a, 17.
100
was the case. Nevertheless, the Soviet decline decreased the utility of maintaining the
South African nuclear arsenal.
The second factor that affected the South African disarmament decision was de
Klerk's policy change of moving the country towards popular rule. In 1989 South
Africa was unstable, but not on the verge of collapse. As such, de Klerk's reasons for
promoting a policy of popular rule remain unclear. However, it seems likely that his
decision was a response to what he saw as one of only the available alternatives. De
Klerk could either preside over a government that continued oppression of blacks,
which would result in future domestic unrest, or gradually change the structure of
government while the Apartheid regime was still stable enough to guide and control
the course of that change. 132
That the Apartheid regime willingly gave up control of South Africa has
interesting implications for regime security theory. It suggests that, when faced with
near-certain eventual overthrow or a slow dispersion of power, illegitimate regimes
may opt for the latter over the former. Choosing slow reform makes sense for a
rational regime if that route has a higher aggregate utility in the long term for members
of the regime than the alternative. Giving up power may seem counterintuitive. This is
not always the case. In situations in which the aggregate utility of members of the
regime is higher if power is given up than if it is kept, giving up power is rational. 133
132 Ibid., 22.
133 For example, if staying in power means possible brutal death for the leading members from revolt,
while slowly giving up power means retiring to a long life of relative luxury, it is absolutely rational for
a regime to give up power. The more gradual and controlled the transition process, and the more the
101
The move towards popular rule also lends support to both the traditional
security and regime security hypotheses. For the traditional security hypotheses,
because it was the policies of the Apartheid regime which caused the security threat in
the first place, the removal of those policies would negate the threat. For regime
security theory, the move towards popular rule makes sense as the primary cause of
disarmament on two levels. Domestic resistance would, for the most part, end with the
move towards popular rule - the goal of the domestic resistance groups. Similarly, as
international ire directed at South Africa was a response to the nature of the Apartheid
regime, popular rule would end international isolation and reduce the incentive for
outside interference. With the removal of both domestic and international pressure
resulting from the end of Apartheid, nuclear weapons, and their associated costs,
would simply no longer necessary. In summary, the timing and major events that
surround the disarmament of South Africa provide support for both the traditional
security and regime security hypotheses.
Conclusions
Traditional security explanations cannot adequately explain the Apartheid's
regime's decision to develop nuclear weapons because of the lack of a credible nuclear
or conventional threat against South Africa. In addition, because of the time-line of
South African weapons development compared to that of the international
actors can ensure themselves of safety after they no longer hold the sole reins of power, the more likely
the regime is to consider such a course of action rational.
102
environment and because of the clandestine nature of South Africa's nuclear weapons,
the case is not a good fit for the democratic or prestige arguments.
Regime security theory provides a strong explanation of the South African
case. The application of regime security theory to the case is supported by the
chronologies of the Apartheid regime's foreign and domestic relations and the
regime's nuclear development. The Apartheid regime was increasingly internationally
isolated, had a preexisting raw material base and technical knowledge set, and both
feared and experienced increasing domestic instability directly aided by a foreign
power prior to its development of nuclear weapons.
103
CHAPTER IV
NORTH KOREA
The Kim regime's nuclear aspirations originated from the threat ofU.S.
intervention on the peninsula following the Korean War. As the Cold War dragged on,
North Korea developed its artillery deterrent, and the attention ofthe U.S. waned.
Soon, the prospect ofwar was restricted to the propaganda machines of the North
Korean airwaves, and the Kim regime proceeded with nuclear development at a
gradual pace. 134
The end ofthe Cold War left the Kim regime internationally isolated and the
North Korean economy in a precarious position. As communist governments began to
fall in eastern Europe, North Korea renewed it efforts to develop nuclear weapons. In
this chapter, I will argue that the Kim regime saw acceleration towards nuclear
weapons development as the solution to its worsening domestic and international
situation.
The first section ofthe chapter provides an outline ofNorth Korean domestic
issues and international relations from the end of the Korean War in 1953 to the
present day. The second section provides a chronology of nuclear decision making and
134 The program progressed gradually enough that the U.S. remained, for the most part, disinterested
until the early 1980's. Reiss, 1995a, 233.
104
development in North Korea. The third section presents tests of how regime security
theory applies to the North Korean case, and summarizes the information from the
case as it relates to the relevant alternative explanations in Chapter II: strategic nuclear
security, conventional security, prestige and international economics. The conclusion
of this analysis is that regime security theory provides the best explanation of North
Korean behavior approaching and following the end of the Cold War.
A Historical Outline
The North Korean nuclear program cannot be treated as an isolated issue with
a clear, constant, limited chain of cause and effect. 135 In order to understand the North
Korean case, it is critical to understand the context and nature of the North Korean
state, as well as the actors and interests involved over the course of the last 50 years.
This section briefly describes the domestic situation in North Korea after the Korean
War, and introduces the major international actors and their interests. The timeline
will focus on the period beginning with the decline of Soviet influence in the late
1980s up through the collapse of the Soviet Union and its aftermath for three reasons.
First, it is the period in which the domestic situation of North Korea began to rapidly
deteriorate. Second, it is the period in which the interests of other actors in the system
changed most suddenly. Third, it is the period in which North Korea accelerated its
nuclear program.
135 Ibid., 234.
105
A Context of Domestic Decay
During the late 1950s and early 1960s North Koreans were introduced to the
political philosophy ofjucheism, a uniquely North Korean form of communism
focusing on independence. Outside observers have described it as xenophobic, fascist,
racist, and nationalistic. In the 1960s, under the tenets ofjuche philosophy, the Kim
regime attempted to eradicate all outside influences from North Korea by cutting off
any communication between private citizens and the outside world. The regime
endeavored to become the sole source of information, banning radios capable of
receiving channels other than the official state channel. By the end of the 1960s, North
Korea was arguably the world's most complete police state. 136
Political repression in North Korea was accompanied by incredibly inefficient
economic policies. Unsustainable and ill informed agricultural practices led to
continually decreasing harvests from the 1970s onward. Large scale industrial
projects, individually costing up to 25% of North Korean GNP, ended in complete
failure. Meanwhile, strict adherence to a command economy meant that North Korean
citizens were for decades forbidden to grow private gardens to supplement their food.
The Kim regime's massive military expenditures exacerbated conditions within the
failing economy. The isolationist nature of the Kim regime was directly responsible
for many of its economic woes. The most rudimentary technical advice from the West
could have prevented the calamitous failures of North Korea's large-scale industrial
136 Becker 2005, 64-66.
106
projects and explained why state agricultural policies, such as increasing seed density
in order to increase long run harvest, were doomed to failure. 137 Economic decline
worsened between 1990 and 1992. The fall of the Soviet bloc caused North Korean
trade to decrease in volume by 50%. When combined with domestic factors, this
resulted in an annual GNP contraction of between three and five percent during the
early 1990s. 138
The continually worsening situation reached a tipping point by the mid-1990s,
and North Korea entered a period of famine that resulted in the death of approximately
two million North Koreans - almost ten percent of the population. 139 The famine only
receded after years of Chinese, South Korean and Western aid shipments to North
Korea. Beginning in 1994, China shipped approximately one million tons of wheat
and rice and five hundred thousand tons of heavy fuel to the Kim regime each year. 140
In recent years, there is evidence that the Kim regime has recognized that it
must accept outside assistance in restructuring its economy. North Korea has
experimented with pockets of economic reform, supporting Special Administrative
Regions along both the Chinese and South Korean border. Additionally, in recent
137 Becker 2005, 101-123.
133 Reiss 1995,241.
139 Becker 2005,20-39; and Park 2005.
140 Shambaugh 2003, 47.
107
years Kim Jong Il himself has received lectures from Chinese economists and
reportedly demonstrated sophisticated knowledge of economics. 141
North Korean International Relations
During the Cold War North Korea remained, to varying degrees over time, a
client state of the large communist patrons, China and the U.S.S.R. While these states
did not ensure absolute safety or constant security, they did provide the Kim regime
with important economic support, international military alliances and a shield from
popular international censure for its policies. North Korea maintained a rough parity in
diplomatic allies with its estranged southern counterpart. The Kim regime relied on the
Soviet Union and China, while South Korea was backed by the Western powers.
When the Cold War ended, so did this parity. With the collapse of the U.S.S.R. and
the fall of the communist bloc, North Korea emerged from the Cold War more isolated
than at any other time in its brief history. 142
The Soviet Union, the largest source of advanced military equipment to the
Kim regime, no longer existed. The newly emerged Russia established diplomatic ties
with South Korea, and South Korea was able to successfully pursue reconciliation
with communist China. North Korea was unable to do the same with the United States
141 Ibid., 50.
142 Reiss 1995a, 231.
108
or Japan. North Korea was further isolated by the internal collapse of its eastern
European allies and the absorption ofEast Gennany by West Gennany. 143
The post Cold War isolation ofNorth Korea, when combined with its failing
economy, pressured the Kim regime into increasing international openness and
participating (to varying degrees) in the proliferation control regimes. 144 The most
prevalent non-proliferation negotiations with North Korea are considered
chronologically in the nuclear timeline section below. Negotiations with the Kim
regime on the nuclear issue continued on and off from the 1980s into 2008 without
resolution, and hit an all time low in 2006 when North Korea successfully tested a
low-yield nuclear weapon. 145
The following sections layout the interests ofthe important international
actors involved in the North Korean case, how those interests have changed over time,
and how those actors have affected the behavior ofNorth Korea.
143 Ibid., 231-237.
144 Ibid., 241.
145 While there is some debate as to whether or not the North Korean test was a success or a failure, the
available information is consistent with a successful low yield explosion and within the acceptable yield
range announced by the Kim regime prior to the explosion. For details of the explosion analysis see,
Hui 2007.
109
A Timeline for Nuclear Decisions: North Korea and the IAEA
North Korea's nuclear program began in the mid 1950s with aid from the
Soviet Union and China. l46 In 1964, a nuclear research center was established at
Yongbyon, accompanied by a four megawatt research reactor which came online the
following year. This first Yongbyon reactor was not placed under IAEA safeguards
until 1977, 12 years later. 147
During the early 1980s the United States' attention turned to the North Korean
nuclear program when Washington detected the construction ofa thirty megawatt
reactor at the Yongbyon facility. 148 The reactor, which came online in January of
1986, was a natural-uranium, graphite moderated reactor. This meant two things. First,
the low burn rate of the reactor was well suited to producing high proportions of
uranium 239 in its spent fuel, which decays into plutonium and is therefore useful for
nuclear weapons. Second, the reactor did not use heavy water, a North Korean import,
meaning that it could be operated independently. 149
On September 12th 1985, during the construction ofthe Yongbyon reactor, the
Kim regime, under pressure from the U.S.S.R., signed The Treaty on the Non-
Proliferation ofNuclear Weapons (NPT), but failed to sign the associated IAEA
146 Reiss 1995,233; and Spector 1991.
147 Spector 1991, 9.
148 Reiss 1995a, 233.
149 Reiss 1995a, 234; and Spector 1991, 9.
110
safeguards agreement. I50 In 1986, the same year that the Yongbyon thirty megawatt
reactor went online, North Korea began construction of a two hundred megawatt
Yongbyon reactor and another reactor at Taechon with a capacity between six and
eight hundred megawatts. Ifboth of these new reactors were to become operational
they could produce enough plutonium to build forty-five nuclear weapons each
year. 151
It was two years later, in 1988, that U. S. satellites found solid evidence of
nuclear weapons development at the Yongbyon facility. The United States discovered
a plant capable of reprocessing the spent fuel rods from the graphite moderated
reactors and extracting fissile plutonium, as well as a high explosives test range. In the
spring of 1989, North Korea shut down the thirty megawatt Yongbyon reactor for
around three months, which was enough time for them to remove the nuclear fuel rods
containing the plutonium necessary for reprocessing. The reactor shutdown, combined
with previous events, led the U.S. 1989 National Intelligence Estimate to conclude that
North Korea was attempting to develop a nuclear weapon. 152
Between 1989 and 1991 the Yongbyon facility expanded to over 100
buildings. In June of 1991 South Korea announced that the Kim regime had conducted
a test ofthe non-nuclear components for a nuclear device at the thirty megawatt test
reactor at Yongbyon. North Korea denied the charge, claiming that the thirty
150 Reiss 1995a, 234; and Kerr 2003.
151 Reiss 1995,234.
152 South Korea had already come to this conclusion. Reiss 1995,236.
111
megawatt reactor was part of its civilian energy program, an unlikely claim
considering that the reactor contained no electrical lines or transformers. 153
On December 31, 1991, North and South Korea announced a "Joint
Denuclearization Declaration" which stated that neither would test, manufacture,
produce, introduce, possess, store, deploy or use nuclear weapons, nor possess
reprocessing or uranium enrichment facilities. In January of the following year, after
Seoul and Washington canceled the annual Team Spirit joint military exercise, North
Korea signed the IAEA safeguards agreement and ratified it in April, allowing IAEA
inspectors into North Korea. 154
Through most of 1992 inspectors operated with relative freedom of movement,
but problems developed. The IAEA was concerned with two sites. The first was
"building 500." U.S. satellite intelligence had shown North Korean workers digging
trenches and laying pipe between it and the Yongbyon reprocessing facility. The
second was an older site that the North Koreans had attempted to hide by covering it
with dirt and landscaping the area. The IAEA demanded inspections of these two sites
based on the fear that together they could contain up to 4000 cubic meters ofliquid
nuclear waste. They were denied. 155
In addition to the two problem sites, the IAEA found a discrepancy between
the lab analyses of the plutonium North Korea had admitted to separating in 1990 and
153 Reiss 1995,234-236.
154 Reiss 1995,239.
155 Reiss 1995, 244-246.
112
the North Korean story that they had only separated plutonium once. Isotope analysis
showed that separation had occurred three times. After additional analysis provided by
a smear on a glove box inside the hot cells, inspectors determined that the North
Koreans had conducted plutonium reprocessing in 1989, 1990 and 1991. This meant
that North Korea had lied about its nuclear activities, and that it had accumulated more
plutonium than it had admitted. I56
The IAEA inspectors, after failing to gain access to the restricted sites in late
1992 and early 1993, forwarded the matter to the IAEA board ofgovernors. North
Korea denied everything, including the IAEA evidence and the U.S. satellite photos
presented. The board ofgovernors issued a resolution calling for North Korea to
accede to the inspection requests. North Korea refused the following day, threatening
to take "self-defensive measures" to safeguard its sovereignty.I57
On March 12, immediately after Team Spirit 1993, the Kim regime announced
that it was pulling out of the NPT, citing Team Spirit and the demand for special
inspections by the IAEA The response of the international community was to move
back from insisting on special inspections to just focusing on keeping North Korea in
the NPT. Over the course of the rest of year the IAEA attempts to gain access to the
restricted sites were met with a series of obstructions and threats by the North
Koreans. On November 1, 1993 the U.N. General Assembly passed a resolution
156 Ibid., 246.
157 North Korea actually cited the movie "Star Wars" as demonstrating that the United States was
capable of"high tech sleight ofhand" and had faked the satellite shots. Reiss 1995,246-250.
113
criticizing North Korea's failure to adhere to safeguards. North Korea was the only
dissenting vote. 158
By signing the Agreed Framework in October of 1994, North Korea agreed to
freeze its nuclear activities in exchange for two light water nuclear reactors from the
United States and heavy fuel shipments until the completion ofthe reactors to offset
the energy lost by leaving inert the graphite moderated reactors that North Korea had
in operation. The agreement also provided for inspections of the new, proliferation
resistant reactors once operational. 159
After several years ofrelative quiet, on August 31, 1998, North Korea
launched a Taepo-Dong 1 missile over the Sea of Japan. The Taepo-Dong 1 test is
argued to have demonstrated progress towards an ICBM delivery system, and,
therefore, evidence of renewed nuclear aspirations. 160 161
In 2002 the U.S. discovered that the Kim regime was engaged in an additional,
covert nuclear program independent of the one at the Yongbyon facility, and
158 Reiss 1995,251-288.
]59 Unlike graphite moderated reactors, light water reactors are proliferation resistant because the spent
fuel byproducts are far more difficult to weaponize. Snyder 2000, 164; Mearsheimer, 1994; and Reiss
1995,288.
160 Perry 2006. It is also possible that the intention of the test was to elicit diplomatic attention from the
U.S., and that the technology was geared toward missile sales abroad. However, the size of the
subsequent 2006 North Korean nuclear test also suggests an attempt to produce a bomb with a viable
intercontinental delivery vehicle.
161 Despite the tensions stoked by the missile test, by 1999, near the end of the Clinton administration,
talks with North Korea had progressed in a positive direction. The Kim regime seemed interested in
engagement and movement toward nonnalization of relations. It appeared that the current level of U.S.
engagement and investment gave the North Koreans the assurances they needed. The diplomatic
environment with North Korea changed in early 2000, as the incoming Bush administration effectively
cut off the dialogue with the Kim regime and proceeded to take a year and a half to formulate a new
policy before re-engaging North Korea. This new policy never came to fruition. Perry 2006.
114
announced that North Korea had admitted in a high level meeting to having a
clandestine highly enriched uranium (HEU) program, a claim disputed by the North
Koreans. 162 In response to U.S. evidence of a clandestine HEU program, in November
2002, the Korean Peninsula Energy Development Organization (KEDO) suspended
the heavy fuel shipments to North Korea that had been promised under the Agreed
Framework. Following the suspension of heavy fuel shipments by KEDO, North
Korea restarted its graphite moderated Yongbyon reactor. The United States
confirmed the restart in early 2003 around the same time that North Korea both
expelled IAEA inspectors and began reprocessing spent fuel rods. On April 23, 2003,
North Korea publicly announced it possessed nuclear weapons. 163
Between 2003 and 2005, six party talks were conducted between the regional
powers and the two Koreas. These talks appeared to finally produce an agreement in
August 2005. North Korea agreed to give up its nuclear weapons program, and the
United States promised to not initiate military force to overthrow the North Korea
regime and to restart the light water reactor program. However, the agreement was
short lived; both the United States and North Korea insisted that the other be the first
to take a substantive move forward. 164 On October 9,2006, North Korea officially
162 Perry 2006; Kerr 2003; and Abromowitz 2003. Bruce Cummings disputes this date, claiming that the
Clinton administration knew about this program in 1997 or 1998 and it was not outed until 2002 for
political reasons. Cummings 2004, 60.
163 Shambaugh 2003, 47; Kerr 2003; and Perry, 2006. Cummings offers an alternative explanation of
the HEU program, claiming that the centrifuges used to create HEU are the same equipment used to
provide LEU for the North's coming light water reactor facilities, and NK has clearly expressed strong
interest in independence, and that because of this the existence of a "second nuclear program" is
unclear. Cummings 2004.
164 Park 2005, 76; and Perry, 2006.
115
declared the completion ofa successful nuclear test. 165 On May 25, 2009 North Korea
declared a second successful nuclear test. 166
There are three conclusions that should be drawn from the timetable of the
North Korean nuclear program. First, the available information suggests a broad range
of time for the original decision to produce nuclear weapons. The development of
North Korean nuclear weapons took place over half a century. The lack of information
about the internal decision making process ofNorth Korea leaves us without a way of
nailing down the date. However, the likely time period for the initial decision to
produce nuclear weapons is between the construction ofthe thirty megawatt
Yongbyon facility in 1985, and the start of construction on the Yongbyon fuel rod
reprocessing plant in 1988. 167
Second, it is likely that the decision was made in stages. For North Korea,
weapons development was likely an ultimate goal from the beginning of its nuclear
program. However, the point at which they chose to accelerate the program by
building both reprocessing facilities and beginning new, larger power plants occurred
in the late 1980s. The physical evidence suggests the late 1980s as the point at which
North Korea stepped up the clock on its nuclear weapons program.
Third, after receiving help from the U.S.S.R. and China as early as the mid
1950s and having a thirty megawatt nuclear reactor running by 1986, North Korea did
165 Hui 2007.
166 Sang-Hun 2009.
167 Reiss 1995,234; and Spector 1991, 10.
116
not produce a nuclear weapon until the late 1990s at the earliest, and 2006 at the latest.
This timeline, like the South African case, belies the idea that any country can simply
throw enough money at its nuclear weapons program and have it materialize
overnight. Like South Africa, even after North Korea had the necessary facilities for
plutonium reprocessing in 1988, it took many years to produce a nuclear weapon. Had
North Korea not had an existing, longstanding domestic nuclear program as it entered
post-Cold War isolation, its decision calculus would likely have been entirely
different.
Analysis of the North Korean Case
This section will summarize the conclusions resulting from testing the
arguments explained in Chapter IT against the North Korean case in three main points.
First, the traditional security arguments do not fully explain the acquisition of nuclear
weapons by North Korea. Second, analysis of the North Korean case strongly supports
the premise that regime security theory best explains North Korean behavior near the
end of the Cold War and during the post Cold War period. Third, neither the
international economics argument nor the systemic prestige argument adequately
explains the North Korean case. Support for these conclusions will be developed in
three subsections. These will:
• Test the strategic nuclear threat and conventional security threat arguments;
• Apply regime security theory to the North Korean case; and
117
• Attempt to apply the international economic and international prestige
arguments to the North Korean case.
Traditional Threat Arguments
The Strategic Nuclear Threat against North Korea
North Korea had a legitimate fear of U.S. nuclear attack during and after the
Korean War. The United States publicly discussed using nuclear weapons against the
Kim regime three times during the Korean War - once at the beginning, when U.S.
forces were pushed back to Pusan, again, right after the Chinese entered the war, and
again, right after Eisenhower took office. After the Korean War the United States
adopted a doctrine of "massive retaliation" which included a corollary that stated its
willingness to use nuclear weapons to resolve regional issues. The strategic nuclear
threat hypothesis provides a strong explanation of North Korean actions in the era of
the Korean War. It is conceivable, even likely, that the U.S. nuclear threat prompted
the North Korean artillery deterrent and the extensive North Korean bunker system. 168
While it is undeniable that the United States did give North Korea cause to
have strategic nuclear fears, these fears do not explain North Korean behavior in the
late and post-Cold War eras. By the late 1980s the situation both in North Korea and
around the world had changed drastically. First, North Korea had successfully placed
168 Perry 2006, 81; and Spector 1991, 9-10.
118
artillery batteries that could level Seoul, a city of 12 million, at a moment's notice. As
South Korea had by that point developed into a key Western ally, the North Korean
artillery alone was sufficient to deter the United States from seriously contemplating
nuclear attack against the Kim regime. Second, even if the Kim regime was not
convinced by its own deterrent capacity, nuclear weapons had not been used in war for
almost fifty years, and the international cost to U.S. interests of launching a nuclear
attack against North Korea vastly outweighed any potential benefits. There are other
reasons why a nuclear attack by the United States was incredibly unlikely, including
rising international and domestic norm against nuclear weapons use and the U.S.
domestic political cost. However, even if we assume that North Korea discounted
these rationales completely or had no knowledge of them, it would be absurd to claim
that the Kim regime was so out of touch with the climate of international relations as
to not understand the costs to the U.S. of such an attack. By the time that North Korea
made the decision to build the reprocessing plant at the Yongbyon facility, the chances
of a United States nuclear attack, realistic in the 1950s, was incredibly remote. 169
Additional problems for a strategic threat explanation of North Korea nuclear
development arose in 1991 when President Bush announced the unilateral withdrawal
of ground and sea launched nuclear weapons from overseas. The remaining U.S.
nuclear weapons on the Korean Peninsula, those designed for aerial delivery, were
removed later that year. By the end of 1991, well before North Korea had completed
169 Robinson, 2004.
119
any nuclear weapons, the United States had unilaterally and without precondition
removed all U.S. nuclear weapons from the Korean peninsula. l7O
The information available in the North Korean case undermines the strategic
threat theory's applicability to that case. The evidence shows that there was an initial
strategic nuclear threat against North Korea in the 1950s, but no similar threat during
the expansion of the weapons program beginning in the early 1980s.
The Overwhelming Conventional Threat against North Korea
The overwhelming conventional threat argument loses credence over time in
the North Korean case. South Korea has not posed a conventional threat to North
Korea since the Korean War. In the time period following the Korean War, North
Korea possessed a far larger military than South Korea, and, long before modem
technology eclipsed North Korea's numerical advantage, the Kim regime had firmly
established the artillery batteries aimed at Seoul. Further, the South Korean military
composition was designed to be uniquely suited for defense. At no time since the
Korean War South Korea posed a conventional threat to North Korea. I71
The conventional military threat theory has some support in the period
following the Korean War when considering the possibility ofconventional attack by
the United States. However, the artillery deterrent stifles the feasibility of conventional
170 Reiss 1995a, 237.
171 Beldecos 1995; and O'Hanlon 1998.
120
u.s. attack in the same way it does nuclear attack, and so the overwhelming
conventional threat argument cannot explain North Korean nuclear weapons
development following the establishment of that deterrent.
Regime Security Theory
This section summarizes the application of available evidence regarding North
Korea's domestic and international context to the timeline of the regime's nuclear
weapons program. It then discusses the results of testing whether that evidence
supports the conclusion that regime security theory provides a strong explanation of
North Korea's acceleration in nuclear development and eventual weapons program.
The inner dialogue of the Kim regime is not public knowledge. However, I
argue that it is likely that the Kim regime understands that, if it possesses a nuclear
deterrent, regime collapse is no longer a viable policy preference for the United States.
I also argue that the regime is aware ofthe potential cost ofthe collapse of a nuclear
state. Further, it understands that, in an effort to prevent a catastrophic collapse, the
United States has every incentive to cooperate with China in providing enough aid to
keep North Korea afloat. The regime understands that the U.S. will be compelled to do
this regardless of the level of North Korean domestic decay or how diplomatically
isolated North Korea remains.
The North Korean case supports regime security theory; however the logic is
slightly different than that applied to South Africa. South Africa sought to change
121
great power preference in order to coerce the Soviet Union to end interference aimed
at destabilizing the state. North Korea sought to change great power preference in
order to coerce the United States to aid in propping up a failing state instead of
supporting its eventual collapse.
Because we already know that North Korea did possess the capacity to develop
and test nuclear weapons, in order for regime security theory to provide an adequate
explanation of the North Korean case three additional things must be true.
• The Kim regime must be dealing with a period of increasing domestic
instability or a strong potential of future increasing domestic instability.
• The Kim regime must perceive that it is experiencing a period of international
isolation, including pressure from at least one major pole for domestic reform
and fear that one or more of the major powers is considering interfering with
its domestic affairs.
• The timing of the decision to build nuclear weapons should correspond to the
changes in the independent variables.
Evidence ofNorth Korean Domestic Instability
Due to the isolationist nature of the Kim regime and the continuing lack of
reliable information coming from North Korea, most of the information about
domestic instability in North Korea necessarily comes from refugees and defectors.
This information cannot be fully trusted. Refugees have a strong tendency to
122
exaggerate the direness of the situation from which they escaped, and political
defectors have a clear interest in portraying the worst possible picture of the state from
which they defected. What we do know is that the North Korean economy and food
infrastructure were on the verge of collapse by, at the latest, the early 1990s. This
crisis was decades in the making. North Korean harvests began declining in the 1970s
with the beginning ofjuche agricultural practices. While it is unclear that the regime
was aware of the full extent of the agricultural decay, as its officials were often fearful
of reporting shortfalls, the famine developed out of a steady, three decade decline in
harvest caused by over-planting, lack of crop rotation, single crop planting and
extensive hillside planting which eroded the hills so that the crops were washed away
by the rain. 172 The aggregate economic situation, the mere existence of high level
political defectors from North Korea, the multitude and similarity of refugee accounts
emerging from the Hermit Kingdom, and the increasing frequency of defectors and
accounts over time, give both sources some credence. While any given individual
refugee cannot be counted on as a credible source, in numbers, the accounts become
more and more difficult to dismiss. 173
During the 1990s, North Korean refugee stories were initially dismissed to a
large degree, despite their numbers, because so many sounded extreme beyond belief.
Of the thousands of refugees that fled the country during the famine, many reported
I72 Becker 2005, 101-123.
173 The highest ranking North Korean defector to date, Hwang Jang yop, the creator of the Juche
ideology and Kim Ii Sung's top advisor on ideology, defected to South Korea in 1997. Hwang claims
that Kim Jong II has perverted Stalinism from a system where the dictator served the people to a feudal
system where the people served the dictator. Becker 2005,64-66.
123
that the famine was so severe that some of the population had resorted to acts of
cannibalism. There are multiple accounts of citizens being shot after they were caught
luring street children into their homes or shops, killing them and then mixing the meat
with pork to sell at the market. 174
Refugee reports detailing the results offood shortage also include accounts of
increasing unrest as the food ran out. Desperation for food led to an increase in violent
attacks, protests, strikes, uprisings, sabotage ofgovernment buildings and the killings
of officials and their families by the starving population. This unrest tells us much
about the condition ofthe North Korean state. Conditions during the famine were
unstable to the point that citizens participated in public uprisings in a country where
the wrong political discourse could result in being sent to a death camp. 175 Kim Jong II
responded to domestic dissent with waves of countrywide purges and summary public
executions. In addition, beginning in 1995, once food shortages became severe enough
to be felt at the higher levels of society, Kim Jong II channeled much of the food aid
coming into the country to the upper class and the military in order to maintain their
loyalty. The Kim regime's efforts, combined with continued food and fuel aid
shipments, successfully kept a lid on the instability resulting from the 1990s famine.
It is apparent, however, that the Kim regime's efforts to maintain stability within the
country have at many times been dependent on foreign largess. 176
174 Becker 2005,29.
175 Ibid., 36.
176 Park 2005~ and Becker 2005,37,64-66.
124
The idea that foreign aid is a necessity for the North Korean regime to avoid
collapse is supported by concerns expressed within the Chinese government. The
Chinese are so concerned with the stability ofNorth Korea that any Chinese action
meant to drive home a point, such as temporarily shutting down an oil pipeline, must
be gauged by how it will affect the limited domestic stability of North Korea. 177
While it bears repeating that much of the evidence of regime instability coming
out of North Korea is not completely reliable, the unmistakable overall picture is of a
country on the brink of collapse. The available information supports the assertion that
there has been a period of increasing domestic instability in North Korea, rooted as far
back as the beginnings of the Kim regime and hitting critical levels in the 1990s.
A Perception ofInternational Isolation
During the Cold War North Korea was strongly aligned with the communist
bloc. The breakup ofthe communist bloc and eventual fall of the Soviet Union was
diplomatically disastrous for North Korea. Well before the Soviet Union finally fell,
North Korea's patron had grown increasingly distant and less cooperative. Following
the fall of communism in eastern Europe and the disappearance of the Soviet Union,
North Korea became more isolated than at any previous point in its history. This era of
increased isolation was characterized by the self-perceived ideological victory ofthe
West, the very real decrease in recipients of Western ire, and the decrease in North
177 Park 2005.
125
Korea's allies. The dearth of allies and patrons, combined with increasing international
attention to the North Korean domestic situation, meant a very frightening landscape
for the Kim regime as it entered the 1990s. General world opinion of the Kim regime
was unsympathetic. On November 1, 1993, the U.N. General Assembly passed a
resolution criticizing North Korea's failure to adhere to IAEA safeguards. North
Korea stood alone, the only dissenting vote. 178
This section will present evidence of increasing regime isolation over the
course of the Kim regime, and then check for the specific conditions of the perception
of isolation variable. Evidence will be subdivided into four categories, and will consist
ofNorth Korea's relationship with the United States, South Korea, China and the
Soviet Union. The specific conditions of the isolation variable that will be checked
are: first, that the Kim regime feared domestic interference by one or more of the great
powers; and second, that, in addition to being isolated internationally, the Kim regime
perceived that it was broadly isolated internationally.
The United States
After the United States and the U.S.S.R. divided the Korean Peninsula
following World War II, the U.S. retained control of South Korea and set up a
government sympathetic to the West. The United States then proceeded to participate
in the Korean civil war on the side of its South Korean allies. U.S. involvement in the
178 Reiss 1995a, 243.
126
Korean War, subsequent entrenchment in South Korea, support of the South Korean
government, and threats ofnuclear attack against the Kim regime determinate1y
established the United States as the primary great power threat to North Korea. 179
The United State's interests in Korea shifted significantly during the period
following the Korean War. First, as the threat of renewed war on the peninsula
decreased, US. attention shifted to combating the spread ofcommunism elsewhere,
and its interests in Korea shifted to maintenance of the status quo - an isolated and
weak North and an economically prosperous, pro-West South. Second, beginning in
the 1980s, the US.'s main interest in North Korea became preventing the Kim regime
from successfully building nuclear weapons. In addition to these two major shifts, the
United States joined South Korea and the world community in condemnation of the
Kim regime's human rights record and the domestic situation in North Korea. It also
condemned North Korean activities against South Korea, and, following a North
Korean bombing of South Korea in 1987, officially added North Korea to the US. list
of state sponsors ofterrorism in 1988. In the late and post-Cold War world it became
feasible for the US. to pressure other countries to cut aid and impose sanctions in a
way that was not possible during the Cold War. The US. had been constantly applying
pressure against North Korea since the early 1980s when their attention returned to the
North Korean nuclear program. Fear ofUS. intervention within North Korean borders
replaced fear ofan actual military strike. 180
179 . During the early 1950's the United States threatened or publicly discussed the use of nuclear
weapons against North Korea on no less than three occasions. Perry 2006,81.
180 Sciolino 1988; and Spector 1991,8.
127
While it is impossible to be certain of the perspective of the North Korean
government towards the U.S., it is safe to say that, even during times of low tension, it
has feared u.s. interventionism on the peninsula, either through or on behalf of South
Korea. This is supported by repeated requests for security assurances from the U.S. as
part of or as a precondition to multiple bargains.
The Team Spirit joint military exercise between the United States and South
Korea exemplifies the perspective of North Korea towards the great power's
relationship with its southern neighbor. North Korea's distaste for military cooperation
between the two countries has allowed the U.S. to use Team Spirit to pressure North
Korea on weapons inspections issues. In 1993 North Korea responded to threats to
resume Team Spirit, which had been canceled in 1992, by threatening to leave the
NPT in response to what it saw as United States interference with North Korean
internal affairs. North Korea claimed that it would be forced to leave the NPT until
"the U.S. stopped its nuclear threats against North Korea" and the IAEA returned to
"its principle of independence and impartiality." While this threat was clear posturing,
as the U.S. had not threatened North Korea with nuclear weapons and there was no
evidence that the IAEA was biased or coerced, it is evident that the Kim regime was
threatened by the prospect of U.S. pressure coercing it into accepting internal
inspections. lSI
181 Reiss 1995, 252.
128
South Korea
South Korea has had a conflicted relationship with its northern counterpart.
North Korea represents the most immediate military threat to South Korea. It has built
a massive artillery deterrent aimed at Seoul and committed numerous acts of
aggression against South Korea since the Korean War. 182 However, the division of
Korea split many families, which left many South Koreans sympathetic to the living
conditions of their northern counterparts, many of whom are literally brothers and
sisters. As a result of these ties, South Korea sent and supported food and energy aid,
as well as repeatedly tried to move toward reconciliation with the North despite
repeated limited engagements initiated by the North. 183
The relationship between North and South Korea has changed over time in
response to the vastly widening gap between the wealth and productive capacity of the
two countries. After the Korean War, South Korea was concerned primarily with
182 These acts include, but are not limited to: a suspected attempt to sabotage the Wolsong nuclear
reactor in 1983, an assassination attempt in 1983 against Chun Doo Hwan which killed members of his
entourage, including four cabinet level officials, and a bombing of a Korean Airlines passenger jet,
killing everyone onboard in 1987. Spector 1991, 8.
183 Two major moves toward reconciliation occurred during the end-of-Cold-War period. The first was
the Sunshine Policy initiated by Kim Dae Jung in 1998. The second was a series of steps toward
reconciliation on nuclear and other issues taken in the early 1990s, spurred by the Sunshine policy and
North Korea's new openness to international negotiations. In 1991, South Korea offered the North a bi-
lateral, intensive inspections regime to be implemented after U.S. nukes were removed from South
Korea. In November of 1991, President Roh Tae Woo unilaterally declared that Seoul would use
nuclear energy for peaceful purposes only and asked Pyongyang to do the same. In December, the two
parties signed a reconciliation agreement that omitted the nuclear issue; however, later that month they
produced the "Joint Declaration on the Denuclearization of the Korean Peninsula." While these moves
towards reconciliation and denuclearization were ultimately unsuccessful they are important evidence
that South Korea (and, to some degree, North Korea), felt that reconciliation was within their interests.
Reiss 1995a, 237-239.
129
deterrence of the North. Over time, the state ofthe North Korean economy increased
South Korean humanitarian interest and added the possibility of the collapse of the
Kim regime to South Korea's concerns. In addition to concern for what would happen
from a military standpoint, the South Korean economy, while relatively strong, would
suffer significantly under the burden of absorbing the millions of people who would
need help should the Kim regime collapse. For both ofthese reasons, as well as
humanitarian ones, while the South Korean government has repeatedly condemned
actions of the Kim regime, the national interests produce a policy preference for a
gradual, controlled opening ofNorth Korea and reform of its economy, a process that
would end in reconciliation, but not produce the same level ofhardship in South
Korea. 184
The separation in the South Korean mind between the Kim regime and the
North Korean people has allowed the Kim regime to exploit the country that is its
competition for control of the peninsula during the past few decades. North Korea has
raised tensions through limited attacks and spy rings and pushed for concessions,
while betting that South Korea is unwilling to go to war and will continue to provide
humanitarian relief regardless ofNorth Korean military actions. 185
184 Policy stance toward North Korea, hard or soft, is one of the most important domestic political issues
in South Korea.
185 South Korea uncovered a North Korean spy ring in 1993. Reiss 1995a, 244-245.
130
China
China's interest in North Korea has been defined by their shared border. In the
1950s this meant fighting in the Korean War in order to prevent a bastion ofcapitalism
from taking hold. Through the Cold War this meant acting as a patron ofthe Kim
regime for the same reason. In the post-Cold War world, proximity created a new
primary interest. Regional instability, influx ofrefugees, and the possible nuclear
repercussions that could be caused by the collapse ofthe Kim regime changed the
defining interest ofChina in North Korea to regime stability.I86
Believing the Kim regime's policies are unsustainable, Beijing has been the
strongest external advocate ofoverhauling North Korea's juche system. 187 China has
extended food and energy aid to the ailing regime and allowed significant refugee
influx from North Korea, despite the cost to northeastern China. This is most likely
because China viewed the influx as a safety valve for the Hermit kingdom. 188
China's relationship with the Kim regime is not one dimensional. Like the
u.s., China seeks and has sought to become the dominant power on the Korean
peninsula. In order to reinforce its role as the key actor on the peninsula, China has
both prevented the international community from undertaking measures that it
considers counterproductive, and sided against North Korea when it feels its
186 Shambaugh 2003, 47.
187 Chinese critics compare North Korea to Maoist China and argue that the only option is reform. Ibid.,
46-50.
188 Ibid., 49.
131
problematic client state is out of line. China has also maintained good economic
relations with South Korea since their reconciliation in the early 1990s. Today, South
Korean Chinese relations dwarfNorth Korea's relations with its closest "ally". 189
China, like the U.S., would prefer that North Korea not possess nuclear
weapons. However, unlike the U. S., this preference is part of its broader goal of
regional stability, not an individual top priority. The current Chinese perspective is
that the North Korean nuclear program is more bark than bite; they view it as a
bargaining chip used to gain concessions rather than as a security threat. 190
Chinese incentives toward stability and against the costs of regime collapse
have meant that, in the past, North Korea has had leeway to act out toward China in a
manner that belies their relative sizes and location. In the post-Cold War era North
Korea continued its nuclear weapons development and belligerent actions toward
South Korea over Chinese objections. However, China's demonstrated willingness to
cut off aid supplies when the Kim regime gets out of hand has acted as a restraint on
North Korean brinksmanship behavior against South Korea and the West.
China did block sanctions against North Korea twice during 1993, saying that
such efforts would only complicate matters. Since 1994 China has said that it views
coercive strategies as counterproductive in gaining North Korean cooperation, fearing
that such strategies could prompt the regime to take desperate actions. 191
189 Shambaugh 2003,51; Reiss 1995a, 243; and Spector 1991, 10.
190 Shambaugh 2003, 52-55; and Park 2005.
191 Reiss 1995a, 250; and Shambaugh 2003, 55.
132
In the early 1990s, when Beijing promised Pyongyang that Chinese
reconciliation with South Korea would not hurt relations with North Korea, China
added a caveat: China would continue to guarantee the existence ofNorth Korea as
long as North Korea abandoned its nuclear weapons program. 192 In early 2003, China
warned North Korea that renewed provocations against the U. S. could strain Chinese-
North Korean relations. To reinforce that point, China shut off an oil pipeline to North
Korea for three days in March, officially citing technical problems. According to a
Western diplomat, the Kim regime got the message - behave. 193
It is fair to say that while China does not wish North Korea to fail, it is not
pleased with the conduct of the North Korean government. The Chinese preference for
regime survival should not be misconstrued as an affinity for the Kim Regime. Beijing
has considered North Korea a "headache" since the 1950s, and, at times, Chinese
officials speak of the North Korean situation with despair and disdain. Finally, while
China claims to want to work against the continued isolation ofthe Kim regime, its
punitive attitude and veiled threats towards the Kim regime indicate that China's
preferences lie in nothing more than dealing with a problem neighbor however
feasible. 194
192 Reiss 19958, 243.
193 Park 2005.
194 Chinese critics have complained about North Korea for the following reasons: the cult of personality
around the Kim dynasty, the Stalinist security state, the command economy, the poverty, the use of
scarce resources by the military, the mass mobilization techniques, the condescension ofNorth Korean
officials to their Chinese counterparts, and the autarkic, xenophobic paranoia. An additional indication
of tense relations is the rarity and superficiality of military officer exchanges between the two
govermnents. Shambaugh 2003, 46-47.
133
Russia
In 1961 the Soviet Union signed a defense pact with the Kim regime which
called for automatic intervention should North Korea be invaded. During the Cold
War period the Soviet Union was an ideological ally of North Korea, but often at odds
with the Kim regime. As early as 1965 there was significant tension in the client-
patron relationship. North Korea worried that the Soviets would sell them out, and the
Soviets accused the Kim regime of being Chinese pawns. The Kim regime's emphasis
of independence from Soviet or Chinese "control," a product ofthejuche philosophy,
caused additional tensions with its Soviet patron. During the late Soviet period,
Moscow repeatedly pressured Pyongyang into complying with non-proliferation
guidelines. In 1990 Soviet Foreign Minister Shevardnadze pressured Pyongyang to
sign the IAEA safeguards agreement, which produced an especially hostile
atmosphere between the two countries. When the Soviets moved toward improving
relations with Seoul, the Kim regime's response to the overtures was to threaten the
Soviets, claiming it would embark on a nuclear development program and would
recognize Japanese claims to four U.S.S.R. occupied islands.
The North Korean threat lacked teeth. The Kim regime already had an active
nuclear program, and recognition of Japanese claims to the islands would not affect
the status of the islands in the slightest. Nevertheless, Moscow responded harshly. The
U.S.S.R. both threatened to cut off all nuclear cooperation if the North Korean nuclear
134
facilities were not placed under IAEA safeguards and pointedly ignored the North
Korean threat, establishing diplomatic relations with Seoul within the month. 195
What relations did exist between Moscow and Pyongyang died with the Soviet
Union. Russia, no longer concerned with propping up an ideological ally, continued
improving the relations already established with South Korea by the collapsing Soviet
Union. Russia allowed South Korea entry into the UN in 1991, and both withdrew its
promise of military aid and renewed its moratorium on nuclear cooperation with North
Korea in 1992. 196 Post-Cold War Russia washed its hands of the problems of the
Hermit Kingdom. It took policy leads from the U.S. and China on the remaining
nuclear issues. The new Russia remained open to continued commercial ventures with
the Kim regime regardless of other issues, but that has been the extent of its
cooperation to date. 197
In November 1992, the new Russian government frankly abandoned any
commitments to North Korean defense. Boris Yeltsin, on a visit to Seoul, declared that
Russia would halt all military assistance to North Korea, and stated that the 1961
defense pact, which called for automatic intervention in case North Korea was
invaded, needed to be "either canceled or drastically revised," adding that Russia did
"not intend to render such military assistance". Yeltsin also announced that Russia
would not resume any nuclear cooperation with the Kim regime until they fully
195 Reiss 1995,237.
196 Reiss 1995,237-243.
197 Park 2005.
135
implemented IAEA safeguards, a statement denounced by Pyongyang as "foolish and
despicable" 198
Checking the Conditions ofIsolation
North Korea clearly feared domestic interference from the United States
following the inception of the North Korean nuclear program. However, it is less
obvious that it was actively concerned with domestic interference before beginning
along the path of nuclear weapons development. Before the acceleration of its nuclear
program North Korea's concerns about U.S intervention were twofold. First, it feared
that the United States would both aid in efforts to undermine the regime by supporting
the erosion of the wall blocking information getting into North Korea through the
mechanisms of South Korea and individuals operating within the North. Second, it
feared that because of the U.S. preference for regime overthrow, the U.S. would use
international pressure to undermine the amount of aid North Korea would receive in
the coming food shortage, a shortage most certainly long predicted within North
Korea. This fear came true as the United States did use cutting off aid and pressuring
other countries to do the same as a coercive diplomatic tool against North Korea
during the early 1990s. Not only did North Korea fear U.S pressure towards
destabilization, (both from within, through subversion, and without, through
198 Reiss 1995,243.
136
diplomacy) but it also had a strong incentive to coerce positive u.s. interference,
needing food imports to stabilize the threat of famine.
It is also clear that the Kim regime understood its international isolation. The
Kim regime's two main allies, the Soviet Union and China, both publicly expressed
distaste for the regime's particular policies early in the Cold War period. The repeated
attempts of the Kim regime to play off the Soviets and Chinese against each other
demonstrated North Korea understood how tenuous its relationship was with both
powers. Further, the movement of both the Soviet Union and China towards economic
relations with South Korea prior to the end of the Cold War was a clear signal to North
Korea that its relatively severe isolation was increasing. The regime's loud objections
and threats towards its patrons attest to its understanding of this shift.
Checking the Timing a/Nuclear Development
The lack ofreliable information coming out ofNorth Korea makes checking
the timing of the regime's nuclear decisions against regime instability and perceived
isolation inexact. The North Korean nuclear program began in the 1950s at a point
when the Kim regime was relatively stable and possessed a solid patron in the Soviet
Union. The nuclear threat from the United States during this period meant that the
strategic security threat explanation could account for all North Korean nuclear
developments. Therefore, regime security theory does not explain the early decision to
embark on a nuclear development program. The likely timetable for the acceleration of
137
the North Korean nuclear weapons program is the late 1980s, in the period leading up
to the fall of the Soviet Union and the Eastern bloc.
Did North Korean nuclear development correspond with a period of increasing
international isolation and a period of domestic instability or future projected domestic
instability? This question requires refinement into four separate questions. First, was
North Korea facing increased domestic instability during the 1980s? Unfortunately, a
reliable answer to that question is unavailable. Second, did North Korea expect to face
future increased domestic instability in the late 1980s? I argue that it is likely that the
Kim regime was aware of the coming period of increased domestic instability. Third,
was North Korea facing increased international isolation during the 1980s? Fourth, did
North Korea expect to face future increased international isolation in the late 1980s? I
argue that North Korea's international isolation remained at a constant, high level
throughout the 1980s, and that, with the beginning of the fall of Soviet satellites in
eastern Europe, the Kim regime understood that its relations were likely to further
deteriorate.
The primary factor that caused domestic instability in North Korea was the
collapse ofthe agricultural economy. The Kim regime may have been unaware of the
full extent of its economic downturn at the beginning. During the late 1970s Kim 11
Sung was receiving news from his subordinates of higher than actual harvest results.
However, based on its behavior, described below, it is unlikely that the Kim regime
was ignorant ofthe continual worsening ofthe agricultural situation in its own country
over the following twenty years. The Kim regime oversaw the large scale industrial
138
projects that failed to increase food or energy production. The regime also pioneered
government programs to supplement harvests with "food substitutes." Perhaps most
telling, the Kim regime stopped publishing agricultural statistics in 1978 after the
predicted harvest came up twenty to thirty percent short. Clearly, by the late 1970s, the
government was aware that its agricultural policies were not providing the intended
results; instead North Korea faced a stark reality of an ever decreasing food supply. 199
It is clear that the Kim regime understood, for the most part, the severity of the
food production issues in North Korea earlier than 1980. With that conclusion in mind,
it seems unlikely that by the late 1980s the regime was unaware that the country was
at the brink of famine. While, again, the limited information available makes a
concrete statement impossible, it is far more likely that the Kim regime feared
increasing future domestic instability caused by food shortages in the late 1980s than
that it did not.
North Korea's level of international isolation has always been relatively high.
The Kim regime was isolated from the developed West, and aligned with the Soviet
block during the Cold War. Directly following the Korean War the Soviet Union was
an active patron, providing a mutual defense pact and supplying the Kim regime with
nuclear power, technology and training. The Soviet relationship quickly degraded. As
early as the mid 1960s the Russians had taken a cool tone toward North Korea who
had repeatedly attempted to play of the Soviet Union and China against each other in
bids for patronage. From the late 1960s onward North Korea's relations with its patron
199 Becker 2005, 108-109.
139
states were cool and antagonistic to varying degrees, though they remained under the
Soviet nuclear umbrella.
Previous to the fall of the Soviet Union, as the Russians began repairing
relations with South Korea over North Korean objections, it became clear that the
alliance with North Korea had degraded to a formality. As the Eastern bloc began to
fall, all Soviet support for North Korea dried up, and the Russian Federation quickly
rescinded any commitments on the peninsula. China has retained a functional
relationship with the Kim regime, though it has been strained since the mid 1960s
onward. The People's Republic's primary interest, at this point, is keeping the ailing
regime from disgorging too many refugees into northern China. While North Korean
international relations did plummet during the late 1980s, it is not clear that this was
foreseeable by the regime. Nevertheless, the constant isolation of North Korea, a
problematic ally to the East and a pariah state to the West, means that it met the
conditions of the international isolation variable during the period of perceived future
instability.
Discussing Alternative Explanations
Evidence in the North Korean case can be applied to the international
economic integration and prestige arguments of nuclear proliferation. As North Korea
has not been a democracy at any point, democracy, democratic transition, and
domestic politics arguments cannot be applied to the case.
140
The North Korea case appears consistent with the argument that high levels of
economic integration undermine the chance that a state will develop nuclear weapons.
North Korea's independence oriented juche philosophy and low levels of international
cooperation have produced a state with extremely low levels of international economic
integration. However, because the Kim regime was never economically integrated,
North Korea can only present a consistent case, and does not directly support or
undermine the international economic integration argument.
The domestic prestige argument, that politicians within a state pursue nuclear
weapons in order to enhance their hold on power, is applicable to North Korea. The
Kim regime has used its nuclear capacity as a propaganda tool on the general
population repeatedly.2°oHowever, the Kim regime has maintained unprecedented
control over media and information. North Korea's propaganda machine claimed for
years that North Korea was a socialist paradise with far higher living standards than
the rest of the world, while much ofthe population lived near starvation. The regime
could advertise a successful nuclear weapons program as public propaganda whether
or not one actually existed. This makes the domestic prestige gains of nuclear weapons
with regards to the general population meaningless.
On the surface, the North Korean case appears to support the international
prestige argument. The Kim regime has often publicly emphasized its nuclear
capacity, threatening "retaliation" for actions of foreign governments to which it
200 Myers 2009; and Becker 2005.
141
objects.2°1 However, the Kim regime's incentives for advertising its nuclear arsenal
are most likely not a push toward regional hegemony, and far more likely a method of
altering the perception of other powers in order to make North Korea a less appealing
target. The geopolitical situation ofNorth Korea, which makes the argument that
North Korea wants prestige in order to make itself a less appealing target more likely,
makes the prestige argument effectively indistinct from traditional security arguments,
and, as discussed in Chapter II, impossible to distinguish.
Conclusions
After reviewing the specifics ofthe North Korean case, traditional security
explanations appear to do a poor job of explaining the Kim regime's decision to
accelerate its nuclear program in the late 1980s. Additionally, neither prestige nor
international economic integration arguments can provide a compelling explanation of
North Korean behavior. Regime security theory does provide an explanation
consistent with the information available regarding North Korea. In the period
concerned, the late 1980s, North Korea had a preexisting nuclear program, was
internationally isolated, feared increasing domestic instability and embarked on an
acceleration of its nuclear weapons program.
201 Sang-Hun 2008; and Reiss 1995a.
142
If regime security theory does explain North Korea's behavior, then we should
expect certain future behavior. Specifically, we should see North Korea unwilling to
relinquish a minimum nuclear deterrent.
143
CHAPTER V
CONCLUSION
Scholars debate whether states develop nuclear weapons for reasons of
security, prestige, or domestic politics. The problem that tends to arise with these
approaches is that, because they come out of individual worldviews, they often
provide blanket explanations for a state's decision to go nuclear. Such decisions are
both complex and context dependent, and may have entirely disparate causes in
separate cases. This same pressure in theory creation tends to produce an
overemphasis on parsimony at the cost of overlooking variables critical to particular
subsets of cases. This is the eternal trade-off of any modeling endeavor - the battle
between abstraction and accuracy.
I argue that in order to adequately explain cases of nuclear development not
satisfactorily accounted for by existing theories, a more contextual, less general theory
is necessary. My thesis has focused on two cases which are significantly similar to
each other and significantly dissimilar to other cases of nuclear proliferation to date.
Neither South Africa nor North Korea has been adequately explained by existing
theories, and both cases are especially problematic for existing security theories of
nuclear proliferation. After comparing the cases, a striking number of similarities were
apparent. During the period of their nuclear development both South Africa and North
144
Korea had domestic regimes which had preexisting nuclear resources, had been
ostracized by the international community, were the targets of ire of at least one major
pole of world power, and experienced a period of domestic instability.
Regime security theory was developed out of this comparison. Regime security
theory argues that in cases of no clear strategic threat, regimes which possess some
preexisting nuclear capacity may respond to international isolation, pressure from a
major power(s), and increasing domestic instability by developing nuclear weapons.
Nuclear weapons allow such a regime to change the preferences of the great powers
because of the costs and risks associated with a nuclear armed state. The great
power(s) will favor maintenance of the status quo and may even prop up a nuclear
armed regime it intensely dislikes.
Regime security theory is useful primarily as a means to explain and
understand the decision making process of small potential nuclear powers. The utility
of understanding this decision making process is that it allows for more informed
foreign policies intended to prevent the completion of a nuclear program. There are
four significant policy implications that come out of this thesis. The first three policy
implications are connected. Each is derived from the causal processes laid out in
regime security theory and explains that if you remove one of the variables which lead
to non-standard nuclear weapons development, it will not occur.
The first preventive policy suggestion is that because small or poor states
without a preexisting resource advantage will not develop nuclear weapons, removing
or denying such an advantage significantly decreases the chances of proliferation.
145
Both South Africa and North Korea had significant nuclear programs, developed via
cooperation or funding from the great powers, before they decided to develop nuclear
weapons. The second policy suggestion is that the marked inclusion of the state in the
international environment and non-coercive, non-threatening interaction from the great
powers may serve to ameliorate the isolation variable. The third policy suggestion is
that working with the regime to ensure domestic stability will remove the main
pressure towards developing nuclear weapons. A stable, secure regime will not have
the same incentives to proliferate.
Finally, regime security theory suggests that coercive diplomacy, trade
sanctions, limited military actions and threats will be ineffective measures in
preventing proliferation. Once a regime feels threatened, alone and unstable, it will be
incredibly reluctant to give away its nuclear potential. The only coercive strategy that
is likely to be effective against such a regime is the costly option of regime overthrow;
anything less will serve to justify the regime's perception of the hostile world around
it. It will back further into a corner, compelled to complete its nuclear program for
protection.
This thesis has beenan exercise in theory creation, not theory testing. Much
further research is necessary to test the applicability of the theory to alternate cases.
Two subsets of cases provide possible candidates. The first subset is cases in which
states without clear strategic security threat developed significant nuclear weapons
146
programs that did not reach fruition.2°2 These cases would be specifically interesting to
look at to determine why they did not reach fruition and check whether that
explanation is consistent with regime security theory. Such cases might include Egypt,
Congo and Libya. The second subset is cases where states which possibly meet all the
conditions of regime security theory have not attempted to develop nuclear weapons.
If cases exist in which the condition and independent variables were met and the state
did not attempt to develop nuclear weapons, regime security theory would be
undermined.
In a broader theoretical context this thesis undermines the idea that regime
interests are analogous with state interests and thus casts doubt on the black box as a
useful construct. Further research which explains instances of "irrational" state
behavior by exploring contexts in which centralized governments make decisions
irrational for the states but rational for the government could be informed by the
observations in this thesis.
Despite the significant task of theory testing which remains to be
accomplished, this thesis has produced and explained a testable theory of nuclear
proliferation which provides an internally consistent explanation of previously
problematic cases. While regime security theory lacks the extreme parsimony of
202 Possible test cases for regime security theory do not include every country which has ever
considered embarking on a program of nuclear development. Walsh, citing a source in the Egyptian
defense department states: "One person, when asked about radiological weapons, responded that
'naturaIly,' Egypt had looked at a 'trash bomb.' He explains that 'in every military there are dreamers
and exploiters' who convince their bosses to back some scheme." Such discussions within a
government are, on their own, not evidence ofa "near miss" for proliferation. Walsh 2001.
147
traditional security theories, it is my hope that what is lost in generality by opening the
black box is made up for with more context specific, useful, policy insights.
148
BIBLIOGRAPHY
Abromowitz, Morton 1., Lane1y, James T., Heginbotham, Eric. 2003. Meeting the
North Korean Nuclear Challenge The Council on Foreign Relations, 2003
[accessed November 19,2004.
Agreed Framework Between the United States of America and the Democratic
Peoples Republic of Korea. October 21,1994. Geneva, Switzerland.
Allison, Graham T. 1971. Essence ofDecision: Explaining the Cuban Missile Crisis,
Boston, MA: Little, Brown.
Becker, Jasper 2005. Kim Jong II and the Looming Threat ofNorth Korea, New York:
Oxford University Press.
Beldecos, Nick, Heginbotham, Eric Spring 1995. "The Conventional Military Balance
in Korea." Breakthroughs, 4 (1).
Chafetz, Glenn 1993. "The End of the Cold War and the Future of Nuclear
Proliferation: an Alternative to the Neorealist Perspective." In The
Proliferation Puzzle: Why Nuclear Weapons Spread and What Results, edited
by Zachary S. Davis and Benjamin Frankel. London: Frank Cass and Co., 127-
158.
Cummings, Bruce, Abrahamian, Ervand, Ma'oz, Moshe 2004. Inventing the Axis of
Evil, New York: The New Press.
Davis, Zachary S. 1993. "The Realist Nuclear Regime." In The Proliferation Puzzle:
Why Nuclear Weapons Spread and What Results, edited by Zachary S. Davis
and Benjamin Frankel. London: Frank Cass and Co., 79-99.
Davis, Zachary S., and Benjamin Frankel, eds. 1993. The Proliferation Puzzle: Why
Nuclear Weapons Spread and What Results, London: Frank Cass and Co.
Fearon, James D, Laitin, David D 2003. "Ethnicity, Insurgency, and Civil War."
American Political Science Review, 97 (1): 75-90.
149
Frankel, Benjamin 1993. "The Brooding Shadow: Systemic Incentives and Nuclear
Weapons Proliferation." In The proliferation Puzzle: Why Nuclear Weapons
Spread and What Results, edited by Zachary S. Davis and Benjamin Frankel.
London: Frank Cass and Co., 37-78.
George, Alexander L., and Andrew Bennett 2005. Case Studies and Theory
Development in the Social Sciences, Cambridge, MA: MIT Press.
Hui, Zhang 2007. "Revisiting North Korea's Nuclear Test." China Security, 3 (3): 119-
130.
Kerr, Paul. 2003. Chronology ofUS.-North Korean Nuclear andMissile Diplomacy
Arms Control Association, 2003 [accessed 4/14 2008]. Available from
http://www.armscontrol.org/pdf7dprkchron.pdf
Krasner, Stephen 1992. "Realism, Imperialism, and Democracy: A Response to
Gilbert." Political Theory, 20 (20): 39.
Maoz, Zeev 2003. "The Mixed Blessing ofIsrael's Nuclear Policy." International
Security, 28 (2).
Mearsheimer, John 1. May 17,1998. Here We Go Again. New York Times, p. 17.
Mearsheimer, John 1. 2001. The Tragedy ofGreat Power Politics, New York: Norton.
Meyers, B. R. 2009. To Beat a Dictator, Ignore Him. New York Times, April 1,2009.
Meyer, Stephen M. 1984. The Dynamics ofNuclear Proliferation, Chicago and
London: The University of Chicago Press.
Moore, 1.D.L 1987. South Africa andNuclear Proliferation, New York: St. Martin's
Press.
OHanlon, Michael Spring 1998. "Stopping a North Korean Invasion: Why Defending
South Korea is Easier Than the Pentagon Thinks." International Security, 22
(4).
Park, John S 2005. "Inside Multilateralism: The Six Party Talks." The Washington
Quarterly, 28 (4): 75-91.
Paul, T.Y. 2000. Power Versus Prudence, Montreal: McGill- Queen's University
Press.
150
Pelletiere, Stephen C. 2003. A War Crime Or an Act of War? New York Times,
January 31.
Perry, William J 2006. "Proliferation on the Peninsula: Five North Korean Nuclear
Crises." The ANNALS ofthe American Academy ofPolitical and Social
Science, 607 (78).
Raas, Whitney and, and Austin Long 2007. "Osirak Redux? Assessing Israeli
Capabilities to Destroy Iranian Nuclear Facilities." International Security, 31
(4).
Reiss, Mitchell 1988. Without the Bomb: The Politics ofNuclear Nonproliferation,
New York: Columbia University Press.
Reiss, Mitchell 1995a. Bridled Ambition: Why Countries Constrain Their Nuclear
Capabilities, Washington, DC: Woodrow Wilson Center Press.
Reiss, Mitchell 1995b. "Nuclear Rollback Decisions: Future Lessons?" Arms Control
Today, July/August. 25: 10-15.
Robinson, Colin. 2004. Stand-ojJwith North Korea: War Scenarios and Consequences
2004 [accessed 2004].
Sagan, Scott D. 1996. "Why Do States Build Nuclear Weapons? Three Models in
Search of a Bomb." International Security, Winter. 21 (3): 54.
Sang-Hun, Choe. 2008. North Korea Threatens to Turn South into 'Ashes'. New York
Times, March 30th, 2008.
Sang-Hun, Choe. 2009. North Korea Claims to Conduct 2nd Nuclear Test. New York
Times.
Scheinman, Lawrence 1965. Atomic Energy Policy in France Under the Fourth
Republic, Princeton: Princeton University Press.
Sciolino. January 21,1988. North Korea Added to Terror List New York Times.
Shambaugh, David 2003. "China and the Korean Peninsula: Playing for the Long
Term." The Washington Quarterly, 26 (2): 43-56.
Shultz, George. 1984. Preventing the Proliferation of Nuclear Weapons, edited by U.S.
Department of State.
151
Snyder, Jed C., and Samuel F. Wells, Jr 1985. Limiting Nuclear Proliferation,
Cambridge, MA: Ballinger Pub. Co.
Snyder, Scott 2000. "Pyongyang's Pressure." Washington Quarterly, 23 (3): 163-170.
Solingen, Etal 1994. "The Domestic Sources of Regional Regimes: The Evolution of
Nuclear Ambiguity in the Middle East." International Studies Quarterly, 38
(2): 305-337.
Solingen, Etel 2007. Nuclear Logics: Contrasting Paths in East Asia and the Middle
East, Princeton: Princeton University Press.
Spector, Leonard S, Smith, Jacqueline R 1991. "North Korea: The Next Nuclear
Nightmare?" Arms Control Today, 22 (2): 8-13.
Walsh, James 1. 2001. Bombs Unbuilt: Power, Ideas and Institutions in International
Politics, Dept. of Political Science, Massachusetts Institute of Technology.
Waltz, Kenneth N. 1981. "The Spread of Nuclear Weapons: More May be Better."
Autumn.
Zolberg, Aristide R, Suhrke, Astri, Aguayo, Sergio 1989. Escapejrom Violence:
Conlict and the Refugee Crisis in the Developing World, New York, Oxford:
Oxford Press.