The Causes of Nuclear Weapons Proliferation: Scott D. Sagan

PL14CH11-Sagan ARI 18 February 2011 17:14
V I E W
E
R
S
Review in Advance first posted online
on March 21, 2011. (Changes may
C E
I N
still occur before final publication

online and in print.)
N
A
D V A
The Causes of Nuclear

Weapons Proliferation
Scott D. Sagan
by Stanford University - Main Campus - Green Library on 04/25/11. For personal use only.
Department of Political Science, Stanford University, Stanford, California 94305;

Annu. Rev. Polit. Sci. 2011.14. Downloaded from www.annualreviews.org
email: ssagan@stanford.edu
Annu. Rev. Polit. Sci. 2011. 14:225–44 Keywords

The Annual Review of Public Health is online at International Atomic Energy Agency (IAEA), nuclear latency,
publhealth.annualreviews.org
Non-Proliferation Treaty (NPT), plutonium reprocessing, uranium
This article’s doi: enrichment
10.1146/annurev-polisci-052209–131042
Copyright c 2011 by Annual Reviews. Abstract

All rights reserved
This critical review of the new political science literature on the causes
1094-2939/11/0615-0225$20.00 of nuclear weapons proliferation consists of four parts. The first section
briefly presents what we know about which states developed nuclear
weapons and which states started but abandoned weapons development
programs. I highlight the problems that result from uncertainty about
the accuracy and completeness of the data. The second and third sec-
tions review the literature on the spread of the technical capability to
develop nuclear weapons. We still lack robust knowledge about the re-
lationship between the development of civilian nuclear power programs
and nuclear weapons acquisition. The next two sections review the liter-
ature on the demand for nuclear weapons. Comparative case studies and
statistical studies have improved our understanding of the diversity of
motives for weapons development and restraints, but serious gaps in our
knowledge remain. The sixth section outlines alternative theories about
the potential impact of the nuclear Non-Proliferation Treaty (NPT) on
nuclear weapons programs decisions. Finally, I lay out a future research
agenda to address the weaknesses in our current understanding of the
causes of nuclear proliferation.
225
Changes may still occur before final publication online and in print
INTRODUCTION this review, I argue that the supply-side litera-

ture has mistakenly focused on inaccurate mea-
Concerns about the potential spread of nuclear
sures of nuclear weapons capabilities, while the
NPT: Non- weapons have increased during the past decade,
Proliferation Treaty demand-side literature has paid inadequate at-
as North Korea tested two nuclear weapons,
tention to how the Non-Proliferation Treaty
Iran was placed under United Nations Secu-
(NPT) and related institutions of the broader
rity Council sanctions for refusing to stop its
nonproliferation regime influence incentives
uranium enrichment program, and more na-
for and against nuclear weapons acquisition. In
tions sought civilian nuclear power plants to
addition, this traditional division of the subfield
produce electricity and combat global warm-
into supply and demand problems has in itself
ing. The connections between the spread of
focused scholarly attention away from under-
civilian nuclear power and the proliferation of
standing the relationship between supply and
nuclear weapons, however, are not clear-cut.
demand, between a state’s increased nuclear

The physical connection is simple: The fuel
“latency” and its incentives for and against de-
and by-product of light-water nuclear power
veloping nuclear weapons.

reactors—enriched uranium and plutonium—
There have been many positive develop-
can also be used to produce nuclear weapons.
ments in the past decade, as new theories have
From a technical perspective, as Nobel
been developed and scholars have increasingly
Laureate Hanes Alfvén noted, “Atoms for peace
used multiple methods to test their theories.
and atoms for war are Siamese twins” (as quoted
Despite this progress, however, the new quan-
in Miller 1979, p. 19). The political connec-
titative and qualitative research projects on the
tions between the growth nuclear power capa-
causes of proliferation have been limited by
bilities and nuclear proliferation risks, however,
the information available about the very small
are more complex. Governments that develop
number of cases of nuclear weapons prolifera-
specific technologies related to civilian nuclear
tion. Indeed, there is a deep and abiding uncer-
power facilities may be more capable of devel-
tainty about the basic data on which states have
oping nuclear weapons, but are they therefore
sought nuclear weapons and when they started
more likely to do so?
and ended their programs. We must therefore
Although it is premature to claim, as some
improveon the historical proliferation data sets,
scholars and nuclear industry officials have
and yet, given the secrecy surrounding nuclear
done, that a “nuclear renaissance” has started
programs, some uncertainties are likely to re-
with respect to the spread of nuclear power
main. To move forward beyond that, we will
plants to new countries around the globe
also need more multidisciplinary research, with
(Miller & Sagan 2009, Lester & Rosner 2009,
political scientists developing better under-
Lavergeon 2009, Socolow & Glaser 2009),
standings of the technology of nuclear power
there has been a renaissance of scholarly inter-
and nuclear weapons and also the effectiveness
est among political scientists in explaining the
of international law and export control regimes.
causes of nuclear weapons proliferation. The
This critical review of the political science
new political science literature on nuclear pro-
literature on the causes of nuclear weapons pro-
liferation and nonproliferation should help us
liferation consists of four parts. The first sec-
understand the crucial questions about states’
tion below briefly presents what we know about
capabilities and motives for building the bomb,
which states developed nuclear weapons, and
but it has thus far been of limited utility. In the
which states started weapons development pro-
literature, the technical capability to develop
grams but then abandoned them. I highlight
nuclear weapons is called the “supply-side” as-
the problems that result from remaining un-
pect of proliferation, and a government’s mo-
certainty about the accuracy and completeness
tivation to develop nuclear weapons is called
of the existing data. The second and third sec-
the “demand-side” aspect of proliferation. In
tions review the supply-side literature. Here I
226 Sagan
argue that political scientists have largely been nuclear programs but nevertheless failed to
going down a dead-end street that has led to lit- develop the bomb. Scholars have thus usefully
tle robust knowledge about the relationship be- divided nuclear proliferation into phases
tween the development of nuclear power pro- (explore, pursue, and acquire), even though
grams and nuclear weapons acquisition. The in practice it can be difficult to determine
fourth and fifth sections review the literature when a state actually falls into these categories.
on the demand for nuclear weapons. Here I Figure 2 (see color insert) displays my esti-
conclude that a mixture of comparative case mates of when states started exploring nuclear
studies and statistical studies has improved our weapons programs, and when they abandoned
understanding of the diversity of motives for their nuclear weapons development efforts
weapons development and restraints, but that (with the exceptions of Iran and Syria, which are
serious gaps in our knowledge remain. The coded as continuing a covert nuclear weapons
sixth section outlines alternative perspectives program). There is no single explanation

on the role of the NPT in nuclear decision for why states give up their bomb programs
making and presents evidence on differences (Campbell et al. 2004, Levite 2003, Paul
between democratic and nondemocratic states’ 2000, Reiss 1995, Rublee 2009). Some of
compliance with the treaty. The final section these “nuclear reversal” cases were U.S. allies
lays out a future research agenda to address the who received strong diplomatic pressure from
weaknesses in our current understanding of the Washington to stop suspected covert programs
causes of nuclear proliferation. (Taiwan, South Korea); another resulted from
“disarmament by force” (Iraq). Some govern-
ments dismantled their bomb programs after
NUCLEAR WEAPONS AND or in anticipation of domestic regime change
NUCLEAR WEAPONS PROGRAMS (Romania, Brazil, South Africa); others ended
Figure 1 (see color insert) presents a famil- their pursuit of nuclear weapons when an ally
iar graph of when different states acquired nu- terminated joint nuclear weapons development
clear weapons, from the first U.S. nuclear test in agreements (Italy, Germany). But many acts of
1945 through the North Korean test in 2006. both nuclear reversal and abstinence appear to
It shows a steady but slow increase. One new be more voluntary in nature and, as discussed
nuclear weapons state has emerged about every below, the emergence of the NPT in 1968
five years, and—with the exceptions of South and the later addition of related institutions,
Africa and the three former Soviet Union states such as the Nuclear Suppliers Group (NSG),
that inherited the weapons (but never had op- appear to have influenced many decisions to
erational control over them)—each state has refrain from developing military applications
kept its nuclear weapons once it has them. of nuclear power programs.
There is virtually no disagreement about what It is important to note, however, that sig-
states have acquired the bomb, although there nificant uncertainty exists about whether and
are still uncertainties about the precise year when to code particular states as exploring or
in which “second-generation” nuclear weapons pursuing the bomb. This uncertainty exists in
states—Israel, India, Pakistan, and North part because many governments pursuing nu-
Korea—first acquired an operational nuclear clear weapons place veils of secrecy around their
weapon. programs; in other cases, however, uncertainty
We are not only interested in which states about a government’s nuclear weapons pro-
eventually acquired nuclear weapons, however, gram exists because the government itself never
but also need to study what governments have determined whether early nuclear-related re-
taken preliminary steps to develop nuclear search was to be used only for future power
weapons and then stopped such efforts, and program development or for future nuclear
what governments had serious and sustained weapons development.
www.annualreviews.org • Causes of Nuclear Weapons Proliferation 227
The small-numbers problem and the lin- THE SUPPLY-SIDE LITERATURE

gering uncertainty about the accuracy of the
Ever since John F. Kennedy predicted in 1963
data combine to create enormous difficulties
that there might be as many as “fifteen, twenty,
for both the quantitative and case-study liter-
or twenty-five nuclear weapons powers” in the
atures on nuclear proliferation. In the quanti-
1970s, statesmen and scholars alike have been
tative literature, the relatively small number of
surprised that the pace of nuclear prolifera-
“confirmed” states that have tried to develop the
tion has been so much slower than predicted
bomb has meant that the discovery of additional
(Potter & Mukhatzhanova 2010). Indeed,
states, or the removal of states from the data
Hymans (2010, p. 13) argues that “the basic
set because of new “disconfirming” evidence,
puzzle facing the study of nuclear prolifera-
can significantly reduce the robustness of many
tion is why there is such a large and persistent
authors’ findings about the causes of prolifera-
gap between the technically nuclear weapons–
tion (Montgomery & Sagan 2009, 2011). For

capable states—at present there are perhaps
example, Montgomery & Sagan 2011) show
as many as fifty—and the small number of
that adding new confirmed cases of states that

actual nuclear weapons states.” Much of the
sought nuclear weapons to the data set devel-
new supply-side literature has been obsessed
oped by Jo & Gartzke (2007a) significantly re-
with this puzzle. The underlying assumption
duces the magnitude of, and in some cases even
here, that more and more states have become
reverses, their statistical findings about correla-
“nuclear weapons capable,” however, depends
tions between a state’s membership in the NPT,
crucially on how one defines and measures nu-
the nuclear threats of its adversary, the diffu-
clear weapons capability.
sion of nuclear technology, and the status of
What does it mean for a state to be capa-
that state’s nuclear weapons program.1
ble of building nuclear weapons? The central
For the qualitative literature, the deep se-
weakness of the coding rules by which political
crecy surrounding many governments’ nuclear
scientists have labeled states “nuclear capable”
weapons decisions means that reasonable schol-
is clear once one delves into the methodology
ars can still defend conflicting interpretations of
used in their studies (Sagan 2010). The current
when and why particular governments chose to
generation of proliferation specialists have built
move forward or contain their nuclear weapons
on the methodology and coding rules used in
programs. For example, scholars disagree about
Meyer’s (1984) pioneering book The Dynamics
whether Argentina’s nuclear program was de-
of Nuclear Proliferation and Stoll’s (1996) up-
signed to fuel naval reactors for a submarine
date of the Meyer data through 1992. Meyer
fleet or for a nuclear weapon (Hymans 2001,
(1984) carefully measured a set of ten tech-
Müller & Schmidt 2010), while other observers
nical and economic indicators—national min-
continue to debate whether the Japanese gov-
ing activity, indigenous uranium deposits, met-
ernment has ever actively pursued a nuclear
allurgists, steel production, construction work
weapons option (Campbell & Sunohara 2004,
force, chemical engineers, nitric acid produc-
Japan Times 2010). In short, the small num-
tion, electrical production capacity, nuclear
ber of cases and remaining uncertainties about
engineers, physicists, chemists, and explosives
key historical decisions present a significant and
and electronics specialists—to produce what he
continuing challenge to understanding when
called “a list of nations with latent capabilities
and why proliferation occurs.
to manufacture nuclear weapons” (p. 41). Be-
cause he could not measure directly whether
the quantity or quality of a state’s nuclear engi-
1
The cases added were Italy and West Germany 1957–1958 neers and its explosives and electronics special-
(Nuti 1993, Müller 2003), Egypt 1960–1967 (Rublee 2009),
Australia 1956–1972 (Walsh 1997), and Libya 1970–2003 ists were sufficient to build a nuclear weapon,
(Bowen 2006, Braut-Hegghammer 2009). Meyer used two proxy indicators: whether the
228 Sagan
state had been operating a research reactor for state with uranium deposits on its territory or a
three years (the proxy for nuclear engineer- state that has acquired “produced uranium” for
ing expertise) and whether the state manufac- a research or power reactor. (See Table 1 for a
IAEA: International
tured automobiles, or assembled automobiles comparison of the various coding schemes.) Atomic Energy
and manufactured radios and television sets (the Yet, although Jo & Gartzke (2007a, p. 169 Agency
proxy for expertise in explosives and electron- correctly note that “states that lack the basic
ics). Based on this particular set of indicators, material capabilities will be excluded from the
Meyer found that 34 states held the latent ca- group of potential proliferators,” their coding
pability to build nuclear weapons in 1982. rules do not in fact treat the possession of fissile
Stoll (1996) updated the Meyer data set, with materials (enriched uranium or plutonium) as a
a hidden yet significant change in coding rules, necessary but not sufficient condition for build-
in the mid-1990s: Whereas Meyer had mea- ing nuclear weapons. Instead, Jo & Gartzke
sured indigenous uranium sources, Stoll sim- implicitly assume that “where there is a will,
ply assumed that all states had access to nu- there is a way” because their coding rules lead
clear materials, claiming that they were now to states being labeled as “nuclear weapons ca-
freely available in the global marketplace. Stoll pable” without having the necessary fissile ma-
thus assumed away the crucial technical bottle- terials to make even one bomb. The evidence
neck of whether a state has access to uranium that Jo & Gartzke’s coding rules do not ade-
that, once enriched, could be used in a nuclear quately capture the necessary conditions for nu-
weapons program. Based on the resulting data clear weapons development is apparent in the
set, Stoll argued that 48 states had a latent nu- fact that their data set coded North Korea as
clear weapons capability in 1992 (Figure 3; see not having full capability to develop nuclear
color insert). weapons in 2001 (it still lacked sufficient chem-
Hymans (2004, p. 4; 2010) accepted the ical engineers, nitric acid production capability,
logic of that argument, updated Stoll’s pro- and explosives specialists), even though North
liferation estimates, and claimed that the data Korea was a major exporter of long-range mis-
demonstrate a “yawning gap between technical siles at the time and the North Koreans were
potential and military reality,” expressed by the known to have separated plutonium from the
contrast between the number of states that have fuel rods of the Yongbyon reactor (May 2001).
the capability to produce nuclear weapons and The Jo & Gartzke coding rules also led to the
the number that have actually done so. A state odd conclusion that South Africa lacked the
cannot make a nuclear weapon, however, unless capability to build nuclear weapons in 2001—
it has highly enriched uranium or plutonium even though it had built six nuclear weapons in
from a large reactor, and Stoll’s hidden assump- the 1980s and, after dismantling the weapons
tion that any state could acquire uranium on the in the 1990s, still maintained 450–600 kg of
open market, coupled with his use of research highly enriched uranium according to Interna-
reactor experience as the measure of required tional Atomic Energy Agency (IAEA) safeguard
nuclear engineering expertise, essentially inspections (Purkitt et al. 2002). (Within Jo &
removed those two technical constraints. Gartzke’s coding rules, South Africa did not
Jo & Gartzke (2007b) alter the Stoll cod- have the “necessary” chemical engineers and ni-
ing scheme by dropping three of the Stoll and tric acid production capacity.)
Meyer indicators (construction workforce, steel By not focusing attention on enriched ura-
production, and previous national mining ac- nium and plutonium, the weak proxy measures
tivity) on the grounds that they are “too eas- for nuclear weapons capability used by Jo &
ily available to be thresholds” and modifying Gartzke clearly lead to bizarre results. Trinidad
the coding for necessary “uranium deposits” and Tobago (which “only” lacks uranium de-
(which, as we have seen, were assumed by Stoll posits, “produced uranium,” and any research
to be available for all states) to include either a reactor experience) is assessed to have had a
Table 1 Comparison of Meyer/Stoll (Meyer 1984, Stoll 1996) and Jo & Gartzke (2007b) nuclear capability indicators
Definition of indicator
Indicator Meyer/Stoll Jo & Gartzke
Mining activity Some fraction of labor force in mining activity ———
Uranium deposits Known uranium deposits (Meyer) Known uranium deposits or produced
Assumed market access (Stoll) uranium already
Metallurgists Production of crude steel Production of crude steel or aluminum
Steel Production of crude steel ———
Construction workforce Production of steel and cement ———
Chemical engineers Production of nitric acid or sulfuric acid Production of nitric acid or sulfuric acid
Nitric acid production capacity Nitric acid production or sulfuric acid Nitric acid production or sulfuric acid
production and nonorganic nitrogenous production and nonorganic nitrogenous
fertilizer production fertilizer production

Electricity production capacity Installed electrical capacity of 200 MWe Installed capacity of 200 MWe or
produces equivalent of 50,000 metric

tons of oil
Nuclear engineers, physicists, chemists Three research-reactor years Three research-reactor years
Specialists in electronics and explosives Manufacture of motor vehicles, or assembly Manufacture or assembly of motor
of motor vehicles and manufacture of radios vehicles and manufacture of radios or
or televisions televisions
higher degree of nuclear weapons latency in science frequently conflates two analytically dif-
2001 than is North Korea, which was only five ferent phenomena under the same labels of “nu-
years away from detonating its first nuclear clear weapons capability” or “nuclear latency”:
weapon. Egypt, which had only two research first, what should properly be called “nuclear
reactors in 2001, is assessed to have a higher de- self-sufficiency”—a measure of how indepen-
gree of nuclear weapons latency than is South dent or self-sufficient a potential long-term nu-
Africa ( Jo & Gartzke 2007b). clear weapons program could be; and second,
These problems are examples of a common “nuclear latency”—a measure of how quickly a
weakness in the quantitative political science lit- state could develop a nuclear weapon from its
erature on proliferation (Montgomery & Sagan current state of technological development if it
2009, 2011). All too often the authors look for chose to do so (Sagan 2010).
the keys under the lamppost: They use “proxy In short, by focusing our attention away
variables” that are easily available, rather than from the acquisition of the necessary fissile ma-
collecting the data that reflect the substantive terials, the supply-side proliferation literature
variables of real interest. For example, scholars has led to a grossly exaggerated estimate of how
interested in measuring nuclear “weaponization many states currently have the technical capa-
capability” could have built original data sets on bility to build nuclear weapons. Indeed, the ap-
related arms-manufacturing experience and ex- propriate proliferation puzzle may not be why
pertise, but instead they used preexisting data so many “nuclear weapons capable” states have
on automobile and television manufacturing. refrained from building nuclear weapons, but
Furthermore, there are often hidden, but cru- why so few states have acquired the facilities and
cial, assumptions in this literature, such as “the technology needed to enrich uranium or repro-
open market enables all states to get nuclear cess plutonium. In 2010, ten states are known
materials,” that have a strong but unacknowl- to have uranium enrichment facilities (World
edged impact on the findings. Finally, the new Nuclear Association 2010), and eight states,
quantitative proliferation literature in political only one of which is a non–nuclear weapons
230 Sagan
state (NNWS), are believed to have national likely to do so” (2009a, p. 176)—is certainly
plutonium reprocessing facilities (International true but hardly counterintuitive. Indeed, 5 of
Panel on Fissile Materials 2010). Why has there Kroenig’s 14 cases include not just assistance
NNWS: non–nuclear
been so little proliferation of facilities for the in building enrichment or reprocessing facili- weapons state
production of fissile materials? Answering this ties but help in the design of nuclear weapons
question would require deeper investigations themselves. The fact that the states that were
into the economics of nuclear power and more given bomb designs were more likely than other
research on the development of nuclear export states to later acquire nuclear weapons is hardly
control regimes than political scientists have surprising.
conducted to date. Kroenig’s important contribution to our
understanding of proliferation, however, is
on the causes of sensitive nuclear assistance
NEW SUPPLY-SIDE RESEARCH:
(2009b, 2010). Why do states sell or give

BRINGING BACK THE FUEL sensitive nuclear assistance to others, given the
Two new contributions have helped improve likelihood that this might encourage weapons
the literature by focusing precisely on the proliferation? Most scholars and policy makers
technologies needed to link nuclear power have assumed that governments do this for
to nuclear proliferation: nuclear exports and economic reasons or because they fail to predict
technology agreements regarding nuclear fuel the nuclear proliferation consequences of their
production, reprocessing technology, nuclear nuclear technology export policy. Kroenig,
engineering training, and power and research however, finds that this is rarely the case.
reactors. First, Kroenig (2009a,b, 2010) argues Instead, governments often provide sensitive
that international sensitive nuclear assistance assistance for political and strategic reasons,
contributes to the spread of nuclear weapons. and this is true because of, rather than despite,
Kroenig defines sensitive nuclear assistance as the anticipated proliferation consequences.
giving information on weapons design or con- Governments are more likely to transfer sen-
struction, providing large amounts of weapons- sitive technologies into regions in which they
grade fissile material, or assisting in the con- cannot project conventional military power
struction of facilities for uranium enrichment and to states that are rivals of their enemies.
or plutonium reprocessing that could be used “In other words, states may calculate that the
to produce weapons-grade fissile material. He enemy of their enemy is their best customer,”
tests his theory both statistically and through Kroenig (2010, p. 38) tersely argues. This is an
three case studies of countries that acquired important (and disturbing) finding.2
nuclear weapons after having received sensitive The second new contribution to the supply-
nuclear assistance: Israel, China, and Pakistan. side proliferation literature is the work of
Kroenig (2009a, 2010) finds that sensitive Fuhrmann (2009a,b). In contrast to Kroenig,
nuclear assistance is positively correlated to ac- Fuhrmann (2009a, p. 8) claims that the spread
quisition of a nuclear bomb, as is gross domestic of all types of peaceful nuclear technology, not
product, the state’s overall industrial capacity, just “sensitive” nuclear technology, increases
rivalry with nuclear weapons states, and regime the likelihood of proliferation: “The conven-
type. He carefully traces the importance of sen- tional wisdom is wrong—and dangerous. All
sitive nuclear assistance in the history of his types of civilian nuclear assistance raise the risks
three case studies and also examines other cases
of states acquiring such assistance and exploring 2
It is also a controversial finding, for it in part depends on the
weapons programs. His central conclusion— accuracy of Kroenig’s argument that the activities of the A.Q.
“states that have the ability to produce nu- Khan network were “state-sponsored by any reasonable def-
inition of the term” and that the common enemy of Pakistan
clear weapons, either through international as- and its nuclear customers was the United States (Kroenig
sistance or domestic capacity, are much more 2010, pp. 135, 140).
of proliferation. Peaceful nuclear cooperation a country’s interest in nuclear weapons tech-

and proliferation are causally connected be- nology would make participation in an NCA
cause of the dual-use nature of nuclear technol- more likely, and thus it could be that NCAs are
ogy and know-how.” Fuhrmann acknowledges not causing later weapons programs, but rather
that the vast majority of states that have re- that plans for nuclear weapons programs are
ceived civil nuclear assistance agreements have motivating NCAs. The Singh & Way (2004)
not acquired weapons (in 99.77% of country- data set, which Fuhrmann uses for coding pur-
year observations, states receiving civilian nu- suit and acquisition, also includes a measure
clear assistance did not acquire the bomb), but for when states first explored nuclear weapons
he also insists that there is a strong statisti- that he does not utilize. Of Fuhrmann’s five
cal and causal link between the number of nu- case studies, three countries—Israel, Pakistan,
clear cooperation agreements (NCAs) and the and India—received substantial nuclear assis-
likelihood that a country will initiate a nuclear tance after they were already coded by Singh
weapons program and eventually acquire the & Way as exploring, but not pursuing, nuclear
bomb. Fuhrmann asserts that “nuclear coopera- weapons. North Korea received aid from the
tion strongly influences whether a country goes Soviet Union at about the same time it began
down the nuclear [weapons] path. Participation exploring. Only South Africa received all of its
in at least one nuclear cooperation agreement aid before exploring in 1969, and it is notable
increases the likelihood of beginning a bomb that this assistance was provided in the 1950s,
program by about 500%” (2009a, p. 32). before the establishment of the NPT regime.
Fuhrmann’s central insight—that a state Third, Fuhrmann uses aggregate NCAs as
may acquire dual-use technology with only the crucial independent variable, which is prob-
peaceful intent, but then succumb to the lematic. This variable places a great deal of
temptation to initiate weapons research when weight on countries, like India, that participated
international threats emerge—is an important in a large number of NCAs prior to acquisition
one. However, there are three problems of nuclear weapons. Fuhrmann also includes a
with his analysis. First, his estimate about number of cases in which NCAs were offered to
the substantive effects of NCAs on initiation a state that was known to be exploring a nuclear
of weapons programs is overstated. In his weapons option in an attempt to provide eco-
crosstabular analysis, 3 states are coded as nomic carrots to keep the state from pursuing
having pursued nuclear weapons without nuclear weapons to completion. And Fuhrmann
NCAs (0.07% of observations) and 12 states includes NCAs that were negotiated and signed
pursued weapons after receiving NCAs (0.42% but then canceled before coming into fruition
of observations). Although technically this (Kroenig in Bluth et al. 2010).
does mean that states with NCAs are >500%
more likely to start nuclear weapons programs
than are states that did not have NCAs, the THE DEMAND-SIDE
likelihood of any state, even after signing an LITERATURE
NCA, pursuing nuclear weapons is still less Different studies of why states are interested in
than half a percent. Moreover, the lag time acquiring nuclear weapons examine different
Fuhrmann includes in his pursuit variable drops incentives and often focus on different actors
Russia and the United States from this analysis who influence government decisions. Early
entirely, further overemphasizing this effect. work on the demand for nuclear weapons
Second, there is an endogeneity prob- focused on the utility of using international
lem: Fuhrmann did not adequately control for relations theories to derive alternative testable
whether the countries in question were already hypotheses. Solingen’s 1994 article “The Polit-
exploring the nuclear weapons option when ical Economy of Nuclear Restraint” contrasted
they signed NCAs. It seems reasonable that neorealist predictions about the importance
232 Sagan
of threats to her own prediction that “ruling (2007a) probit model finds strong support
coalitions pursuing economic liberalization for status-driven motives and somewhat less
are more likely to embrace regional nuclear support for domestic political and international
regimes than their inward-looking, nation- security motives, with an especially interesting
alist, and radical-confessional counter-parts” counterintuitive finding that rivalry with a
(p. 136). Sagan’s 1996 article “Three Models nuclear weapons state is negatively correlated
in Search of a Bomb” compared the influence with both program and acquisition. Bleek’s
of security threats, domestic political interests, (2010) hazard model, using an updated data set
and international norms on nuclear prolifer- of states that explored, pursued, and acquired
ation decisions. The first of the three models the bomb, finds in contrast that there is no
focused on military security motivations; the correlation between states’ decisions to pursue
second emphasized domestic political actors or nuclear weapons or eventually acquire them
bureaucratic coalitions that form to support or and one or more of that state’s enduring rivals
oppose nuclear weapons for parochial reasons; having pursued or acquired nuclear weapons.
and the third focused on norms and prestige Singh & Way (2004), Jo & Gartzke (2007a),
considerations that can encourage or discour- and Bleek (2010) all do find that security
age acquisition of nuclear weapons (Sagan guarantees from a nuclear power have weak
1996/1997, 1999). Sagan found strongest or no effect on a state’s decision to explore a
support for the security model, but he argued nuclear weapons option, yet this may be due
that domestic interests and normative concerns to selection effects: It seems reasonable to
about prestige were “sufficient, but not neces- assume that a nuclear weapons state will be
sary” conditions for proliferation in a limited more likely to offer a security guarantee to
number of cases. a country either facing a significant nuclear
Scholarly debate has continued for two threat or exploring nuclear weapons options.
decades about how to weigh these different Thus, security guarantees may have a stronger
motivations for proliferation. Some scholars causal effect in the real world than is captured
argue for the prime importance of security by the correlations in the observed data.
motivations (Mearsheimer 1990, Frankel 1993,
Frankel & Davis 1993, Thayer 1995, Paul
2000, Hecker 2010), others for the importance NEW DEMAND-SIDE RESEARCH:
of the constraints or incentives stemming from BRINGING BACK LEADERSHIP
domestic economic interests (Lavoy 1993, Three recent case-study works by Abraham,
Solingen 1998, Liberman 2001), and some Hymans, and Solingen have added significantly
constructivist scholars focus on normative to our understanding of specific national and
constraints and changing identities of indi- regional cases of proliferation and nonprolifer-
vidual leaders or governments (Chafetz et al. ation by developing new theoretical constructs.
1996, Katzenstein 1996, Grillot & Long 2000, Each of these scholars focuses attention on the
Tannenwald 2007, Rublee 2009). interests and strategic preferences of govern-
The new quantitative studies of prolifer- ment leaders and tests his or her theories against
ation have assessed states’ motivations using a limited set of cases. It will be especially valu-
different data sets, different coding rules for able, therefore, for future researchers to test
key variables, and different statistical methods, these theories more broadly.
all of which could add to the robustness of their Abraham has not only written one of the
results if there were common findings. Unfor- best studies of the Indian nuclear power and
tunately, that is not the case. Singh & Way’s weapons program but has also introduced the
(2004) hazard model finds strong support for important concept of “nuclear ambivalence”
both enduring rivalries and militarized disputes to the subfield (Abraham 1998, 2006, 2009).
as causes of proliferation, while Jo & Gartzke’s Abraham argues that the field’s obsession with
nuclear weapons proliferation leads to a deep if they had the opportunity to get them; Hy-
misunderstanding of the relationship between mans’ central argument turns this assumption
civilian nuclear power and nuclear weapons. on its head. Very few leaders want nuclear
Most scholars (like most U.S. intelligence an- weapons, he conjectures, because building the
alysts) have examined the kinds of nuclear fa- bomb is “a revolutionary decision” and “a leap
cilities a country developed and have tried to into the dark.” The decision makers cannot
discern whether the government is really build- know whether the state will be able to build
ing only a civilian energy capability or is se- an arsenal, nor whether the strategic effect will
cretly seeking to develop a latent or breakout enhance or damage national security. Nuclear
nuclear weapons capability. Thus, scholars of- proliferation is therefore rare, not because of
ten write about “nuclear ambiguity” (is the gov- institutional constraints or alliance pressures,
ernment seeking weapons or not?) or “nuclear but rather “because few state leaders may in fact
opacity” (is the government hiding nuclear fa- need to be constrained from seeking the bomb”
cilities to cover its intentions?) when assess- (p. 12). Hymans develops a psychological the-
ing the meaning of specific programs (Frankel ory, placing causal importance on what he calls
1991, Frankel & Davis 1993). Abraham argues, the “oppositional nationalist” identity of some
in contrast, that government leaders may not leaders, who see their nation as naturally in a
have strong and well-developed intentions in hostile relationship with other states and see
the nuclear field at all. They may suffer from themselves as inherently superior to (or at least
basic ambivalence; they could be persuaded to the equal of) their enemies. For such leaders,
push for weapons or for civilian power, depend- getting the bomb is not just a matter of a ra-
ing on a wide range of political and social factors tional cost–benefit calculation of state or even
that emerge long after they have constructed party self-interest: “Driven by fear and pride,
the physical and organizational foundations for oppositional nationalists develop a desire for
their nuclear program. Abraham tests his the- nuclear weapons that goes beyond calculation,
ory largely against evidence of the diversity and to self-expression” (Hymans 2006, p. 2).
fluidity of motives seen inside the Indian nu- Hymans’ case studies are particularly rich.
clear program between 1945 and the Pokhran He uncovers, for example, important new his-
nuclear weapons tests in 1998. He places prime torical evidence suggesting that Argentina’s
importance on the domestic power and politi- uranium enrichment program, contrary to the
cal autonomy of what he calls the “strategic en- common coding in the literature, may have
clave” of probomb scientists and government been designed to produce nuclear fuel for a sub-
bureaucrats inside the Indian nuclear establish- marine fleet instead of being part of a covert
ment (Abraham 1998; also see Bajpai 2009). Still nuclear weapons program. This is a fine ex-
he argues that the phenomenon of “nuclear am- ample of how a case study can contribute to
bivalence” is common inside other capitals and building and revising the data sets needed for
nuclear bureaucracies, especially in the less de- broader tests of theories of proliferation. And
veloped world. He further warns that nonpro- to his credit, Hymans does not rely on his his-
liferation policies to dissuade ambivalent actors torical case studies of nuclear decision making
not to do what they have no intention to do can in India, Argentina, France, and Australia to
backfire, setting off national or bureaucratic re- measure his key independent variable, which
sistance to external pressures. would be especially problematic for such an in-
Hymans (2006) makes a similarly creative ductively derived theory. Instead, he develops
theoretical contribution, and his detailed case a complex measure of “national identity con-
histories of rarely studied incidents of prolif- ception” and uses a quantitative content anal-
eration and restraint add significantly to our ysis of political leaders’ major public speeches
knowledge. Realism assumes that most lead- (such as a state-of-the-union address) to assess
ers would want nuclear weapons for security whether the leader can be accurately coded as an
234 Sagan
oppositional nationalist. This provides a check integration” model that focuses on the degree
to reduce the danger that Hymans might use to which the government bases its domestic
his knowledge of the final outcome to influence political survival on economic integration
his coding of the leaders’ national identity. into the global economy—against alternative
Although both Abraham and Hymans have realist (security factors), neo-institutionalist
developed valuable new theories and have (the NPT), and norms-and-constructivist (the
conducted strong, albeit limited, tests of their nuclear taboo and global citizenship) theories.
ideas, their work shares a weakness: Not only Her findings strongly support her global eco-
do they themselves not test their theories on nomic integration theory: States whose leaders
a wider set of cases, but it is also not clear or ruling coalitions advocate integration into
how they or other scholars could do so given the global economy (mostly in East Asia) reject
the deliberately covert nature of many nuclear nuclear weapons development, whereas leaders
weapons programs. Abraham may be right or coalitions that favor inward-looking reliance
about some governments or leaders holding on domestic markets and nationalist values
ambivalent positions on nuclear matters, but (mostly in the Middle East) are more prone to
how could outside scholars measure the degree pursue nuclear weapons.
to which this is the case when decisions are The strength of case studies is the detailed
being discussed and made behind a veil of tracing of the causal influences of economic in-
secrecy? Moreover, scholars have not yet de- terests, but the weakness of this methodology
veloped compelling measures of bureaucratic is its inability to assess broader generalizabil-
power and autonomy that could help quantita- ity of the theory. Other scholars’ quantitative
tive researchers determine whether the Indian tests of Solingen’s global economic integration
“strategic enclave” phenomenon is an outlier model have produced mixed findings. Singh &
or a more common occurrence (Sagan 2009a, Way’s (2004) hazard model tested the effects of
Sasikumar & Way 2009). Similarly, Hymans two independent variables regarding a state’s
may be right about the influence of oppositional integration into the global economy—a state’s
nationalism in the cases he has studied, but “interdependence” as measured by trade ratios
how can an outside analyst accurately measure (exports and imports as a share of GDP) and
such psychological characteristics for specific a state’s “liberalization” as measured by move-
leaders making nuclear decisions who do not ment toward or away from higher trade inter-
make many public announcements, or whose dependence. Singh & Way found that higher
positions are strategically misrepresented to trade ratios were negatively and significantly
foreign or domestic audiences for bargaining correlated to all three measures of prolifera-
purposes? In short, Abraham and Hymans raise tion (exploration, program, and acquisition),
important theoretical questions and provide but that economic liberalization was insignif-
valuable empirical understanding of specific icant at all three stages. In fact, in the explore-
cases, but it will be a major challenge for future and-acquire phase, liberalization had a weak
scholars to assess the broader strength of their positive correlation. Singh & Way’s quantita-
theories about nonproliferation. tive analysis thus offers only tepid support for
Finally, Solingen’s Nuclear Logics: Con- Solingen’s theory. Bleek (2010) uses a slightly
trasting Paths in East Asia and the Middle different data set on states that sought and ac-
East (2007) compares the nuclear behavior quired the bomb, but the same proxy measure
of states in those two regions, which had for economic interdependence and liberaliza-
similar qualities in the early 1970s but have tion as Singh & Way, and also reports mixed
since experienced different nuclear trajecto- findings. However, they are not the same mixed
ries. Solingen’s focused regional comparison findings: Economic interdependence is nega-
methodology uses detailed case studies in tively correlated with the exploration of nu-
order to test her preferred theory—a “global clear weapons, is not significantly correlated to
pursuit, and is positively correlated to nuclear THE NPT, REGIME TYPE, AND
weapons acquisition. Like Singh & Way, how- NUCLEAR PROLIFERATION
ever, Bleek finds that economic liberalization is
Despite this lack of consensus about the central
not significantly correlated with proliferation at
causes of nuclear proliferation, there are two
any stage.
specific agreements in both the case-study and
Solingen’s case studies are impressive and
the quantitative political science literature:
have produced counterintuitive findings. Her
(a) the NPT does not have significant effects on
detailed case study of Japanese nuclear decision
the likelihood of proliferation and (b) regime
making in the 1970s, for example, reverses the
type has only a minimal effect on proliferation.
usual causal arrow regarding the relationship
Because of the (false) consensus that the
between security guarantees and nonprolif-
capability to build nuclear weapons is already
eration commitments. In “a fine example of
widely available, most political scientists
intellectual judo” (Hymans 2010), Solingen

studying proliferation are highly skeptical of
presents evidence that Japanese leaders, who
the potential effect of efforts to control the
were already opposed to Japanese nuclear

supply of nuclear technology. Hymans (2006,
weapons on economic grounds, successfully
p. 220) argues that “given the widespread
sought tighter U.S. security assurances in order
diffusion of nuclear capacities, supply-side
to justify their antinuclear positions in domestic
control measures against potential proliferant
debates in Tokyo. In addition, Solingen’s focus
states are clearly of declining utility.” Jo &
on what she calls changing “world time” in
Gartzke (2007a, pp. 185–86) maintain that “the
proliferation—the 1968 NPT is seen as a water-
inhibiting effect of the NPT is overcome by
shed that alters the incentives and disincentives
the stronger technological diffusion effect” and
for states to seek nuclear weapons—is an
conclude that “enthusiasm for the NPT among
important warning to quantitative researchers
proliferation opponents thus appears to be mis-
not to expect identical patterns of proliferation
placed.” Solingen (2007, p. 31) also expresses
and restraint over the entire nuclear age.
skepticism about the effectiveness of the treaty:
This focus on the post-1968 NPT world
“Would more states have opted for nuclear
time, however, makes it more puzzling that
weapons had the NPT never been concluded?
Solingen denigrates the role of the treaty and
Not necessarily . . . ” Betts (2000, p. 69) makes
does not examine whether the NPT was nec-
the most extreme argument, from a realist
essary for “liberalizing” governments to be
perspective, about the irrelevance of the NPT:
concerned that movement toward a nuclear
weapons program would lead to international As useful as treaties are, it is a misconcep-
sanctions or other restrictions on the poten- tion to see them as a solution. They are
tial benefits from integrating into the global effects of nonproliferation, not causes of it.
economy. In addition, Solingen does not fo- The NPT and CTBT (Comprehensive Test
cus sufficient attention on the nuclear indus- Ban Treaty) reflect the intent of their ad-
try in NNWSs, which can be major engines of herents to abjure nuclear weapons. To date,
global economic integration, if they feel that the countries considered problematic—those
their access to foreign technology and fuel will that might acquire nuclear weapons—simply
be restricted if there is any suspected weapons- did not join the NPT (South Africa stayed
related research in their nations. As noted be- out while it had a nuclear weapons pro-
low, the position of the nuclear industry may gram and joined when it decided to get
be one of the most important determinants of rid of it). Or else they joined and cheated
whether a government eventually moves toward (Iraq and North Korea). . . . If the NPT
developing nuclear weapons options or toward or CTBT themselves prevent proliferation,
becoming an even stronger supporter of nuclear one should be able to name at least one
nonproliferation. specific country that would have sought
236 Sagan
nuclear weapons or tested them, but refrained This relates, however, to the second area of
from doing so, or was stopped, because of ei- agreement in the literature, regarding regime
ther treaty. None comes to mind. type and proliferation. Democracies and autoc-
racies are seen to be similar in their proliferation
Betts’ statement is anachronistic. He ignores behavior; if anything, democracies are found to
how many states not considered “problematic” be slightly more likely to go nuclear than non-
today were in fact feared potential proliferators democracies. Jo & Gartzke (2007a) find that
in the past. In 1963, for example, National Intel- regime type makes no difference in whether
ligence Estimate 4–63 and a Robert McNamara a government initiates a nuclear weapons pro-
Department of Defense report predicted that gram, but that democracies are more likely than
eight countries could develop nuclear weapons nondemocracies to acquire nuclear weapons
by 1973: China, India, Sweden, Australia, Israel, once they have a program. The authors spec-
South Africa, Japan, and West Germany (Lavoy ulate that this is due to one of two causes:
2004). Egypt was considered to have “moder- “democracies are more vulnerable to national-
ate to high” motivation and a capability to ac- ist pressure” ( Jo & Gartzke 2007a, p. 179) or
quire nuclear weapons later in the 1970s; and “partial democratic states use nuclear weapons
Argentina, Brazil, Romania, Bulgaria, Hun- proliferation as a diversion for domestic politi-
gary, and Yugoslavia were all feared to be able cal reasons” (p. 184). Singh & Way (2004) also
to develop the bomb by the 1980s (Lavoy 2004). find that democracies are more likely to acquire
To understand the role of the NPT, one should nuclear weapons. The results of their hazard
not select on the dependent variable and con- model show that democracy makes states 25%
centrate only on recent “problematic” cases of more likely to explore, 95% more likely to ac-
states that acquired or try to acquire the bomb. quire, and slightly more likely to pursue nuclear
Instead, one should also study the broader set weapons. Their multinomial logit model offers
of cases of “nuclear abstinence” (states that re- similar results, with the exception that it sug-
frained from starting a weapons program) and gests that democracies are slightly less likely to
also examine the history of the many states that pursue nuclear weapons. Singh & Way (2004,
gave up their programs before they joined the p. 864) explain their findings on democracy by
treaty and those that cheated on their NPT arguing that “democratic governments may be
commitments, were caught, and were encour- tempted to pander to nationalist populations in
aged or forced to abide by their Article II com- an effort to boost their popularity and retain
mitment not to seek nuclear weapons. power.” Kroenig (2009b, p. 172) similarly notes
Moreover, the NPT included requirements that democracy measures are positively corre-
for NNWS members to accept safeguards— lated with nuclear weapons acquisition, arguing
inspections run by the IAEA—on their nu- that his model “provides support for the idea
clear power facilities as a precondition for re- that democratic states may be more prone to
ceiving peaceful nuclear assistance (Scheinman nuclear proliferation because they may be sub-
1987, Schiff 1984). This IAEA inspection sys- ject to pressure from domestic constituencies
tem has caught a number of states (including that favor nuclear development.” Fuhrmann
Iran, North Korea, South Korea, and Egypt) (2009a, p. 36) finds the coefficient for democ-
either cheating on their NPT commitments or racy to be positive, but not significantly so.
at least engaging in ambiguous but suspicious These findings oversimplify the role of
weapons-related activities (Acton 2009). Thus, regime type in determining nuclear prolif-
the NPT should not be seen as irrelevant be- eration behavior. Democratic countries have
cause some states have not complied with their certainly both pursued and acquired nuclear
commitments, for it is the treaty which sets up weapons, but the new quantitative literature
the inspections that permit us to know when has ignored the important observation that
cheating occurs. no NNWS democracy has cheated on its
commitments under the NPT. In all past Compliance and Violation,” discusses trade
cases, democracies that started and either aban- treaties, human rights treaties, and alliances,
doned or completed nuclear weapons pro- but says nothing about the NPT.
NWS: nuclear
weapons state grams either did so before the NPT came How exactly does the NPT work? Neolib-
into force or did not join the NPT at all eral institutional theory sees the NPT as a so-
(Müller & Schmidt 2010, Miller & Sagan 2009). lution to two collective action problems. The
Figure 4 (see color insert) reproduces the data NWSs find it useful not to provide nuclear
from Figure 2 on when different states started weapons to their allies, provided the other
nuclear exploration and nuclear programs, NWSs likewise refrain, and Article I of the
adding their NPT ratification dates and a mea- treaty enshrines this joint commitment. The
sure of their regime type based on the Polity NNWSs find it in their interest to constrain
IV data set (http://www.systemicpeace.org/ themselves from developing nuclear weapons
polity/polity4.htm). This figure clearly shows only if they can be reassured that their NNWS
that democracies have behaved differently neighbors are practicing similar constraints,
with respect to nuclear weapons prolifera- and Article II and the IAEA safeguards inspec-
tion. Democracies have both sucessfully de- tors provide precisely that reassurance (Sagan
veloped nuclear weapons and started but then 1996/1997, Dai 2007). An alternative “domes-
abandoned nuclear programs. But no demo- tic politics model” sees the NPT as promoting
cratic NNWS has ever started a covert nuclear “responsible” use of civilian nuclear power, and
weapons program after the government ratified the actors who are empowered by the treaty
the NPT. and manage nuclear power plants in their state
have parochial reasons to adhere to their treaty
NEW PROLIFERATION PUZZLES commitments (Sagan 1996/1997). The IAEA
AND RESEARCH AGENDAS creates an inspection mechanism that makes it
easier for governments to monitor their own
Social science research should help us better
scientists’ behavior, and the growth of nuclear
predict the likely nuclear future as well as ex-
power in the state increases the potential eco-
plain the nuclear past. But to do so, scholars will
nomic costs that powerful actors would face if
need to use multi-methods research designs to
sanctions were enacted because their govern-
take advantage of the strengths of both quanti-
ment was accused of having a covert weapons
tative and case-study methods, and also develop
program. In addition, the NPT can tip the do-
more accurate understandings of the technical
mestic balance of power in NNWSs as new
and institutional conditions under which nu-
bureaucratic actors favoring compliance with
clear proliferation and restraint decisions are
treaty obligations are created and strengthened
made. In particular, scholars will need to focus
through the NPT review conferences and re-
more attention on four related puzzles.
lated meetings. Ratifying the treaty can have
special strength in democracies if fear of paying
How the Non-Proliferation domestic audience costs encourages leaders to
Treaty Works maintain treaty commitments.
We have a strong literature on institutions A third, more “realist” approach sees the
dealing with whether international political NPT largely as a way to manage hypocrisy:
economy and human rights treaties constrain The NWSs claim to be committed to nuclear
their members, but much less on international disarmament under Article VI in order to
security treaties, especially nuclear weapons encourage the NWWSs to accept a “second-
proliferation (Chayes & Chayes 1995; Moravc- class” nuclear status. In particular, the NPT
sik 2000; Simmons 2000, 2010; Dai 2007). under this theory was designed in the 1960s
Indeed, it is revealing that Simmons’ (2010) primarily to constrain Germany and India;
Annual Review of Political Science article, “Treaty opening the treaty for universal signature was
238 Sagan
merely a face-saving device to make the con- (Albin 2006). If this is true in the NPT, it would
straints more politically acceptable in Bonn and help explain the renewed focus at NPT Review
New Delhi (Swango 2009). This view sees the Conferences on the issue of whether the NWSs
real strengths of the NPT in the additional in- have been meeting their Article VI treaty com-
ternational institutions that have been created mitments to work in good faith toward eventual
to manage export controls, such as the Nuclear nuclear weapons disarmament (Sagan 2009b).
Suppliers Group (NSG), which coordinates As part of this effort to evaluate the NPT and
decisions about what states can receive what the broader nuclear nonproliferation regime in
kinds of nuclear technologies (Davis 1993). the future, it will also be necessary to focus be-
It will be important, however, not to seek a yond the treaty to understand the origins and
master theory for how the treaty works, for I see effectiveness of related nuclear technology ex-
no reason to assume that the NPT has identical port control institutions. We have strong stud-
pathways of influence for different states. In- ies of the origins of the NPT itself (Shaker
deed, like many treaties, the NPT is a complex 1980, Bunn 1992, Swango 2009). But there are
institution representing a set of compromises no equivalent studies of the origins or effec-

between the governments that negotiated its tiveness of the NSG, or UN Security Council
content. We know very little about why dif- Resolution 1540, or the Proliferation Security
ferent governments joined the NPT and how Initiative. These newer institutions are crucial
their interests and interpretations have shaped elements of the nonproliferation regime and
the patterns of their compliance behavior (see, should not be ignored.
however, Erickson & Way 2011 and Potter
2010). There are likely to be some states who
joined the NPT to cement a nonproliferation How Regime Type Influences
bargain with regional rivals; some cheater states Nuclear Trajectory
that joined while already having or planning We need more work on how regime type influ-
covert nuclear weapons programs; and many ences decisions about both nuclear power and
more states that joined hoping membership nuclear proliferation. I have presented the data
would increase civilian nuclear technology showing that democracies do not cheat on their
transfer to them. There are likely to be even NPT commitments, but we do not know why
more states that joined the NPT as a part of that is the case. Are democracies more con-
joining the world order and have no intention strained than nondemocracies by their treaty
of ever developing a civilian nuclear energy commitments, or are they simply more selec-
infrastructure, much less nuclear weapons, tive in joining only treaties they intend to com-
such as Bhutan, Papua New Guinea, and Fiji. ply with fully (Simmons & Hopkins 2005, von
Indeed, even the Holy See, hardly a candidate Stein 2005)? If democratic leaders do feel more
for nuclear power nor a nuclear weapons constrained than nondemocratic leaders, why
concern, is a member of the NPT. is this the case? Do they fear “audience costs”
Given this complex and diverse set of mo- (punishment by voters) if they are caught cheat-
tives for joining the treaty, we should not ex- ing on agreements, or do they believe the like-
pect to find that a state’s NPT status is re- lihood of being caught cheating is higher be-
lated to its nuclear proliferation behavior in any cause of increased transparency and protection
simple manner. In addition, we should not as- of whistleblowers in a democracy (Tomz 2007)?
sume that states’ interests in the NPT are set With respect to the NPT, the simple divi-
in stone. Legal research on the NPT and other sion between democracies and nondemocracies
international treaties has suggested that devel- should be supplemented with studies on the re-
oping world actors have become less willing in lationship between the political leadership and
the postcolonial period to accept unequal treat- the scientific community and military. Hymans
ment in treaty agreements and implementation (2008), for example, usefully theorizes that
“neopatrimonial” or “sultanistic” regimes— by the severity of its demand for a weapon.

governments characterized by extreme person- Second, a high degree of nuclear capability or
alized rule, use of state resources to buy off latency could influence demand by enabling
clients, and an absence of checks and balances— actors who favor a nuclear weapon to argue that
will take longer to develop advanced levels acquiring a weapon would be a relatively easy
of nuclear technology and will fail more of- undertaking. Here it is worth differentiating
ten in attempts to move from one techno- between “latency,” a measure of how long it
logical threshold to the next. Weeks’s (2008) would take for a state to acquire a bomb, and
work on “autocratic audience costs” in interna- “hedging,” which is a deliberate action by a
tional crises suggests another research question government to make a nuclear weapons pro-
regarding proliferation: Do different types of gram easier in the future. Third, a high degree
nondemocratic regimes show different propen- of latency could make it easier for a pro–nuclear
sities to signal their intent and ability to abide weapons party or individual leaders to imple-
by their NPT commitments? Last, it is possi- ment a decision to acquire nuclear weapons if
ble that democracies may be less likely to move they are in power for only a brief period of time.
successfully toward nuclear disarmament than
nondemocracies because of the increased effect How the Success or Failure
of “nuclear nationalism” among the electorate of Civilian Nuclear Power
(Perkovich 1999). Influences Proliferation
Does the civilian nuclear power industry con-
How Nuclear Supply and Nuclear strain states or does it make nuclear weapons
Demand Influence Each Other proliferation easier? Do expensive failures to
We need more interdisciplinary work on the produce efficient nuclear power increase the
spread of nuclear power and how it influences likelihood of proliferation? An important deter-
“nuclear weapons latency” and the “time lines” minant of the nuclear future will be the degree
needed for states to move from different states to which the spread of nuclear power produces
of civilian power status to nuclear weapons new actors in different states that want to main-
capabilities (Wohlstetter et al. 1977, Zentner tain peaceful programs and oppose turning
et al. 2005, Harney et al. 2006, Sagan 2010). civilian energy programs into nuclear weapons
To conduct this analysis accurately, scholars programs. Indeed, how best to ensure that
will need to understand the history of different civilian nuclear power bureaucracies maintain
states’ open efforts to buy and sell sensitive nu- a strong interest in opposing nuclear weapons
clear technology, and also to evaluate the effects proliferation may be the single most critical
of the development of illicit nuclear technol- question to answer for reducing the potentially
ogy networks such as the A.Q. Khan network dangerous effect of the global spread of nuclear
(Braun & Chyba 2004, Montgomery 2005, power on the likelihood of nuclear weapons
Albright 2010). proliferation. This is ironic, for although some
Future efforts to understand nuclear latency nonproliferation specialists may not want more
and proliferation will be most useful if they countries to start nuclear power programs,
study the temporal connection between supply once a state starts a nuclear power program
and demand for nuclear weapons instead of its nonproliferation behavior may be strongly
treating these two “sides” of nuclear prolifera- influenced by the degree to which its civilian
tion as separate issues. Three potential connec- nuclear industry is a successful contributor to
tions are obvious. First, how hard a government national energy production. The leaders and
works on a nuclear weapons program—the re- bureaucratic organizations that run successful
sources it commits to the program and whether nuclear power enterprises may have increased
it is engaged in a “crash” effort or normal con- incentives to maintain strong ties to the
struction effort—is likely to be strongly affected global nuclear power industry, to international
240 Sagan
capital and technology markets, and to global theoretically informed case studies of prolif-
regulatory agencies—and hence may be more eration and nonproliferation decisions, both
likely to cooperate with the nuclear nonpro- to improve our data sets and to test whether
liferation regime. Leaders of less successful or the causal mechanisms posited in our theo-
struggling nuclear power enterprises, in con- ries are actually present. We need better quan-
trast, may be more likely to support clandestine titative studies to test the generalizability of
or breakout nuclear weapons development our theories. We need more interdisciplinary
programs to justify their existence, prestige, research on the technical bottlenecks and le-
and high budgets within their state. gal restrictions on the spread of nuclear facil-
ities. These complex research agendas should
be linked together to produce more robust
CONCLUSIONS findings about the nuclear past and promote
A renaissance in nuclear proliferation stud- better predictions about the global nuclear
ies is under way. Yet, we need more future.
DISCLOSURE STATEMENT
The author is not aware of any affiliations, memberships, funding, or financial holdings that might
be perceived as affecting the objectivity of this review.
ACKNOWLEDGMENTS
I thank Philipp Bleek, James Fearon, Matthew Fuhrmann, Jacques Hymans, Gaurav Kampani,
Matthew Kroenig, and Etel Solingen for their comments on this piece. Additionally, I am grateful
to Erin Dexter, Samantha Lang-Eppsteiner, Jane Esberg, Astasia Myers, Jaclyn Tandler, and
Jeremy Voss for research assistance.
LITERATURE CITED
Abraham I. 1998. The Making of the Indian Atomic Bomb: Science, Secrecy, and the Postcolonial State. London:
Zed Books
Abraham I. 2006. The ambivalence of nuclear histories. Osiris 21:49–46
Abraham I. 2009. Contra-proliferation: the Indian bomb and nuclear developmentalism. In Inside Nuclear
South Asia, ed. SD Sagan, pp. 106–36. Stanford, CA: Stanford Univ. Press
Acton JM. 2009. Nuclear power, disarmament, and technological restraint. Survival 51(4):101–126
Albin C. 2006. Justice and Fairness in International Negotiation. Cambridge, UK: Cambridge Univ. Press
Albright D. 2010. Peddling Peril: How the Secret Nuclear Trade Arms America’s Enemies. New York: Free Press
Bajpai K. 2009. The BJP and the Bomb. In Inside Nuclear South Asia, ed. SD Sagan, pp. 25–67. Stanford, CA:
Stanford Univ. Press
Betts RK. 2000. Universal deterrence or conceptual collapse? Liberal pessimism and utopian realism. In The
Coming Crisis: Nuclear Proliferation, US Interests and World Order, ed. VA Utgoff, pp. 51–85. Cambridge,
MA: MIT Press
Bleek PC. 2010. Why do states proliferate? Quantitative analysis of the exploration, pursuit, and acquisition
of nuclear weapons. See Potter & Mukhatzhanova 2010, pp. 159–76
Bluth C, Kroenig M, Lee R, Sailor W, Fuhrmann M. 2010. Correspondence: civilian nuclear cooperation and
the proliferation of nuclear weapons. Int. Secur. 35(1):184–200
Bowen WQ. 2006. Libya and nuclear proliferation: stepping back from the brink. Adelphi Pap. 46(380)
Braun C, Chyba C. 2004. Proliferation rings: new challenges to the nuclear nonproliferation regime.
Int. Secur. 29(2):5–49
Braut-Hegghammer M. 2009. Nuclear entrepreneurs: drivers of nuclear proliferation. PhD thesis, Dep. Int. Relat.,
London School Econ. Polit. Sci.
Bunn G. 1992. Arms Control by Committee: Managing Negotiations with the Russians. Stanford, CA: Stanford
Univ. Press
Campbell KM, Einhorn RJ, Reiss MB, eds. 2004. The Nuclear Tipping Point: Why States Reconsider Their Nuclear
Choices. Washington, DC: Brookings Inst.
Campbell KM, Sunohara T. 2004. Japan: thinking the unthinkable. See Campbell et al. 2004, pp. 218–53
Chafetz G, Abramson H, Grillot S. 2007. Role theory and foreign policy: Belarussian and Ukrainian compliance
with the nuclear nonproliferation regime. Polit. Psychol. 17:727–57
Chayes A, Chayes AH. 1995. The New Sovereignty: Compliance with International Regulatory Agreements. Cam-
bridge, MA: Harvard Univ. Press
Cirincione J, Wolfstahl J, Rajkumar M. 2002. Deadly Arsenals: Tracking Weapons of Mass Destruction. Wash-
ington, DC: Carnegie Endow. Int. Peace
Cohen A. 1999. Israel and the Bomb. New York: Columbia Univ. Press
Dai X. 2007. International Institutions and National Policies. New York: Cambridge Univ. Press
Davis ZS. 1993. The realist nuclear perspective. Secur. Stud. 2(3/4):79–99
Erickson JL, Way C. 2011. Membership has its priveleges: conventional arms and influence in the NPT.
In Causes and Consequences of Nuclear Proliferation, ed. R Rauchhaus, M Kroenig, E Gartzke. New York:
Routledge. In press
Forland A. 1997. Norway’s nuclear odyssey: from optimistic proponent to nonproliferator. Nonproliferation
Rev. 4(2):1–16
Frankel B. 1991. Opaque Nuclear Proliferation: Methodological and Policy Implications. London: Frank Cass
Frankel B. 1993. The brooding shadow: systemic incentives and nuclear weapons. Sec. Stud. 2(3/4):37–78
Frankel B, Davis ZS, eds. 1993. The Proliferation Puzzle: Why Nuclear Weapons Spread and What Results. New
York: Routledge
Fuhrmann M. 2009a. Spreading temptation: proliferation and peaceful Nuclear Cooperation Agreements. Int.
Secur. 34(1):7–41
Fuhrmann M. 2009b. Taking a walk on the supply side: the determinants of civilian nuclear cooperation. J.
Confl. Resolut. 53(2):181–208
Harney R, Brown G, Carlyle M, Skroch E, Wood K. 2006. Anatomy of a project to produce a first nuclear
weapon. Sci. Global Secur. 14(3):163–82
Hecker SS. 2010. Lessons learned from the North Korean nuclear crisis. Daedalus 139(1):44–56
Hymans JEC. 2001. Of gauchos and gringos: why Argentina never wanted the bomb, and why the United
States thought it did. Secur. Stud. 10(3):153–85
Hymans JEC. 2006. The Psychology of Nuclear Proliferation: Identity, Emotions and Foreign Policy. Cambridge,
UK: Cambridge Univ. Press
Hymans JEC. 2008. Assessing North Korean nuclear intentions and capacities: a new approach. J. East Asian
Stud. 8(2):259–92
Hymans JEC. 2009. Proliferation implications of Atoms for Peace: a case study of the Yugoslav experience. Presented
at Nobel Inst. Symp. Peace, Stability, and Nuclear Order, Oscarsborg, Drobak, Norway, June 25–27
Hymans JEC. 2010. The study of nuclear proliferation and nonproliferation: toward a new consensus? See
Potter & Mukhatzhanova 2010, pp. 13–37
International Atomic Energy Agency. 2002. Fact sheet: Iraq’s nuclear weapon programme.
http://www.iraqwatch.org/un/IAEA/iaea-facts-042502.htm
International Panel on Fissile Materials. 2010. Limiting national fissile-material production capabili-
ties. http://www.fissilematerials.org/ipfm/pages_us_en/inventories/limiting_production/limiting_
production.php
Japan Times. 2010. Tokyo sought out Bonn on going nuclear in ’69. Nov. 30.
http://search.japantimes.co.jp/cgi-bin/nn20101130a3.html
Jo DJ, Gartzke E. 2007a. Determinants of nuclear weapons proliferation. J. Confl. Resolut. 51(1):167–94
Jo DJ, Gartzke E. 2007b. Data notes for “Determinants of Nuclear Weapons Proliferation.” http://jcr.sagepub.
com/content/vol51/issue1/images/data/167/DC1/Jo&Gartzke_(Feb_07).zip
242 Sagan
Kroenig M. 2009a. Importing the bomb: sensitive nuclear assistance and proliferation. J. Confl. Resol. 53(2):161–
80
Kroenig M. 2009b. Exporting the bomb: sensitive nuclear assistance and proliferation. Am. Polit. Sci. Rev.
103(1):113–33
Kroenig M. 2010. Exporting the Bomb: Technology Transfer and the Spread of Nuclear Weapons. Ithaca, NY:
Cornell Univ. Press
Kuntzel M. 1995. Bonn and the Bomb: German Politics and the Nuclear Option. London: Pluto Press
Lavergeon A. 2009. The nuclear renaissance: an opportunity to enhance the culture of nonproliferation.
Daedalus 138(4):91–99
Lavoy PR. 1993. Nuclear myths and the causes of nuclear proliferation. Secur. Stud. 2(3&4):192–212
Lavoy PR. 2004. Predicting nuclear proliferation: a declassified history. Strategic Insights 3(1).
http://fas.org/man/eprint/lavoy.pdf
Lester RK, Rosner R. 2009. The growth of nuclear power: drivers and constraints. Daedalus 138(4):19–30
Levite A. 2003. Never say never again: nuclear reversal revisited. Int. Secur. 27(3):59–88
Liberman P. 2001. The rise and fall of the South African Bomb. Int. Secur. 26(2):45–86
May M, ed. 2001. Verifying the agreed framework. Cent. Global Secur. Res., Lawrence Livermore Natl. Lab.,
Cent. Int. Secur. and Coop., Stanford Univ., UCRL-ID-142036, CGSR-2001–001

Mearsheimer JJ. 1990. Back to the future: instability in Europe after the Cold War. Int. Secur. 15(1):5–56
Meyer SM. 1984. The Dynamics of Nuclear Proliferation. Chicago: Univ. Chicago Press
Miller M. 1979. The nuclear dilemma: power, proliferation and development. Technol. Rev. 81(6):18–20
Miller SM, Sagan SD. 2009. Nuclear power without nuclear proliferation? Daedalus 138(4):7–18
Montgomery AM. 2005. Ringing in proliferation: how to dismantle an atomic bomb network. Int. Secur.
30(2):153–87
Montgomery AM, Sagan SD. 2009. The perils of predicting proliferation. J. Confl. Resolut. 53(2):302–28
Montgomery AM, Sagan SD. 2011. The perils of predicting proliferation. In Causes and Consequences of Nuclear
Proliferation, ed. R Rauchhaus, M Kroenig, E Gartzke. New York: Routledge. In press
Moravcsik A. 2000 . The origins of human rights regimes: democratic delegation in postwar Europe. Int.
Organ. 54(2):217–52
Müller H. 2003. German national identity and WMD nonproliferation. Nonproliferation Rev. 10(2):1–20
Müller H, Schmidt A. 2010. The little known story of de-proliferation: why states give up nuclear weapons
activities. In See Potter & Mukhatzhanova 2010, pp. 124–58
Nuti L. 1993. ‘Me too please’: Italy and the politics of nuclear weapons, 1945–1970. Diplomacy Statecraft
4(1):114–48
Paul TV. 2000. Power versus Prudence: Why Nations Forgo Nuclear Weapons. Montreal: McGill-Queen’s Univ.
Press
Perkovich G. 1999. India’s Nuclear Bomb. Berkeley: Univ. Calif. Press
Potter WC. 2010. The NPT and the sources of nuclear restraint. Daedalus 139(1):68–81
Potter WC, Mukhatzhanova G, eds. 2010. Forecasting Proliferation in the 21st Century: the Role of Theory. Palo
Alto, CA: Stanford Univ. Press
Potter WC, Mukhatzhanova G. 2010. The role of theory, an introduction. See Potter & Mukhatzhanova
2010, pp. 1–12
Purkitt HE, Burgess SF, Liberman P. 2002. Correspondence: South Africa’s nuclear decisions. Int. Secur.
27(1):186–94
Reiss M. 1995. Bridled Ambition: Why Countries Constrain their Nuclear Capabilities. Washington, DC: Woodrow
Wilson Cent. Press
Rublee MR. 2009. Nonproliferation Norms: Why States Choose Nuclear Restraint. Atlanta: Univ. Georgia Press
Russian Federation Foreign Intelligence Service. 1995. The nuclear potential of individual countries: Romania.
http://www.fas.org/irp/threat/svr_nuke.htm#romania
Sagan SD. 1996/1997. Why do states build nuclear weapons? Three models in search of a bomb. Int. Secur.
21(3):54–86
Sagan SD. 1999. Rethinking the causes of nuclear proliferation: three models in search of a bomb. In The
Coming Crisis: Nuclear Proliferation, U.S. Interests, and World Order, ed. VA Utgoff, pp. 17–50
Sagan SD. 2009a. Introduction. In Inside Nuclear South Asia, ed. SD Sagan, pp. 1–24. Stanford, CA: Stanford
Univ. Press
Sagan SD. 2009b. Shared responsibilities for nuclear disarmament. Daedalus 138(4):157–68
Sagan SD. 2010. Nuclear latency and nuclear proliferation. See Potter & Mukhatzhanova 2010, pp. 13–37
Sasikumar K, Way C. 2009. Testing theories of proliferation in South Asia. In Inside Nuclear South Asia, ed.
SD Sagan, pp. _______. Stanford, CA: Stanford Univ. Press
Scheinman L. 1987. International Atomic Energy Agency and World Nuclear Order. Washington, DC: Resour.
Future
Schiff BN. 1984. International Nuclear Technology Transfer: The Dilemmas of Dissemination and Control. Totowa,
NJ: Rowman & Allanheld
Shaker MI. 1980. The Nuclear Non-Proliferation Treaty: Origin and Implementation 1959–1979. New York:
Oceana
Simmons BA. 2000. Compliance with international agreements. Am. Polit. Sci. Rev. 94(4):819–35
Simmons BA. 2010. Treaty compliance and violation. Annu. Rev. Polit. Sci. 13:273–96
Simmons BA, Hopkins DJ. 2005. The constraining power of international treaties: theory and methods. Am.
Polit. Sci. Rev. 99(4):623–31
Singh S, Way CR. 2004. The correlates of nuclear proliferation: a quantitative test. J. Confl. Resolut. 48(6):859–
85
Socolow RH, Glaser A. 2009. Nuclear energy and climate change. Daedalus 138(4):31–44
Solingen E. 1994. The political economy of nuclear restraint. Int. Secur. 19(2):126–69
Solingen E. 1998. Regional Orders at Century’s Dawn: Global and Domestic Influences on Grand Strategy. Princeton,
NJ: Princeton Univ. Press
Solingen E. 2007. Nuclear logics: contrasting paths in East Asia and the Middle East. Princeton, NJ: Princeton
Univ. Press
Spector LS, Berman DR. 2010. The Syrian nuclear puzzle. In Forecasting Nuclear Proliferation in the 21st Century:
A Comparative Perspective, ed. WC Potter, G Mukhatzhanova, pp. 100–30. Stanford, CA: Stanford Univ.
Press
Stoll JR. 1996. World production of latent nuclear capacity. http://es.rice.edu:80/projects/Poli378/
Nuclear/Proliferation/proliferation.html
Swango DE. 2009. The Nuclear Nonproliferation Treaty: constrainer, screener, or enabler. PhD thesis, Dep. Polit.
Sci., Univ. Calif. Los Angeles.
Tannenwald N. 2007. The Nuclear Taboo: the United States and the Non-Use of Nuclear Weapons. Cambridge,
UK: Cambridge Univ. Press
Thayer B. 1995. The causes of nuclear proliferation and the utility of the nuclear nonproliferation regime.
Secur. Stud. 4(3):463–519
Tomz M. 2007. Domestic audience costs in international relations: an experimental approach. Int. Org.
61(4):821–40
U.S. Department of State. 2010. Adherence to and compliance with arms control, nonproliferation, and disarmament
agreements and commitments. http://www.state.gov/documents/organization/145181.pdf
von Stein J. 2005. Do treaties constrain or screen? Selection bias and treaty compliance. Am. Polit. Sci. Rev.
99(4):611–22
Walsh J. 1997. Surprise Down Under: the secret history of Australia’s nuclear ambitions. Nonproliferation Rev.
5:1–20
Weeks JL. 2008. Autocratic audience costs: regime type and signaling resolve. Int. Organ. 62(1):35–64
Wohlstetter A, Brown TA, Jones G, McGarvey DC, Rowen H, et al. 1977. Swords from Ploughshares: The
Military Potential of Civilian Nuclear Energy. Chicago: Univ. Chicago Press
World Nuclear Association. 2010. Uranium enrichment. http://www.world-nuclear.org/info/
inf28.html
Zentner MD, Coles GL, Talbert RJ. 2005. Nuclear proliferation technology trends analysis. Pacific Northwest
Natl. Lab. http://www.pnl.gov/main/publications/external/technical_reports/PNNL-14480.pdf
244 Sagan
14
UKR, BEL, KAZ

12
-S AFR
-KAZ
10
PAK NK
S AFR
Number of states
8
IND
-UKR, -BEL
ISR
6
CHI
FRA
4
RUS UK
2
USA
Year
Figure 1
Nuclear weapons proliferation (Sagan 2010). Figure c Scott D. Sagan based on data from Singh & Way
(2004). For this graph, the dates of the following countries’ nuclear programs were modified from Singh &
Way on the basis of new research: Israel (Cohen 1999), India (Perkovich 1999, Abraham 1998), North Korea
(Hecker 2010).
IRAN, YUGO
ITA, WGER ROM TAI, -YUGO
8
SK -SK IRAN -TAI

7
AUS, TAI, -EGY -TAI ALG

AUS IRAQ, -BRA
6
-AUS
-ITA, LIB
Number of states
-WGER -IRAQ
5
EGY S AFR, -IRAN
-YUGO,
-AUS -SWI, -SWE
4
SWE, YUGO
-ALG, -ROM SYR

3
-LIB
2
BRA -S AFR
1
SWI
45 50 55 60 65 70 75 80 85 90 95 00 05 10
19 19 19 19 19 19 19 19 19 19 19 20 20 20
Year
Figure 2
Nuclear exploration, program, and reversal (Sagan 2010). Figure
c Scott D. Sagan based on data from
Singh & Way (2004). The dates of the following countries’ nuclear programs were modified for this graph
based on new research: Yugoslavia (Hymans 2009), Argentina (Hymans 2001), Libya (Cirincione et al.
2002), Iraq (International Atomic Energy Agency 2002), Romania (Russian Federation Foreign Intelligence
Service 1995), Italy and West Germany (Kuntzel 1995, Müller 2003, Müller & Schmidt 2010), Syria
(Spector & Berman 2010, U.S. Department of State 2010), and Norway (Forland 1997).
www.annualreviews.org • Causes of Nuclear Weapons Proliferation C-1
60
50
Number of States
40
States with
Nuclear
30 Weapons
Nuclear
20 Capable
States
10
0
50 55 60 65 70 75 80 85 00
19 19 19 19 19 19 19 19 /20
90
Year 19
Figure 3
Stoll’s nuclear latency data for 1992 (Stoll 1996).
Autocracy (-10 – 0)
: Rafied NPT
Semi-Democracy (1 – 6)
: Rafied NPT as ROC
Democracy (7 – 10)
: Rafied NPT, Withdrew in
2003
Transion Period
Switzerland
Brazil
Yugoslavia (1) + (2)
Sweden
Australia (1) + (2)
Italy
(West) Germany
South Korea
Egypt
North Korea
Taiwan (1) + (2)
South Africa
Libya
Iraq
Algeria
Iran (1) + (2)
Romania
Syria
1940 1950 1960 1970 1980 1990 2000
Figure 4
Regime type, date of ratification of the Non-Proliferation Treaty, and nuclear exploration and pursuit. Figure
c Scott D. Sagan based
on data from Sagan (2010).
C-2 Sagan

The Causes of Nuclear Weapons Proliferation: Scott D. Sagan

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Causes of Nuclear Weapons Proliferation: Scott D. Sagan

Uploaded by

Copyright:

Available Formats

PL14CH11-Sagan ARI 18 February 2011 17:14

still occur before final publication

The Causes of Nuclear

Department of Political Science, Stanford University, Stanford, California 94305;

Annu. Rev. Polit. Sci. 2011. 14:225–44 Keywords

Copyright c 2011 by Annual Reviews. Abstract

INTRODUCTION this review, I argue that the supply-side litera-

demand, between a state’s increased nuclear

veloping nuclear weapons.

sixth section outlines alternative perspectives program). There is no single explanation

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 227

The small-numbers problem and the lin- THE SUPPLY-SIDE LITERATURE

tion (Montgomery & Sagan 2009, 2011). For

that adding new conﬁrmed cases of states that

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 229

fertilizer production fertilizer production

produces equivalent of 50,000 metric

(2009b, 2010). Why do states sell or give

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 231

of proliferation. Peaceful nuclear cooperation a country’s interest in nuclear weapons tech-

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 233

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 235

intellectual judo” (Hymans 2010), Solingen

were already opposed to Japanese nuclear

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 237

institution representing a set of compromises no equivalent studies of the origins or effec-

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 239

“neopatrimonial” or “sultanistic” regimes— by the severity of its demand for a weapon.

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 241

Cent. Int. Secur. and Coop., Stanford Univ., UCRL-ID-142036, CGSR-2001–001

www.annualreviews.org • Causes of Nuclear Weapons Proliferation 243

UKR, BEL, KAZ

SK -SK IRAN -TAI

AUS, TAI, -EGY -TAI ALG

-ALG, -ROM SYR

www.annualreviews.org • Causes of Nuclear Weapons Proliferation C-1

1940 1950 1960 1970 1980 1990 2000

You might also like