Nuclear Proliferation, U.S.

Interests,
and World Order
THE COMI NG CRI SI S
T
H
E

C
O
M
I
N
G

C
R
I
S
I
S
U
T
G
O
F
F
,
E
D
I
T
O
R
The Coming Crisis
Nuclear Proliferation, U.S. Interests, and World Order
edited by Victor A. Utgoff
foreword by General Larry D. Welch
How will continued proliferation of nuclear weapons
change the global political order? This collection
of essays comes to conclusions at odds with the
conventional wisdom. Stephen Rosen and Barry
Posen explore how nuclear proliferation may affect
U.S. incentives to confront regional aggression.
Stephen Walt argues that regional allies will likely
prove willing to stand with a strong and ready
United States against nuclear-backed aggression.
George Quester and Brad Roberts examine long-
term strategic objectives in responding to nuclear
attack by a regional aggressor. Richard Betts
highlights the potential for disastrous mistakes
in moving toward and living in a world heavily
populated with nuclear-armed states. Scott Sagan
explains how the nuclear nonproliferation policies
best suited to some states can spur proliferation
by others. Caroline Ziemke shows how the analysis
of a states strategic personality can provide
insights into why it might want nuclear weapons
and how its policies may develop once it gets them.
And Victor Utgoff concludes that the United States
seems more likely to intervene against regional
aggression when the aggressor has nuclear weapons
than when it does not.
,!7IA2G2-cbabfh!:t;K;k;K;k
UTGCH 0-262-21015-0
Victor A. Utgoff is Deputy Director of the
Strategy, Forces, and Resources Division of
the Institute for Defense Analyses.
VI CTOR A. UTGOFF, EDI TOR
foreword by General Larry D. Welch
This volume is an original and important contribution to the growing debate on
nuclear proliferation. It includes chapters by several of the most important American
thinkers on this subject. . . . The book makes clear the enormity of the challenges
we face and presents useful and prudent ideas for dealing with them.
Michael Nacht, Dean, Richard and Rhoda Goldman School of Public Policy,
University of California at Berkeley
An important and much needed study which adds new and deeper insights to an
understanding of the security issues posed by weapons of mass destruction in the
coming decades. The authors are to be greatly commended for their valuable and
timely contributions.
Andrew J. Goodpaster, Senior Advisor to the Board, Institute for Defense Analyses, and
Supreme Allied Commander and Commander-in-Chief, U.S. Forces, Europe, 196974
This book is both welcome and timely in contributing new thinking and scholarly
research into one of the most complex and important questions of our time. . . .
The highest value of this book may well be in awakening Americans, as well as
potential aggressors around the world, that the United States is not likely to retreat
from countering aggression just because weapons of mass destruction might be used.
Ambassador C. Paul Robinson, President and Laboratories Director, Sandia National
Laboratories
BCSIA Studies in International Security
The MIT Press
Massachusetts Institute of Technology
Cambridge, Massachusetts 02142
http://mitpress.mit.edu
cover photo: Atomic Street Sign, Amargosa Valley
Carole Gallagher
Utgoff jacket 2/14/02 11:50 AM Page 1
The Coming Crisis
Nuclear Proliferation, U.S. Interests, and
World Order
The BCSIA Studies in International Security book series is edited at the Belfer
Center for Science and International Affairs at Harvard Universitys John F.
Kennedy School of Government and published by The MIT Press. The series
publishes books on contemporary issues in international security policy, as
well as their conceptual and historical foundations. Topics of particular interest
to the series include the spread of weapons of mass destruction, internal
conict, the international effects of democracy and democratization, and U.S.
defense policy.
A complete list of BCSIA Studies appears at the back of this volume.
The Coming Crisis
Nuclear Proliferation, U.S. Interests, and
World Order
Editor
Victor A. Utgoff
BCSIA Studies in International Security
MIT Press
Cambridge, Massachusetts
London, England
with sponsorship of the Institute
for Defense Analyses
2000 by the Belfer Center for Science and International Affairs
John F. Kennedy School of Government, Harvard University
Cambridge, Massachusetts 02138
(617) 4951400
All rights reserved. No part of this book may be reproduced, stored in
a retrieval system, or transmitted in any form or by any means
electronic, electrostatic, magnetic tape, mechanical, photocopying,
recording, or otherwisewithout permission in writing from the
Belfer Center for Science and International Affairs, 79 John F. Kennedy
Street, Cambridge, MA 02138.
Library of Congress Cataloging-in-Publication Data
The coming crisis : nuclear proliferation, U.S. interests, and world
order / edited by Victor A. Utgoff.
p. cm.(BCSIA studies in international security)
Includes bibliographical references and index.
ISBN 0-262-21015-0 (hc. : alk. paper)
ISBN 0-262-71005-6 (pbk. : alk. paper)
1. Weapons of mass destruction. 2. Nuclear nonproliferation.
3. National securityUnited States. 4. United StatesMilitary
policy. 5. International relations. I. Utgoff, Victor A. II. Series.
U793.C65 2000
327.1747dc21
98-31554
Contents
Foreword General Larry D. Welch, USAF (Retired) vii
Part I Pressures for Nuclear Proliferation and Crises 1
Chapter 1 The Specter of Nuclear, Biological, and
Chemical Weapons Proliferation
Victor A. Utgoff
3
Chapter 2 Rethinking the Causes of Nuclear
Proliferation: Three Models in Search of a
Bomb
Scott D. Sagan
17
Chapter 3 Universal Deterrence or Conceptual Collapse?
Liberal Pessimism and Utopian Realism
Richard K. Betts
51
Chapter 4 The National Myth and Strategic Personality
of Iran: A Counterproliferation Perspective
Caroline F. Ziemke
87
Part II Potential Evolution and Consequences of a
Nuclear Crisis with the United States
123
Chapter 5 Nuclear Proliferation and Alliance Relations
Stephen Peter Rosen
125
Chapter 6 U.S. Security Policy in a Nuclear-Armed
World, or What If Iraq Had Had Nuclear
Weapons?
Barry R. Posen
157
Chapter 7 Containing Rogues and Renegades: Coalition
Strategies and Counterproliferation
Stephen M. Walt
191
Chapter 8 The Response to Renegade Use of Weapons
of Mass Destruction
George H. Quester
227
Chapter 9 Rethinking How Wars Must End: NBC War
Termination Issues in the PostCold War Era
Brad Roberts
245
Chapter 10 The Coming Crisis: Nuclear Proliferation,
U.S. Interests, and World Order
A Combined Perspective
Victor A. Utgoff
279
Contributors 303
Index 307
About the Belfer Center for Science and International Affairs 322
vi the coming crisis
Foreword
The accelerating worldwide advance of technology and knowledge cre-
ates rapidly expanding opportunities for humankindopportunities that
have been of great benet to the developed world, and especially to the
United States. But the advance of technology also provides opportunities
for increasing numbers of states to build weapons that in minutes to
hours can cause levels of destruction that once took years of warfare with
massive forces. Further, more states actually are building such weapons,
or positioning themselves to be able to build them quickly. The evidence
of this half-century trend is clear and compelling. The nuclear weapons
tests carried out by India and Pakistan in May 1998 and Iraqs continued
resistance to UN efforts to eliminate its program for weapons of mass
destruction are recent reminders.
Some hope that the proliferation of weapons of mass destruction will
ultimately lead potential aggressors to conclude that war has become too
dangerous. But centuries of history, including the past ve decades, lead
most observers of the international scene to be deeply skeptical that a
more proliferated world would be more peaceful. It seems more likely
that highly destructive wars would increase as the number of actors
armed with these weapons rises. Thus, efforts to limit or roll back prolif-
eration remain a national priority.
There is reason for some optimism about the outcome of such efforts.
Looking back, international nonproliferation efforts, coupled with the
self-restraint exercised by many nations, have been surprisingly effective.
Predictions made decades ago of the number of states that would have
weapons of mass destruction by 2000 have proven pessimistic. While the
large majority of the worlds states are now capable of building weapons
of mass destruction, only a minority appear to have done so, or to be
purposely moving toward such weapons.
Many factors are involved in explaining this divergence between
capabilities to build such weapons and the choice to do so. Among the
most important is the belief that the major states will continue to play
their postWorld War II role of keeping sovereign states from conquering
or destroying one another. But proliferation raises the risk involved in
intervention, and the end of the global contest for power with the former
Soviet Union causes some to believe that the outcomes of regional wars
are less important to the United States. This combination could under-
mine condence in the capability and the will of the United States to
continue to play the key stabilizing role the world has come to expect
of it.
I believe the United States will continue in its stabilizing role for at
least three reasons. First, U.S. political leaders, whatever their political
philosophy, have always found it difcult to keep the nation on the
sidelines in the face of massive violence or destabilizing developments.
Second, the United States will seldom, if ever, nd it in its national
interest to be deterred from standing up to aggression. Third, I believe
that the United States remains willing to accept riskseven large risks
for an important cause. And the prevention, suppression, and defeat of
aggression backed by the use or threat of weapons of mass destruction
will continue to be seen as important to the peace and stability that serve
U.S. national interests. In addition to the more immediate costs of failing
to deal with such aggression, history tells us that such failures are likely
to lead to a far more proliferated, dangerous, and less cooperative world.
Thus, the United States is and should be committed to deterring and,
if need be, to defending against aggression, especially when backed by
weapons of mass destruction. The military forces needed to do this will
include nuclear forces that provide a credible threat of devastating retali-
ation for the use of weapons of mass destruction. At the same time,
minimizing the risks posed to the forces and citizens of the United States,
its allies, and other nations when confronting such aggression will require
other substantial preparationspreparations that increase the power of
deterrence by further reducing an aggressors condence that the gain
from such attacks is worth their potentially very high costs. These prepa-
rations will include better ways to prevent attacks with weapons of mass
destruction, to interdict such attacks when launched, and to reduce the
effects that such attacks can have on their targets. They are also likely to
include substantial changes in how the United States organizes, deploys,
and operates its forces when faced by such threats.
The preparations that the United States must make are neither cheap
nor easy. The effort required may come at the cost of other things with
more obvious appeal. Thus, making the needed preparations will require
a keen understanding of the larger dangers posed by continued prolif-
viii the coming crisis
eration of weapons of mass destruction, and what must be done to reduce
them.
This collection of essays contributes to that crucial understanding. It
presents a variety of perspectives on important policy and strategy prob-
lems posed by the continued proliferation of weapons of mass destruc-
tion. These problems and their solutions are substantially different from
those posed during the Cold War, when the United States and the Soviet
Union confronted each other with massive arsenals of nuclear weapons.
I hope the reader will reect on the insightful essays presented here.
And beyond that, I hope that other policymakers and experts will be
encouraged to contribute further to the understanding of this broad and
important topic. Better and more up-to-date analysis and understanding
of the challenges posed by the proliferation of weapons of mass destruc-
tion are vital to nding paths to a safer and better world.
General Larry D. Welch, USAF (ret.)
April 1999
Alexandria, Virginia
foreword ix
Part I
Pressures for Nuclear Proliferation
and Crises
Chapter 1
The Specter of Nuclear,
Biological, and Chemical
Weapons Proliferation
Victor A. Utgoff
In the past decade, the United States and other responsible nations have
become increasingly concerned that growing numbers of states and even
sub-state organizations will obtain nuclear, biological, or chemical (NBC)
weapons capable of causing massive destruction. These types of weapons
are spreading.
1
India and Pakistan have both recently carried out multiple
nuclear weapons tests. The agreement under which North Korea sus-
pended its nuclear weapons program appears to be unraveling.
2
And a
number of antagonistic states, such as Iraq, North Korea, Iran, and Libya
are trying to obtain NBC weapons.
3
The publics awareness of the harm these weapons could cause is
being heightened. For example, retired military ofcers who once com-
manded nuclear arsenals have highlighted the dangers of maintaining
these forces.
4
The chemical attacks by Japanese terrorists that caused
1. For a review of the problem of NBC proliferation as seen by the United States
Department of Defense, see Proliferation: Threat and Response (Washington, D.C.: United
States Department of Defense, November 1997). Another useful view of the prolifera-
tion problems is provided by Randall Forsberg, William Driscoll, Gregory Webb, and
Jonathan Dean, Nonproliferation Primer: Preventing the Spread of Nuclear, Chemical, and
Biological Weapons (MIT Press, 1995).
2. See Brad Roberts, The Future of Nuclear Weapons in Asia (Institute for Defense
Analyses, forthcoming), for a comprehensive review of the potential for nuclear
proliferation in Asia in the aftermath of the May 1998 nuclear tests by Indian and
Pakistan.
3. Beyond the ve declared nuclear powers, at least 25 countries already have or
may be developing nuclear, biological, or chemical weapons, or their missile delivery
systems. Report on Activities and Programs for Countering Proliferation and NBC Terror-
ism, Counterproliferation Program Review Committee (Washington, D.C.: United States
Department of Defense, May 1998), p. 3-1.
4. Retired Nuclear Warrior Sounds Alarm on Weapons: Ex-SAC Commander Calls
Policy Irrational, Washington Post, December 4, 1996, p. 1.
nearly 20 deaths and 6,000 injuries captured public attention worldwide.
5
And the horrors of biological weapons have been publicized in a variety
of recent literary works and television programs.
6
The U.S. government is sufciently concerned about the potential for
use of biological weapons on the battleeld to take action. During the
Gulf War, it vaccinated as many troops as possible against anthrax. And
the Department of Defense has begun a program that will ultimately
provide vaccinations against anthrax to all active-duty military service
members and reservists.
7
Most importantly, policymakers and the public sense that the prolif-
eration of NBC weapons may lead the nation to a most difcult dilemma:
If important U.S. overseas interests are challenged by states newly armed
with such weapons, the United States must choose between running the
sharply increased risks of defending its interests, or compromising those
interests, together with its reputation for military preeminence and a
willingness to protect allies and friends.
These concerns have led to new initiatives aimed at slowing or
reversing the proliferation of NBC weapons. In recent years, the U.S.
government has brokered agreements that have led three newly inde-
pendent states to give up the nuclear arsenals they had inherited from
the Soviet Union. Multiyear legislation sponsored by Senators Sam Nunn,
Richard Lugar, and Pete Domenici has provided funds to reduce Soviet
and now Russian nuclear forces and to minimize the prospects that the
materials and expertise necessary for creating nuclear weapons will leak
out of the former Soviet Union.
In addition, the United States continues to support a sputtering and
still incomplete United Nations (UN) program to root out Iraqs NBC
programs. The United States was also instrumental in winning the in-
denite extension of the Nuclear Non-Proliferation Treaty (NPT) in 1995.
And, together with the other declared nuclear powers, the United States
has suspended its nuclear testing program with the expectation that all
5. The Continuing Threat from Weapons of Mass Destruction (Nonproliferation Center,
Ofce of the Director of U.S. Central Intelligence, March 1996), p. 5.
6. For example, see Richard Preston, The Cobra Event (New York: Random House,
1997); and John F. Case, The First Horseman (Fawcett, 1998); in addition, television series
such as Seven Days and The X-Files have dealt with the concept of biological warfare.
Finally, numerous nonctional documentaries and reports have been led by the news
media in relation to domestic anthrax scares, the Iraqi biological weapons program,
and revelations that Russia continues to work on biological weapons.
7. Total Force Anthrax Vaccinations To Begin, DefenseLINK Release No. 430-98,
August 14, 1998 (http://www.defenselink.mil/news/Aug1998/b08141998_bt430-
98.html).
4 the coming crisis
nuclear and nuclear-capable states will eventually join the Comprehen-
sive Test Ban Treaty.
In recent years, the United States has also taken new steps to counter
the capabilities of proliferators to effectively threaten or actually use NBC
weapons. These steps include a multiyear Counter-Proliferation Initiative
(CPI) to develop new technologies that can allow these weapons to be
attacked and destroyed before they can be used, or intercepted after they
have been launched but before they reach their targets.
8
The CPI is also
improving the protection of U.S. forces against chemical and biological
agents that do arrive in their vicinities. In addition, U.S. military planners
are developing operational concepts and plans for employing forces so
that they can perform their missions with minimal risks of defeat or of
suffering historically unprecedented losses from NBC attacks. Finally,
some initial steps have been taken toward cooperative counterprolifera-
tion efforts with key allies.
9
Impressive as these various nonproliferation and counterproliferation
actions may be, they are only a start toward the goal of denying prolif-
erators the potential destructive and coercive power of NBC weapons.
Among the larger efforts that lie ahead, three efforts stand out. First, the
creation of an effective defense against the kinds of NBC capabilities that
proliferators might aspire toespecially considering the many different
forms that these weapons and their means of delivery might takeis a
task with substantial technical difculties and costs.
Second, for political as well as practical reasons, the United States
cannot bear all the burdens of countering NBC weapons. Other states that
can be threatened by these weapons, or that are relatively capable of
contributing to efforts to counter them, must be convinced to participate
and to take the necessary actions, including cooperative efforts to protect
against NBC attacks. In addition, partners will need to be visibly involved
if they are to share adequately the responsibility for military actions that
might be required against an NBC-armed regional challenger. Such in-
volvement requires cooperative efforts to prepare other states forces to
ght effectively alongside those of the United States. It also means in-
volving prospective partners in the key decisions regarding military ob-
jectives and the possible retaliatory use of nuclear weapons, should that
prove necessary.
8. See the ve Reports to the Congress on Activities and Programs for Countering Prolif-
eration (and NBC Terrorism [1998]), Counterproliferation Program Review Committee (Wash-
ington, D.C.: United States Department of Defense, May 1994, 1995, 1996, 1997, and
1998).
9. Proliferation: Threat and Response, pp. 6263.
chapter 1 5
Third, the United States and other cooperating governments must
develop a better public awareness of the need to prevent and counter the
proliferation of NBC weapons. In particular, publics must be prepared to
face the possibility of challenges to important interests by NBC-armed
regional aggressors and to support the necessary political and military
preparations. Waiting until such a challenge materializes to clarify the
potential stakes and risks and the pros and cons of alternative courses of
action increases the chances of political confusion and devastating mis-
takes, and the chances that such challenges would arise in the rst place.
Clearly, the overall political and technical effort required to halt, roll
back, and counter the continued proliferation of NBC weapons is very
substantial. Will the United States prove willing over the long haul to
bear the costs and other burdens involved?
The answer is far from clear. Rather than defend against the Soviet
Unions nuclear capabilities in any signicant way, for decades the United
States accepted a mutual nuclear deterrence relationship. Its willingness
to compromise its policy of punishing Pakistan and India for pursuing
nuclear weapons and to overlook Israels nuclear weapons program dem-
onstrates that nonproliferation is not always the highest priority for the
United States. In addition, while the frightening specters of NBC attacks
on U.S. forces or cities are disturbing, they are hard for the U.S. public to
take too seriouslythe public tolerated such fears for the decades of the
Cold War. Moreover, it is even easier to discount the possibility of such
attacks by renegade states that have not been seen as major powers in
the past, and whose military capabilities are so modest compared to those
of the United States and its allies.
But the possibility of such attacks cannot be discountedand the
preparations that the United States makes to meet such challenges will
strongly affect the outcome of such an attack. Rather than wait until an
NBC-armed state challenges an important regional interest, rather than
wait until the discomforts of accommodating to a world in which NBC
proliferation gives otherwise minor powers inuence disproportionate to
their populations, productivity, or moral considerations, we must nd the
motivation now to face the problem of proliferation more seriously. A
deeper and broader appreciation of the eventual implications of contin-
ued proliferation of NBC weapons will allow the United States and its
allies to trade the risks and discomforts of dangerous confrontations and
twisted world orders for the burdens of preparation and avoidance.
The goal of this book is to help develop such an appreciation. It is an
attempt to anticipate some of the ways in which continued proliferation
of NBC weapons is likely to pose challenges to the United States and
other supporters of a gracefully evolving liberal world order. It is also a
6 the coming crisis
hard look at the kinds of painful dilemmas and actions that will likely be
forced on the responsible world community if strong measures to counter
proliferation are not taken. In this book, six academics join with several
analysts at the Institute for Defense Analyses to explore some of the
implications of continued and uncountered proliferation of NBC weap-
ons. I invited the authors to address any of the following list of questions,
or any alternative question my list suggested:

What changes might continuing proliferation of NBC weapons be

expected to have in the long run on the nature of international
relations? How would such changes affect the interests of the United
States and the larger global community? What could be done to
mitigate these effects if proliferation cannot be halted?

What political-military problems are involved in creating and main-

taining international coalitions for intervening against an NBC-armed
regional challenger?

How must a war against a regional challenger that threatens or

employs NBC weapons end?

What can be learned about a nations biases toward the acquisition

and use of NBC weapons from studying its strategic personality?
Every prospective explorer approached accepted this invitation with
alacrity. Their explorations, presented in the following eight chapters,
provide many arguments and insights that are contrary to the conven-
tional wisdom in this area.
While all the chapters were drafted independently, some have
inuenced others. This resulted from a two-day meeting of the authors
at the Institute for Defense Analyses to present their drafts to a small
group of experts.
The rst part of the book looks at some of the different motivations
states see for acquiring nuclear weapons, and how proliferation is creating
the potential for dangerous crises in which nuclear weapons might get
used. The second part explores other potential consequences of continued
nuclear proliferation, and in particular, how a crisis in which a nuclear-
armed aggressor challenges the United States might evolve. In order of
presentation, then, the main arguments of the chapters are as follows.
Pressures for Nuclear Proliferation and Crises
In Chapter 2, Rethinking the Causes of Nuclear Proliferation: Three
Models in Search of a Bomb, Scott D. Sagan notes the scant attention
paid to the question of why states build nuclear weapons. Sagan argues
chapter 1 7
that this lack of attention follows from a near consensus that nuclear
weapons are only built to meet security threats that cannot be met by
other means (the security model). Sagan challenges this assumption,
presenting evidence that nuclear weapons also serve other less obvious,
parochial purposes. Nuclear weapons are important objects in bureau-
cratic struggles and internal debates (the domestic politics model), and
they can serve as important symbols of a states modernity and identity
(the norms model).
Sagan points out that the most appropriate nonproliferation policies
for a state depend upon the model that best explains why it might seek
nuclear weapons, and that some of the policies called for by different
models can be contradictory. For example, while the security model calls
for extending nuclear deterrence assurances to states facing threats that
might otherwise lead them to build their own weapons, the norms model
argues against giving nuclear weapons the importance that such a role
would suggest. Similarly, the perceived value of nuclear weapons is
raised if the United States, with its great conventional military power,
feels it must deter chemical and biological attacks with the threat of
nuclear retaliation. Finally, Sagan argues that the United States is going
to have to choose either to wean its allies away from extended nuclear
deterrent guarantees or accept the equally difcult task of maintaining a
norm against nuclear proliferation that it does not honor itself.
In Chapter 3, Universal Deterrence or Conceptual Collapse? Liberal
Pessimism and Utopian Realism, Richard K. Betts argues that while the
utopian realists, who see the spread of nuclear weapons as leading to uni-
versal mutual deterrence and military restraint, may correctly predict the
effect of continued proliferation in nearly all cases, the liberal pessimists,
who view the spread of nuclear weapons with alarm, are probably also
right in assuming that increased numbers of nuclear-armed states means
an increased likelihood that nuclear weapons eventually will be used.
Betts further argues that the ramications of a breakdown in the
taboo on nuclear use are too unpredictable for anyone to want to run this
experiment. He describes a variety of ways in which the taboo could
break down. For example, while most states would want nuclear weap-
ons for strictly defensive reasons, a few might become emboldened to try
aggression and end up in a dangerous confrontation with a nuclear-
armed superpower accustomed to intervening in areas of vital interest.
Betts points out that the United States and the Soviet Union took approxi-
mately fteen years to work out ways to avoid dangerous confrontations
and had some very tense moments along the way. He also notes that the
logic of deterrence theory may not be obvious to individuals in countries
8 the coming crisis
for whom these questions are new, and that madness and irrationality do
sometimes occur in the behavior of political leaders.
Finally, Betts argues that while the shock of the next use of nuclear
weapons could lead either to faster proliferation or to far stronger efforts
to roll it back, it seems improbable that the willingness of the United
States and others to rely on nuclear deterrence would remain unshaken.
In Chapter 4, The National Myth and Strategic Personality of Iran:
A Counterproliferation Perspective, Caroline F. Ziemke argues that
every nation has a strategic personality that denes how it is disposed to
behave toward other nations. Ziemke states that this personality can be
discerned by studying a nations public myth, the stories and themes it
uses to illuminate for itself its social and ethical norms and its collective
identity. Thus, an understanding of a proliferators national myth may
provide important insights into why it might want nuclear weapons and
the purposes to which it might put such weapons.
Ziemkes reading of Irans national myth indicates that Iran is su-
premely condent of the superiority of its culture. It sees its troubled
history since the glory of the Persian Empire solely as the result of
invasions and evil inuences from the outside world. Consistent with
this, the United States, with its corrupting material culture and its dec-
ades of meddling in Iranian affairs, is seen as the embodiment of foreign
evil, the Great Satan.
Ziemke employs Irans national myth to interpret its foreign policy
and intentions for nuclear weapons. She argues that Iran wants most of
all to win the respect that its superior culture deserves. It also wants
hegemonic inuence over the Persian Gulf region, which requires that
the United States leave, and it wants to be safe from potential enemies,
particularly Iraq. However, it is not interested in actually conquering its
neighbors. Thus, Ziemke sees Iran wanting nuclear weapons to inspire
respect and fear, and as insurance against invasion, but not as backing
for conventional aggression.
Ziemke also argues that Iran is very unlikely to risk the rst use of
nuclear weapons against the United States or its allies unless it were
about to be overwhelmed. Iran knows the Great Satan is perfectly
willing to annihilate it in response to any rst use of nuclear weapons
against the United States. Its national myth also points to a ruling elite
that will not risk the survival of the Iranian faithful, to whom it sees itself
accountable.
Ziemke points out the contradictory natures of U.S. and Iranian
foreign policies toward each other. Iran wants the United States out of
the Persian Gulf, but threatens its neighbors in ways that increase their
chapter 1 9
interest in U.S. protection. The United States wants Iran to stop support-
ing terrorism and to halt its nuclear program, but reinforces Iranian
paranoia with the dual containment policy, and gives Iran the psycho-
logical victories it craves with every protest against Iranian actions and
every successful penetration of the U.S. arms embargo. Rationalizing
these contradictory policies will take a great deal of time and effort on
both sides, but seems worthwhile given the substantial interests the two
sides actually have in common. Clearly, a good understanding of each
others interests and values will become even more important should Iran
create signicant capabilities to threaten and use NBC weapons.
These three chapters constituting Part I of the book lead me to three
related conclusions. Sagans chapter suggests that effective policies for
stopping nuclear proliferation are going to be even more difcult to nd
and implement than the nonproliferation community has supposed.
Bettss chapter then tells us that there are many ways for dangerous crises
to emerge that would threaten the use of nuclear weapons both as
proliferation continues and even in a fully proliferated world. Third,
Ziemkes chapter suggests that nuclear challenges may be less likely to
emerge from some members of the current rogues gallery, in this case
Iran, than is commonly supposed. Ziemkes chapter is also interesting for
its use of the concept of strategic personality to discern a states likely
behavior. I employ this concept extensively in the concluding chapter.
Collectively, these three chapters suggest that a complete halt to
nuclear proliferation may be even more difcult to achieve than has been
commonly supposed. They also highlight the importance of tailoring
nonproliferation policies to the specic countries and regions in question.
Potential Evolution and Consequences of a Nuclear Crisis with the
United States
In Chapter 5, Nuclear Proliferation and Alliance Relations, Stephen
Peter Rosen argues that with the end of the Cold War, nuclear prolifera-
tion will now weaken alliances rather than strengthen them. He reasons
that the risks of becoming involved in a crisis with a nuclear-armed
regional aggressor have always seemed high, but that the incentives to
become involved are now much weaker. In particular, staying on the
sidelines no longer threatens some disadvantage in a long-term geopo-
litical competition for the highest stakes, nor that the unattended crisis
might somehow catalyze a global nuclear war.
Rosen explores the risks of intervention against a nuclear-armed
regional power by reasoning through a scenario in which Iraq had half
a dozen nuclear weapons when it invaded Kuwait in 1990. Rosen sees
10 the coming crisis
the aggressor as able to employ nuclear weapons to prevent or collapse
support from allies that the United States would need, or to destroy the
ports necessary for a timely intervention before the United States could
reach them. He sees few opportunities for the United States to threaten
or use its nuclear weapons with both comparable military effect and
destruction. Rosen also sees the United States as greatly concerned about
the large numbers of forces it could lose to a nuclear strike against a
crowded port.
Rosen argues that his assessment applies generally to interventions
against nuclear-armed regional states, and believes that it is likely to be
widely understood and appreciated both by regional allies and oppo-
nents. Thus, regional nuclear proliferation will devalue and erode U.S.
alliances. This would happen particularly quickly should a crisis reveal
that the United States is unwilling to intervene against a nuclear-armed
regional aggressor.
In Chapter 6, U.S. Security Policy in a Nuclear-Armed World, or
What If Iraq Had Had Nuclear Weapons? Barry R. Posen also assesses
a counterfactual 199091 scenario in which Iraq had half a dozen surviv-
able and deliverable nuclear weapons. Posen sees this scenario (and thus
its prospective future analogue) as a dening moment for the future of
nuclear weapons, U.S. credibility as a reliable protector, and the nature
of world order.
Posen argues that if the United States had accepted Kuwaits con-
quest by a nuclear-armed Iraq, nuclear weapons would have been re-
dened as effective offensive weapons: they would have deterred a far
stronger United States from rolling back a conquest made with conven-
tional forces. Nuclear weapons would no longer be strictly defenders of
the status quo, as they have been since World War II. Posen argues that
this changed role for nuclear weapons would lead to a hellishly com-
petitive world as aggressors rush to get nuclear weapons and potential
victims, including shaken long-term U.S. allies, scramble to beat them to
it. Incentives would rise for wars to prevent neighbors from getting
nuclear weapons, or to capitalize on military advantages before they are
offset. The world might ultimately settle into equilibrium of many mutual
deterrence relationships, but the long transition would be very exciting.
Posen sees this world as forcing the United States into an uncomfortable
isolation or into adopting very difcult and burdensome policies to
counter the effects of having balked in the rst place.
Thus, Posen argues in favor of intervening. To help minimize the
prospects that Iraq would use its presumed nuclear weapons, he suggests
making ferocious threats of nuclear retaliation, and clearly limiting U.S.
military goals well short of threatening Iraqs total defeat. Posen admits
chapter 1 11
that this reemphasis of nuclear deterrence would do some damage to U.S.
nonproliferation policy, but argues that failure to intervene and to mini-
mize the prospects of nuclear use by the opponent would do even more.
Finally, Posen suggests a campaign to clarify to all states involved in the
crisis the unacceptable long-term consequences for the United States of
failing to roll back Iraqs gains.
Posens analysis of the ofcial strategy debates during the actual
199091 Persian Gulf War suggests that had Iraq actually been nuclear-
armed, his recommendations would have gotten a sympathetic hearing.
Finally, Posen argues the United States will eventually face a dening
moment of this kind and should think it through in advance.
In Chapter 7, Containing Rogues and Renegades: Coalition Strate-
gies and Counterproliferation, Stephen M. Walt argues that history does
not support the pessimism of many strategic planners about prospects
for containing rogue regimes, or their warnings that the spread of nuclear
weapons would have corrosive effects on important U.S. commitments.
Walt says that states have usually been willing to ally with a strong power
to balance threats posed by neighbors having powerful offensive forces
and obvious aggressive intent.
Walt points out that such willingness to confront even powers armed
with nuclear weapons has been well demonstrated by history. He also
points out that great power allies, and particularly the United States, will
be available, given their strong interest in preventing nuclear weapons
from being used for coercion or as a shield for conventional aggression.
Walt argues that defensive coalitions for containing aggressive states
are far easier to form and maintain than offensive coalitions. Defensive
coalitions should prove willing to force an aggressor to relinquish any
conquests, though not to try to overthrow the rogue government. To
facilitate the formation of coalitions for containment, Walt recommends
aggressive collection of intelligence that would reveal the intentions and
capabilities of rogue states. He recommends arrangements to share the
costs and risks in a reasonably equitable manner. He further recommends
assuring potential aggressors that the United States will not seek their
overthrow if they remain peaceful, but will oust them if they attack their
neighbors with nuclear weapons. Walt also highlights the need to main-
tain the strong defensive capabilities required to assure threatened states
that an alliance with the United States can protect them.
In Chapter 8, The Response to Renegade Use of Weapons of Mass
Destruction, George H. Quester argues that if a renegade state were to
use weapons of mass destruction against the United States and its allies,
the U.S. response is more likely to be guided by the norms of the U.S.
law enforcement system and the historical precedents of World War II
12 the coming crisis
than by the Cold War theory that unconditional surrender can never be
sought from a state possessing such weapons.
Quester points out that domestic criminals are jailed for a combina-
tion of four purposes: to disarm them, to make them an example to
others, to impose revenge on behalf of victims, and to reform them.
Similarly, in World War II, criminal regimes in Germany and Japan that
destroyed massive numbers of other states civilians and soldiers were
disarmed. Both states were subjected to very destructive mass bombing
that satised discernible urges for revenge and the creation of examples
for the future. Finally, the political systems of both states were fundamen-
tally reformed. In contrast, had the Soviet Union used nuclear weapons
against the United States and its allies, any realistic pursuit of all four of
these goals would have been unthinkable. A massive attack by the Soviet
Union would have left the United States capable of little more than an
angry and apocalyptic revenge. Limited use might have led to retaliation
aimed more at disarmament, setting an example for the future, and a
lesser revenge. However, pursuit of unconditional surrender and political
reform of the Soviet Union would have risked escalation of the war and
the annihilation of both sides.
Quester argues that renegade states with small to modest capabilities
for mass destruction will be seen as too dangerous to live with once they
use these weapons, but not too dangerous to defeat. Quester notes that
counterproliferation programs that reduce the destruction a renegade
could do would make it even clearer that the United States will not be
bound by Cold War nuclear theory, but will follow its sense that seeking
unconditional surrender of such a renegade is the right thing to do.
Finally, in Chapter 9, Rethinking How Wars Must End: NBC War
Termination Issues in the PostCold War Era, Brad Roberts argues that
in wars against NBC-armed regional challengers, the enormous advan-
tages in power and overall survivability of the United States and its allies
would allow them to choose how these conicts will end. Roberts goes
on to argue that stalemate would likely seem unacceptable, and that
unconditional surrender and political reform of the opponent would
seem necessary, if the opponent does substantial damage with NBC
weapons. Perhaps most important, he argues that the United States will
have to choose a course of action that addresses both the immediate
problems posed by the war and the longer-term U.S. interests in the peace
that follows.
Thus, the United States must not be seen as a nuclear bully that
was overcome by rage and fear and used its weapons in impulsive,
imperious, and excessively destructive ways. Such actions could lead to
a widespread view that U.S. military power is too dangerous and needs
chapter 1 13
to be counterbalanced, and could encourage further proliferation of NBC
weapons. Alternatively, the United States must not be seen as a nuclear
wimp whose fears of staying the course led it to appease an NBC-armed
aggressor. This could encourage tests of U.S. willingness to defend other
interests, undermine U.S. security guarantees, and thus spur further pro-
liferation of NBC weapons by allies. It could also lead to a political
backlash against those who had caused such a U.S. decline.
Instead, Roberts argues that the United States should seek to be seen
as a responsible and just steward of the collective good. This requires
ending the conict in ways that resolve its underlying cause, that remove
the threat posed by the aggressor, and that use nuclear weapons only to
the extent that they appear needed to end the war and to save lives.
Finally, Roberts argues that emerging from such a war as a just and
responsible steward will be easier if the United States does three things.
First, it must reduce its own and help to reduce its allies vulnerabilities
to NBC attack. Second, it must work with allies, the Congress, and the
public to shape the political context for regional conict involving NBC
weapons. And, third, it is especially important to let potential aggressors
understand how the United States would likely see its alternatives in such
a conict.
The main conclusion I draw from Part II of the book is that the United
States seems likely to prove more resolute than much of the community
that is expert on nuclear proliferation and its potential consequences
seems to believe. Rosen argues that the United States would not see a net
advantage in opposing regional aggression backed by nuclear weapons
and thus that potential allies would not trust alliance with the United
States to save them: all the following chapters argue the opposite.
In closing this introductory chapter, let me emphasize that I have
presented only a few of the points made by each of my colleagues, and
even fewer of the strong justications they present in defense of their
points. Moreover, my selections and renditions of their arguments are
surely colored by my own views. Thus, I delay any further comments or
interpretations of their work until Chapter 10, when the reader will
have had a chance to read their work and to form his or her own
opinions.
Chapter 10, which notes the different perspectives that the authors
have brought to their analyses, argues that projections of the likely be-
havior of the United States and other nations can prot from bringing
more consciously balanced mixtures of perspectives to bear. It then adds
a strong measure of the strategic personality perspective to those of the
other authors in an attempt to project likely U.S. behavior in response to
14 the coming crisis
several of the more important questions that would be posed when
confronting nuclear-backed aggression.
* * *
Finally, I would like to thank Larry Welch, Bob Roberts, Phil Major,
and Mike Leonard of the Institute for Defense Analyses for their steadfast
support of this project. I also owe special thanks to Brad Roberts for his
invaluable comments and suggestions on both my introductory and con-
cluding chapters, and to Caroline Ziemke for her comments on the con-
cluding chapter, which draws extensively on her research on strategic
personalities. Rafael Bonoan was very helpful in providing supporting
literature and analysis for Chapter 10. Richard Falkenrath and Leo
MacKay deserve a salute for their frequent encouragement to create and
especially to nish this book. Sean-Lynn Jones and Karen Motley made
excellent suggestions on the organization and presentation of the book,
and Miriam Avinss editorial suggestions have made the book far more
readable and to the point. Kristina Cherniahivsky deserves great credit
for her patient and long-suffering search for a suitable cover photo.
Johnathan Wallis wins my thanks for his careful review of the entire text.
Last, but not least, Olga Alvarado deserves special thanks for incorporat-
ing a myriad of changes into the text, and keeping the manuscript in its
various forms organized and moving.
chapter 1 15
Chapter 2
Rethinking the Causes of
Nuclear Proliferation:
Three Models in Search of a
Bomb
Scott D. Sagan
Why do states build nuclear weapons? An accurate answer to this ques-
tion is critically important both for predicting the long-term future of
international security and for current foreign policy efforts to prevent the
spread of nuclear weapons. Yet given the importance of this central
proliferation puzzle, it is surprising how little sustained attention has
been devoted to examining and comparing alternative answers.
This lack of critical attention is not due to a lack of information: there
is now a large literature on nuclear decision-making inside the states that
have developed nuclear weapons, and a smaller but still signicant set
of case studies of states decisions to refrain from developing nuclear
weapons. Instead, the inattention appears to reect a near consensus that
the answer is obvious. Many U.S. policymakers and most international
relations scholars have a clear and simple answer to the proliferation
puzzle: states will seek to develop nuclear weapons when they face a
signicant military threat to their security that cannot be met through
alternative means; if they do not face such threats, they will willingly
remain non-nuclear states.
1
An earlier version of this essay appeared as Why Do States Build Nuclear Weapons?
Three Models in Search of a Bomb, in International Security, Vol. 21, No. 3 (Winter
1996/97).
1. Among policymakers, John Deutch presents the most unadorned summary of the
basic argument that the fundamental motivation to seek a weapon is the perception
that national security will be improved. John M. Deutch, The New Nuclear Threat,
Foreign Affairs, Vol. 71, No. 41 (Fall 1992), pp. 124125. Also see George Shultz,
Preventing the Proliferation of Nuclear Weapons, Department of State Bulletin, Vol.
84, No. 2093 (December 1984), pp. 1721. For examples of the dominant paradigm
among scholars, see Michael M. May, Nuclear Weapons Supply and Demand,
American Scientist, Vol. 82, No. 6 (NovemberDecember 1994), pp. 526537; Bradley A.
Thayer, The Causes of Nuclear Proliferation and the Nonproliferation Regime,
Security Studies, Vol. 4, No. 3 (Spring 1995), pp. 463519; Benjamin Frankel, The
The central purpose of this chapter is to challenge this conventional
wisdom about nuclear proliferation. I argue that the consensus view,
which focuses on national security considerations as the cause of prolif-
eration, is dangerously inadequate because nuclear weapons programs
also serve more parochial and less obvious objectives. Like other weap-
ons, nuclear weapons are more than tools of national security; they are
political objects of considerable importance in domestic politics and bu-
reaucratic struggles and also serve as international symbols of modernity
and identity.
The chapter examines three alternative theoretical frameworks
what I call models in the very informal sense of the termabout why
states decide to build or refrain from developing nuclear weapons: the
security model, according to which states build nuclear weapons to
increase national security against foreign threats, especially nuclear
threats; the domestic politics model, which envisions nuclear weapons
as political tools used to advance parochial domestic and bureaucratic
interests; and the norms model, under which nuclear weapons acqui-
sition, or restraint in weapons development, is determined by the role of
such weapons as a symbol of a states modernity and identity. Although
many of the ideas underlying these models appear in the case study and
nonproliferation policy literature, they have not been adequately ana-
lyzed, nor placed in a comparative theoretical framework, nor properly
evaluated against empirical evidence. When I discuss these models, there-
fore, I compare their explanations for nuclear weapons development,
present alternative interpretations of the history of some major prolifera-
tion decisions, and contrast the models implications for nonproliferation
policy. The article concludes with an outline of a research agenda for
future proliferation studies and an examination of the policy dilemmas
illuminated by these models.
The crafting of effective nonproliferation policies will remain critical.
Despite the successful 1995 agreement to permanently extend the Nuclear
Non-Proliferation Treaty (NPT), there will be continuing NPT review
conferences assessing the implementation of the treaty every ve years;
each member state can legally withdraw from the treaty, under the su-
preme national interest clause, if it gives three months notice; and many
Brooding Shadow: Systemic Incentives and Nuclear Weapons Proliferation, and Rich-
ard K. Betts, Paranoids, Pygmies, Pariahs, and Nonproliferation Revisited, both in
Zachary S. Davis and Benjamin Frankel, eds., The Proliferation Puzzle, special issue of
Security Studies, Vol. 2, Nos. 3/4 (Spring/Summer 1993), pp. 3738 and pp. 100124;
and David Gompert, Kenneth Watman, and Dean Wilkening, Nuclear First Use
Revisited, Survival, Vol. 37, No. 3 (Autumn 1995), p. 39.
18 the coming crisis
new states can be expected to develop, at a minimum, a latent nuclear
weapons capability over the coming decade. Indeed, some fty-seven
states now operate or are constructing nuclear power or research reactors,
and it has been estimated that about thirty countries today have the
necessary industrial infrastructure and scientic expertise to build nu-
clear weapons on a crash basis if they chose to do so.
2
The NPT encour-
ages this long-term trend by promoting the development of power reac-
tors in exchange for the imposition of International Atomic Energy
Agency (IAEA) safeguards on the resulting nuclear materials. This sug-
gests that while most attention concerning proliferation in the immediate
term has appropriately focused on controlling nuclear materials in the
former Soviet Union, managing the new nuclear relationship between
India and Pakistan, and preventing the small number of suspected active
proliferators (such as Iraq, Iran, Libya, and North Korea) that currently
appear to have vigorous nuclear weapons programs from getting the
bomb, the longer-term and enduring proliferation problem will be to
ensure that the larger and continually growing number of latent nuclear
states maintain their non-nuclear weapons status. This underscores the
policy importance of addressing the sources of the political demand for
nuclear weapons, rather than focusing primarily on efforts to safeguard
existing stockpiles of nuclear materials and to restrict the supply of specic
weapons technology from the haves to the have-nots.
If my arguments and evidence concerning the three models of pro-
liferation are correct, however, any future demand-side nonproliferation
strategy will face inherent contradictions. For, in contrast to the views of
scholars who claim that a traditional realist theory focusing on security
threats explains all cases of proliferation and nuclear restraint,
3
the his-
torical evidence presented here suggests that each theory explains some
past cases quite well and others quite poorly. Unfortunately, since the
models provide different and often contradictory lessons for U.S. nonpro-
liferation policy, policies designed to address one future proliferation
2. See Steve Fetter, Verifying Nuclear Disarmament, Occasional Paper No. 29,
Henry L. Stimson Center, Washington, D.C., October 1996, p. 38; and Afliations and
Nuclear Activities of 172 NPT Parties, Arms Control Today, Vol. 25, No. 2 (March 1995),
pp. 3336. For earlier pioneering efforts to assess states nuclear weapons latent capa-
bility and demand, see Stephen M. Meyer, The Dynamics of Nuclear Proliferation (Chi-
cago: University of Chicago Press, 1984); and William C. Potter, Nuclear Power and
Nonproliferation (Cambridge, Mass: Oelgeschlager, Gunn and Hain, 1982).
3. For example, May, Nuclear Weapons Supply and Demand; Thayer, The Causes
of Nuclear Proliferation and the Nonproliferation Regime; and Frankel, The Brood-
ing Shadow: Systemic Incentives and Nuclear Weapons Proliferation.
chapter 2 19
problem will likely exacerbate others. As I discuss in more detail below,
particularly severe tensions are likely to emerge in the future between
U.S. extended deterrence policies designed to address security concerns
of potential proliferators and U.S. nonproliferator policies designed to
maintain and enhance international norms against nuclear use and acqui-
sition.
The Security Model: Nuclear Weapons and International Threats
According to neorealist theory in political science, states exist in an
anarchical international system and must therefore rely on self-help to
protect their sovereignty and national security.
4
Because of the enormous
destructive power of nuclear weapons, any state that seeks to maintain
its national security must balance against any rival state that develops
nuclear weapons by gaining access to a nuclear deterrent itself. Strong
states will pursue a form of internal balancing by adopting the costly but
self-sufcient policy of developing their own nuclear weapons. Weak
states will join a balancing alliance with a nuclear power, utilizing a
promise of nuclear retaliation by that ally as a means of extended deter-
rence. Acquiring a nuclear ally may be the only option available to weak
states, but the policy inevitably raises questions about the credibility of
extended deterrence guarantees, since the nuclear power would also fear
retaliation if it responded to an attack on its ally.
Although states might also choose to develop nuclear weapons to
serve either as deterrents against overwhelming conventional military
threats or as coercive tools to compel changes in the status quo, a focus
on states responses to emerging nuclear threats is the most common and
most parsimonious explanation for nuclear weapons proliferation.
5
George Shultz once nicely summarized the argument: Proliferation be-
4. The seminal text of neorealism remains Kenneth N. Waltz, Theory of International
Politics (New York: Random House, 1979). Also see Kenneth N. Waltz, The Origins
of War in Neorealist Theory, in Robert I. Rotberg and Theodore K. Rabb, eds., The
Origin and Prevention of Major Wars (New York: Cambridge University Press, 1989),
pp. 3952; and Robert O. Keohane, ed., Neorealism and Its Critics (New York: Columbia
University Press, 1986).
5. The Israeli, and possibly the Pakistani, nuclear weapons decisions might be the
best examples of defensive responses to conventional security threats; Iraq, and pos-
sibly North Korea, might be the best examples of the offensive coercive threat moti-
vation. On the status quo bias in neorealist theory in general, see Randall L. Schweller,
Bandwagoning for Prot: Bringing the Revisionist State Back In, International Secu-
rity, Vol. 19, No. 1 (Summer 1994), pp. 72107; and Richard Rosecrance and Arthur A.
Stein, eds., The Domestic Bases of Grand Strategy (Ithaca, N.Y.: Cornell University Press,
1993).
20 the coming crisis
gets proliferation.
6
Every time one state develops nuclear weapons to
balance against its main rival, it also creates a nuclear threat to another
state in the region, which then must initiate its own nuclear weapons
program to maintain its national security.
From this perspective, the history of nuclear proliferation is a strate-
gic chain reaction. During World War II, none of the major belligerents
was certain that the development of nuclear weapons was possible, but
all knew that other states were already working or could soon be working
to build the bomb. This fundamental fear was the central impetus for the
U.S., British, German, Soviet, and Japanese nuclear weapons programs.
The United States developed atomic weapons rst, not because it had any
greater demand for the atomic bomb than these other powers but because
it invested more heavily in the program and made the right set of
technological and organizational choices.
7
After August 1945, the Soviet
Unions program was reinvigorated because the U.S. atomic attacks on
Hiroshima and Nagasaki demonstrated that nuclear weapons were tech-
nically possible, and the emerging Cold War made a Soviet bomb a
strategic imperative. From the realist perspective, the Soviet response was
perfectly predictable. Josef Stalins reported request to Igor Kurchatov
and B.L. Yannikov in August 1945 appears like a textbook example of
realist logic: A single demand of you comrades. . . . Provide us with
atomic weapons in the shortest possible time. You know that Hiroshima
has shaken the whole world. The balance has been destroyed. Provide
the bombit will remove a great danger from us.
8
Other states decisions to develop nuclear weapons can also be ex-
plained within the same framework. London and Paris are thus seen to
have built nuclear weapons because of the growing Soviet military threat
and the inherent reduction in the credibility of the U.S. nuclear guarantee
to NATO (North Atlantic Treaty Organization) allies once the Soviet
Union was able to threaten retaliation against the United States.
9
China
6. Shultz, Preventing the Proliferation of Nuclear Weapons, p. 18.
7. On the genesis of the atomic programs in World War II, see McGeorge Bundy,
Danger and Survival: Choices about the Bomb in the First Fifty Years (New York: Random
House, 1988) pp. 353; and Richard Rhodes, The Making of the Atomic Bomb (New York:
Simon and Schuster, 1986).
8. A. Lavrentyeva in Stroiteli novogo mira, V mire knig, No. 9 (1970), in David
Holloway, The Soviet Union and the Arms Race (New Haven, Conn.: Yale University
Press, 1980), p. 20; also quoted in Thayer, The Causes of Nuclear Proliferation,
p. 487.
9. Important sources on the British case include Margaret Gowing, Britain and Atomic
Energy, 19391945 (London: Macmillan, 1964); Margaret Gowing, Independence and
Deterrence: Britain and Atomic Energy, 19451952, vols. 1 and 2 (London: Macmillan,
chapter 2 21
developed the bomb because Beijing was threatened with possible nu-
clear attack by the United States at the end of the Korean War and again
during the Taiwan Straits crises in the mid-1950s. Not only did Moscow
prove to be an irresolute nuclear ally in the 1950s, but the emergence of
hostility in Sino-Soviet relations in the 1960s further encouraged Beijing
to develop, in Avery Goldsteins phrase, the robust and affordable secu-
rity of nuclear weapons, since the border clashes again exposed the
limited value of Chinas conventional deterrent.
10
After China developed the bomb in 1964, India, which had just
fought a war with China in 1962, was bound to follow suit. Indias
strategic response to the Chinese test came a decade later, when its
Atomic Energy Commission successfully completed the long research
and development process required to construct and detonate what was
called a peaceful nuclear explosion (PNE) in May 1974. According to
realist logic, India maintained an ambiguous nuclear posture until 1998
building sufcient nuclear materials and components for a moderate-
sized nuclear arsenal, but not testing or deploying weapons into the
eldin a clever effort to deter the Chinese without encouraging nuclear
weapons programs in other neighboring states.
11
After Indias 1974 test,
it was inevitable that Pakistan, which now faced a hostile neighbor with
both a nuclear weapons capability and conventional military superiority,
would seek to produce a nuclear weapon as quickly as possible.
12
By the
late 1990s, Indian ofcials believed that it faced both a small nuclear
1974); and Andrew Pierre, Nuclear Politics: The British Experience with an Independent
Strategic Force, 19391970 (London: Oxford University Press, 1972). On the French case,
see Lawrence Scheinman, Atomic Energy Policy in France Under the Fourth Republic
(Princeton, N.J.: Princeton University Press, 1965), and Wilfred L. Kohl, French Nuclear
Diplomacy (Princeton, N.J.: Princeton University Press, 1971).
10. Avery Goldstein, Robust and Affordable Security: Some Lessons from the Sec-
ond-Ranking Powers During the Cold War, Journal of Strategic Studies, Vol. 15, No. 4
(December 1992), p. 494. The seminal source on the Chinese weapons program, which
emphasizes the importance of U.S. nuclear threats in the 1950s, is John W. Lewis and
Xue Litai, China Builds the Bomb (Stanford, Calif.: Stanford University Press, 1988).
11. For important works on the history of the Indian nuclear program, see Itty
Abraham, The Making of the Indian Atomic Bomb: Science, Society, and the Postcolonial State
(London: Zed Books, 1998); George Perkovich, Indias Ambiguous Bomb (Berkeley, Calif.:
University of California Press, 1999); Ashok Kapur, Indias Nuclear Option: Atomic
Diplomacy and Decision Making (New York: Praeger, 1976); and T.T. Poulose, ed., Per-
spectives of Indias Nuclear Policy (New Delhi: Young Asia Publications, 1978).
12. Valuable sources on Pakistans program include Ziba Moshaver, Nuclear Weapons
Proliferation in the Indian Subcontinent (Basingstoke, U.K.: Macmillan, 1991); Ashok
Kapur, Pakistans Nuclear Development (New York: Croom Helm, 1987); and Samina
Ahmed, Pakistans Nuclear Weapons Program, International Security, Vol. 23, No. 4
(Spring 1999), pp. 178204.
22 the coming crisis
Pakistani capability, developed with the assistance of China, and the
continuing Chinese nuclear threat. These twin security problems forced
the Indian government to move its nuclear capability out of the basement,
testing ve nuclear weapons on May 11 and May 13, 1998. We live in a
world where India is surrounded by nuclear weaponry, Prime Minister
Atal Vajpayee exclaimed soon after he announced the Indian tests.
13
Predictably, Pakistan quickly followed suit: the government in Islamabad
announced that it too had tested nuclear weapons on May 28 and May
30, less than a month after the Indian tests.
explaining nuclear restraint
Given the strong deterrent capabilities of nuclear weapons, why would
any state that is technically capable of developing the bomb voluntarily
give up such powerful tools of security? Cases of nuclear weapons re-
versal can also be viewed through the lens provided by the security
model if one assumes that external security threats can radically change
or be reevaluated. The case of South Africa has most often been analyzed
in this light, with the new security threats that emerged in the mid-1970s
seen as the cause of South Africas bomb program and the end of these
threats in the late 1980s as the cause of its policy reversal. As President
F.W. de Klerk explained in his speech to Parliament in March 1993, the
Pretoria government saw a growing Soviet expansionist threat to south-
ern Africa; the buildup of the Cuban forces in Angola from 1975
onwards reinforced the perception that a deterrent was necessary, as did
South Africas relative international isolation and the fact that it could not
rely on outside assistance should it be attacked.
14
Six atomic weapons
were therefore constructed, but were stored disassembled in a secret
location from 1980 to 1989, when the program was halted. It has been
claimed that South African nuclear strategy during this period was de-
signed to use the bomb both as a deterrent against the Soviets and as a
13. We Have Shown Them That We Mean Business, interview with Prime Minister
Vajpayee, India Today, May 25, 1998 (www.india-today.com/itoday/25051998/
vajint.html).
14. F.W. de Klerk, March 24, 1993, address to the South African Parliament as tran-
scribed in Foreign Broadcast Information Service (FBIS), JPRS-TND-93-009 (March 29,
1993), p. 1 (henceforth cited as de Klerk, Address to Parliament). For analyses that
focus largely on security threats as the cause of the program, see Darryl Howlett and
John Simpson, Nuclearization and Denuclearization in South Africa, Survival, Vol.
35, No. 3 (Autumn 1993), pp. 154173; and J.W. de Villers, Roger Jardine, and Mitchell
Reiss, Why South Africa Gave up the Bomb, Foreign Affairs, Vol. 72, No. 5 (Novem-
ber/December 1993), pp. 98109. For a more detailed and more balanced perspective
see Mitchell Reiss, Bridled Ambition: Why Countries Constrain Their Nuclear Capabilities
(Washington, D.C.: Woodrow Wilson Center Press, 1995), pp. 744.
chapter 2 23
tool of blackmail against the United States. If Soviet or Soviet-supported
military forces directly threatened South Africa, the regime reportedly
planned to announce that it had a small arsenal of nuclear weapons, and
dramatically test one or more of the weapons if necessary by dropping
them from aircraft over the ocean, hoping that such a test would shock
the United States into intervention on behalf of the Pretoria regime.
15
South Africa destroyed its small nuclear weapons arsenal in 1991, realist
theory suggests, because of the radical reduction in the external security
threats to the regime. By 1989, the risk of a Soviet-led or sponsored attack
on South Africa was virtually eliminated. President de Klerk cited three
specic changes in military threats in his speech to Parliament: a cease-
re had been negotiated in Angola; the tripartite agreement granted
independence to Namibia in 1988; and most importantly, the Cold War
had come to an end.
16
Although the details vary, the basic security model has also been used
to explain other examples of nuclear restraint. For example, both Argen-
tina and Brazil refused to complete the steps necessary to join the Latin
American nuclear weaponsfree zone (NWFZ) and began active pro-
grams in the 1970s that eventually could have produced nuclear weap-
ons; however, their 1990 joint declaration of plans to abandon their
programs is seen as the natural result of the recognition that the two
states, which had not fought a war against one another since 1828, posed
no fundamental security threat to each other.
17
Similarly, it has been
argued that the non-Russian former states of the Soviet Union that were
born nuclearUkraine, Kazakhstan, and Belarusdecided to give up
their arsenals because of a mixture of two realist model arguments: their
long-standing close ties to Moscow meant that these states did not per-
ceive Russia as a major military threat to their security and sovereignty,
and increased U.S. security guarantees to these states made their posses-
15. Military planners nonetheless developed nuclear target lists in their contingency
military plans, and research was conducted on the development of the hydrogen bomb
until 1985. See Reiss, Bridled Ambition, p. 16.
16. See de Klerk, Address to Parliament, p. 2.
17. Thayer, The Causes of Nuclear Proliferation, p. 497; and May, Nuclear Weap-
ons Supply and Demand, pp. 534535. For analyses of the Argentine-Brazilian deci-
sion, see Monica Serrano, Brazil and Argentina, in Mitchell Reiss and Robert S.
Litwak, eds., Nuclear Proliferation After the Cold War (Washington, D.C.: Woodrow
Wilson Center Press, 1994), pp. 231255; Jose Goldemberg and Harold A. Feiveson,
Denuclearization in Argentina and Brazil, Arms Control Today, Vol. 24, No. 2 (March
1994), pp. 1014; Reiss, Bridled Ambition, pp. 4588; and John R. Redick, Julio C.
Carasales, and Paulo S. Wrobel, Nuclear Rapprochement: Argentina, Brazil, and the
Nonproliferation Regime, Washington Quarterly, Vol. 18, No. 1 (Winter 1995), pp.
107122.
24 the coming crisis
sion of nuclear weapons less necessary.
18
In short, from a realists per-
spective, nuclear restraint is caused by the absence of the fundamental
military threats that produce positive proliferation decisions.
policy implications of the security model
Several basic predictions and prescriptions ow naturally from the logic
of the security model. First, since states that face nuclear adversaries will
eventually develop their own arsenals unless they secure credible alliance
guarantees from a nuclear power, it is essential that the United States
maintain its nuclear commitments to key allies, including some form of
continued rst-use policy.
19
Other efforts to enhance the security of po-
tential proliferatorssuch as condence-building measures or negative
security assurances that the nuclear states will not use their weapons
against non-nuclear statescan also be helpful in the short run, but will
likely not be effective in the long term since the anarchic international
system forces states to be suspicious of potential rivals.
Under the security models logic, the NPT is seen as an institution
permitting non-nuclear states to overcome a collective action problem.
Non-nuclear states in the NPT would prefer to become the only nuclear
weapons power in their regions, but since that is an unlikely outcome if
one state develops a nuclear arsenal, each is willing to refrain from
proliferation if, and only if, its neighbors remain non-nuclear. The treaty
permits such states to exercise restraint with increased condence that
their neighbors will follow suit, or will at least give sufcient advance
warning before a withdrawal from the regime. It follows, from this logic,
that other elements of the NPT regime should be considered far less
important. Specically, the commitments that the United States and other
nuclear states made under Article VI of the treatythat they will pursue
negotiations in good faith on measures relating to cessation of the
nuclear arms race at an early date and to nuclear disarmamentare
18. Sherman Garnett writes, for example, that for many Ukrainian citizensnot just
the ethnic Russiansit is difcult to conceive of Russia as an enemy to be deterred
with nuclear weapons. Sherman W. Garnett, Ukraines Decision to Join the NPT,
Arms Control Today, Vol. 25, No. 1 (January 1995), p. 8. Garnett also maintains that the
role that security assurances played in the creation of a framework for Ukrainian
denuclearization is obvious. They were of immense importance. Sherman W. Garnett,
The Role of Security Assurances in Ukrainian Denuclearization, in Virginia Foran,
ed., Missed Opportunities? The Role of Security Assurances in Nuclear Non-Proliferation
(Washington, D.C.: Carnegie Endowment for International Peace, 1997).
19. See Lewis Dunn, Controlling the Bomb (New Haven, Conn.: Yale University Press,
1982); May, Nuclear Weapons Supply and Demand, p. 535; and Frankel, The Brood-
ing Shadow, pp. 4754.
chapter 2 25
merely sops to public opinion in non-nuclear countries. The degree to
which the nuclear states follow through on these Article VI commitments
will not signicantly inuence the actual behavior of non-nuclear states,
since it will not change their security status.
Under realist logic, however, U.S. nonproliferation policy can only
slow, not eliminate, the spread of nuclear weapons. Efforts to slow the
process may be useful, but they will eventually be countered by two very
strong structural forces that create an inexorable momentum toward a
world of numerous nuclear weapons states. First, the end of the Cold War
created a more uncertain multipolar world in which U.S. nuclear guar-
antees will be considered increasingly less reliable; second, each time one
state develops nuclear weapons, it will increase the strategic incentives
for neighboring states to follow suit.
20
problems and evidence
Whats wrong with this picture? The security model is parsimonious; the
resulting history is conceptually clear; and the theory ts our intuitive
belief that important events in history (like the development of a nuclear
weapon) must have equally important causes (like national security
threats). A major problem exists, however, concerning the evidence. The
realist history depends primarily on the statements of motivation by the
key governmental leaders, who have a vested interest in explaining that
the choices they made served the national interest, and on a correlation
in time between the emergence of a plausible security threat and a
decision to develop nuclear weapons. Indeed, many realist case studies
observe a nuclear weapons decision and then work backwards, looking
for the national security threat that must have caused the decision.
Similarly, realist scholars too often observe a state decision not to have
nuclear weapons and then work backwards to nd the change in the
international environment that must have led the government to be-
lieve that threats to national security were radically decreasing.
A deeper analysis opens up the black box of decision-making and
reveals in more detail how governments actually made their nuclear
decisions. Moreover, any rigorous attempt to evaluate the security model
of proliferation also requires an effort to develop alternative explanations,
and to assess whether they provide more or less compelling explanations
for proliferation decisions. The following sections therefore develop a
20. See Kenneth N. Waltz, The Emerging Structure of International Politics, Inter-
national Security, Vol. 18, No. 2 (Fall 1993), pp. 4479; and John J. Mearsheimer, Back
to the Future: Instability in Europe after the Cold War, International Security, Vol. 15,
No. 1 (Summer 1990), pp. 556.
26 the coming crisis
domestic politics model and a norms model of proliferation and evaluate
the explanations that ow from their logic, versus the security models
arguments offered above, for some important cases of both nuclear pro-
liferation and nuclear restraint.
The Domestic Politics Model: Nuclear Pork and Parochial Interests
A second model of nuclear weapons proliferation focuses on the domestic
actors who encourage or discourage governments from pursuing the
bomb. Whether or not the acquisition of nuclear weapons serves the
national interests of a state, it is likely to serve the parochial interests of
at least some individual actors within the state. Three kinds of actors are
often important in proliferation decisions: the states nuclear energy es-
tablishment (which includes ofcials in state-run laboratories as well as
in civilian reactor facilities); important units within the professional mili-
tary (often within the air force, though sometimes in navy bureaucracies
interested in nuclear propulsion); and politicians in states where individ-
ual parties or the mass public strongly favor nuclear weapons acquisition.
When such actors form coalitions that are strong enough to control the
governments decision-making processeither through their direct po-
litical power or indirectly through their control of informationnuclear
weapons programs are likely to thrive.
Unfortunately, there is no well-developed domestic political theory
of nuclear weapons proliferation that identies the conditions under
which such coalitions are formed and become powerful enough to pro-
duce their preferred outcomes.
21
The basic logic of this approach, how-
ever, has been strongly inuenced by the literature on bureaucratic poli-
tics and military technology decision-making in the United States and the
Soviet Union during the Cold War.
22
In this literature, bureaucratic actors
are not seen as passive recipients of top-down political decisions; instead,
21. This is a serious weakness shared by many domestic-level theories in interna-
tional relations, not just theories of proliferation. On this issue, see Ethan B. Kapstein,
Is Realism Dead? The Domestic Sources of International Politics, International Or-
ganization, Vol. 49, No. 4 (Autumn 1995), pp. 751774.
22. The best examples of this literature include Morton H. Halperin, Bureaucratic
Politics and Foreign Policy (Washington, D.C.: The Brookings Institution, 1974); Matthew
Evangelista, Innovation and the Arms Race: How the United States and Soviet Union Develop
New Military Technologies (Ithaca, N.Y.: Cornell University Press, 1988); and Donald
MacKenzie, Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance (Cam-
bridge, Mass.: MIT Press, 1990). For a valuable effort to apply insights from the
literature on the social construction of technology to proliferation problems, see Steven
Flank, Exploding the Black Box: The Historical Sociology of Nuclear Proliferation,
Security Studies, Vol. 3, No. 2 (Winter 1993/94), pp. 259294.
chapter 2 27
they create the conditions that favor weapons acquisition by encouraging
extreme perceptions of foreign threats, promoting supportive politicians,
and actively lobbying for increased defense spending. This bottom-up view
focuses on the formation of coalitions within a states scientic- military-
industrial complex. The initial ideas for individual weapons innovations
are often developed inside state laboratories, where scientists favor mili-
tary innovation simply because it is technically exciting and keeps money
and prestige owing to their laboratories. Such scientists are then able to
nd, or even create, sponsors in the professional military whose bureau-
cratic interests and specic military responsibilities lead them also to
favor the particular weapons system. Finally, such a coalition builds
broader political support within the executive or legislative branches by
shaping perceptions about the costs and benets of weapons programs.
Of course, realists also recognize that domestic political actors have
parochial interests, but they argue that such interests have only a mar-
ginal inuence on crucial national security issues. The outcome of bu-
reaucratic battles, for example, may well determine whether a state builds
900 or 1,000 ICBMs (intercontinental ballistic missiles) or emphasizes
submarines or strategic bombers in its nuclear arsenal; but a strong
consensus among domestic actors will soon emerge about the basic need
to respond in kind when a potential adversary acquires nuclear weapons.
In contrast, from the domestic politics perspective nuclear weapons pro-
grams are not obvious or inevitable solutions to international security
problems; instead, nuclear weapons programs are solutions looking for a
problem to justify their existence. In this model, international threats
certainly exist, but are seen as more malleable and more subject to inter-
pretation, and can therefore produce a variety of responses from domestic
actors. Security threats are not the central cause of weapons decisions
according to this model: they are merely windows of opportunity through
which parochial interests can jump.
proliferation revisited: addressing the india puzzles
The historical case that most strongly ts the domestic politics model is
the Indian nuclear weapons experience. A focus on domestic politics in
New Delhi, rather than on the strategic threats posed by China and
Pakistan, leads to a very different interpretation of Indias nuclear weap-
ons decisions. This is true both of Indias rst test of a peaceful nuclear
explosive in 1974 and its May 1998 nuclear weaponization tests.
In contrast to the brief realists account outlined above, a closer look
at the genesis of the Indian bomb program reveals that there was no
consensus among ofcials in New Delhi that it was necessary to have a
nuclear deterrent as a response to Chinas 1964 nuclear test. According
28 the coming crisis
to realist logic, a consensus should have existed and produced one of two
events. First, a crash weapons program could have been initiated; there
is no evidence that such an emergency program was started, however,
and indeed, given the relatively advanced state of Indian nuclear energy
at the time, such an effort could have produced a nuclear weapon by the
mid- to late 1960s, relatively soon after the Chinese test, instead of in
1974.
23
Second, leaders in New Delhi could have made a concerted effort
to acquire nuclear guarantees from the United States, the Soviet Union,
or other nuclear powers. Indian ofcials, however, did not adopt a con-
sistent policy to pursue security guarantees: in diplomatic discussions
after the Chinese test, ofcials rejected the idea of bilateral guarantees
because they would not conform with Indias nonaligned status; refused
to consider allowing foreign troops in India to support a nuclear commit-
ment; and publicly questioned whether any multilateral or bilateral guar-
antee could possibly be credible.
24
Instead of producing a united Indian effort to acquire a nuclear
deterrent, the Chinese nuclear test produced a prolonged bureaucratic
battle, fought inside the New Delhi political elite and nuclear energy
establishment, between actors who wanted India to develop a nuclear
weapons capability as soon as possible and other actors who opposed an
Indian bomb and supported global nuclear disarmament and later Indian
membership in the NPT. Soon after the Chinese nuclear test, for example,
Prime Minister Lal Bahadur Shastri argued against developing an Indian
atomic arsenal, in part because the estimated costs of $4284 million were
deemed excessive; Homi Bhabba, the head of the Atomic Energy Com-
mission (AEC), however, loudly lobbied for the development of nuclear
weapons capability, claiming that India could develop a bomb in eighteen
months and that an arsenal of fty atomic bombs would cost less than
$21 million (a gure that excluded the construction of reactors, separation
plants, and the opportunity costs of diverting scientists from develop-
ment projects).
25
Although Shastri continued to oppose weapons devel-
23. In 1963, U.S. intelligence agencies estimated that India could test a nuclear
weapon in four to ve years (1967 or 1968). By 1965, U.S. estimates were that it would
take one to three additional years. See Peter R. Lavoy, Nuclear Myths and the Causes
of Proliferation, in Davis and Frankel, The Proliferation Puzzle, p. 202; and George
Bunn, Arms Control by Committee: Managing Negotiations with the Russians (Stanford,
Calif.: Stanford University Press, 1992), p. 68.
24. See A.G. Noorani, Indias Quest for a Nuclear Guarantee, Asian Survey, Vol. 7,
No. 7 (July 1967), pp. 490502.
25. Frank E. Couper, Indian Party Conict on the Issue of Atomic Weapons, Journal
of Developing Areas, Vol. 3, No. 2 (January 1969), pp. 192193. Also see Lavoy, Nuclear
Myths and the Causes of Proliferation, p. 201.
chapter 2 29
opment and rebuked legislators in congressional debates for quoting
Bhabbas excessively optimistic cost estimates, he compromised with the
pro-bomb members of the Congress party and the AEC leadership, agree-
ing to create a classied project to develop the ability to detonate a PNE
within six months of any nal political decision.
26
However, even this
compromise was short-lived, as Bhabbas successor at the AEC, Vikram
Sarabhai, opposed the development of any Indian nuclear explosives,
whether they were called PNEs or bombs, and ordered a halt to the PNE
preparation program.
27
After Sarabhais death in 1971, the pro-bomb scientists in the AEC
began to lobby Prime Minister Indira Gandhi, and developed an alliance
with defense laboratories whose participation was needed to fabricate the
explosive lenses for a nuclear test.
28
Unfortunately, rm evidence on why
Gandhi decided to approve the scientists recommendation to build and
test a peaceful Indian nuclear device does not exist: indeed, even
nuclear scientists who pushed for the May 1974 test now acknowledge
that it is impossible to know whether Gandhi was primarily responding
to domestic motivesshe neither asked questions at the critical secret
meetings in early 1974 nor explained why she approved their PNE rec-
ommendations.
29
Three observations about the decision, however, do
suggest that domestic political concerns, rather than international secu-
rity threats, were paramount. First, the decision was made by Prime
Minister Gandhi, with the advice of a very small circle of personal
advisers and scientists from the nuclear establishment. Senior defense
and foreign affairs ofcials in India were not involved in the initial
decision to prepare the nuclear device, nor in the nal decision to test it.
The military services were not asked how nuclear weapons would affect
their war plans and military doctrines; the Defense Minister was report-
edly informed of, but not consulted about, the nal test decision only ten
days before the May 18 explosion; the Foreign Minister was merely given
26. See Shyam Bhatia, Indias Nuclear Bomb (Ghaziabad: Vikas Publishing House,
1979), pp. 120122. The director of the PNE study later wrote that getting the Prime
Minister to agree to this venture must have required great persuasion, as Shastriji was
opposed to the idea of atomic explosions of any kind. Raja Ramanna, Years of
Pilgrimage: An Autobiography (New Delhi: Viking, 1991), p. 74.
27. See Kapur, Indias Nuclear Option, p. 195; Mitchell Reiss, Without the Bomb: The
Politics of Nuclear Nonproliferation (New York: Columbia University Press, 1988), p. 221
and p. 325, n. 42; and Ramanna, Years of Pilgrimage, p. 75.
28. Ramanna, Years of Pilgrimage, p. 89.
29. See Perkovich, Indias Nuclear Bomb; and Ramanna, Years of Pilgrimage, p. 89.
30 the coming crisis
a forty-eight-hour notice of the detonation.
30
This pattern suggests that
security arguments were of secondary importance, and at a minimum,
were not thoroughly analyzed or debated before the nuclear test.
Second, the subsequent absence of a systematic program for either
nuclear weapons or PNE development and testing, and New Delhis lack
of preparedness for Canadas immediate termination of nuclear assis-
tance, suggest that the decision was taken quickly, even in haste, and thus
may have focused more on immediate political concerns rather than on
longer-term security or energy interests.
Third, domestic support for the Gandhi government had fallen to an
all-time low in late 1973 and early 1974 due to a prolonged and severe
domestic recession, the eruption of large-scale riots in a number of regions,
and the lingering effects of the splintering of the ruling Congress Party.
From a domestic politics perspective, it would be highly surprising for a
politician with such problems to resist what she knew was a major
opportunity to increase her standing in public opinion polls and to defuse
an issue about which she had been criticized by her domestic opponents.
31
Indeed, the domestic consequences of the test were very rewarding: the
nuclear detonation occurred during the governments unprecedented
crackdown on striking railroad workers and contributed to a major, albeit
short-lived, increase in support for the Gandhi government.
32
30. See Neil H.A. Joeck, Nuclear Proliferation and National Security in India and
Pakistan, (Ph.D. dissertation, University of California, Los Angeles, 1986), p. 229; and
Kapur, Indias Nuclear Option, p. 198.
31. Although Gandhi denied, in a later interview, that domestic concerns inuenced
her 1974 decision, she did acknowledge that the nuclear test would have been useful
for elections. See Rodney W. Jones, India, in Jozef Goldblat, ed., Non-Proliferation:
The Why and the Wherefore (London: Taylor and Francis, 1985), p. 114.
32. For example, Indian public opinion polls taken in June 1974 reported that a full
91 percent of the literate adult population knew about the explosion and 90 percent
of those individuals answered in the afrmative when asked if they were personally
proud of this achievement. The overall result was that public support for Gandhi
increased by one-third in the month after the nuclear test, according to the Indian
Institute of Public Opinion, leading the Institute to conclude that both she and the
Congress Party have been restored to the nations condence. The Institutes analysis
was that the increase was the result of both the demonstration of Indias atomic
capability and the decisive action on the Railway strike, though the data outlined
above suggest that more emphasis should be placed on the weapons test. See The
Prime Ministers Popularity: June 1974, and Indian Public Opinion and the Railway
Strike, in Monthly Public Opinion Surveys (Indian Institute of Public Opinion), Vol. 19,
No. 8 (May 1974), pp. 56 and pp. 711; and Public Opinion on Indias Nuclear
Device, Monthly Public Opinion Surveys, Vol. 19, No. 9 (June 1974), Blue Supplement,
pp. IIIIV.
chapter 2 31
Similarly, the Indian governments decision to test ve nuclear weap-
ons in May 1998 should be seen in the light of domestic politics in New
Delhi. For more than twenty years after the 1974 PNE explosion, Indian
politicians had refrained from testing nuclear weapons, despite wide-
spread public support for the Indian nuclear option and weaponiza-
tion through further tests.
33
Throughout this period, many government
ofcials and Indian defense analysts argued that Indias national security
would indeed be harmed by nuclear weapons tests, since that would
simply compel Pakistan to conduct nuclear tests. How does one therefore
explain the momentous shift of policy in 1998?
The best explanation focuses on important changes in domestic poli-
tics rather than developments in international relations in South Asia in
the late 1980s. Although Indian political authorities often cited the need
to deter the Chinese nuclear threat as the main purpose of the May 1998
tests, the degree to which Sino-Indian relations had improved during the
1990s raises doubts about this rationale.
34
Indeed, China and India signed
a bilateral agreement in 1993 to maintain peaceful relations in disputed
border areas, implemented a set of condence-building measures, and
negotiated a set of agreements in 1995 and 1996 for mutual reductions of
military units in those regions.
35
There were certainly continuing tensions
in Sino-Indian relations, but no major deterioration in the military balance
or the political relationship had occurred prior to the May 1998 nuclear
tests.
What had occurred just prior to the tests was the March 1998 election,
which brought into ofce an eighteen-party coalition government led by
the Hindu nationalist party, the Bharatiya Janata Party (BJP). This gov-
ernment was highly unstable, however, and a number of coalition partner
parties threatened to leave the coalition soon after Prime Minister Atal
Vajpayee took ofce. The nuclear tests were tremendously popular
throughout India: 86 percent of those polled in May 1998 supported the
test and 44 percent stated that they were now more likely to vote for the
33. See David Cortright and Amitabh Matoo, India and the Bomb: Public Opinion and
Nuclear Options (Notre Dame, Ind.: University of Notre Dame Press, 1996).
34. See, for example, the comments of Defense Minister George Fernandes in China
and Pakistan Present a Collaborated Threat, The Times of India, April 12, 1998
(www.timesondia.com/120498/12edit4.htm).
35. See the discussion in Hua Han, Sino-Indian Relations and Nuclear Arms Con-
trol, in Eric Arnett, ed., Nuclear Weapons and Arms Control in South Asia After the Test
Ban, SIPRI Research Report No. 14, Stockholm International Peace Research Institute
(Oxford: Oxford University Press, 1998), pp. 3552; and Gaurav Kampani, From
Existential Deterrence to Minimum Deterrence, Nonproliferation Review, Vol. 6, No. 1
(Fall 1998), pp. 1224.
32 the coming crisis
BJP in any forthcoming election.
36
The resulting shift in public opinion
silenced the BJPs internal critics, at least temporarily, and may have
broadened the base of support for the BJP over the longer term, by adding
the appeal of scientic progress or nuclear nationalism to its less popu-
lar Hindu nationalist platform.
37
This domestic politics explanation is predicated on the existence of a
widespread belief among Indian voters that the states ability to build a
nuclear weapon is a sign of its scientic prowess. Yet, as noted before, it
is estimated that some thirty countries today could quickly develop
nuclear weapons should they choose to do so. Ironically, the very success
of the NPT has thus produced one counterproductive result: New Delhi
ofcials can proudly proclaim that Indias scientists have made India one
of only eight nuclear weapons states, but only because the NPT has
helped reduce the incentives to go nuclear in many other states that
have more advanced technological capabilities.
development and denuclearization: south africa revisited
The domestic politics model predicts that reversals of weapons decisions
occur not when external threats are diminished, but rather when there
are major internal political changes. There are a number of reasons why
internal changes could produce restraint: a new government has an
opportunity to change course more easily because it can blame the failed
policies of the previous regime; actors with parochial interests in favor of
weapons programs may lose internal struggles to newly empowered
actors with other interests; and the outgoing government may fear that
the incoming government would not be a reliable custodian of nuclear
weapons. Each of these domestic pathways to restraint can be relatively
independent of changes in international security threats.
Such forces can be seen when one reexamines the history of South
Africas weapons program with a focus on domestic political interests
rather than national security. For example, President de Klerks public
explanation for the program stressed that it was caused by the need to
deter a Soviet expansionist threat to Southern Africa, especially after
Cuban military forces intervened in Angola in October 1975. Yet the
preliminary research needed to develop nuclear devices was started in-
side South Africas Atomic Energy Board in 1971, on the independent
authority of the Minister of Mines. A non-nuclear scale model of a gun-
36. The polling data are reported in Solid Support, India Today, May 25, 1998.
37. A thorough analysis of the domestic political factors inuencing the Indian
decision appears in Perkovich, Indias Nuclear Bomb. Also see Kalpana Sharma, The
Hindu Bomb, Bulletin of Atomic Scientists, Vol 54, No. 4 (July/August 1998), pp. 3033.
chapter 2 33
type explosive device was secretly tested in May 1974, and later in 1974,
after the results of this test were known, Prime Minister John Voster
approved plans to construct a small number of explosive devices and to
build a secret testing site in the Kalahari desert.
38
Such evidence strongly
supports the claims of South African scientists that the nuclear program
was originally designed to produce PNEs, and that it was championed
within the government by the South African nuclear power and mining
industries to enhance their standing in international scientic circles and
to be utilized in mining situations.
39
This explanation for the origin of the nuclear program helps to
explain South African nuclear doctrine, which otherwise appears so
strange, as a post hoc development used to exploit devices that were
originally developed for other purposes. (Testing a nuclear device in the
event of a Soviet invasion might reduce the likelihood of U.S. intervention
and would raise great risks that the Soviet Union would use nuclear
weapons.) Senior ofcials in the program have stated, for example, that
the military was not consulted about the bomb design and that opera-
tional considerations, such as the size and weight of the devices, were
not taken into account.
40
As a result, the rst South African nuclear device
was actually too large to be deliverable by an aircraft and had to be
redesigned because it did not meet the safety and reliability standards set
by Armscor, the engineering organization run by the South African mili-
tary, which took over the nuclear program in 1978.
41
The timing and details of actions concerning South Africas decision
to dismantle and destroy its bomb stockpile also suggest that domestic
political considerations were critical. In September 1989, before the Cold
War was unambiguously over (the Berlin Wall fell in November 1989), de
Klerk was elected president and immediately requested a high-level
report on the possibility of dismantling the six nuclear devices. Ofcials
in South Africa considered de Klerks request as a sign that he had
already decided to abandon the weapons program. Although possible
38. See the chronology in Reiss, Bridled Ambition, p. 8 and p. 27; and Waldo Stumpf,
South Africas Nuclear Weapons Program: From Deterrence to Dismantlement, Arms
Control Today, Vol. 25, No. 10 (December 1995/January 1996), p. 4. Also see David
Fischer, South Africa, in Reiss and Litwak, Nuclear Proliferation After the Cold War,
p. 208; and David Albright, South Africas Secret Nuclear Weapons, ISIS Report
(Washington, D.C.: Institute for Science and International Security, May 1994), pp. 68.
39. See Mark Hibbs, South Africas Secret Nuclear Program: From a PNE to a
Deterrent, Nuclear Fuel (May 10, 1993), pp. 36; and Stumpf, South Africas Nuclear
Weapons Program, p. 4.
40. See Reiss, Bridled Ambitions, p. 12.
41. Albright, South Africas Secret Nuclear Weapons, p. 10.
34 the coming crisis
concerns about who would inherit nuclear weapons are rarely discussed
in the public rationales for the dismantlement decision, the de Klerk
governments actions spoke more loudly than its words: the weapons
components were dismantled before IAEA inspections could be held to
verify the activities, and all the nuclear programs plans, history of deci-
sions, and approval and design documents were burned prior to the
public announcement of the programs existence. This highly unusual
step strongly suggests that fear of African National Congress control of
nuclear weapons (and perhaps also concern about possible seizure by
white extremists) was critical in the decision.
42
Domestic politics can also be seen as playing a critical role in other
cases of nuclear restraint. In Argentina and Brazil, for example, the key
change explaining the shift from nuclear competition to cooperative re-
straint in the 1980s could not have been a major reduction of security
threats, since there was no such reduction. Indeed, a traditional realist
view would predict that the experience of the 1982 Falklands Warin
which Argentina was defeated by a nuclear power, Great Britainwould
have strongly encouraged Argentinas nuclear ambitions. Instead, the
important change was the emergence of liberalizing domestic regimes in
both states, governments supported by coalitions of actorssuch as
banks, export-oriented rms, and state monetary agenciesthat value
unimpeded access to international markets and oppose economically
unproductive defense and energy enterprises. Nuclear programs that
were run as efdoms and served the interests of the atomic industry
bureaucrats and the military were therefore abandoned by new civilian
regimes with strong support by liberalizing coalitions.
43
policy implications of the domestic politics model
For U.S. nonproliferation policy, the domestic politics approach both
cautions modest expectations about U.S. inuence and calls for a broader
set of diplomatic efforts. Modest expectations are in order, since the
domestic factors that inuence decisions are largely outside the control
of U.S. policy. Nevertheless, a more diverse set of tools could be useful
42. A rare public hint that concerns about domestic stability played a role in the
decision is the acknowledgment by the head of the Atomic Energy Corporation that
the government discussed issuing an immediate announcement revealing the exist-
ence of the weapons and thus permitting the IAEA to dismantle them. They rejected
this plan, however, because the state of the countrys internal political transformation
was not considered conducive to such an announcement at the time. See Stumpf,
South Africas Nuclear Weapons Program, p. 7.
43. The best analysis is Etel Solingen, The Political Economy of Nuclear Restraint,
International Security, Vol. 19, No. 2 (Fall 1994), pp. 126169.
chapter 2 35
to help create and empower domestic coalitions that oppose the devel-
opment or maintenance of nuclear arsenals.
A variety of activities could be included in such a nonproliferation
strategy focused on inuencing domestic debates. International nancial
institutions are already demanding that cuts in military expenditures be
included in conditionality packages for aid recipients. More direct condi-
tionality linkages to nuclear programssuch as deducting the estimated
budget of any suspect research and development program from the
International Monetary Fund or U.S. loans to a countrycould heighten
domestic opposition to such programs.
44
Providing technical information
and intellectual ammunition for domestic actorsby encouraging more
accurate estimates of the economic and environmental costs of nuclear
weapons programs and highlighting the risks of nuclear accidentscould
bring new members into antiproliferation coalitions.
45
In addition, efforts
to encourage strict civilian control of the military, through educational
and organizational reforms, could be productive, especially in states
where the military has the capability to create secret nuclear programs to
serve its parochial interests (like Brazil in the 1980s). Finally, U.S. attempts
to provide alternative sources of employment and prestige to domestic
actors who might otherwise nd weapons programs attractive could
decrease nuclear incentives. To the degree that professional military or-
ganizations support nuclear proliferation, encouraging their involvement
in other military activities (such as Pakistani participation in peacekeep-
ing operations or the Argentine Navys role in the Persian Gulf) could
decrease such support. Where the key actors are laboratory ofcials and
scientists, assistance in non-nuclear research and development programs
(as in the U.S.-Russian lab-to-lab program) could decrease personal and
organizational incentives for weapons research.
A different perspective on the role of the NPT also emerges from the
domestic politics model. The NPT regime is not just a device to increase
states condence about the limits of their potential adversaries nuclear
programs; it is also a tool that can help empower domestic actors who
are opposed to nuclear weapons development. The NPT negotiations
44. Etel Solingen, The Domestic Sources of Nuclear Postures, Institute of Global Conict
and Cooperation, Policy Paper No. 8 (October 1994), p. 11.
45. On these costs and risks, see Kathleen C. Bailey, ed., Weapons of Mass Destruction:
Costs Versus Benets (New Delhi: Manohar Publishers, 1994); Stephen I. Schwartz,
Four Trillion and Counting, Bulletin of the Atomic Scientists, Vol. 51, No. 6 (Novem-
ber/December 1995); Bruce G. Blair, The Logic of Accidental Nuclear War (Washington,
D.C.: The Brookings Institution, 1993); and Scott D. Sagan, The Limits of Safety: Organi-
zations, Accidents, and Nuclear Weapons (Princeton, N.J.: Princeton University Press,
1993).
36 the coming crisis
and review conferences create a well-placed elite in the foreign and de-
fense ministries with considerable bureaucratic and personal interests in
maintaining the regime. The IAEA creates monitoring capabilities and
enforcement incentives against unregulated activities within a states own
nuclear power organizations. The network of nongovernmental organi-
zations built around the treaty supports similar antiproliferation pressure
groups in each state.
According to this model, progress in arms control agreements among
the ve nuclear powers recognized by the NPT regime is important for
nonproliferation because it can provide political cover at home for actors
in potential and emerging nuclear powers to practice restraint. For exam-
ple, the U.S. commitment under Article VI of the NPT to work for the
eventual elimination of nuclear weapons is important because of the
impact that the behavior of the United States and other nuclear powers
can have on the domestic debates in non-nuclear states. Whether or not
the United States originally signed Article VI merely to placate domestic
opinion in non-nuclear states is not important; what is important is that
the loss of this pacifying tool could inuence outcomes in potential
proliferators. In future debates inside such states, the arguments of anti-
nuclear actorsthat nuclear weapons programs do not serve the interests
of their statescan be more easily countered by pro-bomb actors when-
ever they can point to specic actions of the nuclear powers that highlight
these states continued reliance on nuclear deterrence such as the main-
tenance of nuclear rst-use doctrines.
The Norms Model: Nuclear Symbols and State Identity
A third model of nuclear weapons proliferation focuses on norms con-
cerning weapons acquisition. In this model, nuclear decisions serve im-
portant symbolic functions, both shaping and reecting a states identity.
State behavior is determined not by leaders cold calculations about the
national security interests or their parochial bureaucratic interests, but
rather by deeper norms and shared beliefs about what actions are legiti-
mate and appropriate in international relations.
Given the importance of the subject, and the large normative litera-
ture in ethics and law concerning the use of nuclear weapons, it is
surprising that so little attention has been paid to nuclear symbolism
and the development of international norms concerning the acquisition
of nuclear weapons.
46
Sociologists and political scientists have studied
46. On nuclear ethics, see Joseph S. Nye, Jr., Nuclear Ethics (New York: Free Press,
1986); and Steven P. Lee, Morality, Prudence, and Nuclear Weapons (New York: Cam-
chapter 2 37
the emergence and inuence of international norms in other substantive
areas, however, and their insights can lead to a valuable alternative
perspective on proliferation. Within sociology, the new institutionalism
literature suggests that modern organizations and institutions often come
to resemble each other (what is called institutional isomorphism) not
because of competitive selection or rational learning but because institu-
tions mimic each other.
47
These scholars emphasize the importance of
roles, routines, and rituals: individuals and organizations may well have
interests, but such interests are shaped by the social roles actors are
asked to play, are pursued according to habits and routines as much as
through reasoned decisions, and are embedded in a social environment
that promotes certain structures and behaviors as rational and legitimate
and denigrates others as irrational and primitive.
From this sociological perspective, military organizations and their
weapons can be envisioned as serving functions similar to those of ags,
airlines, and Olympic teams: they are part of what modern states believe
they have to possess to be legitimate, modern states. Air Malawi, Royal
Nepal Airlines, and Air Myanmar were not created because they are
cost-effective ways to travel nor because domestic pressure groups
pushed for their development, but because government leaders believed
that a national airline is something that modern states must have to have
to be modern states. Very small and poor states, without a signicant
number of scientists, nevertheless have ofcial government-sponsored
science boards. From a new institutionalist perspective, such similarities
are not the result of functional logic (actions designed to serve either
international or domestic goals); they are the product of shared beliefs
about what is legitimate and modern behavior.
48
bridge University Press, 1993). For a recent analysis of legal restraints on the use of
nuclear weapons, see Nicholas Rostow, The World Health Organization, the Interna-
tional Court of Justice, and Nuclear Weapons, Yale Journal of International Law, Vol. 20,
No. 1 (Winter 1995), pp. 151185. For a rare analysis of the symbolism of nuclear
weapons, see Robert Jervis, The Symbolic Nature of Nuclear Politics, in Jervis, The
Meaning of the Nuclear Revolution (Ithaca, N.Y.: Cornell University Press, 1989), pp.
174225.
47. Among the most important sources are the essays collected in Walter W. Powell
and Paul J. DiMaggio, eds., The New Institutionalism in Organizational Analysis (Chicago:
University of Chicago Press, 1991); and John W. Meyer and W. Richard Scott, Organ-
izational Environments: Ritual and Rationality, 2nd ed. (Newbury Park, Calif.: Sage
Publications, 1992).
48. The most inuential work developing new institutional approaches for interna-
tional security issues is Peter J. Katzenstein, ed., The Culture of National Security: Norms
and Identity in World Politics (New York: Columbia University Press, 1996). Also see
38 the coming crisis
Within political science, a related literature has evolved concerning
the development and spread of norms within international regimes. Al-
though this norms perspective has rarely been applied to the proliferation
problem, scholars have studied such important phenomena as the global
spread of anticolonialism, the abolition of the African slave trade, the
near-total elimination of piracy at sea, and constraints against the use of
chemical weapons.
49
A diverse set of ideas emerging in this eld is
producing a valuable debate about the role of global norms, but not a
well-developed theory about their causal inuence. Still, as one would
expect of political scientists, coercion and power are seen to play a more
important role in spreading norms than in the sociologists literature.
Normative pressures may begin with the actions of entrepreneurial non-
state actors, but their beliefs only have signicant inuence once power-
ful state actors join the cause. Religious and liberal opposition to slavery,
for example, was clearly important in fueling U.S. and British leaders
preferences in the nineteenth century, but such views would not easily
have become an international norm without the bayonets of the Army of
the Potomac at Gettysburg or the ships of the British Navy patrolling the
Marc C. Suchman and Dana P. Eyre, Military Procurement as Rational Myth: Notes
on the Social Construction of Weapons Proliferation, Sociological Forum, Vol. 7, No. 1
(March 1992), pp. 137161; Martha Finnemore, International Organizations as Teach-
ers of Norms: UNESCO and Science Policy, International Organization, Vol. 47, No. 4
(Autumn 1993), pp. 565598; Francisco O. Ramirez and John Boli, Global Patterns of
Educational Institutionalization, in George M. Thomas, John W. Meyer, Francisco O.
Ramirez, and John Boli, eds., Institutional Structure: Constituting State, Society, and the
Individual (Newbury Park, Calif.: Sage Publications, 1987), pp. 150172. For an excellent
survey and critique, see Martha Finnemore, Norms, Culture, and World Politics:
Insights from Sociologys Institutionalism, International Organization, Vol. 50, No. 2
(Spring 1996), pp. 325348.
49. For rare applications of the norms perspective to proliferation, see Harald Mller,
The Internalization of Principles, Norms, and Rules by Governments: The Case of
Security Regimes, in Volker Rittberger, ed., Regime Theory and International Relations
(Oxford: Clarendon Press, 1995), pp. 361390; and Mller, Maintaining Nonnuclear
Weapon Status, in Regina Cowen Karp, ed., Security With Nuclear Weapons? (New
York: Oxford University Press, 1991), pp. 301339. Also see Robert H. Jackson, The
Weight of Ideas in Decolonization: Normative Change in International Relations, in
Judith Goldstein and Robert O. Keohane, eds., Ideas and Foreign Policy (Ithaca, N.Y.:
Cornell University Press, 1993), pp. 111138; Neta C. Crawford, Decolonization as an
International Norm, in Laura W. Reed and Carl Kaysen, eds., Emerging Norms of
Justied Intervention (Cambridge, Mass.: American Academy of Arts and Sciences,
1993), pp. 3761; Ethan A. Nadelmann, Global Prohibition Regimes: The Evolution of
Norms in International Society, International Organization, Vol. 44, No. 4 (Autumn
1990), pp. 479526; and Richard Price, A Genealogy of the Chemical Weapons Taboo,
International Organization, Vol. 49, No. 1 (Winter 1995), pp. 73104.
chapter 2 39
high seas between Africa and Brazil.
50
Similarly, normative beliefs about
chemical weapons were important in creating legal restrictions against
their use in war; yet, the norm was signicantly reinforced at critical
moments by the fear of retaliation in kind and by the availability of other
weapons that were believed by military leaders to be more effective on
the battleeld.
51
The sociologists arguments highlight the possibility that nuclear
weapons programs serve symbolic functions reecting leaders percep-
tions of appropriate and modern behavior. The political science literature
reminds us, however, that such symbols are often contested and that the
resulting norms are spread by power and coercion, and not by the
strength of ideas alone. Both insights usefully illuminate the nuclear
proliferation phenomenon. Existing norms concerning the nonacquisition
of nuclear weapons (such as those embedded in the NPT) could not have
been created without the strong support of the most powerful states in
the international system, who believed that the norms served their nar-
row political interests. Yet, once that effort was successful, these norms
shaped states identities and expectations and even powerful actors be-
came constrained by the norms they had created.
52
The history of nuclear
proliferation is particularly interesting in this regard because a major
discontinuitya shift in nuclear normshas emerged as the result of the
NPT regime.
Although many individual case studies of nuclear weapons decisions
mention the belief that nuclear acquisition will enhance the international
prestige of the state, such prestige has been viewed simply as a reason-
able, though diffuse, means used to enhance the states international
inuence and security. What is missing from these analyses is an under-
standing of why and how actions are granted symbolic meaning: why
are some nuclear weapons acts considered prestigious, while others pro-
duce opprobrium, and how do such beliefs change over time? Why, for
example, was nuclear testing deemed prestigious and legitimate in the
50. Ethan Nadelman, who stresses this point about power, nonetheless adds that
even among the laggards, indeed especially among the laggards, the consciousness
of being perceived as primitive and deviant surely weighed heavily in the decisions
of local rulers to do away with slavery. Nadelman, Global Prohibition Regimes,
p. 497.
51. See Price, A Genealogy of the Chemical Weapons Taboo; and Jeffrey Legro,
Cooperation Under Fire: Anglo-German Restraint During World War II (Ithaca, N.Y.: Cor-
nell University Press, 1995), pp. 144216.
52. For an excellent analysis of how such a process can work in other contexts, see
Michael Byers, Custom, Power, and the Power of Rules, Michigan Journal of Interna-
tional Law, Vol. 17, No. 1 (Fall 1995), pp. 109180.
40 the coming crisis
1960s, but is today considered illegitimate and irresponsible? The answer
is that the NPT regime appears to have shifted the norm concerning what
acts grant prestige and legitimacy from the 1960s notion of joining the
nuclear club to the 1990s concept of joining the club of nations adhering
to the international nuclear agreements. Moreover, the salience of the
norms that were made explicit in the NPT treaty has shifted over time.
These arguments are perhaps best supported by contrasting two cases
Frances decision to build and test nuclear weapons and the Ukraines
decision to give up its nuclear arsenalin which perceptions of legiti-
macy and prestige appear to have had a major inuence, albeit with very
different outcomes.
proliferation revisited: french grandeur and weapons policy
According to realist theory, the French decision to develop nuclear weap-
ons has a very simple explanation: in the 1950s, the Soviet Union was a
grave military threat to French national security, and the best alternative
to building an independent arsenalreliance on the U.S. nuclear guaran-
tee to NATOwas ruled out after the Soviet development of a secure
second-strike capability reduced the credibility of any U.S. nuclear rst-
use threats. According to this explanation, the need for a French arsenal
was driven home by the 1956 Suez Crisis, when Paris was forced to
withdraw its military intervention forces from Egypt after a nuclear threat
from Russia and under U.S. economic pressure. The Suez humiliation of
1956 was decisive, writes David Yost. It was felt that a nuclear weapons
capability would reduce Frances dependence on the U.S. and her vul-
nerability to Soviet blackmail.
53
The central realist argument for French
nuclear weapons was clearly expressed in the rhetorical question Presi-
dent Charles de Gaulle posed to President Dwight Eisenhower in 1959:
Will they [future U.S. presidents] take the risk of devastating American
cities so that Berlin, Brussels and Paris might remain free?
54
This explanation of French nuclear policy does not stand up very well
against either evidence or logic. Indeed, the two most critical decisions
initiating the weapons programPrime Minister Pierre Mendes-Frances
December 1954 decision to start a secret nuclear weapons research pro-
gram inside the Commissariat lnergie atomique (CEA) and the May
53. David S. Yost, Frances Deterrent Posture and Security in Europe, Part I: Capa-
bilities and Doctrine, Adelphi Paper No. 194 (London: International Institute for
Strategic Studies [IISS], Winter 1984/85), p. 4. Also see Kohl, French Nuclear Diplomacy,
p. 36.
54. Jean Lacouture, De Gaulle: The Ruler 19451970 (New York: W.W. Norton, 1993),
p. 421, as quoted in Thayer, The Causes of Nuclear Proliferation, p. 489.
chapter 2 41
1955 authorization by the Ministry of Defense for funds to be transferred
to the CEA for the development of a prototype weaponpredated the
1956 Suez Crisis.
55
In addition, it is by no means clear why French leaders
would think that the traumatic Suez experience could have been avoided
if there had been an independent French nuclear arsenal, since Great
Britain had also been forced to withdraw from the intervention in Egypt
under U.S. and Soviet pressure, despite its possession of nuclear weap-
ons.
56
A simple exercise in comparative logic also raises doubts about the
security model. If the critical cause of proliferation in France was the lack
of credibility of U.S. nuclear guarantees given the growing Soviet threat
in the mid-1950s, why then did not other nuclear-capable states in
Europe, faced with similar security threats at the time, also develop
nuclear weapons?
57
Of all the nuclear-capable states in Europe that were
both threatened by Soviet military power and had reasons to doubt the
credibility of the U.S. rst-use pledge, France was the only state to
develop a weapon; West Germany, the Netherlands, Italy, Switzerland,
Belgium, Norway, and Sweden all restrained their nuclear programs. This
presents a puzzle for the security model, since the Soviet Unions con-
ventional and nuclear threat to most of these states security was at least
as great as its threat to France; the U.S. nuclear guarantee should not have
been considered more credible by those states that had been U.S. enemies
or neutrals in World War II, than by France, a U.S. ally of long standing,
and one which the United States had strongly aided once it entered the
war in 1941.
A stronger explanation for the French decision to build nuclear weap-
ons emerges when one focuses on French leaders perceptions of the
bombs symbolic signicance. The belief that nuclear power and nuclear
weapons were deeply linked to a states position in the international
system was present as early as 1951, when the rst French ve-year plan
was put forward with its stated purpose being to ensure that in 10 years
time France will still be an important country.
58
France emerged from
World War II in an unusual position: it was a liberated victor whose
military capabilities and international standing were not at all compara-
ble to the power and status it had before the war. It should therefore not
55. See Bertrand Goldschmidt, The Atomic Complex (La Grange Park, Ill.: American
Nuclear Society, 1982), p. 131; and Scheinman, Atomic Energy Policy in France Under the
Fourth Republic, pp. 120122.
56. Scheinman, Atomic Energy Policy in France Under the Fourth Republic, pp. 171173.
57. The British acquisition of nuclear weapons in 1952 predated the Soviet develop-
ment of a secure second-strike capability.
58. The document is quoted in Goldschmidt, The Atomic Complex, p. 126.
42 the coming crisis
be surprising that the governments of both the Fourth and the Fifth
Republics vigorously explored alternative means to return France to its
historical great power status.
59
After the war, the initial French effort to
restore its tarnished prestige focused on the ght to hold onto an overseas
empire, yet as Michel Martin has nicely put it, as the curtain was drawn
over colonial domination, it became clear that the countrys grandeur had
to be nourished from other sources.
60
After 1958, the Algerian crisis contributed greatly to Charles de
Gaulles obsession with nuclear weapons as the source of French gran-
deur and independence. In contrast, de Gaulle appeared less concerned
about whether French nuclear forces could provide adequate deterrence
against the Soviet military threat. For example, during both the Berlin
crisis of 1958 (before the 1960 French nuclear weapons test) and the 1962
Cuban crisis (after the test, but before French nuclear forces were opera-
tional), de Gaulle expressed great condence that the Soviets would not
risk an attack on NATO Europe.
61
Wilfred Kohl also reports on a revealing
incident in which a French military strategist sent de Gaulle a copy of a
book on French nuclear doctrine and de Gaulle replied, thanking the
man for his interesting analysis of strategic questions, but stressing that
for him the central and clearly the only important issue was: Will France
remain France?
62
For de Gaulle, the atomic bomb was a dramatic sym-
bol of French independence and was thus needed for France to continue
to be seen, by itself and others, as a great power. He conded to Dwight
Eisenhower in 1959:
A France without world responsibility would be unworthy of herself, espe-
cially in the eyes of Frenchmen. It is for this reason that she disapproves of
NATO, which denies her a share in decision-making and which is conned
to Europe. It is for this reason too that she intends to provide herself with an
atomic armament. Only in this way can our defense and foreign policy be
independent, which we prize above everything else.
63
59. For detailed analyses of the French nuclear weapons decision that focus on
political prestige as the central source of policy, see Scheinman, Atomic Energy Policy
in France Under the Fourth Republic; and Kohl, French Nuclear Diplomacy. Also see Bundy,
Danger and Survival, pp. 472487, 499503.
60. Michel L. Martin, Warriors to Managers: The French Military Establishment Since 1945
(Chapel Hill, N.C.: University of North Carolina Press, 1981), p. 21.
61. See Philip H. Gordon, Charles de Gaulle and the Nuclear Revolution, Security
Studies, Vol. 5, No. 1 (Autumn 1995), pp. 129130.
62. Kohl, French Nuclear Diplomacy, p. 150, quoted in Bundy, Danger and Survival,
p. 502.
63. Charles de Gaulle, Memoirs of Hope: Renewal and Endeavor (New York: Simon and
chapter 2 43
When the French nuclear weapons arsenal is viewed as primarily
serving symbolic functions, a number of puzzling aspects of the history
of French atomic policy become more understandable. The repeated
Gaullist declarations that French nuclear weapons should have world-
wide capabilities and must be aimed in all directions (tous les azimuts)
are not seen as the product of security threats from all directions, but
rather as the only policy that is logically consistent with global grandeur
and independence. Similarly, the French strategic doctrine of propor-
tional deterrence against the Soviet Union during the Cold Warthreat-
ening more limited destruction in a retaliatory strike than did the United
States under its targeting doctrineis not a product of Frances geo-
graphical position or limited economic resources, but rather an indication
that deterrence of the Soviet Union was a justication, and never the
primary purpose of its arsenal. Finally, the profound French reluctance to
stop nuclear testing in the mid-1990s is seen as being produced not only
by the stated concerns about weapons modernization and warhead safety,
but also because weapons tests were perceived by Parisian leaders as
potent symbols of French identity and status as a great power.
restraint revisited: the npt and the ukraine case
Stark contrasts exist between French nuclear decisions in the 1950s and
Ukrainian nuclear decisions in the mid-1990s. When the Soviet Union
collapsed in 1991, an independent Ukraine was born nuclear with more
than 4,000 nuclear weapons on or under its soil. In November 1994, the
Radathe Ukrainian parliament in Kievvoted overwhelmingly to join
the NPT as a non-nuclear state, and all weapons were removed from
Ukrainian territory by June 1996.
This decision to give up a nuclear arsenal is puzzling from the realist
perspective: a number of prominent realist scholars, after all, maintained
that given the history of Russian expansionist behavior and continuing
tensions over Crimea and the treatment of Russian minorities, Ukraines
independence was seriously threatened. They further argued that nuclear
weapons were the only rational solution to this security threat.
64
The
disarmament decision is also puzzling from a traditional domestic politics
perspective. Despite the tragic consequences of the Chernobyl accident,
Schuster, 1971), p. 209, quoted in Yost, Frances Deterrent Posture and Security in Europe,
pp. 1314.
64. See John J. Mearsheimer, The Case for a Ukrainian Nuclear Deterrent, Foreign
Affairs, Vol. 72, No. 3 (Summer 1993), pp. 5066; and Barry R. Posen, The Security
Dilemma and Ethnic Conict, Survival, Vol. 35, No. 1 (Spring 1993), pp. 4445.
44 the coming crisis
public opinion polls in Ukraine showed rapidly growing support for
keeping nuclear weapons in 1992 and 1993: polls showed support for an
independent arsenal increasing from 18 percent in May 1992 to 36 percent
in March 1993, to as much as 45 percent in the summer of 1993.
65
In
addition, well-known retired military ofcers, such as Rada member
General Volodomyr Tolubko, vigorously lobbied to maintain an arsenal
and senior political leaders, most importantly Prime Minister (then Presi-
dent) Leonid Kuchma, came from the Soviet missile-building industry
and would not therefore be expected to take an antinuclear position.
66
An understanding of Ukraines decision to eliminate its nuclear arse-
nal requires that more attention be focused on the role that emerging NPT
nonproliferation norms played in four critical ways. First, Ukrainian
politicians initially adopted antinuclear positions as a way of buttressing
Kievs claims to national sovereignty. In one of its rst efforts to assert a
foreign policy independent of Moscow, Ukraine tried to accede to the
NPT as a non-nuclear state in early 1990, attempting to use NPT mem-
bership to separate itself from the Soviet Union.
67
In July 1990, this policy
was underscored when the Parliament in Kiev issued its Declaration of
Sovereignty. Embedded in declarations about Ukraines right to partici-
pate as a full member in all agreements concerning international peace
and security was the proclamation that Ukraine would become a neu-
tral state that does not participate in military blocs and that adheres to
three non-nuclear principles: not to maintain, produce, or acquire nuclear
weapons. This extraordinary statement was an expedient designed to
buttress Kievs claim to independence from the Soviet Union, rather than
a blueprint laying out Ukraines long-term strategy: indeed, it was
adopted by a vote of 3554, without extensive debate, by the parliament
in which conservative communists (many of whom would later take
pro-nuclear positions) still held the majority of seats.
68
Nevertheless, the
declaration placed the onus of reneging on an international commitment
on the politicians and scholars who afterwards called for keeping an
arsenal, and it is revealing that even many of the more hawkish analysts
65. See William C. Potter, The Politics of Nuclear Renunciation: The Cases of Be-
larus, Kazakhstan, and Ukraine, Henry L. Stimson Center, Occasional Paper No. 22
(April 1995), p. 49.
66. For a detailed analysis, see Bohdan Nahaylo, The Shaping of Ukrainian Atti-
tudes Toward Nuclear Arms, RFE/RL (Radio Free Europe/Radio Liberty) Research Report,
Vol. 2, No. 8 (February 19, 1993), pp. 2145.
67. Potter, The Politics of Nuclear Renunciation, p. 19.
68. See Nahaylo, The Shaping of Ukrainian Attitudes, pp. 2122.
chapter 2 45
thereafter defensively advocated keeping the arsenal on a temporary
basis until other sources of security could be found.
69
Second, the strength
of the NPT regime created a history in which the most recent examples
of new or potential nuclear states were so-called rogue states such as
North Korea, Iran, and Iraq. This was hardly a nuclear club whose new
members would receive international prestige, and during the debate in
Kiev, numerous pro-NPT Ukrainian ofcials insisted that the renunciation
of nuclear weapons was now the best route to enhance Ukraines inter-
national standing.
70
Third, economic pressures were clearly critical to the
Ukrainian decision: the United States and NATO allies encouraged Kiev
to give up the arsenal not by convincing ofcials that nuclear weapons
could never serve as a military deterrent against Moscow, but by persuad-
ing them that not following the NPT norm would result in very negative
economic consequences.
71
The ability of the West to coordinate such
activities, and credibly to threaten collective sanctions and promised
inducements for disarmament, were signicantly heightened by the NPT
norm against the creation of new nuclear weapons states. Fourth, the
Kiev government and the Ukrainian public could more easily accept the
economic inducements offered by the United Statessuch as Nunn-
Lugar payments to help transport and destroy the weaponswith the
belief that they were enabling Ukraine to keep an international commit-
ment.
As with all counterfactuals, it is impossible to assess with certainty
whether Ukraine would have divested itself of nuclear weapons had the
NPT norms not been in existence. Still, it is valuable to try to imagine
how much more difcult a disarmament outcome would have been in
the absence of the NPT and its twenty-ve-year history. Without the NPT,
a policy of keeping a nuclear arsenal would have placed Ukraine in the
category of France and China; with the NPT, it would place Ukraine in
the company of dissenters like India and Pakistan and pariahs like Iraq
and North Korea. International threats to eliminate economic aid and
suspend political ties would be less credible, since individual states
would be more likely to defect from an agreement. Finally, without the
NPT norm, U.S. dismantlement assistance would have been seen in Kiev
as the crass purchase of Ukrainian weapons by a foreign government,
69. Potter, The Politics of Nuclear Renunciation, pp. 2123; and Nahaylo, The
Shaping of Ukrainian Attitudes.
70. See Potter, The Politics of Nuclear Renunciation, p. 44; and Garnett, Ukraines
Decision to Join the NPT, p. 12.
71. An excellent analysis, of U.S. policy appears in Garnett, Ukraines Decision to
Join the NPT, pp. 1012.
46 the coming crisis
instead of being viewed as friendly assistance to help Kiev implement an
international agreement.
policy implications of the norms model
If the norms model of proliferation is correct, the key U.S. policy chal-
lenges are to recognize that such norms can have a strong inuence on
other states nuclear weapons policies, and to adjust U.S. policies to
increase the likelihood that norms will push others toward policies that
also serve U.S. interests. Recognizing the possibility that norms can
inuence other states behavior in complex ways should not be difcult.
After all, the norms of the NPT have already inuenced U.S. nuclear
weapons policy in ways that few scholars or policymakers predicted
ahead of time: in January 1995, for example, the Clinton administration
abandoned the long-standing U.S. position that the Comprehensive Test
Ban Treaty (CTBT) must include an automatic escape clause permitting
states to withdraw from the treaty after ten years. Despite the arguments
made by Pentagon ofcials that such a clause was necessary to protect
U.S. security, the administration accepted the possibility of a permanent
CTBT because senior decision-makers became convinced that the U.S.
position was considered illegitimate by non-nuclear NPT members, due
to the Article VI commitment to eventual disarmament, and might
thereby jeopardize the effort to negotiate a permanent extension of the
NPT treaty.
72
Adjusting U.S. nuclear policies in the future to reinforce emerging
nonproliferation norms will be difcult, however, because many of the
recommended policies derived from the norms perspective directly con-
tradict recommendations derived from the other models. A focus on NPT
norms raises especially severe concerns about how the U.S. nuclear rst-
use doctrine inuences potential proliferators perceptions of the legiti-
macy or illegitimacy of nuclear weapons possession and use.
73
To the
degree that such rst-use policies create beliefs that making nuclear
threats is what great powers do, they will become desired symbols for
states that aspire to that status. The norms argument against the U.S.
nuclear rst-use doctrine, however, contradicts the policy advice derived
from the security model, which stresses the need for continued nuclear
72. Douglas Jehl, U.S. in New Pledge on Atom Test Ban, New York Times, January
31, 1995, p. 1; and Dunbar Lockwood, U.S. Drops CTB Early Out Plan; Test Mora-
torium May Be Permanent, Arms Control Today, Vol. 25, No. 2 (March 1995), p. 27.
73. On this issue, see Barry M. Blechman and Cathleen S. Fisher, Phase Out the
Bomb, Foreign Policy, No. 97 (Winter 199495), pp. 7995; and Wolfgang K.H. Panof-
sky and George Bunn, The Doctrine of the Nuclear-Weapons States and the Future
of Non-Proliferation, Arms Control Today, Vol. 24, No. 6 (July/August 1994), pp. 39.
chapter 2 47
guarantees for U.S. allies. Similarly, the norms perspective suggests that
current U.S. government efforts to maintain the threat of rst use of
nuclear weapons to deter the use of biological or chemical weapons
would damage the nuclear nonproliferation regime.
74
Leaders of non-
nuclear states are much less likely to think that their own acquisition of
nuclear weapons to deter adversaries with chemical and biological weap-
ons is illegitimate and ill-advised if the greatest conventional military
power in the world can not refrain from making such threats.
Other possible policy initiatives are less problematic. For example, if
norms concerning prestige are important, then it would be valuable for
the United States to encourage the development of other sources of
international prestige for current or potential proliferators. Thus, a policy
that made permanent UN Security Council membership for Japan or
Germany conditional upon the maintenance of non-nuclear status under
the NPT might further remove nuclear weapons possession from consid-
erations of international prestige.
Finally, the norms model leads to a more optimistic vision of the
potential future of nonproliferation. Norms are sticky: individual
and group beliefs about appropriate behavior change slowly, and
over time norms can become rules embedded in domestic institutions.
75
In the short run, therefore, norms can be a brake on nuclear chain
reactions: in contrast to more pessimistic realist predictions that prolif-
eration begets proliferation, the norms model suggests that normative
constraints can deter or at least delay proliferation. The long-term future
of the NPT regime is also viewed with more optimism, for the model
envisions the possibility of a gradual emergence of a norm against all
nuclear weapons possession. The development of such a norm may
well have been inadvertentfor quite understandable reasons, the
United States did not take its Article VI commitment to work in good
faith for complete nuclear disarmament seriously during the Cold
War. But to the degree that other states believe that such commitments
are real and legitimate, their perceptions that the United States is back-
sliding away from Article VI will inuence their behavior over time.
74. For contrasting views on this policy, see George Bunn, Expanding Nuclear
Options: Is the U.S. Negating its Non-Use Pledges? Arms Control Today, Vol. 26,
No. 4 (May/June 1996), pp. 710; and Gompert, Watman, and Wilkening, Nuclear
First Use Revisited.
75. For useful discussions, see Abram Chayes and Antonia Handler Chayes, The New
Sovereignty: Compliance with International Regulatory Agreements (Cambridge, Mass.:
Harvard University Press, 1995); and Andrew P. Cortell and James W. Davis, Jr., How
Do International Institutions Matter? The Domestic Impact of International Rules and
Norms, International Studies Quarterly, Vol. 40, No. 4 (December 1996), pp. 451478.
48 the coming crisis
This emphasis on emerging norms therefore highlights the need for the
nuclear powers to reafrm their commitments to global nuclear disarma-
ment, and suggests that it is essential that the United States and other
governments develop a public, long-term strategy for the eventual elimi-
nation of nuclear weapons.
76
The norms model cannot, of course, predict
whether such efforts will ever resolve the classic risks of nuclear disar-
mament: that states can break treaty obligations in crises, that small
arsenals produce strategic instabilities, and that adequate verication of
complete dismantlement is exceedingly difcult. But the model does
predict that there will be severe costs involved if the nuclear powers are
seen to have failed to make signicant progress toward nuclear disarma-
ment.
Causal Complexity and Policy Tradeoffs
The ideas and evidence presented in this chapter suggest that the widely
held security model explanation for nuclear proliferation decisions is
inadequate. A realist might well respond to this argument by asserting
that evidence is always ambiguous in complex historical events, and that
I underestimate foreign threats and thus provide a poor measure of the
effects of security concerns on decision-makers. Moreover, it could be
argued that the best theories are those that explain the largest number of
cases and that the largest number of positive nuclear weapons decisions
in the past (the United States, the Soviet Union, China, Israel, and perhaps
Pakistan) and the majority of the most pressing proliferation cases today
(Iraq, Libya, and possibly North Korea and Iran) appear to be best
explained by the basic security model.
I have no quarrel with the argument that the largest number of past
proliferation cases, and even many emerging ones, are best explained by
the security model. But explaining the majority of cases should not be
satisfying. The evidence presented above strongly suggests that multi-
causality, rather than measurement error, lies at the heart of the nuclear
proliferation problem. Nuclear weapons proliferation and nuclear re-
straint have occurred in the past, and can occur in the future, for more
than one reason: different historical cases are best explained by different
causal models.
76. For important efforts to rethink the elimination issue, see An Evolving U.S.
Nuclear Posture, Report of the Steering Committee of the Project on Eliminating
Weapons of Mass Destruction, Henry L. Stimson Center, Washington, D.C., December
1995; and Donald MacKenzie and Graham Spinardi, Tacit Knowledge, Weapons
Design, and the Uninvention of Nuclear Weapons, American Journal of Sociology,
Vol. 101, No. 1 (July 1995), pp. 44100.
chapter 2 49
This argument has important implications for future scholarship on
proliferation as well as for U.S. nonproliferation policy. The challenge for
scholars is not to produce increasing numbers of detailed but atheoretical
case studies of states nuclear proliferation and restraint decisions; it is to
produce theory-driven comparative studies to help determine the condi-
tions under which different causal forces produced similar outcomes.
Predicting the future based on such an understanding of the past will still
be problematic, since the conditions that produced the past proliferation
outcomes may themselves change. But future scholarship focusing on
how different governments assess the nuclear potential and intention of
neighbors, on why pro-bomb and anti-bomb domestic coalitions form
and gain inuence, and on when and how NPT norms about legitimate
behavior constrain statesmen will be extremely important.
77
For policymakers, the existence of three different reasons why states
develop nuclear weapons suggests that no single policy can ameliorate
all future proliferation problems. Fortunately, some of the policy recom-
mendations derived from the models are compatible: for example, many
of the diplomatic tools suggested by the domestic politics model, which
attempts to reduce the power of individual parochial interests in favor of
nuclear weapons, would not interfere with simultaneous efforts to ad-
dress states security concerns. Similarly, efforts to enhance the interna-
tional status of some non-nuclear states need not either undercut deter-
rence or promote pro-nuclear advocates in those countries.
Unfortunately, other important recommendations from different mod-
els are more contradictory. Most importantly, a security-oriented strategy
of maintaining a major role for U.S. nuclear guarantees to restrain prolifera-
tion among allies will eventually create strong tensions with a norms-ori-
ented strategy seeking to delegitimize nuclear weapons use and acquisition.
The nal outcome of these alternative strategies, of course, is not under
the control of the United States, as leaders of potential proliferators will
decide for themselves whether to pursue or reject nuclear weapons programs.
U.S. policy will not be without inuence, however, and intelligent
decisions will not emerge if U.S. ofcials refuse to recognize that painful
tradeoffs are appearing on the horizon. U.S. decision-makers will even-
tually have to choose between the difcult nonproliferation task of wean-
ing allies away from nuclear guarantees without producing new nuclear
states, and the equally difcult task of maintaining a norm against nu-
clear proliferation without the U.S. governments facing up to its logical
consequences.
77. See the essays in Peter R. Lavoy, Scott D. Sagan, and James J. Wirtz, eds., Planning
the Unthinkable (Ithaca, N.Y.: Cornell University Press, 2000).
50 the coming crisis
Chapter 3
Universal Deterrence or
Conceptual Collapse?
Liberal Pessimism and Utopian
Realism
Richard K. Betts
Weapons of mass destruction (WMD) tend to turn typical thinking
upside down. For example, many strategic pundits argued during the
Cold War that for nuclear weapons, offense is defense, defense is offense.
Killing people is good, killing weapons is bad.
1
More generally, the very
existence of WMD makes those who are otherwise optimistic about the
course of history cringe in horror, while it leaves some who are otherwise
pessimistic quite cheerful. The former fear that no good can come from
a growing number of bad things; the latter argue that a half-century of
peace among nuclear powers proves the reverse.
Most U.S. citizens view international relations through the prism of
the liberal tradition, which emphasizes the idea of progress from primi-
tive conict toward enlightened cooperation, a view reinforced by the
end of the Cold War. Progress should obviate the quest for instruments
of mass killing, not spur it. The spread of such instruments is a wrench
in the works of the end of history. Thus most normal people in the United
States see the diffusion of WMD as a dire threat to be contained and rolled
back. On this issue, ironically, liberals prove profoundly conservative and
elitist. They enshrine the status quo, seeking to freeze the international
hierarchy indenitely through treaties to prevent others from obtaining
the weapons the United States has held for half a century and will
continue to hold indenitely, despite lofty rhetoric about abolition.
1. Axiom quoted in John Newhouse, Cold Dawn: The Story of SALT (New York: Holt,
Rinehart and Winston, 1973), p. 176. The rationale was that offense dominance would
ensure mutual assured destruction and thus prevent either side from striking; the
development of defenses would mislead them to believe that war could be tolerable;
the ability to target only population centers would make a rst strike useless; and the
capacity to target military forces would be a temptation to strike rst. See also Robert
Jervis, Cooperation Under the Security Dilemma, World Politics, Vol. 30, No. 2
(January 1978).
The most Panglossian view, with equal irony, comes out of interna-
tional relations theorys realist tradition. For the most part, realism is
pessimistic about the international human condition. From Thucydides,
through Machiavelli, Hobbes, Carr, and Morgenthau, this school has
viewed international conict as inevitable as long as there is no world
government to prevent it. History is not progress toward world order and
the rule of international law, but a cycle of conict among states or
groups. Hopes for collective security and lions lying down with lambs
are utopian.
Those who identify with classical realism, with its emphasis on hu-
man fallibility, worry about mass destruction as much as liberals do. Out
of the realist tradition, however, comes a schismatic coterie of academics
led by Kenneth Waltz. More than classical realists, this group emphasizes
the constraining effects of the international power structure and holds
that nuclear weapons can produce the permanent peace that liberals have
always believed in and realists have always said is impossible. In the
paradoxical tradition of theories on nuclear weapons this minority might
be dubbed utopian realists. They foresee the universalization of vulner-
ability compelling the universalization of military restraint.
2
The spread
of murderous weapons will ensure that victims can destroy their attackers
in retaliation, which in turn will mean that resorting to force will not
serve anyones interests. Nuclear weapons offer a technological x to
neutralize the traditional natural dynamics of the international political
jungle. The specter of apocalypse becomes the guarantor of peace. This
benign view derives from standard Cold War deterrence theory, and faith
that national leaders are only irresponsible when stakes are not astro-
nomical.
The utopian realists are right about the statistically probable effect of
proliferation in any specic case, but the liberal pessimists are right about
the ultimate general effect. The rst exception to the utopian rule may
upset all condence in stable order and set off dangerous behavior. But
while the liberal diagnosis is a better guide to the danger, realist norms
relying on incentives based on interest rather than on law or preach-
mentare a better guide to policy. Neither school, however, has consid-
2. The principal works making this argument are Kenneth N. Waltz, The Spread of
Nuclear Weapons: More May Be Better, Adelphi Paper No. 171 (London: International
Institute for Strategic Studies [IISS], Autumn 1981); Shai Feldman, Israeli Nuclear
Deterrence: A Strategy for the 1980s (New York: Columbia University Press, 1982); John
J. Mearsheimer, The Case for a Ukrainian Nuclear Deterrent, Foreign Affairs, Vol. 72,
No. 3 (Summer 1993); Bruce Bueno de Mesquita and William H. Riker, An Assessment
of the Merits of Selective Nuclear Proliferation, Journal of Conict Resolution, Vol. 26,
No. 2 (June 1982).
52 the coming crisis
ered very carefully how much the effects of spreading chemical and
biological WMD will parallel those of nuclear weapons.
How Soon and How Big a Problem?
will the future be like the past?
Not all fear the spread of capabilities for mass destruction, but few wish
to spur it. Most believe that much greater proliferation will occur soon if
some extraordinary solution is not found. Is that expectation warranted?
Condence in predicting proliferation should be tempered by the poor
track record of past estimates.
At the beginning of the nuclear era, hubris caused most U.S. ofcials
to underestimate how quickly the Soviet Union would get its own atom
bomb. Since then, however, it has been more common to overestimate the
rate of spread. John F. Kennedys oft-cited remark about a world of
twenty-ve nuclear-armed states by the 1970s is typical. In later years it
was also common to predict the rapid spread of chemical weapons (CW),
the poor mans atomic bomb. Proliferation of chemical weapons went
into reverse after World War II. Only ve or six countries still kept them
by the early 1960s.
3
They did spread more widely than nuclear arms,
however, in the latter part of the century. Unlike nuclear weapons, which
were only ever used in one week in 1945, they are not bound up in a
taboo, since they have actually been used a number of times over the
years.
Since the 1950s, few observers have underestimated the rate at which
nuclear and chemical weapons would spread (although the prospects for
biological weapons programs have been underestimated in recent years).
Past overestimates of horizontal nuclear proliferation (the spread of
weapons to additional countries) do not mean there is no problem, but
they raise a question about whether there is any consistent trend.
4
Though slower than anticipated, nuclear spread has occurred at the rate
of one net entrant every six or seven years: as of 1998 there were eight
3. Edward M. Spiers, Chemical and Biological Weapons: A Study of Proliferation (London:
Macmillan, 1994), p. 11.
4. Even vertical proliferation of nuclear weapons (the increase in numbers within
the superpowers inventories) was less consistent than folklore had it. Rather than an
ever-upward spiral of procurement, the number of U.S. weapons, their aggregate
megatonnage, and their total destructive power (equivalent megatonnage) actually
declined signicantly over the course of the 1960s. Albert Wohlstetter, Racing For-
ward or Ambling Back? in Defending America (New York: Basic Books, 1977), pp. 135
141. Much of this reected retirement of air defense warheads, though, and deploy-
ment of multiple warheads on offensive missiles after the 1960s reversed the trend.
chapter 3 53
known nuclear weapons states fty-three years after the rst nuclear
detonation. More interesting, the rate has not been accelerating. Six of the
eight entered the club in the rst half of the post-1945 period. The latest
spate of hand-wringing that came with the Indian and Pakistani nuclear
tests of 1998 obscured the fact that these were not new entrants to the
club (India originally tested in 1974, and Pakistan had the capability at
least since 1990), and that they are the only net entrants since Israel in the
1960s (South Africa entered but then left). For the second half of the
nuclear era, the net rate of spread has been one every fteen years.
Concern about the issue has also varied signicantly. During the Cold
War few hawkish strategists got exercised about proliferation; they were
too busy worrying about Soviet capabilities and the East-West competi-
tion. Except for Chinas accession in 1964, they often pooh-poohed con-
cern with proliferation, perhaps because all the other entrants were U.S.
allies. Those who used to wring their hands the most about the prolifera-
tion of weapons of mass destruction were doves, who were viscerally
disposed to regard any such weapons as bad, whoever had them, and
who tended to focus on international legal solutions or export controls
rather than on the strategic dimension of the issue.
5
The track record of predictions suggests three questions: Why did
most analysts think that the spread of WMD would be faster and wider
than it has been? Will the reasons that the spread was slower than
anticipated persist? Does the rate matter?
There are at least ve reasons for overestimation. First, those inter-
ested in the question were those inclined to be worried about it. Second,
because good news is no news, pessimists focus on the reasons that bad
things could happen, rather than on the reasons that they might not.
Third, at least in regard to nuclear weapons, the technical ease of pro-
curement was sometimes exaggerated.
6
Fourth, in the 1970s it was mis-
takenly expected that commercial nuclear facilities for generating electric-
ity would burgeon around the world, providing options to build weapons
as a byproduct. Fifth, two strategic considerations were underestimated.
During the Cold War most of the countries that could produce nuclear
weapons were protected by a nuclear superpower ally who did not want
them to do so. In addition, there have always been quite powerful
disadvantages to acquiring nuclear weapons, costs that countries would
5. There were exceptionswell-known hawks such as Albert Wohlstetter and Fred
Iklwho worried about proliferation during the Cold War, but even they did most
of their campaigning on the subject in the 1970s, the trough between intense phases
of the East-West conict.
6. A prime example is the popularity of John McPhees The Curve of Binding Energy
(New York: Ballantine, 1975).
54 the coming crisis
not wish to bear unless they felt extremely vulnerable or extremely cocky.
Apart from the alienation of a superpower patron, the strategic costs
include aggravating relations with neighbors, and getting ones country
put on some great powers nuclear targeting list. It is conceivable that
some whose incentives for nuclear weapons came more from concern
with prestige than with safety might have been deterred by fear of
international opprobrium. That is doubtful, however, since it is not evi-
dent that countries that did acquire such weapons ever suffered for doing
so. (It is no accident that the ve permanent members of the United
Nations Security Council are the ve declared nuclear powers of the
half-century since World War II.)
The wave of concern about proliferation in the 1990s is the third we
have seen in the nuclear era. The rst was in the 1960s, when debates
about the credibility of the U.S. commitment to deliberate escalation in a
NATOWarsaw Pact War, the development of the British and French
independent deterrent forces, and Gaullism led Americans to worry that
other industrial nations would follow suit. The second was in the 1970s,
when the energy crisis led many to worry that the spread of nuclear
power for generation of electricity would enable Third World countries
to enter the club. The third wave is different, for both technical and
political reasons.
Previously, almost all concern focused on the spread of nuclear weap-
ons. Now there is appreciable attention to chemical and biological arms
as well.
7
The end of the Cold War also means that hawks and defense
experts now focus on proliferation, because without the Soviet Union and
Warsaw Pact to worry about, it is where the action is. This affects the
tenor of debates and proposals for how to deal with the question. More
of the discussion of proliferation now has a strategic avorthat is, terms
of reference oriented to balance of power, national security objectives, and
military strategythan it did when the Cold War was at its height.
Counterproliferation became a new watchword, now respectable in its
connotations of muscular policies. During the Cold War, proposals for
military solutions were not favored by those most worried about the
problem, who looked to diplomacy, propaganda, and arms control trea-
ties.
The main effect of the end of the Cold War in respect to proliferation
is on incentives to acquire WMD. To the countries for whom the end of
that war brings more peace and security, the attractiveness of WMD
should be lower. But who are those countries? For the most part they are
7. See Richard K. Betts, The New Threat of Mass Destruction, Foreign Affairs,
Vol. 77, No. 1 (January/February 1998).
chapter 3 55
members of NATO, but their incentives had already been controlled
during the Cold War by the U.S. umbrella. Since France became a nuclear
power more than three decades ago none of those countries has moved
in that direction. The new peace does little to change the odds in western
Europe, although it may increase incentives for vulnerable states in the
east. The two weightiest of those countries, however, show no signs of
moving for WMD: Poland has gotten into NATO, and Ukraine, which is
more vulnerable to reincorporation in Moscows orbit, bartered away the
nuclear weapons it already had for Western economic aid.
The end of the Cold War produced at least as much insecurity as it
resolved. Most of the new disarray involves not interstate violence but
conict over the formation or disintegration of states and political com-
munities. For those countries or groups whose security problems are
internal contests for control within a state, WMD have less strategic value
than in interstate conict. Large area weapons are not attractive to inter-
mingled contending populations and forces that want to preserve their
own groups lives and the property they are ghting over. For civil wars
in which the contestants and their home territories are clearly separated,
on the other hand, this may be less true. In either case, chemical weapons,
which I classify as only borderline WMD for reasons discussed below,
can remain militarily useful.
Those countries whose security problems are external and more
pressing than during the Cold War, who now feel more alone or less
protected by allies, have stronger incentives to acquire WMD. Along with
prestige, security self-reliance is the biggest incentive for a deterrent, and
has gured in the decisions of all countries that have acquired nuclear
weapons to date.
8
The most likely candidates to seek a serious WMD
capability have long been countries that have strong enemies and that
lack reliable great power protectors, and they will continue to be the most
likely candidatesalthough the end of the Cold War changes some of the
membership in that category.
9
Although insecurity has been an ingredient
in all proliferation so far, it is not a sufcient conditionnot all isolated
states have sought an independent deterrent.
8. South Africas decision to reverse its nuclear weapons program and disarm was
consistent with the change in its security situation. Once the apartheid government
decided to accept majority rule, external military threats were largely neutralized, and
the whites of the country had no interest in bestowing such weapons on a black
government. North Korea is now more politically isolated, and Israel less so, than they
were some years ago.
9. See Richard K. Betts, Paranoids, Pygmies, Pariahs and Non-Proliferation Revis-
ited, Security Studies, Vol. 2, Nos. 3/4 (Spring/Summer 1993), which updates an article
originally published in Foreign Policy, No. 26 (Spring 1977).
56 the coming crisis
Changes in incentives are half of the problem; changes in capabilities
are the other half. Here trends cut across each other. For nuclear options,
the dramatic slowdown in dispersion of peaceful nuclear power genera-
tion facilities limits the infrastructures that countries could choose to
divert toward weapons programs. It is unclear, however, how much this
matters. Every known nuclear weapons program, up to and including
those of Iraq and North Korea, has been a dedicated program, with facili-
ties largely designed and procured for the purpose of developing a
weapon. No country with nuclear weapons has gotten them by lurching
absentmindedly out of a peaceful nuclear power program. Cutting in the
other direction, however, is the loose nukes problem in the former
Soviet Union. There, controls on existing weapons, ssionable material,
and trained scientists have been under tremendous stress.
Does the rate at which WMD spread matter? The answer must be yes
if the costs or benets vary with the number of members of the club. In
the long run we are all dead, as John Maynard Keynes said, so the longer
we can delay the costs, or the sooner we can gain the benets, the better.
The probability that WMD will be used varies directly by some degree
with the number of potential users, and a panicky scramble could en-
courage half-baked, poorly safeguarded programs. It is also widely be-
lieved that there is some threshold of membership beyond which re-
straints will collapse and all but microstates will rush to avoid being left
behind. Similarly, a phase of rapid acceleration in which several members
are added in a brief time could provoke other states that might be willing
to watch and wait as long as additions occurred only every half-dozen
years or so.
Detonations by India and Pakistan in 1998 raised fears of a chain
reaction. Such fears in past waves of concern did not pan out. Several
countries bear watching (especially Iraq, North Korea, Iran, and Libya),
but their ambitions existed before the tests. Unless Pakistan writes off the
United States and deliberately aids Irans nuclear program, it is doubtful
that events on the subcontinent will spur emulation elsewhere.
which weapons?
Posing weapons of mass destruction as a category obscures the impor-
tant differences among nuclear, biological, and chemical arms.
10
Yet it
10. Some observers lump ballistic missiles with WMD. This is awkward since they
are delivery systems rather than weapons themselves, and missiles should not be
especially signicant apart from their role in carrying WMDand for a country that
has WMD, the mode of delivery is far less signicant than having the weapons, which
can be delivered by other means such as aircraft. Despite the psychological effects of
missile attacks with high explosives, missilesespecially ballistic missiles, which un-
chapter 3 57
remains important to think of WMD generically, in terms of what distin-
guishes them from normal weapons: their primary function or strategic
comparative advantage. The main purpose of most normal weapons is to
defeat or destroy military forces in combat, or to attack a specic war-
supporting installation. The comparative advantage of WMD is to inict
maximum injury on a society and its population. WMD should be capable
of producing tens of thousands of casualties in a single attack. (In these
terms, conventional high explosives were WMD in World War II when
they were used in incendiary attacks against residential areas by the
Royal Air Force over Germany and by the U.S. Army Air Force against
Japan. Since 1945 there has been an unbroken taboo not only against the
use of nuclear weapons, but also against the conventional bombing of
large areas of dense civilian population on a scale comparable to what
was done in the 1940s. The closest exceptionsthe U.S. bombing of Hanoi
in 1972, and the Iran-Iraq war of the citieswere not in the same class.)
Functions do overlap. Nuclear weapons can be used to strike other
nuclear weapons, or for small attacks on clumps of conventional forces;
biological weapons could be used against large concentrations of troops
in staging areas far from a front; and chemicals are well recognized as
weapons for the battleeld. Major nuclear powers during the Cold War
also devoted much planning and a signicant portion of their nuclear
resources to tactical applications, although they did so only after they had
elded hundreds of weapons for strategic use against the most vital
targets in each others interiors. Of the weaker countries likely to seek
WMD in the next few decades, few are likely to have so many that they
can be wasted on the battleeld (although Pakistani ofcials have
like air-breathing cruise missiles are expensive and must carry all their own energy
are extremely inefcient means for delivering conventional ordnance. Aircraft can
carry far greater payloads, and are reusable. The war of the cities between Iraq and
Iran in 1988 ranks as one of the smallest strategic bombing campaigns in history.
Hanoi absorbed 500 times as much ordnance from U.S. aircraft in the Linebacker II
campaign in eleven days as Tehran did from Iraqi missiles in seven weeks. The attacks
on Tehran had great effect on public morale, while those on Baghdad did not, because
the Iraqis were winning. See Thomas L. McNaugher, Ballistic Missiles and Chemical
Weapons, International Security, Vol. 15, No. 2 (Fall 1990), pp. 11, 13; and Steve Fetter,
Ballistic Missiles and Weapons of Mass Destruction, International Security, Vol. 16,
No. 1 (Summer 1991), pp. 9, 13. Ballistic missiles have also not been the subject of
efforts at treaty constraint in the same way nuclear, biological, and chemical weapons
have. Strategic arms limitation treaties and the Intermediate-range Nuclear Force
treaty limits U.S. and Russian missiles, but no one elses. The Missile Technology
Control Regime is a voluntary arrangement, not a treaty. This fact led the Bush
administration to oppose the Gore-McCain bill in 1989, which sought mandatory
sanctions on foreigners for transfers of missile-related items that would not be illegal
outside the United States. See Senate Committee on Foreign Relations, National Security
Implications of Missile Proliferation, 101st Cong., 1st sess., 1989, pp. 43, 47, 66.
58 the coming crisis
spoken of plans for tactical use of nuclear weapons). The destructive
effects and operational comparative advantage of these agents are great-
est when used against undefended civilians and economies. The relative
importance of the three types of WMD may be misjudged, however, in
the popular imagination.
Nuclear weapons have long been seen as the principal WMD prob-
lem, ever since they became the essential military ingredient in the titanic
Cold War struggle between the superpowers, the main symbol of strate-
gic independence and world-class clout. But even though the technical
barriers to acquiring ssionable material and weaponizing nuclear ord-
nance have eroded over time, they are still substantial, and still far greater
than the obstacles to getting either chemical or biological weapons.
Chemical weapons are most ambiguous. In the past they have been
unpopular among militaries as counterforce weapons, ever since the
mixed results of World War I (when winds sometimes turned chemical
salvos against the side that launched them), and because it is compara-
tively easy for modern forces to defend against CW given some tactical
warning. Protection against chemicals does impose burdens on a defense,
which could prove decisive where the local balance of power was close
to even. So chemical weapons are tactically signicant, but pound for
pound, they are most lethal when used against unprepared civilians.
Dominated by pictures of dead Kurdish women and children, the image
of chemical ordnance in recent times has been of a terror weapon. Chemi-
cals certainly count as WMD if used in high volume against concentrated
populations, but that is hard to do.
11
The amount of weaponized chemical
agent necessary for killing large numbers of people is hard to stockpile,
transport, and deploy. In some cases aircraft can deliver these weapons
effectively. Otherwise, the best way to deliver large quantities of CW is
by artillery, which can only be done within very close range. An army
that can get that close to its enemys cities is usually winning the war
anyway. On the other hand, chemical weapons can be very effective
against military forces if used in a surprise attack, before the defenders
can suit up. The unfavorable view of CWs battleeld effectiveness also
changed in the course of the Iran-Iraq War.
12
Observers have long recog-
11. With favorable weather conditions and no civil defense, chemical weapons can
be up to 500 times more deadly than conventional explosives. Fetter, Ballistic Missiles
and Weapons of Mass Destruction, p. 23.
12. In World War I, chemical weapons produced 1.6 to ve times as many casualties
as a similar amount of high explosive, but this average came from cases of use against
unprotected troops. McNaugher, Ballistic Missiles and Chemical Weapons, p. 19. In
the last year of the war, in contrast, chemical barrages accounted for 20 percent of the
artillery munitions used, but caused only 15 percent of casualties and 1.4 percent of
chapter 3 59
nized chemicals as more appropriate for tactical use than biological weap-
ons because chemicals can be used with more precision and discrimina-
tion and take quicker tactical effect.
13
Given the limitations of logistics
and delivery capability for covering large areas, CW offer the least
efcient mass destruction of all WMD.
Biological weapons (BW) received far less attention than either of the
other WMD until the late 1990s. They were banned by the superpowers
in 1972, and were largely forgotten by policymakers until recent revela-
tions that the Soviet Union and Russia had cheated and continued their
program. Inattention was also abetted by a widely held but incorrect
notion that, unlike chemical weapons, biological weapons have not been
used in warfare. The purposeful spreading of disease to enemies occurred
in a number of cases in premodern times, and the Japanese conducted
numerous eld tests of biological weapons during World War II, killing
substantial numbers of Chinese.
14
Biological weapons may be the greatest WMD threat of all, for two
simple reasons. First, they are far more effective for mass killing than are
chemical weapons. Second, they are far easier to acquire than are nuclear
weapons. To illustrate the rst point, the U.S. Ofce of Technology As-
sessment estimated that a single airplane delivering 100 kilograms of
anthrax spores by aerosol on a clear, calm night could kill more than 300
times as many people as one delivering 1,000 kilograms of Sarin nerve
gasbetween one and three million fatalities if the target was the Washing-
ton, D.C. area.
15
On the second point, earlier technical obstacles to making
deaths. Effective Iraqi use of chemicals against Iranian human wave assaults changed
attitudes. Brad Roberts, Chemical Disarmament and International Security, Adelphi Paper
No. 267 (London: IISS, Spring 1992), pp. 6, 16.
13. Victor Utgoff, The Challenge of Chemical Weapons: An American Perspective (London:
Macmillan, 1990), p. 98.
14. The Japanese attacks included the contamination of reservoirs, ponds, and wells;
aerial spraying against wheat crops; and the use of infected eas to spread typhus
rickettsia, cholera, and plague. The campaign was limited because delivery systems
worked poorly (for example, pathogens sent by artillery burned up on impact).
Sheldon Harris, Japanese Biological Warfare Research on Humans: A Case Study of
Microbiology and Ethics, Annals of the New York Academy of Sciences, Vol. 666 (Decem-
ber 1992), p. 32. Had the Japanese been able to develop an efcient aerosol dissemi-
nation system, the history of World War II would read quite differently today. Col.
Randall J. Larsen and Robert P. Kadlec, Biological Warfare: A PostCold War Threat to
Americas Strategic Mobility Forces, Viewpoints No. 95-4 (Ridgway Center, University of
Pittsburgh, n.d.), p. 5.
15. U.S. Congress, Ofce of Technology Assessment [OTA], Proliferation of Weapons of
Mass Destruction: Assessing the Risks, OTA-ISC-559 (Washington, D.C.: U.S. Govern-
ment Printing Ofce [GPO], August 1993), p. 54.
60 the coming crisis
biological agents effective have largely been overcome.
16
Problems in
stability, controlling effects, keeping agents alive over long periods, de-
livery, and the danger of infecting the user had made most strategists
uninterested in them. Modern fermentation procedures, aerosol dispens-
ing systems, genetic engineering, and other innovations in biotechnology
have abated many of the problems. Neither knowledge nor equipment
for the mass production of BW agents are esoteric or contained by export
controls. As a result, as long ago as 1988 the U.S. Director of Central
Intelligence reported that at least ten countries were developing biologi-
cal weapons; up to one hundred countries have the capability to do so,
and some estimated at the beginning of the 1990s that more than twenty
had a program.
17
who wants the weapons?
Rational strategic reasons to want WMD include: to deter the use of
WMD against ones own country; to redress inferiority in conventional
military capabilities by threatening to escalate in retaliation against an
16. At the end of the Korean War, the U.S. Joint Strategic Plans Committee concluded
that there were major technical obstacles to producing and disseminating biological
weapons effectively. John Ellis Van Courtland Moon, Biological Warfare Allegations:
The Korean War Case, Annals of the New York Academy of Sciences, Vol. 666 (December
1992), p. 68.
17. Kathleen C. Bailey, Doomsday Weapons in the Hands of Many: The Arms Control
Challenge of the 90s (Urbana: University of Illinois Press, 1991), pp. 8790; and Larsen
and Kadlec, Biological Warfare, pp. 1, 5. Until recently access to pathogens was easy.
Larsen and Kadlec reported that Virtually any agent can be legally acquired from
organizations such as the America Type Culture Collection (ATCC) in Rockville,
Maryland. ATCC is a legitimate corporation that serves the world-wide research
community. . . . One of its customers, however, was the Iraqi government, which
purchased anthrax and four other agents (p. 7). It is possible to tighten restraints on
legal commerce, but it is doubtful that controls can ever be strict enough to prevent
determined buyers from acquiring pathogens through intermediaries, theft, or collec-
tion from natural sources.
Table 3.1. Incentives for Choice Among WMD
Destructiveness
High Low

Availability
High Biological Chemical
Low Nuclear
chapter 3 61
enemys conventional attack; and to coerce an adversary into political
concessions. The rst two reasons make WMD weapons for the weak, the
third makes them weapons for the adventurous. The difference in these
motives matters for determining how dangerous proliferation will prove
to be, and what great powers ought to do about it.
People in the United States often lose sight of the defensive motives
for WMD, or assume that the desire of poor countries to obtain them is
less legitimate than the reasons that the United States itself has wanted
them. But is there any rationale for why the United States ever needed
nuclear weapons that is not an even stronger rationale for India or
Pakistan? India can hardly be deemed alarmist for wanting an option to
deter Chinaa superior power, on its border, with whom it fought a war
in 1962. And whatever is sufcient to deter China is more than enough
to frighten Pakistan. The probability of war between Pakistan and India
has always been higher than it ever was between the United States and
Soviet Union; India demonstrated its nuclear capability more than two
decades ago; and the imbalance of conventional military power in Indias
favor is greater than was the imbalance between NATO and the Warsaw
Pact.
18
Cuba, Vietnam, North Korea, and Taiwan are the most obviously
vulnerable states today, since they face enemies much stronger than they
are, and lack any assured alliance. By the same token, however, they also
have an incentive to avoid provoking their enemies into actionwhich
a nascent nuclear program might do. Other WMD programs are less
likely to be so riskychemical and biological weapons have never ener-
gized as much discussion of preventive attack.
The American tendency to depreciate defensive motives is greatest in
regard to countries considered bad or irresponsible, ones now commonly
called rogue states. Some of these are considered rogues because they
mount what are seen as unprovoked attacks on Western interests, or
because they are run by wild and crazy despots. However, it is not
necessary to make excuses for such governments to recognize that they
have good reason to worry about being put out of business by a super-
power. The more ostracized a regime, the more warranted is that fear.
The more that outside powers do to confront the regime, for however
many good reasons, the more they do to conrm the fear.
At the same time, if an effort to develop WMD is met with diplomatic
fulmination but acquiescence, the wisdom of the effort is conrmed. Take
18. See Richard K. Betts, Incentives for Nuclear Weapons, and Nuclear Defense
Options: Strategies, Costs, and Contingencies, in Joseph A. Yager, ed., Nonproliferation
and U.S. Foreign Policy (Washington, D.C.: The Brookings Institution, 1980).
62 the coming crisis
North Korea. It has been one of the most grotesque and odious of
governments. But who has any better reason for wanting a deterrent than
a country alone, one of the last anachronistic holdouts from the defeated
Second World of communist states, whose legitimacy is challenged by the
vastly richer and more populous southern half of the divided country,
which is in turn supported by the worlds only superpower? And what
penalty did Pyongyang suffer for violating its Nuclear Non-Proliferation
Treaty obligations and moving toward a weapon capability? It was not
attacked. It did not even have to bear economic sanctions. Rather, the
penalty was that it had to accept billions of dollars in aid and the
construction of new reactor facilities for generating energy, and it
achieved a breakthrough in diplomatic exchange with Washington that it
had been seeking for forty years. All of this bolsters the Pyongyang
regime, rather than promoting its collapse. Moreover, North Korea was
not required to rectify its treaty violation immediately by surrendering
or even accounting for the ssionable material it had that was sufcient
for one or two weapons, and was required only to begin moving in the
direction of neutralizing its nuclear option.
19
While the agreement may
have been the best that Washington could do, the point is simply that
Pyongyang was given no reason to regret its nuclear gambit.
Effects on Stability and Strategy
not to worry?
Liberal pessimism requires little exposition, since its logic strikes most
Westerners as self-evident. Utopian realism argues, in contrast, that we
should learn from the long peace between the superpowers during
the Cold War that nuclear weapons have been good for children and
other living things. They stabilized international relations by enforcing
caution, making the danger of attacking another nuclear power so
obvious that none dared take a chance on challenging the status quo by
force.
19. The Agreed Framework Between the United States of America and the Demo-
cratic Peoples Republic of Korea, which many in the West believe has brought North
Koreas nuclear option under control, did not require prompt destruction of the
plutonium reprocessing plant, nor even the immediate transfer of spent fuel to foreign
custody. The new reactors to be supplied will not be completed or turned over until
these matters are resolved to Washingtons satisfaction, but the new reactors, while
less easily applicable to production of ssionable material, are not incapable of being
used for it. The agreement commits the United States to reverse more than four
decades of policyto move toward full normalization of political and economic
relations, and to provide formal assurances to the DPRK, against the threat or use
of nuclear weapons by the U.S.
chapter 3 63
This argument must be taken seriously, at least by anyone who
believes that the mutual deterrence relationship between the United
States and Soviet Union during the Cold War was stable and reduced
their willingness to risk war, or who believes that the prospect of mutual
assured destruction rendered all the elaborate schemes for counterforce
targeting fanciful and made the piling up of tens of thousands of weapons
superuous. As Waltz says, Miscalculation causes wars. One side ex-
pects victory at an affordable price, while the other side hopes to avoid
defeat. Here the differences between conventional and nuclear worlds are
fundamental. In the former, states are too often tempted to act on advan-
tages that are wishfully discerned and narrowly calculated.
20
He also
argues that the reasons that nuclear deterrence kept the Cold War from
turning hot will be even more applicable and evident in the postCold
War world:
Nuclear weapons restore the clarity and simplicity lost as bipolar situations
are replaced by multipolar ones. . . .
Deterrent strategies offer this great advantage: Within wide ranges neither
side need respond to increases in the other sides military capabilities. . . .
This should be easier for lesser nuclear states to understand than it was for
the United States and the Soviet Union. Because most of them are economi-
cally hard-pressed, they will not want to have more than enough. . . .
States can safely shrink their borders because defense in depth becomes
irrelevant. . . .
The problem of stretching a deterrent [to cover allies], which agitated the
western alliance, is not a problem for lesser nuclear states. Their problem is
not to protect others but to protect themselves. . . . Weak states easily estab-
lish their credibility.
21
The Waltz argument cannot be brushed off, but surprisingly few
academic strategists besides Scott Sagan have refuted it in detail. Al-
though most intellectuals as well as normal people oppose proliferation,
writings arguing the benets are more obtrusive in the literature of
international relations theory. Why this difference between conventional
wisdom and some currents of academic fashion? One reason is that
outside of political science departments, people in the United States do
not approach the question as detached analysts shorn of national identity.
20. Kenneth Waltz, More May Be Better, in Scott D. Sagan and Kenneth N. Waltz,
The Spread of Nuclear Weapons: A Debate (New York: W.W. Norton, 1995), p. 6.
21. Waltz, in Sagan and Waltz, The Spread of Nuclear Weapons, pp. 14, 27, 30, 31, 33.
64 the coming crisis
They do not abstract themselves from the policy interests of the United
States, and do not care nearly as much about resolving the security
anxieties of non-nuclear countries as they do about minimizing the
chances that nuclear weapons will ever be used anywhere. The world is
an uncertain place, where parsimonious theories about stability may or
may not prove correct, but where our own country would have less to
worry about if we were the only ones to have weapons of mass destruc-
tion. It is easier for ofcials than for analysts to apply shameless double
standards and recommend policies that are better for the United States
than for other countries that want strategic independence.
High-quality theory is not necessarily a direct guide to good policy.
In the scientically rickety world of social science, any theory that pre-
dicts, say, 90 percent of outcomes on some important matter is an amaz-
ingly good theory. The Waltz argument may be in that category. In the
overwhelming majority of cases, new nuclear states may be more cau-
tious and remain deterred by each other. In the world of policy, on the
other hand, people do not marvel at all the cases where nuclear weapons
will make the world safer, but worry about the exceptions where things
will go wrong. Those wrapped up in policy also take more seriously the
prospects for nonrational or accidental action associated with complex
organizations, problems that Sagan poses as the main grounds for greater
pessimism than Waltz derives from looking at the broader logic of the
international balance of power.
22
In most cases, the logic of deterrence theory that became the bedrock
of U.S. strategic thought in the course of the nuclear era suggests that the
acquisition of nuclear weapons should have a stabilizing effectthat is,
they should make it hard to change the status quo by force. Those who
have a powerful deterrent will be less coercible or conquerable. It is less
clear whether they will coerce non-nuclear neighbors. Rogue states that
start brandishing nuclear threats risk bringing down an international
consensusand more signicantly, U.S. power and countercoercionon
themselves. On the other hand, they may sometimes nd that nuclear
capability makes the outside powers more amenable to negotiation than
they might otherwise be (as in the case of North Koreas diplomatic coup
with the United States).
If nuclear spread enhances stability, this is not entirely good news for
the United States, since it has been accustomed to attacking small coun-
tries with impunity when it felt justied and provoked. The United States
is not accustomed to being deterred by anyone but the Russians or
22. Scott Sagan, More May Be Worse, and Sagan Responds to Waltz, in Sagan
and Waltz, The Spread of Nuclear Weapons, pp. 4792, 115136.
chapter 3 65
Chinese. This is not the main reason, however, that the Waltz argument
fails to command enthusiasm. The main reason is the worry that real
statesmen may not always have the courage of Waltz convictions, that
one exception to the rule may be too many, and that the ramications of
the rst breakage of the half-century taboo on nuclear use are too unpre-
dictable to tempt us to run the experiment. If the probability that nothing
will go seriously wrong in any one case of proliferation is a reassuring
90 percent, the odds that nothing will go seriously wrong in any of them
decline steadily as the number of cases grows. In short, when it comes to
nuclear weapons, very stable in almost all cases is great for purposes
of theoretical clarication, but not good enough for purposes of policy
prescription.
potential effects
Nevertheless, if logic suggests that the most common effect of the spread
of WMD is to increase security and caution, and that more countries
should be able to take care of their own security, there should be declines
in expansive conceptions of core security; the disposition of states armed
with WMD to meddle with each other; the importance of alliances; and
the importance of international legal regulations for WMD.
Most countries security policies are only about their core security
territorial integrity and sovereignty over the country itself. Only great
powers dene their security more broadly, to include buffer zones or the
security of friends and allies. But all of the countries historically consid-
ered great powers (except Germany and Japan, which still recall the
experience of their own states extinction) have already had a nuclear
capability for some time. Some small countries, such as Israel, have been
impelled by geography to include control of adjacent territory within
their concept of core security. Shai Feldman, a student of Waltz, argued
that reliance on nuclear deterrence could allow Israel to relinquish occu-
pied territory.
23
Yet Israel has had nuclear weapons for over a quarter of
a century, long before Feldmans analysis, and long before it proved
willing to cede territory conquered in 1967.
The notion that widespread nuclear capability would inhibit aggres-
sion by creating a world of porcupines or a unit veto system of om-
nilateral deterrence is an old one.
24
The suppression of military interven-
tionism, however, could simply channel impulses to meddle into covert
23. Feldman, Israeli Nuclear Deterrence.
24. Morton Kaplan, System and Process in International Politics (New York: Wiley, 1957),
pp. 5052, 69; and George Liska, Nations in Alliance (Baltimore, Md.: Johns Hopkins
University Press, 1962), pp. 271274.
66 the coming crisis
political action or other less direct methods. These in turn could increase
diplomatic tension and the chances of miscalculation, especially since
many of the political systems of the potential proliferators are likely to
be weak, permeable, and praetorian, unlike the stable institutionalized
governments of the developed world.
25
Internal political weakness and
externally deployable military strength (via WMD) are a volatile combi-
nation. It was reckless enough for the Argentine junta in 1982 to divert
public attention from internal economic problems by grabbing the Falk-
land (or Malvinas) Islandsone of only two cases of a non-nuclear state
initiating combat against a nuclear power (the other being Egypt and
Syria against Israel in 1973). Would that conict have unfolded differently
if Argentina had had its own nuclear weapons?
To the extent that the spread of deterrence capability makes more
countries strategically independent, it should depreciate the importance
of alliance guarantees, by making them both less necessary and less
credible. This was the rationale behind the Gaullist force de frappe and
critique of NATO in the 1960sa rationale that made much more sense
than U.S. policymakers would ever admit. Indeed, the attraction of an
independent deterrent should vary inversely with condence in alterna-
tive means of protecting ones security. States have not acquired nuclear
weapons during periods of growing support from allies. Britain acquired
its nuclear capability in a period of no change in its alliance relationship
with the United States; France got its nuclear force after the Suez crisis
weakened its tie to Washington; China as relations with the Soviet Union
worsened; Israel and India when they were not members of dependable
military alliances; and North Koreas nuclear program took off when rst
U.S.-Soviet dtente and Sino-U.S. rapprochement, and then the end of the
Cold War, weakened external guarantees to Pyongyang.
26
(The lure of
strategic independence of course is a matter of degree.
27
Despite with-
25. On praetorianism, which makes the dynamics of domestic politics in poorly
developed states analogous to those of the international state of nature, see Samuel P.
Huntington, Political Order in Changing Societies (New Haven, Conn.: Yale University
Press, 1968), chap. 4. For reections on the problem of controlling nuclear weapons
held by unstable governments, see Lewis A. Dunn, Military Politics, Nuclear Prolif-
eration, and the Nuclear Coup dEtat, Journal of Strategic Studies, Vol. 1, No. 1 (May
1978).
26. Pakistan is a mixed case, since it brought its nuclear program to fruition in the
1980s when, motivated by the need for Pakistani support in the Afghan guerrillas, the
United States was supplying more foreign aid than before. The program was initiated
and accelerated, however, much earlier, when Washington was refusing to supply even
A-7 aircraft.
27. Waltz exaggerated the extent to which Cold War bipolarity reduced the impor-
tance of alliances. His only explanation for why the United States and Soviet Union
chapter 3 67
drawing from the integrated military commandan institutionalized
mechanism of alliance historically unique in peacetimeFrance remained
a steadfast member of the Atlantic alliance itself.) The spread of inde-
pendent deterrents should also further diminish the already tattered
hopes for collective security through multilateral enforcement by inter-
national organizations like the United Nations. If deterrence by threat of
mass destruction works against a regular alliance, there is no reason it
should not work against a coalition representing the UN.
Even after WMD have spread much more, there will be areas of
conict among countries or groups that do not have their own such
weapons. In those cases the incentives and inhibitions on multilateral
involvement will not necessarily change much. What will be more com-
mon and more awkward are cases in which one local contestant has
WMD, the others do not, and outside powers support these others. We
have already run one experiment of that sort, after Iraq invaded Kuwait.
Baghdads demonstrated chemical and suspected biological capability
did not deter U.S. attack. Was this because U.S. forces had the equipment
with which to defend against chemicals and had vaccinated troops
against anthrax? Would knowledge that Baghdad had a few nuclear
weapons have deterred the Western coalition from invading Iraqi terri-
tory? Is a superpower just not likely to be intimidated by a country that
is no more than a medium power apart from its WMD arsenal? Or will
leaders think of chemical and biological weapons in signicantly different
ways from how they have understood nuclear arms, and thus limit the
transferability of logic from the highly developed deterrence theory of
the Cold War nuclear competition? (See the chapters in this volume by
Stephen Rosen and Barry Posen.)
dubious restraints
Liberals opposed to proliferation have put most of their hope into
promoting a nonproliferation regime based on a bevy of treaties,
especially the Nuclear Non-Proliferation Treaty (NPT), the Comprehen-
sive Test Ban Treaty (CTBT), and international conventions banning
chemical and biological arms. Much energy has been expended in con-
ferences and speeches about reinforcing nonproliferation norms. This
emphasis overlooks the essence of the problem and ignores the nontrivial
scrambled so furiously to line up allies and keep them in line was that they were
overreacting to each other. Similarly, he does not offer a compelling reason for why, if
both superpowers possessed nuclear deterrents that were as solid and dependable for
ensuring their security as Waltz maintained, they did not consider isolationist foreign
policy options more seriously. Kenneth N. Waltz, Theory of International Politics (Read-
ing, Mass.: Addison-Wesley, 1979), chap. 8.
68 the coming crisis
costs of making treaty adherence the centerpiece of nonproliferation
strategy. What are the downsides of treaties aimed at proscribing
weapons?
First, as useful as treaties are, it is a misconception to see them as a
solution. They are effects of nonproliferation, not causes of it. The NPT
and CTBT reect the intent of their adherents to abjure nuclear weapons.
To date, the countries considered problematicthose that might acquire
nuclear weaponssimply did not join the NPT. (South Africa stayed out
while it had a nuclear weapons program and joined when it decided to
get rid of it.) Or else they joined and cheated (Iraq and North Korea). The
inspection obligations of treaty membership did not reveal or reverse the
weapons programs of Iraq and North Korea. Iraq passed International
Atomic Energy Agency (IAEA) inspections, and it was unilateral U.S.
intelligence collection, not the IAEA, that detected the illegal North Ko-
rean activity. If the NPT or CTBT themselves prevent proliferation, one should
be able to name at least one specic country that would have sought nuclear
weapons or tested them, but refrained from doing so, or was stopped, because of
either treaty. None comes to mind.
Second, however useful the treaties are, some of themthe chemical
and biological weapons conventionsare not veriable with any
condence. Facilities needed to develop such weapons are sufciently
compact and similar to legitimate commercial facilities that inspectors
would need much more luck to overcome a dedicated concealment effort
than they would in looking for forbidden nuclear installations. It is better
to have countries under legal inhibitions than not, but such efforts should
not take precedence over other measures to limit countries incentives
and capabilities.
Third, selling the treaties on nuclear limitations in terms of legal and
moral norms is politically counterproductive. When Indians and Paki-
stanis with a handful of nuclear devices hear sanctimonious demands to
give them up coming from Americanswho face far fewer uncertainties
about their core territorial security, and who will continue to possess
thousands of well-tested nuclear weapons even if stockpiles come down
by 90 percent or morethey understandably react with exasperation. The
charge that the NPT attempts to establish an international caste system
is not hyperbole; that is the very essence of the treaty. The NPT is based
on an irrevocable division between the few rich countries who had these
weapons in 1968, and all others who were expected to remain without
them forever. This division is in the interest of the United States and the
other nuclear haves, but it undermines the pursuit of that interest to
pose it as a moral principle. It is less quixotic to deal in terms of a bargain,
offering things in the interest of the targets to convince them that the
chapter 3 69
benets of giving up the capabilities exceed the costs in security or
prestige.
28
Fourth, attempting to coerce countries to observe treaty regimes, even
when they have not signed the treaties, risks damaging other important
interests without achieving the nonproliferation objective. The legislation
that required imposing economic sanctions on India and Pakistan in 1998
put the United States in the bizarre position of being on better economic
terms with and more politically supportive of China, an authoritarian
power that abuses human rights, poses potential problems for U.S. secu-
rity interests, and has many well-tested nuclear weapons, than it is in
regard to India, the worlds largest democracy, which has few and less
frequently tested nuclear weapons, and has never threatened the United
States. It also left Pakistan weakened by sanctions and with little left to
lose from sharing its nuclear technology with richer Islamic states.
In any event, if WMD spread widely, international treaty restraints
will by denition wane to irrelevance. This will matter symbolically, and
for the morale of antiproliferation crusaders. These are not minor consid-
erations, and treaty restraints are useful in other important respects. For
example, the inspection provisions of the NPT can provide warning of
the diversion of ssionable material, or at least complicate the diverters
job of deception. The NPT, as well as treaties on chemical and biological
weapons, do have value as rallying points around which great powers
might mobilize a coalition to move against violators. (This has not hap-
pened yet beyond the case of Iraq, where detected violations reinforced
the case for sanctions.) Without the invasion of Kuwait and the additional
evidence uncovered after the Iraqi surrender, it is unlikely that sanctions
would have been imposed because of the NPT violation alone. The NPT
did provide a legal pretext on which to demand international sanctions
against North Korea, but Washington did not press for them. Instead it
moved to buy off the North Koreans.
back to optimism?
If the nonproliferation regime unravels, should we count on the tradi-
tional deterrence logic of utopian realism? It is based more on logic than
28. The NPT originally did this, by offering a quid pro quo. Article IV of the treaty
guarantees the have-nots access to the fullest possible exchange of technology for
peaceful nuclear energy generation in exchange for not building weapons. This was
undermined by the development of the nuclear suppliers cartel, which functioned to
cut off trade in technology for reprocessing plutonium or enriching uranium. And
since the nuclear energy industry is near collapse now anyway, technology transfer is
not much of an inducement anymore. In contrast to its original rationale, the NPT now
constitutes a simple demand to the nuclear weapons have-nots to remain so.
70 the coming crisis
evidence. Empirical validation for the logic rests on the one experience
of long-routinized deterrence, that between the United States and Soviet
Union in the second half of the Cold War. The early phase of the Cold
War, before crises over Berlin and Cuba worked out the limits to probes
and provocations, is a less reassuring model. Only with hindsight is it
easy to assume that because the superpowers did not go over the edge,
it was foreordained by deterrence that they could not. Only because
China did not take Quemoy and Matsu, the Soviet Union did not take
West Berlin, and the United States did not take Cuba does it appear
inherent in the logic of deterrence that they would not. But then why did
the Soviet Union and Chinaboth nuclear powers in 1969come to
blows on the Ussuri River? Why did Egypt and Syria attack Israeli-
occupied territory in 1973? Why did Argentina attack territory held by a
nuclear power in 1982? Waltz answers that nuclear deterrence only works
to protect core securityhome territorynot allies, occupied territory, or
faraway possessions.
29
But the territory in dispute on the Ussuri was not
a colony or an ally. The U.S.-Soviet model of Cold War deterrence does
not match the geographic conditions of many other conicts. Unlike
China, India, and Pakistan, the United States and Soviet Union did not
share a land border, and had no territorial claims against each others
homelands. And if nuclear deterrence cannot be expected to cover allied
territory, why did Waltz denigrate anxiety about NATOs reliance on U.S.
extended nuclear deterrence, and pooh-pooh U.S. efforts to shift toward
reliance on conventional defense? There are too many contortions neces-
sary to square all these exceptions and contradictions.
If nuclear deterrence makes countries less worried about their core
security, why should it necessarily make them less expansive or adven-
turous? They may feel less compulsion to rely on territorial buffers, but
they may also feel more able to throw their weight around outside their
borders. If U.S. and Soviet nuclear power coincided with their global
competition and expansive conceptions of security, why should nuclear
capability now make smaller powers more content to look no further than
their most fundamental interests? Might it not offer them more options
for wider-ranging foreign policy than they had before?
This should be especially true where mutual deterrent balances do
not emerge immediately, and new nuclear powers are the only ones in
their neighborhoods. If local powers start pushing their neighbors
around, the risks increase that a superpower accustomed to thinking and
acting as a global policeman will get involved. In few such situations
would the rules and limits of the competitive game already have been
29. Waltz, in Sagan and Waltz, The Spread of Nuclear Weapons, p. 16.
chapter 3 71
worked out, as it took Moscow and Washington more than fteen years
to do.
Perhaps a completed Iraqi nuclear capability would have deterred
the United States from rolling back the conquest of Kuwait. This is far
from certain, however, since the other Iraqi WMD capability did not.
There was much public discussion of the danger posed by Iraqs chemical
weapons, and somewhat less of its biological weapons, but negligible
argument that they made it too risky to attack Iraq.
30
Perhaps this just
reects the weakness of chemical weapons as WMD. Or perhaps it reects
the extent to which people in the United States have trouble thinking
about being deterred by a country that is not a great power. On the other
hand, the extent to which Washington rewarded North Korea for violat-
ing the NPT points in the opposite direction: it may be easy not only for
a second-rate power to deter, but also for it to coerce.
There are many technical and political counterarguments to Waltzian
optimism about nuclear spread: poor countries nuclear forces may be
both small and vulnerable, creating the incentive for preventive attack;
conict dyads in many regions are between contiguous adversaries, pre-
cluding the tactical warning time of surprise attack on which U.S.-Soviet
deterrence rested; animosities among contestants in some parts of the
world (for example, Hindus vs. Muslims, Arabs vs. Israelis, Serbs vs.
Albanians) are more intense and visceral than the tensions between
Americans and Russians ever were, and the difference threatens cool
calculation during crises; cognitive orientations, value systems, and
modes of calculation in some non-Western cultures are not conducive to
the economistic logic of Western deterrence theory; and so forth.
The optimists answers boil down to two points. First, deterrence
does not require either the overkill or the elaborate technical aspects of
stability associated with the evolution of the U.S.-Soviet balance. Simple
uncertainty about the ability to escape even a few weapons from a
victims retaliation is enough to deter a country from mounting a nuclear
attack, since the potential loss of a few cities is out of all proportion to
possible gains. It is also easy for even the poorest country to hide a few
weapons and deliver them inefciently, slowly, but on target nonetheless.
The acionados lore of technical stability, counterforce duels, and whiz-
bang command and control systems is beside the point, and mistakes the
details that preoccupy developed nations militaries and professional
strategists for the concerns that govern the decisions of political leaders
30. The Chairman of the Joint Chiefs of Staff at the time reports that biological
weapons were a bigger concern internally than was manifested in the press. General
Colin L. Powell, USA (Ret.) with Joseph Persico, My American Journey (New York:
Random House, 1995), pp. 494, 503504. See also Barry Posens chapter in this volume.
72 the coming crisis
with common sense. Second, worries about irrational and reckless leaders
represent a crypto-racist double standard. Even the wildest and nastiest
prove quite sober when contemplating the extinction of their societies
and of themselves. The magic of nuclear reality is that its risks are so
perfectly stark that no one can goof.
There are two main rebuttals, and they are convincing. First, the fact
of a half-century of nuclear peace between the superpowers leads the
optimists to assume that what was, had to be, and to overestimate how
intrinsically safe the confrontation was. Although U.S. and Soviet leaders
meant to be cautious, there were numerous accidents that raised the risk
of inadvertent escalation.
31
Moreover, the tendency of military elites to
consider preventive war as a solution more readily than civilian politi-
cians do manifested itself even in the United States; in newly nuclear
countries with military governments, these tendencies would not be as
reliably constrained as they have been. Second, irrationality and idiocy
do occur in the behavior of political leaders. What is too risky to contem-
plate is often clear only in hindsight. Waltz himself attributes the mis-
taken overarmament and anxiety about stability in the superpower rela-
tionship to fuzzy thinking, yet does not worry that fuzzy thinking will
lead new nuclear states to build vulnerable forces or make other danger-
ous mistakes. As Sagan puts it:
Waltz and other nuclear proliferation optimists have confused prescriptions
of what rational states should do with predictions of what real states will
do. . . . This is an error that Waltz avoided in Theory of International Politics,
where he noted that the theory requires no assumptions of rationality . . .
the theory says simply that if some do relatively well, others will emulate
them or fall by the wayside . . . Adding this element of natural selection to
a theory of international relations puts less of a burden on the assumption of
rationality. My approach is consistent with this vision. Many nuclear states
may well behave sensibly, but some will not and will then fall by the
wayside. Falling by the wayside, however, means using their nuclear weap-
ons in this case.
32
Finally, experts in both government and academia tend to forget that the
lore of deterrence taken for granted in the United States after decades of
facing the issue on the front burner of policylore that seems so quin-
tessentially logical, transcultural, and, to the optimists, self-evidentis
not so obvious to all responsible individuals in countries to whom the
31. Scott D. Sagan, The Limits of Safety (Princeton, N.J.: Princeton University Press,
1993).
32. Sagan, in Sagan and Waltz, The Spread of Nuclear Weapons, pp. 8687.
chapter 3 73
questions are new. Perhaps the implications of vulnerability to mass
destruction are not subject to ethnocentric interpretation, and will have
forced themselves into the consciousness of nuclear neophytes by the
time they get far down the road to deployment. But cultural or other
subjective reasons for differences in assumptions or sensitivities are not
always evident before some crisis forces them into view.
Two anecdotes heard in conversation years ago illustrate this point.
One came from a U.S. consultant who described a meeting with a group
of the Brazilian political elite (business leaders and military ofcers) in
the mid-1970s, when a deal to acquire full-cycle nuclear power capabili-
ties from Germany was in the works. The American told the Brazilians
that they should think carefully before venturing toward a nuclear weap-
ons program. Oh, why? he said they asked. Well, he responded,
right now Brazil is the most powerful country in the region and has no
external security problems at all. But if you get nuclear weapons, Argen-
tina will get them too. Then for the rst time in Brazilian history all your
cities will be at risk. Some hesitant mumbling among the participants
followed, and one said, You know, we never thought of that. The
second story is of a conversation in the 1980s. An American was spouting
standard arguments about reciprocal fear of surprise attack, survivable
second-strike capability, and the delicate balance of terror, to try to con-
vince a Pakistani air force ofcer of the danger of deploying a small but
vulnerable force. The American emphasized that a few weapons on
aircraft on soft airelds would be a magnet to Indian attack. He insisted
that at the least the Pakistanis should refrain from deploying nuclear
weapons until they conformed to responsible standards of technical sta-
bility. Throughout this monologue, the ofcer kept shaking his head.
Finally the Pakistani said, with great vehemence, You just dont under-
stand. Once we have it, they wont dare attack.
33
Of course these stories do not prove how countries would develop
strategy or deploy forces. But they remind us that there is no reason to
assume that the deterrence theory that became second nature to U.S.
strategists in the Cold War will govern future developments either, any
more than the theory of laissez-faire capitalism that seemed self-evident
to many Western economists governed economic policies in India and in
33. Sagan notes that military organizations, manifesting the typical tendency of
organizations toward goal displacement, resist diverting funds from procurement of
weapons systems to security measures. For example, the U.S. Air Force refused until
long into the nuclear era to allocate substantial resources to reducing the vulnerability
of Strategic Air Command bases, and China left its developing nuclear force vulner-
able for a full decade. Sagan, in Sagan and Waltz, The Spread of Nuclear Weapons,
pp. 6768, 72.
74 the coming crisis
other non-Western countries. Jaswant Singh, a former military ofcer and
principal spokesman on security policy for the new government in India
in 1998, noted in several fora that U.S. obsessions with command and
control arrangements or technical congurations of forces were not a
concern for his government.
34
The perceived stability of superpower mutual deterrence took years
to evolve. Most would estimate that if nuclear spread encourages preven-
tive war, it is most likely to happen in the early phase of force develop-
ment, before redundancy or protective measures secure a striking force
against destruction, or better yet, in the phase when states are developing
ssionable material, when the critical targets (reactor, reprocessing, or
enrichment facilities) cannot be easily hidden. There is a natural contra-
diction, however, between timing that makes sense in strategic terms (the
sooner the better, when the odds of disabling the capability without
suffering retaliation are highest) and the timing that makes sense in
political terms (the later the better, when the danger has become real and
unambiguous, when forming a government consensus for action is easi-
est, but when there is more doubt about escaping retaliation).
As of 1998, about eleven countries had built or come close to produc-
ing nuclear weapons (the United States, Russia, Great Britain, France,
China, Israel, India, Pakistan, South Africa, Iraq, and North Korea). Some
others have ventured toward a nuclear weapons program but not gotten
as far. Only once has an adversary mounted a preventive attack: Israel de-
stroyed Baghdads Osirak reactor in 1981. It is debatable whether that
strike did more to retard Iraqi moves toward nuclear weapons than to
spur them.
35
In any case, Iraq was far closer to getting nuclear weapons
a decade later than it was when the Israelis struck. The United States
attacked a much wider range of Iraqi nuclear facilities in 1991, but only
because a war triggered by events having nothing to do with the nuclear
problem provided the excuse.
Nothing in the record so far suggests that preventive war will be a
frequent choice of countries facing a developing nuclear capability. It is
least improbable in the combination of circumstances that came together
in the case of the Israeli attack on Iraq: the nuclearizing adversary appears
aggressive and risk prone; it lacks signicant conventional military means
to retaliate for a strike on nuclear facilities; the attacker has an expansive
34. We are not replicating the experience of the West, Singh says. Therefore, what
the West constructed in the management of their arsenals is not what India requires.
Kenneth J. Cooper and John Ward Anderson, Walking on Thin Ice, Washington Post
National Weekly Edition, June 8, 1998, p. 6.
35. Richard K. Betts, Nuclear Proliferation After Osirak, Arms Control Today, Vol.
11, No. 7 (September 1981), p. 1.
chapter 3 75
conception of its own core security; and attack with conventional ord-
nance can effectively destroy the existing nuclear capability. However,
since the incentives for preventive war are long-term vulnerability rather
than immediate crisis, there is always time to continue assessing the
question, mulling options, searching for other solutions. There is no
moment at which a decision seems necessary, so the path of least resis-
tance is to wait.
Preemptive attack should be less unlikely than preventive war.
36
In a
crisis, with ambiguous warnings or short mobilization times (especially
for a nuclear striking force), the distinction between preventive and
preemptive attack (which is obscure to some observers anyway) may
blur.
37
During the Cold War there was only one near-war crisis, in October
1962, and the United States prepared its nuclear striking forces for pre-
emption by moving to DEFCON-2 (readiness just short of wartime)
before there was any evidence of Soviet preparation to attack.
38
Future
crises among newer nuclear nations may involve more volatile political
interactions and more vulnerable forces to tempt attack. Preemption
becomes most tempting if it also appears possible to accomplish with
non-nuclear forces.
All of these arguments and judgments are oriented to nuclear weap-
ons. There has been less consideration of how deterrence theory applies
to chemical weapons, and far less in regard to biological. These questions
36. This may not be saying much. See Dan Reiter, Exploding the Powder Keg Myth:
Preemptive Wars Almost Never Happen, International Security, Vol. 20, No. 2 (Fall
1995).
37. Preventive attack is undertaken against a potential and growing threat, lest the
target country become too strong to defeat at a later date. Preemptive attack is spurred
by strategic warning, evidence that the enemy is already preparing an imminent
attack. See Richard K. Betts, Surprise Attack (Washington, D.C.: The Brookings Institu-
tion, 1982), pp. 145147, and Betts, Surprise Attack and Preemption, in Graham T.
Allison, Albert Carnesale, and Joseph S. Nye, Jr., eds., Hawks, Doves, and Owls (New
York: W.W. Norton, 1985), pp. 5658.
38. Consistent with readying the option to strike rst, the U.S. alert served the
purpose of diplomatic coercion. Thus the preparations were not hidden from the
Soviets. See Marc Trachtenberg, The Inuence of Nuclear Weapons in the Cuban
Missile Crisis, International Security, Vol. 10, No. 1 (Summer 1985), pp. 157, 159, 161;
and Richard K. Betts, Nuclear Blackmail and Nuclear Balance (Washington, D.C.: The
Brookings Institution, 1987), pp. 118120. The DEFCON-3 (for Defense Condition 3)
alert in October 1973, and especially the interaction of U.S. and Soviet ships in the
Eastern Mediterranean, was more dangerous than political authorities in Washington
realized. See Joseph F. Bouchard, Command in Crisis: Four Case Studies (New York:
Columbia University Press, 1991), chap. 6. In that case, however, there was no serious
belief within the military establishment that preemption might occur.
76 the coming crisis
become more important as the United States faces less danger from
hostile great powers and more from aggrieved subnational groups.
Preventive or preemptive rst strikes become most probable when
WMD capability is deployed by nongovernmental terrorists. By deni-
tion, terrorists have the greatest interest in using terror weapons.
39
To date
this possibility has commanded a great deal of attention, but while nearly
a dozen states have acquired nuclear weapons and many more have
acquired other forms of WMD, no non-state terrorist organization has yet
used WMD for blackmail.
40
This restraint is ironic, because terrorist groups represent the best
potential exception to the logic of Waltzian optimism. They should be
harder to deter than states. Their locationthe address upon which
retaliation can be visitedis often hard to pin down. If terrorists can be
pinned down, they are likely to be targeted anyway, irrespective of
whether they have WMD. Because they are hard to nd and target, at
least without inicting tremendous collateral damage, terrorists should
have more leeway to use WMD for coercion (it would be easy to kill
certain known cliques if counterterrorists were willing to wipe out all of
Damascus or Tehran). Holding a city hostage beats holding the passen-
gers of an airliner.
The danger that terrorists might obtain nuclear weapons has been the
focus of most concern, but the more probable danger is their use of
chemical or biological weapons, given the easier access to them. This was
evident in 1995 when the Aum Shinrikyo cult released chemical agents
in the Tokyo subway and elsewhere.
41
But why have terrorists not done
39. I dene terrorism in two variants, which overlap: individualistic anomic violence
designed to punish the target for past sins (for example, Unabomber murders); and a
strategy of coercion aimed to induce compliance with demands by threatening to kill
the targets civilian population. (Making the threat credible may require demonstrating
the capability.) The second variant is most likely to be a strategy of the weaker side
in a conict dyad, since the stronger can more easily achieve objectives by using
superior conventional power and defeating its opponents military forces.
40. A borderline case occurred late in 1995, when Chechens reportedly placed a small
amount of radioactive material in a public area in Moscow, and claimed to have
smuggled three other parcels of cesium into Russia. Rebel leader Shamil Basayev was
quoted as saying, People these days . . . think we can no longer hurt the Russians.
So we will give them a little sign . . . that we are completely prepared to commit acts
of terrorism. Michael Specter, Chechen Insurgents Take Their Struggle to a Moscow
Park, New York Times, November 24, 1995, pp. A1, A17. The Aum Shinrikyo chemical
incidents in Tokyo do not count as coercion; no demands preceded or followed the
attacks.
41. See the thorough review of these incidents in Ron Purver, Chemical and Biological
chapter 3 77
more of this? For one thing, any use of WMD might overcome inhibitions
of great powers to attack terrorists whose address is ascertainable, or the
states who tolerate their presence. But perhaps the greatest reason for
optimism is that there is yet little evidence of terrorists who have an
interest in inicting true mass destruction (tens of thousands of casual-
ties). The closest things have been hijackings involving a few hundred
potential victims. For those promoting a political cause, it may seem that
really impressive mass murder would be counterproductive.
The biggest danger would be from a group or state that did not want
to be recognized and negotiated with, one that wanted only to hurt its
victim without getting credit (as with the Lockerbie bombing). And the
biggest danger within this category would be a group or state that sees
its victims whole society or culture as a threat, rather than a particular
government, and could thus want mass destruction for its own sakefor
example, religious or cultural radicals of the sort who bombed the World
Trade Center, those who see liberal Western culture as a threat to their
preferred social order. Since even fanatical governments are not usually
suicidal, deterrence by threat of annihilation in retaliation should still
help the United States.
Other Options
alternatives to deterrence and compellence
The realist tradition offers the best guidance to security policy, on balance,
but this does not include the utopian realist factions approval of WMD.
The best world for the United Statesespecially while it is the only
superpower in conventional military termsis one in which the smallest
possible number of other countries have WMD. Better that some weak
countries be forced to live with more insecurity than that they have the
capacity to catalyze events that injure the United States. But what if more
proliferation nevertheless occurs? Two approaches that have been ne-
glected or rejected in U.S. policy discussions should become less unthink-
able as WMD spread: passive defenses and political appeasement.
There has been much discussion of active defenses, even continuing
proposals for investment in ballistic missile defense despite the end of
the Cold War. An attempt at such an active defense system makes sense
for threats in the middle range, rather than threats at either extreme. That
is, a capability to counter a major power that can deliver more than a
handful of weapons, but remains unable to deliver thousands (as the
Terrorism: The Threat According to the Open Literature (Canadian Security Intelligence
Service, June 1995), pp. 153189.
78 the coming crisis
Soviet Union could), might make sense. At the extremes, in contrast to
this middle range, anti-missile systems offer too little or too muchtoo
little against a sophisticated saturation attack, too much against weaker
enemies that do not rely on ballistic missiles. For poor and weak coun-
tries, other means of delivery will be far more logical than ballistic
missiles for a long time to come. Commercial aircraft, ground-hugging
cruise missiles, or clandestine insertion are easier to use and just as
effective as ballistic missiles for purposes less ambitious than destroying
dozens of targets. Clandestine means are not absolutely easy to use, but
they are comparatively easy to use.
Passive civil defenses should have a higher priority than ballistic
missile defense. They pose a far smaller claim on resources and (at least
against nuclear and chemical arms) offer more anti-bang for the buck.
Civil defenses that would offer little protection against an attack by
hundreds of megatons could be much more effective in coping with a
dozen or so low-yield ssion weapons. It would not help near ground-
zero, but could reduce blast casualties in adjacent areas and fallout casu-
alties in outlying areas.
Until recently it was hard to imagine how politicians could summon
the public to embrace a civil defense program now when it could not be
sold successfully during the Cold War. In the past, arguments for major
efforts to anticipate and block WMD threats from small enemies were
attacked by some as alarmist and by others as offering false hope. As the
Cold War fades, however, so do old prejudices. President Clintons 1998
announcement of initiatives to improve home defense preparations
against biological attack reect the opening in this direction. Policy
should now build momentum to develop various measures for blunting
WMD effects: contingency plans for crisis relocation; simple broadcas-
table instructions on how to build hasty shelters; stockpiling of protective
masks, suits, decontamination equipment, chemical antidotes, and anti-
biotics; standby vaccination programs; and so forth.
Such measures are desirable, especially because they are cheap com-
pared to most programs in the defense budget. But they will only reduce
the problem, not solve it. Biological weapons pose the greatest challenge.
Because they will be easier to manufacture than nuclear weapons and
easier to use for mass killing, they could become a weapon of choice for
a state or group determined to inict mass casualties on the United States.
Defense against BW is possible, but least feasible in the most likely
situation. If the government knows where a BW attack will come, if it
knows exactly what agent will be used, and if it has sufcient advance
notice, it may be possible to produce and distribute appropriate vaccines
or antibiotics in the right communities; these are too many ifs to allow
chapter 3 79
much condence.
42
It is possible to deploy potentially effective protective
masks, and given their comparatively low cost this appears advisable.
43
But would such masks be effective against a surprise night attack? How
long would wearers have to keep the masks on to avert infection? Would
they be able to drink and eat? Our current inability to defend against the
common cold shows the limitations even under ideal conditions of intel-
ligence, and the use of recombinant DNA technologies to alter viruses
limits the possibilities of vaccination even against specic strains.
44
This is a reminder that the mass element in WMD is why such
weapons give an intrinsic advantage to the attacker. It is why they are
associated with apocalyptic consequences, rather than just lots of death,
as conventional warfare is. It is a reminder of why any who have such
weapons have a powerful deterrent, and thus why anyone who faces
themsuch as the United Statesshould be deterred from meddling
with them.
One way to avoid meddling is to retreat from political and strategic
involvement in the affairs of other countries and their enemies. Would
the World Trade Center have been attacked in 1993 had not the radicals
who perpetrated the crime seen the United States as responsible for the
frustration of their compatriots cause in the Middle East? To draw the
lesson that our best defense is retreat would be outrageous and dishon-
orable to many Americans. It would certainly repeal the dominant ethos
of righteous world power that the United States has cultivated since Pearl
Harbor. But as Waltz says, A big reason for Americas resistance to the
42. Some conclude that overall, vaccines are more useful to aggressors (for protecting
themselves against the agents they disseminate) than to defenders. For example,
Erhard Geissler notes, Immunological prophylaxis is highly specic. Protection is
provided only for a particular type of pathogen. He also notes that so many variations
of biological and toxin weapons are possible that the defender needs exact intelligence
in time to research, produce, and disseminate vaccines or antidotes (even assuming
that the particular biological threat in question is one that can be countered). Erhard
Geissler, A New Generation of Vaccines Against Biological and Toxin Weapons, in
Geissler, ed., Biological and Toxin Weapons Today, Stockholm International Peace Re-
search Institute (New York: Oxford University Press, 1986), pp. 6364.
43. See Karl Lowe, Graham Pearson, and Victor Utgoff, Potential Values of a Simple
BW Protective Mask, IDA P-3077 (Alexandria, Va.: Institute for Defense Analyses,
September 1995).
44. Harlee Strauss and Jonathan King, The Fallacy of Defensive Biological Weapon
Programmes, in Geissler, ed., Biological and Toxin Weapons Today, pp. 7071. The ca-
pacity to outrace defensive counters may be great in large WMD establishments
associated with hostile governments, but small in terrorist groups with limited re-
sources and repertoires. For the latter, some measures that could be circumvented in
principlesuch as vaccinations against standard anthrax strainsmight raise the
hurdle high enough to frustrate the effort.
80 the coming crisis
spread of nuclear weapons is that if weak countries have some they will
cramp our style. Militarily punishing small countries for behavior we
dislike would become much more perilous.
45
Has the United States been willing to take responsibility for righting
wrongs or preventing them in faraway places in part because it faced few
serious risks in doing so? In the future, which will be the greater risk:
failing to confront a country or group that is acquiring WMD and thus
allowing our vulnerability to grow, or confronting them, asking for a test?
Options if the Taboo is Broken
Theorizing about deterrence, arms control, and nuclear strategy was a
cottage industry for decades. The emphasis, however, is on cottage and
was. Those accustomed to bandying these ideas easily forget that they
were never on the tips of most peoples tongues. Moreover, the corps of
experts who promoted these terms of reference in the United States is
aging and dwindling.
Common sense and conventional wisdom about WMD remain ab-
stract, because neither nuclear nor biological weapons have been used
against populations since the end of World War II. The common sense
and conventional wisdom is also all very simpleand almost entirely
about nuclear weapons. The closest thing to wide agreement about these
matters in society at large is that these weapons are unusable. We may
need them, but for the purpose of ensuring that no one else uses them
against us. The ne-grained scenarios and schemes for tactical rst-use,
counterforce duels, or graduated escalation and limited nuclear war that
dominated defense policy debates inside the Washington beltway during
the Cold War never resonated outside. What is most impressive about
nuclear and biological weapons is that the unthinkability of their use
seems conrmed by the lengthening tradition of non-use. The taboo has
reinforced itself.
46
As time goes on, and the number of independent decision centers
capable of unleashing the arms grows, some weapons of mass destruction
are likely to be used somewhere. What then? Utopian realists say it will
be too bad, but the effects need not extend beyond whatever faraway
45. Waltz, in Sagan and Waltz, The Spread of Nuclear Weapons, p. 111.
46. In this sense the taboo might be thought of as constituting a regime as the term
has been used in international relations theory since the early 1980san informal rule
that takes on a life of its own as the tradition of observance and mutual interest in its
functioning accumulates. On the notion of a nuclear taboo, see Peter Gizewskis
forthcoming Columbia University dissertation.
chapter 3 81
backwater suffers the tragedy. It is hard to believe, however, that anyone
elsewhere will continue to view the issues and options as they did during
a half-century or more of non-use. If ten thousand or more people are
killed by a nuclear or biological attack, this is likely to alter attitudes
overnight. But which way?
Quite opposite reactions are imaginable. The shock might jar sluggish
statesmen into taking the danger seriously, cutting through diplomatic
and military red tape, and undertaking dramatic actions to push the genie
back in the bottle. Or the shock might prompt panic and a rush to stock
up on WMD, as the possibility of use underlines the need for deterrent
capability, or the effectiveness of such weapons as instruments of policy.
One seldom-noticed danger is that breakage of the taboo could de-
mystify the weapons and make them look more conventional than our
post-Hiroshima images of them. It helps to recall that in the 1930s,
popular images of conventional strategic bombing were that it would be
apocalyptic, bringing belligerent countries to their knees quickly. The
apocalyptic image was fed by the German bombing of Guernica, a com-
paratively small city in Spain. When World War II came in Europe, both
British and Germans initially refrained from bombing attacks on cities.
Once city bombing began and gathered steam, however, it proved to be
far less decisive than many had expected. British and German popula-
tions managed to adjust and absorb it. Over time, however, the ferocity
of Allied bombing of Germany and Japan did approach the apocalyptic
levels originally envisioned. In short, dire assumptions about the
awesomeness of strategic bombing deterred its initiation, but once initi-
ated did not prevent gradual escalation to the devastating level originally
envisioned.
Nuclear weapon inventories of countries like India and Pakistan are
likely to remain small in number and yield for some time. According to
press reports, by some U.S. estimates the yields of the 1998 tests were
only a few kilotons. If the rst weapon detonated in combat is a low-yield
device in a large city with uneven terrain and lots of reinforced concrete,
it might only destroy a small part of the city. A bomb that killed 10,000
to 20,000 people would be seen as a stunning catastrophe, but there are
now many parts of the world where that number would be less than
1 percent of a citys population. The disaster could seem surprisingly
limited, since in the popular imagination (underwritten by the results in
the small and imsy cities of Hiroshima and Nagasaki), nuclear weapons
mean one bomb, one city. Awful destruction that yet seems surprisingly
limited could prompt revisionist reactions among lay elites in some coun-
tries about the meaning of nuclear ordnance. (It is notable that reaction
82 the coming crisis
in the United States to the Aum Shinrikyo attacks in Japan was quite mild,
probably because casualties were surprisingly low.)
If potential results range from shoring up the dam to breaking it,
there is no clear guidance about the consequences of breaking the nuclear
taboo. What seems most improbable is that the willingness to rely on old
concepts of mutual deterrence and stability as solutions to living with
WMD would survive unshaken. If the rst use of WMD is against a
country not comparably armed, the main lesson may seem the validation
of deterrence; doubters or the previously complacent will be more likely
to bank on utopian realism and tank up on their own WMD to prevent
being victimized. (It is notable, though, that Japan did not react this way
to its own experience in August 1945 once it regained independence.) If
the rst use is against a country with comparable arms, however, it will
probably appear to discredit reliance on deterrence. In that case, it is hard
to believe that the rst lapse in the long nuclear peace will not trigger
stampedes in at least one of three directions: toward pacist foreign
policies, however out of character and costly they may be; toward more
energetic development of defensive measures, however limited their ef-
fectiveness may be; or toward brutal interventionist head-cracking by
great powers to get the genie back in the bottle, however risky such
activism could be.
47
Another wild card is uncertainty about whether the logic of deter-
rence concepts that developed almost entirely around nuclear weapons
will apply as well to biological and chemical weapons, neither of which
has been enshrouded in the same taboo or mystique as have nuclear
weapons. Historyin the form of plagues across the millennia or gas
warfare in this centuryhas given people reason to think of chemical and
biological weapons as more normal and more usable than nuclear arms.
47. Many would hope in addition for a fourth possibility, that the shock might make
feasible more effective multilateral regulation by the United Nations or some other
international organization and collective security mechanism. No one can prove this
is impossible, but there is no precedent by which to expect it. As Waltz says in
explaining why so many injunctions for states to subordinate their immediate interests
to the needs of the global ecosystem have gone unheeded: The problems are found
at the global level. Solutions to the problems continue to depend on national policies.
What are the conditions that would make nations more or less willing to obey the
injunctions . . . for the sake of the system? No one has shown how that can be done,
although many wring their hands and plead for rational behavior. The very problem,
however, is that rational behavior, given structural constraints, does not lead to the
wanted results. With each country constrained to take care of itself, no one can take
care of the system. . . . Necessities do not create possibilities. Theory of International
Politics, p. 109.
chapter 3 83
It is also easier for an attacker to conceal responsibility for an epidemic
caused by biological attack. If WMD proliferation takes the form of quests
for more easily obtainable or more anonymous CW and BW, how certain
can it be that the solid deterrence celebrated by utopian realists xated
on the Cold War model will be seen to apply to these other vehicles of
mass destruction?
Utopian realism should not convince us to promote the spread of
WMD, but it may hold out some hope if nonproliferation fails. At the
least, like the rest of the realist tradition, it calls attention to a dimension
of the problem that is woefully neglected by the most ardent opponents
of proliferation: the fact that weapons do not proliferate by accident, but
because countries or groups believe they need them. If they need weap-
ons for prestige, there may be little the great powers can do other than
threaten sanctions and hope that the price will strike the targets as greater
than the prestige is worth, or at the other extreme, attempt to buy them
off (for example, give a permanent UN Security Council seat to India).
If countries need the weapons for security, there may be more the
great powers can do. If nonproliferation is truly a vital interest for the
WMD haves, they should be willing to assuage the security incentives
of weaker states by providing security guarantees. It is hard to believe,
however, that they are more likely to do this after the Cold War than
during it. Today, nonproliferation must compete with incentives to re-
trench military commitments. For committed internationalists, and those
who want to keep the United States a global power for whatever reason,
nonproliferation is a vehicle for resisting isolationism. Robert Art, for
example, argues, Taking out additional insurance against nuclear weap-
ons spread, then, is the prime, indeed the only, security rationale for a
continuing military global role for the United States.
48
In a world where
the United States is the only superpower, however, few others will want
or dare to challenge its core security. The interest of foreigners in hurting
the United States is more likely to be driven by desire to deter the United
States from acting against their core interests. Thus, it is not as easy as it
once was to refute the anti-interventionist argument that the best way to
keep others from trying to hurt the United States is to avoid meddling in
their affairs. Leaving such governments or movements alone would often
comport with utopian realism, but it would negate the liberal impulse to
reform illiberal regimes.
For most of the nuclear era the priority that the United States placed
on nonproliferation was high in principle but low in practice. Washington
48. Robert J. Art, A Defensible Defense: Americas Grand Strategy After the Cold
War, International Security, Vol. 15, No. 4 (Spring 1991), p. 30 (emphasis added).
84 the coming crisis
was always willing to promote nonproliferation when it did not have to
short-change some other objective, but seldom did it prove willing to
sacrice other interests for the cause. The most obvious example was
Pakistan, where Washington applied the Symington Amendment and cut
the country off from U.S. aid in the late 1970s, until the Soviet Union
invaded Afghanistan and made Pakistan a necessary ally in another
causeat which point Washington turned around and gave even more
aid, including F-16 combat aircraft even more advanced than the models
it had earlier refused to sell. Then, when India and Pakistan tested in
1998, sanctions were imposed, but only because legislation allowed no
presidential waiver and contrary interests in relations with the two coun-
tries did not appear as compelling at the moment as during the Cold War.
New solutions for proliferation are not obvious: the problem is not
new, and thinking about solutions has gone on for a long time. What is
new is not the problem, but the amount of concern about it (which is
entirely by default, due to the disappearance of the Soviet threat). As the
essential problem remains the same, so too do the two main barriers to
solution. One is that as long as world politics revolves around sovereign
states, those states (or the groups that seek to create new states or
supplant old ones) will seek means to protect themselves and exert
power. The other is that countries have interests that compete with
nonproliferation for their diplomatic, military, and political capital. Until
either of those conditions change, we are likely to continue sliding, but
slowly, toward a world where more countries and groups have more
means to hurt each other.
For some of these actors, getting a strategic deterrent from WMD may
improve their security, and sometimes the spread may even avert con-
ventional wars that might otherwise occur in unstable regions. But for
traditionally pessimistic realists as well as liberals, what may sometimes
be good for weak and vulnerable states will not ultimately be good for
those whose strength and security can only be compromised by prolif-
eration. The United States should act as if the utopian realists are wrong,
but hope that they are right.
chapter 3 85
Chapter 4
The National Myth and
Strategic Personality of Iran:
A Counterproliferation
Perspective
Caroline F. Ziemke
The proliferation of nuclear weapons is undoubtedly the most complex
and pressing strategic challenge facing the United States and the interna-
tional community in the twenty-rst century. Many of those seeking to
acquire nuclear weapons are states with which the United States has
strained relations and major cultural differences, making the task of
forging effective counterproliferation approaches even more difcult.
This chapter focuses on one such statethe Islamic Republic of Iranto
demonstrate the value of national myth as a tool to understand and
predict state conduct.
There are several reasons for the focus on Iran. First, it tops most lists
of states seeking and likely to acquire nuclear weapons in the not-too-
distant future. Second, it is one of the states that the United States has
singled out in its national strategy as a likely regional troublemaker, one
of what former National Security Advisor Anthony Lake called backlash
states.
1
Third, opinions vary widely, especially between the United
States and many of its European allies, as to how great a danger Iran
poses to the international system and how best to cope with it. Finally,
Iran was formerly one of the United States most important regional allies
and is now a sworn enemy, suggesting that perhaps the 1979 Revolution
marked a dramatic change in Irans strategic personality. Not so. Through
a brief survey of Irans national myth, this chapter demonstrates the
continuities in Irans strategic personality that provide the best basis for
predicting its likely future conduct and national strategies. In the conclu-
sion, I suggest how decision-makers might use such insight into Irans
strategic personality to shape future counterproliferation approaches.
1. Anthony Lake, Confronting Backlash States, Foreign Affairs, Vol. 73, No. 2
(March/April 1994).
Policymakers and analysts thinking about strategies for responding
to the proliferation of nuclear weapons must begin by trying to answer
three basic questions. First, what might motivate any particular state to
devote substantial and often scarce national resources to developing or
acquiring a nuclear weapons capability? Second, where would nuclear
weapons t in that states national strategy, and how and under what
circumstances might it employ them? Third, how can the United States
and the international community persuade such a state not to acquire
nuclear weapons or, in the event that counterproliferation fails, deter their
use? Counterproliferation and deterrence are both essentially forms of
persuasionattempts to modify a states conduct using rational and,
preferably, noncoercive means. But effective persuasion requires some
understanding of how the target state makes its own strategic decisions
how it sees its relationship to the outside world, to what situations and
events its pays the most attention, and what standards it uses for assess-
ing circumstances and arriving at a decision as to how to proceed.
The problem with standard social science models for analyzing stra-
tegic decision-making in specic states is that, in measuring a states
conduct against a general theoretical model, they inevitably pluck events
out of their historical context. But a states conduct always follows from
a whole complex of motives, preferences, beliefs, prejudices, and ways of
thinking that have deep roots in history. As a result, the rationality that
informs the strategic conduct of a real, live, esh-and-blood state with a
past is unlikely to conform precisely to what a hypothetical scientic and
mathematical model would predict, and the policy community tends to
dismiss the states behavior as a willful attempt to aunt international
norms, or somehow irrational. But to draw such conclusions gets us no
closer to answering the important questions: Why would a particular
state want nuclear weapons? What might it do with them if it gets them?
And what can be done to prevent it from acquiring or using nuclear
capabilities?
Nations are the product of their unique historical experience in the
same way that individuals are, in a biological sense, the product of their
genetic background. A states historical experience shapes how it sees
itself, how it views the outside world, and how it makes its strategic
decisions. To make use of their historical experience, nations tend to focus
most on those aspects of their history that have the most meaning and
tell them the most about who they are and what they aspire to be. Those
aspects of a states historical experience are stored in its national myth:
the remembered history of how the nation came to be and what heroes,
demons, traumas, golden ages, and symbols of national identity are most
important in dening the boundaries between us and the others.
88 the coming crisis
National myth is not necessarily factual history; rather, it is a metaphori-
cal representation of a states history that a states people believe to be
true and use as their guide for social and political life. In the realm of
national myth, historical accuracy is less important than motivational
power. National myth presents an exaggerated view of the strengths,
virtues, triumphs, and traumas that make up a states collective self-
image and provides the blueprint for its strategic personality: how it sees
its relationship to the outside world, assesses its options and national
interests, and makes decisions about how to act.
From National Myth To Strategic Personality
the iranian case
This section uses national myth as a tool to decipher Irans rational model
and derive a broad strategic personality prole that provides insight into
how Iran perceives reality, makes judgments about its interests and goals,
and draws conclusions concerning how best to advance those interests.
Three aspects of national myth are of particular utility in assessing
Irans strategic personality: Irans remembered history (including the
history of the ancient Persian Empire), its civic philosophy (including the
civic aspects of Islamism), and its shared culture (including past and
present religious doctrines). The inuence of Shia myth is explicit in the
present Iranian theocracy. Still, the contemporary Iranian national myth
has much deeper roots in the pre-Islamic Zoroastrian and Persian secular
myths than are readily apparent. In this respect, the mythic tradition of
Iran differs fundamentally from that of its Arab neighbors. The Arabs
look to the High Caliphate of 750 to 1258 CE as the golden age of Islam
and the high water mark of Arab (and Islamic) cultural and political
power. But the golden age of Islam was a period of defeat and decline
for the Persian Empire, and an era of persecution for the Shia. Persias
glory days predated Islam, and while Islam is now an integral part of
Iranian culture, even the Shia clerics have not managed to purge the
inuence of the religious, cultural, and political traditions that came
before it.
zoroastrianism and pre-islamic myth
The dominant faith of pre-Islamic Persia was Zoroastrianism, a monothe-
istic faith founded on the teachings of the prophet Zarathustra in the sixth
century BCE and the ofcial religion of the Persian Empire until the rise
of the Safavid Dynasty in the fteenth century CE. Zoroastrian cosmology
centered on the celestial dual between Ahura Mazda (the Lord of Light)
and Angra-Mainyu (the Lord of Darkness) that would culminate in a nal
chapter 4 89
judgment day, when the dead would be resurrected and the wicked
banished to eternal suffering. Ahura Mazda would thereafter preside
over a puried universe and endless age of peace and prosperity on earth
in the land of Turan (Tehran).
2
Zoroastrianism was not a universalist faith and hence did not spread
much outside Persia, but it had a profound impact on the evolution of
later monotheism: the roots of the doctrine of redemptive suffering in
early Judaism, Christianity, and Shia Islam can be traced directly to the
inuence of Zoroastrian teaching.
3
This cosmology left an indelible mark
on the Persian worldview: the scale of evil in the outside world is
alarming and the surest route to safety is to hew to ones community, and
thus remain close to Ahura Mazda and the forces of good.
4
Iranian
Shiism retains elements of Zoroastrian beliefs and rituals that, in large
part, account for its theological distinctions from the Sunni Islamic main-
stream.
Pre-Islamic Persia also left important secular nationalist legacies. The
Shahnemeh is the mythic saga of the reign of Jamshid, who liberated Persia
from the rule of Zahhak, a Faust-like Arab prince who became king of
Persia through a pact with the Lord of Darkness. Zahhaks three-
hundred-year reign is a period of suffering and evil that ends when
Fariydun, a Moses-like gure, defeats the forces of evil to usher in ve
hundred years of peace and prosperity. Persia eventually collapses into
civil war when Fariyduns sons make alliances with central Asian war-
lords, allowing Persia, once again, to fall to the mercy of foreign interlop-
2. The Encyclopedia of Philosophy (New York: Macmillan, 1967), Vol. 8, p. 380, s.v.
Zoroastrianism; and Hans-Joachim Klimkeit, Gnosis on the Silk Road (San Francisco:
Harper, 1993), p. 202.
3. Joseph Campbell, Myths to Live By (New York: Penguin Arkana, 1993), pp. 182183.
4. The Puritanism of early settlers in New England instilled a similarly dualist vein
in the U.S. national myth that today is most commonly expressed as the shining city
on the hill. The Puritans fervently believed that their mission in the New World was
divinely decreed, that its achievement required strict discipline and single-minded
pursuit of denite ends (establishing the shining city on the hill), and that deviation
from those ends was tantamount to mortal sin. The Puritans, like the Shia, saw
themselves as the outpost of righteousness surrounded by a dangerous, heathen world
(state-sponsored Anglo-Catholicism in England; and the Native Americans in the New
World). The Puritan myth instilled a strict spiritual and communal disciplinenot
unlike that of modern Shiismas a means to keep the atomistic inclinations of
individuals under control and the community focused on its shared objective. Just as
the Iranians fear temptation from the materialism and individualism of the West, the
Puritans saw in Native American culturewith its emphasis on individual liberty and
freedoma potentially divisive inuence and a dire threat to their quest for righteous-
ness. See Richard Slotkin, Regeneration through Violence: The Mythology of the American
Frontier, 16001860 (New York: Harper Perennial, 1996), pp. 3637, 45.
90 the coming crisis
ers until Jamshid ends the civil war and reunies Persia. Again, the
xenophobic lesson of the myth is clear: Persian culture thrives when its
leaders use their power to keep it isolated and suffers when they pursue
alliances and power-sharing with the outside world. The historical basis
of the Shahnemeh is probably the reign of Cyrus the Great (550529 BCE),
who established the rst great Persian Empire. Cyrus is remembered as
a benevolent ruler and founder of the worlds rst superpower: a
Persian empire that at its peak extended from Anatolia to China. Persia
thereafter became the economic and cultural crossroads of the ancient
world, the center of the Silk Road that connected the Turkish and Arab
lands to the west with Central Asia and India to the east. The current
Iranian regime has gone to great pains to revive this aspect of Persias
legacy, steeping its efforts to develop economic ties with Central and
South Asia in the mythic image of the Silk Road.
Persian secular history, like the mythic Shahnemeh, is peppered with
court intrigues, coups, counter-coups, and bureaucratic power struggles.
Persian politics, then and now, is obsessed with conspiracies. The secular
national myth that has grown around the remembered history of the
Persian empire, along with the spiritual and cultural superiority that
follow from Zoroastrianism, have instilled the conviction that Irans pres-
ent weakness and vulnerability are unnatural and follow from Persian
attempts to compromise with the outside world. The pre-Islamic Persian
myth imprinted the Iranian national psyche with an us-versus-them
mentality and a preoccupation with conspiracy and externalized evil. In
interpreting events, Iran takes a comprehensively external and fatalist
view: failures are not the fault of any weakness on Irans part; therefore,
changes in behavior are unnecessary and, at any rate, would bring no
change in outcomes. The Shia myth, with its strong sense of martyrdom,
gave Irans fatalism a clear victim mentality. Once the Islamic revolution
radicalized Shia myth, the sense of victimization led to rhetoric that
looked to the outside world like a dangerous thirst for revenge.
shiism and iranian national myth
The importance of pre-Islamic myth notwithstanding, it is impossible to
separate the Iranian national character from Iranian Shiism, the polestar
of modern Iranian nationalism. If Islam is poorly understood in the West,
Shiism is an even greater enigma. This is ironic, since the theology and
cosmology of Shiism and Christianity share common ancestry and are in
many ways more similar to each other than either is to Sunni Islam. The
ecclesiastical structure of Shiism, like Christianity and unlike Sunni Is-
lam, consists of a formal and highly educated clerical hierarchy. The Shia
clergy also share with their Christian counterparts a tradition of abstract
chapter 4 91
theology and a broad exposure to comparative religion, something that
had little appeal for the more pragmatic Arabs. It was, in large part, the
erudition of the Shia clergy that ultimately made them a threat to the
established political and religious order in Iran and, later, the broader
regional order.
5
The Shia take very seriously the Koranic injunction that
Allah gave man not only eyes and ears to receive his word, but also the
power of judgment by which to discern the path of righteousness.
6
Islam probably rst came to Persia in the eleventh and twelfth cen-
turies CE, but it did not become the dominant faith until the rise of the
Safavid dynasty in the mid-sixteenth century. The Safavids were initially
followers of Susm, a school of Islamic mysticism that drew heavily on
local mythologies and cultural identity.
7
Because the messianic Sus
proved difcult to control, the early Safavid shahs saw in Shiism the
basis of a version of Islam that would promote social cohesion and build
the foundation of a Persian national identity, and hence enhance their
political power. Once they established Twelver Shiism as the ofcial
religion of Iran, the Safavids conducted a brutal campaign of persecution
and conversion against the Sunni, Su, and Zoroastrians but buffered the
social trauma by incorporating elements of the suppressed faiths.
8
The
Safavids further isolated Iran by encouraging Shia alternatives to Sunni
symbols of veneration: shrines at Mashad and Qom to rival those at
Mecca and Medinah, the pilgrimage to Karbala (the site of the martyrdom
of the Imam Husayn), and the establishment of an annual observance of
the life and death of Husayn. By the late seventeenth century, Shiism had
become a distinct and comprehensively Iranian alternative to the pre-
dominantly Arab Sunni Islam. The Shia clerics, however, proved less
malleable than the shahs had hoped and, over time, became increasingly
independent and critical of state authority. The clerics gradually with-
drew from their association with the decreasingly popular shahs and
became a parallel authority, often acting as the ringleaders of growing
opposition to the power and abuses of the secular leaders.
Two mythic archetypes clearly distinguish Iranian Shiism from main-
stream Sunni Islam: the martyrdom of the Imam Husayn, and the expec-
5. Olivier Roy, The Failure of Political Islam, trans. Carol Volk (Cambridge, Mass.:
Harvard University Press, 1994), p. 136.
6. Abdullah Yusuf Ali, The Meaning of the Holy Quran (Brentwood, Md: Amana
Corporation, 1992), n. 6138 to Surah 90: 612. This translation and edition is the source
for all Koranic references in this chapter.
7. Karen Armstrong, A History of God (New York: Alfred A. Knopf, 1994), p. 260.
8. Ira M. Lapidus, A History of Islamic Societies (New York: Cambridge University
Press, 1988), pp. 295296.
92 the coming crisis
tation of a messiah, the Mahdi. The martyrdom of the Imam Husaynthe
grandson of the Prophet Mohammedis the central archetype for Shia
devotion and piety. The myth of Husayn has its historical roots in the
second Arab civil war (682991 CE). The urbanized Umayyad Caliphate
sought to solidify its political and religious authority over the Islamic
world, but it met resistance from the rural, nomadic Muslim followers of
Ali Ibn Abi Talibthe cousin and son-in-law of the Prophet. Alis fol-
lowers believed that the caliphs had sacriced their spiritual responsibil-
ity in the quest for temporal power and were, hence, corrupt and illegiti-
mate rulers. Alis son, Husayn, was ambushed and murdered by the
forces of the Caliph Yazid at Karbala (Iraq) in 680 CE. The murder of
Husayn provides Shiism its origin myth: the struggle against the corrupt
followers of the assassin Yazid is the rst skirmish in the apocalyptic
struggle between goodness and corruption. The open display of mourn-
ing for Husayn is an essential element of Shia piety, and is ritually
observed in two public rites: the taziyah majalis, an annual passion play
reenacting the assassination, and the ziyarah, the pilgrimage to Karbala.
Husayn set the standard for Shia piety, idealism, nobility of character,
and aesthetic detachment from worldly concerns, but at the same time
he became a key political role model. Husayn died because he refused to
compromise the principles of Islam for mere political gain, and his mar-
tyrdom established a precedent: defending the purity of the faith and the
Muslim community, no matter what the cost, is the highest religious
responsibility.
Western culture and social and political philosophies are, in modern
Iranian national myth, the sort of corrupting inuences that Husayn died
to resist. To modern Iran, the United States and its Arab allies are the
contemporary embodiment of the caliphate; they seek to weaken and
persecute Iran and the Shia while holding out to the Iranian people the
promise of material comfort if they stray from the path of righteousness
and submit to the authority of corrupt, temporal powers. Two heroes of
modern Iranian nationalism personify Husayns example of incorrupti-
bilityMuhammed Mossadegh and the Ayatollah Khomeini.
The Husayn myth has not remained static, however. For centuries,
the mourning for Husayn focused Iranian national myth on the historical
injustice, persecution, and temporal defeat of the Shia faithful, contrib-
uting mightily to the Iranian tendency to externalize misfortune.
9
Modern
Islamist revolutionaries, led by the Ayatollah Khomeini, revitalized and
9. Haggay Ram and Galia Sabar-Friedman, The Political Signicance of Myth: The
Case of Iran and Kenya in a Comparative Perspective, Cultural Dynamics, Vol. 8,
No. 1 (1996), p. 55.
chapter 4 93
radicalized the Husayn myth into a call to action. Radical Islam no longer
dwelled on Husayns temporal sacrice in the interest of otherworldly
gain, but recast him as a righteous hero willing to sacrice his life in the
struggle against corruption, tyranny, and oppression in this world.
10
It
was the mobilization of myth par excellence.
Shiism segregates the archetypes of martyr and messiah. For the
Shia, the Twelfth Imam is the anticipated mahdi (messiah) whose coming
will herald Gods kingdom on earth. The Imams, descendants of the
Prophet through his son-in-law Ali, were the spiritual leaders of the early
Shia community and venerated as temporal representatives of the con-
tinuation of Gods covenant with Mohammed.
11
The death of the eleventh
Imam in 874 CE without a known heir precipitated the rise of messianic
Shiism.
12
The legend arose of the Twelfth Imamthe secret issue of his
predecessorwho (like Moses and Jesus before him) became the target
of a jealous ruler in his infancy. For his protection, and to maintain his
purity from taint by the corrupt caliphate, the Twelfth Imam was spirited
into celestial hiding, returning only occasionally for forty-day visitations
and otherwise ruling through his designated representatives. This lesser
occultation ended in 941 CE, when direct contact with the Twelfth Imam
was lost, ushering in the greater occultation, which the Shia believe
will end at the judgment day. The greater occultation is to be an era of
challenge and suffering for the faithful, the reward for which will be the
triumphant return of the Twelfth Imam as the mahdi to guide the faithful
in the establishment of Gods kingdom on earth. Although he never made
the claim himself, Khomeini was venerated by many Shia as the incar-
nation of the mahdi. Khomeini and his fellow ayatollahs have claimed
status as the mahdis designated representatives to this generation of
Shia; their pronouncements are considered divinely inspired and, hence,
infallible.
jihad and the great satan
The element of Islam that precipitates the greatest misunderstanding
between the Muslim and Christian worlds is the concept of jihad. In the
West, it is the most universally recognized and widely misunderstood of
all the tenets of Islam, in part because it has taken on a political promi-
nence all out of proportion to its proper place in the Islamic faith. Jihad
is a characterization of the struggle that the faithful face in an imperfect
10. Ibid., p. 59.
11. Armstrong, A History of God, pp. 162163.
12. Lapidus, History of Islamic Societies, pp. 118119.
94 the coming crisis
world. Muslim theology denes two levels of jihad. The lesser jihad is
the external struggle to defend and expand the territory and community
of Islam. The greater jihadthe primary responsibility of the pious Mus-
limis the spiritual warfare that takes place within the individual against
the powers of perversity that tempt him from the path of righteousness.
The greater jihad makes strict and constant demands on the faithful: that
they so x their eyes on God that worldly enterprises lose their impor-
tance; and that they be willing and eager to make earnest and sincere
sacrice of body, property, and if need be, life in Gods service.
13
As the
taxonomy implies, the individual act is more important than collective
action, and peaceful jihadwriting, preaching, charity, or setting an ex-
ample by living according to Islamis far more meritorious in Gods eyes
than participation in brutality and war.
14
The Muslim concept of jihad, in
this sense, differs little from similar notions of spiritual struggle in
Zoroastrian, Judaic, and Christian traditions.
The Iranian brand of Shia Islam, however, downplays the internal
spiritual struggle of the greater jihad and the existence of evil within its
community; it denies the existence of an intrinsic dark side within its
society. Instead, the Iranian-Shia national myth projects all the misfor-
tune, suffering, and evil in Iranian societypast, present, and future
onto others and gives its rhetoric of jihad a distinctly intimidating air. To
the Iranian Shia the jihad is less a spiritual struggle than a defensive
stand against relentless barrages from their enemies, human and super-
natural.
15
In earlier times, the projection of evil was aimed at the Arabs
13. Here again, the parallels with New England Puritanism are strong. Like the Shia,
the Puritans saw themselves as members of the family of God rst, and the family of
man second. They viewed earthly life as a way-station: a period of preparation for the
afterlife. Family life and social structures were a mechanism to become better suited
to Gods service, and bring them nearer to God. Death was a nal rite of passage
that frees believers from all earthly bonds, thus allowing them to return to the Father
in heaven. By this calculation, material comforts and vain individuality only served
to create sources of division and envy and distract both the individual and the
community from their focus on preparation for Gods kingdom. See John R. Gillis, A
World of Their Own Making: Myth, Ritual, and the Quest for Family Values (New York:
Basic Books, 1996), pp. 3132.
14. Ali, The Meaning of the Holy Quran, n. 1270 to Surah 9: 20, p. 444.
15. The need to externalize evil is a common element in puritan movements. In New
England Puritanism, the externalization of evil was most disturbingly manifested in
the witch hunts that peaked in Salem, Massachusetts, in 1692. Historians differ as to
the specic causes of the witch scares. Feminist historians have argued that the accused
were independent-minded individuals, usually women, who threatened their neigh-
bors sense of spiritual well-being by their refusal or inability to conform to strict social
norms. Another school argues that the witch hunts were the product of rising eco-
nomic communal tensions that, likewise, threatened the spiritual unity and well-being
chapter 4 95
and the Central Asian invaders. Today, it is aimed squarely and unblink-
ingly at the West in general and the United Statesthe Great Satanin
particular.
The Iranians mythic image of the United States as the Great Satan is
not without historical foundation. Irans history over the past ve dec-
ades is peppered with instances of U.S. intervention in its internal affairs.
The Central Intelligence Agency (CIA) masterminded the coup that over-
threw the brief revolution led by Mossadegh and restored the shah to the
Peacock throne in 1953. Through the mid-1970s, the United States sup-
ported the repressive Pahlavi regime, and the CIA worked closely with
the Shahs brutal internal security force, SAVAK. The Iranians believe the
United States exploited their countryincluding the promotion of a
costly Iranian military buildupsolely for the benet of its Cold War
strategy against the Soviet Union. The United States and its Western allies
supported Iraq in its war against Iran, remained silent concerning Iraqi
use of chemical weapons against Iran, and imposed a one-sided embargo
on Iran in an attempt to pressure it into ending a war that the Iranians
did not initiate. U.S. rhetoric regarding Iran remains consistently hostile,
and in the Iranian view, unjust.
Iranian demonization of the West is born, in part, from a deep sense
of insecurity. Western ideas about pluralism, democracy, and social equal-
ity appear to conservative Iranians to threaten the traditional social order.
Western consumerism bears the onus for undermining traditional values
by emphasizing material and worldly concerns over spiritual ones. Ira-
nian religious leaders see the United States as a particular threat as it
aggressively exports its popular culture, secularism, consumerism, and
individualism and seeks to create an American Islam among its secu-
larized Muslim client regimes. In the Islamist view, Iranian society can
avoid dilution of its faith only by a strict cultural quarantine.
16
The
Iranian Islamists are convinced that the Western cultural threat to Islam
of the communitytensions between the rural agrarian elements and the increasingly
capitalistic towns, between cities and rural landowners, and within and among con-
gregations and ministers. Again, clear parallels with Shiism emerge: strict gender roles
as the basis for community solidarity, the challenge of harmonizing economic growth
with spiritual purity, conict between the needs and ambitions of the individual and
the pursuit of communal goals, tensions between rural and urban values, and conicts
between various power centers within the ranks of clergy. For the feminist interpreta-
tion, see Carol F. Karlsen, The Devil in the Shape of a Woman: Witchcraft in Colonial New
England (New York: Vintage Books, 1987); for the socio-economic interpretation, see
Paul Boyer and Stephen Nissenbaum, Salem Possessed: The Social Origins of Witchcraft
(Cambridge, Mass.: Harvard University Press, 1974).
16. Graham E. Fuller, The Center of the Universe: The Geopolitics of Iran (Boulder,
Colo.: Westview Press, 1991), pp. 252253.
96 the coming crisis
is not passive but part of an aggressive and intentional program of
manipulation aimed at undermining Irans spiritual and political cohe-
sion. Western rhetoric about the Islamic threat and the clash of civili-
zations fuels this sense that the Judeo-Christian world fears the strength
of Islam and is determined to destroy it.
17
Irans anti-Westernism is about controlling its own fate; yet a recur-
ring pattern of Iranian history has been its inability to do so for more
than short intervals. This paradox leads to an Iranian preoccupation
with saving face that has a dual effect on Iranian national myth. First,
it helps explain Irans endless weaving of elaborate, and often implausi-
ble, conspiracy theories. Irans extreme sensitivity to shame amplies
its disinclination to internalize responsibility for its weaknesses and fail-
ures: misfortune always has an external agent. While maintaining the
conspiracy paranoia, the Islamic Revolution at least empowered the
Iranian masses by giving them a voice in their economic and political
future. But Irans steady economic deterioration, and the spreading
disillusion with a clerical hierarchy whose degree of corruption ap-
proaches that of the shahs elite, is leading to a renewed sense of power-
lessness, especially among young people who increasingly see even the
most basic goalscollege, careers, marriage, or familydrifting beyond
their reach.
Second, because of their keen sensitivity to face, Iranians put great
store in being inscrutable. In the past, they hid a determination never to
accept a subordinate role behind a facade of elaborate courtesy and
rhetorical hyperbole. Today, the extravagance of anti-Western revolution-
ary rhetoric serves a similarly defensive purpose. The political and relig-
ious leaders in Iran make frequent pronouncements that reveal almost
nothing about the substance of Iranian policies and ambitions. In both
internal and external affairs, dissembling has become a fundamental
survival technique in Iran, since information is power in the hands of
ones enemies.
18
The rigid social and political mores of postrevolutionary
Iran have done nothing to reverse the tradition of Iranian difdence; just
as in the past, todays young Iranians have become cynical and quietly
subversive, presenting a falsely pious front in public while privately
embracing fashions and mores in direct conict with the ideals of Is-
lamism.
19
17. Bernard Lewis, The Roots of Muslim Rage, Atlantic Monthly (September 1990).
18. Fuller, The Center of the Universe, pp. 11, 15, 19.
19. Geraldine Brooks, Teenage Indels Hanging Out: In High Tops and Jeans,
Iranian Youths are Quietly Subverting their Parents Revolution, New York Times
Magazine, April 30, 1995, p. 46.
chapter 4 97
the genesis of the iranian revolutionary myth
The Islamic revolution itself is an exercise in mythogenesis. The Islamic
Republic embodies the reconciliation of traditional Islamic behavioral
norms with the demands of the modern world. The architects of Iranian
Islamism do not seek to abandon modernism or return to the social order
of the seventh century. Nothing about Islamic theology is inherently
hostile to technology or modernization. But to Muslim societies experi-
encing the resultant social dislocation, the trappings of traditionthe
reorganization of society based on the sharia (the Islamic legal code), the
restoration of social virtue through dress and comportment, and the
condemnation (fatwa) of those who challenge traditional normscan
provide a sense of continuity and collective security. Even modern Iranian
Islamic dress codes are the stuff of national myth: with the exception
of the chador (the head to toe black veil that never completely disappeared
from rural and traditional Iran), hijab in Iran is a thoroughly modern
expression of Islamic standards of modesty, although symbolically remi-
niscent of more traditional dress.
20
Everyday life in postrevolutionary Iran mirrors the duality between
the desirable, as reected in revolutionary myth, and the actual revolu-
tionary reality. Consumerism in Iran has continued more or less unabated
despite the countrys deep economic troubles. Regular mosque atten-
dance has dropped dramatically since the revolution, and the calls to
prayer bring barely a hiccup in urban activity. There are periodic conser-
vative backlashes in Iran, particularly against the education system that
clerics charge is not sufciently Islamic, but the movement to revive strict
Islamic revolutionary principles has garnered little popular support. The
Islamic piety of the revolution was itself a mythic ritual of sorts. Revolu-
tionary puritanism also has deep Zoroastrian roots: the spiritual battle
between goodexemplied by the outwardly pious and inwardly spiri-
tual lifeand evilthe giving in to the baser, impure physical instincts
and worldly temptations.
While steeped in Islamic myth and norms, the events of 1979 repre-
sented rst and foremost a political revolution. Khomeinis revolutionary
role models were secular and, for the most part, Western. During the
revolution and since, revolutionary political goals have always taken
precedence over religious goals. The centrality of the sharia in Iranian
law is established in the constitution; thus, it is the secular law that
establishes the legitimacy of religious law, and all new legislation must
conform to both Islamic and constitutional standards. Moreover, Iranian
law contains many non-Islamic concepts: legal (if not yet actual) equality
20. Roy, The Failure of Political Islam, pp. 56, 59, 181182.
98 the coming crisis
between the sexes concerning property, employment, and family rights;
the lack of legal religious discrimination; citizenship requirements that
are in no way tied to religious afliation and give no preference to
Muslim over non-Muslim aliens; and the retention of the Persian solar,
rather than the Muslim lunar, calendar.
21
the strategic personality of iran
Irans national myth dwells on two factors that lie at the core of its
strategic personality: its cultural and religious singularity, and its long
history of invasion and manipulation by outsiders it sees as inferior. This
has rendered Iran a society that is introverted to the point of paranoia,
one that holds fast to its cultural traditions, even as it longs to enjoy the
benets of modern technology. Iran is also a culture that has a long
memory and holds grudges: the span of history that directly informs its
decisions and behaviors ranges as far back as the ancient Persian Empire,
and as far forward as Judgment Day. In short, Irans model of time is
much closer to the holistic Eastern cyclical concepts than to the compart-
mented, linear time of the West. Social and organizational structures in
Iran are nebulous, generally consisting of complex webs of primary
groups and conicting centers of power (clergy vs. technocrats, bazaaris
vs. industrialists, professional military vs. revolutionary guards, and nu-
merous clerical and political subgroups). Communication among Iranians
and between Iran and the outside world is heavily laced with metaphor,
symbolism, and hyperbole; there are meanings behind meanings. Iran is
a value-oriented culture, and sees its value structure as uniquely merito-
rious and constantly under siege from the outside. The Iranians are
idealists, even visionaries, and believe their society is perfectible. In fact,
the Iranians are so condent of their moral superiority that they feel
compelled to attribute any shortfalls to outside interference. As a society,
Iranians are only secondarily interested in their material success; Irans
cultural artifacts reect this antimaterialism: its cultural expression tends
toward the etherealpoetry, literature, theologyrather than the con-
crete visual arts, architecture, infrastructure, and technology.
Iranian Foreign Policy and National Security Strategy through
the Prism of National Myth
The central theme of Irans mythic self-image and the key to its strategic
personality is its sense of cultural and moral superioritya superiority
that the outside world has most often failed to honor. Since the earliest
21. Ibid., p. 178.
chapter 4 99
invasions of Persia by Huns, Greeks, and Arabs through the nineteenth-
and twentieth-century pressures from Ottoman and then European
imperialism, Persian sensibilities have been outraged by the abhorrent
spectacle of indels manipulating the fortunes of true believers and
undermining the interests of Persia. It is a humiliation that still grieves
the Iranian psyche and drives its participation in international relations.
Persia has always labored under a certain nationalist narcissism,
imagining that it is the focus of international intrigue, jealousy, and
calculation.
22
Zoroastrianism and Shiism, both fairly insular faiths,
reinforced Irans cultural arrogance with a sense of spiritual and
moral ascendancy over its Arab neighbors. Cultural arrogance toward
its Arab and Turkish neighbors was initially defensive, preventing
political and military domination from translating into foreign cultural
assimilation. After World War II, Iran pursued a foreign policy
that seemed designed to alienate its neighbors and cultivate its iconoclas-
tic image. The shahs regime was openly hostile toward Arab national-
ism, established cordial relations with Israel, and forged a close mili-
tary and economic relationship with the United States. Even the
Islamic revolution put a premium on the distinctions between Iran
and the Arab world: Shiism versus Sunnism, Islamist Iran versus the
secular Arabs, champions of the victims of imperialism versus the
puppets of the West. The stark dualism of the Zoroastrian and Shia
myths are thus clearly reected in foreign policy, giving Iranian diplo-
macy its stark zero-sum approach: We are good, you are bad. We win,
you lose.
Persias once great empire and culture has, over the course of modern
history, been gradually ennervated and humiliated by foreign encroach-
ment and domination, and cultural imperialism. As a result, one of the
most important aims of Iranian foreign policy since the reign of Reza
Shah prior to World War II has been to restore Iran to its rightful place
as a great and feared power in the Persian Gulf and the world. All
developing nations want the West to treat them with respect, even if they
are realistic enough not to expect to be treated as equals. To Iran, however,
which sees itself as the heir to a once glorious and feared empire, to be
treated as a second- or even third-tier power is grievous. The current
mythic role of the United States and the Western powers as the Great
Satan follows from the perception that it was the arrival of the Europe-
ans, their culture, and their ideas that undermined and eventually de-
stroyed Irans autonomy and dignity.
22. Fuller, The Center of the Universe, p. 255.
100 the coming crisis
irans regional strategy
The Persians throughout history have believed themselves surrounded
by expansionist aggressors. That historical paranoia has been enhanced
by recent history: Western dependence on access to oil, Irans strategic
importance in U.S. strategy during the Cold War, its ability to inuence
trafc in and out of the Persian Gulf at the Straits of Hormuz, and the
success with which the Islamic revolution disoriented Western policies
have all served to intensify Western, and particularly U.S., concerns over
Irans hegemonic ambitions. Irans present quest for hegemony is driven
by the desire to force Western power out of the region and establish a
more natural regional balance that extends Iranian inuence but not
Iranian territory. The fundamental difference in the security paradigms
of Iran and its Gulf Cooperation Council (GCC) neighbors is this: Iran
does not believe regional stability and security are possible so long as the
United States retains a military presence in the Persian Gulf; the GCC is
founded on the principle that regional security and stability are impossi-
ble without a sustained U.S. presence.
23
There are few points of contention between Iran and its neighbors
that are likely to result in open conict in the near term. Although the
territorial dispute with the United Arab Emirates (UAE) periodically
takes on a more confrontational air, none of the parties to the conict
Iran, the GCC, or the UAEseem likely to press the issue to the point of
a military conict. Irans pursuit of a blue-water navy (an ambition
inherited from the Pahlavi regime) might also prove destabilizing. While
Iran clearly seeks the ability to block ships entering the Persian Gulf, its
interest in keeping the sea lines-of-communication open there is at least
as great as that of the United States, and recently both sides have taken
care to avoid potential naval confrontations. The Persian Gulf War elimi-
nated Iraq as a potential aggressor for the time being; but legitimate
concern lingers in Iran about the long-term possibility of a resurrected
Iraqi military machine. There are still outstanding differences between
the two states, and as a large, potentially wealthy, and populous state,
Iraq is the most likely challenger to Irans security in the Gulf Region.
Since the 1970s, Iran and Saudi Arabia have engaged in an on-again,
off-again Islamic Cold War over the future conguration of Persian Gulf
security. In the past few years, the relationship has warmed as the Iranian
regime has backed away from its declared intention to export its revolu-
23. Sharam Chubin and Charles Tripp, IranSaudi Arabia Relations and Regional Order,
Adelphi Paper 304 (London: International Institute for Strategic Studies, November
1996), p. 28.
chapter 4 101
tion. Iran probably also senses that if the Arab-Israeli peace process bears
fruit, it may suddenly nd itself the new nemesis for Arabs in search of
someone to replace the Zionists as the source of all misfortune in the Arab
world. The peace process may contribute to a further sense of insecurity
in Iran in another way. The West has made arms sales part of the currency
that moves the peace process forward, building the military capabilities
of states potentially hostile to Iran to spur their cooperation. The strategy
may indeed contribute to Arab-Israeli stability by closing some of the
capabilities gaps, but the United States should recognize that it is creating
a source of anxiety for Iran.
In light of its dual aspiration for regional leadership and territorial
and cultural security, Iran seems to be pursuing a two-track regional
foreign policy. The Iranian regime has been making some conciliatory
gestures toward mainstream Arab statesespecially Saudi Arabia and
Egyptin the hopes of drawing them away from their relatively close
security ties to the United States. Iran has also hinted at the possibility
that it might forge a revisionist axis of Iran, Syria, and Iraq to balance
the increasingly daunting U.S.-Saudi alliance. Syria has been Irans only
real ally in the region; the two regimes have in the past shared a bitter
enmity with Iraq and, more recently, have moved closer in their position
concerning the Arab-Israeli peace process. More surprising has been
Irans improved relationship with Iraq, once a mortal enemy, and the
apparent rise of a mutual sense of shared Iranian-Iraqi interests. Rhetori-
cal darts still y, and Iran continues to support Shia rebellion inside Iraq;
but Iran has also been outspoken in its criticism of continued UN sanc-
tions and has been an important source of black market trade for Iraq.
Talk of a future alliance with Iraq is likely little more than Iranian pos-
turing, though. Iran has the most to lose from a full revitalization of Iraqs
economic and military power and has no powerful allies to shore up its
defensive position; moreover, its sole regional ally, Syria, is unlikely to
share Irans interest in rapprochement with Saddam Husseins Iraq.
The tenor of Irans national myth and the pull of history seem likely
to continue to undermine Irans relations with its Arab neighbors. The
long history of Arab-Iranian enmity carries much weight in cultures with
long memories, and the Arab states will remain suspicious of Irans
intentions in its recent attempts to improve relations. Recent Iranian
behavior in the Persian Gulfa military build-up coupled with an in-
creased tempo of naval exercises in the Gulf, the recent dispute with the
UAE over the mid-Gulf islands, and alleged Iranian connections to the
anti-U.S. terrorist attacks in Saudi Arabiabelies Irans statements of
peaceful and conciliatory intentions. Far from moving the Arab states
toward warmer relations, Irans actions have led to increases in defense
102 the coming crisis
spending among the GCC states and increased reliance on Western secu-
rity guarantees, and they have virtually insured that few of Irans neigh-
bors will join it in resisting the emerging U.S.-Saudi military dominance
of the Persian Gulf region. Irans relative military position has deterio-
rated since 1979, even given the crippling defeat Iraq suffered in the 1991
Persian Gulf War.
24
Irans mobilization of its myth of victimization and Western conspir-
acy holds little water with its Arab neighbors. Most Arab states agree (at
least privately) that Iran and Iraqnot the Gulf Arabs or the United
Statesare responsible for the increased Western military presence
through their long streak of reckless behavior: Irans initiation of the 1988
tanker war in the Gulf; Iraqs invasion of Kuwait and its subsequent
saber-rattling in late 1994; and Irans naval arms build-up, militarization
of the Tumbs islands and Abu-Musa, and acceleration of its nuclear
program. Before it can win even cautious concessions from the rest of the
Gulf states, Iran will have to revise one of the central themes of the
Persian national myth: its natural claim to hegemony in South Asia and
the Persian Gulf based on its moral and cultural superiority. Even then,
the Arabs are likely to proceed only with great caution. They will not
repeat their 1990 mistake of waiting for open aggression to appeal for
outside help against an aggressively expansionist regime (in that case,
Iraq). They are willing to ally themselves with the devil if necessary, to
ensure security until Iran demonstrates a genuine desire for peaceful
relations with its neighbors.
25
One issue, more than any other, stands in the way of improved
relations between Iran and its Arab neighbors: the declared intention of
Irans Islamist regime to export its revolution throughout the Muslim
world, and its particular targeting of conservative, traditional Sunni re-
gimes in the Persian Gulf region. The export of Islamic revolution is
enshrined in the Constitution of Irans Islamic Republic, but since at least
1985, the regime has gradually disengaged from its once robust program
of subversion and violence in favor of building a positive image of Iran
as a role model of the prosperous Islamic state.
26
Irans efforts to export
Islamist revolution provides two key elements of its strategy to force the
24. Dov Waxman, The Islamic Republic of Iran: Between Revolution and Realpolitik,
Conict Studies 308 (Leamington Spa, U.K.: Research Institute for the Study of Conict
and Terrorism, April 1998), pp. 1718.
25. Iran Forcing the Arabs into the Wests Lap, Mideast Mirror, March 17, 1995, and
April 1, 1995.
26. Sandra Mackey, The Iranians: Persia, Islam, and the Soul of a Nation (New York:
Plume, 1998), p. 345.
chapter 4 103
West to fear and respect Iran, and by implication, the rest of the Muslim
world. First, Iran paints itself both domestically and internationally as the
modern-day heir of the great Muslim (albeit Kurdish) general Saladin
who will lead Islam in wresting control of the rightful territory of Islam
from the domination of todays crusaders: the Zionists, the secularists.
Second, Iran stands to gain what it sees as important moral and psycho-
logical victories over the United States and its regional enemies by fos-
tering domestic unrest and potential political upheaval in Arab states like
Egypt and Saudi Arabia, whose regimes have aligned themselves with
the West. In so doing, it hopes also to force other Gulf states to think
twice before aligning themselves with the U.S.-Saudi security axis.
Irans support of Islamist movements is the centerpiece of its effort
to restore something of its ancient glory; it also has roots in another aspect
of Irans national mythits concept of and quest for Islamic justice. Irans
Islamic interventionism is part of a more general revisionist foreign policy
that seeks to overthrow the Western-dominated world system, at least as
it functions in the Persian Gulf and Middle East, and replace it with a
system that is more egalitarianat least for Iran. The establishment of
national systems based on Islamic principles is essential to Irans view of
a desirable future world order, because Muslim societies can only be
treated with the dignity and justice they deserve within an Islamic con-
text.
Despite the obvious mythic status that came with being the rst
successful Islamist revolutionary state, Iran has met with limited success
in extending its inuence beyond what one analyst called its self-
imposed Shia ghetto.
27
Even within Shia Islam, Irans quest for para-
mountcy has met with frustration. The Shia outside Iran, it seems, are a
difcult lot to manipulate: they are either loyal citizens of their home
countries; or they are members of Islamist resistance movements that are
ercely independent. Shia clergy outside Iran have not undergone the
radicalization of Shia myth that, in Iran, transformed the Imam Hussein
into a prototypical Islamic revolutionary.
28
Of late, non-Iranian Shia
who number at least 150 million compared to Irans 60 millionhave
come to resent Iranian efforts to dominate the clerical elite; local Shia
populations in India, Pakistan, and Central Asia have challenged the right
of the Iranian clergy to choose the next marja (the spiritual leader of the
worlds Shia faithful) and have chosen their own indigenous leaders.
29
27. Roy, The Failure of Political Islam, p. 184.
28. Ram and Sabar-Friedman, The Political Signicance of Myth, pp. 6063.
29. Iran after Araki: Things will never be the same, Mideast Mirror, December 1,
1994, p. 21.
104 the coming crisis
Iran has recently attempted to expand its inuence by offering support
to Sunni-dominated Islamist movements in the Arab world and Farsi-
speaking minorities in the Central Asian republics of the former Soviet
Union. These efforts have been only moderately successful because the
Shia inuence is limited with Sunni Arabs, and ethnic and religious ties
with the more secular Central Asian Muslims are tenuous.
Still, while Irans inuence may be limited, it is concentrated in areas
of signicant strategic interest in the West as well as the Arab world. As
of 1988, 57 percent of the population in Iraq, 60 percent in Bahrain, 32
percent in Lebanon, 15 percent in Kuwait, and 12 percent in Syria were
Shia.
30
Iran has made some inroads in areas where the Shia minorities
have been radicalized and where politicized clergy lead resistance move-
ments, particularly in the strategically vital but politically vulnerable Gulf
monarchies like Bahrain and Saudi Arabia.
31
Nor is Irans revolutionary
Islamist myth without appeal outside Shia Islam. Iran established ties
with Islamic resistance movements in Afghanistan and Pakistan, pro-
vided important moral and material support to Palestinian Islamists
opposing the Palestinian-Israeli peace process, has provided nancial
support for radical Islamists in Saudi Arabia, and has been active in
supporting and training Bosnian Muslim forces in the former Yugoslavia.
The Iranian presence in Bosnia proved a major concern for NATO Imple-
mentation Force (IFOR) peacekeepers implementing the 1995 Dayton
Accords. Iran has also succeeded in exercising a major inuence on the
course of events in Lebanon through its support of the Shia Hezbollah
(Army of God) guerrilla forces and may also be playing a role in the
course of the Syrian-Israeli bilateral talks.
Despite Irans status as a role model and its short-term, episodic
success in inuencing Islamist movements, at least three factors argue
against Irans long-term success in its campaign for unchallenged domi-
nance of global Islamist politics. First, while most Islamist opposition
movements are not nationalist in the strict sense, the character of each
follows directly from its own unique national culture and myth, and each
pursues an agenda based on indigenous frustrations and aspirations.
30. Sharam Chubin, Iran and Regional Security in the Persian Gulf, Survival, Vol.
34, No. 3 (Autumn 1992), p. 68.
31. The Saudi government does not release public data on the percentage of Shia
within the Kingdom, as Sunni Islam is the ofcial religion of all Saudi citizens.
However, the Shia, while not a signicant percentage of the Saudi population as a
whole, are known to be either a majority or sizable plurality in small pockets along
the Persian Gulf and Red Sea coasts. In addition, some Shia live in Saudi Arabia as
guest workers (including some from Pakistan), although the number has likely de-
clined since the 1991 Persian Gulf War.
chapter 4 105
Second, most Islamist movements are Sunni. They look for their revolu-
tionary example to the Sunni Islamic Brotherhood rather than the Iranian
revolution, and are unlikely to abandon the traditional Sunni denigration
of Shiism to subordinate themselves to Iran. Islamists will continue to
welcome Irans material support but are likely to balk at attempts at
ideological and political manipulation by Tehran. It is unlikely, in short,
that Iran will become the great puppet master of the Islamist movement.
Finally, the fundamental issues that create tension between Iran and its
neighbors are essentially strategic rather than ideological in origin; these
differences, for the most part, predate the Iranian revolution and would
likely continue even if more Islamically correct regimes replaced the
Gulf monarchies.
32
The fact that Irans myth lies especially close to the surface in its
external relations does not rule out a certain expediency: the Iranians are
nationalists rst, coreligionists second. Iran has already demonstrated its
willingness to set aside its ideological agenda given the right incentives
and constrains. In fact, it has used its support of Islamist uprisings as
a bargaining chip in international relations. Iran has carefully avoided
the appearance of supporting Islamists in former Soviet Central Asia
and Chechnya, and has played the role of peace-broker in the Nagorno-
Karabakh conict, all in the interest of expanding Russian-Iranian rela-
tions, which both seek for economic and strategic reasons. Iran may
eventually have to back further away from its Islamist interventionism
for three other reasons. First, if Irans attempts to improve relations with
its regional neighbors are part of a serious effort at rapprochement, then
abandoning its support of disruptive extremists in the Gulf Arab states
seems a prerequisite. Second, Irans conciliatory overtures to the Mubarak
regime indicate a concession that Islamist revolution is unlikely to over-
throw Egypts secular regime. Islamist movements in Bahrain and Saudi
Arabia face similar frustration, although they are likely to remain active
and may force some response from the existing regimes.
The third and ultimate motive for curtailing its Islamist offensive may
be internal. Iran is a multiethnic nation, tied together by religion rather
than language or homogeneous culture. It has been a remarkably stable
entity over the centuries with comparatively little ethnic unrest, but the
presence of large ethnic minorities (Azeris, Kurds, Armenians, Baluchis,
Turkmens, Arabs, to name a few) along borders with potentially hostile
or at least unstable neighbors could prove a source of future problems,
in combination with Irans deteriorating economy and declining standard
of living. Domestic unrest is likely to spread as the Iranian population
32. Chubin, Iran and Regional Security, p. 53.
106 the coming crisis
grows impatient with the slow pace of economic and political reform and
as the nascent conict between Islamic conservatives and more progres-
sive elements heats up. The frustrations of a highly educated but under-
employed middle class and the disillusioned, religiously zealous veter-
ans; the pressures of the fast (although decelerating) rate of population
growth; and the destabilization of urban populations by an inux of
undereducated rural peasants all add fuel to an already incendiary social
situation. The Iranian regime may be beginning to suspect that meddling
in the internal affairs of other statesat least those with the means to
retaliate in kindis not a regional norm that Iran would like to see
spread. A substantial expatriate resistance movement has made no secret
of its ambition to return to Iran, and its government in exile, while
unpopular inside Iran, has gained allies in Washington, raising the spec-
ter of an Iranian Bay of Pigs. In short, Iran may nd itself the occupant
of a glass house: interference in the domestic affairs of its neighbors could
boomerang on the Islamist regime with disastrous effect.
33
Persian tradition leaves Iran with a very different calculus of the
balance of ends and means in international security than that of the
Western world. Iranians have never been pragmatic realists in their ap-
proach to international relations. Iran, like the United States, is partial to
crusades: both tend to imbue foreign policy and conicts with the moral
baggage of the conict between good and evil. Long before the revolu-
tion, the shah enthusiastically signed on in a partnership with the United
States and Israel in the Holy War against communism. Irans current
Islamist leadership puts great store in moral and psychological victories
and can view what might, to Western eyes, seem a setback as a net victory.
Iran takes great pride in what it sees as two great victories over the
United States: the seizure of the U.S. Embassy in Tehran in 1979 and the
subsequent defeat of President Jimmy Carter in 1980, and the Iran-Contra
affair, which seriously undermined a major Reagan administration for-
eign policy initiative. In both cases, the Iranians believe (not without
cause) that their actions triggered major political crises in the United
States and undermined both its self-condence and its international im-
age. Unlike Saddam Hussein, Iranian leaders understand that they cannot
hope to challenge U.S. military power directly, and have thus opted for
more indirect challenges to U.S. power: psychological and moral offen-
33. Roughly 60 percent of the Iranian population is too young to remember the 1979
revolution; hence, they have little memory of the bad old days under the Shah, or
of the hard times of the early revolutionary years. They compare their lot to the outside
world, and, whether radical Islamists or moderates, nd the present theocracy sclerotic
and wrong-headed. See The Mullahs Balance Sheet, Economist, January 18, 1997,
Survey Iran insert, p. 3.
chapter 4 107
sives rather than head-on military confrontations. Iran can be expected
to continue its efforts to seize any opportunity to make the United States
look weak, to undermine U.S. credibility as a friend of the Arabs, and to
exploit rifts between the United States and its allies (as it has over
economic embargoes and, more recently, punitive air strikes against Iraq).
irans military modernization
A key theme in Iranian foreign policy is its ambition to restore Iran to a
position of regional, if not global, respect; it is logical to assume that its
principle objective in national security strategy should be the construction
of a military capability consonant with such respect. In this regard, the
Islamic Republic has remained consistent with its predecessors. Like the
shah, who wanted to build Iran into a self-reliant power that did not
depend on anyone else for its security and which could, if need be, strike
fear in the hearts of potential aggressors or challengers, the Islamic
Republic seeks to establish a sort of prickly autarchy that would inspire
outsiders to leave it alone but would allow Iran to venture out and
engage with the outside world at times and by means of its choosing.
The shah exploited Irans immense oil wealth and close ties to the West
especially the United Statesto build, among other things, a modern,
professionalized army and air force that could impose Irans (and the
Wests) will throughout the Persian Gulf region. Revolutionary Iran will
likewise strive to build a military capability that strikes fear and respect
into the hearts of its potential enemies, but in a way suited to its goals
and social and political capabilities. The desire to regain something of its
former prestige and to do so by becoming an alternative to rather than a
clone of the West provides much of the motivation for the modernization
of Irans military establishment.
Irans anti-Western, revolutionary national myth provides two impor-
tant hints as to the nature of future military modernization. First, Iranian
Islamists have been highly critical of the extent to which Middle Eastern
regimesrst the shah, now the Gulf Arabsmortgaged the economic
futures of their societies to acquire advanced Western military technolo-
gies. This, the Iranians believe, keeps them dependent on the West for
training and maintenance (as Iran learned the hard way when its 1970-era
air force deteriorated for lack of spare parts and know-how). Second, the
Iranian Islamists have an iconoclastic view of how best to withstand the
economic and military prowess of the West. Irans view of itself and its
relationships to the outside world points to a strategy aimed at deter-
rence; like the porcupine, Iran may not need to win but merely to inict
sufcient pain to cause its adversaries to back away and keep their
distance. Iran will modernize its military capabilities according to its own
108 the coming crisis
unique vision of its strategic environment. The challenge for the United
States and Irans neighbors will be to anticipate what that model will be,
how it might shape the direction of future Iranian military strategies and
capabilities, and what might be done to modify or respond to Irans
strategic vision. The Iranians will have learned from the Persian Gulf
Crisis of 199091 and are likely to strive to avoid Saddam Husseins mis-
steps. The most important of these lessons might be the risk inherent in
constructing a large-scale Western-style military establishment. Most likely,
Iran will attempt to build capabilities that can respond to the kind of con-
ventional aggression that might come from a resurgent Iraq and defend
Irans interests in the Persian Gulf without triggering a massive Western
military intervention. At the same time, the perception that the West is
unable to eliminate Iraq as a nuclear or chemical threat might lead Iran
to step up its own efforts to develop defensive capabilities in those areas.
An understanding of Iranian myth also suggests that the evolution
of a formal, hierarchical military establishment on the European model
is highly unlikely. The character of Persian social and political interac-
tiondifdence, dissembling, and conspiracy-mindednesshave per-
petually undermined its political and military effectiveness.
34
Irans
decision-making structure reects its decentralized social and political
structure: there is no dominant voice in international security policy,
competing power centers often work at cross-purposes, and decisions
usually reect unsatisfying and inefcient compromises.
35
Moreover, the
mullahs who dominate the current regime would be unlikely to welcome
the competition for power that would follow the founding of a robust
military establishment. Political purges since the 1979 revolution have
further undermined the professionalism of the Iranian military estab-
lishment. Irans President Muhammed Khatami, elected in 1997, is likely
to continue reforms initiated under Hashemi Rafsanjanis pragmatist
regime to moderate the emphasis on Islamic correctness over technical
expertise in the Iranian military, as well as in industry, transportation, and
energy production. The pragmatists increased efforts to liberalize society,
professionalize the military, and mend fences with Iranian expatriates
stems, in part, from the realization that Iran has paid a steep price for the
politically driven depletion of its technical and professional classes.
36
Nonetheless, the creation of a large professional militaryeven along the
lines of the pseudo-professionalism of the Iraqi Armyis unlikely.
34. Fuller, The Center of the Universe, pp. 11, 1519.
35. Chubin and Tripp, Iran-Saudi Arabia Relations and Regional Order, p. 51.
36. Iran Seeks to Deter Israeli Threat to Nuclear Facilities, Mideast Mirror, January
10, 1995, pp. 1011.
chapter 4 109
The political and cultural limits on Irans conventional military capa-
bilities provide an important motive to acquire new technologies. Iran
began acquiring ballistic missiles in 1985, when it became obvious that it
was incapable of maintaining and operating an effective air force.
37
Irans
human wave tactics, which relied on mass armies of untrained but
religiously zealous soldiers, were effective in defending against Iraqs
more advanced military technology. Such mass armies are much less
effective offensive forces, but they t well into Irans porcupine strategy
as a disincentive to a conventional attack. An arsenal of nuclear, biologi-
cal, and chemical weapons would complement Irans conventional strat-
egy, providing an additional layer of deterrence in the event that the
human-wave defense failed or Irans human and material resources were
depleted. The nature of Irans foreign policy and national security objec-
tives and its perception of its role in the international community also
leave it highly motivated to acquire both nuclear and biological or chemi-
cal weapons and their delivery systems.
Irans military modernization and nuclear programs serve an internal
political purpose as well. While the general trend toward moderation is
likely to continue, Khatami and the pragmatists still face powerful oppo-
sition from more conservative quarters: hard-line Islamists are angry over
what they regard as the secularization of Iranian society and the aban-
donment of revolutionary principles; moderates are frustrated and disil-
lusioned by the snails pace of democratization and modernization; and
everyone is alarmed at the deterioration of the Iranian economy and the
governments apparent inability to turn things around. Military modern-
ization, especially the acquisition of a nuclear capability, continues to
enjoy great political popularity and is widely perceived as a source of
international respect and prestige. In fact, the military modernization is
the greatest political success of a regime that has seen precious few in
recent years. Irans leadership also remembers the psychological impact
that the mere threat of Iraqi chemical attacks against civilian populations
had on Iranian morale in the nal stages of the Iran-Iraq War, contributing
to the Khomeini regimes decision to accept a negotiated peace in 1988.
However, the ultimate success of Irans quest for a credible nuclear
capability is uncertain, as is its commitment to its nuclear weapons
program. Iran is clearly determined to develop or purchase nuclear ca-
pabilities. Iran currently spends roughly $2 billion a year on nuclear
technology from Russia, China, Pakistan, North Korea, and the former
37. Seth Carus and Janne Nolan, Arms Control and the Proliferation of Ballistic
Missiles, in Alan Platt, ed., Arms Control and Condence-Building in the Middle East
(Washington, D.C.: United States Institute of Peace Press, 1992), p. 68.
110 the coming crisis
Soviet republics. Given Irans economic crisis, however, it is far from
certain that it can maintain this pace of spending, especially given the
resources it has recently devoted to conventional modernization. Irans
nuclear program is also vulnerable to political setbacks, should domestic
economic realities force the diversion of resources away from military
spendingthough U.S. and European expressions of alarm and attempts
to put pressure on Iran have breathed new political life into the program
in the short term and will be a strong incentive to continue however high
the economic price. A signicant proportion of Iranian decision-makers,
including some conservative hard-liners, oppose nuclear weapons on
Islamic grounds. (Khomeini abandoned the shahs nuclear program, de-
nouncing nuclear weapons as inhumane and un-Islamic.)
38
This is not to
say that the Iranians do not intend to keep their options open. But any
nuclear program seems likely to remain a low priority for the time being.
Still, U.S. counterproliferation efforts, even given their limited formal
success, have slowed the progress of the Iranian nuclear program, largely
by discouraging Irans most likely nuclear suppliers from fully support-
ing its proliferation efforts.
39
Ultimately, Irans Islamist leadership most likely believes it need not
build a force capable of challenging the West, just one capable of deterring
its Arab neighborsparticularly Iraqand keeping the Western powers
and their regional allies uncertain about Irans intentions. (Inscrutability
is, after all, a Persian high art.) Both the Persian tradition of conict
(whether interpersonal or international) and Islamist revolutionary myth
put great stock in psychological maneuvering and moral victories. Hence,
Iran will be perfectly capable of declaring victory in defeat, much as
Saddam Hussein did, simply because it survived the onslaught from the
West. Iran and Iraq share a determination to resist what they see as a
Western and Israeli-dominated status quo in Middle East regional secu-
rity. So far, Iran has sent a clear message that it intends to challenge the
status quo, but has conducted its revisionist campaign below the level of
a clear casus belli. If the Iranians have found the effort to rebuild and
modernize their conventional capabilities and acquire nuclear weapons
frustrating, those efforts have nonetheless netted Iran some political capi-
tal. The purchase of Russian-built submarines, the militarization of the
disputed mid-Gulf islands, an increasingly robust tempo of naval exer-
cises in the Gulf, and an aggressive program to procure nuclear weapons
has brought Iran once again to the forefront of Western security concerns.
38. Gary Sick, Rethinking Dual Containment, Survival, Vol. 40, No. 1 (Spring 1998),
p. 17.
39. Israel Digs In: No Signing of the NPT, Mideast Mirror, January 11, 1995, p. 8.
chapter 4 111
After the Iran-Iraq War, the West wrote Iran off for a time as a defanged,
marginal threat. To the extent that at least part of Irans motivation is to
regain status, the Western reaction is a clear message that the strategy is
working.
Irans security objectives, including its pursuit of nuclear weapons,
follow from its strategic personality and its deep sense of insecurity.
Iranian security policy is based on three assumptions: that some form of
U.S-led military aggression against Iran is inevitable in the near future;
that Iraq will eventually rebuild its military capabilities and pose an
equally certain threat; and that Iran could not stand up to either of its
potential adversaries with its projected conventional military capabilities.
Iran also knows it desperately needs to modernize and revitalize its
economy but that it cannot do so without considerable help from Western
Europe and Asia. Irans technological and military backwardness and its
lack of regional inuence conict with its arrogant self-image. The result-
ing dissonance could lead to two policy responses: shame and appease-
ment, or the stepped-up, compensatory bravado of the schoolyard bully.
For the time being, at least, Iran has opted for the latter.
Iranian Nuclear Strategy and the Counterproliferation Agenda
The conventional wisdom says that Iran seeks to acquire nuclear capa-
bilities because they are the great equalizer that would allow it to
challenge Western military dominance directly, and perhaps offensively.
This conventional wisdom overstates the similarities between Iran and
Iraq. While Irans military build-up and hegemonic ambitions do pose a
threat to regional stability, Iran does not share Iraqs aggressive, territo-
rially expansionist ambitions or its personalized and, hence, unpre-
dictable security objectives. Moreover, Iran is ruled by a regime that,
while certainly not a liberal democracy, is nonetheless accountable to its
people and more democratic than the United States Arab allies in the
regionand certainly more than Iraq.
40
That accountability makes it
40. One school of analysis draws conclusions concerning Irans likely nuclear strate-
gies based upon its casualty-intensive human wave tactics during the Iran-Iraq War.
(See, for example, Paula A. DeSutter, Denial and Jeopardy: Deterring Iranian Use of NBC
Weapons [Washington, D.C.: National Defense University Press, 1997], p. 10.) Citing a
so-called cult of death, these analysts assume that Shia religious fervor and the
myth of martyrdom leave Iran less risk-averse and more willing to accept casualty
levels, even among civilians, that most other states would deem unacceptable. Apart
from the cultural arrogance that such a conclusion implies, the facts do not bear it out
for at least two reasons. First, the Iranian human waves were the product of
desperation, not ideology (although Islamic fervor was clearly mobilized to gather
volunteers for what amounted to a suicide mission)the Iranian army had exhausted
112 the coming crisis
much less likely to risk its survival recklessly for self-serving ends, as the
Iraqi regime has done. Iran seems to be constrained by the same sorts of
norms that govern Western concepts of the rules of war, although those
norms are clearly colored by Irans Shia myth. For example, Iran initially
conceived of the Iran-Iraq War as a defense of righteous Islam against an
onslaught from the blasphemous Saddam-Yazid (a mythic reference to
the Umayyad caliph, Yazid, responsible for the murder of Imam Husayn).
Therefore, Iran pursued a military strategy that sought to limit the impact
of the war on the Iraqi people and avoid collateral damage, especially in
the predominantly Shia provinces. The underlying assumption of this
strategy was that the Iraqi people, given the opportunity, would rise in
rebellion against the apostate Baathist regime. Only after several years,
as it became clear that the Iraqi population had no inclination to over-
throw Saddam Hussein, did the Iranian leadership lift restraints on the
conduct of the war. The Iranians stopped viewing the Iraqi population as
innocents and started treating themthe Iraqi Shia includedas willing
followers of Yazid. The redenition of the Iraqi civilian population
paved the way for Irans decision to begin shelling Iraqi cities in 1984.
41
So, in this case, Irans particular interpretation of its war with Iraq, at
least initially, inspired even greater restraint than standard norms might
have dictated.
Iran can produce chemical weapons and will soon be able to produce
ballistic missiles to complement its North Korean Nodong missiles. To
date, however, Iran has treated its chemical weapons as a deterrent
capability and has demonstrated misgivings about their use. In fact, one
of Irans most bitter grievances against the West is that the United Nations
and the Western powers issued no condemnation of or sanctions against
Iraq for its frequent use of chemicals against Iranian forces after 1985.
42
all other options. Public and clerical opinion quickly turned against the human cost
of a war that, once Iran was on the offensive in Iraqi territory, was no longer seen as
defense of the Iranian nation but an ill-considered and unpopular Islamic crusade.
Second, the bombing of Tehran during the War of the Cities, which is a better
indicator of the Iranian response to nuclear attacks, led the Iranian public, the clerics,
and eventually even the Ayatollah Khomeini to conclude that the survival of the
Iranian nation outweighed even the loftiest Islamic objectives. (See Mackey, The
Iranians, pp. 324333.)
41. John Kelsay, Islam and War: A Study in Comparative Ethics (Louisville, Ky.: West-
minster/John Knox Press, 1993), p. 75.
42. Iran probably employed some captured Iraqi chemical weapons shells during the
Iran-Iraq War, but their use was episodic, opportunistic, and on a very small scale.
Anthony Cordesman, Weapons of Mass Destruction in the Middle East (London:
Brasseys, 1991), p. 92.
chapter 4 113
Iran principally seems to see its chemical arsenal as a deterrent to Israeli
nuclear weapons, as do some of the Arab powers. In the name of fairness
and justice, Iran proclaims its right to have chemical weapons so long as
the Western powers and Israel maintain their nuclear capabilities.
What if the Iranian porcupine grows nuclear quills? Three elements
of the Iranian myth should gure prominently in any attempt to counter
an Iranian nuclear strategy. First, Iran believes it is the center of the
universe and the eventual seat of paradise; hence, it is culturally and
morally stronger than any of its adversaries, especially the corrupt West.
Moreover, its Zoroastrian and Shia traditions instill in Iran the condence
that it is destined, sooner or later, to defeat the forces of evil through the
power of its righteousness and the favor of God. Therefore, it is not
necessary or even desirable to pursue extremely risky strategies, espe-
cially ones in which the stakes are high (survival) and the chances of
prevailing nearly nonexistent. Second, Iran will assume (as did Saddam
Hussein) that the Great Satan does not have the mettle to stand up to
pain and sufferingthat the United States is unlikely to risk signicant
casualties in any conict with Iran. Third, Irans concept of victory is
driven by its sense of shame over past foreign domination and the
determination to defend its territorial, cultural, and religious integrity. It
is not necessary that Iran defeat its adversaries, merely that it prevent
their violating Irans frontiers.
Irans national myth will constrain its use of nuclear weapons. Be-
cause it sees the United States as the Great Satan that operates without
moral constraints and with the aim of destroying the Islamic way of life,
Iran has to assume that if it uses its nuclear weapons, the United States
will not hesitate to retaliate in kind. The Iranians also contend that Iranian
lives are expendable in the U.S. view, as demonstrated in its failure to
condemn Iraqi gas attacks against Iran. Given these assumptions, Iran
almost certainly will assume that U.S. retaliation would be far greater
than the degree of damage Iran could inict on the United States, Saudi
Arabia, or Israel. Similarly, Iran (like its Arab neighbors) is acutely aware
of Israels vast military superiority, and its ability and willingness to
punish far in excess of any pain Iran could inict on Israel. Iran is also
aware of Israels national myth: that it will ght to the last Israeli to
defend its right to exist and will be little constrained by international
sanction or criticism.
Iranian nuclear weapons will almost surely be weapons of last resort
aimed at Iraqi cities and military facilities, U.S. military forces in the
Persian Gulf region, Israeli cities, and those Arab states that might coop-
erate with a U.S. military action against Iran. While it is clearly important
to Iran to acquire nuclear weapons, Iran is no more likely than any other
114 the coming crisis
nuclear power has been to consider the employment of nuclear weapons
with anything but the utmost gravity. Iran is not a risk-taker in its efforts
to challenge the West; rather, it fusses, fumes, and tests the waters, while
maintaining enough plausible deniability to back off without losing face
when it meets real resistance.
Will Iran attempt to capitalize on its nuclear status in future saber-
rattling and regional pressure tactics? Almost certainly. Is it likely to
initiate what it certainly knows would be a suicidal rst strike before all
other options had been exhausted and Irans survival was in imminent
danger? There is no reason to think so, any more than for any other
rational power. The message to the West would be if you attempt to
destroy us, we may go down, but we will take as many Americans,
Europeans, and Israelis with us as we possibly can. The message to
potential Arab members of an anti-Iran coalition would be similar: If the
Great Satan invades Iran, and threatens our sovereignty and Islamic way
of life, then we will visit the wrath of God on those who invited them in
and gave them haven. The underlying assumption of an Iranian nuclear
strategy would be that no matter how much the United States, the West,
and the Arabs might vilify Iran, they would not regard its defeat as worth
that kind of risk.
irans incentives to a pursue nuclear weapons capability
Three factors drive Irans quest to develop a nuclear weapons capability.
Only one, the development of peaceful nuclear power, is openly declared.
The ofcial line is that Iran seeks to build a nuclear power infrastructure
that would eventually provide 20 percent of Irans electrical power.
43
Both
Presidents Rafsanjani and Khatami have been vehement in their denial
that Iran seeks nuclear weapons, and the insistence that Islam forbids it,
that Iran hates nuclear weapons, and that it is not a weapon that can be
used in human confrontation.
44
Perhaps this is Iranian politics at its
dissembling best. Few doubt that Irans ultimate if not immediate goal is
to build a nuclear weapons production capability, but peaceful nuclear
power provides a g leaf, both for the Iranians and for their potential
suppliers (especially Russia and China) that do not wish to directly
challenge the U.S. counterproliferation agenda. However, the economic
argument also affords something of a lever for U.S. counterproliferation
policy. The construction of a nuclear power infrastructure makes little
43. Friends and Foes: How Iran Sees Itself in the World, Economist, January 18,
1997, Survey Iran insert, p. 10.
44. Hashemi Rafsanjani interview with Christiane Amanpour, CNN (Cable News
Network) Specials, Transcript #534, July 2, 1995.
chapter 4 115
economic sense considering that Iran sits on one of the worlds largest
reserves of oil and natural gas. The Iranian argument that it needs nuclear
power to modernize its economy could be deated if European and Asian
companies were encouraged to help Iran modernize its conventional
electric power infrastructure. Such an approach could only minimally
affect Irans ambitions, but it could complicate Iranian efforts to nd
suppliers among states with an economic interest in maintaining reason-
ably good relations with the United States.
Irans second objective in seeking a nuclear arsenal is defensive. Iraq
continues to be a major security concern for Iran, despite the latters
recent threats to forge a revisionist alliance. Iran was traumatized by its
militarys inability either to defend against or retaliate for Iraqi missile
attacks in 1988; its leadership is determined not to be caught short again.
Iran feels doubly threatened by the unprecedented level of U.S. military
presence in the Persian Gulf (which even the pro-U.S. shah would likely
have found unacceptable). Quelling Irans defensive concerns would be
a complicated task, but there are steps that can be taken to reduce their
destabilizing potential. One approach would be to recast the rhetoric of
U.S. policy to minimize (or eliminate) its ideological aspects in favor of
stressing international norms and concrete national interests. The United
States and Iran share a number of common security interests in the
regionstarting with the continued containment of Iraqi military power.
They also share a penchant for ideological crusades; but, in this case, the
realpolitik and ideological agendas are irresolvable and work at cross-
purposes. Iranian rhetoric radicalizes when its leadership feels cornered,
but has abandoned its ideological agenda given the proper incentives and
constraints (as in its relations with former Soviet Central Asia). It will
likely be a very long time before the United States and Iran once again
see themselves as strategic allies, but as Irans neutrality in the 1991
Persian Gulf War demonstrated, the Islamic Republic can recognize (al-
though not acknowledge) certain common interests with the United
States. For that to happen, however, the United States will have to mod-
erate its dual containment rhetoric and follow the lead of its European,
Asian, and Arab neighbors that have resigned themselves to the legiti-
macy and likely longevity of the Iranian Islamist regime. The challenge
will be to nd a way to do so that avoids the appearance of backing
down, which Iranians will characterize as weakness and exploit in its
internal and regional propaganda. There are ways for the United States
to decouple its approaches toward Iran from the containment of Iraq
without easing legitimate pressure on Iran to improve its record concern-
ing human rights, terrorism, interference in the internal political affairs
of its neighbors, threatening military activity, and nuclear proliferation.
116 the coming crisis
A second approach to defusing the destabilizing potential of Irans
strategic insecurity might be to revise the current U.S. policy concerning
the embargo on conventional arms transfers to Iran. The current approach
hands Iran a moral victory every time it circumvents U.S. efforts (which
it does with reasonable frequency), and the difculty Iran has had acquir-
ing conventional weapons and spare parts increases the incentive to
acquire weapons of mass destruction, especially nuclear weapons. While
the denial of all Iranian military modernization might be the U.S. ideal,
it is an unrealistic goal. If some modernization is inevitable, better that it
be in conventional forces than in nuclear capabilities, especially since
resources devoted to conventional forces may be resources taken out of
the nuclear weapons program. A conventionally stronger Iran will com-
plicate the regional security challenge, but not nearly so much as would
a nuclear-armed Iran.
Irans nal incentive for becoming a member of the nuclear club is
its desire to be a regional hegemon, a major regional power, or at least
not marginal. Leadership of the Islamic world is a popular objective in
Iran and one of the regimes few reliable sources of political consensus.
Iran, in addition, has always seen itself as the strategic center of gravity
in the Persian Gulf. The Iranians would prefer to be admired as a positive
role model of a righteous Islamic society; but failing that, they will settle
for being feared and pretend that is what they wanted all along. Ironically,
while this is the aspect of Iranian policy that looks most ominous to the
United States and the Gulf Arabs, it may be the least realizable for a
number of reasons. First, Irans ideological excesses consistently work
against its ambitions. The intensely practical Saudis and most other GCC
states have good cause to be highly suspicious of Iran and very skeptical
of Iranian gestures of conciliation; it is a skepticism born of long experi-
ence and likely to be of long duration. Even those Western European
states that have pursued a policy of critical dialogueincluding Ger-
many, until recently Irans best European friendhave lost patience with
the intemperance of Irans ideologically driven misconduct. Second, Iran
is beginning to show signs that it recognizes that it will eventually need
to moderate its ideological stance to have any hope of becoming economi-
cally or militarily competitive. Internally, pragmatists and technocrats
within the Iranian regime recognize the need to balance Islamic correct-
ness with technical competence. Externally, Iran has acted surprisingly
responsibly in Central Asia, largely in order not to alienate Russia, an
important economic partner and military supplier. Likewise, for purely
expedient reasons, Iran has made common cause with China, which treats
its Muslims notoriously badly. In short, Iran is likely to continue to
moderate its external behavioralbeit slowly and erraticallyas the in-
chapter 4 117
centives to do so increase. Finally, even if Iran moderates its external
behavior, regional dominance is likely to remain beyond its grasp. The
pragmatic strategic differences that divide Iran and its regional competi-
torsespecially Saudi Arabiaare wider and more intractable than those
that separate Iran and the United States.
45
In fact, it is not beyond the
realm of possibility that Iran and Saudi Arabia will at some point become
competitors for U.S. support in the region, as they were before 1979.
u.s. and iranian bargaining strengths
Iran has a few advantages in resisting the United States counterprolif-
eration agenda. First, Iran is not alone in its suspicion of the major
powers motives in pressing the counterproliferation agenda. The fairness
issue that Iran consistently raises, especially regarding Israeli nuclear
capabilities, resonates throughout the Middle East and the developing
world. Moreover, Irans status as an uncompensated victim of Iraqi
chemical weapons attacks gives its protestations a moral viability among
the weapons of mass destruction (WMD) have-nots. Irans rhetorical
ourish and penchant for playing the victim have served it well in this
regard. Fairness is an issue for which the antiproliferation major powers
must eventually formulate an answer. Second, the Iranian nuclear and
conventional military modernization programs are politically very popu-
lar in Iran, despite their cost. They seem to cause their Saudi and U.S.
rivals heartburn, providing Iran with a symbolic victory and prestige. In
fact, the revitalization of Iranian military might is a rare political success
for a regime that has had precious few. The declared U.S. dual contain-
ment strategy tends to reinforce the popularity of the modernization
program by lending credence to Irans national paranoia. Finally, experi-
ence has demonstrated the difculty of preventing the acquisition of
nuclear technologies by a state that is determined to have them. U.S.
efforts to cut off Irans access to nuclear technology at the source have
been somewhat successful; but U.S. relations with Irans major supplier,
China, chilled at the same time that Iran and China have begun to see
common cause, especially in their mutual sense of grievance at perceived
U.S. attempts to coerce them into internal reforms.
The United States is not without a few advantages of its own. First
and foremost, Iran desperately needs an economic turnaround, which
will depend on foreign investment and technical assistance. European
and Asian investors are not likely to make signicant capital commit-
ments in Iran until they are sure they do not stand to lose their invest-
ments, as some did during the 199091 Persian Gulf Crisis. To reassure
45. Chubin and Tripp, Iran-Saudi Arabia Relations and Regional Order, p. 74.
118 the coming crisis
them, Iran will have to demonstrate convincingly its ability to function
as a responsible international agent. It cannot do so unless it achieves at
least minimal rapprochement with the United States. The Islamist lead-
ership will continue to protest loudly that it is unconcerned by U.S.
opinion of its regime; at the same time its public ritualdaily ag-
burning (a practice long staged by the Iranian regime, but which Khatami
denounced in a 1998 address to the U.S. people on Cable News Network)
and Death to America displays before the old U.S. embassybetray
their obsession. And the Iranian population will continue in its appar-
ently unquenchable fascination with the United States and its culture, will
continue to learn English, and will keep alive rumors of secret talks to
improve U.S.-Iranian relations.
46
The second U.S. advantage lies in pragmatic strategic and economic
calculations: no state would suffer more than Iran were there to be a
major disruption of trafc in the Persian Gulf, because Iran currently has
no overland outlets for its oil. The United States is the only guarantor of
security and reasonable stability in the region and the only naval power
capable of keeping the Gulf open. Irans mythic image of the United
States (amplied by the 1991 Persian Gulf War) leads it to assume that
Iranian actions will not be met with moderate or even proportional
responses from the United States. Iran harbors little doubt that the United
States will retaliate for any WMD rst strike with catastrophic effect
against which Iran could not hope to defend itself. Iranian leaders also
know that, in that event, world opinion would be unlikely to take their
side; even if it did, the United States would not be deterred.
Countering Nuclear Proliferation in Iran
The real challenge for U.S. counterproliferation policy toward Iran is to
nd the middle way between looking weak, vulnerable, or desperate and
looking ominous and belligerent. Irans national myth points to some
approaches. U.S. policy should avoid actions that trigger Iranian national
myth in ways that run counter to its objectives: arguments that focus on
ideological issues rather than concrete national interest; rhetoric, such as
the Islamic threat; policies or actions that hand Iranian hard-liners
moral victories; or policies that overtly aim to overthrow or destabilize
the current political order, and hence feed Irans xenophobic paranoia. At
the same time, it is possible to craft approaches that mobilize or respond
to the Iranian myth in a way that advances the U.S. agenda. The United
States must continue aggressively to occupy the moral high ground
46. Friends and Foes, Economist, p. 10.
chapter 4 119
concerning weapons of mass destruction: point out the immorality of
wasting scarce resources on expensive weapons of questionable defensive
utility; turn Irans self-righteousness and its antinuclear rhetoric against
it by emphasizing the unjust, inhumane, immoral, and fundamentally
un-Islamic nature of such weapons; and emphasize that Iran, a past
victim of chemical weapons, bears a particular responsibility to lead the
Islamic world in resisting the spread of such weapons. Nor can it hurt to
prick the Iranian ego by championing new norms of international behav-
ior that denigrate WMD as the last resort of failed, corrupt leaders trying
desperately to acquire the empty trappings of power at the expense of
building truly robust and successful societies. The rst tier powers of
the postCold War erathe United States and its European allies, Russia,
and even South Africaare trying to nd ways to denuclearize. In the
process, they are creating a new norm: strong, internally secure, and
externally respected powers do not need the crutch of nuclear dominance
and would certainly not bankrupt their societies to get it. At the same
time, a new regime of international norms should hold up as paragons
those developing countries (like the Asian tigers) that have achieved
international respect and inuence, often without natural sources of
wealth like oil, and that have avoided Westernization by investing in
their economies and keeping their investment in military infrastructure
proportionately low.
Finally, the United States can best avoid the appearance of either
weakness or belligerency by dropping its side of the ideological battle
and tailoring its rhetoric to a measured, realist worldview that focuses on
Irans behavior rather than its ideology and denes benchmarks for
improved relations. In so doing, the United States need not abandon but
only recast the ideological principles that dene its own strategic person-
ality. Instead of insisting that Iran is the new evil empire (which feeds
both its ego and its paranoia) and demanding that our allies conform to
that view, the United States can state that it nds Irans behavior unac-
ceptableits support for international terrorism, its interference in the
internal political stability of its neighbors, and its nuclear and chemical
weapons proliferation. A unilateral stand that withholds U.S. economic
and political approval but also states clear conditions under which the
United States would be willing to relax its pressure on Iran shifts the
burden for improving relations back to the Iranian regime. European
advocates of critical dialogue are coming to the realization that they
have been no more successful than the United States in moderating
Iranian behavior. The Iranian Islamist regime, no matter how important
its economic ties with states like Germany and France, seems unable to
restrain its Jacobean tendenciesas it demonstrated in sending hit
120 the coming crisis
squads (under cover of diplomatic passports) to assassinate exiled dissi-
dents in Germany. Those states, especially in Europe, that are now in-
clined either to buck U.S. policy as a sign of policy independence, and to
attribute Irans more intemperate behavior to U.S. provocation and cul-
tural arrogance, may eventually run out of rationalizations and nd
themselves compelled to put greater emphasis on the critical dimension
of their dialogue.
Upon encountering difculties, those who deal with Middle East
policy often throw their hands up and mutter about inscrutable Orien-
tals and the unpredictability of non-Western cultures. The objective of
this brief exercise has been to offer some insights into the mythic foun-
dations of Iranian culture and strategic personality in an effort to make
its rationality less opaque and its behavior less mystifying. Iran is not
unpredictable; we just have not been very good at reading it. Iran is not
irrational; we just have not fully deciphered the cognitive models that
inform its behavior. The temptation to classify Iran as an irrational rogue
that operates outside the limits of accepted norms is counterproductive.
It leaves us with few positive options, since it is virtually impossible to
devise rational approaches to modifying irrational behavior. The national
mythstrategic personality approach offers a better option: it predicts the
possible range of future Iranian behavior and likely reactions to U.S.
strategic initiatives by seeking insight into how the Iranians think about
their security and relationship to the outside world. To say the Iranian
mind is inscrutable tells us nothing useful; to attempt to put ourselves
inside the Iranian mind, however, can point to some constructive ap-
proaches if not immediate solutions.
chapter 4 121
Part II
Potential Evolution and Consequences of a
Nuclear Crisis with the United States
Chapter 5
Nuclear Proliferation and
Alliance Relations
Stephen Peter Rosen
The recent multiple nuclear tests conducted by India and Pakistan in
1998 showed that countries that have a latent nuclear capability could
move quickly and reliably from bombs in the basement, that is, un-
tested and unassembled weapons, to usable nuclear or thermonuclear
weapons. The possibility is now more obvious that a latent or clandestine
nuclear weapons state might move quickly to the status of a demon-
strated nuclear power after it had initiated a military crisis. The oppo-
nents of that state might then face a state after it had committed an act
of aggression, only to nd out that it was capable of using nuclear
weapons.
How will further nuclear proliferation affect the pattern of alliances
between the United States and regional powers? During the Cold War,
the creation of new nuclear powers tended to draw the superpowers into
closer relations with their regional allies. The possibility of regional nu-
clear weapons use in the context of the U.S.-Soviet competition created
the possibility that U.S. allies and Soviet clients might either be able to
achieve decisive gains by nuclear threats or use, gains that would affect
the U.S.-Soviet balance of power, or that U.S. allies and Soviet clients
might become engaged in a nuclear conict that could escalate and
involve the superpowers in a war in which nuclear weapons were being
used. As a result, the United States and Soviet Union tended to intervene
to stabilize regional conicts and prevent regional nuclear weapons use.
The perception of this dynamic provided the foundations for the de facto
nuclear strategies of several smaller nuclear powers. With the demise of
the Soviet Union, regional nuclear weapons use may still occur, but one
incentive for the United States to become involved has been greatly
decreased. The costs and risks of intervention in a regional nuclear
conict were always high. Now that the need to balance against another
superpower is reduced or gone, U.S. military intervention after nuclear
threats have been made in the context of a regional war may be less likely.
The regional allies of the United States may well have less condence that
the United States will intervene militarily in the face of regional nuclear
threats. The belief that the United States would project military power to
ensure the survival of its regional allies provides the basis of our asym-
metrical alliances with regional powers, alliances in which the United
States extends guarantees to them in return for access to military bases
on their soil. These alliances are unlikely to survive regional nuclear
threats to which the U.S. does not militarily respond. The result will be
a sharp decrease in the military access granted by our former allies to
bases on their soil or to the maritime areas and air space over which they
claim jurisdiction. In the next chapter, Barry Posen argues that the large
and negative long-term consequences of nonintervention might lead the
United States to intervene, despite the near-term risks; this chapter ex-
plores the reasons why and circumstances under which the United States
might not intervene.
To support this argument, I rst review the information that has
emerged about the new nuclear states plans for nuclear weapons use
developed during the Cold War. Second, I try to assess the impact of
regional nuclear proliferation on the incentives of the United States to
intervene militarily in regional conicts by mentally replaying the 1991
war against Iraq with the assumption that Iraq had a small nuclear
arsenal. Based on the air war against Iraq, what can we say about the
United States nuclear strategy in a hostile encounter with a regional
nuclear power? The effectiveness of the U.S. bombing of Iraqi nuclear
facilities can now be evaluated on the basis of some of the reports made
after on-site inspections of those facilities conducted under United Na-
tions auspices, interviews with members of inspection teams, and the
report submitted to the U.S. Secretary of the Air Force, the Gulf War Air
Power Survey.
1
Based on the experiences of 199091, what might have
been the characteristics of an encounter between the United States and a
nuclear-armed Iraq? What might Iraqs nuclear strategy have been, given
what we know about its actual regional political-military strategy? This
chapter examines what the events of August 1990 through August 1991
might have looked like if Iraq had completed a nuclear weapons devel-
opment program before it invaded Kuwait, but did not reveal its capabili-
ties until after it invaded Kuwait. I then develop a partial answer to the
second question by examining in detail what we now know about the
actual effects of the U.S. bombing campaign directed against Iraqi nuclear
1. Thomas A. Keany and Eliot A. Cohen, Gulf War Air Power Survey: Summary Report
(Washington, D.C.: Government Printing Ofce, 1993).
126 the coming crisis
weaponsrelated facilities in 1991. Together, the analysis of the actual
bombing and the review of the hypothetical Iraqi use of nuclear weapons
suggest the nature of the problems the United States might face in en-
counters with regional nuclear powers over the next ve to ten years.
This look at the future is supplemented by surveys of how members of
the U.S. policy community actually responded during 1993 when they
were asked to develop policy options for the United States in simulated
encounters with regional nuclear powers.
The result of this examination can only be tentative, but it is striking.
There appear to be no good non-nuclear military options open to the
United States when facing a regional nuclear power. There is little detect-
able sentiment in the U.S. policy community today to use U.S. nuclear
weapons against aggressive regional nuclear powers. Without the global
strategic need to prevent military gains by clients of the Soviet Union or
to prevent regional nuclear wars from escalating to U.S.-Soviet nuclear
war, there is less incentive for U.S. military intervention. Hostile regional
nuclear powers are, therefore, in a good position to deter U.S. military
intervention against them, and so may be in a better position to threaten
the allies of the United States that do not have nuclear weapons and are
within reach of the non-nuclear forces of the aggressive power. The main
purpose of regional nuclear weapons use is likely not to be against cities
or military targets in the region. Nuclear weapons use is most likely to
follow an act of aggression, and to be purely demonstrative, designed to
deter U.S. military intervention and to convince the other powers in the
region that they cannot rely on U.S. security guarantees, so that the
aggressor can obtain its foreign policy objectives without going to war
with the United States. The asymmetrical U.S. relations with its regional
allies in Europe, the Middle East, and the Pacic are unlikely to survive
if the patterns of behavior forecast in this essay take place. The chapter
concludes with a discussion of what measures the United States might
take to increase its ability to preserve its alliances by increasing its ability
to intervene militarily against small nuclear powers.
Nuclear Weapons Doctrine in New Nuclear States
What do we now know of the ideas concerning nuclear weapons use in
India, Pakistan, South Africa, and Israel?
In 1990 and 1993, I interviewed the retired chief of staff of the Indian
Army, General K. Sundarji, and the head of Secretary of the Defense
Research and Development Organization of the Indian Ministry of De-
fense, Dr. V.S. Arunachalam. The timing of these interviews, which were
given on the initiative of Arunachalam, may have been related to the
chapter 5 127
decline and then the demise of the Soviet Union, and the perceived need
in India to develop better relations with the United States. One element
in the effort to improve U.S.-Indian relations would be some explanation
of the nuclear weapons custody arrangements in India that would reas-
sure the United States that weapons would not be used inadvertently or
in thoughtless ways. Arunachalam has subsequently denied in print the
substance of these interviews. In this paper, I present what he told me in
December 1990, and what he conrmed as accurate in the spring of 1993,
and what I reported to the United States government in December 1990.
In 1990, Arunachalam, though formally denying that India had nu-
clear weapons, adopted the following position. It would be reasonable,
he said, for non-Indians to assume that since the 1974 Indian nuclear
weapons test, the government of India had worked very hard at resolving
the command and control problems for nuclear weapons. In fact, the
civilian leadership had fought a long and difcult struggle with the
Indian military to decide who would control nuclear weapons and how
they would be used. This struggle was resolved in favor of the civilians.
The outcome was that the Indian military would not be and has not
been told how many nuclear weapons India could have, nor will it be
told in peacetime how nuclear weapons would be used in war. The
inescapable implication was that the Indian military does not have cus-
tody of nuclear weapons components in peacetime. Since the military
cannot plan for nuclear war, a set of detailed instructions on how to get
access to nuclear weapons and how to employ them has been prepared
by civilians. These sealed instructions have been given to a military
ofcer with instructions to open them in the event that nuclear weapons
have been used against India and have destroyed the Indian national
command authority at New Delhi. To quote Arunachalam, If New Delhi
goes up in a mushroom cloud, a certain theater commander will go to a
safe, open his book, and begin reading at page one, paragraph one, and
will act step by step on the basis of what he reads. The technical means
for command and control are rudimentary, but adequate for Indian needs.
No provision has been made to harden Indian communications against
the effects of electromagnetic pulse (EMP), since it is thought that the
number of nuclear weapons that would be used against India would be
sufciently small and the campaign sufciently short that EMP blackout
would not be substantial enough to interfere with the execution of Indian
plans.
Much of what Arunachalam said was placed in broader context by
the remarks of General Sundarji in interviews in 1990 and 1993. He said
that India has adopted a pure minimum deterrence posture, for two
reasons. First, India cannot afford to develop a nuclear force to destroy
128 the coming crisis
Pakistani nuclear weapons because the Pakistanis have dispersed and
hidden their nuclear weapons and India does not have enough resources
to nd and strike them. Second, India has sufcient non-nuclear forces
to handle Pakistani non-nuclear attacks and thus does not need to
threaten Pakistan with a nuclear rst strike in order to deter war. When
asked what India would do if Pakistan achieved some limited initial
territorial gains in an invasion of Kashmir, halted, and then threatened
to use nuclear weapons if India tried to push it back, Sundarji said India
would simply proceed with its non-nuclear counterattack.
After the 1998 nuclear weapons tests, the only public statement about
Indian nuclear doctrine was given by Prime Minister Atal Behari Vaj-
payee, who publicly stated that India needed only a credible deterrent,
and would not need a large infrastructure to control or safeguard Indian
nuclear weapons.
2
While consistent with the Arunachalam-Sundarji state-
ments, it does not conrm them.
The alleged Indian posture has several advantages. It would make
the Indian problems of obtaining warning and of executing its nuclear
strike much simpler. There would be no need for India to retaliate quickly.
Although Indias retaliation after a Pakistani use of nuclear weapons
would not be delayed indenitely, it would be acceptable for India to
retaliate in something like twelve to forty-eight hours. Capabilities to
execute a slow-motion nuclear war would be much cheaper and easier
for India to build. Most importantly, the alleged strategy would prevent
military custody of nuclear weapons or any control over nuclear weapons
in peacetime.
3
On the Pakistani side, the retired Chief of Staff of the Pakistani Army,
Mirza Azlam Beg, wrote in April 1994 that Pakistan had acquired a
nuclear weapons capability but had capped it because we thought that
Pakistan had acquired the maximum deterrence level that is needed to
avert the threat that we perceived and . . . could also quantify, for exam-
ple, the correlation between India and Pakistan . . . conventional forces,
forces which he wrote favored India by three to one in ground forces
and ve to one in air forces. As a result, he quoted approvingly the
statement by Sundarji that between India and Pakistan a substantial
degree of nuclear stability exists, and Beg went on to write that the
existence of Indian and Pakistani nuclear weapons certainly has reduced
the possibility of war. The Pakistanis, he wrote, had created a nuclear
2. Kenneth Cooper, Leader Says India Has A Credible Deterrent, Washington Post,
June 17, 1998, p. 21.
3. Interviews conducted by author in Pune, India, December 1990, and Cambridge,
Massachusetts, March 1993.
chapter 5 129
chain of command that was run by the head of state operating through
a National Nuclear Command Authority that issued orders to a military
Joint Operations Center in Rawalpindi.
4
The Indian-Pakistani nuclear relationship, thus, appears to have
emerged as a stable relationship in which non-nuclear military imbal-
ances are neutralized by the existence of nuclear weapons. At the same
time, there are persistent reports that the possibility of nuclear weapons
use in South Asia has been manipulated by the Pakistanis in order to
increase the level of U.S. diplomatic intervention.
Some evidence suggests that other smaller states have explicitly con-
sidered the use of nuclear weapons in ways that would lead to inter-
vention by the United States. For example, the 1993 statement by the
president of South Africa explicitly asserted that while the purpose of
South Africas seven nuclear weapons was related to deterrence, the
South African strategy for nuclear weapons use had nothing to do with
counter-city or counterforce targeting, nor even directly with the regional
adversaries of South Africa. Rather, the strategy was that if the situation
in Southern Africa were to deteriorate seriously, a condential indication
of the deterrent capability would be given to one or more of the major
powers, for example the United States, in an attempt to persuade them
to intervene.
5
This plan for nuclear weapons use resembles that imputed
to Israel. It has long been hinted that the strategy that Israel employed
in the early hours of the 1973 war was similar to the South African nuclear
strategy, and that Israel initiated preparations to turn the nuclear option
into usable nuclear weapons, and then communicated to the United
States Israels immediate need for arms, as well as the desperate alter-
native it contemplated if it did not get help.
6
The concepts concerning nuclear weapons use in the four cases of
India, Pakistan, South Africa, and Israel appear to fall into two categories:
minimum deterrence nuclear strategies and strategies to draw the inter-
vention of the superpowers. They do not appear to consider the use of
nuclear weapons to deter superpower military intervention, but this is a
logical possibility in Northeast Asia or Southwest Asia.
4. Mirza Alam Beg, Who Will Press the Button? The News (Islamabad, Pakistan)
April 23, 1994, p. 6, reprinted in Foreign Broadcast Information Service, FBIS-NES-94-
086, May 4, 1994, pp. 6971. See also the remarks by Beg in Pakistan Observer, December
2, 1993, p. 1, reprinted in FBIS-NES-93230, December 2, 1993.
5. Speech by F.W. De Klerk, March 24, 1993, released by the embassy of South Africa
to the United States, March 26, 1993.
6. Nadav Safran, Israel: The Embattled Ally (Cambridge, Mass.: Harvard University
Press, 1978), p. 488.
130 the coming crisis
There is no hard information that could provide us with certain
estimates about how nuclear weapons might be used by regional powers
to deter intervention against them. Could small and medium-sized nu-
clear powers be condent that their small arsenals could survive a non-
nuclear attack by the United States? If so, would they then be condent
of their ability to deter U.S. military intervention? Given the data avail-
able from the 1991 Gulf War, what can be said about the ability of the
United States to destroy the nuclear weapons of small states by non-
nuclear means?
U.S. Targets in Iraq in the 1991 War
disarming strikes: the air war against iraqi ballistic missiles
Two sets of Iraqi targets that were attacked by the aircraft and missiles
of the United States in the 1991 war were relevant to Iraqi nuclear
war-making capabilities. The rst was the Scud ballistic missile, one of
the possible delivery systems for Iraqi nuclear weapons. The Gulf War
Air Power Survey (GWAPS) analysis of the air campaign against the Iraqi
Scud missile launchers is thorough and candid. It notes that possible Scud
hiding places, suspected xed launch sites, production facilities, and
mobile Scud launchers were attacked. One thousand ve hundred aircraft
strikes of all kinds were launched against Scuds, out of a total of some
42,000 strikes by all Coalition xed-wing (as opposed to helicopter) air-
craft during the war. One thousand aircraft were sent on Scud patrols.
Only 215 strikes were directed at what the pilots thought were mobile
Scud launchers. The Iraqis appear to have relied totally on mobile launch-
ers, and did not employ any of the suspected xed launch sites. The
effectiveness of the air attacks against mobile Scud launchers was ex-
tremely limited, in three senses. First, because many Scuds were launched
at night and the launchers were small and hard to see from the air, even
in the forty-two cases when the are of light from a Scud launch was
observed by a U.S. aircraft pilot in a strike aircraft circling in the area, in
only eight cases was the pilot then able to see a target well enough to
launch a weapon against it. Second, while the number of Scuds the Iraqis
were able to launch does appear to have been reduced because they were
being hunted by U.S. aircraft, this reduction was limited. Although the
number of Scud launches did decrease from the rst to second weeks of
the war, and then dropped further in the third week, the number of
launches then began to increase, and in the last eight days of the war
there were as many Scud launches as in the second week of the war.
Third, in the words of the report:
chapter 5 131
The fundamental sensor limitations of the Coalition aircraft, coupled with the
effectiveness of Iraqi employment tactics (including use of decoys), suggest
that few mobile Scud launchers were actually destroyed by Coalition aircraft
or special forces during the war. . . . Once again, there is no indisputable
proof that Scud mobile launchersas opposed to high-delity decoys, trucks,
or other objects with Scud-like signatureswere destroyed by xed-wing
aircraft.
7
One cannot infer directly from the 1991 campaign against Iraqi Scuds
armed with non-nuclear warheads that a similar campaign waged against
Iraqi missiles armed with nuclear warheads would have looked the same.
(This issue is discussed under the section that considers hypothetical Iraqi
nuclear weapons use.) However, one issue that would have been common
to both the actual non-nuclear ballistic missile force and a hypothetical
nuclear force would have been the physical aspects of the command and
control system used to give orders to missile commanders to launch their
weapons. Though the Iraqi command and control system was subjected
to 840 precision and nonprecision attacks, by the end of the second week
of the war, Coalition air planners and intelligence analysts became in-
creasingly convinced that Iraqs national-level telecommunications sys-
tem had not collapsed as a result of attacks on central switching and
microwave relays. . . . Fiber-optic networks and computerized switching
systems proved particularly tough to put out of action. . . . Moreover, the
Iraqi government had been able to continue launching Scuds during the
nal days of the campaign.
8
the air war against the iraqi nuclear weapons infrastructure
The second target set was composed of facilities that produced the ma-
terials necessary for the production of Iraqi nuclear weapons. A military
campaign to destroy Iraqi nuclear weapons would necessarily have at-
tacked this infrastructure as well in order to prevent Iraqi nuclear re-
armament. How effective was the bombing of the Iraqi nuclear infra-
structure?
To begin with, how should effectiveness be dened? It is routine
for the U.S. Air Force to identify a set of targets, attack them, and then
to assess the effectiveness of the campaign by counting the number of
targets on the list against which attack missions were own, and then,
using aerial photography, to count the number of targets hit and dam-
aged. The percent of targets on the original list damaged constitutes one
measure of effectiveness. A second, more appropriate measure of effec-
7. Keany and Cohen, Gulf War Air Power Survey, pp. 8390.
8. Ibid., pp. 6970.
132 the coming crisis
tiveness is the one adopted by the Economic Objectives Unit of the Ofce
of Strategic Services in its work to support the U.S. Eighth Air Force
during World War II. In that assessment, the measure of effectiveness
adopted was how the industrial systemscomposed of a network of
factories, warehouses and so onwere affected by bombing. Since the
industrial systems that produced aircraft, engines for armored vehicles,
and petroleum could be rebuilt, and since the consumers of the output
of those systems could draw supplies from stockpiles of produced end
items, the relevant question was how many months worth of industrial
production had been destroyed by bombing. This could then be com-
pared to rates of consumption to achieve a meaningful measure of the
effectiveness of the bombing.
9
A modied version of this measure of
effectiveness appears to be appropriate for the Iraqi nuclear weapons
infrastructure. There, too, facilities that were damaged or destroyed by
bombing could be repaired or replaced by stockpiled output of those
facilities or by stockpiled, redundant production capabilities. Since there
was and is no observable rate of consumption for Iraqi nuclear weapons,
the nal measure of effectiveness is simply how many months of delay
would be introduced into the Iraqi nuclear weapons production system.
The air campaign, according to Iraqis interviewed by the International
Atomic Energy Authority/United Nations Commission (IAEA/UN-
SCOM) teams, halted the Iraqi nuclear program in mid-January 1991.
10
But what was the capacity of the Iraqi program to recover and begin work
again once the bombing stopped? On the last day of the air war against
Iraq, how long would it have taken the Iraqis to restore their nuclear
weapons production capacity to where it had been when the air campaign
began?
Unfortunately, this straightforward question has no straightforward
answer. Events only indirectly related to the bombing added to the actual
delays the Iraqis faced at wars end. The victors insisted that IAEA
inspection teams be given unhindered access to all known and suspected
Iraqi nuclear weaponsrelated facilities. This led Iraq to dismantle, hide,
or destroy certain portions of its nuclear infrastructure. In addition, the
IAEA teams were themselves able to order the destruction of equipment
related to the production of nuclear weapons, and to monitor Iraqi sites
after the war. The U.S. victory in the war increased the ability of the
9. Walt W. Rostow, EOU and its Doctrine, Pre-invasion Bombing Strategy (Austin:
University of Texas Press, 1981), pp. 1523.
10. Report by the Director General, IAEA, The Implementation of United Nations
Security Council Resolutions 687, 707, and 715 (1991): Addendum to GC(39)/10,
September 4, 1995, p. 2.
chapter 5 133
United States to restrict the ow of oil revenue and foreign technology
into Iraq, and raised global sensitivities to selling any kind of industrial
and scientic equipment to Iraq. In part, these were indirect results of the
bombing of Iraq, which increased the difculty Iraq faces in trying to
reconstruct its nuclear weapons program. As a result, the answer to the
question of practical policy interest, how many months will it take Iraq
to rebuild its nuclear program to where it was in January 1991? is quite
different from the question, by how many months did the bombing itself
set back the Iraqi nuclear weapons program? It is the answer to the
second question in which we are interested, so that we can understand
the value of bombing isolated from the value of militarily occupying a
hostile regional nuclear power and from the value of alternative postwar
control regimes.
The most important source of information to answer this question
comes from the IAEA inspection reports, which are unclassied.
11
The
members of the IAEA teams were on loan to the United Nations from
various U.S. and European government agencies or from jobs that gave
the team members technical expertise that allowed them to judge the
development of and damage to a nuclear weapons production infrastruc-
ture. Twelve inspections were carried out between the end of the war and
June 1992. There is some disagreement among the IAEA analysts as to
exactly how close the Iraqis were to a bomb before the war, and about
which directions the Iraqi program was likely to take in the years after
the war; however, when asked to provide their best judgment on how
much time Iraq would have needed to rebuild its nuclear weapons pro-
gram on the day the bombing stopped, given where they thought it was
when the war started and assuming the same access to international
resources that was available before the war, the analysts expressed a high
degree of agreement on seven points.
First, the component of the Iraqi nuclear program that was furthest
from completion in January 1991 was the production of the ssile material
for an atomic bomb. There are debates about how long it would have
taken Iraq to have produced enough ssile material for a bomb, but the
analysts agreed that all other problems of bomb manufacture were solved
or were solvable within the shortest period of time that would have been
needed to manufacture the necessary amount of ssile material. Iraq had
in 1991 workable detonators, high-explosive lenses, neutron initiator
technology, and a valid design for an implosion bomb incorporating
11. For a brief summary of the state of the Iraqi nuclear infrastructure in January
1994, after the war and after twenty-two IAEA inspection visits, see IAEA Action Team
for Iraq: Fact Sheet (Vienna: IAEA, January 18, 1994).
134 the coming crisis
design features that the United States introduced into its ssion weapons
in the late 1940s to reduce their size and weight.
12
Making enough
weapons-grade ssile material was the task that determined the pace of
the overall Iraqi program. It was the long pole in the tent.
Second, after the bombing of the Osirak reactor by the Israelis in 1981,
the Iraqis switched from work on a plutonium bomb to an enriched
uranium bomb that did not require a highly visible, vulnerable reactor to
produce weapons-grade ssile material. Large amounts of uranium ore
were obtained from a superphosphate mine in Iraq at Al Qain.
13
Of the
many routes for producing highly enriched uranium in which Iraq had
invested time and money (electromagnetic isotope separation, centrifuge
enrichment, ion-exchange technologies, gaseous diffusion, jet nozzle),
electromagnetic isotope separationknown variously as EMIS and Calu-
trons and Baghdadtronswas the route most likely to have produced
enough ssile material for an Iraqi bomb rst. The other methods may
have proven useful for providing partially enriched uranium for further
enrichment in the EMIS process. The Iraqis manufactured EMIS devices
in Iraq. Some necessary items of electrical equipment to support the EMIS
devices were purchased abroad. Iraq had at least thirty EMIS units run-
ning or being assembled and plans to build forty more. There is uncer-
tainty about when the forty devices would have been nished. One
assessment is that seventy EMIS devices could have been operating by
January 1992; seventy EMIS devices could have produced enough en-
riched uranium for a bomb that was not particularly efcient in its use
of uranium in three years, that is, by January 1995. If the other separation
technologies had been used to provide partially enriched feeds to the
EMIS systems, that time might have been cut to one year, that is, to
January 1993, but there is debate about how far along the other separation
technologies were.
Third, the mine at Al Qain that supplied Iraqi uranium oxide was
12. See the report of the fourth IAEA visit on its trip to Iraq, July 27August 10, 1991,
which found exploding bridge wire detonators, and polonium and beryllium for
neutron initiators at Al Qa Qaa. See also the report of the fth IAEA visit to Iraq of
September 2230, 1991, which found high explosive lenses for an implosion device,
and documents at Al Atheer covering Iraqi activities up to May 31, 1990, which include
an Iraqi design for a weapon incorporating a ying plate core to increase the efciency
with which energy is transferred to the core of ssile material. See also the inde-
pendently prepared brieng charts of IAEA team member David Kay, Iraq and
Beyond: Challenges in Controlling Nuclear Proliferation, February 1992.
13. Uranium oxide (UO
4
) can be produced as a byproduct of superphosphate extrac-
tion. Iraqis reported to the eleventh IAEA team in Iraq in April 715, 1992, that they
had produced 168 tons of uranium oxide powder, and had also purchased natural
uranium from Brazil.
chapter 5 135
heavily bombed during the 1991 war. The rst IAEA team to visit it, in
July 1991, estimated that it would take the Iraqis two to three years to
put it back into operation. In fact, the mine was back in operation
producing superphosphates by September 1991, within six months after
the end of the war. IAEA interviews noted that in any case, enough
uranium oxide was available before the war began and remained after
the war to sustain the enrichment program. Overall, the delay created in
the Iraqi nuclear program by the bombing has to be judged at between
zero and six months. However, uranium oxide must be turned into a gas
before it can be enriched by the EMIS process. The production of that
gaseous form of uranium was performed at Al Jazira at a turn-key plant
built from the ground up by a foreign rm. If the Iraqis had been free to
go back to that rm, and if that rm, or another one like it, had been
willing to do the work, rebuilding the plant would have taken about two
years. It is not known if Iraq had learned something about how to build
the plant after having operated it for a number of years. If Iraq had
learned a lot, and had been free to devote unlimited resources to its
reconstruction, a crash program might have rebuilt it in eighteen months.
Fourth, the EMIS separation process consumes enormous quantities
of electrical power, as do other enrichment processes. The electrical power
generating and distribution system in Iraq was heavily bombed during
the war, and some 88 percent of its installed generating capacity was hit
and damaged. But the impact of this destruction was far less. Iraq had
installed almost twice the generating capacity it needed, and so, in effect,
had redundant electrical power generators. IAEA inspectors also found
that Iraq had numerous stockpiled transformers that could be used to
replace electrical power substations that were damaged by bombing.
Sheds containing over 100 spare substation transformers were observed
at nuclear facilities during initial inspections. They disappeared by the
time of follow-up inspections, and the assumption was that they had been
used to help repair the national power grid. The net effect was that
adequate electrical power production relative to consumption was
quickly restored by May 1991. The summer of 1991 was particularly hot
in Baghdad, and all available air conditioners in that city ran at out,
with no observed electrical brownouts or blackouts. Air Force planners
had thought that it would take two years to restore full power to Bagh-
dad.
14
Fifth, the impact of U.S. bombing on EMIS devices and processes was
limited by the redundancy that Iraq had built into the program, by its
efforts to protect the EMIS devices from bombing, and by the ease with
14. Keeney and Cohen, Gulf War Air Power Survey, pp. 7174.
136 the coming crisis
which the devices could be put back into operation after disruption. The
Iraqis went to considerable lengths to create redundancy in their EMIS
program. The rst EMIS enrichment facility was constructed at Tuwaitha.
Duplicates of that plant were built at Tarmiya and Ash Sharqat.
15
There
may have been additional unidentied sites. The rst IAEA team to visit
known EMIS installations found that some EMIS machines themselves
(the disks, vacuum chambers, and pole magnets) had survived the bomb-
ing, either inside damaged buildings or underground, where they had
been buried to protect them from bombing or from IAEA inspection.
Some EMIS equipment was found to have been destroyed, but inspectors
estimated that 95 percent of the damage observed had been done by Iraq
itself in the process of moving the EMIS machinery in order to conceal it
from IAEA inspectors, not by bombing. IAEA teams reported Iraqi efforts
to recover hidden EMIS equipment in July 1991. Estimates of how long
it would take Iraq in the absence of United Nations monitoring and
controls to dig up and restart hidden EMIS devices vary. There is dis-
agreement about how well the EMIS device survives burial, and disagree-
ment about how long it takes to repair the fragile electronic and hydraulic
equipment that supports the EMIS device. That supporting infrastructure
was more heavily damaged than the EMIS devices themselves. The EMIS
devices themselves are constructed from heavy pieces of steel, are hard
to damage, and are not technologically difcult to build with appropriate
machine tools. Totally rebuilding that infrastructure at all known sites
would take about two years. However, considerable amounts of equip-
ment survived the bombing, and repairs to the damaged EMIS devices
themselves were estimated to take no more than twelve months.
16
Repairs to the EMIS devices or to other components of the Iraqi
nuclear program would have depended on the availability of machine
tools. EMIS devices and centrifuges for uranium isotope separation can
be built only with modern machine tools such as vertical and horizontal
ow forming machines and multiple axis, computer-controlled milling
and boring machines. IAEA inspectors who were experts in machine tools
surveyed the sites where such machine tools were known to have been
in operation. Many of these machine tools, by virtue of the heavy work
they perform, are themselves constructed out of heavy pieces of steel, and
are inherently hardened against bombing. IAEA inspectors visited sites
with bombed buildings that housed machine tools still in operable con-
dition. They also saw sites from which heavy machine tools had obvi-
ously been removed, and large stockpiles of spare machine tools. In April
15. Report of the fourth IAEA inspection trip in Iraq, July 27August 10, 1991.
16. Ibid.
chapter 5 137
1992, IAEA inspectors identied 122 machine tools which have technical
characteristics required for producing key components needed in a nu-
clear program. All ten of the computer-controlled machine tools that
were known to have been delivered to Iraq before the war and that would
be needed to make key parts of centrifuges for isotope separation sur-
vived the bombing. IAEA inspectors found fty-four boring mills that
would be used to make EMIS disks, and estimated that there were an
estimated one hundred such mills in Iraq in 1992. In Al Atheer, the special
presses needed to produce the high explosive lenses for implosion-type
atomic bombs survived the war. The extent to which Iraq protected itself
against bombing by buying machine tools and storing them away from
existing factories is suggested by an IAEA report that an Iraqi factory that
had been leveled by bombing was by June 1992 fully rebuilt and
equipped with 150 theretofore unknown milling machines; it was busy
making artillery shells.
17
Sixth, nuclear reactors still played a role in the Iraqi program, and
were inherently hard to hide, protect, or make redundant. Although the
Iraqis were not relying on reactors for the production of plutonium for
use as the ssile material in a bomb, their reactors were still important
for the production of short-lived isotopes of hydrogen, polonium, and
plutonium, all useful as neutron initiators to begin the ssioning of the
core of an atomic bomb at the proper moment. The Iraqi research reactor
used for the production of those isotopes was destroyed by bombing.
Rebuilding that reactor would take, under favorable conditions, an esti-
mated ve years. The impact of the destruction of that reactor is harder
to estimate. The necessary isotopes could have been produced and stock-
piled before the war, but would decay in a few years or less. However,
devices are commercially available that can produce neutrons for initia-
tion without the use of isotopes. The use of such devices imposes some
design problems on weapons builders. The Iraqis showed IAEA inspec-
tors neutron initiators on which they were working, but it is unlikely that
they were completely candid about the true extent of their competence
in this area. As a result, the full impact of the destruction of the Iraqi
nuclear reactor cannot be condently assessed.
Finally, in many ways, the most important component of the Iraqi
nuclear weapons program remains the highly skilled people working in
it who had solved many problems and gained the expertise that would
aid them in solving others.
18
Iraq invested at least as much time and effort
17. Report of the eleventh IAEA inspection in Iraq, April 715, 1992.
18. For a discussion of the importance of the availability of highly competent scien-
tists and engineers to regional powers, like Iraq, which are working on an atomic
138 the coming crisis
in creating redundancy in this cadre of technical workers as it did in
creating redundancy in its physical plant. For example, teams of Iraqis
were brought in to operate the original EMIS facility at Tuwaitha and
then were moved out to operate another EMIS facility at Tarmiya, while
a new team was put in to train at Tuwaitha.
19
Approximately 20,000
people were estimated to have been employed in the clandestine Iraqi
nuclear weapons program.
20
It is not clear that all key members of that
program have been identied. There is, for example, the possibility that
Iraq had a large program concentrated on the use of centrifuges for
isotope separation that was staffed by a group of people separate from
the major program and facilities. Given the lack of information about the
identities of these people and the ease with which people can be dis-
persed and protected against bombing, the impact of the bombing on the
Iraqi nuclear weapons program personnel must be assumed to have been
minimal.
hypothetical iraqi nuclear weapons in 1990
If the available data suggest that the United States would have had
considerable difculty in destroying or otherwise neutralizing Iraqi nu-
clear weapons capabilities, they do not tell us how Iraq might have
employed any nuclear weapons that it might have had but could not
actually construct before the war began. For the purposes of this exercise,
some arbitrary assumptions must be made about what the Iraqi nuclear
weapons program might have achieved in 1990 and 1991. To avoid debate
about what the Iraqis really might have achieved and when, questions
that are important but not central to this exercise, this chapter simply
stipulates a notional Iraqi nuclear force that represents the kind of re-
gional nuclear threat that the United States might have to face at some
time in the future. The Iraqi nuclear capability will be assumed to have
been between ve and ten ssion weapons that could have been launched
by ballistic or cruise missiles. This particular capability was certainly not
available to Iraq in 1991. (The exact time when Iraq could have achieved
this capability is, once again, an important question but not central to this
paper.) Iraq would have had to have developed a nuclear warhead that
could have been moved around with a mobile missile and then mated
with it when the time for launch came. It is worthwhile noting that the
United States had developed such a weapon, the Mk-7/W-7 weapon,
bomb, see Russell Seitz, Interesting Times, Olin Institute for Strategic Studies, Mono-
graph No. 2.
19. Report of the fourth IAEA inspection team in Iraq, July 27August 10, 1991.
20. David Kay, Iraq and Beyond.
chapter 5 139
based on highly enriched uranium, the same ssile material Iraq was
trying to produce. The missile warhead version of this warhead was
available to the United States in 1954 for delivery by submarine-launched
cruise missiles and land-based ballistic missiles.
21
The Iraqis would have
to have solved to their satisfaction the civil- military problems of com-
mand associated with giving Iraqi military commanders live nuclear
weapons and the means to deliver them over long distances, and then
giving them the freedom to move the weapons about. Ensuring that these
commanders, who may have been harboring grievances toward their
own regime or others, or who might have entrepreneurial ambitions,
would re their weapons only when told to and only at the target
intended by the government, is not a trivial problem. We know from the
Soviet case that despotic regimes have been able to reconcile the problems
of ensuring the survival of nuclear weapons with their particular civil-
military problems of nuclear command and control. Since those methods
usually involved the uses of secret police and military forces to keep the
nuclear forces in check, rather than sophisticated technology, there is no
reason to assume that there would be insuperable technical barriers to
Iraqs doing likewise.
22
What might the deterrent value of nuclear weapons have been to Iraq
at the time of the 1990 invasion of Kuwait and in the 1991 war? The U.S.
Army and Air Force were able to deploy massive forces and the logistics
to support them to a relatively small number of bases in the Persian Gulf
area. The Navy was able to send multiple carrier battle groups into the
Persian Gulf. The existence of Iraqi nuclear weapons could have deterred
or interfered with this deployment in two ways. First, Iraqi nuclear
weapons could have deterred countries like Saudi Arabia from allowing
the United States access to its bases. How could the United States have
reacted if the Saudis had decided not to allow U.S. forces on their soil
because they feared Iraqi nuclear attack? The United States could have
threatened nuclear retaliation against Iraq if it attacked Saudi Arabia with
nuclear weapons, and could have threatened to carry out that threat with
long-range forces not in the theater. Would such a threat have been
21. See the discussion of the 1,700-pound W-7 warhead, which went into production
in August 1954 and was the warhead for the U.S. Armys Corporal and Honest
John ballistic missiles, in Chuck Hansen, U.S. Nuclear Weapons: The Secret History
(New York: Orion Books, 1988), pp. 133136, 180.
22. See Stephen M. Meyer, Soviet Nuclear Operations, in Ashton Carter et al., eds.,
Managing Nuclear Operations (Washington, D.C.: Brookings, 1987), pp. 490493, for a
discussion of the KGB role in controlling nuclear weapons that were under the control
of Soviet military commanders.
140 the coming crisis
credible to the Saudis and to the Iraqis? Such a question cannot be
answered denitively, but some observations can be made.
First, despite cultural and political ties to Western Europe that were
stronger than those to Saudi Arabia, the United States was never able
entirely to reassure the governments of Western Europe that the Strategic
Air Command, by itself, was sufcient to deter Soviet nuclear attack on
Western Europe, once the Soviet Union acquired even very modest and
uncertain means of attacking U.S. cities with nuclear weapons. A combi-
nation of forward deployed non-nuclear and theater nuclear forces were
consistently deployed over a period of almost forty years in order to
neutralize Soviet nuclear threats and to reassure U.S. allies in Europe that
Soviet nuclear attack was deterred; however, the government of Japan
never felt the need for U.S. nuclear weapons on Japanese soil. Allies of
the United States facing a nuclear Iraq, Iran, or North Korea might be
harder to reassure than Great Britain, France, and Germany, particularly
if nuclear threats against U.S. allies were coupled with implicit fears of
nuclear attacks on U.S. cities if the United States used nuclear weapons.
Even if the enemy had only marginal capabilities to attack the United
States, this threat would create considerable uncertainties within its alli-
ances. The United States would be in the difcult position of trying to
prove to its allies either that the common enemy had no means of
attacking the United Statesfor example, that it had never been and
would never be able to introduce a clandestine nuclear weapon into the
United Statesor that the United States would be willing to use nuclear
weapons if Riyadh were attacked, even if it meant losing New York.
Second, even if the United States were able to convince its regional
allies that it was safe for them to allow it to use bases on their soil, the
leadership of the U.S. military would itself have considerable doubts
about the wisdom of concentrating hundreds of thousands of U.S. sol-
diers and airmen in a few bases within range of enemy nuclear weapons.
Even though the United States could have retaliated with nuclear weap-
ons if the Soviets had conducted nuclear attacks against U.S. forces in
Western Europe, the U.S. Army consistently worried about the military
problems created by the presence of enemy nuclear weapons that could
strike at troop concentrations. While the army never developed a totally
satisfactory solution to this problem, it invested considerable time and
money in partial solutions ranging from the Pentomic division to heli-
copter air mobility to Air-Land Battle.
23
Similarly, the U.S. Navy spent
23. Kevin Patrick Sheehan, Preparing for an Imaginary War (Ph.D. dissertation,
Harvard University, 1988).
chapter 5 141
billions of dollars and structured its eet tactics to limit the ability of the
Soviet Union to strike successfully at concentrated naval formations with
nuclear weapons. Despite the deployment of the F-14/PHOENIX system,
and Aegis air defenses, the reluctance of the carrier battle group com-
manders to operate in closed waters when within range of enemy nuclear
weapons suggests that these costly programs were at best partial solu-
tions to the problem of potential nuclear attack. Even if the U.S. govern-
ment had adopted a declaratory policy of striking at Iraq with nuclear
weapons if Iraq used nuclear weapons against bases on land or against
U.S. Navy ships at sea, prudent military planners would very likely have
altered their deployment and operational plans to take into account the
existence of Iraqi nuclear weapons. These alterations would have reduced
the ability of the United States to concentrate its forces, increased the
amount of resources devoted to defense, rather than offensive strike
capabilities, and imposed signicant virtual attrition on the expeditionary
force.
Would the United States have been able to circumvent the difculties
associated with nuclear threats against our regional allies and U.S. mili-
tary personnel within range of a hostile nuclear-armed power by execut-
ing a successful preemptive strike against Iraqi nuclear weapons before
initiating Desert Shield, that is, before the deployment of U.S. tactical
aviation to bases in the region? A preemptive attack conducted by U.S.
military units stationed outside the Persian Gulf area would have faced
considerable difculties. Only B-52 and F-111 bombers had the necessary
range for such missions, and it is not clear that they had the sensors for
the kind of reconnaissance-strike missions later conducted by F-15s and
supporting aircraft during Desert Storm. No F-111s remained in the U.S.
Air Force inventory as of 1998. And even the reconnaissance-strike team
of F-15s, as we have noted above, was unable to suppress Iraqi Scud
ballistic missiles on mobile launchers. As noted above, the lesson of the
1991 war was that if Iraq had been willing to mate its nuclear weapons
with its ballistic missile launchers, mix those launchers with non-nuclear
launchers with similar signatures, and then disperse the ballistic missile
force and keep it mobile, it is difcult to see how the United States could
have had condence in its ability to destroy even a high proportion of
the deployed Iraqi nuclear weapons.
It is entirely conceivable that Saddam Hussein would have been
reluctant to put operational, nuclear-armed ballistic missiles into the
hands of several Iraqi military commanders left to their own devices in
the Iraqi desert. With such power in their hands, and under the command
of a despot who was not well liked, Iraqi military commanders with
nuclear weapons might not have proved completely reliable. It might
142 the coming crisis
well have been difcult for Iraq to implement command and control
arrangements to prevent unauthorized use of nuclear-armed, mobile bal-
listic missiles that would also have ensured prompt execution of plans
while under attack, if necessary. Such considerations might well have led
Iraq to keep its nuclear weapons in concealed, very hard storage sites.
Such sites would be very difcult for the United States to nd. In fact, in
January 1991, the U.S. list of nuclear-related targets in Iraq consisted of
two sites. At the end of the war, UN inspectors operating in Iraq identied
over twenty nuclear-related facilities in Iraq, sixteen of which were des-
ignated as main facilities.
24
A separate question would be the ability of
the United States to have successfully attacked superhardened sites with
non-nuclear weapons. Available evidence suggests that by the end of the
war, the United States had constructed special kinetic energy penetrators
that could destroy deep underground bunkers. Tunnels into mountains
would require a kinetic energy penetrator either to penetrate hundreds,
perhaps thousands of feet of rock, or to execute difcult maneuvers in
order to fall vertically and then turn and y horizontally into the tunnel
entrance. Cruise missiles might y into tunnel entrances and seal the
known entrances into a nuclear weapons storage site. This might be
adequate to neutralize the nuclear weapons for a period, but with enough
time, the weapon could be recovered. Fully neutralizing the weapon
would require occupying the site and taking the weapon out.
A tentative conclusion is that the United States would have faced
considerable difculties in executing a preemptive attack with weapons
based outside the region that would have given the U.S. president high
condence that all or most of the Iraqi nuclear weapons had been de-
stroyed. The inability of the United States to conduct such a preemptive
disarming attack could have deterred U.S. military deployments to the
region, and certainly would have forced U.S. military planners to take
precautions that would have reduced the offensive military capabilities
of the expeditionary force. It also seems fair to conclude that such a
disarming attack would still have had uncertain prospects even after the
U.S. Desert Shield build-up had been completed.
Thus, the overall lessons learned from the 1991 war by regional
powers contemplating military action threatening U.S. interests might be
that developing nuclear weapons creates only a small risk of U.S. military
action before a nuclear weapon is complete, particularly if adroit diplo-
macy is employed; that much nuclear weaponsrelated activity is con-
cealable, even from intrusive inspections, that nuclear weapons can create
considerable uncertainties within U.S. alliance relations that might pre-
24. Keeney and Cohen, Gulf War Air Power Survey, p. 79.
chapter 5 143
vent the U.S. use of regional bases, and that a substantial proportion of
the nuclear weapons of Iraq might not have been easily destroyed by a
preemptive U.S. attack employing non-nuclear weapons. As a result, a
U.S. president might decide not to deploy forces to the region, or might
deploy a force with reduced offensive military power that would require
more time and face less favorable force ratios when engaging a regional
nuclear power. All of these tentative conclusions could and should be
tested by focused studies employing the best available political and
technical intelligence and commentary from U.S. military commanders
with operational responsibilities.
The 1991 War and Hypothetical Nuclear Weapons Use
The previous section addressed what might have happened if Iraq had
simply possessed nuclear weapons when it invaded Kuwait in August
1990. This section explores the problems and issues the United States
would have faced if Iraq had actually used nuclear weapons. In the
absence of any concrete information about Iraqi nuclear weapons em-
ployment doctrine, the discussion proceeds on the assumption that the
government of Iraq would have had the following set of strategic objec-
tives. These objectives would apply as well to other expansionist regional
powers with nuclear weapons. First, Iraq would have sought to deter or
prevent the U.S. military build-up in the region, which gave the United
States effective military options against it. Second, if Iraq had failed to
prevent the U.S. build-up, it would have sought to deter U.S. employment
of its military forces in the region. Third, it would have sought to mini-
mize the likelihood that a U.S. president would use nuclear weapons
against Iraq. Fourth, while pursuing all of these objectives, the govern-
ment of Iraq would have tried maintain central control over Iraqi nuclear
weapons, out of fear that Iraqi commanders might not be entirely reliable
custodians of the few nuclear weapons in the Iraqi arsenal, and to make
sure that if Saddam Hussein issued an order to use the nuclear weapons,
his order would be implemented quickly and faithfully.
To be sure, this set of objectives represents an idealized set of Iraqi
objectives that are not entirely consistent with Iraqs behavior during the
crisis. For example, Iraq did not mount armored attacks on Saudi Arabia
immediately after invading Kuwait, though by doing so Iraq might have
been able to deny the United States the areas in which it later built up
its forces. Saddam Hussein did not fully exploit the confused diplomacy
of the period from August 1990 through January 1991 to deny the United
States a solid political basis for military action. The set of objectives,
therefore, represent a worst case in which a hypothetical nuclear-armed
144 the coming crisis
Iraq acts as a completely informed, rational actor that does everything
within its power to frustrate U.S. strategies, and tests the capacity of U.S.
military forces and strategy to the limit.
How might Iraq have translated the broad objectives listed above into
military plans? Four operational goals for Iraq are consistent with the
objectives listed above.
First, the timing of the revelation of Iraqi nuclear weapons capabili-
ties would have been important. If Iraq had announced its nuclear weap-
ons capabilities before invading Kuwait, the United States presumably
would have become more alert and more focused on the possibility of
Iraqi aggression. U.S. diplomacy and military preparations would have
been affected by the announcement, and Iraq might never have been
presented with what it perceived as a favorable opportunity to invade
Kuwait. But if Iraq invaded Kuwait and then credibly revealed its nuclear
weapons capabilities, the consequences might have been serious. Iraq
might have used nuclear weapons to create the impression within Turkey,
Saudi Arabia, and other countries in the region that it was dangerous to
those countries to allow the United States the use of their military bases
in response to the Iraqi invasion of Kuwait. There would be a natural
tension between the United States and its regional allies as a result of the
fact that the United States would be out of range of nuclear-armed Iraqi
missiles and aircraft, while the allies of the United States were in range
and vulnerable to nuclear attack. Before the Desert Shield buildup was
complete, there were tensions between the United States and Saudi Ara-
bia because of the differential in non-nuclear Iraqi threats to Saudi Arabia
as compared to the United States. Iraqi nuclear weapons capabilities
would have exacerbated that tension.
Second, Iraq might have used nuclear weapons to maximize the
perception in the United States that military action against Iraq was
extremely dangerous and carried the chance of very high U.S. casualties
and ultimate political failure. While U.S. leaders would have been
condent that U.S. nuclear superiority would make possible a nuclear
response to Iraqi nuclear weapons use, such a response might not have
prevented hundreds of thousands of U.S. casualties and a politically
disastrous outcome for the United States in the Middle East. Iraq might
be able to create a perception of such risks without mounting nuclear
attacks on U.S. forces. An Iraqi detonation of a nuclear weapon plus the
dispersal of nuclear-capable delivery systems within Iraq, for example,
could have created the impression that Iraq could inict massive damage
on U.S. forces and that a U.S. nuclear counterforce strike in retaliation
would also have to be massive. This perception could have been maxi-
mized by deceptive Iraqi practices designed to encourage U.S. assess-
chapter 5 145
ments that Iraqi nuclear forces were larger than they were or larger than
they had previously been estimated to be. Multiplying units with the
same signatures as nuclear weapons units and revealing the existence of
previously clandestine nuclear material production sites are two obvious,
hypothetical deceptive options. Imagine the impact that an Iraqi nuclear
test in August 1991 would have had along with the revelation of the
Calutron facilities for uranium enrichment, if outside observers could not
go and assess the production capacity of the facilities.
The fact that U.S. perceptions might have been manipulated by nu-
clear explosions without an attack on U.S. forces is important because the
third Iraqi operational goal would likely have been to avoid the use of
nuclear weapons against U.S. military personnel, the military personnel
of close U.S. allies such as Israel, and, perhaps, to avoid the use of nuclear
weapons against civilians. This operational goal would be adopted if Iraq
chose to minimize the likelihood that the United States would use nuclear
weapons against Iraq. Nuclear attacks against politically salient targets
such as U.S. or Israeli troops would practically guarantee U.S. military
retaliation.
Fourth, and nally, Iraqi operational plans for nuclear weapons use
would likely have been simple, tightly scripted, and preprogrammed in
order to cope with the command and control problems faced by a despot
with potentially unreliable commanders, few nuclear weapons, and a
communications system that is vulnerable to military attack. Nuclear-
armed ballistic missile units operating from xed, concealed locations,
with the coordinates of xed targets locked into the guidance system,
under the command of an elite military unit commanded by a relative of
the ruler, guarded by secret policemen reporting back to the ruler through
separate communications channels, prepared to execute a set of simple
shoot/dont shoot instructions communicated by secure land lines were
more likely to have been characteristic of Iraqi military plans than
nuclear-armed aircraft on airborne alert with orders to search for and
destroy U.S. carrier battle groups at sea or maneuvering ground units.
Given these objectives and constraints, what kinds of nuclear attacks
might Iraq have conducted? The nuclear option with the fewest material
effects might have been the option with the greatest net political benets
to Iraq. Iraq could have detonated a nuclear weapon over the territory of
Saudi Arabia in the period August through October 1990 that inicted no
major damage on military or population targets, but which would have
dramatically demonstrated Iraqi nuclear capabilities before the United
States had been able to build up operationally signicant military capa-
bilities in the region. What would have been the advantages and dis-
advantages to Iraq from such an initial use of nuclear weapons? Such
146 the coming crisis
an attack would certainly have delayed U.S. military intervention, and
would have complicated U.S. military deployments if they had pro-
ceeded. Compared to other conceivable Iraqi uses of nuclear weapons, it
would have given the United States fewer reasons to use nuclear weap-
ons of its own. It would have been well within the assumed limits of the
Iraqi command and control system. The demonstration shot would, at a
minimum, have brought about a pause in the deployment of U.S. and
allied forces to Saudi Arabia while the situation was being claried and
reconsidered by the governments of the United States and Saudi Arabia.
The Saudi government would certainly have asked the U.S. government
for guarantees and protection against subsequent nuclear attacks if the
Saudis continued to allow the military buildup on their soil. In the United
States, as in past military crises, the Congress and the U.S. people would
have rallied around the commander-in-chief and, at least initially, would
have given him their support in whatever policy he chose to pursue. But
if the crisis was not resolved within a week or so, debate within the
United States military, the executive branch, and in the larger political
system would have become intense.
On the negative side for Iraq, the attack would have alerted the world
to its nuclear capabilities and willingness to use them, and this would
very likely have solidied universal diplomatic hostility to Iraq. This, in
turn, might have facilitated the development of the anti-Iraqi coalition.
U.S., Israeli, and perhaps even Soviet inhibitions on the use of nuclear
weapons against Iraq would have dropped.
Once the initial shock had passed, how might the United States have
reacted to such a demonstration shot? One option would have been
simply to continue the buildup of non-nuclear U.S. military forces in the
region, and to declare that any Iraqi nuclear attack on U.S. or allied
civilian or military personnel would be met with a full-scale U.S. nuclear
retaliation. In effect, this strategy would declare that Saddam Husseins
nuclear threat was a bluff, and would call the bluff. This might or might
not have been accompanied by a tit-for-tat demonstration of U.S. nuclear
capabilities. If the Saudi government hesitated or refused to issue an
invitation to the U.S. forces to deploy to Saudi territory, the United States
could have moved in without permission. Deployments would have
continued, but with greater care to disperse and defend U.S. forces
against ballistic or cruise missile attack. At the operational level, the
deployment would have proceeded in a more orderly fashion if U.S.
military planners had already explored the problems of deploying an
expeditionary force against a regional nuclear enemy. If Saddam Hussein
declared in response that he had already managed to introduce a clan-
destine nuclear weapon into the United States, this would also be treated
chapter 5 147
as a bluff, since any nuclear destruction of a U.S. city would certainly
result in the nuclear destruction of Iraq.
This option would call for a high tolerance for the risk of nuclear
destruction, perhaps more tolerance than major U.S. political actors or
our regional allies might possess. If the Saudis were not willing to go
along with the United States, this would add the costs of acting against
the wishes of a friendly Arab state. The ability of the United States
government to reduce the risk of nuclear attack by military preemption
would be problematic. In this scenario, the United States would not yet
have deployed sufcient non-nuclear air power in the region to execute
any effective air strikes against Iraqi nuclear weapons. A U.S. nuclear
strike against Saddams nuclear weapons, while within U.S. capabilities,
might have been politically difcult to execute, particularly if uncertain-
ties about the location and hardness of the Iraqi weapons storage sites
forced the United States to conduct a fairly massive nuclear attack to
ensure itself that most of the Iraqi weapons were destroyed. For example,
let us say the United States delivered 100 nuclear weapons by cruise
missile with yields in the 200-kiloton range to destroy a maximum of ten
estimated actual weapons in individual storage sites, with ve decoy
storage sites for every real site, and two U.S. weapons per target. If the
nuclear weapons storage sites were collocated with Iraqi population
centers, the United States would face a greater dilemma. Other means of
reducing the perceived risk of nuclear attack on U.S. allies might include
the deployment of nuclear offensive systems to Saudi Arabia to couple
any Iraqi nuclear attack on Saudi Arabia to a U.S. nuclear response by
forces in theater, or the deployment of U.S. systems to defend against
Iraqi nuclear attack. In the aftermath of an Iraqi nuclear weapons dem-
onstration, the use of nuclear warheads on U.S. defensive systems, in-
cluding Patriot, Phoenix, and Aegis/Standard 2 systems, might have been
both politically acceptable and necessary to enhance the real and per-
ceived effectiveness of U.S. defenses against Iraqi nuclear attack. If such
warheads were available, if plans had been made to deploy them, and if
exercises had familiarized crews with the modes of employment for
nuclear-armed defensive systems, this option would have been more
readily available to U.S. leaders to help them reduce perceived and actual
risks of Iraqi nuclear attack.
Another option might have been a U.S. ultimatum to Iraq to with-
draw from Kuwait or face U.S. nuclear attack. The kind of nuclear attack
needed to carry out the threat inherent in this response would have been
well within the capabilities of U.S. nuclear weapons outside the region,
but even more politically difcult to execute than continued military
deployments to the region.
148 the coming crisis
Moving one rung up the escalation ladder, once Iraq decided to take
the risk of employing nuclear weapons at all, it might have decided that
its objective of denying the United States the ability and desire to inter-
vene militarily in the region would best be served by nuclear attacks on
the few ports and airelds in the region through which U.S. forces would
have to approach the theater. Given the limited number of harbors in the
region that can unload heavy equipment, such an attack might have
halted the deployment of U.S. armored forces until the ports were decon-
taminated and repaired. Coupled with nuclear attacks on Saudi Arabias
airelds with runways and logistics support facilities to handle C-5 class
aircraft and heavily laden F-15s, this attack could have created physical
barriers to U.S. military intervention that could have lasted for many
months, if not years. Preplanned strikes against xed targets would,
again, have been within the assumed constraints on Iraqi command and
control.
On the downside for Iraq, such an attack would presumably have
killed thousands of Saudi Arabians. While Saddam Hussein displayed
little reluctance to kill his fellow Arabs in Kuwait, and might even have
wished to demonstrate his ruthlessness against his enemies, a nuclear
attack on centers of population would have reduced the political con-
straints on U.S. use of nuclear weapons against Iraq.
What might the U.S. option have been in this scenario? Since the
attack would severely constrain the deployment of conventional forces to
the region, Iraqi forces would still be in control of Kuwait. Nuclear strikes
against Iraq by U.S. forces based in or out of the region might be threat-
ened, and those threats would have had somewhat more credibility than
in the rst scenario because of the civilian casualties in Saudi Arabia. But
what targets might the United States have credibly attacked? Iraqi occu-
pation forces in Kuwait would be inappropriate targets. Iraqi ports or
urban-industrial centers could have been destroyed in tit-for-tat repri-
sal/punishment attacks, though this might well have been politically
difcult for a U.S. president to order, particularly if the initial Iraqi attack
had killed only a few U.S. military personnel in Saudi Arabia. In the rst
days after the Iraqi attack, U.S. blood would be boiling for nuclear
retaliation, but after that, the United States would debate the morality
and strategic utility of killing tens of thousands of Iraqi civilians, and of
weakening the psychological restraints on the use of nuclear weapons
even further, the future of U.S. relations in the region after a nuclear
attack on Arab cities, and so on. There would, therefore, very likely have
been a search for Iraqi military targets that were not near Iraqi centers of
population: nuclear weapons storage sites, ballistic missile deployment
areas, elite military units, lines of communication from Iraq to Kuwait.
chapter 5 149
Once again, there might have been hesitation to use nuclear weapons
against these targets because of the risks outlined in the section above,
particularly if Saddam had chosen to place his nuclear weapons produc-
tion and storage sites near population centers. But even if these hesita-
tions had been overcome and the attacks had been executed, Iraq would
still have been in possession of Kuwait and Baghdad; Iraqi oil production
centers would still have been intact; Saddam Hussein would still have
remained in power; and the United States would still have faced physical
obstacles to near-term non-nuclear military action against Iraq.
What might the United States have done then? There would appear
to be few options other than to conduct whatever nuclear reprisal/pun-
ishment attacks seemed appropriate, use nuclear weapons against Iraqi
nuclear weapons where possible, proceed slowly either to reconstruct the
air and beachheads into Saudi Arabia or build new ones, build up U.S.
military forces in the theater, and then liberate Kuwait. Faced with such
a task stretching over months in a postnuclear war world, it is unclear
what the national interest or the actions of the United States would have
been.
In an unlikely variant of the scenario just reviewed, Iraq could have
attacked not only Saudi ports and airelds, but also the Saudi National
Command Authority. This might even further reduce political constraints
on the use of U.S. nuclear weapons against Baghdad. It would also create
a disturbing problem for the United States: The Saudi government would
have disappeared, and it would be unclear who controlled the country
and who the U.S. allies were. Very few U.S. troops would have been
killed. If the United States had decided to intervene militarily with con-
ventional forces, it would have had to intervene unilaterally, as much to
restore order in Saudi Arabia and to prevent the emergence of a hostile
regime as to liberate Kuwait. In this scenario, a nuclear attack on Iraqi
military forces would not have liberated Kuwait and would have left the
Iraqi government in power. If, despite the absence of U.S. casualties, the
United States had destroyed Baghdad with nuclear weapons in retali-
ation, it would not only have killed tens of thousands of civilians, but
would also have faced the need to intervene in Iraq to restore order there
as well. An Iraqi leader willing to think all this through and to take his
chances might have concluded that the United States would be self-
deterred from taking any military action in response to the nuclear de-
struction of the Saudi government. Indeed, it is difcult to construct any
interesting U.S. policy options for this scenario.
The next level of Iraqi nuclear attack might have involved attacks on
U.S. forces after they had arrived in the region. This is one of the least
likely scenarios because it would virtually have guaranteed U.S. nuclear
150 the coming crisis
attacks on all Iraqi military units and political command posts, for emo-
tional as much as for strategic reasons.
An Iraqi detonation of a man-portable suitcase bomb or bomb in
a merchant ship or light civilian aircraft in a major U.S. city, as opposed
to threats to execute such attacks, represents the least likely of all scenar-
ios. Nuclear weapons, by logic and common agreement, have severely
limited strategic utility, but they are very good at deterring nuclear attack
on cities. This is true whether the weapons are delivered by ballistic
missile or by suitcase. For Iraq to extract any strategic benet from the
use of nuclear weapons smuggled into the United States, Iraq would have
had to identify itself as the party responsible for the destruction of a U.S.
city. It is impossible to imagine any circumstances in which the U.S.
government would not have executed a massive retaliatory strike in
response to such an attack, and impossible to imagine circumstances in
which the certain knowledge of retaliation would not have deterred this
kind of attack by Saddam Hussein. Threats of such an attack to frighten
the United States, to complicate the tasks of the U.S. government, and to
force it to divert resources into searches for such a weapon are, however,
completely plausible.
This extended hypothetical consideration of regional nuclear weap-
ons use utilizes a Middle Eastern enemy and allies, but its logic is not
conned to such allies and enemies. Middle Eastern nuclear enemies may
well try to intimidate U.S. allies in Europe who provide the military bases
on their soil that make U.S. military intervention in the Middle East
feasible. Northeast Asian nuclear enemies of the United States may try to
intimidate Japan to prevent the U.S. use of military bases in that country.
The essential logic of the problem is that U.S. regional military alliances
have been based on credible U.S. military guarantees of the survival of
the regional ally, in return for basing rights. Regional nuclear proliferation
undermines that asymmetrical alliance relation wherever it exists.
Nuclear Weapons Use and U.S. Policy
How is the U.S. policy community responding to the problem of possible
nuclear weapons use? A survey conducted by David Andre reviewed
political military simulations conducted by or for twenty U.S. govern-
ment agencies involving the use of nuclear weapons by hostile regional
powers. A second survey, conducted by RAND, was based on a series of
simulations conducted by RAND specically to elicit the views of the
policy community toward the problem of regional nuclear weapons use.
Twelve named analytical institutions and several unnamed organizations
sent a total of 239 of their members to participate in the RAND simula-
chapter 5 151
tions. Both surveys found an overwhelming reluctance by the partici-
pants in the simulations to recommend the use of U.S. nuclear weapons
in response to a wide variety of simulated nuclear weapons use. There
was a strong preference to seek nonmilitary solutions to nuclear crises,
to avoid engaging U.S. military forces in the crisis at all, and to use
long-range, accurate, non-nuclear strike forces if the United States had to
act militarily. There was a notable lack of consensus that it was in the
interest of the United States to punish nuclear proliferation or regional
nuclear weapons use.
25
The survey data appears to reveal a disinclination by the U. S. defense
policy community to come to grips with the difculties of handling
hostile regional nuclear powers. This disinclination is reected in U.S.
military doctrinal publications in which the problem of hostile regional
nuclear weapons use is generally handled with broad assumptions and
wishful thinking. The most recent version of the basic U.S. Army eld
manual for operations, FM-100-5, published in 1993, simply states that
the problem of nuclear weapons use by a hostile regional power may best
be handled by rapidly bringing the war to a conclusion favorable to the
United States before the enemys nuclear weapons can be brought to
bear.
26
A 1994 U.S. Army Training and Doctrines Command (TRADOC)
publication that looked forward at military operations in the twenty-rst
century, TRADOC pamphlet 525-5, notes the phenomenon of the prolif-
eration of nuclear weapons. Though it deals in some detail with the
changes in military operations that will be the result of improvements in
non-nuclear warfare, and particularly in information technologies, the
pamphlet is silent on how the U.S. Army might conduct military opera-
tions against a regional nuclear power.
27
implications for u.s. policymakers
If the United States is not likely to be able to disarm a hostile nuclear
power by force, if it is reluctant to use military force at all against regional
25. David Andre, The Third World Few Nuclear Weapons Problem: Policy Issues
and Implications for Military Planning Guidance as Derived from Gaming (McLean,
Va.: SAIC, April 1993); and Marc Dean Wilmot, Roger Molander, and Peter Wilson,
The Day After . . . Study: Nuclear Proliferation in the PostCold War World (Santa
Monica, Calif.: RAND, 1993), vols. I and II.
26. U.S. Army, Field Manual 100-5 Operations (Washington, D.C.: Headquarters, De-
partment of the Army, 1993) pp. 610.
27. U.S. Army Future Battleeld Directorate, TRADOC Pamphlet 525-5: Force XXI
Operations (Fort Monroe, Va.: Headquarters U.S. Army Training and Doctrine Com-
mand, 1994), pp. 27, 319.
152 the coming crisis
nuclear powers and particularly unwilling to use nuclear weapons, what
is the result?
If regional states adopt policies of minimum deterrence to each other,
the results appear to be minimal. The increased security and stability of
the military relations between two nuclear-armed states with minimum
deterrence doctrines seem not to affect the interests of the United States
in any signicant way. In addition, U.S. military guarantees, assistance,
and intervention will be much less in demand from states that have
established stable nuclear deterrence. There will be little that the United
States will be able to do militarily to help or hurt nations locked in a
relationship of stable nuclear deterrence, short of offering to provide
credible rst strike capabilities or credible antinuclear defensesand
practical and political constraints on the United States will limit its ability
to do either. The inuence that the United States can exert over states in
this kind of relationship will be likely to decline.
States will be less able to draw the military intervention of the United
States by means of the demonstrative use of nuclear weapons when there
is no competitive superpower that might prot from the failure of the
United States to intervene in the aftermath of nuclear weapons use.
However, regional wars that threaten the existence of a U.S. ally remain
possible, and the use of nuclear weapons as a distress signal to the United
States remains possible.
If regional allies of the United States are presented with regional
nuclear threats, the United States will face the familiar problem of ex-
tended deterrence in a new context. Will the threat of U.S. nuclear retali-
ation protect a third party from nuclear or conventional attacks? In the
past, there were doubts, in the United States and Western Europe, about
whether the United States would actually use nuclear weapons on the
battleeld or against the Soviet Union if Western Europe were invaded
by non-nuclear Warsaw Pact forces or attacked by Soviet nuclear weap-
ons. Yet during the Cold War, the security of Western Europe was under-
stood to be vital to the security of the United States; as a result, the United
States was willing to station troops and nuclear weapons in Western
Europe and to adopt declaratory policies linking nuclear weapons use in
Europe to intercontinental nuclear war. After the Cold War, many of the
areas that are affected by regional nuclear weapons threats are likely to
appear much less vital to the United States, and less worth the risks of
nuclear war. Measures to extend U.S. nuclear deterrence to those regions
are likely to be much less vigorous and credible.
In the cases of both stable regional nuclear deterrence and hostile
regional nuclear threats, the result is likely to be a diminution in the
inuence that the United States can exercise over a regions politics by
chapter 5 153
military means. Over the next decade it is possible that there will be
additional nuclear proliferation, and little in the way of U.S. responses
that effectively reassure our allies that we remain credible and useful
allies. In consequence there could follow a weakening of the military
alliances of the United States with countries in regions where there are
regional nuclear threats, specically, South Asia, Northeast Asia, and
Southwest Asia. This consequence is consistent with the abstract logic of
nuclear deterrence. If nuclear weapons create absolute threats to national
survival, and there are no external protectors, countries will tend to
respond either by capitulation or by developing their own nuclear weap-
ons. This may restore stability, but the process weakens any existing
military alliances. Nuclear proliferation reduces the willingness of allies
to run risks for others, and it reduces the need for allies by states with
nuclear weapons. As a somewhat unexpected consequence of nuclear
proliferation, we should expect not a world full of nuclear war, but a
world in which alliances are weak or nonexistent. The consequences for
U.S. foreign policy may be signicant: U.S. citizens are habituated to
ghting overseas assisted by and to protect our allies. They are increas-
ingly reluctant to pay the costs of unilateral military action, and eager to
nd justication for their military actions in the support of their allies.
Nuclear proliferation and a world without allies may make U.S. military
intervention abroad, for good or ill, increasingly difcult.
implications for u.s. military planning
One possible reaction to the arguments developed in this chapter is that
a fundamental change in the pattern of U.S. military alliances after the
demise of the Soviet Union is natural and acceptable, though possibly
uncomfortable or even fatal for some former allies. Hence, no major
changes in U.S. military policy are called for. Another possible reaction
is that the level of international conict and chaos that would accompany
this change in alliance relations is unacceptable because the direct human
costs to the regional powers would be appalling and the indirect costs to
the United States would be large, though hard to predict. How robust is
the U.S. militarys ability to disarm small nuclear powers?
The ability of the United States to respond effectively to small nuclear
powers with offensive and defensive military measures has been affected
by the virtual denuclearization of U.S. power projection forces. Given the
hardness and position location uncertainties that are currently associated
with anticipated regional nuclear forces, and given the difculty of inter-
cepting regional ballistic and cruise missile capabilities, the U.S. use of
smaller nuclear weapons for military purposes, offensive and defensive,
appears to have some value from an operational military perspective. The
154 the coming crisis
U.S. military, for a variety of reasons, has reduced and is reducing its
deployment of such weapons. A proper assessment of the impact of this
denuclearization on the ability of the United States to respond to regional
nuclear powers is beyond the scope of this paper, but appears to be
negative.
The ability of the United States to respond militarily to smaller
nuclear powers is also being affected by military research and develop-
ment. Over the longer term, U.S. capabilities may be changed by what
has been called a revolution in military affairs. In the near term, the
ability of the United States to defend against medium- and theater-range
ballistic missiles may be improved by the successful completion of pro-
grams such as the Theater High-Altitude Area Defense system, improved
versions of the Aegis system, and airborne laser systems for use against
cruise and ballistic missile targets. Over the longer term, the integration
of offensive and defensive systems into a theater anti-missile system
appears to be of some promise, and is being investigated by both the
United States and Israel.
28
The United States may be able to use new
technologies to collect, process, and distribute information for the pur-
poses of neutralizing a small hostile nuclear power. Sensors on aerial
platforms, both manned and unmanned, unmanned ground sensors, and
other inputs can now be integrated with greater rapidity and coherence
to provide, in the absence of enemy countermeasures, a radically im-
proved picture of the targets the United States may wish to strike. Linked
with long-range precision strike capabilities, deep penetration light
ground forces, and other systems, the combination of improved informa-
tion and coordinated offensive and defensive antimissile forces may yield
greatly improved results; such systems are being explored by the United
States. Radically improved integration, management, distribution, and
utilization of target-tracking data for eet air defenses are being explored
by the United States Navy. Operational concepts for moving an expedi-
tionary force with dramatically fewer soldiers close to a target while
under the defensive umbrella of eet air defenses, then establishing
ground-based antimissile defenses, and then maneuvering in a fashion
designed to force the enemy to deploy and expose its nuclear weapons
are being explored by the U.S. Marine Corps. These experimental techni-
cal and operational capabilities are far from operational status, and may
28. See, for example, General John M. Shalikashvili, Joint Vision 2010 (Washington,
D.C.: Ofce of the Joint Chiefs of Staff, n.d.). pp. 2224. For insights into Israeli
preparation for war against states with nuclear-armed missiles, see Eliot A. Cohen,
Michael J. Eisenstadt, and Andrew Bacevich, Knives, Tanks, and Missiles: Israels Security
Revolution (Washington, D.C.: Washington Institute for Near East Policy, 1998), pp. 93
94, 125, 127.
chapter 5 155
never reach that level. If, however, the United States judges that it is
worthwhile to develop military measures that would increase its ability
to act against small nuclear powers and so increase the ability of the
United States to maintain its alliance relations, these efforts should be
encouraged. Of course, the United States will not be the only country able
to prot from time and improved technology. Hostile countries will be
able to take advantage of technology to improve their ability to hide from
or otherwise thwart U.S. forces. How this interaction between hiders and
nders plays out is very much in question.
What will be the political utility of improved technology? Will im-
proved U.S. antinuclear capabilities, many of which will be secret and
undemonstrated, be credible to our allies in the face of demonstrated
nuclear threats? Will U.S. technology be substitutable for the lives of U.S.
citizens put at risk when the question is the credibility of U.S. nuclear
guarantees? And will the United States be inclined to put itself in harms
way if worst does come to worst, to help bring about a more stable and
secure world? Or will it accept a world in which the United States cannot
use its military power to resolve regional conicts in its favor?
156 the coming crisis
Chapter 6
U.S. Security Policy in a
Nuclear-Armed World, or
What If Iraq Had Had
Nuclear Weapons?
Barry R. Posen
Presuming that the United States remains a global power, it may well
one day confront a situation where a nuclear-armed regional power
threatens to trespass upon, or actually has trespassed upon, a very im-
portant U.S. national interest. The United States may still lack an array
of offensive and defensive weaponry that would permit a senior military
commander to assure the president of the United States that the adver-
sary cannot successfully explode a nuclear weapon on the territory of a
U.S. ally, or over a U.S. military force, or even on the United States itself.
As far as one can tell, the consideration of how the United States would
handle this situation has not proceeded very far. The national security
policy community, in and out of government, is beguiled by the possibil-
ity of a happy ending.
1
It hopes that nonproliferation policy will prevent
most proliferation and that a combination of weapons and tactics will
negate the capabilities of those who slip through the policy net.
2
Special thanks to my research assistant, David Burbach, and to the numerous col-
leagues who have offered comments on the chapter and on presentations based on the
chapter.
1. For a review of United States postCold War nuclear weapons and nonprolifera-
tion policy, and policy debates, see Stephen Cambone and Patrick J. Garrity, The
Future of United States Nuclear Policy, Survival, Vol. 36, No. 4 (Winter 199495),
pp. 7395. They divide policy preferences on nuclear weapons into two rough schools,
traditionalists and Marginalisers. The former is self-explanatory, the latter argue
that the United States has a unique opportunity to place nuclear weapons on the road
to ultimate extinction, pp. 7576.
2. Marc Dean Millot, Roger Molander, and Peter A. Wilson, The Day After . . .
Study: Nuclear Proliferation in the PostCold War World (Santa Monica, Calif., RAND,
1993), vols. I, II, and III. This document is a must-read for students of current U.S.
nonproliferation policy. It reports the results of a series of small-group policy simula-
tions conducted by a representative cross-section of mid-level, mainly U.S., civilian
This chapter assesses why the United States might, in fact, act force-
fully in a confrontation with an expansionist state in possession of a
modest nuclear retaliatory capability, and how it might proceed. The
vehicle for this assessment is a counterfactual historical analysis of the
U.S. reaction to Iraqs invasion of Kuwait in 1990. For heuristic purposes,
I introduce a small nuclear force into the Iraqi arsenal, a half-dozen
weapons and the means to deliver them regionally.
3
These weapons
cannot reliably be located for conventional or nuclear preemption, and
their delivery systems are sufciently capable that extant defensive weap-
onry cannot intercept them reliably. I postulate that Saudi leaders invite
the United States to send forces to help deter an Iraqi attack on the
kingdom immediately following the invasion of Kuwait.
4
I then ask two
questions. Should the United States consider the military liberation of
Kuwait?
5
If so, what strategy should follow?
and military policymakers and practitioners and policy analysts. The study conducted
multiple iterations of four simulations involving nuclear weapons employment in four
different regions, the Persian Gulf, Korea, the former Soviet Union, and South Asia. I
have reviewed the results of the rst two with some care, and the overview report. It
is pretty clear from the simulation that regional nuclear crises are so scary in prospect
that people seem inclined to focus the bulk of their energy on the happy ending.
While many participants recognize the improbability of the happy ending, the docu-
ment shows a relative paucity of recommendations about how the United States ought
to analyze and prepare for the most likely and dangerous contingency, a nuclear-
armed regional adversary against whose nuclear forces we will not have reliable
offensive and defensive options. In the exercises, this kind of threat introduced
enormous conservatism into the calculations of participants contemplating U.S. mili-
tary intervention. Vol. I, p. 15. I do not believe that this conservatism is the fault
of the simulation designers, but rather a fundamental insight gleaned from the exercise
about the mindset of the U.S. policy community. For the authors brief summary of
the insights to be derived from these simulations, see Roger C. Molander and Peter
A. Wilson, On Dealing with the Prospect of Nuclear Chaos, Washington Quarterly,
Vol. 17, No. 3 (Summer 1994), p. 32.
3. Al Venter, How Saddam Almost Built His Bomb, Janes Intelligence Review,
Vol. 9, No. 12 (December 1997), pp. 559566, suggests that Iraq might have completed
a bomb by perhaps the end of 1992, had the Kuwait crisis not occurred.
4. This assumption is arguable, but it is sufciently plausible that it does not vitiate
the value of my analysis. See the essay in this volume by Steven Walt.
5. Many seem to think it self-evident that the United States would be unwilling to
act. For example, Lawrence Freedman writes, One of the advantages of a nuclear
arsenal may be its role in discouraging Western involvement in local conicts, thereby
hastening Western disengagement from the security arrangements in many parts of
the world. One only needs to contemplate the impact of a completed Iraqi nuclear
program on Western calculations during the Gulf crisis to appreciate the importance
of such a step. Freedman, Great Powers, Vital Interests and Nuclear Weapons,
Survival, Vol. 36, No. 4 (Winter 199495), p. 47. See also Michael M. May and Roger D.
Speed, The Role of U.S. Nuclear Weapons in Regional Conicts (Center for Interna-
158 the coming crisis
I develop three recommendations for the U.S. government about the
hypothesized scenario. First, the United States should have tried to lib-
erate Kuwait because of the general strategic consequences of inaction,
not because of the intrinsic strategic value of Kuwait. Indeed, the whole
denition of the crisis should have changed to the rst postCold War
nuclear crisis. The likely consequences of inaction are developed at
length, and provide the core of the argument for action.
Second, to support a liberation campaign, the United States ought to
have pursued a strategy of intrawar deterrence. It would have needed
to make explicit and ferocious threats that nuclear retaliation would occur
if Saddam Hussein used nuclear weapons on any member of the coali-
tion. It should have explained to Saddam Hussein through a very sys-
tematic diplomatic campaign why the United States was compelled to
liberate Kuwait, and would be forced to retaliate with nuclear weapons
if Iraq employed them.
Third, the coalition should have pursued a military strategy of lim-
ited war in this operation: both ends and means ought to have been
restrained. This would have been the best way to control the risks. An
intensive discussion on the nature and scope of these military operations
among and between civilians and soldiers would have been essential.
The rst three sections of the chapter develop these recommenda-
tions. In each section, I rst offer the policy advice that I think I would
have given immediately after Iraq invaded Kuwait. Then I speculate on
the answers to two additional questions: To what extent would my advice
to act have found a sympathetic hearing among some of the actual
participants in the crisis? To what extent would my advice on how to act
have found a sympathetic ear? I show that the participants in the crisis
confronted issues analogous to those I raise. I reason that the decision-
makers actual behavior suggests that the three recommendations I de-
velop would have found a sympathetic hearing.
To complete the counterfactual analysis I briey discuss how success-
ful the conventional war might have been under the tactical constraints
I recommend. I also discuss the strategic benets of such a war. The
chapter concludes with a critical assessment of current U.S. nonprolifera-
tion policies and some advice for possible changes in those policies. My
tional Security and Arms Control, Stanford University, June 1993). Keith Payne, Deter-
rence in the Second Nuclear Age (Lexington: University of Kentucky Press, 1996); Barry
Schneider, Radical Responses to Radical Regimes: Evaluating Preemptive Counter-
proliferation, McNair Paper No. 41 (Washington, D.C.: Institute for National Strategic
Studies, National Defense University, 1995); and Martin Van Creveld, Nuclear Prolifera-
tion and the Future of Conict (New York: Free Press, 1993) all broadly concur with the
view that regional nuclear powers can easily deter the United States.
chapter 6 159
tools throughout are offense-defense theory, deterrence theory, and lim-
ited war theory.
This chapter does not argue that the policies advocated here would
have been followed. There were too many players with too many diverse
interests involved in the crisis to argue comprehensively that this advice
would have become the basis for action. For example, the analytic prem-
ises and tools that I employ do not permit me to speak to the reaction of
the U.S. public, or other publics. Neither can I predict the behavior of the
U.S. Congress, or the legislatures of allies.
Some pieces of the analytic puzzle that might protably be pursued
with my premises and tools are also omitted, largely for the sake of
brevity. For example, if the United States had known that Iraq possessed
nuclear weapons, would the preinvasion crisis have developed as it did?
How should the United States have acted during such a period? Presum-
ing that the invasion of Kuwait did happen, would the leaders of Saudi
Arabia have invited U.S. forces into their country?
6
The Risks of Inaction
the view from 19901991
Iraqs conquest of Kuwait immediately confronts U.S. decision-makers
with one very important fact: This is the rst postCold War nuclear
crisis. It is a dening moment. The actual stakesKuwaits oil and Iraqs
powerare probably secondary to the more fundamental question: What
will nuclear weapons mean? The consensus at the end of the Cold War
is that nuclear weapons deter nuclear attacks on oneself or ones allies,
and arguably deter conventional invasion of ones own territory, and to
a lesser and more debatable extent, ones allies territory.
7
6. Molander and Wilson argue, A regional predator will nd a small nuclear arsenal
a powerful tool for collapsing regional military coalitions that the United States might
craft to oppose such a future opponent. On Dealing with the Prospect of Nuclear
Chaos, p. 32.
7. Robert F. Jervis, The Meaning of the Nuclear Revolution (Ithaca, N.Y.: Cornell Univer-
sity Press, 1989), especially chap. 1, The Theory of the Nuclear Revolution, pp. 145.
He believes that nuclear weapons do all of these things; his essay essentially summa-
rizes the consensus view among nongovernmental deterrence theorists, and what I
suspect was the median view among most analysts, about the impact of nuclear
weapons on international politics in the 1970s and 1980s. There were, of course, many
who held opposite views. Since the end of the Cold War, however, one has heard fewer
challenges to the Jervis view. It is the basis for current U.S. nuclear declaratory
policy, strategic nuclear weapons acquisition and deployment policy, strategic nuclear
arms control policy, and nuclear nonproliferation initiatives. Indeed, current policy
seems to aim to sell the argument that nuclear weapons are good for almost nothing
160 the coming crisis
Will the United States and its allies, through inaction, allow other
countries nuclear forces to become potent instruments of aggression
against non-nuclear powers?
8
If the Iraqi conquest of Kuwait is permitted
to stand, nuclear weapons will come to be viewed as a shield that protects
conventional conquests from any challenger, including a great power
heavily armed with its own nuclear weapons. The new nuclear truism
will be that conventional forces take; nuclear forces hold! The context will
matter as well. Very important interests of the worlds sole surviving
superpower will have been successfully trampled by a state of modest
conventional and tiny nuclear capabilities.
The United States has asserted a vital interest in the independence of
the various Gulf oil states, and in the free ow of oil out of the region,
since the fall of the shah of Iran. The United States has programmed and
deployed military forces and assets for the explicit purpose of protecting
this interest. The United States has employed military force in operation
Earnest Will to defend directly the free ow of oil. Since 1973, the United
States has devoted considerable diplomatic effort and nancial resources
to achieve a settlement of the Arab-Israeli conict. It is fair to say that
there is no place else in the world where the United States has been so
active diplomatically and militarily without formal treaty relationships.
Capitulation to Saddam Husseins coup de main will therefore enhance
the image of nuclear weapons as great equalizers. Indeed, their successful
exploitation to deter a conventional counterattack to liberate ill-gotten
gains would make them seem even more valuable.
If Iraq were to deter a direct challenge to its aggression, how big a
change would it be? The change would be radical because nuclear weap-
ons have not been used by any state in this way since the dawn of the
nuclear age. No state has engaged in large-scale conventional aggression,
and then explicitly or implicitly tried to protect its gains from a conven-
but deterrence of nuclear attack on oneself and ones allies. Following Indias nuclear
tests in 1998, Indian policymakers expressed strikingly similar views about nuclear
weapons, but, of course, they may be telling us what we want to hear. John Burns,
India Defense Chief Calls U.S. Hypocritical, New York Times, June 18, 1998, p. A6.
8. Some do assume that this would occur. For example, May and Speed write,
Regional tactical nuclear deterrence could be difcult for the U.S. to counter, depend-
ing on the actual military and political situation. Thus, if the U.S. wanted to roll back
an aggression, as it did in the Gulf War, it would have to initiate the use of force against
at least a tactical nuclear threat, and potentially a strategic one as well. The feasibility
of doing this successfully and the feasibility of keeping the use of nuclear weapons
local might be so questionable as to deter any crossing of the military lines, just as it
did in Europe for forty years. In addition, any coalition the U.S. were to put together
for this purpose would come under extreme stress. May and Speed, The Role of U.S.
Nuclear Weapons in Regional Conicts.
chapter 6 161
tional counterattack through nuclear deterrent threats. This gambit is so
alien to the ways both theorists and political leaders have understood
nuclear weapons that there is almost nothing written about it even from
a theoretical perspective. The situation would be analogous to a half-
successful Warsaw Pact invasion of West Germany, which stalled after a
100-kilometer gain, going uncounterattacked by NATOs conventional
forces because of a Soviet claim that its nuclear deterrent umbrella now
covered its new real estate. An explicit consideration of this possibility
probably exists somewhere in the classied or open literature, but I have
neither found it nor heard of it.
9
The closest the nuclear deterrence literature comes to an explicit
treatment of the question of how conventional aggression could occur
between nuclear-armed adversaries is the stability-instability paradox.
This predicts that two states that can destroy each other may resort to
conventional warfare with each other because they do not fear escalation.
The strategic balance is so stable, and the consequences of nuclear war
so obvious, that neither side would nd it reasonable to employ nuclear
weapons, and both would know it.
10
But the paradox logically cannot
predict conventional attacks without also predicting conventional coun-
terattacks.
11
9. During the Cold War some military planners and analysts believed that nuclear
weapons would make it difcult to mount an amphibious landing in Europe if initial
enemy successes expelled the United States entirely from the Eurasian landmass. This
was a largely tactical belief, i.e., that amphibious forces were so vulnerable to nuclear
attack during the landing that the United States would never risk such an approach.
General Omar Bradley averred in 1950 that appraising the power of the atomic bomb,
I am wondering whether we shall ever have another large-scale amphibious operation.
Frankly, the atomic bomb, properly delivered, almost precludes such a possibility. See
Roger Hilsman, NATO: The Developing Strategic Context, in Klaus Knorr, ed.,
NATO and American Security (Princeton, N.J.: Princeton University Press, 1959), p. 19,
n. 7. Similar views were expressed by Sir John Slessor, Strategy for the West (New York:
William Morrow, 1954), p. 90; and Glenn Snyder, Deterrence and Defense (Princeton,
N.J.: Princeton University Press, 1961), p. 139.
10. Robert Jervis, The Illogic of American Nuclear Strategy (Ithaca, N.Y.: Cornell Uni-
versity Press, 1984), pp. 2934, 148150. In The Meaning of the Nuclear Revolution, Jervis
argues at greater length that the paradox seems not to operate very powerfully
(pp. 1923). Direct conventional clashes between the superpowers did not occur dur-
ing the Cold War. This may suggest that the paradox is weak. Alternatively, perhaps
neither superpower actually coveted much of what was in the others sphere of
interest. Or, given the nuclear warghting elements of both superpowers strategies,
the nuclear balance was somehow just unstable enough to discourage risk-taking. Or,
the relationship of NATO and Pact conventional forces was sufciently balanced that
neither side had high condence of victory.
11. The Illogic of American Nuclear Strategy, p. 155. Whereas the stability-instability
paradox could lead to relentless probing in the calculation that the risks are low
162 the coming crisis
Even states that achieved large-scale territorial gains through conven-
tional counterattacks did not try to secure their gains with nuclear deter-
rence; nor did the mere possession of nuclear weapons deter challengers.
For example, the United States took North Korea in a counteroffensive,
but did not make nuclear threats to dissuade a Chinese challenge to that
offensive, even after the appearance of Chinese troops in October 1950.
12
And the Chinese were obviously not deterred from entering the war by
the known U.S. nuclear capability.
13
Israel never formally invoked nuclear
enough to be worth running, decision-makers have taken what Patrick Morgan calls
a more sensible approach and have generally been willing to forego the chance of
gains in order to keep the risks of war as low as possible. This might suggest another
alternative pattern to the crisis under consideration; as a nuclear weapons state, and
hence a legitimate nuclear target, Saddam Hussein might have chosen not to invade
Kuwait.
12. Nuclear-congured B-29s were dispatched to the Far East at the end of July 1950,
when United States and Republic of Korea forces had retreated to the Pusan perimeter.
This was publicly announced but was accompanied by no explicit statement of their
purpose. Apparently the force had defensive and deterrent purposes both in Korea,
and against a Chinese attack on Taiwan. Roger Dingman, Atomic Diplomacy During
the Korean War, International Security, Vol. 13, No. 3 (Winter 1988/89), pp. 6065. The
bombers were returned to the United States before the Chinese crossed the Yalu.
Nothing that even looks like a nuclear deterrent threat is made again until President
Trumans confusing allusions to nuclear weapons in a press conference on November
30, ve days into the rst major Chinese offensive of the war, and more than a month
after the rst strong Chinese appearance. This nuclear threat, which was widely
reported in the press, appears to have been unintended, and if anything was driven
by a concern to avert a tactical military disaster. In any case, it precipitated a wave of
concern in the United States, but more particularly in Britain, and was quickly dis-
avowed. Ibid., pp. 6569. A third nuclear threat was effectively made in April, when
aircraft and atomic weapons were dispatched to Guam, in part out of fear of a possible
combined theater-wide Chinese and Soviet offensive, suggested by intelligence indi-
cators. This deployment was accompanied by indirect and somewhat oblique warn-
ings through contacts in Hong Kong that there were limits to United States restraint.
Ibid., pp. 5091. For an even more elaborate account of these events see Rosemary
Foot, The Wrong War (Ithaca, N.Y.: Cornell University Press, 1985). Her account does
not include any explicit U.S. warnings to China or the Soviet Union that they must
accept the unication of Korea. Indeed, the United States issued no such warning even
after it received signicant warnings from China that the United Nations should not
cross the 38th parallel.
13. Thomas J. Christensen, Threats, Assurances, and the Last Chance for Peace: The
Lessons of Maos Korean War Telegrams, International Security, Vol. 17, No. 1 (Summer
1992), pp. 137138, describes Maos complex reasoning as revealed in the telegrams to
Stalin and Zhou Enlai. While they mentioned Maos fear of U.S. military forces camped
on his border as a reason for military intervention in Korea, they do not mention any
fear of nuclear bombing. Ejecting the United States from the peninsula was seen as a
way to prevent a battleeld stalemate, which Mao feared would precipitate intensive
bombing of Chinas cities. If nuclear bombing had been feared, this belief would have
chapter 6 163
weapons to secure its gains of 1967, and its rumored possession of nuclear
weapons was not enough to deter an Egyptian and Syrian challenge.
14
Effectively, the toleration of an Iraqi success in Kuwait would change
nuclear weapons from defenders of the status quo to instruments of
aggression. The consequences for world politics are likely to approximate
those that theorists have associated with offense-dominant worlds, at
least until most states can acquire secure second-strike capabilities.
15
International politics will suddenly turn very competitive.
Ambitious states will be the quickest to learn lessons from the epi-
sode. Aggressors will strive harder to get nuclear weapons because their
utility as conquest protectors will have been demonstrated starkly.
States that get or already have nuclear weapons, and have claims against
their non-nuclear neighbors, would feel substantially freer to enforce
those claims. And because their neighbors understand this, and will
surely try to get their own nuclear weapons to ensure themselves, the
ambitious will rush to push their claims.
The failure to act to reverse Iraqi aggression would thus also increase
the desire of status quo powers for their own nuclear weapons. Status
quo states would have to assume that non-nuclear states with claims
against them would soon embark on nuclear weapons programs. What-
ever deterrent benet non-nuclear states have derived from tacit or for-
mal alliances with any of the ve major declared nuclear powers will also
have been eroded, once the strongest nuclear power is deterred by a few
nuclear weapons from acting to secure a vital interest. Threatened status
quo states will thus nd it prudent to move quickly to try to acquire an
independent nuclear capability.
been senseless, because the United States could have devastated the small number of
important Chinese industrial areas with nuclear bombers from bases outside of Korea.
Mao would likely have understood this.
14. Jonathan Shimshoni, Israel and Conventional Deterrence (Ithaca, N.Y.: Cornell Uni-
versity Press, 1988), pp. 3133, notes that declaratory policy on both sides between
1967 and 1973 was completely devoid of references to Israeli nuclear weapons.
15. Stephen Van Evera, The Cult of the Offensive and the Origins of the First World
War, in Steven Miller, Sean Lynn Jones, and Stephen Van Evera, eds., Military Strategy
and the Origins of the First World War (Princeton, N.J.: Princeton University Press, 1991),
pp. 6467. He lists aggressive foreign policies; rst strike and rst mobilization advan-
tages; wider windows of vulnerability and opportunity; competitive diplomacy to
include brinkmanship and faits accomplis; and tighter secrecy as plausible conse-
quences of widely perceived offense dominance. Because the nuclear offensive advan-
tage discussed here is strategic rather than tactical, the rst-strike and rst mobili-
zation advantages ought not to be a direct consequence of the United States failure
to act. They may arise as an indirect consequence, however, since many of the states
trying to acquire nuclear weapons quickly will lack the resources to eld a secure
retaliatory force.
164 the coming crisis
The utility of distant allies in a conventional war will have been
reduced if not eliminated. It will have been demonstrated that the nuclear
aggressor can deter the intervention of allies to take back lost territory.
If a status quo power is conventionally inferior to a nuclear-armed
challenger, it would have to act quickly to try to redress the conventional
balance on its own. It might also seek the permanent presence of substan-
tial allied conventional forces on its soil. But even if a status quo power
can compete conventionally, it will never be certain that its conventional
defenses will hold in war. Thus, it will want its own nuclear weapons
to ensure that if a nuclear adversary achieves military gains, that
adversary will not be able to deter a conventional counterattack.
Against a nuclear-armed aggressor, only the country that stands to lose
its sovereignty or real estate has the necessary will to risk a nuclear
confrontation.
Ambitious states will see many incentives for preventive war. In
particular, evidence of nascent nuclear weapons programs in states they
see as prospective victims could spark a conventional attack, because this
would be the last opportunity for aggrandizement. Aside from the incli-
nation to take long-coveted land when a temporary military advantage
arises, one can also imagine a wave of Osirak-type attacks, in which
both status quo and aggressor states attempted to wipe out one anothers
nascent nuclear capabilities conventionally as Israel attempted on June 5,
1981, in its air raid on an Iraqi research reactor. To evade preventive
attacks, both conventional and nuclear, even status quo states would have
powerful incentives to pursue secret nuclear weapons programs, and to
try to acquire black market bombs. (Other weapons of mass destruction
that might be secretly developed, such as biological weapons, would also
become attractive.) This would in turn create new opportunities for those
with small nuclear weapons programs in one region to sell weapons to
threatened states in other regions, and it would also create incentives for
theft. Rumors of secret nuclear weapons programs would be rife, causing
still more countries to contemplate nuclear acquisition. The Nuclear Non-
Proliferation Treaty (NPT) would simply be swamped by violations and
suspected violations.
This singular failure to act could thus usher in a dynamic pattern of
international security competition in which even the good guys would
have to play a rough game. It would certainly make for regional nuclear
and conventional arms races. The incentives for preventive war would
intensify. Nuclear weapons states would perceive aggression against non-
nuclear states to be relatively easy whenever their conventional capabili-
ties appeared sufcient to achieve success. After years of trouble, the
world might settle down to many stable relationships of mutual deter-
chapter 6 165
rence, a series of micro-cold wars. This process could take a long time,
and be very exciting.
Taiwan and South Korea will face the starkest choices. These non-
nuclear states face nuclear or near-nuclear rivals with strong claims
against them. Both are competitive conventionally with their adversaries,
but hardly dominant. The South Koreans have a direct U.S. military
presence but they will likely clamor for even more U.S. troopssufcient
to block condently any North Korean attack virtually at its initiation.
Seouls proximity to the border may prove too tempting a target for
North Koreas large standing army, if the United States shows itself to be
too intimidated by a few nuclear weapons to even try to take back lost
territory. Taiwan is not even as well placed as South Korea, since U.S.
forces are not present, and U.S. diplomacy has to some extent conceded
the legitimacy of the Chinese claim to Taiwan. Both Taiwan and South
Korea have the scientic and industrial capability, and the wealth, to
build nuclear weapons; indeed, both once had nascent nuclear weapons
programs that U.S. policy initiatives managed to stop.
16
It is implausible that Japan can hold fast to its non-nuclear status.
Though it is free of claims by any state against most of its territory, can
it ride out a sudden surge of proliferation on its periphery? It seems
unlikely that such a capable state would want to remain dependent on
U.S. deterrent promises in such a competitive world. The United States
will also likely nd itself facing a new round of doubts in the Federal
Republic of Germany about the credibility of the U.S. nuclear commit-
ment.
17
16. See Mitchell Reiss, Without the Bomb (New York: Columbia University Press, 1988),
pp. 86108, who suggests that the Republic of Korea did initiate a nuclear weapons
program in the 1970s. See also Joseph Yager, Taiwan, in Joseph Yager, ed., Nonprolif-
eration and United States Foreign Policy (Washington, D.C.: Brookings Institution, 1980),
pp. 6681, suggesting that the evidence for a deliberate Taiwanese program at that
time was weak. Nevertheless, the United States did insist that a particular reactor be
shut down, which it was. This was not the end of the story, however, as a program
continued that brought the Taiwanese very close to a bomb in 1987, when additional
U.S. restraining pressure was exercised. David Albright and Cory Gay, Taiwan:
Nuclear Nightmare Averted, Bulletin of the Atomic Scientists, Vol. 54, No. 1 (Janu-
ary/February 1998), pp. 5460.
17. At the time of Desert Shield and Desert Storm, the Soviet Union was still intact.
Had it collapsed following a U.S. failure to act against Iraq, the subsequent consolida-
tion of Soviet tactical nuclear weapons would probably not have proceeded so
smoothly. It is implausible, for example, that Ukraine would have contemplated giving
up its nuclear weapons. Given the probability that more than one nuclear power
would have emerged from the wreckage of the Soviet Union, it seems likely that states
on the periphery of the old empire, seeing these new nuclear states arise, would
themselves begin to think more seriously about their own nuclear weapons capability.
166 the coming crisis
The course of world politics can be changed dramatically for the
worse if the United States fails to act to reverse Iraqi aggression. The
relative U.S. power position will be undermined. The perceived utility of
nuclear weapons will be enhanced. Thus, the incidence of war in general
will probably increase, at least in the near term. The probability of nuclear
accidents and ecological disasters associated with crash, low-budget nu-
clear weapons programs will also increase. Most seriously, there will be
a greater risk of regional nuclear war. Can the United States nd a
plausible strategy to live comfortably in such a world? If not, then these
dangers seem sufciently compelling to warrant U.S. action against Iraq.
could the united states live in such a world? Three alternative
U.S. strategies suggest themselves if a new era of intense strategic com-
petition and stronger proliferation incentives is set in motion by a failure
to act against Iraq: an intensied counterproliferation policy; a military
build-up to allow the United States condently to defend allies against
conventional aggression; and a fortress America policy.
Failure to act against Iraqi aggression would show U.S. decision-
makers that a few nuclear weapons in the hands of a prospective enemy
can utterly paralyze U.S. policy. If the United States is to maintain any
freedom of action in the world, then nuclear proliferation must not occur.
The political aftermath of the conquest of Kuwait would make it difcult
to place a new emphasis on the NPT. Too many states would have
incentives to get nuclear weapons to be placated with great power prom-
ises. Instead, the United States would have to concentrate much more on
active measures. States bent on getting nuclear weapons, with policies
inimical to U.S. interests, would have to be denied the completion of their
programs. Any means, fair or foul, would be employed to this end. U.S.
intelligence agencies would be turned loose to sabotage nascent pro-
grams. If these measures failed, then preventive conventional war would
be undertaken. These measures would only be acceptable before aspiring
nuclear weapons states completed a working bomb. States with a work-
ing weapon would have to be left alone.
A second policy would create little NATOsforward-deployed
U.S. capabilities sufcient for a tenacious local defense, plus institution-
alized nuclear guaranteeswherever the United States has important
interests that might be threatened by a nuclear state. To convince pro-
The most likely candidate here would have been Iran, which would have perceived
itself under nuclear pressure from every sideIraq, India, Pakistan, and the Trans-
caucasian nuclear fragments of the former Soviet Union (Azerbaijan, Armenia, or
Georgia). Turkey could not have remained quiescent in the face of these developments.
chapter 6 167
spective victims not to acquire nuclear weapons, the United States or
another superpower would have to guarantee their security with
sufcient force to stop a conventional aggressor. Programs of nuclear
cooperation, similar to those arranged between the United States and its
NATO allies, might also have to be considered. Desert Shield, the little
NATO of the Middle East, might last for decades. U.S. forces in Korea
would have to be beefed up. The planned drawdown of U.S. troops
stationed in Europe may need to be canceled. The fundamental decision
about whether or not to ght for Taiwan could no longer be postponed.
The United States would now be called upon to offer nuclear hos-
tages to any ally threatened by a nuclear aggressorone important func-
tion that U.S. troops served in Germany during the Cold War. If Germany
had been subjected to nuclear attack, U.S. troops (and their dependents)
would also have suffered. This increased the probability that the United
States would retaliate. The likely indignation of U.S. leaders alone raised
the risk of a nuclear response. Failure to respond might also have eroded
the credibility of the U.S. nuclear deterrent to discourage nuclear attacks
on North America. If U.S. soldiers and citizens could be attacked with
nuclear weapons with impunity abroad, U.S. decision-makers would
have reason to fear that the next step would be an ambitious aggressor
misperceiving the freedom to threaten U.S. citizens at home. This fear
would have provided an additional incentive for a strong response.
Extended nuclear deterrence is unlikely to go even as smoothly as
it did during the Cold War. The failure to act to counter aggression by a
small nuclear weapons state like Iraq would surely raise new doubts
about the value of forward-deployed U.S. troops as guarantors of U.S.
nuclear retaliation. Even during the height of the Cold War, many asked
whether the United States would trade Boston for Bonn, if it was called
upon to retaliate for a Soviet nuclear attack. Greater doubts would arise
under these new conditions.
Though the previous two policies could be executed separately, they
would probably be combined. Perhaps, after considerable assertive coun-
terproliferation and a substantial U.S. military buildup, the proliferation
incentives that would arise from a failure in Kuwait could be dampened.
Aggressors would nd it dangerous to initiate nuclear programs; status
quo powers would nd the United States ready to offer the coin of real,
local conventional military power to forestall invasions, and incidentally
provide hostages against coercive nuclear threats.
The human and material costs of this strategy are substantial. One
cannot be sure that even these measures would prove adequate to deter
all future proliferators, so costlier measures could prove necessary. Forc-
168 the coming crisis
ible occupation and disarmament of some countries is the next step.
18
This seems probable in the cases of North Korea and Iran, for example.
These countries have large populations and conscript armies; they will
be difcult to conquer and even more difcult to police.
The little NATOs aspect of the strategy could require a force struc-
ture equal to or greater than that sustained by the United States during
the last years of the Cold War. U.S. forces in NATOs center region might
shrink somewhat, but greater standing forces would be necessary in the
southern regionparticularly in Turkey. Forces stationed in South Korea
might have to grow. A new permanent force in the Persian Gulf region
will prove necessary. Ready forces for new commitments would prove
useful to create a general expectation among states that the United States
would come to their assistance if their neighbors turned nasty. If coupled
with the preventive war option, an additional strategic reserve offen-
sive force would prove necessary.
Fortress America is a distinct, third alternative national strategy.
The task of U.S. policy would be to divert nuclear competition and the
risks of nuclear war away from its own territory. The United States would
simply accept that a hellishly competitive world will emerge. The rst
step would be a change in U.S. foreign policy toward disengagement. By
staying out of overseas political competitions, the United States would
reduce the incentives for new nuclear powers to threaten the country or
its armed forces. U.S. theater forces would not be jeopardized by regional
nuclear forces because there would not be any U.S. theater forces.
19
The United States would rely more forthrightly on its own nuclear
deterrent power; policymakers would stress their intention to respond to
any direct attack on the United States with a devastating retaliation.
Renewed efforts in the realm of strategic defense would be justied; just
in case others doubt U.S. will, they will also face formidable obstacles to
18. For example, it is clear in retrospect that the Iraqi nuclear weapons program that
existed prior to Desert Storm could not have been catalogued, much less destroyed,
without considerable direct access to the entire country. This was only achieved after
Iraq suffered a terric military defeat and became convinced that it faced the threat of
a potentially endless economic embargo.
19. Freedman, Great Powers, Vital Interests and Nuclear Weapons, p. 48, sees this
as a general consequence of extensive nuclear proliferation. The spread of nuclear
weapons, in terms of political control as much as absolute numbers, encourages
strategic disengagement and thus a loss of inuence in regions where important, if
not quite vital, interests are involved . . . proliferation feeds on and then reinforces an
existing tendency to reduce the security links between the declared nuclear powers
and those parts of their far abroads that are not covered by a well-established
alliance.
chapter 6 169
a successful attack. Moreover, in a world where many states have nuclear
weapons and not all are competent, an errant weapon could come the
way of the United States. U.S. intelligence services would focus inten-
sively on reducing the possibility that nuclear weapons might be smug-
gled into the United States. More importantly, intelligence agencies would
act to back U.S. deterrent threats by creating the strong expectation that
the United States would determine the source of any weapon smuggled
into the country and nd some group or place against which a retaliation
can be directed.
This policy would require a major change from the way the United
States has conducted its foreign policy since World War II, but disengage-
ment from world politics is not unprecedented for this country. The
United States would require enormous self-control to stay out of political
disputes abroad that may arouse its passions or idealism. And it would
require steady nerves.
inferences from the actual crisis
The thinking and behavior of U.S. decision-makers during the Desert
Shield crisis suggests that the foregoing analysis, or something similar,
would have been conducted and would have found a ready audience.
Though the nuclear issue apparently did not arise in early discussions,
there is considerable evidence that key decision-makers worried about
the global and regional consequences of letting the Iraqi success stand. It
was feared that Iraqi power would grow, and that the credibility of U.S.
threats to deter future Iraqi action would weaken. More broadly, U.S.
credibility would suffer, perhaps enough to set off a wave of challenges.
Brent Scowcroft, the presidents national security adviser, favored
strong action against Iraqs seizure of Kuwait. In one of the earliest
National Security Council meetings on the Iraqi invasion of Kuwait, on
August 3, Scowcroft reportedly opened the meeting by saying, We have
got to examine what the long-term interests are for this country and for
the Middle East if the invasion and taking of Kuwait become an accom-
plished fact. We have to begin our deliberations with the fact that this is
unacceptable. Yes, its hard to do much. There are lots of reasons why we
cant do things but its our job.
20
The conversation in this meeting
20. Bob Woodward, The Commanders (New York: Simon and Schuster, 1991), paper-
back edition, pp. 217218. See also p. 211, on Scowcrofts critical reaction to the dis-
cussions in the preceding days meeting: Scowcroft indicated to the President that the
meeting had seemed to miss the point about the larger foreign policy questions.
Unfortunately, Woodward does not spell out precisely what Scowcroft meant. It seems
to have been a combination of a failure to meet a longstanding U.S. declaration that
Persian Gulf Oil could not be allowed to fall into the hands of a dangerous adversary
170 the coming crisis
reportedly moved in the direction of a consensus that Iraq had created a
major strategic problem for the United States.
21
U.S. Secretary of Defense Richard Cheney reportedly had ambitious
objectives from the outset of the crisis. He was apparently concerned
about the future political and military threat Hussein could pose if he
controlled Gulf oil. This must be interpreted as fear of what Iraq could
and would do if it became even more powerful. Already on August 2, he
saw the question as whether the U.S. goal should be just the liberation
of Kuwait, or also the overthrow of Saddam Hussein.
22
Dennis Ross, the
State Departments Director of Policy Planning, and a key aid to U.S.
Secretary of State James Baker, saw Saddam Husseins aggression as an
indicator that Iraq had become a threat to the entire Middle East. He
apparently concluded that Saddam Husseins regime would need to be
eliminated.
23
UN Ambassador Thomas Pickering believed that U.S. credi-
bility in the Middle East would suffer if nothing was done about the
invasion of Kuwait.
24
Deputy Secretary of State Lawrence Eagleburger was also a strong
early proponent of rollback. He worried about the precedent that success-
ful Iraqi aggression would set for all the Quaddas and Kim Il Sungs of
the world, indicating to them that the end of the Cold War had created
exploitable power vacuums.
25
In the August 4 Camp David meeting, one
unnamed participant made a mirror image argument. I think if we
succeed this time, the next such crisis might not take all this much agony.
In other words, if potential aggressors believe in advance that the civi-
lized world is going to behave in a certain way, they will tend to tailor
their actions.
26
President George Bush seems to have learned the 1930s lesson on
appeasement. He was concerned that if he practiced appeasement, it
would have the same consequences it had in the 1930s. One unnamed
and a more general desire to show that the United States was not afraid to use military
force to protect its interests. Both boil down to credibility.
21. Ibid., p. 218.
22. Michael Gordon and Bernard Trainor, The Generals War: The Inside Story of the
Conict in the Gulf (Boston: Little Brown, 1995), pp. 3233. Woodward, The Commanders,
p. 208, asserts that Cheney only favored the defense of Saudi Arabia at this point.
23. Gordon and Trainor, The Generals War, p. 35.
24. Ibid., p. 36.
25. Ibid., p. 37.
26. U.S. News and World Report, Triumph Without Victory: The Unreported History of
the Persian Gulf War (New York: Random House, 1992), p. 72.
chapter 6 171
presidential adviser observed, George Bush is deathly afraid of appease-
ment. His generation had to ght a war over it, and he feels that if he
blinks today, he will be leaving a real mess for the next generation to
clean up.
27
President Bush characterized his views this way in sub-
sequent communications with the staff of U.S. News and World Report.
The bottom line was that aggression could not stand. If he was permitted
to get away with that, heaven knows where the world would have gone
and what forces would have been unleashed.
28
This perspective was also expressed by British Prime Minister Mar-
garet Thatcher in a meeting with President Bush in Aspen, Colorado, on
August 2. She analogized explicitly to Nazi aggression in the 1930s, an
analogy that Bush employed repeatedly.
29
Later in the crisis, in a Decem-
ber 24 meeting, Bush indicated that he would no longer settle for an
unconditional Iraqi withdrawal; he wanted a military campaign and
victory. Chairman of the U.S. Joint Chiefs of Staff Colin Powell reports
President Bushs reasoning: If the Iraqis withdrew now, it would be with
impunity for their crimes. A pullback would also mean that Saddam
would leave Kuwait intact with his huge army intact, ready to ght
another day.
30
The president appears to have been concerned that a
lesson needed to be taught; the aggressor must be punished, and de-
prived of capabilities for future mischief.
Had the United States confronted a nuclear-armed Iraq, the choices
would have been difcult. A decision to launch a military operation to
liberate Kuwait would have involved immediate and obvious nuclear
risks. The consequences of inaction outlined above, and the relatively
unpleasant range of possible U.S. alternative policies in the new world
thus created, would surely have been enumerated by someone. But the
costs and risks of inaction would have seemed theoretical and distant.
The risks of confrontation with a nuclear Iraq would have seemed clear
27. Steven J. Wayne, President Bush Goes to War, in Stanley Renshon, ed., The
Political Psychology of the Gulf War: Leaders, Publics, and the Process of Conict (Pittsburgh,
Penn.: University of Pittsburgh Press, 1993), p. 39. The themes of no appeasement
and stopping aggression both surface in the August 2 National Security Council
meeting. Woodward, The Commanders, pp. 235237.
28. U.S. News and World Report, Triumph Without Victory, p. 48. An unnamed adviser
who was with Bush when the invasion occurred observed, He thought that the
aggression should not stand, especially in a place where we had staked out as our
vital interests . . . . That was the Presidents strong view. Ibid., p. 48.
29. Gordon and Trainor, The Generals War, p. 36; U.S. News and World Report,
Triumph Without Victory, p. 72.
30. Colin Powell with Joseph Persico, My American Journey (New York: Random
House, 1995), p. 499.
172 the coming crisis
and imminent. One cannot say with condence how the consensus
among the key decision-makers would have actually evolved. But this
analysis does suggest that the often implied proposition that the United
States would easily have been deterred by Iraq should be subjected to
much closer scrutiny. There were strong reasons for the United States to
act. And there is much circumstantial evidence from the way the crisis
was actually handled to suggest that these reasons would have resonated
with the decision-makers. Yet a nal decision to act would have required
a credible plan for success. Someone would have had to propose a
strategy that promised political and military success, with low risks of
escalation. Below, I try to develop such a strategy.
Planning the Reversal of Iraqi Aggression
the view from 19901991
As we approach the end of autumn, Desert Shield is a great success.
Sufcient Western military power has reached the Persian Gulf to defend
successfully against any new Iraqi conventional attacks. Though the eco-
nomic embargo on Iraq is taking hold, and ought to be given a fair chance
to pressure Saddam Hussein into withdrawal from Kuwait, the United
States cannot count on the success of this strategy. It may fail altogether,
or simply succeed at such a slow pace that Iraq will have plenty of time
to devise diplomatic strategies to erode the integrity of the U.S.-led
coalition. Though Iraq is dependent on many imports, it has vast reserves
of the cheapest and most industrially exible energy sourceoil. This
alone must give the economy a substantial adaptive capacity. The eco-
nomic embargo may weaken. Iraqs reserves of oil, to include its new
Kuwaiti reserves, are a seductive prize for some prospective trading
partner. On the diplomatic side, the Arab-Israeli conict provides a lever
that Saddam will try to exploit diplomatically to ssure the coalition.
Saudi Arabia may feel squeezed by the trade-off between external secu-
rity and internal stability. Elements within Saudi society may begin to ask
how long the Americans intend to stay. Decades may not be an accept-
able answer.
Sooner or later, the United States and its allies will have to consider
military action to eject Iraq from Kuwait. There is no reliable algorithm
that can tell the United States exactly when it should switch from em-
bargo to war. Given that embargo is a slow and insidious weapon, if the
economic embargo does not soon persuade Saddam Hussein to change his
policy, it will take years for its cumulative effects to force him to do so (if
ever.) It is therefore unreasonable to wait until the coalition begins to
suffer erosion to go to war, and more useful to choose a cut-off point for
chapter 6 173
giving the embargo a chance to work. For planning purposes, if Iraq is
not out of Kuwait by autumn of 1991, the United States and its partners
should place themselves in a military position to initiate offensive opera-
tions at that time.
intrawar deterrence
Perhaps the most important issue that the coalition faces is how to
neutralize any attempt by Iraq to exploit its nuclear weapons to deter a
coalition attack of any kind, or to thwart a coalition victory. Two methods
suggest themselves, a splendid rst strike, or intrawar deterrence. It is
clear that a splendid rst strike that knocks out Iraqi nuclear weapons
with conventional ordnance would be best, but it seems improbable that
the military can provide a high-condence option to do this. Thus, the
coalitions strategy must be to wage a conventional war with Iraq while
deterring its resort to nuclear weapons. Deterrence depends on capability
and will. There is no doubt that the United States and its allies command
massive conventional and nuclear capability. It ought not to be impossible
to explain to Iraq what the United States can do in response to an Iraqi
use of nuclear weapons. The problem is convincing Saddam Hussein that
the United States will retaliate in particularly horrible ways if he employs
nuclear weapons. Only diplomacy can convince Saddam Hussein that
U.S. will is stronger than his.
The United States must persuade Iraq that the United States cares
more about the liberation of Kuwait than Iraq cares about holding it.
Deterrent diplomacy is exactly that; explanations matter as much or more
than military aspects. The problem is difcult. Iraq is the local power, has
developed a historical (albeit trumped up) claim to the land, has tried
to seize the country once before, can claim (with some legitimacy) a
number of injuries inicted by Kuwait, and has just seized the country.
Iraq probably perceives that its bargaining position is very strong. Here-
tofore, the United States has cared about Kuwait largely for economic
reasons. Kuwait has a lot of oil, but there is plenty more in Saudi Arabia
and in the Gulf states, and those are now well protected by the forces
entering the region for Desert Shield. Why should Saddam believe that
nuclear threats will not stop a coalition offensive? And, if need be, why
should not a single nuclear explosion cause the United States to recon-
sider its entire position? The United States, and its many allies, obviously
must disabuse Iraqi leaders of this belief.
The United States and its friends must explain to Iraq just how
terrible the political world appears to them, if Iraq is allowed to enjoy its
conquest. The method for accomplishing this is to explain over and over
U.S. expectations about the grave security situation that will result if Iraq
174 the coming crisis
is allowed to deter the coalition from liberating Kuwait. This story is
somewhat analogous to the we dont deal with terrorists and hostage-
takers position that many Western countries, including the United
States, have taken. (Of course, the United States did, to its great discredit,
deviate from this position.) Countries say that they will not deal with
terrorists because it leaves them open to future exploitation. Failure to
counter Iraqi aggression would effectively do the same, and much more.
The task of explaining this position to Iraqi leaders, especially Sad-
dam Hussein, is not a simple one, and must be pursued with enormous
energy and creativity. Both public repetition and private messages must
be employed. Numerous senior U.S. gures, representatives of other
governments, and even representatives of international organizations
such as the United Nations, will have to repeat these arguments publicly.
Go-betweens, such as the Russians and the Jordanians, must convey these
messages privately. In addition, the appropriate historical analogies must
be made publicly. While the Hitler analogy is already proving a useful
public relations tool within the United States, perhaps the Cuba anal-
ogy could prove useful in communicating the seriousness of U.S. intent
to Saddam Hussein.
Given the Cold War experience with coercive diplomacy, it may be
reasonable to move some nuclear weapons toward the theater to enhance
the credibility of the threat of ferocious retaliation in the event of an Iraqi
nuclear attack. There is no particular technical utility to such action. It
would be purely a diplomatic signal. Some joint coalition planning for
the retaliatory use of such weapons probably ought to occur. Rumors of
such consultations would have an additional diplomatic effect. At the
same time, such consultations would lend credibility to the message that
this is a contingent threat. If Iraq does not use nuclear weapons, neither
would the allies.
inferences from the actual crisis
The issue of intrawar deterrence did in fact arise in a muted way dur-
ingthe planning for Desert Storm. U.S. Secretary of State James Baker, in
his January 9, 1991, meeting with Iraqi Deputy Prime Minister Tariq
Aziz, carried a letter from President Bush that threatened underspeci-
ed but nevertheless horrible consequences if Iraq used chemical or
biological weapons.
31
Indeed, Aziz thought that these were nuclear
31. The Bush letter reportedly read, The United States will not tolerate the use of
chemical or biological weapons, support of any kind of terrorist actions, or the
destruction of Kuwaits oilelds and installations. The American people would de-
mand the strongest possible response. You and your country will pay a terrible price
chapter 6 175
threats.
32
Though Iraq had plenty of chemical weapons, it did not employ
them.
33
It seems plausible that even more explicit retaliatory threats
would have been leveled against a nuclear threat.
Oddly, the private message to Saddam Hussein was preceded by
what appears to have been a deliberate public message indicating that
the use of nuclear weapons was not under consideration. A January 7,
1991, Washington Post story quotes several named and unnamed senior
civilian and military ofcials to the effect that nuclear use was not under
consideration. Ironically, this message was probably an effort to quell the
fears of allies that the United States was considering the use of nuclear
weapons, fears that may have been created by deterrent diplomacy. For
example, U.S. Secretary of Defense Richard Cheney said during a visit to
the Gulf on December 23, Were Saddam Hussein foolish enough to use
weapons of mass destruction, the U.S. response would be absolutely
overwhelming and it would be devastating.
34
Both from the Post story,
and from other accounts, it appears that a variety of oblique yet ferocious
messages had been sent.
35
I suspect that the administration suddenly
if you order unconscionable action of this sort. Lawrence Freedman and Ephraim
Karsh, The Gulf Conict: 199091 (Princeton, N.J.: Princeton University Press, 1993),
p. 255.
32. Aziz informed Rolf Ekeus, the chief UN investigator of Iraqs nuclear, biological,
and chemical weapons programs, that Iraq decided not to employ biological weapons
because of a strong, if somewhat ambiguous, warning received from the Bush admini-
stration on January 9, 1991, which Iraqi leaders interpreted as a nuclear threat. In a
four-hour meeting with Aziz in Geneva, Secretary of State James A. Baker III threat-
ened a U.S. response that would set Iraq back years by reducing its industry to
rubble. R. Jeffrey Smith, U.N. Says Iraqis Prepared Germ Weapons in Gulf War,
Washington Post, August 26, 1995, p. A1. Other accounts suggest that Baker threatened
the overthrow of the Iraqi government in the event of a use of chemical or biological
weapons. See Freedman and Karsh, The Gulf Conict, p. 257; Baker says that he did
hope to create the impression that nuclear weapons might be employed. James Baker
with Thomas DeFrank, The Politics of Diplomacy (New York: Putnam and Sons, 1995),
p. 359.
33. On current views of the extent of Iraqs chemical weapon arsenal, see Christopher
Wren, U.N. Arms Inspector Firm on Iraq Nerve Gas, New York Times, June 25, 1998,
p. A10.
34. Jeffrey Smith and Rick Atkinson, U.S. Rules Out Gulf Use of Nuclear, Chemical
Arms, Washington Post, January 7, 1991, p. A1.
35. According to Lawrence Freedman, In the 1991 Persian Gulf War, Britain and
France were ready to rule out nuclear use prior to the conict: the Americans were
more hesitant, although they worked to deter Iraqi chemical-weapon use largely
through threats to the regime in Baghdad. Great Powers, p. 41. See also Molly
Moore, A Woman at War (New York: Scribners Sons, 1993), p. 318. During his
speaking engagements, Boomer [Lieutenant General Walter Boomer, the U.S. Marine
176 the coming crisis
became concerned that deterrent diplomacy had perhaps gotten out of
control. Perhaps certain coalition partners were beginning to believe that
the United States did intend to employ nuclear weapons.
Nuclear weapons were apparently discussed more frequently than
policymakers admitted at the time. In November of 1990, Cheney is
quoted as having told a proponent of nuclear weapons use that nobody
had suggested such a thing in the entire three months of the crisis.
36
In
an interview in 1996, former President George Bush asserted that nuclear
use was not something we really considered at all.
37
Yet General Colin
Powell says in his memoirs that Cheney had asked him to take a look at
tactical nuclear options in early October 1990.
38
General Scowcroft stated
in a 1995 interview that the administration did indeed discuss nuclear
weapons, but determined that their use would be unnecessary.
39
Some-
time in late fall, General Schwarzkopf suggested to General Powell a
diplomatic demarche that the United States would use nuclear weapons
if the Iraqis used chemicals.
40
Newsweek magazine reported in its January
14, 1991, issue that Schwarzkopf had earlier asked for permission to use
a nuclear weapon to generate an electromagnetic pulse to wreck Iraqi
Corps commander in Desert Storm] talked rather candidly about how the media had
been used to intimidate Saddam Hussein. As I watched from the back of one confer-
ence room, Boomer told the audience that reporters like Molly Moore . . . frequently
asked him during the buildup: What will be the American response if Saddam uses
chemical weapons? Boomer said his standard reply was: Its going to be something
worse, something terrible, implying that perhaps nuclear weapons would be consid-
ered. Boomer smiled when he admitted to the audience, I just made that up; I didnt
know what the hell we were going to do. According to two RAND analysts, an
unnamed British general told reporters that chemical use by Iraq would mean a
nuclear response by the coalition forces, though they provide no citation. Paul K.
Davis and John Arquilla, Deterring or Coercing Opponents in Crisis, Lessons from
the War with Saddam Hussein (Santa Monica, Calif.: RAND, National Defense Re-
search Institute), p. 59, n. 35.
36. Jeffrey Smith, U.N. Says Iraqis Prepared Germ Weapons in Gulf War, Washing-
ton Post, August 26, 1995, p. A1.
37. See A Gulf War Exclusive: Talking With David Frost, Transcript No. 51, January
16, 1996, p. 5, cited in Payne, Deterrence in the Second Nuclear Age, pp. 138139.
38. Powell, My American Journey, p. 486.
39. General Brent Scowcroft, Meet the Press, NBC News, August 27, 1995, tran-
script. Mr. McManus: Did you deliberately give the Iraqis the impression that you
were ready to use nuclear weapons? Answer: We kept it very ambiguous, which I
think was the right thing to do . . . . Im just telling you what we had privately
decided. But, no, we never ruled out the use of nuclear weapons and should not, I
dont think.
40. Rick Atkinson, Crusade: The Untold Story of the Persian Gulf War (Boston: Houghton
Mifin, 1993), p. 86.
chapter 6 177
radars and communications devices at the outset of a campaign, and that
the Pentagon actually hired some consultants to examine the option. And
it was suggested during the planning discussions for attacks on Iraqi
biological weapons storage sites that nuclear weapons would be the
surest way to generate enough heat to actually kill the viruses with high
condence. Thus, the use of nuclear weapons seems indeed to have been
considered by the U.S. Secretary of Defense, and by the Commander-in-
Chief of U.S. forces in the Persian Gulf, and not solely for reasons of
deterrence. Nuclear weapons use was not ruled out a priori. Because
nobody in authority at that time believed that Iraq had a nuclear weapons
capability, or even a chemical capability that could seriously disrupt U.S.
conventional operations, these nuclear ruminations did not go anywhere.
Indeed, one of the reasons most often suggested for the absence of serious
consideration of a nuclear response was condence that coalition forces
could effectively operate even if Iraq employed chemical weapons.
41
Powell himself was drafting an explicitly non-nuclear deterrent mes-
sage to Iraq on January 15 listing a host of conventional attacks the United
States and its coalition partners would mount in the event of a chemical
or biological attack. Specic targets such as the merchant eet, the road
and rail network, ports, oil facilities, commercial airports, and even dams
on the Tigris and Euphrates were on the list.
42
These targets, particularly
the dams, are similar to those suggested the preceding autumn by the
principal air commanders in the theater as possible responses to an Iraqi
chemical attack.
43
I argue above that dealing with a nuclear-armed Iraq would have
demanded a political and military strategy of intrawar deterrence. Such
a strategy depends both on will and capability. Will, in particular, must
be systematically communicated to the adversary. Evidence from the
actual crisis on the matter of Iraqi chemical weapons use shows that such
a strategy was followed. Numerous messages, public and private, formal
and informal, were deliberately sent to Iraq. These messages threatened
a massive response if Iraq employed chemical weapons. The threats were
sufciently frequent and ferocious that Iraqi ofcials claim after the fact
that they were taken as nuclear threats. Circumstantial evidence suggests
41. See especially Powell, My American Journey, p. 468. The Iraqi chemical threat was
manageable. Our troops had protective suits and detection and alarm systems. In
battle, we would be fast-moving and in the open desert, not trapped as civilians might
be. A chemical attack would be a public relations crisis, but not a battleeld disaster.
42. Powell, My American Journey, p. 504.
43. Atkinson, Crusade, p. 86.
178 the coming crisis
that even some U.S. allies may have been concerned that the United States
was contemplating a nuclear response. The only thing missing from the
record is evidence that a comprehensive explanation accompanied the
threat to Saddam. The closest thing to such an explanation is the passage
in President Bushs January 9 letter to Saddam, The American people
would demand the strongest possible response, a proposition reiterated
verbally to Tariq Aziz by Secretary of State Baker. For purposes of my
argument, one would have liked to see a statement such as We will have
no choice but to retaliate massively, because we must expect to encounter
adversaries in the future who are armed with these weapons, and who
must be made to understand the grave consequences of using them
against the United States. Nevertheless, it seems fair to conclude that
the concept of intrawar deterrence and the mechanics of deterrent diplo-
macy were well understood by the Bush administration, and were em-
ployed. This lends credibility to the proposition that the kind of strategy
I outlined above would have fallen on a receptive audience of high-level
policymakers.
The Conduct of the War
the view from 19901991
Intrawar deterrence is one key means to control the risks of a large-scale
conventional war with nuclear-armed Iraq. A second potential means is
by limiting the coalitions war aims, and by constraining the coalitions
military operations consistent with these aims. In stark terms, intrawar
deterrence aims to discourage Iraq from employing nuclear weapons out
of calculation. We must also consider how to avoid provoking the Iraqis
into using their nuclear weapons out of desperation.
What should coalition war aims be? It is the coalitions irreducible
requirement to eject Saddam Hussein from Kuwait. Moreover, if this must
be done by force, it is reasonable to strive for the maximum destruction
of Iraqi military capabilities in and near Kuwait. The Iraqis must be
taught that aggression does not pay. And it is a good idea to take this
opportunity to whittle down Iraqi military power to help stabilize the
region.
There are other plausible and tempting objectives beyond these, but
it may be imprudent to pursue them. It would be wonderful to be rid of
Saddam entirely. If Saddams regime cannot be eliminated, it would be
benecial not only for the region, but for the future of U.S. nonprolifera-
tion policies, to eliminate Iraqs weapons of mass destruction. Unhappily,
for the sake of avoiding Iraqi resort to nuclear weapons out of despera-
chapter 6 179
tion, the coalition probably must forego these objectives. The goals of the
war needed to be depersonalized. The coalition should not even hint
that Saddam Hussein is the target of the war; indeed, it should explicitly
say that neither he, nor the sovereignty of Iraq within its prior bounda-
ries, is at issue. Only Kuwait is at issue. A cornered rat with a nuclear
weapon is a pretty dangerous animal. Leave it an escape route.
Can Iraqs nuclear weapons program safely be included as a military
objective of the coalitions conventional military attack? Since Iraq al-
ready has a bomb, this project seems too dangerous; it might put Iraq in
a use or lose situation. During the Cold War, it was generally believed
that U.S. and Soviet nuclear forces were the family jewels. Though each
side targeted the others nuclear forces, neither sides leadership doubted
the grave risks associated with the initiation of a counterforce campaign.
Though Iraq is new to the nuclear club, and possesses only a few weap-
ons, it is reasonable to assume that it places a high value on these
weapons. The coalition ought not to count on Saddam Husseins accept-
ing his gradual nuclear disarmament. It is plausible that he would fear
that once the nuclear weapons were gone, the coalition would feel free
to do anything it wantedperhaps even use a nuclear weapon on him.
Though the last possibility might seem slight, an attempt to topple the
regime would seem quite plausible. Saddam might calculate that the use
of a single nuclear weapon, perhaps only as a demonstration, would be
enough to cause the coalition to cease its military campaign to disarm
him.
The coalitions military strategy should be amended to help limit
escalatory risks. In particular, we must ensure that conventional opera-
tions do not create unintended damage that would be construed either
as an attack on the regime, or as an attack on the weapons of mass
destruction that Saddam Hussein may view as his secure second-strike
capability. If U.S. military planners could be very sure that these capabili-
ties could be quickly destroyed, then it might be reasonable to risk such
an attack. But it seems unlikely that U.S. military planners could offer
such assurances.
To ensure that Saddam does not overestimate the hostile intent of the
coalition, considerable restraint in military operations seems in order.
Most notably, the coalition will have to forego the bulk of the strategic
bombing campaign now under consideration. The coalitions political
leadership will need to draw some notional lines to constrain its military
operations, perhaps the boundaries of what is now called the Kuwait
Theater of Operations (KTO), Kuwait and Southern Iraq. Intense mili-
tary operations will be permitted inside the KTO; only limited operations
would be permitted in the rest of Iraq.
180 the coming crisis
Because Saddam will not be forced to pay a cost for his aggression
in terms of damage to his country, the full cost would be extracted from
Iraqi eld forces. Annihilation of Iraqi combat forces in the theater should
be an explicit military goal. The troops can go home, but the equipment
stays. The message to Saddam, and to those who might follow, is that the
forces one invests in aggressive war simply are lost.
inferences from the actual crisis
What is the likelihood that these notions of limited war would have
found their way into actual coalition behavior? In the aftermath of Viet-
nam, the generally negative appraisal of constraints placed both on mili-
tary means and political ends in that war has made limited war a very
problematical concept in security policy circles, both civilian and military.
U.S. planning for conventional war with the Soviet Union did not dem-
onstrate any great sensitivity to the question of whether conventional
threats against nuclear forces might be escalatory.
44
Yet, both the objec-
tives and conduct of Desert Storm were wrapped in political constraints.
Positive and negative civilian intervention into military operational plan-
ning on both diplomatic and domestic political grounds occurred. Indeed,
Secretary of Defense Richard Cheney went to considerable lengths to
educate himself to facilitate such intervention.
A limited war aim was the principal form of restraint practiced
during the war, though the precise political reasons for this limitation are
murky. Specically, it was understood from the beginning that the coali-
tion would not invade and occupy Iraq with an eye toward changing its
government. Former Secretary of State James Baker states that a march
to Baghdad was never on the table, because everyone understood it was
costly and unnecessary.
45
General Powell suggested that the Hitler rheto-
ric be toned down because it could imply that Saddam would in fact be
ousted, which was not a primary war aim.
46
At the February 27, 1991,
Oval Ofce meeting where the decision to end the war was made, Sec-
retary of State Baker argued, We have achieved our aims. We have gotten
them out of Kuwait.
47
Powell believed that the war had a specic and
limited objective, and in the controversial end-game supported a rapid
termination of the ground war since this limited objectivethe liberation
44. Barry R. Posen, Inadvertent Escalation, Conventional War and Nuclear Risks (Ithaca,
N.Y.: Cornell University Press, 1991). See especially chaps. 2 and 4.
45. James Baker with Thomas DeFrank, The Politics of Diplomacy (New York: Putnam
and Sons, 1995), p. 437.
46. Powell, My American Journey, p. 491.
47. Gordon and Trainor, The Generals War, p. 416.
chapter 6 181
of Kuwaithad been achieved.
48
The decision to end the war was made
even though Saddam Hussein was still in power. Moreover, though the
key decision-makers were more optimistic about the damage to Iraqi
weapons of mass destruction than ultimately proved warranted, they
knew then that Iraq still possessed chemical weapons stocks.
49
Initial accounts of the war suggested that the United States was
somehow restrained from intervention inside Iraq by its regional coalition
partners. Two of the more thorough scholars of the war argue the oppo-
site; both Saudi Arabia and Turkey supported intervention during the
Iraqi civil war that followed Desert Storm.
50
Therefore, in spite of the
ambitious rhetoric that provided the public rationale for the war, and
local support for direct intervention in Iraq, it seems reasonable to con-
clude that the United States was not willing to pay very much to over-
throw Saddam Hussein. The U.S. war aims were limited by its sensitivity
to costs.
If aims were limited, so were means. Secretary of Defense Richard
Cheney went to great lengths to develop his own ability to assess profes-
sional military advice. He asked for, and received, what was essentially
a short course in military operational planning.
51
He apparently told Paul
Wolfowitz that though he did not intend to micromanage the preparation
of the war plan, he did intend to own it when its nished.
52
And he
was not shy about making his own suggestions. The initial briengs by
the Central Command (Centcom) planners for a possible counteroffensive
into Kuwait disturbed him. Colin, I have been thinking about this all
night. I cant let Norm do this high diddle diddle up-the-middle plan, I
just cant let him do it.
53
Cheney proceeded to make his own inquiries
and became an advocate of a large-scale two-division raid into western
Iraq. Though this suggestion was ultimately deected by Schwarzkopf
and Powell, it did contribute to the development of the ultimate left-hook
plan.
Limitations were also suggested on the air war. President Bush was
particularly concerned that targets of special historical or cultural sig-
nicance not be struck. He raised his concerns at two briengs, one in
48. Powell, My American Journey, pp. 519527.
49. Gordon and Trainor, The Generals War, p. 414.
50. Ibid., pp. 454, 456.
51. Woodward, The Commanders, p. 317.
52. Ibid., p. 316.
53. Gordon and Trainor, The Generals War, p. 141.
182 the coming crisis
October and one in January.
54
A different kind of discussion arose over
the issue of direct attacks on presumed Iraqi biological weapons storage
facilities. Out of fear of the collateral damage that could have been
produced had the viruses been inadvertently released, Cheney wanted to
be quite sure that there was a very high probability of complete destruc-
tion.
55
The actual conduct of the crisis does suggest that limited war
concepts were not entirely alien to the thinking of the principal decision-
makers. Constraints were placed on both ends and means. Nevertheless,
there is no evidence of fear that attacks on any of Saddams weapons of
mass destruction could conceivably produce use or lose behavior.
Thus, there is only qualied support for the proposition that
decision-makers would have followed the advice outlined above, and
avoided attacks on Iraqi nuclear weapons in order to avoid escalation.
Iraqs nuclear, biological, and chemical weapons programs and its missile
delivery capabilities were a military objective of the coalition. Had U.S.
decision-makers understood that these had only been partially damaged
by the forty-three-day air war, it is possible that the air campaign would
have continued after the liberation of Kuwait. Yet it was then understood
that some chemical weapons and missiles had survived the war. The costs
or risks of additional attacks upon them were apparently deemed to have
exceeded the potential benets. That this tradeoff was made suggests that
decision-makers would at least have considered seriously the costs and
risks of direct attacks on Iraqi nuclear weapons.
War Outcome
How would the war have gone if the coalition had fought a limited war
against a nuclear-armed Iraq? Would Iraq have used nuclear weapons?
54. After the October 11 brieng to the president on air and ground war plans, Bush
told Powell to make doubly sure that no targets of religious or historic signicance
were on the target list. He apparently did not want to anger unnecessarily the Iraqi
people or Arabs throughout the Middle East. Ibid., p. 136. On the evening of January
13, Cheney reviewed the target list with the president, to ensure that Bush was aware
of potential points of controversy. The president asked that one group of targets be
dropped, statues of Saddam and triumphal arches thought to be of great psychologi-
cal value to the Iraqi people as national symbols. Woodward, The Commanders, p. 353.
55. Gordon and Trainor, The Generals War, pp. 191193. The authors say that the sites
in question were not in fact biological weapons sites; Atkinson, Crusade, pp. 8890,
suggests that there was a six-week debate on whether or not to attack these targets.
Both Schwarzkopf and General Horner, his air commander, favored the attack and
ultimately prevailed with a technical argument on how it could be done successfully.
chapter 6 183
Would the conventional victory have been as splendid? What message
would have been sent to other aggressors? What message to those con-
templating a nuclear weapons program?
I doubt that Iraq would have used nuclear weapons under these
conditions. There would have been no political or military incentive, and
many disincentives.
56
Saddam did not use chemical or biological weap-
ons under conditions of conventional disaster; why would he have used
nuclear weapons?
57
The more important question, however, is whether
Western leaders would have believed that they could deter nuclear esca-
lation by Iraq. It is difcult to judge. My argument is that once they had
taken reasonable steps to control the risks, the costs and risks of inaction
would have seemed greater than the costs of action.
The war would probably have gone about as well as it did, maybe
better. But this is a statement that benets entirely from hindsight. It is
now clear that the bombing of Iraq proper did not play the decisive role
in the ejection of Iraqi troops from Kuwait. Bombing in the KTO did play
a decisive role. Iraqi command and control was degraded but not sev-
ered.
58
Since Iraq had many months to prepare the theater, it turned out
that the interdiction of supplies into the theater did not matter much.
There were plenty of weapons, munitions, food, fuel, and troops in
southern Iraq and even Kuwait. The hungry troops were at the front; they
were hungry because the supply lines to their in-theater depots were
severed, not because the supply lines to Baghdad were severed. The Iraqi
air force was so bad that it would have been blown out of the sky had it
tried to affect the ground battle. It is not clear that killing the air force on
56. Aharon Levran shrewdly observes that during the war with Iran, Saddam did
not employ chemical weapons indiscriminately; they were employed mainly in situ-
ations where the risk of decisive Iranian ground attacks was high, and largely on Iraqi
soil. Thus, even against an adversary with no in-kind retaliatory capability, Saddam
saw it in his interest to observe some limits, and employed chemicals as weapons of
last resort. See Lerran, Israeli Strategy After Desert Storm: Lessons of the Second Gulf War
(Frank Cass: London, 1997), pp. 6871.
57. The Pentagon concedes that there was one inadvertent release of chemical agents
immediately following Desert Storm as a result of U.S. demolition of an Iraqi weapons
dump. A White House panel appointed to study the question found substantial
evidence that some U.S. Marines were exposed to chemical agents during the war.
Philip Shenon, Panel Says Pentagon Ignored Signs of Poison Gas, New York Times,
October 31, 1997. Jonathan Tucker, a researcher dismissed from the White House panel,
believes that considerable evidence suggest that the Iraqi forces engaged in sporadic,
uncoordinated chemical warfare during the gulf war. Philip Shenon, Weapons Ex-
pert Tells of Possible Iraqi Gas Attacks in Gulf War, New York Times, April 25, 1997.
58. Thomas A. Keaney and Eliot A. Cohen, Gulf War Air Power Survey: Summary
Report (Washington, D.C.: U.S. Department of the Air Force, 1993), p. 70.
184 the coming crisis
the ground, in its shelters, was preferable to killing it in the air.
59
More-
over, it was the air forces inability to compete in the air that drove it to
hide in shelters. Attacking the shelters simply drove aircraft to Iran, a
bizarre and not wholly benecial outcome. An irony here is that many
sophisticated aircraft and munitions would have been diverted from their
early campaign against bunkers, bridges, and aircraft shelters, to the more
serious business of wrecking Iraqs theater forces; this might have has-
tened their collapse.
Offensive air attacks on Iraqi Scud missiles destroyed few if any of
them.
60
The pressure of sustained air attacks probably did complicate and
hasten launch procedures, eroding the already poor accuracy of the sys-
tem. Moreover, to avoid detection and destruction, Scud crews seem to
have launched less frequently than they might have otherwise. And salvo
ring was inhibited. In the end, this reduced but did not eliminate the
effectiveness of Scud missiles as terror weapons.
An Iraqi nuclear weapons program would have complicated Sad-
dams calculations about conventional Scud launches. Scuds might have
been his preferred nuclear delivery system. Would he have wanted to
bring down Western air power on them if he did not have to do so?
Perhaps concerns about the survival of nuclear Scuds would have dis-
couraged him from ring conventional Scuds. On the other hand, he
might have felt safe behind his nuclear shield, and launched the conven-
tional Scud attacks in the same fashion as actually occurred in the Gulf
War. He seems to have hoped to provoke Israeli counterattacks that
would in turn detach the Arab members of the coalition. The United
States would have had to retaliate in some way against conventional Scud
launches to prevent the Israelis from doing so. This would have under-
mined a limited war strategy, but it would have been the lesser evil.
59. One could argue, however, that attacks on the airforce bases all over Iraq exerted
debilitating effect on Iraqi offensive air capability in excess of what one might infer
from the total count of aircraft destroyed or own to Iran, about half the total
inventory. Air war planners would have argued that it was unduly risky to allow the
Iraqis a sanctuary where they might plan and organize massed, chemically armed air
attacks. A possible compromise would have been to attack and suppress the Iraqi
airelds within effective unrefuelled, tactical, range of the KTO, particularly those
airelds hosting the most effective Iraqi attack aircraft. Naturally, air defense and
command and control assets protecting these bases would also need to be suppressed.
60. Keaney and Cohen, Gulf War Air Power Survey, write, Although Iraqs average
weekly launch rate of modied Scuds during Desert Storm (14.7 launches per week)
was lower than it had been during the 1988 war of the cities, and while launch rates
generally declined over the course of the Gulf War, the actual destruction of any Iraqi
mobile launchers by xed-wing Coalition aircraft remains impossible to conrm
(p. 83). See also pp. 8990.
chapter 6 185
Important lessons about what nuclear weapons are not good for
would have emerged if the coalition had fought a limited war against a
nuclear-armed Iraq. They are not shields for conventional conquests.
Where important great power interests are engaged, they are not the great
equalizer. Moreover, the often expensive conventional forces that one
must bet in wars of conquest face a great risk of annihilation. This lesson
might affect those who are meant to take orders, not merely those who
give them. Field commanders might see the odds of a successful coup at
home to be better than the odds of surviving a war with the United States.
Nevertheless, a lesson would have been taught about what nuclear
weapons are good for. They are great instruments to deter threats against
ones homeland, even conventional threats, even by a great power, even
when you are in the wrong. Moreover, by invoking nuclear deterrence,
the coalition would itself have demonstrated in the broadest sense the
continued utility of nuclear weapons in international politics.
61
Thus, any
hopes of delegitimizing these weapons, or sustaining a condence
game regarding their supposed inutility, would be dashed. Some damage
would have been done to the cause of nonproliferation, but not nearly as
much as would have been done by a failure to act.
The Future
It can be argued that all I have done here is to tell a theoretically informed
alternative story, a different story than is usually implied by the question
What if Iraq had had nuclear weapons? Given the nature of my meth-
odology this is unavoidable. Nevertheless, the analysis does help us think
about the future.
The United States is indeed working energetically to limit the number
of nuclear powers in the world. Policymakers struggled to ensure that
only one nuclear power emerged from the wreckage of the Soviet Union,
with apparent success. They have worked with considerable effectiveness
to dismantle Iraqs nuclear weapons program, and have at least limited
North Koreas nuclear materials production capability.
Some U.S. policy analysts call for the abolition of nuclear weaponry,
in part to strengthen the legitimacy of the NPT. The problem of fullling
the Article VI treaty pledges of the nuclear weapons states to seek com-
61. This has apparently occurred. Commenting on the 1995 revelations by Tariq Aziz
that Iraq was deterred from chemical and biological weapons use by what its leaders
perceived to be a nuclear threat, a French defense ofcial averred, Did Tariq say they
were deterred? Whatever the reason, we like that kind of testimonial. Joseph Fitchett,
Nuclear States See Vindication: Threat of Annihilation Deterred Iraq, They Say,
International Herald Tribune, September 12, 1995.
186 the coming crisis
plete nuclear disarmament was a key issue of the 1995 NPT review
conference.
62
The U.S. government hopes that there will be further acces-
sions to the treaty, especially by India and Pakistan, which so dramatically
revealed their possession of such weapons in 1998, and by Israel, which
is widely suspected to have nuclear weapons.
The Pentagon has a counterproliferation initiative, which includes
among other projects research and development on various offensive and
defensive measures to use conventional means to neutralize the nuclear
weapons of other countries in the event of crisis.
63
These means must be
conventional rather than nuclear in order to protect the devaluation of
nuclear weaponry now presumed to be necessary to support nonprolif-
eration diplomacy.
64
Some hint that these capabilities could be employed
62. George Bunn, Roland Timerbaev, and James Leonard, Nuclear Disarmament:
How Much Have the Five Nuclear Powers Promised in the Non-Proliferation Treaty?
(Washington, D.C.: Lawyers Alliance for World Security, Committee for National
Security, and the Washington Council on Non-Proliferation, June 1994), pp. 115,
provides an exhaustive discussion of Article VI, which reads, Each of the Parties to
the Treaty undertakes to pursue negotiations in good faith on effective measures
relating to cessation of the nuclear arms race at an early date and to nuclear disarma-
ment, and on a treaty on general and complete disarmament under strict and effective
international control. The authors tend toward a literal interpretation of the clause.
63. See Cambone and Garrity, The Future of United States Nuclear Policy, pp. 87
90, for a summary of the counterproliferation initiative.
64. See, for example, former Secretary of Defense William Perrys statement, I cant
envision the circumstances in which the use of nuclear weapons would be reasonable
or prudent military action. Meet the Press transcript, NBC News, April 3, 1994,
pp. 78, cited in Payne, Deterrence in the Second Nuclear Age, p. 139. New Pentagon
guidelines for the use of nuclear weapons, however, seem to include options for
retaliating against smaller rogue states that might use chemical or biological weap-
ons. See Steven Lee Myers, U.S. Updates All-Out Atom War Guidelines, New York
Times, p. A3; see also Hans Kristensen, Targets of Opportunity, Bulletin of the Atomic
Scientists (September/October 1997), pp. 2228. A subtle change can also be observed
between the 1996 and 1997 versions of the Pentagons Proliferation: Threat and Response.
In the 1996 version the possibility of nuclear retaliation against an adversary that
employed weapons of mass destruction against the United States is only hinted at in
the preface, and, as far as I can discern, discussed not at all in the body of the report.
Ofce of the Secretary of Defense, Proliferation: Threat and Response 1996 (Washington,
D.C.: U.S. Government Printing Ofce [GPO], April 1996). In the 1997 version,
however, a more explicit statement is offered in the body of the text. In the introduc-
tory passages of the section DOD Response, the following appears: DOD is under-
taking a variety of programs and activities to deter the use of NBC weapons against
U.S. and allied forces, as well as against the territories of the United States and its
friends and allies. The effectiveness of these efforts will depend on the perceptions
and assessments of potential aggressors who possess NBC weapons regarding the
resolve of the United States to deal with such threats. Indeed the knowledge that the United
States has a powerful and ready nuclear capability is a signicant deterrent to the use of these
weapons. Effective deterrence will depend on a range of nuclear and conventional
chapter 6 187
preventively, to ensure that nascent nuclear weapons states never com-
plete their programs. Operational and tactical adaptations that might
reduce the vulnerability of U.S. military forces to weapons of mass de-
struction are also under study. The coalition diplomacy of confrontations
with new nuclear powers has also received some attention.
65
Although a combination of international legal instruments and uni-
lateral military means may do much to reduce the proliferation problem,
it is unlikely that the success will be total. North Koreas future coopera-
tion is not a sure thing. New nuclear hopefuls will surely arise, with Iran
as perhaps the most likely candidate. Though Ukraines program to
relinquish the nuclear weaponry it inherited from the Soviet Union is
complete as of this writing, its substantial civil nuclear energy industry
provides a foundation for a future weapons program. The NPT nuclear
weapons states will continue to reduce the number of nuclear weapons
they hold, but real progress to complete nuclear disarmament is unlikely.
Nuclear weapons are the sole source of modern Russias great power
status, and the sole bulwark of the security of eastern Russia against
1,200,000,000 Chinese. It is fatuous to imagine that Russia would support
complete nuclear disarmament. Thus, the complaints of bad faith by
the NPT non-nuclear states against the nuclear states are likely to in-
crease rather than diminish, degrading somewhat the legitimacy of the
treaty.
The counterproliferation initiative will undoubtedly yield some inter-
esting technology and weapons. But it is improbable that a high-
condence capability to neutralize even small nuclear forces with con-
ventional means will emerge. Coalition air power destroyed few if any
mobile Scud missiles in Desert Storm, suggesting that a moderately
well-organized mobile ballistic missile force can be incredibly elusive.
66
response capabilities, as well as active and passive defenses and supporting command,
control, communications, and intelligence (my italics). See Ofce of the Secretary of
Defense, Proliferation: Threat and Response 1997 (Washington, D.C.: U.S. GPO, Novem-
ber 1997), http://www.defenselink.mil/pubs/prolif97/secii.html# prevention.
65. See, for example, Thomas J. Hirschfeld, The Impact of Nuclear Proliferation:
Final Report, Center for Naval Analysis, July 1995, which devotes considerable
attention to force structure and operational requirements of confrontations with
nuclear-armed countries. It pays less attention to alliance diplomacy.
66. Keaney and Cohen, Gulf War Air Power Survey, pp. 7890. The report estimates
that Iraq had, at most, thirty or so mobile [Scud] launchers at the start of the war . . .
few mobile Scud launchers were actually destroyed by Coalition aircraft or special
forces during the war . . . Coalition air power does not appear to have been very
effective against this militarily insignicant target category (pp. 8990). Roughly
188 the coming crisis
The Patriot missiles erratic performance in Desert Storm also provides
some cause for doubt about the likely success of future defensive pro-
grams, though it was not designed for this role, and purpose-built theater
ballistic missile systems would likely do better.
67
Finally, a series of preventive wars to neutralize nascent nuclear
programs is unlikely; the United States would not have waged such a
war against Iraq in the absence of its attempted conquest of Kuwait. The
United States probably will not do so against Korea. Even if it does, one
doubts that the United States can muster the political will to do so
repeatedly.
One day, the United States will probably face a crisis caused by
aggression of a nuclear-armed regional power against an important U.S.
interest. The United States would be unlikely to have a combination of
offensive and defensive capabilities that could eliminate an adversarys
nuclear forces. Thus, the United States will face a dening moment of the
kind I have outlined. If so, it will have to assess the very factors discussed
here. While the highly competitive political world that could follow U.S.
inaction may be perceived as uninviting, more thinking must be done
about how the United States might live in that world. Are any of the three
alternative strategies outlined abovepreventive war, little NATOs, or
isolationismacceptable alternatives for U.S. foreign and security policy?
Perhaps I have overlooked other diplomatic and military strategies that
would be preferable to the risks of a nuclear confrontation. If the United
States chooses to act, however, it is difcult to see any other foundation
upon which to rest its action but intrawar nuclear deterrence, and limited
war. These expedients require as much attention and analysis as the rest
of the proliferation agenda, but there seems to be no natural constituency
for this work in the policy community. Intrawar deterrence raises the
political salience of nuclear forces, and thus weakens the nuclear devalu-
ation tactic that is now an element of U.S. nonproliferation policy. Limited
war strategies, especially limits on military means, go against the grain
of military wisdom, and exacerbate civil-military conict in the United
States. If both the costs and risks of inaction, and the costs and risks of
action seem unpalatable, then diplomatic strategies must be devised that
offer some hope of redening the situation to make inaction appear
1500 strikes were carried out against targets associated with Iraqi ballistic missile
capabilities. Perhaps another 1000 Scud Patrol sorties dropped on other targets
(pp. 8384).
67. Ballistic missile designers may devise simple penetration aids as well, compli-
cating the task of Anti-tactical Ballistic Missile systems more advanced than Patriot.
chapter 6 189
differently than how I have portrayed it.
68
This would seem unusually
creative diplomacy.
68. For example, when Iraq invaded Kuwait, Saddam Hussein argued that Kuwait
was a lost province of Iraq, which he was merely reclaiming. Few would credit this
assertion. But if the United States wished neither to risk nuclear escalation in a military
campaign of liberation, nor to risk the lessons that putative aggressors might draw
from the episode, U.S. diplomacy might simply have found it expedient to accept,
indeed to argue, the legitimacy of the Iraqi claim. In this way, the basic principle that
nuclear weapons deter attacks on ones own territory, but do not protect conquests,
would have been sustained. Such a diplomatic campaign might have limited the
lessons other states might have drawn from the U.S. concession of Kuwait to Iraq, and
thus reduced the nuclear proliferation damage of the concession. This might not work
more than once.
190 the coming crisis
Chapter 7
Containing Rogues and
Renegades: Coalition
Strategies and
Counterproliferation
Stephen M. Walt
Since the end of the Cold War, U.S. defense planners have emphasized
the threat from a set of countries whose values and ambitions appear to
be sharply at odds with U.S. goals and international norms. In addition
to a deep-seated hostility toward the United States, so-called rogue or
backlash states such as Iraq, Iran, North Korea, and Libya are believed
to harbor aggressive designs toward one or more of their neighbors and
are regarded as especially willing to use violence to advance their foreign
policy goals.
1
Each of these states has taken steps to acquire weapons of
mass destruction (or WMD, usually dened as nuclear, biological, or
chemical weapons), with varying degrees of success. The threat from
these potential nuclear rogues, which was rst acknowledged in the
Department of Defenses Bottom-Up Review in 1993, led to the Defense
Counterproliferation Initiative that same year and continues to guide U.S.
efforts to isolate these regimes and retard or reverse their nuclear pro-
grams.
2
1. While serving in the Clinton administration, former U.S. National Security Adviser
W. Anthony Lake characterized rogue (or backlash) states as undemocratic re-
gimes that promote radical ideologies, repeatedly violate human rights, and whose
behavior is often aggressive and deant. He also described them as increasingly
isolated and guided by a siege mentality. See Lake, Confronting Backlash States,
Foreign Affairs, Vol. 73, No. 2 (1994), pp. 4546. Current U.S. defense plans call for
military forces sufcient to help defeat aggression in two nearly simultaneous major
regional conicts, specically mentioning North Korea and Iraq in this context. See
A National Security Strategy for a New Century (Washington, D.C.: The White House,
1998); and for background, Michael T. Klare, Rogue States and Nuclear Outlaws (New
York: Hill and Wang, 1995).
2. In 1993, Secretary of Defense Les Aspin identied four main threats to U.S. security
and stated that the one that most urgently and directly threatens Americans at home
and American interests abroad is the new nuclear danger. . . . [It] is perhaps a handful
The emergence of rogue states armed with WMD is believed to be
especially dangerous for at least four reasons.
3
First, such regimes are
believed to be more likely to use these capabilities than other states would
be, either because they are ideologically committed to altering the status
quo or because they are less sensitive to the human costs that the use of
such weapons might entail.
4
Second, a rogue state armed with WMD might be able to deter other
states from intervening against it, thereby facilitating its efforts to coerce
or conquer its neighbors. In particular, some U.S. ofcials suggest that the
threat of retaliation with WMD might deter the United States from using
its superior conventional capabilities to counter conventional aggression,
thereby placing current U.S. allies in jeopardy. According to former U.S.
Secretary of Defense Les Aspin, for example, Today, the United States is
the biggest kid on the block when it comes to conventional military forces
and it is our potential adversaries who may attain nuclear weapons. So
nuclear weapons may still be the great equalizer; the problem is the
United States may now be the equalizee.
5
of nuclear weapons in the hands of rogue states or even terrorist organizations. See
Secretary of Defense Les Aspin, The Defense Departments New Counterproliferation
Initiative, Address to National Academy of Sciences Committee on International
Security and Arms Control, Washington, D.C., December 7, 1993; Aspin, The Bottom-Up
Review: Forces for a New Era (Washington, D.C.: U.S. Department of Defense, 1993); and
Ofce of the Deputy Secretary of Defense, Report on Non-Proliferation and Counterpro-
liferation Activities and Programs (Washington, D.C.: U.S. Department of Defense, May
1994). More recently, Secretary of Defense William Cohen has warned that the United
States faces a heightened prospect that regional aggressors, third-rate armies, terrorist
cells and even religious cults will wield disproportionate power by usingor even
threatening to usenuclear, biological, or chemical weapons against our troops in the
eld and our people at home. See Ofce of the Secretary of Defense, Proliferation:
Threat and Response (Washington, D.C.: U.S. Government Printing Ofce [GPO], 1997).
3. The potential dangers created by the spread of weapons of mass destruction are
summarized in Peter R. Lavoy, The Strategic Consequences of Nuclear Proliferation,
Security Studies, Vol. 4, No. 2 (1995).
4. For example, an ofcial in the Defense Departments Ofce of Counterproliferation
Policy has argued that many states anxious to acquire nuclear, biological, chemical
weapons and missiles (NBC/M) . . . are led by undemocratic, unpredictable leaders
whose motivations are largely opaque to outside observation and understanding.
Citing North Korea, the same ofcial noted that the relatively opaque character of
the North Korean leadership increases the difculty of persuading the regime not to
acquire NBC/M, to roll back its capabilities, or not to use them. See Paul R.S.
Gebhard, Not by Diplomacy or Defense Alone: The Role of Regional Security Strate-
gies in U.S. Proliferation Policy, Washington Quarterly, Vol. 18, No. 1 (1994), pp. 170
171.
5. Similarly, U.S. Senator John McCain once warned that a nuclear North Korea might
be more inclined to attack South Korea, because its nuclear arsenal would deter the
192 the coming crisis
Third, and following from the second point, the emergence of rogue
states armed with WMD might make threatened states more reluctant to
join alliances against them, for fear of becoming the victim of a highly
destructive attack. Such fears could inhibit U.S. efforts to contain these
regimes and could defeat multilateral efforts to moderate their interna-
tional conduct. For example, Roger Molander and Peter Wilson argue that
a regional predator will nd a small nuclear arsenal a powerful tool for
collapsing regional military coalitions that the United States might craft
to oppose such a future opponent, and former Defense Department
ofcial Zalmay Khalilzad suggests that rogue states such as Iraq and Iran
might use WMD to deter the United States and its allies from [acting]
. . . [or to] intimidate the GCC [Gulf Cooperation Council] states into not
inviting U.S. or other Western forces to intervene, thereby facilitating
renewed aggression in the Persian Gulf.
6
Finally, many experts argue that a rogue states unconventional arse-
nal would lack adequate safeguards and command-and-control devices,
thereby increasing the risk of theft, accidents, or unauthorized use. Even
if rogue leaders proved more rational than many believe, the danger
of inadvertent attacks would grow as more states acquire WMD capa-
bilities.
7
United States from counterattacking. See Aspin, Defense Departments Counterpro-
liferation Initiative; and McCain, letter to the New York Times, March 28, 1994, p. A10.
In the same spirit, John Arquilla suggests that the acquisition of nuclear weapons by
rogue states would create a situation similar to Europe in the 1970s and early 1980s,
when it began to grow clear that the U.S. threat to use nuclear weapons to defend
against Soviet conventional aggression was hollow. See Arquilla, Bound to Fail?
Regional Deterrence after the Cold War, Comparative Strategy, Vol. 14, No. 2 (1995),
p. 133. Even Kenneth Waltz, who is usually sanguine about the possible spread of
nuclear weapons, has conceded that if weak countries have some it will cramp our
style. Militarily punishing small countries for behavior we dislike would become much
more perilous. See Kenneth N. Waltz and Scott D. Sagan, The Spread of Nuclear
Weapons: A Debate (New York: Norton, 1995), p. 111.
6. See Roger C. Molander and Peter A. Wilson, On Dealing with the Prospect of
Nuclear Chaos, Washington Quarterly, Vol. 17, No. 3 (1994); and Zalmay Khalilzad,
The United States and the Persian Gulf: Preventing Regional Hegemony, Survival,
Vol. 37, No. 2 (1995), pp. 9899, 105106; and also Marc Dean Millot, Roger Molander,
and Peter A. Wilson, The Day After . . . Study: Nuclear Proliferation in the PostCold
War World (Santa Monica, Calif.: RAND Corporation, 1993); Aspin, Bottom Up Review,
p. 5; and Steven David, Why the Third World Still Matters, International Security,
Vol. 17, No. 3 (Winter 1992/93), pp. 145155.
7. See especially Scott Sagans contributions to Waltz and Sagan, Spread of Nuclear
Weapons; and also Peter Feaver, Neooptimists and Proliferations Enduring Prob-
lems, Security Studies, Vol. 6, No. 4 (1997). Important challenges to this view include
Bradley Thayer, The Risk of Nuclear Inadvertence: A Review Essay, Security Studies,
chapter 7 193
Is all this concern justied? The fear that nuclear rogues would
have profoundly destabilizing effects on world politics lies at the heart
of present efforts to fashion ambitious military responses, ranging from
sophisticated disarming attacks to elaborate types of strategic defense.
Even so, many strategic planners seem pessimistic about the prospects
for containing these regimes, and warn that the spread of WMD would
have corrosive effects on important U.S. security commitments.
This chapter offers a different assessment. The central question is
straightforward: will the acquisition of WMD by so-called rogue states
encourage or impede the formation of defensive coalitions to contain,
coerce, or in extreme cases, disarm them? Contrary to much of the
conventional wisdom, I argue in this chapter that it will not be especially
difcult to forge effective alliances against so-called rogue states, even if
they were to acquire WMD. Indeed, overstating the difculties could be
self-fullling, if it encouraged potential partners to expect more protec-
tion than is necessary or if it convinced U.S. leaders that the prospects
for success were not worth the costs and risks.
The rst section of this chapter examines the characteristics of so-
called rogue states and the difculties involved in forecasting their be-
havior. The second section draws on the theoretical literature on alliance
formation to explain why states join military coalitions, and identies the
main obstacles that can impede the formation of effective alliances and
undermine their value as a deterrent or their performance in war. The
third section examines the specic problems that rogue states armed with
WMD pose to the formation and effectiveness of coalitions, and argues
that a strategy of containment will minimize these obstacles most effec-
tively. Finally, the conclusion outlines several ways to facilitate efforts to
contain these states, assuming that this remains an important U.S. inter-
est.
What Is a Rogue State? Why Are They Dangerous?
While experts disagree about the speed with which WMD will spread,
there is a widespread consensus that many states could acquire
these weapons if they so desired.
8
In spite of renewed efforts to limit the
Vol. 3, No. 3 (1994); David Karl, Proliferation Pessimism and Emerging Nuclear
Powers, International Security, Vol. 21, No. 3 (Winter 1996/97); and Jordan Seng, Less
is More: Command and Control Advantages of Minor Nuclear States, Security Studies,
Vol. 6, No. 4 (1997).
8. See, among many others, Steve Fetter, Ballistic Missiles and Weapons of Mass
194 the coming crisis
spread of WMD (symbolized by the agreement to extend the Nuclear
Non-Proliferation Treaty), the ability of even weak states to acquire some
type of WMD has been enhanced by the growing diffusion of techni-
cal knowledge; the break-up of the Soviet Union and the resulting
uncertainty about custodianship of nuclear materials there; the
permeability of existing international safeguards; and the willingness
of several states to provide key components of nuclear, biological,
and chemical weapons technology. The recent nuclear tests by India
and Pakistan and the reconstitution of Iraqs WMD capability despite
the intrusive United Nations Special Commission (UNSCOM)
inspections regime conrm that these capabilities are within the reach of
many states. Thus, part of the concern about these rogue states
stems from the recognition that there may be little that outside powers
can do to prevent a state from acquiring WMD if it wants them badly
enough.
9
The acquisition of WMD by rogue states seems dangerous because
these states are believed to harbor particularly aggressive goals, an as-
sessment that rests on a set of core assumptions about their motivations
and likely behavior. First, virtually all analysts regard such regimes as
fundamentally hostile to the United States. Second, rogue states are as-
sumed to have inherently revisionist aims, meaning that they seek to alter
the status quo for reasons other than a desire to improve their own
security. Third, they are often regarded as especially willing to take risks,
because they are highly autocratic, inspired by a radical ideology, or
Destruction: What is the Threat? What Should be Done? International Security, Vol. 16,
No. 1 (1991); Brad Roberts, From Non-Proliferation to Antiproliferation, International
Security, Vol. 18, No. 1 (1993); Mitchell Reiss, Without the Bomb: The Politics of Nuclear
Non-Proliferation (New York: Columbia University Press, 1988), p. 23; William C. Martel
and William T. Pendley, Nuclear Coexistence: Rethinking U.S. Policy to Promote Stability
in an Era of Proliferation (Montgomery, Ala.: U.S. Air War College, 1994), pp. 79; Aspin,
New Counterproliferation Initiative; Peter D. Zimmermann, Technical Barriers to
Nuclear Proliferation, Security Studies, Vol. 2, Nos. 34 (1993); and Proliferation:
Bronze Medal Technology is Enough, Orbis, Vol. 38, No. 1 (1994); and Mark D.
Mandeles, Between a Rock and a Hard Place: Implications for the U.S. of Third World
Nuclear Weapon and Ballistic Missile Proliferation, Security Studies, Vol. 1, No. 2
(1991).
9. The motivations for nuclear acquisition are examined in Bradley A. Thayer, The
Causes of Nuclear Proliferation and the Utility of the Nuclear Non-Proliferation
Regime, Security Studies, Vol. 4, No. 3 (1995); Zachary S. Davis and Benjamin Frankel,
eds., The Proliferation Puzzle: Why Nuclear Weapons Spread (and What Results),
special issue of Security Studies, Vol. 2, Nos. 34 (1993); Stephen M. Meyer, The Dynam-
ics of Nuclear Proliferation (Chicago: University of Chicago Press, 1984); and Martel and
Pendley, Nuclear Coexistence, pp. 1827.
chapter 7 195
prone to make decisions in an irrational fashion. As a result, such
regimes are believed to be especially difcult to deter.
10
In short, the
perceived threat from rogue states rests on the belief that they are unusu-
ally bellicose, and would become even more aggressive were they to
acquire weapons of mass destruction.
11
This characterization should not be accepted uncritically. Portraying
rogue states as a looming international threat obscures their relative
weakness, international isolation, and internal vulnerabilities.
12
It also
discounts the possibility that their aggressive behavior stems largely from
insecurity rather than ideological conviction or a desire for glory or
material gain.
13
States that see their security as precarious are often more
willing to run risks (including the use of force) to improve their positions;
one need not approve of the actions of Wilhelmine and Nazi Germany,
Imperial Japan, or Bathist Iraq to recognize that intense perceptions of
threat lay at the core of their expansionist behavior. German and Japanese
elites saw themselves as tightly encircled, and believed their regimes
were destined to decline unless they took action to reverse the trend.
Similarly, Iraqs attempt to seize Kuwait probably arose less from
Husseins megalomania or from Bathist ideology than from the profound
vulnerabilities facing the Iraqi regime. In addition to the serious long-
term threat from Iran (which possesses far more latent power), Iraq was
burdened by the debts it incurred during the Iran-Iraq war, the falling
price of oil, the OPEC (Organization of Petroleum Exporting Countries)
10. As U.S. Senator John McCain has noted, the rapids of internal politicsand
external rivalriesin broad stretches of the developing world confound any condent
projections regarding stable and restrained, as contrasted with unstable and poten-
tially unrestrained actors on the proliferation stage. See his Proliferation in the
1990s: Implications for U.S. Policy and Planning, Strategic Review, Vol. 17, No. 3 (1989),
p. 12.
11. The Bottom-Up Review emphasized the danger from states set on regional domina-
tion through military aggression while simultaneously pursuing nuclear, biological, and
chemical weapons capabilities. See Aspin, Bottom-Up Review, p. 1 (emphasis added).
12. In 1996, the U.S. population was approximately 268 million, its gross national
product (GNP) was $7.6 trillion, and its defense expenditures were roughly $265
billion. By contrast, Iraq, Iran, Libya, and North Korea have a combined population of
120 million, a combined GNP of $128 billion, and their combined defense spending
was $11.4 billion. Thus, the U.S. population was 2.22 times larger than the combined
population of the four main rogue states, its GNP was nearly sixty times greater
than theirs, and it spent over twenty-three times as much on defense as they did. These
calculations are based on data in International Institute for Strategic Studies, The
Military Balance: 199798 (Oxford: Oxford University Press, 1997).
13. Bradley Thayer argues that insecurity is the principal motive behind the Iraqi,
Iranian, and North Korean nuclear programs. See his Causes of Nuclear Prolifera-
tion, pp. 495496.
196 the coming crisis
states refusal to adhere to prior price agreements, and deep internal
divisions. Seizing Kuwait promised a quick x to these problems, and the
apparent absence of an outside commitment to protect the sheikhdom
made the temptation impossible to resist.
14
The point is not to defend Iraqi expansionism, of course; the point is
simply to recognize that insecurity is often a major source of aggression.
If insecurity is the main cause of a rogue states bellicose behavior, then
the acquisition of WMD could make a seemingly rogue state less
dangerous. By contrast, if the desire to expand arises from other sources
(such as religious or ideological convictions, long-standing territorial
irredenta, etc.), then obtaining WMD might easily make a state more
bellicose. This suggests that efforts to deal with rogue states should be
informed by a systematic and dispassionate analysis of their underlying
motivations.
15
Finally, the assumption that such regimes are either irrational or
insensitive to costs is problematic. Although rogue regimes often treat
their own populations with great brutality, that is quite different from
taking steps that are likely to lead to the destruction of the regime itself.
Even if a rogue state were somewhat more likely to use WMD than a
non-rogue state, the likelihood of use is still quite low. Thus, it is by no
means obvious that nuclear rogues pose a special sort of danger.
Indeed, the alleged threat from so-called rogue states is not new.
The Soviet Union was the rst modern rogue insofar as its ruling
ideology and foreign policy behavior revealed a deep-seated commitment
to altering the global status quo.
16
U.S. policymakers were alarmed by
the Soviet acquisition of nuclear capabilities in 1949, and their rhetoric
was strikingly similar to the language now used to describe the threat
from contemporary rogue states. In 1949, for example, the National Se-
curity Council Memorandum U.S. Objectives for National Security
14. See Efraim Karsh and Inari Rautsi, Why Saddam Hussein Invaded Kuwait,
Survival, Vol. 33, No. 1 (1991); and Lawrence Freedman and Efraim Karsh, The Gulf
Conict, 199091: Diplomacy and War in the New World Order (Princeton, N.J.: Princeton
University Press, 1993), chap. 2.
15. Robert Blackwill and Ashton Carter recommend understanding the strategic
personality of a new proliferator, noting that a wider variety of proliferator person-
alities is possible than is sometimes acknowledged in writings about nuclear strategy.
See Blackwill and Carter, The Role of Intelligence, in Robert D. Blackwill and Albert
Carnesale, eds., New Nuclear Nations: Consequences for U.S. Policy (New York: Council
on Foreign Relations, 1993), p. 236.
16. The Soviet Union was treated as a pariah state by the Western powers throughout
the 1920s, and the United States did not even establish diplomatic relations with
Moscow until 1934.
chapter 7 197
(NSC-68) warned that Soviet possession of nuclear weapons would pro-
vide it with great coercive power for use in time of peace and would
discourage the victims of its aggression from taking any action . . .
which would risk war.
17
Similar fears accompanied the Chinese acqui-
sition of a nuclear capability in 1964, an event that led both U.S. and
Soviet leaders to consider preemptive strikes against Chinese nuclear
facilities.
18
As with todays rogue states, U.S. concerns about a nuclear
China were exacerbated by the unwarranted belief that the communist
government in China was highly aggressive, inherently unpredictable,
and therefore difcult to deter. As U.S. Secretary of State Dean Rusk
warned in 1966, a country whose behavior is as violent, irascible, un-
yielding, and hostile as that of Communist China is led by leaders whose
view of the world and of life itself is unreal. Such rhetoric could be
applied to contemporary rogue states virtually unchanged.
19
Not only are rogue states not a new phenomenon, but the historical
record suggests that the existing nuclear states acquired these weapons
largely for defensive purposes (i.e., in response to legitimate security
fears), or for reasons of status, and have not behaved more recklessly
since acquiring these capabilities.
20
The introduction of nuclear weapons
into South Asia ended the costly cycle of conventional war between India
and Pakistan, for example, and both regimes recognize that these weap-
ons are useful only in a deterrent role.
21
Iraqs decision to refrain from
17. See U.S. Objectives for National Security (NSC-68), in Thomas H. Etzold and
John Lewis Gaddis, eds., Containment: Documents on American Policy and Strategy,
19451950 (New York: Columbia University Press, 1978), p. 398.
18. See Gordon H. Chang, Friends and Enemies: The United States, China, and the Soviet
Union, 19481972 (Stanford, Calif: Stanford University Press, 1990), chap. 8; and Henry
A. Kissinger, White House Years (Boston: Little, Brown and Co., 1979), pp. 183187.
19. See Franz Schurmann and Orville Schell, eds., The China Reader, vol. 3: Communist
China (New York: Vintage Books, 1967), p. 508. On Chinas nuclear program, see Avery
Goldstein, Understanding Nuclear Proliferation: Theoretical Explanation and Chinas
National Experience, Security Studies, Vol. 2, No. 34 (1993); on its rather cautious
approach to the use of force, see Allen S. Whiting, The Use of Force in Foreign Policy
by the Peoples Republic of China, Annals of the American Academy of Political and Social
Science, Vol. 402 (1972), pp. 5566.
20. The Soviet acquisition of nuclear weapons seems to have had little effect on
Stalins decision to approve the North Korean invasion of South Korea in 1950. See
Sergei N. Goncharov, John W. Lewis, and Xue Litai, Uncertain Partners: Stalin, Mao, and
the Korean War (Stanford, Calif.: Stanford University Press, 1993), chap. 5.
21. For example, Indian Defense Minister George Fernandes remarked in 1998 that
it is an established fact that any country that has nuclear weapons cannot use them.
By denition, they can be used only as a deterrent. If we had to go nuclear, it was for
the purpose of possessing a nuclear deterrent that would enable us to tackle some of
the threats that we faced, only that. Quoted in New York Times, June 18, 1998, p. A6.
198 the coming crisis
using chemical weapons during the Gulf War suggests that even a leader
as risk-acceptant as Saddam Hussein can be deterred by a powerful
retaliatory capability.
22
Thus, there seems to be little reason to regard
todays rogue states as an especially novel or grave danger.
The problem, of course, is that one cannot be sure what a regime
would do were it to attain WMD. Furthermore, motivations and inten-
tions can change without warning, and weapons that were acquired for
one purpose could be used for other ends later. Thus, even if todays
rogue states are probably not as dangerous as the more alarmist rheto-
ric suggests, there are still good reasons to worry about their weapons
programs and foreign policy objectives, and good grounds for thinking
about how the United States and its allies should respond. In particular,
how will the acquisition of WMD by a perceived rogue state affect the
willingness of other states to form alliances against it? To answer this
question, let us begin by considering why states form alliances in the rst
place.
When Do States Form Alliances?
Alliances are formed so that their members can combine their capabilities
in ways that further their respective interests. Alliances can take many
forms (offensive or defensive, symmetrical or asymmetrical, formal or
informal, etc.), but the common thread in virtually all such arrangements
is each members desire to increase its security by collaborating with at
least one other state.
balancing behavior
The principal reason to form an alliance is to oppose an external threat.
23
In particular, states form military coalitions in order to balance against
the most threatening state or coalition that they face. The rationale for
this tendency is straightforward: because no supreme authority exists to
22. During the Gulf War, several U.S. ofcials made statements that were clearly
designed to deter Iraq from using its chemical arsenal against the coalitions forces.
See Freedman and Karsh, Gulf Conict, pp. 257, 288289.
23. This discussion is based on Stephen M. Walt, The Origins of Alliances (Ithaca:
Cornell University Press, 1987), esp. chap. 2. See also Stephen M. Walt, Testing
Theories of Alliance Formation: The Case of Southwest Asia, International Organiza-
tion, Vol. 43, No. 2 (1988); Alliances, Threats and U.S. Grand Strategy: A Reply to
Kaufman and Labs, Security Studies, Vol. 1, No. 3 (1992); Multilateral Collective
Security Arrangements, in Richard Shultz, Roy Godson, and Ted Greenwood, eds.,
Security Studies for the 1990s (New York: Pergamon-Brasseys, 1993); and Why Alli-
ances Endure or Collapse, Survival, Vol. 39, No. 1 (1997).
chapter 7 199
protect states from each other, states facing a possible threat will join
forces with others in order to amass sufcient power to deter or defeat
an attack.
This explanation is often framed in terms of power; states with lesser
capabilities are presumed to combine against a stronger state in order to
prevent the stronger power from dominating.
24
But it is more accurate to
say that states form coalitions to balance against threats, and power is
only one element in their calculations (albeit an important one). In gen-
eral, the level of threat that states face will be a function of four distinct
factors: aggregate power, geographic proximity, offensive capabilities,
and aggressive intentions.
aggregate power. Other things being equal, the greater a states total
resources (population, industrial capacity, military strength, etc.), the
greater harm it can inict on others. Thus, the greater a states aggregate
power, the larger the potential threat. Recognizing this, the traditional aim
of U.S. grand strategy has been to prevent any single power from con-
trolling the combined resources of Eurasia, because such an agglomera-
tion of power could pose a serious threat to U.S. security. Englands
traditional reliance on a balance of power policy reected similar moti-
vations, and contemporary concerns over Chinas economic and military
growth betray much the same sensitivity. A states aggregate power is not
the sole determinant of threat, of course, but it is always an important
one.
geographic proximity. Because the ability to project power declines
with distance, nearby states are usually more dangerous than those that
are far away.
25
As a result, states are more likely to form coalitions in
response to threats from their neighbors, and will prefer allies that are
geographically separate. This factor helps explain why, during the Cold
War, regional powers usually preferred to rely on support from a distant
superpower rather than on cooperation with other regional actors. First,
the superpowers could do more to help. Second, allying with a neighbor
can be dangerous if the neighbor becomes too strong. This motive also
24. For classic expressions of this view, see Hans J. Morgenthau, Politics Among
Nations (New York: Alfred A. Knopf, 1967); and Kenneth N. Waltz, Theory of Interna-
tional Politics (Reading, Mass.: Addison-Wesley, 1978).
25. On the so-called power-distance gradient, see Kenneth Boulding, Conict and
Defense: A General Theory (New York: Harper Torchbooks, 1962); and Albert Wohlstetter,
Illusions of Distance, Foreign Affairs, Vol. 46, No. 2 (1968). See also Kautilya,
Arthasastra, in Paul Seabury, ed., Balance of Power (San Francisco: Chandler Publish-
ing, 1965).
200 the coming crisis
explains why the superpowers efforts to recruit regional clients during
the Cold War were only partly successful. Although Moscow and Wash-
ington sought allies primarily to counter each other, their clients in the
developing world were usually more worried about regional or internal
threats and relatively unconcerned about the global balance.
26
offensive capabilities. States with large offensive capabilitiesde-
ned as the ability to threaten the sovereignty or territorial integrity of
other statespose a greater threat than states whose capabilities are
designed primarily to defend their own territory. As a states offensive
capabilities increase (either because its military expenditures are growing
or because its armed forces are being tailored for offensive warfare), other
states will be more inclined to ally against it. The ability to undermine
other regimes via propaganda or subversion can be a potent source of
threat as well, which explains why revolutionary states that see them-
selves as a model for others are usually regarded as dangerous, even
when they are militarily weak.
27
offensive intentions. States with aggressive aims are obviously more
dangerous than states that are satised with the status quo. Even a
relatively weak state is likely to face a defensive coalition if other states
view its intentions as especially bellicose. Other things being equal, there-
fore, the belief that a state harbors offensive intentions will increase the
likelihood that a countervailing coalition will form against it.
Together, these four factors explain why potential hegemons like
Napoleonic France, Wilhelmine Germany, and Nazi Germany eventually
faced overwhelming opposing coalitions: each of these states was a great
power lying in close proximity to others, and each combined large offen-
sive capabilities with extremely aggressive aims. The same factors explain
why the Soviet Union was badly overmatched during the Cold War: it
was the second most powerful state in the system, it was geographically
close to the medium powers of Europe and Asia, it possessed a large
military establishment whose forces and doctrine were tailored for offen-
sive wars of conquest, and it never abandoned its public commitment to
world revolution. Although the United States had greater overall capa-
bilities (including a large network of overseas bases and a signicantly
26. See Walt, Origins of Alliances, pp. 158165, 266; and Steven David, Choosing Sides:
Alignment and Realignment in the Third World (Baltimore, Md.: Johns Hopkins Univer-
sity Press, 1991).
27. See Stephen M. Walt, Revolution and War (Ithaca: Cornell University Press, 1996);
and Stephen M. Walt, Revolution and War, World Politics, Vol. 44, No. 2 (1992).
chapter 7 201
greater capacity for global military intervention), its distance from the
other centers of world power made it the perfect ally for the medium
powers that were most directly threatened by the Soviet Union.
Balance-of-threat theory also explains why the United States was
able to assemble such a large and cohesive coalition during the Gulf War.
Although Iraq was not a great power, its military forces were growing,
its conquest of Kuwait threatened Western oil supplies, and Saddam
Husseins behavior suggested that he had extremely aggressive ambitions
and was willing to run impressive risks to achieve them. Adroit diplo-
macy helped bring the coalition into existence, but the essential precon-
ditions for an effective coalition were certainly present.
28
The theory also
tells us why it has been more difcult to maintain the Gulf War coalition
after its smashing victory over Iraq. Because Iraqs power is greatly
diminished, other states feel less need to join forces to oppose it. More-
over, states like Russia and France are now worried about excessive U.S.
inuence, and so are increasingly reluctant to support U.S. policy whole-
heartedly.
bandwagoning behavior
Although the desire to balance threats is the main reason why states form
coalitions, some states may prefer to jump on the bandwagon and join
forces with the dominant (or most threatening) power rather than attempt
to form a coalition against it. This may be done either for defensive or
offensive reasons: some states may seek the stronger side to appease it;
others join the strong side to prot from its victory.
Several factors will affect a states propensity to bandwagon. First,
weak states are more inclined to bandwagon than are strong states; they
can do little to affect the outcome and must choose the winning side at
all costs.
29
The likelihood of bandwagoning also increases when a state
faces an imminent threat and potential coalition partners are unavailable.
Even if such a state would prefer to balance, it may be forced to band-
wagon (or adopt a neutral position) if it cannot nd adequate support.
Third, bandwagoning is more likely to occur in the latter stages of a war,
when the outcome is clear and states rush to curry favor with the victors
28. See David Garnham, Explaining Middle Eastern Alignments During the Gulf
War, Jerusalem Journal of International Relations, Vol. 13, No. 3 (1991); and Freedman
and Karsh, Gulf Conict, chaps. 6, 7, 25.
29. Eric Labs suggests that even weak states are reluctant to bandwagon, though they
will do so as a last resort. See his Do Weak States Bandwagon? Security Studies,
Vol. 1, No. 3 (1992). See also Michael Handel, Weak States in the International System
(London: Frank Cass, 1981); and Robert L. Rothstein, Alliances and Small Powers (New
York: Columbia University Press, 1968).
202 the coming crisis
or to extract spoils from the vanquished. Lastly, revisionist states will be
more inclined to bandwagon, because they are especially interested in
reaping the benets from victory and must take pains to pick the likely
winners.
30
Mussolinis behavior following the German attack on France
in 1940 offers a nice example of all three factors; Italy was a weak state
with revisionist aims, and Mussolini waited until Frances defeat was
apparent before declaring war.
Although great powers often fear that potential allies will align with
the strongest or most threatening state, this fear receives little support
from international history. Every modern attempt to achieve hegemony
in Europe has been thwarted by a powerful balancing coalition. More
recent examples of balancing behavior include the coalition that fought
the Gulf War against Iraq, the resurgence of the Association of Southeast
Asian Nations (ASEAN) following the Vietnam war, the Sino-U.S. rap-
prochement in the 1970s, the formation of the Gulf Cooperation Council
and the U.S. tilt against Iran during the Iran-Iraq war, and the coalition
of Front-Line States against South Africa during the 1970s and 1980s.
31
This tendency should not surprise us; balancing is usually preferred for
the simple reason that no national leader can be completely sure what
others will do. Bandwagoning is dangerous because it increases the ca-
pabilities available to a threatening power and requires that the weaker
state trust in the continued benevolence of the stronger state.
32
In effect,
the state must subordinate itself to the will of another power (and at
worst completely lose its sovereignty), and most leaders, irrespective of
their relative power, regime type, or personal inclinations, will be loath
to embrace this possibility if other options are available. Because percep-
tions of intent are often unreliable and intentions can change overnight,
it is usually safer to balance against potential threats rather than take their
continued forbearance on faith.
implications for the united states
The ability of the United States to forge and manage future coalitions will
be primarily a function of the external security environment. States do
not join alliances or ght wars for altruistic or sentimental reasonssuch
30. See Randall K. Schweller, Bandwagoning for Prot: Bringing the Revisionist
State Back In, International Security, Vol. 19, No. 1 (1994).
31. For other examples, see Walt, Testing Theories of Alliance Formation.
32. For this reason, partners in an offensive coalition are prone to quarrel once victory
is achieved, often over the division of spoils. Examples include the victorious coalition
of Serbia, Bulgaria, Macedonia, and Greece after the First Balkan War, or the Nazi-
Soviet agreement to divide Poland in 1939.
chapter 7 203
vital decisions ordinarily reect careful considerations of self-interest.
Although shared values and similar political systems can facilitate align-
ment and smooth potential rifts, U.S. leaders should not expect assistance
from other states unless these governments believe it is in their interest
to help. If other states do not see themselves as threatened, it will be
extremely difcult for the United States to entice them into a military
coalition.
33
The U.S. ability to form and lead future coalitions will depend on its
own capacity to contribute to them. Although states generally prefer to
balance against threats, they will nd this option less appealing if they
believe that the effort is unlikely to succeed. Thus, even if the United
States comes to rely less on formal alliances such as NATO and more on
ad hoc coalitions, it must still possess sufcient military assets to make
itself an attractive partner. In short, coalition support can be an adjunct
to U.S. military power but not a substitute for it.
obstacles to effective alliance formation
When the balancing process works smoothly, and when states tendency
to form balancing coalitions is widely understood, aggression will be
discouraged because even highly ambitious leaders will expect opposi-
tion to any aggressive moves. For example, the rapid establishment of
the Western alliance in 194950, which made it clear to the Soviet Union
that any overt act of aggression would be met by a combined Western
response, enhanced the security of the West and helped preserve peace
between East and West for more than four decades. It is also likely that
Saddam Hussein would not have acted as he did in August 1991 had he
known that the seizure of Kuwait would generate a vast coalition of
powers seeking to expel his forces and disarm his regime.
By contrast, aggressors are harder to deter when balancing takes
place slowly or inefciently, and the probability of war will perforce
increase. For instance, although Britain and France did balance against
Nazi Germany in the 1930s, their efforts to do so were neither efcient
nor swift. Not only were Britain and France distracted and diverted by
threats in several separate theaters (which made it hard to focus their
attention on Germany), but Adolf Hitler was a master of dissimulation
who carefully cloaked the scope of his ambitions. As a result, his future
opponents did not recognize the true magnitude of the threat until very
33. Because clever aggressors try to conceal their aims until their preparations are
well advanced, status quo powers may want to coordinate possible responses in
advance, to facilitate the formation of a defensive coalition in the shortest possible
time.
204 the coming crisis
late. Efforts to form a countervailing coalition were also impaired by
ideological antipathies between Britain and France and the Soviet Union,
by rivalries among Germanys East European neighbors, and by the East
European states reluctance to ally with Stalin (whose embrace they
correctly judged to be as dangerous as Hitlers). For these and other
reasons, the effort to balance against Germany fell short of the level
necessary for successful deterrence, especially given Hitlers extraordi-
nary desire for war.
34
The efciency of the balancing process also plays an important role
in determining success in wartime. The ability to coordinate strategy and
maintain an adequate effort will maximize the likelihood of victory, while
disputes over wartime priorities and attempts to pass the costs of the war
onto ones allies will undermine any coalitions efforts and may lead to
an unnecessary defeat. In the wars of the French Revolution, for example,
the rst and second coalitions against revolutionary France failed in part
because the member states found it difcult to coordinate their military
activities and because they focused on maximizing their individual gains
instead of concentrating on the common aim of defeating France. By
contrast, although the victorious coalitions in World Wars I and II repeat-
edly quarreled over strategy and relative levels of effort, these disagree-
ments did not undermine their battleeld performance signicantly.
35
These examples reveal that one cannot assume that balancing coali-
tions will always form or will always operate well enough to succeed.
What are the main obstacles to the formation of effective and efcient
alliances?
conflicting interests. States are unlikely to form a coalition if their
interests are directly opposed, or if they interpret a particular strategic
situation in very different ways. Obviously, forming an effective coalition
will be impossible if one state regards another as especially dangerous
but other states do not share its view. For example, European allies of the
United States did not support its effort to overthrow the Sandinista
34. On these points, see Walt, Alliances, Threats and U.S. Grand Strategy.
35. On the tensions within the rst and second Coalitions, see John M. Sherwig,
Guineas and Gunpowder: British Foreign Aid in the Wars with France (Cambridge, Mass.:
Harvard University Press, 1969); Paul Schroeder, The Transformation of European Politics
(Oxford: Clarendon Press, 1994); and Steven Ross, European Diplomatic History, 1789
1815: France against Europe (Garden City, N.J.: Anchor Books, 1969), pp. 8792. On the
Grand Alliance in World War II, see William H. McNeill, America, Britain and Russia:
Their Cooperation and Conict, 19411946 (New York: Johnson Reprint Co. reprint of
1953 ed., 1970); and Robert Beitzell, The Uneasy Alliance: America, Britain and Russia,
19411943 (New York: Alfred A. Knopf, 1972).
chapter 7 205
regime in Nicaragua, largely because they did not regard the Sandinistas
as a serious threat. Alternatively, other states may recognize that a par-
ticular power is dangerous, but conclude that their own interests are not
affected. For example, U.S. efforts to isolate Iran during the 1980 hostage
crisis were weakened by the unwillingness of key U.S. allies to go beyond
token sanctions, largely because they did not regard the seizure of the
U.S. embassy as a threat to their own interests. More recently, U.S. efforts
to isolate Tehran and Baghdad have also been undermined by the fact
that medium powers such as Germany and France did not share the U.S.
appraisal of Iran and Iraq and were interested in pursuing commercial
opportunities with both states.
36
Lastly, states may refrain from joining a
coalition simply because they prefer to support the other side. During the
Cold War, for example, the United States and Soviet Union interpreted
most international events in a strictly zero-sum fashion. Because their
interests were in conict on so many issues, they rarely saw each other
as prospective coalition partners.
37
In general, the greater the divergence
in state interests among a set of potential allies, the more difcult it will
be to form a coalition among them.
competing prescriptions. Even when states agree that a particular
power is dangerous and that they share an interest in countering it, they
may disagree over how to respond. If a particular state appears to have
aggressive aims, for instance, some states may favor a strategy of accom-
modation while others advocate efforts to contain or overthrow it. For
example, French ofcials at the Paris Peace Conference in 1919 repeatedly
advocated joint Western intervention against the Bolshevik regime in
Russia, but were unable to persuade British and U.S. leaders that such a
step was necessary or advisable.
38
Similarly, although the United States
and the members of the so-called Contadora Group (Mexico, Venezuela,
Colombia, and Panama) generally agreed that the Sandinista regime in
Nicaragua posed a threat to regional stability, they disagreed completely
on both the magnitude of the threat and the best way to deal with it.
Similar dynamics can be found in the efforts in the 1990s to formulate
36. See Charles Lane, Germanys New Ostpolitik, Foreign Affairs, Vol. 74, No. 6
(1995).
37. One of the clearest signs of the Cold Wars passing was Russias willingness to
acquiesce in the United Nations action during the Gulf War, despite its long-standing
alignment with Iraq and its reservations about U.S. policy toward Baghdad.
38. See John Thompson, Russia, Bolshevism and the Versailles Peace (Princeton, N.J.:
Princeton University Press, 1966), chap. 6.
206 the coming crisis
a unied policy toward North Koreas nuclear weapons program or in
U.S. and European differences over the best approach to take toward the
civil war in Bosnia, and the problem has been especially stark in the
protracted debate over how to respond to Iraqs repeated efforts to evade
the UNSCOM inspections regime. Because many U.S. allies believed that
a punitive response (such as airstrikes) would be ineffective or counter-
productive, they were unwilling to support U.S. efforts to compel Iraq to
comply fully with the sanctions regime. In each case, the question is
less one of conicting interests than competing prescriptions; the poten-
tial partners generally want the same thing, but do not agree on how to
get it.
39
collective action. The problems just noted are compounded by the
familiar problem of collective action.
40
Even if all members of a coalition
agree on what the main threat is and on how they should respond, their
efforts to do so may be undermined by disagreements over who should
bear the burden. The danger, of course, is that each members efforts to
pass the buck to its partners may lead the coalition to do too little to
achieve its common objective, even when the coalition possesses greater
overall resources than its adversaries. This problem clearly aficted the
coalitions against revolutionary France and Napoleon, whose members
were constantly wrangling over subsidies, spoils, and strategy. Similar
problems weakened the effort to balance against Hitler during the
1930s.
41
The familiar debates over burdensharing in NATO are another
39. In the February 1998 confrontation between Washington and Baghdad, only Great
Britain and Kuwait supported the use of military force to compel Iraqi compliance
with the sanctions regime. Turkey refused to let the United States use its bases to
conduct air strikes, France and Russia openly opposed U.S. policy, and the Arab
League (including Egypt and Saudi Arabia) opposed any use of military force. The
general problem of competing prescriptions is discussed in Charles T. Allan, Ex-
tended Conventional Deterrence: In from the Cold and Out of the Nuclear Fire,
Washington Quarterly, Vol. 17, No. 3 (1994), pp. 222223.
40. See Mancur Olson, The Logic of Collective Action (Cambridge, Mass.: Harvard
University Press, 1965); and Mancur Olson and Richard Zeckhauser, An Economic
Theory of Alliances, Review of Economics and Statistics, Vol. 48, No. 3 (1966).
41. On the French case, see Schroeder, Transformation of European Politics, and Paul
Schroeder, The Collapse of the Second Coalition, Journal of Modern History, Vol. 59,
No. 2 (1987). On the 1930s, see Barry R. Posen, The Sources of Military Doctrine: France,
Britain and Germany between the World Wars (Ithaca, N.Y.: Cornell University Press,
1984); and Thomas J. Christensen and Jack Snyder, Chained Gangs and Passed Bucks:
Predicting Alliance Patterns in Multipolarity, International Organization, Vol. 44, No. 2
(1990).
chapter 7 207
example of this problem, and U.S. complaints about the distribution of
costs during the Gulf War eventually convinced Japan to contribute
roughly $12 billion to the allied war effort.
As the latter two examples indicate, the problem of collective action
is not insurmountable. But resolving issues of collective action are likely
to be more difcult when the members of a coalition face different
strategic problemssome risk annihilation while others do notand so
cannot adjust the burdens by the relatively simple expedient of transfer-
ring funds.
the problem of uncertainty. A nal obstacle to effective alliance for-
mation is uncertainty. Inefcient balancing is more likely when the threat
is ambiguous, since it is more difcult for states to agree on what the
main threat is and on how they should respond. After the Bolshevik
revolution, for example, British Prime Minister David Lloyd George ar-
gued against allied intervention in Russia by warning that the facts
required to make such a decision had never been ascertained and were
probably unascertainable.
42
As noted earlier, it was difcult to form an
effective coalition against Nazi Germany in part because the medium
powers of eastern Europe could not tell whether Nazi Germany or the
Soviet Union posed the greater danger, and Great Britain could not decide
whether the main danger to its interests was Japan in the Far East, Italy
in the Mediterranean, or Germany on the continent. More recently, a lack
of information has plagued efforts to respond to the North Korean nu-
clear program and to Iraqs efforts to evade the UNSCOM inspections
regime. Because potential coalition members disagreed about the circum-
stances in each case and could not obtain denitive information either
about current conditions or the likely consequences of alternative courses
of action, it was more difcult to assemble multilateral support. By
contrast, the German occupation of the rump of Czechoslovakia in March
1939 and Iraqs invasion of Kuwait in August 1991 provided unambigu-
ous evidence of their aggressive aims and facilitated the formation of
strong countervailing coalitions.
Uncertainty is a constant feature of world politics; fortunately, the
lack of perfect information is never an absolute barrier to an effective
42. Similarly, U.S. President Woodrow Wilson told a condant that his impressions
of Russia were based on indenite information, and his doubts helped convince him
that the Allies should leave it to the Russians to ght it out among themselves.
Quoted in Richard H. Ullman, Anglo-Soviet Relations, vol. 2: Britain and the Russian Civil
War (Princeton, N.J.: Princeton University Press, 1968), pp. 9697; and Frederick S.
Calhoun, Power and Principle: Armed Intervention in Wilsonian Foreign Policy (Kent, Ohio:
Kent State University Press, 1986), pp. 232, 238.
208 the coming crisis
alliance. Still, ambiguity about the nature or extent of the threat will make
it more difcult to bring a coalition into being and will fuel disputes
among the members, lowering its deterrent power and impairing its
wartime performance.
summary
While each of the obstacles identied above impedes the formation of an
effective alliance, each barrier is likely to shrink as the level of threat
increases. As rival states become more powerful or aggressive, the danger
they pose overrides other interests and encourages states that are other-
wise deeply suspicious to join forces, as the United States and the Soviet
Union did in World War II or as the United States and Syria did during
the Gulf War. Similarly, disagreements over how to respond decline as
the threat becomes manifest, if only because a strategy of appeasement
will be exposed as infeasible and rmer measures (including the use of
force) will come to be seen as necessary. The dilemmas of collective action
operate with less force when states are facing a truly grave threat, because
guaranteeing victory will become more important than trying to mini-
mize ones own contribution.
43
Finally, although ambitious aggressors
may be able to disguise their ambitions for some time, they cannot hide
them forever. Thus, uncertainty about the level of threat is likely to
diminish as the level of threat increases, thereby facilitating the formation
of a countervailing coalition.
Containing Rogue States: Implications and Obstacles
Do these same dynamics operate when states face a threat from a rogue
state armed with WMD? I answer this question below.
will other states be willing to balance?
Concerns about the possible emergence of rogue states armed with WMD
rest on the belief that other states will be reluctant to stand up to them
for fear of inviting a devastating response. Such fears are compounded
by the worry that great powers such as the United States will be more
reluctant to help out, thereby allowing rogue states to commit acts of
aggression or intimidation with impunity, and reinforcing the temptation
43. An important exception arises when the threat is clearly conned to only one
member of a coalition. Such circumstances tempt that states allies to remain aloof so
as to avoid the costs of ghting, possibly in the hope that a war will leave both
antagonists worse off. Obviously, such opportunistic behavior makes sense only when
one is condent that the aggressor will either be weakened by defeat or sated by its
victory.
chapter 7 209
to bandwagon. Such fears are almost certainly overstated. Rogue states
that acquire WMD are likely to face formidable balancing coalitions.
Although these countervailing alliances will have to overcome the obsta-
cles identied in the previous section, doing so should not be especially
difcult, for four reasons.
First, rogue states are invariably regarded as threatening even before
they acquire WMD, usually because they are believed to harbor highly
revisionist intentions; neighboring powers will already be on their guard.
As a result, it will require little or no effort for the United States to
convince them that the danger is real. The acquisition of WMD will
merely increase the rogue states capabilities (thereby increasing the level
of threat) and provide an additional incentive to balance. The rogues
acquisition of WMD is also likely to strengthen other states belief that
the rogue state holds aggressive ambitions, further reinforcing their de-
sire to join forces against it.
44
Second, neighboring powers have little incentive to cave in to a rogue
states threats, even when the enemy acquires WMD. Its threats will be
effective only if the victims are convinced that the rogue state is willing
to launch an unprovoked attack, and the historical record suggests that
this sort of blackmail rarely succeeds.
45
Acceding to such a threat would
invite additional demands, and in effect means relinquishing ones own
sovereignty and independence without a ght. Few ruling elites would
be willing to accept this sort of subordination, particularly when it would
leave them at the mercy of a regime that had shown a tendency to
brutalize its own citizens. Would anyone voluntarily accept Saddam
Husseins authority while other options were available, particularly if he
were trying to compel compliance by threatening annihilation?
46
44. These effects might not be as large as is sometimes supposed, however. Nuclear
weapons are not as usable as conventional capabilities, and would not be effective
instruments of conquest unless they were married to a large conventional military
capability. Accordingly, they will not increase a rogue states offensive power as much
as a comparable increase in conventional military capabilities would. Similarly, be-
cause rogue states are by denition regarded as having highly revisionist aims, the
discovery that they had obtained WMD would at most reinforce the existing belief
that they were aggressively inclined. The historical record supports this view: U.S.
relations with Great Britain, France, and Israel remained positive after these states
obtained nuclear weapons, just as U.S. relations with the Soviet Union and China were
hostile both before and after they joined the nuclear club.
45. See Richard K. Betts, Nuclear Blackmail and Nuclear Balance (Washington, D.C.:
Brookings Institution, 1987).
46. The precarious existence of the Iraqi Kurds provides a sobering reminder of what
subordination to Iraqi hegemony entails.
210 the coming crisis
Third, a rogue states threats to use WMD against its neighbors will
not be very credible. Those who fear that rogue states will use WMD
aggressively adopt a very pessimistic double standard. They assume that
outside powers like the United States will be deterred by a rogue state
with a handful of unreliable weapons of mass destruction. They also
assume that the rogue states leaders will be utterly unfazed by the vast
nuclear and conventional capabilities of the United States.
47
The credibil-
ity of a rogue states nuclear threats is further reduced by the possibility
that an attack on a neighbor would inundate its own territory with
damaging radioactive fallout, a problem heightened by the lack of a
long-range missile capability. Even if these states do obtain WMD, in
short, these weapons will be no more usable than the superpower arse-
nals were.
Finally, although neighboring states might be forced to bandwagon
if they were utterly isolated, the great powers (and especially the United
States) have an enduring interest in preventing other states (and espe-
cially relatively weak states) from using WMD for blackmail, coercion, or
as a shield for conventional aggression.
48
Nuclear weapons have been a
profoundly defensive force throughout the nuclear age, and this tendency
has been an important cause of peace between the major powers.
49
To
permit a rogue state to alter the status quo through the use of nuclear
blackmail would reverse this equation, and would send a powerful signal
that these weapons had become an effective instrument of expansion or
aggression. The consequences for world politics would be tremendous:
incentives to proliferate would grow apace, revisionist powers would be
quick to repeat their efforts to intimidate others, and war as a result of
miscalculation would become much more likely. Neither the United
States nor the other great powers has any interest in allowing such a
world to emerge, which means that they have ample reason to bolster
states facing threats from a well-armed rogue.
50
47. On this point see Waltz, Spread of Nuclear Weapons, p. 40.
48. This line of argument follows Barry Posens chapter in this volume.
49. On the defensive nature of nuclear weapons, see Robert Jervis, The Meaning of the
Nuclear Revolution: Statecraft and the Prospect of Armageddon (Ithaca, N.Y.: Cornell Uni-
versity Press, 1989) and The Illogic of U.S. Nuclear Strategy (Ithaca, N.Y.: Cornell
University Press, 1984); and Kenneth N. Waltz, Nuclear Myths and Political Realities,
American Political Science Review, Vol. 85, No. 4 (1991).
50. One could even argue that the United States has a general interest in punishing
the use of WMD for aggressive purposes, even if its own specic interests were not
affected.
chapter 7 211
Thus, states facing a rogue states blackmail should be able to obtain
U.S. backing.
51
And because helping to thwart nuclear blackmail is in the
larger U.S. interest, it should not be too difcult for U.S. leaders to
convince both potential allies and possible challengers that its pledges are
genuine.
Since nearby states have little reason to kowtow to a rogue states
blackmail, and little reason to fear that they will be utterly abandoned,
there is also little reason to fear that a rogues acquisition of WMD would
immediately undermine existing security commitments or trigger a cas-
cade of capitulations. Once again, the history of the nuclear age provides
reassuring evidence on this point. U.S. leaders once worried that Soviet
acquisition of nuclear weapons would allow the Kremlin to conduct
piecemeal aggression against others, counting on our unwillingness to
engage in atomic war, and cause our allies to lose their determina-
tion.
52
Similar fears accompanied the growth of Soviet nuclear and
conventional forces in the 1970s, which many experts believed cast doubt
on U.S. credibility and placed NATO in danger of collapse.
In fact, most non-nuclear states have been more than willing to
balance against hostile nuclear adversaries. For example, non-nuclear
powers such as West Germany, Belgium, Holland, Italy, and Japan joined
defensive alliances against the Soviet Union throughout the Cold War,
just as Cuba and Nicaragua aligned themselves with Moscow to gain
protection from the United States. Pakistan strengthened its alignment
with China following Indias peaceful nuclear explosion in 1974, and
Israels not-so-secret development of nuclear weapons did not deter a
number of Arab states from allying against it. Indeed, Syria and Egypt
launched a conventional attack on Israel in October 1973 despite their
awareness that Israel already possessed a number of nuclear bombs.
Syrian troops opposed the Israeli invasion of Lebanon in 1982 despite
their conventional inferiority and Israels nuclear arsenal, and Iraqs
chemical weapons capability did not deter the United States or its coali-
tion partners from ousting it from Kuwait in 1991.
53
Thus, the historical
record is admirably clear: an opponents possession of WMD does not
make other states unwilling to defend their own vital interests, or deter
them from using force when necessary.
51. It is worth noting that the United States has been willing to undertake a variety
of new military commitments in the late 1990sincluding Bosnia, Macedonia, and the
expansion of NATOeven though its own security was not directly at stake.
52. See U.S. Objectives for National Security (NSC-68), p. 414.
53. On the latter example, see Quester and Utgoff, No-First-Use and Proliferation,
pp. 109110.
212 the coming crisis
Two caveats should be noted at this point. First, the willing-
ness of non-nuclear states to balance against nuclear powers during the
Cold War may have been partly an artifact of bipolarity, insofar as
the U.S.-Soviet rivalry made each superpowers commitments to its allies
more credible. Because each superpower had an obvious interest in con-
taining expansion by the other, other states could be fairly condent
that their patrons nuclear umbrella would remain open. With the
passing of bipolarity, the credibility of existing commitments may wane,
and it may be somewhat harder for the United States (or other great
powers) to convince threatened states that it will be willing to help. But
as noted above, the United States has both a general interest in deterring
nuclear-backed aggression and specic interests in a number of regions.
Although it may take a bit more effort to reassure potential victims, both
their own desire to remain independent and the U.S. interest in prevent-
ing nuclear coercion should combine to provide a sufcient level of
credibility.
Second, the acquisition of WMD could enhance a rogue states ability
to deter other states from trying to conquer and overthrow it. Yet because
such a regime would not be as fearful of losing power should a war go
badly, its WMD might make it more adventurous than it would otherwise
be. However, the rogue state armed with WMD might actually be less
willing to use force to advance its foreign policy objectives, out of a fear
that inadvertent or deliberate escalation might trigger an overwhelming
nuclear response. Thus, the net effect of WMD on a rogue states behavior
is not obvious.
States are unlikely to view a rogue states acquisition of WMD with
equanimity, if only because they cannot be sure how these new capabili-
ties might be used in the future. States facing a threat from a well-armed
rogue will prefer to balance against them, and they should be able to
count on adequate allied support. What goals might such an alliance seek,
and how will the main obstacles to effective alliance formation affect its
cohesion and effectiveness?
objectives
Coalitions formed in response to a rogue states acquisition of WMD
could pursue several different objectives. First, other states might join
forces to persuade the rogue state to abandon its WMD ambitions by
threatening it with diplomatic isolation, economic sanctions, or military
encirclement. By increasing the political, economic, and strategic costs of
acquiring WMD, this strategy seeks to convince the rogue state that it
will be better off if it forgoes them. Such a policy could also provide a
lesson to other states that might consider obtaining their own WMD
chapter 7 213
arsenals.
54
The problem, of course, is that efforts to pressure a rogue state
are likely to reinforce its sense of insecurity and increase its desire for a
more powerful deterrent.
55
Second, states may ally against a rogue state to deter it from taking
aggressive actions in the future. Here the goal is one of containment;
instead of trying to halt or reverse the acquisition of a WMD arsenal,
threatened states would seek to persuade the rogue state not to use its
new capabilities to intimidate others or to gain tangible benets (such as
additional territory).
Third, if containment failed to work (or was not attempted), a defen-
sive coalition might try to reverse the outcome of the rogue states
aggression, as the United States and its allies did following the Iraqi
seizure of Kuwait. This objective is more demanding than containment
(particularly if achieving it requires direct attacks against the homeland
of a country that possesses WMD), but might be necessary either to
protect specic interests or to reinforce the norm that WMD cannot be
used as a shield for conventional aggression.
Finally, in extreme cases, two or more states could join forces to
conduct a preventive strike against the rogue states WMD arsenal. Such
a policy is likely to be adopted only when there is strong evidence that
the rogue state might actually use its new capabilities. As we shall see,
this goal would place the greatest strains on an alliances solidarity.
possible obstacles
Which obstacles to an effective alliance will operate most powerfully
when states seek to join forces against a rising rogue state, and which
objectives are likely to minimize these difculties?
conflicting interests. In general, the potential threat from a rogue
states WMD should reduce conicts of interest among its potential op-
ponents. However, the interests of allies are unlikely to be fully congru-
54. Most states will weigh the costs and risks of obtaining WMD before they begin
such a program, however, and only a regime that was highly motivated is likely to
make such a decision in the face of high costs and risks. Unfortunately, persuading a
highly motivated adversary to abandon this goal will not be easy. Interestingly, al-
though several states (e.g., Sweden, Brazil, Argentina, and South Africa) appear to
have abandoned their nuclear programs voluntarily, outside pressure seems to have
played little role in their decisions. See Mitchell Reiss, Bridled Ambition: Why Countries
Constrain Their Nuclear Capabilities (Washington, D.C.: Woodrow Wilson Center, 1995).
55. Pakistans experience supports this point, insofar as U.S. efforts to punish Paki-
stan by withholding conventional weaponry probably encouraged it to continue its
nuclear program.
214 the coming crisis
ent, if only because geography places potential partners in different
circumstances. States that are near the rogue state will have a greater
interest in containing its expansionist tendencies, but they will also be
more vulnerable to conventional or nuclear retaliation by the rogue
power and less willing to take direct action against it. States outside the
region will be less concerned by the spread of WMD (and thus less willing
to act to stop it), especially because none of the current so-called rogue
states is a global power or about to become one. Thus, anti-rogue alliances
will inevitably exhibit certain conicts of interest, even when all parties
are aware of the threat.
Alliances against a rogue state may also be impaired by other differ-
ences among the potential partners. Although the end of the Cold War
has made it easier to isolate contemporary rogue states, which can no
longer turn to the Soviet Union, other political divisions could undermine
efforts to contain or coerce these regimes. For example, support from
China helped North Korea resist U.S. efforts to pressure it into abandon-
ing its nuclear program, and U.S. efforts in the 1990s to contain Iran and
Iraq simultaneously have been undermined by its own reluctance to use
either one as a counterweight to the other.
56
Another potential conict of interest could arise over the goal of
nonproliferation itself. In the Far East, for example, efforts to reverse the
North Korean nuclear program were undermined by the ambiguous
motivations of South Korea and Japan. Not only do these states worry
about a possible military confrontation in the region, but both may be
ambivalent about the North Korean nuclear program itself. South Korean
strategists may hope to inherit the North Korean program after reunica-
tion, while some Japanese ofcials may regard the North Korean program
as a pretext that could eventually allow Tokyo to acquire a nuclear arsenal
of its own. Thus, although neither Japan nor South Korea is comfortable
with the idea of a nuclear North Korea, they have been reluctant to do
much to prevent it.
57
These conicting interests highlight the problem of credibility. Local
powers will prefer to balance against a rogue state, especially if they
believe that other states are willing to help. If they fear that they will be
left in the lurch, however, they may be more inclined to adopt a position
of neutrality. The question of credibility was a recurring (albeit exagger-
56. The United States did tilt toward Iraq during the 1980s, but has clearly been
unwilling to do so since the Gulf War.
57. Evidence of conicting Korean, Japanese, and U.S. interests can be found in Leon
V. Sigal, Disarming Strangers: Nuclear Diplomacy with North Korea (Princeton, N.J.:
Princeton University Press, 1998), especially pp. 9, 5859, 74.
chapter 7 215
ated) issue throughout the Cold War, but was mitigated by the United
States obvious interest in helping contain Soviet power. Because none of
the current rogue states poses a similar threat to the United States, it will
have more difculty convincing potential allies that Washington will still
fulll its commitments.
58
In short, although many states are likely to regard a rogue states
acquisition of WMD as a dangerous development, estimates of the seri-
ousness of the threat will vary greatly. A clever rogue government could
do much to exacerbate this problem, either by offering concessions to
mollify potential opponents, by masking its true ambitions, or by con-
cealing the full extent of its weapons programs. As the Iraqi case illus-
trates all too clearly, it is difcult to organize extensive counterprolifera-
tion activities unless or until a rogue state commits clear and
unmistakable acts of aggression. In 1991, Iraqs naked incorporation of
Kuwait made it relatively easy to assemble a large coalition for the
purpose of expelling an aggressor. In 199899, by contrast, Iraqs repeated
efforts to thwart the UNSCOM inspectors did not pose as grave a threat
and thus did not provoke a united allied response. Current and future
rogue states are likely to learn from Husseins unhappy experience; they
will avoid overly provocative actions until their WMD programs are
further advanced.
competing prescriptions. Even when several states agree that a par-
ticular rogue state is a serious threat, they may disagree on how the threat
should be met. If the rogue state has not yet obtained WMD, uncertainty
about its actual capabilities will discourage some states from taking
forceful action, based on the belief that the use of force may not be
necessary.
59
Once such a state does acquire WMD, the fear that preventive
action might not be completely successful could undermine allies cohe-
sion and render a forceful response impossible.
60
Here again, the confron-
tation between the United States and Iraq over the UNSCOM mission is
revealing. Because even the proponents of a military response questioned
58. See Marc Dean Millot, Facing the Emerging Reality of Regional Nuclear Adver-
saries, Washington Quarterly, Vol. 17, No. 3 (1994), p. 57.
59. See Michle A. Flournoy, Implications for U.S. Military Strategy, in Blackwill
and Carnesale, New Nuclear Nations, pp. 155156.
60. Israels preventive attack on the Iraqi nuclear facility is an exception that proves
the rule, as the operation was conducted unilaterally. By contrast, the international
community failed to respond to repeated reports about Iraqs renewed nuclear pro-
gram until Baghdad foolishly provided an opportunity by seizing Kuwait.
216 the coming crisis
whether airstrikes would eliminate Iraqs WMD capacity or persuade
Hussein to permit the UNSCOM inspectors to return, it proved impossi-
ble to assemble widespread international support for the use of force.
A policy of containment minimizes these problems. Although some
members of the opposing coalition may want to do more, any state that
regards the rogue state as dangerous is likely to agree on the need to
prevent it from expanding. States within the region have an obvious
interest in containing potential aggressors as they are the most likely
objects of attack, while the United States and some other great powers
have a general interest in deterring aggression, especially when it in-
volves the threat or use of WMD. By placing the onus for starting a war
on the rogue state itself, a strategy of containment also minimizes the
danger of escalation and the potential risk to each member of the coali-
tion, thereby reducing incentives to defect.
61
What if containment were to fail? Would an anti-rogue coalition be
able to summon the political will to reverse the results of aggression,
especially if the enemy already possessed WMD? The most likely answer
is yes. A deliberate act of aggression by a rogue state would remove
any lingering doubts about its bellicose aims and its willingness to take
risks. Such an act would strengthen both its neighbors desire to weaken
or overthrow it and the great powers interest in punishing a forceful
challenge to the status quo. Thus, it should not be too difcult to convince
a defensive coalition to force an aggressor to relinquish its gains, although
the risk of escalation would probably make some members reluctant to
overthrow the rogue government itself.
By the same logic, it will be very difcult for an anti-rogue coalition
to take preventive or preemptive action. As the recent confrontations with
North Korea and Iraq reveal, such a policy requires a consensus that the
rogue state is sufciently dangerous to justify starting a war, with all its
attendant risks. Not only are democratic states generally disinclined to
ght preventive wars, but allies are likely to face different levels of risk
and will probably disagree about the level of threat and the probability
of success.
62
For all these reasons, therefore, a strategy of containment
will generally be the preferred response.
61. There is another reason why states may favor different responses. If taking action
against a rogue state would require a large foreign military presence, and if such a
presence would have worrisome domestic repercussions, then local powers are more
likely to favor strategies of deterrence or accommodation rather than prevention.
62. See Randall K. Schweller, Domestic Structure and Preventive War: Are Democ-
racies More Pacic? World Politics, Vol. 44, No. 2 (1992).
chapter 7 217
collective action. Even if the United States and its allies agreed on
the need to confront a rogue state and were united on the strategy to
follow, their efforts to contain, compel, or disarm it might be undermined
by each states effort to pass the costs and burdens of action onto its allies.
These conicts cannot be met simply by negotiating more equitable
nancial arrangements (as the United States and its NATO allies were
able to do), because the costs of membership in an anti-rogue coalition
will be difcult to allocate evenly. Nearby states will face a greater risk
of retaliation than allies who are far away, while outside powers (and
especially the United States) will have to provide the bulk of the military
forces needed to deter or reverse aggression. When each members con-
tributions are difcult to compare, the danger grows that each member
will feel it is bearing an unfair burden. It will be difcult even to obtain
agreement for modest steps like economic sanctions, given most states
unwillingness to forego prots and market share simply because an
economic partner has chosen to acquire WMD.
63
Because a policy of containment will usually be the least expensive,
this approach will minimize quarrels over burden sharing. Neighboring
states will have to bear the social and economic costs of joint exercises or
the presence of foreign troops, and the United States will have to provide
the bulk of the military muscle, but these burdens should not be too
onerous for either side. Moreover, because a rogue states neighbors will
face a more direct threat than the United States, they will need its help
more than it needs them. As a result, the U.S. ability to insist on adequate
allied contributions should be substantial.
If the alliance adopts more ambitious objectives, however, the temp-
tation to free-ride will inevitably increase. All members might have an
interest in disarming a dangerous rogue, or in reversing a deliberate act
of aggression, but most members are likely to prefer that someone else
bear the costs and run the risks. This problem can be reduced by meas-
ures designed to protect the alliance members from retaliation (such as
theater defense systems), and by strategies designed to share the political
and military burdens. For example, the United States might be reluctant
to take preventive action unilaterally, for fear of acquiring a reputation
63. Responses to the Indian and Pakistani nuclear tests are instructive. Although the
United States and others have vowed to impose sanctions on both sides, the actual
penalties are likely to be modest. For example, although the Arms Control Export Act
of 1994 requires the U.S. government to cut off trade credits and loan guarantees to
states that conduct nuclear tests, the Clinton administration quickly requested an
exception for wheat sales by U.S. farmers. As Senator Slade Gordon warned, every
day that sanctions are in place well lose sales, and they are sales we cant recover.
See New York Times, June 15, 1998, p. A17.
218 the coming crisis
for bullying weaker states. But this political cost would be reduced if
other states were actively involved, even if their participation were
largely symbolic. Similarly, neighboring states might be reluctant to take
direct action if U.S. troops were not in the front lines, because the fear of
a massive U.S. response would probably discourage the opponent from
using its WMD in the course of a conventional war.
The bottom line is that the problem of free-riding will be most serious
when a coalition contemplates offensive action against a rogue state, and
will be lowest when implementing a strategy of containment.
the impact of uncertainty. Each of the obstacles just discussed would
be exacerbated by the inherent uncertainty involved in appraising the
actual threat that a rogue state represents. Because such regimes will go
to great lengths to conceal the extent of their achievements, it will be
difcult to estimate their capabilities. In particular, rogue states may
adopt a policy of deliberate ambiguity regarding their capabilities (some-
times referred to as a strategy of opaque proliferation) to reap the bene-
ts of a WMD deterrent while minimizing the political or military costs.
64
Even when other states are convinced that the regime has acquired WMD,
the rogue is likely to conceal the size, location, and capabilities of its
arsenal in order to complicate efforts to destroy it. And the less that
outsiders know, the harder it will be for them to agree on what to do.
Uncertainty about intentions will be equally troublesome, even when
the rogue state appears extremely bellicose. As noted earlier, the problem
is the inherent difculty of determining how the acquisition of WMD will
affect the new regimes propensity to use force. Not only would the
possession of a reliable deterrent reduce the insecurity that encourages
expansionist behavior, but achieving such a capability might invite pre-
emption or retaliation if war broke out. Thus, even seemingly irrational
or bellicose rogues will probably behave with great caution as their
capabilities increase, which means that efforts to contain them stand a
good chance of succeeding.
65
Overall, uncertainties about both capabilities and intentions will
make it difcult for a coalition to reach agreement on offensive action
against a proliferating rogue state. The intelligence requirements for a
military solution would be extremely demanding, given the need to
64. See Benjamin Frankel, ed., Opaque Nuclear Proliferation: Methodological and Policy
Implications (London: Frank Cass, 1991); and Thayer, Causes of Nuclear Proliferation,
pp. 508517.
65. This argument has been made by a number of scholars, most notably Kenneth
Waltz. See his contribution to Waltz and Sagan, Spread of Nuclear Weapons.
chapter 7 219
locate virtually all potential weapons sites and production and research
facilities and the relative ease with which these targets can be concealed.
66
As a result, coalition support for a strategy of preventive war is likely to
be forthcoming only when there is unambiguous evidence that a nascent
rogue was planning to attack. Unfortunately, history warns that this sort
of evidence will rarely, if ever, be available. Indeed, the Wests poor track
record in gauging the scope of North Koreas and Iraqs WMD programs
and its failure to anticipate Indias resumption of nuclear testing merely
underscore the inherent difculty of acquiring information that foreign
governments are eager to conceal. Given these uncertainties, it would be
folly to expect other states to agree to preempt a rogue state merely on
the basis of U.S. intelligence estimates.
67
These difculties will be compounded by the need for secrecy. A
preventive strike is more likely to succeed when it is conducted without
warning (as Israel did in 1981), but the advantage of surprise is lost if an
extended diplomatic campaign is necessary in order to round up allied
support.
68
And a clever rogue can exacerbate virtually all of these prob-
lems by masking its intentions, downplaying its radical beliefs, and re-
fraining from overt acts of aggression until its weapons are completed
and dispersed.
69
Taken together, these obstacles suggest that containment should be
the primary objective for any anti-rogue alliance, with reversing aggres-
sion as a secondary goal should events require it. Containment will reect
the members shared interests most accurately; is most likely to generate
consensus within the alliance and to minimize collective action problems;
and will require the least amount of reliable information. Reversing
conventional aggression will be more difcult but still feasibleassum-
ing that the act of aggression is unambiguousbut preventive military
action is unlikely to garner signicant allied backing.
66. It is worth noting that the U.S.-led coalition apparently destroyed no more than
a handful of Iraqs Scud missile launchers, despite an intensive air campaign against
them. See Thomas A. Keaney and Eliot A. Cohen, Gulf War Air Power Survey: Summary
Report (Washington, D.C.: U.S. GPO, 1993), pp. 7890.
67. On the failure to gauge Iraqs WMD programs, see Freedman and Karsh, Gulf
Conict, p. 320; and David Albright, Masters of Deception, Bulletin of the Atomic
Scientists, Vol. 54, No. 3 (1998).
68. See Millot, Facing the Emerging Reality, p. 48.
69. Once again, the Gulf war provides an apt contrast. The formation of the coalition
was facilitated by Iraqs overt aggression against Kuwait, which gave clear evidence
of Iraqs revisionist aims. Moreover, the coalitions war plans did not rely heavily on
secrecy regarding the timing of the attack or whether it would take place at all.
220 the coming crisis
Recommendations for U.S. Policy
The aquisition of WMD by any rogue state will highlight its threatening
nature and make its containment seem more necessary. At the same time,
it also makes steps beyond containment more dangerous. These insights
lead to two broad conclusions and a number of specic recommenda-
tions.
First, despite the various obstacles to coalition formation, it should
not be difcult to organize defensive coalitions against contemporary
rogue states, even if they were to obtain WMD. Indeed, containment is
by far the most feasible response to a rogue state that is in the process of
obtaining these capabilities. Containment places the onus for conict on
the rogue state, lowers the risk that the United States or its partners will
be dragged into a war that might otherwise be avoided, and minimizes
the various obstacles to an effective alliance. It also responds to the fear
that rogue states will act more aggressively once they obtain WMD, while
leaving the door open to improved relations should the rogue state feel
more secure once it acquires its own deterrent. And such a policy might
help deter the acquisition of WMD in the rst place, by forcing potential
proliferators to weigh the benets of possession against the political and
military costs of isolation.
Second, efforts to organize offensive coalitions to compel or disarm a
rogue state are unlikely to succeed. The U.S. efforts to pressure North
Korea and Iraq into abandoning their WMD ambitions are quite revealing
on this point, because widespread allied support for military action was
lacking in both cases. The obstacles identied above will be especially
severe should the United States seek allied support for preventive mili-
tary action; even when potential allies secretly favor such a step, the
temptation to free-ride will be especially high. Among other things, it
should not surprise us that there was strong allied support to reverse
Iraqi aggression in 1991, or that support for the UNSCOM regime has
deteriorated since then.
recommendations
In light of the various incentives and obstacles identied in this chapter,
U.S. efforts to organize military coalitions against a rogue state should
emphasize several elements.
reemphasize intelligence. Timely and accurate intelligence is a prereq-
uisite for forming an effective coalition; the United States cannot expect
others to help if they do not believe that a rogue state is dangerous.
chapter 7 221
Similarly, mounting a credible deterrent to aggression by a rogue state
armed with WMD will require accurate information about its present
capabilities and the likely objects of its revisionist aims.
70
In addition to
monitoring rogue states capabilities, therefore, U.S. intelligence efforts
should also focus on identifying the underlying sources of each rogues
ambitions, and their susceptibility to change. In particular, efforts to
ascertain how the acquisition of WMD is likely to affect the rogue states
behavior will be especially valuable as a safeguard against either com-
placency or paranoia.
share the costs and risks. Assuming that other states share similar
perceptions of the threat, the United States will have to nd ways to
demonstrate its own commitment and to share the costs and risks. Po-
tential allies need to be condent that they will not be left in the lurch
should deterrence fail. As discussed above, this step will not be as
difcult as many experts fear, in part because the United States has
specic interests in most regions of the world and also because the United
States has a general interest in reinforcing the defensive character of
WMD. Obviously, reiterating these interests will be an important element
of U.S. diplomacy toward potential coalition partners.
In addition, efforts to contain rogue states are likely to require some
tangible expressions of U.S. commitment, even if only of a symbolic
nature. Joint military exercises nicely reinforce both commitment and
capability, and can be bolstered by public declarations, naval port calls,
diplomatic visits, and in extreme cases, the permanent deployment of a
modest number of U.S. troops. By ensuring that the United States will be
involved should force be used, such measures diminish the odds that a
rogue state would take action in the rst place.
This sort of forward policy would also give the United States a
bargaining chip in its subsequent dealings with the rogue state; for
example, the United States could offer to reduce its military presence if
the rogue state abandoned its WMD programs. If a large U.S. presence is
infeasible, pre-positioned stocks and other preparations for rapid rein-
forcement can convey commitment and enhance military effectiveness.
71
Another possibility is the deployment of theater ballistic missile defense
systems, which can both protect U.S. military forces in the area and
70. See Henry Sokolski, Fighting Proliferation with Intelligence, Orbis, Vol. 38, No.
2 (1994); and Blackwill and Carter, The Role of Intelligence.
71. There is a potential tradeoff here; a large stockpile of pre-positioned equipment
might be an inviting target in the event of war. On the other hand, destroying that
stockpile would probably ensure that the United States would become involved.
222 the coming crisis
reassure potential victims who may fear a WMD attack, thereby reinforc-
ing containment in two mutually reinforcing ways.
72
Over time, public support for an active counterproliferation policy
will depend on the willingness of other states to contribute to the com-
mon goal. Most vital U.S. interests are only indirectly affected by contem-
porary rogue states (because none of them is in a position to attack the
United States directly), and U.S. taxpayers will not support a major effort
to contain rogue states if the states it seeks to protect are unwilling to aid
its efforts. As in any alliance, in short, success will depend on each
members ability to convince its partner or partners that they are not
being exploited.
identify allies carefully. The importance of maintaining credibility
and distributing costs suggests that the United States needs to think
clearly about which states actually merit a U.S. commitment and why.
Should the United States try to protect any state that is threatened by an
aggressive rogue state, or should it do so only when U.S. interests are
more directly engaged? If the latter, then the United States would commit
itself to aid Saudi Arabia against Iraq or Iran but might not commit itself
to oppose Libyan interference in Chad. The need to set priorities will
become more acute as U.S. military capabilities decline. The general sense
that rogue states are potentially dangerous does not mean that they are
equally threatening or that the United States must respond to each one.
73
dont overstate the threat. U.S. leaders should be careful not to
exaggerate the danger posed by rogue regimes. Ironically, efforts to limit
proliferation will be less effective if rogue states conclude that developing
WMD is the best way to get a lot of international attention or to get the
United States to focus on regional security issues. Potential allies might
conclude that these states are so powerful or dangerous that it would be
foolish to stand up to them. As I have already argued, todays rogue states
are rather weak and insecure regimes whose nuclear ambitions are based
primarily on a deep sense of insecurity.
74
There seems to be little reason
72. See Utgoff and Quester, No-First-Use and Proliferation, pp. 107108.
73. This criterion is complicated by the general U.S. interest in discouraging the use
of WMD for aggressive purposes, which might lead the United States to punish any
state that uses WMD to alter the international status quo even when the state in
question is not hostile to the United States itself. But this problem does not alter the
value of carefully identifying places where the United States should make a special
effort to contain rogue states.
74. The worst case would be if a rogue states sense of insecurity were so great that
chapter 7 223
to fear that these states would be suicidally bellicose even if they did
acquire such capabilities, which suggests that prompt efforts to contain
them are both necessary and likely to succeed. In seeking allies, therefore,
U.S. strategic planners should strive to portray the threat as accurately as
possible, without indulging in the hyperbole that all too often accompa-
nies efforts to rally public or international support.
75
combine deterrence and reassurance. The West won the Cold War
by containing the Soviet Union while avoiding all-out war, until a com-
bination of external pressures and internal developments brought the fall
of communism. Similar objectives should guide U.S. behavior toward
most rogue states: it should strive to contain them while emphasizing
that its aims are limited. No matter how objectionable such regimes may
be, it will rarely be in the U.S. interest to back them into a corner: States
that see others as irrevocably hostile are prone to extremely risky behav-
ior. Unless the United States is willing to take more forceful action,
containment should be tempered with contingent forbearance. In other
words, the United States should communicate that it will not try to
overthrow these regimes so long as they abstain from aggression; but will
act to oust them if they attack their neighbors, and especially if they use
WMD.
create firebreaks against the use of wmd. The United States should
consider how to erect rebreaks against the use of WMD in the event
that deterrence does break down.
76
One way is to reemphasize the taboo
against nuclear weaponry, making it clear that the United States and its
allies would regard the use of such weapons as a qualitatively different
act and would adjust their evaluations of the regime and their conduct
toward it accordingly.
77
Continued reductions in the U.S. arsenal will
reinforce this norm, and the ratication of the chemical and biological
it felt compelled to expand even after it acquired WMD, and attempted to use its new
capabilities to accomplish that objective.
75. However tempting it is to magnify external threats in a era of budgetary pressure,
the strategy can be counterproductive if it suggests that the problem is so grave as to
allow for no solution.
76. On the nature and construction of such thresholds, see Thomas C. Schelling,
Bargaining, Communication, and Limited War, and Nuclear Weapons and Limited
War, in The Strategy of Conict (Cambridge, Mass.: Harvard University Press, 1960).
77. For this reason, proposals that the United States respond to this threat by devel-
oping micro-nuclear capabilities would probably be counterproductive, as they
would merely serve to blur the distinction between conventional and nuclear weapons
and open the way for large-scale nuclear use by others. For an example of this sort of
224 the coming crisis
weapons conventions could begin to expand the norm to cover these
technologies as well. Finally, the United States should also remind the
international community that using a weapon of mass destruction against
U.S. territory, military forces, or allies will inevitably invite overwhelming
U.S. retaliation. Because the United States now possesses overwhelming
conventional superiority (and is likely to retain it for some time), it is in
its interest to keep WMD as unusable as possible.
maintain adequate u.s. defense capabilities. Any serious effort to
combat the spread of WMD and to contain rogue states will require the
United States to maintain a robust set of military capabilities. No state
can expect to wield international inuence without possessing substantial
power, and efforts to forge anti-rogue coalitions will be more difcult if
the United States lacks the strength that has made it a valuable ally in
the past. Although the programmed decline in U.S. military forces could
make the United States more dependent on its allies support, its ability
to obtain this support would evaporate if its capabilities erode too far. To
repeat: allied support can be an invaluable adjunct to U.S. power, but it
is not a substitute for it.
78
final thoughts
If the main arguments advanced here are correct, then the alleged danger
from nuclear rogues is overstated. Instead of enabling otherwise weak
and unpopular regimes to expand with impunitysecure in the belief
that neither their neighbors nor the great powers will dare resistthe
acquisition of WMD is more likely to spark the formation of countervail-
ing coalitions committed to containing or reversing any attempt by the
rogue to expand. Although these coalitions will face the usual obstacles
to effective balancing, these barriers can be overcome with fairly modest
efforts. In particular, costly programs to develop theater defenses or to
argument, see Thomas W. Dowler and Joseph S. Howard II, Stability in a Proliferated
World, Strategic Review, Vol. 23, No. 2 (1995).
78. Of course, one could argue that the absence of a major great power challenger
means that the United States has few if any overseas interests at present, and thus
little need either for large military forces or an extensive alliance network. By this
logic, states that are directly threatened should bear the costs of containing todays
rogue states, and the United States can abandon its efforts to promote international
order and concentrate on domestic issues. This position is defensible (although I do
not share it), but it requires a willingness to abandon any effort to shape the future
evolution of the international system. International inuence cannot be had on the
cheap; if the United States wants to wield substantial leverage, it will have to pay
the price.
chapter 7 225
acquire the capacity for a disarming conventional attack are probably
unnecessary and might even be counterproductive.
This conclusion does not imply that the United States can retreat to
isolationism, secure in the belief that other states will do all that is
needed. U.S. military power will be an important component in these
defensive coalitions, and will be especially necessary to deter the actual
use of WMD against states that lack their own deterrent.
Indeed, how the United States responds to the spread of WMD could
have far-reaching effects on the future of world politics. Although the
United States cannot prevent some states from acquiring WMD, it can do
much to affect the implications of this trend, and thus its pace and scope.
If the United States does not act to contain rogue states armed with WMD,
or if it allows them to use these weapons to coerce or to conquer other
states, then it will be helping reinforce the belief that these weapons are
a potent instrument for altering the international status quo. The demand
for WMD will grow apace, and the danger of war is likely to rise, at least
in the short run. By contrast, if the United States arrays itself against both
the spread of WMD and their use for coercion or conquest, it will help
discredit them as instruments of aggression and diminish incentives to
acquire them.
As the Cold War fades into history, therefore, the United States stands
at a crossroads. Down one path lies a world where WMD remain scarce
and unusable; down the other lies a world where such weapons are
plentiful, where they are seen as potent sources of political inuence, and
where they are somewhat more likely to be used. Because the latter world
is clearly less desirable, the United States should try to prevent it. Such
a strategy is likely to garner considerable allied backing. The more per-
tinent question is whether U.S. citizens are willing to support active
involvement in international affairs, even when it costs large sums of
money, and occasionally the lives of its citizens as well.
226 the coming crisis
Chapter 8
The Response to Renegade
Use of Weapons of Mass
Destruction
George H. Quester
How the United States responds to an adversary that used nuclear,
biological, or chemical weapons to attack U.S. forces at bases abroad, a
U.S. city, or the city of a major ally would play a critical role in determin-
ing how such weapons are seen afterward. The U.S. reaction to an attack
would inuence whether other rogue states would make attacks with
weapons of mass destruction (WMD). Indeed, if the United States lets
other states know what its response would be, the prospect of U.S.
retaliation could deter the use of WMD, if the response is of the proper
form and magnitude.
This chapter explores two models for how the United States could
respond to a renegades use of WMDa law enforcement model, as
illustrated by the Allies actions after World War II, and the Cold War
deterrence model. I argue that unconditional surrender is the appropriate
objective if a country such as Iraq or North Korea uses nuclear weapons.
In contrast, the Cold War theory of deterrence does not apply to a rogue
state that has used WMD. If a renegade state uses one or several such
weapons, it should not assume that the rest of its WMD arsenal can ward
off extreme punishment. The treatment that was applied to Hitlers re-
gime after it had killed millions of people is the appropriate treatment
for a rogue state that uses WMD today.
The next section examines the law enforcement model: for what
reasons does society imprison criminals? The second section examines
how well these motivations explain U.S. policies toward Germany and
Japan after World War II. The third section surveys the usefulness of Cold
Warera ideas about deterrence and the argument that no surrender is
truly unconditional. The conclusion offers policy recommendations.
From the Korean War to the response to the Iraqi invasion of Kuwait,
and as far back as U.S. persistence in the Pacic during World War II, the
United States has failed to advertise the vehemence with which it will
respond to foreign aggression. As doubts are expressed today about U.S.
commitments after the end of the Cold War, it would be wise for the
United States, and its adversaries, to take account of this pattern.
The Law Enforcement Model
Analogies with domestic law enforcement can help to illuminate U.S.
military choices at the international level, since, as Thomas Schelling
noted in the 1950s, a criminal justice system applies parallel strategic
approaches.
1
The law enforcement model is also valuable in that it high-
lights the role of justice and legitimacy; it will always be easier to deter
a potential renegade, and to enlist cooperation in punishing such a rene-
gade, when other countries share the U.S. sense of justice and outrage.
Why do we retaliate against criminalsmurderers or embezzlers
by placing them in prisons that are quite costly to the taxpayer? There
are at least four reasons, each of them relevant to how the United States
should respond to the use of WMD by a renegade state.
One reason for putting criminals into prisons is to disarm them, to
render them incapable of committing similar crimes again. Murderers are
a threat to citizens besides those they have already killed. At the mini-
mum murderers must be denied weapons; since kitchen knives and other
deadly weapons are so easy to come by, this may require locking such
people in prisons. (This reasoning applies much more to a murderer than
to an embezzler; once the identity of an embezzler is known, banks will
not hire him or her again, and the monetary assets of the rest of society
would be fairly safe, even if the individual were never to be imprisoned.)
Since so many tools, such as axes and kitchen knives, can also be
instruments of crime, we cannot ban all weapons; we must wait until
someone commits a crime. Once someone is identied as an axe-
murderer, he or she will have to be denied all use of axes or other
potential weapons. A similar approach may be useful with the dual-use
technologies of chemistry, biology, and nuclear physics, where what
serves civilian purposes also produces weapons. For example, Iraq cannot
be allowed the same facilities as states with less of a history of aggression.
A second reason to imprison criminals is to make an example of
the violator, thus deterring other potential law-breakers. Imprisonment is
a way to punish the criminal. (This, to use the jargon of military strategy,
is a countervalue rather than a counterforce argument for imprison-
1. Some interesting discussion of analogies between domestic justice systems and
international strategy can be found in Thomas Schelling, The Strategy of Conict (Cam-
bridge, Mass.: Harvard University Press, 1960).
228 the coming crisis
ment.) For this purpose, being in prison must be made a decidedly less
pleasant experience than being free.
There are limits to how painful U.S. citizens and other civilized
people want their prisons to be. The horror stories of southern state
prisons of the early twentieth century are not something we would want
to bring back, in part because this is too close to torture or cruel and
unusual punishment, and also because it may conict substantially with
the goal of reforming the criminal. Yet, the U.S. public also objects when-
ever penitentiaries come to be labeled country club prisons, because
the prospect of a penitentiary with color television sets and gymnastic
equipment and good food may not deter a potential law-breaker who is
living in an environment of urban poverty.
The major reason we imprison people who embezzle large amounts
of money may thus be to make an example of them, in the hope that
others will be deterred from stealing money if they see that this can lead
to the discomfort of imprisonment.
Just as we cannot let a law-breaker get away with a crime in
domestic society, lest everyone else feel free to commit the same offense,
countries must respond to violations of their rights in the international
arena; North Korea will be watching to see what Iraq can get away with,
and vice versa.
A third argument for imprisoning criminals is justice. Most people
grow up with an elementary sense of justice, by which wrongs should
be retaliated, simply to rectify the score. We are angry when our rights
are violated; we become calmer, we feel right about the world again, if
we are able to inict punishment on the violators.
Justice is an important word in most cultures, and enormous argu-
ments (and wars) can erupt about exactly what is just.
2
Indeed, some
of the worst conicts are those in which each side feels itself the victim
of an injustice inicted by the other, and where each thirsts for retribution.
It is very human, in such debates about justice, for each sides sense of
rightness to intensify. Thus, conicts where each side brings along its own
version of history, such as that between Croats and Serbs in the former
Yugoslavia, are exacerbated.
The goal of justice may seem closely related to the goal of setting
an example, but it is not identical. Justice is an end in itself, something
sought not because of any conscious calculation of consequences, such as
deterring a potential adversary from misbehavior. The relatives of a
murder victim typically feel relief when the murderer is convicted and
2. On the manifold disagreements of denitions of justice, see Carl J. Friedrich, Justice
(New York: Atherton Press, 1963).
chapter 8 229
given a stiff sentence, not so much because they are concerned about their
own immediate safety, or because they are concerned to deter other
murderers in the abstract, but because they wish for revenge.
The instinct for revenge is very human. As an assurance that the
guilty will be punished, the existence of such instincts are indeed valuable
for deterring crimes and aggressions. The common instinct for revenge
helps to make the prospect of retaliatory punishment credible.
But taking revenge may not always make sense after a crime has been
committed or after an aggression has occurred. Such an abstract psycho-
logical sense of retribution, however emotionally important, cannot be
the only consideration in domestic law enforcement decisions, or in the
international response to a WMD use. Feelings of justice are grounded in
deep emotions, which tend to confuse rational calculations. Justice may
come at too high a price in other, more practical, considerations.
A fourth purpose for prisons is to reform criminals, to change their
behavior, to teach them that they can support themselves without violat-
ing the laws. U.S. citizens worry that their prison systems are not very
successful at rehabilitating inmates, but many convicts do indeed leave
prison and never violate the law again. (Of course, the process of aging
reduces some of the energies required for criminal activity. If law-breakers
are held out of reach of society until they mature, they are less of a risk
thereafter. This rationale combines arguments one and four for imprison-
ment.)
Reform is certainly among the goals pursued internationally. The
United States is gratied by the change in Germany and Japan since 1945,
and hopes for a similar change as the former Soviet Union becomes a
democratically governed Russia.
In setting sentences and in designing prisons, it is often difcult to
determine which approachdisarming, reforming, or punishing the
criminal, or establishing a clear deterrent precedent for othersis the
most feasible.
In addition, each of the goals of imprisonment can conict with each
of the others. What exacts the greater punishment may not always
achieve the maximum of disarmament. (The 1943 bombing of Hamburg
provides a very clear example in the eld of military strategy, as more
German workers showed up at factories after the bombing, because the
delicatessens, etc., that had exempted them from war work had been
leveled.) For example, excessive punishment, or excessive security to
ensure that inmates cannot arm themselves, may get in the way of
reforming the criminals. And the abstract pursuit of justice or revenge
could come at too high a price for our more practical goals.
230 the coming crisis
The End of World War II
The United States has responded to international instances of mass
destruction in the past. In World War II, the Nazi regime in Germany
killed millions of people in concentration camps.
3
The Japanese army,
after the capture of the Chinese capital at Nanking in 1937, reportedly
killed several hundred thousands of people by bayonet.
4
The lethal gas
used in gas chambers and bayonets are not what we normally classify as
WMD, but such weapons facilitated destruction of the magnitude that we
might experience if rogue proliferators used WMD today.
How did the outside world respond to these acts of mass destruction?
What lessons can we learn from what happened to the Nazis or the
Japanese military leaders, and to their countries?
disarming the renegade
The Allies original intention after World War II was that the Axis powers
should be totally disarmed, that is, rendered incapable of ever again
militarily threatening and inicting mass destruction on their neighbors.
Japan was denied any armed forces in the constitution drafted for it by
the staff of U.S. General Douglas MacArthur, and the Allies agreed that
Germany was not to have any military forces.
For a time, such a total disarmament was indeed applied. The Japa-
nese and German military forces were disarmed and disbanded, and the
Axis states lost their sovereignty to the Allied occupying powers.
After some years, however, the exigencies of the Cold War led in
Japan to the emergence of three Self-Defense Forces, Ground, Air, and
Naval, in place of an Army, Air Force, and Navy.
5
In the German case, an
escalatory race emerged; the Western allies established a Federal Border
Police while the Soviets established a Peoples Police, each of which
looked remarkably military. They were followed by openly acknowl-
edged armed forces on each side.
6
3. On the Holocaust, see Raul Hilberg, The Destruction of the European Jews (New York:
Holmes and Meier, 1985).
4. See Edward Russell, The Knights of Bushido (London: Corgi, 1960) for an account
of Japanese behavior in Nanking and elsewhere in Asia.
5. The emergence of the Japanese Self-Defense Forces is discussed in Michael
Schaller, The American Occupation of Japan: The Origins of the Cold War in Asia (New
York: Oxford University Press, 1985).
6. The escalatory process of East and West German rearmament is discussed in Hans
Speier, German Rearmament and Atomic War (Evanston, Ill.: Row, Peterson, 1957).
chapter 8 231
If the victims of Axis aggression had expected that Germany and
Japan were to be totally and permanently disarmed, they would have
been disappointed. But from their very creation, the West German mili-
tary forces were tied closely to NATO, and the Germans were not to have
their own general staff. And the constraints on the Japanese Self-Defense
Force also amount to a compromise of sovereignty greater than most
nations would accept.
In one most important aspect the Axis powers have not been allowed
to rearm at all: with weapons of mass destruction.
If one asks the basic question today about the Nuclear Non-Prolifera-
tion Treaty (NPT)why some countries are allowed to possess nuclear
weapons and others are notthe answer is often evasive, amid assur-
ances that a total nuclear disarmament of all the powers of the world is
the goal. The more honest answer might be that the non-use of nuclear
weapons may require that there be at least two nuclear forces deterring
each other, given the inability of the world to forget how to produce such
weapons, and given the inherent difculties of the safeguards required
to assure each country that no other country is cheating.
The ve states that had weapons when the NPT was drafted may be
more than are necessary for mutual deterrence, for mutual assured
destruction, but this is where, by accident of history, the nonproliferation
barrier was drawn.
To justify this barrier further, one sometimes hears references to the
political instability of third world countries, of countries such as Libya or
Iraq; but note is then taken that such instability has appeared also in
Beijing and Moscow, while it has not in Sweden or Australia. It is dan-
gerous to base the nonproliferation argument on the premise that Mos-
cow and Washington are more qualied to handle nuclear weapons,
while New Delhi and Islamabad are not, for this will simply unite the
South Asians in indignation at the slur on their character.
7
Even fty years after World War II, many still feel that Germany, Italy,
and Japan, the parties guilty for the barbarism that was at the heart of
World War II, are particularly disqualied to handle nuclear, biological,
or chemical weapons.
The Federal Republic of Germany was required to renounce nuclear,
biological, and chemical weapons by the Western European Union Treaty
that legally sanctioned its rearmament in 1955. It grudgingly signed the
7. For this authors more extensive view of South Asian attitudes on nuclear prolif-
eration, see George H. Quester, Nuclear Pakistan and Nuclear India: Stable Deterrent or
Proliferation Challenge? (Carlisle, Penn.: Army War College Strategic Studies Institute,
1992).
232 the coming crisis
NPT at the end of the 1960s, renewing this commitment to forego nuclear
weapons. Japan has maintained a nuclear taboo, even while its Self-
Defense Forces had become a substantial conventional army, navy, and
air force by the end of the 1980s.
punishing the renegade
In addition to disarming the Axis powers, in 1945 the Allies intended to
impose some punishment. In practice, there was indeed an important
setting of an example here, so that anyone contemplating emulating
Hitler or Tojo would have to think twice.
One cannot forget, of course, that nuclear destruction was imposed
on Japan in 1945. It is more often forgotten that the physicists and
engineers who worked so feverishly to produce nuclear weapons in the
Manhattan Project visualized them as being readied for use against Ger-
many, more than against Japan.
8
The focus on Nazi Germany stemmed, of course, from fears that the
Germans might be racing to acquire nuclear weapons themselves. The
bomb might thus be used to defeat Germany before Hitler could use such
weapons against the West. But another reason for the focus was that so
many of the nuclear scientists were direct or indirect victims of Nazi
policies, who had been driven to ee Germany or Europe, guessing or
knowing that their relatives were being murdered in concentration
camps.
9
If the atomic bomb had been ready in time, it almost surely
would have been dropped on German cities before Japanese cities. In-
stead, German cities were exposed to deadly conventional bombardment,
with the casualties in Hamburg and Dresden reaching into the hundreds
of thousands.
Of course, these attacks were primarily intended to force the Ger-
mans and the Japanese to surrender, rather than to set an example for
other aggressors in the future. That is, it was an attempt at intrawar
compellence of the Axis, rather than future deterrence of others. Germany
did not surrender because of the punishment of World War II aerial
bombardment, while Japan most probably did.
Still, some of the motivation for punishing the Axis populations went
beyond speeding the Allied victory, and this motivation persisted after
the surrenders. The Soviets surely punished the Germans (and even the
Japanese, with whom they were at war for a mere three days) through
8. For a contrary interpretation, see Arjun Makijani, Japan Always the Target,
Bulletin of the Atomic Scientists, Vol. 51, No. 3 (May/June, 1995), pp. 2327.
9. On the motives driving the Manhattan Project, see Thomas Powers, Heisenbergs
War (New York: Knopf, 1992).
chapter 8 233
their treatment of prisoners of wara great number did not survive to
return to their homes, and the rest were held in forced labor for many
years.
10
Some of the U.S. plans for the occupation of Germany after the
war, for example the Morgenthau Plan, likewise amounted to punishment
or revenge, albeit these were also labeled as disarming the Germans by
eliminating their heavy industry.
11
Policies of nonfraternization and the
maintenance of austere food rations could similarly be justied less by
the logic of disarmament and more by that of punishment. One did not
have to execute leaders like Goering or Tojo to render them incapable of
launching wars again; rather, their execution achieved a feeling that
justice was done and a good precedent had been set. The war crimes
trials in Nuremburg and in Tokyo could be advertised, like any other
judicial process, as the setting of an enduring legal example; critics of
their legality claimed that these processes simply allowed the victors to
take revenge.
12
Thus there was some punishment of the Germans and Japanese,
whether for the sake of justice or for setting an example, but much less
than these defeated peoples might have had reason to expect, especially
given how their own armies and governments behaved. Both of the
defeated Axis powers were on short rations immediately after the war,
but so were the rest of Europe and East Asia. It took time to overcome
the economic disruption that the war inicted.
The punishments were surely greater while these countries were still
holding out against surrender, and less very soon thereafter. This suggests
that the punishments were being inicted on the Axis states because they
were still engaging in behavior threatening to the outside world. Simi-
larly, the punishments a judge might impose on someone who is refusing
to testify when he has been given immunity are often much more severe
and open-ended than the punishments imposed on a robber who has
been arrested and disarmed, and so can no longer engage in robbery.
But the reduction in the level of punishment after 1945 also shows
how the desire for revenge can cool, at least in a democracy, with its
humane instincts. For example, in a public opinion poll taken among U.S.
citizens in November 1945, immediately after the atomic bombings and
Japans surrender, some 22 percent of the respondents regretted that the
10. Soviet treatment of prisoners of war is detailed in Helmut Fehling, One Great
Prison (Boston: Beacon Press, 1951).
11. On the Morgenthau Plan, see Warren Kimball, Swords or Ploughshares? (Philadel-
phia: Lippincott, 1976).
12. The Nuremberg trials are discussed in Telford Taylor, The Anatomy of the Nurem-
berg Trials (New York: Knopf, 1992).
234 the coming crisis
United States had not inicted more nuclear destruction on the Japa-
nese.
13
Most U.S. citizens would today be shocked by this 1945 expression
of opinion.
Passions also cooled after 1946 because of the new drives of Cold War
competition between Moscow and the West. But, in the U.S. and British
zones of Germany, and in Japan, occupied entirely by the United States,
the occupying forces by and large did not have their heart in being
punitive, even if it still seemed quite plausible that the German and
Japanese crimes should be blamed not only on the leaderships, but on
the peoples as a whole.
Indeed, within two years the United States was donating substantial
economic assistance to its former enemies through the Marshall Plan and
related aid efforts.
14
Perhaps this was because the economic situation in
the former war zones had deteriorated to the extent that no additional
punishment was necessary, but the aid also demonstrated an unusual
generosity, where the winner in a war delivers gifts to the loser.
The Allied bombings of Germany substantially exceeded the volume
of the German bombings of Britain, and hence did more than enough to
achieve justice for Coventry and London.
15
But they were not enough, by
any eye for an eye, tooth for a tooth reasoning, to make up for the
concentration camps and the Holocaust and the totality of the Nazi
occupation of Europe, and all the deaths and destruction suffered in
ordinary combat.
It is hardly surprising that most Jewish people in the United States
are less inclined to turn the page on German history than are other
citizens, or that the Chinese and Koreans, and indeed almost all East
Asians, are much more bitter than are the average U.S. citizens toward
Japan.
The more direct victims of wartime atrocities might thus feel bitter
that less than justice was achieved. Perhaps too many war criminals
received light sentences or were not punished at all.
Perhaps, where mass destruction is involved, it may be difcult
to inict commensurate revenge, unless it is inicted very rapidly and
all at once. A civilized people will not stomach the task of being
13. Poll of November 30, 1945, cited in Public Opinion Quarterly, Vol. 9, No. 4 (Winter,
19451946), p. 530.
14. On the Marshall Plan, see Armand Clesse and Archie Epps, eds., Present at the
Creation (New York: Harper and Row, 1990).
15. The magnitudes of the German and British-U.S. bombing campaigns are com-
pared in Charles Webster and Noble Frankland, The Strategic Air Offensive Against
Germany (London: Her Majestys Stationery Ofce, 1961).
chapter 8 235
cruel in retaliation, once the transgressor has ceased to defy the norms
of the worldonce the immediate compellence task has been accom-
plished.
Where it is necessary, for longer-term considerations, to teach a
lesson to deter others, it will not be easy to persist in this retaliation.
When the enemy is no longer deantly engaged in aggression or atroci-
ties, when it is no longer challenging the democratic powers in a contest
of wills, the revenge motive is reduced, and the punishments soon cease.
Many analysts argue that there should be no nuclear retaliation for a
chemical or biological attack on a U.S. military base or U.S. city, because
the robust array of U.S. conventional weapons will allow for more than
sufcient retaliation. (And a few analysts, with logical consistency, also
suggest that there should be no U.S. nuclear response even if a country
such as Iran uses a nuclear weapon against the United States.) However,
if the United States sees its task as making an example of the transgressor,
so as to deter others, conventional retaliation could take too long, and
might not be pursued for long enough to impose a commensurate pun-
ishment. To respond tellingly to a use of weapons of mass destruction, to
really set an example, it may be necessary to respond with WMD.
The United States may thus have to retain an array of nuclear weap-
ons for a variety of reasons. First, such weapons will be required to
respond to another powers use of nuclear weapons. Second, some tacti-
cal situations might arise where conventional defenses could not protect
an ally; rather than have the ally obtain its own nuclear weapons, there
may be a need to maintain U.S. extended deterrence. Third, the threat of
a U.S. nuclear response may be needed to deter some renegade state from
using chemical or biological WMD.
16
reforming the renegade
After 1945, the Allies emphasis shifted to reforming Germany and Japan,
instead of very severely punishing the Axis countries, or totally and
permanently disarming them.
In 1919, the French government under Georges Clemenceau shared
none of U.S. President Woodrow Wilsons enthusiasm and optimism for
achieving democracy in Germany, fearing that a democratically governed
German neighbor would be just as hostile as one governed by the Kaiser.
With the Treaty of Versailles, France was intent entirely on disarming
16. For an analysis urging U.S. nuclear retaliation threats against another states
chemical or biological attacks, but not against a purely conventional attack, see David
Gompert, Kenneth Watman, and Dean Wilkening, Nuclear First Use Revisited,
Survival, Vol. 37, No. 3 (Autumn 1995), pp. 726.
236 the coming crisis
Germany, and on punishing the Germans to set a deterrent example for
the future.
But in the total defeat and unconditional surrender imposed on the
Axis powers in 1945, the Germans and Japanese did not just have their
governments defeated and removed; their political cultures were success-
fully reformed.
17
The decades since 1945 might suggest that it is foolish
to pursue revenge or justice (as in Germany after World War I), but that
it is possible to inculcate democracy, and that democracies are more
peaceful and internationally cooperative than are other forms of govern-
ment.
Of course, some analysts still voice doubts about whether Germany
and Italy and Japan have been reformed. They note that Italy in 1994
elected a governing coalition that included neo-Fascists.
18
Germany ex-
periences neo-Nazi skinhead attacks on resident foreigners. The same
kinds of marginal anti-social behavior can be found in places like France
and Sweden, perhaps with no less frequency on a per capita basis, but it
understandably attracts more notice in Germany.
19
Scholars debate whether Japans openness and liberalization and
democracy are supercial.
20
For example, it is disturbing that the Japa-
nese school system does far less than the German to address the basic
guilt of the Axis powers for the outbreak of World War II.
21
Yet, the
democratization of these defeated Axis powers now looks like a general
success, far exceeding what most U.S. citizens and others might have
dared hope for during World War II.
Prisons often fail to reform convicts, which is an argument for in-
creasing the punishments to deter crime, or for disarming such criminals
by locking them up, rather than pursuing a reform-school mentality.
By contrast, the handling of the defeated Axis powers after World War II
suggests that the reform of governments and their foreign policies is
attainable. Thus, reform efforts could be as important as setting an
example or disarming a defeated enemy. The possibility of reform
should not be forgotten in the passion for justice or revenge.
The United States has felt generally successful in its response to the
mass destruction perpetrated by the Axis powers in World War II. But
17. See John D. Montgomery, Forced to be Free (Chicago: University of Chicago Press,
1957).
18. See New York Times, April 16, 1994, p. 1.
19. See, for example, New York Times, October 14, 1994, p. A-12.
20. See Karel Van Wolferen, The Enigma of Japanese Power (New York: Knopf, 1989).
21. See Saburo Ienaga, The Glorication of War In Japanese Education, International
Security, Vol. 18, No. 3 (Winter 1993/94), pp. 113133.
chapter 8 237
the modes of achieving the success were not what would have been
predicted in 1943.
As with law enforcement, it is not easy to predict which of the four
basic kinds of response to a transgression will offer success. And difcult
choices must be confronted, as one approach will conict with another.
The Cold War Deterrence Model
The situation that the United States faced in its Cold War standoff with
the Soviet Union is very different from the situation it would face if a
renegade were to use WMD, in two important ways.
First, had the Soviet Union made an initial use of weapons of mass
destruction, the damage could have been innitely more severe than
what any renegade state can launch in the near future. A Soviet missile
and bomber surprise attack might have killed most of the U.S. popula-
tion.
It thus was not really clear why the United States should have retali-
ated to set an example here (for it would be difcult, after this disaster,
to imagine any comparable situation emerging). The holocaust that Mos-
cow launched would have brought the United States to the point where
many analysts predicted that the survivors would envy the dead, and
retaliation against the cities and civilian population of the Soviet Union
would mainly have increased the aggregate calamity suffered by all when
deterrence had failed.
22
But everyone knew that the United States would of course retali-
ate for such an attack, and the leaders in Moscow surely knew this. The
iniction of 100 million casualties on the peoples of the Soviet Union
would have achieved the satisfaction of revenge, matching the 100 million
Americans killed in this nuclear Pearl Harbor.
Here is one place where the distinction between seeking justice and
revenge for its own sake and seeking to set an example is very impor-
tant, since there might not have been enough relevant political leaders
around to remember the example. The United States would indeed
have retaliated to achieve some sense of justice, even if there were less
other reason to do so. Everyone knew that there would be retaliation,
and, for this reason if for no other, no such sneak attack ever came.
By contrast, if the rogue proliferators we are addressing here were to
utilize WMD, the numbers of Americans killed would be far smaller. In
addition to containing our worst-case fears, this also resurrects the more
22. For a discussion of why retaliation would allegedly make no sense here, see
Philip Green, Deadly Logic (Columbus: Ohio State University Press, 1966).
238 the coming crisis
practical motivation for a retaliatory response. In addition to the simple
satisfaction of revenge and justice, we would again be setting an exam-
ple for the deterrence of similar acts in the future.
But there is a second, even more important distinction with the years
of the Cold War. What if the Soviet Union had decided to wage a limited
nuclear war, detonating perhaps only three or four nuclear warheads,
while holding thousands in reserve? It would have been very difcult for
the United States and the rest of the world to respond by pursuing the
total elimination of Soviet weapons of mass destruction, or the complete
defeat and reform of the Soviet Union.
There surely would have had to be some punitive U.S. response
against such a Soviet limited nuclear attack. Moscow could not have been
allowed to get away with such a nuclear escalation, lest it feel free to do
the same whenever it felt so inclined in the future. Yet the response would
have had to be measured, in accord with the dictates of graduated
deterrence and intrawar deterrence. It was generally assumed that the
United States could never put its adversary into a position of using or
losing its strategic nuclear forces, the forces that could wipe out the
United States.
In contrast, a confrontation with a renegade user of WMD in the next
decades will not threaten the national existence of the United States. As
in World War II and Desert Storm, there will be more immediate risks for
U.S. allies; but, in confronting a seeming lunatic, a state that is already
inclined to inict mass destruction on the world, the main priority will
be to disarm that renegade, and disarm it totally.
the end of cold war intrawar deterrence
It is easiest to retaliate for a renegades use of WMD if it has nothing left
in its WMD arsenal. The choices are more difcult when the renegade
retains even a few additional warheads, and can inict additional de-
struction.
While the renegades the United States may face will possess some
nuclear warheads or other weapons of mass destruction, they will not
have thousands nor even hundreds for a long time to come, and a
meaningful counterproliferation program will reduce the destructive po-
tential any such proliferator can deploy. Because they can brandish so
much less destructive potential than the former Soviet Union, it would
be a poor prescription for policy to project the constraints of the Cold
War.
World War II may be much more relevant. What if Hitler or the
Japanese had acquired some nuclear warheads at about the same time as
the United States, and had used one or two against London or
chapter 8 239
Chungking? Would the United States have settled for a containment of
the adversarys forces in such a case, merely pushing the Nazis inside
Germany, and the Japanese Army to its home islands? Or would it have
demanded unconditional surrender, as it did? By the Cold War conven-
tional wisdom of limited war, there would have been denite constraints
on how the United States could respond to any use of nuclear weapons
of mass destruction when the opponent retained any reserve of atomic
bombs.
However, I argue that there would be much less restraint on how far
the United States felt able to go. A grim analogy is the U.S. and British
responses to the early reports of the Nazi death camps.
23
The idea of
negotiating with the Nazis to end these operations was rejected out of
hand, on the argument that the only acceptable objective was the total
defeat of Hitlers regime. Yet millions of people perished as the Nazis
continued their program of mass destruction. And a great number of
ordinary West and East Europeans perished in the Allied military offen-
sives directed to their liberation. The decision to press for a total defeat
of the Axis had taken these losses into account. Thus, whenever an
adversary has demonstrated a willingness to inict horrendous aggres-
sive acts, the threats of what else the rogue adversary can inict until it is
defeated should not be allowed to lead to intrawar deterrence.
The motivation might be simple anger and a thirst for revenge. Or it
might be more practical; a Hitler or a Saddam Hussein who had dropped
a nuclear bomb could never be trusted not to return to mass destruction,
so that disarming the rogue-state would be more relevant than revenge.
In any event, the world simply could not tolerate the continuation in
power of leaders who had killed so many people using weapons of mass
destruction.
If the Axis powers had mounted a nuclear attack, the United States
and its allies would have had to push on to Berlin and Tokyo in 1945 so
as to decisively eliminate their regimes, even as the few Axis nuclear
warheads were still being red back in retaliation.
unconditional surrender
The issue of constraints during wartime relates closely to the concept of
unconditional surrender, an idea that requires our scrutiny.
The Allied demand for unconditional surrender was used by the Nazi
and Japanese leaderships to reinforce the ghting will of their forces, on
23. The controversial Allied policies toward the Holocaust are given critical attention
in Arthur D. Morse, While Six Million Died (New York: Random House, 1968).
240 the coming crisis
the basic argument that surrender to the Allies now offered no limits at
all to the punishments to be inicted.
In 1958, Paul Kecskemeti argued that unconditional surrender is
always a contradiction in terms.
24
If an adversary is to choose to cease
re, it must be promised something in exchange for this, at the least a
guarantee that one will also cease shooting, and that the defeated popu-
lation will not be massacred. Kecskemetis study detailed how the Italian,
German, and Japanese surrenders in World War II, along with the French
1940 surrender to the Germans, all involved some important promises by
the winner to the loser.
Kecskemeti and many others drew the logical deduction that the
additional retaliation a defeated adversary could still inict precludes all
thought of unconditional surrender. Kecskemetis argument was received
as plausible, in large part because it seemed to t the mutual assured
destruction situation of the Cold War. Since the Soviet nuclear arsenal
could destroy the national existence of the United States, and vice versa,
neither power could dare to push the other to total defeat, or to ask for
unconditional surrender.
But the resources of a renegade like Saddam Hussein would be far
smaller than what the Soviets possessed in 1958. The problem with
Saddam Hussein will less resemble the Cold War and more the problem
with Hitler in World War II, and the problem in law enforcement.
Renegade Use of WMD: The Unconditional Surrender Formula
What should the United States do if weapons of mass destruction are
used by Iraq, North Korea, or a similar renegade state in the future?
Two important inferences can be drawn from this chapter. First,
rather than being an automatic and unavoidable consequence, the con-
straining fears of intrawar deterrence and limited war may not take
effect if a renegade actually uses WMD and still retains more; it may
instead seem too urgent to wrest all these WMD away from the renegade.
Second, rather than being an unreasonable posture and a logical
oxymoron, unconditional surrender should be the demand of the civ-
ilized world after a renegades acts of mass destruction. A use of WMD
could be so outrageous, so much beyond the normal standards of gov-
ernments around the world, that the offending government would have
to be replaced from the top down.
24. Paul Kecskemeti, Strategic Surrender (Stanford, Calif.: Stanford University Press,
1958).
chapter 8 241
If Saddam Hussein had clearly used chemical or biological weapons
against UN forces in Desert Storm, the case for advancing all the way to
Baghdad might have become overwhelming. And the same logic may
apply in the future, if the regimes in Tehran or Pyongyang should use
such weapons, or if Iraq should somehow reacquire, and then use, such
WMD.
Such a pursuit of unconditional surrender would be driven by one
or all three of the major motives for response noted above.
First, perhaps most urgently, it would seem necessary to render the
regime incapable of repeating its acts of mass destruction. Military con-
quest as far as the capital may not be the only way to accomplish this; the
hope of President George Bush and the rest of the UN coalition in 1991
was that the peace terms imposed on Saddam Hussein at the end of
Desert Storm would sufce to deny Iraq any WMD. But a total occupation
of the guilty country, and the supplanting of its government, might seem
the best way of achieving disarmament.
The most appropriate response to any renegade use of WMD might
thus simply be a maximum of disarmament. The United Nations Security
Council resolutions on Iraq (when Saddam Hussein had of course not yet
used his WMD, but had engaged in blatant conventional aggression)
established a useful legal precedent.
There would be a need to strive for an even more rigorous version
of the nuclear disarmament imposed on Iraq, going beyond normal
International Atomic Energy Agency (IAEA) procedures under the NPT,
and for a removal of most or all of the dual-use nuclear and biological
and chemical infrastructure that lends itself to weapons of mass destruc-
tion. Given the inherent problems posed by such dual-use technologies,
an unconditional surrender might be required to put them all under
effective control.
The second major focus of a policy of unconditional surrender, after
a renegade states use of WMD, would be as important: to show any
similar state around the world that using such weapons leads to punish-
ments that outweigh the gains.
In addition to striving for the same disarmament imposed on Ger-
many and Japan, it will be important to set a precedent. For example,
punishing the leaders in a war crimes trial would make a great deal of
sense, not so much to reform the regime, which may remain difcult, but
to set an example that would be watched by others.
The United States might also consider counterpower nuclear tar-
getingan idea that emerged during the Reagan administration
about how best to deter the Soviet leadership. It was argued that
the proper targeting objectiveone that would truly punish the Kremlin
242 the coming crisis
leaders, and more importantly deter them from aggression in the rst
placewas not to kill millions of Soviet citizens, but rather to undermine
the Communist Partys control over the Soviet Union.
25
It is reasonable to assume that Kim Jong-Il or Saddam Hussein or
the ayatollahs in Iran also care about power; the best way to punish
any use of WMD by such leaders might be to eliminate their regimes.
For North Korea, this would mean merging it with South Korea (some-
thing long overdue on the merits of the wishes of the Korean people,
in any event). For Iraq, this would mean breaking it into pieces, as
Kurds in the north and Shiites in the south would be allowed to secede,
and in Iran it would perhaps mean imposing a secular and democratic
government. All of this might be undertaken not so much to pro-
duce peace-loving governments, but to confront other potential WMD-
wielding renegades with the punishment they fear the most, losing
power.
The third possible goal of demanding unconditional surrender would
be to reform the regime, perhaps to install a Western-style democracy, or
at least to eliminate the militarism or fanaticism that caused such a threat
to the world. The lesson of Germany and Japan is that such reform is
possible, and that it is accomplished by a total takeover of power, rather
than by stopping short of an occupation of the country.
Given how unwilling the North Korean and Iraqi regimes have been
to risk facing opposition in any free elections, there is no reason to
suppose that someone like Saddam Hussein or Kim Jong-Il really speaks
for his people. But the question remains: If one frees the Kurds and
Shiites, and arrests Saddam Hussein to be tried for war crimes, would
the successor regime be easier to deal with? And if the punishment for
Pyongyangs use of WMD were the reunication of Korea, would a
united Korea renounce nuclear weapons?
Yet one must remember how unsure the United States was in 1945
about the prospects for reforming and democratizing Germany and Ja-
pan. When asked in November 1945 whether they expected Germany or
Japan to give up ideas of ruling the world (or Asia), 63 percent of U.S.
respondents said that Japan would try again, and 46 percent felt this
way about Germany; but, as a sign of a trend, 60 percent had been
pessimistic about Germany on the same poll question in July of 1945.
26
25. On counterpower theories of nuclear targeting, see Bernard Albert, Construc-
tive Counterpower, Orbis, Vol. 19, No. 3 (Summer 1976), pp. 343366.
26. Polls for July 27, 1945, cited in Public Opinion Quarterly, Vol. 9, No. 3 (Fall 1945),
p. 386; polls for November 30, 1945, cited in Public Opinion Quarterly, Vol. 9, No. 4
(Winter 194546), p. 534.
chapter 8 243
The question of Germany or Japan trying to conquer the world or Asia
is so settled today that no public opinion poll would pose it.
In sum, though the Allies were unsure in 1943 what their policy of
unconditional surrender, in response to the mass destruction inicted by
the Axis, was to achieve, in retrospect the policy made sense and was a
success.
In face of an act of mass destruction in 2003, demanding an uncon-
ditional surrender might again make senseand again, it might be easier
to identify the policy than to specify exactly in what way it would succeed.
But by one logic or another, an axe-murderer must be denied access to
tools; by one logic or another, the rogue user of WMD may have to be
forced to surrender.
The primary effect of an aggressors use of WMD will not be to
convince us that we must live with this, but instead that we cannot live
with it. The constraints on political goals that seemed so self-evident
during the Cold War would no longer apply, as an actual renegades use
of WMD might require that the perpetrators entire political system be
overturned and reformed.
244 the coming crisis
Chapter 9
Rethinking How Wars Must
End: NBC War Termination
Issues in the PostCold War Era
Brad Roberts
How might a major theater war involving the use of nuclear, biological,
or chemical (NBC) weapons end? How should it end? How should the
United States think about, plan for, and attempt to shape the termination
phase of such wars?
These questions, and speculative answers to them, are generally
missing from the debate about proliferation, counterproliferation, and the
emerging interstate security order. Yet most everyone in that debate
accepts that a major theater war involving the use of NBC weapons
would be a watershed event, not just for the nations involved and their
immediate neighbors, but also for the nonproliferation project and indeed
for the world order such as it now exists. This chapter examines the war
termination issues associated with NBC regional wars, both to shed some
light on the nature of that watershed, and to identify steps that can be
taken to reduce the likelihood of such wars or to halt them before they
have escalated to their full and horrible potential.
The chapter begins with a critique of the intellectual inheritance on
NBC war termination: the Cold War provided a number of points of
reference that continue to surface in analysis of future war termination
issues, and the Persian Gulf War of 199192 provided some additional
points of reference.
1
Unfortunately, this inheritance includes a body of
analysis largely suited to a strategic era now past and a collection of
sentiment and assumption that ought not be mistaken for systematic
analysis.
The second section elaborates a more systematic approach to the war
termination subject. To predict what perceptions and interests are likely
1. For a comprehensive look at Cold Warvintage thinking on war termination, see
William T.R. Fox, ed., How Wars are Ended, Annals of the U.S. Academy of Political
and Social Science, Vol. 392 (November 1970).
to shape the thinking of decision-makers in the termination phases of a
regional NBC war, it is necessary to project how such a war might unfold,
examining the set of decisions associated with the use and non-use of
NBC weapons, not the structure and character of the larger conict. The
purpose here is to illuminate the factors of signicance to the termination
phase of a war in which NBC weapons have been used.
The third section postulates alternative outcomes and weighs them
for their likely impact on U.S. interests. It argues that asymmetries of
interest, power, vulnerability, and leverage will shape outcomes, but that
those asymmetries will be uid in war and that perceptions of them will
change as the war unfolds. The fourth section emphasizes the fundamen-
tal differences between achieving battleeld victory and winning the
postwar peace.
The chapter closes with a review of the policy implications of the
analysis. This includes a discussion of the essential military and political
ingredients of successful war termination by the United States.
The Intellectual Inheritance
Start a debate on war termination issues, and most U.S. scholars turn
immediately to Herman Kahn and Fred Ikl. Writing in the early 1960s,
Herman Kahn sought to come to terms with the particular problems
associated with the unique dynamics of wars of mass destructionwars
made possible by the advent of the atomic arsenals then being assembled
by the major powers. Such wars involved the risk, indeed the seeming
certainty, of Armageddon. Such wars became essentially unthinkable
from a political perspective and, as Kahn observed, the task of ending
such wars was truly unthinkable.
2
In a world of mutual assured de-
struction (MAD), wars terminated themselves with the spasm that
produced global cataclysm.
Although there was more subtlety to Kahns argument than pre-
sented here, the subtleties have fallen away from the way people utilize
Kahns ideas today. Many people subscribe to the view that wars gone
nuclear are wars whose ending cannot be conceived. The problem with
this way of thinking about contemporary major theater NBC wars is that
such wars, even those waged by an aggressor with nuclear weapons, are
2. To quote: Many people consider nuclear warhow it might start, how it might
be foughtas unthinkable. But even more, the problem of how such a war might
terminate is unthinkable. See Herman Kahn, Issues of Thermonuclear War Termi-
nation, Annals of the U.S. Academy of Political and Social Science, Vol. 392 (November
1970), p. 134. See also Kahn, Thinking About the Unthinkable (New York: Horizon Press,
1962), and On Escalation (New York: Praeger, 1965).
246 the coming crisis
not MAD wars. The assured destruction is not mutual. Such wars do not
threaten the United States with assured destruction: only the United
States (and potentially its coalition partners) would have the capacity to
inict such damage. The regional aggressor may be able to tear off an
arm by imposing signicant costs, military or human, on its opponents,
but not to inict mass destruction. Of course, these costs make avoiding
such wars important. This disparity of destructive power implies that the
United States would win any such war, unless it chooses not to for some
reason. This obliges the United States to think about what outcomes to
such wars it can and cannot tolerate, and what price it is willing to pay
to secure its preferences.
Writing a decade after Kahn, Fred Ikl sought to draw lessons from
the traumatic effort to extricate the nation from the war in Vietnam. In
his book Every War Must End, Ikl explores the factors other than the
balance of military power that shape the nal outcome of wars, including,
for example, the wars impact on domestic politics and the degree to
which outside powers will intervene. His particular focus was limited
wars, in contrast to total wars involving the full application of all power
resources available to the belligerents.
While Ikls arguments have continuing relevance to regional wars
in which the United States plays a role, weapons of mass destruction
feature hardly at all in his analysis. In a new preface to the book written
in 1991, Ikl observes that the strategic concepts of the Cold War era
cannot be relied upon in an era characterized by challenges by NBC-
armed tyrants. He concludes that a new strategy [is needed] to super-
sede the concepts we developed for the bipolar confrontation.
3
Ikls legacy is a particular way of thinking about the NBC war
termination problem that is widespread today among civilian and mili-
tary analysts in the United States. It holds that any war that goes
nuclear moves from the realm of the military to the political, and that
its termination is strictly the presidents job, as he or she searches with
the adversary for points of compromise and accommodation. By this
view, how the next war will end is unknowable. It might go well or badly
from the point of view of the interests of the United States. But its
outcome and termination will depend primarily on the temperament and
skill of the president and the presidents immediate political advisers.
After Ikl, the next seminal statement in this debate was by General
K. Sundarji, the former chief of staff of the Indian military credited with
asserting that the Persian Gulf War showed that if you are going to take
3. Fred Charles Ikl, Every War Must End, 2nd ed. (New York: Columbia University
Press, 1991), p. xv.
chapter 9 247
on the United States you had better have a nuclear weapon.
4
This
assertion implies that in any regional war involving NBC weapons, the
regional power will be able to utilize its mass destruction assets to win.
What does this actually entail? Adherents of this view make three basic
arguments. First, a basic asymmetry of interests would characterize any
such warinterests would be vital for the regional actor and peripheral
for the United Statesand the regional power should be able to exploit
this asymmetry to its advantage. Second, the United States has shown
itself to be very sensitive to casualties; it is deterrable by a state capable
of driving human costs beyond what the U.S. public will bear. Third, the
Persian Gulf War showed the United States to be unbeatable by conven-
tional military means, but vulnerable to nuclear or biological means.
These arguments, correct or not, are used to support the expectation that
the rst major theater war with NBC weapons will end very badly from
the point of view of U.S. intereststhat the United States will be coerced
into acquiescing to aggression and even perhaps backing down from
stated security guarantees.
This view of war termination has generated something of a backlash,
especially within the U.S. military. Adherents also offer three basic argu-
ments. First, the United States can dominate at every level of escalation
initiated by an aggressor. Second, no regional aggressor possesses a nu-
clear arsenal capable of inicting assured destruction on the United
StatesMAD is gone. Third, U.S. superiority implies that an NBC-armed
aggressor could not hope to exploit comparatively paltry assets to deter
the United States from intervening and staying the course. Some adher-
ents of this way of thinking, expressing their sense of U.S. supremacy,
speak loosely of turning the aggressors capital into a glass parking lot.
These arguments are used to support the expectation that the rst major
theater war with NBC weapons will end with unconditional surrender
by the regional aggressor.
These three different ways of thinking about regional NBC war ter-
mination might be deemed schools. The term suggests a coherence the
views lack, but it accurately conveys the way in which different expecta-
tions have begun to coalesce around particular views of the lessons of
the past, assumptions about the present, and expectations of the future.
The three schools certainly imply very different views of the conse-
quences of the rst major theater NBC war. Those in the third school (the
United States has all the trump cards) predict a more orderly world after
4. This view was expressed to a conference of the Defense Nuclear Agency in June
1993. See Proceedings, Defense Nuclear Agency Second Annual Conference on Control-
ling Arms, Richmond, Virginia, June 1993.
248 the coming crisis
any NBC war, as the worlds only superpower extends its security to
others. Those in the second school (the aggressors nuke is the trump
card) expect the rst major theater war involving NBC weapons to result
in a breakdown of a U.S.-dominated global security structure, as allies
lose condence in the United States and adversaries are emboldened to
test new limits.
5
Those in the rst school (its the presidents job) see
the future as a mystery, as dictated by whim and personality and the
domestic politics of the moment.
These schools share one striking feature: they posit war outcomes as
driven by certain inevitable forces of power, history, or personality. Out-
comes are foreordained. But of course history does not work that way.
Certainly, the history of war is full of surprisesof results that were not
predicted ahead of time by one or both sides. Wars unfold in a series of
choices by the belligerents, choices made on the basis of available military
capabilities; what is seen as militarily and politically necessary, viable, or
suicidal; and perceptions of what is at stake in each stage. The combat-
ants sense of honor can play a decisive role in their decisions.
6
By
thinking through these factors and their impact on the dynamic of an
NBC war, it is possible to get beyond mechanistic predictions of the
outcomes of such wars to a better understanding of how the central
decision-makers will conceive their interests and stakes and the means to
protect themand how these will play out in the war escalation and
termination phases. This is the purpose of the following section.
NBC Wartime Choices: Why, When, and How?
For purposes of this analysis, the focus here is on three primary choices:
the aggressors choice to initiate NBC warfare, the U.S. choice about how
to respond, and the aggressors choice about how to respond to U.S.
actions or inaction.
why initiate nbc warfare?
The rst choice is the aggressors: whether or not to initiate the use of
NBC weapons. For many in the U.S. strategic community, the choice to
use NBC seems bafing or irrational. But such use might appear to an
aggressor as logical and necessary. An aggressor might be motivated to
initiate the use of NBC weapons for a variety of reasons. It might believe
that such use would cripple U.S. political will to prosecute or escalate a
5. This argument is elaborated in Stephen Rosens chapter in this volume.
6. Donald Kagan, On the Origins of War and the Prescription of Peace (New York: Anchor
Books, 1995).
chapter 9 249
conict by inducing great fear among the U.S. public (and among poten-
tial coalition partners). It might hope to achieve prompt battleeld victory
over a militarily superior United States by using massively destructive
weapons at an early stage of a conict. Or it might conclude that it can
employ such weapons to create a fait accompli in regions where the
United States is not engaged militarily and lacks historical involvement.
These three purposes t under the general rubric of deterring, defeating,
or preempting U.S. action.
If we credit the leaders of NBC-armed regional states with both
rationality and at least a modicum of strategic calculus, how might an
aggressor think about the risks associated with these uses of NBC weap-
ons?
7
Might the leaders convince themselves that the risk-benet balance
favors their interests? For uses aimed at deterring U.S. action, the aggres-
sor must recognize a possibility that the use of NBC weapons to intimi-
date the United States might backre. Rather than inducing accommoda-
tion and appeasement, such use could generate anger or acts of reprisal,
or strengthen U.S. will to permanently expunge the threat. Coalition
partners threatened by NBC weapons might well urge the United States
to decisive action, if they come to believe that an aggressor has grown
intolerably menacing.
8
The taboo against all these weapons remains
strong in the countries that an aggressor might target for intimidation.
But an aggressor might well conclude that U.S. interests at stake in the
region are peripheral, while its own stake is vital; this is the basis of the
expectation that the United States will disengage in response to NBC
threats and attacks. Moreover, noting U.S. withdrawal from conicts in
places like Somalia and Lebanon once the human costs began to escalate,
an aggressor might expect to manipulate U.S. decision-making processes
and institutions by threatening large human costs. The head of an insular
and authoritarian regime in particular might well believe that a U.S.
public made fearful through shows of indiscriminate violence, and the
hint of more to come, will back down. Such a leader might conclude that
the risks of rousing potential U.S. coalition partners are outweighed by
the likelihood that their vulnerabilities will result in efforts by them to
exert leverage on Washington to make compromises that seem to prevent
7. For a discussion of the connection between psychopathology and deterrence and
a historical review of failures of deterrence, see Keith Payne, Deterrence in the Second
Nuclear Age (Lexington: University Press of Kentucky, 1996), especially chap. 4.
8. For more on this line of argument, see Brad Roberts and Victor Utgoff, Coalitions
Against NBC-Armed Regional Aggressors: How Are They Formed, Maintained, and
Led? Comparative Strategy, Vol. 16, No. 3 (JulySeptember 1997), pp. 233252.
250 the coming crisis
further loss of life. An aggressor might also calculate that the risks of
violating the taboo on biological and chemical weapons are outweighed
by the advantages such weapons offer over nuclear weapons. These
advantages are their ability to kill invisibly, the fear and revulsion they
generate, and the possibility that their use for purposes of coercion might
be less likely than the use of nuclear weapons to generate undesirable
responses by the United States.
An aggressor that opts to use NBC weapons in pursuit of battleeld
victory seems likely to go through a similar risk calculus. One risk is that
the use of NBC weapons might not derail U.S. military action. Perhaps
the aggressor has overstated U.S. military vulnerabilities or has failed to
use its own weapons effectively. Another risk is that even militarily
crippling attacks may lead not to the desired disengagement of U.S. forces
but to heightened political resolve in Washington to see the war through
to unconditional surrender by the aggressor. But an aggressor might
convince itself that there is at least a reasonable chance that the United
States will not try to reverse the aggressors gainsor that the United
States will not threaten the regime in doing so. Or it might believe that
the risks of using NBC weapons are less than the risks of not doing
sothe failure to try to take the initiative in a looming conict with the
United States may threaten the survival of the regime or compel it to live
with a regional status quo that is itself threatening to the regime or is
otherwise intolerable.
An aggressor that opts to try to preempt U.S. action would also
confront some risks. The primary risk is that the United States sometimes
nds its interests engaged by events in regions where any interests had
previously been explicitly discounted. U.S. interests are not always clearly
perceived or well dened, but catalytic events can illuminate them and
induce the United States to act decisively. Washingtons decision in 1950
to come to the defense of South Korea shortly after its explicit exclusion
from a catalogue of U.S. interests is a case in point. The use of nuclear
weapons anywhere could well lead U.S. leaders and the public to believe
that major issues were at stake, which the United States could not allow
to play out from the sidelines.
9
An aggressor opting to use NBC weapons
must judge the United States as lacking both the interests and the incen-
tive to risk massively destructive war to reverse the aggressors gains.
9. As Lawrence Freedman has argued, even in regions remote from U.S. interests and
where the United States lacks deployed military forces, an aggressors use of WMD is
likely to create a compelling national interest for the United States, and for the nuclear
weapons states generally. See Freedman, Great Powers, Vital Interests, and Nuclear
Weapons, Survival, Vol. 34, No. 4 (Winter 19941995), pp. 3552.
chapter 9 251
Thus an aggressor might well deem the use of NBC weapons to be
strategically sound, if risky. Given those risks, an aggressor opting to use
those weapons will likely limit their use to the lowest possible level seen
to be likely to achieve the desired result without stimulating an overre-
action. Or, if the regime anticipates the impossibility of surviving any
confrontation with the United States unless the latter is paralyzed mili-
tarily and politically, it might develop a go-for-broke attitude.
10
how might the united states respond?
If the rst choice is the aggressors, the second is that of the United States:
to respond or not to the aggressors use of NBC weapons and if so, by
what means. The United States might choose to withdraw its forces and
disengage from the conict. Or it might strike back in reprisal or retali-
ation. Its war aims might expand. It might opt to limit its military
response to conventional weapons, or to employ nuclear ones. Choices
among these options are likely to be shaped by many factors. Seven of
these factors are discussed here: the level and character of destruction;
the potential for continued destruction by the aggressor; the mood of the
U.S. public; congressional debate and action; the international political
context; military advice; and the U.S. stake.
the level and character of destruction inflicted by the aggressor
with its use of nbc weapons. If the destruction inicts few casualties,
or is inconsequential for U.S. military operations, the United States may
opt to prosecute the war without specic reply or reprisal, while offering
perhaps a commitment to remove the offending regime. However, if the
aggressors use of NBC weapons generates high casualties or cripples
U.S. and coalition military operations, then the United States may feel
compelled to use its own nuclear weapons in response. If the destruction
were limited to battleeld targets, the United States might opt to limit its
own nuclear response to similar targets. If the destruction reaches into
the territories or cities of neighboring states or U.S. allies, the United
States may feel compelled to inict greater punishment on the aggressor
state. If the destruction reaches into the United States itself, U.S. leaders
10. Such a reading of U.S. will is not unknown in history. For example, Japan risked
a politically crippling blow at Pearl Harbor, though the United States was far more
capable of war mobilization than Japan. States unlikely to prevail in a long-term war
sometimes conclude that war can be made and won in the short term with a decisive
attack on the national resolve of the larger power. See Barry Wolf, When the Weak
Attack the Strong: Failures of Deterrence, RAND Note N-3261-A (Santa Monica,
Calif.: RAND, 1991).
252 the coming crisis
would likely feel compelled to respond immediately, sharply, and deci-
sively.
the potential for continued destruction by the aggressor. If the
potential for continued destruction is known to be virtually nonexistent
because the aggressor has used most or all of a small NBC force in its
initial salvo, the United States might press on with existing war plans and
aims. But if some NBC potential remains, a counterforce strike is likely
to be seen as an urgent necessity. If a limited potential remains, the United
States might opt to reply with a limited strike of its ownlimited either
in the types and numbers of targets struck or in the weapons chosen. But
if such a limited strike is seen as unlikely to deprive the aggressor of the
ability to inict signicant damage, perhaps for extended periods of time
and beyond the theater of combat, then such restraint may be seen as
unwise. Instead, the United States might opt to deliver as hard a blow as
possible, perhaps even with nuclear weapons, in the hope that this would
eliminate the aggressors potential for continued destruction. Even if such
a strike does not promise complete success, it may be seen as useful for
various purposes: breaking the aggressors will to continue to use NBC
weaponry, punishing the rst use, demonstrating U.S. willingness to
escalate, and reassuring coalition partners that the war will end quickly
and with defeat for the aggressor. Such a strike would also be useful for
impairing the aggressors NBC capabilities to the point where active and
passive U.S. defenses deprive them of any meaningful operational im-
pact.
11
However, if the end of the confrontation seems sufciently close,
the United States might refrain from reprisal, acting instead to seek a
political settlement that induces the aggressor not to use its reserve
capability.
the mood of the u.s. public. The U.S. publics sensitivity to casualties
is oft remarked, as is the role of the media in magnifying the impact of
casualties on the national mood. The U.S. public may well cringe and
shrink in fear after an aggressors use of NBC weapons, but its sensitivity
is often misunderstood. Commentators who expect the U.S. public to
cringe in fear as casualties mount seem to take little cognizance of the dif-
erence between peripheral and vital interests in the reactions of the U.S.
11. For a discussion of the utility of such defenses in inuencing the outcome of wars
against WMD-armed regional adversaries, see Victor Utgoff and Johnathan Wallis,
Major Regional Contingencies Against States Armed with Nuclear, Biological, and
Chemical Weapons: Rising Above Deterrence (Alexandria, Va.: Institute for Defense
Analyses, June 1996).
chapter 9 253
public and the historically well-demonstrated willingness of the public to
suffer casualties in the name of a signicant national interest.
12
Similarly, the antinuclear sentiment prevalent today in the United
States suggests that the president could face strong pressures not to use
U.S. nuclear weapons except in response to attacks on targets in the
United States itself. But once an aggressor breaks the nuclear taboo, and
especially if large numbers of U.S. casualties result, an explosion of
outrage and anger would likely ensueespecially when television brings
home pictures of U.S. victims. The result could be entirely unanticipated
by those who make predictions based on the U.S. record in Somalia and
Lebanon. Herman Kahn has written of the historic tendencies of a U.S.
public inclined to a messianic worldview to apply extravagant force
when roused to action.
13
Rage could combine with moral fervor to pro-
duce a U.S. military campaign to expunge an aggressor seen in the United
States as evil, a campaign that would draw on all available military
capabilities to seek the most immediate and decisive end to the war at
hand and a permanent removal of any threat of future war by this regime.
Moreover, because appeasement has a particularly pejorative mean-
ing in the United States (where it is associated with cowardice, not
compromise), the president would likely face strong pressure not to back
down from the defense of U.S. interests in the face of threats from what
the public might view as a nuclear pygmy. Nuclear attack on U.S. forces
could well produce strong popular pressure for a nuclear reply by the
United States, one aimed not just at reprisal or escalation containment,
but at immediately ending the conict and vanquishing the foe. It is
useful to recall the private intervention with Saddam Hussein by Japa-
nese Prime Minister Yasuhiro Nakasone, as reported in the press at the
time. After U.S. Secretary of State James Baker delivered the famous letter
to Iraqi Deputy Prime Minister Tariq Aziz threatening terrible conse-
quences for certain Iraqi actions if war came, Nakasone reportedly sought
out Hussein to attest to the fact that the United States is the kind of
country that would make good on its promise of retaliation and retribu-
tionas Japan knew better than most.
12. See Eric V. Larson, Casualties and Consensus: The Historical Role of Casualties in
Domestic Support for U.S. Military Operations, RAND Report MR-726-RC (Santa Monica,
Calif.: RAND, 1996); Benjamin C. Schwarz, Casualties, Public Opinion, and U.S. Military
Intervention: Implications for U.S. Regional Deterrence Strategies, RAND Report MR-431-
A/AF (Santa Monica, Calif.: RAND, 1994); and Mark Lorell and Charles Kelley,
Casualties, Public Opinion, and Presidential Policy During the Vietnam War, RAND Report
R-3060-AF (Santa Monica, Calif.: RAND, March 1985).
13. Kahn, On Escalation, p. 17.
254 the coming crisis
congressional debate and action. Members of Congress would cer-
tainly have strong views about what to do in a nuclear crisis, views that
would have weight not least because of the congressional power of the
purse and its ability to shape public opinion. Sharply different views
could be presentfor every voice of compromise or retreat, there could
be others shouting dont tread on me and calling for decisive U.S.
action. Or Congress might rise up in near unanimity to urge a particular
course of action upon the president. Whether Congress would clarify or
obfuscate the choices before the United States and its national command
authority is an open question. The answer would have much to do with
the character of leadership to be found among key committees, their
chairpersons, and the political parties. The skill of the executive branch
and especially the president in shaping the legislative debate would be
critical. The result might be the building of consensus from dissension,
as seen in the lead-up to Desert Storm. Or consensus may not be
achieved. The president would value consensus, which would strengthen
the administrations political hand, but dissension could preserve the
freedom to maneuver.
the international political context. If the United States has assem-
bled an international coalition to face an NBC-armed aggressor, its basic
strategic choices would have to be made with the coalition in mind. This
might strengthen the tendency to back down in a confrontation, if the
coalition is held in check by its most cautious members; or a sense of
acute vulnerability among coalition members could stimulate calls for
immediate, decisive action. The United States might nd that its exibil-
ity to choose among options is limited by its leadership of a coalition in
which terms of engagement have been well negotiated as a price of
coalition formation, or it might feel less constrained than when acting
alone if the international legitimacy offered by the coalition works to
expand its options. Deciding how to respond to an aggressors use of
NBC weapons would be especially complex when other coalition mem-
bers have the capacity to respond independently with NBC attacks of
their own. If the coalition were to include states willing and able to use
nuclear, biological, or chemical weapons of their own, the United States
would nd itself in the difcult position of needing to act to forestall or
perhaps condone such use at the same time that the weapons of mass
destruction (WMD) capabilities of the aggressor are a focus of interna-
tional fear and condemnation. As in the Persian Gulf War, when the
United States persuaded Israel not to retaliate for Iraqi missile strikes on
its cities, it seems likely that the United States would try to maintain the
chapter 9 255
maximum degree of control over both the local dynamics of the conict
and the international politics of the confrontation. Furthermore, in the
capitals of allied and coalition nations, concern about the durability of
U.S. security guarantees and about U.S. credibility more generally would
add to the brew. Even in wars where the United States nds itself ghting
without a coalition, many of these concerns would nevertheless still be
likely to affect the United States as it calculates the long-term conse-
quences of its choices on U.S. alliance relations and its more general
reputation and political standing.
14
military advice. The inuence of a presidents military advisers peaks
in time of military crisis. Particularly on the issue of a possible U.S.
nuclear reply, the precise nature of military advice is difcult to predict.
The disposition against nuclear weapons is strong in the U.S. military
today, reecting not least its condence in its ability to carry out its
missions without resort to weapons of mass destruction.
15
But the mili-
tarys antinuclearism might perish once U.S. forces or interests are
attacked with weapons generating mass casualties. In reply, U.S. nuclear
use might be deemed necessary and appropriate to punish the enemy, to
prevent its further use of NBC weapons, to collapse its military apparatus
to prevent further aggression, or to save lives, whether U.S. or allied.
Some military advisers might argue that the restoration of the WMD
taboo requires U.S. nuclear retaliation to punish the aggressor and to
substantiate the notion that nuclear weapons are not useful as instru-
ments of aggression or even as ultimate guarantors of the sovereignty of
aggressive regimes.
the u.s. stake. The point is not trite. It seems reasonable to assume that
the United States would not nd itself in confrontations that might
involve NBC weapons except in instances of compelling national interest.
But, as many have noted, U.S. history is replete with episodes in which,
as casualties mount, its level of military engagement has been deemed
inappropriate for the stake involved, with a consequent U.S. withdrawal
and loss of face (e.g., from Lebanon and Somalia). Thus, the United States
14. For more on this theme, see Roberts and Utgoff, Coalitions Against NBC-Armed
Regional Aggressors.
15. This antinuclear disposition has been identied by RAND analysts in follow-up
to the Day After . . . study of nuclear futures. See Marc Dean Millot, Facing the
Emerging Reality of Regional Nuclear Adversaries, Washington Quarterly, Vol. 17, No.
3 (Summer 1994), pp. 4171; and Roger C. Molander and Peter A. Wilson, On Dealing
with the Prospect of Nuclear Chaos, Washington Quarterly, Vol. 17, No. 3 (Summer
1994), pp. 1939.
256 the coming crisis
might somehow be drawn into an NBC confrontation over issues not of
central importance to it. Such a confrontation would likely present U.S.
leaders with an unpalatable choice between risking an unpopular war
fought with unpopular weapons and trying to cut its losses.
Of course, the NBC component may reveal U.S. interests related to
perceptions of its power and credibility that neither it nor the aggressor
had conceived. Those interests could lead the United States to take actions
not expected by an aggressor that believes the U.S. stake in the conict
to be low. Even if the aggressor only threatens to use NBC weapons,
compelling interests could well be generated for the United States. Wash-
ington would likely feel that a perception that the United States had
backed down in the face of NBC threats from a regional aggressor would
have terrible consequences for the United States, both politically and
militarily.
likely u.s. preferences. These seven factors, and others, as magnied
by the media and played out through the personalities of the national
command authority, would shape the choices made by the United States.
Given the variety and contingent character of these factors, simple pre-
dictions of U.S. war-time choices, whether by U.S. analysts or by leaders
in an aggressor state, seem foolhardy. But likely preferences can be illus-
trated.
A decision by the United States to back down or withdraw from a
confrontation when an aggressor has used nuclear weapons seems un-
likely. The president, the U.S. public, and the international community
would have to see little benet in punishing the aggressor. Key constitu-
encies would have to be deeply divided, and the stake widely seen as
disproportionately low compared to the costs of continued action. This
conclusion seems valid even for the mere threat of use. Many constitu-
encies would likely feel that it was very important not to allow the
precedent to stand. Moreover, the president would have to choose to
allow blackmail by an aggressor, and it is far from clear that the domestic
or international political context would permit this to happen.
A decision to reply with military strikes aimed at annihilating the
aggressor state and society is only slightly less unlikely. Key domestic and
foreign constituencies might be gripped by rage or fear, and thus might
seek such a reply to punish the act of nuclear aggression and remove the
threat of continued aggression. Decision-makers might be deluged with
pleas to seek an immediate cessation of the conict in order to protect
the broader interests of the United States and the international commu-
nity. But the president would likely also be cautioned to act with restraint,
and could opt to use U.S. force cautiously in the hope that a settlement
chapter 9 257
might become possible without U.S. actions that cause enormous addi-
tional casualties.
A decision to ignore an aggressors nuclear use and to continue the
military confrontation on previously existing terms appears likely in
certain circumstances. The victims of the rst attack must be few, the
aggressors capacity for continued nuclear use must be low, the demands
for retribution must be few, and a means of sufciently expeditious
military victory at minimal human cost must be available.
A variant of this option would be the continued utilization of con-
ventional weaponry in pursuit of expanded war aims. All wars unfold
within the context of an agreed battle by which adversaries accept what
is and is not at stake; changes to the agreed battle by expanding it to
additional fronts, whether military, political, or economic ones, might be
alternatives to a U.S. nuclear reply.
16
Such strategies may be successful in
imposing new costs upon the aggressor sufcient to secure an outcome
acceptable to the United States. But such expansion may not be feasible.
If an interventionary force had already thrown its full resources into the
conventional battle it would have no further units to deploy, at least in a
timely fashion, no new fronts to open, and no new targets to attack.
Expansion of war aims to include removal of the regime in power might
offer some leverage. But some regimes may already believe that their
survival is at stake in a war with the United States, no matter what the
U.S. declaratory policy. Moreover, by spelling out plans to remove the
regime, the United States may well remove the last barriers of restrainta
regime with nothing to lose may go for broke.
A decision to meet an aggressors use of nuclear weapons with a
limited nuclear reply by the United States seems at least as likely as these
last two replies. Especially if victims are numerous, the aggressors ca-
pacity for continued nuclear use remains, publics are angry and vulner-
able, and the political groundwork has been done for a concerted coali-
tion response, such limited use appears likely.
The use of nuclear weapons would likely be opposed by those who
believe that conventional weapons can do the necessary reprisal, retali-
ation, and defeat. However, while conventional weapons have taken on
many of the strategic targets traditionally assigned to nuclear weapons,
and may yet take on more, the speed with which they can achieve U.S.
war aims is considerably less than that of nuclear weapons. Carrying on
the battle by conventional means alone could come at a high cost, if the
aggressor can continue to inict mass casualties on the forces and popu-
16. Agreed battle is Max Singers term, as discussed in Kahn, On Escalation,
pp. 48.
258 the coming crisis
lations of coalition members before the coalition beats its forces on the
battleeld.
If the conventional option seems likely to be expensive, the president
would face a difcult choice: employ nuclear weapons, hoping that this
will reduce the casualties suffered by the coalition, even if their employ-
ment might cause many casualties within the aggressor state; or refrain
from their use, hoping that overall casualties will be kept to a minimum,
even if a disproportionate share is borne by the coalition members. The
president would no doubt be advised to do everything possible to protect
U.S. forces and to minimize the casualties borne by the regional victims
of aggression, even while tallying the possible political costs associated
with U.S. nuclear use. Of course, the United States made such a choice
in 1945. Faced with a clear choice between using nuclear weapons to try
to end a war quickly and relying on slower but still certain non-nuclear
means, a U.S. president would probably not want to risk the sacrice of
further U.S. lives, or lives of those within the region already victimized
by aggression, to resurrect a non-nuclear taboo that the enemy had
already broken.
These predictions are all based on the aggressors use of nuclear
weapons. What about biological and chemical weapons? Can they be
expected to shape U.S. perceptions and choices similarly?
Chemical weapons seem unlikely to produce the same effects. Gen-
erating large casualties against well-protected forces takes a great deal of
sustained effort by a chemically armed aggressor. Moreover, chemical
weapons do not have anything like the political symbolism of nuclear
weapons. But it is also true that chemical weapons can be used to mas-
sively destructive effect. An aggressor equipped with the means to
efciently deliver a chemical agent on unprotected population centers or
military targets of campaign signicance such as ports, airelds, and
other logistic centers could signicantly disrupt an intervention force,
especially in the early stages of deployment (when such weapons might
generate relatively few casualties). Furthermore, a very strong taboo is
attached to such weapons in the mind of the U.S. public, which might
well feel that strong punishment of such use would be necessary. So a
U.S. nuclear reply to an aggressors use of chemical weapons cannot be
ruled out entirely.
Biological weapons seem likely to be closer to nuclear than chemical
weapons in their effects on U.S. choices. Their potential for mass destruc-
tion of unprotected populations is indisputable, as is their utility in
attacks on unprotected military targets of campaign signicance. Wash-
ington would expect an aggressor that uses them once to use them again,
and even the most potent preemptive counterforce attacks are unlikely
chapter 9 259
to remove the fear that the aggressor might be able to inict signicant
continued destruction with such weapons. The U.S. public could be
particularly enraged by the use of biological weapons, given the deep
abhorrence for them. Internationally, there would likely be many calls to
reply to and punish such use in order to demonstrate to all states with
biological weapons programs that such weapons cannot be used success-
fully and will bring down the full wrath of the international community.
Indeed, the United States will have a very strong interest in ensuring that
the right lesson is drawn from the rst major biological confrontation,
a lesson that leads away from and not toward such weapons. On the other
hand, biological attacks that generate few casualties and marginal mili-
tary or political consequences would be less certain than even limited
nuclear attacks of stimulating strong U.S. responses.
In considering possible U.S. responses to an aggressors use of chemi-
cal or biological weapons, due consideration must be given the fact that
the United States has forsworn both chemical and biological weapons.
Accordingly, it no longer has an option to reply in kind to attack on
U.S. forces with such weapons.
17
The absence of an in-kind capability
may be inconsistent with the desire for maximum U.S. exibility, but the
U.S. military sees no advantage in making in-kind chemical or biological
retaliatory strikes to rst-use by an enemy, preferring instead a mix of
other capabilities, including self-protection, defenses against delivery sys-
tems, conventional preponderance, and nuclear threats. A dilemma arises
when these latter capabilities might prove inadequate to meet the require-
ments created by the seven factors enumerated here. For example, if an
aggressors use of chemical or biological weapons generates huge casu-
alties, cripples U.S. or coalition military operations, enrages public and
political sentiment, and if the potential for continued destruction is sig-
nicant, the U.S. command authority seems likely to consider a nuclear
reply. It would then face a clear choice between continuing the war by
conventional means, thus suffering large numbers of casualties, and in-
itiating nuclear use aimed at bringing that risk to an end.
Thus the United States (and the coalition) could face decisions about
the propriety of making the rst use of nuclear weapons, albeit not the
17. Biological weapons were renounced unilaterally in 1969 as a prelude to a bilateral
U.S.-Soviet biological weapons disarmament agreement and ultimately the multilat-
eral Biological and Toxin Weapons Convention of 1975. Production ceased and the U.S.
stockpile was destroyed by order of the Nixon White House. Chemical weapons were
renounced in 1992 in the endgame to the negotiation of the Chemical Weapons
Convention. A residual stockpile of chemical munitions remains in the U.S. arsenal but
is slowly being destroyed.
260 the coming crisis
rst use of weapons of mass destruction. If the aggressor state is a party
to the Nuclear Non-Proliferation Treaty (NPT) and is considered not to
be in violation of that treaty, U.S. choices would be constrained by the
assurances it has made to NPT parties. But questions of legality and of
justice may prove separable. Legally, the United States is obliged not to
use nuclear weapons against states that are party to the NPT (except for
those aligned with a nuclear weapons state).
18
But the failure to use
nuclear weapons may be deemed unjust in time of war, if the legal
stricture comes to be seen as the only barrier to a quick end to a war of
aggressionespecially if the aggressor is using other banned weapons to
inict mass casualties on its neighbors and on the members of the coali-
tion formed to reverse the aggression. Both biological and chemical weap-
ons are the subject of a global arms control regime. Of course, if the state
is not a party to the NPT or is in violation, the legal constraint disappears,
but the presumption against U.S. nuclear use in reply to non-nuclear
challenges would remain.
Just as the aggressors choices are not risk-free, the United States
would also face risks as it calculates its choices. It may miscalculate the
ability of the aggressor to survive a limited nuclear attack and to continue
to inict damageor to escalate that damage to the U.S. populace or
sustain that damage with acts of vengeance against U.S. leaders or the
U.S. public in the months or years after battleeld confrontation ends.
The United States may misread the intent or strategic personality of the
aggressor, only to nd that actions intended to deter escalation instead
raise its ambitions. An NBC-armed aggressor with nothing to lose may
become even more dangerous, abandoning a strategy of strategic bargain-
ing aimed at escaping war in favor of a strategy of punishment and
retribution in the nal stage of defeator even after losing the war.
Biological weapons may be seen as particularly attractive and effective
for this purpose. Indeed, the fear that an aggressor might use biological
agents surreptitiously after the war could emerge as an important factor
in Washingtons calculation of an adversarys capacity to inict continued
destruction. It is instructive to recall that the United Nations Special
Commission on Iraq (UNSCOM) discovered that Saddam Hussein dis-
persed biological weapons from their storage facilities as the war began,
18. See Treaty on the Non-Proliferation of Nuclear Weapons, in Arms Control and
Disarmament Agreements: Texts and Histories of the Negotiations (Washington, D.C.: Arms
Control and Disarmament Agency, 1990), pp. 89102. See also Michael Wheeler, Posi-
tive and Negative Security Assurances, Paper No. 9, Project on Rethinking Arms Control,
Center for International and Security Studies at Maryland (College Park: School of
Public Affairs, University of Maryland at College Park, February 1994).
chapter 9 261
with orders that they be launched against coalition military forces and
neighboring states in the event that the coalition destroyed Baghdad.
19
the aggressors response
Once an aggressor uses NBC weapons and the United States makes its
response, what then? Would the aggressor opt to continue the ght with
weapons of mass destruction, accept defeat, or something in between?
An aggressors decision to continue the WMD conict would likely
surprise those in the United States who believe in the dissuasive effect of
the overwhelming escalatory potential of the U.S. arsenal. But what might
an aggressor that calculates risks and tries to shape the ow of events
think? How might it perceive its options? If the United States had chosen
not to respond with nuclear weapons to the original attack, the aggressor
may believe that U.S. weapons would be used only when the very
survival of the nation is threatened, and thus may believe that it is free
to operate with NBC weapons below that thresholdeven to the point
of inicting massive destruction on its immediate neighbors. If the United
States has responded with nuclear weapons in a limited way, the aggres-
sor may perceive that judicious continued use, calculated not to excite
stronger U.S. responses, may be possible. It may believe that calibrating
its WMD attacks to the U.S. action by not expanding target sets or levels
of destruction might stabilize the conict while also giving the United
States strong incentive to seek a compromise solution. Moreover, how-
ever rational its calculations of costs and benets of various options, the
leader may face keen pressures from within his or her regime or else-
where to strike a decisive, last-ditch blow at the United States, calculating
that only the collapse of U.S. will might allow the regime to survive.
The risks for the aggressor are obvious. Continuation of an NBC
conict with the United States would probably remove any lingering
opposition in Washington and elsewhere to a strategy aimed at uncondi-
tional surrender by the aggressor and removal of the regime in power (if
these were not already coalition war aims).
20
An aggressors continued
use of NBC weapons after a U.S. nuclear reply would certainly discredit
those within the U.S. decision-making process who had argued for re-
straint in the U.S. reply, in the hope that an aggressor might reciprocate.
19. Report of the Secretary-General on the status of the implementation of the
Special Commissions plan for the ongoing monitoring and verication of Iraqs
compliance with relevant parts of section C of Security Council resolution 687 (1991),
United Nations Reference S/1995/864, October 11, 1995.
20. For more on this line of argument, see George Questers chapter in this volume.
262 the coming crisis
Remaining constraints on U.S. actions would rapidly erode, leading to
U.S. efforts to promptly and decisively end the war. At this point in a
conict, the regimes control of its military and society could begin to
break down. A leadership core committed to a war of national suicide
may nd that the military prefers another course, perhaps to ght another
day. It may nd that the public turns against it, especially elites that
heretofore had supported the regime out of a commitment to nationalism
and personal status. Loss of control by the aggressive regime would likely
create opportunities for negotiated settlements. The fear of a loss of
control might well be an important incentive for the regime not to choose
to prolong an NBC exchange with the United States. Moreover, even if a
state somehow succeeded in inicting defeat on the United States within
the theater, it would have to contemplate the possibility that other states,
including nuclear-armed ones, would not stand idly by, given their own
stake in an international system in which the United States is the locus
of initiative and the provider of security guarantees.
The Military Outcomes of NBC Wars
What does this analytical framework contribute to an understanding of
NBC war termination issues? At the very least, it reveals a greater range
of outcomes than one-side-takes-all. It is possible to envisage at least six
different outcomes of any regional war in which NBC weapons have been
used.
First, the aggressor might realize all goals. This might happen if the
United States refrains from using its nuclear weapons after an aggressor
uses NBC weapons to deny the United States or a U.S.-led coalition
military victory, or uses those weapons to attack cities of coalition mem-
bers, including perhaps the United States, successfully breaking the will
of the coalition to proceed.
Second, the aggressor might achieve some goals and retain some
spoils of war, but not its full original ambition. This could happen if the
aggressors use of NBC weapons has limited effectswhether on the
battleeld, to stalemate or defeat an intervention by U.S. forces or by a
U.S.-led coalition, or politically, in hobbling or crushing that coalition.
Nuclear use by the United States is unlikely to be a part of this outcome.
Third, the status quo ante might be restored. WMD use by one or
both sides produces stalemate on the battleeld and politically. The in-
tervening powers resort to long-term political strategy rather than mili-
tary solutions to contain further aggression and punish past behavior.
chapter 9 263
Thus, the aggression is overturned but the aggressor pays no substantial
price.
Fourth, the aggression could be overturned with the aggressor pay-
ing a substantial price, perhaps the loss of the invasion force, but the
regime remains in power. This might happen if the aggressors use of
NBC weapons fails to deter the formation of a coalition or defeat its
efforts to reverse the original aggression. In this scenario, the United
States would make limited or no use of its nuclear weapons.
Fifth, the aggression might be overturned and the aggressor regime
removed from power in a war that the United States opts to pursue by
conventional means alone, and in which its forces and allies accordingly
suffer substantial WMD-inicted casualties.
Finally, the aggressor might be defeated and removed from power in
a war where both sides use weapons of mass destruction, perhaps sub-
stantially.
Which outcomes are most likely? Which least likely? Why? At this
point, it is useful to return to the schools of thought sketched at the
beginning of this chapter. The view that aggressors will win because their
NBC assets are a trump card embodies an important truth. There would
be political pressures to end the war before it escalates, or before a second
exchange of weapons of mass destruction, and thus to nd points of
compromise. The undisputed U.S. capacity to inict decisive military
defeat on a regional aggressor combined with the public rage and urge
to inict punishment on it will be juxtaposed with the need to keep a war
from getting out of hand. The high human costs associated with all
outcomes, even outright U.S. victory, and the uncertain politics of nuclear
use, suggest that the United States will be constrained in dening its
responses to NBC use and will be pressured to accept outcomes that leave
some major issues poorly resolved.
But this school overlooks the many U.S. trump cards, and the fact
that it would be likely to enjoy more political and military latitude in
meeting the dilemmas of conict than an aggressor. An aggressors mo-
ment of maximum inuence will probably be its rst hints at the possible
use of NBC weapons, when the fear can be reaped for political conces-
sions. But once the aggressor engages the United States in direct military
confrontation with weapons of mass destruction, its war aims are likely
to shift from gaining something to survival. The weapons will be risky
to use because of the response they may generate. A coup may remove
the regime that initiated the aggression. Or that regime may struggle to
maintain a domestic power base while abandoning the interests of terri-
tory or prestige that seem to be leading the state to a war of national
264 the coming crisis
suicide. An uncertain period could come if a leader turns toward
peacewhen it has eschewed victory but has not accepted defeat, and
decides on a strategy that maximizes its chances of surviving to ght
another day.
21
Thus the school of thought that the United States will win no matter
what also embodies an important truth. The United States would enjoy
many advantages of political position and military exibility and reach
over a regional aggressor equipped with a small arsenal of NBC weapons.
It did not enjoy such advantages when it faced the Soviet Union, which
was armed with what were seen as superior conventional forces and a
capacity to inict assured nuclear destruction on the United States. Nu-
clear choices by the United States can shape war outcomes in ways that
a regional aggressors cannot.
But if regional NBC war seems somehow easier than a MAD war, it
is likely to prove neither easy nor cost-free. Aggressors options would
be far more limited than those of the former Soviet Union, but they would
have options nonetheless: to use weapons of mass destruction, perhaps
to continue to do so as a conict evolves, and to target civilians in or
beyond the zone of combat. The U.S. ability to prevail at any level of
confrontation chosen by the aggressor does not mean that prevailing will
be easy, as there will be certain inescapable costs, human and other.
Furthermore, the U.S. ability to control an unfolding conict depends on
its military preparations to minimize the leverage of potential opponents
or political work to create durable coalitions, actions that it may prove
unable to execute satisfactorily.
Basic asymmetries in power, vulnerability, and thus leverage point to
the conclusion that how NBC wars will end depends fundamentally on
how the United States wants them to end. What conditions might the
United States nd necessary in dening acceptable war outcomes? At a
minimum, the United States would want to extract something from the
war termination process that it can call a win. Stalemates would likely
be unacceptable to senior policymakers, who would be concerned that
the aggressor and its use of NBC weapons be punished so as not to
encourage future aggression, particularly aggression backed by weapons
of mass destruction. If the aggressors use of NBC weapons is seen as
more incidental than egregious, the United States might settle for an
outcome that reliably eliminates the weapons potential of the state. But
U.S. demands would escalate as the damage increases, and the United
21. The term is from Fox, Causes of Peace and Conditions of War, p. 6. See also
Ikl, Every War Must End, especially chaps. 4 and 5.
chapter 9 265
States and especially coalition partners neighboring the aggressor would
probably also push for the removal of the regime and the reform of the
state.
Control of the regime, disarmament of the military, and reform of the
state are not modest goals. They are obviously more than battleeld
victory: they constitute unconditional surrender. Such surrender is of
course well known in history but, in the nuclear era, pushing for uncon-
ditional surrender of an NBC-armed opponent has so far been seen as
impossible. During the Cold War, the realities of mutual assured destruc-
tion and the suicidal nature of waging total war ruled out any possibility
of seeking unconditional surrender by the Soviets in any possible nuclear
war. But as George Quester argues in this volume, in wars where assured
destruction of the United States is not a possibility, it is possible to
contemplate unconditional surrender by an enemy equipped with weap-
ons of mass destruction. The United States and the international commu-
nity should be able to pursue the elimination of regimes whose behavior
transgresses global norms and whose continued existence threatens both
its neighbors and those norms. They have done so previously in defeating
fascism.
To be sure, a skillful aggressor would attempt to steer conicts in
ways that minimize U.S. advantages and exploit U.S. vulnerabilities. But
if the foregoing review of likely war outcomes is on the mark, the
aggressor seems relatively unlikely to be able to force outcomes on the
United States that are distinctly contrary to core U.S. national interests.
Outcomes involving some compromise by the United States appear pos-
sible, but those inicting major defeat on the United States are remote
possibilities.
Winning the War versus Winning the Peace
The argument to this point implies that successful war termination by the
United States requires the exploitation of U.S. advantages in power and
leverage. But this is only a part of the puzzle. Fred Ikl again offers a
useful perspective, noting
the intellectual difculty of connecting military plans with their ultimate
purpose. Battles and campaigns are amenable to analysis as rather self-
contained contests of military power and, to some extent, are predictable on
the strength of rigorous calculation. By contrast, the nal outcome of wars
depends on a much wider range of factors, many of them highly elusive . . . .
In part, governments tend to lose sight of the ending of wars and the nations
interests that lie beyond it, precisely because ghting a war is an effort of
266 the coming crisis
such vast magnitude. Thus it can happen that military men, while skillfully
planning their intricate operations and coordinating complicated maneuvers,
remain curiously blind in failing to perceive that it is the outcome of the war,
not the outcome of the campaigns within it, that determines how well their
plans serve the nations interests.
22
There are important differences between defeating enemies and win-
ning wars. The latter requires not just battleeld victory but a winning
of the peace that follows. Especially in wars involving weapons of
mass destruction, this seems inescapable. As Herman Kahn noted dec-
ades ago, even in a war forced upon it by a malevolent and aggressive
regime, the United States will be held to a higher political standard
than the aggressor.
23
After any war involving weapons of mass destruc-
tion, the United States would likely nd itself subjected to a type of
scrutiny the aggressor would not. This is the price of its status as the
more powerful actor with the stronger obligationand historical claim
to acting in defense of common interests and to using force only by
and for agreed principles. As Kahn argued, if the aggression is not pun-
ished, if the level of destruction is seen as disproportionate to the stake,
if the coalition disintegrates, or if the United States is seen to have
acted rashly and in anger, the United States will be blamedwhatever
happens to the aggressor. Would U.S. actions be seen as both wise and
just by the international communityand by the U.S. public? If the
United States fails on either score, it could lose the peace that follows the
war.
Kahn could not have anticipated the particular historical epoch in
which we now consider this argument. Writing in the 1950s, he was
describing a United States leading a new alliance of nations against an
aggressive and nuclearizing Soviet Union. But bipolarity has given way
to unipolarityto the moment in history following the disintegration of
the Soviet Union when the United States nds itself cast as the worlds
premier military power, as the only power capable of mobilizing major
international institutions such as the United Nations for peace and secu-
rity purposes, yet somehow less than the superpower it once was with
the emergence of other centers and sources of power on the world
scene.
24
Today, some fear a United States whose power is unfettered by
22. Ikl, Every War Must End, pp. 12.
23. Fox, The Causes of Peace and Conditions of War, p. 10.
24. Charles Krauthammer has written of the United States in its unipolar moment,
in which its power is unchallenged but its world role is uncertain. Krauthammer, The
Unipolar Moment, Foreign Affairs, Vol. 70, No. 1 (1991), pp. 2333.
chapter 9 267
that of a rival peer.
25
Others fear a United States that will give in to its
historic temptations of isolationism and withdraw from the world stage,
to reappear impulsively and unreliably, if at all. Still others look to the
United States as the only country capable of leading the international
community and sustaining existing world order institutions.
Thus to win the peace after a major theater NBC war, Washington
must ensure that the record of that conict sends the proper signals about
the United States. Put succinctly, it will have to make choices and shape
outcomes in ways that avoid the casting of the United States as a nuclear
bully or as a cowardly wimpand that succeed in casting it as a respon-
sible steward of a just world order.
The United States would be cast as a nuclear bully if it were seen to
have stumbled unwittingly into nuclear confrontation, allowed itself to
be overcome by rage and fear, and then to have used its weapons impul-
sively and in ways that look heavy-handed or imperious, perhaps impos-
ing destruction later seen as disproportionate to the issues at stake. The
unipolar moment heightens the risk that U.S. nuclear use in a regional
war might be interpreted this way, whatever the aggressiveness and
malevolence of the offending state. Those who believe that the United
States will grow increasingly arrogant in its use of power and capricious
in its world role would interpret any U.S. nuclear use that seemed unduly
harsh or overbearing as conrming their expectation. A reputation as a
nuclear bully would certainly erode the willingness of other states to side
with the United States on international political issues or security con-
cerns. The United States would have to expect a diminution of its stand-
ing and a relative decline of its power as other states, whether friendly
or not, seek to arm themselves with their own strategic weapons.
Legitimacy won or lost abroad would likely have an impact at home
as well, given the U.S. publics sensitivity to how others judge national
behavior. However just or necessary U.S. nuclear use might be seen in
the White House, the event would likely unleash uid, chaotic, and
contradictory emotions at homefrom righteousness to shame, from
relief to guilt. The result could be highly unsettling within the U.S. body
politic (and damaging to the presidents political prospects). The process
of coming to terms with U.S. actions could well occupy the nation for
years to come, as it has after the Vietnam war.
If the United States comes to be seen abroad and domestically as a
nuclear bully, and if U.S. power is delegitimized, the results may prove
far-reaching. Herman Kahn anticipated a deep public revulsion toward
25. For a discussion of this perspective, see Shahram Chubin, The South and the
New World Order, Washington Quarterly, Vol. 16, No. 4 (Autumn 1993), pp. 87108.
268 the coming crisis
old ways of doing business in the wake of superpower thermonuclear
exchange and a surge of opinion that such wars should never again be
allowed to happen; he believed this would lead to strong pressures for
world government.
26
A limited nuclear war might produce a similar
revulsion but different political impulses. Abroad, U.S. leadership might
meet growing resistance, as all things U.S., from the military to business
and cultural institutions, fall into disfavor. At home, the impulse to
isolationism might rebound. At the very least, if the United States were
to come to be seen as a bully, strong pressures would likely build to
counterbalance or roll back U.S. military advantages.
It would be ironic indeed if in prosecuting a war against a rogue
state, the United States were itself to be seen as a belligerent nation, as
a state feared because of its nuclear weapons and seemingly driven by a
messianic worldview colored by self-delusion and ambition. This out-
come seems highly unlikely, but its very possibility is a cautionary re-
minder that U.S. power and inuence rest as muchif not moreon its
reputation as a benign power as on its singular military status.
What about the obverse of this problem? How might the United
States come to be seen as cowardly wimp? Such a perception might result
if the United States were seen to have refrained from using nuclear
weapons for fear of a domestic or international political backlash, and
therefore compromised a signicant national or international interest to
appease, in the most venal sense, the ambitions of the aggressor. The
possibility that U.S. non-use would be perceived in this way has arguably
increased; the passing of the Soviet threat has led some to expect that the
United States will again retreat from the world stage. U.S. non-use in a
situation where nuclear use was widely seen as necessary and legitimate
could fuel this expectation.
The cowardly wimp reputation would also have far-reaching nega-
tive consequences among allies, adversaries, and citizens of the United
States. Among allies it would raise basic questions about the value of U.S.
security guarantees and the reliability of U.S. power in defense of an
array of security, political, and economic interests. Among adversaries,
the most aggressive would likely be emboldened to test the new limits
of U.S. will and endurance. And among the body politic, a political
backlash against the prophets of decline would surely result.
Between bully and wimp lies a large middle ground where, it seems
likely, U.S. leaders would seek to steer the nation in time of crisis. To
demonstrate that the United States remains a nation capable of using its
powerand capable of doing so well and for the collective goodthe
26. Herman Kahn, Issues of Thermonuclear War Termination, pp. 133180.
chapter 9 269
United States is likely to make choices seen and defended in terms of its
stewardship of that collective good.
This sense of stewardship might shape U.S. choices in a number of
ways. If its use of power is to be seen as responsible, the United States
must show exibility in managing crises by making every effort to avoid
especially costly outcomes. But it must also press for a resolution of the
root causes underlying the conict and a long-term removal of the threat
if its use of power is also to be seen as purposeful. If it opts to use nuclear
weapons, that use must appear as necessary to end the war, proportionate
to the issues at stake, and just in terms of lives saved. It must act with
the understanding that the record of how the conict began and how it
played out will be scrutinized, even after a victorious war, to assess how
well the United States behaved as leader and guarantor and how well it
offered its opponents an acceptable way to avert NBC confrontation.
Articulating U.S. choices and building a case for U.S. actions would
require that the president help the nation and the world to interpret the
crisis and to understand the military, political, and moral context of U.S.
choices. Establishing a tenable connection between means and ends, and
building some measure of consensus with Capitol Hill, the media, and
the U.S. public (and perhaps even within the executive branch) on the
chosen course of action would likely be a principal challenge for the
president and his or her advisersespecially if that course is nuclear use.
Many of the same tasks would be necessary with friends and allies abroad
and in major international institutions such as the United Nations and
North Atlantic Treaty Organization (NATO). The Cold War debate about
nuclear weapons revealed a deep division in U.S. society and internation-
ally about nuclear weapons, with a strong body of opinion that their use
would be inconsistent with just war traditions. With so little discussion
until now of the role of nuclear weapons in the postCold War world,
the president would likely face a major challenge in seeking to establish
U.S. nuclear retaliation as both necessary and just.
27
This is a political campaign that cannot be cobbled together at short
notice. The United States cannot answer at the last minuteor alonethe
questions of power and legitimacy that WMD crises will inevitably pose.
Finding the best decisions, actions, and words at the moment of crisis is
no substitute for those other initiatives that are always essential to main-
taining the U.S. reputation as a reliable and benign power and responsible
27. For a discussion of some of these themes, see Brad Roberts, NBC-Armed Rogues:
Is There a Moral Case for Preemption? in Elliot Abrams, ed., Close Calls: Intervention,
Terrorism, Missile Defense, and Just War Today (Washington, D.C.: Ethics and Public
Policy Center, 1998), pp. 83108.
270 the coming crisis
steward of common interests. That is, the political requirements of future
major theater NBC wars oblige U.S. leaders to attend in the present to
the foundations of U.S. power, to U.S. credibility as a great power, to its
responsibilities for initiative and leadership, and to the challenge of
adapting to a role of rst-among-equals even at a time of military su-
premacyall tasks that garner slight interest in an inwardly focused
nation.
What then is successful war termination? Success means winning the
war in a way that answers basic questions about the United States in
ways the country would prefer. Victory requires both restraint and the
purposeful use of power. It requires recognition now that to decide how
a war must end is also to decide the nature of the peace that succeeds it,
and that winning the peace after a war in which NBC weapons have been
used by one or both sides may well prove more difcult than winning
the war. Too little thought has been given to the negative consequences
for the peace of winning such wars the wrong way.
What are the lessons for the three schools of thought noted above?
For those who believe that the aggressors NBC assets can trump any
U.S. action, it is important to recognize the interests created for the United
States if it opts to back down in the face of coercive threats. An aggres-
sors NBC assets will certainly generate pressures on the United States to
compromisebut successful war termination requires the purposeful use
of U.S. power. This implies that the United States must not allow itself to
be deterred or dissuaded by an aggressors NBC threats and that nding
long-term solutions that remove the necessity of coexisting with NBC-
armed renegade states will be essential.
For those who believe that America has all the trump cards, asym-
metries in power will certainly imply that wars will end as the United
States wants them to. But winning the long-term peace requires that the
United States not use overwhelming force.
For those who believe that its the presidents job to end such wars,
it seems likely that the president would be very sensitive to the political
result of the war (and to his or her own place in history). The president
is very likely to see the imperative of steering the nation between the
poles of bully and wimp on a course of stewardship. The president is
unlikely to perceive a clear demarcation between political and military
interests and thus is likely to require that military advisers overcome the
intellectual difculty of connecting military plans with their ultimate
purpose observed by Ikl. In particular, the military should be prepared
to endow the president with the options to do the right thing in a regional
crisis, and to ensure its own ability to operate within likely political
choices and limits set by the White House.
chapter 9 271
No tool of social sciences enables us to predict the future. We cannot
know today how a future regional NBC war will end. But we can know
how it might end. We can certainly know how it should end. It might end
with clear-cut victory for one or the other side, but it is more likely to
end after a large cost has been paid by one or both sides to protect basic
interests. It should end in a way that secures the peace that follows by
teaching the right lessons about the utility of NBC weapons and the
viability of a U.S.-led international security orderamong many other
things. From the U.S. point of view, such wars must end in ways that
protect key regional interests while also sending the right political signals
about the United States as a reliable security guarantor and benign power.
Policy Implications
The outcomes of wars are not dictated by certain inevitable forces of
technology, history, or personality. This line of argument puts a large
burden on policymakers in Washington and allied capitals to shift the
balance of capabilities and perceptions so that when and if war comes
they will be able to do what seems right and necessary at reasonable cost.
One way to approach the policy problem is to focus on the seven factors
enumerated above, which are likely to shape the U.S. choice about how
to respond to an aggressors use of NBC weapons. On each of the seven,
policies available today can improve the U.S. position in a future conict.
The international context can be shaped ahead of time by working
with regional allies on political-military strategies to meet the challenges
of regional aggressors; with the other members of the UN Security Coun-
cil to bolster the councils role in replying to NBC aggression; and with
other prospective members of international coalitions that might be
brought together in response to aggression.
28
Congressional perspectives can be shaped through an executive-led
dialogue with key committees and chairpersons on issues of regional and
international security, proliferation concerns, and U.S. foreign and de-
fense policy. It could also be productive to reach out to constituencies
with which the Congress interacts on such matters, such as the think tank
and academic communities.
The public mood will be shaped largely by events of the moment,
but setting out some benchmarks in U.S. declaratory policy and articu-
lating U.S. interests and policies to constituencies in academe and the
media could pay dividends in time of crisis.
28. Roberts and Utgoff, Coalitions Against NBC-Armed Aggressors.
272 the coming crisis
The U.S. strategic stake can be kept well focused and well understood
internationally by a foreign policy that clearly denes U.S. national inter-
ests and foreign policy priorities, and engages both allies and potential
adversaries in a dialogue on U.S. perspectives.
Limiting the level and character of destruction of an initial NBC
attack requires various approaches. One is working with the probable
regional targets of an aggressors NBC weapons to put in place both
active and passive defenses. Another is undertaking international threat
reduction measures, including the effective implementation of arms con-
trol and nonproliferation mechanisms so that they help constrain the
future evolution of the regional NBC threat.
Limiting the capacity for continued destruction is also a task with
many parts. One is the development of a better understanding of the
strategic personalities of aggressive regimes, so that their ambitions,
values, and decision mechanisms can be understood and engaged in time
of crisis. Caroline Ziemkes contribution to this volume offers one such
effort. This is an intelligence problem as much as a problem of military
planning. Another part is the elding of active and passive defenses by
the U.S. military, for the protection of its own forces and as a supplement
to the efforts of regional allies.
Preparing the military voice to participate in dialogue at the White
House in times of crisis requires careful thought about how WMD wars
might unfold, the likely perception of interests both in Washington and
in the aggressors capital, and a readiness to articulate how operational
plans can be shaped to help the president achieve acceptable political
outcomes.
This list of policies is entirely illustrative; it is not an exhaustive
review of the tools available to U.S. policymakers to shape war outcomes.
Rather, it illustrates how much the United States can do ahead of time to
shape the context in which an NBC crisis will play out, and thus to
promote outcomes conducive to U.S. interests, and that it would be
difcult or unhelpful to try to do some or all of these things at the very
last minute.
Many of the tasks fall under the rubric of the defense counterprolif-
eration initiative.
29
The degree of vulnerability of U.S. military forces to
an aggressors NBC attacks would be a critical determinant of the risk
29. This initiative was announced by Secretary of Defense Les Aspin in a speech to
the National Academy of Sciences on December 7, 1993. Its purposes are to prevent
the proliferation of weapons of mass destruction and roll it back where possible, to
deter the use of such weapons, and to change how the United States ghts wars in
chapter 9 273
calculus on both sides. A less vulnerable U.S. force would open up more
options for the United States, while a more vulnerable one opens up more
options for the aggressor. Indeed, the United States could be compelled
to act as a wimp if gaps in U.S. capabilities help the aggressor to ght a
war largely on its terms and to its strengths. Similarly, it could be com-
pelled to act as a bully if it is left with no more effective option than
strategic nuclear attack to meet the aggressors use of NBC weapons.
Moreover, counterproliferation initiatives jointly pursued with regional
allies are the basis of joint action and help to build the political condence
in each other that would be necessary to weather crises. If the United
States is capable of sustaining military operations on a battleeld con-
taminated with chemical and biological weapons, of protecting rear areas
and population centers from NBC attack, and of conducting exible
nuclear operations, U.S. action is not foreclosed, and the aggressor may
well conclude that an NBC war against the United States or a regional
U.S. ally is unwinnableand thus should not be risked. Thus a successful
counterproliferation initiative can change the relative prices of alternative
outcomes in ways that make it easier for the United States to do what is
right and necessary, and harder for the aggressor to begin an NBC war.
Two more specic policy-related questions also follow from this
analysis. What choices in war best secure the U.S. interest in preventing
the further proliferation of nuclear weapons? And what does this analysis
imply about the problem of deterrence?
What is best for the future of the nuclear nonproliferation effort is
not obvious. Desert Storm has left a strong conviction that the cause of
nuclear nonproliferation is best served by prosecuting wars against NBC-
armed adversaries with only conventional weapons. After an aggressor
uses weapons of mass destruction, however, convictions may change.
This seems especially likely to be so if nuclear retaliation is the only
means to inict commensurate punishment on that aggressor or to deny
its war aims at costs in human terms acceptable to the victims of aggres-
sion and to those called upon to reverse it. Moreover, whether or not
nuclear weapons are used by the United States seems likely to prove a
less signicant determinant of the nuclear future than whether other
states come to view the United States as a bully, wimp, or steward.
How should the United States seek to deter an NBC-armed aggres-
sor? This chapter has illustrated a variety of ways that deterrence might
not work as hoped, as an aggressors risk calculus downplays or disre-
the presence of NBC threats so that it can achieve its war aims despite that presence.
See Proliferation: Threat and Response (Washington, D.C.: Ofce of the Secretary of
Defense, November 1997).
274 the coming crisis
gards U.S. capabilities or declaratory policies. It demonstrates that there
is benet to be reaped in rising above deterrence to assemble the
capability to successfully prosecute wars against NBC-armed regional
adversaries without merely relying on the dissuasive effects of U.S. con-
ventional and nuclear superiority.
30
But it also points to some specic
lessons for specic situations.
In deterring an aggressors rst use of NBC weapons, perceptions of
U.S. will are likely to affect the aggressors choice more than its percep-
tions of U.S. capabilities. The United States could strengthen deterrence
of NBC rst use by an aggressor by working to dispel the notion that
retreats from conict, as in Somalia and Lebanon, imply that the United
States will back down in a conict over a major interest. Developing a
better understanding of the U.S. publics sensitivity to casualties could
also contribute to deterrence. More generally, an aggressors expectations
about how U.S. leaders will behave in crisis is shaped fundamentally by
its reading of U.S. culture and politics. An aggressor that perceives a
complacent and vulnerable United States may stimulate a crisis and then
initiate NBC use in the hope of reaping local gains. One that perceives a
nervous but unprepared United States may opt to strike before U.S.
advantages are secured. One that sees the U.S. president as timid or
foolish, or the U.S. body politic as divided and manipulable, may seek
confrontation in the hope of dealing a sharp blow to U.S. prestige and
inuence. But a potential aggressor that has been led through how the
United States would assess its alternative courses of action in an unfold-
ing crisis may conclude that any route that might lead to confrontation
with the United States is suicidal. Thus, deterrence of rst-use would be
strengthened by clarication of the interests that would be created or
magnied for the United States by any act of regional aggression backed
by NBC weapons.
Deterring an aggressors second use of WMD in reply to U.S. retali-
ation entails a different set of problems. The task of coercing leaders who
do not hold their populations dear poses difcult nuclear use questions
for the United States. Countervalue strategies seem to offer less promise
than counterforce oneswhich is to say that the United States seems
likely to use nuclear weapons not to coerce but to defeat. The United
States may nd this stage of conict ripe for trying to separate a war-
making leader from his or her military and societal sources of support,
as noted above. Deterrence will be well served by conveying the capacity
of the United States (and its coalition partners) to survive the worst blows
that the aggressor can contemplate delivering and the reality that every
30. Utgoff and Wallis, Rising Above Deterrence.
chapter 9 275
step up the escalation ladder gives the United States more room to
maneuver while worsening the aggressors own situation.
Conclusions
The variety of policy directions suggested above demonstrates that the
United States can indeed take actions ahead of time that will inuence
how a war fought with NBC weapons will end. But these actions require
a strategic view of the problem well beyond the current conventional
wisdoms that clutter the debate. No longer a problem that is truly
unthinkable in a war that cannot be won and therefore must not be
fought, such wars require serious analytical scrutiny in an effort to
prevent them or minimize their costs. The United States should plan for
the termination phase of such wars by consulting with allies and prospec-
tive coalition partners on the kinds of interests likely to shape war-time
coalition dynamics, and working backwards from that to develop the
right military capabilities. It should attempt to shape the termination
phase of such wars by limiting the vulnerabilities that an aggressor might
target with NBC weapons, by dispelling myths about the strategic per-
sonality of the United States, and by putting together a posture of military
force and declaratory policy suited to the deterrence requirements of
specic operational theaters. Our preference to focus on how to secure
the nuclear peace does not relieve us of the obligation to think through
these choices about nuclear warindeed, the two tasks are integral.
What does this analysis of the war termination problem imply for the
larger questions raised in this volume? First, it is another reminder of the
way in which wishful thinking, conventional wisdom, and political fash-
ion narrow the necessary debate. The wishful thinking is that major war,
especially war that threatens the United States with mass casualties, is an
anachronism. The conventional wisdom is that nuclear wars will not be
fought because they cannot be won. The political fashion dictates that
problems of nuclear war are somehow beyond the bounds of polite and
proper discourse. Were it only so. Wars employing nuclear, biological, or
chemical weapons and involving U.S. forces are more likely today than
in decades past, given the growing prevalence of asymmetric counters to
superior U.S. conventional military power in precisely those regions
where the United States offers security guarantees. These issues belong
on the policy agendahowever unpopular or seemingly obscure.
Second, this analysis reinforces a theme in many of the chapters in
this volumethat the rst major regional war with NBC weapons, if it
occurs, will be a dening event. At the very least, ideas based on pre-war
perceptions, especially those carried over from the Cold War, seem likely
276 the coming crisis
to be driven aside by new concerns. Perceptions of the utility of weapons
of mass destruction, of the nature of the retaliatory problem, and of the
international role of the United States would be catalyzed by events.
Indeed, any confrontation seems likely to be in large part a political effort
by the combatants to shape those perceptions. Mishandled, it would cast
the United States as either bully or wimp, thus weakening the founda-
tions of world order, by teaching states that they must organize either
against or without the United States. Handled well, it would solidify U.S.
leadership, conrm its stewardship of common interests, and deter future
regional challengers.
chapter 9 277
Chapter 10
The Coming Crisis: Nuclear
Proliferation, U.S. Interests,
and World OrderA Combined
Perspective
Victor A. Utgoff
This collection of essays has addressed a fundamental question: What
changes might the proliferation of nuclear, biological, and chemical (NBC)
weapons create in the long run for international relations? While only
time will answer this question completely, the importance of anticipating
the possibilities grows with each new NBC-armed state.
Our main concern with this question is that proliferation will even-
tually lead to a crisis in which an NBC-armed state tries to force the
United States to sacrice a vital interest. As Rosen, Posen, Quester, and
Roberts all argue, how the United States and its allies act in the face of
such a challenge will say a great deal about the likely nature of the
subsequent world order. Wisdom and resolve could preserve the pros-
pects for a reasonably safe and improving world order; their absence
could turn the world toward further NBC-backed aggression, accelerated
proliferation, and the breakdown of important forms of international
cooperation. This book has sought to present insights that can help to
foster the needed wisdom and resolve.
Addressing the question of how NBC weapons may affect the world
order requires exploration of an enormous expanse of intellectual terrain,
much of which remains terra incognita. As one would expect from sea-
soned explorers, the trip reports presented by the authors in this volume
are lled with descriptions of important and interesting features of this
terrain. Equally interesting are the very different perspectives through
which they viewed the terrain. Most explored primarily with the eyes of
realists, concerning themselves with relative balances of power among
states involved with NBC weapons. One viewed the terrain in terms of
strategic personality, a nations historically derived sense of how it should
interact with other nations. Others saw the terrain through mixed per-
spectives including those of realism, international norms, bureaucratic
interests, and the attractions of international prestige.
What can be learned from these trip reports? The most important
message is that the conventional wisdomwhich posits that the contin-
ued proliferation of weapons of mass destruction will cause the United
States to become a far more reluctant sheriff in the international arena
than it has been in the pastseems wrong.
1
While the essays do provide
some arguments that are consistent with this view, the larger weight of
their arguments points in the opposite direction. A second important
message is that most of the unconventional wisdom presented by the
authors arises from perspectives other than realism. This raises an impor-
tant question: Should realism continue to play so strong a role in assess-
ments of likely U.S. international behavior as it has in the past?
I believe it should not. Throughout its history, the United States has
disdained Europes historical emphasis on balance of power international
politics. While U.S. international behavior during the Cold War was
strongly conditioned by concerns about the enormous military power of
the Soviet Union, the collapse of that empire has largely freed the United
States to revert to type. The United States is the worlds strongest power.
It has far more capability than ever before to exert its will around the
globe, and thus to preserve and improve the aspects of the global order
that it believes are right. Accordingly, projections of future U.S. behavior
in the international arena should now focus more attention on what this
nation sees as right behavior in any particular circumstance.
2
How can we understand what kind of behavior the United States is
disposed to see as right?
3
One way is to adopt the approach taken by
Caroline Ziemke in her essay on Iran and construct a picture of the U.S.
collective view or strategic personality. Such a construct must necessar-
ily include a concept of the nations most important values and how those
values enter into its perceptions and decision-making.
This chapter summarizes the main features of the strategic personal-
ity type the United States shares with several other nations. It describes
1. As will be argued below, the United States does not seem that reluctant a sheriff.
See Richard N. Haass, The Reluctant Sheriff: The United States after the Cold War (Council
on Foreign Relations, 1997).
2. Kissinger argues that without the threat of the Soviet Union to leaven its Wilsonian
idealism, the United States needs to pay more conscious attention to balance of power
considerations in framing its postCold War foreign policy. This chapter argues in
favor of understanding and being more conscious of the role of the U.S. value-oriented
strategic personality in motivating key foreign policy choices. Henry Kissinger, Diplo-
macy (New York: Simon & Schuster, 1994).
3. Unfortunately, neither the theory nor the practice of political policy analysis has
yet developed rigorous tools for combining realism with other perspectives. See
Gregory A. Raymond, Problems and Prospects in the Study of International Norms,
Merschon International Studies Review, No. 41 (1997), pp. 205245.
280 the coming crisis
how that personality type is manifested by the United States, emphasiz-
ing characteristics that will inuence its behavior in the face of challenges
involving NBC weapons. It then explores how this strategic personality
seems likely to reinforce or counterbalance other more commonly consid-
ered factors that would inuence the outcome of a confrontation between
the United States and an NBC-armed regional aggressor. And, nally, it
draws some larger policy implications from this analysis.
Characterizing the U.S. Strategic Personality
Caroline Ziemke argues that a states strategic personality describes its
distinctive ways of thinking, and how its ways of thinking inuence its
international conduct.
4
She argues that the strategic personality of a state
reects its entire history of adapting to its physical surroundings, main-
taining the internal stability of its society, and meeting the challenges that
come from interactions with other states. Thus, the strategic personalities
of long-established states are very stable.
Ziemke divides strategic personalities into types based on three char-
acteristics. The rst characteristic is how a state views interactions with
others. The second is what information it selects and remembers from the
innity of events that constantly inundate it. The third is the basis upon
which it selects courses of action. Ziemke shows that the U.S. strategic
personality type is extroverted, intuitive, and feeling.
5
Extroverted states extol individualism and competition as the engine
for domestic progress. Similarly, they see interactions of their state with
others as desirable for themselves and for the international system as a
whole. Extroverted states view history as an irreversible march toward
some common progressive end goal for all humankind. They believe that
they must move forward, grow, and change, or that they will stagnate
and be overtaken by the forces of progress in history.
In their interactions with other states, extroverted states commonly
challenge, push, and prod until they encounter resistance or get pushed
back. That is, they have a weak sense of boundaries, which is perhaps
the most consistent source of international conict for such states. The
social structures of extroverted states call on individuals to adapt their
needs and ambitions to those of the others with whom they interact. This
4. Caroline F. Ziemke, The Strategic Personality Types: Using National Myth to Project
State Conduct, forthcoming.
5. Ziemke identies Portugal and the Netherlands as having the same strategic
personality type as the United States. Ziemke, Strategic Personality Types. Clearly any
specic personality type can manifest itself in a wide variety of ways.
chapter 10 281
internal social norm translates into the expectation that the international
system also should employ negotiation and compromise rather than
violence to resolve interstate differences. Over the course of their histo-
ries, the extroverted states have arrived at a consensus that war is not the
way to settle their differences. Finally, extroverted states see themselves
as relatively adaptable and can fail to appreciate that not all states are.
They care very much about how the rest of the international system sees
them.
Intuitive states tend to have national myths dening their origins and
ultimate goals in terms of a particular idea or doctrine. They usually see
themselves as working toward some future golden age and their national
myths often contain a promised land, paradise, or utopia. Accordingly,
they pay most attention to the patterns and meanings in unfolding events
that seem to provide insights into whether the future is evolving as
desired. These states dene evil as any force that stands in the way of
achieving their ideals. For intuitive states, patriotic rituals are important
for reinforcing the guiding ideal. Residence and citizenship requirements
are usually dened in terms of acceptance of the guiding ideal. Physical
boundaries are less important than ideological ones.
Because they seem to have an instinctive sense of where events are
headed, extroverted intuitive states are natural leaders. They are particu-
larly effective in promoting new approaches to addressing problems,
inspiring enthusiasm, and drawing other states to follow their vision,
even if their followers do not completely grasp or accept it.
Feeling states judge alternative conclusions or courses of action in
terms of their consistency with a hierarchy of national values. While logic
enters their assessments, they are suspicious of it as a guide to conduct.
Thus, where logic and values conict, feeling states are more likely to act
in accordance with their values.
6
When extroverted feeling states are forced by circumstance to follow
the dictates of logic or realism rather than their values, they feel uncom-
fortable, and can be expected to change their positions if an opening
presents itself. With respect to the international system, the extroverted
feeling state is guided more by values dening how the system should
work than by any logical picture of how the system does work. While
the value structures of extroverted feeling states are rooted in both their
6. Huntington notes that throughout the history of the United States a broad
consensus has existed among the American people in support of liberal, democratic,
individualistic and egalitarian values. He also states that being human, Americans
have never been able to live up to their ideals; being Americans, they have been unable
to abandon them. In Samuel P. Huntington, American Ideals versus American
Institutions, Political Science Quarterly, Vol. 97, No. 1 (Spring 1982).
282 the coming crisis
basic culture and history, they are constantly adapted to their experience
in interacting with others and the implications of that experience for their
values. This constant evolution in values and especially their context-
dependent application can frequently make extroverted feeling states
seem ckle and sometimes hypocritical.
For feeling states, righteousness comes from conducting ones affairs
virtuously and from good works. The legal codes in feeling states are
generally designed to promote and enforce a particular hierarchy of
values, and penal codes often go beyond mere justice and seek rehabili-
tation or retribution. Government in the feeling states is often primarily
a caretaker of the cherished values and arbiter of differences. For the
feeling state, being admired and recognized internationally for its virtue
is important to its sense of national worth and strength.
u.s. manifestation of its strategic personality type
The United States manifests its extroverted, intuitive, feeling personality
type in specic ways that say a great deal about how it would be disposed
to act toward aggression backed with NBC weapons.
7
First, the fundamental shared ideal upon which the United States as
an intuitive feeling state is based is liberty for the individual. This pre-
eminent value can explain most of the signicant political-economic fea-
tures of the United States. It is the basis of U.S. law. It explains the U.S.
emphasis on negotiation and compromise as the only legitimate means
for resolving disputes. Its best protection is democratic government. It
leads directly to a free market economy and rigorous protection of prop-
erty rights.
The key theme of U.S. domestic history has been the pursuit of a
more perfect liberty. Most of the larger initial aws in how liberty for the
individual was dened, practiced, and protected have been corrected.
Slavery ended in the Civil War, initial restrictions on suffrage have been
largely eliminated, and the threats to liberty that came with unregulated
capitalism have been contained.
Most important, individual liberty is seen as the fundamental key to
the enormous success of the United States as a society. Liberty opens great
opportunities to create wealth, institutions, and other values, and im-
poses primary responsibility on the individual for determining his or her
7. This section was informed by: Walter A. McDougall, Promised Land, Crusader State:
The American Encounter with the World Since 1997 (Boston: Houghton Mifin, 1997);
Alan Brinkley, The Unnished Nation: A Concise History of the American People, 2nd ed.
(New York: Knopf, 1997); George Wilson Pierson, Tocqueville in America (Baltimore,
Md.: Johns Hopkins University Press, 1938); and Richard Slotkin, Regeneration through
Violence: The Mythology of the American Frontier, 16001860 (New York: Harper, 1972).
chapter 10 283
own fate and meaning. This combination of opportunity and responsibil-
ity is seen as the internal engine for continuous U.S. innovation and
growth. Not surprisingly, the United States sees itself in the lead on
humankinds long march toward the perfect society.
Religious liberty for the individual is a particularly important com-
ponent of the U.S. strategic personality. It places greater responsibility on
the individual to decide the larger questions of right and wrong than felt
by individuals within states with an ofcial religion. While a majority of
U.S. citizens may accept the teachings of the New Testament as an ideal,
they seem most often to practice the eye-for-an-eye philosophy of the
Old. Indeed, most historical U.S. heroes did not turn the other cheek
when insulted or assaulted, but gave back as good as they got. Justice
and the need for revenge usually require commensurate punishment in
U.S. society.
As an extroverted state, the United States is very active in the inter-
national arena. It sees interaction and competition among states as im-
portant for healthy adaptation and growth for itself and for others. More
generally, the United States has strong views about how the international
system of states should worka guiding vision of an international City
on the Hill. In particular, as an extroverted state, the United States wants
the world system to mirror its own main characteristics.
8
For example,
the world economic system should have open global markets. The rule
of law should be the basis for global political stability rather than balances
of power among nations. Negotiation and compromise should be the
means by which states settle their differences. And the world should be
composed of democracies that cherish liberty and human rights.
9
For many decades the United States has been active in creating and
participating in numerous institutions to support such a world. Providing
for international security came rst, with the United Nations, NATO, the
Conference on Security and Cooperation in Europe, and a variety of other
types of security regimes. The search for global economic security has led
to institutions such as the World Trade Organization, the World Bank, the
8. Ruggie argues that there is a certain congruence between the vision of the world
order invoked by American leaders when founding a new international order has
been at stake, and the principles of domestic order at play in Americas understanding
of its own founding, in its own sense of political community. John Gerard Ruggie,
The Past as Prologue? Interests, Identity, and American Foreign Policy, International
Security, Vol. 21, No. 4 (Spring 1992). See also David C. Hendrickson, In Our Own
Image: The Sources of American Conduct in World Affairs, The National Interest
(Winter 1997/98).
9. Tony Smith, Americas Mission: The United States and the Worldwide Struggle for
Democracy in the Twentieth Century (Princeton, N.J.: Princeton University Press, 1994).
284 the coming crisis
International Monetary Fund, and the Group of Eight. And, reecting the
extroverts minimal regard for national and cultural boundaries, the
United States has pursued an international campaign for human rights
that has made many nations extremely uncomfortable.
Perhaps most important, the United States has been exporting
and defending democracy abroad for the past hundred years.
10
For ex-
ample, when the United States became a colonial power with the Spanish-
American War of 1898, President William McKinley set as U.S. goals to
establish liberty, justice and good government in our new possessions.
And, after proposing a postwar League of Nations to maintain the peace,
President Woodrow Wilson took the United States into World War I to
make the world safe for democracy. President Franklin Roosevelt took
the United States into World War II and led the effort to defeat and then
make democracies of the totalitarian regimes that began it. And for more
than forty years after World War II, the United States led an international
campaign to contain the global challenge to democracy mounted by the
Soviet Union. That this campaign ended without another global war, but
with freedom and budding democracy for many of the states that had
been part of the Soviet Unions empire and for Russia itself, is an enor-
mous source of satisfaction for the United States and its allies.
Since the end of the Cold War, the defense and enlargement of the
community of democracies has been a stated goal of the United States.
The United States has carried out at least three military interventions
abroad to protect, restore, or establish democracy in the postCold War
era.
11
And the Clinton administration supports the expansion of NATO
to assure new allies that it will protect their newly won freedom and
democracy.
Looking back, the United States takes pride in the results of its
campaign to transform the world into a closer image of itself. To be sure,
it has passed up a variety of possible opportunities to protect and expand
liberty and democracy abroad. Some efforts it did undertake failed mis-
erably. Still, U.S. support for liberty and democracy, most often simply by
providing an example of its potential advantages, has had a remarkably
positive effect on the world. Most of the nations, with a majority of the
global population, are democratic today, where only a few were a hun-
dred years ago. And the world seems a closer image of the United States
10. Quotes from McDougall, Promised Land, Crusader State.
11. The three interventions were in Somalia, Bosnia, and Haiti. The 199091 interven-
tion against Iraq might be considered a fourth such intervention. Although neither
Kuwait nor Saudi Arabia is a democracy, the purpose of the intervention was also to
protect the larger democratic community. Many other lesser interventions have taken
place in addition to these relatively high-prole interventions.
chapter 10 285
in other important ways. Free markets have spread widely. Much of the
world has adopted a technical culture similar to that seen in the United
States.
In sum, the preeminent value for the United States in its international
role has been the extension to the rest of the world of its own system of
liberty, democracy, free markets, and the rule of law. The United States
has campaigned for these values for more than a century at considerable
cost in treasure and blood, and the campaign appears to have been very
successful.
reliability of the u.s. strategic personality as a predictor of
u.s. behavior preferences
If the U.S. strategic personality is to be useful as a predictor of its
preferred behavior, at least two things must be true. First, the mechanisms
by which it makes itself felt must be reasonably reliable. Second, it must
be reasonably stable.
12
on reliable mechanisms The U.S. strategic personality makes its in-
uence felt on the nations decisions through communication, debate, and
leadership. The communications media, which constantly searches for
issues that will capture the publics attention, is drawn to issues that
trigger fundamental U.S. values. In turn, debate on such issues inevitably
surfaces a wide range of suggestions for how to resolve the issue. Those
that resonate best with U.S. values are the most likely to be selected.
Decisions that the nations leaders make in secret do not benet from
wide public debate, but a wise leadership will do its best to act as it
understands the larger body politic would want. Further, whatever inter-
pretations and actions are decided upon are likely to become public
before long. This also argues for decisions that will appeal to U.S. values.
12. How rapidly the U.S. strategic personality may be changing is disputed. Arthur
Schlesinger, Jr., argues that in its essential, the national character will be recognizable
much the same as it has been for a couple of centuries. People seeking clues to the
American Mystery will still read, and quote, Tocqueville. In Arthur Schlesinger, Jr.,
Has Democracy A Future? Foreign Affairs (SeptemberOctober 1997). Samuel Hunt-
ington has also argued that despite their seventeenth- and eighteenth-century origins,
American values and ideals have demonstrated tremendous persistence and resiliency
in the twentieth century. Dened vaguely and abstractly, these ideals have been
relatively easily adapted to the needs of successive generations. Samuel Huntington,
American Ideals versus American Institutions. William H. McNeill provides an
interesting essay on the state of the U.S. public myth (a concept closely related to
strategic personality) in The Care and Repair of Public Myth, in William H. McNeill,
Mythistory and Other Essays (Chicago: University of Chicago Press, 1986).
286 the coming crisis
The presidents leadership also tends to reinforce U.S. behavior that
is consistent with its strategic personality. Especially when risking war, a
president wants the broadest and deepest support possible, and the most
effective way to win it is to appeal to interpretations and values that are
consistent with the U.S. strategic personality. Presidents are also very
conscious of how they will be compared to their predecessors in the
futureand the greatest presidents have articulated appealing values for
the nation, led the country in heroic actions to further those values, and
thus helped to strengthen and shape the U.S. strategic personality.
Of course, a president who has great public credibility can occasion-
ally deviate from the interpretations and actions called for by the U.S.
strategic personality. However, doing so risks the quick evaporation of
public support if the chosen action does not go well. In contrast, actions
that are consistent with U.S. values may be supported long after they
have been shown to be unwise.
13
In sum, then, it seems likely that the
mechanisms that conform U.S. behavior to its strategic personality are
fairly reliable.
on the stability of the u.s. strategic personality. The most fun-
damental characteristics that make the United States an extroverted-
intuitive-feeling state, and the more basic ways in which the United
States manifests its personality type, change very slowly.
14
The U.S. per-
sonality is the product of over 200 years as an independent nation, and
nearly 400 years experience with the special challenges and opportunities
of life in the New World. Fundamental changes to the personality of such
a well-established state usually require comparable lengths of time to
evolve. Less fundamental changes can happen more quickly. However,
there are good reasons to believe that any such changes to the U.S.
strategic personality are evolving gracefully over a matter of many dec-
ades. And they may be offset in part by other trends.
First, the outside world does not seem to be the source of large
pressures for change in the U.S. strategic personality. The United States
has been and remains relatively successful in the international arena, and
many important states seem to depend upon U.S. global involvement.
Some observers worry that the United States will ruin its wide acceptance
13. As examples, President Dwight D. Eisenhowers credibility as a military leader
was very important in bringing the United States to understand and accept the notion
of limited war in the aftermath of World War II. The credibility of Presidents Kennedy,
Johnson, and Nixon worked to extend the U.S. involvement in Vietnam.
14. Ziemke, Strategic Personality Types.
chapter 10 287
and inuence by abusing its current preeminence of power. This seems
unlikely; reecting its basic extroverted nature, the United States seems
too sensitive to world opinion to let itself become a pariah, and it attaches
fundamental value to the concept of democracy among the states.
While internal change could also lead to a shift in the U.S. strategic
personality, this does not seem a signicant worry either. The values that
are the basis for the effective functioning of U.S. society are remarkably
few and simple, as indicated above. Most of the body politic takes them
for granted, and most groups that do not generally seem to be seeking
to have those values applied more benecially to them.
There is some worry that families, schools, churches, service organi-
zations, job environments, and the media generally are not inculcating
traditional U.S. values effectively. Troubling signs include a reduced sense
of responsibility, as reected in more out-of-wedlock births, and a re-
duced respect for solving problems without violence, as reected in
increased violence in public schools. At the same time, however, there are
good reasons to believe that the core U.S. values continue to be accepted
by the great majority of the public. Vastly improved communications
seem on balance to be a positive force for building and reinforcing
common values.
15
The public applauds traditional heroic behavior that
seems no less frequent in todays news than in decades past. The con-
tinuing demand and respect for traditional heroes seems evident in posi-
tive public reactions to well-attended motion pictures featuring the likes
of Bruce Willis, Tom Hanks, Sigourney Weaver, Jodie Foster and others
playing roles in which they perform impressive feats and make great
sacrices to protect the lives, freedom, dignity, and general welfare of
their fellow citizens.
Some observers argue that the United States is becoming more di-
vided along ethnic lines, particularly as a result of high rates of immigra-
tion from south of the border. While this immigration is having some
effect on U.S. culture, it seems doubtful that it will lead to substantial
changes in the larger U.S. strategic personality. The great majority of the
immigrants are attracted to the United States by features that are funda-
mental to this personality. Further, leaving home requires an intuitive
15. The role of communications in encouraging a sense of nationhood among large
numbers of people who know only a minute fraction of their fellow citizens is
discussed in Anderson Benedict, Imagined Communities: Reections on the Origin and
Spread of Nationalism (London: Verso, 1983). In particular, Benedict analyses the con-
nections between the appearance of inexpensive newspapers and modern nations. A
pessimistic assessment of the current state of some of the social mechanisms that
nourish U.S. democracy is provided by Robert D. Putnam, Social Capital and Democ-
racy, Braudel Papers, Fernand Braudel Institute of World Economics, No. 9 (1995).
288 the coming crisis
persons imagination and willingness to take risks, which biases immi-
gration toward those who can readily t into U.S. society. And, as with
all previous waves of immigrants, subsequent generations t in even
better.
Francis Fukuyama has recently suggested that the increasing number
of women in inuential positions in the United States may be leading
toward less aggressive U.S. international behavior.
16
He notes that this
could put the United States at a disadvantage in dealing with upstart
states that are led by more aggressive men. However, such change is
occurring very slowly, particularly in the U.S. national security apparatus.
Further, women who win senior positions do not seem signicantly less
competitive than their male counterparts, or likely to misunderstand the
imperatives of dealing effectively with serious external aggression.
Edward Luttwak has argued that the U.S. willingness to go to war
has been reduced because U.S. families usually have far fewer children
than they did decades ago, and parents with few offspring are less willing
to risk their loss to war.
17
On the other hand, the prospects of dying in
battle are far smaller than they were in past generations as a result of
increased investments in protecting the combatants and the improved
speed and quality of medical care. Perhaps equally important, volunteers
are more justiably sent into combat than draftees were.
Finally, yet another stabilizing factor is at work. The United States
considers itself a highly adaptive problem-solving society. Consistent
with its intuitive feeling character, whenever the U.S. public becomes
seriously concerned that its society is losing its competitive edge or other
important qualities, political support develops for corrective measures
that often have useful effect.
In summary, then, it seems likely that the U.S. strategic personality
as we have known it for many decades will remain a reasonably reliable
predictor of U.S. behavioral preferences.
Potential Inuence of the United States Strategic Personality on a
Crisis with an NBC-Armed Regional Aggressor
Ziemkes approach to typing the personalities of states shows great prom-
ise for projecting the more likely behavior of states. While her approach
is new, this characterization of the U.S. strategic personality type and its
16. Francis Fukuyama, What if Women Ran the World? Foreign Affairs (Septem-
ber/October 1998).
17. Edward N. Luttwak, Where Are the Great Powers? At Home with the Kids,
Foreign Affairs (July/August 1994).
chapter 10 289
manifestation spelled out above ring true. It suggests that the United
States has a strong attachment to its vision of how the international
system should work, and to the states that have come to accept that
vision. This strong attachment seems likely to weigh heavily in U.S.
decision-making on how to deal with aggression by regional states armed
with weapons of mass destruction.
I examine this hypothesis by considering four questions that seem
particularly important in characterizing likely U.S. behavior in this kind
of situation: First, would the United States likely prove resolute in facing
future aggression by a regional challenger armed with NBC weapons?
Second, would other regional nations prove willing to stand with the
United States against such aggression? This question also depends upon
the U.S. strategic personality and how potential allies perceive it. Third,
how might the United States retaliate for an aggressors devastating use
of NBC weapons? And, nally, would the use of such weapons lead the
United States and its allies to commit to the total defeat of such an
aggressor? In each case, I will juxtapose the answers that are suggested
by the U.S. strategic personality with those suggested by other perspec-
tives represented in this volume, especially realism.
u.s. resolve
Would the United States likely prove resolute in facing aggression by a
regional challenger armed with NBC weapons? The U.S. strategic person-
ality suggests that it would. Consider how the United States might feel
about failing to come to the aid of a state, especially an ally or a democ-
racy that is threatened by NBC-backed aggression. Tolerating such ag-
gression would be seen as a rst step toward abandoning the largely
democratic world order that the United States sees itself as having been
instrumental in creating. Not intervening would also be seen as deeply
compromising the U.S. role as the preeminent world leader, and spe-
cically its role as the defender of the democratic faith. Such a failure
would be felt as a source of shame for the United States.
18
The United
States has generally played the role of sheriff in democracy town, and
is stronger than anyone else. Thus, other states and the United States itself
would expect it to form a posse to go after the villain, if not to face the
challenger single-handed.
Fears of being thought a coward by itself or by other states would
drive the United States toward intervention as well. A United States that
18. For an interesting discussion of the powerful role of honor in motivating national
behavior, see Elliot Abrams, ed., Honor Among Nations: Intangible Interests and Foreign
Policy (Washington, D.C.: Ethics and Public Policy Center, 1998).
290 the coming crisis
has the capacity to extinguish any state would feel that it was cowardly
to back down from confronting an evil that at worst can impose great
pain on the United States, but not destroy it. Paradoxically, U.S. fears of
being attacked with nuclear weapons could also prove a source of cour-
age: the United States could readily believe that the offender would never
escalate to NBC warfare, as it must be even more fearful of U.S. nuclear
capabilities to annihilate it in retaliation.
In sum, the U.S. strategic personality suggests that it would feel
strongly that standing up even to NBC-backed aggression is the right
thing to do. In fact, one could argue that, all else being equal, the United
States could nd it even more difcult to tolerate aggression when it is
backed with NBC weapons than when it is not.
19
Rosen and Posen address this question directly from their realist
perspectives. Rosen argues that the risks to the United States of confront-
ing a regional power armed with nuclear weapons are very high and
that the end of the superpower competition with the Soviet Union
has dramatically reduced U.S. incentives to accept such risks. He
also states that the United States would see no good options for prose-
cuting such a confrontation, and that the U.S. military leadership would
argue to the president that the risks of intervening against such an
aggressor are unacceptably high. However, he also observes that protect-
ing other nations overseas has been the stated purpose of U.S. armed
forces. While military leaders may see high risks in confronting an NBC-
armed aggressor, they could also worry that reluctance to do so could
raise serious questions about why the United States should maintain
expensive intervention forces. In other words, the United States could
have signicant bureaucratic interests in not ducking this kind of chal-
lenge.
Posen takes a very different position on the question of U.S. resolve.
He maintains that the United States would judge as unacceptable the
long-run consequences of not standing up to an NBC-armed challenger
and thus demonstrating that nuclear weapons can enable aggression.
Posen also remarks that if the United States wished to avoid the painful
dilemma of either confronting a nuclear challenger or of accepting the
painful consequences of not doing so, it would require diplomatic strate-
gies that make inaction appear differently. Developing and implementing
such strategies would be no small feat given the obligation the United
States feels to protect at least the democracies, and the honor that it sees
in this role.
19. See Lawrence Freedman, Great Powers, Vital Interests and Nuclear Weapons,
Survival (Winter 199495), pp. 3552.
chapter 10 291
Posens position that the United States would likely see that it must
be resolute seems the more persuasive of these two realpolitik perspec-
tives; when we consider the behavior suggested by the U.S. strategic
personality, resolute behavior seems far more likely than not.
regional allies
Would other regional states prove willing to stand with the United States
against NBC-backed aggression? The U.S. strategic personality inuences
the answer to this question in several ways. First, Walt argues that
regional states that do not have their own nuclear weapons and are
threatened by those who do would almost certainly be willing to ally
with the United States, provided several conditions are met. The most
important of these conditions are, rst, that the United States does not
propose to initiate attacks on the threatening state. Second, they would
need to see that the United States can be counted on as a reliable and
capable protector.
Threatened regional states considering alliance with the United States
should derive comfort from a good appreciation of its strategic personal-
ity. First, the United States places a very high premium on not resorting
to violence against any state that has not yet broken the peace. Second,
the United States attaches a very high value to honoring commitments;
it sees abandoning an ally to aggression as shameful. Third, it cares
deeply about its reputation in the international community.
The second way in which the U.S. strategic personality should affect
the willingness of other states to join a U.S.-led coalition arises from the
substantial mismatch between the strategic personality of the United
States and those of many of the regional states that might seek its help.
The great majority of the states outside the Western Hemisphere and
Europe are introverted, and most of them are sensing, feeling states. Their
highest values are the safety and internal stability of their societies. They
see the Western emphasis on freedom of the individual and on progress
through competition as serious threats to that stability. They have histori-
cally dealt with corrupting inuences from the outside by withstanding,
excluding, and eventually expelling foreigners.
20
Such states can be expected to be quite reluctant to allow the United
States to station its forces among their populations. Thus, they would
likely delay any such action until the threat of aggression became unmis-
takable and imminent. And when U.S. troops are on their soil, these states
can be expected to keep them at arms length. Some of the more common
mechanisms are isolation, efforts to suppress displays of behavior that
20. Ziemke, Strategic Personality Types.
292 the coming crisis
are normal in the United States but offensive to the host society, and
active reinforcement of the host populations belief in the superiority of
its own culture. After the threat has passed, these countries can be ex-
pected to control contact with their populations even more stringently
and to encourage the United States to go home. However understandable
such actions are, they can suggest a lack of appreciation for its culture
that the United States can nd grating, particularly given the importance
the United States attaches to being appreciated and respected. Saudi
Arabias treatment of U.S. forces deployed on its territory during and
after the Persian Gulf War of 199091 is a good example.
Thus the need to form alliances can create conicts among U.S.
values. Clearly, the United States wants to slow NBC proliferation that
could lead to a very threatening and violent world. This suggests that
alliance with the United States should be made as easy as possible for
countries that would accept U.S. protection in lieu of their own NBC
forces. But the price of alliance can be acceptance of the host countrys
resistance to U.S. culture, which limits U.S. opportunities to win converts
to its creed.
A second conict of values would arise when the United States
considers how to assemble a coalition to stand up to NBC-backed re-
gional aggression. Consistent with its democratic ideals, the United States
attaches a high value to legitimizing important international actions by
seeking the support of as many members of the international community
as possible. At the same time, it would be deeply upset if their involve-
ment in a U.S.-led action were to lead to massive destruction for those
supporters. Balancing these conicting values may require the United
States to minimize the number of active coalition partners it would seek
in such a confrontation and to settle for less direct or delayed political
support from other sympathetic states.
Taken together, these observations suggest that the nature of the U.S.
strategic personality should encourage alliance with the United States,
but that the United States should not expect a close embrace that would
indicate gratitude from its regional allies. The United States may want to
limit the numbers of allies it accepts into the coalition to those that have
essential contributions to make or can protect themselves reasonably well
from NBC attacks.
The realpolitik arguments some of the authors address to this ques-
tion are contradictory. Several suggest that continued nuclear prolifera-
tion could make the United States seem a less reliable protector than
during the Cold War. Rosen further suggests that most of the world
would no longer see its protection as a vital U.S. interest, at least in
realpolitik terms. Thus, they would see little prospect that the United
chapter 10 293
States would take the kinds of measures that might make an offer of
protection credible, such as stationing U.S. forces on their territory during
or even before a crisis.
As noted above, Walts arguments point in the opposite direction.
Walt also notes Posens key pointthe United States would not want
nuclear weapons to be demonstrated as capable of enabling successful
conventional aggression. He points out that this obvious U.S. realist
interest should help to convince potential allies of the U.S. commitment
to stand by them, and argues that great power allies should share this
basic concern. Finally, Walt points out that many states have been willing
to stand up to nuclear-armed powers in the past, and have even attacked
them.
Walts realpolitik arguments seem very persuasive. Non-NBC-armed
regional states facing an NBC-armed aggressor would be at a tremendous
disadvantage in terms of the local balance of power, and realpolitik
arguments would weigh heavily in their decisions. Alliance with a willing
and reliable state that is strong enough to overwhelm the regional ag-
gressor but would not start a war itself is an obvious choice under the
circumstances. And the U.S. strategic personality suggests that it will
continue to meet these conditions very well.
u.s. retaliation
Would the United States retaliate in kind if an opponents use of NBC
weapons were to cause great damage to U.S. forces or to U.S. allies? The
U.S. strategic personality suggests in three ways that the United States
would be disposed to strike back with nuclear weapons if an opponent
were to make highly destructive use of NBC weapons. First, reecting its
intuitive nature, the United States seems likely to conceive of its role in
confronting NBC-backed aggression as one of good against evil: once
again it would be required to defend the hard-won realm of democracy
against totalitarian evil. This characterization could be accentuated to the
extent that the opposing leadership had been demonized and to the
extent that the use of NBC weapons had become even more stigmatized
than it is today. Highly destructive use of NBC weapons in such a
situation would seem to provide clear illumination of an evil state that
richly deserves nuclear retaliation, at least against its ruling elite.
Second, the U.S. strategic personality includes a strong sense of jus-
tice and the need for revenge that would motivate commensurate pun-
ishment. Consistent with the U.S. conception that criminal offenders
should make amends to society by accepting punishment, an offending
state will be seen to deserve nuclear retaliation for the sin of rst use,
294 the coming crisis
and having suffered it, may nd quicker forgiveness from the United
States and other like-minded societies.
Third, the United States seems likely to have made at least implicit
promises to its regional allies that it would retaliate in kind for any
devastating NBC attacks made against them. In the event of such attacks,
the allies could call on the United States to honor such promises. As a
feeling nation, the United States would want to choose a course of action
that is consistent with the values it cherishes, which include honor and
reliability. And, as an extroverted feeling state, the United States values
highly its reputation with its allies and the larger international commu-
nity. Nuclear retaliation seems more consistent with such values than any
other course of action.
In contrast, the conventional wisdom regarding U.S. willingness to
retaliate with nuclear weapons appears to point in the opposite direction.
Rosen notes an extensive series of RAND Corporation war games played
with a cross-section of experts and ofcials in the United States.
21
The
war games posed situations in which a regional aggressor had employed
nuclear weapons to oppose a U.S.-led intervention.
RANDs analyses of the play of these war games suggests an over-
whelming reluctance by the U.S. policy community to recommend retali-
ation with nuclear weapons, provided that great damage had not been
done to the United States itself. The players also exhibited a remarkable
lack of consensus that it was in the U.S. interest to punish nuclear
proliferation, or even the use of nuclear weapons. Moreover, faced with
the prospect of a regional confrontation that could lead to the use of
nuclear weapons, the players showed a marked reluctance even to rec-
ommend that U.S. forces become engaged.
Rosen suggests that these games may point to a signicant shift in
the thinking of U.S. policymakers. He also suggests that they may reect
a disinclination in the U.S. policy community to come to grips with the
difculties of handling hostile regional powers armed with nuclear weap-
ons. Finally, he notes that an unwillingness to address these problems is
also evident in the scant attention given to them by published U.S.
military doctrine.
Of course, these war games, which probably constitute the broadest
and most systematic U.S. postCold War effort to address the problems
of intervening against nuclear-armed regional aggressors, have sampled
the opinions of only a tiny fraction of the community with an interest
21. Marc Dean Millot, Roger Molander, and Peter Wilson, The Day After . . . Study:
Nuclear Proliferation in the PostCold War World (Santa Monica, Calif.: RAND Corpora-
tion, 1993).
chapter 10 295
and expertise in such matters. Moreover, very few of the players have
held such senior positions as to be actually responsible for the safety of
the nation. Still, arguments that nuclear weapons are becoming unusable
even under the most extreme circumstances are becoming increasingly
common. Perhaps the most vocal supporter of this position is General
Lee Butler, an excommander of the U.S. Strategic Air Command and its
successor, the Strategic Command.
22
Rosen himself sees American blood boiling for retaliation in the rst
few days if a regional opponent were to initiate the use of nuclear
weapons to cause great damage. But he argues that other considerations
would come into play afterwards, such as concerns for the morality of
U.S. actions, or the potential effects of future relations with allies.
Quester and Roberts make a variety of points suggesting U.S. actions
would be strongly inuenced by its values, though they do not employ
the strategic personality idea. Quester suggests that the U.S. concepts of
justice and revenge embedded in its criminal justice system would point
toward nuclear retaliation. He notes that the U.S. instinct for revenge
can heat quickly and take considerable time to cool. He cites the substan-
tial support shown by U.S. citizens in the months after Japans surrender
in World War II for the proposition that more atom bombs should have
been dropped. He also argues that had the United States been destroyed
as a result of a nuclear attack by the Soviet Union, massive nuclear
retaliation would have been virtually certain just for revenge. Finally,
Quester argues that the primary effect of an aggressors use of nuclear
weapons against the United States or its allies would likely be to convince
them that they could not live with such a state, a view that would also
seem to reduce U.S. inhibitions against nuclear retaliation.
Roberts sees U.S. nuclear retaliation against a regional opponent who
causes great damage with nuclear weapons as a real possibility, primarily
for reasons that seem to be a matter of the U.S. personality. He argues
that U.S. nuclear retaliation could be motivated by deep anger, demands
that might come from allies and coalition partners, and the prospect that
nuclear retaliation might suppress the opponents further use of its weap-
ons. It could also be motivated by a desire to restore a broken taboo
against the use of nuclear weapons, or to expunge an opponent whose
use of such weapons had marked it as the epitome of evil. Roberts doubts
that a U.S. president would be willing to refrain from nuclear retaliation
if it is seen as likely to reduce signicantly the number of U.S. lives lost
22. See General Lee Butler, The Risks of Deterrence: From Super Powers to Rogue Leaders,
An Address to the National Press Club, Washington, D.C. (February 2, 1998).
296 the coming crisis
in meeting necessary military goals. He argues that the president might
see not retaliating as politically impossible.
Roberts also argues that the United States should decide on the
nature of its retaliation in light of how it wants to judge its own actions,
and have them judged by others, in the aftermath of the conict. Thus,
the United States should not succumb to the opposite temptations of
apocalyptic revenge, or of quitting the ght, but rather should aim to-
ward a middle course of being a responsible steward of its own and the
worlds long-term security interests.
Walt recommends that the United States make clear to any regional
aggressor that it would suffer overwhelming retaliation if it were to make
highly destructive use of nuclear weapons. He also argues that in the
aftermath of such use, any revulsion at the idea of nuclear retaliation
seems likely to disappear.
In summary, then, while no one can know for certain, the United
States seems far more likely than not to retaliate in kind if an opponent
were to use NBC weapons to do great damage to the U.S. or its allies.
Many strong incentives to retaliate have to do with U.S. values as
reected in its strategic personality. Indeed, staying cool enough to keep
the response measured in any sense would be a challenge for U.S. leaders,
especially given the likely perception of the perpetrator of NBC use as
the embodiment of evil.
unconditional surrender
Would the United States and its allies commit to the total defeat of a
regional aggressor that had used NBC weapons to cause great damage?
The U.S. strategic personality would strongly inuence the answer to this
question, which is closely related to the last one. Again, the opponents
use of NBC weapons in support of aggression would be a most dramatic
insult to the democratic world order, which the United States sees as an
extension of itself. By threatening the fundamental U.S. concept of how
it is doing good for the world, the opponent would be casting itself as
evil in the eyes of the United States; as noted above, that image would
be intensied enormously by an aggressors devastating use of NBC
weapons.
Thus, the United States would see making peace as an accommoda-
tion with the devil. And peace could leave the United States and its allies
feeling that justice for so large a crime had not been done. Moreover, for
states with extroverted strategic personalities such as the United States,
human progress is supposed to be linear and irreversible. Accordingly,
problems are supposed to be solved once and for all, rather than through
chapter 10 297
the acceptance of partial solutions with their likelihood of repeated
trouble and continued costs for the indenite future. These arguments
suggest that the United States would want to pursue the opponents total
defeat.
The United States regional allies could feel very differently. They
would bear the greatest risks of further NBC attacks by the aggressor,
and the probability of such attacks would likely rise as the opponent sees
its defeat approaching. For many possible regional allies, even a few NBC
attacks against key undefended targets could end the states existence.
And the introverted nature of most of the states that are neither European
nor descendants of European states does not point toward their favoring
the total defeat of the opponent. In contrast to the extroverted United
States, introverted states tend to see life as cyclical with few if any nal
solutions short of paradise.
23
Questers answer to the question of total defeat relies more on U.S.
values than on realist considerations. He argues that so long as the
existence of the United States is not at risk, it would want to defeat,
disarm, and reform aggressors that initiate highly destructive use of
nuclear weapons. Quester posits that such crimes would convince the
allies that they cannot live with such aggressors. In supporting his posi-
tion, Quester points to the analogous policies of the U.S. criminal justice
system for dealing with armed robbers. He notes the unconditional sur-
render strategy that was followed in World War II, even though the Allies
knew that they could avoid the near-term loss of millions more lives by
accepting the continued existence of Nazi Germany and Imperial Japan.
Roberts and Walt take similar positions. Roberts argues that the
opponent should suffer a major loss for its aggression, and that if it had
caused great damage with its nuclear weapons, it should be treated as
Quester suggests: with defeat, disarmament, and reform. Walt states that
while the United States should let such opponents know that it will not
try to overthrow them so long as they refrain from using nuclear weap-
ons, it should assure opposing leaders that it will oust them if they use
nuclear weapons to cause great destruction.
Other largely realpolitik considerations could help to motivate a
decision to seek the total defeat of the opponent. U.S. and allied leaders
could appreciate the large costs, risks, and political complexities of reas-
sembling the forces needed to defeat the offending state if it were to
engage in aggression again. And the prospects of further aggression
would seem high if the same offending government remains in power
and nurses the resentment generated by its original failure. Further,
23. Ziemke, Strategic Personality Types.
298 the coming crisis
choosing not to seek total defeat of the aggressor for fear of spurring
additional use of NBC weapons would send a general message that such
weapons can provide a safety net that at least limits the risk of offensive
military campaigns.
Still, all of these points do not produce an obvious answer to the last
question. While the United States would seem inclined to seek total defeat
for an aggressor that makes highly destructive attacks with NBC weap-
ons, its regional allies may prefer lesser goals. Anticipating this possible
divergence of views, the United States might want to accept the minimum
number of regional allies. On the other hand, regional allies might see
that gaining inuence over the U.S. strategy, and this decision in particu-
lar, would be an important reason to try to join the U.S. coalition.
Observations and Conclusions
So, will the world order be fundamentally changed if the proliferation of
NBC weapons continues? Yes, but probably not as most people seem to
believe. The conventional wisdom says that obtaining NBC weapons
would enable the smaller powers to veto interventions by the United
States and the other major powers. The United States and the other major
powers would be forced to compromise or abandon their regional inter-
ests, and aggressive regional states armed with NBC weapons would be
free to dominate if not conquer their less well-armed neighbors.
This book tells a different story. That story suggests that the United
States is very likely to understand and reject the outcomes projected by
the conventional wisdom. It will be motivated by a realists perception
of the dangerous consequences of allowing conventional aggression
backed by NBC weapons to go unopposed. And it will be motivated to
reject them because to do otherwise would mean abandoning its long-
standing role as promoter and defender of the democratic faith. Retreat-
ing could also require the United States to repudiate other more specic
values such as its sense of its honor, bravery, and reliability. These are
deep-seated values. Together, the essays in this volume, including this
one, suggest that the United States is not nearly so reluctant to play its
role as global sheriff as its occasional complaints about the burdens of the
role might indicate.
This unconventional picture of the consequences of NBC proliferation
and aggression backed by such weapons has some important implica-
tions. First, the United States needs to develop a deeper and richer
understanding of the consequences of continued NBC proliferation, one
that combines and balances all the relevant perspectives. The strategic
personality concept can provide a better understanding of how U.S.
chapter 10 299
values bear on its likely behavior in an increasingly proliferated world,
how other involved nations may perceive that world, and how they
might be inclined to act in it. It should also facilitate an improved under-
standing of how the states involved in this problem perceive each others
words and actions.
Second, major differences in how the United States, potential chal-
lengers, and other involved states see the stakes at risk in a future
confrontation where NBC use is a possibility are a recipe for catastrophe.
All international actors must be made aware of the depth of the U.S.
commitment to the evolving democratic world order, and that it is likely
to see such a confrontation as a mortal threat to that order. Thus, the
United States should not only develop the clearest understanding of the
stakes it would see at risk, but communicate this understanding to all
who might become involved.
Third, to the extent that the U.S. stakes in opposing NBC-backed
aggression are as high and its bias toward defending them as strong as
argued here, the United States should abandon any delusions that it can
simply decide not to become engaged in such confrontations. Instead, it
needs to be well prepared to defend against all forms of aggression,
including those where the aggressor is armed with NBC weapons. Not
surprisingly, the U.S. government has made a good start in this direc-
tion.
24
Still, it has a long way to go to meet the minimum goal of
protecting its intervention forces well enough to avoid their defeat should
a well-armed regional aggressor make substantial use of NBC weapons.
And other states that the United States might need as allies will need to
be protected as well. This goal is feasible. Defending against the modest
nuclear arsenals that a new proliferator might plausibly afford is far
easier than defending against a nuclear superpowers arsenal. And de-
fending against attacks with chemical and biological weapons is far easier
than defending against a modest nuclear arsenal.
25
Protecting U.S. and allied intervention forces from defeat with NBC
weapons has a variety of advantages. It could discourage an opponent
from attempting aggression and could undermine some of the incentives
to obtain NBC weapons in the rst place. Especially important, strong
24. See Proliferation: Threat and Response, Ofce of the Secretary of Defense (Washing-
ton, D.C.: U.S. Department of Defense, November 1997), and Report on Activities and
Programs for Countering Proliferation and NBC Terrorism, Counterproliferation Program
Review Committee (Washington, D.C.: U.S. Department of Defense, May 1998).
25. See Joshua Lederberg, ed., Biological Weapons: Limiting the Threat, BCSIA Studies
in International Security (Cambridge, Mass.: MIT Press, 1999); and Victor A. Utgoff,
Nuclear Weapons and the Deterrence of Biological and Chemical Warfare, Henry L.
Stimson Center Occasional Paper No. 36 (October 1997).
300 the coming crisis
defenses could limit the damage done by an opponents NBC weapons
to a small fraction of the damage that might otherwise be possible. As a
result, the pressures for nuclear retaliation would be lower, and the
amount of retaliatory damage that might seem necessary would be re-
duced. Further, by reducing the potential risks of further NBC use, strong
defenses could open the door to imposing total defeat on a challenger
that had initiated the use of such weapons.
Fourth, the United States should rededicate itself to preventing and
rolling back proliferation in any way it can. Past nonproliferation efforts
have led to many important successes. The current efforts to get India
and Pakistan to halt further development of their nuclear capabilities are
important. And there will be future successes. Still, nonproliferation has
too often been sacriced to other interests. Such sacrices could cost the
world dearly if they increase the potential for the kinds of dangerous
confrontations that are our concern here.
Finally, while the United States should continue to work toward
reducing its dependence on nuclear weapons, it must be careful not to
get the cart before the horse. If the United States is as likely to retaliate
with nuclear weapons for an aggressors highly destructive use of NBC
weapons as suggested here, it should be very leery about suggesting
otherwise. And unless and until the world nds a safe way to eliminate
all NBC weapons, the United States should never let doubts develop
about its capability to impose nuclear punishment when it is justied. In
the end, while proliferation of NBC weapons remains a threat to peace,
the threat of retaliation in kind remains an essential underpinning for the
stability of the world order as we know it.
chapter 10 301
Contributors
Richard K. Betts is the Leo A. Shifrin Professor and Director of the Institute
of War and Peace Studies at Columbia University, and Director of National
Security Studies at the Council on Foreign Relations. He was formerly a
Senior Fellow at the Brookings Institution and has taught at the Johns Hop-
kins Nitze School of Advanced International Studies and Harvard University.
Among his books are Military Readiness (Brookings, 1995); Nuclear Blackmail
and Nuclear Balance (Brookings, 1987); Surprise Attack: Lessons for Defense
Planning (Brookings, 1982); and Soldiers, Statesmen, and Cold War Crises (Har-
vard University Press, 1977). In January 1999, Senate Minority Leader Tom
Daschle appointed Betts a member of the National Commission on Terrorism.
Barry R. Posen is Professor of Political Science at the Massachusetts Institute
of Technology. He has written two books, Inadvertent Escalation (Cornell
University Press, 1991), and The Sources of Military Doctrine (Cornell Univer-
sity Press, 1984). Other publications include Nationalism, the Mass Army
and Military Power, in International Security, Vol. 18, No. 2 (Fall 1993);
Competing U.S. Grand Strategies, with Andrew L. Ross, in Eagle Adrift:
American Foreign Policy at the End of the Century, edited by Robert J. Lieber
(HarperCollins, 1996); and Military Responses to Refugee Disasters, in
International Security, Vol. 21, No. 1 (Summer 1996). Dr. Posens current activi-
ties include research on innovation in the U.S. Army from 197080. He is
afliated with the MIT Security Studies Program.
George H. Quester is a Professor of Government and Politics at the Univer-
sity of Maryland, where he teaches courses on international relations, Ameri-
can foreign policy, and international military security. He has taught pre-
viously at Cornell and Harvard Universities, at UCLA, and in the Department
of Military Strategy at the National War College. From 1991 to 1993, he served
as the Olin Visiting Professor at the U.S. Naval Academy. Dr. Quester is the
author of a number of books and articles on international security issues, and
on broader questions of international relations, and has served as a consultant
for the Institute for Defense Analyses (IDA).
Brad Roberts is a member of the research staff at IDA, where he contributes
to studies for the Ofce of the Secretary of Defense and the joint military
staff. His areas of expertise are counterproliferation, nonproliferation, coun-
terterrorism, and NBC weapons. He is also an adjunct professor in the Elliott
School of International Studies at George Washington University, chairman
of the research advisory council of the Chemical and Biological Arms Control
Institute, a member of the executive committee of the Council for Security
Cooperation in the Asia Pacic, and a consultant to Los Alamos National
Laboratory.
Stephen Peter Rosen is the Kaneb Professor of National Security and Military
Affairs and Associate Director of the Olin Institute for Strategic Studies at
Harvard University. Previously, he was the civilian assistant to the Director,
Net Assessment in the Ofce of the Secretary of Defense; the Director of
Political-Military Affairs on the staff of the National Security Council; and a
professor in the Strategy Department at the Naval War College. He was a
consultant to the Presidents Commission on Integrated Long Term Strategy,
and to the Gulf War Air Power Survey sponsored by the Secretary of the Air
Force. He is the author of Societies and Military Power: India and its Armies
(Cornell University Press, 1996) and Winning the Next War: Innovation and the
Modern Military (Cornell University Press, 1991). He is currently working on
a project on strategy and the biology of cognition.
Scott D. Sagan is Associate Professor of Political Science and Co-Director of
Stanfords Center for International Security and Cooperation. Dr. Sagan is the
co-author (with Kenneth N. Waltz) of The Spread of Nuclear Weapons: A Debate
(W.W. Norton, 1995) and the author of The Limits of Safety: Organizations,
Accidents, and Nuclear Weapons (Princeton University Press, 1993) and Moving
Targets: Nuclear Strategy and National Security (Princeton University Press,
1989).
Victor A. Utgoff is Deputy Director of the Strategy, Forces and Resources
Division of IDA. During 199899 he established the Advanced Systems and
Concepts Ofce for the Defense Threat Reduction Agency. He is the author
of The Challenge of Chemical Weapons: An American Perspective (MacMillan,
1990), co-author (with Barry M. Blechman) of Fiscal and Economic Implications
of Strategic Defenses (Westview, 1986), and has published widely on issues
posed by the proliferation of nuclear, biological and chemical weapons. In
1999, he received the Andrew J. Goodpaster Award for Excellence in Research.
Stephen M. Walt is Kirkpatrick Professor of International Affairs at the John
F. Kennedy School of Government at Harvard University. Previously, he was
304 the coming crisis
Professor of Political Science and Master of the Social Science Collegiate
Division at the University of Chicago. He is the author of The Origins of
Alliances (Cornell University Press, 1987), which received the Edgar S. Furniss
National Security Book Award, and numerous articles on international rela-
tions and security studies.
General Larry D. Welch, USAF (Ret.) is the President and Chief Executive
Ofcer of IDA. He took this position after a long military career that included
assignments as Commander in Chief of the Strategic Air Command and Chief
of Staff of the U.S. Air Force. He participates in many senior defense advisory
activities including the Commission to Assess the Ballistic Missile Threat to
the United States, the Commission on Maintaining U.S. Nuclear Weapons
Expertise, and the U.S. Strategic Command Advisory Group. He is chairman
of the Presidents Security Policy Advisory Board, and the Department of
Defense Threat Reduction Advisory Committee. He chaired the 1998 Defense
Science Board Task Force on Nuclear Deterrence.
Caroline F. Ziemke is a Research Staff Member at IDA. She has a doctorate
in Military History and Strategic Studies from Ohio State University. She is
working on a book on the inuence of historical experience on the strategic
conduct of states entitled The Strategic Personality Types: Using National Myth
to Project State Conduct.
contributors 305
Index
A
Afghanistan, 105
alliance formation
to augment offensive capa-
bilities, 201
to balance against threaten-
ing enemy, 199200
bandwagoning, 202203
to create aggregate power,
200
effect of states competing
prescriptions, 206207
obstacles to, 204205
with offensive intentions,
201202
as response to threats from
neighboring states, 200201
with threat of rogue states
WMD, 209213
with uncertainty, 208209
Andre, David, 151
anthrax vaccinations, 4
Art, Robert, 84
Arunachalam, V.S., 127128
Aspin, Les, 191n2, 192
Aum Shinrikyo cult, 34, 77, 83
Aziz, Tariq, 175, 179
B
Baker, James, 175, 179, 181
ballistic missiles
as carriers of WMDs, 57n10
Irans acquisition of, 110,
113
U.S. attacks during Gulf War
on, 131132
See also Scud missiles
Beg, Mirza Azlam, 129130
Berlin crisis (1958), 43
Bhabba, Homi, 2930
Bharatiya Janata Party (BJP),
India, 3233
biological weapons
countries developing, 61
home defense against attack
by, 7980
potential for use of, 4
Soviet and Russian pro-
grams of, 60
taboo against using, 8185
use in warfare, 58, 60
Bosnia, 105
Bush, George, 171172, 177,
179, 182183
Butler, Lee, 296
C
chemical weapons
effect against military forces,
59
effective use of, 59
Irans capability to produce,
113114
use by Iraq, 176, 178
Cheney, Richard, 171, 176177,
181183
China
development of atomic
bomb, 2122, 54, 67
as supplier of nuclear tech-
nologies, 118
U.S. concerns about WMD
acquisition by, 198
Clinton, Bill, 79
coalition formation
in response to rogue states
acquisition of WMD, 213
219
states rationale for, 199
203
states reasons for not form-
ing, 205206
by U.S. to contain rogue
states with WMDs, 221226
Cold War
circumstances for military
forces in Japan and Ger-
many, 231232
debate about nuclear weap-
ons during, 270
deterrence model, 227230,
238241
deterrence theory of, 52
effect of nuclear deterrence
on, 64
effects of end of, 5556
inuence of nuclear weap-
ons during, 6364
near-war crisis (1962), 76
nuclear deterrence model
during, 6, 71
postCold War incentives to
acquire WMDs, 55
command and control system,
Iraqi, 132
comparative advantage of
WMDs, 58
compellence, 7881
Comprehensive Test Ban
Treaty (CTBT), 5
reecting intent, not proscrib-
ing, 69
shift in U.S. escape clause po-
sition, 47
containment policy, U.S., 221
226
Counter-Proliferation Initiative
(CPI),
U.S., 5, 191
counterproliferation policy, U.S.
effect in North Korea, 186
Irans resistance to, 118
against Iraq, 186
as postCold War policy, 55
as preparation for NBC at-
tacks, 273274
response to countries prolif-
eration incentives, 167170
strategy toward Iran, 119121
Cuban crisis, 43
D
decision-making
domestic politics model of
nuclear weapons develop-
ment, 2737
308 the coming crisis
norms model of nuclear
weapons development, 37
49
security model of nuclear
weapons development, 20
27
de Gaulle, Charles, 41, 43
44
de Klerk, F.W., 2324, 3334
Department of Defense
(DOD), Ofce of
Counterproliferation Policy,
192n4
Desert Shield crisis. See Per-
sian Gulf War
deterrence
alternatives to, 7881
during Cold War, 6364
Cold War deterrence model,
238241
effect of nuclear, 71
of nuclear weapons in India
and Pakistan, 198
superpower mutual, 75
deterrence, intrawar
in actual U.S. planning for
Desert Storm, 175179
Cold War, 239240
with a nucleararmed Iraq,
178
U.S. plan from 1990 perspec-
tive, 174175
uses for, 179
deterrence theory
applied to biological and
chemical weapons, 7677,
83
of Cold War, 52, 6364
logic of, 6572
of utopian realists, 84
Domenici, Pete, 4
E
Eagleburger, Lawrence, 171
electromagnetic isotope separa-
tion (EMIS)
estimates of time to repair
bomb-damaged, 137138
impact of U.S. bombing on
devices and processes, 136
137
Iraqi enrichment facilities,
137
manufacture of devices in
Iraq, 135138
ethnic minorities, Iran, 106
F
Feldman, Shai, 66
foreign policy, Iranian, 99102,
108
foreign policy, U.S.
advantages in dealing with
Iran, 118119
containment policy, 118, 221
226
See also counterproliferation
policy, U.S.
France
development of nuclear
weapons, 21, 4144, 56, 67
G
Gandhi, Indira, 3031
Germany
disarmed at end of World
War II, 231
military forces in occupied,
231233
non-nuclear status, 232233
index 309
Germany (continued)
postWorld War II reform of,
236238
punishment for World War
II actions, 233236
WMDs not permitted in, 232
Goldstein, Avery, 22
Gulf Cooperation Council
(GCC)
basis for founding, 101
concerns related to Irans ac-
tions, 1023
Gulf War Air Power Survey
(GWAPS)
analysis of air campaign
against Iraqi Scud missiles,
131
H
hegemony
Irans quest for regional, 101
102, 108, 117
I
Ikl, Fred, 246, 247, 266267,
271
India
development and detonation
of nuclear weapons, 6, 22,
2829, 54, 57, 67, 85, 128,
195
domestic politics related to
nuclear weapons, 2833
economic sanctions against,
70
events prior to 1998 nuclear
tests, 3233
nuclear weapons as deter-
rent, 198
nuclear weapons relations
with Pakistan, 129130
International Atomic Energy
Agency (IAEA)
inspection of Iraqi EMIS in-
stallations, 137
inspection of suspected Iraqi
nuclear program, 133134
monitoring and enforcement
incentives of, 37
safeguards for nuclear mate-
rials, 19
Iran
accountable regime of, 112
113
advantages in resisting coun-
terproliferation policy, 118
demonization of the West by,
9697, 100, 114115
intent to export revolution-
ary ideas, 103 104
national myth of, 8990
need for foreign investment
and technical assistance,
118119
nuclear technologies sup-
plied by China, 118
quest for regional hegemony,
101102, 108, 117
as rogue or backlash state,
191
Iran-Iraq War
Irans military strategy in,
113
use of chemical weapons, 59
60
Iraq
actual behavior during Per-
sian Gulf War, 144
benets from seizing Ku-
wait, 197
310 the coming crisis
coalition against, 204
with hypothetical nuclear
weapons (1990), 139144,
172173
hypothetical objectives in us-
ing hypothetical nuclear
weapons, 144151
Irans military defense
against, 110
Irans relationship with, 102
Israel destroys Osirak reac-
tor (1981), 75,135
manufacture of EMIS de-
vices, 135
under NPT, 69
as rogue or backlash state,
191
sources of uranium ore and
enriched uranium, 135
136
U.S. planning reversal of
Iraqi aggression from 1990
perspective, 173179
U.S. targets during Gulf War,
131144
WMD capability, 72, 195
Israel
destroys Iraqi nuclear reac-
tor (1981), 75, 135
nuclear program, 6, 67
Japan
disarmed at end of World
War II, 231232
eld tests of biological weap-
ons, 60
non-nuclear status, 166, 233
postWorld War II reform of,
236238
punishment for World War
II actions, 233 236
Self-Defense Forces, 232233
WMDs not permitted in, 232
K
Kahn, Herman, 246, 267269
Kecskemeti, Paul, 241
Kennedy, John F., 53
Keynes, John Maynard, 57
Khalilzad, Zalmay, 193
Khatami, Muhammed, 109
110, 115, 119
Kohl, Wilfred, 43
KTO. See Kuwait Theater of
Operations (KTO)
Kuchma, Leonid, 45
Kuwait
in hypothetical Iraqi attack
using hypothetical nuclear
weapons, 144151, 158
159
U.S. view of Iraqi conquest
of (1990), 160167
Kuwait Theater of Operations
(KTO)
effect of bombing in, 184
intense military operations
within, 180
L
Lake, Anthony, 87
Latin America
nuclear weaponsfree zone,
24
law enforcement model, 227
Lebanon, 105
Libya, 191
Lugar, Richard, 4
index 311
M
McKinley, William, 285
Marshall Plan, 235
Martin, Michael, 43
Mendes-France, Pierre, 41
military sector
Iranian capability in, 108112
Iranian strategy in Iran-Iraq
War, 113
Irans nuclear and conven-
tional modernization pro-
grams, 118
South Korea, 166
Taiwan, 166
missiles with nuclear war-
heads, 139140
Molander, Roger, 193
mutual assured destruction,
51n1
N
national myth
of intuitive states, 282
Israeli, 114
of a state, 8889
national myth, Iranian
concept of Islamic justice, 104
in development of U.S. coun-
terproliferation policy, 119
120
dress code, 98
effect on relations with other
Arab countries, 102
pre-Islamic Persian, 8991,
103
revolutionary myth, 9899,
108109
Shiism as part of, 9195, 113
strategic personality related
to, 99
view of foreign policy and
national security through,
99100
NBC (nuclear, biological, and
chemical) weapons
aggressors choice to use,
249252, 271
aggressors response to U.S
response to use of, 262263
continued proliferation of, 3
7
countries trying to obtain, 3
creation of defenses against,
5
developing public awareness
of need to counter NBC
weapons, 6
initiatives to slow prolifera-
tion of, 4
means to deny NBC weap-
ons to proliferators, 56
policies to prepare for use in
potential war, 272273
potential effect of prolifera-
tion, 4
potential for attacks by, 6
potential inuence of United
States on aggressors using,
289299
predicting end of regional
war, 272
spread of, 3
termination of warfare, 245,
247249
U.S. response to aggressors
use of, 252
vulnerability of U.S. military
forces to attacks by, 273274
312 the coming crisis
See also WMDs (weapons of
mass destruction)
new institutionalism, 38
nonproliferation
inuence of domestic politics
on U.S. policy, 3536
treaties proscribing weapons,
6970
norms
within international regimes,
3940
of new institutionalism, 38
Noth Korea
effect of U.S. counterprolif-
eration policy in, 186
under NPT, 69
nuclear program, 67
as rogue or backlash state,
191
violation of NPT, 63
NPT. See Nuclear Non-Prolif-
eration Treaty
(NPT)
nuclear commitments, rst-use
policy, 25
nuclear deterrence
effect on Cold War, 64
logic of utopian realism, 70
78
Soviet-U.S. relationship, 6
Stability-instability paradox,
162
Nuclear Non-Proliferation
Treaty (NPT)
Article IV, 70n28, 186187
conditions for violation of,
165
in domestic policy model, 36
37
extension (1995), 4
inuence on proliferation
norms, 41, 4547
North Koreas violation of, 63
nuclear and non-nuclear
states under, 2526, 44,
166, 232233
potential for emerging
norms against nuclear
weapons, 4849
promotion of development
of power reactors, 19
reecting intent, not proscrib-
ing, 69
treaty review conferences,
18, 187
U.S. commitment under Arti-
cle VI, 37, 48
violation by North Korea, 63
nuclear power
construction or operation of
reactors, 19
Irans declaration to develop
peaceful, 115116
nuclear restraint
Argentina and Brazil, 24, 35,
74
cause of, 25
of most European countries,
42
South Africa, 24, 3335
Ukraine, Kazakhstan, and
Belarus, 24
nuclear technologies
Chinas as supplier to Iran,
118
nuclear weapons
disadvantages to acquisition
of, 5455
elements in countering Irans
strategy for, 114
index 313
nuclear weapons (continued)
hypothetical deterrence of
Iraqi (1990), 139144
hypothetical question of use
in Gulf War, 144151
hypothetical U.S. problems
related to use of hypotheti-
cal, 144151
international norms concern-
ing acquisition of, 3738
Irans constraint related to
use of, 114115
Irans incentives to pursue,
115
as political objects, 18, 59
potential for use in Persian
Gulf War, 174179
in regional conict, 125126,
157
sources of political demand
for, 19
South Africas plan for use
of, 130
taboo against using, 8185
U.S. policy response to use
of, 151156
U.S. retention of, 236
See also India; nuclear weap-
ons programs; Pakistan
nuclear weaponsfree zone
(NWFZ), Latin America, 24
nuclear weapons programs
cases of reversals of, 2325
dedicated, 57
domestic politics model of
states, 18, 2737
estimates of time to repair
Iraqs bomb-damaged facili-
ties, 133139
IAEA inspection teams, 133
134
Iran, 11015
norms model of states acqui-
sition and development,
18, 3749
as political tools, 18
security model of states, 18,
2027, 49
as symbolic functions, 40, 42
44, 48
theories for why states de-
velop, 1820
U.S. Gulf air war against
Iraqi, 132139
U.S. suspension of testing, 4
5
nuclear weapons proliferation
consensus view of causes of,
18
history of, 21
inuence of NPT norms on,
41
initiatives to slow, 4
issues in developing re-
sponse to, 88
liberals opposed to, 6869
of NBC weapons, 37
shift in norms related to, 40
41
waves of concern (1960s,
1970s, 1990s), 55
See also counterproliferation
policy
Nunn, Sam, 4
O
Ofce of Strategic Services,
Economic Objectives Unit,
133
oil and gas reserves, Iranian,
116
314 the coming crisis
P
Pakistan
development and detonation
of nuclear weapons, 6, 22
23, 54, 57, 74, 85, 195
economic sanctions against,
70
Iranian ties with Islamic re-
sistance in, 105
nuclear weapons as deter-
rent, 198
nuclear weapons relations
with India, 129130
Persian Gulf War
hypothetical Iraqi behavior
with hypothetical nuclear
weapons, 144151
hypothetical question of nu-
clear weapons used dur-
ing, 144151
inferences from, 170173
Irans neutrality in, 116
Iraqi behavior during, 144
U.S. targets in Iraq during,
131144
Pickering, Thomas, 171
Poland
as NATO member, 56
policy issues
domestic politics model of
nuclear weapons develop-
ment, 3537, 50
of norms model of nuclear
weapons development, 47
50
related to actions if war
comes, 272273
of security model of nuclear
weapons development, 25
26, 50
U.S policy related to use of
nuclear weapons, 151156
See also foreign policy, Ira-
nian; foreign policy, U.S.
Posen, Barry, 126
Powell, Colin, 172, 177178,
181, 182
preemptive attack
conditions for, 77
likelihood of, 76
U.S. preparation for (1962),
76
preventive war
conditions for, 77
incentives for, 76
probability of, 7576
See also preemptive attack
proliferation. See counterprolif-
eration policy, U.S.; nonpro-
liferation; Nuclear Non-
Proliferation Treaty (NPT);
nuclear weapons prolifera-
tion
Q
Quester, George, 266
R
Rafsanjani, Hashemi, 109, 115
RAND Corporation
survey of U.S. policy re-
sponse to nuclear weapons
use, 151156, 157n2
rogue or backlash states
with acquired WMDs, 191
193
characteristics of, 194199
containing, 209226
perceived threat from, 196
index 315
rogue or backlash states (con-
tinued)
potential for development of
WMDs, 6272
threats to use WMD, 209226
Roosevelt, Franklin D., 285
Ross, Dennis, 171
Rusk, Dean, 198
S
Sagan, Scott, 6465, 73
Sarabhai, Vikram, 30
Saudi Arabia
in hypothetical Iraqi attack
using hypothetical nuclear
weapons, 145151
relationship with Iran, 102
103
support for direct interven-
tion in Iraq, 182
Schwarzkopf, Norman, 177,
182
Scowcroft, Brent, 170, 177
Scud missiles
attacks against Iraqi launch
sites (199091), 131132, 185
security, national
interests shared by Iran and
United States, 116
Iranian policy and strategy
for, 99100, 112
threats to U.S., 191n2
Shastri, Lal Bahadur, 2930
Shia Islam, 104
Shultz, George, 2021
Singh, Jaswant, 75
South Africa
domestic politics related to
nuclear weapons, 3335
nuclear weapons program,
2324, 54
strategy for nuclear weap-
ons, 130
South Korea
response against nuclear ag-
gressor, 166
Soviet Union
development of nuclear
weapons, 21
disintegration of, 267
as rogue state, 197
Stalin, Josef, 21
states
alliance to create aggregate
power, 200
augmenting offensive capa-
bilities, 201
bandwagoning behavior, 202
203
characteristics of extro-
verted, 281282, 284
characteristics of feeling
states, 282283
characteristics of intuitive,
282
circumstances for not form-
ing coalition, 205206
coalitions in response to
threats from neighboring
states, 200201
coalitions to balance against
threats, 199200, 204205
with offensive intentions,
202202
See also rogue or backlash
states
states, nonnuclear
Germany, Japan, Ukraine,
Iraq, 44, 72, 166, 232233
316 the coming crisis
in NPT, 25, 44, 6970, 166,
232233
states, nuclear
in NPT, 2526, 6970, 232
nuclear weapons doctrine in
new, 127131
See also China; France; India;
Pakistan; South Africa;
United Kingdom; United
States
strategic personality
potential inuence on NBC
aggressor of U.S, 289299
types based on charac-
teristics, 281
of United States, 280289, 294
Suez Crisis (1956), 4142
Sundarji, K., 127129, 247248
Sunni Islam, 105
Syria, 102
T
Taiwan, 166
terrorist groups
as exception to utopian real-
ism, 77
potential use of WMDs by,
7778
Thatcher, Margaret, 172
Tolubko, Volodomyr, 45
Turkey, 182
U
Ukraine
bartering of nuclear weap-
ons for economic aid, 56
nuclear restraint, 24, 4447,
188
uncertainty
in assessing threat by rogue
state, 219220
as reason for alliance forma-
tion, 208209
unipolarity supplants bipolar-
ity, 267268
United Arab Emirates (UAE),
102
United Kingdom
development of nuclear
weapons, 21, 67
United Nations
IAEA inspectors operating in
Iraq under, 134
program to root out Iraqs
NBC programs, 4
United States
ability to form and lead coali-
tions, 203204
advantages in dealings with
Iran, 118119
commitment under Article
VI of the NPT, 37, 48
counterproliferation efforts,
111112, 118
creation of defenses against
NBC weapons, 5
decision not to intervention
directly in Iraq, 182
development of atomic
bomb, 21
dual containment strategy,
118
efforts to isolate Iran and
Iraq, 206
as hypothetical cowardly
wimp, 269, 274
as hypothetical nuclear bully,
268269, 274
index 317
United States (continued)
hypothetical problems and
actions related to hypo-
thetical nuclear weapons
use, 144151, 158160, 167
170
inuence of NPT norms on
nuclear policy, 47
interventions in Iranian inter-
nal affairs, 96
missile with nuclear war-
head, 139140
nuclear rstuse doctrine, 47
48
planning reversal of Iraqi ag-
gression from 1990 perspec-
tive, 173179
policy under operation Ear-
nest Will, 161
potential inuence on NBC
armed aggressor, 289299
predictions of behavior of,
286289
retention of nuclear weapons
by, 236
strategic personality of, 280
289
support for states facing
threats of rogue state,
212
suspension of nuclear test-
ing,
45
targets in Iraq during Gulf
War, 131144
using its stewardship in man-
aging crises, 270271
view of Iraqi conquest of Ku-
wait (1990), 160167
as winner of WMD war and
peace, 267268
as worlds premier military
power, 267
U.S. Air Force
effectiveness measurements,
132133
V
Vajpayee, Atal, 23, 129
W
Waltz, Kenneth, 52, 6466, 71,
73, 77, 8081
war, conventional
between India and Pakistan,
198
between nuclear-armed ad-
versaries, 162165
wars
nal outcome, 266267
military outcomes of NBC,
263266
termination by United States,
266
winning, 267
Western European Union
Treaty (1955), 232233
Wilson, Peter, 193
Wilson, Woodrow, 285
WMDs (weapons of mass de-
struction)
acquired by rogue, or back-
lash states, 191192, 195
199
biological weapons as, 6061
CW as least efcient, 5960
effect of continued prolifera-
tion of, 280
effect of widespread develop-
ment of, 70
318 the coming crisis
if used by renegade state,
241244
not permitted in former Axis
states, 232
nuclear, biological, and
chemical, 5758
overestimation of spread of,
54
postCold War incentives to
acquire, 5557
probable response to deploy-
ment of, 77
rate of spread of, 5768
reasons for wanting, 6168
states with incentives to ac-
quire, 5657, 6163, 7273,
195199
what if they are used?, 8185
See also NBC (nuclear, bio-
logical, and chemical)
weapons
Wolfowitz, Paul, 182
Y
Yost, David, 41
Z
Ziemke, Caroline, 280, 281, 289
index 319
BCSIA Studies in International Security
Published by The MIT Press
Sean M. Lynn-Jones and Steven E. Miller, series editors
Karen Motley, executive editor
Belfer Center for Science and International Affairs (BCSIA)
John F. Kennedy School of Government, Harvard University
Allison, Graham T., Owen R. Cot, Jr., Richard A. Falkenrath, and Steven E.
Miller, Avoiding Nuclear Anarchy: Containing the Threat of Loose Russian Nuclear
Weapons and Fissile Material (1996)
Allison, Graham T., and Kalypso Nicoladis, eds., The Greek Paradox: Promise vs.
Performance (1996)
Arbatov, Alexei, Abram Chayes, Antonia Handler Chayes, and Lara Olson,
eds., Managing Conflict in the Former Soviet Union: Russian and American
Perspectives (1997)
Bennett, Andrew, Condemned to Repetition? The Rise, Fall, and Reprise of
Soviet-Russian Military Interventionism, 19731996 (1999)
Blackwill, Robert D., and Michael Strmer, eds., Allies Divided: Transatlantic
Policies for the Greater Middle East (1997)
Brom, Shlomo, and Yiftah Shapir, eds., The Middle East Military Balance
19992000 (1999)
Brown, Michael E., ed., The International Dimensions of Internal Conflict (1996)
Brown, Michael E., and Sumit Ganguly, eds., Government Policies and Ethnic
Relations in Asia and the Pacific (1997)
Elman, Miriam Fendius, ed., Paths to Peace: Is Democracy the Answer? (1997)
Falkenrath, Richard A., Shaping Europes Military Order: The Origins and
Consequences of the CFE Treaty (1994)
Falkenrath, Richard A., Robert D. Newman, and Bradley A. Thayer, Americas
Achilles Heel: Nuclear, Biological, and Chemical Terrorism and Covert Attack (1998)
Feldman, Shai, Nuclear Weapons and Arms Control in the Middle East (1996)
Forsberg, Randall, ed., The Arms Production Dilemma: Contraction and Restraint
in the World Combat Aircraft Industry (1994)
Hagerty, Devin T., The Consequences of Nuclear Proliferation: Lessons from South
Asia (1998)
Heymann, Philip B., Terrorism and America: A Commonsense Strategy for a
Democratic Society (1998)
Kokoshin, Andrei A., Soviet Strategic Thought, 191791 (1998)
Lederberg, Joshua, Biological Weapons: Limiting the Threat (1999)
Shields, John M., and William C. Potter, eds., Dismantling the Cold War: U.S.
and NIS Perspectives on the Nunn-Lugar Cooperative Threat Reduction Program
(1997)
Tucker, Jonathan B., ed., Toxic Terror: Assessing Terrorist Use of Chemical and
Biological Weapons (2000)
Utgoff, Victor A., ed., The Coming Crisis: Nuclear Proliferation, U.S. Interests, and
World Order (2000)