Nuclear strategy in the twenty-first century

Nuclear strategy in the twenty-first century

Nuclear Strategy in

the Twenty-First Century

Stephen J. Cimbala

Library of Congress Cataloging-in-Publication Data

Cimbala, Stephen J.

Nuclear strategy in the twenty-first century / Stephen J. Cimbala.

p. cm.

Includes bibliographical references and index.

ISBN 0–275–96869–3 (alk. paper)

1. Nuclear warfare—Forecasting. I. Title.

U263.C493 2000

355.02'17'0905—dc21 99–052984

British Library Cataloguing in Publication Data is available.

Copyright 2000 by Stephen J. Cimbala

All rights reserved. No portion of this book may be

reproduced, by any process or technique, without

the express written consent of the publisher.

Library of Congress Catalog Card Number: 99–052984

ISBN: 0–275–96869–3

First published in 2000

Praeger Publishers, 88 Post Road West, Westport, CT 06881

An imprint of Greenwood Publishing Group, Inc.

Printed in the United States of America


The paper used in this book complies with the

Permanent Paper Standard issued by the National

Information Standards Organization (Z39.48–1984).



Acknowledgments vii

Introduction ix

I. Nuclear History and Its Lessons 1

1. Limited War in the Nuclear Age: Military Frustration and

U.S. Adaptation 3

2. Marching Beyond Marx: The Red Army and Nuclear

Weapons 25

3. The Cuban Missile Crisis and Its Legacy 45

II. Nuclear Policy and Strategy in the Future 67

4. Nuclear Proliferation: Fortuitous Past, Uncertain Future 69

5. Russia and Nuclear Weapons After the Cold War: A

Potemkin Village? 93

6. Nuclear Weapons and ‘‘Third Wave’’ Warfare 129

Conclusion 181

For Further Reading 203

Index 207


It is with pleasure that I acknowledge the following persons for their

ideas, insights and patience in critiquing portions of this manuscript

and/or providing pertinent references: John Arquilla, Robert Batcher,

Richard Betts, Jerome Bracken, Paul Davis, Peter Feaver, Raymond Garthoff,

Daniel Geller, Colin Gray, Jacob Kipp, Fred Nyland, Keith Payne,

James Scouras and Timothy Thomas.

I am also grateful to the following journals for permission to use portions

of my articles: Frank Cass Publishers, ‘‘Information Warfare and

Nuclear Conflict Termination,’’ European Security, Vol. 7, No. 4 (Winter

1998), pp. 69–90; ‘‘Nuclear Crisis Management and Information Warfare,’’

Parameters, No. 2 (Summer 1992), pp. 117–128 (Chapter 6); and

Frank Cass Publishers, ‘‘The Cold War and Soviet Military Strategy,’’

The Journal of Slavic Military Studies, Vol. 10, No. 3 (September 1997),

pp. 25–55 (Chapter 2).

I gratefully acknowledge administrative support from Penn State Delaware

County Campus staff: Charele Raport, George Franz and Edward


The encouragement of and support for this project by Dr. James Sabin,

Greenwood Publishing Group, is very much appreciated.

My wife and children have been the center of my life and the source

of many blessings. They have endured still another professional distraction.

The opinions expressed here do not represent the views of Penn State

University, nor of any U.S. government agency, nor of any person named



No theory or combination of theories is ever going to provide us

with the paradigmatic equivalent of a ‘‘crystal ball,’’ in which we can

perceive the future with the same clarity we take for granted when

we view the past. But theories that successfully explain a system’s

past do not normally lose their validity as they approach, and even

proceed beyond, the present.

—John Lewis Gaddis, The United States

and the End of the Cold War, p. 191 1

Nuclear strategy has an oxymoronic ring. The idea that something as

subtle and nuanced as ‘‘strategy’’ could be related to the use of instruments

as deadly as nuclear weapons sounds almost obscene to experts

and to lay persons alike. ‘‘Nuclear strategy’’ calls to mind the jokes about

‘‘military intelligence’’ or ‘‘smart warfare.’’ Despite understandable skepticism

about their value, nuclear weapons had to be faced. With the

advent of atomic bombs near the end of World War II and their use

against Hiroshima and Nagasaki, the world of military technology had

been changed forever. What changes nuclear weapons would bring

about in the world of military strategy was, in 1945, still arguable.

We now have considerable distance and military hindsight on the

question of how nuclear weapons influenced U.S., Soviet-Russian and

other military strategies and defense policies. The Cold War has ended,

and the Soviet Union is in the past tense. We can look backward more

or less objectively and evaluate what nuclear weapons were good for,

and what not, at least between 1946 and 1991. This is not just an aca-

x Introduction

demic exercise. The purpose of any policy study, likewise this one, is to

draw from past experience in order to learn for the future.

Nuclear weapons remain in the arsenals of the Cold War nuclear powers:

Great Britain, China, France, Russia and the United States. In addition,

in 1998 two states were added to the list of openly acknowledged

nuclear powers: India and Pakistan. Israel is also widely believed to possess

nuclear weapons, and Iran, Iraq, North Korea, Syria and Libya have

been reported in various sources as seeking to acquire nuclear and other

weapons of mass destruction (WMD). The spread of nuclear weapons to

these and possibly other currently non-nuclear states might upset the

applecart of U.S. and allied military strategies and, arguably, make for

a more dangerous and precarious world.

Chapter 1 argues that the first problem presented to U.S. policy makers

in the Cold War was that of limited war. The outbreak of war in Korea

caught U.S. planners, still fixated on World War II experience and programmed

for global conflict against the Soviet Union, off guard. The

United States was obliged in Korea to fight a limited war for limited

objectives under the shadow of nuclear weapons. Some contend that a

veiled threat of nuclear use by the Eisenhower administration might

have expedited the Korean armistice, but other factors were also important,

including the death of Stalin and his succession by a regime less

interested in military confrontation with capitalism. For the remainder

of the Cold War, the United States struggled to define the scope of American

commitments for which conventional war was worth waging on a

large scale. Always in the background was the U.S. relationship with the

Soviet Union and, after 1972, with China. U.S. peacetime defense expenditures

in the Cold War years were unprecedented in American history,

but they might have been even larger without nuclear weapons as an

ultimate deterrent.

The Soviets were also required to adapt painfully to the realities of the

first nuclear age. Chapter 2 covers this story by charting developments

in Soviet military theory and in security policy related to the new conditions

imposed by nuclear weapons. Inevitable war with the capitalist

world was supplanted by the continuing standoff of peaceful coexistence

without war. Under the communist party chairman, Leonid Brezhnev,

the Soviet political and military leadership eventually accepted the reality,

if not the desirability, of nuclear deterrence. Mikhail Gorbachev,

the reformist Soviet president and communist party chairman, moved

further in the direction of military detente based on arms control after

he assumed power in 1985. Gorbachev, influenced by his assessment that

nuclear war was inadmissible as an instrument of Soviet policy, declared

that the international class struggle had been preempted by a new condition

of security interdependence between East and West. Soviet military

leaders, like their American counterparts, acquired much of their

Introduction xi

nuclear learning in fits and starts, as during Cold War confrontations or


Chapter 3 considers once again the already overwritten episode known

to historians as the Cuban missile crisis of 1962. The author cannot claim

to have discovered new archives, or evidence as such, pertinent to the

crisis. But a stronger case can be made for rethinking some of the conclusions

drawn by other investigators, including expert historians and

political scientists who have devoted a great deal of attention to the

missiles of October. In Chapter 3, we suggest that the crisis lends itself

to roseate misconstruction about the utility of nuclear weapons for brinkmanship

and for diplomatic coercion. Cuba 1962 was a crisis that never

should have come about except for considerable political mismanagement

in Moscow and in Washington. Once it came about, it was handled

with less than glittering dexterity by Kennedy and Khrushchev, despite

the laurels showered upon the American president and his advisors in

many studies. Cuba 1962 was a narrow and fortuitous escape from mutual

disaster: understood as such by policy makers, the experiment of

poorly-thought-out nuclear brinkmanship was not repeated for the

remainder of the Cold War.

The Cold War was dangerous enough, but the degree of danger was

limited by the small number of nuclear powers and by the fact that the

two ‘‘superpowers,’’ the United States and the Soviet Union, presided

over an essential bipolar international system. The end of the Cold War

and the probable spread of nuclear weapons adds uncertainty to the

stability of the new world order, as discussed in Chapter 4. Despite the

experience of the Cold War nuclear arms races and occasional military

confrontations, some draw the lesson that nuclear deterrence worked to

stabilize international relations between 1946 and 1991 and that it will

work in the same fashion into the next century. This optimistic view of

an international environment permissive of the spread of nuclear weapons

and yet marked by deterrence and arms race stability is wrong. The

theoretical arguments in favor of proliferation as compatible with stability

are flaccid, and the policy implications of those arguments cannot

be sustained. Instead of a carefully managed nuclear proliferation

marked by stable deterrence and an absence of nuclear fears, the next

century is likely to see nuclear weapons used as ‘‘equalizers’’ against

high-tech conventional powers such as the United States and its NATO


Chapter 5 examines a case of qualitative nuclear proliferation. Russia

has inherited the strategic nuclear arsenal of the former Soviet Union.

We refer to this shift from ‘‘Soviet’’ to ‘‘Russian’’ nuclear forces as a case

of proliferation for the following reason. The United States knew with

some confidence that the nuclear forces of the former Soviet Union were

at all times under firm party and regime political control, at least until

xii Introduction

the abortive coup of August 19–21, 1991. On the other hand, the future

of civil-military relations in the new Russia is open ended. Russia’s nuclear

weapons are post–Cold War political and military two-way streets.

Since Russia’s conventional forces have disintegrated for want of virtually

everything since 1991, nuclear weapons remain Russia’s sole claim

to major power status in Eurasia. But Russia retains nuclear weapons far

in excess of her own military needs for deterrence, even if Russia’s new

military doctrine allows for nuclear first use in the face of some forms

of conventional attack. Russia should conclude START arms reduction

agreements with the United States that could contribute to stable deterrence

at lower levels, and both states should seek to improve political

relations to the point at which deterrence, at least the nuclear kind, becomes


Chapter 6 considers the coexistence of nuclear deterrence and ‘‘third

wave’’ or postindustrial warfare and wonders how peaceful that coexistence

can be. Nuclear deterrence theory and nuclear strategic thinking

were products of second wave civilization, or the industrial age. As we

enter the ‘‘third wave’’ society, driven by information and electronics,

nuclear and other weapons of mass destruction seem as obsolete as Tyrannosaurus

Rex, and so they are, at least in terms of waging discriminate

war while limiting collateral damage against targets not expressly

aimed at. The Gulf War of 1991 was the first window on the future of

long-range, precision-guided, advanced conventional weapons, and on

the knowledge-intensive technologies for seeing and comprehending the

battlefield that will support those weapons in major conflicts. This is a

war form in which the United States and a few other high-tech states

have progressed beyond the present capabilities of any aspiring peer

competitors. This advantage in information-based warfare may have a

dark side, however. The attributes of information warfare may, in time

of crisis between nuclear armed opponents, play havoc with the requirements

for stable deterrence. Eager ‘‘infowarriors’’ could contribute to

a failure of deterrence based on a mistaken decision for preemption, an

accidental or inadvertent nuclear escalation or other undesired side effects

of combining nuclear deterrence with info-compellence.

The conclusion advances a number of arguments that make inferences

about the nuclear future based on the nuclear past. Some of these judgments

are very counterintuitive to current thinking about policy and

strategy, at least as it appears in the counsels of government and in the

prestige journals. So be it. The future of nuclear weapons, as spelled out

here, is that their main significance is not to threaten Armageddon in an

instant. Instead, the uses to which the more numerous nuclear arsenals

of the future will be put, compared to the past, will be highly nuanced

and situation specific. 2 Deterrence will be only one among many combinations

of policy and strategy pursued by a more heterogeneous cast

Introduction xiii

of actors armed with weapons of mass destruction. 3 The goals for which

their WMD will be threatened or used will be as diverse as are the reasons

for state and non-state actors going to war in the past: power, revenge,

hatred, glory, reasons of state so defined, ethnic rivalry, religious

intolerance and all the rest. 4 The world after the Cold War returns to

‘‘normalcy,’’ (i.e., a condition of Hobbesian competition and frequent

outbreak of war in the seams of the international system). The pertinent

question is whether the center can hold.


1. John Lewis Gaddis, The United States and the End of the Cold War: Implications.

Reconsiderations. Provocations (New York: Oxford University Press, 1992).

2. A more complex international system, compared to the Cold War, will demand

of theorists and policy makers a willingness to juggle more than one model

of causation or paradigm simultaneously. This may result in paradigm redundancy,

but as historian John Lewis Gaddis has noted, paradigm redundancy is

better than paradigm ‘‘fratricide’’ in dealing with complex social behavior. See

Gaddis, The United States and the End of the Cold War, p. 191.

3. Keith B. Payne, Deterrence in the Second Nuclear Age (Lexington: University

Press of Kentucky, 1996), passim.

4. On this point see Samuel P. Huntington, The Clash of Civilizations and the

Remaking of World Order (New York: Simon and Schuster, 1996), esp. pp. 95ff.;

and Martin Van Creveld, The Transformation of War (New York: The Free Press,

1991), passim.

Part I

Nuclear History and Its Lessons

Chapter 1

Limited War in the

Nuclear Age: Military Frustration

and U.S. Adaptation


One of the most important challenges posed for U.S. and Soviet military

thinkers was the adjustment from the total war mentality that had been

required to defeat the Axis coalition in World War II. As the numbers

of U.S. and Soviet nuclear weapons and intercontinental delivery systems

multiplied, it became apparent that a world war or a war in Europe

could not be fought to a politically acceptable outcome. This meant that

generals and admirals on both sides of the Cold War had to face the

wrenching possibility of fighting wars for limited political objectives with

limited military means. The idea did not go down easily in either the

East or the West.

In this chapter I review some of the challenges presented by limited

war to U.S. political leaders, military planners and academic strategists

by the unexpected setting of the Cold War. (The Soviet adaptations in

strategy as a result of the nuclear age, and what their nuclear learning

portends for the future, will be addressed in the next chapter.) I first take

the story more or less chronologically from Korea through the Gulf war

of 1991. It was a clear case of on-the-job learning, and very much against

the instinctive American way of war. 1 It also holds portents for the future

of American military art under an equally stressful transition forced by

technology: from reliance upon weapons of mass destruction, and deterrence

by threat of punishment, to reliance upon weapons of precision

aim and reduced collateral damage, with deterrence based on threat of

denial of opposed military aims.

4 Nuclear History and Its Lessons


The extension of U.S. peacetime defense commitments to Western Europe,

followed by the stationing of permanent American garrisons there,

was a politico-military strategy for Cold War competition. But it was

also a strategy for freezing the status quo in the center of Europe, thereby

reducing the risk of inadvertent war between the United States and the

Soviets. 2 NATO was to symbolize the absence of proclivities among the

British, French and Germans to fight with one another as a by-product

of its importance for deterring Soviet attack. NATO was also to reassure

the Europeans against another U.S. cop-out if Europe’s calm were threatened

by another aspiring hegemon. Although not fully appreciated even

now, NATO’s political roles were as important as its military one. Most

U.S. foreign policy influentials did not anticipate an actual shooting war

in Europe during the latter 1940s or early 1950s. As George F. Kennan

had anticipated, what was more probable was the slow squeeze of Kremlin

pressure against American and allied interests both directly, as in the

Berlin crisis of 1948, and through surrogates, as in Korea in 1950.

Prior to the outbreak of the Korean War, the Truman administration

had a hard sell for military buildup, including a rapid expansion of the

U.S. nuclear arsenal. NSC-68, a high-level policy study calling for major

U.S. rearmament in view of an imminent Soviet military threat to Europe

and Asia, had been completed shortly before the eruption of North Korea’s

forces across the 38th parallel in June 1950. 3 Chinese entry into the

war only convinced many Americans that a Sino-Soviet bloc now threatened

U.S. global interests. However, Korea was an improbable war for

which American strategic planners had scarcely prepared. Expecting a

global war against the Soviet Union begun in Europe, planners had given

little consideration to the possibility of U.S. involvement in limited wars

supported by the Soviet leadership but fought by other governments and

forces outside of Europe.

Korea posed strategic and policy-making dilemmas in Washington.

The Truman adminstration’s decision to fight a limited war was controversial

on several grounds. Field commander Douglas MacArthur chafed

at political restrictions on military operations. Truman neglected to ask

for a formal declaration of war against North Korea or against China

after Chinese troops later entered the fighting on the Korean peninsula.

The war was fought under the auspices of a United Nations collective

security operation. Since the precedent had been set for commitment of

U.S. forces to limited war without a congressional declaration of war,

the precedent would be repeated to disastrous effect in Vietnam. Nor

was the U.S. intervention in Korea exemplary of truly multilateral collective

security operations, since it was in fact a U.S. military operation

terminated according to U.S. requirements. Thus it provided no model

Military Frustration and U.S. Adaptation 5

Table 1.1

Historical Trends in U.S. Defense Spending (DOD Budget Authority in

Billions of Constant 1989 Dollars)

Source: Lawrence J. Korb, ‘‘Where Did All the Money Go?,’’ ch. 1 in Stephen J. Cimbala,

ed., Mysteries of the Cold War (Aldershot, England: Ashgate Publishers, 1999).

for future uses of U.S. military power on behalf of collective security

missions. To the contrary, the Korean War led to the militarization of

containment and to the hardening of Cold War fault lines between the

communist and capitalist worlds.

The war in Korea also opened the door to unprecedented U.S. peacetime

defense budgets. Despite the fact that it was deliberately limited in

geographical scope and in military escalation, the war was expensive,

and defense spending never really reverted to the fiscal austerity of the

Truman administration prior to 1950. Table 1.1 depicts the growth of

U.S. defense budgets from FY 1950 through 1980.

The Korean War was treated in American military thought and doctrine

as an exception and an aberration, and few appropriate lessons

about the attributes of limited war fighting were drawn from it. The

availability during the Eisenhower administration of larger numbers of

nuclear weapons supported the shift to a declaratory strategy for general

war of massive retaliation. While administration officials were eventually

forced to retreat from this formulation in cases of less than total war, for

global war against the Soviet Union, Eisenhower defense planning relied

mainly upon promptly delivered and massive air atomic offensives. Special

study committees such as the Gaither Committee pointed to the need

for a larger menu of military responses, and Army officials chafed at the

allocation of defense resources within arbitrary ceilings and under plan-

6 Nuclear History and Its Lessons

ning assumptions favoring Air Force and Navy procurement. NATO’s

declared objective of 96 active duty and reserve divisions was far beyond

any commitment its members were actually willing or able to provide.

Thus, reliance on nuclear weapons for extended deterrence became all

the more necessary as a result of allied as well as U.S. domestic budgetary


The Army emerged from the 1950s as the fourth wheel of a defense

establishment whose preferred military doctrines favored the more technical

and less manpower-intensive arms of service. Under the Kennedy

administration things would soon change. Kennedy preferred the strategy

which became known as flexible response, calling for improved U.S.

conventional forces for crisis response, forward presence and, if necessary,

actual war fighting in order to raise the nuclear threshold in Europe.

This last rationale was pushed hard within NATO by Defense

Secretary Robert McNamara, to the detriment of alliance solidarity on

doctrine until the French departure from NATO’s military command

structure in 1966 and the promulgation of flexible response in 1967. Flexible

response arguably allowed a greater role for the ground forces in

U.S. military doctrine and force planning, but by the time flexible response

became official NATO doctrine, the lines between Cold War

‘‘East’’ and ‘‘West’’ had solidified and neither side seemed interested

even in limited probes against the other.

Civilian and military strategists, as well as some policy makers who

recognized the inappropriateness of massive retaliation for other than

all-out nuclear war, struggled during the latter 1950s and early 1960s to

define a concept of selective military strategy suited to the variety challenges

to U.S. security. Robert Endicott Osgood, for example, called for

increased sophistication in U.S. academic and public understanding of

the requirements of limited war in the nuclear age. 4 Henry A. Kissinger,

academic strategist later to serve as national security advisor to President

Richard M. Nixon and as Nixon’s secretary of state, examined the potential

role of nuclear weapons in U.S. limited war strategy. 5 Thomas C.

Schelling applied bargaining theory to the study of military strategy in

several influential works. 6 William W. Kaufmann explained the military

strategy of the Kennedy administration as an effort to provide an extended

menu of military options even for nuclear war. 7 Maxwell Davenport

Taylor, writing as former U.S. Army chief of staff, critiqued

Eisenhower’s strategy as negligent of preparedness for limited war and

as insufficiently attentive to the needs of U.S. ground forces. 8 These and

other varieties of limited war theory were not without shortcomings:

Limited-war theory had numerous flaws. It was primarily an academic, rather

than a military concept, and it drastically misunderstood the dynamics of war.

Its authors seemed to say that since limited war was mainly about bargaining

Military Frustration and U.S. Adaptation 7

and diplomacy, it required no knowledge of military matters and indeed military

considerations should not affect its conduct...In terms of bargaining theory,

moreover, they (limited war theorists) assumed a greater capacity than was warranted

on the part of a gigantic bureaucracy like the United States government

to send clear, precise signals, and they reduced the behavior of potential enemies

to that of laboratory rats. 9

If strategic deadlock reigned in Europe, Khrushchev’s insistence that

wars of national liberation could be unleashed against Third World regimes

supportive of U.S. policy called forth from the Kennedy administration

a burst of doctrinal innovations. Special operations and

low-intensity conflict studies, as the term was later denoted, led to an

emphasis on subconventional warfare, psychological operations and nation

building as constituent elements of U.S. military strategy. 10 But only

a fringe of the armed forces officer corps, such as the Green Berets, committed

themselves to careers along these lines. The more traditional arms

of service lacked serious interest in special operations and regarded their

counterinsurgency brethren with undisguised distaste. As the U.S. commitment

to Vietnam escalated well beyond the engagement of special

operations forces and intelligence operatives, conventional military

mind-sets displaced the political side of the politico-military equation on

which special operations had been predicated. U.S. conventional forces

in Vietnam, on the evidence, fought well against North Vietnamese conventional

forces and Viet Cong units when the latter were willing to

stand and fight pitched battles.

However, it became apparent by 1968 even to the Department of Defense

that the United States could not win the counterinsurgency or conventional

wars at an acceptable cost: Johnson’s resignation and Nixon’s

phased disengagement followed. Having decided that escalation from

limited commitment to a major U.S. military campaign in South Vietnam

was necessary, President Johnson nonetheless sought to balance the requirement

for military escalation against his other priorities in domestic

politics, especially his cherished Great Society programs recently passed

by Congress. Johnson’s ‘‘guns and butter’’ policy filled the armed forces’

enlisted personnel requirements by expanded draft calls instead of mobilizing

the reserves. The result of this approach was to create nationwide

dissent against the war, first across U.S. college campuses, and then

in other segments of the general population.

The domestic turbulence on the home front, in part due to Johnson’s

lack of any apparent strategy for victory, brought the U.S. military escalation

in Vietnam to a stopping point. When U.S. Commander-in-Chief

William Westmoreland asked for several hundred thousand additional

troops in 1968, then Secretary of Defense Clark Clifford suggested to

Johnson that he pull the plug. Johnson did so, announcing his intention

8 Nuclear History and Its Lessons

not to seek another term of office and thereby conceding the failure of

U.S. policy and strategy in Vietnam. However, Johnson left the nation

with a major force and policy commitment to a war which would continue

without complete U.S. disengagement until 1973, and with war

between Vietnamese until 1975. With military disengagement from Vietnam

went another look at U.S. conscription policy, and the Gates Commission

recommendation to end conscription was adopted and ordered

into effect beginning in 1983. In effect, the United States had come full

circle to its pre-twentieth-century peacetime standard of raising armed

forces by voluntary enlistment (except for the American Civil War, when

both sides drafted).

The onset of the All-Volunteer Force coincided with post-Vietnam doctrinal

revisionism. The Nixon administration changed the 1960s strategy

of being able to fight two and one-half wars simultaneously to one and

one-half wars, and Nixon emphasized that U.S. support for besieged

allies would stop short of involving American ground forces. Voluntary

enlistment dictated a strategy of selective rather than ubiquitous military

engagement. Selective engagement was also facilitated by the full-blown

emergence of U.S.-Soviet detente during the 1970s and Sino-American

rapprochement. It was perceived by U.S. foreign and defense policy elites

that diplomatic containment of Moscow’s ambitions was more cost effective

than overpromising of U.S. military involvement in regional conflicts.

U.S. and Soviet leaders worked to stabilize the Middle East and to

create new expectations about their mutual interests in avoiding nuclear

war and inadvertent military escalation. In addition, under the direction

of Chief of Staff Gen. Creighton Abrams, Army planners during the early

1970s configured the ‘‘total force’’ concept so that future presidents could

not avoid substantial reserve call-ups during any national mobilization

for war. 11

U.S. military planners who contemplated how to prevail in a war between

NATO and the Warsaw Pact, and who were confounded by the

commingling of conventional and nuclear forces in Europe, found a more

amenable theater of operations for the application of U.S. military power

in the Gulf War of 1991. A five months’ period of grace for military

buildup in Saudi Arabia did no harm to U.S. readiness for war in January

1991, and U.S. air-land battle doctrine played successfully before a

packed house. The results of the Gulf War of 1991 seemed to vindicate

not only U.S. conventional military strategy and technology, but also the

decision in favor of the all-volunteer force taken decades earlier. (The

subject of selective military strategy and the Gulf War is taken up specifically

in Chapter 3). Columnist Charles Krauthammer, celebrating the

‘‘unipolar moment’’ in which the United States had allegedly found itself

by virtue of the collapse of the Soviet Union, noted that

Military Frustration and U.S. Adaptation 9

Table 1.2

Department of Defense Personnel (End of Fiscal Year, in Thousands)

Source: Adapted from William J. Perry, Secretary of Defense, Annual Report to the President

and the Congress (Washington, D.C.: U.S. Government Printing Office, 1996), p. 254,

table V-4.

in 1950 the U.S. engaged in a war with North Korea: it lasted three years, cost

54,000 American lives, and ended in a draw. Forty-one years later, the U.S. engaged

in a war with Iraq: it lasted six weeks, cost 196 American lives, and ended

in a rout. If the Roman Empire had declined at this rate, you would be reading

this in Latin. 12

However, experts recognized the ironical character of the vindication of

U.S. strategy. The United States was already in a defense build-down

initiated by President George Bush prior to Iraq’s attack into Kuwait;

revisions resulting from President Bill Clinton’s ‘‘Bottom-Up Review’’

would diminish active duty forces even more. For example, Table 1.2

summarizes U.S. military active and reserve personnel end strengths

from 1987 through 1997 (as projected in 1996).

U.S. forces were nevertheless required, according to Clinton’s national

military strategy published in 1996, to deter or, if necessary, to fight with

allies and ‘‘decisively win’’ in two ‘‘nearly simultaneous’’ major regional

conflicts (MRCs). 13 The option of fighting and holding on one front while

prevailing on another, the initially proferred one and one-half or ‘‘winhold-win’’

strategy first adumbrated by former Defense Secretary Aspin,

was shot down by Pentagon and congressional opponents. Some dubbed

it ‘‘win-lose-lose’’ or ‘‘win-hold-oops.’’ 14 In addition, the Congress

and some politico-military strategists in the executive branch were also

10 Nuclear History and Its Lessons

planning to employ U.S. military capability for nontraditional or noncombat

missions, including operations designed to preserve sanctuary

from attack for besieged ethnic or national populations (such as operation

Provide Comfort for the Kurds in Iraq). As well, both Bush and

Clinton strategies for more traditional uses of U.S. military power emphasized

the performance of forward presence and crisis response missions

intended for regional contingency operations outside of Europe,

not for global warfare or for large interstate wars in Europe.

Some experts contended that the Clinton defense program based on

Aspin’s Bottom-Up Review was already underfunded by 1995, placing

an expected crimp in future modernization and other aspects of support

necessary for the two nearly simultaneous MRCs strategy. 15 Real

(inflation-adjusted) Department of Defense (DOD) spending between

1990 and 1995 fell by 25 percent, around $85 billion in 1995 dollars. The

Clinton Future Years Defense Program (FYDP) introduced in February

1994 proposed cutting real defense spending by another 10 percent between

1995 and 1999, a 1995–1999 reduction of about $25 billion. If enacted

exactly as proposed, the February 1994 FYDP would have

decreased the share of U.S. gross domestic product (GDP) devoted to

national security to 2.9 percent—the lowest share of GDP devoted to

national security since before World War II. 16 The Clinton administration

in December 1994 asked Congress for an additional $25 billion for defense

in the period 1996–2001, of this, some $10 billion would be spent

during the 1996–1999 period against any shortfall. This and other proposed

adjustments to the original FYDP would, according to the Congressional

Budget Office, reduce its estimate of the DOD shortfall to

around $47 billion, about 4 percent of total planned spending in 1995–

1999. 17 The U.S. military force structure projected for September 1997 and

expected to meet the requirements for two nearly simultaneous MRCs is

summarized in Table 1.3, compared to earlier baseline years.


Whence the Threat?

According to military historian Martin Van Creveld, large scale conventional

war is mostly obsolete. The future of warfare lies in low intensity

conflict, terrorism and the like. 18 The reasons are that large scale

warfare does not pay political and military dividends relative to its costs.

Van Creveld argues that this trend represents the dismantling of the

Clausewitzian paradigm which dominated military strategy formation

from the Peace of Westphalia until the end of World War II. In Clausewitz’s

model of the relationship between war and policy, a trinitarian

unity among people, government and army supported the conduct of

Table 1.3

U.S. Conventional Force Structure (Selected Years, Compared with Bottom-

Up Review)

Note: Army reserve totals exclude fifteen enhanced readiness brigades; Marine Corps totals

include one reserve Marine Corps division.

Source: Adapted from U.S. Congress, Congressional Budget Office, Reducing the Deficit:

Spending and Revenue Options (Washington, D.C.: U.S. Government Printing Office,

August 1996), p. 99, table 3-1.

12 Nuclear History and Its Lessons

war on behalf of state interests. War was something done on behalf of

the state, and only on behalf of the state. This was an important marker

in Western military and political history, for it made possible, along with

the political theories of Machiavelli and Hobbes, the ‘‘realist’’ tradition

in international politics which informed generations of scholars and students.

This prejudgment of the obsolescence of Clausewitz’s theory of war,

and the assumption that the territorial state is headed for the ash heap

of military history, should not be accepted without question. As to Clausewitz’s

theory, as explained in his great work, Vom Kriege (On War), it

is not a ‘‘theory’’ in the form of a list of timeless axioms or answers

about war. It is, instead, a ‘‘theory’’ as a set of guiding questions for

instruction. Clausewitz’s purpose is to open the mind of the soldier to

the subtleties of the scholar, and that of the scholar, to the uncertainties

and hard realities of soldiering. Therefore, his description of the essence

of war as consisting of violence, chance and reason remains pertinent to

the conduct of war in modern times, whether by state or non-state actors.

19 Clausewitz’s insistence upon the subordination of military ends

and means to policy loses none of its relevance either, even if the monopoly

of the state over the use of violence becomes less habit forming

after the Cold War. 20

Two points are noteworthy here. First, the monopoly of the territorial

state over the legitimate use of military force is not the same as its ability

to monopolize all uses of force. State assertions of legal monopoly have

coexisted with antistatist and other violence on the part of individuals

and groups, notwithstanding state claims to the exclusive right to use

military power. Internal wars of various descriptions, including rebellions

intended to overthrow the state itself, have contested the legitimacy

of states and their military monopolies from the very founding of the

modern state system to the present.

Second, it is not entirely clear that wars between states have been

supplanted by wars within states, if the criterion being used is one of

the overall significance for the international system as a whole. It cannot

be denied that, since 1945, there have been many more wars within states

than between states. 21

But it cannot be inferred from the greater frequency of internal wars

that states are less important politically: in some ways, the rising frequency

of intrastate wars is testimony to the growing, not the weakening,

significance of states as international actors. Many of these intrastate

wars have resulted from the desire of ethno-national, religious and other

groups who feel displaced within the confines of their existing political

order for a state of their own. Palestinians and Kurds provide excellent,

contemporary examples. The state of not suffering from apparent lack of

demand, but from insufficiency of supply relative to demand. In addi-

Military Frustration and U.S. Adaptation 13

tion, interstate war has changed its cast of characters in the last half of

the twentieth century: relatively speaking, it is no longer the resort of

the great powers, but of the medium and small powers. 22 Whether this

is an artifact of the Cold War that will continue indefinitely into the post–

Cold War international system, or a more permanent feature of international

politics, is a subject for reasonable disagreement.

Unless the next few decades are less turbulent than many suppose,

U.S. policy makers will be forced to consider the involvement of U.S.

forces in both conventional and unconventional wars. These ‘‘wars’’ may

include multinational peace operations, for which the U.S. Army has

now written official doctrine. Whether the U.S. armed forces will be able

to wage high intensity warfare, low intensity conflict, both, or neither in

the future is as much dependent on the American public’s understanding

of the American way of war as it is on the military-technical issues such

as force size and weapons modernization. Public perceptions of international

threats are notoriously fickle, and Cold War policy makers deliberately

exaggerated the degree of threat in order to obtain defense

commitments from the United States Congress. One defense scholar argues

that the United States was virtually free from serious threat of invasion

and conquest, slow strangulation through global blockade or

nuclear attack during the Cold War years. 23 Robert J. Art compared

threats to U.S. security during three eras, summarized in Table 1.4. 24

Many would argue that Art’s definitions of possible dangers in the

Cold War and post–Cold War eras are adequate to address worst-case

scenarios. But at levels of threat or potential conflict below those worst

cases, Cold War experience and events since 1990 offer less reassurance

that other security objectives can be guaranteed at an acceptable cost. As

John Lewis Gaddis has noted:

Victories in wars—hot or cold—tend to unfocus the mind. They encourage pride,

complacency, and the abandonment of calculation; the result is likely to be disproportion

in the balance that always has to exist in strategy, between what one

sets out to do, and what one can feasibly expect to accomplish. It can be a dangerous

thing to have achieved one’s objectives, because then one has to decide

whattodonext. 25

What Kind of War?

There are some cautions that we can derive from Cold War history.

One caution is that forces which are optimized for high intensity conflict

against industrial strength armies cannot simply be reduced in size and

reassigned to low intensity warfare. During the 1960s and prior to the

Vietnam escalation, for example, it was assumed by planners that forces

adequate for war between NATO and the Warsaw Pact would easily

14 Nuclear History and Its Lessons

Table 1.4

Threats to U.S. Security in Three Eras

Source: Robert J. Art, ‘‘A Defensible Defense: America’s Grand Strategy After the Cold

War,’’ International Security, No. 4 (Spring 1991); p. 11, table 2. I have slightly revised

Art’s category labels with no effect on his meaning.

brush aside smaller and less heavily armed foes. 26 It is now acknowledged

that low intensity conflict or unconventional warfare, including

counterinsurency and counterterrorism, is qualitatively different from

larger scale warfare. 27 One result of the U.S. military being forced to face

the unique environment for unconventional warfare was the reorganization

of special operations forces, under congressional prodding, as depicted

in Table 1.5.

The creation of a unified U.S. military command for special operations

forces was a long, drawn-out battle that terminated in a series of compromises

between DOD and Congress in 1987. The way was prepared

for a unified command for special operations (USSOCOM) by the landmark

Goldwater–Nichols legislation of 1986. Goldwater–Nichols was a

congressional mandate to the Pentagon to get its act together on ‘‘jointness.’’

It increased the role of the Chairman, JCS as the principal military

advisor to the President and the Secretary of Defense and made the Joint

Staff responsible to the Chairman. The same legislation also empowered

the various unified commanders-in-chief (CINCs) with more complete

authority over forces assigned to their commands, regardless of service

Table 1.5

U.S. Special Operations Command Organization

Note: SEAL refers to elite Navy commando units (SEa, Air, Land).

Sources: Adapted from U.S. Special Operations Command, History (Tampa, Fla.: MacDill

AFB, HQ USSOCOM, September 1998), pp. 3–6; U.S. Special Operations Forces, Posture

Statement: 1998 (Washington, D.C.: Office of the Assistant Secretary of Defense (Special

Operations/Low Intensity Conflict), 1998), pp. 47–53; and Sam C. Sarkesian, ‘‘Special

Operations, Low Intensity Conflict (Unconventional Conflicts), and the Clinton Defense

Strategy,’’ in Stephen J. Cimbala, ed., Clinton and Post–Cold War Defense (Westport,

Conn.: Praeger Publishers, 1996).

16 Nuclear History and Its Lessons

of origin. Norman Schwartzkopf as theater commander and Colin Powell

as Chairman, JCS would later put this legislation to good effect in the

Gulf War of 1991.

USSOCOM was also backed up with institutional support in the Pentagon,

an assistant secretary of defense for special operations and low

intensity conflict (ASD/SOLIC). Additional support was provided by the

establishment of a new Major Force Program (MFP 11) specifically for

SOF (its ‘‘checkbook’’). 28 The creation of USSOCOM and ASD/SOLIC

was propelled by necessity as well as by choice. The failed attempt to

rescue U.S. hostages in Iran in 1980 showcased problems of force preparedness,

command/control and ‘‘interoperability’’ (force elements

from different arms of service being able to coordinate the timing and

character of their military operations). Some of these problems repeated

themselves during the U.S. invasion of Grenada in 1983 (Operation Urgent

Fury). An alliance of former special operations ‘‘mafia,’’ members

of Congress dissatisfied with the status quo, and well-placed officials in

the executive branch worked the system to establish a supportive and

promotive niche for special operations forces under a unified command.

29 These developments in special operations organization and command

amounted to more than moving the foxes to different boxes. The

changes in organization and command made possible improvements in

training and performance that paid dividends through the remainder of

the century.

A second cautionary note derived from Cold War experience is that

low intensity conflicts involve ambiguous political missions for which

U.S. popular support cannot be assumed and must be assiduously built.

A third lesson is that the U.S. armed forces’ sense of military professionalism

may be compromised by missions outside the competency of military

training and experience. 30 Assigning to military forces the mission

of ‘‘nation building’’ confuses a military mission with a broader political

one, to the probable detriment of both military and political objectives.

Nonetheless, the post–Cold War world has already found the U.S. military

involved in a number of peacekeeping, peace enforcement, humanitarian

assistance and other nontraditional operations.

The need to fight syncretic wars which are simultaneously conventional

and unconventional in one sense drives U.S. military historians

and planners back to the Revolutionary War roots of American military

practice. As historian Russell Weigley has noted, General George

Washington preferred to model the Continental Army along the lines

of eighteenth-century European military forces. 31 Washington feared that

irregular forces could not be counted on against British regular forces,

and he also remained wary of the potential costs to the American social

fabric of guerrilla warfare. Even his postwar efforts to shape the peacetime

U.S. armed forces favored a small regular army supported by a

Military Frustration and U.S. Adaptation 17

compulsory service and federally regulated militia. 32 On the other hand,

America’s revolutionary war against Britain also included successful U.S.

unconventional campaigns against British regulars, such as the guerrilla

attacks on Burgoyne’s lines of communication and flanks contributory

to his defeat at Saratoga. 33 U.S. professional military heritage from the

War of 1812 was also a mixed estate. On one side stood the Battle of

New Orleans, suggesting that citizen-soldiers could fight with distinction

against regular British forces. On the other side stood the battles of Chippewa

and Lundy’s Lane, in which American regulars acquitted themselves

well against their British counterparts in open field battles without

use of unconventional tactics. 34

A Large and Diverse Menu: Recipe for Indigestion?

Major theater wars in the Middle East/Southwest Asia and/or Northeast

Asia remain among the dominant scenarios that drive Pentagon

planners, even a decade after the end of the Cold War. The Gulf War of

1991 was either the concluding campaign of the Cold War or the first of

the post–Cold War era. It was, at the operational-tactical level, a one

sided campaign: crushing blows against Iraqi armed forces, command

systems and military infrastructure from the air, followed by a short

ground campaign based on strategic flanking movements. 35 The war

demonstrated that the demise of the Soviet Union left the United States

in a unique position of undisputed conventional and military superpower.

36 No state could threaten the United States with large scale military

defeat, although arms control was still necessary to winkle out of

the former Soviet Union most of its residual nuclear arms. If U.S. nuclear

and conventional military power made it all but invulnerable to any

campaign of annihilation, its maritime supremacy as of 1991 seemed to

ensure against the success of any hostile war of attrition. The odor of

unipolarity was indeed in the air as the Bush administration vacated the

White House, but Bush initiated an ill-fated intervention in Somalia that

brought American self-assessments back to earth.

A large menu of problems remained for U.S. policy makers after the

Cold War. Many of these problems will call for military forward presence,

peacetime engagement and other missions not previously tasked

during the Cold War fixation on a one variant war. The end of the Cold

War requires that the United States place less emphasis upon forces designed

to deter global war or to fight a major coalition war in Europe.

Post–Cold War U.S. forces will be smaller in size, contingency oriented

instead of scenario dependent, and not infrequently committed to peacekeeping

and other collective security missions, under UN or other auspices.

37 According to RAND defense expert Paul K. Davis, the concept

18 Nuclear History and Its Lessons

of ‘‘environment shaping’’ now moves to the center of adaptive U.S.

defense planning:

The United States is now concerned at least as much with environment shaping

(e.g., encouraging regional stability or peaceful change, and reducing inventives

for other nations to seek superpower status) as with more traditional military

missions. Consistent with that, Cheney (Bush’s Secretary of Defense Dick Cheney)

and Aspin (Clinton’s first Secretary of Defense, Les Aspin) reaffirmed the

need for significant U.S. forward presence in critical regions worldwide because

of the unique opportunities and responsibilities the nation now has and because

it is so strongly in the U.S. interest to avoid the kinds of regional instabilities

that might ensue if power vacuums arose—e.g., a military competition among

Japan, China and Korea in the Far East, or various worrisome possibilities in

Europe. 38

The concept of environment shaping implies that the United States

should take a combination of political and military proactive measures

to head off crisis and wars before they happen. The focus of defense

planning on specific threats that the United States might want to deter

shifts to flexible preparedness for uncertain destabilizing events that policy

makers want to avoid or contain. For example, neither by itself nor in

cooperation with allies (NATO) could the United States have deterred

the outbreak of war among ethno-national communities in former Yugoslavia

in 1991–1992. On the other hand, NATO’s peace enforcement

operation Joint Endeavour, begun in 1995 in Bosnia, and its Operation

Allied Force military campaign against Yugoslavia in 1999 did attempt

to establish limits on politically motivated genocide and its derivative

refugee problems.

One problem with environmental shaping is that it does not necessarily

require fewer forces compared to some Cold War levels. Environmental

shaping presupposes some U.S. forward presence, especially

maritime, in areas of potential instability, and a variety of composite

forces with variable schemes of assembly for rapid insertion and rapid

reaction. Forces of this sort are not necessarily cheap, especially per unit.

In addition, the size of U.S. forces by fiscal year 1997, according to Department

of Defense projections, will be significantly less than the status

of active duty forces in 1990 (see Table 1.2).

The difficulty of using U.S. military force under any circumstances

may increase, relative to Cold War precedent. In the early years of the

twenty-first century, political leaders may expect to use military force in

at least five kinds of situations:

• major regional conflicts (MRCs) or theater wars (e.g., in the Persian Gulf or


Military Frustration and U.S. Adaptation 19

• operations ‘‘other than war’’ (OOTW), including traditional peacekeeping,

peace enforcement, humanitarian rescue, and complex contingency operations

(complex contingency operations are those in which a humanitarian operation

is combined with a requirement for some other operation);

• counterinsurgency or insurgency support;

• counter-proliferation efforts;

• anti-terrorist raids in order to free hostages, to capture terrorist leaders or to

destroy identifiable bases of support for terrorist organizations.

Of these, the involvement in major regional conflicts has the greatest

potential for prolonged fighting and high casualties. Historical U.S. experience

and recent (1993) events in Somalia suggest that policy makers

and military planners can count on little forgiveness in Congress or in

the American public once U.S. casualties begin to accumulate, especially

in wars for which an apparent national interest is not obvious. Concern

about lack of public tolerance for high casualty levels was one reason

why NATO’s Operation Allied Force campaign against Yugoslavia in

1999 excluded the option of a ground offensive, although doing so publicly

was unnecessarily informative to the Serbs. Allied Force was something

in between a major regional conflict (air power from sixteen of

NATO’s nineteen member states was involved, with the United States

deploying some 700 of about 1,000 allied planes) and a peace enforcement

mission (to compel the regime of President Slobodan Milosevic to

desist in ethnic cleansing and to permit repatriation of Albanians, previously

displaced from Kosovo, under international supervision). As

such, it had fragile U.S. public support and uncertain congressional reaction

to the Clinton administration’s commitment of U.S. policy to

NATO’s most extensive military operation ever.

A U.S. military strategy based on the deterrence of major regional

conflicts may be as controversial as the containment strategy which preceded

it. 39 Declining U.S. defense budgets will create difficult trade-offs

among the desired goals of preserving force size, modernizing weapons

and C3, maintaining readiness for crisis response, and preserving sustainability

for protracted conflict. On account of constrained U.S. resources,

American forces will be dependent upon international coalitions

for the conduct of major military operations, as they were in Desert

Storm and in Allied Force. In addition, ‘‘to the extent that deterrent strategies

form part of the response to today’s threats and challenges, they

will increasingly be implemented multilaterally and economically.’’ 40

Beyond resource constraints, Desert Storm also suggests a qualitative

dilemma for U.S. military planners. Precision guided munitions, improved

communications and command/control, and reconnaissancestrike

complexes will make possible, in theory, the selective targeting of

20 Nuclear History and Its Lessons

enemy military assets while minimizing collateral damage. 41 In practice,

as opposed to theory, the economies and social fabrics of Third World

states may be so fragile that the ‘‘precision’’ possible in high-technology

warfare is of no meaning to those on the receiving end, nor to observers

of U.S. and allied efforts. Thus, efforts to maintain specific political and

military limitations in war, and to derive from those limitations carefully

laid down strike plans, could fail unavoidably and to the detriment of

coalition management. 42

The larger question of when and where the United States and allied

NATO forces should intervene remained unresolved at century’s end.

Military intervention during the Cold War was largely dictated by requirements

flowing from U.S. containment policy and by the requirements

of deterrence and defense in Europe. The post–Cold War world

is already a wild card of nonspecific and geographically distributed

threats. In the aftermath of NATO’s eleven-week air war against the

security forces of Yugoslavia, the Clinton administration sought to make

explicit a revised doctrine on military intervention. The Clinton security

team sought to develop new guidelines that went beyond the ‘‘Powell

doctrine’’ (Gen. Colin Powell, formerly chairman of the JCS under Bush).

The Powell doctrine called for intervention only when vital interests

were threatened, when overwhelming force would be committed, and

when broad public and congressional support could be guaranteed. In

contrast, a post-Kosovo ‘‘Clinton doctrine,’’ according to national security

advisor Samuel R. Berger, would call for intervention when (1) a

compelling national interest exists, (2) there is a moral imperative to act

and (3), the United States and/or its allies have the capacity to act. 43

The Clinton doctrine was designed to address the question of intervention

in NATO out-of-area situations and other wars of choice. The

Powell doctrine was directed more toward wars of necessity and was

entirely based on a realist paradigm of international relations. Clinton’s

amendments to the Powell doctrine sought a more nuanced spectrum of

challenges to U.S. and allied security and a prescription that would justify

intervention on idealist as well as realist doctrines. The Powell doctrine

was directed to the world as seen by Thomas Hobbes; the Clinton

amendments sought to incorporate the world as perceived by Immanuel

Kant. One thing was agreed upon by both schools of thought: intervention

required the capability to act effectively as well as the will to do so.

Interventions undertaken beyond the reach of U.S. or allied capabilities,

however noble in purpose, would in all likelihood crash and burn.


Limited war was the last thing that the U.S. defense establishment

expected to have to deal with in the decade following World War II.

Military Frustration and U.S. Adaptation 21

Once it was clear that the wartime alliance was doomed to postwar rivalry

and confrontation, American and Soviet leaders prepared for another

war in Europe, assuming that it would entail a world war. But the

outbreak of war in Korea and the plethora of nuclear weapons that first

Americans, and then Soviets, brought into their arsenals changed military

thinking about war. Global war, or even a large scale war in Europe,

became unthinkable. The only ‘‘acceptable’’ conflicts were those limited

to a single theater of operations and restricted by ground rules that precluded

escalation to any direct, U.S.-Soviet shooting war. Within these

constraints, the Americans and Soviets played hardball for 45 years, and

some of the limited wars fought in those years were very costly in lives

and money.

The Cold War did something else for the U.S. military. It established

the precedent of a large and diverse tri-service military capability maintained

in peacetime, even at costs that no pre–World War II Congress

could have imagined. Kennedy’s and Johnson’s military planners imagined

that a ‘‘two and one half’’ war capability would allow simultaneous

wars against Russia, China and a third, but smaller, state. This omnivorous

definition of U.S. military reach was blown apart by the Vietnam

War. U.S. commitments after Vietnam became more selective. It was not

until the Gulf War of 1991 that another major theater war tested U.S.

military capability, with a very different outcome. U.S. military strategy

after the Cold War calls for an ability to fight two, almost simultaneous

‘‘major regional conflicts.’’ But the additional expectation of heavy U.S.

commitment to multilateral peace operations, including Balkan peace impositions,

stretched post–Cold War operating tempos, personnel and

training to the limit by the end of the second Clinton term.


1. Russell F. Weigley, Towards an American Army: Military Thought from Washington

to Marshall (New York: Columbia University Press, 1962); Samuel P.

Huntington, The Soldier and the State (Cambridge, Mass.: Belknap Press/Harvard

University Press, 1957); Stephen J. Cimbala, ‘‘United States,’’ in Constantine P.

Danopoulos and Cynthia Watson, eds., The Political Role of the Military: An International

Handbook (Westport, Conn.: Greenwood Press, 1996), pp. 420–439.

2. For pertinent documentation, see Walter Lafeber, America, Russia and the

Cold War, 1945–1975, 3rd ed. (New York: John Wiley and Sons, 1976). On the

development of U.S. Cold War policy, see John Lewis Gaddis, The United States

and the Origins of the Cold War, 1941–1947 (New York: Columbia University Press,

1972). For Soviet policy, see Vojtech Mastny, Russia’s Road to the Cold War: Diplomacy,

Warfare, and the Politics of Communism, 1941–45 (New York: Columbia

University Press, 1979).

3. John Lewis Gaddis, The Long Peace: Inquiries into the History of the Cold War

(New York: Oxford University Press, 1987), p. 114.

22 Nuclear History and Its Lessons

4. Robert Endicott Osgood, Limited War: The Challenge to American Strategy

(Chicago: University of Chicago Press, 1957).

5. Henry A. Kissinger, Nuclear Weapons and Foreign Policy (New York: Harper

and Row, 1957).

6. Thomas C. Schelling, The Strategy of Conflict (Cambridge, Mass.: Harvard

University Press, 1960). See also Schelling, Arms and Influence (New Haven,

Conn.: Yale University Press, 1996).

7. William W. Kaufmann, The McNamara Strategy (New York: Harper and

Row, 1966).

8. Maxwell D. Taylor, The Uncertain Trumpet (New York: Harper and Row,


9. Robert A. Doughty et al., Warfare in the Western World, Vol. II: Military

Operations since 1871 (Lexington, Mass.: D. C. Heath, 1996), p. 913.

10. A critique of U.S. experience is provided in D. Michael Shafer, Deadly Paradigms:

The Failure of U.S. Counterinsurgency Policy (Princeton, N.J.: Princeton University

Press, 1988). See also Douglas S. Blaufarb, The Counterinsurgency Era: U.S.

Doctrine and Performance, 1950 to the Present (New York: The Free Press, 1977).

For evaluations of Amerian experiences with covert action, see John Prados, Presidents’

Secret Wars: CIA and Pentagon Covert Operations Since World War II (New

York: William Morrow, 1986), and Roy Godson, ed., Intelligence Requirements for

the 1980s, Vol. 4: Covert Action (Washington, D.C.: National Strategy Information

Center, 1983). An assessment of the impact of low intensity conflict on American

military professionalism appears in Sam C. Sarkesian, Beyond the Battlefield: The

New Military Professionalism (New York: Pergamon Press, 1981), chs. 4–7.

11. Harry G. Summers, Jr., On Strategy II: A Critical Analysis of the Gulf War

(New York: Dell Publishing Co., 1992), pp. 72–73.

12. Charles Krauthhammer, ‘‘The Unipolar Moment,’’ in Graham Allison and

Gregory F. Treverton, eds., Rethinking America’s Security (New York: W. W. Norton,

1992), p. 298.

13. William J. Perry, Secretary of Defense, Annual Report to the President and

the Congress (Washington, D.C.: U.S. Government Printing Office, 1996), p. 5.

14. Michael Klare, Rogue States and Nuclear Outlaws: America’s Search for a New

Foreign Policy (New York: Hill and Wang, 1995), pp. 109–110.

15. For diverse views on this, see the chapters by Don M. Snider and Andrew

J. Kelly, Paul K. Davis and Lawrence J. Korb in Stephen J. Cimbala, ed., Clinton

and Post–Cold War Defense (Westport, Conn.: Praeger Publishers, 1996).

16. U.S. Congress, Congressional Budget Office, An Analysis of the Administration’s

Future Years Defense Program for 1995 through 1999 (Washington, D.C.: CBO

Papers, Congressional Budget Office, January 1995), p. 18.

17. Ibid., pp. 4–5.

18. Martin Van Creveld, The Transformation of War (New York: The Free Press,


19. Carl von Clausewitz, On War, edited and translated by Michael Howard

and Peter Paret (Princeton, N.J.: Princeton University Press, 1976).

20. For an excellent discussion, see John A. English, Marching through Chaos:

The Descent of Armies in Theory and Practice (Westport, Conn.: Praeger Publishers,

1996), pp. 36–37.

Military Frustration and U.S. Adaptation 23

21. K. J. Holsti, The State, War and the State of War (Cambridge: Cambridge

University Press, 1996), pp. 24–25.

22. Ibid., p. 25.

23. Robert J. Art, ‘‘A Defensible Defense: America’s Grand Strategy after the

Cold War,’’ International Security, No. 4 (Spring 1991), pp. 5–53.

24. Ibid., passim. Art offers a deliberately narrow and specific definition of

security: the ability of the United States to protect its homeland from attack,

conquest, invasion or destruction (ibid., p. 7).

25. John Lewis Gaddis, The United States and the End of the Cold War: Implications,

Reconsiderations, Provocations (New York: Oxford University Press, 1992),

pp. 193–194.

26. Andrew J. Krepinevich, Jr., The Army and Vietnam (Baltimore, Md.: Johns

Hopkins University Press, 1986), esp. pp. 164–193.

27. Sam C. Sarkesian suggests that the term unconventional conflict is preferable

to low intensity conflict. Unconventional conflicts are nontraditional and not in

conformity with the American way of war. These kinds of conflicts emphasize

social and political variables, especially the problem of revolution and counterrevolution,

instead of the military dimensions of conflict. See Sarkesian, ‘‘U.S.

Strategy and Unconventional Conflicts: The Elusive Goal,’’ in Sarkesian and John

Allen Williams, eds., The U.S. Army in a New Security Era (Boulder, Colo.: Lynne

Rienner Publishers, 1990), pp. 195–216.

28. U.S. Special Operations Command, History (MacDill AFB, Tampa, Fla.: HQ

USSOCOM, September 1998), p. 5.

29. Susan L. Marquis, Unconventional Warfare: Rebuilding U.S. Special Operations

Forces (Washington, D.C.: Brookings Institution, 1997), esp. pp. 107–147.

30. Sarkesian, Beyond the Battlefield, chs.4–6.

31. Russell B. Weigley, ‘‘American Strategy from Its Beginnings through the

First World War,’’ in Peter Paret, ed., Makers of Modern Strategy, pp. 410–412, and

Weigley, The American Way of War: A History of United States Military Strategy and

Policy (New York: Macmillan, 1973).

32. Weigley, ‘‘American Strategy from Its Beginnings through the First World

War,’’ p. 412.

33. Ibid., p. 410; Sarkesian, America’s Forgotten Wars, p. 107.

34. Sarkesian, America’s Forgotten Wars, p. 110.

35. See Thomas A. Keaney and Eliot A. Cohen, Revolution in Warfare? Air Power

in the Persian Gulf (Annapolis, Md.: Naval Institute Press, 1995), pp. 1–21.

36. For an assessment of threats to U.S. security remaining after the Cold War,

see Art, ‘‘A Defensible Defense: America’s Grand Strategy after the Cold War,’’

pp. 23ff. Art contends that traditional geopolitical logic exaggerated the degree

of threat to U.S. interests from 1945–1990. Especially after 1960, the United States

was faced with neither credible threats of invasion nor slow strangulation subsequent

to the conquest of Eurasia by an adversary hegemon.

37. A collection of essays speaking to these and other issues pertinent to future

U.S. defense planning is Paul K. Davis, ed., New Challenges for Defense Planning:

Rethinking How Much Is Enough (Santa Monica, Calif.: RAND Corporation, 1994),

esp. chs. 2 and 4.

38. Paul K. Davis, ‘‘Planning Under Uncertainty Then and Now: Paradigms

Lost and Paradigms Emerging,’’ in Paul K. Davis, ed., New Challenges for Defense

24 Nuclear History and Its Lessons

Planning: Rethinking How Much Is Enough (Santa Monica, Calif.: RAND Corporation,

1994, p. 31.

39. The point is emphasized in Michael J. Mazarr, Don M. Snider and James

A. Blackwell, Jr., Desert Storm: The Gulf War and What We Learned (Boulder, Colo.:

Westview Press, 1993), pp. 162–168.

40. Ibid., p. 175.

41. The term reconnaissance-strike complexes originated in Russian/Soviet military

discourse to describe combinations of increasingly accurate conventional

munitions, improved target identification and location, and enhanced control and

communications systems for directing the employment of munitions against selected

targets. See V. G. Reznichenko, I. N. Vorob’iev and N. F. Miroshnichenko,

Taktika (Tactics) (Moscow: Voenizdat, 1987), p. 24, which notes: ‘‘In the opinion

of foreign specialists, reconnaissance-strike (fire) complexes are the most effective

form of high precision weapon. High precision reconnaissance resources and

high precision weapons are coordinated by an automated control system making

it possible to carry out reconnaissance and destruction missions practically in

real time.’’ The issue continues to preoccupy the Russian military, although it

will struggle to maintain the military-industrial base for future modernization.

See John Erickson, ‘‘Quo Vadis? The Changing Faces of Soviet/Russian Forces

and Now Russia,’’ in Stephen J. Blank and Jacob W. Kipp, eds., The Soviet Military

and the Future (Westport, Conn.: Greenwood Press, 1992), pp. 33–58.

42. According to one source, ‘‘Various offices in the Pentagon are battling for

control of a dramatic new initiative for nonlethal warfare. By using blinding

lasers and chemical immobilizers to stun foot soldiers and munitions with ‘‘entanglement’’

warheads to stop armored vehicles on land or ships at sea, it is

hoped by some that the United States could some day fight a war that did not

involve death, or at least few deaths.’’ Mazarr, Snider and Blackwell, Desert

Storm, p. 172. Life imitates art, or at least, simulation.

43. Interview with Samuel R. Berger, national security advisor to President

Clinton, cited in Jodi Enda, ‘‘Coming Soon: Clinton Doctrine,’’ Philadelphia Inquirer,

June 20, 1999, p. A18.

Chapter 2

Marching Beyond Marx:

The Red Army and Nuclear Weapons


Soviet hubris in the aftermath of victory in World War II concealed a

great fear: that the postwar world might evolve in ways that renewed

the threat of surprise attack launched from beyond Russia’s borders. Stalin

publicly deprecated the value of atomic weapons so long as the

United States maintained a nuclear monopoly: privately, he urged on his

own scientists and feared U.S. nuclear intimidation. 1 As Soviet military

thinkers adapted to the availability of plentiful nuclear weapons and

long range delivery systems, they developed concepts of deterrence and

of crisis management considerably different in emphasis from those familiar

to Americans (U.S. and Soviet crisis management concepts are

considered in the next chapter). The facts of Russian and Soviet historical

experience, the role of the professional armed forces in setting down the

‘‘military science’’ aspects of military doctrine, and the geopolitical setting

for Soviet postwar foreign policy making all implied a uniquely

Russian context for thinking about the role of nuclear forces and of nuclear

dissuasion in defense policy. 2

This chapter emphasizes two aspects of Soviet adaptation in military

strategy between 1945 and 1989: (1) the impact of nuclear weapons on

Soviet military strategy for fighting or deterring a world war (and, logically

enough, for fighting or deterring a major war in Europe that might

possibly, although not inevitably, become a world war); and (2) the implications

of nuclear weapons, and of the emergence of nuclear-strategic

parity, for the likelihood of conventional war in Europe and for the way

in which such a war might have been fought if deterrence had failed.

26 Nuclear History and Its Lessons

In facing these questions, the party and military leaders of the Soviet

Union had to confront the first ‘‘revolution in military affairs’’ to test the

ingenuity and resolve of the victors of the Great Patriotic War. (See

Chapter 4 for a discussion of the second ‘‘revolution in military affairs’’

and its implications for post–Soviet Russia.) The combination of acknowledged

nuclear stalemate and increased activeness of conventional

defenses made possible a variety of East-West arms limitation agreements

that helped to bring the Cold War to an end. 3


Three generalizations about the Soviet view of deterrence set it apart

from the standard American and NATO perspective of the Cold War

years. First, Soviet thinking was less centered on the military balance per

se than it was conditioned by the idea of ‘‘correlation of forces,’’ (sootnoshenie

sil), a more inclusive construct that took into account political,

social, economic, and moral-psychological factors of the competing systems.

Second, the Soviet view of deterrence was not as specifically oriented

to nuclear weapons and nuclear war as was the American. Soviet

leaders were equally concerned with avoiding a conventional war in

Europe on unfavorable terms, among other reasons because such a conflict

might go nuclear under some conditions. Third, the Soviet political

leadership inherited a ‘‘defense of Mother Russia’’ or ‘‘Barbarossa’’ complex

from Russia’s history of invasions from Poles, Swedes, French, Germans

and others who ravaged across her extended and vulnerable

borders. This explains why, despite the crudity of available technology

and in the face of American arms control biases against defenses, Russians

took anti-missile defense of the homeland quite seriously. 4 (Other

aspects of Soviet views of deterrence, more pertinent to the specific problem

of crisis avoidance or management, are noted in the next chapter.)

Stalin’s successors in the party leadership and their military advisors

acknowledged after 1953 that nuclear weapons had brought about a revolution

in military affairs. This acknowledgment came in stages and arrived

in full force only after Khrushchev had finally disposed of serious

rivals for the party leadership in 1957. Khrushchev adopted no proposal

or line of thinking without excessive enthusiasm. He fully embraced the

idea of nuclear missile warfare as the centerpiece of Soviet military planning

guidance. 5 He was left with the problem of explaining how socialism

would survive and triumph in the aftermath of a global nuclear war

with the Americans and their NATO allies. After Khrushchev’s departure

from the top posts in party and government in 1964, party guidance

continued to insist that socialism would prevail even in nuclear war. But,

under Brezhnev and his successors, military doctrine was adjusted to

The Red Army and Nuclear Weapons 27

allow the problem of nuclear escalation to be considered as a variable

instead of a constant.

Soviet military literature in the early 1960s emphasized that war would

probably begin with a massive nuclear surprise attack and involve the

Soviet Union and its allies against the West in a major coalition war.

Following Khrushchev’s ouster and changes in U.S. and allied NATO

strategy toward ‘‘flexible response’’ in the second half of the 1960s, Soviet

military strategists showed increased interest in scenarios other than

massive retaliation and global conflict. 6 As the decade progressed there

was, on the part of a variety of authoritative political and military thinkers,

greater acknowledgment that even large scale war could be waged

with conventional weapons only. In addition, the potential payoff from

Soviet support for wars of ‘‘national liberation’’ against pro-Western

states outside of Europe directed additional military thinking toward

local and limited wars. The development of Soviet military thinking in

the 1950s and 1960s can be summarized by enumerating the five types

of war conceived by leading theoreticians and military strategists during

that time:

• a massive nuclear, but relatively short, war in which strategic nuclear weapons

played the major role in deciding the outcome;

• a more protracted war including nuclear strikes but also involving all of the

armed forces;

• a major war in which nuclear weapons are used in a restricted or limited manner

in one or several theaters of military action;

• a major war limited to the use of conventional weapons;

• a local war involving conventional weapons. 7

The willingness of Soviet military analysts to entertain multiple scenarios

as equally valid possibilities increased in the 1970s as a result of

detente between the United States and the Soviet Union. New threat

assessments were also mandated by the geopolitical repositioning of

China, hitherto included in the Soviet camp but henceforth treated as an

ally of the United States and NATO. 8

The period of threat preceding war, as explained by lecturers at the

Soviet General Staff Academy during the l970s, was a time in which the

Soviet armed forces were in great danger of being caught by surprise. 9

In a nuclear war the ability to seize the strategic initiative in the first

minutes would have, according to these estimates, a decisive impact on

the development of military action and on the duration and outcome of

the conflict. Soviet forces must be prepared for transition from conventional

to nuclear operations at any moment. The first volume of the lecture

materials from the General Staff Academy discussed the ‘‘forms of

28 Nuclear History and Its Lessons

initiation of war by the aggressor’’ and, with regard to NATO, specified

the following possibilities:

• surprise strikes with unlimited use of nuclear weapons;

• a strike with initially limited use of nuclear weapons and subsequently going

over to full use of the complete nuclear arsenal;

• strikes by groupings of armed forces deployed in the TSMAs (TVDs) without

the use of nuclear weapons;

initiation of war by gradual expansion of local wars. 10

Initiation of war by the United States and its NATO allies through a

general nuclear attack was studied in the 1970s as ‘‘the basic form of

initiating war, with respect to American doctrine.’’ 11 This form of initiating

war was the most dangerous and could have had the greatest consequences

if unexpected and not reacted to promptly. The Soviet

lecturers described at some length how the United States would have

orchestrated a nuclear surprise attack. 12 Although the most dangerous

from the Soviet standpoint, this course of action was not judged to be the

most likely, according to the Voroshilov lecture materials. More likely

was the limited use of nuclear weapons, followed by the unlimited use

of the U.S. and NATO complete arsenals. 13

On the threshold of nuclear first use, the moment at which NATO had

taken the decision for nuclear escalation was of vital importance for Soviet

intelligence to establish. Subsequent reactive movements of troops,

command posts, logistics, and other assets and the preparation of counterstrikes

would have to take place during the time between detection

of NATO’s decision to go nuclear and the launching of the first salvos.

Otherwise the Soviet and allied forces thrusting into Central Europe

would be confronted with a Barbarossa on the move, leading to a disruption

of their attack plans, tempo of operations and combat stability.

Soviet sensitivity to these possibilities was apparently acute in the

early 1980s. During NATO command post exercise Able Archer conducted

in November 1983, Warsaw Pact intelligence monitored the flow

of events according to tasking laid down by experience and precedent.

The exercise included practice with NATO nuclear release procedures. 14

British and American listening posts detected unusual Soviet sensitivity

to unfolding events, with a significant increase in the volume and sense

of urgency in Warsaw Pact message traffic. 15 On November 8–9, according

to Soviet defector Oleg Gordievsky, an ‘‘Operation RYAN’’ message

was sent from Moscow Center to KGB residencies abroad. Operation

RYAN (for raketno yadernoye napadeniye, or rocket nuclear attack) had

been established in 1981 for intelligence gathering and strategic warning

The Red Army and Nuclear Weapons 29

with regard to the possibility of nuclear surprise against the Soviet Union

and its East European allies. 16

The danger of reciprocal alerting of nuclear forces leading to crisis

instability was not hypothetical, for several reasons. First, the Soviet concept

of an alerted or generated force might have been entirely different

from the U.S. concept. The Soviet conceptual framework for military

planning emphasized mobilization, readiness and concentration of all

force components, regarding the generation of nuclear forces as one part

of that process. 17 Second, Soviet political leaders and military planners

were disinclined to use nuclear alerts as political signalling devices during

a crisis. The Soviet view of crisis was that crisis was an objective

condition or period of threat during which states actually prepare for

war. 18 (Soviet, compared to U.S., views of crisis avoidance or management

receive more specific treatment in the next chapter.)

Third, the skewing of intelligence and warning toward the extreme

case of ‘‘bolt from the blue’’ attacks may have been less pronounced in

Soviet nuclear planning compared to American. Although Soviet historical

experience emphasized the importance of avoiding strategic military

surprise, the totally unexpected and massive, surprise nuclear strike was

not judged as the most probable of events by Soviet military thinkers of

the 1970s and subsequently. Not only the previously cited Voroshilov

materials confirm this, but so, too, do Soviet force deployments and crisis

management behaviors. 19 Soviet perceptions of strategic nuclear inferiority

in the 1950s and early 1960s eventually gave way in the face of

improved U.S.-Soviet political relations and more diverse and survivable

Soviet force deployments. The political leadership would have expected

to receive from various intelligence sources at least some advance warning

of any hostile intent to attack, and this warning would allow some

time to prepare a response. Moreover, attainment of strategic nuclear

parity with the United States provided to the Soviet leadership the options

of preemption, launch under attack/launch on tactical warning, or

second strike ride out. 20

Strategic warning indicators provided by various Soviet intelligence

agencies would have been derived from monitoring of Western communications

traffic, troop movements, diplomatic initiatives and any

other apparent preparations for higher levels of military alert or for

war. 21 The Soviet view of combat readiness recognized that not all forces

can be alerted at the same rate, or need be. Strategic retaliatory forces,

air defense troops, and ground and air forces deployed in the first operational

echelon had to be maintained ‘‘at full wartime strength and

should be able to advance themselves to a level of full combat readiness

in the shortest time for the accomplishment of assigned missions.’’ 22 The

decision for strategic deployment (strategicheskoye razvertyvaniye) and for

the transition of the Soviet armed forces from a peacetime to a wartime

30 Nuclear History and Its Lessons

standing was one to be made by the highest political leadership. 23 One

of the most important aspects of strategic deployment of forces, according

to authoritative Soviet sources, was continuous control of the armed

forces by use of military communications networks and other links. 24 The

reasons for this included not only the avoidance of enemy surprise, but

also the possibility of communications and command system breakdowns,

and of the subsequent fragmentation of the command system

into uncoordinated parts:

The complex situation in which strategic deployment of the Armed Forces is

conducted requires centralized control. At the same time, considering the limited

capabilities of control elements to furnish a wide range of timely information

and taking into account the possible interruption of control, particularly in a

nuclear war, special importance is given to the initiative of commanders at all

levels on the basis of overall concepts and plans. 25

There is little evidence in published Soviet military doctrine to suggest

that their armed forces or political leadership would have been interested

in fighting a strategic or theater nuclear war (strategic for them) in the

controlled and selective manner sometimes envisioned in U.S. academic

literature. 26 In the event of nuclear war, Soviet strategic forces ‘‘would

be used massively rather than sequentially, and against a wide range of

nuclear and conventional military targets, command-and-control facilities,

centres of political and administrative leadership, economic and industrial

facilities, power supplies, etc., rather than more selectively.’’ 27

Urban areas would not have been attacked gratuitously or in pursuit of

some arbitrary number of fatalities, but ‘‘neither would they be avoided

if they were near military, political or industrial targets.’’ 28

Soviet long range ballistic missile forces (land based and submarine

launched) would have been hard pressed to satisfy the targeting requirements

of a cautious Soviet war planner for most of the Cold War.

Target arrays in the Transoceanic TVD (Theater of Military Action), essentially

North America, would have included both ‘‘hard’’ and ‘‘soft’’

targets. Hard targets are those that are heavily protected and must be

attacked by the most accurately delivered warheads. Such targets include

missile silos and launch control centers, nuclear weapons storage depots,

and other reinforced command, control, communications and intelligence

(C3I) facilities. Soft targets of interest to Soviet planners were presumably

those related to the destruction of U.S. conventional military

power or the disabling of U.S. war-related economy: airfields, ports, bases,

depots, electric power plants, petroleum refineries, chemical plants

related to military use, and other facilities not specially protected against

the effects of nuclear blast. A comparison of Soviet targeting require-

The Red Army and Nuclear Weapons 31

Table 2.1

Soviet Cold War Nuclear Targeting Requirements and Available Weapons:

Transoceanic TVD

*Soviet SS-7 and SS-8 ICBMs would not have been effective against hard targets such as

Minuteman silos, but they could have been targeted against the softer Atlas and Titan

launchers still in the U.S. inventory.

**The 1980 versions of SS-18 and SS-19 Soviet ICBMs would not have been effective against

upgraded U.S. Minuteman ICBM silos rated at about 2,000 psi. Thus Soviet countersilo

attacks in 1980 would have required much more than the three on one targeting that

the later SS-18 Mod 4 and the SS-19 Mod 3 made feasible by the mid-1980s.

Source: Adapted from estimates by William T. Lee. See William T. Lee and Richard F. Staar,

Soviet Military Policy since World War II (Stanford, Calif.: Stanford University/Hoover

Institution Press, 1986), p. 160.

ments for selected years with strategic missile warheads available in the

same years against hard and soft targets is provided in Table 2.1.

It became as self-evident to the Soviet political leadership as to their

NATO opponents, especially subsequent to the Cuban missile crisis, that

nuclear weapons were not really ‘‘usable’’ in battle, although the same

weapons might be useful in deterrence. 29 This acceptance of mutual deterrence

and its underlying arms control construct, strategic nuclear parity,

became canonical during the SALT I negotiating period in the late

1960s and early 1970s. Mutual deterrence was a fact of life, from the

Soviet standpoint, although not a preferred condition: it did not preclude

32 Nuclear History and Its Lessons

the possibility of nuclear war entirely. 30 Another marker of the Soviet

view of deterrence was Brezhnev’s speech at Tula in January 1977 in

which the Soviet leader renounced the aim of nuclear superiority as an

objective of Soviet policy. Subsequent to Brezhnev’s Tula speech, both

political and military leaders consistently acknowledged that Soviet policy

was defensive and designed to prevent attacks on the USSR, not for

superiority over the United States or for fighting and winning a nuclear

war. 31

The acceptance of nuclear parity and the inadmissability of nuclear

war as deliberate policy were not acts of charity on the part of Moscow.

Soviet military analysts on the General Staff told their political leaders

in the Kremlin that the realities of modern war precluded victory in a

large scale conventional war, such as might occur in Europe, as well as

in any nuclear war. The conventional forces of the Soviet empire and of

NATO, were they to clash on the northern and central European fronts

and somehow avoid escalation to nuclear attack, would nevertheless destroy

the very social values the two sides were thought to be defending.

A repeat of World War II to an acceptable, although costly, outcome was

simply not feasible by any measure of military planning. The superfluity

of large scale conventional warfare was even more threatening to the

Soviet state psyche than was the recognition, commonplace among elites

by the latter 1960s and by scientists even earlier, that a nuclear war

would have no winners. The cult of the Great Patriotic War in Russia

bolstered the regime against its many failures in economics and in public

recititude. If the Red Army was in fact no longer a usable instrument

against the main enemy as defined by Moscow and at an acceptable cost,

why should Soviet citizens continue to tolerate empty stores and shelves

in order to devote 15 percent of their GNP to defense? Table 2.2 summarizes

indicators of U.S. and Soviet nuclear and conventional military

power for selected years of the Cold War.

During the Gorbachev era after 1985 theorists from civilian research

institutes were encouraged to develop concepts of arms control and ‘‘reasonable

sufficiency’’ in defense more in keeping with Gorbachev’s desire

to reduce defense expenditures and to stabilize the arms race. 32 Other

defense intellectuals argued that numerical parity in nuclear arms, even

at lower levels, was insufficient to guarantee stability. Managing the

problem of inadvertent war or escalation was equally significant. 33 Some

authors contended that nuclear war stood apart from any relationship

between war and politics as previously posited by Clausewitz or Lenin. 34

The agreement to entirely eliminate so-called long-range, intermediate

nuclear forces (LRINF) in Europe, reached by Presidents Reagan and

Gorbachev, testified not only to the shared recognition of the absurdity

of nuclear war, but also to the changed perceptions of the two sides’

Table 2.2

U.S. and Soviet Force Levels, Selected Years

*U.S. and Soviet divisions are not directly comparable. Soviet divisions are made equivalent to those of the United States in this comparison.

Source: Lawrence J. Korb, ‘‘Where Did All the Money Go?’’ in Stephen J. Cimbala, ed., Mysteries of the Cold War (Brookfield, Vt.: Ashgate Publishers,

forthcoming), ch. 1.

34 Nuclear History and Its Lessons

intentions that had taken root between Reagan’s first and second terms

in office.

The insistence of Russia’s post-Soviet leadership on adherence to the

ABM Treaty of 1972 showed that the concept of nuclear forbearance by

mutual deterrence had carried forward from the Cold War into an uncertain

future. Post–Cold War U.S.-Russian cooperation on CIS denuclearization,

on nonproliferation and on a Comprehensive Test Ban (CTB)

agreement opened for signature in 1996 suggested continuing recognition

that nuclear weapons were simply different, and excess numbers of

them suitable for discard. 35



The great Russian commander and military theorist Aleksandr Suvorov

was not enamored of defensive actions. ‘‘The very name defense,’’

he once wrote, ‘‘already proves weakness, and so it incites timidity.’’ 36

But a Suvorov in command of Soviet forces in the 1970s and 1980s would

have recognized that new weapons and command/control systems made

possible more activeness in offensive and defensive battle, below the

nuclear threshold. In addition to new technologies, Soviet revisitation of

their own World War II history, and influential thinking about operational

art by prominent commanders, called forth new appreciations of

the defensive as a necessary or expedient form of war. Prior to 1985 these

modern views of the defensive carried no implication of disjunction between

the politico-military and military-technical levels of Soviet military

doctrine. After Gorbachev, reconciliation of the two levels became more

problematic. 37

As early as 1982, Marshal N. V. Ogarkov (then Chief of the Soviet

General Staff) had noted that ‘‘The previous forms of employment of

combined units and formations have in large measure ceased to correspond

to present conditions.’’ 38 A new U.S. military strategy, according

to Ogarkov, called for ‘‘preparing the armed forces to wage a war with

the employment of solely conventional weaponry.’’ 39 This contention

was repeated by Ogarkov in a 1985 publication. After discussing the

nuclear strategy of the Reagan administration as one which was offensively

oriented and designed to make possible a preemptive first strike,

Ogarkov noted that the U.S. military strategy ‘‘also envisions training its

armed forces to wage a war with the use of only conventional means of

destruction.’’ 40 What this might imply for the concept of the offensive

engagement was noted in an authoritative study of tactics published in

1984 under the editorship of Lt. Gen. V. G. Rezhichenko:

The offensive engagement today is more dynamic than in the last war. Being

fully motorized and amply equipped with tanks, forces can attack with smaller

The Red Army and Nuclear Weapons 35

densities of personnel and equipment than before, and yet in considerably greater

depth and with greater momentum. 41

In the offensive engagement using only conventional weapons, the enemy’s

first and second echelons and reserves were to be attacked sequentially

while moving rifle and tank subunits into the depth of his

defense. The employment of modern weaponry ‘‘increases the decisiveness

of an offensive engagement.’’ 42 Decisiveness resulted from the continuous

increase in troop capabilities and their ability to defeat the

enemy without having overall superiority in forces and equipment. 43 Although

decisiveness was important in offensive operations in past wars,

in modern conditions of the 1980s offensive operations were thought to

be even more decisive. According to Soviet tactical assessments, modern

rifle and tank subunits with highly effective combat equipment and

weapons were capable of breaching quickly a deeply echeloned defense

even if the defenders were well equipped with nuclear weapons and

high technology conventional forces, including ground-launched antitank

weapons, artillery, reconnaissance-strike complexes, airborne and

amphibious assault units, and helicopters. 44

Note that this mid-1980s assessment was offered of Soviet capabilities

for conducting offensive engagements using conventional weapons only.

Its apparent optimism concealed a concern on the part of Soviet planners

that these objectives might not be attainable in the event of actual war,

as opposed to the conduct of military exercises. Much depended on the

correct timing and coordination of efforts. Although total surprise

against NATO was not probable during any crisis or period of tension,

partial surprise was not precluded. And partial surprise was necessary

if the Soviet Union were to have had any hope of breaching a fully

prepared defense in depth which NATO was increasingly more capable

of presenting. 45 After the dramatic political events of 1989 in which the

fall of communist governments in Eastern Europe disestablished the

Warsaw Pact as a cohesive military alliance, the prospects for a short

warning, unreinforced attack against NATO seemed to have dropped

from improbable to impossible. Henceforth the Soviet Union might have

to assume the defensive, not the offensive, in the initial period of war.

Colonel-General M. A. Gareev, then deputy chief of the Soviet General

Staff, wrote in 1985 that Soviet military theory and operational plans on

the eve of World War II gave insufficient attention to the proper conduct

of the operational and strategic defensive. He noted that the ‘‘idea of the

continuous shifting of war at its very outset to enemy territory (and the

idea was unsound both scientifically and backed up neither by an analysis

of the actual situation or by operational calculations) had so beguiled

certain leading military workers that the possibility of conducting military

operations on our own territory was virtually excluded.’’ 46 This

36 Nuclear History and Its Lessons

same assessment was offered by Andrey Kokoshin and Valentin Larionov

in their discussion of the battle of Kursk as a model for the implementation

of the doctrine of defensive sufficiency. 47 The authors did

not dwell on the fact that Kursk was an example of very active defense

nor that it was based on the acquisition of very precise intelligence about

the opponent’s intentions. Kursk was also a case of an operational counteroffensive

planned and conducted after war had been declared and

fought for several years. Thus it provided little in the way of guidance

for harried Soviet planners who might be tasked to defend expansive

borders on the basis of force posture and military doctrine which excluded

preemption or even defensive activeness as an option.

However, the problem of military stability has always been two sided:

the prevention of accidental/inadvertent war was as important as the

deterrence of deliberate aggression. According to prominent Soviet military

theorists of the 1980s, previous military planning did not always

take into account all aspects of the complex relationship between mobilization

and deterrence. In a departure from the precedent set by V. D.

Sokolovskiy’s Voennaya strategiya (Military Strategy) in the 1960s, Gareev

in his M. V. Frunze—voennyi teoretik (M. V. Frunze—Military theorist)

doubted whether mobilization of all essential forces and means prior to

war was either necessary or desirable. 48 The Sokolovskiy volume reflected

the shared conviction by the Soviet military leadership in the

1960s that any war between East and West would shortly become nuclear

and all out. By the time of Gareev’s M. V. Frunze (the mid-1980s) Soviet

planning guidance and military theory had changed considerably, beginning

with a major shift in 1966 toward a priority for the prevention

of world war and the nuclear destruction of the USSR. 49

Prevention of the nuclear destruction of the Soviet Union could not be

guaranteed once fighting had expanded to include strikes against the

territorial homelands of the superpowers. Even the limited use of nuclear

weapons in Europe carried incalculable risks of expansion into total war.

Therefore, to the extent possible, escalation to the level of nuclear warfare

would have to be forestalled, and the Soviet ground and tactical air

forces in Europe would have to fight below the nuclear threshold. Soviet

theater and strategic nuclear forces would assume the roles of counterdeterrents

to NATO’s theater and strategic nuclear deterrents, opening

the highway for a conventional test of strength. 50

Even without nuclear escalation, the problem of military stability and

the possibility of first strike fears leading to war demanded further attention.

Gareev referred to mobilization as ‘‘tantamount to war’’ in the

sense that mobilization sends signals to the other side, raising its level

of awareness and raising its sensitivity to any indicators of planning for

surprise attack. 51 The Soviet General Staff would have preferred to have

authorization for prewar mobilization which was proof even against

The Red Army and Nuclear Weapons 37

worst-case surprises, but the likelihood is that the post-1985 political

leadership would have denied them this. 52 Therefore, the armed forces

had to plan for war under disadvantageous conditions and allow for the

possibility of enemy preemption with conventional deep strike. According

to the guidelines provided by Soviet political leadership (Gorbachev)

in the latter 1980s, military doctrine emphasized the prevention (predotvrashchat’)

of war, and for this purpose the political leadership had to

avoid provocative mobilizations inviting enemy preemption. 53 For some

Soviet military planners of the latter 1980s, the problem of optimizing

preparedness while avoiding unnecessary provocation of potential enemies

was a reminder of the months immediately preceding Barbarossa:

The Soviet leadership did not want to provoke hostilities at a time of complex

secrecy in the Soviet system, so that there was a fundamental dysfunction—a

cybernetic dysfunction—between two Soviet systems of information. One was

among the initiated, and that included a good portion of the General Staff and

certainly the political leadership who thought that war was coming, although

how close they could not say. The second was an attitude among the rank-andfile

that war was not close and therefore should not be anticipated. 54

For both prospective attackers and defenders using large and technologically

well-equipped forces in the last two decades of the Cold War,

the capability for rapid mobilization, concentration and deployment of

combat and combat support elements, once war was judged likely, became

an important deterrent. This importance was emphasized in lecture

materials from the Soviet General Staff Academy during the 1970s. 55 The

capabilities of modern reconnaissance systems, weapons and control

made possible as never before the seizure of initiative by the defense. It

was precisely this possibility, of NATO seizing the initiative from an

initially defensive posture and inflicting deep strikes on Soviet reinforcements

and logistics, which worried Soviet planners in the 1980s.

In short, current and future weapons technology could make it possible

to overturn the enemy’s plans, to attack enemy forces at great

depth, and to inflict decisive losses whether fighting from the offensive or

the defensive. The complexity of even small high-technology conventional

forces, let alone larger ones, made the struggle for information even more

important for the defender who must seize the initiative as soon as the

prospective attacker’s plans are successfully gleaned. However, in an era

of nuclear and highly destructive conventional weapons, provocation of

an attack which was not actually being planned had to be avoided. The

line between deterrence and provocation could be maintained by prospective

defenders only if their command and control systems were intact

and functioning with high effectiveness.

Under the relentless pressure of modern battle, command and control

38 Nuclear History and Its Lessons

systems and other aspects of defensive combat stability (zhivuchest’) were

almost certain to be stressed to the ultimate, even if nuclear escalation

could have been avoided. Authoritative Soviet assessments of NATO

potential to conduct an ‘‘air-land operation’’ in the latter 1980s warned

that the objective of such an operation would be ‘‘destroying the enemy

throughout the entire depth of his army’s operational formation.’’ 56 In

these conditions of uninterrupted battle, defensive war fighting could be

as dynamic as offensive war fighting.

Defensive combat within the framework of an air-land operation is a combination

of static and dynamic actions by combined-arms formations and units, coupled

with growing fire pressure upon the advancing, deploying and attacking enemy.

It presupposes integrated application of the principles of positional and mobile

defense with the purpose of halting an offense and seizing the initiative. In this

case defense is conducted no less decisively than offense. 57

The increased activeness of offensive and defensive combat, should

deterrence fail, posed a problem for those who sought in the latter 1980s

to revise Soviet military theory in a direction more declaredly defensive.

In an article in the December 1989 issue of Kommunist, Soviet Defense

Minister Dmitri Yazov outlined his view of a change in the relationship

between the political or socio-political and military-technical levels of

Soviet military doctrine. 58 Yazov acknowledged that a contradiction had

marked the past development of Soviet military doctrine, between its

political and military-technical aspects. Whereas in its political aspects

military doctrine was always defensive stipulating the rejection of military

attack on anyone at all, on the military-technical plane it relied on

‘‘decisive offensive actions’’ in case war was unleashed against the Soviet

Union and its allies. It was also assumed that, the higher the capability

of the Soviet armed forces for such actions, the more solid the defense,

and the less likely an attack by the enemy. Eventually this resulted in a

contradiction which had to be acknowledged and resolved; the defensive

thrust of the political aspect of military doctrine was in contradiction,

according to Yazov, to its military-technical emphasis on offensive action.

Therefore, ‘‘in the contemporary contents of our doctrine, brought

into action in 1987, this contradiction is completely eliminated.’’ 59 The

contradiction was resolved, according to Yazov, and in conformity with

authoritative party guidance at the time, by movement toward a posture

of reasonable sufficiency for defense which would become evident

through changes in Soviet defense budgets and in force structure.

Gorbachev’s effort to impose defensiveness to Soviet military doctrine

was, as we now know, only successful in part. 60 The parlous state of the

Soviet economy in the latter 1980s drove Gorbachev toward decisions

favoring nuclear and conventional arms reductions and toward redefin-

The Red Army and Nuclear Weapons 39

ition of the East-West competition along the lines of common European

security. 61 The rapid collapse of the Warsaw Treaty Organization and

Soviet control over their former military satellites in East Central Europe

resulted in premature closure of the debate over ‘‘defensive sufficiency’’

in favor of matters more urgent. In February 1988 Gorbachev announced

that the Limited Contingent of Soviet Forces in Afghanistan would be

completely withdrawn within one year. In one sense the Soviet military,

rid of Afghanistan, breathed a sigh of relief; from another standpoint, it

was a serious blow to professional military self-esteem. The Soviet army

had ‘‘lost’’ a war for the first time, and the setbacks could not be blamed

on ‘‘Third World’’ clients. 62 Military participation in the attempted overthrow

of Gorbachev in August 1991 provided additional evidence that

some members of the Soviet officer corps had become more politicized

in favor of ‘‘Soviet’’ military doctrine than had their nominal party and

government civilian superiors. 63 The collapse of the Soviet Union in December

1991 brought a halt to the further development of a uniquely

‘‘Soviet’’ military theory, strategy and doctrine, but much of it would

obviously be carried forward into the armed forces of newly democratic



Faced with a bipolar, nuclear world for more than four decades, the

Soviet political and military leadership adjusted authoritative views on

the nature of a future war, on the relationship between technology and

military power, and on other socio-political and military-technical aspects

of doctrine. Eventually recognizing that victory in a general nuclear

war was impossible and that the limitation of any nuclear war to a particular

theater of operations was unlikely, Soviet military thinkers accepted

the fact of mutual deterrence based on survivable retaliatory

forces. However, in the Soviet view, an essentially equivalent balance of

nuclear forces between Russia and its enemies could not guarantee peace

and security. War could break out for reasons having little or nothing to

do with rational calculation or with the prewar correlation of military

and other forces. Therefore, authoritative and expert Soviet views on the

prevention of war, by deterrence or by other means, had to take into

account the possible outbreak of crisis leading to war.

The results of Soviet military adaptation to Cold War demands were

successful in the short run, although at dizzying cost to an already stagnant

economic system in the longer term. Soviet military adaptation created

the image of a superpower equivalent in power and potential to the

U.S. and its NATO allies. But this putative military equivalency was, in

the Russian tradition, mostly Potemkinism. Behind the facade of Russia’s

superpower status lay its dormant economy, its dissident nationalities

40 Nuclear History and Its Lessons

and its sclerotic party and military leadership. When it all crashed between

1989 and 1991, the Russian core of Soviet military power was

forced to reinvent itself with truncated conventional military forces and

residual nuclear weapons of uncertain purpose.


1. David Holloway, Stalin and the Bomb: The Soviet Union and Atomic Energy,

1939–1956 (New Haven, Conn.: Yale University Press, 1994), p. 253. Although

Stalin emphasized the development of delivery vehicles for Soviet atomic weapons

and defense against atomic attack, he did not regard the atomic bomb as a

decisive weapon. Ibid., p. 250.

2. Harriet Fast Scott and William F. Scott, Soviet Military Doctrine: Continuity,

Formulation, and Dissemination (Boulder, Colo.: Westview Press, 1988), esp. pp. 3–


3. Raymond L. Garthoff, The Great Transition: American-Soviet Relations and the

End of the Cold War (Washington, D.C.: Brookings Institution, 1994), esp. pp. 551–


4. Garthoff, Deterrence and the Revolution in Soviet Military Doctrine (Washington,

D.C.: Brookings Institution, 1990), pp. 6–28, is especially helpful on these

points. Other sources pertinent to Soviet deterrence concepts are cited in notes

that follow.

5. V. D. Sokolovskiy, ed., Voennaya Strategiya (Military Strategy) (Moscow:

Voenizdat, 1962). Later editions of this work were published in 1963 and 1968.

See also Harriet Fast Scott and William F. Scott, Soviet Military Doctrine: Continuity,

Formulation and Dissemination (Boulder, Colo.: Westview Press, 1988),

pp. 34–41.

6. Andrei A. Kokoshin, Soviet Strategic Thought, 1917–91 (Cambridge, Mass.:

MIT Press, 1998), p. 124.

7. Ibid., p. 126.

8. Ibid., p. 127.

9. Ghulam Dastagir Wardak, comp., and Graham Hall Turbiville, Jr., gen. ed.,

The Voroshilov Lectures: Materials from the Soviet General Staff Academy, Vol. 1

(Washington, D.C.: National Defense University Press, 1989), pp. 233–254.

10. Ibid., pp. 244–245.

11. Ibid., p. 245.

12. Ibid., pp. 245–246.

13. Ibid., p. 247. This is one clue to apparent Soviet unwillingness to accept

U.S. versions of controlled nuclear escalation, as for example in National Security

Decision Memorandum (NSDM) 242 later referred to as the ‘‘Schlesinger doctrine.’’

14. Ben B. Fischer, A Cold War Conundrum: The 1983 Soviet War Scare (Washington,

D.C.: U.S. Central Intelligence Agency, Center for the Study of Intelligence,

September 1997), pp. 24–26, provides an informative account of Able

Archer and its presumed relationship to Soviet fears of war.

15. Gordon Brook-Shepherd, The Storm Birds: Soviet Postwar Defectors (New

York: Wiedenfeld and Nicolson, 1989), p. 329.

The Red Army and Nuclear Weapons 41

16. Ibid., pp. 330–331.

17. See Graham H. Turbiville, Jr., ‘‘Strategic Deployment: Mobilizing and

Moving the Force,’’ Military Review, Vol. 68 (December 1988), pp. 41–49. According

to the first volume of the Voroshilov Military Academy of the General Staff

lecture materials of the 1970s, ‘‘the notification by alert given to bring the strategic

nuclear forces to full combat readiness, is only some minutes, while, for the

units, large units and operational formations of the various Services of the Armed

Forces, such notification will be longer to a greater extent.’’ Wardak and Turbiville,

Voroshilov Lectures, Vol. 1, p. 181. On the levels of combat readiness (boevoy

gotovnost’) and the process of bringing the Soviet armed forces to full combat

readiness, see ibid., ch. 4. Although the process of bringing the armed forces to

the highest level of combat readiness ‘‘should be conducted under all conditions

in close consideration of the employment of nuclear weapons by the enemy’’ and

‘‘control should ensure centralized and simultaneous communication of signals

and instructions’’ according to measures developed by the General Staff and

made known to the troops (ibid., pp. 193–194).

18. Stephen Shenfield, ‘‘Crisis Management: the Soviet Approach,’’ in Carl Jacobsen,

ed., Strategic Power: USA/USSR (New York: Macmillan, 1990), p. 200. See

also the discussion in the next chapter on Soviet versus American views of crisis


19. Bruce G. Blair, The Logic of Accidental Nuclear War (Washington, D.C.:

Brookings Institution, 1993), esp. pp. 25–26.

20. Stephen M. Meyer, ‘‘Soviet Nuclear Operations,’’ in Ashton B. Carter, John

D. Steinbruner and Charles A. Zraket, eds., Managing Nuclear Operations (Washington,

D.C.: Brookings Institution, 1987), pp. 470–534.

21. Ibid., pp. 513–516.

22. Wardak and Turbiville, Voroshilov Lectures, vol. 1, p. 181.

23. Ibid., p. 229. Soviet military writers also use the term ‘‘strategic leadership’’

(stratigicheskaya rukovodstva) to refer to the highest political and military leaders

as a group. Prior to 1987 this was assumed to be the Defense Council, a smaller

body within the Politburo, and it was supposed that in wartime the Defense

Council assumed the same functions and authority as had the GKO (State Committee

for Defense) in World War II. Under Gorbachev realignments of power

among party, government and armed forces after 1987, there was more guesswork

involved in extrapolating from Soviet peacetime to wartime command arrangements.

For pertinent background, see Harriet Fast and William F. Scott, The

Soviet Control Structure: Capabilities for Wartime Survival (New York: Crane, Russak/National

Strategy Information Center, 1983), pp. 45–58.

24. Wardak and Turbiville, Voroshilov Lectures, Vol. 1, p. 229.

25. Ibid., p. 231.

26. For a discussion of this, see Edward L. Warner III, Soviet Concepts and

Capabilities for Limited Nuclear War: What We Know and How We Know It (Santa

Monica, Calif.: RAND Corporation, February 1989). This study contains many

important primary source references. See also Lt. Gen. A. I. Yevseev, ‘‘O nekotorykh

tendentsiyakh v izmenenii soderzhaniya i kharaktera nachal’nogo perioda

voiny’’ Voenno-istoricheskii zhurnal (Military-Historical Journal), No. 11 (November

1985), pp. 10–20. Writing of the impact of a massive nuclear strike at the

outset of a war, Yevseev notes that ‘‘the initial period of a future nuclear-rocket

42 Nuclear History and Its Lessons

war may be the fundamental and decisive period which in large measure predetermines

the further development of armed conflict, and in certain conditions

the outcome of war’’ (Yevseyev, ‘‘O nekorotykh tendentsiyakh ...,’’ p. 17).

27. Desmond Ball, Soviet Strategic Planning and the Control of Nuclear War (Canberra:

Strategic and Defence Studies Centre, Australian National University, November

1983), p. 5. See also p. 8 for his notional RISOP table.

28. Ibid., p. 5.

29. Garthoff, Deterrence and the Revolution in Soviet Military Doctrine, esp. p. 52

and passim., and David Holloway, The Soviet Union and the Arms Race, 2nd ed.

(New Haven, Conn.: Yale University Press, 1983), esp. pp. 43–52.

30. See the opening Soviet statement at the SALT I negotiations, as quoted in

Holloway, The Soviet Union and the Arms Race, p. 46.

31. For example, Marshal of the Soviet Union N. V. Ogarkov, Istoriya uchit

bditel’nosti (History Teaches Vigilance) (Moscow: Voenizdat, 1985), pp. 75–77. See

esp. p. 77 for Ogarkov’s emphasis on Soviet ‘‘no first use’’ policy with regard to

nuclear weapons and his pejorative reference to U.S. nomenclature of strategic

‘‘offensive’’ forces. See also his discussion of U.S. plans for limited nuclear war

in Europe and for building up U.S. strategic forces under Reagan. Ibid., p. 68.

32. See, for example: Andrei Kokoshin, ‘‘Nuclear Arms Reduction and Strategic

Stability,’’ SShA: Ekonomika, politika, ideologiya, No. 2 (February 1988), pp. 3–

12, FBIS-SOV-88–051, March 16, 1988; Alexei Arbatov, ‘‘Parity and Reasonable

Sufficiency,’’ International Affairs (in English), No. 10 (October 1988), pp. 75–87;

Arbatov, ‘‘How Much Defence Is Sufficient?’’ International Affairs (in English),

No. 4 (1989), pp. 31–44; Lev Semeiko, ‘‘Razumnaya dostatochnost’—put’ knadezhnomy

miry’’ (Reasonable sufficiency—path to reliable peace), Kommunist,

No. 7 (May 1989), pp. 112–121.

33. Vitaliy Zhurkin, Sergei Karaganov, and Andrei Kortunov, ‘‘Security Challenges:

Old and New,’’ Kommunist, No. 1 (January 1988), pp. 42–50, JPRS-UKO-

88–006, March 24, 1988.

34. Daniil Proektor, ‘‘Politics, Clausewitz and Victory in Nuclear War,’’ International

Affairs (in English), No. 5 (1988), pp. 74–80. A large debate about politics

and nuclear war had been stimulated by an earlier article in the same journal:

see Boris Kanevsky and Pyotr Shabardin, ‘‘The Correlation of Politics, War and

a Nuclear Catastrophe,’’ International Affairs (in English), No. 2 (1988), pp. 95–


35. Graham Allison, Ashton B. Carter, Steven E. Miller and Philip Zelikow,

Cooperative Denuclearization: From Pledges to Deeds (Cambridge, Mass.: Center for

Science and International Affairs, Harvard University, January 1993). See also

William C. Martel and William T. Pendley, Nuclear Coexistence: Rethinking U.S.

Policy to Promote Stability in an Era of Proliferation (Montgomery, Ala.: Air War

College, April 1994), esp. pp. 49–66.

36. Suvoroy, quoted in V. Ye. Savkin, Osnovnye printsipy operativnogo iskusstva

i taktiki (Basic Principles of Operational Art and Tactics) (Moscow: Voenizdat,

1972), translated in U.S. Air Force Soviet Military Thought Series (Washington,

D.C.: U.S. Government Printing Office, n.d.), p. 244.

37. See, in particular, Scott and Scott, Soviet Military Doctrine: Continuity, Formulation,

and Dissemination, pp. 110–115.

The Red Army and Nuclear Weapons 43

38. Marshal N. V. Ogarkov, Vsegda v gotovnosti k zashchite Otechestva (Always

in Readiness to Defend the Homeland (Moscow: Voenizdat, 1982), p. 34.

39. Ibid., p. 16.

40. Ogarkov, Istoriya uchit bditel’nosti, pp. 68–69.

41. V. G. Reznichenko, I. N. Vorob’iev and N. F. Miroshnichenko, Taktika (Tactics)

(Moscow: Voenizdat, 1984), translated by CIS Multilingual Section, National

Defense Headquarters, Ottawa, Canada, p. 69.

42. Ibid., p. 67.

43. Ibid.

44. Ibid.

45. On partial surprise, see Vigor, Soviet Blitzkrieg Theory, pp. 156–157.

46. M. A. Gareev, M. V. Frunze—Voennyi teoretik (M. V. Frunze—Military theorist)

(New York: Pergamon/Brassey’s, 1988, in English), p. 208.

47. Andrey Kokoshin and Valentin Larionov, ‘‘Kurskaya bitva v svete sovremennoye

oboronitel’noye doctriny’’ (The Battle of Kursk in View of Contemporary

Defensive Doctrine), Mirovaya ekonomika i mezhdunarodnyye otnosheniya, No.

8 (1987), pp. 32–40.

48. Gareev, M. V. Frunze, p. 216. Gareev cites V. D. Sokolovskiy, ed., Voennaya

strategiya (Military Strategy) (Moscow: Voenizdat, 1963), p. 22.

49. Michael McGwire, Military Objectives in Soviet Foreign Policy (Washington,

D.C.: Brookings Institution, 1987), chs. 2 and 3.

50. See John G. Hines and Phillip A. Petersen, ‘‘The Changing Soviet System

of Control for Theater War,’’ International Defense Review, No. 3 (March 1986),

revised in Stephen J. Cimbala, ed., Soviet C3 (Washington, D.C.: AFCEA International

Press, 1987), pp. 191–219; John G. Hines, Phillip A. Petersen and Notra

Trulock III, ‘‘Soviet Military Theory from 1945–2000: Implications for NATO,’’

The Washington Quarterly, No. 4 (1986), pp. 117–137; and Raymond L. Garthoff,

‘‘Mutual Deterrence, Parity and Strategic Arms Limitation in Soviet Policy,’’ in

Derek Leebaert, ed., Soviet Military Thinking (London: Allen and Unwin, 1981),

pp. 92–124.

51. Gareev, M. V. Frunze, p. 216.

52. Ibid.

53. See Marshal of the Soviet Union and former Defense Minister Sergei Akhromeyev,

‘‘Doktrina predotvrashcheniya voiny, zashchity mira i sotsializma’’

(Doctrine for the prevention of war, the defense of peace and socialism), Problemy

mira i sotsializma, Vol. 12 (December 1987), pp. 23–28.

54. Kipp, ‘‘Soviet War Planning,’’ in Glantz, ed., The Initial Period of War on the

Eastern Front, p. 49.

55. Wardack and Turbiville, Voroshilov Lectures, Vol. 1, pp. 205–232.

56. Reznichenko et al., Taktika, p. 20.

57. Ibid., p. 21.

58. D. T. Yazov, ‘‘Novaya model’ bezopasnosti i vooruzhennyye sily’’ (A new

model of security and the armed forces), Kommunist, Vol. 18 (December 1989),

pp. 61–72.

59. Ibid., p. 66.

60. Dale R. Herspring, The Soviet High Command, 1967–1989: Personalities and

Politics (Princeton, N.J.: Princeton University Press, 1990), pp. 265–276. See also

Garthoff, The Great Transition, pp. 528–530.

44 Nuclear History and Its Lessons

61. Nicolai N. Petro and Alvin Z. Rubinstein, Russian Foreign Policy: From

Empire to Nation-State (New York: Longman, 1997), p. 151.

62. George M. Mellinger, ‘‘Survey: The Military Year 1989 in Review,’’ in

George M. Mellinger, ed., Soviet Armed Forces Review Annual, Vol. 13, 1989 (Gulf

Breeze, Fla.: Academic International Press, 1995), p. 2.

63. Thomas M. Nichols, The Sacred Cause: Civil-Military Conflict over Soviet National

Security 1917–1992 (Ithaca, N.Y.: Cornell University Press, 1993). On the

failed coup of August 1991, see John B. Dunlop, The Rise of Russia and the Fall of

the Soviet Empire (Princeton, N.J.: Princeton University Press, 1993), pp. 186–255,

esp. pp. 247–250.


Chapter 3

The Cuban Missile Crisis

and Its Legacy

Almost forty years after the Cuban missile crisis and at least a decade

into the post–Cold War era, scholars and policy makers might doubt that

there is anything new to be learned from the Cuban missile crisis of 1962.

The crisis has been the most studied confrontation of the nuclear age,

and for good reason. Neither before nor after the United States and Soviet

Union clashed in October 1962, were the two Cold War superpowers

as close to a failure of negotiation and an outbreak of general war. Despite

the abundance of historical and other studies of the crisis, conflicting

interpretations of its causes and the meaning of its outcome remain. 1

Explanations for the unprecedented and daring Soviet missile deployments

into the Western Hemisphere, and for the way in which the United

States responded once the missiles had been discovered, leave a great

deal of room for controversy, as we shall see.

If you believe most historians and political scientists who have studied

the Cuban missile crisis, an otherwise risk-averse Soviet leadership suddenly

decided in the spring of 1962 on a preposterous gamble that had

no chance of avoiding detection and, even if undetected prior to completion,

could only have provoked the Americans into a crisis or war.

Other accounts aver that the Kennedy administration, despite its confused

foreign policy and mixed messages to the Soviet Union about U.S.

resolve, improvised a model of ‘‘coercive diplomacy’’ and deterrence

that extricated the United States from the crisis on favorable terms and

forced Khrushchev to back down without humiliation. 2 These explanations

strain credulity because they are, in the main, not convincing.

46 Nuclear History and Its Lessons

The truth is more painful. Kennedy and Khrushchev stumbled into the

crisis and, by luck and dead reckoning, barely stumbled out of it without

war. Their impromptu nuclear ballet should be a warning, not a consolation,

with regard to a future of additional nuclear powers. My arguments

proceed in four steps. First, I show how both sides in the years

immediately preceding the crisis failed to appreciate the other’s perspectives

on the nuclear balance. Second, I ask how important nuclear

deterrence really was in determining the outcome of the crisis, compared

to other possible factors. Third, I consider the implications of U.S. Jupiter

missiles deployed in Europe. Intended as interim measures to shore up

nuclear deterrence for NATO, the Jupiters helped to cause the Cuban

contretemps and became unexpected nuclear hostages during the crisis.

Fourth, I discuss the possibility that nuclear weapons contributed to resolving

the missile crisis, not by the logic of deterrence, but by their

existential capacity to create fear of escalation and loss of control over



The U.S.-Soviet relationship throughout the Cold War years was

marked by simplification and misperception by each side of the political

objectives and military doctrine of the other. 3 As only one example, U.S.

policy analysts and government officials often described Soviet military

doctrine, including doctrine for the use of strategic nuclear weapons, as

first strike oriented and aimed at seeking victory through nuclear war. 4

U.S. analysts who sought to make this case could draw from some statements

made by party officials, military-technical literature including important

publications in the Officer’s Library Series, and some evidence of

Soviet research and development on future generations of nuclear offensive

and defensive weapons. Officials in the Carter and Reagan administrations

were especially concerned with Soviet ICBM capabilities for

preemptive attack on U.S. missile silos and command centers, and with

Soviet ballistic missile defenses already deployed and in development.

Had the same information included a larger component of U.S. awareness

of the Soviet strategic view, U.S. assessments might not have

seemed so ominous. Soviet interest in nuclear offensive and defensive

forces could have been interpreted as components of a strategy which

emphasized the deterrence of war and the limitation of damage, should

nuclear war occur. This interpretation might have been supported by the

recognition that Soviet interest in first strike strategies waned as their

military planners became less dependent on preemption for survivability

of their land-based forces. Then, too, the Soviet force configuration differed

from the American: U.S. capabilities spread over three legs of the

strategic ‘‘triad’’ placed most survivable U.S. striking power in subma-

The Cuban Missile Crisis and Its Legacy 47

rines and bombers. Soviet retaliatory potential as well as first strike capability

resided in the timely launch of their ICBMs, carrying a

disproportionate share of their hard target warheads. The argument that

Soviet ICBMs were targeted against U.S. ICBMs and therefore intended

as first strike weapons ignored the equally plausible inference that second

strike counterforce was considered by Soviet military planners, as

by U.S. defense officials, as a requirement for credible deterrence. 5

It was ironical that during the latter 1950s and early 1960s, both the

United States and the Soviet governments were marked by fears of strategic

insecurity, based on misperceptions of actual capabilities and of

military intentions. Khrushchev’s atomic diplomacy of the latter 1950s

had sought to exploit U.S. fears that Soviet competency in nuclear rocket

weapons greatly exceeded U.S. ability to develop and to deploy those

weapons. Khrushchev used extravagant claims of Soviet nuclear superiority

to buttress otherwise weak foreign policies and to fend off domestic

and foreign critics of his detente policies and military budget

cuts. 6 Addressing the Supreme Soviet in January 1960, Khrushchev asserted,

not for the first time, his claim to strategic superiority, argued

that world economic trends were moving in favor of socialism, and contended

that nuclear war, although certainly devastating for both sides,

would result in victory for socialism. 7

By the fall of 1960, Khrushchev had begun to retreat from some of his

more extravagant claims about Soviet nuclear superiority and about the

ability of socialism to prevail in nuclear war with comparatively few

casualties. His speeches more and more after the fall of 1960 emphasized

that the consequences of nuclear war for both sides would be highly

destructive. The same thematic focus on the mutually destructive effects

of nuclear war appeared in an article in the Communist party theoretical

journal, Kommunist, by General Nikolai Talensky in July of 1960, and

Talensky enlarged his presentation of the same themes in a later article

in Mezhdunarodnaya zhizn’ (International Affairs). 8 According to Talensky’s

later article, calculations showed that casualties in a world war

would be approximately 500 million to 600 million in the main theater

of military action (presumably Europe), of an estimated total population

of 800 million. 9 One reason for this retreat of Khrushchev and military

leaders from previous assertions of nuclear superiority and war winning

capability was the Soviet view that Chinese leaders were far too cavalier

about the consequences of nuclear war, challenging the Soviets for leadership

of the world communist movement on the basis of ideological

claims which disputed Soviet willingness to stand firm in confrontations

with the West.

Another reason for a pulling back in Soviet nuclear assertiveness was

a recognition by the Soviet leadership that the United States was much

more aware by 1960, and thereafter, of the actual state of the strategic

48 Nuclear History and Its Lessons

nuclear balance of power. Eisenhower’s last State of the Union address

provided an opportunity for the U.S. president to note that ‘‘the ‘missile

gap’ shows every sign of being a fiction.’’ 10 Almost immediately after

the Kennedy administration took office, press reports appeared which

claimed, on the basis of Pentagon studies, that there was actually no

‘‘missile gap.’’ 11 Citing new U.S. intelligence estimates, press reports in

September 1961, acknowledged that actual Soviet ICBM deployments in

1961 would fall far short of the maximum possible number projected in

earlier U.S. estimates. Therefore, the new intelligence estimates, according

to press reports, eliminated completely any notion of a missile gap

unfavorable to the United States. 12

Nor was this all. Beginning in October 1961 Kennedy administration

officials launched an offensive in public diplomacy to dispel the existence

of any missile gap favorable to the Soviet Union. Further, U.S. government

officials publicly proclaimed that the United States had now (by

autumn 1961) attained strategic nuclear superiority over the Soviet Union.

The public relations offensive began with the speech by Deputy

Secretary of Defense Roswell Gilpatric on October 21, 1961, and was

followed by similar statements from other high Kennedy administration

officials. Gilpatric noted that the United States, even after absorbing a

Soviet surprise first strike, would probably retain second strike forces

which were greater than the forces used by the Soviet Union in its attack.

‘‘In short,’’ according to Gilpatric, ‘‘we have a second-strike capability

which is at least as extensive as what the Soviets can deliver by striking

first.’’ 13 The actual balance of forces at the time of the Cuban missile

crisis (including forces becoming available at the very end of the crisis

period, through October 28) appears in Table 3.1.

As Richard Ned Lebow and Janice Gross Stein have noted, the reaction

of Khrushchev and his military advisors to this U.S. public diplomacy

was understandably one of concern, even alarm. 14 U.S. intelligence had

to have mapped correctly the locations of Soviet ICBMs (SS-6s) in order

to determine with such precision that the nuclear strategic balance was

so lopsidedly in the U.S. favor. Therefore, Soviet land-based missile

forces might be vulnerable to a U.S. first strike. The meaning of U.S.

strategic nuclear superiority might be not only the existence of a U.S. relative

advantage in nuclear striking power, but also the U.S. ability to

jeopardize the survival of the Soviet deterrent. The Soviets were quick

to respond to the Gilpatric speech. Two days later, Soviet Defense Minister

Malinovskiy, addressing the Twenty-Second Party Congress in Moscow,

charged that Gilpatric, with the concurrence of President Kennedy,

was ‘‘brandishing the might of the United States’’ and had ‘‘threatened

us with force.’’ 15 Malinovskiy added that ‘‘this threat does not frighten

us,’’ but obviously it did.

Soviet leaders gave similar and negative appraisals to statements on

Table 3.1

Strategic Nuclear Forces, U.S.-Soviet Balance, Cuban Missile Crisis

Note: Totals include Soviet MRBMs and IRBMs scheduled for initial deployment in Cuba

but not Soviet MRBMs or IRBMs deployed in the Soviet Union.

Sources: Author’s estimates from Raymond L. Garthoff, Reflections on the Cuban Missile Crisis,

rev. ed. (Washington, D.C.: Brookings Institution, 1989), p. 208; Garthoff, ‘‘New

Evidence on the Cuban Missile Crisis: Khrushchev, Nuclear Weapons and the Cuban

Missile Crisis,’’ Cold War International History Project Bulletin, No. 11 (Washington, D.C.:

Woodrow Wilson International Center for Scholars, Winter 1998), pp. 251–262.

50 Nuclear History and Its Lessons

the subject of nuclear weapons and nuclear war made by U.S. leaders,

including President Kennedy, subsequent to the Gilpatric speech. A

Spring 1962 interview with President Kennedy published in the Saturday

Evening Post was interpreted in the Soviet press as an attempt to intimidate

the Soviet leadership by threatening a U.S. nuclear first strike under

some conditions. 16 Taking note of the assertively optimistic trends in U.S.

official statements on the nuclear balance of power in July 1962, Khrushchev

described the new U.S. appraisals as meaningless. He argued that

the real military balance of power could only be determined in the course

of a war. 17 This was obvious backing and filling. The general trend in

Soviet statements about the strategic balance from mid-1961 was one of

the acknowledgment of, and the acceptance of, parity as the basis for

political relations between the two powers. 18

Khrushchev’s rocket rattling of the immediate post-Sputnik period had

set the stage for his own humiliation when the facts were revealed about

the true nature of the nuclear-strategic balance in 1961. Khrushchev’s

strategy of nuclear bluff was annoying to the United States and helped

to provoke a U.S. response which appeared to the Soviet Union as one

based on nuclear bullying. The result of Soviet nuclear bluffing followed

by U.S. nuclear bullying was that both sides moved further from a shared

understanding of the security dilemma created by their military competition,

and especially by their strategic forces.

A more accurate understanding of Soviet perspective might have suggested

to the Americans in 1962 that the Soviets were less concerned

with the ‘‘bean count’’ of U.S. compared to Soviet nuclear weapons, and

more concerned with the broader correlation of social and political

forces. The role of nuclear weapons and other forces in Soviet military

strategy had been to support Soviet policy, including the spread of revolutionary

Marxism-Leninism to states outside of the Soviet bloc (at least,

until Gorbachev). This could hardly be accomplished by nuclear adventurism

against an opponent with superior forces, as Khrushchev was

reminded by the Politburo when it decided it no longer required his

services. Soviet political strategy is not always compatible with the most

risky or assertive military strategy, as their willingness to adhere to the

SALT I and SALT II agreements, including the ABM Treaty, attested.

Soviet comprehension of the American standpoint might have helped

to avoid the misjudgment that the United States would accept the Soviet

missile deployments in Cuba. Although we do not know as much as we

would like to know about the Soviet interpretion of U.S. failure at the

Bay of Pigs, it seems safe to infer that the episode could not have impressed

them favorably with U.S. determination and sagacity. Khrushchev

must have wondered why Kennedy authorized the expedition

and then failed to rescue the situation when the chips were down. This

was a reasonable doubt of Kennedy’s resolve on Khrushchev’s part;

The Cuban Missile Crisis and Its Legacy 51

many Americans doubted it too. Kennedy’s reluctance to follow through

in the Bay of Pigs might have seemed to Khrushchev as a characteristic

propensity for hesitation in crises, instead of a singular uncertainty on

the part of a new president facing an unexpected debacle. Khrushchev

also seems to have erred in assuming that Kennedy perceived only a

domestic policy problem with regard to possible Soviet missile deployments

in Cuba. Soviet assurances to U.S. officials in September 1962 suggested

that no offensive missiles or other objectionable weapons would

be deployed in Cuba which might complicate matters for the U.S. president

during an election campaign.

President Kennedy and his advisors, in turn, were insufficiently sensitive

to the problem of how the Soviets might (wrongfully) interpret

U.S. domestic policy debates. Kennedy’s reassurances to members of

Congress that there were no ‘‘offensive’’ Soviet weapons in Cuba and

that he would not accept the deployment of offensive weapons in the

future drew a fine and legalistic line between offense and defense. Soviet

leaders might well have interpreted this distinction as a loophole which

could be exploited to justify the deployment of nuclear weapons in Cuba.

After all, whether weapons are defined as offensive or defensive is an

issue of purpose as much as it is an issue of technology. If, from the

Soviet perspective, the purpose of the missile deployments was to contribute

to the deterrence of an attack on Cuba, or to the defense of Cuba

if attacked, then the weapons could from that perspective be described

as defensive.

Some U.S. students of the crisis have concluded that only the U.S.

ultimatum and threat of an immediate air strike or invasion, conveyed

to the Soviet government through Ambassador Dobrynin October 27 by

Robert Kennedy, forced Khrushchev to agree to remove the missiles. An

equally plausible argument could be made that the game was up for

Khrushchev once the United States obtained unambiguous, photographic

evidence of the Soviet deployments. From then on, Khrushchev’s agenda

was to save as much political face as possible and to withdraw the missiles

while precluding a U.S. attack on Cuba. Obviously, from the standpoint

of sheer military power, there was little the Soviet Union could do

to prevent the United States from using its conventional force superiority

to overthrow Castro. On the other hand, the United States was not eager

to repeat the Bay of Pigs fiasco, so any U.S. invasion decision would

have had to commit major forces against a significant Soviet and Cuban

conventional defense. Khrushchev’s missile deployments almost gave the

United States a rationale for a difficult and costly military undertaking

which Kennedy would have been hard put to justify without the symbolism

of Soviet nuclear power deployed in the Caribbean. Khrushchev

turned a potentially dissuasive conventional force, effective against all

52 Nuclear History and Its Lessons

but massive invasion, into a lightning rod which would justify exactly

that kind of U.S. attack on Cuba.


The significance of deterrence in the Cuban missile crisis is not necessarily

what it appears to be. Some participants in the crisis and some

widely read accounts attributed Khrushchev’s willingness to withdraw

the missiles as a concession to U.S. nuclear superiority. The truth is more

complicated, and less reassuring to proponents of deterrence and nuclear

crisis management.

The United States was able to use deterrence to its advantage in Cuba,

once the Soviet missiles had been discovered and the U.S. president determined

to have them removed. But nuclear deterrence entered into the

picture only as a backdrop to the successful application of conventional

deterrence and a willingness to engage in crisis bargaining based on the

appearance of reciprocal concession. The credible threat to destroy Soviet

offensive missile emplacements in Cuba by air strike and land invasion,

and the corollary threat to remove the Castro regime from power, could

be accomplished with conventional forces alone. The burden of geographical

war widening or nuclear escalation would be Khrushchev’s,

not Kennedy’s.

The favorable outcome for U.S. crisis management should not obscure

the fact that nuclear and conventional deterrence failed prior to the crisis.

On the basis of what he must have known about the military balance in

nuclear and conventional forces, Khrushchev took an extreme risk in

placing Soviet missiles into Cuba. The explanation that he did so in order

to adjust an unfavorable strategic balance is consistent with U.S. deterrence

theory to a point, but at other points is not. 19

A strategic nuclear balance of power tilted seventeen to one in favor

of the United States should have deterred Khrushchev from his Cuban

initiative, according to both orthodox and heterodox schools of nuclear

deterrence strategy. The orthodox school argues that mutual vulnerability

and second strike capability are necessary and sufficient conditions

for the preservation of deterrence stability. The heterodox school contends

that mutual second strike capability is not enough for credible

deterrence when push comes to shove. The United States, in this second

model of credible deterrence, also requires for crisis management a significant

relative advantage in nuclear striking power or, in what amounts

to the same thing, in capability for damage limitation. 20

Neither the orthodox nor the heterodox model of deterrence would

allow for the kind of challenge Khrushchev made in the face of overwhelming

U.S. conventional and nuclear superiority. According to orthodox

logic, the United States in 1962 possessed a second strike

The Cuban Missile Crisis and Its Legacy 53

Table 3.2

U.S. and Soviet Survivable and Deliverable Forces, October 1962*

*This table compares U.S. survivors of a Soviet first strike and Soviet survivors of a U.S.

first strike. It is not therefore a classical ‘‘exchange model’’ but a statistical comparison

derived from an exchange model. Information about the exact model used is available

from the author.

capability against the Soviet Union; the Soviet Union did not have a

similar capability against the United States. And the heterodox requirements

for deterrence were also fulfilled: the United States had significant

advantages in nuclear striking power and in the ability to impose a relatively

favorable war outcome (if not an absolutely acceptable one). Making

reasonable assumptions about the performance parameters of Soviet

and U.S. weapons, Table 3.2 shows the plausible outcome of any nuclear

exchange in the last days of the Cuban missile crisis.

One can argue that Khrushchev was ‘‘irrational’’ according to the logic

of U.S. deterrence theory, but the observation elides the central issue of

whether deterrence logic has any explanatory power. Deterrence nomenclature

is pervasive in the literature, but demonstrating the explanatory

or predictive power of a deterrence model is something else. However,

54 Nuclear History and Its Lessons

deterrence supported by the credible ability to prevail in battle at an

acceptable cost to the threatener is another matter. The United States was

in this position in Cuba, unless nuclear weapons were brought into the

picture by the Soviet Union in the Caribbean or elsewhere. The United

States had established conventional ‘‘escalation dominance’’ in that it

could remove the missiles forcibly if it chose without nuclear escalation

or geographical war widening. The burden of further escalation was

placed upon Khrushchev, but no step that Khrushchev could have taken,

subsequent to a U.S. invasion of Cuba, could have saved Soviet missile

sites from destruction or, in all likelihood, the Castro regime from military

defeat. 21

Khrushchev attempted to implement his own model of ‘‘extended deterrence,’’

but it was more of a political than a military model. The ties

between Cuba, as a standard bearer of socialist community buried within

the U.S. sphere of influence, and its Soviet benefactor were not those of

a military guarantee. Castro sought an explicit Soviet defense guarantee

and wanted to go public with the news of Soviet missile deployments

in Cuba, but the Soviet leadership demurred on both counts. Cuba was

a prize worth keeping in the Soviet camp so long as the risks of doing

so fell well short of actual military conflict with the United States. Khrushchev

was not prepared to give Castro a blank check in the form of

excessive leverage over Soviet decisions for war and peace in the Caribbean.

Evidence for this comes from Soviet behavior before and during the

Cuban crisis, and some of the most interesting recent evidence appears

in crisis correspondence between Khrushchev and Castro recently published

in the December 2, 1990 issue of Granma. In a message to Khrushchev

on October 26, 1962 (two days before the crisis was resolved),

Castro tells Khrushchev that ‘‘aggression is almost imminent within the

next 24 or 72 hours’’ in the form of a U.S. air attack or invasion. 22 Castro

then conveys his ‘‘personal opinion’’ that if ‘‘the imperialists invade

Cuba with the goal of occupying it, the danger that that aggressive policy

poses for humanity is so great that following that event the Soviet Union

must never allow the circumstances in which the imperialists could launch the

first nuclear strike against it.’’ 23 Castro added in this message that if the

United States actually invaded Cuba, then ‘‘that would be the moment

to eliminate such danger forever through an act of clear legitimate defense,

however harsh and terrible the solution would be, for there is no

other.’’ 24

Khrushchev’s response to this request for a Soviet nuclear first strike

on the United States following any U.S. invasion of Cuba (sent October

28, the day that the Soviet Union agreed to remove the missiles in return

for a U.S. noninvasion-of-Cuba pledge) was to urge Castro ‘‘not to be

carried away by sentiment and to show firmness.’’ 25 Khrushchev argued

The Cuban Missile Crisis and Its Legacy 55

in response to Castro that the Soviet Union had settled the issue in Castro’s

favor by obtaining a noninvasion pledge from the United States

and by preventing war from breaking out. Khrushchev also offered the

argument that Pentagon ‘‘militarists’’ were now trying to frustrate the

agreement that he and Kennedy had reached. This was why, according

to Khrushchev’s response to Castro, the ‘‘provocative flights’’ of U.S.

reconnaissance planes continued. Khrushchev scolded Castro for shooting

down a U.S. reconnaissance plane October 27: ‘‘Yesterday you shot

down one of these, while earlier you didn’t shoot them down when they

overflew your territory.’’ 26 The Soviet leader implied that such triggerhappiness

would play into the hands of those in U.S. government circles

who wanted war: ‘‘The aggressors will take advantage of such a step for

their own purposes.’’ 27

The United States, despite its apparent military superiority at the nuclear

and conventional levels, was as ready to terminate the crisis without

war as was the Soviet Union. The U.S. objective was not to sever

completely the ‘‘extended deterrence’’ connection between the Soviet Union

and Cuba. U.S. crisis management objectives did emphasize, nonetheless,

two aspects of the U.S. view of the Soviet-Cuban connection. The

first was that, from Washington’s standpoint, the Cuban-Soviet relationship

was perceived as one of client and patron, or dependency. This was

emphasized in U.S. insistence upon dealing only with Khrushchev on

the conditions for removing the Soviet missiles. Second, the United States

and the Soviet Union resolved the crisis on terms which called for United

Nations inspection and verification of the Soviet missile withdrawal. Fidel

Castro objected on both counts. He disliked the willingness of Khrushchev

to arrange for crisis termination without having consulted Cuba

first. And, Castro refused to cooperate in permitting United Nations or

other on-site inspection of missile launcher dismantling and removal.

The United States and the Soviet Union worked around this obstacle by

arranging for the removal and shipment of the missiles in such a way

that the process could be verified by U.S. aircraft surveillance and by

other means. Castro objected to the terms on which the crisis was ended

on the grounds that they implied a relationship between Havana and

Moscow of one-way dependency, instead of two-way exchange.


Nuclear deterrence operated in the Cuban missile crisis as more than

a two-way street. It was not only a question of the balance of nuclear

forces based in North America or Russia and capable of striking directly

into one another’s homelands. Also important were the U.S. Jupiter intermediate-range

ballistic missiles (IRBMs) deployed in Turkey and Italy

during the Eisenhower administration as interim deterrents, while the

56 Nuclear History and Its Lessons

U.S. awaited its first deployment of intercontinental ballistic missiles

(ICBMs). By the time of the Cuban missile crisis, the Jupiters deployed

earlier in Turkey, liquid fueled and slow to respond, were technologically

obsolete, and yet politically provocative of Soviet first strike fears.

The Jupiters do speak for the relevancy of nuclear deterrence in the Cuban

missile crisis, but they do not speak favorably. The U.S. missiles in

Turkey helped to provoke the Soviet decision to deploy their own offensive

missiles to Cuba, and the Jupiters became hostages that could be

squeezed by the Soviets’ own version of coercive diplomacy once the

crisis was under way.

The decisions for U.S. IRBM deployments in Turkey and Italy were

taken in the aftermath of the Suez crisis of 1956, which shook allied

NATO confidence in American guarantees of European security, and in

the context of post-Sputnik American concerns about the viability of the

U.S. nuclear deterrent. 28 Some arguments used by U.S. proponents of the

Thor (in Great Britain) and Jupiter IRBMs were not too dissimilar from

those used by the Soviets on behalf of MRBM deployments (and planned

IRBM deployments) to Cuba in 1962. U.S. leaders feared after the initial

test launches of Soviet ICBMs in 1957 that they needed an interim fix for

a perceived status of missile inferiority (although not overall force inferiority,

given the size of U.S. bomber forces in the latter 1950s). The

Jupiter missiles deployed in Turkey were liquid fueled and used ‘‘soft’’

(above ground) launchers, which made them vulnerable to first strikes

or prompt retaliatory launches.

U.S. leaders saw the Jupiter missile deployments as a concession to

the requirements for NATO alliance unity. The host European nation

would ‘‘own’’ the missiles and launchers, but the United States would

maintain control over warhead dispersal and launch decisions (presumably

in consultation with the host state). For the Turkish government,

this meant that they had accepted a share of the U.S. nuclear deterrent

despite the obvious provocation this would provide for Moscow. The

strategic rationale for the Thor and Jupiter deployments was vitiated by

technology which made possible sea-based missile deployments and

ICBMs based in North America which could cover the same target base

in the Soviet Union or in Eastern Europe. The Kennedy administration

recognized that the Jupiters in Turkey constituted a technological dinosaur

and a potential political provocation. The president had decided in

principle to order the removal of the Jupiter missiles from Turkey prior

to the development of the 1962 Cuban crisis, but he had not pressed the

issue assertively after initial approaches to Turkey were rebuffed by that


The Cuban missile crisis thus caught the Kennedy administration with

obsolete nuclear missiles deployed in a forward, exposed position, obviously

vulnerable to Soviet conventional as well as nuclear attacks. More-

The Cuban Missile Crisis and Its Legacy 57

over, Soviet attacks against U.S. missiles in Turkey, in response to any

U.S. attack on Cuba, could draw the entire NATO alliance into a war

with Moscow. Kennedy and McNamara, during deliberations of the

ExComm, recognized the political irony that obsolete missiles deployed

in Turkey were now potential hostages to Soviet horizontal and vertical

escalation. In addition, McNamara was especially conscious of the danger

of escalation once a U.S. NATO ally was attacked in the aftermath

of fighting in Cuba.

McGeorge Bundy, special assistant to the president (Kennedy) for national

security, and James G. Blight transcribed and edited tapes of the

October 27, 1962 meetings of the ExComm, and portions of this material

appeared in the Winter 1987/1988 issue of the journal International Security.

29 In these transcripts, President Kennedy continually returns to

the theme that the Jupiter missiles offered Khrushchev an attractive way

out of his predicament that Kennedy might not be able to refuse. Khrushchev’s

‘‘second’’ letter of October 27 toughened the terms suggested

in his ‘‘first’’ letter of October 26, wherein he agreed to remove Soviet

offensive missiles from Cuba in return for a U.S. noninvasion pledge.

The October 27 letter (which may have been composed and sent first for

reasons still not fully known) insisted upon a trade of U.S. Jupiter missiles

in Turkey for Soviet missiles in Cuba. Kennedy was bothered by

the apparent symmetry of the trade, in the eyes of world, allied NATO

and U.S. opinion.

The president’s principal advisors, on the other hand, emphasized the

potential damage to NATO solidarity, to U.S.-Turkish relations, and to

future credibility of extended deterrence in Europe if the United States

made an obvious missile trade under the pressure of the Cuban crisis.

As the president kept returning to the apparent plausibility of a missile

trade, his advisors sharpened their cautionary notes about the impact

on NATO and future deterrence. One example is cited below, during

ExComm discussions on how to respond to Khrushchev’s two apparently

contradictory letters:

Kennedy: How much negotiation have we had with the Turks?

Dean Rusk, Secretary of State: Wehaven’t talked with the Turks. The Turks have

talked with us—the Turks have talked with us in—uh—NATO.

Kennedy: Well, have we gone to the Turkish government before this came out

this week? I’ve talked about it now for a week. Have we had any conversation

in Turkey, with the Turks?

Rusk: We’ve not actually talked to the Turks.

George W. Ball, Undersecretary of State: We did it on a basis where if we talked to

the Turks, I mean this would be an extremely unsettling business.

Kennedy: Well, this is unsettling now, George, because he’s got us in a pretty good

58 Nuclear History and Its Lessons

spot here, because most people will regard this not as an unreasonable proposal,

I’ll just tell you that. In fact, in many ways—

Bundy: But what most people, Mr. President?

Kennedy: I think you’re going to find it very difficult to explain why we are going

to take hostile military action in Cuba, against these sites—what we’ve been

thinking about—the thing that he’s saying is, ‘‘If you’ll get yours out of Turkey,

we’ll get ours out of Cuba.’’ I think we’ve got a very tough one here. 30

Kennedy’s advisors continue to express hostility to the idea of a missile

trade throughout the remainder of this discussion. Rusk comments that

‘‘The Cuba thing is a Western Hemisphere problem, an intrusion into

the Western Hemisphere.’’ Paul Nitze argues that the president should

try to get the missiles out of Cuba ‘‘pursuant to the private negotiation’’

(the terms of the first Khrushchev letter). Bundy cautions that a missile

trade, if accepted at this stage of the crisis, means that ‘‘our position

would come apart very fast.’’ Ball notes that if we talked to the Turks

about an immediate missile deal, they would take it up with NATO and

‘‘our position would have been undermined.’’ He adds that the United

States ‘‘persuaded them (the Turks) that this was an essential requirement’’

and now Turkey feels that a matter of prestige is involved. Bundy

argues that a missile trade would create the impression of trying to sell

out U.S. allies for American interests, adding that ‘‘that would be the

view in all of NATO.’’ 31

Despite this consensus of his advisors against the concept of a missile

trade, Kennedy held open until the very end of the crisis the option of

a missile trade. He approved a backchannel initiative from Dean Rusk

to the UN secretary general that would have resulted in a ‘‘UN’’ proposal

for a missile trade as the basis for resolving the crisis. The Rusk initiative

was developed as an option; the president had not made up his mind at

the time whether he would accept a missile trade if Khrushchev refused

to deal on the basis of the latter’s first letter. 32 It seems apparent from

the bulk of the historical evidence, especially from the evidence of the

past decade, that Kennedy would have traded Jupiter missiles in Turkey

for Russian missiles in Cuba publicly in order to get an agreement that

would have avoided war with the Soviet Union over Cuba.


The second counterargument to my assertion that nuclear deterrence

remained secondary to conventional deterrence in the Cuban missile crisis

is the importance of the perceived role of escalation in bringing the

crisis to a conclusion. Pressure to end the crisis in a timely manner came

not only from the possibility of a limited war in the Caribbean, but also

from the possible expansion of the fighting into general U.S.-Soviet con-

The Cuban Missile Crisis and Its Legacy 59

flict. Absent nuclear weapons, the crisis may have been much more prolonged,

and the terms on which it was resolved more ambiguous.

The counterargument that nuclear deterrence, in addition to conventional

deterrence, made a difference in the Cuban missile crisis assumes

a connection between nuclear weapons and nuclear deterrence which is not

necessarily proved. Whereas conventional deterrence may have operated

asymmetrically to support crisis management in favor of U.S. policy objectives,

the impact of nuclear weapons on decision making may have

been symmetrical. The Soviet and U.S. leaderships might equally have

feared loss of control more than either feared a deliberate first strike by

the opponent. Notice that this is different from conventional deterrence,

in which leaders’ hopes and fears are almost directly correlated with

expected battlefield outcomes. Leaders planning a conventional war may

still guess incorrectly, with disastrous results. Nevertheless, expected battlefield

outcomes in conventional war can be projected with more reliability

than they can for nuclear war scenarios, however subject to error

the former are necessarily going to be. The disconnection between nuclear

deterrence and military victory makes the effort to control crisis a

military as much as a political objective.

Evidence for this counter-counter argument would be fears on the part

of a side with a great deal of nuclear ‘‘superiority’’ in numbers of second

strike weapons that despite this superiority, loss of control could result

in nuclear war with unacceptable outcomes for the superior power.

Members of the U.S. ExComm decision-making group, including the

president and his leading cabinet officers, do show this perceptual inclination

to fear loss of control leading to nuclear escalation despite apparent

U.S. strategic nuclear superiority. Secretary of Defense Robert

McNamara, estimating the U.S. numerical advantage in strategic nuclear

weapons at seventeen to one in October 1962, nevertheless doubted that

this relatively advantageous position was meaningfully related to the

attainment of U.S. crisis management objectives without war. In his interview

with James G. Blight in May 1987, McNamara explained his reasoning

in terms that indicate the irrelvance of relative advantage for a

cost-benefit calculus in which unknown risks of absolute destruction are


Look, in my judgment, in fundamental terms, the so-called strategic balance

hasn’t shifted since 1962. The significant question isn’t: How many weapons did

we have then and now, relative to the Soviets? The question you should ask is:

What did each side have in its arsenal then and now that was, or is, militarily

useful? Let me put it another way: What is the likelihood then and now that

either side might initiate the use of nuclear weapons and come away with a net

gain? The answer to both questions is: Zero! Then and now, for both the U.S.

and the Soviet Union, there are no militarily useful nuclear weapons in their

arsenals and thus there is no advantage in using them. 33

60 Nuclear History and Its Lessons

Others on the ExComm did not agree with McNamara’s pessimism

about the irrelevancy of the nuclear balance of power. C. Douglas Dillon,

secretary of the treasury under Presidents Kennedy and Johnson and a

member of the ExComm, shared the view of Paul Nitze and other

‘‘hawks’’ that U.S. nuclear superiority was decisive in forcing Khrushchev

to back down. Dillon recalled in 1987 that as the crisis wore on

he became progressively less worried, in contrast to other ExComm participants

who became more nervous about possible war and nuclear escalation.

Dillon noted that in the Treasury Department he had not been

fully current for the last several years on the details of U.S. force structure.

He added:

I was not, when I first heard about it, fully aware of the extent of the nuclear

superiority that we had. And, when I became aware of that, then I changed my

view entirely and, of course, I agree totally with Nitze and think the McNamara

thesis that our nuclear superiority made little or no difference is dead wrong.

Our nuclear preponderance was essential. That’s what made the Russians back

off, plus the fact of our total conventional superiority in the region. 34

Dillon engages in transference here, of the logics of nuclear deterrence

and conventional dissuasion, and as the interview continues he advances

two supporting points to explain why others, including McNamara, discounted

U.S. nuclear superiority. First, the more experienced policy makers

on the ExComm were hawkish, according to Dillon, because they

had been through crisis management and decision-making situations before.

As he explained, ‘‘I think simple inexperience led to an inordinate

fear of nuclear damage, the fear of what might happen. McNamara, in

particular, felt that way, I guess, although I wasn’t so conscious at the

time that that was his reason.’’

One reason for the greater concern on the part of McNamara and other

ExComm ‘‘doves’’ about the risks of escalation was undoubtedly the

higher sensitivity of the Defense and State Departments to the implications

for Europe and NATO of a failure in crisis containment. McNamara

illustrates this sensitivity to the European implications of risk assessment

when he pushes his ExComm colleagues on October 27 to consider the

aftermath of a U.S. air strike and invasion of Cuba. McNamara persists

in raising the troubling issue of what U.S. responses will be if the Soviets

strike at Jupiter missile bases in Turkey. 35 Most other ExComm members

do not see the point, so McNamara drives home the danger of nuclear

escalation by sketching a plausible scenario. The ‘‘minimum’’ military

response by NATO to a Soviet attack on the Jupiter missiles in Turkey,

according to McNamara, would involve conventional war fighting in and

near Turkey, including strikes by Turkish and U.S. aircraft against Soviet

warships and/or naval bases in the Black Sea area. McNamara empha-

The Cuban Missile Crisis and Its Legacy 61

sizes that such exchanges would be ‘‘damned dangerous’’ and the implication

of imminent escalation to nuclear war fighting is obvious. 36 He

then argues that the United States defuse the Turkish missiles before any

invasion of Cuba (presumably making this public) so that the Turkish

missiles are removed from their hostage status.

In making this argument about the risks of escalation in an alliance

context, McNamara is not necessarily breaking faith with his earlier emphasis

on the priority of shared nuclear risk and the irrelevance of putative

nuclear superiority. But the acknowledgment of the hostage status

of the Jupiters in Turkey and their potentially catalytic role in crisis or

wartime escalation is an acknowledgment of the mistake made in deploying

those missiles. They were deployed, among other reasons, in

order to create deterrrence ‘‘coupling’’ between theater forces and strategic

nuclear forces. The assumption was that coupling would make extended

deterrence more credible than it would otherwise be, by adding

additional levels of U.S. force deployments in Europe between conventional

and all-out nuclear war. The same assumption helped to drive U.S.

and allied NATO rationales for the decision taken in 1979 to deploy

Pershing II and GLCMs in Western Europe (begun in 1983, and disbanded

as a result of the INF Treaty of December 1987).

As the Jupiter missiles in Turkey in 1962 became nuclear crisis management

hostages and potential catalysts of escalation, so, too, did the

NATO ‘‘572’’ deployments become hostages which slowed the momentum

of arms control and detente in Europe during the 1970s. The reason

for the irrelevance of Jupiters in 1962, as for the Pershing II and GLCMs

in the 1980s, had little to do with their technical characteristics (such as

the Jupiters’ vulnerability and long launch preparation, or the Pershings’

range). The political issue is that, because nuclear dissuasion cannot be

substituted for nuclear deterrence, ‘‘intermediate’’ nuclear weapons do

not necessarily support the successful management of crisis and the control

of escalation.

Instead, such weapons deployments can contribute to the deterioration

of crisis and to the loss of control over escalation. They can do this, as

the Jupiters did, by commingling horizontal escalation, or geographical

war widening, with vertical escalation, the expansion of conventional

into nuclear war fighting. Khrushchev’s attack on Jupiter missiles in Turkey

would have been a ‘‘nuclear’’ war even if he had only used conventional

weapons: nuclear weapons would be destroyed in the attack, and

perhaps fired back at the Soviet Union if its conventional first strike were

unsuccessful. (The same potential problem faced U.S. air strike planners

once Soviet MRBMs in Cuba were thought to be operational). This ‘‘vertical’’

expansion of the fighting could have been compounded in 1962

by ‘‘horizontal’’ extension of combat to Berlin or Turkey.

Against my arguments here, it might be contended that the issue of

62 Nuclear History and Its Lessons

intermediate nuclear weapons deployed in Europe was actually irrelevant

to the resolution of the Cuban missile crisis. One could take the

strict position that the missiles in Turkey were not a clandestine U.S.

deployment but a publicly acknowledged agreement under NATO auspices.

The Turkish missiles were a red herring introduced into the Cuban

missile crisis by Khrushchev, in search of a face-saving exit.

I concede that this argument has some validity, but it misses the distinction

between precision of policy objective (getting the missiles out of

Cuba without introducing irrelevant issues) and the potential for U.S.

nuclear weapons deployed abroad to contribute to inadvertent escalation.

Kennedy was right to keep the policy focus on the removal of Soviet

missiles from Cuba without a publicly acknowledged linkage to subsequent

removal of Jupiter missiles from Turkey. 37 On the other hand, transcripts

of ExComm deliberations and other evidence suggest that Kennedy also

recognized that complications created for his management of the missile

crisis by the presence of vulnerable, nuclear capable missiles deployed

so close to Soviet borders. Ironically, the Turkish missiles also served as

part of Khrushchev’s justification for deploying Soviet MRBMs and

IRBMs to Cuba: he would pose to the Americans a threat similar in scope

and in geographical proximity to that presented by U.S. IRBMs in Turkey

and in other European countries. 38

The irrelevance of Turkish missiles can be asserted only on the assumption

that what mattered in the resolution of the Cuban missile crisis,

and perhaps in the instigation of it also, was the strategic nuclear

balance of power. Although some members of the ExComm do assert

that this balance was of primary importance, other key policy makers,

including the U.S. president and secretary of defense, did not assume so

direct a connection between nuclear superiority and crisis management

prevalence. Resolution of the Cuban missile crisis in 1962 may suggest

that nuclear deterrence is loosely coupled to crisis management even in

a two-sided U.S.-Soviet confrontation, and perhaps even less relevant in

multisided crisis among nuclear armed states with less experience in

conflict resolution and intracrisis communication. 39


The ‘‘Cuban missile crisis’’ is not one point in time: it is misunderstood

as a confrontation between two nuclear-armed states over thirteen days

in October 1962. The Cuban missile crisis was the endgame of a series

of political and military blunders committed by governments in possession

of large arsenals and small brains. The Soviet leadership persuaded

itself that a nuclear violation of the Western Hemisphere close to American

shores would call forth no serious U.S. military response. The

United States stationed obsolete nuclear missiles in Europe that could

The Cuban Missile Crisis and Its Legacy 63

only have served to provoke Soviet insecurity and justify Khrushchev’s

nuclear adventurism. Especially in the latter 1950s and early 1960s, both

states engaged in rocket reductionism: when policy had no clear response

to a foreign security challenge, wave the nuclear sword.

Explanations for the Cuban missile crisis and its outcome as a successful

test of U.S. nuclear deterrence miss the point. Faith in nuclear

deterrence before the crisis contributed to its eruption when, and how, it

did. Once the crisis was in progress, both Kennedy and Khrushchev

sought to improvise around deterrence and find a path toward mutual

conciliation based on trade-offs. They did so in an explicit and tacit bargain

that required immediate Soviet withdrawal of offensive missiles

from Cuba and eventual American removal of Jupiter missiles from Europe.

To the extent that deterrence was relevant to the outcome of the

Cuban missile crisis, U.S. conventional deterrence based on the credible

threat to prevail in any Caribbean military confrontation was more important

than nuclear deterrence.

These conclusions speak not only to the nuclear past, but also to the

nuclear future. Kennedy and Khrushchev at the time represented the

world’s two largest military powers and nuclear states. This was the

World Series of Cold War confrontations between the ‘‘A’’ teams of

world politics. Each state had had more than a decade to adjust to the

idea that both were armed with weapons of unprecedented destructiveness.

Yet the performances of the political and military high commands

in both states, prior to and during the crisis, were marked by erratic

personalities, uncertain decision-making processes, preposterous military

advice, and wishful thinking in favor of preferred scenarios. Can we

expect better performances from the nuclear-armed states of the future,

and if not, can we assume equally fortuitous escapes from disaster? For

how long?


1. See Raymond L. Garthoff, ‘‘New Evidence on the Cuban Missile Crisis:

Khrushchev, Nuclear Weapons and the Cuban Missile Crisis,’’ Cold War International

History Project Bulletin, No. 11 (Washington, D.C.: Woodrow Wilson International

Center for Scholars, Winter 1998), pp. 251–262.

2. Alexander L. George, ‘‘The Cuban Missile Crisis, 1962’’ in George, David

K. Hall and William E. Simons, eds., The Limits of Coercive Diplomacy: Laos, Cuba,

Vietnam (Boston: Little, Brown, 1971), pp. 86–143. Coercive diplomacy, according

to George, is a defensive strategy that is used to deal with the efforts of an

adversary to change the status quo in its favor. The objectives of coercive diplomacy

can include persuading an opponent to stop short of its goal, convincing

the opponent to undo an action already taken or persuading the opponent to

make changes in its government or regime that will bring about the desired policy

change. See George, ‘‘Coercive Diplomacy: Definition and Characteristics,’’ in

64 Nuclear History and Its Lessons

George and William E. Simons, eds., The Limits of Coercive Diplomacy, 2nd ed.

(Boulder, Colo.: Westview Press, 1994), esp. pp. 8–9, and in the same volume,

George, ‘‘The Cuban Missile Crisis: Peaceful Resolution Through Coercive Diplomacy,’’

pp. 111–132.

3. Raymond L. Garthoff, Deterrence and the Revolution in Soviet Military Doctrine

(Washington, D.C.: Brookings Institution, 1990), esp. pp. 6–28.

4. The various editions of the Pentagon’s Soviet Military Power during the

Reagan years provide ample evidence for my point here. For an overview of this

issue, see Robert L. Arnett, ‘‘Soviet Attitudes Towards Nuclear War: Do They

Really Think They Can Win?’’ Journal of Strategic Studies, No. 2 (September 1979),

pp. 172–191.

5. For evidence, see President’s Commission on Strategic Forces (Scowcroft

Commission), Report (Washington, D.C., April 1983).

6. John Lewis Gaddis, We Now Know: Rethinking Cold War History (Oxford:

Clarendon Press, 1997), esp. pp. 222–223 and 235–236 on Khrushchev’s strategy

for exploiting fictive nuclear superiority.

7. Pravda, January 15, 1961.

8. Talenskiy, in Kommunist, No. 7 (1960), pp. 31–41 said that it was necessary

to ‘‘emphasize that a future war, if the aggressors succeeded in unleashing it,

will lead to such an increase in human losses on both sides that its consequences

for mankind might be catastrophic.’’ In his International Affairs article, Talenskiy

drew an explicit comparison between the destruction of Soviet cities at the hands

of Nazi Germany and the destruction attendant to nuclear rocket war, arguing

that the degree of destruction in nuclear war would be ‘‘magnified a thousand

times’’ compared to World War II and extended over whole continents. Talenskiy,

Kommunist, No. 7 (1960), pp. 31–41, and Talenskiy, Mezhdunarodnaya zhizn’,

No. 10 (1960), p. 33, both cited in Arnold L. Horelick and Myron Rush, Strategic

Power and Soviet Foreign Policy (Chicago: University of Chicago Press, 1966),

pp. 78–79.

9. Horelick and Rush, Strategic Power and Soviet Foreign Policy, pp. 78–79.

10. Ibid., p. 80.

11. Ibid.

12. Ibid., p. 83.

13. Gilpatric, quoted in ibid., p. 84.

14. Richard Ned Lebow and Janice Gross Stein, We All Lost the Cold War

(Princeton, N.J.: Princeton University Press, 1994), ch. 2.

15. Horelick and Rush, Strategic Power and Soviet Foreign Policy, p. 85.

16. Ibid., pp. 86–87. Kennedy had actually said that he would not rule out the

possibility of a U.S. first strike under some conditions, which was consistent with

previous U.S. policy guidance for nuclear weapons employment in the Eisenhower


17. Khrushchev, quoted in Pravda, July 11, 1962, cited in Horelick and Rush,

Strategic Power and Soviet Foreign Policy, p. 87.

18. Horelick and Rush, Strategic Power and Soviet Foreign Policy, p. 87.

19. On the Soviet decision for deployment of missiles to Cuba, see Garthoff,

Reflections on the Cuban Missile Crisis, rev. ed. (Washington, D.C.: Brookings Institution,

1989), pp. 6–42, and Ernest R. May and Philip D. Zelikow, The Kennedy

The Cuban Missile Crisis and Its Legacy 65

Tapes: Inside the White House during the Cuban Missile Crisis (Cambridge, Mass.:

Harvard University Press, 1997), pp. 666–671.

20. I have admittedly collapsed a wide spectrum of opinion into two boxes

here. For more complete discussion, see Robert Jervis, The Meaning of the Nuclear

Revolution (Ithaca, N.Y.: Cornell University Press, 1989); Colin S. Gray, Nuclear

Strategy and National Style (Lanham, Md.: Hamilton Press, 1986); and David W.

Tarr, Nuclear Deterrence and International Security: Alternative Nuclear Regimes

(White Plains, N.Y.: Longman Publishing Group, 1991).

21. See Gen. Anatoli I. Gribkov and William Y. Smith, Operation ANADYR:

U.S. and Soviet Generals Recount the Cuban Missile Crisis (Chicago: Edition Q

Books, 1994), passim. With regard to nuclear force loadings, Soviet tactical nuclear

weapons (in addition to those intended for SS-4 and SS-5 launchers) numbered

between 98 and 104: 80 for two regiments of front cruise missiles; 12 for Luna

surface to surface, short range missiles; 6 for gravity bombs for IL-28 bombers;

and possible additional charges for nuclear-armed naval mines. The approximate

maximum range for the cruise missiles, fired from Cuban shore points nearest

to U.S. territory, was 90 miles: warheads for these cruise missiles were in the

five to twelve kiloton range. The 60 warheads deployed for SS-4 and SS-5 launchers

in Cuba (the latter never actually reached Cuba, having been turned back by

the U.S. quarantine) were in addition to these tactical weapons. All but the SS-5

launchers apparently reached Cuba in a single shipment on October 4. The existence

of a Soviet tactical nuclear force of this size, unknown to U.S. invasion

planners, indicates that for Moscow the psychological investment in defense of

Cuba was apparently much higher than Americans, then and subsequently during

the Cold War, believed. The potential for inadvertent escalation was obviously

much greater than crisis participants could have known. I am very grateful

to Raymond Garthoff, Brookings Institution, for updated information on Cuba,

based on his extensive discussions with Russian defense experts and many years

of study devoted to this issue.

22. Granma, December 2, 1990, p. 3. Letter from Castro to Khrushchev, October

26, 1962. I am grateful to Ned Lebow for first calling this to my attention.

23. Ibid. Italics supplied.

24. Ibid. Italics supplied.

25. Letter from Khrushchev to Castro, October 28, 1962, Granma, December 2,

1990, p. 3.

26. Ibid.

27. Ibid.

28. David N. Schwartz, NATO’s Nuclear Dilemmas (Washington, D.C.: Brookings

Institution, 1983), esp. pp. 62–66 and 73–74. The Gaither Report had also

advocated overseas U.S. IRBM deployments even before Sputnik (ibid., p. 65).

U.S. Jupiter missiles were formally handed over to the Turks on October 22, the

day of President Kennedy’s televised address announcing U.S. discovery of the

Soviet missiles in Cuba and the decision to impose the quarantine in response.

James G. Blight and David A. Welch, On the Brink: Americans and Soviets Reexamine

the Cuban Missile Crisis (New York: Hill and Wang, 1989), p. 172.

29. McGeorge Bundy, transcriber, and James G. Blight, ed., ‘‘October 27, 1962:

Transcripts of the Meetings of the ExComm,’’ International Security, No. 3 (Winter

1987/88), pp. 30–92.

66 Nuclear History and Its Lessons

30. Ibid., pp. 36–37.

31. Ibid., pp. 38–39.

32. Blight and Welch, On the Brink, pp. 170–171 and 173.

33. McNamara, quoted in Blight and Welch, On the Brink, p. 187.

34. Ibid., p. 153.

35. Bundy and Blight, ‘‘October 27, 1962,’’ pp. 72ff.

36. Ibid., p. 75.

37. Numerous arguments to this effect by members of the ExComm appear in

Blight and Welch, On the Brink, p. 110.

38. In his memoirs, Khrushchev noted that ‘‘In addition to protecting Cuba,

our missiles would have equalized what the West likes to call ‘the balance of

power.’ The Americans had surrounded our country with military bases and

threatened us with nuclear weapons, and now they would learn just what it feels

like to have enemy missiles pointing at you; we’d be doing nothing more than

giving them a little of their own medicine.’’ Strobe Talbott, ed. and trans., Khrushchev

Remembers (Boston: Little, Brown, 1970), p. 494. A summary of ExComm

deliberations on Khrushchev’s possible motives for the deployment appears in

Roger Hilsman, The Cuban Missile Crisis: The Struggle over Policy (Westport, Conn.:

Praeger Publishers, 1996), pp. 79–81. A more recent appraisal of pertinent evidence

on Soviet decision making in the Cuban missile crisis appears in May and

Zelikow, The Kennedy Tapes, esp. pp. 666–671. May and Zelikow note that most

historians have favored the ‘‘defense of Cuba’’ hypothesis for Khrushchev’s decision,

although ExComm participants regarded the impact on the strategic nuclear

balance as more important. They argue that Kennedy and his advisors ‘‘did

not make any serious misjudgments about the Soviets’’ although the Soviet decision-making

process was veiled in obscurity. ‘‘The more we learn about Soviet

decision-making in the Khrushchev era, the less confidence we can feel in any

analyses that explain decisions in terms of a hierarchy of interest calculations’’

(pp. 666, 668). My analysis accepts the second point, about the obscurity of Soviet

decision making, but rejects the argument that U.S. decision making avoided

serious misjudgments, both before and during the crisis.

39. It will be important for U.S. and Soviet scholars to work together in order

to establish more confidence in such arguments, admittedly tentative. Some Soviet

scholars are now applying modelling and simulation techniques to the analysis

of the Cuban missile crisis. See, for example, V. P. Akimov, V. B. Lukov, P. B.

Parshin and V. M. Sergeyev, ‘‘Karibskiy krizis: opyt modelirovaniya’’ (The Caribbean

crisis: Experience of modelling), SShA: Politika, ekonomika, ideologiya, No.

5 (1989), pp. 36–49.

Part II

Nuclear Policy and Strategy

in the Future

Chapter 4

Nuclear Proliferation: Fortuitous

Past, Uncertain Future


In the 1960s, experts predicted that within several decades nuclear weapons

would be as abundant as mosquitoes and that the likelihood of nuclear

war was, therefore, inevitably rising. Things did not turn out that

way. Many explained the absence of major war in Europe for 45 years

as the result of political realism and nuclear deterrence. Some theorists

and policy makers now predict that the slow spread of nuclear weapons

can be made compatible with future international peace and stability by

mixing the same ingredients: realism and deterrence. 1 These optimistic

expectations about the role of nuclear weapons in the post–Cold War

world are likely to be disappointed.

Post–Cold War confidence in the perpetual success of nuclear deterrence

in avoiding war is even more misplaced. The reasons why are both

theoretical and practical. The theories on which the case for peaceful

coexistence between stability and nuclear proliferation rest, once examined

with care, provide an inadequate platform to support optimism.

Practice also cautions against optimism. The non-nuclear states that seek

to acquire nuclear weapons, other weapons of mass destruction (WMD),

and ballistic missile delivery systems include an unrepresentative sample

of antisystemic rogues, including some with regional scores to settle.

Nuclear optimists assume that these states will reason as past nuclear

powers have about the costs and risks of nuclear brinkmanship or war.


The argument that the post–Cold War world may be compatible with

a hitherto unknown, and unacceptable, degree of nuclear weapons

70 Nuclear Policy and Strategy in the Future

spread rests on some basic theoretical postulates about international relations.

These basic assumptions are derived from the ‘‘realist’’ or neorealist

school of international political thought. 2 The temptation to parse

these arguments for fine points that distinguish among old realists, new

realists and hemi-, demi- and semi-realists will be resisted here. We are

interested in the realist-derived assumptions that are specifically related

to nuclear proliferation. Realist principles have considerable explanatory

power and predictive utility at a very high level of abstraction, thus their

appeal to scholars. Realism also has an inherent pessimism about some

aspects of international relations, thus its appeal to worldly heads of state

and military planners.

Proponents of international realism confronted nuclear technology

with mixed reactions. The nuclear revolution separated the accomplishment

of military denial from the infliction of military punishment. The

meaning of this for strategists was that military victory, defined prior to

the nuclear age as the ability to prevail over opposed forces in battle,

now was permissible only well below the level of total war. And less

than total wars were risky as never before. Nuclear realists admit that

these profound changes have taken place in the relationship between

force and policy. They argue, however, that the new relationship between

force and policy strengthens rather than weakens some perennial

principles of international relations theory. Power is still king, but the

king is now latent power in the form of risk manipulation and threat of

war, instead of power actually displayed on the battlefield. Peace is now

guaranteed by threat of war unacceptable in its social consequences, instead

of being dependent upon the defender’s credible threat to defeat

the attacker’s armed forces in battle.

Despite some attempts to reconcile political realism with nuclear

weapons, the union remained a shotgun marriage. Military professionals

remained unimpressed with the substitution of manipulation of risk for

the actual ability to prevail in combat. Strategic theorists were frustrated

by mutual vulnerability based on assured retaliation and the apparent

futility of any defenses against nuclear weapons during the Cold War.

Political leaders kept demanding refinements in nuclear war plans that

would somehow make them less than totally self-defeating, including

options for graduated attacks and for the preservation of some measure

of deterrence into nuclear war itself. Fortunately, the Cold War ended

before either the United States or the Soviet Union could test the assumption

of limited nuclear war.

Notwithstanding these tensions, realism supported nuclearism during

the Cold War by affirming the significance of international power politics

for global stability. Realists are skeptical that international stability and

peace can depend upon negotiation, trust or the exploitation of common

interests among states. Peace is the result of fear of war, or, where victory

Nuclear Proliferation 71

in combat at an acceptable cost is not possible, fear of unacceptable societal

consequences due to war. Nuclear deterrence thus closed with realism

on the ability of a threatened state or alliance to pose the credible

threat of assured retaliation against any attacker. Retaliation having been

assured, no sensible or rational attacker would strike. This logic would

apply across the board to any possessor of nuclear weapons regardless

of its culture, ideology, or leaders.


The nuclear version of international realism has a number of intellectual

and policy-prescriptive weaknesses. There are four problem areas:

whether the realist view is based on exceptional cases, whether the economic

theories on which some realist arguments about deterrence stability

are based can be transferred from economics to international

politics, whether realism can account for both general and immediate

deterrence situations, and whether rational decision making as conceived

by realists is dependable for explanation and prediction of arms race and

war-provoking behavior.

Atypical Cases

First, realist arguments for the possibility of a stable, nuclear, multipolar

world are based on the Cold War experiences of the United States

and the Soviet Union. The supposition is that, just as the U.S. and the

Soviet political and military leaderships worked out rules of the road for

crisis management and the avoidance of nuclear war, so, too will new

nuclear powers among the current non-nuclear states. However, there

are reasons to doubt whether the U.S. and Soviet experiences can be

repeated after the Cold War. First, the U.S.-Soviet nuclear relationship

between 1945 and 1990 was also supported by bipolarity and by an approximate

equality, although an asymmetrical one, in overall U.S. and

Soviet military power. Neither bipolarity nor, obviously, U.S.-Russian

global military equity is available to support stable relations in the post–

Cold War world; in fact, both are irrelevant so long as Russia evolves in

a democratic, capitalist direction and prefers cooperative U.S.-Russian

foreign relations.

A second reason why the U.S. and Soviet Cold War experiences are

unlikely to be repeated by future proliferators is that the relationship

between political legitimacy and military control was secure in Moscow

and in Washington during the Cold War. A politically tame military

cannot be assumed for some plausible future additions to the list of nuclear

powers. The issue here is not whether democracies are less warlike

than dictatorships are. The question is whether the regime can impose

72 Nuclear Policy and Strategy in the Future

either assertive or delegative military control over its armed forces and,

if it does, the consequences for its crisis management and normal nuclear


Assertive control implies a great deal of civilian intervention in military

operations and management, delegative control, more willingness to let

the military have their own way on operational and organizational issues.

Strict rules about nuclear custody are an example of assertive control;

for example, in the early years of the nuclear age, atomic weapons

were withheld from the military under normal conditions. An example

of delegative control is the understanding among U.S. Cold War policy

makers that, in the event of a nuclear attack disabling the president and/

or the civilian chain of command, the U.S. deterrent would not be paralyzed.

Military commanders could, under carefully defined and

admittedly drastic conditions, launch in response to unambiguous indications

of attack. 3

Organizational process factors and other decision-making attributes of

states with small, new nuclear arsenals may push their militaries toward

doctrines that favor nuclear preemption. 4 First strike–vulnerable forces

may invite attack on themselves. Newly acquired nuclear arsenals may

not be ‘‘fail safe’’ against accidental launch or military usurpation of civil

command prerogative. Among nuclear aspirants in 1998, several states,

including North Korea, Iran, Iraq (temporarily thwarted by UN inspections)

and Libya, the distinction between ‘‘civil’’ and ‘‘military’’ was as

opaque to many outside observers as it was to some of their own poorly

informed citizens. In Pakistan, a declared nuclear power since May 1998,

the military has run the nuclear weapons development program from

the time of its inception to the present. Neither Indian nor Pakistani

nuclear release protocols are clear to outside observers, and uncertainty

marks U.S. understanding of ‘‘first use’’ or ‘‘no first use’’ doctrines in

New Delhi and Islamabad. The possibility cannot be excluded that nuclear

command authority rests de facto in the hands of brass hats, unaccountable

to civil control, in any one or more of the new nuclear

powers or nuclear aspiring states. For example, according to one U.S.

study prepared for Lawrence Livermore National Laboratory in 1991:

The nuclear chain of command is likely to reflect the military’s dominance in

Pakistani decision making. Thus, although formal authority to launch a Pakistani

nuclear strike could be expected to reside with the President (as does control

over the nuclear program today), the Pakistani military is all but certain to obtain

de facto control over nuclear weapons and the decision to use them. A future

civilian government could ignore the military’s advice on nuclear use only at its

own peril. 5

Another reason why the U.S.-Soviet experience may not be normative

for newer nuclear powers is that there were no pieces of territory or

Nuclear Proliferation 73

other vital interests for which one of the sides was committed to go to

war rather than to suffer defeat or stalemate. The two sides were generally

satisfied by bloc consolidation and by internal power balancing instead

of external adventurism and zero sum competition for territory or resources.

The preceding observation does not imply that Cold War crises,

such as those which occurred over Berlin and Cuba, were not dangerous.

They were dangerous, but the danger was mitigated by the awareness

that neither state had to sacrifice a vital piece of its own territory or its

own national values (allies were another matter) in order to avoid war.

What was at stake in the most dangerous U.S.–Soviet Cold War confrontations

was ‘‘extended’’ deterrence, or the credibility of nuclear protection

extended to allies, and not defense of the homeland per se.

Who Calls the Shots?

The second major set of theoretical problems with nuclear realism lies

in the adaptation of arguments from microeconomic theory to theories

of interstate relations. Kenneth Waltz explicitly compares the behaviors

of states in an international system to the behavior of firms in a market.

As the market forces firms into a common mode of rational decision

making in order to survive, so, too, does the international system, according

to Waltz, dictate similar constraints upon the behavior of states.

The analogy, however, is wrong. The international system does not dominate

its leading state actors: leading states define the parameters of the

system. The international system, unlike the theoretical free market, is

sub-system dominant. The ‘‘system’’ or composite of interactions among

units is the cross product of the separate behaviors of the units. 6

International politics is a game of oligopoly, in which the few rule the

many. Because this is so, there cannot be any ‘‘system’’ to which the

leading oligopolists, unlike the remainder of the states, are subject

against their wishes. The system is driven by the preferred ends and

means of its leading members on issues that are perceived as vital interests

to those states or as important, although not necessary vital. 7 Realists,

especially structural realists who emphasize the number of powers

and their polarities as determinants of peace and war, assume that some

‘‘system’’ of interactions exists independently of the states which make

it up. This is a useful heuristic for theorists, but a very mistaken view of

the way in which policy is actually made in international affairs. Because

realists insist upon reification of the system independently of the principal

actors within the system, they miss the sub-systemic dominance

built into the international order. Napoleon Bonaparte and Adolf Hitler,

for example, saw the international order not as a system that would

constrain their objectives and ambitions, but as a series of swinging

doors, each awaiting a fateful, aggressive push.

74 Nuclear Policy and Strategy in the Future

An important test of whether meaningful theory can proceed on the

basis of the realist, or realpolitik, premise of ‘‘system’’ separateness, or

whether domestic political forces must also be taken into account by

theorists, is to test realist and domestic/constrained hypotheses against

historical evidence. According to Bruce Bueno de Mesquita and David

Lalman, the realist perspective as formalized in their models is not supported

by the past two centuries’ experience of interstate behavior. 8 The

authors deduce an ‘‘acquiescence impossibility’’ theorem which shows

that, in a logically developed game structure based on realist assumptions,

it is impossible for one state to acquiesce to the demands of another

‘‘regardless of the beliefs held by the rivals, regardless of the initial demand

made by one of the states, and regardless of initial endowments

of capabilities, coalitional support, propensities to take risks, or anything

else.’’ 9 None of the deductions derived from the realist or neorealist versions

of their international interactions game, according to Bueno de

Mesquita and Lalman, were supported in the empirical data set that

included 707 dyadic interactions. 10

One might argue, in defense of realists on this point, that the assumption

of system determinism is a useful falsehood. It allows for parsimony

in expression and in focus on the essential attributes of the international

system. But, again, the assumption of ‘‘apartness’’ of the system and its

essential state or non-state actors is only useful, and methodologically

defensible, if it leads to insights which are both accurate and not otherwise

attainable. Neither exceptional accuracy nor exceptional attainability

of insight has been demonstrated by realists for the assumption of

system and actor ‘‘apartness.’’ This is probably one reason why some

realist thinkers did not exclude what Waltz, in another study, refers to

as first and second image variables. 11 Realism fails to explain the high

degree of international cooperation which takes place despite a legally

anarchic international order because of the biased manner in which realism

deals with imperfect information. According to Bueno de Mesquita

and Lalman:

In the realist world, imperfect information can only encourage violence. Incorrect

beliefs about the intentions of rivals can only steer disputes away from negotiation

(or the status quo) and toward the blackmail inherent in a capitulation or

the tragedy inherent in a war. Incorrect beliefs, secrecy, misperception, misjudgment,

and miscalculation are routine features of human intercourse. In that sense,

a realist world could be a dangerous world indeed. 12

As Robert Jervis has noted, one can divide international systems theorists

according to whether the ‘‘system’’ is treated as an independent

variable, as a dependent variable, or as both. 13 Waltz contends that the

Nuclear Proliferation 75

most important causes of international behavior reside in the structure

of the international system, i.e., in the number of powers and in their

positions relative to one another. 14 Jervis notes that Waltz’s structure

omits some important variables and processes that are neither at the

system or actor level, for example, technology and the degree and kind

of international interdependence. 15

There is another difficulty inherent in Waltz’s approach. System polarity

is virtually identical with system structure in his analysis. But this

near-identity of polarity and structure is flawed. Polarity is more the

result of past state and non-state actor behaviors than it is the cause of

future behaviors. Cold War bipolarity was the result of World War II, of

nuclear weapons, and of the fact that leaders perceived correctly the futility

of starting World War III in Europe. Leaders’ perceptions of the balance

of power are an intervening variable between polarity and outcomes

such as stability, including peace or war. In other words, leaders’ perceptions,

including their risk aversion or risk acceptance, are the efficient

causes for international behavior; ‘‘systems’’ and polarity are formal


The difference between efficient and formal causes is important for

theories that purport to be empirically testable. Formal causes are proved

by an abstract process that follows a deductive chain of reasoning. Efficient

causes are demonstrated by observation of temporal sequences and

behavioral effects. International systems theorists who emphasize the importance

of structure have been more successful at proving formal than

efficient causes. There is merit in doing so, and Waltz and others who

have argued from this perspective deserve credit for their rigor and for

the insights derived from their perspective. 16

The danger for international systems theorists lies in transferring inferences

from the realm of deductive logic to the world of policy explanation

and prediction. For example, Waltz argues both that (1) because

there were only two Cold War superpowers, each had to balance against

the other at virtually any point, and (2) disputes among their allies could

not drag the United States and Soviets into war because they could satisfy

their deterrence requirements through internal balancing, rather than

alliance aggregation. 17 The first argument is at least partly inconsistent

with the second, and neither is confirmed by Cold War evidence. The

United States and Soviets sometimes conceded important disputes to one

another in order to avoid the possibility of inadvertent war or escalation,

as in the U.S. refusal to expand the ground war in Vietnam on account

of expected Soviet and Chinese reactions. And allies sometimes did drag

the superpowers into crisis and under credible threat of war, as the Israelis

and Egyptians did in 1973.

76 Nuclear Policy and Strategy in the Future

General Versus Immediate Deterrence

The preceding discussion also points to the third general set of problems

with realist theories and nuclear weapons spread. The structure of

the international system is not related to general deterrence in the same

way as it is related to immediate deterrence. According to Patrick M.

Morgan, the need for general deterrence is inherent in the normal dayto-day

relations of states, based on the distribution of power and states’

assumptions about one another’s intentions. General deterrence is the

latent possibility that many states may opt for war within an anarchic

or non-hierarchical international order. 18 Immediate deterrence is a situation

in which one side has actually made specific threats against another,

the second side perceives itself threatened, and a significant

likelihood of war exists in the minds of leaders in at least one of the two

states. For example, the onset of a crisis often signifies a failure of general

deterrence, but as yet immediate deterrence has not failed because states

have not yet abandoned diplomacy and crisis management for battle.

It makes sense to assume that there might be a strong correlation between

success or failure in general deterrence and system attributes such

as distributions of actor capabilities and objectives. However, the relationship

between international systems and failures of immediate deterrence

is much more indirect. State and sub-state variables, including the

attributes of individuals, groups and bureaucratic organizations, are

among the filters through which any ‘‘system’’ forces must pass before

those forces are manifest in state decisions and policies. The distinction

between general and immediate deterrence helps to explain why perfectly

logical deductions from deterrence theory based on rationality postulates

often fly in the face of states’ actual behavior. 19

The significance of the distinction between general and immediate deterrence

is illustrated by the Cuban missile crisis. (A detailed discussion

of this crisis and its implications for the future of nuclear deterrence

appears in Chapter 3). The decision by Khrushchev to put Soviet medium-

and intermediate-range ballistic missiles into Cuba was intended,

among other objectives, to diminish the publicly acknowledged (by U.S.

government officials) gap between U.S. and Soviet strategic nuclear capabilities.

Khrushchev’s decision, made in the spring of 1962 after consulting

very few key advisors, represented a failure of general

deterrence. The Soviet leadership had decided to risk the emplacement

of its nuclear weapons outside of Soviet territory and in the Western

Hemisphere for the first time. However, it was not yet a failure of immediate

deterrence. Immediate deterrence was not involved in Khrushchev’s

clandestine deployment program because the deployments

were deliberately kept secret. Had Khrushchev carried through his orig-

Nuclear Proliferation 77

inal plans, he would have completed the missile deployments and then

announced their existence.

In that eventuality, the mere existence of Khrushchev’s missiles on

Cuban soil, however threatening it seemed to U.S. policy makers, would

not have created a situation of immediate deterrence. Only the completion

of deployments followed by a coercive threat would move the situation

from a failure of general deterrence (Soviets make a dangerous

move in the arms race) to one of immediate deterrence (for example,

Soviets now demand that U.S. and allies leave West Berlin immediately).

The preceding supposition is of the ‘‘what if’’ or counterfactual kind. We

may never know the full story of Khrushchev’s motives for the missile

deployments. 20 The actual shift from a general to an immediate deterrence

situation took place on October 22, when President Kennedy ordered

the Soviet missiles removed from Cuba, announced that the U.S.

was imposing a quarantine on Soviet shipments to Cuba, and stated that

a nuclear missile launched from Cuba against any target in the Western

Hemisphere would call forth a full U.S. retaliatory response against the

Soviet Union.

Realist perspectives help to explain the background to general deterrence

failure in this instance, but they do little to clarify why the U.S.

and Soviet political leaderships chose as they did. If the international

power positions of states yield unambiguous inferences about their crisis

management strategies, Khrushchev should never have dared to put missiles

into Cuba. And the United States, once the missiles had been discovered,

need not have hesitated to invade Cuba or to launch an air

strike to destroy the missile sites, collocated air defense sites, and other

nuclear-capable weapons platforms deployed in Cuba by Moscow. 21 Realists

would argue, against the preceding statement, that nuclear weapons

made the Soviets and the Americans cautious during the Cuban

missile crisis. The danger created by nuclear weapons helped to end the

crisis without war, following the logic and against my earlier argument.

However, realist arguments will not work in this context. Nuclear

weapons did not make the crisis easier to manage, but harder. They

added to the risk of escalation, to be sure, and leaders were well aware

of these risks. The U.S. deliberately and, some would say, successfully

manipulated the risk of escalation and war in order to force Khrushchev’s

withdrawal of the missiles. But the argument, for nuclear coercion

as the path to Cuban crisis settlement, will not work because

nuclear weapons, and the Soviet sense of inferiority in the nuclear arms

race, were major causes for the crisis. 22 If it is argued that nuclear weapons

helped to resolve the crisis, that is true only as a historical tautology:

having caused it or helped to cause it and by making it more dangerous,

they played a part in ending it.

Realism aided by an appreciation for historical serendipity and inde-

78 Nuclear Policy and Strategy in the Future

terminacy fares better in explaining the Cold War relationship between

nuclear weapons and peace. For example, John Lewis Gaddis contends

that nuclear weapons influenced post–World War II international relations

in at least four ways. First, nuclear weapons helped to support an

already existing reluctance of the great powers to wage war against one

another. Second, states that possessed nuclear weapons became more

risk averse. Third, nuclear weapons did not create bipolarity after World

War II, but they did prolong its life, and so, too, helped to prolong stability.

Fourth, nuclear weapons helped to perpetuate the Cold War by

saving the United States, the Soviet Union and their allies’ military expenditures

on conventional forces, expenditures which if necessary

might have forced rethinking of Cold War assumptions sooner. 23

This summary shows that nuclear stability asserted by realists was

purchased with significant trade-offs. For example, Gaddis’ fourth form

of nuclear influence acknowledges that political relations between the

United States and the Soviet Union remained adversarial longer than

necessary, in part due to ingrained habits of military hangover. To the

realists’ contention that nuclear weapons made war less likely because

fighting became more dangerous, Gaddis’ fourth argument for nuclear

relevancy points to the downside of that contention. The very weapons

of mass destruction which some would contend were instruments of

deterrence or peace were also causes of U.S. and Soviet leaders’ fears of

devastating surprise attack. The capabilities of these weapons were so

unprecedented that the very fact of their being targeted at another state

made a relationship hostile in military-operational terms even when it

had passed into a stage of nonhostility in policy.

Decision Rationality

As anticipated in the comments above, a fourth problematical aspect

of realism as the basis for optimism about nuclear weapons spread is the

question of rational decision making. The assumption of rational decision

making is a necessary condition for making testable hypotheses and verifiable

generalizations about social behavior, In and of itself, the rationality

postulate does no harm. It becomes dangerous, however, when it

is assumed that particular notions of rational decision making can be

transferred from one culture or society to another. U.S. policy makers

have on more than one occasion substituted assumptions for evidence

or intelligence about the behavioral propensities or mind-sets of foreign

leaders. For example, throughout the summer and early autumn of 1962,

U.S. leaders simply assumed that Khrushchev would not dare to put

Soviet missiles into Cuba because it was illogical and too risk acceptant,

by U.S. reasoning, for the Soviet leader to have done so. As another

example, American policy makers between 1965 and 1968 assumed that

Nuclear Proliferation 79

selective bombing of targets in North Vietnam would increase the pressure

on the regime in Hanoi to withdraw its support from the National

Liberation Front in South Vietnam.

When assumptions based on U.S. decision rationality are not supported

by experience, leaders sometimes cling to the assumptions or to

their supporting logic and blame the other side for ‘‘irrational’’ or illogical

behavior. Khrushchev’s deployment of missiles in Cuba in the face

of U.S. warnings against doing so has been described as irrational by

many American, and even some Soviet, sources. Yet in his memoirs

Khrushchev gives two reasons, equalizing the balance of nuclear power

and deterring U.S. attack on Cuba, which make plain sense from his

political and military vantage point. Similarly, the North Vietnamese reaction

to U.S. bombing from 1965 to 1968 was to increase their support

to the NLF and their commitment to ultimate victory over the government

of South Vietnam and its American supporters. The U.S. bombing

could destroy value targets in North Vietnam, but it could not remove

from Hanoi its capability to support insurgency in the south. Nor could

bombing impose any unacceptable cost to North Vietnamese military

capabilities for large scale, conventional ground warfare, later put to use

in the final push by Hanoi against Saigon in 1975.

U.S. policy makers assumed in July 1990 that it would not be prudent

for Saddam Hussein to attack and occupy Kuwait. The Iraqi leader was

thought by most American prewar assessments to be using coercive diplomacy

against Kuwait on account of its uncooperative oil pricing behavior.

Saddam Hussein also miscalculated Bush administration

perceptions of U.S. and allied interests in the region, and he misestimated

U.S. domestic politics as still being caught up in a Vietnam syndrome

that would preclude President Bush from the actual use of force. Even

after weeks of pounding from the U.S. and allied coalition air forces in

January 1991, Saddam disbelieved that the United States would initiate

a ground war on account of fear of excessive numbers of American and

allied casualties.

Looking inside the heads of enemy leaders, especially those of idiosyncratic

and impulsive dictators, is never easy. But the preceding examples

hold some pertinent social science lessons. Explanatory and

predictive approaches that may suffice for issue areas such as welfare,

urban development, education and other largely domestic matters are

not necessarily optimal for explaining behavior pertinent to war and

peace. In these other issues with less than ultimate stakes, it makes sense

to base predictions of future behavior on typical past behavior and on

culturally shared norms and values. However, in international behavior

related to war and peace, it is more important to be able to explain and

predict atypical behaviors between states and leaders who do not share

cultural norms and values. In other words, the marginal utility of being

80 Nuclear Policy and Strategy in the Future

able to explain typical, as opposed to atypical, behavior declines as the

situation moves from one of general to immediate deterrence (see above).


The Present Condition

The spread of nuclear weapons is not an isolated danger, but one

heavily bound up with the proliferation of other weapons of mass destruction

(chemical and biological) and of the means for their delivery.

Arguments that the spread of nuclear weapons will contribute to military

gridlock in East Asia or in the Middle East, as in Europe between 1946

and 1990, ignore the synergistic threat to stability brought about by the

lethal combination of chemical, biological and nuclear weapons, and improved

air and land based delivery systems, among aspiring regional

hegemons or dissatisfied states. That having been said, the problem of

nuclear proliferation is sufficiently serious in its own right. Table 4.1

summarizes the status of nuclear proliferation, groups states into pertinent

categories and offers supplementary notations about selected actors.

The status of North Korea has been mired in a complicated shell game

of U.S. political relations with both Koreas, of bureaucratic politics on

the American home front, and of a confrontationally oriented U.S. government

approach to North Korea, up to the very edge of a near outbreak

of war in 1994. 24 North Korea’s standoff with the United States and the

International Atomic Energy Agency (IAEA) over its nuclear production

program, culminating in the Framework Agreement of 1994 intended to

cap that program, is well known. 25 During the 1980s and 1990s, North

Korea was able to develop a complete nuclear fuel cycle, including a

capability for the production of plutonium at its Yongbyon nuclear research

center. A plutonium production reactor became operational in

1986 with refueling in 1989, making available weapons grade plutonium

for at least one nuclear weapon. North Korea was also building a 50megawatt

reactor at Yongbyon and a 200-megawatt reactor at Taechon

before construction was halted under the Framework Agreement. The

50-megawatt reactor could have produced enough plutonium for North

Korea to make between seven and ten nuclear weapons per year. 26

North Korea was also developing a diversified industry for the production

of ballistic missiles of various ranges, including missiles for export.

North Korea in 1996 deployed with its forces several hundred

Scud-B and Scud-C ballistic missiles with maximum ranges of 300 and

500 kilometers, respectively. Its Nodong 1 intermediate-range (estimated

1,000 kilometers with a 1,000 kilogram payload) ground mobile, liquidpropelled

missile was first tested in 1990 and entered service in 1994. By

December 1994, according to some reports, between twelve and eighteen

Table 4.1

Status of Nuclear Proliferation

1. India and Pakistan declared themselves nuclear powers after each completed a series of

tests in May 1998. India is estimated to have 60 to 80 weapons and Pakistan 10 to 15.

Neither state is a member of NPT.

2. Israel is thought to have between 70 and 125 weapons. Israel is not a signatory of NPT.

3. North Korea’s nuclear program is supposedly frozen under International Atomic Energy

Agency (IAEA) safeguards. In 1994 Pyongyang signed the Agreed Framework with the

United States calling for North Korea to freeze and subsequently give up all parts of its

nuclear weapons program. In return, the United States promised to arrange for North

Korea to receive two 1,000 megawatt light-water reactors, plus annual allotments of

500,000 tons of heavy fuel oil until the first LWR is completed. Implementation of the

Agreed Framework has been assigned to the Korean Peninsula Energy Development

Organization (KEDO), also including South Korea, Japan and the European Union.

4. Iran is a member of NPT. The United States suspects that Iran seeks a nuclear weapons

program and has tried to prevent other states from providing Teheran with pertinent

technology or know-how. Russia agreed in 1995 not to sell uranium enrichment technology

to Iran, and China promised in 1997 to end civil nuclear cooperation with Iran.

5. According to UN Security Council Resolution 687, the UN Special Commission for Iraq

(UNSCOM) and IAEA were to verify the complete elimination of Iraq’s nuclear, chemical

and biological weapons, its ballistic missiles, and its means for producing these

weapons and delivery systems. After U.S. bombing attacks on Iraq in late 1998, Iraqi

head of state Saddam Hussein ejected UNSCOM from the country, and it is unclear as

of this writing when, or if, inspections can resume.

6. Libya is a member of NPT, but the United States maintains that the regime nevertheless

wants to acquire nuclear weapons.

Sources: Updated and adapted from Arms Control Association, Fact Sheet, The State of

Nuclear Proliferation (Washington, D.C., May 1998). For Iraq’s weapons of mass destruction,

see Scott Ritter, Endgame: Solving the Iraq Problem—Once and for All (New York:

Simon and Schuster, 1999), esp. pp. 217–224. See also Commission to Assess the Ballistic

Missile Threat to the United States (Rumsfeld Commission), Report (Executive

Summary) (Washington, D.C., July 15, 1998).

82 Nuclear Policy and Strategy in the Future

Nodong-1 missiles were in service. 27 A follow-on Nodong 2 with increased

range and reduced payload has been reported. Estimated range

is 1300 to 1500 kilometers with a payload from 500 to 750 kilograms. 28

Also in development are two missiles of longer range than Nodong, the

Taepodong 1 and 2, with estimated ranges greater than 1,500 and 4,000

kilometers, respectively. 29

During the summer of 1998 North Korea test fired a three-stage ballistic

missile rocket over the Sea of Japan. The DPRK government described

the test as an intended satellite launch that was less than

completely successful, but some U.S. observers drew the conclusion that

North Korea had demonstrated a prototype capability for missile attacks

well beyond the tactical or theater range. According to the Commission

to Assess the Ballistic Missile Threat to the United States in its 1998

report to Congress, a North Korean decision to rapidly deploy the Taepodong

2 ballistic missile might not be known to U.S. intelligence very

far in advance of the decision to launch. The capabilities of Taepodong

2 once deployed, according to the commission, are potentially strategic

in reach and impact:

This missile could reach major cities and military bases in Alaska and the smaller,

westernmost islands in the Hawaiian chain. Light-weight variations of the TD-2

could fly as far as 10,000 km, placing at risk western U.S. territory in an arc

extending northwest from Phoenix, Arizona to Madison, Wisconsin. 30

A summary of some attributes of North Korea’s missiles of longer than

tactical ranges appears in Table 4.2.

An important aspect of North Korea’s ballistic missile program is that

it is designed for export as well as for Pyongyang’s own defense needs.

Hundreds of Scud missiles have been provided by North Korea to countries

in the Middle East, including Iran, and North Korea is already marketing

the Nodong for export. The U.S. Department of Defense estimates

that, thus far, North Korea has not become an international supplier of

nuclear, chemical or biological weapons technology, despite its aggressive

marketing of missiles and missile technology. Pyongyang has a substantial

chemical weapons capability and limited facilities for producing

biological weapons. 31

North Korea’s ballistic missile export program has enhanced the threat

to stability in Southwest Asia. Iran acquired Scud-B missiles from Libya

and North Korea and Scud-C missiles from the latter in the 1980s. During

just three years of the Iran-Iraq war, between 1985 and 1988, Iran fired

almost 100 Scud-B missiles at Iraq. In addition to obtaining ballistic missiles

from North Korea, Iran is attempting to set up its own missile production

capability. Acquisition of the North Korean Nodong missile

would permit Iran to attack targets in Israel, much of Saudi Arabia and

Nuclear Proliferation 83

Table 4.2

DPRK Selected Ballistic Missile Capabilities, May 1999

(1.) A Space Launch Vehicle (SLV) capable of placing a 100-kilogram payload into Low

Earth Orbit (LEO) is, in theory, able to deliver a 200-kilogram warhead to a range of

about 10,000 kilometers.

Sources: Excerpts from chart originally prepared by Joseph S. Bermudez, Jr., May 11, 1999,

via Ed Ivanhoe Website, and Jane’s Strategic Weapons Systems, September 30, 1998. janes/.

the Trucial States, Turkey, Russia and other former Soviet states, Pakistan

and India. 32 Iran’s original reason for acquiring ballistic missiles was to

employ them in its protracted war against Iraq during the 1980s. Having

acquired the taste for technology and having taken note of Iraq’s post–

Desert Storm weakness in the 1990s, Iran now reasonably aspires to the

status of first among equals among Gulf states. In addition to its ballistic

missile capabilities, Iran has Chinese-supplied cruise missiles, artillery

and aircraft capable of delivering chemical and biological weapons and

Russian-built SU-24 fighter-bombers that can deliver nuclear weapons. 33

Rumors of Russian-Iranian nuclear cooperation and of leakage of nuclear

weapons experts from the former Soviet Union into Iran are frequent in

U.S. and other media sources.

Can Deterrence Still Work, and When?

The marriage of nuclear and other weapons of mass destruction to

long-range ballistic or cruise missiles expands the capabilities of aspiring

regional hegemons and of state or non-state actors determined for various

reasons to undo the international status quo. Even with short-range

ballistic missiles of the Scud-B or Scud-C range and accuracy, states

84 Nuclear Policy and Strategy in the Future

capable of producing chemical and biological, not to mention nuclear,

weapons can wreak havoc against their regional neighbors. Iraq’s military

buildup prior to the Gulf War of 1991 is a case in point. The world

was fortunate that Saddam Hussein’s exhausted exchequer motivated his

attack on Kuwait in 1990 instead of five years later. Iraq’s massive military

establishment included a multipronged strategy for acquiring nuclear

weapons and a substantial chemical and biological weapons

arsenal. Iraq’s nuclear program was, according to authoritative sources,

‘‘massive’’ and ‘‘for most practical purposes fiscally unconstrained.’’ 34

The pre–Desert Storm Iraqi nuclear program was also ‘‘closer to fielding

a nuclear weapon, and less vulnerable to destruction by precision bombing,

than coalition air commanders and planners or U.S. intelligence specialists

realized before Desert Storm.’’ 35 Coalition target lists on January

16 included two suspect nuclear production facilities; postwar UN inspectors

uncovered more than twenty sites related to Iraq’s nuclear program,

including sixteen described as ‘‘main facilities.’’ 36

In addition to the uncertainties surrounding Iraq’s prewar nuclear

weapons program, Saddam Hussein’s mobile Scud missiles played havoc

with coalition intelligence during Desert Storm and threatened to cause

a political crisis within the anti-Iraqi alliance. Scud attacks on Israeli cities

created the possibility that Tel Aviv might retaliate, thus bringing Israel

directly into the war and giving Saddam a wedge issue to divide Arab

members of the U.S.-led coalition from others. According to the U.S. Air

Force–commissioned Gulf War Air Power Survey:

Efforts by Coalition air forces to suppress Iraqi launches of Scud missiles against

Israel, Saudi Arabia, and other Persian Gulf nations during Desert Storm ran into

many of the same problems evident in the case of the Iraqis’ nuclear weapons

program. Key portions of the target set—notably the pre-surveyed launch sites

and hiding places used by the mobile launchers—were not identified before 17

January, and, even in the face of intense efforts to find and destroy them, the

mobile launchers proved remarkably elusive and survivable. 37

Soviet exercises with Scuds in Eastern Europe and Iraqi practices during

the Iran-Iraq war suggested to coalition air war planners that a sufficient

number of prelaunch signatures and adequate time might be

available to permit attacks on mobile launchers, before they fired, by

patrolling aircraft. Iraqi countermeasures disappointed these expectations:

their Gulf War use of mobile Scuds, compared to earlier cases,

reduced prelaunch set-up times, avoided telltale electromagnetic emissions

that gave away locations and deployed large numbers of decoys

in order to confuse coalition targeters. 38

The case of Iraq is instructive for optimists about the stability of a new

world order marked by proliferation of weapons of mass destruction and

Nuclear Proliferation 85

modern delivery systems. Consider how different the problem facing the

United States would have been if Iraq had invaded Kuwait in 1996 instead

of 1990. The United States in 1990 did not face an Iraqi adversary

already equipped with usable nuclear weapons. The United States had

available for Desert Storm the large, forward-deployed forces built up in

the Cold War years for a theater-strategic campaign against the Soviet

Union and its Warsaw Pact allies. The Soviet Union under Gorbachev

decided with some reluctance to support the UN authorization for the

forcible expulsion from Kuwait of its former ally in Baghdad. Iraq’s unwillingness

to employ chemical or biological weapons against the United

States was probably related to its expectation that a U.S. nuclear retaliation

might follow, to which Iraq could not respond in kind. An Iraq in

1996 (absent Desert Storm) possessing nuclear charges capable of delivery

by air or missile, even over distances of several hundred kilometers,

could have posed a threat against outside intervention by the United

States and its NATO allies, or against regional antagonists like Saudi

Arabia and Israel, very different from the threat it posed in 1990.

The preceding statement seems almost self-evident, but there is more

than a self-evident point built into it. Iraq successfully concealed from

the most technically complex intelligence systems in the world the prewar

location of most of the installations related to its nuclear weapons

program. Iraqi mobile Scuds confounded coalition air war planners to

the extent that there exists not even a single documented case of mobile

Scud destruction by coalition fixed-wing aircraft. 39 Notably, this level of

frustration marked the efforts of the winning side in a very one-sided

military contest, an essentially post-industrial strategy for warfare

against a static defensive strategy accompanied by political ineptitude in

Baghdad of the highest order. 40 In addition, the United States and its

allies had five months to build up forces, collect intelligence and plan

countermeasures against Saddam’s anticipated moves while Iraqi forces

inexplicably squatted down in the Kuwaiti theater of operations. All

these considerations point to the uniqueness of the environment surrounding

Desert Storm and contain tacit warnings about the potential

mischief of a future Saddam, strategically tutored and more decisive.

The U.S. inability or unwillingness to deter Iraq’s invasion of Kuwait

in 1990 contains another warning about realist optimism and proliferation.

The basic maxims of deterrence learned during the Cold War years

may have to be rethought, or in some cases rejected outright, in the

remainder of the century and thereafter. Nuclear weapons and war

avoidance worked together during the Cold War because U.S.-Soviet

strategic nuclear bipolarity enforced a connection between basic and extended

deterrence. One could predict the degree of vulnerability to coercion

to be expected of U.S. allies by deduction from the stability of the

U.S.-Soviet relationship itself. U.S. strategic nuclear forces were coupled

86 Nuclear Policy and Strategy in the Future

to the fates of European and Japanese allies who could not then be coerced

into submission by the Soviets, nor by third parties allied to the

Soviets, without accepting unknown risks of escalation into confrontation

with the U.S. deterrent.

The collapse of bipolarity after the Cold War diminishes the link between

basic and extended deterrence. One can make fewer reliable predictions

about states’ behaviors on the basis of ‘‘system’’ variables. The

significance of this theoretical construct for the practical problem of nonproliferation

is illustrated by then Secretary of Defense William J. Perry’s

comment that future terrorists or rogue regimes ‘‘may not buy into our

deterrence theory. Indeed, they may be madder than MAD.’’ 41 Deterrence

theory a la the Cold War, based on realist premises that assume

risk-averse and cost-benefit-sensitive leaders, may no longer hold tenable

for leaders armed with weapons of mass destruction and motivated by

‘‘irrational’’ or ‘‘illogical’’ objectives by at least U.S. standards. As Keith

B. Payne has explained:

Assuming that deterrence will ‘‘work’’ because the opponent will behave sensibly

is bound to be the basis for a future surprise. I do not know whether our

expectations of a generically sensible opponent will next be dashed by a so-called

rogue state, such as North Korea, or by another challenger. That they will be

dashed, however, is near certain. As we move into the second nuclear age and

confront opponents with whom we are relatively unfamiliar, assumptions of a

generically sensible foe almost certainly will ensure surprises. 42

In addition, most academic or policy analyses have focused on dyadic

relationships that are not complicated by triangulation or indirect deterrence.

43 Current U.S. strategy relies upon the ability to conduct two

nearly simultaneous major theater wars against rogue states or aspiring

regional hegemons by exploiting the superior reach, battlefield knowledge

and striking power of U.S. high-technology, conventional weapons.

Future rogues or other regional aggressors might seek to neutralize the

U.S. conventional deterrent by acquiring weapons of mass destruction

and by threatening U.S. forces, regional allies, or even noninvolved

states. Threats to use nuclear weapons against noninvolved states’ cities

might play against U.S. or allied fears of civilian casualties. As Robert

Harkavy has noted:

But conceivably, an Iraq or Iran could threaten to use nuclear weapons against

other countries—perhaps those not aligned with the United States or nominally

under its extended deterrence protection—banking on American reluctance to

countenance massive civilian casualties anywhere. Either Iran or Iraq could indeed

threaten to attack the other with nuclear weapons. Iraq could so threaten

its own Kurdish cities; Iran could threaten cities to its north in the new Central

Asian nations. 44

Nuclear Proliferation 87

Another reason why deterrence might not work in a post–Cold War

proliferated world is that reliable and timely intelligence and warning

might not be available about the intentions or capabilities of rogues with

WMD and ballistic missile capabilities. According to the Rumsfeld Commission,

the U.S. intelligence community had great difficulty assessing

the pace and scope of North Korea’s Nodong missile program and may

have very little advance warning of the deployment of Taepodong 2. 45

The commission report states that Iran has a nuclear weapons program

intended to produce nuclear weapons as soon as possible, and the technical

capability to demonstrate an ICBM–range ballistic missile similar

to the TD-2 within five years of a decision by Iran to do so. Unfortunately,

according to the Rumsfeld Commission, the United States is unlikely

to know whether Iran has produced nuclear weapons ‘‘until after

the fact.’’ 46 An Iranian ballistic missile with a 10,000-kilometer range

‘‘could hold the United States at risk in an arc extending northeast of a

line from Philadelphia, Pennsylvania, to St. Paul, Minnesota.’’ 47 The

Rumsfeld Commission concluded, with regard to the possible dangers

presented by short or no warning ballistic missile attacks and WMD


A new strategic environment now gives emerging ballistic missile powers the

capacity, through a combination of domestic development and foreign assistance,

to acquire the means to strike the U.S. within about five years of a decision to

acquire such a capability (10 years in the case of Iraq). During several of those

years, the U.S. might not be aware that such a decision had been made. Available

alternative means of delivery can shorten the warning time of deployment nearly

to zero. 48

The preceding arguments do not prove that deterrence, old style, can

never work in the new world order, including deterrence based on nuclear

weapons. But deterrence in the next century will be more conditional,

culturally driven and less technology oriented than it was during

the Cold War. States owning weapons of mass destruction and ballistic

missiles will present a mosaic of hard to read intentions that defy easy

characterization by standard intelligence collectors. Deterrence, having

been overdetermined in the Cold War, may lead the pack of underachievers

before the twenty-first century is very old.


The slow spread of nuclear weapons during the Cold War has led

some analysts and policy makers to unjustified optimism about the manageability

of a faster rate of nuclear proliferation in the post–Cold War

world. The conceptual arguments for a proliferation-acceptant world,

88 Nuclear Policy and Strategy in the Future

compatible with stability, are based on a model of political realism that

relies on photocopy policy makers. Optimists about proliferation also

expect that deterrence will work as well after the Cold War as it did

between 1946 and 1991. Even realists must concede, however, that the

spread of nuclear weapons among a more heterogenous cast of actors

after 1991 makes the probability of nuclear first use, and the forces causing

leaders to take that plunge, less predictable than hitherto. As the case

of India and Pakistan shows, not only are more states acquiring nuclear

weapons, but those weapons are showing up in dyadic relationships

already strained to the limit by prior conventional wars and mutual suspicions.

This is not to argue that proliferation is all of a piece. It is not. There

is indeed a case to be made for well-managed, as opposed to poorly

managed, proliferation in a less than perfect world. 49 It does make a

difference whether Iraq or Belgium gets nukes, and the difference is not

a matter of discriminatory ‘‘we can’t trust them’’ snobbery on the part

of Europeans. Europe is now empty of wars among great powers as

never before in modern times, and a variety of security regimes, including

some formidable military capabilities, can be called forth from the

vasty deep to deal with troublemakers.

Outside of Europe, those architectures and institutions are less effective

or nonexistent. Since weapons don’t make war but states and leaders

do, international institutions that make it easier to conciliate grievances,

and more transparent who is cheating on military agreements, make proliferation

relatively manageable short of disaster. That having been said,

this author is a pessimist about human nature: he therefore prefers fewer

nuclear-armed states to more, and a successful conclusion to the efforts

by current nuclear powers to reduce the sizes and hair-trigger dependencies

of their arsenals.


1. Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-

Wesley, 1979). See also, and more specifically on Waltz’s views of the relationship

between nuclear weapons and stability: The Spread of Nuclear Weapons: More May

Be Better, Adelphi Papers No. 171 (London: International Institute of Strategic

Studies, 1981); ‘‘Nuclear Myths and Political Realities,’’ American Political Science

Review, No. 3 (September 1990), pp. 731–745; and his chapters in Scott D. Sagan

and Kenneth N. Waltz, The Spread of Nuclear Weapons: A Debate (New York: W. W.

Norton, 1995). Other arguments for a positive association between the spread of

survivable nuclear forces and international stability appear in Martin Van Creveld,

Nuclear Proliferation and the Future of Conflict (New York: The Free Press,


2. Realists contend that power is based on tangible resources such as population,

economic capacity and territory, and the most influential among them

Nuclear Proliferation 89

also believe that power is both a means and an end in international politics. See

Hans J. Morgenthau, Politics among Nations: The Struggle for Power and Peace (New

York: Alfred A. Knopf, 1948). Neorealists hold, as do realists, that the structure

of the international system, especially system polarity, is the most important

determinant of the context for state decision making. Neorealists, in contrast to

realists, are more likely to acknowledge sources of power other than tangible

ones, and to treat power as a means but not as an end in itself. Paul R. Viotti

and Mark V. Kauppi divide international political theories into realist, pluralist

and globalist schools, a taxonomy similar to that offered by Kalevi J. Holsti. See

Viotti and Kauppi, eds., International Relations Theory: Realism, Pluralism, Globalism

(New York: Macmillan, 1993), esp. pp. 61–227, and Holsti, Peace and War: Armed

Conflicts and International Order (Cambridge: Cambridge University Press, 1991),

p. 328. See also Holsti’s comments on the roots of realism and neorealism,

pp. 329–330. An excellent summary and critique of neorealist views is provided

by Robert O. Keohane, ‘‘Theory of World Politics: Structural Realism and Beyond,’’

in Ada W. Finifter, ed., Political Science: The State of the Discipline (Washington,

D.C.: American Political Science Association, 1983), and reprinted in

Viotti and Kauppi, eds., International Relations Theory, pp. 186–227.

3. The distinctions between assertive and delegative control are explained in

Peter Douglas Feaver, Guarding the Guardians: Civilian Control of Nuclear Weapons

in the United States (Ithaca, N.Y.: Cornell University Press, 1992), pp. 3–28. On

U.S. nuclear predelegation during the Cold War, see Bruce G. Blair, The Logic of

Accidental Nuclear War (Washington, D.C.: Brookings Institution, 1993), pp. 46–


4. For arguments about the impact of organizational and decision-making

variables on the relationship between nuclear weapons and stability, see Sagan,

in Sagan and Waltz, The Spread of Nuclear Weapons, pp. 47–91. See also Scott D.

Sagan, The Limits of Safety: Organizations, Accidents and Nuclear Weapons (Princeton,

N.J.: Princeton University Press, 1993), esp. pp. 28–44.

5. Lewis A. Dunn, Sarah A. Mullen, Gregory F. Giles, Joseph A. Yager and

James S. Tomashoff, The Next Nuclear-Weapon States? (McLean, Va.: Science Applications

International Corporation, 1991), ch. 2, p. 51.

6. The term ‘‘system’’ has many uses in international politics and in political

science. Structural-realist theories of international politics emphasize the causal

importance of system structure—numbers and types of units in the system and

the distribution of military and other capabilities among those units. Other variations

of systems theory emphasize the interactions among components of the

system, including the interdependence of the actors or units. For a concise discussion

of systemic theories of international politics, see James E. Dougherty and

Robert L. Pfaltzgraff, Jr., Contending Theories of International Politics, 4th ed. (New

York: Longman, 1997), pp. 100–134.

7. Vital interests as used here refers to interests over which states resist compromise

and for which they are willing to go to war. See Donald M. Snow,

National Security: Defense Policy in a Changed International Order (New York: St.

Martin’s Press, 1998), pp. 173–180.

8. Bruce Bueno de Mesquita and David Lalman, War and Reason: Domestic and

International Imperatives (New Haven, Conn.: Yale University Press, 1992).

9. Ibid., p. 267.

90 Nuclear Policy and Strategy in the Future

10. Ibid., pp. 267–268.

11. Kenneth N. Waltz, Man, the State and War (New York: Columbia University

Press, 1959). The first image includes human nature and individual psychological

attributes pertinent to decision making. The second image refers to state level

decisions and behaviors.

12. Bueno de Mesquita and Lalman, War and Reason, p. 269. The authors acknowledge

that, under conditions of imperfect information, states might mistakenly

stumble into war as a result of misjudgments based on inaccurate

information. But in a domestically constrained as opposed to a realist model of

strategic rationality, leaders may also ‘‘mistakenly’’ avoid war and ‘‘stumble into

negotiation or other peaceful solutions to their differences.’’

13. Robert Jervis, System Effects: Complexity in Political and Social Life (Princeton,

N.J.: Princeton University Press, 1998), pp. 92–93.

14. Waltz, Theory of International Politics, p. 80.

15. Jervis, System Effects, p. 109.

16. See Morton Kaplan, System and Process in International Politics (New York:

John Wiley and Sons, 1957) for an early and pioneering effort. International systems

theories are classified and critiqued in Jervis, System Effects, ch.3.

17. Waltz, ‘‘The Stability of a Bipolar World,’’ Daedalus, Vol. 93 (Summer 1964),

pp. 881–909, and Waltz, Theory of International Politics, pp. 170–171, cited in Jervis,

System Effects, p. 118.

18. For the distinction between general and immediate deterrence, see Patrick

M. Morgan, Deterrence: A Conceptual Approach (Beverly Hills, Calif.: Sage

Publications, 1983). Or, as Hobbes explained it, it is a precept or general rule of

reason that ‘‘every man, ought to endeavour Peace, as farre as he has hope of

obtaining it; and when he cannot obtain it, that he may seek, and use, all helps,

and advantages of Warre.’’ Thomas Hobbes, Leviathan (New York: Washington

Square Press, 1964), p. 88.

19. As Jervis explains, rationality assumptions are not necessarily falsified by

cases in which leaders have chosen poorly. But in many other instances ‘‘the

beliefs and policies are so removed from what a careful and disinterested analysis

of the situation reveals that the failure is hard to fit into the framework generated

by rationality.’’ Jervis, introduction in Jervis, Richard Ned Lebow and Janice

Gross Stein, Psychology and Deterrence (Baltimore: Johns Hopkins University

Press, 1985), p. 6.

20. See Raymond L. Garthoff, Reflections on the Cuban Missile Crisis, rev. ed.

(Washington, D.C.: Brookings Institution, 1989), pp. 6–42.

21. U.S. officials at the time of the Cuban missile crisis underestimated significantly

the size of the Soviets’ conventional forces deployed on that island (actually

some 40,000). Nor did they realize that, in addition to warheads for

medium- and intermediate-range missiles deployed in Cuba, the Soviets also

deployed nuclear warheads for tactical weapons launchers. At the time of the

crisis, U.S. leaders were uncertain whether any Soviet warheads had actually

arrived in Cuba. Raymond L. Garthoff, ‘‘The Havana Conference on the Cuban

Missile Crisis,’’ Cold War International History Project Bulletin (Washington, D.C.:

Woodrow Wilson Center, Spring 1992), pp. 1–4. According to Bruce Blair, actual

orders to the senior Soviet commander in Cuba specifically precluded the use of

Nuclear Proliferation 91

any nuclear weapons without prior approval from Moscow (Blair, The Logic of

Accidental Nuclear War, p. 109).

22. U.S. leaders were not well informed about the actual Soviet nuclear force

deployments in Cuba in October 1962. See Gen. Anatoli I. Gribkov and William

Y. Smith, Operation ANADYR: U.S. and Soviet Generals Recount the Cuban Missile

Crisis (Chicago: Edition Q Books, 1994).

23. Gaddis, ‘‘The Essential Relevance of Nuclear Weapons,’’ in The United

States and the End of the Cold War, pp. 105–132.

24. Leon V. Sigal, Disarming Strangers: Nuclear Diplomacy with North Korea

(Princeton, N.J.: Princeton University Press, 1998), esp. pp. 3–14.

25. See Mazarr, North Korea and the Bomb, passim.

26. Office of the Secretary of Defense, Proliferation: Threat and Response (Washington,

D.C.: U.S. Government Printing Office, 1996), pp. 6–7 provides background

on the North Korean nuclear program.

27. Jane’s Strategic Weapons Systems, September 30, 1998. http://janes. ...h_janes/.

28. Ibid.

29. Ibid., p. 9.

30. Commission to Assess the Ballistic Missile Threat to the United States

(Rumsfeld Commission), Report (Executive Summary) (Washington, D.C., July 15,

1998), p. 9. Pagination may be inexact due to variations in electronic transmission.

31. Ibid.

32. See map, Ranges of Current and Future Ballistic Missile Systems (Iran), in

Office of the Secretary of Defense, Proliferation: Threat and Response, p. 17.

33. Ibid., p. 16.

34. Thomas A. Keaney and Eliot A. Cohen, Revolution in Warfare? Air Power in

the Persian Gulf (Annapolis, Md.: Naval Institute Press, 1995), p. 67. This is a

revised version of the official U.S. Air Force Gulf War Air Power Survey (GWAPS)

first published in 1993.

35. Keaney and Cohen, Revolution in Warfare, p. 67.

36. Ibid.

37. Ibid., p. 72.

38. Ibid., p. 75.

39. Ibid., p. 78. Special forces teams may have destroyed some mobile Scuds.

40. Jeffrey Record, Hollow Victory: A Contrary View of the Gulf War (Washington,

D.C.: Brassey’s, 1993), pp. 71–73.

41. Secretary of Defense William J. Perry, On Ballistic Missile Defense: Excerpt

from a Speech to the Chicago Council on Foreign Relations, March 8, 1995, p. 1

(mimeo), cited in Keith B. Payne, Deterrence in the Second Nuclear Age (Lexington:

University Press of Kentucky, 1996), p. 58.

42. Payne, Deterrence in the Second Nuclear Age, pp. 57–58.

43. Robert E. Harkavy, ‘‘Triangular or Indirect Deterrence/Compellence:

Something New in Deterrence Theory,’’ Comparative Strategy, No. 1 (1998),

pp. 63–82.

44. Ibid., p. 74.

45. Rumsfeld Commission, Report (Executive Summary), p. 10.

46. Ibid, p. 11.

47. Ibid.

92 Nuclear Policy and Strategy in the Future

48. Ibid., p. 22. The Rumsfeld Commission cannot be dismissed by skeptics as

an alarmist group. Its membership included, in addition to former Secretary of

Defense Donald Rumsfeld, who chaired it, noted experts on nuclear technology,

strategy and policy representing a variety of policy views and professional backgrounds.

49. This point has been argued effectively by Mearsheimer and Waltz: see

previous citations.

Chapter 5

Russia and Nuclear Weapons After

the Cold War: A Potemkin Village?

Russia’s nuclear arsenal is seen by some Russians, and by some American

experts in arms control and security policy, as a continuation of the

Soviet one from the Cold War era. Policy makers and expert analysts

continue to talk within the old frame of reference dominated by calculations

of relative force size and of expected damage inflicted under various

wartime scenarios. Russia’s military cling to nuclear weapons as

talismans of past imperial glory, and U.S. officials negotiate with Russia

over force reductions within a framework that assigns priority to maintaining

stable deterrence and nuclear parity. It is, by and large, a Potemkin

village in a fictive world.

Russia’s security problem is not maintaining ‘‘deterrence’’ against any

presumed encroachment from enemies, something easily accomplished

with far fewer nuclear weapons than it has. The real problem is that

Russia’s economy, society and military are melting down. This meltdown

brings with it the problem that Russia may be unable to rebuild credible

conventional military forces as the backbone of future fighting power.

At best, a twenty-first century Russian nuclear force, modernized and

properly controlled, would be a possible deterrent against nuclear intimidation

or invasion of Russian territory.

This chapter first considers Russia’s nuclear posture and its possible

short-term futures from the traditional standpoint: balances of power,

START-driven arms control reductions, and stable deterrence, including

the avoidance of accidental or inadvertent war. We then discuss the long

view that confronts Russia’s political and military leadership, of an information

driven ‘‘Revolution in Military Affairs’’ creating new generations

of non-nuclear weapons and control systems. Russia must consider

94 Nuclear Policy and Strategy in the Future

how much money it can afford to pour down the ratholes of weapons

of mass destruction when future great power status will be defined by

smart weapons and smarter control, reconnaissance and assessment systems.


Force Structures

The United States has agreed with Russia that Russia should accept

the strategic arms control obligations of the former Soviet government,

undertaken in the START I and II agreements signed in 1991 and 1993,

respectively. The second agreement called for the two sides to reduce

their holdings of strategic nuclear weapons to ceilings in the range of

3,000 to 3,500 warheads, with additional limitations on launchers, especially

MIRVed ICBMs (land-based missiles with multiple, independently

targeted warheads). As late as June 1998 there was considerable controversy

in Russia whether START II would be put on hold, or even, according

to some pessimists in the Duma, put into mothballs.

The reasons given by critics were both strategic and economic. The

strategic issue was that START II was a bad deal favorable to the United

States Russia was giving up its MIRVed ICBMs, playing into the U.S.

comparative strength in bomber and submarine-launched weapons. The

economic concern was that Russia might not be able to afford the expensive

ICBM dismantlement and SLBM force buildup that a shift in

dependency from land- to sea-based missiles and/or bombers would

entail. From an economic standpoint, Russia would be better off under

a START III regime that required less conversion and modernization

compared to START II.

The United States and Russia concluded several agreements in 1997

with the objective of firming up START II and increasing the probability

of its successful ratification in Russia. First, Washington and Moscow

agreed to delay final implementation of the treaty-required reductions

until December 31, 2007, instead of January 1, 2003. Related to this step,

they also committed themselves to prompt negotiations on a follow-up

START III agreement that would reduce each side’s strategic nuclear

warheads to 2,000–2,500 by 2007. 1

Another reassurance for Russia was provided in bilateral agreements

with regard to U.S. deployment of highly capable Theater Missile Defense

(TMD) systems. Both agreed to ban testing of TMD systems against

ballistic missile targets with speeds above 5 km/sec. or ranges in excess

of 3,500 kilometers. The United States and Russia also agreed not to

develop, test or deploy space-based TMD interceptors and will exchange

information on theater missile defense plans and programs. 2 A third

Russia and Nuclear Weapons After the Cold War 95

agreement thought useful in expediting Russian ratification of START II

was the NATO-Russian Founding Act creating a Permanent Joint Council

as a consultative forum for security issues of mutual interest. The

Founding Act and Permanent Joint Council helped to assuage Russian

concerns about the 1997 decisions taken on NATO enlargement to include

the Czech Republic, Hungary and Poland by 1999. 3

Not all has gone favorably since the NATO-Russian Founding Act for

U.S.-Russian political relations, nor for the arms control incentives that

derive from those relationships. First, the U.S. Congress in 1998 and 1999

accelerated its demands for early deployment of a National Missile Defense

(NMD) system, and some of the congressional initiatives would

require radical revision or U.S. abrogation of the ABM Treaty. Second,

the Clinton administration modified its position on BMD in January 1999

and moved perceptibly toward congressional Republican demands for

an NMD system designed to protect against accidental launches or small

attacks from rogue states. 4 Third, the NATO military operation against

Yugoslavia beginning March 24, 1999, and its timing so near the Washington

summit of April 1999 marking the addition of three new member

states, allowed nationalists and communists in the Russian Parliament

to stoke the fires of anti-Western imperialism and to postpone ratification

of START II once again.

Table 5.1 summarizes the status of U.S. and Russian strategic nuclear

forces at the beginning of 1999.

If the preceding were some of Russia’s arms control discontents near

century’s end, a serious U.S. concern about cooperative threat reduction

was the status of the Russian program for the custodial management and

security of its nuclear weapons and fissile materials. Russians have expressed

confidence that the 20,000 or so nuclear weapons (i.e., warheads)

are appropriately guarded by the 12th Main Directorate of the Russian

Ministry of Defense. However, on at least one occasion, several officers

of the 12th Main Directorate took part in an armed rebellion that included

several murders, an attempted aircraft hijacking, and the taking

of hostages. 5

Even if the warheads are secure, the stockpile of fissile material is

another matter. Some 650 tons of weapons grade material are spread

among several hundred buildings in the former Soviet Union. Security

at most of these installations is deplorable by current U.S. or former

Soviet standards. Russia has inherited the old Soviet system of safeguards

that was designed to keep foreign spies out of installations, not

to monitor the behavior of employees or to measure accurately the inventories

of weapons grade material. In addition, financial crises in Russia

have made destitute many of the security personnel guarding these

institutions. Insecurity is compounded by excessive secrecy: U.S. estimates

of Russia’s nuclear warheads have an uncertainty range of plus

Table 5.1

U.S. and Russian Strategic Nuclear Forces (MOU of January 1999)

Source: Arms Control Association, April 1999, provided by request and based on January

1999 Memorandum of Understanding, U.S. State Department.

Russia and Nuclear Weapons After the Cold War 97

or minus 5,000; and, although the United States has verified the actual

numbers of Russian missiles destroyed persuant to arms agreements, it

has not been able to verify the actual destruction of a single Russian

nuclear warhead. 6

The United States has responded to these concerns with programs under

the rubric of ‘‘cooperative threat reduction’’ (subsequently ‘‘Expanded

Threat Reduction’’) to prevent theft or diversion of materials, to

provide alternate and more stable employment for nuclear workers, to

improve the quality and transparency of information about stockpiles,

and to reduce the currently excessive stockpiles of fissile material. Some

progress has been made on improving security at facilities (via alarms

and detectors, for example), but a system of reliable inventory is still far

away. The United States has an agreement to purchase 500 tons of highly

enriched uranium (HEU) from Russia in order to create incentives for

the destruction of warheads and of excessive fissile materials. But many

additional tons of additional HEU remain outside the reach of this agreement,

and little progress has been made toward any deal to purchase

Russia’s excess plutonium. With regard to progress on transparency, the

picture is mixed. U.S. and Russian scientists have been cooperating to

develop technology for improving inventories of warheads and materials

and for warhead dismantlement. Russian security services have not always

been user friendly to these efforts, and their future attitudes are

unpredictable. 7

This resume of cooperative threat reduction and denuclearization is

given here because it is related to the political climate that makes possible,

or precludes, progress on strategic arms reduction. Cooperative

threat reduction is also germane to interstate arms control because of

Russia’s economy. If Russia’s economic slide of the latter 1990s is not

slowed down and its nuclear materials and weapons complex is spun

off into bureaucratic never-never land, Russia will, regardless of the intentions

of its government, become a major proliferator by dint of undetected

smuggling, theft and diversion of its fissile materials. Will its

warheads be far behind?

Maintaining Parity and Stability

Can Russia maintain essential equivalence or strategic nuclear parity

with U.S. forces while undergoing denuclearization? As a first step in

this analysis we will consider the outcomes of nuclear exchanges between

START I, START II and START III–compliant U.S. and Russian

strategic nuclear forces, making plausible assumptions about operational

parameters for each stage. The model used here permits comparison of

the surviving forces available to each side after each has launched a

single strike (first or retaliatory strike). We also compare the surviving

98 Nuclear Policy and Strategy in the Future

and deliverable warheads available to each party under conditions of

generated alert (forces poised to attack or retaliate) or day-to-day alert

(forces at peacetime or normal alertness). For comparative purposes, the

numbers of surviving and deliverable warheads for the United States

and for the Soviet Union in 1983, when strategic nuclear forces were

much larger and more threatening than now, are also shown in Table 5.2.

The results shown in Table 5.2 indicate that both the United States and

Russia retain a large number of retaliatory warheads after absorbing a

first strike even with forces reduced to START-III levels. Numbers of

surviving and deliverable warheads provide some guaranty of assured

retaliation, but they do not indicate how stable the various force structures

might be. The stability of a nuclear strategic balance is influenced

by, in addition to the numbers of surviving and deliverable warheads

for each state, their relational properties. First strike stability is one example

of a relational property.

We will take two perspectives on first strike stability. 8 First, we calculate

a simple index of first strike stability for each force posture (1983,

and START I–III) and, within each force posture, for each of four conditions:

(1) both sides have generated forces, (2) only Russian forces are

on generated alert, (3) only U.S. forces have been generated, and (4) both

forces are on day-to-day alert. The stability index for each of sixteen

posture/conditions is provided in Table 5.3. The index is calculated by

computing the difference between each side’s first strike warheads and

its surviving warheads in each posture/condition. For each of the sixteen

possibilities, the Russian ‘‘difference’’ is divided by the U.S. ‘‘difference.’’

The most stable possibility is a ratio of Russian to U.S. differences of

unity 1.0. As the ratio rises or falls from 1.0, instability of the relationship

between the two forces gets higher.

The first strike stability index measures how similar or different are

the ‘‘first strike advantages’’ of two forces. The closer the index is to 1.0,

the more similar are their first strike advantages. This approach differs

from other attempts to measure stability as inherent in the number of

survivors per se. This index is an admittedly simplistic measure of first

strike stability, but it shows that there is no necessary relationship between

larger and more diverse forces, on the one hand, and a more stable

(i.e., less first strike prone) U.S.-Russian nuclear-strategic relationship.

The various START forces are all more stable than the 1983 forces by

this measure, although the 1983 forces offered the two sides a virtual

feast of missile types (for the then Soviet forces) and bomber types (for

the United States).

A related question is whether stability could be maintained if the

United States and Russia not only reduced force sizes, but also realigned

their forces to require fewer kinds of delivery systems. As an example,

the United States might eliminate its land-based missiles and rely on

submarine-launched missiles and air-delivered weapons. Russia might

Table 5.2

Surviving and Deliverable Retaliatory Warheads: START I–START III and

1983 Baseline

Gen. Generated Alert; Day Day-to-Day Alert.

Source: Author. Methodology adapted from statistical model first developed at U.S. Naval

Postgraduate School by Dr. James John Tritten. Dr. Tritten is not responsible for its

application here. The model is an abstract, statistical model of force attrition and thus

does not capture the finer nuances of targeting and the resulting variations in outcomes

across categories of alertness that would otherwise result.

100 Nuclear Policy and Strategy in the Future

Table 5.3

First Strike Stability Indices for Various Force Structures

Source: Author.

eliminate its intercontinental bomber force and depend upon its landbased

missiles and submarines. How would such ‘‘dyads’’ or two-leg

forces compare with three-sided, or ‘‘triads?’’ I attempted to simulate

this impact for START I, START II and START III force structures as

previously given. These U.S. and Russian forces were modified into dyads

by using the same ceilings on total number of warheads as in the

case of three-sided forces, but allowing freedom to mix different arms of

Russia and Nuclear Weapons After the Cold War 101

Table 5.4

Russian and U.S. Outcomes in Various START Environments: Dyads

Compared to Triads

Source: Author. Forces for dyads are listed in the Appendix at the end of this chapter.

service within each dyad. Thus, for example, the United States might

prefer to eliminate its Minuteman III launchers from its START II force

and to make up the 500 warheads by adding to its submarine-launched

missile warheads. On the other hand, Russia might prefer in START II

to eliminate its bomber warheads and to substitute additional warheads

on land-based missiles. Russia has relatively more confidence and more

investment in its land-based missiles compared to its sea-based force; the

United States favors sea-based warheads to modernizing or augmenting

its land-based missiles.

Following through on these assumptions and using the model employed

previously, we can calculate the numbers of surviving and deliverable

warheads for each state under the limitations of each START

environment, comparing triads with dyads and resulting in Table 5.4.

As above, the model used is an abstract, statistical model of force attrition

and thus does not capture the finer nuances of targeting and the

resulting variations in outcomes across categories of alertness that would

otherwise result.

102 Nuclear Policy and Strategy in the Future

The results show that neither side would be worse off under a dyad,

and in many cases both sides would retain more, instead of fewer, surviving

and deliverable warheads. The findings result from the fact that,

in shifting to a dyad, each side (in this simulation) chose to eliminate its

most vulnerable component: air forces for Russia and land-based missiles

for the United States. The result was more efficient, but smaller, forces.

Even under START III overall limitations on force loadings, both states

retain 600–700 survivable and deliverable warheads.

Another perspective on stability is to ask what might happen if either

of the two sides deployed missile defenses. This is no longer a hypothetical

issue. U.S. secretary of defense William Cohen announced in

January 1999 an adjustment in the ‘‘three plus three’’ program that all

but committed the United States to the eventual deployment of an NMD

system against rogue nation attacks. According to Cohen, the United

States would commit $6.6 billion to a ‘‘three plus five’’ program that

would produce a system ready for deployment by the year 20005. 9 A

final decision on deployment of any U.S. NMD system will be made in

the year 2000, permitting additional technology development and testing

of proposed system components. As envisioned by DOD and BMDO

(Ballistic Missile Defense Organization) these components would be

space-based detectors for missile launch, long-range radars to track missile

flight paths, other radars for intercept tracking, and non-nuclear kill

interceptors. 10

The relationship between defenses and national security objectives in

different technology environments is far from obvious. The distinction

between the objectives of ‘‘central deterrence’’ and ‘‘first strike stability’’

is important. As explained by Glenn A. Kent and David E. Thaler, central

deterrence is concerned with the prevention of a deliberate first strike

by an adversary who has concluded that the ‘‘benefits’’ from a nuclear

attack outweigh the benefits of peace. 11 With the kinds of forces deployed

by the Americans and Soviets during most of the Cold War, central deterrence

was easily accomplished. First strike stability, on the other hand,

deals with the dilemma of leaders who, in a crisis, fear preemption by

the other side. Leaders in this situation may strike first due to their fear

of incurring a first strike and being forced to retaliate after having suffered

devastating losses. 12

The distinction between central deterrence and first strike stability as

security objectives can be related to three possible strategic environments:

(1) an offense-dominant environment, in which offensive technology

is more advanced than defensive and in which major powers

deploy large offensive forces, (2) a defense-dominant environment, in

which highly capable defensive technologies are superior to the state of

the art in offenses, and offenses are therefore reduced to small forces,

and (3) an offense-defense competitive environment, in which advanced

Russia and Nuclear Weapons After the Cold War 103

Table 5.5

First Strike Stability and Central Deterrence in Three Environments

Source: Adapted from Glenn A. Kent and David E. Thaler, First Strike Stability and Strategic

Defenses (Santa Monica, Calif.: RAND Corporation, 1990), p. 4. I have reworded some

cell entries in keeping with my purposes but not changed the thrust of the Kent-Thaler


technologies push forward in both offense and defense, leading to large

force structures in both categories for anyone who wants to remain in

the major leagues. 13 The relationship between the kind of strategic environment

and the quality of deterrence or first strike stability found in

each environment may be described in tabular form, as in Table 5.5.

How much difference would defenses make in the stability of the U.S.-

Russian strategic nuclear relationship in transition to a partnership? During

the Cold War, the prospect of a transition from deterrence, based

exclusively on offensive retaliation, to a mixed-force structure employing

both offenses and defenses, was held back by mutual suspicion in Washington

and Moscow. Transition to a mixed-deterrent force was also inhibited

by the primitive state of defense technology compared to offense.

Improved technologies and better U.S.-Russian political relations now

reopen the question whether defenses mixed with offenses would improve

stability, under the assumption of mutually agreed and deployed


Russia remains warily skeptical that any U.S. missile defense deployment

could be consistent with stable deterrence. The Supreme Commander

of the Russian Strategic Missile Forces (RVSN), Colonel-General

Vladimir Yakovlev, called in January 1999 for a global ‘‘strategic stability

treaty’’ that would include, in addition to the United States and Russia,

Britain, France and China. 14 According to Yakovlev, such an agreement

would include reductions in U.S and Soviet strategic nuclear warheads

104 Nuclear Policy and Strategy in the Future

even to START III levels and agreement between the two states on ‘‘the

inviolability of space.’’ 15 He specified, in regard to space arms control,

the need for a pledge not to create space vehicles capable of attacking

warning systems to detect missile attacks. Additional Russian skepticism

about a U.S. limited national defense system was voiced by Ministry of

Defense official Colonel-General Igor Valynkin, who contended in early

February 1999 that a U.S. revision of the ABM Treaty to permit missile

defenses would upset stability and that Russia would ‘‘undoubtedly respond.’’


Defense, even granted the assumption of technologies better than Cold

War defenses, is still difficult to do with high assurance of effectiveness.

Space-based defense interceptors are prohibited by the ABM Treaty that

remains in force; the same agreement also limits the extensiveness of

ground-based deployments to selected sites, precluding national territorial

coverage. The military tasking of defenses under any revised

U.S.-Russian arms control regime will thus be restricted to accidental

launches or limited attacks from rogue states armed with ballistic missiles.

Even against attacks of modest sizes by Cold War standards, defenses

that are very good (i.e., allow very little ‘‘leakage’’ of attacking

warheads through the system) will not preclude historically unprecedented

levels of societal damage. Consider, in this regard, the figures in

Table 5.6. This table shows the numbers of surviving and deliverable

warheads that will still ‘‘leak’’ through a defense that is 90 percent effective

against the retaliatory strike of the other side.

Crisis Stability and Targeting

The previous section sketched some measures of stability with and

without defenses. The reader has already noticed that for each force posture,

we varied our assumptions about the readiness status of each side’s

retaliatory forces. An additional dynamic is introduced by uncertainty

about the firing doctrines resorted to by U.S. or Russian leaders during

a crisis. The assumption of retaliation after ride out, U.S. declaratory

policy and the operational mode preferred by many arms control experts,

may be too optimistic.

The Cold War model of first strike surprise and second strike retaliation

after riding out an attack was apparently based on incorrect academic

assessments of the operational configuration of the U.S. and Soviet command

and control systems. According to Bruce G. Blair’s analysis of U.S.

and Soviet Cold War nuclear operations, each side’s war planners gradually

moved toward a de facto ‘‘launch on warning’’ posture in order to

protect vital command and control elements and vulnerable forces. 17 Although

U.S. declaratory policy from the 1960s until the end of the Cold

War insisted upon ‘‘second strike ride out’’ as the criterion for force

Russia and Nuclear Weapons After the Cold War 105

Table 5.6

90 Percent Defense Effectiveness against Russian, U.S. Retaliation (Cell

Entries Surviving and Deliverable Warheads Leaking Through Defense)

Source: Author.

sizing, nuclear targeting and warning functions were programmed in the

expectation that LOW was the preferred mode of response to surprise

first strike. 18 According to Blair:

Retaliation after ride-out was an abstraction remote from operational predilections.

From that standpoint it qualified as one of the great myths of the nuclear

106 Nuclear Policy and Strategy in the Future

Table 5.7

Target Coverage for U.S. and Russian Warhead Inventories, 1991 and 2003

Total Wpns. Total strategic nuclear warheads available.

Gen. RO Surviving and arriving warheads, generated alert.

Day RO Surviving and arriving warheads, day-to-day alert.

Nuc Targs. nuclear forces related target set.

OMT Targs. other military targets.

C31 Targs. command, control, communications and intelligence targets.

U/I Targs. urban/industrial targets.

Source: Adapted from Richard Mesic, Roger Molander and Peter A. Wilson, Strategic Futures:

Evolving Missions for Traditional Strategic Delivery Vehicles (Santa Monica, Calif.:

RAND Corporation, 1995), p. 19. I have omitted explanatory notes provided by the


age. And though launch on warning did complicate Soviet attack planning, its

role in the posture was central—the command and control system’s preferred

and best prepared mode of wartime operation. 19

For U.S. and Russian military planners, the significance of reliance upon

prompt launch relates not only to survivability, but also to the spectrum

of targets that surviving and retaliating forces can be expected to attack.

Discussions of strategic nuclear targeting generally have assumed that

four classes of targets are important to planners: (1) the other side’s nuclear

forces, especially those on strategic launchers and capable of

prompt response, (2) so-called Other Military Targets (OMT), or conventional

forces and supporting infrastructure, (3) command, control, communications

and intelligence (C3I) targets, and military-related–

urban-industrial (U/I) targets. As force levels are reduced, the numbers

of targets that U.S. or Russian forces can hold at risk in a first or retaliatory

strike must also diminish. This diminished target coverage is offset,

of course, by the smaller numbers of forces that might need to be targeted

after arms control agreements have taken hold. One study, shown

in Table 5.7, contrasted U.S. and Soviet/Russian 1991 surviving and ar-

Russia and Nuclear Weapons After the Cold War 107

riving warheads, relative to the four targets sets listed above, with the

anticipated year 2003 capabilities of each state’s forces.

Too many discussions of nuclear targeting have treated the subject as

a mathematical exercise of considerable abstraction. But targeting is

planned for, and carried out by, reactive military organizations. We can

agree that Cold War nuclear operations showed the following assumptions

to be reasonable: (1) both sides’ command and control systems, like

all bureaucratic organizations, would have survival of the organization

itself at or near the top of their value hierarchies, (2) each side’s command/control

and communications system for nuclear warning and response

would be more vulnerable than its strategic retaliatory forces

would be, and (3) an attacker would not have to destroy most of the

physical infrastructure of the C3 system in order to disable it sufficiently

to prevent a coordinated retaliation based on highly complicated prewar plans.

In other words, a command/control system cut up into pieces could still

provide for some retaliation, but not according to any coherent plans

conjured up in the comparative luxury of peacetime. 20

U.S. officials contend that Clinton revisions in nuclear doctrine include

explicit assertions that ‘‘launch on warning’’ is neither declaratory nor

action policy. Referring to the Presidential Decision Directive (PDD) of

November 1997 on U.S. nuclear strategy and doctrine, Robert Bell, senior

director for defense policy and arms control at the National Security

Council, noted that:

in this PDD we direct our military forces to continue to posture themselves in

such a way as to not rely on launch on warning—to be able to absorb a nuclear

strike and still have enough force surviving to constitute credible deterrence. 21

In September 1998 Presidents Clinton and Yeltsin agreed to create a

permanent, joint early warning center to be located in Russia. The proposed

center would make possible a reciprocal flow of early warning

data from each side’s warning sensors to the national command center

of the other side. 22 The permanent center would be staffed by Americans

and Russians. Related to this, the United States and Russia would push

within the larger international community for agreements on prelaunch

notification of any long-range missile tests and space launches. On a

separate track, U.S. defense officials announced in February 1999 that

they would invite a Russian delegation to a specially designed facility in

Colorado Springs (near Space Command) from mid-December to mid-

January. The expert Russian delegation would take part in monitoring

U.S. warning systems and data displays during this crossover to the

millennium. 23 These confidence building and transparency measures, as

they are known to arms control experts, would be designed to reassure

both U.S. and Russian leaders that no glitch in the automated command

108 Nuclear Policy and Strategy in the Future

and control systems for alerting nuclear forces would result in an accidental


Transparency between military establishments can contribute to the

clarification of misunderstandings during tense moments of potential or

actual confrontation. For example, after several weeks of NATO air attack

against Yugoslavia in the spring of 1999, Russian President Boris

Yeltsin issued a sharp warning that Russia could not disinterest itself

from the situation on the ground. Referring April 9 to the possibility of

a NATO ground offensive into Yugoslavia, Yeltsin stated that ‘‘we cannot

permit that’’ although he failed to specify what Russia might do in

response. 24 On the same day, Speaker of the Russian Parliament Gennady

Seleznyov told reporters that President Yeltsin had ordered Russian missiles

to be ‘‘pointed at countries which are fighting against Yugoslavia.’’ 25

The reference was to the 1997 NATO-Russian Founding Act that established

a consultative framework for future relations between NATO and

Russia. One of its spillovers was a public announcement that in the future

neither NATO nor Russian strategic nuclear missiles would be targeted

against one another. This announcement was more symbolism

than substance, and so was Yeltsin’s attributed threat to retarget NATO.

Modern Russian strategic nuclear missiles can be retargeted electronically

and remotely within minutes of an order to do so.

The post–Cold War Russian nuclear command system has not fully

worked itself out. The ‘‘aurora borealis’’ incident in 1995, in which Russian

warning systems temporarily mistook a Norwegian scientific rocket

for a U.S. SLBM launch, suggests that the gaps in Russia’s warning and

response systems require some temporary fixes in systems and procedures,

as well as permanent replacement for outdated and deteriorating

equipment. Even then, the problem of interactive complexity between

people and systems remains. Consider, as an illustration, a hypothetical

military command and control system with one central command post,

ten major regional commands, a linked communications system among

the various command posts, a direct connection between each command

post and a space satellite warning (launch detection) system, and a computer/information

system with distributed data for all commands. Each

of these elements can be abbreviated as follows:

Centcom Central Command

Regcom Regional Command

Netcom Communications System

Alarm Satellite Warning System

Cyber Computer/Information System

We can use a simple matrix to summarize some of the possible relationships

among these various components of the national military com-

Russia and Nuclear Weapons After the Cold War 109

mand decision system. Numbers in the matrix indicate whether any

individual point in the system is properly ‘‘connected’’ to another. Being

‘‘connected’’ implies accurate receipt of, and transmission of, messages

back and forth, and correct comprehension of those messages not distorted

by enemy or friendly fog. Suppose the matrix below:

At the simplest level, we might use 1s and 0s to represent a state of

adequate or inadequate connectness, as defined above. The matrix above

might then appear, at some future and hypothetical time, as shown below:

In this example, Centcom has lost contact with or cannot correctly interpret

Regcom, remains connected to Netcom and Cyber, but is disconnected

from Alarm. Now imagine how much more complicated this

‘‘status of comprehension’’ matrix might be if, instead of 1s and 0s, we

substituted a scale for degrees of connectedness in each cell of the matrix,

say, from 1 through 10. As an additional complication, we might relax

the assumption that the various nodes in the system would know accurately

their degree of connectedness or isolation from one another. The

result would be a matrix of enormous complexity despite the deliberately

simplistic example on which it is based. Of course, the Russian as well

as the U.S. decision-making system for nuclear alert would require a

much more complicated model for its representation. 26

The point of this demonstration is this: interactive complexity between

people and warning/communications/control systems ensures that even

in normal times, parts of the system are never in complete synchroni-

110 Nuclear Policy and Strategy in the Future

zation and operators as well as policy makers may be blissfully unaware

of their lack of connectedness until crisis ensues. Standard operating procedures

may have evolved which allow each component to function efficiently

in peacetime despite what is happening in other parts of the

system (in the same way that we all develop ‘‘workarounds’’ for anticipated

shortfalls of our colleagues in the workplace or for roadblocks

thrown up by administrators and managers to prevent lateral collaboration

among knowledge workers). SOPs that are expedient for carrying

on day-to-day business, despite the interruptions of outside agencies and

bureaus, may be dysfunctional when higher levels of alert are suddenly

called for and channels of communications and control have to be improvised.

Presidents Clinton and Yeltsin agreed in the spring of 1997 to a notional

START III target of 2,000 warheads on each side, continuing the

START II assumption of de-MIRVed ICBMs and with additional downloading

of U.S. submarine-launched ballistic missiles compared to

START II. Some in the United States favor a realignment of strategic force

structure to a dyad (of submarine-launched missiles and air-delivered

weapons) if a START III agreement is ratified, but most U.S. defense

planners favor the diversity inherent in a triad. Compared to START II,

START III would save Russia considerable amounts of money in maintenance

and modernization. Since START III appears as a win-win proposition

for both sides and has no obviously negative effect on stability,

no one should be surprised if a START III agreement is formally concluded

in 1999.


Revolution or Evolution?

Both Russian and U.S. defense experts now agree that we are on the

threshold of a Military-Technological Revolution or a Revolution in Military

Affairs that will eventually change strategy, doctrine and organization

in the next century. 27 The Russian and U.S. paths for arriving at

this destination, better defined as an expectation, make an interesting

contrast. The U.S. Department of Defense took a new look at American

history of the 1920s and 1930s and discovered a period of astonishing

innovation in thinking about airpower, sea control, amphibious warfare

and so on. 28 Although the resources were not yet available to bring these

concepts to fruition, World War II offered fertile ground for the realization

of these ideas in the wars against Germany and Japan. More recently,

the United States in the latter 1970s began development of a

number of technologies that fortuitously came together in the latter 1980s

and were given their baptism of fire against Iraq in 1991. Among these

Russia and Nuclear Weapons After the Cold War 111

technologies were improvements in seeing, aiming and intelligence mapping

of a deep battlefield.

For the Soviets, their theorists in the 1920s and 1930s, primarily V. K.

Triandafillov and M. N. Tukhachevskiy, anticipated offensively oriented,

combined-arms warfare as it was later fought in World War II and explicitly

referred to the concept of ‘‘deep operation.’’ 29 For example, Triandafillov’s

1929 study of The Character of Operations of Modern Armies has

an interesting and far-sighted discussion of ‘‘shock armies,’’ selfsustaining

for operations of great depth and with decisive goals, that

prefigured the attributes of World War II Soviet battles fought simultaneously

over several fronts, including Visla-Oder, Belorussia and Berlin.

30 And a more recent contribution to post–Cold War thinking about

military-technological revolution, noted earlier, was the work of Soviet

military writers in the 1980s explicitly anticipating Western developments

in reconnaissance-strike warfare.

On both sides of the aisle, before and after the Cold War, there was

some expectation that the military-technological revolution (MTR,

hereafter) will have implications for the role of U.S. and Russian nuclear

weapons. 31 For the United States, some already tout Desert Storm as a

first-generation, conventional strategic air-land operation growing out of

U.S. Army AirLand Battle and NATO Follow-on Forces Attack (FOFA)

concepts designed for war in Europe. There is a clear sense that Soviet

1980s-vintage appreciation of the MTR owed much to their scrutiny of

U.S. and NATO concepts and also to their awareness that the United

States and its allies could afford the modernization to bring these concepts

into touch with force structure.

Russian military experts were particularly interested in Desert Storm

because Soviet military writers had foreseen part of the new information

war so clearly in the 1980s. Marshal N. V. Ogarkov, then chief of the

Soviet General Staff, noted in the mid-1980s the high reliance of NATO

‘‘air-land operations (engagements) for a theater wide, combined-arms

campaign making wide use of high precision weapons and reconnaissance-strike

complexes. 32 V. G. Reznichenko’s 1987 edition of Taktika

(tactics) explained in some detail the future of ‘‘reconnaissance-strike’’

and ‘‘reconnaissance-fire’’ complexes of very accurate, long-range weapons

with automated command and control systems supported by improved,

real-time reconnaissance. 33 If imitation is the greatest form of

flattery, it is noteworthy that a former U.S. secretary of defense used the

term ‘‘reconnaissance-strike’’ to describe the role the United States

should play in the future of multinational peace operations. 34

The Russians of the 1990s and hereafter are looking upward at MTR

from the vantage point of a stripped-down and demoralized military

profession and a financially strapped economy. 35 Having recognized the

growing significance of MTR in their theoretical writing, the Russian

112 Nuclear Policy and Strategy in the Future

General Staff must be pained to realize how far removed from their

eventual grasp such a force must be. 36 The Russians are arguably

stretched to the limit to conduct World War II style operations against

Chechens. A Russian military modernized along the lines of MTR would

require a rethinking of basic defense concepts and a realignment of the

relationship between military-industry and the armed forces.

MTR also implies, if the Russians choose to compete in this direction,

a smaller, high-tech conventional force based on voluntary recruitment

in addition to, or in lieu of, conscription. One implication of equipping

a force with smarter technology is that the people who make use of it

must be smarter as well. The mute conscript of the Cold War Soviet

armed forces who might not even have spoken Russian provides an obviously

incompetent model for the future Russian soldier operating

within an MTR force. Nor is this all. In any transition from current to

MTR-type forces, the nature of command itself will undergo considerable


Russian imperial forces and their Soviet successors operated within a

tradition of top-down, rear-forward command that will be hard put to

keep pace with the fast-moving and information-dominated warfare of

the next century. Although the Russians’ forward-looking thinkers recognize

this point in the abstract, there is more to accomplish than the

adjustment of military operations and training rituals. There is a cultural

barrier that must be broken through to get from where the Russians are

now to a system more like the venerable German tradition of Auftragstaktik,

loosely translated as mission orders. A system of that type allows

considerable latitude to lower commanders to define the ‘‘how’’ of their

mission after they have been given the ‘‘what.’’

Assuming that new generations can break cultural barriers against

mission orders and that resources can somehow be obtained for modernization,

a high-tech Russian conventional force may permit less reliance

on nuclear deterrence or on actual nuclear use prompted by

conventional force deficiencies. This would be welcome news to arms

controllers and disarmers, but not necessarily to the U.S. military. The

U.S. Department of Defense realized only after Desert Storm how far

ahead the United States was in the technologies and tactics of

reconnaissance-strike warfare. The United States reigned in the 1990s as

a singular global superpower in its capacity to wage large scale, multitheater

conventional warfare, even after expected reductions in defense

spending through 1997. Russia was forced to sit by as NATO, led by

U.S. long-range air power, ravaged Yugoslavia’s military and command

assets for more than a month in the spring of 1999 with virtual impunity.

NATO’S ability to have its way in reconnaissance-strike warfare is evident

from the Defense News Service press summary of one night’s activity

in March, following.

Russia and Nuclear Weapons After the Cold War 113


L@DTIC.MIL From: Press Service afisnews_sender@DTIC.MIL Subject:Allied

Force Aircraft Continue Bombing Campaign


By Jim Garamone

American Forces Press Service

WASHINGTON—Most NATO air attacks on Yugoslavia hit the country’s air

defense network, but some 20 percent struck army and special police units in

Kosovo, Pentagon spokesman Ken Bacon said March 25.

The air and sea barrage against Yugoslavia was in its second day and, Bacon

told reporters, it would continue as long as needed.

He summarized first-day NATO strikes: U.S. B-52 bombers began the operation

with air-launched cruise missiles, and U.S. and British ships and submarines

in the Adriatic Sea followed with dozens of Tomahawk cruise missiles. NATO

pilots flew 156 sorties against ‘‘about three dozen’’ targets. The U.S. B-2 Spirit

stealth bomber flew in combat for the first time.

NATO spokesman Jamie Shea said March 26 that the second night’s attacks

continued to ‘‘meet essential mission objectives.’’ He said 64 aircraft flew 400

sorties against 50 targets. All allied aircraft returned safely both nights; NATO

officials released photos and gun camera footage showing first-night attacks.

News reports Mar. 25 stated the Serbs had stepped up their campaign against

the ethnic Albanians in Kosovo. Bacon was unable to confirm the Serbs’ activities,

but called them in character. ‘‘If our military strikes continue, they will focus

more and more on achieving our primary goal, which is to reduce the ability of

the Yugoslav forces to target or repress the Kosovar Albanians,’’ he said.

Shea rejected the notion that the NATO strikes instigated the Serb actions

against the Kosovars. ‘‘These [Serb] atrocities have been going on for more than

a year,’’ he said. He said once the Yugoslavian air defense system is neutralized,

NATO pilots will concentrate on hitting army heavy artillery, tanks and armored

personnel carriers.

Pilots reported no surface-to-air missiles on the first night. NATO spokesman

Royal Air Force Air Commodore David Wilby said the Yugoslavians fired one

SA-6 SAM during the second night, but it was fired ‘‘ineffectively.’’

Bacon said the reason may be a degraded Serb air defense system or the Serbs

are hoarding missiles. ‘‘We just don’t know at this stage,’’ he said. ‘‘But it is clear

that when we fly in such a hostile environment we take steps involving the attack

or suppression of air defenses, as well as various evasive actions to avoid or

make it difficult for our planes to be attacked.’’

NATO aircraft shot down three Serb MiG-29 Fulcrum aircraft the first night—

two by U.S. pilots, one by a Dutch pilot. Pilots met no aircraft the second night.

Bacon said some Serb aircraft were destroyed on the ground.

Bacon and Shea said the bombing campaign will end once Yugoslavia complies

with NATO demands. The person who can define how long campaign lasts is

Yugoslav President Slobodan Milosevic, Bacon said.

‘‘He can at any minute decide that peace is the best way to resolve this issue.

He can go back to the table and sit down and say that he’s going to agree to a

114 Nuclear Policy and Strategy in the Future

cease fire, say he’s going to pull his forces out and end this,’’ Bacon said. ‘‘That

is the quickest way to end [the aerial campaign], and it could happen any time

he decides to do it.’’

Shea said after a March 26 North Atlantic Council meeting that the 19 NATO

nations were demonstrating ‘‘unity, resolve and determination.’’ ‘‘We want to

stop, once-and-for-all, the brutal acts of repression in Kosovo,’’ he said.

The unique American status in high-tech conventional warfare creates

an ironical situation. Nuclear weapons, once the queens of the chessboard

in Washington as in Moscow, now become the preferred weapons

of the weak, not the strong. The stronger military powers of the next

century will be those who can excel in ‘‘post-nuclear’’ warfare. 37 This is

no small order. It means more than just the development of concepts for

reconnaissance-strike or, in broader compass, space-land warfare. It also

implies bringing weapons from the development stage into procurement

and from procurement into the force structure. As well, it implies realistic

training on the new systems so that reasonable guesstimates can be

obtained about their wartime performances. Recall that the U.S. ‘‘stealth’’

fighter-bomber (F-117) outperformed everyone’s expectations in the Gulf

War of 1991, much to the distress of Iraqi leader Saddam Hussein and

the delight of U.S. President George Bush. 38

Nuclear Weapons and the RMA

The assumption that the revolution in military affairs automatically

favors the United States is not necessarily correct. If one version of the

theory of hegemonic change in international relations is correct, any

dominant hegemon is eventually going to be challenged by a peer competitor

or by a coalition of hostile states that accumulates sufficient

power to do so. 39 Highly competent challengers to the status quo will

eventually emerge and force the hegemon to reactive and defensive

moves, including military ones. With regard to the RMA, for example,

Chinese strategic thinkers have already indicated considerable interest

in the impact of new information technology on warfare. Citing studies

led by Andrew Marshall of the Pentagon’s Office of Net Assessment and

other U.S. military service analysis, Major General Wang Pufeng, formerly

director of the Strategy Department, Academy of Military Science,

Beijing, stated quite categorically in 1995:

In the near future, information warfare will control the form and future of war.

We recognize this developmental trend of information warfare and see it as a

driving force in the modernization of China’s military and combat readiness. This

trend will be highly critical to achieving victory in future wars. 40

Russia and Nuclear Weapons After the Cold War 115

Where Russian, or for that matter, U.S. nuclear weapons fit into this

picture is yet to be determined. There is the ‘‘optimist’s future’’ in which

U.S. and Russian forces are reduced to the size of minimum deterrents,

capable of second strike retaliation against a small number of countervalue

target sets. 41 In addition, the qualities of these forces would change

dramatically. Targeting would be a la carte and might be decentralized

to the regional or functional CINCs, with oversight from above. Further,

in the optimist’s future, nuclear weapons would be taken off ready-alert

status. Most warheads would no longer be mated with launchers. The

connections between warning, intelligence and response would be decoupled

so that a slowing down of the response mechanism, instead of

a speeding up, would be progammed into the decision-making process.

Both the United States and Russia would proceed on the assumption

that any nuclear weapon originating from one side’s deployed forces and

heading toward a destination in the other’s territory was an accidental

or inadvertent launch. Each would agree to willingly run a greater risk

of a Type I error (undetected attack) than of a Type II error (falsely

detected attack), contrary to the Cold War practice. 42

Would these changes be meaningful, at or below START II levels of

deployment? The changes are meaningful not only for the relationship

between the United States and Russia, but also for the message that they

send to other nuclear powers and to potential proliferators. The vast

majority of states have voted to extend indefinitely the Nuclear Non-

Proliferation Treaty (NPT); the United States, among other advocates of

indefinite extension, took some consolation from this outcome. The Clinton

administration also felt justifiably positive about the strong degree

of international support it was able to line up behind the draft Comprehensive

Test Ban (CTB) treaty opened for signature in 1996. Further to

the cause of anti-nuclearism, there was considerable international support

for multinational agreements to cut off the production of fissile materials.

On the other hand, the announced nuclear tests of India and Pakistan

in May 1998, and the dubiously successful test launch of a medium-range

ballistic missile (MRBM) by Iran two months later, amounted to a temporary

but serious setback for nuclear nonproliferation and for efforts to

prevent the spread of missile technology. The Cold War U.S. and Soviet

potlatch over nuclear weapons could not be blamed entirely for the latest

additions to the nuclear club: India and Pakistan both argued that their

regional security problems dictated an explicit nuclear status. But there

can be no denial that the U.S. and Soviet military behaviors showing

faith in nuclear deterrence between 1945 and 1990 set an example that,

sooner or later, others would find persuasive. The question was whether

deterrence could work on the Indian subcontinent or between Iran and

116 Nuclear Policy and Strategy in the Future

its various regional enemies in the same way that it ‘‘worked’’ between

Washington and Moscow during the Cold War. 43

In addition to sending the signal that nuclear proliferation is to be

discouraged, present nuclear powers should also operate and deploy

their own forces as a demonstration of their own commitment to nuclear

stability and security. Slow-flying launchers are to be preferred to fast

flyers; delayed-reaction decision norms to prompt reaction; multiplesensor

designation to single-source dependency; and, above all, the

reliable political control of armed forces, established prior to the reality

of nuclear competency, to a politico-military context in which the authority

for nuclear authentication and enablement is still being contested.

It is at this point that three issues tie together: the issue of U.S. and

Russian nuclear arms reductions, the issue of military-technological revolution,

and the issue of political accountability of the armed forces. Two

aspects of the new information warfare have been noted already: the

struggle to control the electronic spectrum, and the technological and

other pressures tending to push effective command downward, and forward.

A third attribute of the new information wars is this: high-precision

weapons, combined with automated control systems and with almost

total transparency of the battlefield, will enable as never before the preemptive

destruction of command and control systems at the very outset

of a war without nuclear weapons. 44 Prompt destruction of command

posts, communications networks, sensors and other C4I necessities could

lead to immediate loss of combat stability—the ability to maintain cohesion

and control under the stress of military action. 45

In other words, the revolution in military affairs could help to undo

itself if it creates sufficient fears on the part of new nuclear nations that

their capabilities will be subject to timely and decisive preemption

whether or not they have threatened explicitly to use nuclear weapons against

regional opponents. The Clinton counterproliferation strategy carries this

message, at least by implication, to North Korea and Iraq. Thus far, the

message has not produced too much international controversy on account

of the assumption that counterproliferation would only be directed

against outlaw states and rogue regimes. However, counterproliferation

runs the risk of creating anticipatory U.S. allies and dependencies beyond

the reach of timely American reinforcement. 46 Will the next aspiring

regional hegemon allow the United States about five months to

regroup and respond, as did Saddam Hussein? Counterproliferation also

runs military risks. If it is less than totally successful, a nuclear riposte

against U.S. forces or allied territory is certainly possible. A third risky

aspect of counterproliferation is that, as a declaratory policy, it invites

aspiring proliferators to be even more secretive about their state of the

art. Fourth, counterproliferation opens the door to another debate about

Russia and Nuclear Weapons After the Cold War 117

theater defenses and about the costs of TMD deployments, and on whose


Recognizing that improved theater missile defenses could be part of

the revolution in military affairs is another illustration of the connections

between post-nuclear innovation and nuclear residual forces. Since the

politics of things suggest that nuclear weapons, if only on account of

inertia, will not soon be got rid of, the military-technological revolution

in conventional weapons acts as a new overlay on an old substructure

for deterrence. Nuclear weapons can lurk in the basement for delayed

and ultimate retaliation while the initial stages of a war are fought

‘‘above’’ the nuclear threshold (the metaphor is a deliberate contrast to

the assumptions of the Cold War) by high-technology conventional

forces. The question then becomes what one side will do when the tide

has turned against its forces and the Faustian temptation to use nuclear

weapons reappears.


Western observers of the Russian armed forces found themselves unexpectedly

concerned after 1991 about Russia’s military weakness instead

of its potential strength. The disintegration of the Russian armed

forces between 1991 and 1995 caused increased reliance on nuclear weapons

as the makeweights of Russia’s deterrence strategy. And Russia’s

declaratory doctrine, emphasizing nuclear first use in response to conventional

military attacks on strategic military targets in Russia, was

given credence by the sorry state of Russia’s conventional forces: Russia’s

leadership might be forced into early nuclear use or threat in spite of


The role of nuclear weapons in future Russian military strategy is

partly a carryforward from the Cold War past and partly open ended,

subject to future Russian capabilities and threat assessments. Russia’s

willingness to inherit Soviet accountability for ratifying and implementing

START agreements avoided premature derailment of that process.

Russian equivocation on the ratification of START II does not necessarily

spell permanent difficulty for the future of strategic arms reductions.

Reductions below START II levels would still leave both states with survivable

nuclear forces adequate to guarantee assured retaliation against

a variety of target sets. The future competency of Russia’s warning and

control systems is less certain, and Russia’s security forces, including

those that protect and command its nuclear weapons, are melting down

along with the rest of its economy. During the Cold War Soviet nuclear

forces were thought to be its paragons of deterrence; now, they are

mostly vestiges of an old order that still drags Russia down.

Appendix: Force Structures



Appendix (continued)



Appendix (continued)



124 Nuclear Policy and Strategy in the Future

Appendix (continued)


1. Jack Mendelsohn, ‘‘The U.S.-Russian Strategic Arms Control Agenda,’’

Arms Control Today (November/December 1997), p. 12.

2. Ibid.

3. Ibid.

4. See the comments by John Rhinelander in Arms Control Today (March 1999),

pp. 10–12.

5. See the remarks by Matthew Bunn in Arms Control Today (March 1999),

pp. 12–14.

6. Ibid., p. 13.

7. Ibid.

8. I am grateful for the opportunity to review two papers by Frederic S.

Nyland, U.S. Arms Control and Disarmament Agency, in which the subject of

measuring first strike stability is discussed with more mathematical sophistication

than it is here. Nyland is not responsible for any of the measures in this


9. Douglas J. Gillert, ‘‘Cohen Announces National Missile Defense Plan,’’

Armed Forces Press Service, January 21, 1999, DEFENSE-PRESS-SERVICE-


10. Ibid.

11. Glenn A. Kent and David E. Thaler, First Strike Stability and Strategic Defense:

Part II of a Methodology for Evaluating Strategic Forces (Santa Monica, Calif.:

RAND Corporation, 1990), p. 2.

12. Ibid.

13. Ibid., p. 4. I have modified the Kent-Thaler discussion to add an explicit

reference to the relative state of offensive and defensive technologies, in addition

to the relative force sizes in offense and defense. This strengthens their original

argument, in my judgment.

14. Interfax, Moscow, January 26, 1999, via, February 2,


15. Ibid.

Russia and Nuclear Weapons After the Cold War 125

16., February 4, 1999.

17. Bruce G. Blair, The Logic of Accidental Nuclear War (Washington, D.C.:

Brookings Institution, 1993), p. 173.

18. Ibid., pp. 166–167 and 173.

19. Ibid., p. 175. Soviet operational practice evolved in the same direction (see

esp. ibid., pp. 208–210). Blair notes that despite the drift toward LOW in both

sides’ operational predilections, LOW was a thin reed to lean on for a variety of

reasons (pp. 174 and 211–213).

20. Documentation of these generalizations would be tedious, given a large

and redundant literature. Examples include: Bruce G. Blair, Strategic Command

and Control: Redefining the Nuclear Threat (Washington, D.C.: Brookings Institution,

1985); Paul Bracken, The Command and Control of Nuclear Forces (New Haven,

Conn.: Yale University Press, 1983); and Kurt Gottfried and Bruce G. Blair, eds.,

Crisis Stability and Nuclear War (New York: Oxford University Press, 1988). Most

of this analysis is based on studies of the U.S. system; inferences about the Soviet

system relied on a less open database. However, see Desmond Ball, The Soviet

Strategic Command, Control, Communications and Intelligence (C3I) System, Reference

Paper No. 133 (Canberra, Australia: Strategic and Defence Studies Centre,

Australian National University, 1985); Blair, The Logic of Accidental Nuclear War,

chs. 4 and 5; Ghulam Dastagir Wardak, comp. and Graham Hall Turbiville, Jr.,

gen. ed., The Voroshilov Lectures: Materials from the Soviet General Staff Academy,

Vol. 1 (Washington, D.C.: National Defense University Press, 1989), pp. 244–250

and 261–262; Stephen M. Meyer, ‘‘Soviet Nuclear Operations,’’ in Ashton B. Carter,

John D. Steinbruner and Charles A. Zraket, eds., Managing Nuclear Operations

(Washington, D.C.: Brookings Institution, 1987), pp. 470–534; William T. Lee and

Richard F. Staar, Soviet Military Policy Since World War II (Stanford, Calif.: Hoover

Institution Press, 1986); and Harriet Fast Scott and William F. Scott, The Soviet

Control Structure: Capabilities for Wartime Survival (New York: Crane, Russak,


21. Crag Cerniello, ‘‘Clinton Issues New Guidelines on U.S. Nuclear Weapons

Doctrine,’’ Arms Control Today (November/December 1997), p. 23.

22. DOD News Briefing, Dr. Edward L. Warner, Assistant Secretary of Defense

for Strategy and Threat Reduction, February 25, 1999, DODTRANSCRIPTS-


23. Ibid.

24. The Guardian/Observer Press Association, Friday, April 9, 1999, Guardian

Website via Netscape.

25. Agence France-Presse, April 9, 1999, Webpage via Netscape.

26. I am grateful to Paul Bracken for suggesting this illustration. He bears no

responsibility for the analysis or argument here.

27. Evidence of Russian awareness of the significance of RMA appears in General

Makhmut Gareev, If War Comes Tomorrow: The Contours of Future Armed Conflict

(London: Frank Cass Publishers, 1998), esp. pp. 49 and 57–66.

28. For example, much of the founding theory and prophecy pertinent to U.S.

air power was developed in this period. See Carl H. Builder, The Icarus Syndrome:

The Role of Air Power Theory in the Evolution and Fate of the U.S. Air Force (New

Brunswick, N.J.: Transaction Publishers, 1994), esp. pp. 39–80. I am grateful to

Eliot A. Cohen, Johns Hopkins University, and Steven Metz, U.S. Army War

126 Nuclear Policy and Strategy in the Future

College, for calling my attention to the significance of this period. They bear no

responsibility for arguments here.

29. Col. David M. Glantz, The Motor-Mechanization Program of the Red Army

During the Interwar Years (Ft. Leavenworth, Kans.: U.S. Army Command and

General Staff College, Combined Arms Center, March 1990), esp. pp. 5–8 and13–


30. V. K. Triandafillov, Kharakter operatsii sovremennykh armii (Nature of operations

of modern armies), trans. William A. Burhans (Moscow-Leningrad, 1929),

pp. 109–110.

31. For elaboration, see Paul Bracken and Raoul Henri Alcala, Whither the

RMA: Two Perspectives on Tomorrow’s Army (Carlisle Barracks, Pa.: U.S. Army War

College, Strategic Studies Institute, July 1994).

32. Marshal of the Soviet Union N. V. Ogarkov, Istoriva uchit bditel’nosti (History

Teaches Vigilance) (Moscow: Voyenizdat, 1985), p. 69.

33. V. G. Reznichenko, Taktika (Tactics) (Moscow: Voenizdat, 1987), p. 24.

34. William J. Perry, ‘‘Military Action: When to Use It and How to Ensure Its

Effectiveness,’’ in Janne E. Nolan, ed., Global Engagement: Cooperation and Security

in the 21st Century (Washington, D.C.: Brookings Institution, 1994), pp. 235–242.

35. See, for elaboration, Jacob W. Kipp, ‘‘The General Staff Looks at ‘Desert

Storm’: Through the Prism of Contemporary Politics,’’ in Stephen J. Blank and

Jacob W. Kipp, eds., The Soviet Military and the Future (Westport, Conn.: Greenwood

Press, 1992), pp. 115–144.

36. John Erickson, ‘‘Quo Vadis?: The Changing Faces of Soviet/Russian Forces

and Now Russia,’’ in Blank and Kipp, eds., The Soviet Military and the Future,

pp. 33–58. See especially his comments in note 11, p. 55 on the reactions of individual

arms of service to the experience of the Gulf War of 1991.

37. I owe this locution to Edward Luttwak, although he is not responsible for

its use here.

38. Michael R. Gordon and Gen. Bernard E. Trainor, The Generals’ War: The

Inside Story of the Conflict in the Gulf (Boston: Little, Brown, 1995), p. 117.

39. Robert G. Gilpin, War and Change in World Politics (Cambridge: Cambridge

University Press, 1981).

40. Major General Wang Pufeng, ‘‘The Challenge of Information Warfare,’’ in

National Defense University, Institute for National Strategic Studies, Part Four,

The Revolution in Military Affairs (Ft. McNair, Washington, D.C.: National Defense

University, 1995).

41. The feasibility of this in comparison to higher levels of U.S. and Russian

strategic nuclear arms is explored in my article, ‘‘Deterrence Stability with

Smaller Forces: Prospects and Problems,’’ Journal of Peace Research, No. 1 (February

1995), pp. 65–78.

42. Blair, The Logic of Accidental Nuclear War, pp. 230 and 237–238.

43. According to Richard Ned Lebow and Janice Gross Stein, the Cold War

superpowers ‘‘overdosed’’ on deterrence, causing unnecessary disruptions in

their political relationship and bringing about some of the very confrontations

and crises that deterrence was supposed to have avoided. See Lebow and Stein,

We All Lost the Cold War (Princeton, N.J.: Princeton University Press, 1994),

pp. 324–376.

44. Two genres of Soviet literature are pertinent here. On command and con-

Russia and Nuclear Weapons After the Cold War 127

trol vulnerability under modern combat conditions, see D. A. Ivanov, V. P. Savelyev

and P. V. Shemanskii, Osnovy upravleniya voiskami v boyu (Fundamentals of

troop control in battle) (Moscow: Voenizdat, 1977), pp. 52–53. For the significance

of the initial period of war in Soviet military thinking, see S. P. Ivanov, Nachal’nyi

period voiny (The initial period of war) (Moscow: Voenizdat, 1974); Lt. Gen. A. I.

Yevseev, ‘‘O nekotorykh tendentsiyakh izmenii soderzhaniya i kharaktera nachal

‘nogo perioda voiny’’ (On certain tendencies in the changing content and character

of the initial period of war), Voenna-istoricheskii zhurnal, No. 2 (November

1985), pp. 10–20.

45. The Russian term for this is zhivuchest’. See C. N. Donnelly et al., The Sustainability

of the Soviet Army in Battle (The Hague: SHAPE Technical Center, September


46. The U.S. Defense Counterproliferation Initiative (CPI) is explained in William

J. Perry, Secretary of Defense, Annual Report to the President and the Congress

(Washington, D.C.: U.S. Government Printing Office, February 1995), pp. 71–82.

Chapter 6

Nuclear Weapons and

‘‘Third Wave’’ Warfare

Will the information revolution make nuclear weapons obsolete? Military

analysts and academic experts in security studies have described a

post–Cold War world dominated by a ‘‘Revolution in Military Affairs’’

based on high-technology, conventional weapons and nearly complete

knowledge of the wartime environment. 1 Other experts have suggested

that nuclear and other weapons of mass destruction are the past tense

of military art: information-based non-nuclear weapons, in this vision,

are the future of war. 2 The assumption of entirely separate paths for

nuclear weapons and for information warfare may be correct in general

but incorrect for some kinds of situations, including crises between

nuclear-armed states. In nuclear crises or arms races, nuclear power and

information strategies may come together to create a new, and potentially

terrifying, synthesis.

For present purposes, information warfare can be defined as activities

by a state or non-state actor to exploit the content or processing of information

to its advantage in time of peace, crisis and war, and to deny

potential or actual foes the ability to exploit the same means against

itself. 3 This is intended as an expansive, and permissive, definition, although

it has an inescapable bias toward military and security-related

issues. Information warfare can include both cyberwar and netwar. Cyberwar,

according to John Arquilla and David Ronfeldt, is a comprehensive,

information-based approach to battle, normally discussed in terms

of high-intensity or mid-intensity conflicts. 4 Netwar is defined by the

same authors as a comprehensive, information-based approach to societal

conflict. 5 Cyberwar is more the province of states and conventional

130 Nuclear Policy and Strategy in the Future

wars: netwar, more characteristic of non-state actors and unconventional

wars. 6


There are at least three reasons why the issues of information warfare

and its potential impact upon nuclear deterrence are important. First,

there is the growing significance of information warfare as a speculative

construct for military-strategic thinking, especially in countries with

high-technology militaries. Along with this trend is the emergence of

technologies which have made policy makers and planners more sensitive

to the significance of communications, computers and networks of

information in time of peace, crisis and war.

Second, there is some appreciation on the part of the U.S. defense

community, at least, that infowar may have strategic potential. This

means that infowar could, by itself, bring about fundamental or decisive

changes in the peacetime balance of power, or in friendly or enemy perceptions

of the balance. In addition, in time of war, infowar could be a

major, if not decisive, factor in determining the outcome of a military

conflict. The Gulf War of 1991 has been called, not without reason, the

first information war. 7 However, compared to what is conceivable or

potentially available for future infowarriors, Desert Storm was an earlygeneration

dress rehearsal. 8 The United States and its allies must assume

that other states will be able to acquire and to use infowar technologies:

leadership in present day infowar is no assurance of future preeminence.

This warning against infowar complacency is especially timely now,

when direct military challenges to U.S. security are seemingly nonexistent.

Military-strategic surprise is, by definition, almost never anticipated

by the victim. 9

The assertion that information warfare is an issue of potentially strategic

significance for U.S. planners and policy makers is not an endorsement

of the idea that the United States is vulnerable to an imminent

info-driven Pearl Harbor. The U.S. national information infrastructure

(NII) ‘‘has yet to suffer a major attack or anything close to it despite

numerous smaller attacks.’’ 10 Although some military functions are vulnerable

to attacks against part of the NII, a ‘‘strategic’’ disruption of the

entire U.S. computer and communications system by hackers would require

fictional capabilities not yet available. The more realistic aspect is

that partial attacks could distort policy makers’ and commanders’ perceptual

universes or information flows in crises or wartime situations,

leading to a faulty decision. 11 Even the carrying out of successful attacks

is not easy. As Martin C. Libicki has noted:

Can communications be sufficiently disrupted to retard or confound the nation’s

ability to respond to a crisis overseas? An enemy with precise and accurate

Nuclear Weapons and ‘‘Third Wave’’ Warfare 131

knowledge of how decisions are made and how information is passed within the

U.S. military might get inside the cycle and do real damage—but the enemy

must understand the cycle very well. Even insiders can rarely count on knowing

how information is routed into a decision; in an age in which hierarchical information

flow is giving way to networked information flow the importance of any

one predesignated route is doubtful. 12

A third reason for the significance of the infowar/deterrence relationship

is the continuing reality of nuclear deterrence in the world after the

Cold War. Contrary to some expectations, nuclear weapons and arms

control issues have not vanished over the horizon in a post–Desert Storm

euphoria. There are at least four reasons for this:

First, Russia still has many thousands of nuclear weapons, including

those of intercontinental range. U.S. officials anxiously await the ratification

by the Russian State Duma (lower house of the Russian legislature)

of the START II agreement that would reduce permitted strategic

nuclear warheads for each side to between 3,000 and 3,500 force loadings.

13 U.S. Department of Defense funds have been authorized by congressional

Nunn-Lugar legislation for the destruction, dismantlement

and secure storage of much of the Cold War Soviet nuclear arsenal. Russia’s

willingness to ratify START II and to proceed to a Clinton-sought

START III is related to two other important security issues: NATO enlargement,

and continued adherence to the ABM Treaty of 1972. 14

A second reason why nuclear deterrence remains important in the

post–Cold War world is that the other acknowledged nuclear powers, in

addition to the United States and Russia, show no inclination to abandon

nuclear weapons as ultimate deterrents. China is, in fact, by all accounts

engaged in a significant modernization of its military-technology base,

including the base that supports improved delivery systems for nuclear

weapons. France maintains the force de dissuasion as its ultima ratio and,

despite increased cordiality with NATO of late, continues its Cold War

tradition of reserving only to French decision the right to launch nuclear

weapons in retaliation. British modernization of nuclear-strategic delivery

systems for ballistic missile submarines continued into the 1990s,

with U.S. assistance. With regard to declaratory strategy for nuclear use,

while NATO has pushed nuclear weapons into the background as weapons

of last resort, Russia’s conventional military weakness has propelled

its nuclear weapons into the forecourt.

A third reason for the continued importance of nuclear deterrence is

the spread of nuclear weapons among ‘‘opaque’’ proliferators and the

potential for additional non-nuclear states to acquire these and other

weapons of mass destruction. The ‘‘opaque’’ proliferators whom virtually

everyone now includes in the category of nuclear competent actors

are India, Israel and Pakistan. Experts disagree about the sizes of their

132 Nuclear Policy and Strategy in the Future

weapons inventories or deployment capabilities on short notice; more

significant, even their military and scientific experts have stopped denying

the all-but-certain nuclear-acquired status of these three countries.

In addition to these opaque proliferators, other candidate non-nuclear

states thought to be pursuing nuclear capability include Libya, Iran, Iraq

and (until 1994) North Korea. The potential for regional destabilization

once any of these currently non-nuclear states acquires nuclear weapons

is enhanced given their apparent interest in other weapons of mass destruction

and in delivery systems for nuclear, chemical and/or biological

weapons. 15

Fourth, nuclear deterrence remains important because non-state actors,

including terrorists, interstate criminal organizations (ICOs), and revolutionary

actors of various sorts, may acquire nuclear or other weapons

of mass destruction. Although for some of their purposes nuclear weapons

would be superfluous, for other objectives they would, even in small

numbers and puny yields, be quite appropriate. Terrorists who could

present a plausible threat to detonate even a small nuclear device within

a target state could raise the potential risk of hostage rescue operations,

for hostages and for armed forces of the target state. 16 Terrorists allied

with a state actor and equipped with nuclear weapons could gain from

their ally valuable intelligence, sanctuary and diplomatic cover. Criminal

cartels could replenish their coffers by trafficking in nuclear weapons or

weapons grade materials as middlemen, even if they were not the end

users themselves. Consider the increased degree of freedom for U.S. opponents

in Panama (Just Cause) or in Iraq (Desert Shield and Desert

Storm) if Noriega’s agents in the Canal Zone or Saddam’s operatives in

Kuwait had had available even a small nuclear device, well timed and

strategically located.

A fifth reason for the continuing significance of nuclear deterrence in

the post–Cold War system is, somewhat paradoxically, Russia’s military

and economic weakness for at least the remainder of this century. There

are two aspects of this weakness that might contribute to nuclear deterrence

failure based on failed crisis management, mistaken preemption or

accidental/inadvertent war. First, Russia’s conventional military weakness

makes it more reliant on nuclear weapons as weapons of first choice

or first use, instead of last resort. Second, Russia’s economic problems

mean that it will have difficulty maintaining personnel morale and reliability.

In addition, Russia’s military will also be lacking in funds to

modernize and properly equip its early warning and nuclear command,

control and communciations systems. Russia’s military and economic

weaknesses may encourage, at least in the near term, reliance on prompt

launch doctrines for nuclear retaliation and emphasis on nuclear first use

in situations where conventional forces are in jeopardy. In addition, at

least some Russian thinkers have noted the potentially strategic signifi-

Nuclear Weapons and ‘‘Third Wave’’ Warfare 133

cance of information warfare and have connected the consequences of

information attacks to potentially nuclear responses:

From a military point of view, the use of information warfare means against

Russia or its armed forces will categorically not be considered a non-military

phase of a conflict, whether there were casualties or not...considering the possible

catastrophic consequences of the use of strategic information warfare means

by an enemy, whether on economic or state command and control systems, or

on the combat potential of the armed forces,...Russia retains the right to use

nuclear weapons first against the means and forces of information warfare, and

then against the aggressor state itself. 17

The preceding argues the significance of the relationship between infowar

and deterrence, in general. The sections that follow offer four more

specific perspectives on the relationship between infowar and nuclear

deterrence. We consider the implications of the nuclear deterrenceinformation

warfare nexus in situations of (1) crisis management, (2) preemption,

(3) accidental-inadvertent war, and (4) conflict termination.

These are not mutually exclusive categories, and some illustrative material

could be just as easily provided under one heading as under another.

Since no nuclear weapons have been fired in anger since Nagasaki,

placement of illustrations or cases is more like a moving finger of historical

analogy than a scientific experiment. For example, it is quite reasonable

that the same incident could illustrate pertinent points about

crisis management or the risk of accidental/inadvertent war, or both.


Crisis management, including nuclear crisis management, is both a

competitive and cooperative endeavour between military adversaries. A

crisis is, by definition, a time of great tension and uncertainty. 18 Threats

are in the air, and time pressure on policy makers seems intense. Each

side has objectives that it wants to attain and values that it deems important

to protect. During crisis, state behaviors are especially interactive

and interdependent with those of another state. It would not be too farfetched

to refer to this interdependent stream of interstate crisis behaviors

as a system, provided the term ‘‘system’’ is not reified out of

proportion. The system aspect implies reciprocal causation of the crisis

time behaviors of ‘‘A’’ by ‘‘B,’’ and vice versa.

One aspect of crisis management is the deceptively simple question:

what defines a crisis as such? When does the latent capacity of the international

order for violence or hostile threat assessment cross over into

the terrain of actual crisis behavior? It may be useful to separate traffic

jams from head-on collisions. A traffic jam creates the potential for col-

134 Nuclear Policy and Strategy in the Future

lisions but leaves individual drivers with some steering capacity for the

avoidance of damage to their particular vehicles. A breakdown of general

deterrence in the system raises generalized threat perceptions among

salient actors, but it does not guarantee that any particular state to state

relationship will deteriorate into specific deterrent or compellent threats.

Therefore, Patrick Morgan’s concept of ‘‘immediate’’ deterrence failure

is useful in defining the onset of a crisis: specific sources of hostile intent

have been identified by one state with reference to another, threats have

been exchanged, and responses must now be decided upon. 19 The passage

into a crisis is equivalent to the shift from Hobbes’ world of omnipresent

potential for violence to the actual movement of troops and

exchanges of diplomatic demarches.

The first requisite of successful crisis management is clear signaling.

Each side must send its estimate of the situation to the other in a way

that the other correctly interprets. It is not necessary for the two sides to

have identical or even initially complementary interests. But a sufficient

number of correctly sent and received signals are prerequisite to effective

transfer of enemy goals and objectives from one side to the other. If

signals are poorly sent or misunderstood, steps taken by the sender or

receiver may lead to unintended consequences, including miscalculated


Clear signaling for effective crisis management includes high-fidelity

communication between adversaries, and within the respective decisionmaking

structures of each side. High-fidelity communication in a crisis

can be distorted by everything that might interfere physically, mechanically

or behaviorally with accurate transmission. Electromagnetic pulses

that disrupt communication circuitry or physical destruction of communication

networks are obvious examples of impediments to highfidelity

communication. Cultural differences that prevent accurate

understanding of shared meanings between states can confound deterrence

as practiced according to one side’s theory. As Keith B. Payne

notes, with regard to the potential for deterrence failure in the post–Cold

War period:

Unfortunately, our expectations of opponents’ behavior frequently are unmet,

not because our opponents necessarily are irrational but because we do not understand

them—their individual values, goals, determination and commitments—

in the context of the engagement, and therefore we are surprised when their

‘‘unreasonable’’ behavior differs from our expectations. 20

A second requirement of successful crisis management is the reduction

of time pressure on policy makers and commanders so that no untoward

steps are taken toward escalation mainly or solely as a result of a misperception

that ‘‘time is up.’’ Policy makers and military planners are

Nuclear Weapons and ‘‘Third Wave’’ Warfare 135

capable of inventing fictive worlds of perception and evaluation in which

‘‘h hour’’ becomes more than a useful benchmark for decision closure.

In decision pathologies possible under crisis conditions, deadlines may

be confused with policy objectives themselves: ends become means, and

means, ends. For example, the war plans of the great powers in July 1914

contributed to a shared self-fulfilling prophecy among leaders in Berlin,

St. Petersburg, and Vienna that only by prompt mobilization and attack

could decisive losses be avoided in war. Plans so predicated, on the

rigidity of mobilization timetables, proved insufficiently flexible for policy

makers who wanted to slow down the momentum of late July/early

August toward an irrevocable decision in favor of war and away from


A third attribute of successful crisis management is that each side

should be able to offer the other a safety valve or a ‘‘face-saving’’ exit

from a predicament that has escalated beyond its original expectations.

The search for options should back neither crisis participant into a corner

from which there is no graceful retreat. For example, President John F.

Kennedy was able, during the Cuban missile crisis of 1962, to offer Soviet

Premier Khrushchev a face-saving exit from his overextended missile

deployments. Kennedy publicly committed the United States to refrain

from future military aggression against Cuba and privately agreed to

remove and dismantle Jupiter medium-range ballistic missiles previously

deployed among U.S. NATO allies. 21 Kennedy and his ExComm advisors

recognized, after some days of deliberation and clearer focus on the Soviet

view of events, that the United States would lose, not gain, by public

humiliation of Khrushchev that might, in turn, diminish Khrushchev’s

interest in any mutually agreed solution to the crisis.

A fourth attribute of successful crisis management is that each side

maintains an accurate perception of the other side’s intentions and military

capabilities. This becomes difficult during a crisis because, in the

heat of a partly competitive relationship and a threat-intensive environment,

intentions and capabilities can change. As Robert Jervis has explained,

the process by which beliefs that war was inevitable might have

created a self-fulfilling Cold War prophecy:

The superpowers’ beliefs about whether or not war between them is inevitable

create reality as much as they reflect it. Because preemption could be the only

rational reason to launch an all-out war, beliefs about what the other side is

about to do are of major importance and depend in large part on an estimate of

the other’s beliefs about what the first side will do. 22

Intentions can change during a crisis if policy makers become more

optimistic about gains or more pessimistic about potential losses during

the crisis. Capabilities can change due to the management of military

136 Nuclear Policy and Strategy in the Future

alerts and the deployment or other movement of military forces. Heightened

states of military readiness on each side are intended to send a twosided

signal: of readiness for the worst if the other side attacks, and of

a nonthreatening steadiness of purpose in the face of enemy passivity.

This mixed message is hard to send under the best of crisis management

conditions, since each state’s behaviors and communications, as observed

by its opponent, may not seem consistent. Under the stress of time pressure

and of military threat, different parts of complex security organizations

may be taking decisions from the perspective of their narrowly

defined, bureaucratic interests. These bureaucratically chosen decisions

and actions may not coincide with policy makers’ intent, nor with the

decisions and actions of other parts of the government.

If the foregoing is accepted as a summary of some of the attributes of

successful crisis management, information warfare has the potential to

attack or to disrupt crisis management on each of the preceding attributes.

First, information warfare can muddy the signals being sent from one

side to the other in a crisis. It can do this deliberately or inadvertently.

Suppose one side plants a virus or worm in the other’s communications

networks. 23 The virus or worm becomes activated during the crisis and

destroys or alters information. The missing or altered information may

make it more difficult for the cyber-victim to arrange a military attack.

But destroyed or altered information may mislead either side into thinking

that its signal has been correctly interpreted, when it has not. Thus,

side A may be intending to signal ‘‘resolve’’ instead of ‘‘yield’’ to its

opponent on a particular issue. Side B, misperceiving a ‘‘yield’’ message,

decides to continue its aggression, meeting unexpected resistance and

causing a much more dangerous situation to develop.

Infowar can also destroy or disrupt communication channels that impede

successful crisis management. One way infowar can do this is to

disrupt communication links between policy makers and force commanders

during a period of high threat and severe time pressure. Two

kinds of unanticipated problems, from the standpoint of civil-military

relations, are possible under these conditions. First, political leaders may

have predelegated limited authority for nuclear release or launch under

restrictive conditions: only when these few conditions obtain, according

to the protocols of predelegation, would force commanders be authorized

to employ nuclear weapons distributed within their command. (See

the discussion of conflict termination, below, for related points about

predelegation). Clogged, destroyed or disrupted communications could

prevent top leaders from knowing that force commanders perceived a

situation to be far more desperate, and thus permissive of nuclear initiative,

than it really was. For example, during the Cold War, disrupted

communications between U.S. National Command Authority and ballis-

Nuclear Weapons and ‘‘Third Wave’’ Warfare 137

tic missile submarines, once the latter came under attack, could have

resulted in a joint decision by submarine officers and crew to launch in

the absence of contrary instructions.

Second, information warfare during a crisis will almost certainly increase

time pressure under which political leaders operate. It may do

this actually, or it may affect the perceived time lines within which the

policy-making process can take its decisions. Once either side sees parts

of its command, control and communications (C3I) system being subverted

by phony information or extraneous cyber-noise, its sense of panic

at the possible loss of military options will be enormous. In the case of

U.S. Cold War nuclear war plans, for example, disruption of even portions

of the strategic command, control and communications system

could have prevented competent execution of parts of the SIOP (the strategic

nuclear war plan). The SIOP depended upon finely orchestrated

time-on-target estimates and precise damage expectancies against

various classes of targets. Partially misinformed or disinformed networks

and communications centers would have led to redundant attacks

against the same target sets and, quite possibly, unplanned attacks on

friendly military or civil ground zeros.

A third potentially disruptive impact of infowar on nuclear crisis management

is that infowar may reduce the search for available alternatives

to the few and desperate. Policy makers searching for escapes from crisis

denouements need flexible options and creative problem solving. Victims

of information warfare may have diminished ability to solve problems

routinely, let alone creatively, once information networks are filled with

flotsam and jetsam. Questions to operators will be poorly posed, and

responses will (if available at all) be driven toward the least common

denominator of previously programmed standard operating procedures.

The propensity to search for the first available alternative that meets

minimum satisfactory conditions of goal attainment (‘‘satisficing’’) is

strong enough under normal conditions in nonmilitary bureaucratic organizations.

24 In civil-military command and control systems under the

stress of nuclear crisis decision making, the first available alternative may

quite literally be the last. Or, so policy makers and their military advisors

may persuade themselves. Accordingly, the bias toward prompt and adequate

solutions is strong. During the Cuban missile crisis, for example,

a number of members of the ExComm Presidential advisory group favored

an air strike and invasion of Cuba during the entire thirteen days

of crisis deliberation. Had less time been available for debate and had

President Kennedy not deliberately structured the discussion in a way

that forced alternatives to the surface, the air strike and invasion might

well have been the chosen alternative. 25

Fourth, and finally, on the issue of crisis management, infowar can

cause flawed images of each side’s intentions and capabilities to be con-

138 Nuclear Policy and Strategy in the Future

veyed to the other, with potentially disastrous results. This problem is

not limited to crisis management, as we shall see. A good example of

the possible side effects of simple misunderstanding and noncommunication

on U.S. crisis management will suffice for now. At the most tense

period of the Cuban missile crisis, a U-2 reconnaissance aircraft got off

course and strayed into Soviet airspace. U.S. and Soviet fighters scrambled

and a possible Arctic confrontation of air forces loomed. Khrushchev

later told Kennedy that Soviet air defenses might have

interpreted the U-2 flight as a prestrike reconnaissance mission or

bomber, calling for a compensatory response by Moscow. 26 Fortunately,

Moscow chose to give the United States the benefit of the doubt in this

instance and to permit U.S. fighters to escort the wayward U-2 back to

Alaska. The question why this scheduled U-2 mission had not been

scrubbed once the crisis began has never been fully answered. The answer

may be as simple as bureaucratic inertia compounded by noncommunication

down the chain of command by policy makers who failed to

appreciate the risk of ‘‘normal’’ reconnaissance under these extraordinary


All crises are characterized to some extent by a high degree of threat,

short time for decision and a ‘‘fog of crisis’’ reminiscent of Clausewitz’s

‘‘fog of war’’ that confuses crisis participants about what is happening.

Before the discipline of ‘‘crisis management’’ was ever invented by modern

scholarship, historians had captured the rush-to-judgment character

of much crisis-time decision making among great powers. 27 The influence

of nuclear weapons on crisis decision making is therefore not so dramatic

as has been sometimes supposed. The presence of nuclear forces obviously

influences the degree of destruction that can be done, should crisis

management fail. Short of that catastrophe, the greater interest of scholars

is in how the presence of nuclear weapons might affect the decisionmaking

process in a crisis. The problem is conceptually overdetermined:

there are so many potentially important causal factors relevant to a

decision with regard to war or peace.

Despite the risk of explanatory overkill for successful cases of crisis

management, including nuclear ones, the firebreak between crises under

the shadow of potential nuclear destruction, and those not so overshadowed,

remains important. Infowar can be unleashed like a pit bull

against the command, control, communications, computers, intelligence,

surveillance and reconnaissance (C4ISR) of a potential opponent in conventional

conflicts with fewer risks compared to the case of a nucleararmed

foe. The objective of infowar in conventional warfare is to deny

enemy forces battlespace awareness and to obtain dominant awareness

for oneself, as the United States largely was able to do in the Gulf War

of 1991. 28 In a crisis with nuclear weapons available to the side against

Nuclear Weapons and ‘‘Third Wave’’ Warfare 139

which infowar is used, crippling the intelligence and command and control

systems of the potential foe is at best a necessary, and certainly far

from a sufficient, condition for successful crisis management or deterrence.

And under some conditions of nuclear crisis management, crippling

the C4ISR of the foe may be simply unwise. What do we think the

Russians will do when their bunkers go bonkers? The answer lies only

partly in hardware: the mind-sets of people, their training and their

military-strategic culture must also be factored in. 29

One aspect of crisis management is the deceptively simple question:

what defines a crisis as such? When does the latent capacity of the international

order for violence or hostile threat assessment cross over into

the terrain of actual crisis behavior? It may be useful to separate traffic

jams from head-on collisions. A traffic jam creates the potential for collisions

but leaves individual drivers with some steering capacity for the

avoidance of damage to their particular vehicles. A breakdown of general

deterrence in the system raises generalized threat perceptions among

salient actors, but it does not guarantee that any particular state-to-state

relationship will deteriorate into specific deterrent or compellent threats.

Therefore, Patrick Morgan’s concept of ‘‘immediate’’ deterrence failure

is useful in defining the onset of a crisis: specific sources of hostile intent

have been identified by one state with reference to another, threats have

been exchanged, and responses must now be decided upon. 30 The passage

into a crisis is equivalent to the shift from Hobbes’ world of omnipresent

potential for violence to the actual movement of troops and

exchanges of diplomatic demarches.

With or without nuclear weapons, it bears repeating that states are not

simply seeking to avoid war once they have entered into a crisis: they

are trying to accomplish their objectives without war, if possible, or to

position themselves during the crisis so that if war occurs they will be

favorably disposed to fight it. These two objectives are in tension at the

margin. Nuclear weapons and the complexity of modern command and

control systems make this tension more acute. Other aspects of these

systems are noted below, but especially pertinent to crisis management

are several attributes of nuclear weapons and delivery systems in addition

to their high lethality. First, once launched they cannot be recalled.

Second, no defenses against large nuclear attacks are presently feasible.

Third, governments and their command/control apparatuses are more

fragile than forces. Governments and military commanders steering

through a nuclear crisis are aware as never before that they are plausible

specific and prompt targets of enemy war plans and that these attacks,

once launched, cannot be blunted. This concentrates the mind wonderfully,

but how effectively? 31

140 Nuclear Policy and Strategy in the Future


Preemption is the taking of a decision for nuclear first strike in the

expectation that the opponent has already ordered its forces into attack.

Preemption thus differs from preventive, or premeditated, war to forestall

a possible attack at a future time. 32 Although preventive war was

occasionally advocated by high U.S. officials during the Cold War, no

American president ever approved a war plan based on that assumption.

Preemption was, for the United States and for the Soviet Union, a more

serious and continuing concern. 33

The first requirement for the avoidance of nuclear preemption is that

neither side conclude that war is ‘‘inevitable’’ or that the opponent has

already begun it. The ‘‘war is inevitable’’ conclusion and the decision

that the opponent has already begun a war are related, but not necessarily

the same. Leaders with a fatalistic bent might conclude that war

was inevitable well before actual troop movements or shooting started.

Something of this sort seems to have gripped the Austro-Hungarians and

the Germans in the weeks immediately preceding the outbreak of World

War I. The judgment that one’s possible adversaries have already begun

to fight, correct or not, can also lead to a decision for war. In late July

1914 Germany judged Russian mobilization on the basis of Germany’s

own mobilization planning: mobilization for Germany was tantamount

to war. 34 Russia’s mobilization envisioned a concentration of forces near

the frontiers in order to hold a defensive position, but not necessarily a

decision to attack into enemy territory. Understanding Russia’s mobilization

through Germany’s own conceptual lenses, the Kaiser and his

advisors judged that war had, in fact, begun, although Russia’s Tsar was

still vacillating. 35

A second requirement for the avoidance of nuclear preemption is that

each side’s forces and strategic command, control and communications

(C3) be survivable against any conceivable surprise attack. This is a tall

order, and only the U.S. and Soviet systems in the Cold War years could

aspire to meet the requirement against a large and diverse arsenal. The

second part of the requirement, survivable command and control systems,

is much harder to attain. Much depends upon the standard of

survivability expected of the components of any command and control

system: physical infrastructure (bunkers, computers, landlines), organizational

hierarchy (chain of command), communications networks (vertical

and horizontal or lateral), and, not least, people (personal reliability,

professional ethos). 36 One of these components could fail while others

performed up to the expected standard, whatever that was. For example,

the destruction of physical infrastructure would not necessarily disrupt

the entire chain of command nor change the personal reliability of individuals.

Nuclear Weapons and ‘‘Third Wave’’ Warfare 141

Table 6.1

U.S. Cold War Views of Nuclear Deterrence

Source: Based on and adapted from Richard Ned Lebow and Janice Gross Stein, We All

Lost the Cold War (Princeton, N.J.: Princeton University Press, 1994), pp. 348–368. See

also David W. Tarr, Nuclear Deterrence and International Security (London: Longman,

1991), chs. 5–7.

The standard of survivability for forces and command systems can be

set high or low. At one extreme, the simple requirement to strike back

with a minimum retaliatory force against the cities of the opponent might

satisfy the requirements of policy makers. At another extreme, policy

makers might demand nuclear forces and command systems sufficient

to fight an extended nuclear war through many stages. Richard Ned

Lebow and Janice Gross Stein, among others, classify U.S. schools of

thought about the military requirements of nuclear deterrence into three

groups: (1) mutual assured destruction, (2) nuclear war fighting, and (3)

finite deterrence. 37 Table 6.1 summarizes the tenets of the various schools

of thought in terms of each school’s images of Soviet foreign and security

policy objectives, weapons and C3 requirements, and presumptions

about the character of U.S.-Soviet nuclear deterrence.

The model of minimum deterrence never appealed to U.S. presidents

once nuclear weapons and delivery systems became plentiful in number:

they wanted options in addition to several hundred retaliatory strikes

against cities. 38 On the other hand, despite considerable hubris in declaratory

policy, no U.S. forces and command systems ever met the standard

of being able to conduct an ‘‘extended’’ large-scale nuclear war. 39

For the most part, strategists and government officials judged that this

was neither necessary nor advisable for deterrence. Most U.S. strategists

142 Nuclear Policy and Strategy in the Future

and policy makers agreed, notwithstanding other differences of opinion,

that forces and command systems had to be survivable enough to guarantee

‘‘unacceptable’’ retaliation against a variety of enemy target sets.

Arguments about degrees of damage that would meet the standard of

‘‘unacceptable’’ continued throughout the Cold War. 40

A third requirement for the avoidance of preemption is that the management

of nuclear alerts must be competent. This means that the alerting

process itself must allow for gradations of military activity, and that

such activities must take place according to the guidance previously established

by political leaders for the operation of nuclear forces. 41 Alerts

also send to the other side, intentionally or not, signals about intent: is

the alert of side A a precautionary measure on account of its fears of

side B, or is it an ominous sign that side A is preparing to attack? Alerts

are open to ambiguous interpretation. For example, as German forces

massed on the western borders of the Soviet Union in the spring of 1941,

Soviet and foreign intelligence sources warned the Kremlin of Hitler’s

plans to attack. Stalin chose to ignore those warnings, and the launch of

Operation Barbarossa on June 22, 1941, caught Soviet border forces premobilized

and unprepared. 42 On account of security, nuclear alerts may

be more opaque than prenuclear ones, with equal possibilities of ambiguity

in intelligence assessment of their significance.

A fourth requirement for the avoidance of preemption is the timely

and accurate communication of the activities and status of military forces

at the sharp end of the spear to higher political and military authorities.

Political leaders are, on the evidence, often ignorant of the operational

details and standard operating procedures by which their own militaries

alert, move, feed, transport and hide. 43 Two problems here are breadth

versus narrowness of vision, and time lines. At the strategic level of

command, one is concerned with the ‘‘big picture’’ of fleets, armies and

air wings moving in combined-arms formations. At the tactical level,

things are more immediately dangerous and personal. The term ‘‘GI’’

may have survived as slang because it expressed succinctly the worldview

of the grunt. Differences in time horizon also matter in ascertaining

the status of one’s own forces during times of peace, crisis or war. The

time lines of the high command are extended; those of tactical commanders,

dominated by the imminent likelihood of being under fire.

The operation of U.S. maritime forces during the Cuban missile crisis

offers one illustration of this difference in time and perspective between

center and periphery. President Kennedy and his advisors were concerned

to avoid if possible a direct confrontation at sea between Soviet

and American forces. To that end, they exercised a degree of close supervision

over U.S. Navy forces in the theater of operations that some

in the Navy chain of command, including the chief of naval operations,

found objectionable. From the perspective of Admiral Anderson and

Nuclear Weapons and ‘‘Third Wave’’ Warfare 143

some of his tactical commanders, operating a blockade was a dangerous

exercise that had to be left to military experts if it was to be carried out

successfully. The standpoint of Kennedy and McNamara was that this

was no ordinary tactical operation: shooting between Soviet and American

naval forces in the Caribbean could have strategic consequences. 44

A fifth requirement for the avoidance of nuclear preemption is that

flexible options be available to policy makers and to force commanders.

This requirement has two parts. First, policy makers and their military

advisors must be aware of flexible options and believe that they can be

carried out under the exigent conditions. Second, those actually holding

custody of nuclear weapons or operating nuclear forces must understand

the options and be prepared actually to carry them out once told to do

so. Each of these requirements is not necessarily easy to meet. Policy

makers may not know the availability of options or may be reluctant to

order the military into unpreferred choices under duress. For example,

Kaiser Wilhelm II, in the late stage of the crisis preceding the outbreak

of World War I, ordered his chief of the General Staff, General Helmuth

von Moltke, to reverse the direction of the main German mobilization,

from west to east (against Russia, instead of France). The astonished Von

Moltke responded that ‘‘It cannot be done.’’ Moltke meant that it could

not be done without disrupting the intricate railway timetables for moving

forces and supplies that had been painfully worked out according to

the prewar Schlieffen plan, based on an assumed prompt offensive westward

against France through Belgium.

A second aspect of this requirement is that those holding custody of

nuclear weapons, including nuclear force commanders once weapons

have been released to them, understand those flexible options of interest

to policy makers and are prepared to implement them. Policy makers

have sometimes engaged in wishful thinking about available military

options: because flexibility is deemed necessary, it is therefore assumed

to be possible. An example of misbegotten assumptions of this sort is

provided by the behavior of Russia’s high command in the last week of

the July 1914 crisis. Foreign Minister Sergei Sazonov sought a partial

mobilization, including only Russia’s western military districts, directed

against Austria-Hungary but not against Germany. The intent was to

send a message to Austria of Russia’s determination not to permit further

Austrian aggression against Serbia. Sazonov was unaware that the Russian

General Staff had in fact done no serious planning for a partial

mobilization under these or similar circumstances of possible multifront

war against Germany and Austria. Neither Russia’s war minister nor the

chief of the General Staff warned Sazonov to this effect. When the Tsar

at Sazonov’s urging ordered into effect a partial mobilization, the General

Staff and the rest of the military chain of command sat on their

144 Nuclear Policy and Strategy in the Future

hands, hoping it would be superseded by an order for general mobilization

that they were prepared to implement. 45

Nuclear war plans of the Cold War were equally elaborate as the

Schlieffen plan, and policy makers’ lack of familiarity with them is well

documented. Few presidents received more than a once-and-done briefing

on the SIOP. 46 Faced with a serious nuclear crisis or actual war,

presidents would have been presented with a short menu of options and

advised that other options could not be improvised on the spot. U.S.

presidents and Soviet premiers of the Cold War years thus had political

control only over the actual decision to start a nuclear war, but little

effective control over the military execution of that command once given.

Information warfare might contribute to a failure on the part of policy

makers or the command system to meet each of these requirements for

the avoidance of preemption.

First, infowar might raise first strike fears based on the mistaken assumption

by one side that the other had concluded that ‘‘war is inevitable.’’

It might do this in one of two ways. First, deliberate attack on

the information systems of the other side in a crisis might lead the victim

to conclude mistakenly that war had already begun. The United States

has had some experience with infowar against itself of this sort. In 1979

and in 1980, misplayed tapes or failed computer chips resulted in false

warnings of attack at the North American Aerospace Defense Command

(NORAD). In the June 3, 1980, incident, indicators of attack from Soviet

land and sea-based missiles spread from NORAD to other key nuclear

command posts. U.S. military commands prepared for retaliation in case

of valid attack warning: Minuteman launch control officers were alerted

to be ready for possible launch orders, and bomber crews on bases

throughout the United States ran to alert aircraft and started their engines.

47 Fortunately, no crisis was in progress at the time, and operators

were quickly able to calm fears throughout the nuclear chain of command.

A second way in which infowar might raise first strike fears is by

confusion of the enemy’s intelligence and warning (see also below for

the discussion of this factor in relation to accidental/inadvertent war),

but in unexpected ways. If warning systems and the fusion/analysis centers

to which they are connected overreact to confusion in their shared

information nets, interpreters might conclude that the other side has already

launched a first strike or an ‘‘infoattack’’ preparatory to preemptive

attack. And this logic might not, in the abstract, be totally flawed.

From the perspective of traditional military strategy, it makes sense to

attack the enemy’s command and control system, including warning systems,

in the early stages of a war. A blinded and misinformed opponent

can be defeated in battle faster and at lower cost, as U.S. air strikes in

the Gulf War of 1991 demonstrated.

Nuclear Weapons and ‘‘Third Wave’’ Warfare 145

Weaknesses in the contemporary Russian warning and information

systems for nuclear conflict suggest that the danger of feared attacks

against Russia’s command and control system is not hypothetical. Russia’s

military leadership has acknowledged slippage in the reliability of

personnel in the Strategic Nuclear Forces, including the Strategic Rocket

Forces responsible for land-based, long-range missiles. Inadequate pay

and poverty living conditions for the troops are compounded by insufficient

funding for replacements and upgrades to computers and electronics

that tie together warning systems, communications and

commanders. Cleavage of Ukraine and the Baltics from Russia’s western

perimeter cost Moscow important warning radar sites, and key satellite

tracking stations formerly under Moscow’s control now reside in the

newly independent states of Georgia, Kazakhstan and Ukraine. These

technical and personnel problems in Russia’s warning and response system

are compounded by Russia’s proclivities in military strategy related

to nuclear weapons. Russia’s approved military doctrine expressly permits

nuclear first use under some conditions, and Russia’s Strategic

Rocket Forces remain on a hair-trigger, launch-on-warning posture capable

of nuclear preemption. 48

Third, infowar might contribute to preemption by complicating the

management of alerting operations. Even the alerting of conventional

forces on the part of major powers requires many complex interactions

among force components. Nuclear forces must be alerted under separate

protocols that ensure safety and security of those weapons and that

alerted nuclear forces do not invite attack on themselves. Groundlaunched

and most air-delivered weapons must be moved from storage

sites and mated with launch vehicles. Submarines will be surged from

port enroute to their holding stations or probable launch positions. As

one side’s forces and command systems surge to higher levels of activity,

the other side’s intelligence sensors will be taking in more and more raw

information. With strategic nuclear forces and command systems as complicated

as those of the Cold War Americans’ and Soviets’, the potential

for mischievous misconstruction of alerting operations existed even without

present and foreseeable potential for infowar.

What, for example, were the Soviets to make of the massive U.S. military

alert of U.S. conventional air, ground and naval forces during the

Cuban missile crisis, poised for an invasion of Cuba if the president so

decided? U.S. leaders at the time viewed the preparations as precautionary,

should the Soviets refuse to remove the missiles and the crisis turn

uglier. Soviet leaders understandably saw things differently. They concluded

by October 27 that the United States was definitely preparing for

an invasion of Cuba. U.S. participants in Cuban missile crisis decision

making aver that there were preparations for invasion, but no plan for

invasion, as such, had been approved. This distinction, from the Soviet

146 Nuclear Policy and Strategy in the Future

and Cuban standpoint, might have been distinction without a difference.


Now imagine a rerun of this crisis with more up-to-date information

warfare techniques, a ‘‘holodeck’’ version with the same political setting

but with technology of the year 2010. In the holodeck version, U.S. planners

might insert corruptions into Soviet information networks that simulated

an overwhelming attack force about to strike at Cuba, when in

fact such a force was far from being fully prepared. And Khrushchev,

aided by manipulation of U.S. information systems, might have simulated

a full-scale nuclear alert of Soviet forces in order to intimidate

Washington. We know now that, if the United States had actually invaded

Cuba or if the Soviets had actually gone to full nuclear alert in

the Cuban missile crisis, the probability of a mutually acceptable outcome

short of war would have been reduced. U.S. planners were unaware

for much of the crisis period of certain tactical nuclear weapons

deployed with Soviet ground forces and authorized for use in case of an

American invasion of Cuba. 50 And it was helpful to the avoidance of fear

of preemption that the U.S. alerting of its nuclear forces, including at

least one alert broadcast en clair, was not matched by as boisterous a

military statement from Moscow. 51

Related to the third factor in avoiding preemption, careful alert management,

is the fourth, timely and accurate communication of force

status. Leaders must know the status of their own forces and must correctly

communicate this to the adversary (unless, of course, they are deliberately

attempting to conceal force status because they are actually

planning to attack). For example, during the Soviet-orchestrated invasion

of Czechoslovakia in 1968 by forces of the Warsaw Pact, NATO was

careful to avoid provocative overflights or troop movements that could

be misunderstood as a responsive military intervention. A disquieting

example during the Cuban missile crisis was the apparent firing of a test

ICBM from Vandenberg Air Force base over the Pacific. Although this

missile was not weaponized, Soviet observers could be forgiven if they

had assumed otherwise in the middle of the worst crisis of the Cold

War. Fortunately, Soviet air defense or other watchers never detected the

launch or, if they did, chose not to make an issue of it. 52 One can, on the

other hand, overdo the issue of making clear to the opponent the preparedness

of one’s forces. The U.S. SAC commander who broadcast alerting

orders in the clear during the Cuban missile crisis, doubtless to

impress his Soviet interlocuters, was trying too hard. Such braggadocio

was unintended by his superiors and, under more trying circumstances,

could only have contributed to Soviet fears of attack (or of an out-ofcontrol

U.S. commander with nuclear weapons, thus overlapping with

the problem of accidental/inadvertent war).

Nuclear Weapons and ‘‘Third Wave’’ Warfare 147

A fifth requirement for the avoidance of preemption, the availability

of flexible options, might also be put at risk by information warfare.

Logic bombs, worms or viruses that attack information might also deny

policy makers and high commanders of the other side an accurate reading

on their own command system and forces. It was hard enough to

convince Cold War U.S. leaders, on the evidence, that they had options

other than those of massive nuclear retaliation. SIOP planners, convinced

that nuclear flexibility was the road to defeat in war by overcomplicating

the command system, had little or no interest in preparing mini-nuclear

strike packages. As a result, in an actual crisis, the president might have

wanted smaller or more selective options than actually existed on the

available menu. Former Secretary of Defense Robert McNamara recalled,

with reference to the Cuban missile crisis, that he considered the SIOP

options then available irrelevant to the crisis. 53 This is quite a statement,

considering that McNamara had just overseen a major overhaul of nuclear

war plans with the very object of introducing selective nuclear options.

The issue of flexibility can be misconstrued and that is to be avoided

in this context. Flexibility is a two-edged sword. Nuclear flexibility does

not imply that any nuclear war would or could be waged at an acceptable

cost. It means, to the contrary, that however terrible a smaller nuclear

war would be, a larger one would be that much worse. This issue

should not sidetrack us. The present point is that an infowarrior could

‘‘persuade’’ his opponent that the opponent’s information system is full

of electronic junk or that his command/control system for nuclear response

is about to die off. So persuaded, preemption for fear of death

could appeal to leaders even before the other side had made its own

irrevocable decision for war.

Preemption for want of information on account of cyber-distortion,

intended by the other side as intimidation, is a possible path to war in

an age of information complexity. Even the Cuban missile crisis, taking

place in an environment of comparative information simplicity, was dangerous

enough in this regard. For example, U.S. leaders at one stage

wondered whether Khrushchev was still in control in Moscow. 54 Fear

existed that he had been overruled by a more hard-line faction within

the Politburo, after a relatively conciliatory message from Khrushchev

on Friday, October 26 was followed by another message from the Soviet

premier the next day, harsher in tone and more demanding in substance.

In Moscow, some Soviet leaders feared that Kennedy was a virtual prisoner

of hawkish forces that might overthrow his regime unless he acquiesced

to an air strike or invasion of Cuba. Fortunately, these mistaken

images were not compounded by infowarriors using perceptions management

to make them more convincing.

148 Nuclear Policy and Strategy in the Future


Preemption, however mistaken, is a deliberate choice made by desperate

policy makers. Accidental/inadvertent nuclear war is the result

of a process that nobody intended. Since war grows out of a process, it

may erupt in ways that were not anticipated by policy makers or planners.

The following discussion outlines some of the requirements of

avoiding accidental/inadvertent war and explains how infowar might

lead actors to fall short of meeting those requirements. There is, admittedly,

some overlap between issues and factors that relate to preemption

and crisis management, on one hand, and those involved in a discussion

of accidental/inadvertent war. The overlap is only partly the result of

the approach taken by this investigator: nuclear weapons and the speed

with which they can inflict unprecedented destruction create overlap

among the ‘‘real world’’ causes of nuclear-related policy dysfunction.

The first requirement for the avoidance of accidental/inadvertent war

is the ability to balance the requirements for positive and negative control.

This can be discussed as two separate requirements, although they

are intimately related. Negative control is the prevention of any nuclear

release or launch except by duly authorized command. Positive control

means that forces are promptly and reliably responsive to authorized

commands. In some discussions this is characterized as the ‘‘always–

never’’ problem: always ensuring that forces are responsive when required,

but never permitting actions that are unintended by political

and/or military leaders. 55 Confusion can be introduced into the

discussion by the term ‘‘positive control launch’’ which is actually a form

of negative control, as described here. Positive control launch, as in SAC

Cold War ‘‘fail-safe’’ procedures, restrains an attack from taking place

unless a specific coded message authorizes the attack even after bomber

aircraft have been loaded with weapons and routed to preliminary airborne


There is inherent tension between the requirements for negative and

positive control at the margin, especially during the alerting of military

forces. 56 Nuclear weapons make this tension acute. Steps taken to make

forces ready for prompt retaliation after enemy attack can remove some

of the controls against accidental or unauthorized use. More significant

than changes in hardware are changes in the expectations of the people

who operate forces and their associated command and control systems.

As a confrontation between two states looms, military operators will shift

their expectations from latent to manifest awareness of the worst possible

performance of the system for which they are responsible. Forces will be

exercised to guard against worst-case possibilities. However, too much

alerting can wear out forces and reduce their actual readiness for war.

Nuclear Weapons and ‘‘Third Wave’’ Warfare 149

Forces and command systems can only be maintained at high alert levels

for a very short time before performance degradation sets in.

If alerts are extended beyond the ‘‘knee of the curve’’ for ready forces,

then deteriorated performance, including performance capable of causing

accidents, is a possible outcome. People get tired, machinery wears,

and nerves are frayed. Stress levels rise. Interpretations of events are

influenced by the strained condition of physical and human systems.

Introduce the additional complexity of an attack by one side on the information

bytes of the other, exacerbating stress and tension. One possible

result of stress compounded by misinformation is the tendency to

rationalize or falsely explain whatever action was recently taken. Another

possible result is cognitive or motivational bias in assessments of

the status of forces and command/control systems, including information

networks. 57 A cognitive bias is a bias in the logical explanation for

an event or forecast of a future event. Motivational bias is based on

emotional needs of the observer that distort his or her perception of what

is being observed.

First strike fears compounded by information malaise and motivational

or cognitive bias can also contribute to an outbreak of accidental/

inadvertent war. If leaders believe that the opponent will attack at their

weakest moment of preparedness, and if they are further persuaded by

infowar that their command system is becoming a sewer pipe of enemy

disinformation, they may shut down channels or networks that maintain

negative control under stressful conditions. Consider, for example, the

problem of crisis-time communication with ballistic missile submarines.

The possible disruption of these communications in time of war, and the

equally strong possibility of prompt enemy attacks against the other

side’s SSBN force, led Cold War U.S. policy makers to enable submarines

to launch their weapons under approved ‘‘fail deadly’’ procedures. This

meant that, faced with disrupted communications and under presumed

attack from enemy attack submarines, U.S. ballistic missile submarines

(under certain restrictive guidelines) might fire at predesignated targets.

PALs (permissive action links, essentially electronic locks that could only

be bypassed by encoded messages from the U.S. National Command

Authority) were not installed on U.S. SSBNs during the Cold War, although

they were in place on land-based and air-delivered weapons. 58

The U.S. Navy argued that the environment in which maritime operations

were carried out precluded PALs or other devices that depended

upon the fidelity of shore-to-ship communications in wartime. In addition,

the Navy contended that procedural safeguards against accidental/

inadvertent war were more important than mechanical or electronic

locks. Navy training and tradition were the guarantors against nuclear

usurpation. 59 This argument was not entirely self-serving. In the largest

sense, the entire U.S. government depends upon the training and tradi-

150 Nuclear Policy and Strategy in the Future

tion of its military as the fundamental guarantees of civil supremacy in

time of peace or war. The modern military have always had the physical

power or capability to overthrow the government (or, at least, to overthrow

the civilian leadership in Washington, D.C., temporarily), but

never the inclination since the end of the Civil War.

The problem of accidental/inadvertent escalation or war is more complicated

than simple military overthrow of civil power. Slippage of negative

control in the direction of accidental or inadvertent war or

escalation can occur in stages and without any lapse of military loyalty

to civil authority. Disjunction between the intent of political leaders and

military operators can take place when commanders are carrying out

logical procedures under unusual, but possible, conditions. 60 Consider

the management of U.S. naval forces during the Cuban missile crisis.

Standard operating procedure called for the U.S. Navy to force any Soviet

submarines within the quarantine line to the surface. Commanders

proceeded quite logically from this standpoint to do exactly that. It was

not fully appreciated by policy makers or by Navy commanders that this

could lead to inadvertent escalation and war. 61 Soviet submarines signalled

by depth charges might respond not by surfacing as required but

by attacking U.S. vessels. At one point in the crisis this was perceived

by President Kennedy, who reportedly exclaimed ‘‘almost anything but

that’’ when the possibility of military clashes between U.S. forces and

Soviet submarines was mentioned. 62

A second requirement for the avoidance of accidental/inadvertent war

is the validity of warning and attack assessment. Leaders must have

confidence that they can distinguish between false and true warning of

attack. They must also expect, once have received valid warning of attack,

that they will have time to respond appropriately. U.S. nuclear

warning and attack assessment evolved during the Cold War into a

tightly coupled system of warning sensors, analysis and fusion centers,

communications links, commanders and command posts. The nerve center

of U.S. Cold War warning and assessment was NORAD, located in

an underground and hardened shelter complex at Cheyenne Mountain,

Colorado. NORAD even after the Cold War is the chef d’oeuvre of the

elaborate U.S. warning system for surprise attack.

The problem of warning is related to the timing and character of response.

Although warning and response can be separated for purposes

of analysis, operational warning and response are closely related. The

development of long-range, nuclear-tipped missiles required rethinking

of many of the basic premises of warning and response on the part of

U.S. officials. The time between launch of Soviet ICBMs and their detonations

on North American soil would be twenty minutes or less;

submarine-launched missiles might arrive even sooner, depending upon

their assigned targets and launch positions. Warning had therefore to be

Nuclear Weapons and ‘‘Third Wave’’ Warfare 151

automated to some extent, but the ‘‘person in the loop’’ could not be

extracted without control passing to a machine. In addition, with such

short time lines for warning and response, the initial warning decision

and the responsive forces should be assigned to different commands.

Thus, for example, the United States assigned to NORAD the strategic

warning function, and to SAC and the Navy the responsive forces and

weapons. The function of NORAD was to establish the plausibility of

warning within certain parameters and to initiate a series of conferences

among political and military leaders.

The problem of mistaken warning and retaliation based on that mistake

was taken very seriously by U.S., and presumably by Soviet, Cold

War military and political leaders. The compensatory approach to possibly

mistaken warning chosen by the United States was phenomenal

redundancy, or ‘‘dual phenomenology.’’ This locution meant that indicators

of a possible attack would have to be confirmed from more than

a single source of input data. For example, U.S. early warning (DSP)

satellites would first detect the exhaust plume of land- or sea-based missiles

rising from their Soviet points of origin. Minutes later, ballistic missile

early warning radars (BMEWS) at Fylingdales Moor, England; Thule,

Greenland; and Clear, Alaska would confirm the initial observations by

satellite and provide additional details about the size and character of

the attack. The United States also deployed specialized radars for the

detection of submarine-launched ballistic missiles, presumably off the

Atlantic and Pacific coasts of the continental United States. The presumption

of phenomenal redundancy was that even if a single part of

the system gave wrong indications or was out of operation at a particular

time, the remainder of the system could effectively confirm, or disconfirm,

that an attack was under way. See Table 6.2 for a summary of kinds

of sensors and their targets for warning of attack.

Warning having been confirmed by more than one source, the problem

of assessment remained. What kind of attack was in progress: was it a

massive surprise strike against a comprehensive target set, or was it a

‘‘limited’’ strike intended for coercive purposes with follow-up attacks

held in reserve? U.S. and Soviet leaders in the Cold War also would have

been concerned about two possible kinds of errors in attack assessment.

A Type I error results in delayed or flawed launch in response to actual

attack. A Type II error results in premature or mistaken launch although

no actual attack is in progress. 63 The significant thing about Type I and

Type II error probabilities is that they are relational. Steps taken to minimize

the likelihood of one kind of error often increase the probability

of the other type of mistaken decision. For example, building in more

elaborate and redundant checks against false warning of attack (Type I)

may slow down the decision-making process, thereby increasing the

152 Nuclear Policy and Strategy in the Future

Table 6.2

Targets and Sensors for Attack Warning

Source: Adapted from John C. Toomay, ‘‘Warning and Assessment Centers,’’ in Ashton B.

Carter, John D. Steinbruner and Charles A. Zraket, eds., Managing Nuclear Operations

(Washington, D.C.: Brookings Institution, 1987), p. 313.

chance of a reaction too late for retaliatory forces to carry out their assigned


The problem of valid warning and appropriate response is complicated

by the tight coupling of sensors, assessment centers and response

system. Certain high-technology organizations are especially prone, according

to sociologist Charles Perrow, to ‘‘normal accidents.’’ 64 So-called

normal accidents occur in these kinds of organizations when individual

component failures cause other components also to fail, but in unex-

Nuclear Weapons and ‘‘Third Wave’’ Warfare 153

pected ways. The result is a systemic dysfunction not anticipated by the

designers of the system. Accident-prone high-technology organizations,

according to Perrow, share two attributes: interactive complexity and

tight coupling. Interactive complexity of these organizations increases the

frequency of unexpected or seemingly anomalous interactions among the

parts, including the human parts. Tight coupling implies: (1) parts of the

system interact quickly, (2) sequences of activity are invariant—there is

only one right way to do things, (3) little organizational ‘‘slack’’ is available

to compensate for error, (4) safety devices, including redundant

checks and balances against failure, are limited to those planned for, and

designed into, the system. 65 The Three Mile Island nuclear power disaster,

airline crashes, space shuttle malfunctions, electric power grid

brownouts, and other ‘‘normal accidents’’ may be inevitable in organizations

dependent upon advanced technology and highly interactive

parts. The Cold War U.S. nuclear warning, assessment and response systems

are also examples of high-risk, high-technology systems prone to

normal accidents. Yet, despite this documented failure propensity for

these kinds of organizations, normal accidents in the U.S. nuclear command

and control system were apparently rare, and obviously none were

catastrophic. 66

Three possible explanations exist for the avoidance of catastrophic failure

in nuclear warning and attack assessment during the Cold War. First,

the situation was overdetermined. Even the most dullard national political

leadership would search for any way out before authorizing a nuclear

attack, or a retaliation based on warning of attack. Second,

redundancy built into the system ensured against any retaliation based

on mistaken warning. Indeed, numerous false warnings became routine

business during the Cold War for both U.S. and Soviet organizations

charged with warning. Some might argue that the two sides became so

habituated to false warning and the low likelihood of actual attack that

they dropped their guards. This charge was laid by Gorbachev against

the Soviet military after an intrepid West German pilot flew his Cessna

aircraft into Red Square, through and under Soviet radar nets and interceptor

squadrons. U.S. military leaders were not lulled into complacency

about the possibility of accidental nuclear war. But evidence suggests

that their confidence in the ability of the decision-making process to compensate

for errors in warning and attack assessment, possibly contributory

to accidental nuclear war, was excessive. 67

A third possible explanation is that the United States and the Soviets

got lucky. No lethal combination of mistaken warning and a nuclear crisis

occurred simultaneously. The interactive complexity of both sides’ warning

systems was never fully road tested under Gran Prix conditions. Paul

Bracken makes a related and significant point with regard to the judgment

here. The interaction between the U.S. warning and assessment

154 Nuclear Policy and Strategy in the Future

system and its Soviet counterpart became a separate and dangerous part

of the Cold War nuclear complex; that is, the cat watched the rat, and

the rat watched the cat. All of this took place in very quick time. The

potential for the two sides’ interactively complex systems to set one another

off, like new-wave smoke detectors that reacted not only to smoke

but also to the alarm of another detector, was considerable.

Thus far we have established two requirements for the avoidance of

accidental or inadvertent nuclear war or escalation: managing the tradeoff

between positive and negative control, and coping with the complexity

of nuclear warning and attack assessment. A third attribute necessary

for the avoidance of accidental/inadvertent war is solution of the

problem of delegation of authority and, related to that, the issue of devolution

of command.

The president, as head of the executive branch of the U.S. government,

may delegate any of his or her responsibilities almost without limit. The

U.S. Constitution and the Presidential Succession Act of 1947 have established

a constitutional order of succession in case of presidential

death, disability or removal from office. The vice president stands at the

head of this order of succession, followed by the Speaker of the House

of Representatives and the president pro tempore of the Senate. After

that, the heads of cabinet departments are enumerated from the oldest

department to the youngest: State, Treasury, Defense (War originally),

and so forth. The president is the only lawful person who can authorize

nuclear release to force commanders and retaliatory attacks using nuclear

weapons. However, the president is not a singular actor in this

regard: he or she requires the cooperation of various other levels of

command and responsibility. This need for more than a singular center

of competency for nuclear decision, in case of presidential death or disability,

overlaps the military chain of command with the civilian chain of

succession in nuclear matters.

The death of a president or even the entire civilian political leadership

in a surprise strike on Washington, D.C., cannot be permitted to paralyze

U.S. retaliation, for the obvious reason that such vulnerability might invite

attack. Therefore, predelegated or devolved command arrangements

for nuclear authorization (or authentication) and enablement must be possible

from other sources in extremis. Authorization means that the person

or office conveying a command to a subordinate unit is lawfully

entitled to do so. Enablement provides the necessary release mechanisms

that allow operators to circumvent mechanical or electronic locks assigned

to weapons in peacetime to prevent accidental or unauthorized

use. 68 For example, an authorization code from the ‘‘football’’ or suitcase

carried by a presidential aide tells the receiver that the president has,

indeed, sent the indicated command. Enabling commands include the

Nuclear Weapons and ‘‘Third Wave’’ Warfare 155

‘‘unlock’’ codes to bypass electronic locks such as PALs or other use

control devices.

Devolution is a complicated matter because, at least in the U.S. Cold

War case, authorization and enablement codes were usually not held by

the same persons or at the same levels of command. The president and

the secretary of defense, for example, can authorize nuclear release, but

they have neither physical possession nor effective custody of nuclear

weapons. Possession is related to custody, but not necessarily identical

to it. Custody implies control over the weapon, whether in one’s possession

or not. For example, tactical nuclear weapons may be deployed

in storage sites guarded by personnel other than those who would actually

use those weapons once authorized and enabled to do so. In theory,

authorization and enablement commands would be unambiguous

and those lower on the chain of command should automatically carry

out orders from higher echelons. The matter is not so simple in practice:

human operators, not automatons, are in the chain of command. And

actual custody or physical possession of the weapons is the responsibility

of lower echelons who are, in any military situation, capable of resistance

of various kinds in response to orders. Even nuclear-age Willie and Joe

have some discretion and may decide, mutatis mutandis, to think for

themselves, once nuclear charges begin to move from storage sites to

launch platforms. 69

As orders cascade downward and outward through bureaucratic organizations,

honest misconstructions and organizational self-interest are

inevitable concomitants of crisis-time mobilization. Organizations will do

what they have been prepared to do; this normative model dictates that

nuclear force commanders will follow orders to retaliate once given without

even thinking about what they are doing. This model works for a

situation of unambiguous clarity: warheads have begun detonating on

U.S. soil, the president has identified the transgressor and authorized

retaliation, and public support for nuclear response can be assumed.

Prior to the actual arrival of nuclear destruction on American targets,

the expectations become more confused, within and among bureaucratic

compartments. One can imagine some elements taking the ‘‘wait and

see’’ position given the Cold War history of false alarms. Other elements

might delay response and demand additional authentication for so significant

an order. Psychological paralysis on the part of some persons in

the face of orders to unleash nuclear death and destruction, knowing

that their own kith and kin were at risk from enemy strikes of a similar

kind, would be almost predictable, and entirely human. Table 6.3 depicts

various loci in the U.S. nuclear command and control system by function

and tabulates their respective kinds of influence.

The U.S. nuclear command system works, according to one expert analyst,

like a revolver. The function of the presidential center is ‘‘not to

156 Nuclear Policy and Strategy in the Future

Table 6.3

U.S. Nuclear Weapons: Command Functions and Capabilities

1. Some levels of command may hold authorization codes as well, for example, commanders-in-chief

of unified and specified commands.

2. Some levels of command may hold enablement codes as well, for example, commandersin-chief

of unified and specified commands.

Source: Adapted from Peter Douglas Feaver, Guarding the Guardians: Civilian Control of Nuclear

Weapons in the United States (Ithaca, N.Y.: Cornell University Press, 1992), p. 40.

act as a trigger to launch nuclear weapons, but as a safety catch preventing

other triggers from firing.’’ 70 In time of peace and relaxed tension, the trigger

safety is ‘‘on’’ so that the weapon cannot be fired: negative control

reigns supreme. As a crisis develops the controls are progressively relaxed

to permit faster reaction to emergency: positive control becomes

more important. This process becomes apparent in the U.S. DefCon gradations

for management of alerts from level five (lowest, peacetime conditions)

to level one (highest, ready for imminent war). As policy makers

Nuclear Weapons and ‘‘Third Wave’’ Warfare 157

authorize the military to proceed from lower to higher levels of alert,

restraints on force movement and preparedness are ‘‘unwrapped’’ and

the criterion of readiness for combat takes precedence relative to peacetime

safety and security. Alert management is a tricky business, and the

United States, with as much experience as anybody, had its own share

of snafus despite high levels of military personnel reliability and numerous

checks and balances in the system.

Each of the requirements for the avoidance of accidental/inadvertent

war (balancing positive and negative control, making an action upon

valid warning and attack assessment, and maintaining authoritative and

responsive delegation of authority and/or devolution of command) is

potentially at risk from information warfare.

First, the balance between positive and negative control becomes a

more complicated juggling act as alert levels are raised. Some components

of the force, say ICBMs, are permanently at high levels of readiness

for prompt launch. Others, such as the bomber force, require a great deal

of care and feeding under stressful conditions before they are launch

ready. Nuclear-armed, sea-based cruise and ballistic missiles can be readied

to fire in a short time provided that the submarine is on station; but

some submarines may need to proceed from port to station, and others

may be moved in connection with targeting requirements or possible

threats to their survivability. Elements of the command system also require

synchronized movement across disparate services and civilian departments.

If NATO alerts are involved in addition to U.S. forces, as they

would have been during any Cold War confrontation with the Soviets,

the management of alert phasing and timing becomes even more complicated.

Infoweapons introduced into this alerting process have the potential

to disestablish the desirable balance between positive and negative control

as forces are gradually empowered to go to war. From an enemy

perspective, this might be considered a good thing: confuse the American

alerting process and make the wartime command system only partly

ready for battle. That is conventional, not nuclear, logic. In a nuclear

crisis, the two sides have a shared interest in avoiding nuclear war as

well as a competitive desire to prevent one another’s gains. Accordingly,

each will want the other to maintain assured control over the balance

between unlocking the cocked pistol for retaliation and preserving control

over the military movements and actions that might trigger inadvertent

war. And those military movements and actions are dangerous

precisely because their inherent danger might not be so obvious. As Thomas

Schelling has noted, war can begin not as a deliberate decision by

policy makers, but as the result of a process over which neither side has

full control. The possible loss of control to be feared here is not military

usurpation of civil authority or military disregard of authorized com-

158 Nuclear Policy and Strategy in the Future

mands. Instead, it is a lack of correct foresight that results in a sequence

of events foreseen by neither side, creating a new and more adverse climate

of expectations about future behavior. 71

If the two sides in a nuclear crisis get into a sequence of events not

correctly foreseen by either and seek to interpret those events correctly,

information warfare will be harmful, not helpful, to correct interpretation.

An example is perceptions management by one side designed to

suggest to the media, public and legislature of the other side that the

first side’s intentions are only honorable. 72 The second side, according to

this carefully orchestrated set of perceptions fed from one side to another,

is really the ‘‘aggressor’’ or the ‘‘uncooperative’’ partner. And the

media and political elites of the second side might believe the image

created by enemy perceptions management, calling upon leaders to

stand down forces and accept the demands of the opponent. Or, leaders

of the second side might be outraged at the cyber-propaganda of the first

side, escalate their demands, and become more intransigent. As Robert

Jervis has noted:

A state tends to see the behavior of others as more planned, coordinated, and

centralized than it is. Actions that are accidental or the product of different parts

of the bureaucracy following their own policies are likely to be perceived as part

of a coherent, and often devious, plan. In a nuclear crisis, the propensity to see

all of the other side’s behavior as part of a plan would be especially likely to

yield incorrect and dangerous conclusions. 73

Messages may be received as intended, but their political effect is not

necessarily predictable. Hitler’s propaganda efforts to dissuade British

and French military reactions to his occupation of Czechoslovakia

worked as intended in the short run, but established in the Nazi dictator’s

mind misleading impressions of the two states’ weakness in resisting

any further tearing up of the map of Europe. Boomeranged

perceptions management appeared in Soviet Cold War forgeries of diplomatic

and military communications between the United States and its

allies. Most of these were transparent, and once found out, gave additional

evidence of Soviet diplomatic perfidy to hawkish U.S. politicians

and commentators.

Forces poised for immediate retaliation to possible surprise attack also

interact with warning systems looking outward for indications of such

an attack. False indicators planted by the other side’s infowarriors to

bring down air defense systems and missile attack warning radars are

harbingers of disaster once forces and command systems have been

alerted to war-expectant levels. Under those conditions, the vanishing of

information from radar screens or fusion centers could be assumed as

the first infowave of a nuclear attack, calling forth a preemptive response.

Nuclear Weapons and ‘‘Third Wave’’ Warfare 159

Blank screens and obstructed communications leaving personnel in the

dark for orders helped the United States in Desert Storm to clear the way

for its air strikes against targets in Baghdad and at other locations within

Iraqi state territory. The same phenomena might create unacceptable

panic if the stakes were vulnerability to nuclear instead of conventional

attack. Terrible pressures would rise from lower to higher command levels

to ‘‘use them or lose them’’ before communications between and

among NCA, CINCs and geographically dispersed force commanders

were completely severed. The U.S. president might be loath to order into

effect any retaliation under these or similar circumstances, but this legal

consolation might not make much practical difference.

High-altitude detonation of a small nuclear weapon by an enemy over

its own territory and causing no prompt fatalities to U.S. or allied military

personnel could nevertheless destabilize U.S. ground and space assets

and confuse decisions based on those assets. For example, a nuclear explosion

of about 50 kilotons above 100 kilometers in altitude would disrupt

space assets by prompt radiation effects and by creating

atmospheric disturbances. Unhardened satellites in low earth orbit

(LEO), even after they survived the initial radiation, would be vulnerable

to total dose destruction as they passed through aggravated radiation

belts. Within weeks to months, all unhardened LEO satellites within a

theater of operations could be degraded or inoperative. For example, the

electromagnetic pulse (EMP) from a single nuclear burst at an altitude

as low as 40 kilometers has the potential to weaken electronics across

the entire Persian Gulf region, or throughout the Korean peninsula. Ironically,

today’s more sophisticated microelectronic systems are more vulnerable

to radiation than the cruder and simpler (less integration density,

slower operating speed) electronic systems of yesteryear. 74

In theory, no president would authorize the firing of nuclear forces on

the basis of ambiguous warning information. But in the exigent circumstances,

while clarification of the status of schizoid information systems

was sought, force commanders would not be idle. At DefCon 3 or higher

they would be taking appropriate measures to protect their divisions,

fleets and air wings from enemy surprise. These moves would almost

certainly be noticed by the other side’s intelligence and warning, even if

no U.S. or allied nuclear weapons were released, moved or loaded. Few

if any military experts ever thought that a war would begin with a nuclear

‘‘bolt from the blue,’’ absent preceding events. By definition, ‘‘bolt

from the blue’’ is not accidental or inadvertent. Any accidental/inadvertent

nuclear war in the years between 1946 and 1990 would have almost

certainly begun at the level of foot patrols wandering off the map into

restricted areas, maritime collisions on the heels of U.S.-Soviet ‘‘dodgeem’’

games at sea, or inadvertent strays by one side into restricted air

space of the other. 75 Confrontations at the tip of the lance have an entirely

160 Nuclear Policy and Strategy in the Future

different meaning higher up the chain of command if links in the chain

have been deliberately confused, or their interlink communications distorted,

by cyber-pathological strikes. As Bruce G. Blair has noted:

Central authorities cannot reasonably expect military organizations simply to

carry out orders, however rational the orders may be. Organizations are not that

pliant. Any attempt to assert positive or negative control in a way that requires

major abrupt changes in operating procedures, a situation more likely to occur

if operating routines escape attention in peacetime, would invite confusion and

disorder. 76

Consider, for example, the impression created by Presidents Clinton

and Yeltsin, pursuant to their agreement of 1994, that U.S. and Russian

strategic nuclear missiles are no longer aimed at targets on one another’s

state territory. The agreement may have symbolic value, but it did little

to change essential procedures for missile targeting. Although Russian

missiles are supposedly set on ‘‘zero flight plan,’’ the missiles’ memory

banks still store wartime targets. The Russian General Staff can, from

command posts in Moscow and elsewhere, override the de-targeting by

means of a Signal-A computer network, reaiming silo-based missiles at

U.S. targets within ten seconds. 77 U.S. missiles can be retargeted against

Russia just as rapidly.

In addition, Russian operational practice permits its command system

about three or four minutes for detection of an attack, and another three

or four minutes for high-level decision making. Russia’s command system

and early warning network have obviously deteriorated since the

end of the Cold War, along with the rest of Russia’s military, raising the

potential for accidents or inadvertency. For example, in January 1995, the

firing of a Norwegian scientific rocket triggered the first-ever Russian

strategic alert of their prompt launch forces, an emergency conference

including the Russian president and other principal leaders and activation

of the Russian equivalent to the U.S. nuclear ‘‘football’’ or suitcases

assigned to national command authorities. 78 According to U.S. press reports

in April 1997, Russia’s present-day nuclear command and control

system was in danger of breaking down. A retired Strategic Rocket

Forces officer charged that at least one recent system malfunction caused

parts of the nuclear arsenal to go spontaneously into ‘‘combat mode.’’ 79

Finally, on the problem of infowar and the avoidance of accidental or

inadvertent war or escalation, the problem of delegation of authority or

devolution of command is exacerbated by information distortion in the

command system of either side. In the case of the U.S. nuclear command

system, for example, the ‘‘cocked pistol’’ analogy implies that the pistol

will fire back at the highest levels of alert unless authorized commands

hold back retaliation. 80 An enemy seeking to paralyze U.S. retaliation by

Nuclear Weapons and ‘‘Third Wave’’ Warfare 161

confusing the information networks for nuclear command would, instead,

have an equal chance of cutting the command and control system

into pieces. Each piece and the forces with which it could connect would

then be a sovereign, post-attack entity, firing back with whatever resources

remained to it at targets of opportunity. Commanders unable to

communicate with one another might assume that their counterparts

were already dead or hopelessly cut off from proper orders. A dumb,

blinded and disaggregated nuclear response system would then take

over from what had been a singular chain of nuclear command and


The preceding point applies both to accidental/inadvertent war or escalation

and to conflict termination, explored in the next section. Stylized

descriptions of nuclear warfare in the Cold War literature followed predictable

scripts. One side’s first strike subtracted weapons from the second

side’s inventory; the second strike fired back with its survivors; the

first side struck again with its remaining forces, and so forth. These stylistic

descriptions of force exchanges overlooked the differences among

peacetime, crisis and wartime assessments. Assessment is a qualitative

process, dependent upon accurate and timely flow of information. It demands

both central coordination, so that all arms of the military machine

are operating according to a shared concept of alertness or battle, and

decentralized operation or execution of orders. The loss of information

necessary for crisis or wartime assessment may not take place in discrete

and measurable stages, but in weird nonlinear ways. Accordingly, leaders

in a crisis or early wartime situation literally may not know what it is

that they do not know. 81

An important aspect of the difference between conventional and nuclear

deterrence is that the strategic problematique is so much more linear

in the conventional case. The object of the military in a conventional

war is to obtain dominant battlespace awareness and to deny the opponent

the opportunity to use cyberspace effectively. 82 On the other

hand, the problem of avoiding accidental/inadvertent nuclear war is

more nonlinear, perhaps even chaotic, with regard to the information

spectrum. 83 Exploitation of information, including electronic, is both a

cooperative and a competitive activity when the avoidance of nuclear

war is equally as significant as the accomplishment of one’s military

object, should war occur. One side does not necessarily want to drive

the other side’s information regime off the field, or to corrupt it into a

Mad Hatter’s Tea Party of misinformation. Further, the distinction between

‘‘offensive’’ and ‘‘defensive’’ infowar may be more muddled in

the case of nuclear deterrence and the avoidance of accidental war or

escalation compared to the more straightforward task of prevailing in

war, should it occur.

162 Nuclear Policy and Strategy in the Future


Conflict termination as used here refers to war termination: hostilities

have already begun, and a point is reached at which at least one side

prefers to cease firing. Conflict or war termination is difficult enough to

accomplish in a conventional war, especially one between great powers

that are more or less equivalent in military capability. A nuclear war

would obviously differ from a non-nuclear conflict with respect to war

termination, although it is not obvious how. 84 The destruction attendant

to a small number of nuclear explosions, especially against heavily populated

areas, would be total or nearly so for small powers. For large

countries with diverse and survivable arsenals like the Cold War Americans

and Soviets, the choice of war termination might present itself after

the initial salvos. Although historically unprecedented damage would

have been done, much additional damage could be prevented if leaders

came to their senses in time.

War termination was the Achilles heel of U.S. and arguably Soviet

nuclear strategy in the Cold War, for different reasons. Much U.S. nuclear

strategic planning was oriented to the worst-case possibility, a ‘‘bolt

from the blue’’ against the military forces, war supporting industry and

economic infrastructure of North America. Although planners acknowledged

that other kinds of attacks were more probable, this extreme condition

drove much policy, force planning and targeting. 85 Soviet

planners, although not unmindful of the possibility of a massive U.S.

surprise attack, did not consider that this was the main, or most probable,

contingency against which they had to prepare. 86 Instead, Soviet

expectations were for an outbreak of conventional war in Europe pitting

NATO against Soviet and allied Warsaw Pact forces. This war would

begin conventionally but in all likelihood would escalate rather quickly

to a nuclear conflict, first in the Western TVD (theater of military action)

and then globally. Soviet military strategists averred that it was NATO’s

dependency on forward deployed, tactical nuclear weapons that would

cause prompt escalation from conventional to nuclear warfare, and not

a Soviet preference for that means of fighting a theater military campaign

in Europe. 87

Thus the political backdrop for war termination was somewhat different,

in terms of state policy objectives, as between the Soviet Union and

the United States. 88 Notable policy differences also existed within the

NATO alliance. The United States preferred a nuanced strategy of flexible

nuclear response even after the firebreak between conventional and

nuclear war had been crossed. Many Europeans argued that the idea of

any firebreak, between large scale conventional war on the Eurasian

mainland and global war against the Soviet Union, would weaken deterrence.

Many Europeans also felt unwilling or unable to pay for the

Nuclear Weapons and ‘‘Third Wave’’ Warfare 163

improved conventional defenses that might make it possible to prolong

the conventional phase of a war before the decision for nuclear first use

had to be taken. 89 France departed from the NATO military command

structure in 1966 and declared that the decision to use French nuclear

weapons would be made independently of the alliance and only in

French national interest.

Given this political background, bringing any war to a conclusion

would have challenged alliance unity (in the West) and defied the suspicious

mind-sets that had grown up between Cold War antagonists.

Notwithstanding these difficulties, let us review some of the requirements

for nuclear war termination and consider how information warfare

might make it more difficult to meet those requirements. These are

accurate and timely communication of a willingness to cease fire, military

compliance with settlement, stable command and control during war,

and verification of behavior called for in armistice. In any real situation

of war termination, these processes would be commingled, and the ensuing

discussion reflects the fact.

The first requirement for war termination is the accurate and timely

communication of an interest in cease-fire. Unless and until most forces

in any war can be made to cease fire, policy makers’ desire to stop a war

is more theory than reality. In a conventional war fought mostly or entirely

between professional armed forces accountable to civil authorities,

this should be a comparatively simple matter of communication between

heads of state. Even in this case, however, there are many variables to

consider. Armies may or may not be promptly responsive to directives

from their superiors to stop shooting. Disengagement of forces and virtual

disarmament of troops on both sides may be necessary before serious

negotiations over policy issues can begin. The problem is

exacerbated if, within the country of one or more combatants, there exist

factions opposed to any peace settlement short of victory or death. For

example, in World War II a die-hard faction in the Imperial Japanese

military leadership resisted surrender even after Hiroshima and Nagasaki.

Germany’s surrender in World War II was made more difficult

because the ‘‘die-hard’’ was Hitler himself, who preferred to see his

country destroyed along with his own life rather than agree to terms. 90

The problem of communicating an intent to cease fire and a willingness

by the sides to disarm recalcitrant troops and noncompliant field

commanders is exacerbated if the conflict is an unconventional one. In

revolutionary or guerrilla warfare (not identical, but similar in their dissimilarity

from conventional war) or in terrorism, obtaining the proper

‘‘address’’ to which to send communications can be difficult. Guerrilla

and revolutionary leaders not infrequently have difficulty enforcing labor

discipline over their forces, and for understandable reasons. 91 The

essence of unconventional warfare includes reliance on intelligence and

164 Nuclear Policy and Strategy in the Future

combat support from paramilitary elements that may have few, if any,

of the rudiments of professional soldiering. They are civilians armed for

a cause, and once the cause has taken on a momentum of its own, military

discipline may not be reimposed without excessive costs. In Bosnia

between 1992 and 1995 the difficulty of obtaining cease-fire and disarming

regulars, partisans, ruffians and renegades finally called forth a major

NATO military intervention to impose at least a temporary peace.

A nuclear conflict between major powers would have aspects of both

conventional and unconventional warfare. It would in all likelihood be

fought between states, with militaries supposedly under the operational

control of duly constituted civil authorities. However, once a nuclear war

had begun the empowered nuclear commanders have been converted

into potential nuclear ministates. This is especially so in the case of nuclear

force commanders in a state without a democratic tradition of political

accountability of the armed forces. More will be said about the

problem of politico-military control below, but the present point is about

the ability to communicate a timely and accurate sense of one side’s

willingness to stop fighting. If there are plural instead of singular effective

fighting forces in the field with whom to communicate, cease-fire

becomes harder to obtain. Russia’s frustration in attempting to negotiate

a cease-fire with rebels in Chechnya, and not only in this century, makes

the point.

Therefore, a second requirement for war termination is military compliance

with the terms of armistice and, ultimately, the peace settlement.

If disarmament and peace can be imposed upon the losers by the victors,

as in World War II, then the issue of military compliance is solved by

diktat. Under somewhat different conditions, the issue of military compliance

may be solved for the defeated military of the loser, but may

remain as a sore point among one or more military elements of the winning

side. Militaries bound by an alliance agreement may find their commanders

straining at the leashes in advocacy of their preferred strategies

for victory as the war nears its termination and the writing of history is

anticipated. In World War II, the allied armies from Normandy to the

Rhine were marked by contention, caused both by personality differences

and by serious differences in operational appreciations of the situation.

These differences were not only apparent in the field, but in

Whitehall and in Washington. Only Eisenhower’s brief to deal with

Churchill as something approaching a military proconsul for Europe (pro

tem, and with occasional intrusive oversight) was able to resolve the

differences that sparked disagreements among high-strung field commanders

and among allied chiefs of service in Europe.

The U.S. and allied militaries did comply with the terms of settlement

in Europe, of course, in part because everyone had had enough of fighting

by this time and because the last symbol of Hitler’s resistance had

Nuclear Weapons and ‘‘Third Wave’’ Warfare 165

been crushed by the Soviets in Berlin. Also, Japan remained to be dealt

with, and U.S. leaders assumed that Soviet cooperation would be needed

to finish the war against Japan. Therefore, issues that might have been

provocative of immediate postwar disagreement, as between the Western

allies and the Soviets, were papered over for later argument. However,

a good historian could easily concoct a ‘‘counterfactual’’ scenario in

which U.S. and Soviet forces fought at the Elbe instead of shaking hands

across it. Where that would have led is speculative. If Field Marshal

Montgomery’s preference for a sudden and spectacular thrust into Germany’s

vital regions in late 1944 had been followed, instead of Eisenhower’s

preference for a slower and surer pushing of Germany’s

defenses across a broad front, would faster-moving U.S. and British

forces have threatened Stalin’s plans for an east-west dividing line? And

if they had, what would Stalin have done about it?

The termination of World War II in Europe did not involve military

noncompliance with civil authority. The latter stages of World War I are

another story. The mutinies of the French army in 1917 forced a change

to a less aggressive and casualty-intensive operational art on the Western

front, and a reshuffling of the French military high command. 92 The

widely politicized and demoralized Russian army walked away from the

field of battle and many soldiers joined workers in Russia’s cities in revolt

against the provisional government. The soldiers’ and workers’ soviets

that seized power in Petrograd in November 1917 prolonged the

war by withdrawing the Soviet Union from the confict in the spring of

1918, according to the terms of Brest-Litovsk. This defection from the

entente almost dealt a mortal blow to Germany’s enemies. Russia’s willingness

to sue for peace counted as a rationale for later British, U.S. and

Japanese intervention in the Russian civil war, prolonging and exacerbating

that conflict and poisoning U.S.-Soviet relations for more than a

decade. 93

An even more interesting case is the situation in Germany after the

armistice. Since Germany’s military had not been annihilated in the field,

many Germans believed that a dishonorable peace had been agreed to

by an insufficiently patriotic civil government. Germany’s military leadership

played this ‘‘undefeated in battle’’ card effectively in the period

between the two world wars. The German army became a state within

a state even within a supposedly liberal-democratic political order. In

the 1920s the army and the social democrats proved to be the pillars on

which a reconstructed state could be built: the army had respect, and the

social democrats, more or less, could govern under the constraints of the

Versailles peace treaty. Ironically, the military combined this institutional

autonomy with ideological neutrality. They would serve any duly

elected or appointed civil masters, including Hitler, but not as his political

enforcers. Thus Hitler was, before and during World War II, forced

166 Nuclear Policy and Strategy in the Future

to turn to paramilitaries other than the Wehrmacht for the suppression

of his political enemies.

The endgame of the American Civil War provided two interesting episodes

related to the problem of military compliance with settlement.

Robert E. Lee surrendered his forces to Ulysses S. Grant at Appomattox

Court House on April 9, 1865. Grant decided on his own to parole Lee,

his officers and his men of the Army of Northern Virginia. Grant’s parole

decision amounted to a de facto amnesty for Lee and his forces. Lincoln’s

presurrender orders to Grant had forbade the latter to make any political

arrangements, but the president acquiesced to his field commander’s

proffered arrangements.

General William Tecumseh Sherman did not receive as much forgiveness

later that month when, in the aftermath of Lincoln’s assassination,

he negotiated with Confederate General Joseph E. Johnston, commanding

what remained of the rebel army in North Carolina. Unlike Lee’s

forces, Johnston’s troops retained some significant fighting power. Over

two days from April 17–18, Sherman and Johnston engaged in expansive

discussions about ending the entire war and emerged with their own

version of a peace pact. This Sherman–Johnston version of a peace agreement

called for disbanding Confederate armies, recognized Confederate

legislatures if they would take an oath to the U.S. Constitution, and left

unmentioned the problem of slavery. Sherman’s pact was repudiated by

President Andrew Johnson, Secretary of State Edwin M. Stanton and

General Grant, although Sherman honestly believed he was acting consistently

with former President Lincoln’s public speeches and private

discussions on the subject of terms for peace. 94

The preceding cases are not historical excursions off the map, but germane

to the second requirement for nuclear, or other, war termination—

military compliance. The contrast between the first and second world

wars shows that, whereas a more ambitious war aim (unconditional surrender)

may be harder to impose in battle, it leaves fewer ambiguities

both during and after the fighting. Disarmed and dead troops and deposed

cabinets can no longer resist. On the other hand, the proclamation

of unconditional surrender as a wartime policy objective may cut the

ground out from under peace proponents within the government or military

of the other side.

In the case of a nuclear conflict between powers in possession of large

arsenals and ample numbers of delivery systems, conditional surrender

may be difficult if not impossible to arrange. Nuclear weapons not yet

fired must be rounded up and cantoned. Troops capable of firing those

weapons will have to stand down. Commanders will have to yield their

weapons and codes to some representatives of the other side or to neutral

arbiters, despite less than complete knowledge of the degree of disarmament

and compliance by their opposite numbers in uniform. One of

Nuclear Weapons and ‘‘Third Wave’’ Warfare 167

the most death-defying scenarios to write is the regathering of nuclear

weapons among Cold War NATO forces in Europe, after they had been

given out during a crisis that led to conventional fighting but before

nuclear escalation had taken place, once governments had decided to

cease fire.

Related to the second requirement for war termination (military compliance

with settlement) is the third—stable command and control during

war. There are at least two aspects to this requirement: stability of

central control, and preservation of continuing control over widely dispersed

forces on the periphery of the command, control and communications

links. Of course, if central control falls apart, so, too will control

over the peripheral actions of armed forces engaged in combat. Central

control could fail due to destroyed command posts or communications,

or both, among other possibilities. Peripheral control might fail if embattled

commanders and troops were cut off from central authority, as

is quite conceivable under many circumstances in war and has occurred

more than once in military history. There is a mistaken assumption that,

with the plethora of communication links that exist today, forces of great

powers on one side of the globe could not be disconnected from their

owners and operators on the other side. This view assumes incorrectly

that only simple messages need to get through from center to periphery,

and vice versa.

For central and regional commands to make intelligent decisions about

fighting or about war termination, they would need to exchange with

field commanders highly precise and often technical information about

status of forces and other variables. In addition, it might be prudent or

even necessary for these kinds of information exchanges to take place

between or among opposed field commanders, and between each of

them and their respective central commands. Sorting out confusion

within the same chain of command is difficult enough as a war nears its

presumed end. For example, the timing of the cease-fire in the Gulf War

of 1991 caught the commanders of the U.S. XVIII Airborne Corps and

VII Corps off balance. Deputy CENTCOM Lt. Gen. Calvin Waller regarded

the decision as premature on operational grounds. 95 CENTCOM

Gen. Norman Schwartzkopf, according to one account, was prepared ‘‘to

subordinate the final destruction of the Republican Guard to the administration’s

political goals’’ such as avoiding images of U.S. brutality and

avoiding American and Iraqi casualties. 96 Some Washington policy makers

were operating under the mistaken assumption that the crack Republican

Guard Iraqi armored divisions had in essence been destroyed

by the time the cease-fire took place.

Intelligence failures can contribute to wartime confusion between center

and periphery in the latter stages of a war. The Battle of the Bulge

beginning in December 1944 is an example. Neither field commanders

168 Nuclear Policy and Strategy in the Future

nor allied command Europe anticipated that seemingly exhausted and

virtually defeated German forces would attempt one final, desperate offensive

that prolonged the war and shook allied confidence. Initial reports

from forward units suddenly caught off guard and committed to

furious fire fights were at first disbelieved or misconstrued by Army

intelligence as a result of preconceptions about German capabilities,

based on recent performance. 97 The assumption that Germany was down

to its shoe leather was correct; what was missed was that resources remained

to be called out, including previously untapped manpower both

old and young, for Hitler’s final throw of the dice. 98

A fourth requirement for war termination can be deduced easily from

what has been argued so far: verification of behavior in compliance with

a cease-fire and subsequent peace settlement. In conventional wars this

is done by exchanges of military observer personnel. Doubtless this

would be difficult to accomplish in a fast-paced and highly destructive

nuclear conflict. Substitutes for observers, especially in the early stages

of any war termination, would have to be found. Overhead reconnaissance

by satellites and aircraft, if any satellites remained functional and

aircraft could count on flying safely, would provide some necessary verification

of compliance with agreed cease-fire protocols. However, any

reconnaissance or surveillance is highly selective. One must have a conceptual

roadmap for what to look at, and in wartime this is all the more

significant than in peace. Just as moving targets are harder to hit than

stationary ones, so, too are mobile and movable missiles and command

posts harder to verify than fixed forces or sites. In addition, some platforms

used for nuclear launch might not be locatable at all: the virtue of

submarines’ (in time of peace) stealthy defiance of location detection,

could become an obstacle to locking down a truce or cease-fire in war.

Illustrative of the overlap between the requirements for war termination

is the essential nature of reliable and uncorrupted communications

links for sending a message of intent to terminate, and for monitoring

compliance with termination agreements. Quite obviously, then, each

side would have to be able to verify that its own communication links

were operating with fidelity and fully under its own control at the endgame

of negotiation as well as when negotiation feelers are first put out.

Ghosts in the machine that spoofed or distorted information could slow

down or bring to a halt the process of arranging cease-fire.

Information warfare has been identified previously as a threat to successful

crisis management and as a potential cause of preemption or

accidental/inadvertent war. Infowar may also make war termination

impossible or more difficult to attain than it would otherwise be.

First, information warfare could prevent the timely and accurate communication

of any interest in cease-fire or peace settlement. In a nuclear

war between the Cold War Americans and Soviets, the targeting of com-

Nuclear Weapons and ‘‘Third Wave’’ Warfare 169

mand posts and command/control assets, including communications

links between U.S. or Soviet NCA and retaliatory forces, was a two-sided

coin. Military strategy a la Clausewitz would call for early strikes against

critical command and control systems in order to paralyze as much of

the enemy’s nuclear stinger as possible. Strikes against command posts,

communications and other aspects of C3 might be defended as ‘‘economy

of force’’ operations providing a nullifying effect equivalent to much

more costly counterforce strikes. The downside of this approach is that

it might work too effectively. A knockout blow against the other side’s

command and control system could, as noted previously, make it more

difficult for the opponent to organize a stable postattack government

with which reliable negotiations could be carried on.

Of course, this latter concern is only pertinent to a limited, as opposed

to a total, nuclear war, thus the apparently greater interest on the part

of U.S., as opposed to Soviet, military theorists in the entire concept of

limited nuclear war. Soviet forces had the tools with which to fight a

limited nuclear war, but neither Soviet military nor political leadership

exhibited any serious inclination to do so. This disinterest in limited nuclear

war may have reflected Marxist–Leninist precepts about wars between

capitalism and