Professional Documents
Culture Documents
You must not circulate this work in any other form and you must impose this same
condition on any acquirer.
Library of Congress Cataloging-in-Publication Data
Names: Koch, Lisa (Lisa Langdon), author.
Title: Nuclear decisions : changing the course of nuclear weapons programs / Lisa Langdon
Koch.
Description: New York, NY : Oxford University Press, [2023] | Includes bibliographical
references and index.
Identifiers: LCCN 2022062279 (print) | LCCN 2022062280 (ebook) | ISBN 9780197679531
(hardback) |
ISBN 9780197679548 (epub) | ISBN 9780197679555 | ISBN 9780197679562
Subjects: LCSH: Nuclear nonproliferation—Government policy—Case studies. |
Military policy—Decision making—Case studies.
Classification: LCC JZ5675.K62 2023 (print) | LCC JZ5675 (ebook) |
DDC 327.1/747—dc23/eng/20230216
LC record available at https://lccn.loc.gov/2022062279
LC ebook record available at https://lccn.loc.gov/2022062280
DOI: 10.1093/oso/9780197679531.001.0001
Contents
Acknowledgments
Appendix
Notes
Bibliography
Index
Acknowledgments
I began conducting the research that led to this book when I was a
graduate student at the University of Michigan. Allan Stam supported
and encouraged my work, helped me think about the big picture,
and never failed to provide wise counsel. James Morrow, Philip
Potter, and Robert Franseze each taught and advised me in
important ways as I pursued this research in its early form and in
the years after I finished my graduate work. I remember in
particular the times I was lucky enough to be able to talk about the
project with Al, Jim, Phil, and Rob all together, and I thank them for
their invaluable insights and advice. I am also grateful for Charles
Shipan’s scholarly guidance, and for his continuing mentorship.
Chuck’s graduate seminar on American political institutions
influenced the way I think about key institutional players and the
relationships among them. I thank Cameron Thies, Thorin Wright,
and many other generous scholars at the School of Politics and
Global Studies at Arizona State University for their support and
assistance as I finished my dissertation.
I am indebted to Scott Sagan, Vipin Narang, and Branislav
Slantchev, whose insights shaped my thinking about the manuscript.
I reflected upon our lively discussion many times when making the
revisions that have led to a better book. I thank Scott in particular
for his mentorship, which began several years ago when I
introduced myself after a conference panel. Scott invited me to sit
down then and there to tell him about my work, and I have
benefited from his generous guidance and insightful critiques ever
since.
My colleagues at Claremont McKenna College made the
development and completion of this book possible in many different
ways. I thank Hilary Appel and the Keck Center for Strategic and
International Studies for supporting the development of the
manuscript at key moments. The Government Department has
enthusiastically supported my research since I arrived at CMC, and I
thank especially Hilary Appel, William Ascher, Mark Blitz, Hicham Bou
Nassif, Jordan Branch, Andrew Busch, Roderic Camp, Minxin Pei,
Jack Pitney, Shanna Rose, Jon Shields, Aseema Sinha, Jennifer Taw,
and George Thomas for their insights and advice. When I arrived at
CMC, I had the good fortune to be assigned the office next to the
other assistant professor in the department, Emily Pears. I thank
Emily for her friendship, for many conversations about the process of
writing a book, and for helping me think through theoretical tangles
that arose as I wrote.
For generating and sustaining a faculty writing community, I am
grateful in particular to Peter Uvin, Adrienne Martin, Esther Chung-
Kim, Ellen Rentz, Sharda Umanath, Heather Ferguson, Emily Pears,
and Janice Heitkamp. I thank the outstanding students who provided
excellent research assistance, including Katrina Frei-Herrmann,
Daniel Krasemann, Tallan Donine, Marcia Yang, Alexander Li,
Johnson Lin, Charles Warren, and my many nuclear politics seminar
students, in particular Henrietta Toivanen and Stuart Brown. Katrina
deserves special recognition for working with me on various projects
over three years and for executing the first polished drawings of the
proliferation curves.
Alexander Lanoszka, Sarah Croco, Matthew Fuhrmann, and
Leanne Powner each offered valuable advice during the writing
process. I thank Matthew Wells for many discussions and
conversations as the project evolved, and most of all for many years
of friendship. At the University of Michigan, I relied on the
professional knowledge and experience of political science librarian
Catherine Morse, and on Sofia Rosenberg, who volunteered to
translate Swedish writings into English so that I could puzzle out the
characteristics of Swedish nuclear institutions. I thank David McBride
and two anonymous reviewers for Oxford University Press for their
valuable guidance, and Sharon Langworthy for expert copyediting.
I am indebted to Donald Hafner, who taught me about nuclear
weapons strategy when I was an undergraduate student at Boston
College. His teaching and his ideals continue to inspire me, and I
greatly value his ongoing mentorship. For their scholarly advice and
their friendship, I thank Katja Favretto, Vanessa Cruz-Nichols, and
Ida Salusky. I thank Rebecca Martinez for significantly influencing my
approach to the process of research and writing. I am grateful to my
faith community at the Claremont Colleges, in particular Steve Davis,
Esther Chung-Kim, TJ Tsai, George Montanez, and Dave Vosburg for
steadfast support.
My family deserves the most thanks, starting with my mom and
dad, Janice and John Langdon, and my sister, Heather, each of
whom has always provided me with unconditional love and support.
From the start, they have been enthusiastic about this project and
its development into a manuscript, and I am deeply grateful for our
many conversations and for their advice. My mother-in-law and
father-in-law, Paige and Joseph Koch, have also showered me with
love and support ever since I had the good fortune to join their
family. They have read my work, sent me articles related to my
research, and thoughtfully asked me about the manuscript’s
progress.
My dad is professor emeritus of history at Le Moyne College, and
I thank him in particular for the many, many hours he has spent
reading and commenting on various drafts of this manuscript over
the years. Everyone should be so lucky as to have a world historian
on call while conducting case research, not least because the
conversations are such great fun.
I completed much of this manuscript in 2020 and 2021, during
the global pandemic. My husband, Matt, and I worked hard to try to
adapt to a time of significant disruption, including the loss of in-
person school for our three children for more than a year. Writing
during this time was tremendously challenging, and I am truly
fortunate to be part of a wonderful family of five that sustains me.
Thank you, Matt, for your love and support over many years. You
have been on the entire journey with me, start to finish. And finally,
I thank our children, Audrey, Paul, and Timmy, who sometimes
permit me to sneak in a few more minutes to write, sometimes
distract me, and always fill our lives with a special joy.
1
Introduction to Nuclear Decisions
Nuclear Decisions
Nuclear decisions offer the answer. State leaders make decisions
within different information environments that affect their beliefs and
preferences about nuclear weapons. These decisions to accelerate or
reverse progress toward a nuclear weapons capability define each
state’s course. Whether or not a state ultimately acquires nuclear
weapons depends to a large extent on those nuclear decisions.
I argue that two crucial features of the political environment
affect nuclear decision-making. Leaders make decisions not in a
vacuum but in changing international and domestic contexts. First, in
different proliferation eras, changes to international political and
structural conditions constrain or free states to pursue nuclear
weapons development. These conditions are imposed from the top
down. Second, across these eras, domestic scientific and military
organizations may intervene to bring about, or prevent, a nuclear
decision that could redefine a state’s course to the bomb. The
conditions under which scientific and military experts are able to
influence state leaders from the bottom up are thus a critically
important aspect of this story.
Nuclear Goals
The historical record demonstrates that states do not initiate nuclear
weapons programs and then uniformly follow linear paths to a
singular goal. One possible explanation for erratic progression is
political meddling in scientific research and development. Jacques E.
C. Hymans argues that leaders who are unconstrained by state
institutions often interfere in nuclear weapons programs,
unintentionally disrupting progress toward the bomb. Whether
scientists are free to pursue their work in ways that will advance
good research and development or instead face strong incentives to
appease repressive leaders through shortcuts and false reporting
should affect a program’s timeline.9
This compelling argument about time-to-outcome, however,
cannot explain the form a nuclear weapons program takes. Implicit
in Hymans’s argument are the assumptions that states have a
common goal—to quickly produce nuclear weapons—and take linear
paths to the bomb. The observation that few states had obtained a
speedy outcome led Hymans to conclude that something had gone
wrong. However, while racing to the bomb was more common during
the early Cold War, for most of the nuclear age the full-speed-ahead
approach has been the exception, not the rule.10 Rather, leaders
have exhibited a range of preferences regarding the importance and
necessity of quickly acquiring a nuclear arsenal.
If these nonlinear pathways are not a deviation—if nuclear
weapons development is instead typically nonlinear—then
interference with project management cannot be a sufficient
explanation. I argue that the paths to the bomb are rarely linear
because they are interrupted and reformed by nuclear decisions.
Leaders may allow a nuclear weapons program to maintain the
course it is on or even decide to slow or suspend its development.
Domestic organizations are a key source of expert information that
shapes the leader’s perception of the value and strategic purpose of
the nuclear program.
Another possible explanation is that changes in the security
environment prompt a state to move toward or away from the
bomb. Security concerns are an important motivator for the initial
decision to start a nuclear weapons program.11 Yet the security
explanation, too, is insufficient. States that do decide to begin a
program may exist in insecurity for years before choosing the
nuclear path. And once a nuclear weapons program is underway,
many leaders appear to make nuclear decisions that are not based
on either stable or changing external security environments. If
security were the sole driver of nuclear decisions, we would expect
to see acceleration decisions during times of high insecurity and
reversal decisions during times of low insecurity.
The case studies I conduct in this book do not indicate the
presence of such a dynamic. For example, India’s program slowed
significantly in the mid- to late 1970s, despite nuclear weapons
progress in its regional rivals, China and Pakistan. South Korea did
not accelerate its program when its security environment worsened.
Brazil gave up its pursuit of nuclear weapons despite little to no
change in its security environment. South Africa sprinted toward a
nuclear arsenal despite its significant regional military superiority.
Perhaps deep concerns over Soviet interference in southern Africa,
or even fears of invasion, could explain South Africa’s proliferation
curve instead—but then why did Pretoria implement two different
program reversals, well before the fall of the Soviet Union?12
Within the context of an ongoing nuclear weapons program, the
threat environment is not the only important factor that affects
leaders’ perceptions of the costs and benefits of the nuclear
weapons effort. A nuclear weapons program is one of many options
available to a government that faces serious security concerns. A
state could instead decide to arm conventionally, seek military
assistance from an ally, or enter into a defense pact. Or a leader
may decide to gain leverage over adversaries by hedging: pursuing
nuclear development to achieve a latent nuclear weapons capability
without progressing all the way to the weapons themselves. Fears of
a preventive war aimed at the nuclear program could prompt either
a reversal decision to remove the cause of the threat or an
acceleration decision, in hopes of acquiring nuclear weapons to deter
future attack. A threatening security environment could therefore
lead to either type of decision or no decision at all. Security cannot
fully explain states’ proliferation pathways.
Because states consider different policy options in response to the
strategic environment, and each option has its own potential
benefits and drawbacks, nuclear weapons programs are situated
within a political context. Leaders consider many possibilities beyond
the simple binary outcomes of acquisition or termination, and they
do so within a complex information environment that affects how the
value of a nuclear weapons program is understood. They must
weigh the benefits of state security against drawbacks like domestic
resource trade-offs, potential damage to strategic international
relationships, and the likelihood of program success.
Implications
This book joins the growing literature on nuclear proliferation and
reversal, offering a systematic analysis of the process and politics of
nuclear decision-making. I approach this subject from a different
conceptualization of nuclear weapons programs: that they are
defined by decisions to accelerate or reverse nuclear development.
In doing so, I investigate the strategic decisions that create the form
of nuclear weapons programs rather than focusing on the time
between program initiation and the outcome of a nuclear bomb.
Pursuing nuclear weapons, whether in Iran, North Korea, India, or
Pakistan, is a long process punctuated by political decisions that can
change the course of nuclear development. Rather than examining
the conditions present when a milestone program outcome is
realized, I examine the conditions present at the time the nuclear
decision was made.
This analysis reveals that both international structural conditions
and domestic coalitions matter. Even in wartime, whose voices are
heard from within the state and what preferences they express can
change how a leader understands the international environment.
Those domestic experts can highlight or downplay the advantages
and disadvantages of steps to change the course of nuclear weapons
development. The relative balance of power among the key domestic
organizations, which can change as they interact with each other
and their political environment, affects the ability each expert group
has to influence the leader. These organizations may prefer to push
the state either toward or away from nuclear weapons. If we ignore
the domestic environment and instead assume that states pursue
nuclear weapons along uniform and consistent paths, we
underestimate the importance of the nuclear decisions that
determine whether a state ultimately acquires nuclear weapons.
Finally, a central argument of this book is that we should not
study the decision to start a nuclear weapons program as if the
state’s ultimate goal is to quickly produce the weapons. Not only do
nuclear aspirants pursue different goals, but changing circumstances
may also lead a state to later deviate from the original goal. And
because the end results of nuclear decisions are realized months or
years later, programs may reach milestones that are the product of
decisions made by leaders who were responding to conditions that
have since changed. Because a nuclear weapons program outcome
will occur at some period of time after a nuclear decision was made,
examining the conditions at the time of the outcome will be
misleading. We should instead seek to understand the conditions at
the time of the decision that paved the way to the outcome. This
shift in focus could allow states to respond more productively to
changes in their adversaries’ or allies’ nuclear weapons development
and create better nonproliferation policy tools.
Proliferation Curves
If nuclear decisions shape states’ proliferation pathways, what do
different paths to and away from the bomb look like? In the
following discussion I represent six states’ paths, or proliferation
curves, with basic graphs. Along the proliferation curve, nuclear
decisions are represented by vertical lines, each appearing during
the year the decision was made. Dashed lines represent a nuclear
reversal decision, with double-dashed lines representing a
permanent reversal decision. A solid line indicates a nuclear
acceleration decision. Each graph measures time in years along the
x-axis and nuclear development milestones along the y-axis. The
data I used to generate each proliferation curve, and the sources I
relied on, are reported in the appendix.
The milestones I have selected for these graphs indicate key
stages of nuclear development. As a state progresses toward, or
away from, nuclear weapons, the proliferation curve travels through
these different stages. States establish a research and development
program, obtain uranium or plutonium, build a nuclear reactor,
operate the reactor, enrich fissile material so it can be used in a
bomb, test a bomb, and develop delivery systems. It is not always
obvious when a state moves from one milestone to the next, in part
because these accomplishments are often closely guarded secrets.
The dates and milestones I use in these graphs are accurate to the
best of my knowledge as of this writing. Furthermore, the
progression along the path of nuclear weapons development is not
always linear. Different decisions made in the course of a program
have caused some states to skip over a milestone, such as
accumulating a stockpile of nuclear weapons and developing delivery
systems before testing a prototype bomb. A black diamond marks
the time and level of each nuclear development milestone a state
reaches.
The lowest stage of nuclear development is classified as
“nascent,” meaning the state does not yet have an operational
nuclear reactor of any size. The proliferation curve begins in the year
the state government makes the political decision to start a nuclear
weapons program. Of the six proliferation curves shown in this
chapter, only one state—China—began its weapons program at the
lowest stage of nuclear development. Each of the other five states
had existing nuclear infrastructure, at least at the reactor level, in
place prior to the start decision.10
Stages 2 through 5 indicate progressive levels of enrichment and
reprocessing capacity. The ability to produce fissile material to power
nuclear weapons is a key hurdle to overcome in a nuclear weapons
program. Whether a state decides to produce uranium or plutonium
weapons—and in many cases both options are explored—this is the
most difficult step in producing nuclear bombs, as well as one of the
most expensive.11 Most importantly, a state cannot produce nuclear
weapons without highly enriched uranium or plutonium from a large
reactor. For stages 2 and 3, I distinguish between small and large
reactors using the power output threshold of 1 megawatt (MW).12
Reactors with design power greater than 1 MW have significant fuel
requirements, and a large reactor is a necessary condition for the
production of nuclear weapons.13 Several states, including India and
Israel, produced the fissile material needed for nuclear weapons
from large research reactors rather than from industrial-scale
reactors.14 For stages 4 and 5, I distinguish between small and large
uranium enrichment or plutonium reprocessing facilities. Large,
industrial-scale facilities that use commercial technologies can
produce significantly more weapons-grade uranium or plutonium
than small, laboratory, or pilot facilities can.15
The last three stages indicate different levels of weaponization: a
fission test, followed by deployable nuclear weapons, and finally a
fusion (thermonuclear) test. These stages are separate from one
another, as readers will note from the graphs; a state may test an
atomic bomb but not build a nuclear arsenal, as India initially did, or
a state may proceed directly to deployable nuclear weapons without
first testing a device, as South Africa did. If a state reverses to the
extent that the program regresses to an earlier stage of nuclear
development, that is indicated on the graph.
Finally, dark, medium, and light gray shadings indicate the status
of the program before and after each nuclear decision. Dark gray
represents progress toward the bomb, while medium gray indicates
that the program is continuing but is holding steady at its current
level or is progressing very slowly due to reductions or a lack of
interest from the government. Light gray indicates the program has
been suspended, either permanently or temporarily.
I selected the six states represented here to provide a visual
demonstration of proliferation curves that exhibit variation in several
dimensions: the historical era of pursuit, whether the state ultimately
acquired nuclear weapons, and regime type. I divide the nuclear
weapons age into three time periods: the “permissive” decades prior
to the 1964 shock of the Chinese nuclear fission test, when major
world powers—and a few minor powers—sought the bomb with few
external constraints; the “transition period,” beginning with the 1965
start of the negotiations that would culminate in the Nuclear Non-
Proliferation Treaty (NPT), during which a nuclear nonproliferation
regime was emerging; and the “nonproliferation regime period,”
characterized by the deepening of the nonproliferation regime,
dating from the shock of the 1974 Indian peaceful nuclear explosion
(PNE). Each of these periods constitutes a different global
environment with different political and structural constraints on
nuclear aspirants. I develop arguments for the significance of each
historical period in Chapter 4.
I display two proliferation curves from each period in Figure 2.1.
France and China were two of the original five countries to pursue
nuclear weapons, beginning their programs within ten years of the
end of World War II. India and South Korea were most active during
the transition period between the early nuclear weapons states’
pursuits and the consolidation of the nuclear nonproliferation
regime. India acquired nuclear weapons, while South Korea
abandoned its program. Brazil and South Africa pursued nuclear
weapons in the years after the Indian PNE, a shock that generated
the political will to deepen the emerging nonproliferation regime.
Brazil terminated its program, and South Africa acquired nuclear
weapons and then later dismantled them.
Figure 2.1 Proliferation curves by state
Strikingly, each of these graphs is noticeably different from the
others. Progress to the bomb is neither uniform nor clearly
predictable. Brazil, South Korea, and South Africa chose to slow or
suspend their programs, and those programs eventually ended. But
India made reversal decisions that were themselves later reversed
by acceleration decisions, and it ultimately acquired and deployed a
nuclear arsenal. In that successful case, the outcome—a nuclear
weapon—was not simply achieved after the initial decision to start a
weapons program. Several other nuclear decisions were part of the
proliferation process.
In each case, numerous decisions were made in order to achieve
nuclear weapons status. Limiting one’s attention to only one part of
the proliferation curve may thus lead to mistaken conclusions about
nuclear development. Leaders make decisions that change the shape
of the curve. There is no one set, linear path to the bomb, but rather
multiple paths formed by nuclear decisions.
While nuclear decisions occur throughout these proliferation
curves, it is also clear that they are relatively infrequent. This fact is
unsurprising and serves to underscore the importance of the
decisions. State leaders do not make dozens of nuclear reversal and
acceleration decisions over the lifetime of a program. Significant
nuclear decisions require political will. Both acceleration and reversal
decisions are made at high levels, not only because nuclear weapons
are so strongly tied to national interests, but also because they bring
about policy changes, resource reallocations, and sometimes
structural shifts within or among state institutions.
The role of the NPT, which opened for signature in 1968 and
entered into force in 1970, is not obvious at first glance. These
curves do not indicate that either 1968 or 1970 was a watershed
year for program development. However, the proliferation curves in
the second and third time periods do stand in contrast to the French
and Chinese curves in the first time period. In those later periods,
reversal decisions suspend or halt programs and are sometimes
followed by acceleration decisions. The paths appear bumpier and
less predictable. By the mid- to late 1970s nuclear development
seems to be proceeding in a substantively different way. In Chapter
4 I address how, over time, the consolidation of the NPT and
evolution of the broader nonproliferation regime, as structuring
features of the international environment, affected nuclear decision-
making in these different time periods.
The graphs also clearly show that nuclear weapons development
milestones are achieved after nuclear decisions are made, and at
what seem to be irregular intervals. Further, each decision paves the
way for future decisions. States neither stumble into a nuclear
weapons capability nor accidentally fall out of a nuclear weapons
program. Even in states like India, known for its powerful “enclave”
of nuclear scientists, key stages of nuclear development were
neither reached nor abandoned unless governments made political
decisions.16 The formal decision in 1972 to begin working toward a
nuclear test, for example, was made by Prime Minister Indira
Gandhi, who had consulted with Indian nuclear scientists. The
Atomic Energy Commission formed a committee to manage that
effort in response to Gandhi’s order.17 That political decision, and the
resulting preparations, made the 1974 explosion possible. Nuclear
decisions are what set the program toward or away from the
outcomes, and nuclear decisions are thus the focus of this book.
Nuclear Decision-Making
I offer two theoretical mechanisms to explain why leaders make
nuclear decisions. They make decisions within different information
environments that affect their beliefs and preferences about nuclear
weapons. In different proliferation eras, from the top down changes
to international political and structural conditions constrain or free
states to pursue nuclear weapons development. Within and across
these eras, from the bottom up domestic organizations may
intervene to bring about, or prevent, a nuclear decision that has the
potential to reshape the curve.
The first mechanism, which I discuss in this chapter, comes from
the bottom up. The domestic organizations that house a nation’s
nuclear experts can, under certain conditions, influence a leader’s
beliefs about the costs and benefits of a nuclear weapons program.
The state’s security environment alone cannot explain leaders’
preferences regarding nuclear weapons, because the pursuit of
nuclear weapons is far from the only option available to an insecure
state. Leaders can respond to security conditions by selecting from a
range of policy responses. I argue that military and nuclear scientific
organizations may intervene to shape how leaders understand and
interpret the security environment. Under certain conditions, these
expert organizations may work together or compete with each other
to limit or expand the set of available policy options presented to the
leader.
Second, the political context in which leaders make decisions is
shaped from the top down by the external proliferation environment.
Over time the structure and politics of the international system have
created permissive or constraining environments in which states
might pursue nuclear weapons. I argue that the benefits and costs
of pursuing nuclear weapons have been understood differently
within each of three distinct proliferation environments, bookended
by systemic shocks. In each proliferation environment, leaders are
faced with different sets of available options.
Therefore, my investigation of how domestic organizations
attempt to intervene in a state’s nuclear pursuits must include the
limits imposed from above by the different proliferation
environments. The story of Japan’s early pursuit of nuclear weapons,
which I briefly related in the opening pages of this book, helps
illuminate why these changing proliferation contexts are so
important. While Japan’s scientific and military experts fumbled in
the terrible darkness and confusion of August 1945, trying to
determine what had happened at Hiroshima, it would not be long
before states would never again have reason to question the
existence or power of nuclear weapons. Nuclear weapons hopefuls
now understood that the ultimate goal could be achieved. Whether
states faced external constraints and costs in pursuing nuclear
weapons, however, was a different matter. Those constraints and
costs, shaped by changing international political interests, have
created different proliferation conditions across three distinct time
periods. I discuss this top-down mechanism further, and define and
describe the three proliferation environments, in Chapter 4.
Alternative Explanations
I offer a novel explanation as one way to understand why leaders
might make these decisions, and especially why they might make
puzzling decisions. However, I do not claim that these theoretical
mechanisms determine every nuclear decision. Other factors will also
affect the path of a state’s nuclear weapons program. I examine
several possible alternative explanations for proliferation pathways
and discuss why each is insufficient, before turning to my theory of
nuclear decision-making.
Security
Many states choose not to pursue nuclear weapons, despite
possessing the requisite knowledge and resources. A state must also
be willing to proceed down the costly and controversial path to the
bomb. Over the past two decades, multiple studies have
demonstrated the importance of the threat environment in
prompting states to start a nuclear weapons program, although
some measures of threat are less predictive across these studies
than others.1 Scott D. Sagan identifies three different theoretical
motivators that may drive states to start a nuclear weapons
program: security, domestic political interests, and norms.2 As the
case histories of nuclear weapons–seeking states indicate that one
or more of these motivators can be identified as contributing to each
state leader’s decision to start a program, a rich literature has
emerged that investigates the wide range of possible explanations.
But the security-based approach to understanding proliferation has
been, and continues to be, dominant in the field.
If the international system is anarchic, neorealist scholars have
argued, then states will seek to help themselves by pursuing nuclear
weapons.3 Whether balancing against nuclear rivals or responding to
emerging threats, insecure states should seek the deterrent benefits
conferred by these extraordinarily powerful weapons. As Sagan
notes, the security explanation has the benefit of appearing to be
the most obvious: states pursue nuclear weapons in response to
insecurity. Yet in practice this parsimonious approach has
overpredicted proliferation. Insecure states do not necessarily
address a security threat—even a nuclear threat—by pursuing
nuclear weapons. And states that do decide to begin a program may
exist in insecurity for many years prior to starting down a
proliferation pathway. Early in the nuclear weapons age, many
predicted that countries with sufficient technological capacity and
material wealth would likely tumble into proliferation, one after
another. But a world in which countries fell like metaphorical nuclear
dominoes has not come to pass.
One of the most important reasons we do not observe capable
states regularly responding to threats with nuclear acceleration
decisions is that those decisions do not bring about immediate,
useful results. Nuclear acceleration decisions incur very high costs
and are implemented within bureaucratic structures that require time
to adapt to altered circumstances. A program decision may require
building new facilities, hiring new personnel, expanding a
partnership with a foreign government, or finding new sources of
material or equipment. Deciding to increase work on an existing
project to move up the timeline, for example, or to open the
plutonium path to the bomb, would be long-term, highly secret
initiatives that would take months or years to come to fruition. In
fact, I argue in the case studies that security concerns could also
prompt reversal decisions, as leaders may decide to meet emerging
threats by reducing the state’s investment in nuclear weapons and
reallocating resources to conventional arming.
Despite its limitations, the security approach continues to be a
starting point for much of the nonproliferation literature. This has
three broad implications for the study of nuclear weapons. First, the
security approach generates explanations at the state level. States
are conceptualized as unitary actors that will respond to threats in
predictable ways within the structure of the international system.
Domestic-level factors rarely enter into the analysis. In a notable
exception, Vipin Narang’s security-based model of how states pursue
proliferation gives a nod to the domestic environment. He writes:
The inconsistency with which domestic actors accurately assess and respond
to the external security environment means, however, that variation in unit-
level consensus can exert independent influence on a state’s strategy.4