(Professor Barry Buzan, Professor Lene Hansen) Int PDF

SAGE L ~ B ~ A ROF---
Y
[ N T E R N A T I ~ N ARELATIONS
L
INTERNATIONAL SECURITY
SAGE LIBRARY OF INTERNATIONAL RELATIONS
INTERNATIONAL
SECURITY
VOLUME I
The Cold War and Nuclear Deterrence
Barry Buzan and Lene Hansen
@SAGEPublications
Los Angeles London New Delhi Singapore
Introduction and editorial arrangement 0 Barry Buzan and Lene Hansen 2007
First published 2007
Apart from any fair dealing for the purposes of research or

private study, or criticism or review, as permitted under the
Copyright, Designs and Patents Act, 1988, thrs publication
may be reproduced, stored or transn~ittedin any form, or by
any means, only with the prior permrssion in writing of the
publishers, or in the case of reprographic reproduction, In
accordance with the terms of licences issued by the Copyright
Licensing Agency. Enquiries concerning reproduction outside
those terms should be sent to the publishers.
Every effort has been made to trace and acknowledge all the
copyright owners of the material reprinted herem. However, if
any copyright owners have not been located and contacted at
the time of publication, the publishers will be pleased t o make
the necessary arrangements at the first opportunity.
SAGE Publications Ltd

1 Oliver's Yard
5 5 City Road
London E C l Y ISP
SAGE Publications Inc.

2455 Teller Road
Thousand Oaks. California 91320
SAGE Publications India Pvt I,td

B 111 1, Mohan Cooperative Industrial Area
Mathura Road
New Delhi 110 044
SAGE Publications Asia-Pacific Pte Ltd

33 Pekin Street #02-01
Far East Square
Singapore 048763
British Library Cataloguing in Publication data
A catalogue record for this book is available

from the British Library
ISBN: 978-1-4129-2139-8 (set of four volumes)
Library of Congress Control Number: 2006938798
Typeset by Televijay Technologies (P) Limited, Chennai

Printed on paper from sustainable resources
Printed and bound in Zrinski d.d Croatia
Appendix of Sources ix
Editors' Introduction xvii
VOLUME I
The Cold War and Nuclear Deterrence
I . The Sources of Soviet Conduct X 1
2. "National Security" as an Ambiguous Symbol
Arnold Wolfers 15
3. Another "Great Debate": The National Interest of the
United States Hans J . Morgenthau 30
4. The Delicate Balance of Terror Albert Wohlstetter 55
5. The Stability of a Bipolar World Kenneth N. Waltz 74
6. The Sharing of Nuclear Respons~b~lities:
A Problem In
Need of Solution Andre' Beaufre 97
7. S t r ~ t e g ~Studies
c and Its Cr~tics Hedley Bull 106
8. Arms Control and World Order Hedley Bull 117
9. Cooperation Under the Security D~lemma Robert Jervls 130
10. Rational~tyat the Brmk: The Role of C o g n ~ t ~ vProcesses
e In
Failures of Deterrence lack I,. Snyder 171
11. Contrasts in American and Soviet Strategic Thought
Fritz W. Ermarth 19 1
12. Why Nuclear Superiority Doesn't Matter Robert Jervis 207
Colin S. Gray
13. Strategic Stability Reconsidered 223
14. Managing Nuclear Multipolarity John J. Weltman 246
15. Common Security: A Programme for Disarmament -
The Report of the Independent Commission on Disarmament
and Security Issues 258
16. Redefining Security Richard H. Ullman 296
17. Security in the Third World: The Worm About to Turn?
Mohammed Ayoob 317
18. Nuclear War and Climatic Catastrophe: Some Policy
Implications Carl Sagan 330
vi Contents
19. The OffensiveIDefensive Balance of Military Technology:

A Theoretical and Historical Analysis Jack S. Levy 362
20. Why Even Good Defenses may be Bad Charles L. Glaser 386
21. Transarmament: From Offensive to Defensive Defense
Johan Galtung 413
VOLUME I1
The Transition to the Post-Cold War Security Agenda
22. Economic Structure and International Security: The Limits of

the Liberal Case Barry Buzan
23. Sex and Death in the Rational World of Defense Intellectuals
Carol Cohn
24. Geopolitical Discourse: The Soviet Union As Other
Simon Dalby
25. International Security Studies: A Report of a Conference on
the State of the Field Joseph S. Nye, Jr. and
Sean M . Lynn-Jones
26. Base Women Cynthia Enloe
27. The Case Against Linking Environmental Degradation and
National Security Daniel Deudney
28. Security, Sovereignty, and the Challenge of World Politics
R.B.J. Walker
29. How the West was One: Representational Politics of NATO
Bradley S. Klein
30. Soft Power Joseph S. Nye, Jr.
31. Security and Emancipation Ken Booth
32. The Renaissance of Security Studies Stephen M. Walt
33. The Quagmire of Gender and International Security
Rebecca Grant
34. Renaissance in Security Studies? Caveat Lector!
Edward A. Kolodziej
35. A Tale of Two Worlds: Core and Periphery in the
Post-Cold War Era James M. Goldgeier and Michael McFaul
36. Japan's National Security: Structure, Norms, and Policies
Peter J. Katzenstein and Nobuo Okawara
37. The Security Dilemma and Ethnic Conflict Barry R. Posen
38. The Clash of Civilizations? Samuel P. Huntington
Contents vii
39. The Emerging Structure o f International Politics

Kenneth N . Waltz
40. Common, Comprehensive, and Cooperative Security
David Dewitt
4 1 . New Dimensions of Human Security Human
Dezwlopment Report 1994
42. 'Message in a Bottle'? Theory and Praxis in Critical
Security Studies Richard W y n Jones
VOLUME I11
Widening Security
43. What is Security? Emma Rothschild
44. A Genealogy of the Chemical Weapons Taboo Richard Price
45. Securitization and Desecuritization Ole W m e r
46. Security Studies and the End of the Cold War
David A. Baldwin
47. Identity and Security: Buzan and the Copenhagen School
Bill McStueeney
48. Broadening the Agenda of Security Studies: Politics and
Methods Keith Krause and Michael C. Williams
49. Collective Identity in a Democratic Community:
The Case of NATO Thomas Risse-Kappe~
SO. Insecurity and State Formation in the Global Military Order:
The Middle Eastern Case Keith Krause
5 1. Constructing National Interests Jutta Weldes
52. Multiple Identities, Interfacing Games: The Social Construction
of Western Action in Bosnia K.M. Fierke
53. Competing Visions for U.S. Grand Strategy
Barry R. Posen and Andrew L. Ross
54. Imagined (Security) Communities: Cognitive Regions in
International Relations Emanuel Adler
VOLUME IV
Debating Security and Strategy and the Impact of 9-11
55. Should Strategic Studies Survive? Richard K. Betts 1
56. Identity and the Politics of Security Michael C. Williams 25
viii Contents
57. Revisiting Copenhagen: Or, on the Creative Development of a

Security Studies Agenda in Europe Jef Huysmans
58. After Pax Americana: Benign Power, Regional Integration,
and the Sources of a Stable Multipolarity Charles A. Kupchan
59. States of Insecurity: Plutonium and Post-Cold War Anxiety in
New Mexico, 1992-96 Joseph Masco
60. Human Security and the Interests of States Astri Suhrke
61. The Lonely Superpower Samuel P. Huntington
62. The Little Mermaid's Silent Security Dilemma and the Absence
of Gender in the Copenhagen School Lene Hansen
63. Nuclear Order and Disorder William Walker
64. Global Governance, Development and Human Security:
Exploring the Links Caroline Thomas
65. Human Security: Paradigm Shift or Hot Air? Roland Paris
66. Security and Immigration: Toward a Critique of the
Governmentality of Unease Didier Bigo
67. Power and Weakness Robert Kagan
68. Feminist Responses to International Security Studies
J. Ann Tickner
69. On the Pedagogy of 'Small Wars' Tarak Barkawi
70. Knowledge as Power: Science, Military Dominance, and
US. Security Robert L. Paarlberg
71. The 'War on Terror': Good Cause, Wrong Concept
Gilles Andreani
72. Imaging Terror: Logos, Pathos and Ethos James Der Derian
73. Should HIVIAIDS Be Securitized? The Ethical Dilemmas of
Linking HIVIAIDS and Security Stefan Elbe
74. From Social to National Security: On the Fabrication of
Economic Order Mark Neocleous
Appendix of Sources
Grateful acknowledgement is made to the following sources for permis

sion to reproduce material in this book.
1. 'The Sources of Sov~etConduct', X

Eorezgn Affairs, 25(4) (1947): 566-82.
"Reprinted with permmion of Foreign Affam. Copyright (1947) by
the Council on Foreign Relat~ons,Inc."
2. ' "National Security" as an Ambiguous Symbol', Arnold Wolfers
Political Science Quarterly, LXVII(4) (1952):481-502.
0 The Academy of Political Science. Reprinted by permission from
Political Science Quarterly.
3. 'Another "Great Debate": The National Interest of the United States',
Hans I. Mougenthau
The American Political Science Review, XLVl(4) (1952): 96 1-88.
0 American Political Science Association, published by Cambridge
University Press, reprinted with permission.
4. 'The Delicate Balance of Terror', Albert Wohlstetter
Foreign Affairs, 37(2) ( 1 9 5 9 ) :211-34.
"Reprinted with permission of Foreign Affairs. Copyright (1959) by
the Council on Foreign Relations, Inc."
5. 'The Stability of a Bipolar World', Kenneth N. Waltz
Dadalus, 93(3) (1964): 881-909.
0 1964 by the American Academy of Arts 81 Sciences. Reprinted
with permission o f MIT.
6. 'The Sharing of Nuclear Responsibilities: A Problem in Need of
Solution', Andre' Beaufre
International Affairs, 41(3) (1965): 41 1-1 9.
0 Blackwell Publishing. Reprinted with permission.
7. 'Strategic Studies and Its Critics', Hedley Bull
World Polit~cs,20(4) ( 1 9 6 8 ) :593-605.
0 The Johns Hopkins University Press. Reprinted with permission
of The Johns Hopkins University Press.
8. 'Arms Control and World Order', Hedley Bull
Internatzonal Security, l ( 1 ) ( 1 976): 3-16.
0 1976 by the Pres~dentand Fellows of Harvard College and the
Massachusetts Inst~tuteof Technology. Reprinted w ~ t hpermlsslon.
x Appendix of Sources
9. 'Cooperation Under the Security Dilemma', Robert Jeruis

World Politics, 30(2) (1978): 167-214.
O The Johns Hopkins University Press. Reprinted with permission of
The Johns Hopkins University Press.
10. 'Rationality at the Brink: The Role of Cognitive Processes in Failures
of Deterrence', Jack L. Snyder
World Politics, 30(3) (1978): 345-65.
21. 'Contrasts in American and Soviet Strategic Thought', Fritz %! Ermarth
International Security, 3(2) (1978): 138-55.
O 1978 by the President and Fellows of Harvard College and the
Massachusetts Institute of Technology. Reprinted with permission.
12. 'Why Nuclear Superiority Doesn't Matter', Robert Jeruis
Political Science Quarterly, 94(4) (1979-80): 617-33.
O The Academy of Political Science. Reprinted by permission from
Political Science Quarterly.
13. 'Strategic Stability Reconsidered', Colin S. Gray
D ~ d a l u s Journal
: of the American Academy of Arts and Sciences,
109(4) (1980): 135-54.
O 1980 by the American Academy of Arts & Sciences. Reprinted
with permission of MIT.
14. 'Managing Nuclear Multipolarity', John J. Weltman
International Security, 6(3) (198 1-82): 182-94.
15. 'Common Security: A Programme for Disarmament - The Report of
the Independent Commission on Disarmament and Security Issues',
Common Security: A Programme for Disarmament - The Report of
the Independent Commission on Disarmament and Security Issues
(London: Pan Books, 1982), Chapter 1, 'Common Survival' pp. 1-13,
Chapter 6, 'Recommendations and Proposals' pp. 138-76.
O The Independent Commission on Disarmament and Security Issues
1982. Reprinted with permission.
16. 'Redefining Security', Richard H. Ullman
17. 'Security in the Third World: The Worm About to Turn?',
Mohammed Ayoob
International Affairs, 60(1) (1984): 41-51.
O Blackwell publishing. Reprinted with permission.
Appendix of Sources xi
18. 'Nuclear War and Climatic Catastrophe: Some Policy Implications',

Carl Sagan
Foreign Affairs, 62(2) (1983-84): 257-92.
Published by Foreign Affairs. Copyright O 1983 by Carl Sagan.
Reprinted with permission from Democritus Properties, LLC.
19. 'The OffensiveIDefensive Balance of Military Technology:
A Theoretical and Historical Analysis', Jack S. Levy
lnternational Studies Quartevly, 28(2) (1984): 219-38.
0 Rlackwell Publishing. Reprinted with permission.
20. 'Why Even Good Defenses may be Bad', Charles L. Glaser
0 1984 by the President and Fellows of Harvard College and the
21. 'Transarmament: From Offensive to Defensive Defense', ]ohan Galtung
/ournal of Peace Research, 21(2) (1984): 127-39.
0 PRIO. Reprinted with permission.
22. 'Economic Structure and International Security: The Limits of the
Liberal Case', Barry Buzan
International Organization, 38(4) (1984): 597-624.
0 1984 by the World Peace Foundation and the Massachusetts
Institute of Technology. Reprinted with permission.
23. 'Sex and Death in the Rational World of Defense Intellectuals',
Carol Cohn
Signs: Journal of Woman in Culture and Society, 12(4) (1987):
687-718.
0 1987 by The University of Chicago. All rights reserved. Reprinted
with permission.
24. 'Geopolitical Discourse: The Soviet Union As Other', Simon Dalby
Alternative, XIII(4) (1988): 4 15-42.
Copyright 0 1988 by Alternatives. Used with permission of the
World Order Models Project and Lynne Rienner Publisher, Inc.
25. 'International Security Studies: A Report of a Conference on the State
of the Field', Joseph S. Nye, ]r. and Sean M . Lynn-]ones
International Security, 12(4) ( 198 8): 5-27.
26. 'Base Women', Cynthia Enloe
Cynthia Enloe, Bananas, Beaches & Bases: Making Feminist Sense o f
lnternational Politics (Berkeley: University of California Press, 1989),
pp. 6-5-92, 216-20.
0 University of California Press. Reprinted by permission via the
Copyright Clearance Center.
xii Appendix of Sources
27. 'The Case Against Linking Environmental Degradation and National

Security', Daniel Deudney
Millennium: Journal of lnternational Studies, 19(3)( 1990): 46 1-76.
0 Millennium: Journal of lnternational Studies. This article first
appeared in Millennium, and is reproduced with the permission of the
publisher.
28. 'Security, Sovereignty, and the Challenge of World Politics',
R.B.J. Walker
Alternatives, XV(1) (1990): 3-27.
Copyright O 1990 by Alternatives. Used with permission of the
World Order Models Project and Lynne Rienner Publishers, Inc.
29. How the West was One: Representational Politics of NATO',
Bradley S. Klein
International Studies Quarterly, 34(3) (1990): 31 1-25.
O Blackwell Publishing. Reprinted with permission.
30. 'Soft Power', Joseph S. Nye, Jr.
Foreign Policy, 80: 153-71.
Reprinted by permission of PUBLIC AFFAIRS, a member of Perseus
Books Group.
31. 'Security and Emancipation', Ken Booth
Review of International Studies, 17(4) (1991): 313-26.
O British International Studies Association, reproduced with permission.
32. 'The Renaissance of Security Studies', Stephen M. Walt
International Studies Quarterly, 35(2) (1991): 21 1-39.
33. 'The Quagmire of Gender and International Security', Rebecca Grant
V. Spike Peterson (ed.), Gendered States: Feminists (Re) Vision of
International Relations Theory (Boulder i k London: Lynne Rienner
Publishers, 1992), pp. 83-97.
Copyright O 1992 by Lynne Rienner Publishers, Inc. Used with
permission.
34. 'Renaissance in Security Studies? Caveat Lector!', Edward A. Kolodziej
International Studies Quarterly, 36(4) ( 1992): 421-38.
35. 'A Tale of Two Worlds: Core and Periphery in the Post-Cold War
Era', James M. Goldgeier and Michael McFaul
International Organization, 46(2) (1992): 467-9 1.
O 1992 by the World Peace Foundation and the Massachusetts
Institute of Technology. Reprinted with permission.
36. 'Japan's National Security: Structure, Norms, and Policies',
Peter J. Katzenstein and Nobuo Okawara
Appendix of Sources xiii

37. 'The Security Dilemma and Ethnic Conflict', Burry R. Posen
Survival, 35(1) ( 1 993): 27-47.
O 1993 Taylor & Francis Ltd. Reprinted by permission of Taylor 81
Francis Ltd. http://www.informaworld.com. Reprinted by permission.
38. 'The Clash of Civilizations?', Samuel P. Huntington
Foreign Affairs, 72(3) (1993): 22-49.
"Reprinted with permission of Foreign Affairs. Copyright (1993) by
39. 'The Fmerg~ngStructure of Internatwnal Polit~cs',Kenneth N. Waltz
lnternatlonal Securtty, 18(2) (1993):44-79.
0 1993 by the Pres~dentand Fellows of Harvdrd College and the
Massachusetts Institute of Technology. Reprmted with
permission.
40. 'Common, Comprehensive, and Cooperative Security', David Dewitt
The Pacific Review, 7 ( 1 ) (1994): 1-15.
0 1994 Taylor & Francis Ltd. Reprinted by permission of Taylor 8i
4 1 . 'New Dimensions o f Human Security', Human Development Report
Human Development Report 1994 (Oxford: Oxford University Press,
1994), pp. 22-40.
Published by Oxford University Press. Reprinted by permission of
Oxford University Press, Inc.
42. ' "Message in a Bottle"? Theory and Praxis in Critical Security
Studies', Richard Wyn Jones
Contemporary Security Policy, 16(3) (1995):299-319.
O 1955 Taylor & Francis Ltd. Reprinted by permission of Taylor &
4.3. 'What is Security?', Emma Rothschild
Dgdalus, 124(3) (1995): 53-98.
01995 by the American Academy of Arts and Sciences. Reprinted
with permission of MIT.
44. 'A Genealogy o f the Chemical Weapons Taboo', Richard Price
Intcrnational Organization, 49(1) ( 1995): 73-103.
0 1995 by the I 0 Foundation and the Massachusetts Institute of
Technology. Reprinted with permission.
45. 'Securitization and Desecuritization', Ole W ~ v e r
Ronny Lipschutz (ed.), O n Security (New York: Columbia University
Press, 1995), pp. 46-86.
O Columbia University Press. Reprinted with permission.
xiv Appendix of Sources
46. 'Security Studies and the End of the Cold War', David A. Baldwin
World Politics, 48(1) (1995): 117-41.
47. 'Identity and Security: Buzan and the Copenhagen School',
Bill McSweeney
Review of International Studies, 22(1) (1996): 8 1-93.
O British International Studies Association, reproduced with permission.
48. 'Broadening the Agenda of Security Studies: Politics and Methods',
Keith Krause and Michael C. Williams
Mershon International Studies Review, 40(2) (1996): 229-54.
49. 'Collective Identity in a Democratic Community: The Case of NATO',
Thomas Risse-Kappen
Peter J. Katzenstein (ed.), The Culture of National Security: Norms
and Identity in World Politics (New York: Columbia University
Press, 1996), pp. 357-99.
O Columbia University Press. Reprinted with permission.
50. 'Insecurity and State Formation in the Global Military Order: The
Middle Eastern Case', Keith Krause
European Journal of International Relations, 2(3) (1996): 3 19-54.
O Sage Publications Ltd. Reprinted with permission.
51. 'Constructing National Interests', Jutta Weldes
European Journal of lnternational Relations, 2(3) (1996): 275-318.
52. 'Multiple Identities, Interfacing Games: The Social Construction of
Western Action in Bosnia', K.M. Fierke
European Journal of International Relations, 2 ( 4 ) (1996): 467-97.
0 Sage Publications Ltd. Reprinted with permission.
53. 'Competing Visions for U.S. Grand Strategy', Barry R. Posen and
Andrew L. Ross
lnternational Security, 21(3) (1996-97): 5-53.
54. 'Imagined (Security) Communities: Cognitive Regions in International
Relations', Emanuel Adler
Millennium: Journal of International Studies, 26(2) (1997):249-77.
O Millennium: Journal of International Studies. This article first
appeared in Millennium, and is reproduced with the permission of
the publisher.
55. 'Should Strategic Studies Survive?', Richard K. Betts
World Politics, 50(1) (1997): 7-33.
Appendix of Sources xv
0 The Johns Hopkins University Press. Reprinted with permission of

56. 'Identity and the Politics of Security', Michael C. Williams
European Journal of International Relations, 4 ( 2 ) (1998 ): 204-25.
57. 'Revisiting Copenhagen: Or, on the Creative Development of a
Security Studies Agenda in Europe', Jef H u ~ s m a n s
European Journal of International Relations, 4 ( 4 ) (1998):479-505.
58. 'After Pax Americana: Benign Power, Regional Integration, and the
Sources of a Stable M ~ l t i p o l a r i t ~Charles
', A. Kupchan
59. 'States of Insecurity: Plutonium and Post-Cold War Anxiety in
New Mexico, 1992-96', Joseph Masco
Jutta Weldes, Mark Laffey, Hugh Gusterson and Raymond Duvall
(eds), Cultures of Insecurity: States, Communities, and the
Production of Danger (Minneapolis: University of Minnesota Press,
1999), pp. 203-3 1.
O University of Minnesota Press. Reprinted with permission.
60. 'Human Security and the Interests of States', Astri Suhrke
Security Dialogue, 30(3) (1999): 265-76.
0 Sage Publications. Reprinted with permission.
6 1 . 'The Lonely Superpower', Samuel P. Huntington
Foreign Affairs, 78(2) (1999): 35-49.
"Reprinted with permission of Foreign Affairs. Copyright ( 1 999) by
62. 'The Little Mermaid's Silent Security Dilemma and the Absence of
Gender in the Copenhagen School', Lene Hansen
Millennium: Journal of International Studies, 29(2) (2000): 28.5-306.
O Millennium: Journal of International Studies. This article first
appeared in Millennium, and is reproduced with the permission of
the publisher.
63. 'Nuclear Order and Disorder', William Walker
0 Blackwell Publishing. Reprinted with permission.
64. 'Global Governance, Development and Human Security: Exploring
the Links', Caroline Thomas
Third World Quarterly, 22(2) (2001): 159-75.
O 2001 Taylor & Francis Ltd. Reprinted by permission of Taylor &
Francis Ltd. http://www.tandf.co.uk/journals.
xvi Appendix of Sources
65. 'Human Security: Paradigm Shift or Hot Air?', Roland Paris

O 2001 by the President and Fellows of Harvard College
and the Massachusetts lnstitute of Technology. Reprinted with
permission.
66. 'Security and Immigration: Toward a Critique of the Governmentality
of Unease', Didier Bigo
Alternatives, 27(Special Issue) (2002): 63-92.
Copyright O 2002 by Lynne Rienner Publishers, Inc. Used with
permission.
67. 'Power and Weakness', Robert Kagan
Policy Review, 113 (2002): 3-28.
O 2002 by the Board of Trustees of the Leland Stanford Junior
University. Reprinted with permission.
68. 'Feminist Responses to International Security Studies', J. Ann Tickner
Peace Review, 16(1) (2004): 43-48.
O 2004 Taylor & Francis Ltd. Reproduced by permission of Taylor &
Francis Group, LLC., http://www.taylorandfrancis.com
69. 'On the Pedagogy of "Small Wars" ', Tarak Barkawi
70. 'Knowledge as Power: Science, Military Dominance, and U.S.
Security', Robert L. Paarlberg
O 2004 by the President and Fellows of Harvard College and
the Massachusetts Institute of Technology. Reprinted with
permission.
71. 'The "War on Terror": Good Cause, Wrong Concept', Gilles AndrPani
Survival, 46(4) (2004-05): 31-50.
Reprinted by permission of Taylor & Francis Ltd.
http://www.tandf.co.uk/journals.
72. 'Imaging Terror: Logos, Pathos and Ethos', James Der Derian
Third World Quarterly, 26(1) (2005): 23-37.
O Taylor & Francis Ltd. http:Nwww.tandf.co.uk/journals.
73. 'Should HIVIAIDS Be Securitized? The Ethical Dilemmas of Linking
HIVIAIDS and Security', Stefan Elbe
International Studies Quarterly, 50(1) (2006): 119-44.
74. 'From Social to National Security: On the Fabrication of Economic
Order', Mark Neocleous
Security Dialogue, 37(3) (2006): 363-84.
O Sage Publications. Reprinted with permission.
Editors' Introduction
Barry Buzan and Lene Hansen
0
ne of the concepts in International Relations (IR) is security. Since
the primary unit of IR is the state, the field of International Security
Studies (ISS) was formed around the security of states, somewhat mis-
leadingly labeled 'national security'. That states strive to be secure, and that
security involves not only territorial integrity, but the protection of a particu-
lar set of political and cultural values, is thus one of the axioms of inter-
national politics. Yet, if there is agreement that states seek security, there is
no consensus on what this implies for war and conflict. The Realist school
argues that states are by nature driven by their self-interest and a prudent
suspicion towards others: that they seek their own security first and foremost,
and that they put it above all other goals, including economic prosperity. To
be a strong state requires a healthy economy, technological know-how, and a
united citizenry, but what ultimately counts is a state's ability t o project
military force. To Realists the so-called security dilemma implies that states
strive to enhance their security, particularly by maintaining military capabil-
ity adequate t o meet potential challenges. Yet although conceived as a defen-
sive means, to other states this can easily look like an offensive move. States
become trapped in a spiral where each attempts t o protect its security only
to find others raising the stakes in the attempt t o improve theirs. Alliances
might be forged t o balance the capabilities of threatening states, but they
are, argue Realists, fragile and temporary arrangements that dissolve with
the external threats that brought them together.
The discipline of ISS has Realism and its understanding of the state and
military force at its center. Realism is however far from uncontested, indeed
as many of the articles in this Reader show it is challenged on descriptive,
analytical as well as normative grounds. From the birth of ISS a t the end of
World War 11, Idealist or Liberal approaches held a fundamentally different
view of the state and its security. Idealists agreed with Realism that states
feared for their security, but they disagreed that an inbuilt fear of others drove
states to see security only through the lens of narrow self-interest. Hence, it
is possible t o break the escalating logic of the security dilemma: states do have
the capacity to understand the consequences of their own military strategies
and acquisitions, and international institutions can help solve disputes and
build trust. The state, in short, is not a fixed entity; some might act in a
Realist manner, but others might not. If the state is not necessarily Realist,
neither is the international system: states can form institutional arrangements
xviii Editors' Introduction
whose positive purpose stretches beyond defensive alliances and which make
states realize that their own security is deeply interdependent with that of
others.
The debate between Idealists and Realists centered on the status of the state
and the ensuing dynamics of the international system. This debate was and is
central in that it goes to the core of basic analytical and normative assump-
tions about human nature, political community, sovereignty, authority, legit-
imacy and order. Can security only be provided by the state, as Realists argue,
should international or global security be guaranteed by states cooperating,
or is a radical dismantling of the state a precondition for 'real security'? To
trace the evolution of ISS is to uncover a series of conceptual debates on
whether security should be defined in narrow or in broader terms: whether
the security of the state or the security of other 'referent objects' (crucially, the
individual, humanity, ethnic, religious, racial and gendered groups) should be
privileged, whether the concept should be confined to include only military
threats and organized violence or whether it should include for example envir-
onmental dangers, hunger, poverty and epidemics. These are also debates over
what constitutes good social science and whether normative questions can and
should be raised.
The rest of this Introduction is organized into two main sections. The
first considers the birth of ISS and the main forces that have driven its devel-
opment. The second traces the evolution of ISS, looking first at the main-
stream debates during the Cold War, and at the challenges to them; then at
the widening and deepening of the ISS literature that began later in the Cold
War but blossomed during the 1990s; and finally at the impact, still unfold-
ing, of 9/11 and the 'global War on Terror'. References in bold are to articles
included in the Reader.
I. The Origins a n d Dynamics of International Security Studies
The Birth of International Security Studies
ISS emerged as a sub-field of IR after World War 11. It was new in the sense
that while 'security' had long been used by political theorists and military
strategists, 'international security' had never before been the organizing
concept for a field of study. In addition, post-World War I1 writers on inter-
national security began to see themselves as forming a discipline devoted to
a defined set of questions. ISS was nevertheless related to earlier writings in
two important ways.
First, because it drew upon the older discipline of political theory where
writers from classical Greece through to Hobbes, Machiavelli, John Stuart
Mill and Montesquieu had been concerned with the questions of sovereignty
and political authority. It also drew on diplomatic history which traced the
development of relationships between the European states system and the
rest of the world, and military strategists writing on defense, war and power
(on the latter see Baldwin, 1995: 119-120j. The articles included in this
Editors' Introduction xix
reader begin with the birth of the discipline of ISS after World War 11, which
means that these forerunners are not included. But as R.B.J. Walker (1990)
and Michael C. Williams (1998) show, these writings were and are signifi-
cant to ISS precisely because this classical heritage is constantly referred to
and debated.
Second, 'security' was a key political concept prior to 'national security'
gaining center stage after World War 11. Emma Rothschild (1995: 61-4)
traces the many and quite different meanings the word 'security' had before
the French Revolution. She notes that what unites modern uses of 'security'
is a concern with the relationship between the individual on the one side and
the larger collective on the other. Moving from the French Revolution to the
American Depression of the 1930's, Mark Neocleous (2006) and Arnold
Wolfers (1952:481-2) point out that this period linked 'security' to questions
of welfare and economics, not to the national and military security of the state.
What happened after World War I1 was thus that 'security' went from
being a term used within the field of social policy as well as more loosely
within the study of war, defense and diplomacy, to being a term that struc-
tured an entire field. Writing in 1952, Wolfers (1952: 483) observed that,
'The term national security, like national interest, is well enough established
in the political discourse of international relations to designate an objective
of policy distinguishable from others.' The articles included in this four-
volume reader are all part of the development of this field, although they
differ in that some explicitly question the concept of security, while others
take some given version of it for granted. The main example of the latter
practice is most of the works on nuclear deterrence during the Cold War,
which simply analyzed the importance of particular themes within a given
understanding of USIWestern military security (e.g. the distinction between
offensive and defensive weapons, Levy, 1984; or the applicability of a pol-
icy of flexible response or a policy of mutually assured destruction, Jervis,
1979-80; or key events such as the Cuban Missiles Crisis, Snyder, 1978).
Explicit discussions of the concept of security calling for its expansion beyond
nationallstate security and military threats emerged during the 1980's
(Common Security, 1982; Buzan, 1984; Ullman, 1983) and gained ground
as the Cold War and the bipolar international structure dissolved.
When ISS emerged it addressed a classic question - how could states pro-
tect themselves and how did they know what and who were threatening
them? -yet the shift of guiding concepts from 'war' and 'defense' to 'security'
also implied crucial changes in that it opened up the study of a broader set of
political issues, including the importance of societal cohesion and the rela-
tionship between military and non-military threats and vulnerabilities. Indeed
the first three articles in Volume I by George F. Kennan (1947), Arnold
Wolfers (1952) and Hans J. Morgenthau (1952) are clear indications of this
development and together they delineate five themes central to ISS.
First, that the key political question that drove the evolution of LSS was
the concern with the threat posed by the Soviet Union and the ch~llenges
of bipolarity.
xx Editors' Introduction
Second, while military capabilities were crucial to assessing national

security, they were by no means the only important factors, as economic
performance and domestic cohesion were deemed crucial to the con-
tainment of the Soviet Union (Kennan, 1947: 578-81; Morgenthau,
1952: 973).
Third, the interpretation of enemies - or 'knowing the Other' (Dalby,
1988) - and their intentions was also an important theme (Kennan,
1947). Did states have basically similar identities and national interests
or were they products of such distinct histories and traditions that they
needed to be understood in their own right? If states were the product
of historical processes, their identities and definitions of security would
also be open to change.
Fourth, Wolfers argued (1952: 485) that security has both an objective
('the absence of threats t o acquired values') as well as a subjective ('the
absence of fear' that values be attacked) dimension. Assessing security
objectively would be very difficult and if possible then only with hind-
sight (Wolfers, 1952: 485).Yet while objective security would be hard to
define, it should, argued Morgenthau (1952: 965-6) be the goal of ISS
to construct a scientific theory that let the analyst determine a rational
course of action and then assess whether facts complied thereto.
The fifth and final theme introduced was the question of morality.
Wolfers' stance was that 'no policy, or human act in general, can escape
becoming a subject for moral judgment - whether by the conscience of the
actor himself or by others - which calls for the sacrifice of other values,
as any security policy is bound to do' (Wolfers, 1952: 498). Yet, both
Wolfers and Morgenthau (1952: 987) stressed the impossibility of judg-
ing security policy in the abstract and that a choice always had to be
made between different moral principles. Ending on a more Realist note
than Wolfers, Morgenthau (1952: 987) claimed that, 'The realist will
choose the national interest on both moral and pragmatic grounds; for
if he does not take care of the national interest nobody else will.'
T h e Driving Forces of International Security S t u d i e s
Although they were to make a comeback as the explicit conceptual debates

picked up by the mid-1980s, by the mid-1950's these five themes were to a
large extent overridden by more concrete concerns related to nuclear weapons
and deterrence. 'National security' - the security of the state - defined in
politico-military terms held a virtually complete analytical monopoly. As
David Baldwin (1995: 122) noted four decades later, 'It is as if the field came
to be so narrowly defined in later years that the questions addressed during
these early years were no longer considered to belong to the field of security
studies.' But before surveying this historical development and the return of
wider debates on international security, we must answer the question: what
drove this process? Obviously there is not one single factor that can explain the
evolution of ISS and we suggest that five are particularly important, although
Editors' Introduction xxi
not necessarily equally important at all times: great power politics, technology,
key events, the internal dynamic of academic debates, and institutionalization.
Great Power Politics
Great power politics subdivides into three elements: the distribution of

power among the leading states (the polarity of the international system); the
patterns of amity and enmity among the great powers; and the degree of
involvement or interventionism by the great powers. As is immediately clear
from the articles in Volume 1, security analysis during the Cold War was
almost synonymous with studying US-Soviet relations, a bipolar system
with enmity between two great powers whose direct and covert influence
stretched around the globe. Other phenomena appeared on the research
agenda, for instance the question of Third World security (Bull, 1976), b ~ ~ t
these were seen as structured (if not determined) by bipolar relations. The
importance of great power politics is also evident from the debate over
which polarity replaced bipolarity after the end o f the Cold War, with sug-
gestions ranging from uni- to multipolarity (Waltz, 1993; Kupchan, 1998;
Huntington, 1999). Until the end of the 1980s superpower relations had
heen frozen at only slightly fluctuating levels of enmity and engagement, but
with the dissolution of the Soviet Union came not only a reconsideration of
the polarity of the international system, but also of the relations between the
great powers. Was the US going to face enemies or would its deployment of
'soft power' or 'co-optive power' stabilize the system (Nye, 1990)? And
what level of resources was the US prepared to devote to security problems
outside of its own immediate sphere of interest (Posen and Ross, 1996-7)?
The attacks on September 11 led US policy makers and many security ana-
lysts to define a new era. Whether the War on Terror will ultimately boost
or weaken the relative power of the US as well as exacerbate or ameliorate
patterns of amity and enmity remains to be seen, but great power politics is
still a key question on the agenda. To point to great power politics as a driv-
ing force is also to note that ISS began as an American discipline, focused on
American security and written by Americans (although some had emigrated
from Europe to the US before or during World War 11) (Kolodziej, 1992:
434). European approaches might have gained more ground after the end of
the Cold War, but as Ayoob (1983-84) and Krause (1996) point out, it is still
the Western model of the state which forms the core of ISS.
Technology
Technology drives ISS through its impact on the threats, vulnerabilities and
(in)stability of strategic relationships. The arrival during the mid-1940s of
the atom bomb was pretty much the foundational event for Strategic
Studies and the impact of nuclear - and nuclear- related - technology dur-
ing the Cold War can hardly be exaggerated. Nuclear weapons provided a
xxii Editors' Introduction
huge surplus capacity of destructive power for the first time in military his-
tory. Long-range ballistic missiles speeded up delivery times and were cap-
able of carrying nuclear warheads, a technological development that liberated
nuclear weapons from vulnerable bomber delivery systems, and greatly
increased the capacity to make a first strike against opponents. Whereas
nuclear warheads and intercontinental missiles were real developments feed-
ing huge quantities of ISS literature, the enormous and ongoing literature on
Anti Ballistic Missiles (ABM)/Ballistic Missile Defense (BMD) reveals that
even potential technology developments could have major impacts on both
strategic relations and ISS.
Technology need not be exclusively military in kind to make an impact
on ISS. The history of military and civilian technologies is often one of
interplay and 'dual-use'. The Internet for instance was originally developed
as a military technology, as a distributed network transmitting information
under a nuclear attack. Nuclear technology to take another example has a
military as well as a civil side (energy and medicine) that can be difficult to dis-
tinguish, a fact which also complicates the assessment of nuclear proliferation.
The same dilemma is also applicable to biological and chemical weapons or to
the communications technologies applied in both civilian consumer electron-
ics and battlefield management.
If the concept of security is expanded beyond the military sector the list
of technological factors that can drive security debates grows as well. If
AIDSIHIV is seen as a threat to regional security in parts of Africa and Asia,
the retroviral technology for treating those infected is key to the spread and
consequences of the decease (Elbe, 2006). Or, if the environment is threat-
ened by the effects of industrialization, then the technologies implicated in
these threats and their solution become central. The attacks on September 11
and the War on Terror show that technology and the identification of
threats and enemies are intimately linked and that the list of technologies
central t o ISS changes over time. First, the technology of digital networks
(the electronic movement of money, cellphones and Internet communication)
is significant for how physically dispersed terrorist networks communicate
while simultaneously erasing (most of) their traces. Second, the spectacular
character of the attack could be seen as deeply connected to the images
created and the instantaneity with which they were globally broadcast allow-
ing the world to watch in real time as the World Trade Center collapsed
(Der Derian, 2005). Arguably the power of terrorist networks like Al-Qaeda
lies not in their quantitative capabilities (no one considers an invasion likely),
but in their qualitative ability to unsettle Western societies' feeling of security.
Hence the production and distribution of iconic images through new media
technologies that propel this feeling becomes a key security concern.
The question how technology impacts economic, political, military and
cultural developments has been a topic of great debate in the social sciences
and to speak about technology as a driving factor thus raises the specter
of technological determinism (Levy, 1984; Paarlberg, 2004). Yet while tech-
nology is undoubtedly a main driving force in the development of ISS, it is by
no means a determining one, first, because technology is itself influenced by
Editors' Introduction xxiii
the other driving forces; and second, because there are human agents (civilian
and military; commercial and public) who make decisions about which
technology to develop. Decisions on nuclear technology during the Cold War
were hugely impacted by the bipolar confrontation between the US and the
Soviet Union. Once in the world, technology creates pressures of its own,
which again impacts the political process, but this is a complex process
of feedbacks between technology and the other driving forces and hurnan
decisions, not one of determinism.
Key Events
Events come in various forms, and they can change not only relationships
among the powers, but also the academic paradigms used to understand
those relationships. The most dramatic are specific, focused crises that not
only become objects of study in their own right, but which change existing
understandings, relationships and practices in the wider strategic domain.
Two key examples of this type are the Cuba missile crisis in 1962 (Snyder,
1978; Weldes, 1996) and the terrorist attacks on the US on September I I ,
2001 (Barkawi, 2004; Der Derian, 2005). And, the ending of the Cold War
was of course the single most important event in the history of ISS. Other
events take the form of steady processes unfolding over time that change the
knowledge, understanding and consciousness that support existing practices.
A good example of this is the rise of environmental concerns and the move
o f the environment from a background variable to a foreground one (Ullman,
1983; Deudney, 1990). There was no specific crisis that put environmental
issues into the foreground, but rather a steady drip of new information, new
understandings, and a rising public consciousness that grew sufficiently wide
and deep to open a place for environmental security in policy debates and the
ISS literature. The identification of key events might often seem common-
sensical: it is not hard to see the impact of the Soviet Union gaining nuclear
weapons, its dissolution in 1991, or the attacks on September 11. Yet, in ana-
lytical terms one should note that events are in fact politically and inter-
subjectively constituted. It is the acknowledgen~entby politicians, institutions,
the media and the public that something is of such importance that it should
be responded to, possibly even with military means, that makes it an 'event'.
T h e internal Dynamic of Academic Debates
An idealized model of how academic knowledge is created would p e d i c t

that ISS has evolved progressively in response to key events, new tech~mlogies
and great power politics. Hypotheses would be derived, falsified or verified,
and theories revised, expanded or abandoned in response. The actual devel-
opment of IIS is however much more complicated due to the absence
of consensus on what scientific model should be adopted, that is whether
positivist epistemologies and methodologies from the natural sciences and
xxiv Editors' Introduction
economics or interpretative epistemologies from the humanities (sociology,

philosophy, history and anthropology among others) should be used.
This driving force situates ISS within the larger debates on epistemology,
methodology and choice of research focus that are conducted within the
social sciences in general. As shown above, the call for objective measures
and rational science was part of ISS since the early 1950s, and discussions
of epistemology and methodology have been central ever since. To call ISS
an 'applied' discipline would not be completely fair, but it is true that aca-
demic debates in ISS have relied upon significant imports from other discip-
lines, and not just mathematics and economics which delivered most of
the first generation of nuclear strategists: game theory Uervis, 1978);cognitive
psychology (Snyder, 1978); linguistics (Cohn, 1987; Fierke, 1996; Wxver,
1995); social theory (Dalby, 1988; Price, 1995; Jones, 1995; Krause, 1996;
Hansen, 2000; Bigo, 2002; Der Derian, 2005); political theory (Walker, 1990;
Williams, 1998); development and postcolonial studies (Ayoob, 1983-4;
Krause, 1996; Thomas, 2001; Barkawi, 2004); and feminist theory (Cohn,
1987; Grant, 1992; Hansen, 2000; Tickner, 2004). The impact of these
disciplines and their debates have been felt in ISS both in terms of how
security should be conceptualized/what should fall under the rubric of ISS,
and in terms of how it should be analyzed, that is the epistemology and
methodology to be applied.
During the Cold War, ISS had close links to (neo)realism, a link fortified
by the division of labor between ISS and International Political Economy
(IPE)which emerged as the other great sub-field of IR in the 1970s. IPE partly
defined itself against ISS, the two sub-fields carving up the terrain of IR so
that IPE claimed the cooperative, joint-gains side of the subject, with ISS
claiming the conflictual, relative-gains one. As the Cold War drew to a close,
a wave of critical articles emerged which contested both Realist ISS's focus on
national security as well as the rationalist methodologies usually adopted to
its study. Poststructuralists, Constructivists, Critical Theorists and Feminists
introduced interpretative, linguistic, sociological and deconstructive episte-
mologies, and a split between rationalists and 'reflectivists' was coined by
Robert Keohane (1988) to indicate the key dividing line in IR.
Institutionalization
To identify institutionalization as a driving force is to further highlight that

academic debates do not unfold in an economic and structural vacuum.
Research is influenced by the availability of funding, for ISS in general as
well as for particular forms of research; by access to top journals and pub-
lishers; by the processes of hiring and institution building that take place in
universities and by think tanks. For example hiring policies at universities
are extremely central (Betts, 1997). Tenured professors make decisions on
what future students learn, they decide which graduate students are accepted
and on what topics, they have at least part of their time allocated to research,
and the security of tenure provides them with a significant degree of freedom
Editors' htroduction xxv
for research. Successful graduate programs are characterized by the ability

of graduating Ph.D.s to find jobs at respectable universities and policy insti-
tutes. The very fact that courses on Strategic Studies were taught from the
late 1960s onwards further institutionalized ISS within academe as well as
in the policy world.
The level and kind o f funding provided by governments and foundations
is equally important. In countries characterized by high levels of public
spending on education and research it might make a significant difference
that resources are invested not only in university education and research,
but in research centers such as the Copenhagen Peace Research Institute
(COPRI), home of the Copenhagen School (Huysmans, 1998), Stockholm
International Peace Research Institute (SIPRI),International Peace Research
Institute, Oslo (PRIO) and Norwegian Institute of International Affairs
(NUPI). Foundations have significant discretion to target their financial
support towards particular programs, and to foster or inhibit new direc-
tions. The growth in widening approaches in the 1990s was for instance
aided by a series of American foundations (Kolodziej, 1992: 437; for a crit-
ical assessment of this development see Nye and Lynn-Jones, 1988: 21).
The influence of think tanks operating in the gray zone of policy advice,
lobbying and consulting is another aspect of institutionalization, with the
recent influence of neo-conservative think tanks on the policy of the George
W. Bush administration as a case in point. In addition to influencing gov-
ernment policy and public debate, think tanks might also function as a
reservoir for politically motivated appointments (Dalby, 1988).
This account of institutionalization shows that this 'driving' force might
speed up as well as slow down the effect of other forces. The institutionaliza-
tion of ISS through generations of hiring practices might easily breed a certain
conservatism as far as broadening the concept of security is concerned, bal-
ancing the impact of progressive foundations. Institutionalized conceptions
might also 'slow down' the impact of key events, as when neo-realism man-
aged to reinvent itself after its failure to predict the end of the Cold War.
11. T h e Evolution of International Security S t u d i e s
The themes set out by the very first writers on security worked together with
the driving forces to found a field of Strategic Studies that was simultan-
eously productive, influential and fashionable as well as committed to a nar-
row, state-centric, military-political view of ISS. Broader questions, including
those of economic security and domestic cohesion, put on the research agenda
by Wolfers and Morgenthau were marginalized. Normative questions were
not explicitly pursued either as mainstream ISS (Strategic Studies) saw itself
as being on the right side of a conflict whose existence and legitimacy was
beyond questioning. Cold War ISS was driven largely by the policy problems
facing the US, and to a lesser extent those of its allies (e.g. Beaufre, 1965).
xxvi Editors' Introduction
Developments in Soviet capabilities and doctrine was the overwhelming

concern (Ermarth, 1978). Epistemological questions were not at the fore-
front either as an objective, rationalist and positivist approach was shared
by theoretical as well as empirical scholarship.
It was thus the driving forces of superpower politics and new military
technologies that were most conspicuous in forming mainstream ISS. While
bipolarity was thought to be relatively stable (Waltz, 1964) the technology
surrounding nuclear weapons was anything but. Nuclear weapons technol-
ogy underwent very rapid and dramatic development throughout the Cold
War, generating an ongoing military dynamic that defined the heart of ISS.
Additional uncertainties were introduced by misinformation about who
had deployed what: the bomber and missile 'gaps' of the mid- and late
1950s, in which Soviet secrecy and American domestic politics combined to
produce huge US overreactions to non-existent Soviet 'leads'. Sputnik in
1957 helped launched the 'Golden Age' of Strategic Studies, when the basic
rules and dynamics of mutual nuclear deterrence were worked out in detail
(Wohlstetter, 1959), though many had been anticipated by earlier writers
responding to the first arrival of nuclear weapons.
Technological developments locked the superpowers into a fierce arms
race with both quantitative (how many and how powerful missileslwar-
heads?) but also qualitative (how accurate, how quickly delivered, how well
protected against pre-emptive attacks?) dimensions. Arms racing not sur-
prisingly became a staple topic in the literature, giving yet more weight to
what seemed to be the material driving forces defining the strategic agenda.
The superpower arms race was driven not only by improvements in tech-
nology (and the lobbying power of arms industries), but also by fears that
failure to keep up would make one's nuclear forces vulnerable to first strike
by the enemy. From this foundational insecurity arose a huge and elaborate
body of theory and argument about incentives to attack (or not) under vari-
ous conditions of nuclear balance (Wohlstetter, 1959; Jervis, 1979-80; Gray,
1980; Weltman, 1981-2). Formal theoretic ways of thinking were called in
to help in understanding the 'game' of deterrence and bipolar superpower
rivalry (Jervis, 1978).
Much of this literature was heavily dependent on assumptions of ration-
ality (Snyder, 1978) to work out the great chains of if - then propositions that
characterized deterrence theory: if A attacks B in a given way, what is B's best
response, and what would A then do in reply, and then ... . The fear of first
strike was real in the early phases of the Cold War when nuclear arsenals
were small and vulnerable. It diminished from the later 1960s as nuclear ar-
senals got larger and much harder to attack (particularly when missiles were
put into submarines), giving an effective so-called 'secure second strike'. But
it was replaced by another, more subtle, fear known at the ex-ante ex-post
dilemma. This envisaged a counterforce attack by one side against the other
in which the attacked states loses more of its nuclear weapons than the
attacker uses in his first strike. Did rationality suggest that the attacked state
should not retaliate because to do so would be suicidal, thereby having to
Editors' Introduction xxvii
accept the attack without making any retaliation? Understanding these great
chains of reasoning was seen as crucial to developing the best military options
that would deter the enemy from attacking in the first place.
Much ink was spilled among other technological choices over the costs
and benefits of putting multiple warheads on missiles; of pursuing high
levels of accuracy with so-called precision-guided munitions; of developing
supersonic bombers; of deploying cruise missiles; and of building elaborate
protected hiding places for land-based intercontinental ballistic missiles.
One o f the fiercest, and still ongoing, arguments of this sort was about ABM,
aka BMD, systems, under argument since the 1970s (Common Security,
1982; Glaser, 1984; Walker, 2000). Part of the argument was about whether
or not it could be done with existing or likely technology. But the more inter-
esting theoretical part was about what impact deployment of an effective, or
even partly effective, BMD would have on nuclear strategic stability. With its
promise of escape from the whole logic of deterrence, and especially from hav-
ing one's population held hostage under the grim, but supposedly stabilizing,
logic of the Cold War's most notorious acronym, MAD (mutually assured
destruction), RMD proved particularly attractive in US domestic politics,
helped there by its appeal to enthusiasm for technological fixes, and its
amenability to being staged as defensive (the protests of strategists about its
costs, technical difficulty and destabilizing consequences notwithstanding).
In addition to the pressures from rapidly evolving technologies, there was
an ongoing fundamental disagreement about the basic nature of nuclear
deterrence itself, and whether that made it easy or difficult to achieve (Jervis,
1979-80; Gray, 1980). Some thought that nuclear weapons made deterrence
easy, because any even half-rational actor would be given serious pause by the
prospect of obliteration. In other words, possession of a nuclear arsenal suf-
ficient for 'assured destruction' would basically suffice, leading to a so-called
'minimum deterrence' strategy. Others calculated that a ruthless rational
actor (as Kennan, 1947 had postulated the Soviet Union to be) would require
not only a threat of high damage, but also a near-certain probability that such
a retaliation would be delivered, before deterrence could be effective.
Minimum deterrence offered a kind of stability in easy parity, and also
economy, but at some risk of vulnerability to utterly ruthless opponents. Its
logic also provided incentives for nuclear proliferation, making it seem fairly
straightforward for lesser powers to acquire a great equalizer. Proliferation
threatened to complicate not only the core elements of the nuclear debate-
deterrence, arms control, and escalation - all of which were much more dif-
ficult with three or more parties than with just two (Weltman, 1981-2;
Walker, 2000), but also bipolarity itself. Consequently, the US and the Soviet
Union led the way in promoting a nuclear nonproliferation regime, and
nuclear proliferation became a large and elaborate subject in its own right
within the ISS literature.
In contrast to minimum deterrence, maximum deterrence thinking offered
higher entrance costs to would-be nuclear weapons states, and an expensive,
open-ended arms race to existing nuclear weapon states. The supposed gain
xxviii Editors' Introduction
was to close loopholes against extreme aggressors who might take risks along
the lines of the ex- ante ex- post dilemma. Dealing with this contingency
generated demands for huge and elaborate forces capable of responding to
aggression at any level, and of maintaining 'escalation dominance' through-
out the spectrum of conventional and nuclear warfighting. Maximum
deterrence thinking rested on the assumption of a highly aggressive, risk-
taking and opportunistic opponent, but it was also driven by the problem
of Extended Deterrence (ED) that arose from the US guarantees to protect
Europe embodied in the NATO alliance.
ED links the technological driver to the great power politics theme.
Extending the US nuclear umbrella was uncomplicated when the US nuclear
monopoly made deterrence easy even in the face of much superior Soviet
conventional military strength in Europe. But it became fiendishly difficult
when the Soviets also acquired the capability to threaten the US with nuclear
weapons. H o w could the European allies believe that the US would retaliate
against the Soviet Union for, say, an attack on West Germany, when the con-
sequence could be Soviet retaliation against American cities? This question
and its many variants haunted Western strategic thinking from Sputnik
onwards (Beaufre, 1965; Jervis, 1979-80; Gray, 1980). It was central to the
literature on NATO and its recurrent discontents over especially nuclear
strategy, which was another major theme in the ISS literature.
Extended deterrence and flexible response fed another concern intrinsic to
the whole logic of maximum deterrence, and also linked to rival superpower
interventions in crises and conflicts in the Third World (Cuba, Southeast
Asia, the Middle East): escalation and how to control it. The practice of ED
inevitably led to scenarios about low- level warfighting in response to local
aggression, and how to respond if the opponent raised the ante by moving
t o higher levels of force, especially to the use of so-called 'tactical' nuclear
weapons. Maximum deterrence logic required that rationality prevail, and
that limited nuclear war be containable, but there were real doubts about
whether such cool-headedness and fine-tuning would be possible once
command and control systems came under the intense and unpredictable
pressures of actual nuclear warfighting. Maximum deterrence logic and ED
thus pushed deterrence theory into fantastic complications. The great
chains of if-then propositions became so long, and rested on so many ques-
tionable assumptions about both technological performance and human
behavior, that the credibility of the theory itself came into question. Even if
the Cold War had not ended, deterrence theory was in trouble, and even at
its peak it was never without challengers.
Cold War Challengers to Mainstream Strategic Studies
Strategic Studies definitely provided the foundation for ISS, but it was not
unchallenged, nor was it an entirely uniform block as shown by for instance
Snyder's and Bull's critical discussions of rational actor assumptions (Snyder,
1978; Bull, 1968). Peace Research (PR) had to a large extent constituted
itself in opposition to Strategic Studies with Peace Researchers arguing that
Editors' Introduction xxix
SS's assumptions of (elite) actor rationality and bipolar stability were not only
wrong but dangerous. The devastation brought about by a nuclear exchange
would be so enormous that one should not rely upon small elites to make such
decisions (Sagan, 1 9 8 3 4 ) . Peace Researchers sometimes challenged the frame-
work of bipolarity, mainly arguing that arms control, disarmament, a change
from offensive t o defensive forces, and the acknowledgement of each other's
legitimate existence could lower the levels of tension (Galtung, 1984;
Common Security, 1982). Much of PR thus shared SS's conceptualization of
security in political-military and state-based terms, but argued for a different
analysis of the threats and dangers arising from nuclear deterrence. Perhaps
the most conspicuous area of overlap was in the literature on arms control,
which aimed at trying to stabilize the superpower-nuclear relationship by
managing the types and numbers of weapons deployed (Bull, 1976). Arms
control and much PR, can be seen as trying to give concerns about inter-
national security equal status alongside the mainly national security concerns
of the early Strategic Studies literature. Underlying both was a normative cri-
tique of mainstream Strategic Studies for its failure to question sufficiently (or
in some cases at all) the ethical implications of its state-centric focus on
national security under nuclear conditions.
Epistemologically, many Peace Researchers also adopted a rationalist
and objectivist position: the Journal of Conflict Resolution and the journal
of Peace Research featured studies built on clear assumptions of causality
and quantitative analysis of large data sets. Carl Sagan's (1983-4) 'hard-
science' yet evocative account of the climatic consequences of a nuclear war
shows the objective and positivist nature of ISS challengers as well as the
scope of critics. The Bulletin of the Atomic Scientists was also devoted t o
'hard scientists' engaged in trying t o control and contain the consequences
of the nuclear arms race.
Other Peace Researchers worked more critically with the concept of
security arguing that to privilege state security was t o leave out the security
problems faced by many of the world's individuals. Individual security prob-
lems were numerous, but the most important (particularly in the Third World)
were linked t o economic deprivation (poverty, malnutrition and hunger)
and underdevelopment, t o 'structural violence' in Galtung's (1969) termin-
ology. The report on Common Security: A Programme for Disarmament
chaired by Olof Palme was key in this respect as its concept of Common
Security successfully laid out a wider non-SS security agenda (Common
Security, 1982; Dewitt, 1994). The majority of the report dealt with disarma-
ment and was thus in line with the traditional PR approach to international
security, but the report's first chapter also established a link between domes-
tic and international security and North-South relations arguing that the
Third World was highly negatively affected by global militarization which
exacerbated the risk of 'economic failure and social disruption' (Common
Security, 1982: 5 ) .
Not a11 challengers to ISS self-identified as Peace Researchers and during
the 1980s a series of approaches emerged which t o varying degrees argued
xxx Editors' introduction
for a revision of military-state-centric security. Barry Buzan's People, States

and Fear: The National Security Problem in International Relations (1983)
provided the first in-depth monograph on the concept of security with chap-
ters on individual security and the international economic system, and
Richard Ullman (1983; see also Buzan, 1984; Deudney, 1990) called for an
expansion of security to include environmental and economic threats. These
works helped define the basic structure of the widening debate which was to
gain full force by the early 1990s: first, should the traditional focus of ISS on
military-political threats (and means) be expanded to include environmental
and economic threats as these were indeed more threatening to state survival
than the nuclear weapons stabilized through successful deterrence? Second,
should the privilege accorded to the state by the concept of national security
be supplemented or replaced by a concept of internationallglobal security,
individual security or non-state collectivities such as national, ethnic, racial
and gendered groups? Many different answers have been given since the
early 1980's: some wanted to expand the type of threats while being conser-
vative in terms of keeping the state as the referent object, while others argued
in favor of expanding to include non-state actors as well. The specific
answers notwithstanding, the most significant change in ISS was the emer-
gence of a distinct and explicit conceptual debate.
One group of challengers argued that SS was an ethnocentric discipline
in that it left out the Third World (unless it seriously impacted on super-
power relations as during the Korean and Vietnamese Wars and the Cuban
Missiles Crisis), built exclusively on American (or at least Western) authors,
and perhaps most importantly relied upon a particular Western model of
the state (Bull, 1976; Ayoob, 1983-84; Kolodziej, 1992; Krause, 1996;
Barkawi, 2004). This presumably universal, but in fact particularly Western
state, was uncritically applied to the non-Western world. But the legacy of
colonialism, including the often arbitrary drawing of borders, implied that
Third World states experienced distinct forms of security problems. One
crucial difference was that internal security problems played a much larger
role as states suffered from weak domestic political structures and poorly
developed national communities (Ayoob, 1983-84; Krause, 1996). Most of
what was written on Third World security during the 1980s was however
concerned with national security, that is the domestic security problems the
state was facing and the external consequences thereof. Only after the end
of the Cold War did a distinct ISS literature develop that shifted the refer-
ent object from the state t o the individual and sub-state groups.
Criticisms that more explicitly challenged the status of the state were
argued by Poststructuralists and Feminists joining ISS in the latter half of the
1980s. Poststructuralists such as R.B.J. Walker, Richard Ashley, Michael
Shapiro, Simon Dalby, James Der Derian and David Campbell held that the
privilege accorded to the state in ISS rested upon the dominance that state
sovereignty had gained from the Peace of Westphalia onwards. State sover-
eignty was not an inevitable condition, but constantly reproduced through
political practices (with 'practices' covering both foreign and security pol-
icies and the academic disciplines of IR and ISS). State sovereignty offered a
Editors' Introduction xxxi
particular resolution to the questions of authority and security, but it was

simultaneously a solution which made 'security' the prerogative of the state.
Individuals gave up their right to define security - and pursue it - and invested
it in the sovereign state, but this 'solution' meant that it became very difficult,
if not impossible, to formulate a meaningful conception of 'individual secur-
ity' or a collective conception of security below, above or across 'national
security' (Walker, 1990; Williams, 1998). To challenge the 'national security'
of ISS would require not simply calls for 'individual' or 'global' security as
had been the case in parts of PR, but that the complexities at the heart of state
sovereignty and the liberal theory on which it was based were reworked and
reconstructed (Walker, 1990; Wazver, 1995).
Poststructuralism in the 1980s was - as ISS in general - written in the light
of the Cold War and the antagonistic relationship between the superpowers
was the political context in which the Poststructuralist theory on identity and
security developed. Poststructuralists argued that foreign and security policies
were fundamentally about identity, that security policies had to formulate a
narrative depicting a national Self against one or more threatening Others
which might be concrete, as in the case of the Soviet Union, or abstract, as in
the case of 'instability' (Dalby, 1988). Poststructuralism's concept of identity
was relational, social (not a property of the individual, but argued in a public
language and arena), and discursive (articulated through written and spoken
statements). The Cold War and the construction of the Soviet Union provided
Poststructuralists with many examples of the threatening Other (Dalby, 1988)
and it was stressed how the construction of threatening Others on the outside
was simultaneously part of producing domestic consensus and legitimacy
around particular visions of 'America', 'the West' and 'NATO' (Klein, 1990).
Poststructuralism's turn to discourse implied an important epistemol-
ogical shift ( W w e r , 1995). Wolfers' distinction between security's objective
and subjective dimensions had so far underpinned ISS in that discussions of
rationality under deterrence had included subjective factors, but as 'filters'
or 'misperceptions' distorting objective assessments of capabilities and inten-
tions. The discursive conception of security introduced by Poststructuralism
broke with this distinction, pointing neither to material capabilities nor beliefs
as the explanatory variable, but to the way in which both were constituted
and given meaning by ~oliticalactors (Cohn, 1987). Methodologically, this
implied that the words and interpretations derided by scholars as diverse as
Morgenthau (1952) and Galtung (1984) gained centerstage: it was the con-
stitution of 'security', threats and enemies by politicians and institutions
and the legitimation of policies and behavior which were to be examined.
As Poststructuralism emerged as a distinct, critical perspective so did
Feminism, in part as a consequence of how the two approaches were mak-
ing a general impact on the Humanities and the Social Sciences (the driving
force of academic debates), in part as a product of the political opposition
to antagonistic policies on both sides of the Iron Curtain. Feminist critiques
were made of the gendered and disembodied language of deterrence
(Cohn, 1987). Jean B. Elshtain (1987) showed how stereotypical construc-
tions of men as warriors and women as supportive mothers and wives had
xxxii Editors' Introduction
underpinned Western thought and politics from classical Greece onwards,

and it was argued that women faced concrete security problems overlooked
by mainstream ISS, for instance as a consequence of the militarization of
everyday life created by foreign bases (Enloe, 1989). Feminist work often
took a multidimensional and multi-level approach tracing the links between
military security, domestic and global economic structures, and gendered
norms from particular groups of women in specific locales all the way up to
the regional, national and international level (Enloe, 1989; Tickner, 2004).
Conceptually as well as normatively this challenged Strategic Studies'
approach to security: Strategic Studies had focused predominantly on the
interactions between states, and to some extent on the state's ability to
generate domestic order. Feminists stressed that 'national security' excluded
a consideration of 'women's security', but that it also depended upon particu-
lar constructions of masculinity. Hence the goal was not to pitch 'women's
security' against that of men's, but to show how men as well as women were
cast into different identities that sustained a war-prone, Realist system.
Epistemologically and methodologically the shift was equally striking: SS
had focused on material capabilities, state behavior, and in some cases mili-
tary and strategic rationality (Snyder, 1978) using quantitative methods or
logical hypothetical reasoning when spinning out the scenarios of nuclear
deterrence. Feminists argued in favor of making the definition of security
dependent upon individuals' experiences (Tickner, 2004). On the one hand,
'experiences' were based on the individual's subjective definition, on the other
it had to presuppose a collective subject ('women') for individual experiences
to provide a meaningful concept of 'women's security'. As Rebecca Grant
(1992) pointed out, this inherent tension always risked overwriting differ-
ences between women. These tensions aside, Feminist analysis broke import-
ant g o u n d for other approaches to emerge after the Cold War, in particular
for Critical Security Studies (Booth, 1991; Wyn Jones, 1995) and anthro-
pological studies linking micro- and macro-levels of analysis (Masco, 1999).
After the Cold War: Traditional versus Wider Security
Agendas and Understandings
The ending of the Cold War was a benchmark event in the evolution of ISS.
Much of the mainstream military-political agenda simply disappeared.
Bipolarity - and the nuclear deterrence which surrounded it - had been the
cornerstone of mainstream ISS for almost 40 years when the Cold War
came to a halt. As a result mainstream ISS faced something of an institu-
tional crisis, and approaches calling for a widening of security to include
other sectors than the military and referent objects besides the state gained
ground. The period from the early 1990s until September 11,2001 did not
have one dominant event or political problem as had the Cold War to tie
the discipline together. The war against Iraq in 1991, the rise of Japan or
China, nuclear proliferation, the wars following the dissolution of Yugoslavia,
Editors' Introduction xxxiii
humanitarian intervention in Somalia, and the war against Serbia in 1999 con-
stituted some of the key events, but none of them structured an overarching
internat~onalsecurlty agenda.
The impact of this on ISS was not simply to end one line of approach and
open the way to another. As the problematique of international security lost
its dominant core and fragmented into multiple concerns, so also ISS evolved
to match. The traditionalist core did not disappear, but it did lose ground to
other approaches. There is thus quite a lot of continuity, but also a substantial
change in the distribution of interest in various approaches. We characterized
Cold War ISS as having one dominant (traditionalist) main stream with a
number of challengers (Peace Research, Arms Control, Poststructuralism,
Feminism, Postcolonialism) arrayed around it. During the 1990s this pattern
changes to something more like a river delta where the main stream branches
off into several streams, with none having a clear claim to be the main one.
Arms Control, like the traditional approach generally, loses relevance because
the strategic military threat has dropped away. Feminism, Postcolonialism and
Poststructuralism continue to evolve as branches of ISS, and new branches
based o n non-military agendas and Constructivist approaches join in. In this
section we will look first at the continuities from the previous two sections,
and then at the newer branches.
The traditionalists remained robust, continuing to claim primacy, and
pursuing a diverse post-Cold War military-political agenda. Their first con-
cern, closely linked to neo-realism, was the debate over which polarity
(multi- or uni-?) had replaced bipolarity, and the consequences thereof for
US grand strategy (Goldgeier and McFaul, 1992; Huntington, 1993; Waltz,
1993; Posen and Ross, 1996-7; Kupchan, 1998; Huntington, 1999; Kagan,
2002). At the beginning of the Cold War this type of question had been set-
tled relatively early, and the nature of the rival to the US became quickly
and deeply sedirnented. The ending of the Cold War produced a much
murkier international situation in which the nature and identity of the chal-
) the US, if any, remained unclear: Japan and China were given
l e n g e r ( ~to
particular concern (Waltz, 1993; Katzenstein and Okawara, 1993) as was
the potential rise of an Islamic civilization (Huntington, 1993). The ques-
tion of appropriate US strategy remained open for a longer time.
Traditionalists also remained concerned with the strategic impact of tech-
nology, though now largely delinked from Cold War worries about nuclear
deterrence. Discussions about defenses against BMD, and US policy debates
ahout this, continued. So also did the debate about nuclear proliferation
(Walker, 2000), though increasingly both of these themes were linked to
so-called 'rogue states' (especially Iran, Iraq, North Korea) unwilling to
abide by Western rules of the game. The technology theme was also featured
in debate about the Revolution in Military Affairs (RMA, see Paarlberg,
2004), which was mainly concerned with the impact of computing, surveil-
lance and communications technologies on battlefield management.
Conceptually an important shift within traditionalist ISS was to expli-
citly integrate the importance of sub-state conflicts and actors. Cold War
Strategic Studies had only to a limited extent examined the impact of (or on)
xxxiv Editors' Introduction
non-state actors, an analytical decision supported by the fact that while both
the US and the Soviet Union worked to create allies, they did not generally
intervene in defense of populations caught in the throes of civil war or mur-
derous dictators. The end of the Cold War changed this as intra-state and eth-
nic conflicts became more prominent than inter-state wars and cases like the
former Yugoslavia and Somalia extensively covered by the global media.
Humanitarian interventions were undertaken in response with the goal of pro-
tecting civilian populations, although not necessarily to end the war itself
(Posen and Ross, 1996-7; Fierke, 1996). (neo)realists responded by arguing
that their theory was as adequate for analyzing conflicts at the intra-state level
as at the international one (Posen, 1993). It was not tied to state security, but
defined by the question of organized violence (Walt, 1991; Krause, 1996).
Traditionalists did not just pursue their own agenda, but mounted a counter-
attack against the drive towards widening the security agenda that had begun
during the 1980s (see previous section) and gathered strength during the
1990s (Nye and Lynn-Jones, 1988; Walt, 1991; Kolodziej, 1992; Baldwin,
1995; Betts, 1997). A significant change from the 1980s was that a more
explicit debate on security emerged. The concepts of 'wideners' and 'deepen-
ers' were coined (Krause and Williams, 1996), writers discussed where they
fitted into the disciplinary landscape and more schools and labels were intro-
duced in addition to Poststructuralism and Feminism which continued to
grow.
The least controversial wideners were those who followed the earlier
lead of taking a conception of national (state) security and linking military-
political security to questions of the environment, ethnicity, economy or health
while taking a largely objective and materialist (and often rather empirical)
approach. Another group of challengers was made up of mostly American
Constructivist scholars who worked with the concepts of culture, norms,
ideas and identity to show how these constrained and enabled state behav-
ior in ways that material explanations could not account for (Katzenstein
and Okawara, 1993; Risse-Kappen, 1996; Krause, 1996; Price, 1995; Adler,
1997). This form of Constructivism grew out of the larger epistemological
debate in IR between Rationalists and Reflectivists already under way by the
late 1980s and did not directly challenge state-centrism: it allowed for
ideational factors and non-state actors, particular international institutions,
but this was with the aim of explaining the pursuit of state security (usually
through the critique of Rationalist [Realist] Realist approaches), not to chal-
lenge its definition.
Gradually the Constructivist camp divided into a 'mainstream' and a
'critical' position and the latter shifted the focus from the ideational and
state-centric approach of 'mainstream Constructivism' to the study of lin-
guistic and narrative structures in foreign policy texts and to the way they
legitimized and produced security and identity for example in the American
responses to the Cuban Missiles Crisis or the Western policies towards the
Bosnian War (Weldes, 1996; Fierke, 1996). Critical Constructivists also
showed how competing representations were possible, but marginalized by
Editors' Introduction xxxv
state discourse. This move from Conventional to Critical Constructivism

implied a stronger concern with the politics of representation and hence a
substantial and epistemological affinity t o Poststructuralism.
Both forms of Constructivism were significant for bringing the study of
ideational, linguistic and discursive issues to the ISS mainstream, yet, Con-
structivists have not to a similar extent as the Copenhagen School, Critical
Security Studies, or the Human Security approach explicitly addressed the
concept of security. Most of the explicit widening debate was thus carried
out in Europe, in part perhaps due to a stronger PR influence and an ensu-
ing concern with 'security' rather than social science epistemologies. One
interpretation of this development is that ISS becomes less dominated by
American scholars, another that the discipline bifurcates into two (increas-
ingly separate) sets of debates, concepts and epistemologies.
Probably the most successful attempt to capture the attention of the ISS
mainstream while simultaneously reworking and widening the concept of
security has been made by the Copenhagen School. Its first major concep-
tual innovation was the concept of 'societal security' (Wasver et al., 1993)
which had 'society' as its referent object rather than the state (McSweeney,
1996; Krause and Williams, 1996; Huysmans, 1998). Societal security
referred to questions of collective identity (national, ethnic, racial, religious)
and was intended to provide a solution to the long-standing and unresolved
debate between the traditional concept of 'national' (state) security on the
one hand and the PR concepts of individual and global security on the
other. 'National security' only identified threats to the state, while 'individ-
ual security' had no theory of how individual insecurities were to be collect-
ively decided and prioritized (Wzver, 1995). 'Societal security' pointed to
threats posed to societal groups by the state and to the importance of soci-
etal cohesion and identity: individuals were part of society, not separate
from it, but their security was not identical to that of the state either.
The second major conceptual innovation was Ole Waxer's 'securitiza-
tion' theory which argued that 'security' was a speech act with a 'particu-
lar rhetorical and semiotic structure' that constituted something as 'an
existential threat' with a particular 'priority and urgency' that allowed 'the
securitizing actor' 'to break free of procedures or rules he or she would other-
wise be bound by' (Buzan et al., 1998: 25). As 'security' is a particular -
and exceptional - form o f politics, the Copenhagen School argued, it is not
necessarily preferable to constitute something as a security problem as this
might easily mobilize an us-them logic and take the issue out of 'normal
politics' and into the closed corridors of the Foreign Office. The politics and
ethics involved in securitization need therefore to be carefully considered in
the light o f exactly how security discourses constitute threats and subjects
and with what policy result (Elbe, 2006).
The ambition of 'securitization' paralleled 'societal security' in that it
too sought to solve the individual-national security dilemma. Security could
in principle be anything as long as it was discursively articulated as so by
securitizing actors. Securitizing actors articulate threats in reference to a
xxxvi Editors' Introduction
referent object usually located at the 'middle scale of limited collectivities',

larger than the individual and smaller than humanity (Buzan et al., 1998: 36).
However, since 'societal security is about large, self-sustaining identity
groups' (Buzan et al., 1998: 119), it becomes difficult to analyze the poten-
tial insecurity of groups or individuals whose identity is produced through
links to other identities (Hansen, 2000). Securitizing actors are usually top
politicians, representatives of the armed forces or international institutions
(NATO, the EU, the UN); the media and more marginal political actors
might in some cases succeed to push governments into action. The
Copenhagen School could thus be said to take a rather conventional view
on who defines security and hence also on how societal interests and iden-
tities are constructed and represented (McSweeney, 1996; Huysmans,
1998). The School has also been criticized for overlooking how an institu-
tionalized field of security comprised by the everyday practices of the police
and governmental bureaucracies is implicated in national security policies
(Bigo, 2002).
Securitization theory has spawned a growing empirical research agenda,
yet it has also generated criticism, particularly from approaches that advocate
a more radical expansion of the concept of security. One of the most signifi-
cant challengers has been CSS, specifically its so-called 'Welsh contingent',
whose most prominent scholars are Ken Booth and Richard Wyn Jones. To
Ken Booth, the Copenhagen School is 'state-centric, discourse-dominated,
and conservative' (Booth, 2005: 271). CSS can in some ways be seen as an
extension of the normative agenda of PR, but with Critical Theory instead of
positivist methodologies (Jones, 1995). In Booth's (1991: 319-20) view, 'indi-
vidual humans are the ultimate referent' for security as states are unreliable
providers of security and too diverse to provide for 'a comprehensive theory
of security'. For Wyn Jones (1995: 309), it is 'the experience of those men and
women and communities for whom the present world order is a cause of in-
security rather than security' and 'common humanity' which should be at the
heart of ISS. Central to CSS is the concept of emancipation, by Booth (1991:
3 19) defined as 'the freeing of people (as individuals and groups) from those
physical and human constraints which stop them carrying out what they
would freely choose to do'.
'Individual security' creates a link from the older PR traditions through
CSS to Human Security, a perspective that started out as a concept formu-
lated by UNDP (1994), but which has subsequently gained academic atten-
tion and a certain prominence in the policy world. The UNDP report clearly
opted for a widening of security arguing that, 'The concept of security has for
too long been interpreted narrowly: as security of territory from external
aggression, or as protection of national interests in foreign policy or as global
security from the threat of a nuclear holocaust. It has been related more to
nation-states than to people' (UNDP, 1994: 22). In response, Human Security
is 'a universal concern', it emphasizes interdependence between different
forms of threats, it encourages early prevention, and it is 'people-centred'
(UNDP, 1994: 22-23). Human Security in its original formulation was an
attempt to link development to security and its adoption by ISS has been
Editors' introduction xxxvii
predominantly as part of a CSS agenda or in an attack on neo-liberal eco-

nomics and globalization which are seen as deepening already stark global
inequalities (Thomas, 2001).
The concept of Human Security has been the subject of widespread aca-
demic critique most of which evolves around the vagueness of making 'the
human' the referent object (Paris, 2001). As 'individual security' before it,
it confronts the of how a globe of individual insecurities are to he
added up: if security is about assigning priorities t o particular issues, who
is going to define which individual insecurities are more important? And
who should be enforcing these decisions? Since 'humans' never appear as
atomistic entities, stripped of their (collective) identities, to define 'human
security' must imply a vision of how collective security is formulated and by
whom. These conceptual weaknesses have not however prevented Human
Security from becoming a prominent concept on the policy agenda of the
UN, the EU, and of active internationalist states like Canada and Norway
(Suhrke, 1999),and it has been supported by a series of foundations. What
we have is thus an interesting case of a security concept (much like Common
Security before it) making its way into ISS, not primarily through the schol-
arly community, but as a result of the pressure from (parts of) the driving
force of Institutionalization.
The terrorist attacks on the US on September 11, 2001 was definitely a key
event but whether or not the subsequent 'global War on Terror' (CWoT)
defines a new era of international security remains an open question.
Whether it does or not hinges on the American ability to securitize 'terror-
ism' as a ,global threat and have this securitization accepted by at least its
major allies and the great powers. If so (and the possibility is plausible if
not certain), then the GWoT would provide a new core framing for ISS of
a kind that has been absent since the end o f the Cold War. Certainly there
is a noticeable new preoccupation with security actors other than the state,
mainly terrorists (Barkawi, 2004; Der Derian, 2005), but also private mili-
tary companies. Concern with terrorism is of course not new, with the lit-
erature stretching back into the Cold War. But the earlier literature dealt
with terrorism as a peripheral problem to the main core of ISS concerns, not
the central one. The situation however is nothing like that at the early stages
of the Cold War when the identity of 'the enemy' crystallized quickly and
attracted a broad consensus in the West. The GWoT itself, and particularly
the characterization of 'terrorism' and the identity of 'terrorists' remain
heavily contested, and the Bush administration's portrait of itfthem has
done as much to divide the West as to unite it (Andreani, 2004-5). At the
time of writing it remains to be seen whether the unfolding disaster of the
US-led intervention in Iraq will do more to discredit or reinforce the GWoT.
Against the idea of a new era in ISS is the fact that the traditional pre-
occupations of great power politics remain strong and that this might make
other states reluctant to comply with the American articulation of the
xxxviii Editors' introduction
terrorist threat and in particular in granting the US a privileged position in

combating it. The US in turn might question the sincerity and ability of
other states to engage in the war, and thus embark on a classical balance of
power politics against rising states. The debate about US grand strategy
cited above is much less concerned with terrorism than with the thoroughly
traditional fixation on the balance of power, and the possible rise of great
power challengers to the US, principally China and the EU. Great power
politics could easily return to dominate the security agenda, though given
the arguments for democratic peace, that is not inevitable either. Terrorists
certainly do pose a potentially severe threat to public peace and order. But
they do not represent a plausible alternative political order in the way that the
Soviet Union did, and future great power challengers might do. Unless the
direst predictions about terrorists and weapons of mass destruction come to
pass, it may well be that the concern with terrorism that dominated the
agenda post September 11, might prove to be a transitory obsession rather
than something that defines a strategic era.
If the GWoT does prove durable, what does this signify for the direc-
tions of ISS? Was the agenda widening of the 1990s just a response to the
temporary eclipse of military concerns, or do the ideas of democratic peace,
societal and human security and the use of constructivist and other post-
positivist approaches suggest a deeper and more durable transformation? It
may well be too early to tell a mere six years from 2001, but so far the signs
are that ISS will retain its multi-stream mode. There is of course an adjust-
ment to take on board the GWoT as a topic, but far from returning ISS to
a single dominant mainstream, this has just given a new subject that all of the
approaches to ISS can and do address. All have to deal with the fact that
the securitization legitimating the GWoT evolves around an elusive subject -
'the networked Islamic terrorist' - which the Western states have difficulties
comprehending. Yet it is a subject that allows the state to suspend public
and private rights, to restrict immigration and to securitize a wide range of
'foreign' peoples and places.
Looking at the ISS literature since 2001 does not give the impression
that the form it took on during the 1990s is going to change in any major
way. Traditionalists are still concerned about US Grand Strategy (Kagan,
2002), nuclear proliferation, the RMA, and BMD. The Copenhagen School
continues to develop its ideas about securitization and regional security
(Buzan and Wzver, 2003). Poststructuralists pit their analytical techniques
against the GWoT (Der Derian, 2005) and the boundaries between domes-
tic and international security (Bigo, 2002). Feminists have analyzed the
gendering of the GWoT (Tickner, 2002; Agathangelou and Ling, 2004), and
advocates of Human Security and CSS continue their criticism of state-
centrism. Wideners point out the salience of threats from the environment and
elsewhere (Elbe, 2006), and those concerned with the conceptual histories of
security probe its origins and their significance for how its political usage has
evolved (Neocleous, 2006).
Editors' Introduction xxxix
ISS will continue to adapt, as it has done since its birth, to changes in the
public policy environment that define what constitutes international security
for any given time and place. It will thus continue to be sensitive to changes
in great power politics and the distribution of power, and to changes in tech-
nology that affect the use of force. It will be shoved and shaped by military-
political events such as 911 1 and the interventions in Iraq and Afghanistan, as
well as by events in other sectors, be it rises in sea level, the spread of a new
plague, or changes in the way identities play into the realm of security. It
seems likely that for some time yet the US will remain central to ISS (as both
subject and main generator of ISS literature), meaning that its domestic pol-
itics will continue to influence the field in a big way. And so long as academic
life goes on, one can be sure that new fashions and new ways of thinking will
make their impact on how ISS evolves.
What seems unlikely to us is that ISS will ever return to its Cold War form
of a single dominant sector and a single dominant mode of analysis. The
widening and deepenings that started during the Cold War and flowered dur-
ing the 1990s are here to stay, even though the balance of influence among
them will vary from place to place and time to time. All the new approaches
have put down roots, made recognized contributions to understanding, estab-
lished audiences for their work, and begun to reproduce themselves through
the training o f next generations. ISS has become, and will remain, the multi-
channeled field that was hinted at when the concept of security first rose to
prominence, but which got temporarily lost during the obsession with nuclear
deterrence that marked the 'golden age' of the Cold War.
References
~ ~ , M. and Ling, L.H.M. (2004) 'Power, Borders, Security, Wealth: Lessons

A g ~ t h a n g e l o Anna
of Violence ~ l n d Des~re from September I l ' , lntrrnational Studies Qirarterly, 4813):
517-38.
Hooth, Ken ( r d . ) (2005) Critrcal Security Studies and World Polltrcs. B o ~ ~ l d eCo.:r 1.ynne
K~enner.
Huzan, Rarry (1983) Peof~le,States 8 Fear: T h c Nat/onizl .Set-rrrrt)' Problcwz in Internatroni71
RcL7tions. Brighton: Wheatsheaf.
K u ~ a n ,Barry and W ~ v e r ,Ole (200.3) Regior~sand P(jtoers: The Structure of l ~ t ~ r n a t ~ o t z ~ z l
Security. C ~ ~ n b r i d g Cambridge
e: IJniversity Press.
Buyan, Karl-y, W m e r , Ole and de Wilde, J m p (1998) .Security: A Nezc~Frurnezlwrk for
A ~ z u l ~ s iBoulder
s. Co.: Lvnne Rienner.
Elshta~n,Jean B. (1987) W o m e n and War. Ch~cago:The Univers~cyof Chicdgo Press.
Galrung, Johan ( I 969) 'Violence, Peace, and Peace Research', Journal of Peace Research, 6(3):
167-91.
Keoh.ine, Robert 0. ( 1 988) 'Internat~onalInstitutions: Two Approaches', Intertzati~~nnl Studies
Qutrrterly, 32(4): 379-96.
Tickner, J. A n n (2002) 'Feminist Perspectives on 911 l', lnternatzond Studies P c r ~ p ~ r t i v e3(4):
s,
333-50.
Wsver, Ole, Buzan, Barry, Kelstrup, Morten and Lemaitre, Pierre (1993) Identity, Migmtion
m i l the Nrtij S ~ c ~ i r iAgenda
ty in Europe. London: Pinter.
XI Editors' Introduction
Recommended Books on ISS
Allison, Graham (1971) Essence of Decis~on:Explaining the Cuba Missile Crisis. Boston:
Little Brown.
Bigo, Didier (2007) Polrcing Insecurity Today: Defense and Internal Security. Bas~ngstoke:
Palgrave Macmillan.
Brodie, Bernard (1946) The Absolute Weapon: Atomic Pouter and World Order. New York:
Harcourt Brace.
Bull, Hedley (1961) The Control of the Arms Race. London: Weidenfeld and Nicholson.
Buzan, Barry (1987) An lntroductron t o Strategic Studies: Military Technology m d
International Relations. London: Macmillan.
Campbell, David (1998) Writing Security: United States Foreign Policy arid the Politrcs of
Identity. Minneapolis: University of Minnesota Press, 2nd edition.
Evangelists, Matthew (1988)Innovation and the Arms Race: How the United States and the
Soviet Union Develop N e w Military Technologies. Ithaca: Cornell University Press.
Freedman, Lawrence (1981)The Evolution of Nuclear Strategy. London: ~Macmillan.
Gray, Colin (1982) Strategic Studies: A Critical Assessment. London: Aldwych Press.
Hansen, Lene (2006) Security as Practice: Discourse Analysis and the Bosnian War. Imndon:
Routledge.
Jervis, Robert (1976) Perception and Misperception in lnternatronal Politics. Princeton, NJ:
Princeton University Press.
Kahn, Herman (1962) ~ h i n k r About
n~ the Unthinkable. New York: Horizon Press.
Kaplan, Fred M. (1983) The Wizards of Armageddon. New York: Simon and Schuster.
Kapstein, Ethan B. and Mastanduno, Michael (eds.) (1999) Unipolar Politics: Reolisnr and
State Strategies after the Cold War. New York: Columbia University Press.
Katzenstein, Peter (ed.) (1996) The Culture of Natiotzal Security. New York: Columbia
University Press.
Knorr, Klaus (1966) O n the Uses of Military Power In the Nuclear Age. Princeton, N J :
Kolodziej, Edward A. (2005) Security and International Relations. Cambridge: Cambridge
University Press.
Mearsheimer, John (2001) The Tragedy of Great Power Politics. New York: Norton.
Morgan, Patrick (1977, 1983) Deterrence: A Conceptrral Analysrs. London: Sage.
Rapoport, Anatol (1964) Strategy and Conscience. New York: Harper and Row.
Russett, Bruce (1983) The Prisoners of Insecurrty. San Francisco: Freeman.
Schelling, Thomas (1966)Arms and Influence. New Ha5en: Yale University Press.
Sheehan, Michael (2005) International Security: A n Analytical Survey. Boulder Co.: Lynne
Rienner.
Singer, J. David (1962) Deterrence, Arms Control and Disarmament. Columbus Ohio: Ohio
State University Press.
Tickner, J. Ann (2001) Gendering World Politrcs: Issues and Approaches rn the Post-Cold War
Era. New York: Columbia University Press.
The Sources o f Soviet Conduct
T he political personality of Soviet power as we know it today is the

product of ideology and circumstances: ideology inherited by the
present Soviet leaders from the movement in which they had their
political origin, and circumstances of the power which they now have exer-
cised for nearly three decades in Russia. There can be few tasks of psycho-
logical analysis more difficult than to try to trace the interaction of these
two forces and the relative r6le of each in the determination of official
Soviet conduct. Yet the attempt must be made if that conduct is to he under-
stood and effectively countered.
It is difficult to summarize the set of ideological concepts with which the
Soviet leaders came into power. Marxian ideology, in its Russian-Communist
projection, has always been in process of subtle evolution. The materials on
which it bases itself are extensive and complex. But the outstanding features
of Communist thought as it existed in 1916 may perhaps be summarized as
follows: ( a ) that the central factor in the life of man, the factor which deter-
mines the character of public life and the "physiognomy of society," is the
system by which material goods are produced and exchanged; ( b ) that the
capitalist system of production is a nefarious one which inevitably leads
to the exploitation of the working class by the capital-owning class and is
incapable of developing adequately the econonlic resources of society or of
distributing fairly the material goods produced by human labor; (c)that cap-
italism contains the seeds of its own destruction and must, in view of the
inability of the capital-owning class to adjust itself to economic change, result
eventually and inescapably in a revolutionary transfer of power to the work-
ing class; and ( d )that imperialism, the final phase of capitalism, leads directly
to war and revolution.
The rest may be outlined in Lenin's own words: "Unevenness of eco-
nomic and political development is the inflexible law of capitalism. It fol-
lows from this that the victory of Socialism may come originally in a few
capitalist countries or even in a single capitalist country. The victorious
proletariat of that country, having expropriated the capitalists and having
Source: korezgn Affnrrs, 25(4) (1947):566-82.
2 T h e Cold War a n d Nuclear D e t e r r e n c e
organized Socialist production at home, would rise against the remaining

capitalist world, drawing to itself in the process the oppressed classes
of other countries."' It must be noted that there was no assumption that
capitalism would perish without proletarian revolution. A final push was
needed from a revolutionary proletariat movement in order to tip over the
tottering structure. But it was regarded as inevitable that sooner or later
that push be given.
For 50 years prior to the outbreak of the Revolution, this pattern of
thought had exercised great fascination for the members of the Russian
revolutionary movement. Frustrated, discontented, hopeless of finding self-
expression - or too impatient to seek it - in the confining limits of the
Tsarist political system, yet lacking wide popular support for their choice of
bloody revolution as a means of social betterment, these revolutionists
found in Marxist theory a highly convenient rationalization for their own
instinctive desires. It afforded pseudo-scientific justification for their impa-
tience, for their categoric denial of all value in the Tsarist system, for their
yearning for power and revenge and for their inclination to cut corners in
the pursuit of it. It is therefore n o wonder that they had come to believe
implicitly in the truth and soundness of the Marxian-Leninist teachings, so
congenial t o their own impulses and emotions. Their sincerity need not be
impugned. This is a phenomenon as old as human nature itself. It has never
been more aptly described than by Edward Gibbon, who wrote in "The
Decline and Fall of the Roman Empire": "From enthusiasm to imposture
the step is perilous and slippery; the demon of Socrates affords a memorable
instance how a wise man may deceive himself, how a good man may
deceive others, how the conscience may slumber in a mixed and middle
state between self-illusion and voluntary fraud." And it was with this set of
conceptions that the members of the Bolshevik Party entered into power.
Now it must be noted that through all the years of preparation for revo-
lution, the attention of these men, as indeed of Marx himself, had been cen-
tered less on the future form which Socialism2 would take than on the
necessary overthrow of rival power which, in their view, had to precede the
introduction of Socialism. Their views, therefore, on the positive program
to be put into effect, once power was attained, were for the most part nebu-
lous, visionary and impractical. Beyond the nationalization of industry and
the expropriation of large private capital holdings there was no agreed
program. The treatment of the peasantry, which according to the Marxist
formulation was not of the proletariat, had always been a vague spot in the
pattern of Communist thought; and it remained an object of controversy
and vacillation for the first ten years of Communist power.
The circumstances of the immediate post-revolution period - the exist-
ence in Russia of civil war and foreign intervention, together with the obvi-
ous fact that the Communists represented only a tiny minority of the Russian
people - made the establishment of dictatorial power a necessity. The experi-
ment with "war Communism" and the abrupt attempt to eliminate private
production and trade had unfortunate economic consequences and caused
Y The Sources of Soviet Conduct 3
further bitterness against the new revolutionary regime. While the tempor-
ary relaxation of the effort to communize Russia, represented by the New
Economic Policy, alleviated some of this economic distress and thereby
served its purpose, it also made it evident that the "capitalistic sector of
society" was still prepared to profit at once from any relaxation of govern-
mental pressure, and would, if permitted to continue to exist, always con-
stitute a powerful opposing element to the Soviet rCgime and a serious rival
for influence in the country. Somewhat the same situation prevailed with
respect to the individual peasant who, in his own small way, was also a pri-
vate producer.
Lenin, had he lived, might have proved a great enough man to reconcile
these conflicting forces to the ultimate benefit of Russian society, though this
is questionable. But be that as it may, Stalin, and those whom he led in the
struggle for succession to Lenin's position of leadership, were not the men to
tolerate rival political forces in the sphere of power which they coveted.
Their sense of insecurity was too great. Their particular brand of fanaticism,
unmodified by any of the Anglo-Saxon traditions of compromise, was too
fierce and too jealous to envisage any permanent sharing of power. From the
Russian-Asiatic world out of which they had emerged they carried with them
a skepticism as to the possibilities of permanent and peaceful coexistence
of rival forces. Easily persuaded of their own doctrinaire "rightness," they
insisted on the submission or destruction of all competing power. Outside of
the Communist Party, Russian society was to have no rigidity. There were to
be no forms of collective human activity or association which would not be
dominated by the Party. N o other force in Russian society was to be per-
mitted to achieve vitality or integrity. Only the Party was to have structure.
All else was to be an amorphous mass.
And within the Party the same principle was to apply. The mass of Party
members might go through the motions of election, deliberation, decision
and action; but in these motions they were to be animated not by their own
individual wills but by the awesome breath of the Party leadership and the
overbrooding presence of "the word."
Let it be stressed again that subjectively these men probably did not seek
absolutism for its own sake. They doubtless believed - and found it easy to
believe - that they alone knew what was good for society and that they
would accomplish that good once their power was secure and unchallenge-
able. But in seeking that security of their own rule they were prepared to
recognize no restrictions, either of God or man, on the character of their
methods. And until such time as that security might be achieved, they
placed far down on their scale of operational priorities the comforts and
happiness of the peoples entrusted to their care.
Now the outstanding circumstance concerning the Soviet rkgime is that
down to the present day this process of political consolidation has never
been completed and the men in the Kremlin have continued to be predom-
inantly absorbed with the struggle to secure and make absolute the power
which they seized in November 1917. They have endeavored to secure it
4 T h e Cold War a n d Nuclear Deterrence
primarily against forces at home, within Soviet society itself. But they have
also endeavored to secure it against the outside world. For ideology, as we
have seen, taught them that the outside world was hostile and that it was
their duty eventually to overthrow the political forces beyond their borders.
The powerful hands of Russian history and tradition reached up to sustain
them in this feeling. Finally, their own aggressive intransigence with respect
to the outside world began to find its own reaction; and they were soon
forced, to use another Gibbonesque phrase, "to chastise the contumacy"
which they themselves had provoked. It is an undeniable privilege of every
man to prove himself right in the thesis that the world is his enemy; for if
he reiterates it frequently enough and makes it the background of his con-
duct he is bound eventually t o be right.
Now it lies in the nature of the mental world of the Soviet leaders, as well
as in the character of their ideology, that no opposition to them can be offi-
cially recognized as having any merit or justification whatsoever. Such oppos-
ition can flow, in theory, only from the hostile and incorrigible forces of dying
capitalism. As long as remnants of capitalism were officially recognized as
existing in Russia, it was possible to place on them, as an internal element,
part of the blame for the maintenance of a dictatorial form of society. But as
these remnants were liquidated, little by little, this justification fell away; and
when it was indicated officially that they had been finally destroyed, it dis-
appeared altogether. And this fact created one of the most basic of the com-
pulsions which came to act upon the Soviet regime: since capitalism no longer
existed in Russia and since it could not be admitted that there could be ser-
ious or widespread opposition to the Kremlin springing spontaneously from
the liberated masses under its authority, it became necessary to justify the
retention of the dictatorship by stressing the menace of capitalism abroad.
This began at an early date. In 1924 Stalin specifically defended the
retention of the "organs of suppression," meaning, among others, the army
and the secret ~ o l i c e ,on the ground that "as long as there is a capitalist
encirclement there will be danger of intervention with all the consequences
that flow from that danger." In accordance with that theory, and from that
time on, all internal opposition forces in Russia have consistently been por-
trayed as the agents of foreign forces of reaction antagonistic to Soviet power.
By the same token, tremendous emphasis has been placed on the original
Communist thesis of a basic antagonism between the capitalist and Socialist
worlds. It is clear, from many indications, that this emphasis is not founded
in reality. The real facts concerning it have been confused by the existence
abroad of genuine resentment provoked by Soviet philosophy and tactics
and occasionally by the existence of great centers of military power, notably
the Nazi rCgime in Germany and the Japanese Government of the late
1930's, which did indeed have aggressive designs against the Soviet Union.
But there is ample evidence that the stress laid in Moscow on the menace
confronting Soviet society from the world outside its borders is founded not
in the realities of foreign antagonism but in the necessity of explaining away
the maintenance of dictatorial authority at home.
\ T h e S o u r c e s of Soviet Conduct 5
Now the maintenance of this pattern of Soviet power, namely, the pursuit
of unlimited authority domestically, accompanied by the cultivation of the
semi-myth of implacable foreign hostility, has gone far to shape the actual
machinery of Soviet power as we know it today. Internal organs of admin-
istration which did not serve this purpose withered on the vine. Organs
which did serve this purpose became vastly swollen. The security of Soviet
power came to rest on the iron discipline of the Party, on the severity and
ubiquity of the secret police, and on the uncompromising economic monop-
olism of the state. The "organs of suppression," in which the Soviet leaders
had sought security from rival forces, became in large measure the masters
of those whom they were designed to serve. Today the major part of the
structure of Soviet power is committed to the perfection of the dictatorship
and to the maintenance of the concept of Russia as in a state of siege, with
the enemy lowering beyond the walls. And the millions of human beings
who form that part of the structure of power must defend at all costs this
concept of Russia's position, for without it they are themselves superfluous.
As things stand today, the rulers can no longer dream of parting with
these organs of suppression. The quest for absolute power, pursued now for
nearly three decades with a ruthlessness unparalleled (in scope at least) in
modern times, has again produced internally, as it did externally, its own
reaction. The excesses of the police apparatus have fanned the potential
opposition to the regime into something far greater and more dangerous
than it could have been before those excesses began.
But least of all can the rulers dispense with the fiction by which the
maintenance of dictatorial power has been defended. For this fiction has
hcen canonized in Soviet philosophy by the excesses already committed in
its name; and it is now anchored in the Soviet structure of thought by bonds
far greater than those of mere ideology.
So much for the historical background. What does it spell in t e r m of the

political personality of Soviet power as we know it today?
Of the original ideology, nothing has been officially junked. Relief is
maintained in the basic badness of capitalism, in the inevitability of its
destruction, in the obligation of the to assist in that destruction
and to take power into its own hands. But stress has come to be laid pri-
marily on those concepts which relate most specifically to the Soviet regime
itself: to its position as the sole truly Socialist rkgime in a dark and mis-
guided world, and to the relationships of power within it.
The first of these concepts is that of the innate antagonism between cap-
italism and Socialism. We have seen how deeply that concept has become
imbedded in foundations of Soviet power. It has profound implications for
Russia's conduct as a member of international society. It means that there
can never be on Moscow's side any sincere assumption of a community of
6 The Cold War and Nuclear Deterrence
aims between the Soviet Union and powers which are regarded as capitalist.
It must invariably be assumed in Moscow that the aims of the capitalist
world are antagonistic to the Soviet regime, and therefore to the interests of
the peoples it controls. If the Soviet Government occasionally sets its signa-
ture t o documents which would indicate the contrary, this is to be regarded
as a tactical maneuvre permissible in dealing with the enemy (who is with-
out honor) and should be taken in the spirit of caveat emptor. Basically, the
antagonism remains. It is postulated. And from it flow many of the phe-
nomena which we find disturbing in the Kremlin's conduct of foreign policy:
the secretiveness, the lack of frankness, the duplicity, the wary suspicious-
ness, and the basic unfriendliness of purpose. These phenomena are there to
stay, for the foreseeable future. There can be variations of degree and of
emphasis. When there is something the Russians want from us, one or the
other of these features of their policy may be thrust temporarily into the
background; and when that happens there will always be Americans who
will leap forward with gleeful announcements that "the Russians have
changed," and some who will even try to take credit for having brought
about such "changes." But we should not be misled by tactical maneuvres.
These characteristics of Soviet policy, like the postulate from which they
flow, are basic to the internal nature of Soviet power, and will be with us,
whether in the foreground or the background, until the internal nature of
Soviet power is changed.
This means that we are going to continue for a long time to find the
Russians difficult to deal with. It does not mean that they should be consid-
ered as embarked upon a do-or-die program to overthrow our society by a
given date. The theory of the inevitability of the eventual fall of capitalism has
the fortunate connotation that there is no hurry about it. The forces of
progress can take their time in preparing the final coup de grice. Meanwhile,
what is vital is that the "Socialist fatherland" - that oasis of power which has
been already won for Socialism in the person of the Soviet Union - should be
cherished and defended by all good Communists at home and abroad, its for-
tunes promoted, its enemies badgered and confounded. The promotion of pre-
mature, "adventuristic" revolutionary projects abroad which might embarrass
Soviet power in any way would be an inexcusable, even a counter-revolutionary
act. The cause of Socialism is the support and promotion of Soviet power, as
defined in Moscow.
This brings us to the second of the concepts important to contemporary
Soviet outlook. That is the infallibility of the Kremlin. The Soviet concept
of power, which permits no focal points of organization outside the Party
itself, requires that the Party leadership remain in theory the sole repository
of truth. For if truth were t o be found elsewhere, there would be justifica-
tion for its expression in organized activity. But it is precisely that which the
Kremlin cannot and will not permit.
The leadership of the Communist Party is therefore always right, and
has been always right ever since in 1929 Stalin formalized his personal
X The Sources of Soviet Conduct 7
power by announcing that decisions of the Politburo were being taken

unanin~ously.
O n the principle of infallibility there rests the iron discipline of the
Communist Party. In fact, the two concepts are mutually self-supporting.
Perfect discipline requires recognition of infallibility. Infallibility requires the
observance of discipline. And the two together go far to determine the behav-
iorism of the entire Soviet apparatus of power. But their effect cannot be
understood unless a third factor be taken into account: namely, the fact that
the leadership is at liberty to put forward for tactical purposes any particular
thesis which it finds useful to the cause a t any particular moment and to
require the faithful and unquestioning acceptance of that thesis by the meni-
hers of the movement as a whole. This means that truth is not a constant but
is actually created, for all intents and purposes, by the Soviet leaders them-
selves. It may vary from week to week, from month to month. It is nothing
absolute and immutable - nothing which flows from objective reality. It is
only the most recent manifestation of the wisdom of those in whom the
ultimate wisdom is supposed to reside, because they represent the logic of his-
tory. The accumulative effect of these factors is to give to the whole subordin-
ate apparatus of Soviet power an unshakeable stubbornness and steadfastness
in its orientation. This orientation can he changed at will by the Kremlin but
hy n o other power. Once a given party line has been laid down on a given
issue of current policy, the whole Soviet governmental machine, including the
mechanism of diplomacy, moves inexorably along the prescribed path, like a
persistent toy automobile wound up and headed in a given direction, stop-
ping only when it meets with some unanswerable force. The individuals who
are the components of this machine are unamenable t o argument or reason
which comes to them from outside sources. Their whole training has taught
them to mistrust and discount the glib persuasiveness of the outside world.
Like the white dog before the phonograph, they hear only the "master's
voice." And if they are to be called off from the purposes last dictated to
them, it is the master who must call them off. Thus the foreign representative
cannot hope that his words will make any impression on them. The most that
he can hope is that they will be transmitted to those a t the top, who are cap-
able of changing the party line. But even those are not likely to be swayed by
any normal logic in the words of the bourgeois representative. Since there can
be n o appeal t o common purposes, there can be n o appeal to common men-
tal approaches. For this reason, facts speak louder than words to the ears of
the Krenllin; and words carry the greatest weight when they have the ring of
reflecting, or being backed up by, facts of unchallengeable validity.
But we have seen that the Kremlin is under no ideological compulsion to
accon~plishits purposes in a hurry. Like the Church, it is dealing in ideo-
logical concepts which are of long-term validity, and it can afford t o be
patient. It has no right to risk the existing achievements of the revolution for
the sake o f vain baubles of the future. The very teachings of Lenin himself
require great caution and flexibility in the pursuit of Communist purposes.
8 The C o l d War a n d N u c l e a r D e t e r r e n c e
Again, these precepts are fortified by the lessons of Russian history: of cen-
turies of obscure battles between nomadic forces over the stretches of a vast
unfortified plain. Here caution, circumspection, flexibility and deception are
the valuable qualities; and their value finds natural appreciation in the
Russian or the oriental mind. Thus the Kremlin has no compunction about
retreating in the face of superior force. And being under the compulsion of
no timetable, it does not get panicky under the necessity for such retreat. Its
political action is a fluid stream which moves constantly, wherever it is per-
mitted to move, toward a given goal. Its main concern is to make sure that
it has filled every nook and cranny available to it in the basin of world
power. But if it finds unassailable barriers in its path, it accepts these philo-
sophically and accommodates itself to them. The main thing is that there
should always be pressure, unceasing constant pressure, toward the desired
goal.
- There is no trace of any feeling in Soviet psychology that that goal must
be reached at any given time.
These considerations make Soviet diplomacy at once easier and more
difficult to deal with than the diplomacy of individual aggressive leaders
like Napoleon and Hitler. O n the one hand it is more sensitive to contrary
force, more ready to yield on individual sectors of the diplomatic front
when that force is felt to be too strong, and thus more rational in the logic
and rhetoric of power. O n the other hand it cannot be easily defeated or dis-
couraged by a single victory on the part of its opponents. And the patient
persistence by which it is animated means that it can be effectively coun-
tered not by sporadic acts which represent the momentary whims of demo-
cratic opinion but only by intelligent long-range policies on the part of
Russia's adversaries - policies no less steady in their purpose, and no less
variegated and resourceful in their application, than those of the Soviet
Union itself.
In these circumstances it is clear that the main element of any United
States policy toward the Soviet Union must be that of a long-term, patient
but firm and vigilant containment of Russian expansive tendencies. It is
important to note, however, that such a policy has nothing to d o with out-
ward histrionics: with threats or blustering or superfluous gestures of out-
ward "toughness." While the Kremlin is basically flexible in its reaction to
political realities, it is by no means unamenable to considerations of pres-
tige. Like almost any other government, it can be placed by tactless and
threatening gestures in a position where it cannot afford to yield even
though this might be dictated by its sense of realism. The Russian leaders
are keen judges of human psychology, and as such they are highly conscious
that loss of temper and of self-control is never a source of strength in polit-
ical affairs. They are quick to exploit such evidences of weakness. For these
reasons, it is a sine qua non of successful dealing with Russia that the for-
eign government in question should remain at all times cool and collected
and that its demands on Russian policy should be put forward in such a
manner as to leave the way open for a compliance not too detrimental to
Russian prestige.
The Sources of Soviet Conduct 9
In the light of the above, it will be clearly seen that the Soviet pressure
against the free institutions of the western world is something that can be
contained by the adroit and vigilant application of counter-force a t a series
of constantly shifting geographical and political points, corresponding to
the shifts and manceuvres of Soviet policy, but which cannot be charmed o r
talked out of existence. The Russians look forward t o a duel of infinite
duration, and they see that already they have scored great successes. It must
be borne in mind that there was a time when the Communist Party repre-
sented far more of a minority in the sphere of Russian national life than
Soviet power today represents in the world community.
But if ideology convinces the rulers of Russia that truth is on their side
and that they can therefore afford to wait, those of us o n w h o m that ideol-
ogy has n o claim are free to examine objectively the validity o f that prem-
ise. The Soviet thesis not only implies complete lack of control by the west
over its own economic destiny, it likewise assumes Russian unity, discipline
and patience over an infinite period. Let us bring this apocalyptic vision
down to earth, and suppose that the western world finds the strength and
resourcef~~lness to contain Soviet power over a period of ten to fifteen years.
What does that spell for Russia itself?
The Soviet leaders, taking advantage of the contributions of modern
technique to the arts of despotism, have solved the question of obedience
within the confines o f their power. Few challenge their authority; and even
those w h o d o are unable t o make that challenge valid as against the organs
of suppression of the state.
The Kremlin has also proved able t o accomplish its purpose of building
up in Russia, regardless of the interests of the inhabitants, an industrial
foundation of heavy metallurgy, which is, t o be sure, not yet complete but
which is nevertheless continuing t o grow and is approaching those of the
other major industrial countries. All of this, however, both the maintenance
of internal political security and the building of heavy industry, has been
carried out a t a terrible cost in human life and in human hopes and ener-
gies. It has necessitated the use of forced labor o n a scale unprecedented in
modern times under conditions of peace. It has involved the neglect o r
abuse of other phases of Soviet economic life, particularly agriculture, con-
sumers' goods production, housing and transportation.
To all that, the war has added its tren~endoustoll of destruction, death
and human exhaustion. In consequence of this, we have in R L I S Stoday ~~ a
population which is physically a n d spiritually tired. The mass of the people
are disillusioned, skeptical and n o longer as accessible as they once were t o
the magical attraction which Soviet power still radiates to its followers
ahroad. The avidity with which people seized upon the slight respite
accorded t o the Church for tactical reasons during the war was eloquent
testimony to the fact that their capacity for faith and devotion found little
expression in the purposes of the rigime.
In these circumstances, there are limits to the physical and nervous strength
of people themselves. These limits are absolute ones, and are binding even for
the cruelest dictatorship, because beyond them people cannot be driven. The
forced labor camps and the other agencies of constraint provide temporary
means of compelling people to work longer hours than their own volition or
mere economic pressure would dictate; but if people survive them at all they
become old before their time and must be considered as human casualties to
the demands of dictatorship. In either case their best powers are no longer
available to society and can no longer be enlisted in the service of the state.
Here only the younger generation can help. The younger generation,
despite all vicissitudes and sufferings, is numerous and vigorous; and the
Russians are a talented people. But it still remains to be seen what will be
the effects on mature performance of the abnormal emotional strains of
childhood which Soviet dictatorship created and which were enormously
increased by the war. Such things as normal security and placidity of home
environment have practically ceased to exist in the Soviet Union outside of
the most remote farms and villages. And observers are not yet sure whether
that is not going to leave its mark on the over-all capacity of the generation
now coming into maturity.
In addition to this, we have the fact that Soviet economic development,
while it can list certain formidable achievements, has been precariously
spotty and uneven. Russian Communists who speak of the "uneven develop-
ment of capitalism" should blush at the contemplation of their own national
economy. Here certain branches of economic life, such as the metallurgical
and machine industries, have been pushed out of all proportion to other sec-
tors of economy. Here is a nation striving to become in a short period one of
the great industrial nations of the world while it still has no highway network
worthy of the name and only a relatively primitive network of railways.
Much has been done to increase efficiency of labor and to teach primitive
peasants something about the operation of machines. But maintenance is still
a crying deficiency of all Soviet economy. Construction is hasty and poor in
quality. Depreciation must be enormous. And in vast sectors of economic life
it has not yet been possible to instill into labor anything like that general cul-
ture of production and technical self-respect which characterizes the skilled
worker of the west.
It is difficult to see how these deficiencies can be corrected at an early date
by a tired and dispirited population working largely under the shadow of fear
and compulsion. And as long as they are not overcome, Russia will remain
economically a vulnerable, and in a certain sense an impotent, nation, capable
of exporting its enthusiasms and of radiating the strange charm of its primi-
tive political vitality but unable to back up those articles of export by the real
evidences of material power and prosperity.
Meanwhile, a great uncertainty hangs over the political life of the Soviet
Union. That is the uncertainty involved in the transfer of power from one
individual or group of individuals to others.
': The Sources of Soviet Conduct II
This is, of course, outstandingly the problem of the personal position of

Stalin. We must remember that his succession to Lenin's pinnacle of pre-
eminence in the Communist movement was the only such transfer of indi-
vidual authority which the Soviet Union has experienced. That transfer took
12 years to consolidate. It cost the lives of n~illionsof people and shook the
state to its foundations. The attendant tremors were felt all through the inter-
national revolutionary movement, to the disadvantage of the Kremlin itself.
It is always possible that another transfer of preeminent power may
take place quietly and inconspicuously, with no repercussions anywhere.
But again, it is possible that the questions involved may unleash, to use some
of Lenin's words, one of those "incredibly swift transitions" from "delicate
deceit" to "wild violence" which characterize Kussian history, and may shake
Soviet power to its foundations.
But this is not only a question of Stalin himself. There has been, since
1938, a dangerous congealment of political life in the higher circles of Soviet
power. The All-Union Congress of Soviets, in theory the supreme body of the
Party, is supposed to meet not less often than once in three years. It will soon
be eight full years since its last meeting. During this period membership in
the Party has nun~ericallydoubled. Party mortality during the war was enor-
mous; and today well over half of the Party members are persons who have
entered since the last Party congress was held. Meanwhile, the same small
group of men has carried on at the top through an amazing series of national
vicissitudes. Surely there is some reason why the experiences o f the war
brought hasic political changes to every one of the great governments of the
west. Surely the causes of that phenomenon are basic enough to be present
somewhere in the obscurity of Soviet political life, as well. And yet no recog-
nition has been given to these causes in Russia.
It must be surmised from this that even within so highly disciplined an
organization as the Communist Party there must be a growing divergence
in age, outlook and interest between the great mass of Party members, only
so recently recruited into the movement, and the little self-perpetuating
clique of men at the top, whom most of these Party members have never
met, with whom they have never conversed, and with whom they can have
no political intimacy.
Who can say whether, in these circumstances, the eventual rejuvenation of
the higher spheres of authority (which can only be a matter of time) can take
place smoothly and peacefully, or whether rivals in the quest for higher power
will not eventually reach down into these politically immature and inexperi-
enced masses in order to find support for their respective claims? If this were
ever to happen, strange consequences could flow for the Communist Party:
for the membership at large has been exercised only in the practices of iron
discipline and obedience and not in the arts of compromise and accommoda-
tion. And if disunity were ever to seize and paralyze the Party, the chaos and
weakness of Russian society would be revealed in forms beyond description.
For we have seen that Soviet power is only a crust concealing an amorphous
mass of human beings among whom no independent organizational structure

is tolerated. In Russia there is not even such a thing as local government. The
present generation of Russians have never known spontaneity of collective
action. If, consequently, anything were ever to occur to disrupt the unity and
efficacy of the Party as a political instrument, Soviet Russia might be changed
overnight from one of the strongest to one of the weakest and most pitiable
of national societies.
Thus the future of Soviet power may not be by any means as secure as
Russian capacity for self-delusion would make it appear to the men in the
Kremlin. That they can keep power themselves, they have demonstrated.
That they can quietly and easily turn it over to others remains to be proved.
Meanwhile, the hardships of their rule and the vicissitudes of international
life have taken a heavy toll of the strength and hopes of the great people on
whom their power rests. It is curious to note that the ideological power of
Soviet authority is strongest today in areas beyond the frontiers of Russia,
beyond the reach of its police power. This phenomenon brings to mind a
comparison used by Thomas Mann in his great novel "Buddenbrooks."
Observing that human institutions often show the greatest outward bril-
liance at a moment when inner decay is in reality farthest advanced, he
compared the Buddenbrook family, in the days of its greatest glamour, to
one of those stars whose light shines most brightly on this world when in
reality it has long since ceased to exist. And who can say with assurance
that the strong light still cast by the Kremlin on the dissatisfied peoples of
the western world is not the powerful afterglow of a constellation which is
in actuality on the wane? This cannot be proved. And it cannot be dis-
proved. But the possibility remains (and in the opinion of this writer it is a
strong one) that Soviet power, like the capitalist world of its conception,
bears within it the seeds of its own decay, and that the sprouting of these
seeds is well advanced.
It is clear that the United States cannot expect in the foreseeable future to
enjoy political intimacy with the Soviet regime, It must continue to regard
the Soviet Union as a rival, not a partner, in the political arena. It must con-
tinue to expect that Soviet policies will reflect no abstract love of peace and
stability, no real faith in the possibility of a permanent happy coexistence of
the Socialist and capitalist worlds, but rather a cautious, persistent pressure
toward the disruption and weakening of all rival influence and rival power.
Balanced against this are the facts that Russia, as opposed to the western
world in general, is still by far the weaker party, that Soviet policy is highly
flexible, and that Soviet society may well contain deficiencies which will
eventually weaken its own total potential. This would of itself warrant the
United States entering with reasonable confidence upon a policy of firm con-
tainment, designed to confront the Russians with unalterable counter-force
The Sources of Soviet Conduct I3
a t every point where they show signs of encroaching upon the interests of a
peaceful and stable world.
But in actuality the possibilities for American policy are by n o means
limited t o holding the line and hoping for the best. It is entirely possible for
the United States t o influence by its actions the internal developments, both
within Russia and throughout the international Communist movement, by
which Russian policy is largely determined. This is not only a question of
the modest measure of informational activity which this government can
conduct in the Soviet Union and elsewhere, although that, too, is important.
It is rather a question of the degree t o which the United States can create
among the peoples of the world generally the impression of a country which
knows w h a t it wants, which is coping successfully with the problems of its
internal life and with the responsibilities of a World Power, and which has
a spiritual vitality capable of holding its o w n among the major ideological
currents of the time. To the extent that such an impression can be created
and maintained, the aims of Kussian Communism must appear sterile and
quixotic, the hopes and enthusiasm of Moscow's supporters 111ust wane,
and added strain must be imposed on the Kremlin's foreign policies. For the
palsied decrepitude of the capitalist world is the keystone of Communist
philosophy. Even the failure of the United States t o experience the early eco-
nomic depression which the ravens of the Red Square have been predicting
with such complacent confidence since hostilities ceased would have deep
and important repercussions throughout the Communist world.
Ky the same token, exhibitions of indecision, disunity and internal dis-
integration within this country have a n exhilarating effect o n the whole
Communist movement. At each evidence of these tendencies, a thrill of hope
and excitement goes through the Communist world; a new jauntiness can be
noted in the Moscow tread; new groups of Foreign supporters climb on t o
what they can only view as the band wagon of international politics; and
Russian pressure increases all along the line in international affairs.
It would be an exaggeration t o say that American behavior unassisted
atid alone could exercise a power of life and death over the Communist
movement and bring about the early fall of Soviet power in Russia. But the
United States has it in its power t o increase enormously the strains under
which Soviet policy must operate, t o force upon the Kremlin a far greater
degree o f moderation and circumspection than it has had t o observe in
recent years, and in this way t o promote tendencies which must eventually
find their outlet in either the break-up o r the gradual mellowing of Soviet
power. For n o mystical, Messianic movement - and particularly not that of
the Kremlin - can face frustration indefinitely without eventually adjusting
itself in one way o r another t o the logic of that state of affairs.
Thus the decision will really fall in large measure in this country itself.
T h e issue of Soviet-American relations is in essence a test of the over-all
worth o f the United States a s a nation among nations. To avoid destruction
the United States need only measure up t o its own best traditions and prove
itself worthy of preservation as a great nation.
Surely, there was never a fairer test of national quality than this. In the light
of these circumstances, the thoughtful observer of Russian-American relations
will find no cause for complaint in the Kremlin's challenge to American soci-
ety. He will rather experience a certain gratitude to a Providence which, by
providing the American people with this implacable challenge, has made their
entire security as a nation dependent on their pulling themselves together and
accepting the responsibilities of moral and political leadership that history
plainly intended them to bear.
Notes
1. "Concerning the Slogans of the United States o f Europe," August 1915. Official Sowet
edition of Lenin's works.
2. Here and elsewhere in this paper "Socialism" refers to Marxist or Leninist Communism,
not t o liberal Socialism of the Second International variety.
"National Security" as an Ambiguous Symbol
Arnold Wolfers
S
tatesman, publicists and scholars who wish to be considered realists,
as many d o today, are inclined to insist that the foreign policy they
advocate is dictated by the national interest, more specifically by the
national security interest. It is not surprising that this should be so. Today
any reference to the pursuit of security is likely to ring a sympathetic chord.
However, when political formulas such as "national interest" or "national
security" gain popularity they need to be scrutinized with particular care.
They may not mean the same thing to different people. They may not have
any precise meaning at all. Thus, while appearing to offer g ~ ~ i d a n cand
e a
basis for broad consensus they may be permitting everyone to label what-
ever policy he favors with an attractive and possibly deceptive name.
In a very vague and general way "national interest" does suggest a direc-
tion of policy which can he distinguished from several others which may
present theniselves as alternatives. It indicates that the policy is designed to
promote demands which are ascribed to the nation rather than to individ-
uals, sub-national groups or mankind as a whole. It emphasizes that the
policy subordinates other interests to those of the nation. But beyond this,
it has very little meaning.
When Charles Beard's study of The Idea of National Interest was puh-
lished in the early. years
. of the New Deal and under the impact of the Great
Depression, the lines were drawn differently than they are today. The ques-
tion at that time was whether American foreign policy, then largely eco-
nomic in scope and motivation, was aimed not at promoting the welfare
interests of the nation as a whole but instead at satisfying the material inter-
ests of powerful sub-national interest or pressure groups. While it was
found hard to define what was in the interest of national welfare or to dis-
cover standards by which to measure it, there could be no doubt as to what
people had in mind: they desired to see national policy makers rise above
the narrow and special economic interests of parts of the nation to focus
their attention on the more inclusive interests of the whole.
Source: I'olrtlc-'11 Scicnce Q~rczrterly,I.XVII(4) (1952):48 1-502.

Today, the alternative t o a policy of the national interest to which people

refer is of a different character. They fear policy makers may be unduly con-
cerned with the "interests of all of mankind". They see them sacrificing the
less inclusive national community to the wider but in their opinion chimeric
world community. The issue, then, is not one of transcending narrow group
selfishness, as it was a t the time of Beard's discussion, but rather one of
according more exclusive devotion to the narrower cause of the national self.
There is another difference between the current and the earlier debate.
While it would be wrong to say that the economic interest has ceased to
attract attention, it is overshadowed today by the national security interest.
Even in the recent debates on the St. Lawrence Seaway, clearly in the first
instance an economic enterprise, the defenders of the project, when seeking
to impress their listeners with the "national interest" involved, spoke mainly
of the value of the Seaway for military defense in wartime while some oppon-
ents stressed its vulnerability to attack.
The change from a welfare to a security interpretation of the symbol
"national interest" is understandable. Today we are living under the impact
of cold war and threats of external aggression rather than of depression and
social reform. As a result, the formula of the national interest has come to
be practically synonymous with the formula of national security. Unless
explicitly denied, spokesmen for a policy which would take the national
interest as its guide can be assumed to mean that priority shall be given to
measures of security, a term to be analyzed.' The question is raised, there-
fore, whether this seemingly more precise formula of national security
offers statesmen a meaningful guide for action. Can they be expected to
know what it means? Can policies be distinguished and judged on the
ground that they do or d o not serve this interest?
The term national security, like national interest, is well enough estab-
lished in the political discourse of international relations to designate an
objective of policy distinguishable from others. We know roughly what
people have in mind if they complain that their government is neglecting
national security or demanding excessive sacrifices for the sake of enhanc-
ing it. Usually those who raise the cry for a policy oriented exclusively
toward this interest are afraid their country underestimates the external
dangers facing it or is being diverted into idealistic channels unmindful of
these dangers. Moreover, the symbol suggests protection through power
and therefore figures more frequently in the speech of those who believe in
reliance on national power than of those who place their confidence in
model behavior, international cooperation, or the United Nations to carry
their country safely through the tempests of international conflict. For these
reasons it would be an exaggeration to claim that the symbol of national
security is nothing but a stimulus to semantic confusion, though closer
analysis will show that if used without specifications it leaves room for
more confusion than sound political counsel or scientific usage can afford.
The demand for a policy of national security is primarily normative in
character. It is supposed to indicate what the policy of a nation should be in
"National Security" 17
order t o be either expedient - a rational means toward a n accepted end - o r

moral, the best o r least evil course of action. The value judgments implicit in
these normative exhortations will be discussed.
Before doing so, attention should be drawn t o a n assertion of fact which
is implicit if not explicit in most appeals for a policy guided hy national secur-
ity. Such appeals usually assume that nations in fact have made security their
goal except when idealism o r utopianism of their leaders has led them t o
stray from the traditional path. If such conformity of behavior actually
existed, it would he proper t o infer that a country deviating from the estab-
lished pattern of conduct would risk being penalized. This would greatly
strengthen the norn~ativearguments. The trouble with the contention of fact,
however, is that the term "security" covers a range of goals so wide that
highly divergent policies can be interpreted as policies of security.
Security points t o some degree of protection of values previously acquired.
In Walter Lippmann's words, a nation is secure t o the extent t o which it is
not in danger of having t o sacrifice core values, if it wishes t o avoid war,
and is able, if challenged, t o maintain them by victory in such a war.' What
this definition implies is that security rises and falls with the ability of a
nation t o deter a n attack, o r t o defeat it. This is in accord with common
usage of the term.
Security is a value, then, of which a nation can have more o r less and
which it can aspire t o have in greater o r lesser measure.' It has much in
common, in this respect, with power o r wealth, t w o other values of great
importance in international affairs. But while wealth measures the a m o u n t
of a nation's material possessions, and power its ability t o control the
actions of others, security, in a n objective sense, measures the absence of
threats t o acquired values, in a subjective sense, the absence of tear that
such values will be attacked. In both respects a nation's security can run a
wide gamut from almost complete insecurity o r sense of insecurity a t one
pole, t o almost complete security o r absence of fear a t the other.-'
T h e possible discrepancy between the objective and subjective connota-
tion of the term is significant in international relations despite the fact that
the chance of future attack never can be measured "objectively"; it must
always remain a matter of subjective evaluation and speculation. However,
when the French after World War I insisted that they were entitled t o addi-
tional guarantees of security because of the exceptionally dangerous situa-
tion which France was said t o be facing, other Powers in the I.eague
expressed the view that rather than t o submit t o what might be French hys-
terical apprehension the relative security of France should be objectively
evaluated. It is a well-known fact that nations, and groups within nations,
differ widely in their reaction t o one and the same external situation. Some
tend t o exaggerate the danger while others underestimate it. With hindsight
it is sometimes possible t o tell exactly h o w far they deviated from a rational
reaction t o the actual o r objective state of danger existing a t the time.
Even if for n o other reasons, this difference in the reaction t o similar threats
suffices t o make it probable that nations will differ in their efforts t o obtain
more security. Some may find the danger to which they are exposed entirely
normal and in line with their modest security expectations while others con-
sider it unbearable to live with these same dangers. Although this is not the
place t o set up hypotheses on the factors which account for one or the other
attitude, investigation might confirm the hunch that those nations tend to
be most sensitive to threats which have either experienced attacks in the
recent past or, having passed through a prolonged period of an exception-
ally high degree of security, suddenly find themselves thrust into a situation
of danger." Probably national efforts to achieve greater security would also
prove, in part at least, to be a function of the power and opportunity which
nations possess of reducing danger by their own effort^.^
Another and even stronger reason why nations must be expected not to
act uniformly is that they are not all or constantly faced with the same
degree of danger. For purposes of a working hypothesis, theorists may find
it useful at times to postulate conditions wherein all states are enemies -
provided they are not allied against others - and wherein all, therefore, are
equally in danger of attack.' But, while it may be true in the living world,
too, that no sovereign nation can be absolutely safe from future attack,
nobody can reasonably contend that Canada, for example, is threatened
today to the same extent as countries like Iran or Yugoslavia, or that the
British had as much reason to be concerned about the French air force in
the twenties as about Hitler's Luftwaffe in the thirties.
This point, however, should not be overstressed. There can be no quar-
rel with the generalization that most nations, most of the time - the great
Powers particularly - have shown, and had reason t o show, an active con-
cern about some lack of security and have been prepared to make sacrifices
for its enhancement. Danger and the awareness of it have been, and con-
tinue t o be, sufficiently widespread to guarantee some uniformity in this
respect. But a generalization which leaves room both for the frantic kind of
struggle for more security which characterized French policy a t times and
for the neglect of security apparent in American foreign policy after the
close of both World Wars throws little light on the behavior of nations. The
demand for conformity would have meaning only if it could be said - as it
could under the conditions postulated in the working hypothesis of pure
power politics - that nations normally subordinate all other values to the
maximization of their security, which, however, is obviously not the case.
There have been many instances of struggles for more security taking the
form of an unrestrained race for armaments, alliances, strategic boundaries
and the like; but one need only recall the many heated parliamentary debates
on arms appropriations to realize how uncertain has been the extent to
which people will consent to sacrifice for additional increments of security.
Even when there has been no question that armaments would mean more
security, the cost in taxes, the reduction in social benefits or the sheer dis-
comfort involved has militated effectively against further effort. It may be
worth noting in this connection that there seems to be no case in history in
which a country started a preventive war on the grounds of security - unless
"National Security"
IIC~* 19
Hitler's wanton attack on his neighbors be allowed to qualify as such -

although there must have been circumstances where additional security
could have been obtained by war and although so many wars have been
launched for the enhancement of other values. Of course, where security
serves only as a cloak for other more enticing demands, nations or anlbitious
leaders may consider no price for it too high. This is one of the reasons why
very high security aspirations tend to make a nation suspect of hiding more
aggressive aims.
Instead of expecting a uniform drive for enhanced or maximum security,
a different hypothesis may offer a more promising lead. Efforts for security
are bound to be experienced as a burden; security after all is nothing but
the absence of the evil of insecurity, a negative value so to speak. As a con-
sequence, nations will be inclined to minimize these efforts, keeping them
at the lowest level which will provide them with what they consider
adequate protection. This level will often be lower than what statesmen,
military leaders or other particularly security-minded participants in the
decision-making process believe it should be. In any case, together with the
extent of the external threats, numerous domestic factors such as national
character, tradition, preferences and prejudices will influence the level of
security which a nation chooses to make its target.
It might be objected that in the long run nations are not so free to choose
the amount of effort they will put into security. Are they not under a kind o f
compulsion to spare no effort provided they wish to survive? This objection
again would make sense only if the hypothesis of pure power politics were a
realistic image of actual world affairs. In fact, however, a glance at history will
suffice to show that survival has only exceptionally been at stake, particularly
for the major Powers. If nations were not concerned with the protection of
values other than their survival as independent states, most of them, most of
the time, would not have had to be seriously worried about their security,
despite what n~anipulatorsof public opinion engaged in mustering greater
security efforts may have said to the contrary. What "compulsion" there is,
then, is a function not merely of the will of others, real or imagined, to
destroy the nation's independence but of national desires and ambitions to
retain a wealth of other values such as rank, respect, material possessions
and special privileges. It would seem to be a fair guess that the efforts for
security by a particular nation will tend to vary, other things being equal,
with the range of values for which protection is being sought.
In respect to this range there may seem to exist a considerable degree of
uniformity. All over the world today peoples are making sacrifices to pro-
tect and preserve what to them appear as the minimum national core
values, national independence and territorial integrity. But there is deviation
in two directions. Some nations seek protection for more marginal values as
well. There was a time when United States policy could afford to be con-
cerned mainly with the protection of the foreign investments or markets of
its nationals, its "core values" being out of danger, or when Britain was
extending its national self to include large and only vaguely circumscribed
"regions of special interest". It is a well-known and portentous phenom-

enon that bases, security zones and the like may be demanded and acquired
for the purpose of protecting values acquired earlier; and they then become
new national values requiring protection themselves. Pushed to its logical
conclusion, such spatial extension of the range of values does not stop short
of world domination.
A deviation in the opposite direction of a compression of the range of
core values is hardly exceptional in our days either. There is little indication
that Britain is bolstering the security of Hong Kong although colonies were
once considered part of the national territory. The Czechs lifted no finger to
protect their independence against the Soviet Union and many West
Europeans are arguing today that rearmament has become too destructive
of values they cherish to be justified even when national independence is
obviously at stake.
The lack of uniformity does not end here. A policy is not characterized
by its goal, in this case security, alone. In order to become imitable, the
means by which the goal is pursued must be taken into account as well.
Thus, if two nations were both endeavoring to maximize their security but
one were placing all its reliance on armaments and alliances, the other on
meticulous neutrality, a policy maker seeking to emulate their behavior
would be at a loss where to turn. Those who call for a policy guided by
national security are not likely to be unaware of this fact, but they take for
granted that they will be understood to mean a security policy based on
power, and on military power at that. Were it not so, they would be hard put
to prove that their government was not already doing its best for security,
though it was seeking to enhance it by such means as international coopera-
tion or by the negotiation of compromise agreements - means which in one
instance may be totally ineffective or utopian but which in others may have
considerable protective value.
It is understandable why it should so readily be assumed that a quest for
security must necessarily translate itself into a quest for coercive power. In
view of the fact that security is being sought against external violence -
coupled perhaps with internal subversive violence - it seems plausible at first
sight that the response should consist in an accumulation of the same kind
of force for the purpose of resisting an attack or of deterring a would-
be attacker. The most casual reading of history and of contemporary experi-
ence, moreover, suffices to confirm the view that such resort to "power
of resistance" has been the rule with nations grappling with serious threats
to their security, however much the specific form of this power and its
extent may differ. Why otherwise would so many nations which have no
acquisitive designs maintain costly armaments? Why did Denmark with her
state of complete disarmament remain an exception even among the small
Powers?
But again, the generalization that nations seeking security usually place
great reliance on coercive power does not carry one far. The issue is not
whether there is regularly some such reliance but whether there are no
c ~ g n ~ f r c a nd~fferences
t between natrons concerning t h e ~ rover-c~llc h o ~ c eof
the means upon which they place their trust. The controversies concerning
the best road t o future security that are so typical o f coalition partners a t the
close of victorious wars throw light on this question. France in 19 19 and all
the Allies in 1945 believed that protection against another German attack
co~rldhe gained only by means of continued military superiority based o n
German military impotence. President Wilson in 1919 and many observers
in 194.5 were equally convinced, however, that more hope for security lay in
a conciliatory and fair treatment of the defeated enemy, which would rob
him of future incentives t o renew his attack. While this is not the place t o
decide which side was right, one cannot help drawing the conclusion that, in
the matter of means, the roads which are open may lead in diametrically
opposed directions.The choice in every instance will depend o n a multitude
of variables, including ideological and moral convictions, expectations con-
cerning the psychological and political developnients in the camp of the
opponent, and inclinations of individual policy makers."
After all that has been said little is left of the sweeping generalization that
in actual practice nations, guided by their national security interest, tend
to pursue a mif form and therefore imitable policy of security. Instead, there
a r e numerous reasons why they should differ widely in this respect, with
some standing close t o the pole of complete indifference t o security or coni-
plete reliance on nonmilitary means, others close t o the pole of insistence on
absolute security or of complete reliance on coercive power. It should be
added that there exists still another category of nations which cannot be
placed within the continuum connecting these poles because they regard
security of any degree as a n insufficient goal; instead they seek t o acquire
new values even a t the price of greater insecurity. In this category must be
placed not only the "mad Caesars", w h o are o u t for conquest and glory a t
any price, hut also idealistic statesmen w h o would plunge their country into
war for the sake of spreading the benefits of their ideology, for example, of
liberating enslaved peoples.
The actual behavior o f nations, past and present, does not affect the nor-
niativc proposition, t o which we shall n o w turn o u r attention. According t o
this proposition nations are called upon t o give priority t o national security
and thus t o consent t o any sacrifice of value which will provide an addi-
tional increment of security. It m a y be expedient, moral o r both for nations
t o d o so even if they should have failed t o heed such advice in the past and
for the most part are not living up t o it today.
The first question, then, is whether some definable security policy can be
said t o he generally expedient. Because the choice of goals is not a matter of
expediency, it would seem t o make n o sense t o ask whether it is expedient
for nations t o be concerned with the goal of sec~rrityitself; only the means
used t o this end, so it would seem, can he judged as t o their fitness - their
instrumental rationality - t o promote security. Yet, this is not so. Security,
like other aims, may he an intermediate rather than a n ~lltimategoal, in
which case it can be judged as a means t o these more ultimate ends.
Traditionally, the protection and preservation of national core values

have been considered ends in themselves, a t least by those who followed in
the footsteps of Machiavelli or, for other reasons of political philosophy,
placed the prince, state or nation at the pinnacle of their hierarchy of values.
Those who d o so today will be shocked at the mere suggestion that national
security should have to be justified in terms of higher values which it is
expected to serve. But there is a large and perhaps growing current of opin-
ion - as a matter of fact influential in this country for a long time - which
adheres to this idea. We condemn Nazis and Communists for defending their
own totalitarian countries instead of helping to free their people from
tyranny; we enlist support for armaments, here and in Allied countries, not
so much on the grounds that they will protect national security but that by
enhancing such security they will serve to protect ultimate human values like
individual liberty. Again, opposition in Europe and Asia t o military security
measures is based in part on the contention that it would help little to make
national core values secure, if in the process the liberties and the social wel-
fare of the people had to be sacrificed; the prevention of Russian conquest,
some insist, is useless, if in the course of a war of defense a large part of the
people were to be exterminated and most cities destroyed.I0
While excellent arguments can be made to support the thesis that the
preservation of the national independence of this country is worth almost any
price as long as no alternative community is available which could assure the
same degree of order, justice, peace or individual liberty, it becomes necessary
to provide such arguments whenever national security as a value in itself is
being questioned. The answer cannot be taken for granted.
But turning away now from the expediency of security as an intermedi-
ate goal we must ask whether, aside from any moral considerations which
will be discussed later, a specific level of security and specific means of
attaining it can claim to be generally expedient.
When one sets out to define in terms of expediency the level of security
to which a nation should aspire, one might be tempted to assume that the
sky is the limit. Is not insecurity of any kind an evil from which any rational
policy maker would want to rescue his country? Yet, there are obvious rea-
sons why this is not so.
In the first place, every increment of security must be paid by additional
sacrifices of other values usually of a kind more exacting than the mere
expenditure of precious time on the part of policy makers. At a certain
point, then, by something like the economic law of diminishing returns, the
gain in security n o longer compensates for the added costs of attaining it.
As in the case of economic value comparisons and preferences, there is fre-
quently disagreement among different layers of policy makers as to where
the line should be drawn. This is true particularly because absolute security
is out of the question unless a country is capable of world domination, in
which case, however, the insecurities and fears would be "internalized" and
probably magnified. Because nations must "live dangerously", then, t o
some extent, whatever they consent to d o about it, a modicum of additional
4 I "National Security" 23
but only relative security may easily become unattractive to those who have
to bear the chief burden. Nothing renders the task of statesmen in a democ-
racy more difficult than the reluctance of the people to follow them very far
along the road to high and costly security levels.
In the second place, national security policies when based on the accu-
mulation of power have a way of defeating themselves if the target level is
set too high. This is due to the fact that "power of resistance" cannot be
unmistakably distinguished from "power of aggression". What a country
does to bolster its own security through power can be interpreted hy oth-
ers, therefore, as a threat to their security. If this occurs, the vicious circle
of what John Herz has described as the "security dilemma" sets in: the
efforts of one side provoke countermeasures by the other which in turn tend
to wipe out the gains of the first. Theoretically there seems to be no escape
from this frustrating consequence; in practice, however, there are ways to
convince those who might feel threatened that the accumulation of power
is not intended and will never be used for attack." The chief way is that of
keeping the target level within moderate bounds and of avoiding placing
oneself in a position where it has to be raised suddenly and drastically. The
desire to escape from this vicious circle presupposes a security policy of
much self-restraint and moderation, especially in the choice of the target
levef." It can never be expedient to pursue a security policy which by the
fact of provocation or incentive to others fails to increase the nation's rela-
tive power position and capability of resistance.
The question of what means are expedient for the purpose of enhancing
security raises even more thorny problems. Policy makers must decide how
to distribute their reliance on whatever means are available to them and,
particularly, how far to push the accumulation of coercive power. N o
attempt can he made here to decide what the choice should be in order to
be expedient. Obviously, there can be no general answer which would meet
the requirements of every case. The answer depends on the circumstances.
A weak country may have no better means at its disposal than to prove to
stronger neighbors that its strict neutrality can be trusted. Potentially strong
countries may have a chance to deter an aggressor by creating "positions of
strengthv. In some instances they may have no other way of saving them-
selves; while in others even they may find it more expedient to supplement
such a policy, if not to replace it, by a policy intended to negotiate their
opponent out of his aggressive designs.
The reason why "power of resistance" is not the general panacea which
some believe it to be lies in the nature of security itself. If security, in the
objective sense of the term at least, rises and falls with the presence or
absence of aggressive intentions on the part of others, the attitude and
behavior of those from whom the threat emanates are of prime importance.
Such attitude and behavior need not be beyond the realm of influence by
the country seeking to bolster its security. Whenever they do not lie beyond
this realm the most effective and least costly security policy consists in
inducing the opponent to give up his aggressive intentions.
While there is no easy way to determine when means can and should be
used which are directed not at resistance but at the prevention of the desire
of others to attack, it will clarify the issue to sketch the type of hypotheses
which would link specific security policies, as expedient, to some of the
most typical political constellations.
One can think of nations lined up between the two poles of maximum
and minimum "attack propensity", with those unalterably committed to
attack, provided it promises success, at one pole and those whom no
amount of opportunity for successful attack could induce to undertake it at
the other. While security in respect to the first group can come exclusively
as a result of "positions of strength" sufficient to deter or defeat attack,
nothing could d o more to undermine security in respect to the second group
than to start accumulating power of a kind which would provoke fear and
countermoves.
Unfortunately it can never be known with certainty, in practice, what
position within the continuum one's opponent actually occupies. Statesmen
cannot be blamed, moreover, if caution and suspicion lead them to assume
a closer proximity to the first pole than hindsight proves to have been justi-
fied. We believe we have ample p o o f that the Soviet Union today is at or
very close to the first pole, while Canadian policy makers probably place
the United States in its intentions toward Canada a t the second pole.
It is fair to assume that, wherever the issue of security becomes a matter
of serious concern, statesmen will usually be dealing with potential oppon-
ents who occupy a position somewhere between but much closer to the first
of the two poles. This means, then, that an attack must be feared as a pos-
sibility, even though the intention to launch it cannot be considered to have
crystallized to the point where nothing could change it. If this be true, a
security policy in order to be expedient cannot avoid accumulating power
of resistance and yet cannot let it go at that. Efforts have to be made simul-
taneously toward the goal of removing the incentives to attack. This is only
another way of saying that security policy must seek to bring opponents to
occupy a position as close to the second pole as conditions and capabilities
permit.
Such a twofold policy presents the greatest dilemmas because efforts to
change the intentions of an opponent may run counter to the efforts to
build up strength against him. The dangers of any policy of concessions,
symbolized by "Munich", cannot be underestimated. The paradox of this
situation must be faced, however, if security policy is to be expedient. It
implies that national security policy, except when directed against a country
unalterably committed to attack, is the more rational the more it succeeds
in taking the interests, including the security interests, of the other side into
consideration. Only in doing so can it hope to minimize the willingness of
the other to resort to violence. Rather than to insist, then, that under all
conditions security be sought by reliance on nothing but defensive power
and be pushed in a spirit of national selfishness toward the highest targets,
111 "NationalSecurity" 25
it should be stressed that in most instances efforts to satisfy legitimate

demands of others are likely to promise better results in terms of security."
That is probably what George Kennan had in mind when he advised policy
makers to use self-restraint in the pursuit of the national interest. While in
the face of a would-be world conqueror who is beyond the pale of external
influence it is dangerous to be diverted from the accumulation of sheer
defensive power, any mistake about his true state of mind or any neglect of
opportunities to influence his designs, where it has a chance of being suc-
cessful, violates the rules of expediency. It should always be kept in mind
that the ideal security policy is one which would lead to a distribution of
values so satisfactory to all nations that the intention to attack and with it
the problem of security would be minimized. While this is a utopian goal,
policy makers and particularly peacemakers would d o well to remember
that there are occasions when greater approximation to such a goal can be
effected.
We can now focus our attention on the moral issue, if such there be.I4
Those who advocate a policy devoted to national security are not always
aware of the fact - if they d o not explicitly deny it - that they are passing
moral judgment when they advise a nation to pursue the goal of national
security or when they insist that such means as the accumulation of coer-
cive power - or its use - should be employed for this purpose.''
Nations like individuals or other groups may value things not because
they consider them good or less evil than their alternative; they may value
them because they satisfy their pride, heighten their sense of self-esteem or
reduce their fears. However, no policy, or human act in general, can escape
becoming a subject for moral judgment - whether by the conscience of the
actor himself or by others -which calls for the sacrifice of other values, as
any security policy is bound to do. Here it becomes a matter of comparing
and weighing values in order to decide which of them are deemed suffi-
ciently good to justify the evil of sacrificing others. If someone insists that
his country should d o more to build up its strength, he is implying, know-
ingly or not, that more security is sufficiently desirable to warrant such evils
as the cut in much-needed social welfare benefits or as the extension of the
period of military service. I h
Many vivid examples of the moral dilemma are being supplied by current
controversies concerning American security policy. Is a "deal with fascist
Spain" morally justified, provided it added an increment to our security,
though principles valued highly by some were being sacrificed? Should we
engage in subversive activities and risk the lives of our agents if additional
security can be attained thereby? Should we perhaps go so far as to start a
preventive war, when ready, with the enormous evils it would carry with it,
if we should become convinced that no adequate security can be obtained
except by the defeat of the Soviet Union? In this last case, would not the
exponents of amoralism have some moral qualms, at least to the point of
rationalizing a decision favoring such a war by claiming that it would serve
to satisfy not primarily an egotistical national demand for security but an

altruistic desire to liberate enslaved peoples? It is easier to argue for the
amorality of politics if one does not have t o bear the responsibility of choice
and decision!
Far be it from a political scientist to claim any particular competence in
deciding what efforts for national security are or are not morally justified.
What he can contribute here is to point t o the ambiguities of any general
normative demand that security be bought at whatever price it may cost. He
may also be able to make it more difficult for advisers or executors of policy
to hide from themselves or others the moral value judgments and prefer-
ences which underlie whatever security policy they choose to recommend or
conduct.
The moral issue will be resolved in one of several ways depending on the
ethical code upon which the decision is based. From one extreme point of
view it is argued that every sacrifice, especially if imposed on other nations,
is justified provided it contributes in any way to national security. Clearly
this implies a position that places national security at the apex of the value
pyramid and assumes it t o constitute an absolute good to which all other
values must be subordinated. Few will be found t o take this position
because if they subscribed to a nationalistic ethics of this extreme type they
would probably go beyond security - the mere preservation of values - and
insist that the nation is justified in conquering whatever it can use as
Lebensraum or otherwise. At the opposite extreme are the absolute pacifists
who consider the use of coercive power a n absolute evil and condemn any
security policy, therefore, which places reliance on such power.
For anyone who does not share these extreme views the moral issue
raised by the quest for national security is anything but clear-cut and sim-
ple. He should have n o doubts about the right of a nation to protect and
preserve values to which it has a legitimate title or even about its moral duty
to pursue a policy meant to serve such preservation. But he cannot consider
security the supreme law as Machiavelli would have the statesman regard
the ragione di stato. Somewhere a line is drawn, which in every instance he
must seek to discover, that divides the realm of neglect, the "too-little",
from the realm of excess, the "too much". Even Hans Morgenthau who
extols the moral duty of self-preservation seems to take it for granted that
naked force shall be used for security in reaction only to violent attack, not
for preventive war.
Decision makers are faced with the moral problem, then, of choosing
first the values which deserve protection, with national independence rank-
ing high not merely for its own sake but for the guarantee it may offer to
values like liberty, justice and peace. He must further decide which level of
security to make his target. This will frequently be his most difficult moral
task though terms such as adequacy or fair share indicate the kind of stand-
ards that may guide him. Finally, he must choose the means and thus by
scrupulous computation of values compare the sacrifices, which his choice
of means implies, with the security they promise to provide.
\\ oiicr\ "National Security" 27
It follows that policies of national security, far from being all good or
all evil, may be morally praiseworthy or condemnable depending on their
specific character and the particular circumstances of the case. They may be
praised for their self-restraint and the consideration which this implies for
values other than security; they may instead be condemned for being inade-
quate to protect national values. Again, they may be praised in one instance
for the consideration given to the interests of others, particularly of weaker
nations, or condemned in another because of the recklessness with which
national values are risked on the altar of some chimera. The target level falls
under moral judgment for being too ambitious, egotistical and provocative
or for being inadequate; the means employed for being unnecessarily costly
in other values or for being ineffective. This wide range of variety which
arises out of the multitude of variables affecting the value computation
would make it impossible, and in fact meaningless, to pass moral judgment,
positive or negative, on "national security policy in general".
It is this lack of moral homogeneity which in matters of security policy
justifies attacks on so-called moralism, though not on moral evaluation.
The "moralistic approach" is taken to mean a wholesale condemnation either
of any concern with national security - as being an expression of national
egotism - or of a security policy relying on coercive and therefore evil power.
The exponent of such "moralism" is assumed to believe that security for all
peoples can be had today by the exclusive use of such "good" and altruistic
means as model behavior and persuasion, a spirit of conciliation, international
organization or world government. If there are any utopians who cling to this
notion, and have influence on policy, it makes sense to continue to disabuse
them of what can surely be proved to be dangerous illusions.
It is worth emphasizing, however, that the opposite line of argument,
which without regard for the special circumstances would praise everything
done for national security or more particularly everything done for the
enhancement of national power of resistance, is no less guilty of applying
simple and abstract moral principles and of failing to judge each case real-
istically on its merits.
In conclusion, it can be said, then, that normative admonitions to con-
duct a foreign policy guided by the national security interest are no less
ambiguous and misleading than the statement of fact concerning past behav-
ior which was discussed earlier. In order to be meaningful such admonitions
would have to specify the degree of security which a nation shall aspire to
attain and the means by which it is to be attained in a given situation. It may
be good advice in one instance to appeal for greater effort and more arma-
ments; it may be no less expedient and morally advisable in another instance
to call for moderation and for greater reliance on means other than coercive
power. Because the pendulum of public opinion swings so easily from extreme
complacency to extreme apprehension, from utopian reliance on "good will"
to disillusioned faith in naked force only, it is particularly important to be
wary of any simple panacea, even of one that parades in the realist garb of
a policy guided solely by the national security interest.
Notes
1. Hans Morgenthau's In Defense of the National Interest (New York, 1951) IS the most
explicit and impassioned recent plea for an American foreign policy which shall follow "but
one guiding star - the National Interest". While Morgenthau is not equally explicit in regard
t o the meaning he attaches t o the symbol "national interest", it becomes clear in the few pages
devoted to an exposition of this "perennial" interest that the author is thinking in terms of the
national security interest, and specifically of security based on power. The United Stares, he
says, is interested in three things: a unique position as a predominant Power without rival in
the Western Hemisphere and the maintenance of the balance of power in Europe as well as in
Asia, demands which make sense only in the context of a quest for security through power.
2. Walter Lippmann, U. S. Foreign Policy (Boston, 1943), p. 51.
3. This explains why some nations which would seem to fall into the category of status
quo Powers par excellence may nevertheless be dissatisfied and act very much like "imperial-
ist" Powers, as Morgenthau calls nations with acquisitive goals. They are dissatisfied with the
degree of security which they enjoy under the status quo and are out to enhance it. France's
occupation of the Ruhr in 1923 illustrates this type of behav~or.Because the demand for more
security may induce a status quo Power even to resort to the use of violence as a means of
attaining more security, there is reason to beware of the easy and often self-righteous assump-
tion that nations which desire to preserve the status quo are necessarily "peace-loving".
4. Security and power would be synonymous terms if security could be attained only through
the accumulation of power, which will be shown nor to be the case. The fear of attack -security
in the subjective sense - is also not proportionate to the relative power position of a nation. Why,
otherwise, would some weak and exposed nations conslder themselves more secure today than
does the United States?
Harold D. Lasswell and Abraham Kaplan, Power and Society (New Haven, 1950), d e f ~ n -
ing security as "high value expectancy" stress the subjective and speculative character of
security by using the term "expectancy"; the use of the term "h~gh",while mdicating no defin-
ite level, would seem to imply that the security-seeker aims at a position in which the events
he expects - here the continued unmolested enjoyment of his possessions - have considerably
more than an even chance of materializing.
5. The United States offers a good illustration and may be typical in this respect. For a long
time this country was beyond the reach of any enemy attack that could be considered prob-
able. During that period, then, it could afford to dismiss any serious preoccupation w t h
security. Events proved that it was n o worse off for having done so. However, after this happy
condition had ceased to exist, government and people alike showed a lag in their awareness of
the change. When Nicholas J. Spykman raised his voice in the years before World War I1 to
advocate a broader security outlook than was indicated by the symbol "Western Hemisphere
Defense" and a greater appreciation of the r6le of defenswe mtlitary power, he was dealing
with this lag and with the dangers implied in it. If Hans Morgenthau and others raise t h e ~ r
warning voices today, seemmgly treading in Spykman's footsteps, they are addressing a narlon
which after a new relapse into wishful thinking in 1945 has been rad~callydisillusioned and
may now be swinging toward excessive security apprehensions.
6. Terms such as "degree" or "level" of security are not intended to indicate merely
quantitative differences. Nations may also differ in respect to the breadth of their security per-
spective as when American leaders at Yalta were so preoccupied with security agalnst the then
enemy countries of the United States that they failed or refused to consider future American
security vis-a-vis the Soviet Union. The differences may apply, instead, to the time range for
which security is sought as when the British at Versailles were ready to offer France short-run
security guarantees while the French with more foresight ins~sredthat the "German danger"
would not become acute for some ten years.
7. For a discussion of this working hypothesis - as part of the "pure power" hypothesis -
see my article on "The Pole of Power and the Pole of Indifference" in World Pol~tics,vol. IV,
No. 1. October 1951.
8. Pvlyres S. Mcllougal ("Law a n d Peace" in the A m e r m m /o~irtznlof lntcrn~7tron~zI 1.~7~:

vol. 46. No. 1, January 19.52, pp. 102 et seq.) rightly criticize\ H a n s Morgenthau ( a n d George
t i e n t u n tor what tiennan himself wrongly believes t o he h ~ os w n point o t view in the matter;
see fn, 15 rr~frtr)for his failure t o appreciate the rhle which non-power methods, such as legal
procedures and moral a p p e d s , may a t times successfully play in the pursuit of security. But ~t
i \ s u r p r ~ s l n gh o w llrrle aware McDougal appears t o be of the d~sappoinringmodesry of the
c o n t r ~ h u t i o n swhich these "other means" have actually made t o the enhancement of security
a n d the quite insignificant c o n t r ~ b u t ~ o nthey
s have made t o the promotion of ch.inges of the
status quo. T h ~ slatter failure signif~esthat ther have heen unable t o remove the m ~ i nc~luses
of t h e attacks which security-minded peoples rlghtly fear.
9. On the problem of security policy (Sicl~erheitspolrt~k) with special reference t o "collective
secur~ty"see the comprehensive and illuminating study of Heinrich Rogge, "Kollektivsicherheit
Buendnispolitik Voelkerhund", Theorre der natzonulerz ttnd rnternntionizlctr S~cherhcrt(Berlin,
19371, which deserves attention despite the fact that ~t was written a n d puhl~shedin N u t
Germany and bean J distinctly L'revisioni~t"slant.
10. Raymond Dennett goes further in making the generalization that, "if economlc pressures
become great enough, almost any government, when put t o the flnal test, will rnoder,~teo r ahan-
don a ool~ticalassociation" (such as the alliance svstern of the United States with its usefulness
t o nation,d security) "if only a n alteration of policy seems t o offer the possibility of maintaining
o r .~chievitigliving standards ,idequate enough t o permit the reglme t o survive". "Danger Spots
In the Pattern of American Security", In World Politzcs, vol. IV, No. 4, July 19.52, p. 449.
I I. N o t everyone agrees that this can be done. Jeremy Benrham wrote that "measures of
mere self defense are naturally taken for projects of aggression" with the result t h , ~ t "each
ninkes h a t e t o hegin for fe,~rof heing forestalled." Prmczples of Internatronal 1 . a ~ Essay ~. IV.
12. T h e Quakers, in ,I book on T b r Unrted Stntrs m d t h r Sol& Urzion: Some Quczkrr
Propos'11s for Prncr ( N e w Haven, l 9 4 9 ) , p. 14, state that "it is highly q u e s t ~ o n a b l ewhether
securlty c,ln be achieved in the tnodern world through a n attempt t o estahllsh ,111 overwhelm-
ing preponderance of military power." This c a n be read t o mean that 3 less a m b ~ r ~ o milit'lry us
target than overwhelming preponderance r n ~ g h tbe a means o f achieving security.
13. As A.D. Lindsay puts it, "The search for perfect security ... defeats its o w n ends.
Playing for safety 1s the most dangerous way t o live." Introduction t o T h o m a \ Hobbes,
/.~1'1~7t/~dl2, p. XYII.
14. O n the moral problem in international r e l a t ~ o n ssee 11iy article o n "St:~tesmansh~p J I ~
M o r a l Choice" in World Politics, vol. I, N o . 2, January 1949, pp. 176 et seq.. especially
p. 18.5. In o n e of h ~ most s recent statements o n the subject, Reinhold Nlehuhr, T ~ J Iro~zy L, of
Att~cv~cm Hlstory ( N e w York, 19451, points specifically t o the moral problem ~nvolvedin
security policy - " n o imperiled nation", he writes, "is morally able t o dispense with weapons
which might insure its sul-vivd" ( p . 39).
15. It IS not w ~ t h o u irony
t that of the t w o authors w h o have rrcently c o m e o u t tor a policy
of the national interest, the one, George F. Kennan, w h o calls for a policy of n a t ~ o n a lself-
restraint a n d humility, ~ ~ s u a l ~dentlfied
ly with morality, should deny "that state behavior is 1'
fit subject for moral judgment" (Amerrcan DlpIomi7cy. 1900-1950, Chicago, 1952, p. IOO),
while the other, H a n s Morgenthau (op. cit.), calling for a policy of ~ ~ n a d u l t e r ~ l tnational
ed ego-
tlsm, claims t o speak In the n a m e of morality.
.
16. It would he unrealistic t o assume that ~ o l i ,c vmakers divide their attention strictlv
between ends a n d means a n d only after having chosen a speclf~ctarget level as heing morally
iustified decide whether the means by which ~t can be attamed are morally acceptable. Moral
judgment is more llkely t o be passed o n the t o t a l ~ t yof a course of action which embraces both
the desired end , ~ n dthe means which lead t o it.
Another "Great Debate": The National Interest
o f the United States
Hans J. Morgenthau
T he controversy which has arisen on the occasion of Ambassador

Kennan's and my recent publications differs from the great historical
debates on American foreign policy in two significant respects. It
raises a n issue more fundamental to the understanding of American foreign
policy and of all politics than those with which the previous "great debates"
were concerned, and it deals with the issue largely in terms which are not
conducive to understanding.
The great debates of the past, such as the one over intervention vs. neu-
trality in 1793, expansion vs. the status quo before the Mexican and after
the Spanish-American War, international cooperation vs. isolation in the
'twenties, intervention vs. abstention in the late 'thirties - all evolved
around clear-cut issues of foreign policy. In 1793 you were in favor of going
to war on the side of France or of remaining neutral. In the 1840's you
approved of the annexation of Texas or you did not. At the turn of the
century you supported overseas expansion or you were against it. In the
'twenties you advocated joining the League of Nations or staying out of it.
In the late 'thirties you wanted to oppose the Axis Powers by all means
short of war or you wanted t o abstain from intervening. What separates the
"utopian" from the "realist" position cannot be so sharply expressed in
terms of alternative foreign policies. The very same policies can be and are
being supported by both schools of thought. What sets them apart is not
necessarily a matter of practical judgment, but of philosophies and stand-
ards of thought.
The issue which the present debate raises concerns the nature of all politics
and, more particularly, of the American tradition in foreign policy. The history
of modern political thought is the story of a contest between two schools
which differ fundamentally in their conception of the nature of man, society,
and politics. One believes that a rational and moral political order, derived
from universally valid abstract principles, can be achieved here and now.
Source: The American Political Science Review, XLVI(4) ( 1 9 5 2 ) :961-88.

liorgi nlhdu The National Interest of The U.S. 31
It assumes the essential goodness and m f ~ n ~ malleabhty

te of human nature
standards to lack of knowledge and understanding, obsolescent social institu-

tions, or the depravity of certain isolated individuals or groups. It trusts in
education, reform, and the sporadic use of force to remedy these deficiencies.'
The other school believes that the world, imperfect as it is from the
rational point of view, is the result of forces which are inherent in human
nature. To improve the world one must work with those forces, not against
them. This being inherently a world of opposing interests and of conflict
among them, moral principles can never be fully realized, but at best approxi-
mated through the ever temporary balancing of interests and the ever pre-
carious settlement of conflicts. This school, then, sees in a system of checks
and balances a universal principle for all pluralist ~ o c i e t i e s It
. ~ appeals to
historic precedent rather than to abstract principles, and aims at achievement
o f the lcsser evil rather than of the absolute good.
This conflict between two basic conceptions of man and politics is at the
bottom of the present controversy. It is the same conflict which found its
classic expression in the polemic of Burke against the philosophy of the
French Revolution. Given the sad state of political thought in our time, it
would be vain to expect the spokesmen of political realism to speak with
the voice of Burke and the defenders of political utopianism to measure up
to the standards of Condorcet and Rousseau. Yet one has a right to expect
that scholars discuss the issue without resort to invective and with proper
regard for established facts.'
In order to refute a theory which pretends to be scientific, it is first ne-

cessary to understand what a scientific theory is. A scientific theory is an
attempt to bring order and meaning to a mass of phenomena which with-
out it would remain disconnected and unintelligible. Any one who disputes
the scientific character of such a theory either must produce a theory super-
ior in these scientific functions to the one attacked or must, at the very
least, demonstrate that the facts as they actually are do not lend themselves
to the interpretation which the theory has put upon them. When a historian
tells us that the balance of power is not a universal principle of politics,
domestic and international, that it was practiced in Europe only for a limited
period and never by the United States, that it ruined the states that practiced
it,4 it is incumbent upon him to tell us how we can dispose by means of the-
ory of the historic data by which, for instance, David Hume demonstrated
the universality of the balance of power and Paul Scott M o w r e r h n d Alfred
Vagts%ts practice by the United States; what Kautilya was writing about
in the fourth century B.C. when he summarized the theoretical and practical
tradition of Indian statecraft in terms of the balance of power; what the
Greek city states, the Roman republic, and the medieval emperors and
32 T h e Cold War and Nuclear Deterrence
popes were doing if they did not apply the principles of the balance of power;
and how the nations which either neglected these principles or applied them
wrongly suffered political and military defeat and even extinction, while the
nation which applied these principles most consistently and consciously, that
is, Great Britain, enjoyed unrivalled power for an unparalleled length of time.
The historian who wishes to replace the balance of power as the guiding
principle of American foreign policy with the "humanitarian and pacific
traditions" of the "coordinate state"' must first of all explain how it has
come about that the thirteen original states expanded into the full breadth
and a good deal of the length of a continent, until today the strategic fron-
tiers of the United States run parallel to the coastline of Asia and along the
River Elbe. If such are the results of policies based upon "humanitarian and
pacific traditions," never in the history of the world has virtue been more
bountifully rewarded! Yet our historian must explain not only the great
sweep of American expansion, but also the specific foreign policies which
in their historic succession make up that sweep. Is it easier to explain the
successive shifts of American support from Great Britain to France and
back again from the beginning of King George's War in 1744 to the War of
1812 in terms of the "coordinate state" than in terms of the balance of
power? The same question might be asked about the postponement of the
recognition of the independence of the Spanish colonies until 1822, when
the Floridas had been acquired from Spain and Spain had thereby been
deprived of the ability to challenge the United States from within the hemi-
sphere. The same question might be asked about the Monroe Doctrine
itself, about Lincoln's policies toward Great Britain and France, and about
our successive policies with regard to Mexico and the Caribbean. One
could go on and pick out a t random any foreign policy pursued by the
United States from the beginning to 1919 and one would hardly find a pol-
icy, with the exception perhaps of the War of 1812, which could not be
made intelligible by reference to the national interest defined in terms of
power - political, military, and economic - rather than by reference to the
principle of the "coordinate state." This inevitable outcome of such an
inquiry is well summarized in these words:
Ease and prosperity have made us wish the whole world to be as happy
and well to do as ourselves; and we have supposed that institutions and
principles like our own were the simple prescription for making them so.
And yet, when issues of our own interest arose, we have not been
unselfish. We have shown ourselves kin to all the world, when it came to
pushing an advantage. Our action against Spain in the Floridas, and
against Mexico on the coasts of the Pacific; our attitude toward first
the Spaniards, and then the French, with regard to the control of the
Mississippi; the unpitying force with which we thrust the Indians to the
wall wherever they stood in our way, have suited our professions of
peacefulness and justice and liberality no better than the aggressions of
I I ii,itr T h e National Interest of T h e U S . 33
other nations that were strong and not to be gainsaid. Even Mr. Jefferson,
philanthropist and champion of peaceable and modest government though
he was, exemplified this double temper of the people he ruled. "Peace is
our passion," he had declared; but the passion abated when he saw the
mouth of the Mississippi about to pass into the hands of France. Though
he had loved France and hated England, he did not hesitate then what
language to hold. "There is on the globe," he wrote to Mr. Livingston at
Paris, "one single spot the possessor of which is our natural and habitual
enemy. The day that France takes possession of New Orleans seals the
union of two nations, who, in conjunction, can maintain exclusive pos-
session of the sea. From that moment we must marry ourselves to the
British fleet and nation." Our interests must march forward, altruists
though we are; other nations must see to it that they stand off, and d o
not seek to stay us.
This realist appraisal of the American tradition in foreign policy was published
in 1901 in the Atlantic Monthly. Its author was a professor of jurisprudence
and political economy at Princeton by the name of Woodrow Wilson."
Nothing more needs to be said to demonstrate that facts d o not support
a revision of American diplomatic history which tries to substitute "humani-
tarian and pacifist traditions" and the "coiirdinate state" for power politics
and the balance of power as the guiding principle of American foreign
policy. What, then, does support it? Three things: the way American states-
men have spoken about American foreign policy; the legal fiction of the
"coiirdinate state"; finally, and foremost, an emotional urge to justify
American foreign policy in humanitarian, pacifist terms.
It is elementary that the character of a foreign policy can be ascertained
only through the examination of the political acts performed and of the
foreseeable consequences of these acts. Thus we can find out what states-
men have actually done, and from the foreseeable consequences of their acts
we can surmise what their objectives might have been. Yet examination of
the facts is not enough. To give meaning to the factual raw material of his-
tory, we must approach historical reality with a kind of rational outline,
a map which suggests to us the possible meanings of history. In other words,
we put ourselves in the position of a statesman who must meet a certain
problem of foreign policy under certain circumstances and ask ourselves,
what are the rational alternatives from which a statesman may choose who
must meet this problem under these circumstances, presuming always that
he acts in a rational manner, and which of these rational alternatives was
this particular statesman, acting under these circumstances, likely to
choose? It is the testing of this rational hypothesis against the actual facts
and their consequences which gives meaning to the facts of history and
makes the scientific writing of political history possible.
In the process of writing the history of foreign policy the interpretations
by statesmen of their own acts, especially if they are made for public
34 The Cold War a n d Nuclear Deterrence
consumption, must needs have a strictly subsidiary place. The public self-
interpretation by actors on the political scene is itself, of course, a political
act which seeks to present a certain policy to its presumed supporters in
terms of their moral and political folklore and to those against which it is
directed in terms which intend to embarrass and deceive. Such declarations
may indeed shed light upon the character and objectives of the policy pur-
sued if they are considered in conjunction with, and in subordination to,
rational hypotheses, actions, and likely consequences. Yet it is quite a dif-
ferent matter to interpret the American tradition of foreign policy in the
light of a collection of official statements which, like most such statements,
present humanitarian and pacifist justifications for the policies pursued. If
anybody should be bold enough to write a history of world politics with so
uncritical a method he would easily and well-nigh inevitably be driven to
the conclusion that from Timur to Hitler and Stalin the foreign policies of
all nations were inspired by the ideals of humanitarianism and pacifism.
The absurdity of the result is commensurate with the defects of the method.
It is only from a method which accepts the declarations of statesmen
as evidence of the character of the policies pursued, that the principle of
the "coordinate state" receives a semblance of plausibility. Statesmen and
international lawyers have been wont to speak of the "equal dignity" of all
states, regardless of "wealth, power, size, population or c u l t ~ r e , "which
~ I
take the principle of the "coordinate state" to mean. It is also referred to as
the principle of "federalism in international r e l a t i ~ n s . " ' As
~ its prime ex-
amples are cited the relations amongst the states of the Union, the states of
the American system, the members of the Commonwealth of Nations, and
the members of the Swiss Confederation. If the whole world were organized
in accordance with this principle, as are already these four political entities,
it is assumed that the freedom, dignity, and peace of all nations would then
be assured.
There is no need to examine the theoretical and practical merits of the
principle of the "coordinate state," because for none of the four political
entities mentioned does the idea of the "coordinate state" provide the prin-
ciple of political organization. The equality of the states as the political
foundation of the United States became obsolescent when Chief Justice
Marshall's Supreme Court resolved the ambiguity of the Constitution in
favor of the federal government, and it became obsolete when the Civil War
proved Chief Justice Marshall's point. The equality of the states survives
today only in the shadow and by virtue of the federal government's polit-
ical supremacy, and without the cohesive force of that supremacy there
would be no union of equal states to begin with. That these powers of the
federal government are limited and qualified by the principle of federalism,
that is, by the constitutionally granted powers of the states, is quite a dif-
ferent matter; it concerns the distribution of powers between federal gov-
ernment and states within a general system of checks and balances, but has
nothing to do with the equality of the states as the alleged political foun-
dation of the American system of government. With the exception of the
hloip,t.nth~u The National Interest of The U S . 35
equality of senatorial representation, the principle of the equality of the

states is today, as it has been for almost a century, devoid of political con-
tent. It serves only as a principle of regional organization, of administrative
decentralization, and, above all, of constitutional rhetoric. What it really
signifies was pointed out more than fifty years ago by W.A. Dunning when
he summarized his answer to the question "Are the states equal under the
Constitution?" by saying that "the theory of equal states falls to the
ground." I '
Similarly, the federalism of Switzerland is the result of a long series of civil
wars, the last one fought a little more than a century ago, which established
the predominance of the German-speaking cantons within the confederation.
Here too, it is the existence of predominant power, located in one segment of
the federal system, which makes federalism possible in the first place.
By the same token, the unchallengeable supremacy of the United States
within the Western Hemisphere has throughout been the backbone of the
system of American states. As long as this supremacy is secure, there is, on
the one hand, no need for the United States to assert it in the political and
military sphere, and, taking it for granted, the United States can well afford
to pursue a policy of the Good Neighbor; and there is, on the other hand,
no opportunity for the other members of the system to challenge that
supremacy effectively. This is what the principle of the "coiirdinate state"
amounts to in the Western Hemisphere. Consequently, whenever there was
even a remote possibility that the supremacy of the United States might be
challenged, generally through instigation from outside the hemisphere, the
United States asserted its superior power within the hemisphere and acted
as all states must act under similar conditions.
Whatever possibility for common political action there remains among
the members of the Commonwealth of Nations is the result of the interests
which these members may have in common. In other words, the member
states may work together or each of them may work with other nations, as
their interests dictate. Their membership in the Commonwealth, as the
examples of India, South Africa, Australia, and New Zealand clearly show,
has no influence upon this decision; that membership is but a faint remem-
brance of the times when Great Britain could secure cooperation among the
member states on its terms by virtue of its superior power.
What, then, have these four examples of the "coiirdinate state" in com-
mon which would establish them as a distinct type of interstate relation-
ship, and what conclusions can be drawn from them for the organization of
the world? The only thing that these four examples seem to have really in
common is the legal stipulation of the equality of the members of the
respective systems and this characteristic is not peculiar to them, but a gen-
eral principle of international law applicable to all sovereign states. In the
political sphere they seem to have nothing in common at all. What they tend
to show, however, is the decisive importance of the distribution of political
power for the operation of federal and egalitarian relations among states.
The political cohesion of a federal system is the result of superior power
located in some part of it. It is by virtue of its superior power that the pre-
dominant part can afford to grant the other members of the federal system
a measure of equality in the non-political sphere. These observations bring
us back to power politics and the balance of power to which the principle
of the "coordinate state" was supposed t o be the alternative.
In truth, it is not the disinterested consideration of facts which has given
birth to the theory of the "coordinate state." That theory is rather the response
to an emotional urge, and since this emotion is not peculiar to a particular
author but typical of a popular reaction to the new role which the United
States must play in world affairs, it deserves a brief analysis.
One of the great experiences of our time which have impressed them-
selves upon the American mind is the emergence of the United States as a
nation among other nations, exposed to the same opportunities, tempta-
tions, risks, and liabilities to which other nations have been traditionally
exposed. This experience becomes the more shocking if it is compared with
the expectation with which we fought the Second World War. We expected
from that war a reaffirmation of the secure, detached, and independent
position in world affairs which we had inherited from the Founding Fathers
and which we had been successful in preserving at least to the First World
War. By avoiding what we thought had been Wilson's mistakes, we expected
t o emerge from that war if not more independent, certainly more secure
than we were when we entered it. In fact, ~ r o b a b not l ~ even in the early
days of the Republic were we more exposed to danger from abroad than we
are today, and never had we less freedom of action in taking care of our
interests than we have today.
It is naturally shocking to recognize that a happy chapter in the history
of the nation and in one's own way of life has come to an end. There are
those who reconcile themselves to the inevitable, albeit with sorrow rather
than with glee, and try to apply the lessons of the past to the tasks at hand.
There are others who try to escape from a disappointing and threatening
reality into the realm of fantasy. Three such escapist fantasies have arisen in
our midst in response to the challenge of American world leadership and
power: the fantasy of needless American participation in war, the fantasy of
American treason, and the fantasy of American innocence.
The first of these fantasies presumes that the present predicament is a
result not of necessity but of folly, the folly of American statesmen who
needlessly intervened in two world wars. The second of these fantasies
attributes the present predicament to treason in high places whereby the
fruits of victory were handed to the enemy. The third of these fantasies
denies that the predicament is real and prefers to think of it as an intellec-
tual fraud perpetrated upon the American people. To support this fictional
denial of the actualities of the present, it draws upon a fictional account of
the past. The United States does not need to bear at present the intellectual,
moral, and political burdens which go with involvement in power politics
and the maintenance of the balance of power; for it has never borne them in
the past, never having been thus involved. The golden age of past political
I hi 1 T h e National interest of T h e U.S. 37
innocence sheds its glow upon a but seemingly less innocent present and
promises a future in which all the world will follow the example of America,
forswear power politics and the balance of power, and accept the principle
of the "coiirdinate state." Our rearmament program, as exemplified in the
Atlantic Security Pact, we are told, has nothing to do with the balance of
power but aims at the "organization of as much of the world as we can upon
the basis of the coiirdinate state. . . . It may prove impossible under present
conditions to build such a system without having to fight a war with Russia,
but then at least we will be fighting, as we did before, for the thing we con-
sider worth defending with our lives and treasure."" Thus a fictional
account of the American past, begun as an act of uncalled-for patriotic piety,
issues in an ideology for a third world war. Escape we must from the unfamil-
iar, unpleasant, and dangerous present, first into the political innocence of
the past and from there into the immediate future of a third world war,
beyond which the revived and universalized innocence of the more distant
future will surely lie.
We have said that to present the American tradition in foreign policy as
having been free from concern with power politics and the balance of
power is not warranted by the facts of American history. Yet it might still
be argued, and it is actually being argued, that, regardless of the evidence
of history, the American people will not be reconciled to power politics and
the balance of power and will support only policies based upon abstract
moral principles. While in the past the United States might have pursued
balance of power policies and while it might be a good thing if it did do so
again, the American people will not stand for it. Here the emotional appeal
to patriotic piety is joined by calculations of political expediency. Yet the
case for misrepresenting American history has nothing to gain from either.
There is a strong tendency in all historiography to glorify the national
past, and in popular presentations that tendency takes on the aspects of the
jingoist whitewash. Even so penetrating a mind as John Stuart Mill's could
deliver himself of an essay in which he proved, no doubt to the satisfaction
of many of his English readers but certainly of few others, that Great Britain
had never interfered in the affairs of European nations and had interfered
in those of the Indian states only for their own good.'' Yet it is the measure
of a nation's maturity to be able to recognize its past for what it actually is.
Why should we not admit that American foreign policy has been generally
hardheaded and practical and at times ruthless? Why should we deny
Jefferson's cunning, say, in the Puget Sound affair, the cruelty with which
the Indians were treated, and the faithlessness with which the treaties with
the Indians were cast aside? We know that this is the way all nations are
when their interests are at stake - so cruel, so faithless, so cunning. We know
that the United States has refrained from seeking dominions beyond the
seas not because it is more virtuous than other nations, but because it had
the better part of a continent to colonize.
As has been pointed out elsewhere at greater length, the man in the
street, unsophisticated as he is and uninformed as he may be, has a surer
grasp of the essentials of foreign policy and a more mature judgment of its
basic issues than many of the intellectuals and politicians who pretend to
speak for him and cater to what they imagine his prejudices to be. During
the recent war the ideologues of the Atlantic Charter, the Four Freedoms,
and the United Nations were constantly complaining that the American sol-
dier did not know what he was fighting for. Indeed, if he was fighting for
some Utopian ideal, divorced from the concrete experiences and interests of
the country, then the complaint was well grounded. However, if he was
fighting for the territorial integrity of the nation and for its survival as a free
country where he could live, think, and act as he pleased, then he had never
any doubt about what he was fighting for. Ideological rationalizations and
justifications are indeed the indispensable concomitants of all political
action. Yet there is something unhealthy in a craving for ideological intoxi-
cation and in the inability to act and t o see merit in action except under the
stimulant of grandiose ideas and far-fetched schemes. Have our intellectuals
become, like Hamlet, too much beset by doubt to act and, unlike Hamlet,
compelled to still their doubts by renouncing their sense of what is real? The
man in the street has n o such doubts. It is true that ideologues and dema-
gogues can sway him by appealing to his emotions. But it is also true, as
American history shows in abundance and as the popular success of
Ambassador Kennan's book demonstrates, that responsible statesmen can
guide him by awakening his latent understanding of the national interest.
Yet what is the national interest? How can we define it and give it the content
which will make it a guide for action? This is one of the relevant questions to
which the current debate has given rise.
It has been frequently argued against the realist conception of foreign
policy that its key concept, the national interest, does not provide an accept-
able standard for political action. This argument is in the main based upon
two grounds: the elusiveness of the concept and its susceptibility to inter-
pretations, such as limitless imperialism and narrow nationalism, which are
not in keeping with the American tradition in foreign policy. The argument
has substance as far as it goes, but it does not invalidate the usefulness of
the concept.
The concept of the national interest is similar in two respects to the "great
generalities" of the Constitution, such as the general welfare and due process.
It contains a residual meaning which is inherent in the concept itself, but
beyond these minimum requirements its content can run the whole gamut of
meanings which are logically compatible with it. That content is determined
by the political traditions and the total cultural context within which a nation
formulates its foreign policy. The concept of the national interest, then, con-
tains two elements, one that is logically required and in that sense necessary,
and one that is variable and determined by circumstances.
\lo~ci.~itl-iau T h e National Interest of T h e U.S. 39
Any foreign policy which operates under the standard of the national
interest must obviously have some reference to the physical, political, and cul-
tural entity which we call a nation. In a world where a number of sovereign
nations compete with and oppose each other for power, the foreign policies
of all nations must necessarily refer to their survival as their minimum
requirements. Thus all nations d o what they cannot help but do: protect their
physical, political, a n d cultural identity against encroachments by other
nations.
It has been suggested that this reasoning erects the national state into the
last word in politics and the national interest into an absolute standard for
political action. This, however, is not quite the case. The idea of interest is
indeed of the essence of politics and, as such, unaffected by the circum-
stances of time and place. Thucydides' statement, born of the experiences
of ancient Greece, that "identity of interest is the surest of bonds whether
between states or individuals" was taken up in the nineteenth century by
Lord Salisbury's remark that "the only bond of union that endures" among
nations is "the absence of all clashing interests." The perennial issue between
the realist and utopian schools of thought over the nature of politics, to which
we have referred before, might well be formulated in terms of concrete inter-
ests vs. abstract principles. Yet while the concern of politics with interest is
perennial, the connection between interest and the national state is a product
of history.
The national state itself is obviously a product of history and as such
destined t o yield in time to different modes of political organization. As
long as the world is politically organized into nations, the national interest
is indeed the last word in world politics. When the national state will have
been replaced by another mode of organization, foreign policy must then
protect the interest in survival of that new organization. For the benefit of
those who insist upon discarding the national state and constructing supra-
national organizations by constitutional fiat, it must be pointed out that
these new organizational forms will either come into being through con-
quest or else through consent based upon the mutual recognition of the
national interests of the nations concerned; for no nation will forego its
freedom of action if it has n o reason to expect proportionate benefits in
compensation for that loss. This is true of treaties concerning commerce or
fisheries as it is true of the great compacts, such as the European Coal and
Steel Community, through which nations try to create supranational forms
of organization. Thus, by a n apparent paradox, what is historically relative
in the idea of the national interest can be overcome only through the pro-
motion in concert of the national interest of a number of nations.
The survival of a political unit, such as a nation, in its identity is the irre-
ducible minimum, the necessary element of its interests vis-a-vis other units.
Taken in isolation, the determination of its content in a concrete situation is
relatively simple; for it encompasses the integrity of the nation's territory, of
its political institutions, and of its culture. Thus bipartisanship in foreign pol-
icy, especially in times of war, has been most easily achieved in the promotion
of these minimum requirements of the national interest. The situation is

different with respect to the variable elements of the national interest. All the
cross currents of personalities, public opinion, sectional interests, partisan
politics, and political and moral folkways are brought to bear upon their
determination. In consequence, the contribution which science can make to
this field, as to all fields of policy formation, is limited. It can identify the dif-
ferent agencies of the government which contribute to the determination of
the variable elements of the national interest and assess their relative weight.
It can separate the long-range objectives of foreign policy from the short-term
ones which are the means for the achievement of the former and can tenta-
tively establish their rational relations. Finally, it can analyze the variable elem-
ents of the national interest in terms of their legitimacy and their compatibility
with other national values and with the national interest of other nations.
We shall address ourselves briefly to the typical problems with which this
analysis must deal.
The legitimacy of the national interest must be determined in the face
of possible usurpation by subnational, other-national, and supranational
interests. O n the subnational level we find group interests, represented par-
ticularly by ethnic and economic groups, who tend to identify themselves
with the national interest. Charles A. Beard has emphasized, however one-
sidedly, the extent to which the economic interests of certain groups have
been presented as those of the United States.14 Group interests exert, of
course, constant pressure upon the conduct of our foreign policy, claiming
their identity with the national interest. It is, however, doubtful that, with
the exception of a few spectacular cases, they have been successful in deter-
mining the course of American foreign policy. It is much more likely, given
the nature of American domestic politics, that American foreign policy,
insofar as it is the object of pressures by sectional interests, will normally be
a compromise between divergent sectional interests. The concept of the
national interest, as it emerges from this contest as the actual guide for for-
eign policy, may well fall short of what would be rationally required by the
overall interests of the United States. Yet the concept of the national inter-
est which emerges from this contest of conflicting sectional interests is also
more than any particular sectional interest or their sum total. It is, as it
were, the lowest common denominator where sectional interests and the
national interest meet in an uneasy compromise which may leave much t o
be desired in view of all the interests concerned.
The national interest can be usurped by other-national interests in two
typical ways. The case of treason by individuals, either out of conviction or
for pay, needs only to be mentioned here; for insofar as treason is committed
on behalf of a foreign government rather than a supranational principle, it is
significant for psychology, sociology, and criminology, but not for the theory
of politics. The other case, however, is important not only for the theory of
politics but also for its practice, especially in the United States.
National minorities in European countries, ethnic groups in the United
States, ideological minorities anywhere may identify themselves, either
\I ltLlillt The National Interest of The U.S. 41
spontaneously or under the direction of the agents of a foreign government,

with the interests of that foreign government and may promote these inter-
ests under the guise of the national interest of the country whose citizens
they happen to be. The activities of the German-American Bund in the
United States in the 'thirties and of Communists everywhere are cases in
point. Yet the issue of the national interest vs. other-national interests mas-
querading as the national interest has arisen constantly in the United States
in a less clear-cut fashion.
A country which had been settled by consecutive waves of "foreigners"
was bound to find it particularly difficult to identify its own national inter-
est against alleged, seeming, or actual other-national interests represented
by certain groups among its own citizens. Since virtually all citizens of the
United States are, as it were, "more or less" foreign-born, those who were
"less" so have frequently not resisted the temptation to use this distinction
as a polemic weapon against late comers who happened to differ from them
in their conception of the national interest of the United States. Frequently,
this rationalization has been dispensed with and a conception of foreign
policy with which a writer happened to disagree has been attributed out-
right to foreign sympathy or influence or worse. British influence and inter-
ests have served as standard arguments in debates on American foreign
policy. Madison, in his polemic against Hamilton on the occasion of
Washington's Neutrality Proclamation of 1793, identified the Federalist
position with that of "the foreigners and degenerate citizens among us, who
hate our republican government, and the French revolution,"" and the
accusation met with a favorable response in a majority of Congress and of
public opinion. However, these traditional attempts to discredit dissenting
opinion as being influenced by foreign interests should not obscure the real
issue, which is the peculiar vulnerability of the national interest of the
United States to usurpation by the interests of other nations.
The usurpation of the national interest by supranational interests can
derive in our time from two sources: religious bodies and international
organizations. The competition between church and state for determination
of certain interests and policies, domestic and international, has been an
intermittent issue throughout the history of the national state. Here, too,
the legitimate defense of the national interest against usurpation has fre-
quently, especially in the United States, degenerated into the demagogic
stigmatization of dissenting views as being inspired by Rome and, hence,
being incompatible with the national interest. Yet here, too, the misuse of
the issue for demagogic purposes must be considered apart from the legit-
- - - -
imacy of the issue itself.

The more acute problem arises at the present time from the importance
which the public and government officials, at least in their public utterances,
attribute to the values represented and the policies pursued by international
organizations either as alternatives or supplements to the values and policies
for which the national government stands. It is frequently asserted that the
foreign policy of the United States pursues no objectives apart from those of
the United Nations, that, in other words, the foreign policy of the United
States is actually identical with the policy of the United Nations. This asser-
tion cannot refer to anything real in actual politics to support it. For the con-
stitutional structure of international organizations, such as the United
Nations, and their procedural practices make it impossible for them to pursue
interests apart from those of the member-states which dominate their policy-
forming bodies. The identity between the interests of the United Nations and
the United States can only refer to the successful policies of the United States
within the United Nations through which the support of the United Nations
is being secured for the policies of the United States.16The assertion, then, is
mere polemic, different from the one discussed previously in that the identi-
fication of a certain policy with a supranational interest does not seek to
reflect discredit upon the former, but to bestow upon it a dignity which the
national interest pure and simple is supposed to lack.
The real issue in view of the problem that concerns us here is not
whether the so-called interests of the United Nations, which d o not exist
apart from the interests of its most influential members, have superseded
the national interest of the United States, but for what kind of interests the
United States has secured United Nations support. While these interests
cannot be United Nations interests, they do not need to be national inter-
ests either. Here we are in the presence of that modern phenomenon which
has been variously described as "utopianism," "sentimentalism," "moral-
ism," the "legalistic-moralistic approach." The common denominator of all
these tendencies in modern political thought is the substitution for the
national interest of a supranational standard of action which is generally
identified with an international organization, such as the United Nations.
The national interest is here not being usurped by sub- or supranational
interests which, however inferior in worth to the national interest, are never-
theless real and worthy of consideration within their proper sphere. What
challenges the national interest here is a mere figment of the imagination, a
product of wishful thinking, which is postulated as a valid norm for inter-
national conduct, without being valid either there or anywhere else. At this
point we touch the core of the present controversy between utopianism and
realism in international affairs; we shall return to it later in this paper.
The national interest as such must be defended against usurpation by
non-national interests. Yet once that task is accomplished, a rational order
must be established among the values which make up the national interest
and among the resources to be committed to them. While the interests which
a nation may pursue in its relation with other nations are of infinite variety
and magnitude, the resources which are available for the pursuit of such
interests are necessarily limited in quantity and kind. N o nation has the
resources to promote all desirable objectives with equal vigor; all nations
must therefore allocate their scarce resources as rationally as possible. The
indispensable precondition of such rational allocation is a clear understand-
ing of the distinction between the necessary and variable elements of the
national interest. Given the contentious manner in which in democracies
h l o r e c . i i ~ h ~ r r The National Interest of The U.S. 43
the variable elements of the national interest are generally determined, the
advocates of an extensive conception of the national interest will inevitably
present certain variable elements of the national interest as though their
attainment were necessary for the nation's survival. In other words, the neces-
sary elements of the national interest have a tendency to swallow up the
variable elements so that in the end all kinds of objectives, actual or poten-
tial, are justified in terms of national survival. Such arguments have been
advanced, for instance, in support of the rearmament of Western Germany
and of the defense of Formosa. They must be subjected to rational scrutiny
which will determine, however tentatively, their approximate place in the
scale of national values.
The same problem presents itself in its extreme form when a nation pur-
sues, or is asked to pursue, objectives which are not only unnecessary for
its survival but tend to jeopardize it. Second-rate nations which dream of
playing the role of great powers, such as Italy and Poland in the interwar
period, illustrate this point. So d o great powers which dream of remaking
the world in their own image and embark upon world-wide crusades, thus
straining their resources to exhaustion. Here scientific analysis has the
urgent task of pruning down national objectives to the measure of available
resources in order to make their pursuit compatible with national survival.
Finally, the national interest of a nation which is conscious not only of
its own interests but also of that of other nations must be defined in terms
compatible with the latter. In a multinational world this is a requirement of
political morality; in an age of total war it is also one of the conditions for
survival.
In connection with this problem two mutually exclusive arguments have
been advanced. O n the one hand, it has been argued against the theory of
international politics here presented that the concept of the national inter-
est revives the eighteenth-century concept of enlightened self-interest, pre-
suming that the uniformly enlightened pursuit of their self-interest by all
individuals, as by all nations, will of itself be conducive to a peaceful and
harmonious society. O n the other hand, the point has been made that the
pursuit of their national interest by all nations makes war the permanent
arbiter of conflicts among them. Neither argument is well taken.
The concept of the national interest presupposes neither a naturally har-
monious, peaceful world nor the inevitability of war as a consequence of
the pursuit by all nations of their national interest. Quite to the contrary, it
assumes continuous conflict and threat of war, to be minimized through the
continuous adjustment of conflicting interests by diplomatic action. N o
such assumption would be warranted if all nations at all times conceived of
their national interest only in terms of their survival and, in turn, defined
their interest in survival in restrictive and rational terms. As it is, their
conception of the national interest is subject to all the hazards of misinter-
pretation, usurpation, and misjudgment to which reference has been made
above. To minimize these hazards is the first task of a foreign policy which
seeks the defense of the national interest by peaceful means. Its second task
is the defense of the national interest, restrictively and rationally defined,

against the national interests of other nations which may or may not be thus
defined. If they are not, it becomes the task of armed diplomacy to convince
the nations concerned that their legitimate interests have nothing to fear
from a restrictive and rational foreign policy and that their illegitimate
interests have nothing to gain in the face of armed might rationally
employed.
We have said before that the utopian and realist positions in international
affairs d o not necessarily differ in the policies they advocate, but that they
part company over their general philosophies of politics and their way of
thinking about matters political. It does not follow that the present debate
is only of academic interest and without practical significance. Both camps,
it is true, may support the same policy for different reasons. Yet if the rea-
sons are unsound, the soundness of the policies supported by them is a mere
coincidence, and these very same reasons may be, and inevitably are,
invoked on other occasions in support of unsound policies. The nefarious
consequences of false philosophies and wrong ways of thinking may for the
time being be concealed by the apparent success of policies derived from
them. You may go to war, justified by your nation's interests, for a moral
purpose and in disregard of considerations of power; and military victory
seems t o satisfy both your moral aspirations and your nation's interests. Yet
the manner in which you waged the war, achieved victory, and settled the
peace cannot help reflecting your philosophy of politics and your way of
thinking about political problems. If these are in error, you may win victory
on the field of battle and still assist in the defeat of both your moral prin-
ciples and the national interest of your country.
Any number of examples could illustrate the real yet subtle practical
consequences which follow from the different positions taken. We have
chosen two: collective security in Korea and the liberation of the nations
that are captives of Communism. A case for both policies can be made from
both the utopian and realist positions, but with significant differences in the
emphasis and substance of the policies pursued.
Collective security as an abstract principle of utopian politics requires
that all nations come to the aid of a victim of aggression by resisting the
aggressor with all means necessary to frustrate his aims. Once the case of
aggression is established, the duty to act is unequivocal. Its extent may be
affected by concern for the nation's survival; obviously no nation will com-
mit outright suicide in the service of collective security. But beyond that ele-
mental limitation no consideration of interest or power, either with regard
to the aggressor or his victim or the nation acting in the latter's defense, can
qualify the obligation to act under the principle of collective security. Thus
high officials of our government have declared that we intervened in Korea
\ ( I ii: T h e National Interest of T h e U S . 45
not for any narrow interest of ours but in support of the moral principle of
collective security.
Collective security as a concrete principle of realist policy is the age-old
maxim, "Hang together or hang separately," in modern dress. It recognizes
the need for nation A under certain circumstances to defend nation B against
attack by nation C. That need is determined, first, by the interest which A
has in the territorial integrity of B and by the relation of that interest to all
the other interests of A as well as to the resources available for the support
of a11 those interests. Furthermore, A must take into account the power
which is at the disposal of aggressor C for fighting A and B as over against
the power available to A and B for fighting C. The same calculation must
he carried on concerning the power of the likely allies of C as over against
those of A and B. Before going to war for the defense of South Korea in the
name of collective security, an American adherent of political realism would
have demanded an answer to the following four questions: First, what is our
interest in the preservation of the independence of South Korea; second,
what is our power to defend that independence against North Korea; third,
what is our power to defend that independence against China and the Soviet
Union; and fourth, what are the chances for preventing China and the Soviet
Union from entering the Korean War?
In view of the principle of collective security, interpreted in utopian
terms, our intervention in Korea was a foregone conclusion. The interpret-
ation of this principle in realist terms might or might not, depending upon
the concrete circumstances of interest and power, have led us to the same
conclusion. In the execution of the policy of collective security the utopian
had to be indifferent to the possibility of Chinese and Russian intervention,
except for his resolution to apply the principle of collective security to any-
body who would intervene on the side of the aggressor. The realist could
not help weighing the possibility of the intervention of a great power on the
side of the aggressor in terms of the interests engaged and the power avail-
able on the other side."
The Truman administration could not bring itself to taking resolutely the
utopian or the realist position. It resolved to intervene in good measure on
utopian grounds and in spite of military advice to the contrary; it allowed
the military commander to advance to the Yalu River in disregard of the risk
of the intervention of a great power against which collective security could
be carried out only by means of a general war, and then refused to pursue
the war with full effectiveness on the realist grounds of the risk of a third
world war. Thus Mr. Truman in 1952 is caught in the same dilemma from
which Mr. Baldwin could extricate himself in 1936 on the occasion of the
League of Nations sanctions against Italy's attack upon Ethiopia only at an
enormous loss to British prestige. Collective security as a defense of the status
quo short of a general war can be effective only against second-rate powers.
Applied against a major power, it is a contradiction in terms, for it means ne-
cessarily a major war. Of this self-defeating contradiction Mr. Baldwin was as
unaware in the 'thirties as Mr. Truman seems to be in 1952. Mr. Churchill
put Mr. Baldwin's dilemma in these cogent terms: "First, the Prime Minister
had declared that sanctions meant war; secondly, he was resolved that there
must be no war; and thirdly, he decided upon sanctions. It was evidently
impossible to comply with these three conditions." Similarly Mr. Truman
had declared that the effective prosecution of the Korean War meant the
possibility of a third world war; he resolved that there must be no third
world war; and he decided upon intervention in the Korean War. Here, too,
it is impossible to comply with these three conditions.
Similar contradictions are inherent in the proposals which would sub-
stitute for the current policy of containment one of the liberation of the
nations presently the captives of Russian Communism. This objective can
be compatible with the utopian or realist position, but the policies designed
to secure it will be fundamentally different according to whether they are
based upon one or the other position. The clearest case to date for the
utopian justification of such policies has been made by Representative
Charles J. Kersten of Wisconsin who pointed to these four "basic defects"
of the "negative policy of containment and negotiated coexistence":
It would be immoral and unchristian to negotiate a permanent agreement

with forces which by every religious creed and moral precept are evil. It
abandons nearly one-half of humanity and the once free nations of Poland,
Czechoslovakia, Hungary, Rumania, Bulgaria, Albania, Lithuania, Latvia,
Esthonia and China to enslavement of the Communist police state.
It is un-American because it violates the principle of the American
Declaration of Independence, which proclaims the rights of all people to
freedom and their right and duty to throw off tyranny.
It will lead to all-out World War 111 because it aligns all the forces of the
non-Communist world in military opposition to and against all the forces
of the Communist world, including the 800,000,000 peoples behind the
Iron Curtain.
The policy of mere containment is uneconomic and will lead to
national bankruptcy.I8
This statement is interesting for its straightforwardness and because it

combines in a rather typical fashion considerations of abstract morality and
of expediency. The captive nations must be liberated not only because their
captivity is immoral, unchristian, and un-American, but also because its
continuation will lead to a third world war and to national bankruptcy. To
what extent, however, these considerations of expediency are invalidated by
their utopian setting will become obvious from a comparison between the
utopian and the realist positions.
From the utopian point of view there can be no difference between the
liberation of Esthonia or Czechoslovakia, of Poland or China; the captivity
of any nation, large or small, close or far away, is a moral outrage which
cannot be tolerated. The realist, too, seeks the liberation of all captive
nations because he realizes that the presence of the Russian armies in the
Ilorqrr1:lid~i T h e National Interest of T h e U.S. 47
heart of Europe and their cooperation with the Chinese armies constitute
the t w o main sources of the imbalance of power which threatens our secur-
ity. Yet before he formulates a program of liberation, he will seek answers
to a number of questions such as these: While the United States has a gen-
eral interest in the liberation of all captive nations, what is the hierarchy of
interests it has in the liberation, say, of China, Esthonia, and Hungary? And
while the Soviet Union has a general interest in keeping all captive nations
in that state, what is the hierarchy of its interests in keeping, say, Poland,
Eastern Germany, and Bulgaria captive? If we assume, as we must on the
historic evidence of two centuries, that Russia would never give up control
over Poland without being compelled by force of arms, would the objective
of the liberation of Poland justify the ruin of western civilization, that of
Poland included, which would be the certain result of a third world war?
What resources does the United States have at its disposal for the liberation
of all captive nations or some of them? What resources does the Soviet
Union have at its disposal to keep in captivity all captive nations or some
of them? Are we more likely to avoid national bankruptcy by embarking
upon a policy of indiscriminate liberation with the concomitant certainty of
war or by continuing the present policy of containment?
It might be that in a particular instance the policies suggested by the
answers to these questions will coincide with Representative Kersten's pro-
posals, but there can be no doubt that in its overall character, substance,
emphasis, and likely consequences a utopian policy of liberation differs fun-
damentally from a realist one.
The issue between liberation as a utopian principle of abstract morality
vs. the realist evaluation of the consequences which a policy of liberation
would have for the survival of the nation has arisen before in American his-
tory. Abraham Lincoln was faced with a dilemma similar to that which con-
fronts us today. Should he make the liberation of the slaves the ultimate
standard of his policy even at the risk of destroying the Union, as many
urged him to do, or should he subordinate the moral principle of universal
freedom to considerations of the national interest? The answer Lincoln gave
to Horace Greeley, a spokesman for the utopian moralists, is timeless in its
eloquent wisdom. "If there be those," he wrote on August 22, 1862,
who would not save the Union unless they could at the same time save
slavery, I do not agree with them. If there be those who would not save
the Union unless they could at the same time destroy slavery, I d o not
agree with them. My paramount object in this struggle is to save the
Union, and is not either to save or to destroy slavery. If 1 could save the
Union without freeing any slave I would d o it, and if I could save it
by freeing all the slaves, I would do it; and if I could save it by freeing
some and leaving others alone I would also do that. What I do about slav-
ery, and the colored race, I do because 1 believe it helps to save the Union;
and what I forbear, I forbear because I do not believe it would help to save
the Union. I shall do less whenever I shall believe what I am doing hurts
the cause, and I shall do more whenever I shall believe doing more will
help the cause. I shall try to correct errors when shown to be errors; and
I shall adopt new views so fast as they shall appear t o be true views.
I have here stated my purpose according to my view of official duty;
and I intend no modification of my oft-expressed personal wish that all
men everywhere could be free.
The foregoing discussion ought to shed additional light, if this is still

needed, upon the moral merits of the utopian and realist positions. This
question, more than any other, seems to have agitated the critics of realism
in international affairs. Disregarding the voluminous evidence, some of
them have picked a few words out of their context t o prove that realism in
international affairs is unprincipled and contemptuous of morality. To men-
tion but one example, one eminent critic summarizes my position, which he
supposes t o deny the possibility of judging the conduct of states by moral
criteria, in these words: "And one spokesman finds 'a profound and neg-
lected truth,' to use his words, in the dictum of Hobbes that 'there is nei-
ther morality nor law outside the state.'"I9 These are indeed my words, but
not all of them. What I actually said was this:
There is a profound and neglected truth hidden in Hobbes's extreme dic-

tum that the state creates morality as well as law and that there is neither
morality nor law outside the state. Universal moral principles, such as
justice or equality, are capable of guiding political action only to the
extent that they have been given concrete content and have been related
to political situations by society.20
It must be obvious from this passage and from all my other writings on
the subject2' that my position is the exact opposite from what this critic
makes it out t o be. I have always maintained that the actions of states are
subject to universal moral principles and I have been careful to differentiate
my position in this respect from that of Hobbes. Five points basic to my
position may need to be emphasized again.
The first point is what one might call the requirement of cosmic humility
with regard to the moral evaluation of the actions of states. To know that
states are subject to the moral law is one thing; to pretend to know what is
morally required of states in a particular situation is quite another. The
human mind tends naturally to identify the particular interests of states, as
of individuals, with the moral purposes of the universe. The statesman in the
defense of the nation's interests may, and at times even must, yield to that
tendency; the scholar must resist it at every turn. For the light-hearted
assumption that what one's own nation aims at and does is morally good and
that those who oppose that nation's policies are evil is morally indefensible
+ilcl 1 The National Interest of The U S . 49
and intellectually untenable and leads in practice to that distortion of judg-

ment, horn of the blindness of crusading frenzy, which has been the curse of
nations from the beginning of time.
The second point which obviously needs to be made again concerns the
effectiveness of the restraints which morality imposes upon the actions of
states.
A discussion of international morality must guard against the two extremes

either of overrating the influence of ethics upon international politics or
else of denying that statesmen and diplomats are moved by anything else
but considerations of material power.
O n the one hand, there is the dual error of confounding the moral
rules which people actually observe with those they pretend to observe
as well as with those which writers declare they ought to observe. ...
O n the other hand, there is the misconception, usually associated
with the general depreciation and moral condemnation of power politics,
discussed above, that international politics is so thoroughly evil that it is
no use looking for ethical limitations of the aspirations for power 011 the
international scene. Yet, if we ask ourselves what statesmen and diplo-
mats are capable of doing to further the power objectives of their
respective nations and what they actually do, we realize that they do
less than they probably could and less than they actually did in other
periods of history. They refuse to consider certain ends and to use certain
means, either altogether or under certain conditions, not because in the
light of expediency they appear impractical or unwise, but because cer-
tain moral rules interpose an absolute barrier. Moral rules d o not permit
certain policies to be considered at all from the point of view of expedi-
ency. Such ethical inhibitions operate in our time on different levels with
different effectiveness. Their restraining function is most obvious and most
effective in affirming the sacrednese of human life in times of peace."
In connection with this passage we have given a number o f historic

examples showing the influence of moral principles upon the conduct of
foreign policy. An example taken from contemporary history will illustrate
the same point. There can be little doubt that the Soviet Union could have
achieved the objectives of its foreign policy at the end of the Second World
War without antagonizing the nations of the West into that encircling
coalition which has been the nightmare of Bolshevist foreign policy since
1917. It could have mitigated cunning for its own sake and the use of force
with persuasion, conciliation, and a trust derived from the awareness of a
partial community of interests and would thereby have minimized the dangers
to itself and the rest of the world which are inherent in the objectives o f its
policies. Yet the Soviet Union was precluded from relying upon these trad-
itional methods of diplomacy by its general conception of human nature,
politics, and morality. In the general philosophy of Bolshevism there is no
room for honest dissent, the recognition of the intrinsic worth of divergent
interests, and genuine conciliation between such interests. O n all levels of

social interaction opposition must be destroyed by cunning and violence,
since it has no right t o exist, rather than be met half way in view of its
intrinsic legitimacy. This being the general conception of the political
morality of Bolshevism, the foreign policy of the Soviet Union is limited to
a much more narrow choice of means than the foreign policies of other
nations.
The United States, for instance, has been able, in its relations with the
nations of Latin America, to replace military intervention and dollar diplo-
macy with the policy of the Good Neighbor. That drastic change was made
possible by the general conception of political morality which has been
prevalent in the United States from its very inception. The United States is
a pluralist society which presupposes the continuing existence and legit-
imacy of divergent interests. These interests are locked in a continuing strug-
gle for supremacy to be decided by force only as a last resort, but normally
through a multitude of institutional agencies which are so devised as to
allow one or the other interest a temporary advantage but none a perman-
ent supremacy a t the price of the destruction of the others. This morality
of pluralism allows the United States, once it is secure in that minimum of
vital interests t o which we have referred above, t o transfer those principles
of political morality to the international scene and to deal with divergent
interests there with the same methods of genuine compromise and concili-
ation which are a permanent element of its domestic political life.
The third point concerns the relations between universal moral prin-
ciples and political action. I have always maintained that these universal
moral principles cannot be applied t o the actions of states in their abstract
universal formulation, but that they must be, as it were, filtered through the
concrete circumstances of time and place. The individual may say for him-
self: "Fiat justitia, pereat mundus"; the state has no right to say so in the
name of those who are in its care. Both individual and state must judge
political action by universal moral principles, such as that of liberty. Yet
while the individual has a moral right to sacrifice himself in defense of such
a moral principle, the state has no moral right to let its moral disapproba-
tion of the infringement of liberty get in the way of successful political
action, itself inspired by the moral principle of national survival. There can
be n o political morality without prudence, that is, without consideration of
the political consequences of seemingly moral action. Classical and
medieval philosophy knew this and so did Lincoln when he said: "I d o the
very best I know how, the very best I can, and I mean to keep doing so until
the end. If the end brings me out all right, what is said against me won't
amount to anything. If the end brings me out wrong, ten angels swearing I
was right would make n o difference." The issue between utopianism and
realism, as it bears on this point, has been put most succinctly by Edmund
Burke, and what he has to say in the following passage about revolution,
that is, civil war, may well be applied mutatis mutandis to all war.
XZorg,vnili,i~~ The National Interest of The U.S. 51
Nothing universal can be rationally affirmed on any moral or any political

subject. Pure metaphysical abstraction does not belong to these matters.
The lines of morality are not like the ideal lines of mathematics. They
are hroad and deep as well as long. They admit of exceptions; they
demand modifications. These exceptions and modifications are not
made by the process of logic, but by the rules of prudence. Prudence is
not only the first in rank of the virtues political and moral, but she is the
director, the regulator, the standard of them all. Metaphysics cannot live
without definition; but Prudence is cautious how she defines. O u r courts
cannot be more fearful in suffering fictitious cases to be brought before
them for eliciting their determination on a point of law than prudent
moralists are in putting extreme and hazardous cases of conscience upon
emergencies not existing. Without attempting, therefore, to define, what
never can be defined, the case of a revolution in government,
- this, 1
think, may be safely affirmed - that a sore and pressing evil is to be
removed, and that a good, great in its amount and unequivocal in its
nature, must be probable almost to a certainty, before the inestimable
price of our own morals and the well-being of a number of our fellow-
citizens is paid for a revolution. If ever we ought to be economists even
to parsimony, it is in the voluntary production of evil. Every revolution
contains in it something of evil."
Fourth, the realist recognizes that a moral decision, especially in the

political sphere, does not imply a simple choice between a moral principle
and a standard of action which is morally irrelevant or even outright
immoral. A moral decision implies always a choice among different moral
principles, one of which is given precedence over others. To say that a polit-
ical action has no moral purpose is absurd; for political action can be
defined as an attempt to realize moral values through the medium of polit-
ics, that is, power. The relevant moral question concerns the choice among
different moral values, and it is at this point that the realist and the utopian
part company again. If an American statesman must choose between the
promotion of universal liberty, which is a moral good, at the risk of
American security and, hence, of liberty in the United States, and the pro-
motion of American security and of liberty in the United States, which is
another moral good, to the detriment of the promotion of universal liberty,
which choice ought he to make? The utopian will not face the issue squarely
and will deceive himself into believing that he can achieve both goods at the
same time. The realist will choose the national interest on both moral and
pragmatic grounds; for if he does not take care of the national interest
nobody else will, and if he puts American security and liberty in jeopardy
the cause of liberty everywhere will be impaired.
Finally, the political realist distinguishes between his moral sympathies
and the political interests which he must defend. He will distinguish with
Lincoln between his "official duty" which is to protect the national interest
52 T h e Cold War a n d N u c l e a r D e t e r r e n c e
and his "personal wish" which is t o see universal moral values realized
throughout the world.
The issue has been admirably put by Father Wilfred Parsons of Catholic
University in defending Ambassador Kennan's position:
Mr. Kennan did not say state behavior is not a fit subject for moral judg-
ment, but only that it should not sway our realization of the realities with
which we have to deal. Msgr. Koenig continues: "Should we accept power
realities and aspirations without feeling the obligation of moral judg-
ment?" And he appeals to the present writer and other political scientists
to say whether this doctrine agrees with Pope Pius XII's messages on peace.
I am sure that most political scientists, and also Mr. Kennan, would
agree with the Monsignor that we should not accept those realities "with-
out feeling the obligation of moral judgment." But there is a difference
between feeling this obligation (and even expressing it) and allowing this
feeling to sway our actions in concrete negotiations that deal with the
national or world common good. We can still feel and yet deal.
To make my meaning clearer, I understood Mr. Kennan to hold that
we went off the beam with Woodrow Wilson, when we began to make
our moral disapprobation an essential part of our foreign relations, even
sometimes at the expense of our own and the world's common good.
Logically, such an attitude would inhibit our dealing with Britain,
France and a host of countries. Pius XI, speaking of Mussolini after the
Lateran Treaty, said he would deal with the devil himself if he must.
Here was moral disapprobation, but it was not "carried over into the
affairs of states."
This relative position, and not the absolute one of Msgr. Koenig
(with which in itself I agree), is, I think, the issue raised by Mr. Kennan,
and it is worth debating on that basis.14
The contest between utopianism and realism is not tantamount to a contest

between principle and expediency, morality and immorality, although some
spokesmen for the former would like to have it that way. The contest is rather
between one type of political morality and another type of political morality,
one taking as its standard universal moral principles abstractly formulated, the
other weighing these principles against the moral requirements of concrete
political action, their relative merits to be decided by a prudent evaluation of
the political consequences to which they are likely to lead.2"
These points are re-emphasized by the foregoing discussion. Which atti-
tude with regard to collective security and to the liberation of the captive
nations, the utopian or the realist, is more likely to safeguard the survival of
the United States in its territorial, political, and cultural identity and at the
same time to contribute the most to the security and liberty of other nations?
This is the ultimate test - political and moral - by which utopianism and
realism must be judged.
11 1 .rr8 The National Interest of The U.S. 53
Notes
1. 7'his IS tlie ideal type ot the u t o p ~ mp o s i t ~ o nrather than the e m p i r ~ c a descr~prion

l of any
p n r t i c u l ~ rh ~ s t o r ~type.
c In actualit!: a n d this is true particularly of the present, tlie utopian
position in internat~onalaffairs is not always consistent with its philosophic premises.
2. It o ~ ~ g not h t t o need special emphasis that a p r i n c ~ p l eof socidl conduct, In contrast t o a
law of ncirure, a l l o w of, 'ind even presupposes, conduct in violarion of the pr~nciple.Robert
W. Tuckel; In "Professor Morgenthau's Theory of Political 'Kealism"' in this Rtvn w, Vol. 46,
pp. 214-224 ( M a r c h , 1952), has missed this a n d many other polnts in his zeal t o find contr'v
dictions where there 'ire none.
3. ‘‘This Itlie realist] doctrine," write\ o n e historian - Frank TLinnenhaurn,"The Balance of
Power versus the Coordinate State,"Pol~trralScience Qrtartrrly, Vol. 67, p. 173 (June, 1952) -
"is contessedl!; nay gleefully, amoral. It prides itself upon heitig real~stica n d takes Machiavell~
as its gre.ir teacher. It I S contemptuous of the s ~ r n p l eheltefs of honest men, jeer\ ,it the wntl-
mcntal~srnof those w h o believe that men may strive for peace a m o n g nations, a n d looks upon
democracy 3s a li~ndrancet o skilled d~plornacy.It looks wlth 1' certaln d e r i s ~ ~s ue p e r i o r ~ t )
upon the great leciders o t this narlon from Jetterson a n d J o h n Quincy Adams t o Woodrow
Wilson a n d Franklin Delano Roosevelt a n d describes them as moralistic and sentimental, and
suggests that o u r models ought t o he Richelieu, Clernenceau and Bismarck. Its adherents
bellebe that international wars instead of being made hy men a n d supported by institutions
humanly c o n r r ~ v e dhave their orlgln ~n the nature of m a n h ~ m s e l f,lnd are i n e \ i t ~ h l e . "
Another historian, Arthur Schlesinger, Jr., in "Policy a n d National Interest." Partisan
Rrlww: Vol. 18, p. 7 0 9 (Nov.-Dec.. 1951), however, gives Ambassaclor Kennan a clean hill ot
moral health. "But w h ~ rd~fferentiates,"he writes, "the Kennan approach from that of, for
example, the followers of Professor H a n s J. Morgenthau is that he takes the revelat~onsof inter-
national amorality in his srr~de;more than that, he comprehends them in his understanding o f
the tragedy of history. Mr. Kennan, in other words, is deeply moral, rather than moralistic, like
ludge Hull, o r i~nrnoral,like the boys w h o have just discovered that politics involve power."
"This dreadful doctrine," we Lire told ( b y Tannenhaum, pp. 173-174), "has n o w won wide
acceptance hy teachers and scholars in tlie field o f internat~onalrelations and li.is, in f , ~ become ,
the leuding theme in such circles in many of our largest un~vers~ties. It has become the sc-lenre of
international relations - and w h o w o ~ ~ quarrel ld with science, especially when it comes packaged
in good cle~lrEnglish and from high sources? Rut ~t is not science. It is, in fact, only poor l o g ~ c
hased upon false premises, a n d its claim t o be a science is only a bit of unlioly conce~t."
It niay be remarked in p'lssing that t o dispose of a scientific theory as "fashionable" o r 1'
" f ~ d , " as some do with regard t o political realism, may reveal something a b o u t the starc of
mind of the writer, hut reveals n o r l i ~ n ga t all about the 5cientific v a l ~ ~ofe the theory.
4. l'annenbaum. in the article cited above, a n d in "The American Tradition in Foreign
Relat~ons." Foreign Affzirs, Vol. 30, pp. 31-50 (Oct., 1951).
5. O n r Foreign Affairs ( N e w York, 1924), pp. 2 4 6 ff.
6 . "The United States a n d the Balance of I'ower," The / o u r n a l of P o l ~ t ~ c sVol. , 3,
pp. 401-449 (Nov., 1941).
7. T'innenhaum, "The Balance of Power verws the Coijrdinate State," (cited above, note 3 ) ,
p. 173.
8. "Democracy a n d Efficiency," Atlrzntic Monthly, Vol. 87, pp. 293-294 (March, 190 1).
9. 1Bnnenbaum. p. 177.
10. Ihrd.
1 1. W~lliarnArchibald Dunning, Essays o n the C i ~ dWar a n d Rcconstrrtction rznd Rrlated
Topics ( N e w York, 1 9 3 1), p. 3 5 I.
12. Tannenbaum, pp. 195-1 96.
13. "A Few Words on Non-interventioti," D m e r t a t i o n s a n d Discnssior~s: l'olitrc~rl,
/'/~f/oso[)/~rc-i~/,
a n d H ~ s t o r i r d I(Imndon, 18751, pp. 15.3-1 78.
14. 'me ldea of Nrrt~orzal Interest: An Atzalytictrl Study 111 Anzericon F o r e ~ ~ qI'oIic-y
n
( N e w York, 19.34).
15. "Helvidius, in Answer to Pacificus, on President Washington's Proclamation of

Neutrality," in Letters and other Writings of James Madison (Philadelphia, 1867), Vol. 1, p. 61 1.
16. See, o n this point, Hans J. Morgenthau, "International Organizations and Foreign
Policy," in Foundations of World Organization: A Political a n d Cultural Appraisal, Eleventh
Symposium of the Conference o n Science, Philosophy and Religion, edited by Lyman Bryson,
Louis Finkelstein, Harold D. Lasswell, R.M. MacIver (New York, 1952), pp. 377-383.
17. The difference in these two attitudes is well illustrated by the following passage from a
recent Moon Mullins cartoon. An elderly representative of the utopian school asks little Kayo:
"Remember the golden rule. Now, supposing that boy slapped you on the right cheek, what
would you do?" Whereupon Kayo replies realistically: "Jest how big a boy are you supposin'?"
18. New York Times, August 14, 1952, p. 1.
19. A.H. Feller, "In Defense of International Law and Morality," The Annals of the
American Academy of Political and Social Scrence, Vol. 282, p. 80 (July, 1952).
20. In Defense of the National Interest: A Critical Exarnrnation of American Foreign
Policy (New York, 1951), p. 34.
21. See, for instance, "The Machiavellian Utopia," Ethics, Vol. 55, pp. 145-147 (Jan., 1945);
"Ethics and Politics," in Approaches to Group Understanding, Sixth Symposium of the
Conference on Science, Philosophy and Religion, edlted by Bryson, Finkelstein, and MacIver
(New York, 1947), pp. 319-341; "The Escape from Power in the Western World," in Conflicts
of Power in Modern Culture, Seventh Symposium of the Conference on Science, Philosophy and
Religion, edited by Bryson, Finkelstein, and MacIver, pp. 1-12; Scientific Man vs. Power Politics
(Chicago, 1946), Chaps. 7, 8; "Views of Nuremberg: Further Analysis of the Trial and Its
Importance," America, Vol. 76, pp. 266-267 (Dec. 7, 1946); "The Twilight of International
Morality," Ethics, Vol. 58, pp. 79-99 (Jan., 1948); "The Political Science of E.H. Carr," World
Politics, Vol. 1, pp. 127-134 (Oct., 1948); Politics Among Natlons (New York, 1948), Ch. 14;
"National Interest and Moral Principles in Foreign Policy: The Primacy of the National Interest,"
The American Scholar, Vol. 18, pp. 207-212 (Spring, 1949); "The Pathology of Power,"
American Perspective, Vol. 4, pp. 6-10 (Winter, 1950); "The Moral Dilemma in Foreign Policy,"
in The Year Book of World Affairs, 1951 (London, 1951), pp. 12-36.
22. Morgenthau, Politics Among Nations, pp. 174-175.
23. The Works of The Right Honorable Edmund Burke, 4th ed. (Boston, 1871), Vol. 4,
pp. 80-81. Cf. also Burke, "Speech on A Bill for Shortening the Duration of Parliaments," May 8,
1780, in Works; Vol. 7, p. 73: "I must see, to satisfy me, the remedies; I must see, from their
operation in the cure of the old evil, and in the cure of those new evils which are inseparable
from all remedies, how they balance each other, and what is the total result. The excellence of
mathematics and metaphysics is, to have but one thing before you; but he forms the best judge-
ment in all moral disquisitions who has the greatest number and variety of considerations in
one view before him, and can take them in with the best possible consideration of the middle
results of all."
24. America, Vol. 86, p. 700 (March 29, 1952). See also Algernon Cecil, "The Foreign
Office," in The Cambrrdge History of Brrtish Forergn Policy, 1783-1 91 9 (New York, 1923),
Vol. 3, p. 605, concerning Lord Salisbury: "Always, however, the motive of his policy was to
be found in the political interests as opposed to the political sympathies of Great Brltaln; and
in this way his treatment of Foreign Affairs is at the opposite policy from that of Palmerston
or Gladstone." Cf. also the general remarks in Alexander H. Leighton, Human Relations rn a
Changing World (New York, 1949), pp. 155 ff.
25. See, on this point, Shlrley R. Letwin, "Rationalism, Prmciples, and Politics," The Review
of Politics, Vol. 14, pp. 367-393 (July, 1952); L. Susan Stebbing, Ideals and lllusions (London,
1941); Vernon H. Holloway, Religious Ethics and the Politics of Power (New York, 1951); and
Dorothy Fosdick, "Ethical Standards and Political Strategies," Political Scrence Quarterly, Vol. 57,
pp. 214 ff. (1942).
The Delicate Balance of Terror
Albert Wohlstetter
T he first shock administered by the Soviet launching of sputnik has

almost dissipated. The flurry of statements and investigations and
improvised responses has died down, leaving a small residue: a slight
increase in the schedule of bomber and ballistic missile production, with a
resulting small increment in our defense expenditures for the current fiscal
year; a considerable enthusiasm for space travel; and some stirrings of interest
in the teaching of mathematics and physics in the secondary schools. Western
defense policy has almost returned to the level of activity and the emphasis
suited to the basic assumptions which were controlling before sputnik.
One of the most important of these assumptions - t h a t a general thermo-
nuclear war is extremely unlikely - is held in common bv most of the
critics of our defense policy as well as by its proponents. Because of its cru-
cial r d e in the Western strategy of defense, I should like to examine the sta-
bility of the thermonuclear balance which, it is generally supposed, would
make aggression irrational or even insane. The balance, I believe, is in fact
precarious, and this fact has critical implications for policy. Deterrence in
the 1960s is neither assured nor impossible but will be the product of sus-
tained intelligent effort and hard choices, responsibly made. As a major
illustration important both for defense and foreign policy, I shall treat the
particularly stringent conditions for deterrence which affect forces based
close to the enemy, whether they are U.S. forces or those of our allies, under
single or joint control. I shall comment also on the inadequacy as well as
the necessity of deterrence, on the problem of accidental outbreak of war,
and on disarmament.'
11. The Presumed Automatic Balance
I emphasize that requirements for deterrence are stringent. We have heard

so much about the atomic stalemate and the receding prohability of war
which it has produced that this may strike the reader as something of an
Source: Eore~gnAffizirs, 3 7 ( 2 ) ( 1 9 5 9 ) :21 1-34
exaggeration. Is deterrence a necessary consequence of both sides having a

nuclear delivery capability, and is all-out war nearly obsolete? Is mutual
extinction the only outcome of a general war? This belief, frequently
expressed by references to Mr. Oppenheimer's simile of the two scorpions
in a bottle, is perhaps the prevalent one. It is held by a very eminent
and diverse group of people - in England by Sir Winston Churchill,
P. M. S. Blackett, Sir John Slessor, Admiral Buzzard and many others; in
France by such figures as Raymond Aron, General Gallois and General
Gazin; in this country by the titular heads of both parties as well as almost
all writers on military and foreign affairs, by both Henry Kissinger and his
critic, James E. King, Jr., and by George Kennan as well as Dean Acheson.
Mr. Kennan refers to American concern about surprise attack as simply
o b s e ~ s i v e and
; ~ many people have drawn the consequence of the stalemate
as has Blackett, who states: "If it is in fact true, as most current opinion
holds, that strategic air power has abolished global war, then an urgent
problem for the West is to assess how little effort must be put into it to keep
global war abolished."' If peace were founded firmly on mutual terror, and
mutual terror on symmetrical nuclear capabilities, this would be, as Churchill
has said, "a melancholy paradox;" none the less a most comforting one.
Deterrence, however, is not automatic. While feasible, it will be much
harder to achieve in the 1960s than is generally believed. One of the most
disturbing features of current opinion is the underestimation of this difficulty.
This is due partly to a misconstruction of the technological race as a problem
in matching striking forces, partly to a wishful analysis of the Soviet ability
to strike first.
Since sputnik, the United States has made several moves to assure the
world (that is, the enemy, but more especially our allies and ourselves) that we
will match or overmatch Soviet technology and, specifically, Soviet offense
technology. We have, for example, accelerated the bomber and ballistic missile
programs, in particular the intermediate-range ballistic missiles. The problem
has been conceived as more or better bombers - or rockets; or sputniks; or
engineers. This has meant confusing deterrence with matching or exceeding
the enemy's ability to strike first. Matching weapons, however, misconstrues
the nature of the technological race. Not, as is frequently said, because only a
few bombs owned by the defender can make aggression fruitless, but because
even many might not. One outmoded A-bomb dropped from an obsolete
bomber might destroy a great many supersonic jets and ballistic missiles. To
deter an attack means being able to strike back in spite of it. It means, in other
words, a capability to strike second. In the last year or two there has been a
growing awareness of the importance of the distinction between a "strike-
first" and a "strike-second" capability, but little, if any, recognition of the
implications of this distinction for the balance of terror theory.
Where the published writings have not simply underestimated Soviet
capabilities and the advantages of a first strike, they have in general placed
artificial constraints on the Soviet use of the capabilities attributed to them.
They assume, for example, that the enemy will attack in mass over the
\i il11~ii.t1 i . t The Delicate Balance of Terror 57
Arctic through our D ~ s t a n tEarly Warning line, with bombers refueled over
Canada - all resulting in plenty of warning. Most hopefully, it is sometimes
assumed that such attacks will be preceded by days of visible preparations
for moving ground troops. Such assumptions suggest that the Soviet leaders
will be rather bumbling or, better, cooperative. However attractive it may
be for us to narrow Soviet alternatives to these, they would be low in the
order of reference of any reasonable Russians ~ l a n n i n gwar.
I l l . The Quantitative Nature of the Problem and the Uncertainties
In treating Soviet strategies it is important to consider Soviet rather than

Western advantage and to consider the strategy of both sides quantitatively.
The effectiveness o f our own choices will depend on a most complex
numerical interaction of Soviet and Western plans. Unfortunately, both the
privileged and unprivileged information on these matters is precarious. As
a result, competent people have been led into critical error in evaluating the
prospects for deterrence. Western journalists have greatly overestimated
the difficulties of a Soviet surprise attack with thermonuclear weapons and
vastly underestimated the complexity of the Western problem of retaliation.
One intelligent commentator, Richard Rovere, recently expressed the
common view: "If the Russians had ten thousand warheads and a missile for
each, and we had ten hydrogen bombs and ten obsolete bombers, ... aggres-
sion would still be a folly that would appeal only to an insane adventurer."
Mr. Rovere's example is plausible because it assumes implicitly that the
defender's hydrogen bombs will with certainty be visited on the aggressor;
then the damage done by the ten bombs seems terrible enough for deter-
rence, and any more would be simply redundant. This is the basis for the
common view. The example raises questions, even assuming the delivery of
the ten weapons. For instance, the targets aimed at in retaliation might be
sheltered and a quite modest civil defense could hold within tolerable limits
the damage done to such city targets by ten delivered bombs. But the essen-
tial point is that the weapons would not be very likely to reach their targets.
Even if the bombers were dispersed at ten different points, and protected by
shelters so blast resistant as to stand up anywhere outside the lip of the
bomb crater - even inside the fire ball itself - the chances of one of these
bombers surviving the huge attack directed at it would be on the order of
one in a million. (This calculation takes account of the unreliability and in-
accuracy of the missile.) And the damage done by the small minority of these
ten planes that might be in the air at the time of the attack, armed and ready
to run the gauntlet of an alert air defense system, if not zero, would be very
small indeed compared to damage that Russia has suffered in the past. For
Mr. Rovere, like many other writers on this subject, numerical superiority is
not important at all.
For Joseph Alsop, on the other hand, it is important, but the superiority
is on our side. Mr. Alsop recently enunciated as one of the four rules of
available when he was writing in 1956 on weapons for all-out war. But
much of his analysis was based on the assumption that H-bombs could not
be made small enough to be carried in an intercontinental missile. It is now
widely known that intercontinental ballistic missiles will have hydrogen
warheads, and this fact, a secret at the time, invalidates Mr. Blackett's
calculations and, I might say, much of his optimism on the stability of the
balance of terror. In sum, one of the serious obstacles to any widespread
rational judgment on these matters of high policy is that critical elements of
the problem have to be protected by secrecy. However, some of the principal
conclusions about deterrence in the early 1960s can be fairly firmly based,
and based on public information.
IV. The Delicacy of the Balance of Terror
The most important conclusion is that we must expect a vast increase in the
weight of attack which the Soviets can deliver with little warning, and the
growth of a significant Russian capability for an essentially warningless
attack. As a result, strategic deterrence, while feasible, will be extremely dif-
ficult to achieve, and at critical junctures in the 1960s, we may not have the
power to deter attack. Whether we have it or not will depend on some dif-
ficult strategic choices as to the future composition of the deterrent forces
as well as hard choices on its basing, operations and defense.
Manned bombers will continue to make up the predominant part of our
striking force in the early 1960s. None of the popular remedies for their
defense will suffice - not, for example, mere increase of alertness (which
will be offset by the Soviet's increasing capability for attack without signifi-
cant warning), nor simple dispersal or sheltering alone or mobility taken
by itself, nor a mere piling up of interceptors and defense missiles around
SAC bases. Especially extravagant expectations have been placed on the
airborne alert - an extreme form of defense by mobility. The impression is
rather widespread that one-third of the SAC bombers are in the air and
ready for combat at all times.8 This belief is belied by the public record.
According t o the Symington Committee Hearings in 1956, our bombers
averaged 31 hours of flying per month, which is about 4 percent of the
average 732-hour month. An Air Force representative expressed the hope
that within a couple of years, with an increase in the ratio of crews to
aircraft, the bombers would reach 45 hours of flight per month - which is
6 percent. This 4 to 6 percent of the force includes bombers partially fueled
and without bombs. It is, moreover, only an average, admitting variance
down as well as up. Some increase in the number of armed bombers aloft
is to be expected. However, for the current generation of bombers, which
have been designed for speed and range rather than endurance, a continu-
ous air patrol for one-third of the force would be extremely expensive.
O n the other hand, it would be unwise t o look for miracles in the new
weapons systems, which by the mid-1960s may constitute a considerable
'\ $1 iii iIL The Delicate Balance of Terror 61
portion of the United States force. After the Thor, Atlas and Titan there are
a number of promising developments. The solid-fueled rockets, Minuteman
and Polaris, promise in particular to be extremely significant components
of the deterrent force. Today they are being touted as making the problem of
deterrence easy to solve and, in fact, guaranteeing its solution. But none of
the new developments in vehicles is likely to d o that. For the complex job
of deterrence, they all have limitations. The unvaryingly immoderate claims
for each new weapons system should make us wary of the latest "techno-
logical breakthroughs." Only a very short time ago the ballistic missile itself
was supposed to be intrinsically invulnerable on the ground. It is now more
generally understood that its survival is likely to depend on a variety of
choices in its defense.
It is hard to talk with confidence about the mid and late-1960s. A sys-
tematic study of an optimal or a good deterrent force which considered all
the major factors affecting choice and dealt adequately with the uncertain-
ties would be a formidable task. In lieu of this, I shall mention briefly why
none of the many systems available or projected dominates the others in any
obvious way. My comments will take the form of a swift run-through of the
characteristic advantages and disadvantages of various strategic systems at
each of the six successive hurdles mentioned earlier.
The first hurdle to be surmounted is the attainment of a stable, steady-
state peacetime operation. Systems which depend for their survival on
extreme decentralization of controls, as may be the case with large-scale dis-
persal and some of the mobile weapons, raise problems of accidents and over
a long period of peacetime operation this leads in turn to serious political
problems. Systems relying on extensive movement by land, perhaps by truck
caravan, are an obvious example; the introduction of these on European
roads, as is sometimes suggested, would raise grave questions for the govern-
ments of some of our allies. Any extensive increase in the armed air alert will
increase the hazard of accident and intensify the concern already expressed
among our allies. Some of the proposals for bombardment satellites may
involve such hazards of unintended bomb release as to make them out of the
question.
The cost to buy and operate various weapons systems must be seriously
considered. Some systems buy their ability to negotiate a given hurdle - say,
surviving the enemy attack - only at prohibitive cost. Then the number that
can be bought out of a given budget will be small and this will affect the
relative performance of competing systems at various other hurdles, for
example penetrating enemy defenses. Some of the relevant cost comparisons,
then, are between competing systems; others concern the extra costs to the
enemy of canceling an additional expenditure of our own. For example, some
dispersal is essential, though usually it is expensive; if the dispersed bases
are within a warning net, dispersal can help to provide warning against some
sorts of attack, since it forces the attacker to increase the size of his raid and
so makes it more liable to detection as well as somewhat harder to coijrd-
inate. But as the sole or principal defense of our offensive force, dispersal has
First, since thermonuclear weapons give an enormous advantage to the

aggressor, it takes great ingenuity and realism at any given level of nuclear
technology to devise a stable equilibrium. And second, this technology itself
is changing with fantastic speed. Deterrence will require an urgent and
continuing effort.
V. The U s e s and Risks of B a s e s Close t o t h e Soviets
It may now be useful to focus attention on the special problems of deterrent

forces close to the Soviet Union. First, overseas areas have played an impor-
tant r d e in the past and have a continuing though less certain r61e today.
Second, the recent acceleration of production of intermediate-range ballistic
missiles and the negotiation of agreements with various NATO powers for
their basing and operation have given our overseas bases a renewed impor-
tance in deterring attack on the United States - or so it would appear at first
blush. Third, an analysis can throw some light on the problems faced by our
allies in developing an independent ability to deter all-out attack on them-
selves, and in this way it can clarify the much agitated question of nuclear
sharing. Finally, overseas bases affect in many critical ways, political and
economic as well as military, the status of the alliance.
At the end of the last decade, overseas bases appeared to be an advanta-
geous means of achieving the radius extension needed by our short-legged
bombers, of permitting them to use several axes of attack, and of increasing
the number of sorties possible in the course of an extended campaign. With
the growth of our own thermonuclear stockpile, it became apparent that a
long campaign involving many re-uses of a large proportion of our bombers
was not likely to be necessary. With the growth of a Russian nuclear-delivery
capability, it became clear that this was most unlikely to be feasible.
Our overseas bases now have the disadvantage of high vulnerability.
Because they are closer than the United States to the Soviet Union, they are
subject to a vastly greater attack by a larger variety as well as number of
vehicles. With given resources, the Soviets might deliver on nearby bases a
freight of bombs with something like 50 t o 100 times the yield that they
could muster at intercontinental range. Missile accuracy would more than
double. Because there is not much space for obtaining warning - in any case,
there are no deep-warning radar nets - and, since most of our overseas bases
are close to deep water from which submarines might launch missiles, the
warning problem is very much more severe than for bases in the interior of
the United States.
As a result, early in the 1950s the U.S. Air Force decided to recall many
of our bombers to the continental United States and to use the overseas
bases chiefly for refueling, particularly post-strike ground refueling. This
reduced drastically the vulnerability of U.S. bombers and at the same time
retained many of the advantages of overseas operation. For some years now
SAC has been reducing the number of aircraft usually deployed overseas.
\1 11 i:itttt I The Delicate Balance of Terror 65
The purpose is to reduce vulnerability and has little to do with any increas-
ing radius of SAC aircraft. The early B-52 radius is roughly that of the
B-36; the B-47, roughly that of the B-50 or B-29. In fact the radius limitation
and therefore the basing requirements we have discussed will not change
substantially for some time to come. We can talk with comparative confi-
dence here, because the U.S. strategic force is itself largely determined for
this period. Such a force changes more slowly than is generally realized. The
vast majority of the force will consist of manned bombers, and most of
these will be of medium range. Some U.S. bombers will be able to reach
some targets from some U.S. bases within the 48 states without landing on
the way back. O n the other hand, some bomber-target combinations are
not feasible without pre-target landing (and are therefore doubtful). The
Atlas, Titan and Polaris rockets, when available, can of course do without
overseas bases (though the proportion of Polaris submarines kept at sea can
be made larger by the use of submarine tenders based overseas). But even
with the projected force of aerial tankers, the greater part of our force,
which will be manned bombers, cannot be used at all in attacks on the
Soviet Union without at least some use of overseas areas.
What of the bases for Thor and Jupiter, our first intermediate-range ballis-
tic missiles? These have to be close to the enemy, and they must of course be
operating bases, not merely refueling stations. The Thors and Jupiters will be
continuously in range of an enormous Soviet potential for surprise attack.
These installations therefore re-open, in a most acute form, some of the ser-
ious questions of ground vulnerability that were raised about six years ago in
connection with our overseas bomber bases. The decision to station the Thor
and Jupiter missiles overseas has been our principal public response to the
Russian advances in rocketry, and perhaps our most plausible response.
Because it involves our ballistic missiles it appears directly to answer the
Russian rockets. Because it involves using European bases, it appears to make
up for the range superiority of the Russian intercontinental missile. And most
important, it directly involves the NATO powers and gives them an element
of control.
There is no question that it was genuinely urgent not only to meet the
Russian threat but to d o so visibly, in order to save the loosening NATO
alliance. Our allies were fearful that the Soviet ballistic missiles might mean
that we were no longer able or willing to retaliate against the Soviet Union in
case of an attack on them. We hastened to make public a reaction which
would restore their confidence. This move surely appears to increase our own
power to strike back, and also to give our allies a deterrent of their own, inde-
pendent of our decision. It has also been argued that in this respect it merely
advances the inevitable date at which our allies will acquire "modern"
weapons of their own, and that it widens the range of Soviet challenges which
Europe can meet. But we must face seriously the question whether this move
will in fact assure either the ability to retaliate or the decision to attempt it,
on the part of our allies or ourselves. And we should ask at the very least
whether further expansion of this policy will buy as much retaliatory power
precisely who their legal owner is will not affect the retaliatory power of the
Thors and Jupiters one way or the other. They would not be able to deter an
attack which they could not survive. It is curious that many who question
the utility of American overseas bases (for example, our bomber bases in the
United Kingdom) simply assume that, for our allies, possession of strategic
nuclear weapons is one with deterrence.
There remains the view that the provision of these weapons will broaden
the range of response open to our allies. In so far as this view rests on the
belief that the intermediate-range ballistic missile is adapted to limited war,
it is wide of the mark. The inaccuracy of an I.R.B.M. requires high-yield
warheads, and such a combination of inaccuracy and high yield, while quite
appropriate and adequate against unprotected targets in a general war,
would scarcely come within even the most lax, in fact reckless, definition of
limited war. Such a weapon is inappropriate for even the nuclear variety of
limited war, and it is totally useless for meeting the wide variety of provo-
cation that is well below the threshold of nuclear response. In so far as these
missiles will be costly for our allies to install, operate and support, they are
likely t o displace a conventional capability that might be genuinely useful
in limited engagements. More important, they are likely to be used as an
excuse for budget cutting. In this way they will accelerate the general trend
toward dependence on all-out response and so will have the opposite effect
to the one claimed.
Nevertheless, if the Thor and Jupiter have these defects, might not some
future weapon be free of them? Some of these defects, of course, will be over-
come in time. Solid fuels or storable liquids will eventually replace liquid oxy-
gen, reliabilities will increase, various forms of mobility or portability will
become feasible, accuracies may even be so improved that such weapons can
be used in limited wars. But these developments are all years away. In con-
sequence, the discussion will be advanced if a little more precision is given
such terms as "missiles" or "modern" or "advanced weapons." We are not
distributing a generic "modern" weapon with all the virtues of flexibility in
varying circumstances and of invulnerability in all-out war. But even with
advances in the state of the art on our side, it will remain difficult to maintain
a deterrent, especially close in under the enemy's guns.
It follows that, though a wider distribution of nuclear weapons may be
inevitable, or a t any rate likely, and though some countries in addition to
the Soviet Union and the United States may even develop an independent
deterrent, it is by n o means inevitable or even very likely that the power to
deter all-out thermonuclear attack will be widespread. This is true even
though a minor power would not need t o guarantee as large a retaliation as
we in order to deter attack on itself. Unfortunately, the minor powers have
smaller resources as well as poorer strategic location^.^ Mere membership
in the nuclear club might carry with it prestige, as the applicants and nom-
inees expect, but it will be rather expensive, and in time it will be clear that
it does not necessarily confer any of the expected privileges enjoyed by the
two charter members. The burden of deterring a general war as distinct
I r1 The Delicate Balance of Terror 69
from limited wars is still likely to be on the United States and therefore, so
far as our allies are concerned, on the military alliance.
There is one final consideration. Missiles placed near the enemy, even if
they could not retaliate, would have a potent capability for striking first by
surprise. And it might not be easy for the enemy to discern their purpose.
The existence of such a force might be a considerable provocation and in
fact a dangerous one in the sense that it would place a great burden on our
deterrent force which more than ever would have to guarantee extreme risks
to the attacker - worse than the risks of waiting in the face of this danger.
When not coupled with the ability to strike in retaliation, such a capability
might suggest - erroneously, to be sure, in the case of the democracies - an
intention to strike first. If so, it would tend to provoke rather than to deter
general war.
I have dealt here with only one of the functions of overseas bases: their use
as a support for the strategic deterrent force. They have a variety of impor-
tant military, political and economic rhles which are beyond the scope of this
paper. Expenditures in connection with the construction or operation of our
bases, for example, are a form of economic aid and, moreover, a form that is
rather palatable to the Congress. There are other functions in a central war
where their importance may be very considerable and their usefulness in a
limited war might be substantial.
Indeed nothing said here should suggest that deterrence is in itself an ade-
quate strategy. The complementary requirements of a sufficient military policy
cannot be discussed in detail here. Certainly they include a more serious devel-
opment of power to meet limited aggression, especially with more advanced
conventional weapons than those now available. They also include more ener-
getic provision for active and passive defenses to limit the dimensions of the
catastrophe in case deterrence should fail. For example, an economically fea-
sible shelter program might make the difference between 50,000,000 survivors
and 120,000,000 survivors.
But it would be a fatal mistake to suppose that because strategic deter-
rence is inadequate by itself it can be dispensed with. Deterrence is not dis-
pensable. If the picture of the world I have drawn is rather bleak, it could
none the less be cataclysmically worse. Suppose both the United States and
the Soviet Union had the power to destroy each others' retaliatory forces
and society, given the opportunity to administer the opening blow. The sit-
uation would then be something like the old-fashioned Western gun duel. It
would be extraordinarily risky for one side not to attempt to destroy the
other, or to delay doing so, since it not only can emerge unscathed by strik-
ing first but this is the sole way it can reasonably hope to emerge at all.
Evidently such a situation is extremely unstable. O n the other hand, if it is
clear that the aggressor too will suffer catastrophic damage in the event of
his aggression, he then has strong reason not to attack, even though he can
administer great damage. A protected retaliatory capability has a stabiliz-
ing influence not only in deterring rational attack, but also in offering every
inducement to both powers to reduce the chance of accidental war.
they are serious problems and some sorts of limitation and inspection agree-
ment might diminish them. But if there is to be any prospect of realistic and
useful agreement, we must reject the theory of automatic deterrence. And we
must bear in mind that the more extensive a disarmament agreement is, the
smaller the force that a violator would have to hide in order to achieve com-
plete domination. Most obviously, "the abolition of the weapons necessary in
a general or 'unlimited' war" would offer the most insuperable obstacles to an
inspection plan, since the violator could gain an overwhelming advantage
from the concealment of even a few weapons. The need for a deterrent, in this
connection too, is ineradicable.
VII. Summary
Almost everyone seems concerned with the need to relax tension. However,
relaxation of tension, which everyone thinks is good, is not easily distin-
guished from relaxing one's guard, which almost everyone thinks is bad.
Relaxation, like Miltown, is not an end in itself. Not all danger comes from
tension. To be tense where there is danger is only rational.
What can we say then, in sum, on the balance of terror theory of auto-
matic deterrence? It is a contribution to the rhetoric rather than the logic of
war in the thermonuclear age. The notion that a carefully planned surprise
attack can be checkmated almost effortlessly, that, in short, we may resume
our deep pre-sputnik sleep, is wrong and its nearly universal acceptance is
terribly dangerous. Though deterrence is not enough in itself, it is vital.
There are two principal points.
First, deterring general war in both the early and late 1960s will be hard
at best, and hardest both for ourselves and our allies wherever we use forces
based near the enemy.
Second, even if we can deter general war by a strenuous and continuing
effort, this will by no means be the whole of a military, much less a foreign
policy. Such a policy would not of itself remove the danger of accidental
outbreak or limit the damage in case deterrence failed; nor would it be at
all adequate for crises on the .periphery.
-
A g~nerallyuseful way of concluding a grim argument of this kind
would be to affirm that we have the resources, intelligence and courage to
make the correct decisions. That is, of course, the case. And there is a good
chance that we will d o so. But perhaps, as a small aid toward making such
decisions more likely, we should contemplate the possibility that they may
not be made. They are hard, do involve sacrifice, are affected by great
uncertainties and concern matters in which much is altogether unknown
and much else must be hedged by secrecy; and, above all, they entail a new
image of ourselves in a world of persistent danger. It is by no means certain
that we shall meet the test.
I "r : c The Delicate Balance of Terror 73
Notes
1. 1 w'tnt t o thank C.J. Hitch, M.W. Hoag, W.W. Kautnian. A.W. Marshall, H.S. Rowen
and W.W. Taylor for suggestions in preparatton of t h ~ sarticle.
2. (icorge t Kennan, "A Chance t o Withdraw O u r Troops in Europe," H i l r p r ' s
Mirg~rzrnc,,February 1958, p. 4 1.
3. P.M.S. Blackett, "Atomic Weapons a n d East-West RelLttions" ( N e w York: C:,lrnhridgc
l i n ~ v e r y ~ tPress,
y 19561, p. 32.
4. Joseph Alsop, "The New Balance of Power," Encotinter, M'iy 1958, p. 4. It s h o ~ ~ he ld
added that, since these lines were written, Mr. Alsop's vlews have altered.
5. The Nrtv York Timcs, Septernher 6, 1958, p. 2.
6 . (;enera1 Pierre M . Gallois, "A French General Analyzes Nuclear-Age Strategy," Rinlrtk,
Nov. 1 9 58. p. 19; "Nuclear Aggresston a n d National Suicide," Tlw Reportrr., Sept. 18, 19.58,
p. 2 3 .
7. See footnote 9.
8 . See, for example, "NATO, A Critical Appraisal," hy Gardner Patterson ~ n Edg,~r d
S. F u r n ~ s s ,Jr., Princeton University Conference on NATO, Princeton, June 1957, p. 32:
"Although n o one pretended t o know, the hypothesis that one-third of the striking force of the
United States S t r a t e g ~ cAir C o m m m d w a s in the air a t all times was regarded hy most ,IS re:]-
son.lhle."
9 . Gcneral Gallois argues that, while alllances will offer no guarantee, "a small number of
honlbs and a small number of carrlers suffice tor a threatened power t o protect ~tselfagainst
atomic destruction." (Rhrlitis, op. cit., p. 7 1.) H I Snurner~calillustr~tionsgive the defender some
4 0 0 underground launching sites (ihid., p. 22, and The Reporter, o p crt., p. 2 5 ) and suggest t h ~ t
their elimmatton would require between 5,000 a n d 25,000 missiles - w h ~ c his "more o r less
irnpos\ihle" - and that in any case the aggressor would not survive the fallout from his o w n
w c a p o n s WWhier these ,ire 1,irge numhers of targets from the standpoint of the .lggressor will
depend o n the accurac): yield and reliability of offense weapons as well '1s the r e ~ i s t ~ i n cofe the
defender's shelter.; and a number of other matters not spec~fiedin the argument. Gener,il G,lllois
15 ,iw,ire that the expectatton of surviv.il depends o n distance even in the hallist~cmts\ile age and
that o u r .illies are not so fortunate in this respect. Close-in missiles h ~ v ehetrer bomb y~eldsand
accuracies. Moreover, manned a ~ r c r a f - t with still better y~eldsa n d ,~ccuracies- c,in be ~ ~ s by cd
a n aggressor here since warning of their approach is very short. Suffice it t o sciy that the numer-
ic.11 adv'lntage General C;,lllois cites is greatly exaggerated. Furthermore, lie exaggerates the
destructiveness of the retaliatory blow against the aggressor's c ~ t i e sby the renin,ltirs ot tlie
defender's missile forcr -even assuming the aggressor would rake n o special m e a s ~ ~ r teos protcct
hi, cities. Rut pxticulrlrly tor the aggressor - w h o does not lack warntng - ,I civil defense pro-
gram c,in moderate the darnage d o n r by a poorly organized attack. Finally, the sugge\tion that
the a g g r e w u would not surbive the tall-out from his own weapons 1s simply In rrror. The r a p ~ d -
decay fission products w h ~ c hare the ni,ljor lethal problem in the local~tyof ,I surface burst ,Ire
not a serious difficulty for the aggressor. The a m o u n t of the slowdecay products, s t r o n t i ~ ~ m - 9 0
a n d cesium-1.37, In tlie atmosphere would rise considerably. If nothing were done t o counter it,
thts ~ntglir,for ertaniple, increase by many tlmes the incidence of such rel,lti\ely r'ire diseases '1s
hone cancer and leukemia. However, such a cahmity, implying ,111increase of. s ~ y20,000 , deaths
per ).ear for a nation of 200,000,000, 1s of a n entirely d~fferenrorder from the c.lt,istrophe
~ n b o l w n gtens of ~ n ~ l l ~ oofn deaths,
s w h ~ c hGeneral Gallois c o n t e n ~ p l ~ ~elsewhere.
tes And there
'ire measures t h ~ might t reduce even this effect drast~cally.(See the R A N D C o r p o r a t m i Report
R-322-KC:, Report o n iz Stlrri)~of N o ? ~ - M ~ l ~ tDL,~L'MSC,
l ~ r y J L I I 1.
~ 1958.)
10. Aerial reconnaissance, of course, could have a n rndirrrt util~tyhere for surveying large
drect\ t o determine the n ~ ~ m h and e r location of observation posts needed t o p r o v ~ d emore
timely warning.
1 I. J-lmes F. K ~ n g Jr.,
, "Arms and M a n rn the Nuclear-Rocket Era," The Nclc' Keplihl~r,
Seprernhe~-1, 19.58.
5
The Stability of a Bipolar World
Kenneth N.Waltz
T here is a conventional wisdom, accumulated over the centuries, upon

which statesmen and students often draw as they face problems in
international politics. One part of the conventional wisdom is now
often forgotten. Many in Europe, and some in America, have come to
regard an alliance as unsatisfactory if the members of it are grossly unequal
in power. "Real partnership," one hears said in a variety of ways, "is pos-
sible only between equals."' If this is true, an addendum should read: Only
unreal partnerships among states have lasted beyond the moment of press-
ing danger. Where states in association have been near equals, some have
voluntarily abdicated the leadership to others, or the alliance has become
paralyzed by stalemate and indecision, or it has simply dissolved. One may
observe that those who are less than equal are often dissatisfied without
thereby concluding that equality in all things is good. As Machiavelli and
Bismarck well knew, an alliance requires an alliance leader; and leadership
can be most easily maintained where the leader is superior in power. Some
may think of these two exemplars as unworthy; even so, where the un-
worthy were wise, their wisdom should be revived.
A second theorem of the conventional wisdom is still widely accepted. It
reads: A world of many powers is more stable than a bipolar world, with
stability measured by the peacefulness of adjustment within the international
system and by the durability of the system itself. While the first element
of the conventional wisdom might well be revived, the second should be
radically revised.
Pessimism about the possibility of achieving stability in a two-power
world was reinforced after the war by contemplation of the character of the
two major contenders. The Soviet Union, led by a possibly psychotic Stalin,
and the United States, flaccid, isolationist by tradition, and untutored in the
ways of international relations, might well have been thought unsuited to
the task of finding a route to survival. H o w could either reconcile itself to
coexistence when ideological differences were great and antithetical inter-
ests provided constant occasion for conflict? Yet the bipolar world of the
Source: D ~ d a l u s 93(3)
, (1964): 881-909
l\
dl t : The Stability of a Bipolar World 75
postwar period has shown a remarkable stability. Measuring time from the
termination of war, 1964 corresponds to 1937. Despite all of the changes
in the nineteen years since 1945 that might have shaken the world into
another great war, 1964 somehow looks and feels safer than 1937. Is this
true only because we now know that 1937 preceded the holocaust by just
two years? O r is it the terror of nuclear weapons that has kept the world
from major war? O r is the stability of the postwar world intimately related
to its bipolar pattern?
Stability within a Bipolar System
Within a bipolar world, four factors conjoined encourage the limitation of

violence in the relations of states. First, with only two world powers there
are n o peripheries. The United States is the obsessing danger for the Soviet
Union, and the Soviet Union for us, since each can damage the other to an
extent that no other state can match. Any event in the world that involves
the fortunes of the Soviet Union or the united States automatically elicits
the interest of the other. Truman, at the time of the Korean invasion, could
not very well echo Chamberlain's words in the Czechoslovakian crisis and
claim that the Koreans were a people far away in the east of Asia of whom
Americans knew nothing. We had to know about them or quickly find out.
In the 1930's, France lay between England and Germany. England could
believe, and we could too, that their frontier and ours lay on the Rhine.
After World War 11, no third power could lie between the United States and
the Soviet Union, for none existed. The statement that peace is indivisible
was controversial, indeed untrue, when it was made by Litvinov in the
1930's. It became a truism in the 1950's. Any possibility of maintaining a
general peace required a willingness to fight small wars. With the competi-
tion both serious and intense, a loss to one could easily appear as a gain to
the other, a conclusion that follows from the very condition of a two-power
competition. Political action has corresponded to this assumption.
Communist guerrillas operating in Greece prompted the Truman doctrine.
The tightening o f Soviet control over the states of Eastern Europe led to the
Marshall Plan and the Atlantic Defense Treaty, and these in turn gave rise
to the Cominform and the Warsaw Pact. The plan to form a West ('~ e r m a n
government produced the Berlin blockade. Our response in a two-power
world was geared to Soviet action, and theirs to ours, which produced a n
increasingly solid bipolar balance.
Not only are there no peripheries in a bipolar world but also, as a sec-
ond consideration, the range of factors included in the competition is
extended as the intensity of the competition increases. Increased intensity is
expressed in a reluctance to accept small territorial losses, as in Korea, the
Formosa Strait, and Indo-China. Extension of range is apparent wherever
one looks. Vice President Nixon hailed the Supreme Court's desegregation
decision as our greatest victory in the cold war. When it became increasingly
clear that the Soviet economy was growing at a rate that far exceeded our
own, many began to worry that falling behind in the economic race would
lead t o our losing the cold war without a shot being fired. Disarmament
negotiations have most often been taken as an opportunity for propaganda.
As contrasted with the 1930's, there is now constant and effective concern
lest military preparation fall below the level necessitated by the military
efforts of the major antagonist. Changes between the wars affected differ-
ent states differently, with adjustment to the varying ambitions and abilities
of states dependent on cumbrous mechanisms of compensation and realign-
ment. In a multipower balance, who is a danger to whom is often a most
obscure matter: the incentive t o regard all disequilibrating changes with
concern and respond to them with whatever effort may be required is con-
sequently weakened. In our present world changes may affect each of the
two powers differently, and this means all the more that few changes in the
national realm or in the world at large are likely to be thought irrelevant.
Policy proceeds by imitation, with occasional attempts to outflank.
The third distinguishing factor in the bipolar balance, as we have thus
far known it, is the nearly constant presence of pressure and the recurrence
of crises. It would be folly to assert that repeated threats and recurring
crises necessarily decrease danger and promote stability. It may be equally
wrong to assert the opposite, as Khrushchev seems to appreciate. "They
frighten us with war," he told the Bulgarians in May of 1962, "and we
frighten them back bit by bit. They threaten us with nuclear arms and we
tell them: 'Listen, now only fools can d o this, because we have them too,
and they are not smaller than yours but, we think, even better than yours.
So why do you d o foolish things and frighten us?' This is the situation, and
this is why we consider the situation to be good."2 Crises, born of a condi-
tion in which interests and ambitions conflict, are produced by the deter-
mination of one state t o effect a change that another state chooses to resist.
With the Berlin blockade, for example, as with Russia's emplacement of
missiles in Cuba, the United States decided that to resist the change the
Soviet Union sought to bring about was worth the cost of turning its action
into a crisis. If the condition of conflict remains, the absence of crises
becomes more disturbing than their recurrence. Rather a large crisis now
than a small war later is an axiom that should precede the statement, often
made, that to fight small wars in the present may be the means of avoiding
large wars later.
Admittedly, crises also occur in a multipower world, but the dangers are
diffused, responsibilities unclear, and definition of vital interests easily
obscured. The skillful foreign policy, where many states are in balance, is
designed to gain an advantage over one state without antagonizing others
and frightening them into united action. Often in modern Europe, possible
gains have seemed greater than likely losses. Statesmen could thus hope in
crises to push an issue to the limit without causing all the potential opponents
to unite. When possible enemies are several in number, unity of action among
states is difficult to secure. One could therefore think - or hope desperately,
The Stability of a Bipolar World 77
as did Bethmann Hollweg and Adolph Hitler - that no united opposition

would form.
In a bipolar world, on the other hand, attention is focused on crises by
both of the major competitors, and especially by the defensive state. To
move piecemeal and reap gains serially is difficult, for within a world in
confusion there is one great certainty, namely, the knowledge of who will
oppose whom. One's motto may still be, "push to the limit," but limit must
he emphasized as heavily as push. Caution, moderation, and the management
of crisis come to be of great and obvious importance.
Many argue, nonetheless, that caution in crises, and resulting bipolar
stability, is accounted for by the existence of nuclear weapons, with the
number of states involved comparatively inconsequent. That this is a doubt-
ful deduction can be indicated by a consideration of how nuclear weapons
may affect reactions to crises. In the postwar world, bipolarity preceded the
construction of two opposing atomic weapons systems. The United States,
with some success, substituted technological superiority for expenditure on
a conventional military. system
. as a deterrent to the Soviet Union during the
years when we had first an atomic monopoly and then a decisive edge in
quantity and quality of weapons. American military policy was not a mat-
ter of necessity but of preference based on a calculation of advantage. Some
increase in expenditure and a different allocation of monies would have
enabled the United States to deter the Soviet Union by posing credibly the
threat that any Soviet attempt, say, to overwhelm West Germany would
bring the United States into a large-scale conventional war [I].For the Soviet
Union, war against separate European states would have promised large
gains; given the bipolar balance, n o such war could be undertaken without
the clear prospect of American entry. The Russians' appreciation of the situ-
ation is perhaps best illustrated by the structure of their military forces. The
Soviet Union has concentrated heavily on medium-range bombers and
missiles and, to our surprise, has built relatively few intercontinental
weapons. The country of possibly aggressive intent has assumed a posture
of passive deterrence vis-&vis her major adversary, whom she quite sens-
ibly does not want to fight. Against European and other lesser states, the
Soviet Union has a considerable offensive capability [2]. Hence nuclear cap-
abilities merely reinforce a condition that would exist in their absence:
without nuclear technology both the United States and the Soviet Union
have the ability to develop weapons of considerable destructive power. Even
had the atom never been split, each would lose heavily if it were to engage
in a major war against the other.
If number of states is less important than the existence of nuclear power,
then one must ask whether the world balance would continue to be stable
were three or more states able to raise themselves to comparable levels of
nuclear potency. For many reasons one doubts that the equilibrium would be
so secure. Worries about accidents and triggering are widespread, but a still
greater danger might well arise. The existence of a number of nuclear states
would increase the temptation for the more virile of them to maneuver, with
defensive states paralyzed by the possession of military forces the use of

which would mean their own destruction. One would be back in the 1930's,
with the addition of a new dimension of strength which would increase the
pressures upon status quo powers to make piecemeal concessions.
Because bipolarity preceded a two-power nuclear competition, because
in the absence of nuclear weapons destructive power would still be great,
because the existence of a number of nuclear states would increase the
range of difficult political choices, and finally, as will be discussed below,
because nuclear weapons must first be seen as a product of great national
capabilities- rather than as their cause, one is led to the conclusion that
nuclear weapons cannot by themselves be used to explain the stability - or
the instability - of international systems.
Taken together, these three factors - the absence of peripheries, the
range and intensity of competition, and the persistence of pressure and cri-
sis -are among the most important characteristics of the period since World
War 11. The first three points combine to produce an intense competition in
a wide arena with a great variety of means employed. The constancy of
effort of the two major contenders, combined with a fourth factor, their
preponderant power, have made for a remarkable ability to comprehend
and absorb within the bipolar balance the revolutionary political, military,
and economic changes that have occurred. The Soviet Union moved for-
ward and was checked. Empires dissolved, and numerous new states
appeared in the world. Strategic nuclear weapons systems came into the
possession of four separate countries. Tactical nuclear weapons were de-
veloped and to some extent dispersed. The manned bomber gave way to the
missile. Vulnerable missiles were hardened, made mobile, and hidden. A
revolution in military technology occurred on an average of once every five
years and at an accelerating pace."wo "losses" of China, each a qualified
loss but both traumatic, were accommodated without disastrously distort-
ing - or even greatly affecting - the balance between America and Russia.
The effects of American-Soviet preponderance are complex. Its likely
continuation and even its present existence are subjects of controversy. The
stability of a system has to be defined in terms of its durability, as well as
of the peacefulness of adjustment within it. In the pages that follow, some
of the effects of preponderance will be indicated while the durability of the
system is examined.
T h e End of t h e B i p o l a r Era?
In a bipolar world, by definition each of two states or two blocs over-

shadows all others. It may seem that to write in 1964 of bipolarity is merely
to express nostalgia for an era already ending. Richard Rosecrance, referring
t o the period since the war, describes the world as " t r i p ~ l a r . " Walter
~
Lippmann, in a number of columns written in late 1963 and early 1964,
assesses the recent initiatives of France and Communist China, their ability
The Stability of a Bipolar World 79
to move contrary to the desires of the United States and the Soviet Union,
as marking the end of the postwar world in which the two superpowers
closely controlled the actions of even their major associates [3].Hedley Bull,
in a paper prepared for the Council on Foreign Relations in the fall of 1963,
tentatively reaches the conclusion that between now and 197.5 "the system
of polarization of power will cease to be recognizable: that other states will
count for so much in world politics that the two present great powers will
find it difficult, even when cooperating, to dominate them."'
If power is identical with the ability to control, then those who are free
are also strong; and the freedom of the weak would have to be taken as an
indication of the weakness of those who have great material strength. But
the weak and disorganized are often less amenable to control than those
who are wealthy and well disciplined 141. The powerful, out of their strength,
influence and limit each other; the wealthy are hobbled by what they have
to lose. The weak, on the other hand, bedevil the strong; the poor can more
easily ignore their own interests. Such patterns endure and pervade the rela-
tions of men and of groups. Lnited States Steel enjoys less freedom to vary
the price of its products than d o smaller producers. The United States gov-
ernment finds it easier to persuade large corporations and the great labor
unions to cooperate in an anti-inflationary policy than to secure the com-
pliance of small firms and independent unions. The political party in oppos-
ition is freer to speak irresponsibly than is the government. Power corrupts
and renders its possessors responsible; the possession of wealth liberates
and enslaves. That similar patterns are displayed in international relations
is hardly surprising. It is not unusual to find that minor states have a con-
siderable nuisance value in relation to states greatly their superiors in power.
A Chiang Kai-shek, a Syngman Rhee, or a Mossadegh is often more diffi-
cult to deal with than rulers of states more nearly one's equal in power.
The influence and control of the two great powers has stopped short of
domination in most places throughout the postwar period. The power of
the United States and of the Soviet Union has been predominant but not
absolute. To describe the world as bipolar does not mean that either power
can exert a positive control everywhere in the world, but that each has
global interests which it can care for unaided, though help may often be
desirable. To say that bipolarity has, until recently, meant more than this is
to misinterpret the history of the postwar world. Secretary Dulles, in the
middle 1950's, inveighed against neutralism and described it as immoral. His
judgment corresponded to a conviction frequently expressed in Communist
statements. P. E. Vyshinsky, in a 1948 issue of Problems of Philosophy,
declared that "the only determining criterion of revolutionary proletarian
internationalism is: are you for or against the U.S.S.R., the motherland of
the world proletariat? . . . The defense of the U.S.S.R., as of the socialist
motherland of the world proletariat, is the holy duty of every honest man
everywhere and not only of the citizens of the U.S.S.R."' The rejection of
neutralism as an honorable position for other countries to take is another
example of intensity of competition leading to an extension of its range.
By coming to terms with neutralism, as both the United States and the
Soviet Union have done, the superpowers have shown even their inability
to extend their wills without limit.
Bearing in mind the above considerations, can we say whether the recent
independent action of France and Communist China does in fact indicate
the waning of bipolarity, or does it mean merely the loosening of bipolar
blocs, with a bipolar relation between the United States and the Soviet
Union continuing to dominate? By the assessment of those who themselves
value increased independence, the latter would seem to be the case. The Earl
of Home, when he was Secretary of State for Foreign Affairs, thought he
saw developing from the increased power of the Soviet Union and the
United States a nuclear stalemate that would provide for the middle states
a greater opportunity to maneuver.' De Gaulle, in a press conference
famous for other reasons, included the statement that uncertainty about
their use "does not in the least prevent the American nuclear weapons,
which are the most powerful of all, from remaining the essential guarantee
of world p e a ~ e . "Communist
~ China's calculation of international political
and military forces may be highly similar. "Whatever happens," Chou En-lai
has said recently, "the fraternal Chinese and Soviet peoples will stand
together in any storm that breaks out in the world arena."' Ideological dis-
putes between China and Russia are bitter; their policies conflict. But inter-
ests are more durable than the alliances in which they sometimes find
expression. Even though the bonds of alliance are broken, the interest of the
Soviet Union could not easily accommodate the destruction of China if that
were to mean that Western power would be poised on the Siberian border.
That strategic stability produces or at least permits tactical instability is
now a cliche of military analysis. The axiom, transferred to the political
realm, remains true. Lesser states have often found their opportunity to
exist in the interstices of the balance of power. The French and Chinese, in
acting contrary to the wishes of their principal partners, have certainly
caused them some pain. Diplomatic flurries have resulted and some changes
have been produced, yet in a more important respect, France and China
have demonstrated not their power but their impotence: their inability to
affect the dominant relation in the world. The solidity of the bipolar bal-
ance permits middle states to act with impunity precisely because they
know that their divergent actions will not measurably affect the strength of
the Soviet Union or the United States, upon which their own security con-
tinues to rest. The decisions of Britain, France, and China to build nuclear
establishments are further advertisements of weakness. Because American
or Soviet military might provides adequate protection, the middle powers
need not participate in a military division of labor in a way that would con-
tribute maximally to the military strength of their major associates.
The United States is inclined to exaggerate the amount of strength it can
gain from maintaining a system of united alliances as opposed to bilateral
arrangements. The exaggeration arises apparently from vague notions about
the transferability of strength. Actually, as one should expect, the contribution
1 ,ili The Stability of a Bipolar World 81
o f each ally is notable only where it believes that its interests require it to
make an effort. In resisting the invasion of North Korean and, later, Chinese
troops, roughly 90 percent of the non-Korean forces were provided by the
United States."' In South Vietnam at the present time the United States is the
only foreign country engaged. British and French military units in West
Germany, under strength and ill equipped, are of little use. Western Europe
remains, to use the terminology of the 19303, a direct consumer of security.
The only really significant interest of the United States, as is nicely conveyed
by Arnold Wolfers' dubbing us "the hub power,"" is that each country that
may be threatened by Soviet encroachment be politically stable and thus able
to resist subversion, be self-dependent and thus less of an expense to us, and
be able at the outset of a possible military action to put up some kind of a
defense.
On these points, the American interest in Western Europe is precisely the
same as its interest in the economically underdeveloped countries. In the
case of the European countries, however, losses are harder to sustain and
there are advantages clearly to be gained by the United States where our
interests and theirs overlap. It would be difficult to argue that the foreign-aid
programs undertaken by Britain, France, and West Germany transcend a
national purpose or have been enlarged in response to our insisting upon
their duty to share the military and economic responsibilities that the
United States has assumed. The protection of persons, property, and the
pound sterling required Britain to resist Communist guerrillas in Malaya,
which was after all still her dependency. In such a case, the bearing of a
heavy burden by another country serves its interests and ours simultan-
eously. If anything, the possibility of a transfer of strength has decreased in
the past fifteen years, along with a decline in usable military power in
Britain. Britain had in her army 633,242 men in 1948; by 1962 she had
209,500, with further reductions anticipated. The comparable figures for
France are 465,000 and 706,000." France, with a system of conscription for
a comparatively long term and at relatively low pay, has maintained mili-
tary forces impressively large when measured as a percentage of her POPLI-
lation 151. As France takes the first steps along the route followed by England,
her military planning runs parallel to the earlier English calculations; she
will seek to cope with the pressures of large money requirements by mak-
ing similar adjustments. According to present French plans, the total of men
under arms is to be reduced by 4 0 per cent [6].
To compensate for the loss of influence that once came from making a
military contribution outside their own borders, the one country has tried
and the other is now attempting to build nuclear establishments that sup-
posedly promise them some measure of independence. The British effort
remains dependent on American assistance, and the French effort to build
an effective nuclear weapons system is in its infancy. The independence of
recent French policy cannot have been grounded on a nuclear force that
barely exists. It is, rather, a product of intelligence and political will exer-
cised by President de Gaulle in a world in which bipolar stalemate provides
the weak some opportunity to act. Independence of action by France and

by the People's Republic of China is at once a product of loosening alliances -
the lesser dependence of principals upon their associates - and a protest
against it.
In the wake of the war, the countries of Western Europe derived a con-
siderable influence from their weakness and our inability to let them succumb
to internal difficulties or external pressures without thereby disadvantaging
ourselves in relation to Russia. We were less free then because they were so
dependent upon our support. The Soviet challenge made it important to
recreate strength in Western Europe, a purpose that could best be achieved
cooperatively. From about 1960 onward, the dependence of each of the
nuclear giants upon its associates lessened. The earlier postwar pattern was
one of interdependence with consequent influence for junior partners. More
recently a lesser interdependence has permitted and produced assertions of
independence, which must be understood in part as efforts to recapture
influence once enjoyed.
The Durability of the Bipolar World
Bipolarity as a descriptive term remains appropriate as long as there is a

great gap between the power of the two leading countries and the power of
the next most considerable states. When one looks in this light at
Communist China, he is likely to be mesmerized by the magic of numbers.
Surely 750 million Chinese must enable their Communist government to d o
some things very damaging to the United States or the Soviet Union, or to
both of them. When one considers the West European states, he may be
struck by their rapid movement from economic and military dependence
upon the United States to positions of some independence. It is natural to
ask whether this is part of a trend that will continue, or simply a movement
from nearly zero on the scale of independence to a threshold that can hardly
be passed. It is easy t o think that the trend will continue until, again in the
words of Hedley Bull, "over the next decade the Soviet Union and the
United States will find themselves still the principal powers in opposed sys-
tems of alliances, but, like Britain and Germany 1907-1914, aware that
their allies are not irrevocably committed t o their cause and able to cooper-
ate themselves against their lesser allies on particular issues."'Qut this is
an analogy that can mislead. The allies of Britain and of Germany were of
an order of power, as measured by a combination of territory, population,
and economic capability, similar t o that of their principals. That many
important changes have occurred in the past fifteen years is obvious. That
the changes that have occurred and others that are likely will lift any pres-
ent state t o the level of Soviet or American capabilities is all but impossible.
In 1962, the gross national product of the Soviet Union was $260 bil-
lion, of the United States $555 billion, of West Germany $84 billion, and of
Communist China roughly $50 billion. If one projects from these figures,
i
:, r i l . , T h e Stability of a Bipolar World 83
the following picture emerges: the Soviet Union, at an assumed growth rate
o f 5 per cent, will have in the year 2004 a gross national product of $2,080
billion; the United States, at a growth rate of 3 per cent, will have by 2000
a gross national product of $2,220 billion; West Germany, if it grows at a
sustained rate of 6 per cent yearly, will have by 1998 a gross national prod-
uct of $672 billion, and Communist China, projected at 7 per cent, will
have a gross national product in 2002 of $800 billion 171. The growth rates
assume2 are unlikely to be those that actually prevail. The rat& chosen are
those that will narrow the gap between the greatest and the middle powers
to the largest extent presently imaginable. Even on these bases, it becomes
clear that the Soviet Union and the United States to the end of the millennium
will remain the preponderant powers in the world unless two or more of the
middle powers combine in a way that gives them the ability to concert their
political and military actions on a sustained and reliable basis.
The gap that exists can be described in other ways which are more frag-
mentary but perhaps give a still sharper picture. The United States has been
spending on its military establishment yearly an amount that is two-thirds
or more of the entire West German or British or French gross national prod-
uct. In 1962, the Europe of the Six plus Great Britain spent on defense less
than a quarter of the military expenditure of the United S t a t e s . ' T h e United
States spends more on military research and development than any of the
three largest of the West European countries spends on its entire military
establishment.
The country that would develop its own resources, military and other, in
order to play an independent role in the world, faces a dreadful problem. It
is understandably tempting for such countries to believe that by developing
nuclear weapons systems, they can noticeably close the gap between them-
selves and the superpowers. The assumption that nuclear weapons will
serve as the great equalizers appeared early and shows an impressive per-
sistence. "The small country," Jacob Viner wrote in 1946, "will again be
more than a cipher or a mere pawn in power-politics, provided it is big
enough to produce atomic bombs."" Stanley Hoffmann, writing in the
present year, reflects a similar thought in the following words: "True, the
French nuclear program is expensive; but it is also true that conventional
rearmament is not cheaper, and that a division of labor that would leave all
nuclear weapons in United States hands and specialize Europe in conven-
tional forces would earmark Europe for permanent dependence (both mili-
tary and political) in the cold war and permanent decline in the
international competition."'"
It is difficult to know just what is meant by saying that "conventional
rearmament is not cheaper" than a nuclear program, but it is clear that
nuclear programs are very expensive indeed [8 1. France and Britain now spend
about 7 per cent of their gross national products on defense. If this were
increased to the American level of approximately 1 0 per cent, or even if it
were doubled, the defense spending of each country would remain com-
paratively small. The inability to spend large sums, taken together with the
costs of research, development, production, and maintenance, leads one to

the conclusion that the French government is betting that Kahn's revolution
in military technology every five years will no longer take place. The French
might then hope that Polaris submarines, with their missiles, would remain
invulnerable. It is doubtful that they are truly invulnerable even now.
The point is a complicated one. By confusing the tracking mechanism of
a hunter-killer submarine, an easy accomplishment, one submarine can
escape from another. A Soviet submarine, however, may be able to meet and
quietly destroy a French submarine as it comes out of port. It is unlikely
that the French would in such an event say anything at all; surely they
would not wish to draw attention t o the loss of what might be one-third of
their strategic nuclear system [9]. To prevent this, France could choose to
operate her submarine fleet entirely from the Mediterranean, a sea from
which the Soviet Union is militarily excluded. But limiting the direction
from which missiles may come will make it easier for the Soviet Union
to defend against them. Khrushchev's claim that the Soviet Union's rockets
can hit a fly in the sky, which strikes Americans as an irrelevant boast,
has an important implication for the country that would build a small
nuclear force." Missile defenses, almost useless against large numbers, may
be highly successful against the approach of only a few missiles. Furthermore,
a single command and control system can easily be obliterated. Middle
powers will have to concentrate on a single system or a very small number
of systems, and thus deny to themselves the invulnerability gained by the
United States from dispersion of the weapons of any one system and the
existence of multiple systems. Were military innovation to cease, a force
such as that projected by France could gradually be built up to a level of
military significance. If, however, a future French Polaris force should begin
to look dangerous to the Soviet Union, the increased French capability
would itself become an incentive for Russia to move faster. And if Russia
does, so must we too. Far short of America or Russia using nuclear weapons
for the surgical excision of any country's embryonic nuclear capability, the
opportunity t o develop a nuclear force to a level of usefulness exists, if it is
present at all, only on sufferance of the two nuclear giants.
To look upon nuclear weapons systems as the great equalizers is to see
them as causes of the increased power of states. It is more accurate and
more useful to look upon them as the products of great scientific and eco-
nomic capability. The railway age brought a great increase in military
mobility, which the elder von Moltke brilliantly exploited in the wars for
German unification. So long, however, as war power took the form of great
masses of men and material, railways were not able to deliver the whole
force of a nation to a front or concentrate it upon a point. Even in trans-
porting a portion of a country's military power, railways were not able to
cross the front. Thus in 1914, German armies marched through Belgium.'8
In World War 11, the wedding of high explosives and air transport still did
not make it possible to aggregate a nation's whole power and deliver it sud-
denly and decisively to designated military targets. World War I1 was won
i,ilr, The Stability of a Bipolar World 85
slowly and largely on the ground. Nuclear technology produced a change

decisive in one respect. The power of a nation can now be distilled. Like the
French chef who boils down a -pig- for three days until he has a pint of li-
quid that represents the very essence of pig, the country that produces
nuclear warheads and the requisite delivery systems is distilling the power
of a whole nation. But the power has to be there before it can be distilled.
The stills of such countries as Britain, France, and Communist China are
simply not large enough.
Nuclear weapons systems are not the great equalizers, but they are,
rather, in all of their complexity and with all of their tremendous cost, out-
ward signs of the Soviet and American ability to outstrip all others. If other
countries should nevertheless be able to build nuclear systems capable of
doing great damage on second strike to any attacker, they would then, as the
Soviet Union and United States now do, participate in a nuclear stand-off.
Con~petitionwould shift to other means, which t o some extent has already
happened, and traditional criteria of power, including economic and mili-
tary capability, would once again take on their old significance.
This is not to say that nuclear diffusion makes no difference. It is useful
to consider briefly some of the possibilities.
(1)A threat by Britain, France, or Communist China to use nuclear force
in order to deter a conventional attack launched by a great nuclear power is a
threat to do limited damage to the invading state at the risk of one's own
annihilation. It is a radically different way of assuming the deterrent-defensive
posture of Switzerland and should be interpreted as a move to bow out of the
great-power picture. In part the desire for an independent nuclear deterrent
derives, as the late Hugh Gaitskell put it, "from doubts about the readiness
of the United States Government and the American citizens to risk the
destruction of their cities on behalf of Europe."" The nuclear superiority
enjoyed by America in the early 1950's created in Europe a fear that the
United States would too easily s u c c ~ ~ mtoba temptation to retaliate massively.
The arrival of strategic stability has produced the opposite worry In the
words of a senior British general: "McNamara is practically telling the
Soviets that the worst they need expect from an attack on West Germany is
a conventional counterattack."1° Behind the difference on strategy lies a
divergence of interest. A policy of strategic nuclear threat makes the United
States the primary target. A policy of controlled response would shift some of
the danger as well as additional burdens to Europe. The countries of Europe,
separate or united, have an incentive to adopt destabilizing military pro-
grams. Where Britain has led, France now follows. While it is understandable
that lesser powers should, by threatening or using nuclear weapons, want to
he able to decide when the United States or the Soviet Union should risk their
own destruction, it is also easy to see that both the United States and the
Soviet Union would resist such an outcome. The more badly a country may
want to be able to trigger someone else's force, the more difficult it becomes
to do so, which is another way of saying that the Soviet Union and the United
States have something close to invulnerable second-strike systems.
(2) If a middle power were engaged in a conventional military action

against a state of comparable or lesser size, the Soviet Union or the United
States might threaten a nuclear strike in order to bring about a withdrawal.
It is sometimes thought that the possession of a small nuclear force by the
middle power would make such a threat ineffective. In the Suez adventure,
for example, military action by Britain and France called forth Soviet rocket
threats against them. Against states having no strategic nuclear forces, such
threats would be more readily credited, and thus more likely t o exert pres-
sure successfully against the conventional action itself. A small military
action, however, is not worth and does not require nuclear interference by
a great power, for it can be stopped in other ways. The onus of threatening
t o use nuclear weapons first, in order t o interdict conventional interference,
is then placed upon the smaller power. Such a threat would not be credible.
Both the first and second uses presuppose the adequacy of the small
country's nuclear threat when directed against the United States or the
Soviet Union. A capability that is small compared t o America's or Russia's
may be adequate to its task; a certain minimum, doubtfully achievable in
the foreseeable future, is nevertheless required. When Hedley Bull says that
the French ambition is "to become strong enough t o choose deliberately to
act alone,"21 he may have in mind the second use mentioned above, or the
one following, which is seldom discussed.
( 3 ) As the United States and the Soviet Union have opened up a gap in
military power between themselves and all others, so Britain, France, the
People's Republic of China, and states who may follow them can differen-
tiate themselves from non-nuclear nations. Great Britain has placed nuclear
weapons in the Middle and Far East. Let us suppose Indonesia were to
move militarily against Malaysia. A British threat t o use nuclear weapons
could conceivably follow, which might cause Indonesia to stop short or
might persuade the United States to offer the support of the Seventh Fleet
and American Marines in order to avoid the use of nuclear weapons.
The effects of nuclear diffusion are necessarily uncertain, but one point
can sensibly be made: Building a small nuclear force is an unpromising way
of seeking to maintain the integrity of one's state, even though it may enable
that state to act positively against equal or lesser powers.
There can be approximate equality among states even where there is
considerable disparity in the material bases of their power. Whether or not
effective power is fashioned from the material available depends upon ad-
equacy of national organization, wisdom of policy, and intensity of effort.
In the 1920's, France sought t o maintain a greater military strength than
Germany in order to compensate for a lesser French productivity and smaller
population. Where the material differences are relatively small or where
countries of immensely larger capacity are quiescent, it may be possible to
"mobilize" a nation in peacetime in order to build on a lesser material base
a superior military strength. Germany and Japan in the 1930's began to
play the game from which France was withdrawing. The Soviet Union,
since the war, has been able to challenge the United States in many parts of
iidlti The Stability of a Bipolar World 87
the world by spending a disproportionately large share of her smaller

income on military means. There is in the West a quiet nightmare that the
People's Republic of China may follow such a path, that it may mobilize the
nation in order to increase production rapidly while simultaneously acquiring
a large and modern military capability. It is doubtful that she can do either,
and surely not both, and surely not the second without the first, as the data
previously given clearly indicate. As for France and Great Britain, it strains
the imagination to the breaking point to believe that in a world in which
scientific and technological progress has been rapid, either of them will
be able to maintain the pace [ l o ] . Unable to spend on anywhere near the
American or Russian level for work in research, development, and production,
middle powers will, once they have gained an initial advantage, constantly
find themselves falling behind. France and Britain are in the second-ranking
powers' customary position of imitating, with a time lag, the more advanced
weapons systems of their wealthier competitors [ 1I].
From the above analysis, it is clear that the time when other states can
compete at the highest levels of power by a superiority of effort in mobiliz-
ing their resources lies far in the future. Unless some states combine or oth-
ers dissolve in chaos, the world will remain bipolar until the end o f the
present century.
S o m e Dissenting Opinions
The fact remains that many students of international relations have con-
tinued to judge bipolarity unstable as compared to the probable stability o f
a multipower world. Why have they been so confident that the existence of
a number of powers, moving in response to constantly recurring variations
in national power and purpose, would promote the desired stability?
According to Professors Morgenthau and Kaplan, the uncertainty that
results from flexibility of alignment generates a healthy caution in the foreign
policy of every country." Concomitantly, Professor Morgenthau believes
that in the present bipolar world, "the flexibility of the balance of power
and, with it, its restraining influence upon the power aspirations of the main
protagonists on the international scene have disappeared."13 One may agree
with his conclusion and yet draw from his analysis another one unstated by
him: The inflexibility of a bipolar world, with the appetite for power of
each major competitor at once whetted and checked by the other, may pro-
mote a greater stability than flexible balances of power among a larger
number of states.
What are the grounds for coming to a diametrically different conclusion?
The presumed double instability of a bipolar world, that it easily erodes or
explodes, is to a great extent based upon its assumed bloc character. A bloc
improperly managed may indeed fall apart. The leader of each bloc must be
concerned at once with alliance management, for the defection of an allied
state might be fatal to its partners, and with the aims and capabilities of the
opposing bloc. The system is more complex than is a multipower balance,

which in part accounts for its fragility 1121. The situation preceding World War
I provides a striking example. The dissolution of the Austro-Hungarian
Empire would have left Germany alone in the center of Europe. The approx-
imate equality of alliance partners, or their relation of true interdependence,
plus the closeness of competition between the two camps, meant that while
any country could commit its associates, no one country on either side could
exercise control. By contrast, in 1956 the United States could dissociate itself
from the Suez adventure of its two principal allies and even subject them to
pressure. Great Britain, like Austria in 1914, tried to commit, or at least
immobilize, its alliance partner by presenting him with a fait accompli.
Enjoying a position of predominance, the United States could, as Germany
could not, focus its attention on the major adversary while disciplining its
ally, The situations are in other respects different, but the ability of the United
States, in contrast to Germany, to pay a price measured in intra-alliance terms
is striking.
It is important, then, to distinguish sharply a bipolarity of blocs from a
bipolarity of countries. FCnelon thought that of all conditions of balance the
opposition of two states was the happiest. Morgenthau dismisses this judg-
ment with the comment that the benefits Fenelon had hoped for had not
accrued in our world since the war, which depends, one might think, on
what benefits had otherwise been expected 1131.
The conclusion that a multipower balance is relatively stable is reached
by overestimating the system's flexibility, and then dwelling too fondly
upon its effects 1141. A constant shuffling of alliances would be as dangerous
as an unwillingness to make new combinations. Neither too slow nor too
fast: the point is a fine one, made finer still by observing that the rules should
be followed not merely out of an immediate interest of the state but also for
the sake of preserving the international system. The old balance-of-power
system here looks suspiciously like the new collective-security system of
the League of Nations and the United Nations. Either system depends for its
maintenance and functioning upon a "neutrality of alignment" at the moment
of serious threat. To preserve the system, the powerful states must overcome
the constraints of previous ties and the pressures o f both ideological pref-
erences and conflicting present interests in order t o confront the state that
threatens the system.24
In the history of the modern state system, flexibility of alignment has
been conspicuously absent just when, in the interest of stability, it was most
highly d e ~ i r a b l e . ~A' comparison of flexibility within a multipower world
with the ability of the two present superpowers to compensate for changes
by their internal efforts is requisite, for comparison changes the balance of
optimism and pessimism as customarily applied t o the two different sys-
tems. In the world of the 19303, with a European grouping of three, the
Western democracies, out of lassitude, political inhibition, and ideological
distaste, refrained from acting or from combining with others at the advan-
tageous moment. War provided the pressure that forced the world's states
!,\ ~ii, The Stability of a Bipolar World 89
into two opposing coalitions. In peacetime the bipolar world displays a

clarity of relations that is ordinarily found only in war. Raymond Aron
has pointed out that the international "systeme depend de ce que sont,
concretement, les deux p6les, non pas seulement du fait qu'ils sont deux."'"
Modifying Aron's judgment and reversing that of many others, we would say
that in a bipolar world, as compared to one of many powers, the international
system is more likely to dominate. External pressures, if clear and great
enough, force the external combination or the internal effort that interest
requires. The political character of the alliance partner is then most easily
overlooked and the extent to which foreign policy is determined by ideol-
ogy is decreased.
The number of great states in the world has always been so limited that
two acting in concert or, more common historically, one state driving for
hegemony could reasonably conclude that the balance would be altered by
their actions. In the relations of states since the Treaty of Westphalia, there
have never been more than eight great powers, the number that existed, if
one is generous in admitting doubtful members to the club, on the eve of
the First World War. Given paucity of members, states cannot rely on an
equilibrating tendency of the system. Each state must instead look to its
own means, gauge the likelihood of encountering opposition, and estimate
the chances of successful cooperation. The advantages of an international
system with more than two members can at best be small. A careful evalu-
ation of the factors elaborated above indicates that the disadvantages far
outweigh them.
Conclusions That Bear upon Policy
If the preceding explanations are correct, they are also of practical import-
ance. Fixation upon the advantages of flexibility in a multipower balance has
often gone hand in hand with an intense anxiety associated with bipolarity:
the fear that a downward slide or a sudden technological breakthrough by
one great state or the other would decisively alter the balance between them.
Either occurrence could bring catastrophic war, which for the disadvantaged
would be a war of desperation, or world domination from one center with
or without preceding war. The fear is pervasive, and in American writings
most frequently rests on the assumption that, internally dissolute and tired
o f the struggle, we will award the palm to the Soviet Union. Sometimes this
anxiety finds a more sophisticated expression, which turns less upon internal
derangements. In this view, the United States, as the defensive power in the
world, is inherently disadvantaged, for the aggressive power will necessarily
gain if the competition continues long enough. But a conclusion derived from
an incomplete proposition is misleading. One must add that the aggressive
state may lose even though the state seeking to uphold the status quo never
take the offensive. The Soviet Union controls no nation now, except possibly
Cuba, that was not part of its immediate postwar gains. It has lost control in
Yugoslavia and the control it once seemed to have in China. The United
States, since the time it began to behave as a defensive power, has seen some
states slip from commitment to neutralism, but only North Vietnam and
Cuba have come under Communist control. One would prefer no losses at
all, but losses of this magnitude can easily be absorbed. On balance, one
might argue that the United States has gained, though such a judgment
depends on the base line from which measurement is made as well as upon
how gains and losses are defined.
That the United States and the Soviet Union weigh losses and gains
according to their effect upon the bipolar balance is crucial, but there are
many changes in Africa, or Asia, or Latin America that are not likely to be
to the advantage of either the Soviet Union or the United States. This judg-
ment can be spelled out in a number of ways. The doctrine of containment,
for example, should be amended to read: defend, or insulate so that one loss
need not lead to another. The habits of the cold war are so ingrained and
the dangers of a bipolar world so invigorating that the defensive country is
easily led to overreact. In Southeast Asia, since no gain for Communist
China is likely to benefit the Soviet Union, American concern should be
confined to maintaining its reputation and avoiding distant repercussions.
If one goes further and asks how great a gain will accrue to the People's
Republic of China if it extends its territorial control marginally, the answer,
in any of the areas open to it, must be "very little." Neutralization moves
by President de Gaulle, if they can obscure the responsibility for unwanted
events, may in fact be helpful. It is important to realize that the bipolar
world is continuing lest we worry unnecessarily and define the irrelevant
gesture or even the helpful suggestion of lesser powers as troublesome.
A 5 per cent growth rate sustained for three years would add to the
American gross national product an amount greater than the entire gross
national product of Britain or France or West Germany. Even so, the accre-
tion of power the Soviet Union would enjoy by adding, say, West Germany's
capabilities to her own would be immensely important; and one such gain
might easily lead to others. Most gains from outside, however, can add rela-
tively little to the strengths of the Soviet Union or the United States. There
are, then, few single losses that would be crucial, which is a statement that
points to a tension within our argument. Bipolarity encourages each giant
to focus upon crises, while rendering most of them of relative inconse-
quence. We might instead put it this way: Crisis is of concern only where
giving way would lead to an accumulation of losses for one and gains for
the other. In an age characterized by rapidity of change, in many respects
time is slowed down - as is illustrated by the process of "losing" Indo-China
that has gone on for nineteen years without a conclusive result. Since only
a succession of gains could be decisive, there is time for the losing state to
contrive a countering action should it be necessary to do so.
Intensity and breadth of competition and recurrence of crises limn a

picture of constant conflict verging on violence. At the same time, the relative
Walt7 The Stability of a Bipolar World 91
simplicity of relations within a bipolar world, the great pressures that are
generated, and the constant possibility of catastrophe produce a conservatism
on the part of the two greatest powers. The Soviet Union and the United
States may feel more comfortable dealing a deux than in contemplating a
future world in which they vie for existence and possible advantage with
other superpowers. While there is naturally worry about an increase of ten-
sions to intolerable levels, there is also a fear that the tensions themselves will
lead America and Russia to seek agreements designed to bring a relaxation
that will be achieved at the expense of lesser powers. The French general, Paul
Stehlin, commenting on American opposition to Nth-country nuclear forces,
which he interprets as part of an American-Russian effort to maintain a bi-
polar world, asks wistfully: "Does Europe have less political maturity than
the Big Two credit each other with?" With some bitterness he criticizes
America for placing "more faith in the ability of the Russians to control their
tremendous stockpiles of offensive weapons than they do in my country's
capacity to use with wisdom and moderation the modest armaments it is
working so hard to develop for purely deterrent purposes.""
Worries and fears on any such grounds are exaggerated. The Soviet
Union and the United States influence each other more than any of the states
living in their penumbra can possibly hope to do. In the world of the pres-
ent, as of the recent past, a condition of mutual opposition may require
rather than preclude the adjustment of differences. Yet first steps toward
agreement have not led to second and third steps. Instead they have been
mingled with other acts and events that have kept the level of tension quite
high. The test ban was described in the United States as possibly a first great
step toward wider agreement that would increase the chances of maintain-
ing peace. In almost the same breath it was said that we cannot lower our
guard, for Soviet aims have not changed.18 Larger acts than agreement to
halt testing under the sea and above the ground are required to alter a situ-
ation that congealed long ago. The Soviet Union and the United States
remain for the foreseeable future the two countries that can irreparably dam-
age each other. So long as both possess the capability, each must worry that
the other might use it. The worry describes the boundaries that have so far
limited both the building up of tensions and the abatement of competition.
Where weapons of horrible destructive power exist, stability necessarily
appears as an important end. It will not, however, be everyone's highest
value. One who accepts the analysis of bipolarity and the conclusions we
have drawn may nevertheless prefer a world of many powers. The unity and
self-dependence of Europe may, for example, rank higher as goals than
international stability. Or, one may think of European unity as a means of
melding American power with the strength of a united Europe in order to
achieve Western hegemony. Unipolarity may be preferable, for those peoples
who then become dominant, to a competition between two polar states. It
may even promise a greater stability. The question is too complicated to
take up at the moment, but some words of caution are in order.
The United States has consistently favored the unification of Europe, for
adding the strength of a united Europe to the existing power of America
would be sufficient to establish a world hegemony. But there is a confusion

in American rhetoric that accurately reflects a confusion in thought. We
have wanted a Europe united and strong and thus able to share our burdens
with us, but a Europe a t the same time docile and pliant so that it would
agree on which burdens are to be assumed and how duties should be
shared. The enchanting dream of Western hegemony has many implica-
tions, some of them possibly unpleasant. A Europe of the Seven, or even the
Six, could, given time to put its combined resources to work, become a third
power in the world on the largest scale. President de Gaulle has entertained
the fear that such a Europe, if it were to be born under Anglo-Saxon aus-
pices, would serve as an instrument of American foreign policy. One may
have doubts of what would necessarily fol10w.~' De Gaulle is a useful
instructor. If we find the weak troublesome, will the strong be more easily
controlled? A united Europe would represent a great change in the world;
because the change would be great, its effects are difficult to foresee. If
Europe were t o be stable, strong, and cooperative, one might be delighted;
but surely it would be dangerous t o predict that a new Europe would rap-
idly find internal stability and develop political maturity. It would be more
dangerous still to assume that the old American and the new European state
would find their policies always in harmony. It is seemingly a safe assump-
tion that a clear and pressing interest of a new state of Europe would be to
stand firm against any Soviet attempts to move forward. But interests must
be taken in relation to situations. In a world of three great powers, identi-
cal interests may logically lead and in the past have led to dangerously dis-
parate policies. European history of the twentieth century makes optimism
difficult. Nor could one be serene about America's reaction. Typically,
Americans have insufficiently valued the prize of power. The yearning for a
Europe united and thus strong enough to oppose the Soviet Union unaided
is but one example. The pressures of bipolarity have helped to produce
responsibility of action. A relaxation of those pressures will change the situ-
ation t o one in which it will no longer be clear who will oppose whom. Two
considerations then should give one pause: the necessarily unpredictable
quality of the third power and the greater instability of a multipower
world.
A system of small numbers can always be disrupted by the actions of a

Hitler and the reactions of a Chamberlain. Since this is true, it may seem
that we are in the uncomfortable position of relying on the moderation,
courage, and rationality of men holding crucial positions of power. Given
the vagaries of men and the unpredictability of die individual's reaction to
events, one may at this point feel that one's only recourse is to lapse into
prayer. We can, nonetheless, take comfort from the thought that, like other
men, those who are elevated t o power and direct the activities of great states
are not wholly free agents. Beyond the residuum of necessary hope that men
will respond sensibly lies the possibility of estimating the pressures that
i The Stability of a Bipolar World 93
encourage and constrain them to do so. In a world in which two states

united in their mutual antagonism far overshadow any other, the incentives
to a calculated response stand out most clearly, and the sanctions against
irresponsible behavior achieve their greatest force. Not only how the lead-
ers will think but also who they may be will be affected by the presence of
pressures and the clarity of challenges. One may lament Churchill's failure
to gain control of the British government in the 1930's, for he knew what
actions were req~~ireci to maintain a balance-of-pwer system. Churchill did
come to power, it is interesting to note, as the world began to assume the
bipolar form familiar in wartime. If a people representing one pole of
the world now indulges itself by selecting inept rulers, it runs clearly
discernible risks. Leaders of the United States and the Soviet Union are
presumably chosen with an eye to the tasks they will have to perform.
Other countries can enjoy, if they wish, the luxury of selecting leaders who
will most please their peoples by the way in which internal affairs are
managed. The United States and the Soviet Union cannot.
It is not that one entertains the utopian hope that all future Premiers of
the Soviet Union and Presidents of the United States will combine in their
persons a complicated set of nearly perfect virtues, but rather that the pres-
sures of a bipolar world will strongly encourage them to act in ways better
than their characters might otherwise lead one to expect. It is not that one
possesses a serene confidence in the peacefulness, or even the survival of
the world, but rather that cautious optimism may be justified as long as the
pressures to which each must respond are so clearly present. Either country
may go beserk or succumb to in a nation and debility. That necessities are
clear increases the chances that they will be met, but there can be no guar-
antees. Dangers from abroad may unify a state and spur its people to heroic
action. Or, as with France facing Hitler's Germany, external pressures may
divide the leaders, confuse the public, and increase their willingness to give
way. It may also happen that the difficulties of adjustment and the necessity
for calculated action simply become too great. The clarity with which the
necessities of action can now be seen may be blotted out by the blinding
flash of nuclear explosions. The fear that this may happen has reinforced
the factors and processes described in the preceding pages.
By making the two strongest states still more powerful and the emer-
gence of third powers more difficult, nuclear weapons have helped to con-
solidate a condition of bipolarity. It is to a great extent due to its bipolar
structure that the world since the war has enjoyed a stability seldom known
where three or more powers have sought to cooperate with each other or
have competed for existence.
Notes
1. T h e point has been m ~ d bye Raymond Aron, a m o n g others. "Even ~f it had not had the
bomb, would the U n ~ t e dStates have tolerated the expansion of the Soviet empire a5 tar as the
Atlantic? And would Stalin have been ready to face the risk of general war?" Raymond Aron,
The Century of Total War (Boston: Beacon Press, 1955), p. 151.
2. Hanson W. Baldwin, from information supplied by Strategic Air Command headquar-
ters, estimates that Russian intercontinental missiles are one-fourth to one-fifth as numerous
as ours, though Russian warheads are larger. The Russians have one-sixth to one-twelfth the
number of our long-range heavy bombs, with ours having a greater capability (New York
Times, November 21, 1963). In medium range ballistic missiles Russia has been superior.
A report of the Institute of Strategic Studies estimated that as of October, 1962, Russia had
700 such missiles, the West a total of 250 (New York Times, November 9, 1962). British
sources tend to place Russian capabilities in the medium range higher than do American
estimates. Cf. P.M.S. Blackett, "The Real Road to Disarmament: The Military Background to
the Geneva Talks," New Statesman (March 2, 19621, pp. 295-300, with Hanson
W. Baldwin, New York Times, November 26, 1961.
3. See, for example, Walter Lippmann, "NATO Crisis - and Solution: Don't Blame De
Gaulle," Boston Globe, December 5, 1963, p. 26: "The paramount theme of this decade, as
we know it thus far, is that we are emerging from a two-power world and entering one where
there are many powers."
4. Cf. Georg Simmel, "The Sociology of Conflict, 11," The American Journal of Sociology,
IX (March, 1904), 675: "when one opposes a diffused crowd of enemies, one may oftener gain
isolated victories, but it is very hard to arrive at decisive results which definitely fix the rela-
tionships of the contestants."
5. In 1960, 1.5% of total population for France; 1.01% for the United Kingdom; 1.39%
for the United States. M.R.D. Foot, Men in Uniform (London: Weidenfeld and Nicolson, for
the Institute for Strategic Studies, 19611, pp. 162, 163.
6. The reduction is figured from the level of military manpower in 1960. Ministre des
Armkes, Pierre Messmer, "Notre Politique Militaire," Revue de Difense Natronale (May, 1963),
p. 754.
7. To complete the picture, Britain in 1962 had a gross national product of $79 billion and
France of $72 billion. Gross national product figures for all of the countries mentioned, except
China, are from the New York Times, January 26, 1964, E8. The figure of $50 billion for
China in 1962, though it is a figure that is widely given, is necessarily a crude estimate. As a
close and convenient approximation, I have taken 3, 5, 6, and 7% as doubling in 24, 14, 12,
and 1 0 years, respectively.
8. Albert Wohlstetter has estimated that the first one hundred Polaris missiles manufac-
tured and operated for five peacetime years will cost three to five times as much as the cost of
the first one hundred B-47s ("Nuclear Sharing: NATO and the N + l Country," Foreign Affairs,
XXXIX [ A ~ r i l 19611.
& & ,
. >,
364).
,
9. France plans to have three nuclear submarines of sixteen missiles each, the first to be
operating in 1969, the others following at two-year intervals (Messmer, "Notre Politique
Militaire," p. 747).
10. It is not wholly absurd for British and French governments to proclaim, as they fre-
quently do, that an embryonic capability brings an immediate increase of strength; for further
expenditures are not likely to bring much of an additional payoff. Cf. President de Gaulle's
message to his minister-delegate at Reggane upon the explosion of France's first atomic device:
"'Hurrah for France! From this morning she is stronger and prouder!"' Leonard Beaton and
John Maddox, The Spread of Nuclear Weapons (London: Chatto & Windus, 1962), p. 91.
11. The experiences of Chinese Communists prior to 1949 and of the People's Republic of
China since that date suggest that attempts to outflank may bring a greater success than efforts
to imitate! Or, applying an economist's term to military matters, would-be Nth-countries
would d o well t o ask, where d o we have a comparative advantage?
12. Morton A. Kaplan, System and Process rn International Politics (New York: Wiley,
1957), p. 37; and "Bipolarity in a Revolutionary Age," in Kaplan, ed., The Revolution rn
World Politics (New York: Wiley, 1962), p. 254. The difficulties and dangers found in a bi-
polar world by Kaplan are those detected by Hans J. Morgenthau in a system of opposing
alliances. It is of direct importance in assessing the stability of international systems t o note
Waltz The Stability of a Bipolar World 95
that Morgenthau finds "the opposition of two alliances ... the most frequent configuration
within the system of the balance of power" (Politrcs Among Nations [3d ed.; New York:
Knopf, 1961, part 41, p. 189). Kaplan, in turn, writes that "the most likely transformation o f
the 'balance of power' system is to a h~polarsystem" (System and Process, p. 36).
13. Kaplan, though hc treats the case almost as being trivial, adds a statement that IS at least
suggestive: "The tight b~polarsystem 1s stahle only when both bloc actors are h~erarchically
organized" (System and Process, p. 43).
14. Kaplan, e.g., by the fourth and sixth of his rules of a balance-of-power system, requires
a state to oppose any threatening state and to be wllling to ally with any other (System and
Prowss, p 23).
References
I. Henry Kissinger, "Strains on the Alliance," Forergn Affairs, XLI (January, 1963), 284. Cf.
Max Kohnstamm, "The European Tide," D~edalus,XCIII (Winter, 19641, 100-1 02;
McCeorge Bundy's speech to the Economic Club of Chicago, New York Trmes, December
7, 1961; John F. Kennedy, "Address at Independence Hall," Philadelphia, July 4, 1962,
Puhlzc Papers of the I'rcsrdents of the United States (Washington, D.C.: Government
Printing Office, 19631, pp. 537-539.
2. Quoted in V.D. Sokolovskii, ed., Sovzet Military Strategy, Herbert S. Dinerstein, I.eon
( h u r t , and Thomas W. Wolfe, translators and English editors (Englewood Cliffs: Prentice-
Hall, 19631, p. 43.
3. Herman Kahn, On Thermonucleur War (Princeton: Princeton University Press, 19601,
p. 315.
4. Richard N. Rosecrance, Action iwrd Reaction m World Politics (Boston: L~ttle,Brown, 1963),
pp. 210-211.
5. Hedley Bull, "Arlant~cMilitary Problems: A Preliminary Essay." Prepared for the Council
on Foreign Relations meeting of November 20, 1963, p. 21. Quoted with permission of
the author.
6. P. E. Vyshinsky, "Con~munismand the Motherland," 3s quoted in The Kremlin Speirks
(Department of State Publ~cation,4264, October, 1951), pp. 6, 7.
7. Nat~onalUnion of Conservative and Unionist Associations, Officral Report, 8 1st Annual
Conference, 1.landudno (October 10-13, 1962). p. 93.
8. Arnhassadc de FI-ance, Speeches and Press Conferences, No. 185 (January 14, 1963), p. 9.
9. In a statement taped in i'eklng before his African trip In January of 1964, Neu, York fines,
February 4, 1964, p. 2. Cf. the message sent by Communist China's leaders to Premier
Khru~hchevupon the occasmn of his seventieth birthday. After referring to differences
between them, it is stated that: "In the event of a maior world cris~s,the two parties, our two
peoples will undoubtedly stand together agamst our common enemy," New York Times,
April 17, 1964, p. 3.
10. L.eland M. Goodr~ch,"Korea: Collective Measures Against Aggression," Internationd
Conciliat~m,No. 494 (October, 1953), 164.
I I . "Stresses and Strains in 'Going It With Others,"' in Arnold Wolfers, ed., Allrance l'olicy
itr the Chld War (Baltimore: Johns Hopkins Press, 1959), p. 7.
12. The Statesmank Year-Bouk, S.H. Steinberg, ed. ( I a n d o n : Macm~llan,1948), p. 50. Ibid.
(IYSl), p. 99 1. [lid. (196.31, pp. 103, 104, 1003. The figures for Great Britain exclude
the women'^ services, Territorial Army, and colonial troops. Those for France exclude the
gendarmes.
13. Bull, "Atlantic Military Problems," p. 24.
14. Alastair Buchan and Philip Windsor, Arms and Stizbilrty in Europe (New York: Praeger,
lY6?1), p. 205.
15. Jacob Viner, "The implications of the Atomic Bomb for International Relations."
t'roc-eedi?l#s of the American P / ? ~ ~ J s ( J ~Society,
~ ~ c u XC:
/ ( 1946), .55.
16. Stanley Hoffmann, "Cursing de Gaulle Is Not a Policy," The Reporter, XXX (January 30,
1964), 40.
17. Cf. Malcolm W. Hoag, "On Stability in Deterrent Races," In Morton A. Kaplan, ed., T i ~ e
Revolution in World Politics (New York: Wiley, 19621, pp. 408, 409.
18. Cf. a forthcoming book by Victor Basiuk, Institute of War and Peace Studies. Columb~a
University.
19. House of Commons, Parliamentary Debates (March I , 1960),cols. 1 136-1 138. Compare
Hugh Gaitskell, The Challenge of Co-Existence (London: Methuen, 19571, pp. 45-46.
20. Quoted by Eldon Griffiths, "The Revolt of Europe," The Saturday Etlening Post, CCLXIII
(March 9, 1963), 19.
21. Bull, "Atlantic Military Problems," p. 29.
22. Hans J. Morgenthau, Politics Among Natrons (3d ed.; New York: Knopf, 1961 ), part 4.
Morton A. Kaplan, System and Process rn International Politrcs (New York: Wiley, 1957),
pp. 22-36. I shall refer only to Morgenthau and Kaplan, for their writings are widely
known and represent the majority o p i n ~ o nof students in the field.
23. Morgenthau, Politics Among Nations, p. 350. Cf. Kaplan, System and Process, pp. 3 6 4 3 ;
and Kaplan, "Bipolarity in a Revolutionary Age," In Kaplan, ed., The Revolution in
World Polrtics (New York: Wiley, 1962), pp. 2 5 1-266.
24. The point is nicely made in an unpublished paper by Wolfram E Hanrieder, "Actor
Objectives and International Systems" (Center of International Studies, Princeton
University, February, 1964), pp. 4 3 4 4 .
25. For a sharp questioning of "the myth of flexibility," see George Liska's revlew article
"Continuity and Change in International Systems," World Politics, XVI (October, 19631,
122-123.
26. Raymond Aron, Paix et Guerre entre les Natrons (Paris: Calmann-Lev); 1962), p. 156.
27. Gen. Paul Stehlin, "The Evolution of Western Defense," Foreign Affairs, XLII (October,
1963), 81, 77.
28. See, for example, Secretary Rusk's statement before the Senate Foreign Relat~ons
Committee, New York Times, August 13, 1963.
29. Ambassade de France, Speeches and Press Conferences, No. 175 (May 15, 1962), p. 6.
The Sharing of Nuclear Responsibilities: A Problem in
Need of Solution
Andre Beaufre
T he problem of the sharing of nuclear responsibilities has been NATO's

main stumbling block in recent years. It has become obscured by
opinions that are too narrow and too conservative - to the extent that
it appears almost insoluble. However, in this article, I shall try to analyse the
present state of this complex problem and to show that, if it is examined
objectively, real and effective solutions to it can be envisaged.
In recent years, this problen~has been dominated by the differences of attitude

towards it which have reflected both the aims and anxieties of the various
allied nations.
For the Americans, the leading partners in the alliance and controllers of
98 per cent of its nuclear capability, the desire in Europe to share in decisions
relating to the planning and use of nuclear weapons has posed a specific
problem: they have recognised that desire as politically and psychologically
legitimate but, at the technical level, they have been unwilling to accept any
solution to the problem other than the retention of nuclear co~itroli11 the
hands of the President of the United States. The Americans have been com-
mitted to this attitude by the McMahon Act - which is categorical on this
point - and for obvious reasons of national security. Moreover, they have
regarded as hardly fair any idea that they should abandon the principle of
their freedom of nuclear action when they have provided virtually the entire
nuclear arsenal. Politically, they have not been able to refuse discussion of the
problem, but they have been convinced that they need accept only a small
share by their allies in nuclear planning, and that those allies could be satis-
fied by concessions more apparent than real.
The attitude to the problem of Britain, the second nuclear power in the
alliance, has rested on two basic aims. One of these has reflected her great
Source: Internatronol Affirm, 4 l ( 3 ) ( 1965): 41 1-19

vulnerability to nuclear attack, the strong current of domestic opinion

against nuclear weapons and her fear of finding herself dragged into a
nuclear conflict. That aim has been to search for the best possible means of
controlling nuclear weapons to guard as far as possible against a 'war by
accident'. Her second aim, an outcome of her experience in nuclear matters,
has been to find less costly ways of maintaining and developing the British
nuclear deterrent by including it in an inter-allied system in which Britain,
by virtue of her contribution to it, could have a special place.
The Germans, conscious both of providing the core of the alliance's
ground defence and that their territory was likely to be the possible battle-
field, have felt that they were being excluded from any share in the major
decisions upon which their fate might depend. They have wanted, therefore,
not only to have a guarantee of support from the American arsenal to cover
every eventuality, but also to be promoted from what they have regarded as
a discriminating status in the alliance.
For the French, the nuclear debate began a long time ago, when it was
planned to install 1.R.B.M.s and to establish stockpiles of tactical nuclear
weapons in France. The French Government had never considered the
'double-key' principle as adequate because it would have given it no more
than a local veto over the use of weapons based on French territory. What
they insistently demanded was that France should share in the formulation
of plans and in the decisions for use of those weapons, and when they failed
to obtain any satisfaction they refused to allow the installation of 1.R.B.M.s
and nuclear stockpiles on French soil. When General de Gaulle came to
power, he reasserted the French position in a memorandum in which he
proposed that the three nuclear powers in the alliance should devise a joint
nuclear strategy. When this memorandum was virtually ignored, the French
Government reflected its displeasure by a partial withdrawal from NATO
in a series of measures which were more spectacular than substantial.
The various solutions to the basic problem of the sharing of nuclear respon-
sibilities which have been proposed naturally reflect these different national
aims and attitudes.
The Americans thought they had found a solution to the problem in the
plan for a multilateral force. This concept, however, was originally no more
than a secondary product of the Nassau Conference, of which the principal
outcome was the project for a multi-national force embracing the United
States, Britain and France. As a result of France's refusal to take part that
idea was stillborn, and the whole force of American diplomacy was then
concentrated behind the proposal for a multilateral force originally devised
primarily to meet German anxieties. This proposal had the advantage of
completely safeguarding the Americans' freedom of nuclear decision while
at the same time including the European forces in a system which gave them
a semblance of consultation. In addition, it implied the creation of a tightly
t+<rrilre The Sharing of Nuclear Responsibilities 99
integrated inter-allied force very closely linked both to the NATO command
and to American technology. At the same time, a few minor alterations were
made in NATO's nuclear organisation in Europe to provide for a wider
share by European officers in local nuclear planning.
The disadvantages of this scheme were that it would be costly and that
it would, in practice, merely result in reinforcing what was already a super-
abundant American nuclear capability without providing the European
partners with anything more than a nominal right of control. In general, it
had only a lukewarm reception, except by the Germans; they saw it as a
step towards the removal of the discrimination from which they felt them-
selves to be suffering, and as an opportunity to express their support of the
United States whose nuclear protection seemed essential to them. But in
order to guard against what they considered to be the dangerous hypothesis
that the land defence of Europe should depend on purely conventional
weapons, they also insisted on the absolute need for that defence to be
based on tactical nuclear weapons.
France clearly expressed her opposition in principle to the MLF, both
because it offered only illusory possibilities of any real sharing of nuclear
responsibilities and because it irrevocably integrated Europe in the
American nuclear system, thereby excluding any hope that a real European
nuclear force might eventually be established. Instead, France continued to
build up her small nuclear force and to reiterate her wish that the allies
should co-ordinate their strategies in an effective way.
In this situation, the new British Labour Government thought it a clever
move, in the hope of gaining support from the European opponents of the
MLF, to propose an alternative to it. The essential characteristics of this
new ANF proposal were the creation of a multi-national force, composed
of an equal number of British and American Polaris submarines, and of a
mixed-manned Polaris surface fleet on the MLF pattern, but now without
British participation. It was proposed that this force should be placed at the
disposal of NATO to carry out 'interdiction' missions of a more or less
long-range nature in support of the defence of Europe. In addition, certain
arrangements would allow the European Powers to exercise a minimum of
control over the planning and use of this force.
The most obvious result of the British proposal was to deprive the MLF
of what little impetus remained to it. The collision of so many opposing
views on the problem of the sharing of nuclear responsibilities resulted in a
general hesitation to deal with it, and by the beginning of this year it had
reached a dead-end.
During this period, however, it has become clear that the strategic situation
has so far been transformed that the need for an important revision of the
concepts on which the strategy of NATO had been based has become
clearly indicated. For instance, it has become obvious that the hypothesis of
the threat of a Soviet invasion of Western Europe has now, to all intents and
purposes, lost any validity. This development has, of course, been brought
about in large part by the value of NATO as a deterrent, but it has also been
due t o the political and psychological evolution within the U.S.S.R. In any
case, the fact is that not only does such an invasion now seem improbable,
but that the hypothesis of a nuclear war has become unthinkable - because
a nuclear war would certainly entail reciprocal destruction of such propor-
tions that no political objective could justify it. In modern strategic jargon,
the 'bilateral deterrence' of both sides is 'bistable absolue'.
This new situation, which I have analysed in detail in my book,
Dissuasion et Stratkgie,' must have some highly important consequences -
in that it must lead t o a complete change in the ideas to which we have
become accustomed.
These ideas were based on the proposition that, in order to deter an
attack, it was necessary to develop a defensive and retaliatory capacity
which could win the war, or, at least, deprive the enemy of any hope of win-
ning it. This conception implied the need to foresee the course of a nuclear
war and the means by which victory could be gained in it; hence the
detailed study devoted to the problem of launching nuclear weapons and to
the use of tactical and strategic weapons. But such exercises in foresight lost
any purpose from the moment that the enemy possessed sufficient means
of nuclear retaliation. In present circumstances, both at the tactical and
strategic levels, a nuclear exchange would rapidly lead to irrational situations
in which none of the belligerents would be able to maintain its control over
operations. The extent of the destruction that would be caused on both
sides would make this kind of war what might be termed suicide 'by return'.
Given that probability, it cannot but be concluded that the unleashing of a
nuclear war would be the signal for a major catastrophe which should be
avoided at all costs. The course of events immediately after such a war had
begun would render the war itself futile. Thus, contrary to what has been
assumed hitherto, the kernel of the problem does not lie in devising a defen-
sive strategy t o be used in time of war, but in the strategy employed in time
of peace, before any use of nuclear weapons, to deter any open conflict. The
main concern has thus shifted from defence to deterrence and from wartime
to peacetime.
But the application of this new strategy of deterrence is a particularly
delicate matter. When nuclear war seemed plausible, in the Foster Dulles
era, the mere threat of it was an easy and effective gesture. Today, when
each of the two major opponents is convinced that anything is preferable to
a resort t o nuclear weapons, any threat to use them tends to lose all cred-
ibility. O n the other hand, if one of those opponents were to think that such
weapons would not be used in any circumstances, the resulting situation
would be even more dangerous, for there would then be a risk of the loss of
the advantage of the stability which the mere fear of a nuclear conflict has
imposed upon Europe. The political instability resulting from the violent
division of Europe and Germany, and from the subjection of the satellite
The Sharing of Nuclear Responsibilities 10 1
states, cannot avoid leading to conflicts over which, even though they might
he of secondary importance, any governmental control might be lost, so
bringing about a new great war. Even if this serious possibility could he
avoided we would still have to face dangerous complications. That is why
special efforts need to be made to preserve for nuclear weapons that capac-
ity for engendering fear from which we have hitherto so greatly benefited.
Several methods have been proposrd to attain this end. The first - that
of Mr. McNamara, Phase One - was to declare that, if she launched the first
strike, the United States possessed the means to reduce the enemy's capacity
for retaliation very considerably. This thesis is perfectly logical and would
be completely convincing if this 'counter-force capability', which in any
case entails considerable resources, did not appear increasingly less easy to
maintain in face of the development of missile-firing submarines and rockets
protected in concrete silos.
O n the other hand, the Soviet approach to the problem has hitherto been
to declare that war could not be limited, and that any important conflict
would escalate to the nuclear threshold. This is an irrational position and
one which tends to give the nuclear threat a minimum of credibility. In the
same spirit, France has refused to accept the principle of flexible response
as a demonstration of her willingness to face the prospect of escalation to
the strategic nuclear threshold in the event of aggression.
A variation of these attitudes has been provided by the Germans: they
have proposed that tactical nuclear weapons should be placed along the
whole length of the Iron Curtain, so that the aggressor would be thus
assured that any invasion would result in escalation to the nuclear threshold.
The French and German attitudes appear dangerous to American experts,
who retnain haunted by the fear of a nuclear conflict and who, because of the
risks to American territory that such a conflict would now imply, wish to
guard against any 'catalytic' escalation. They have therefore been prompted
to put forward an original and adroit proposal by which the risk of nuclear
conflict would be preserved but maintained at a limited level. This is Mr.
McNamara's Phase Two solution, in which he asserts the possibility that the
United States might respond to an act of aggression by nuclear warning shots,
intended to show her willingness to resort to nuclear weapons, hut only
demonstratively and without actually beginning an all-out nuclear conflict.
The enemy would thus be deterred from resorting to nuclear war by the
threat of massive retaliation against his population, a threat substantiated by
an important fleet of missile-firing submarines and a large number of concrete-
encased Minutemen. As Mr. ~ c ~ a m a himself
r a has put it, damage is to be
limited by the threat of 'insured destruction'.
Incontestably, each of these attitudes reflects the logic of the different
capabilities of the nuclear powers - a clear indication that deterrence cannot
be applied in the same way by all the nuclear powers and that American
policies cannot be followed by lesser nuclear partners. But given the fact that
all these attitudes reflect attempts to grapple with the difficult problem o f
restoring the credibility of a threat which everyone knows to have hecome
unreal, it is very important that everything possible should be done to ensure

that that threat should retain that minimum of spontaneous risk which leads
to that prudence indispensable to the maintenance of peace. To achieve this
end, different methods must simultaneously be used to keep the enemy's mind
in that state of uncertainty which alone can render deterrence effective; thus,
to enable several methods of deterrence to be used simultaneously, there must
be several centres of decision. This basic conclusion is contrary to American
policies; in fact these have for too long been focused upon the stabilisation
of the nuclear threshold although it is today so stable that it needs to be, so
to speak, 'destabilized' to restore its deterrent effect.
But what emerges above all else from this analysis is the fact that deter-
rence can no longer rest exclusively upon the existence of military forces
and retaliatory capacity whose use is foreseen only in time of war. Such
strength remains necessary, but it must be exploited, and its peace time
value enhanced, by a suitable manipulation of the threat it represents, so
that in time of crisis the hand of deterrence can be played to the full, so
keeping the enemy uncertain as to the reactions he could unleash. Such
manipulation, of which we saw an example in the Cuban crisis, forms the
essence of the present deterrent posture. It is a peacetime operation.
Yet present NATO arrangements relate only to time of war. Its inte-
grated command would be only effective from the moment of any enemy
aggression. Contrary to widespread belief, there is no forward planning
within NATO to deal with any crisis that might develop in peacetime. Only
in relation to Berlin has there been forward planning, and on a hypothet-
ical basis other than that on which NATO was created - the possibility of a
defensive war. It is this latter fact, indeed, which makes the idea of an inter-
allied nuclear force placed permanently under NATO command completely
unrealistic. Hitherto, no one has succeeded in placing the smallest conven-
tional unit under NATO command in peacetime. How, then, is it possible
to conceive that such a development would be possible in terms of a nuclear
force - given the much more delicate problems involved? It is therefore neces-
sary to re-examine the organisation of the alliance so that it may be adapted
to new circumstances.
In these new circumstances, the guiding principles of deterrence appear to

be these:
(a) Since it is desirable to increase the uncertainty felt by the enemy by more
than one method and by basing a strategy of deterrence on more than
one centre, it is quite useless to create a new force, such as the MLF or
the ANF, if this is to be dependent directly and openly upon American
nuclear strategy, which already possesses a super-abundant capacity. To
be effective as a deterrent, any new force must, in peacetime, be subject to
an autonomous centre of decision. On the other hand, a nuclear centre of
I3edutrc The Sharing of Nuclear Responsibilities 103
decision cannot be international in character. If it is to be credible, it can

only be national.
(b) The control of deterrent strategy cannot be collective in peacetime. In such
circumstances, only the principle of national independence can be applic-
able, an independence which naturally applies to nuclear no less than to
conventional forces. This principle has not the dangerous implications
often associated with it, because the nuclear threshold has become far too
stable. As we have already seen, a minimum instability must be restored
to it if the deterrent is to retain any value.
(c) In these circumstances, the peacetime problem of the sharing of nuclear
responsibilities no longer lies in the search for an integration of nuclear
capabilities - which is, in any case, impossible to achieve - but in the
organisation of a n effective co-ordination of national deterrent strategies.
Such co-ordination must not aim merely at persuading each of the allies
t o adopt a single method of deterrence. O n the contrary, it must aim at
promoting a mutual understanding of the needs and fears of each, and at
making the best use, by common agreement and in the common interest,
of the specific strategies of each of the partners in relation to the interests
o f the alliance as a whole.
( d ) Forward planning for a war, however unlikely, remains necessary. This
must deal essentially with the organisation of a possible system of nuclear
command, and with the allocation of responsibilities among the various
national forces. Apart from its deterrent value, this forward planning
would provide an opportunity to begin the organisation of inter-allied
consultations in the sphere of nuclear ~ l a n n i nand
~ , the preparation of the
framework for their European nuclear force which must be established
when Europe has eventually taken shape as a political entity.
The ideas which I have put forward above, and which I believe to be soundly
based, are none the less too novel to gain immediate and universal accept-
ance. Some time will no doubt elapse before the need will be recognised for
such an adaptation of our concepts to accord with that evolution of the
strategic situation at present unevenly understood within the alliance. It may
be thought, therefore, that this is too early a stage for positive proposals for
the reorganisation of the alliance to be put forward.
Nevertheless, even if the overall thesis which I have outlined is received
with reservations, two facts are unquestionable:
I . There are no arrangements within the alliance (except for Berlin) to pro-
vide for possible peacetime crises - which can arise at any time in
Central Europe as a consequence of its political evolution; nor does
there exist any means to co-ordinate action by the allied powers in the
event of such a crisis. It is absolutely essential that this gap in forward
planning should be filled.
2. Whether or not one approves of the nuclear situation within the alliance,
there are at present three national nuclear forces whose organisation
must be co-ordinated to provide for the event of war.
These two axioms imply the need to set up two kinds of organisation.
The first would have the task of carrying out, in Europe, what the Americans
call 'crisis management' - no doubt in the form of a 'contingency planning
committee' intended to examine possible crises and their solutions. Such a
committee would certainly need to work for some considerable time before
it was able t o produce practical results. This is a further reason for not
postponing the formation of such a group, for it would have the advantage
of bringing about close co-operation among the major allies - including
Germany - on problems of immediate concern.
The second body would be charged with developing, within the Atlantic
framework, the necessary basis for the co-ordination of national nuclear
forces in time of war. Such studies would allow for the discussion of the
principles of nuclear strategy - something which has hitherto, to all intents
and purposes, been confined to a purely American framework. It seems that
the proposals lately put forward by Mr. McNamara in Paris amount to a
first step in this direction.
The main purpose of these two committees would be to produce a better
mutual understanding of the viewpoints of the different allies and thereby to
pave the way for the development of positive policies acceptable to all. At a
later stage, it would be possible to envisage the creation of a body designed
to improve the co-ordination of peacetime deterrent strategies and wartime
nuclear strategies. I do not propose to mention now any of the schemes
which could be put forward in present circumstances, because the strategic
position is constantly changing and no one can precisely foresee its shape by
the time such studies might have been completed.
It is on this note that I would like to conclude. The nuclear history of the
last 15 years has shown that the situation has constantly changed and that,
on each occasion, the changes have been far-reaching. As I have tried to
point out in this article, it is some time since we entered upon a new phase,
marked by an excess of stability at the nuclear threshold. We have not yet
taken sufficient account of the consequences of this changed situation in
our strategic nuclear ideas, for these are still based on the concept of deter-
rence although the use of nuclear weapons has become unthinkable. It is
therefore urgent that we should adapt ourselves to this new situation.
But, equally, this situation can itself change rapidly. Already, the Vietnam
crisis is leading to an important change of perspective towards Europe on
the part of the United States. Failing a relatively rapid compromise, these
developments can lead to an end of the essentially bipolar system which has
controlled world strategy since 1945. The consequences in Europe of such
The Sharing of Nuclear Responsibilities 105
an event c o d d be considerable and could profoundly alter the strategic

assumptions to which we have become accustomed.
In other words, the bases of any strategic diagnosis are changing more
and more rapidly and, in order to be effective, any such diagnosis must be
characterised by constant observation and adaptation, and approached
without preconceived notions and in a frame of mind receptive to every
kind o f change.
Note
I . I'.ir~s: Armand Colin. 1964. An English edition will he published in the ~ u t u m nby F,lher &
Faller.
Strategic Studies and Its Critics
Hedley Bull
T he civilian strategic analysts who now constitute a distinct profession

in the Western world have from the first been subject to criticism that
has called in question the validity of their methods, their utility to
society, and even their integrity of purpose.' Some of it is directed at par-
ticular strategists or at particular techniques they employ, but much of it
purports to expose deficiencies that are characteristic of the genre. Some of
this is of so scurrilous a nature as not to deserve a reply, but some raises
issues of real importance. What are in fact the distinguishing features of the
new style of strategic analysis? What has given rise to the criticisms that
have been made of it? And what substance d o the criticisms have?
Strategy in its most general sense is the art or science of shaping means
so as to promote ends in any field of conflict. In the special sense in which
I am using it here, the sense in which "strategy" is interchangeable with
"military strategy," it is the art or science of exploiting military force so as
to attain given objects of policy. If we contrast the strategic thinking of con-
temporary military analysts with the classical tradition of strategic thought
from Clausewitz to Douhet, certain of its peculiarities are at once apparent.
First, strategic thinking at the present time is no longer exclusively con-
cerned with the efficient conduct of war. From the time of Napoleon to that
of Hitler, strategy was conceived of as an aspect of war. Contrasting it with
tactics, which was the art of winning battles, Clausewitz defined strategy as
"the art of employment of battles as a means to gain the object in war."] How
to gain the object in war remains a central preoccupation of contemporary
strategic thinking, but it is no longer the only one or necessarily the most
important. Attention has shifted away from war as an instrument of policy
toward the threat of war, and studies of actual violence have given place to
analyses of "deterrence," "crisis management," "the manipulation of risk" -
or, as we call it when it is pacticed by our opponents rather than ourselves,
"blackmail." Moreover, gaining the object in war, even when it remains the
concern of strategists, is no longer always seen in Clausewitz' sense of attain-
ing victory by imposing our will on the adversary. In discussions of the
Source: World Politics,20(4) (1968):593-605.
i-irli Strategic Studies and Its Critics 107
conduct of strategic nuclear war the object of victory over the opponent has
in fact taken second place to that of our own survival. It has sometimes been
argued that the chief mission of United States strategic nuclear forces in the
event of general war is that of the limitation of damage suffered by the United
States and its allies - an object that is not relative to the amount of damage
suffered by the enemy, but absolute.
Second, strategic thinking is no longer the preserve of the military. The great
strategic writers of the past, like Liddell Hart, Fuller, and Mao Tse-tunb' in our
'
own time, were soldiers (or sailors or airmen) or ex-soldiers. They were often
quite bad soldiers; and they had qualities of mind that soldiers, good and bad,
do not often have. But underlying all their theorizing was the assunlption that
strategy was in some sense a practical business, that experience of the man-
agement of forces and weapons in war, even if it was not a sufficient condition
of strategic understanding, was at least a necessary one.
The military profession today is very far from having vacated the field of
strategy; in wide areas of strategic policy the chiefs of staff responsible to
governments remain the preponderant influence. But in the United States
and to a lesser extent elsewhere in the Western world, the civilian experts
have made great inroads. They have overwhelmed the military in the qual-
ity and quantity of their contributions to the literature of the subiect; no one
would now think of turning to the writings of retired officers rather than to
the standard academic treatments of deterrence, limited war, or arms control
for illumination of the problems of the nuclear age. They increasingly dom-
inate the field of education and instruction in the subject - the academic and
quasi-academic centers of strategic studies have displaced the staff colleges
and war colleges, except in narrow fields of professional knowledge. And,
most prominently in the United States, the civilian strategists have entered
the citadels of power and have prevailed over military advisers o n major
issues of policy.
A third peculiarity of strategic thinking at the present time is its abstract
and speculative character. There has not yet been a nuclear war, and the pos-
sibility that there will be one has not yet existed long enough for it to have
become clear how the structure of international life will be affected. Anyone
who has embarked upon a discussion of what the conditions are under which
one country can deter another from doing something, of whether or not
limitations are possible in nuclear war, of whether the nuclear stalemate
makes conflict at lower levels more likely or less, or of whether one country
can credibly threaten to use nuclear weapons on behalf of another must have
experienced the sense of being at sea in an argument i11 which, it seems,
almost any position can be plausibly defended and almost none is safe from
attack.
Strategic thinking, of course, has always been speculative. It has always
had to deal with the future, and it has always involved the making of plans,
the fulfilment of which depends on decisions taken by the opponent as well
as o n those we take ourselves. And the conditions of war and crisis under
which these decisions have to be made make them peculiarly difficult to
anticipate and peculiarly unlikely to follow the lines of assumed standards

of rationality. The advent of nuclear and missile technology, however, has
rendered strategic thinking speculative to a degree that it had not previously
attained. It is not the physical effects of nuclear explosives and missiles that
are speculative; indeed, in this respect war has become more predictable
and measurable than before. What cannot be confidently foreseen is how
statesmen, governments, and societies will behave under the stress of the
use of these weapons or the threat to use them. In a period of vast changes
in warfare and its place in human affairs, the relevance of history and ex-
perience, and the competence of those whose expertise is founded in them,
have rightly been called in question.
A fourth characteristic of strategic thinking at the present time is its sophis-
tication and high technical quality. Many students of strategy today take the
view that until our own time military affairs escaped sustained scientific study
and received only the haphazard attention of second-rate minds. Accordingly
they see themselves as presiding over the birth of a new science, eliminating
antiquated methods and replacing them with up-to-date ones. Some take the
view that there is a close analogy between strategic studies and economics, and
they hold out the hope that the former subject, when it emerges from its birth
pangs, will enable us to rationalize our choices and increase our control of our
environment to the same extent that the latter has done.
This view does less than justice to the classical tradition of strategic
thinking, while it also fails t o recognize the very slight extent t o which the
new scientific rigor in strategic studies has so far circumscribed the domain
of speculation. Nevertheless, it is clear that the intellectual resources now
being devoted to strategic studies are without precedent and that this has
resulted in a literature of higher technical quality and a discussion of a
higher standard of sophistication than have existed before. One incidental
consequence of this is the emergence of strategic studies as an appropriate
subject for inclusion in university curricula. Although I do not myself believe
that it is desirable to separate strategic studies from the wider study of
international relations, it can be argued that it compares very favorably
with some other branches of political science both in its moral and social
relevance and as an intellectual discipline.
A number of factors account for the barrage of criticism that the civilian
strategists have had to face. For those who feel guilt about modern war or
have fear of it - and in some degree this includes all persons who are sensi-
tive and aware - the strategists have undoubtedly provided a scapegoat. The
political influence that the civilian strategists have come to command, espe-
cially though not exclusively in the United States, has caused resentment - on
the one hand among the older generation of soldiers and civil servants whose
influence they have displaced, and on the other hand among their fellow intel-
lectuals who have remained outsiders. Their willingness to treat strategy as a
specialist's subject, even as an esoteric one, has irritated those who are unable
to understand or to emulate them. Their insistence on the complexity of the
problems of strategy and arms control has been unwelcome to purveyors of
Strategic Studies and Its Critics 109
simple solutions of one kind or another. Most basically, perhaps, the position
of the professional strategist is and will remain controversial because the
legitimacy of the question he sets himself - What shall the state do with its
military force? - is itself controversial. While there continues to be disagree-
ment in modern society as to whether or not the state should ever use mili-
tary force or possess it at all, there will not be general agreement ahout the
worth and utility of students of strategy, in the way in which there is now
(although there has not always been) about that of students of medicine,
architecture, or economics.
To show that the motives that underlie criticism of the strategist are some-
times discreditable is not, of course, to say that it is only from these sources
that criticism arises, still less to provide a rebuttal of the criticisms themselves.
Many of the criticisms are worth sympathetic consideration. In my view they
do not constitute, either singly or collectively, a valid indictment of the work
of the civilian strategists. But we should be grateful that they have been made,
for they do draw attention to some false paths along which strategists might
stray and sometimes have strayed. Here 1 shall consider five of the charges.
The first and most common complaint is that the strategists leave morality
out of account. Strategists are often said to be technicians and calculators who
are indifferent as to the moral standing of the causes for which war is under-
taken or of the means by which it is carried on.
There is a sense in which strategic thinking does and should leave morai-
it): out of account. Strategy is about the relationship between means and
ends, and an exercise in "pure" strategy will exclude consideration of the
moral nature of the means and the ends, just as it will exclude anything else
that is extraneous. If what is being said is that strategic judgments should
be colored by moral considerations or that strategic inquiry should be
restricted by moral taboos, this is something that the strategist is bound to
reject. If what the critics of Herman Kahn have in mind is that he should
not have thought about the unthinkable or that he should have thought
about it with his heart instead of his head, then they are obstructing him in
his essential task.
What can be said, however, is that while strategy is one thing and morals
are another, the decisions that governments take in the field of military policy
should not be based on considerations of strategy alone. If the charge against
the strategists is that their advice to governments is drawn up in purely strate-
gic terms, as if strategic imperatives were categorical imperatives, or that they
themselves have no other dimension in their thinking than the calculation of
means and ends, then this is a serious and legitimate complaint.
But so far as one can judge, such a charge is not true of any of the strate-
gists. It is easy to see that their works, dealing as they d o with strategy and
not with other subjects, might give the impression that decisions should be
determined by the logic of this subject alone, but there is no reason to
believe that this impression is correct. Strategists as a class, it seems to me,
are neither any less nor any more sensitive to moral considerations than are
other intelligent and educated persons in the West.
1 10 T h e Cold War a n d Nuclear Deterrence
Why, then, is the charge so frequently made? Can all the critics be
wrong? Surely as between Herman Kahn and his critic James Newman, as
between Irving Horowitz and those he calls "the new civilian militarists,"
or as between Anatol Rapoport and the various unnamed strategists who
are his targets there is some sort of moral disagreement. I believe that there
is, but that what is at issue is not whether or not moral questions should
be asked before decisions are taken but what the answers to the moral
questions are.
In almost any disagreement as to whether or not to resort to war or to
threaten it, or as to how a war should be conducted or what risks in it should
be run, there are moral arguments to be advanced on both sides. What the
critics take t o be the strategists' insensitivity to moral considerations is in
most cases the strategists' greater sense of the moral stature of American
and Western political objectives for which war and the risk of war must be
undertaken. The notion that virtue in international conduct lies simply in
avoiding risk of war and never in assuming it, always in self-abnegation and
never in self-assertion, only in obeying the rules a world community might
legislate if it existed and never in pursuing the different moral guides that
are appropriate in a situation in which it does not - such a notion is of
course untenable. But it forms part of the perspective of many of the critics.
What chiefly characterizes the so-called idealist school to which they belong
is not (as is often said) that it exaggerates the force of moral considerations,
still less that it alone is endowed with moral vision, but that it fails to
appreciate the full range of the moral argument, that it embraces what
Treitschke called "the monkish type of virtue" without being able to see
that there is any other.
There is, I think, a related moral disagreement between the strategists and
their critics, which concerns the role of the strategist as an adviser to govern-
ments. It is said that there is something unbecoming to an intellectual - or at
all events to a university man, with his allegiance to the universal republic of
science - in bestowing the fruits of his strategic advice upon any particular
government. Since governments use this advice to further their conflicts with
one another, the strategic adviser is in a different position from the scholar or
scientist who gives advice about the economy or health or education, since in
these fields the interests of one nation may be advanced without injuring
those of others. The scholar may legitimately ~ r o f f e radvice, if he has any,
about the conditions of peace, so the argument goes, but he is disloyal to his
calling if he provides advice about war.3
Some of this criticism may be met readily enough. One may point out
that the strategic interests of nations are not wholly exclusive of one
another and that contemporary strategists have been inclined to draw atten-
tion to the common interests that nations have in avoiding nuclear war and
in limiting it if it occurs. One may say that one of their contributions has
been the systematic study of arms control, which may be defined as co-
operation among antagonistic states in advancing their perceived common
interests in military policy. Arms-control policy is, I should say, subsumed
t:tiIi Strategic Studies and Its Critics III
under strategy as a special case. It may also be pointed out that it is facile
to regard war and peace as alternative objects of policy, as if peace did not
need to be enforced or war were not an outgrowth of diplomacy.
Yet it remains true that the strategic adviser does assist the government
he serves to advance its objectives at the expense of those of other govern-
ments. But whether or not there is anything in the position of such an
adviser unbecoming to a scholar or a scientist will depend on what we take
the moral nature of that government
- and its objectives to be. Few of the
critics would, 1 think, argue that the scientists who assisted the British and
American governments during the Second World War, and whose position
the contemporary strategists have inherited, were acting in an improper
way. Not everyone will agree that the position is the same now; but at least
it is not possible to maintain that there is any general incompatibility
between assisting a state to augment its relative military position and
remaining faithful to scholarly or scientific values.
The second criticism that I wish to discuss is that strategists take for
granted the existence of military force and confine themselves to considering
how to exploit it, thereby excluding a whole range of policies such as
disarmament or nonviolent resistance that are intended to abolish military
force or to provide substitutes for it.
It is true that strategists take the fact of military force as their starting
point. The question is whether any other starting point is possible at all,
whether the doctrine of disarmament that is implicit in this complaint is not
inherently untenable. The capacity for organized violence between states is
inherent in the nature of man and his environment. The most that can be
expected from a total disarmament agreement is that it might make arma-
ments and armed forces fewer and more primitive.
If what is meant by "total disarmament" is a state of affairs in which war
is physically impossible, in which states cannot wage war even when they
want to (this is what Litvinov meant when he first put forward the proposal
in 1927), then we must say that such a state of affairs cannot be. If, on the
other hand, what is meant is a situation in which military force has been
reduced to very low qualitative and quantitative levels, then this is some-
thing that can in principle occur and may well seem worth trying to bring
about. But the view that security against war is best provided by a low level
of armaments rather than a high one is a particular strategic theory; the
arguments for it and against it belong to the same mode of discourse as that
we apply in evaluating any other proposition about the relationship between
military force and possible ends of policy.
Either, then, the second criticism is a nonsense or it represents an attempt
to contribute to strategic reasoning, not a statement about it from outside.
In fact, it would seem to me, proposals for radical disarmament and
for nonviolent resistance have received a fair hearing within the Western
community of strategists. No doubt strategists are inclined to think too
readily in terms of military solutions to the problems of foreign policy
and to lose sight of the other instruments that are available. But this is the
occupational disease of any specialist, and the remedy for it lies in entering
into debate with the strategist and correcting his perspective.
The third criticism is that strategists are inclined to make unreal assump-
tions about international politics and that in comparing alternative strategies
and computing their costs and benefits they make assumptions that simplify
and distort political reality, that d o not allow for change, and that in the
course of the subsequent analysis become lost to sight.
This is a complaint that has a great deal of force. The technical rigor and
precision of much strategic analysis has been achieved at the cost of losing
touch with political variety and change. If the political terms in the strategists'
equations were more complex and were changed more frequently, the beauty
of much of the ratiocination would be destroyed.
Some of the now-classic analyses of America's problem in choosing her
weapons and military posture were founded upon the assumption that there
was only one significant relationship in nuclear international politics, that
between the United States and the Soviet Union, and that this consisted only
of hostility. Not only, as it were, was the game two-person and zero-sum,
but the two persons were assumed to be identical twins, Country A and
Country B. Even when these analyses were first made they were a simplifi-
cation of reality, but with their survival into the age of the Soviet-American
dktente and of the disintegration of the Atlantic Alliance and the Communist
bloc, they became dangerously unreal. The greatest absurdities of this sort
in recent times formed part of the debate that took place in the United
States during the Kennedy administration about the control of nuclear
weapons in NATO. The various solutions were set out in programmatic
form - a United States nuclear monopoly, national nuclear forces, a NATO
nuclear force, a European nuclear force - and their advantages and
disadvantages were spelled out on the basis that NATO was a single person
and that the sole requirement of that person was to deter attack by the
Soviet Union. Not all those who contributed to the debate, of course,
formulated the problem in this way, but many a weighty treatise appeared
that did so. General de Gaulle has now demonstrated what was perhaps all
along clear, that Paris and London are not Washington and that nuclear
forces have diplomatic functions as well as military ones; but it is extraor-
dinary for how long, under its own momentum, this strange logic persisted.
All that one can say in defense of the strategists against this charge is
that follies of this sort are not inherent in what they do, that technical pre-
cision must often be sacrificed so as to allow for political variety and
change, and that enough of the strategists are aware of this to ensure that
the corrections can come from inside the strategic community.
The fourth criticism is that the civilian strategists are pseudoscientific in
their methods, that specialist techniques they employ - such as game theory,
systems analysis, simulation, and the writing of scenarios - are bogus when
used to arrive a t strategic decisions and serve to give an air of expertise to
positions arbitrarily and subjectively arrived at. This is the theme of the
book Deadly Logic, by Philip Green, and it is also part of the meaning of
I Strategic Studies and Its Critics 1 13
the wrong-headed but subtle and powerful book Strategy and Conscience,
by Anatol Rapoport.
The crux of the matter is the attack on game theory, which more clearly
than any of the other techniques mentioned does represent an impressive
expertise. Rapoport presents some strong arguments against the application
of game theory to strategic decision-making. Exercises in game theory, he
says, deal in numerical probabilities, but these cannot be assigned to unique
events. Such exercises assume the unlimited ability of each party to think
and compute with no limit of time - which actual decision-makers cannot
do. The exercises assume that the goals of each party are single, simple, and
unchanging, whereas historical individuals and groups have objectives that
are plural, complex, and subject to constant revision. And so on.
This attack on the use of game theory is bewildering. As Donald Brennan
has pointed out in a review of Rapoport, the great majority of civilian strate-
gists do not use game theory and indeed would be at a loss to give any
account of i t . T h e r e are, certainly, a number of strategists, like Thomas
Schelling, who have mastered this technique, but in their work exercises in
game theory serve only to illustrate points that are independently arrived at;
they have not employed game theory in order to determine solutions
to strategic problems. As far as I know, the only person who has claimed
that game theory presents a method of solving strategic problems is Oskar
Morgenstern of Princeton University. Morgenstern collaborated with J o h n
von Neumann in producing Theory of Games and Economic Behavior and
has also written a book on strategy, The Question o f National Defense.' But
even in Morgenstern's book, which contains much rhetoric about the value
of game theory, it is not possible to find an instance in which he makes use
of it. I do not despair of finding an example of what Green and Rapoport
are talking about, but I must say 1 have not so far come across one. It may
be that although game theory is not an essential or even a significantly used
technique of the civilian strategists, some of the logic of game theory is
implicit in the way some strategists do their thinking, and a critique of the
former is a way of providing a critique of the latter.h The basic point of Philip
Green's book, that the technique and rigor that the civilian strategists have
brought to the subject do not provide a means of circumventing political
choices and that they can be and sometimes are employed as a political
weapon in support of one arbitrarily chosen policy or another, is undoubt-
edly correct. This, however, is an argument for recognizing the limits of rigor
and precision and for being on guard against their misuse, not for abandon-
ing rigor and precision in favor of something else.
Both in the domestic defense debates in Western countries today and in
international rivalries over arms control or the sharing of military burdens
within alliances, the strategist is constantly finding that his works are
pressed into the service of political objectives that are pursued on different
grounds. The army, the navy, the air force, each has its strategic ideology; the
United States, France, and Great Britain, in contending with one another as
to how nuclear weapons shall be controlled in NATO, as to where and in
what way a war in Europe would be fought, as to what contributions shall

be made to the shield forces in central Europe - each develops a strategic
doctrine that points to the end it has in view, and each is anxious to exploit
the authority of studies independently undertaken and scientifically followed
through.
The strategist himself, however, cannot be held responsible for the use that
others make of his ideas. Moreover, the fact that strategic expertise has come
to have a political function as an ideology is inevitable and, I believe, by no
means wholly regrettable. Scientific expertise has become the idiom of debate,
within governments and between them, not only in the strategic field but in
many others. If it is pressed into service by one party, the other parties must
acquire it themselves or go under. The British Foreign Office now finds it ne-
cessary to employ its own economic experts to d o battle with the Treasury and
the Department of Economic Affairs, its own scientific experts to deal with the
Ministry of Technology and the Atomic Energy Authority, and its own strate-
gic experts t o contend with the Ministry of Defence. The governments of
Western Europe in the last decade have found themselves constantly at a most
serious political disadvantage in relation to the United States in defense mat-
ters because they have not had a body of strategic expertise of their own with
which to frustrate American attempts to overawe them. That they will acquire
such expertise there can be no doubt.
These developments are not wholly to be regretted because they d o raise
the standard and tone of strategic debate at the highest levels of decision;
the necessity under which governments, and departments of governments,
labor of developing strategic ideologies does show that somewhere in the
process of decision, independent and expert studies are being carried out
and that these cannot be ignored.
Is it the case that the civilian strategists in America have been consistent
endorsers of the main lines of United States foreign policy and that they have
hidden this policy outlook beneath a pretense of objectivity? The work of the
most prominent of these persons originated in criticism and questioning of
the established policies of the Eisenhower Administration. N o doubt it still
has proceeded on the basis of assumptions held in common with official
thinking, which looked at from the outside appear as orthodox. In the writ-
ings of some of the strategists, more particularly those Rapoport calls the
"neo-traditionalists," the assumptions (e.g., about the existence of a "threat,"
the need for military strength, the morality of providing it, and so on) are
spelled out and defended. In the writings of others they are not, and in these
cases it is important that critics should identify the assumptions and ques-
tion them. But the shaping of United States military policy is not an exercise
in philosophy or theology; at some point firm assumptions have t o be made,
and on the basis of them the costs and benefits of alternatives worked out.
It is inevitable that in this process the assumptions will be taken as read and
also that books and papers will be written in which the authors address
themselves to others who make the same assumptions, rather than to the
public at large.
i
W Strategic Studies and Its Critics 115
The fifth criticism, although it also comes from Anatol Rapoport, is in

some ways at loggerheads with the fourth. It is that the sin of the strategist,
far from being his covert commitment to political purposes, is his objectivity.
This is really the distinctive contribution of Rapoport's book. The strategist
is detached and aloof, but he has no right to be. The effect of his cold
appraisal of the world as he sees it is to perpetuate the nightmare around him
or t o create it where it does not exist. Given the dangers of the world as it is
now, the appropriate attitude is not to describe it but to go to work on it. The
strategists, who have the ear of the powerful, might accomplish great things
if they abandoned the strategic mode of reasoning for the conscientious; but
instead they are collaborators in the system and are speeding up its movement
toward catastrophe.
If there is a kernel of truth in what Rapoport says it is that the strategist,
like all students of social affairs, is related to what he studies not only as sub-
ject to object but also as cause to effect. It is always important to recognize
in foreign policy, as in the conduct of Western policy toward Russia and
China now, that the intentions and goals of a country, whether they are
peaceful or aggressive, are not fixed and given, but are always in part the
product of our own action toward them.
But this basis of truth does not sustain the strange construction that
Rapoport erects upon it. There are certain things that Kussia will do whatever
policies the United States follows toward her, certain conflicts in the world that
simply have to be taken as given. Arthur Burns has pointed out in a review
that one of Rapoport's errors is to make the common American assumption
of a fundamentally two-person situation.' From the perspective of a small
country on the sidelines of the international arena, the Soviet-American con-
flict simply appears as a datum, something that has arisen quite independently
of anything that small country did or might have done. It is, to say the least,
greatly to exaggerate the influence of the strategists to hold them responsible
for the rise of Russian power and for the overflowing beyond Russian borders
of the revolution of 19 17.
But even if Rapoport is right and America, if not her strategists, has it
in her power to mold the behavior of Russia or China, this does not neces-
sarily support the conclusions that Rapoport would like to draw from this.
If United States policy in recent years has contributed to the changes that
have made the Soviet Union 3 more satisfied power and a more conserva-
tive influence in world affairs, this may have as much to d o with America's
strength and firmness as will her overtures of conciliation or readiness to
make concessions.
The doctrines that the civilian strategic analysts in the West have evolved
in the last decade are scarcely the last word on strategy in the nuclear age, but
should he seen as first, faltering steps in defining a problem that will be with
us for as far into the future as we can see. The three notions that have been
most central in these doctrines - "deterrence," "limited war," and "arms con-
trol" - have all been elaborated chiefly in the context of the Soviet-American
confrontation, and their implications for the more polycentric diplomatic field
that now exists have not been thought out. They have all been put forward in
relation to classical international conflict between states that are internally
stable and armed with the most advanced weapons, and they have not been
adapted to the different but now more prominent circumstances of civil con-
flicts within unstable states with primitive military equipment. Moreover, even
in the narrow field in which, quite rightly, the civilian strategists have concen-
trated their efforts, their most fundamental assumptions are open to challenge,
as the debate about ballistic missile defense is now showing.
Yet the work of the civilian strategists has at least charted some reasoned
course where otherwise there might well have been only drift. It has pro-
vided some solid intellectual fare that subsequent generations, even though
they reject it, are at least likely to recognize as a serious attempt to come to
grips with the problem. When one asks oneself what the history of strate-
gic policy in the West might have been in the last ten years had this influ-
ence not been brought to bear, or when one contemplates the moral and
intellectual poverty of the debate about nuclear affairs (or of that part of it
we are able to see) in the Soviet Union where in fact no such influence
exists, it is difficult to escape the conclusion that even though the civilian
strategists have sometimes committed the errors I have been exploring, they
have served us well.
Notes
1. See, e.g., James R. Newman, review of Herman Kahn's O n Thernionrrrlear Kjar,

Scientific American, cciv (March, 1961), 197-98; P.M.S. Blackett, Studies of War, Nuclear and
Conventional (New York 1962), chap. 10; Sir Solly Zuckerman, Scientists and War: The
Impact of Science on Military and Civil Affairs (New York, 1967), chap. 5; Irving 1.. Horo\virz.
The War Game: Studies of the Neiv Civilian Militarists (New York, 1963); Anatol Rapoport,
Strategy and Conscience (New York, 1964); Philip Green, Deadly I.ogic: The Theory of
Nuclear Deterrence (Columbus 1966).
2. O n War, Book 111, chap. 1.
3 . See, e.g., Max Teichmann, "Strategic Studies or Peace Research?" A r e n ~(Melbourne),
No. 1 2 (Autumn, 1967), 9-16.
4. Bulletin o f the Atomic Scientists, XXI (December, 1965), 25-30.
5. Theory of Games and Economic Behavior (Princeton, 1944); The Question of National
Defense (Princeton, 1959), esp. 61, 164, 269.
6. In his reply to Brennan's review, Rapoport says, "My complaint against the strategists
was not that they use or misuse game theory (although one of my earlier articles was so entitled).
On the contrary, my complaint was that they have not learned some important lessons o f game
theory" (Bulletrn of the Atomic Sc~entzsts,XXI [December, 19651, 31-36). This is a slippery
reformulation that does not answer Brennan's charge, viz., that Rapoport implies that strategists
use game theory, whereas they do not.
7. Arthur Lee Burns, "Must Strategy and Conscience Be Disjoined?" World Politics, XVII
(July, 19651, 687-702.
Arms Control and World Order
Hedley Bull
L I ~present theory and practice of arms control rests on a set of

assumptions - sometimes explicit, more often implicit - as to what
kind of world order is desirable and feasible. It is inevitable that
this should be so, for to raise questions about the quality and quantity of
arms that should exist in international society, about who sho~lldpossess
them, where they should be deployed, for what objectives and in what ways
they should be used, is to raise questions about the political structure of the
world and the distribution of power within it. But the set of assumptions
ahout world order which at present underlies the enterprise of arms control
c o m n ~ a n d slittle support outside the circle of the United States and the
Soviet Union and their closest allies. This is in itself sufficient reason for
raising the questions with which this essay is concerned, uir.
I . What assun~ptionsconcerning a desirable and feasible world order are

implicit in our present theory and practice of arms control?
2. What assumptions about world order should inform our approach to
arms control?
3. Given answers to the above questions, what consequences follow for
arms control policy?
Present Theory and Practice
By "our present theory and practice" I mean the body of theoretical writings
about arms control that arose in the West in the late 1950s and early 1960s
and the body of unilateral policies, tacit understandings and formal agree-
ments, chiefly involving the United States and the Soviet Union, that have
grown LIPabout arms control since that time. What we should notice about
this theory and practice is the extent to which it assumes or implies that
world order can and should be founded upon the present political structure
of the world and the existing distribution of power within it.
Source: Iwternntionnl S ~ c ~ m t 1y( ,1 ) ( 1976): 3- 16.

1 18 The Cold War a n d Nuclear Deterrence
First, there is the definition of arms control itself: "Arms control in its
broadest sense comprises all those acts of military policy in which antag-
onistic states cooperate in the pursuit of common purposes even while they
are struggling in the pursuit of conflicting ones."' When two antagonistic
states pursue common purposes in their military policy - as the United States
and the Soviet Union have sometimes done - these purposes may be univer-
sal ones, accepted as valid by international society as a whole, but they may
also be purely bilateral ones, the special purposes of the cooperating powers
themselves. "If two states," I wrote in 1964, "were to achieve their common
goals in this field by bringing about the ruin of other nations there would
seem no reason to deny that what they were engaged in was arms control,
except for the common but quite unnecessary assumption that arms control
has about it an aura of spiritual rectitude, instead of being a temporal
process like any otheran2
Our present definition of arms control does not in itself entail any bias
either for or against the present political structure of the world. But there is
a tendency in present-day thinking to regard cooperation between the
United States and the Soviet Union as the chief embodiment of arms control,
to see in the field of relations between these two powers both the principal
dangers with which arms control has to contend and the principal means of
coping with them. While (as I shall argue) Soviet-American cooperation in
arms control serves universal purposes it inevitably serves special or bilateral
purposes also. These special or bilateral purposes reflect the preference of
the two great powers for a world order in which they continue to enjoy a
privileged position.
Secondly, there are the objectives proclaimed for arms control. These are
taken to be ~ r i m a r i concerned
l~ with security: to make war, and especially
nuclear war, less likely, and to make it less catastrophic in terms of death
and destruction, if it should occur. A secondary objective is taken to be to
reduce the economic costs of military programs. A tertiary objective has
sometimes been added: the moral and social one of combating "the milita-
rization of society.""
N o doubt these objectives command a wide degree of support in inter-
national society, but the concrete meaning they have acquired serves to
rationalize the existing distribution of power. The list of objectives does not
include goals such as the promotion of just international and internal change,
which in the view of a large section of international society requires an assault
on the prevailing distribution of power, and should be pursued even at the
price of reduced security, an increased economic burden of armaments and a
greater "militarization of society." There is a tendency to confuse the national
security of the United States and the Soviet Union with international security,
the security of international society as a whole; it is the latter objective, not
the former, that should be the cardinal one in assessing arms control p o l i ~ i e s . ~
In some respects Soviet-American cooperation in arms control promotes the
national security of the United States and the Soviet Union at the expense of
the security of other states: it results, for example, in understandings about
Birll Arms Control a n d World Order 1 19
spheres of influence within which local states are left vulnerable to coercion
by one or the other of the great powers, in the redirection of conflict between
the great powers to "gray areas" in which wars are fought "by proxy," and
in attempts to deny third parties arms which they regard as necessary for their
security. Where Soviet-American cooperation evidently promotes inter-
national security and not merely the national security of the great powers -
for example, contributing to the avoidance of global nuclear war - it does so
in ways that leave the existing political structure of the world intact.
Thirdly, there is the notion that the chief proximate goal of arms control
is to stabilize the relationship of mutual nuclear deterrence between the
United States and the Soviet Union. This is a notion that is basic to most con-
temporary thinking about arms control, whether we envisage the stabilization
of mutual nuclear deterrence as achieved through "high" Soviet and United
States arms levels or "low," as implying acceptance of "mutual assured
destruction" or not implying it, as accompanied by "parity" or by some form
of "superiority" for one side, as bound up with some political program for
the promotion of "detente" or as independent of it, as a "first step" towards
nuclear disarmament or as a goal sufficient in itself. Whatever the merits or
demerits of this goal might be, we should recognize that the attempt to stabil-
ize the relationship of mutual nuclear deterrence between the United States
and the Soviet Union, while it is logically consistent with a variety of patterns
of power international politics, confines our attention to measures which
allow the two superpowers to retain arms levels sufficient for mutual nuclear
deterrence, and excludes measures (such as nuclear disarmament, or general
and complete disarmament) which carry the risk of radical change in the dis-
tribution of power as between the superpowers and the rest. Moreover, in
practice Soviet-American cooperation in this field has been accompanied by
the attempt to legitimize very high ceilings of strategic arms, by political
cooperation directed against third parties and by enunciation of a principle of
parity whose effect is to formalize the claims of these two states to a special
position in the hierarchy of military power.
Fourthly, there is the idea that it should he a proximate goal of arms con-
trol to stop or to contain the ,geographical or horizontal diffusion of military
power. This is an idea whose most notable expression is the attempt to com-
bat nuclear proliferation, but it may be seen also in the concern that has been
expressed about the proliferation of chemical and biological weapons and of
conventional arms, especially through the arms trade. Whatever validity
there may be in the argument that international security is endangered by the
spread of nuclear weapons, chemical and biological weapons and conven-
tional arms among a wider circle of states, it is an argument which serves to
rationalize the existing distribution of power. When we decide that it is the
horizontal spread of nuclear weapons rather than their vertical spread that
calls for urgent preventative action, that biological weapons should he fore-
gone by the rich powers because they are "the poor man's atom bomb," or
that there should be measures to stop poor countries from buying conven-
tional arms from rich countries, but not measures to stop rich countries from
producing these arms for themselves, we are choosing arms control arrange-
ments which leave those countries which now possess preponderant military
power secure in the enjoyment of their position.
Fifthly, there is the idea that the United States and the Soviet Union, so
as to minimize the risk of general nuclear war, should observe a series of
tacit rules for the avoidance and control of crises arising out of their con-
flicting objectives in many parts of the world. This is the idea that lies
behind the attempts of the two great powers to restrain allies and clients
that might embroil them on opposite sides in local conflicts, to disengage
from such allies and clients in cases where they cannot be adequately
restrained, to demarcate spheres of influence in which each can intervene
without fear of counter-intervention by the other. Whatever merits we may
find in this idea, we should recognize that what it implies in practice is
the maintenance of a political structure in which the two great powers
cooperate to frustrate the objectives of others: of allied states which seek
to divest themselves of great power restraints, of client states which seek to
engage great power support for their private goals, of fettered states which
seek t o break free of the spheres of influence to which they have been
assigned, and of aspirant great powers which seek to stake out new spheres
of influence of their own.
In the Western countries at present there is some disposition to question
established theories and policies about arms control. Thus there has been
much discussion of the relationship between SALT and political detente, of
the feasibility of establishing Soviet-American strategic parity by agreement,
of the implications of studies of "bureaucratic politics" for the theory of
arms races, of the bearing on arms control of the cruise missile, and of the
relative merits of negotiation and unilateral action as means of advancing
the objectives of arms control."ut there has been little discussion of the
question whether the assumptions about world order that are so central to
our present approach to arms control, and are so decisively rejected by
China and the aspirant powers of the third world, are valid."
An Approach to World Order
If our present theory and practice of arms control proceed on the assumption
that world order can and should be founded on the present political structure
of the world and distribution of power within it we should not leap to the
conclusion that it is an undefendable assumption. Our present theory and
practice have arisen in response to the perception of overwhelming dangers
of nuclear war between the two great powers, and of a need to curb these
dangers as a matter of urgent priority. Those who tell us that some different
order of priorities should have been followed must ask themselves whether
an approach to world order which did not begin with the attempt to find
common ground between the two powers with the capacity to destroy the
world as we know it was ever a possible or an honorable one.
titiil Arms Control and World Order 12 1
If the search for common ground between the United States and the Soviet
Union has resulted in arrangements which confirm their privileged position
in the hierarchy of power we may also point out that the dangers of nuclear
war between them threaten not only the two great powers but international
society as a whole; that these arrangements have in fact served to reduce the
dangers, however inadequately and imperfectly; and that it is not immedi-
ately obvious how the United States and the Soviet Union could have drawn
together in arms control arrangements without also involving themselves in
political cooperation against third parties. If the international order con-
firmed by our present arms control arrangements is one in which certain
powers claim special privileges and responsibilities the question may be asked
whether any international order has ever existed in which this was not so;
and whether, if the United States and the Soviet Union were to forego their
claims to a special position so as to make room for others, there is any rea-
son to suppose that these others would be more willing or able to take on the
responsibilities of such a privileged position than they have been in the past.
But the vision of world order that is projected by our present arms con-
trol arrangements is one against which the majority of states are in revolt.
It is true that among countries such as China, India, Iran, Indonesia, Egypt,
Libya, Saudi Arabia, Brazil, Argentina, Nigeria, one finds different degrees
of opposition to these arrangements, deep mutual divisions without any
agreement about an alternative conception of world order. However, they
all see the emphasis on Soviet-American bilateral goals - in arms control, in
the treatment of security as the commanding value, in the preoccupation
with stabilization of the great power balance, in the efforts to control pro-
liferation, and in the network of tacit understandings between the great
powers - as part of a system of hegemony which they wish to break down
in spite of the fact they have nothing in mind with which to replace it.
It is sometimes argued that the dissent of China and the third world
states from the existing international order need not be fatal to it; that these
states are too weak and divided among themselves to provide any serious
challenge; that particular recalcitrant powers among them c a n be bought
off with favors conferred by one or another of the great powers or their
allies - as even India's opposition to the existing order has been moderated
by its dependence on the Soviet Union, and China's by its sense of strategic
interests shared with the United States. But the shift in the distribution of
power toward the countries of the third world has already begun, and its
impact has already been dramatic. China is a nuclear power, India has con-
ducted a nuclear explosion and half a dozen more third world states have
the potential to develop a nuclear capability. The oil-producing countries
have not only brought about a shift of wealth in their favor and a global
realignment on the Arab-Israeli dispute; more importantly, perhaps, they
have demonstrated by their example what dividends are to be reaped by
overcoming internecine disputes in the interests of a united front, and by
abandoning a position of conciliation for one of confrontation. At the present
time the levers of power available even to the strongest of third world countries
are only such as are conferred by rudimentary military power, the prestige
of numbers, the possession of raw materials which others need, and the
appeal of ideology. In the long run we must expect that some of them will
also have a t their disposal the levers conferred by advanced technology and
high industrial capacity, which will not remain permanently the monopoly
of the so-called northern countries.
The idea of the disunity of the third world, that "there is no such thing
as the third world," overlooks the fact that on certain basic issues this ram-
shackle coalition of states has held together to a remarkable degree, and
that in the last 30 years (as relating to the legitimacy of colonialist and
white supremacist governments, the legality of "wars of national libera-
tion," the duty to transfer wealth from rich countries to poor, the right to
expropriate foreign assets, the law of the sea) it has drastically changed the
prevailing norms in international society. In much of this endeavor the third
world countries have been powerfully supported by the Soviet Union. This
should remind us that the Soviet Union's own commitment to the existing-
international order is a half-hearted and perhaps merely tactical one; that
the Soviet Union, while on some issues (nuclear proliferation, the law of the
sea) it stands arrayed with the "North" against the "South", on others ("wars
of national liberation" against colonialist and white supremacist govern-
ments) it is the ally of the third world against the Western powers.
The conception of world order implicit in our present approach to arms
control is based too narrowly upon the elements of consensus between the
United States and the Soviet Union and their allies. At the same time this con-
ception of world order embodies too narrow a range of goals: it treats the
security of the two great powers as prior to that of international society as a
whole, and it fails to recognize goals of just international and internal change.
What conception of world order, then, should we put in its place? Not,
I think, one which proclaims that a viable world order can be constructed
only if we move "beyond the states system." There is no convincing evi-
dence that the system of states is in decline and about to give place to some
different form of universal political organization; nor, I believe, should we
be impressed by the argument that the states system has become "obsolete"
in the sense of being dysfunctional in relation to basic goals such as peace
and security, eocnomic and social justice and the control of the human envir-
onment. It is only in the Western world that the cry is heard that we should
transcend the states system; the socialist and third world countries clearly
seek to work within its framework. The problem of world order is not that
of how to move beyond the states system, but that of how to make it work.'
Making the states system work must involve the attempt to preserve and
ultimately t o extend the element of consensus among states about common
interests, common rules and common institutions - the consensus whose
existence in the past has entitled us to say that states form not only an inter-
national system but also an international society. It is sometimes argued that
as a consequence of the technological unification of the world, or the growth
of economic "interdependence" or the multiplication of transnational social
KIIII Arms Control and World Order 123
ties consensus in our times has been growing. Thus international lawyers tell
us with satisfaction that it is now widely agreed that rights and duties in
international law are enjoyed not only by states but by individual human
beings, that the scope of international law has widened beyond a rudimen-
tary "international law of coexistence" to become an "international law of
cooperation," and that the source of international law is no longer the con-
sent of states but the consensus or general will of the international commu-
nity as a whole. But there are strong grounds for thinking that at the global
or universal level consensus about the basic framework of orderly inter-
national coexistence has not been growing but shrinking - as the consequence
of ideological divisions, the revolt of subject peoples and the geographical
expansion of the states system beyond its originally European confines.
Making the states system work is a matter of preserving and nurturing what
remains of a rudimentary consensus about "minimum order," not of advan-
cing towards some "optimum order" about which, at the global level, no
consensus exists o r is in prospect.
The consensus which has to be nurtured is not one simply between the
United States and the Soviet Union, or these two plus China, or a wider
consort that would include Japan and Western Europe if and when they
demonstrate that they are great powers. N o consensus is likely to be ade-
quate for a viable states system that does not embrace the countries which
now form the third world. These countries represent a majority of states
and of the world's population and are too large and potentially powerful a
segment of international society to be assigned the status merely of an
object or series of objects of its governing rules.
It follows from this that our conception of world order should not be
shaped by prescriptions for a more centralized system, expressed in an
expanding United Nations or upon "non-territorial centralized direction.""
The third world countries are opposed to centralizing tendencies in world
politics, perceiving correctly that if more powerful centralized institutions
were to be established now, they would rob ably be controlled by the pre-
sent great powers and would reflect their special interests. It is more likely
that world order will continue to rest upon a decentralized system, and that
if a greater role is to be played by international institutions these will be
regional rather than global ones. The question of the establishment of more
powerful institutions at the global level remains, as it were, on the agenda
of world politics: but it is not likely to be discussed seriously by the third
world or indeed by the socialist states until there has first been a shift of
power in their favor.
Whatever notions we entertain about a desirable and feasible world
order, they have to take account the third world's alienation from the
arrangements for world order that exist now. This alienation is rooted not
merely in grievances about colonialism or racism or the distribution of wealth,
consumer goods or technology: it is rooted also in the third world's lack of
power, including military power - its sense of impotence and vulnerability in
relation to the Western countries and the Soviet Union. It may be one of the
conditions of a more viable world order that the shift of power towards the
third world, whose beginnings we noted above, should first be expanded.
It is true that the third world countries at present are preoccupied not with
order but with change; that as they become more powerful they are likely to
create disorder in the course of bringing this change about; and the present
frame work of world order, reflecting as it does the preferred values of the
West and of the Soviet Union, will not be strengthened but is bound to be
weakened by a shift in the distribution of military power towards the third
world countries. But the objection of the third world countries is not to the
quality of order in the present international arrangements; it is rather to the
way in which these arrangements discriminate against them. Once the changes
they are seeking have been effected, and new arrangements have replaced the
old ones, it is possible that they will come to sense a stake in them.
The task of carrying out this redistribution of power must fall to the
third world countries themselves: it is too much to expect that the great
powers and their allies will be willing to carry out this task for them, or
even that they could if they wished to d o so. It is for the third world coun-
tries t o mobilize their resources, to combine with one another and to chal-
lenge the elements of discrimination in the present system. But the Western
powers and the Soviet Union should recognize that such a challenge is nat-
ural and inevitable. They should also recognize that while some of the pre-
sent perceived interests will be injured in the process, they themselves have
a stake in the emergence of a world order of which the majority of states
and of the world's population feel themselves to be a part.
There are difficulties and risks in this approach. It is true that as the major
countries of the third world acquire more military power they are likely to
seek to exploit it in relation to one another and not merely in relation to the
present great powers. It is true that there is no agreement among the third
world states as to which of them are to be the beneficiaries of the process of
redistribution. It is true that world order requires that we attempt to limit and
contain military power; the creation of new centers of military power is bound
to make that attempt more difficult.
C o n s e q u e n c e s f o r A r m s Control
I should not argue that our approach to arms control should be determined
in detail by some precise vision of a desirable or feasible world order. To do
this would be to treat visions of world order more seriously than they deserve;
we cannot be sure enough about the desirability or feasibility of any one of
them to regard it as the legislator of policy here and now. It does appear to
me, however, that we should try to sever the close connection which now
exists between the theory and practice of arms control and attempts to pre-
serve the existing distribution of power. Not only does this connection make
arms control an obstacle to changes' which may be necessary if a more viable
world order is to be evolved, but it also serves to discredit arms control, and
I Arms Control and World Order 125
to obscure the role which it has to play in promoting the purposes of those
seeking to challenge the present distribution of power.
First, we should maintain that the cooperation involved in arms control
should promote universal purposes and not merely bilateral ones. Cooper-
zation that promotes special or bilateral purposes may count as arms con-
trol, but we should be clear that the proper purpose of arms control is to
advance objectives endorsed by international society as a whole. In the case
of United States-Soviet cooperation we should distinguish between the pur-
pose of avoiding nuclear war, which is generally endorsed, and the purpose
of preserving United States and Soviet ascendancy, which is not. The meas-
ures that lead to the one are not always easily distinguishable from those
that lead to the other, and it would be naive to suppose that the two great
powers could pursue the former while wholly abstaining from the latter. But
at least we can distinguish these purposes in our minds, and consider how
far in practice the political structure of American and Soviet ascendancy is
essential to the political structure of peace.
Secondly, while recognizing that security against war, and especially
nuclear war, is the prime goal of arms control, we should distinguish
between the national security of the United States and the Soviet Union, and
the security of international society as a whole; and we should insist that it
is the latter that is the overriding test. It is clear that the national security of
the two great powers is served by arrangements, like the tacit understanding
about spheres of influence, that has left Eastern Europe and Central America
vulnerable to great power intervention; by the export of conflicts to
Southeast Asia or the Persian Gulf or Africa; or by their opposition to
nuclear proliferation in China and India and elsewhere. But it is also clear
that these arrangements do not promote the national security of the other
countries concerned. Whether the security of international society as a whole
is more helped than hindered by these arrangements is a moot point. But
even if we were to conclude that it is more helped than hindered by them we
should still need to distinguish between the security of international society
as a whole and that of its two most powerful members, and we should still
need to insist that there may be other avenues to the former besides those
that have been ordained by the United States and the Soviet Union.
At the same time we should take account of the fact that the goal
of security comes into conflict with that of the promotion of international
and internal change very widely regarded as just. I d o not mean that arms
control arrangements should be designed so as to promote black liberation
in southern Africa, or revolution in Chile or the establishment of a state of
Palestine. It is better to recognize that arms control is concerned chiefly
with only one dimension of world order, viz. peace and security, and more
particularly with its military aspects, than to saddle it with responsibility
for every dimension. But we need to be aware, in pursuing arms control, of
the existence of other dimensions of world order, and particularly of the
role of military force in effecting changes that express emerging principles
of international legitimacy.
Thirdly, we should note the extent to which the process of negotiation

about the great power strategic balance promotes objectives that are bilat-
eral rather than universal. In the short run the attempt to stabilize the
Soviet-American relationship of mutual nuclear deterrence is necessary for
the security of international society as a whole. That this relationship is
more or less stable, and that the two great powers are formally committed
to a process of negotiation aimed at making it more stable at lower levels
of strategic arms, is the chief concrete achievement of arms control and the
chief basis for whatever claim the United States and the Soviet Union can
make to being trustees for mankind.
But even if there were signs that the strategic arms limitation talks were
leading in the direction of a more stable relationship of mutual nuclear
deterrence and of lower levels of strategic arms this goal would be an insuf-
ficient one in relation to the wider considerations of world order sketched
out above: it does not take account of the demand for a wider distribution
of military nuclear power. In fact of course, with the possible exception of
the 1972 ABM Treaty the strategic arms limitation talks have not increased
the stability of the strategic relationship, nor can it very well be argued that
they have reduced the level of strategic arms. What they have done is to ratify
a principle of parity or equality which, however necessary it may be as a
benchmark in discussion, does not express any known goal of arms control;
to contribute to a wider process of political cooperation between the great
powers, now under threat; and to provide a means of rationalizing the reten-
tion of the present high levels of armaments by pointing to ongoing negoti-
ations aimed at their reduction. Whatever may be said for these achievements
they do not offer much to those who threaten horizontal proliferation in
return for vertical.
Fourthly, in our attitude towards the control of the horizontal diffusion
of military power we should seek to distinguish between the universal and
the special interests which it serves, and to take account of its bearing on
the wider issue of world order. What is at stake in the attempt to control
horizontal nuclear proliferation is a universal interest in the security of
international society as a whole, which is likely to be jeopardized as the
number of nuclear weapon states increases, and even more than jeopardized
if proliferation proceeds beyond the sovereign state and nuclear weapons
become the instrument of non-state g r o ~ p sThis . ~ proposition, indeed, is
not denied even by those states which, like China, France and India, are
opposed to the Non-Proliferation Treaty: even if, as in the case of China,
their statements sometimes imply that the nuclear club should be expanded
to include not only themselves but others as well, they do not seek to make
out a case in favor of general or universal proliferation. The argument
between the supporters and the opponents of the Non-Proliferation Treaty
is not about whether or not nuclear proliferation endangers international
security, but about where to draw the line. Nor is our proposition in any
way undermined by the fact that there has been some proliferation or the
likelihood that there will be some more.
Bull Arms Control and World Order 127
But we must recognize that the control of horizontal proliferation can-

not he separated from the control of vertical. If among the potential
nuclear-weapon states the idea is to be combated that nuclear weapons are
effective instruments of policy and sources of prestige this can only be the
consequence of a demonstration by the actual nuclear-weapon states that
they themselves are ceasing to regard nuclear weapons in this way. So long
as they retain their nuclear weapons and go on developing them then no
matter what statements they put out they will not be able to make such a
demonstration. The nuclear-weapon states have, however, had some success
in the sense that if they had not practiced the measure of restraint they have
in fact observed (if, for example, they had actually used nuclear weapons,
or explicitly threatened their use habitually instead of occasionally) hori-
zontal proliferation is likely to have proceeded at a faster rate.
We should also recognize that the attempt to draw the line between
nuclear weapon and non-nuclear-weapon states at the point at which it lies
now, in the absence of any convincing demonstration that nuclear weapons
are of diminishing political and strategic utility, must confirm the privileged
position of the present nuclear-weapon states, and that this flies in the face
of the need on wider grounds for a more even distribution of power.
It is true that once we admit the legitimacy of the demand for a more
"just" or "even" distribution of military nuclear power we are on danger-
ous ground. We do not escape from the charge that we are upholding the
special privileges of a particular group of states merely by substituting
another line of division for the present one; a nuclear club of 10 or 20 or
50 states will be no less vulnerable to the charge that it is discriminatory
than is the present nuclear club of five or six. Perfect justice in this sense can
be achieved only by general nuclear disarmament or by universal nuclear
proliferation; and to spell this out is to provide a reductio ad ahsurdurn of
the idea that justice in relation to the possession of nuclear weapons is iden-
tical with equality in the distribution of them to all states, neglecting quali-
tative differences among states.
But whatever the requirements of "justice," the requirements of a world
order should lead us to doubt whether the present line of division is a viable
one. World order would best be served by efforts to promote - immediately
through steps such as a reduction in great power strategic arms, the adoption
of no first use positions by the nuclear powers and their acceptance of a
comprehensive nuclear test ban - the idea of the diminishing political and
strategic utility of nuclear weapons. But while such efforts are not made, the
present line of division not only will but should be challenged in the inter-
ests of reducing the present dominance of the advanced industrial states.
In the case of conventional arms the process of diffusion, at least of the
more sophisticated kinds of arms (tanks, destroyers, fighter aircraft, elec-
tronic equipment) is the consequence of supply by the advanced countries -
not, as in the case of nuclear weapons, of the development of indigenous
capacities. The policy of the advanced countries is, of course, to supply
sophisticated conventional arms to non-producing countries either so as to
create relationships of dependence that can be manipulated for political

purposes, or for the sake of economic gain. It is, however, often argued, at
least in the Western countries, that it would be desirable to restrict the sup-
ply of these arms to non-producing countries if these counter-vailing consid-
erations could be overcome, and if general agreement among the supplying
countries could be reached.
Here, broadly similar considerations apply as those we noted in the case
of the diffusion of nuclear weapons. The horizontal diffusion of conven-
tional arms is, no doubt, injurious to international security. But the interest
of the recipient countries in acquiring these arms for national defense, as
symbols of national identity, t o appease the demands of their armed forces
or for other reasons, will scarcely abate while the donor countries remain
free to develop their own arms without restriction. For so long as there exist
gross disparities between the advanced industrial countries and the rest of
international society in terms of their capacity t o produce sophisticated con-
ventional arms it does not seem likely that attempts to restrict the inter-
national trade in arms, while not restricting the right of countries to produce
arms, will prove feasible. Nor, in terms of the present arguments would such
attempts advance the prospects of world order, even if they were feasible.
Fifthly, we need to distinguish the universal from the merely special
interests that are served by the tacit rules evolved by the United States and
the Soviet Union for the avoidance and control of crises. There is no doubt
that the universal interest in averting a Soviet-American nuclear war has
been served by the evolution of this system of tacit rules, whatever the cost
in terms of the suppression of other states and peoples. The maintenance of
order in international society has always required that the security of the
whole of international society be treated prior to the security of its parts,
the maintenance of the general balance of power prior to the maintenance
of local balances, the avoidance of war between the great powers prior to
the avoidance of other kinds of wars. For so long as international society
continues to contain a hierarchy of military power, the present one or some
other, it does not seem likely that order can be preserved without rules that
reflect these priorities.
But we should beware of assuming that whenever an ally of one of the
great powers seeks to break free of restraints, whenever a client government
seeks t o dissuade its great power patron from deserting it, or whenever a
country within the sphere of influence of a great power seeks to change its
internal political character or to alter the direction of its foreign policy, it is
the peace of the world that is at stake rather than merely the political con-
venience of the United States, or the Soviet Union.
Notes
1. Hedley Bull, "Introduction t o the Second Edition," The Control of the Arms Race
( N e w York: Praeger, 1965), p. xiv.
2. Bull, The Control of the Arms Race, p. xxxv.
ic I Arms Control and World Order 1 29
3. Bull, The Control of the Arms Kore, p. 3-4.

4. Bull, The Control of t l ~ eAnns Roc-c, p. 28-29.
5. Sec, for example, "Arms, I k f e n s e Policy and Arms Control", I)izedirlrts. 104, 3 ,
Sumnier 1975.
6. An exception is Professor Richard A. Falk who, while in other respects he has fallen into
error, correctly perceives that "Arms control measures have served mainly t o rarity the bipolr
dominance of international polit~csa n d t o maximize the stahiliry o f t h ~ sdomin,lnce from 1'
managerial standpoint." See h ~ s"Arms Control, Global Policy a n d Global Reform",
D~zehnlzls,104, 3 S ~ ~ m m 1975, er p. 40.
7. 1 h.lve spelt o u t this argument in "Models of Future World Order", Indiiz Qztizrterly,
January-March 1975; a n d more full! in Thr Anizrchrcrzl Society. A Study of Orrier in World
l'olitrcs. (Macmillan, f o r r h c o m ~ n g ) .
8. A mystifying phrase of I'rofessor Falk's.
9 . T h ~ sp r o p o s i t ~ o nis n o t self-ev~dent,hut I have argued it in T / J Control
~ of the Arms
Krrce, o p cit.; a n d In "Rethinking Non-l'roliferation", Intcrnotionirl Affirm, Vol. 5 1, No. 2,
April 1975, pp. 175-1 89.
Cooperation Under the Security Dilemma
Robert Jervis
I. Anarchy and the Security Dilemma
T he lack of an international sovereign not only permits wars to occur,

but also makes it difficult for states that are satisfied with the status
quo to arrive at goals that they recognize as being in their common
interest. Because there are no institutions or authorities that can make and
enforce international laws, the policies of cooperation that will bring
mutual rewards if others cooperate may bring disaster if they do not.
Because states are aware of this, anarchy encourages behavior that leaves all
concerned worse off than they could be, even in the extreme case in which
all states would like to freeze the status quo. This is true of the men in
Rousseau's "Stag Hunt." If they cooperate to trap the stag, they will all eat
well. But if one person defects to chase a rabbit - which he likes less than
stag - none of the others will get anything. Thus, all actors have the same
preference order, and there is a solution that gives each his first choice: (1)
cooperate and trap the stag (the international analogue being cooperation
and disarmament); (2) chase a rabbit while others remain at their posts
(maintain a high level of arms while others are disarmed); ( 3 ) all chase rab-
bits (arms competition and high risk of war); and (4) stay at the original
position while another chases a rabbit (being disarmed while others are
armed).l Unless each person thinks that the others will cooperate, he him-
self will not. And why might he fear that any other person would do some-
thing that would sacrifice his own first choice? The other might not
understand the situation, or might not be able to control his impulses if he
saw a rabbit, or might fear that some other member of the group is unreli-
able. If the person voices any of these suspicions, others are more likely to
fear that he will defect, thus making them more likely to defect, thus mak-
ing it more rational for him to defect. Of course in this simple case - and in
many that are more realistic - there are a number of arrangements that
Source: World Politics, 30(2) (1978):167-214.
lerv15 Security Dilemma 13 1
could permit cooperation. But the main point remains: although actors may
know that they seek a common goal, they may not be able to reach it.
Even when there is a solution that is everyone's first choice, the inter-
national case is characterized by three difficulties not present in the Stag
Hunt. First, to the incentives to defect given above must be added the potent
fear that even if the other state now supports the staus quo, it may become
dissatisfied later. N o matter how much decision makers are committed to
the status quo, they cannot bind themselves and their successors to the same
path. Minds can be changed, new leaders can come to power, values can
shift, new opportunities and dangers can arise.
The second problem arises from a possible solution. In order to protect
their possessions, states often seek to control resources or land outside their
own territory. Countries that are not self-sufficient must try to assure that
the necessary supplies will continue to flow in wartime. This was part of the
explanation for Japan's drive into China and Southeast Asia before World
War 11. If there were an international authority that could guarantee access,
this motive for control would disappear. But since there is not, even a state that
would prefer the status quo to increasing its area of control may pursue the
latter policy.
When there are believed to be tight linkages between domestic and for-
eign policy or between the domestic politics of two states, the quest for
security may drive states to interfere pre-emptively in the domestic politics
of others in order to provide an ideological buffer zone. Thus, Metternich's
justification for supervising the politics of the Italian states has been sum-
marized as follows:
Every state is absolutely sovereign in its internal affairs. But this implies
that every state must d o nothing to interfere in the internal affairs of any
other. However, any false or pernicious step taken by any state in its
internal affairs may disturb the repose of another state, and this conse-
quent disturbance of another state's repose constitutes an interference in
that state's internal affairs. Therefore, every state - or rather, every sov-
ereign of 3 great power - has the duty, in the name of the sacred right of
independence of every state, to supervise the governments of smaller
states and to prevent them from taking false and pernicious steps in their
internal affairs."
More frequently, the concern is with direct attack. In order to protect

then~selves,states seek to control, or at least to neutralize, areas on their bor-
ders. But attempts to establish buffer zones can alarm others who have
stakes there, who fear that ind desirable precedents will be set, or who believe
that their own vulnerability will be increased. When buffers are sought in
areas empty of great powers, expansion tends to feed on itself in order to
protect what is acquired, as was often noted by those who opposed colonial
expansion. Balfour's complaint was typical: "Every time I come to a discus-
sion - at intervals of, say, five years - I find there is a new sphere which we
have got to guard, which is supposed to protect the gateways of India. Those
gateways are getting further and further away from India, and I do not know
how far west they are going to be brought by the General Staff.""
Though this process is most clearly visible when it involves territorial
expansion, it often operates with the increase of less tangible power and influ-
ence. The expansion of power usually brings with it an expansion of respon-
sibilities and commitments; to meet them, still greater power is required. The
state will take many positions that are subject to challenge. It will be involved
with a wide range of controversial issues unrelated to its core values. And
retreats that would be seen as normal if made by a small power would be
taken as an index of weakness inviting predation if made by a large one.
The third problem present in international politics but not in the Stag
Hunt is the security dilemma: many of the means by which a state tries to
increase its security decrease the security of others. In domestic society, there
are several ways to increase the safety of one's person and property without
endangering others. One can move to a safer neighborhood, put bars on the
windows, avoid dark streets, and keep a distance from suspicious-looking
characters. Of course these measures are not convenient, cheap, or certain of
success. But no one save criminals need be alarmed if a person takes them.
In international politics, however, one state's gain in security often inadvert-
ently threatens others. In explaining British policy on naval disarmament in
the interwar period to the Japanese, Ramsey MacDonald said that "Nobody
wanted Japan to be i n ~ e c u r e . "But
~ the problem was not with British desires,
but with the consequences of her policy. In earlier periods, too, Britain had
needed a navy large enough to keep the shipping lanes open. But such a navy
could not avoid being a menace to any other state with a coast that could be
raided, trade that could be interdicted, or colonies that could be isolated.
When Germany started building a powerful navy before World War I,
Britain objected that it could only be an offensive weapon aimed at her. As
Sir Edward Grey, the Foreign Secretary, put it to King Edward VII: "If the
German Fleet ever becomes superior to ours, the German Army can conquer
this country. There is no corresponding risk of this kind to Germany;
for however superior our Fleet was, no naval victory could bring us any
nearer to Berlin." The English position was half correct: Germany's
navy was an anti-British instrument. But the British often overlooked what
the Germans knew full well: "in every quarrel with England, German
colonies and trade were ... hostages for England to take." Thus, whether she
intended it or not, the British Navy constituted an important instrument of
coer~ion.~
11. What M a k e s Cooperation M o r e Likely?
Given this gloomy picture, the obvious question is, why are we not all
dead? Or, to put it less starkly, what kinds of variables ameliorate the
impact of anarchy and the security dilemma? The workings of several can
It 1 c Security Dilemma 133
be seen in terms of the Stag Hunt or repeated plays of the Prisoner's

Dilemma. The Prisoner's Dilemma differs from the Stag Hunt in that there is
no solution that is in the best interests of all the participants; there are
offensive as well as defensive incentives to defect from the coalition with the
others; and, if the game is to be played only once, the only rational response is
to defect. But if the game is repeated indefinitely, the latter characteristic no
longer holds and we can analyze the game in terms similar to those applied
to the Stag Hunt. It would be in the interest of each actor to have others
deprived f; the power to defect; each would be willing to sacrifice this abil-
ity if others were similarly restrained. But if the others are not, then it is in
the actor's interest to retain the power to d e f e ~ tThe
. ~ game theory matrices
for these two situations are given below, with the numbers in the boxes
being the order of the actors' preferences.
STAG HUNT PRISONER'S DILEMMA
COOPERATE DEFECT COOPERATE DEFECT
1 COOPERATE
3
4
DEFECT
R
We can see the logical possibilities by rephrasing our question: "Given

either of the above situations, what makes it more or less likely that the play-
ers will cooperate and arrive at CC?" The chances of achieving this outcome
will be increased by: ( I ) anything that increases incentives to cooperate by
increasing the gains of mutual cooperation (CC) and/or decreasing the costs
the actor will pay if he cooperates and the other does not (CD); (2) anything
that decreases the incentives for defecting by decreasing the gains of taking
advantage of the other (DC) and/or increasing the costs of mutual noncooper-
ation (DD); ( 3 ) anything that increases each side's expectation that the other
will cooperate.'
The fear of being exploited (that is, the cost of CD) most strongly drives the
security dilemma; one of the main reasons why international life is not more
nasty, brutish, and short is that states are not as vulnerable as men are in a
state of nature. People are easy to kill, but as Adam Smith replied to a friend
who feared that the Napoleonic Wars would ruin England, "Sir, there is a
great deal of ruin in a nation.""he easier it is to destroy a state, the greater
the reason for it either to join a larger and more secure unit, or else to be
especially suspicious of others, to require a large army, and, if conditions
arc favorable, to attack at the slightest provocation rather than wait to be
attacked. If the failure t o eat that day - be it venison or rabbit - means that
he will starve, a person is likely to defect in the Stag Hunt even if he really
likes venison and has a high level of trust in his colleagues. (Defection is
especially likely if the others are also starving or if they know that he is.) By
contrast, if the costs of CD are lower, if people are well-fed or states are
resilient, they can afford to take a more relaxed view of threats.
A relatively low cost of CD has the effect of transforming the game from
one in which both players make their choices simultaneously to one in which
an actor can make his choice after the other has moved. He will not have to
defect out of fear that the other will, but can wait to see what the other will
do. States that can afford to be cheated in a bargain or that cannot be
destroyed by a surprise attack can more easily trust others and need not act at
the first, and ambiguous, sign of menace. Because they have a margin of time
and error, they need not match, or more than match, any others' arms in
peacetime. They can mobilize in the prewar period or even at the start of the
war itself, and still survive. For example, those who opposed a crash program
to develop the H-bomb felt that the U.S. margin of safety was large enough so
that even if Russia managed to gain a lead in the race, America would not be
endangered. The program's advocates disagreed: "If we let the Russians get
the super first, catastrophe becomes all but certain."'
When the costs of CD are tolerable, not only is security easier to attain
but, what is even more imporant here, the relatively low level of arms and
relatively passive foreign policy that a status-quo power will be able to
adopt are less likely to threaten others. Thus it is easier for status-quo states
to act on their common interests if they are hard to conquer. All other
things being equal, a world of small states will feel the effects of anarchy
much more than a world of large ones. Defensible borders, large size, and
protection against sudden attack not only aid the state, but facilitate coop-
eration that can benefit all states.
Of course, if one state gains invulnerability by being more powerful than
most others, the ~ r o b l e mwill remain because its security provides a base
from which it can exploit others. When the price a state will pay for DD is
low, it leaves others with few hostages for its good behavior. Others who are
more vulnerable will grow apprehensive, which will lead them to acquire
more arms and will reduce the chances of cooperation. The best situation is
one in which a state will not suffer greatly if others exploit it, for example,
by cheating on an arms control agreement (that is, the costs of CD are low);
but it will pay a high long-run price if cooperation with the others breaks
down - for example, if agreements cease functioning or if there is a long war
(that is, the costs of DD are high). The state's invulnerability is then mostly
passive; it provides some protection, but it cannot be used to menace others.
As we will discuss below, this situation is approximated when it is easier for
states to defend themselves than to attack others, or when mutual deterrence
obtains because neither side can protect itself.
The differences between highly vulnerable and less vulnerable states
are illustrated by the contrasting policies of Britain and Austria after the
l e r ~ ~Security
s Dilemma 135
Napoleonic Wars. Britain's geographic isolation and political stability allowed

her to take a fairly relaxed view of disturbances on the Continent. Minor wars
and small changes in territory or in the distribution of power did not affect her
vital interests. An adversary who was out to overthrow the system could be
stopped after he had made his intentions clear. And revolutions within other
states were no menace, since they would not set off unrest within England.
Austria, surrounded by strong powers, was not so fortunate; her policy had to
be more closely attuned to all conflicts. By the time an aggressor-state had
clearly shown its colors, Austria would be gravely threatened. And foreign rev-
olutions, be they democratic or nationalistic, would encourage groups in
Austria to upset the existing order. So it is not surprising that Metternich pro-
pounded the doctrine summarized earlier, which defended Austria's right to
interfere in the internal affairs of others, and that British leaders rejected this
view. Similarly, Austria wanted the Congress system to be a relatively tight
one, regulating most disputes. The British favored a less centralized system. In
other words, in order to protect herself, Austria had either to threaten or to
harm others, whereas Britain did not. For Austria and her neighbors the secu-
rity dilemma was acute; for Britain it was not.
The ~~ltirnate cost of CD is of course loss of sovereignty. This cost can
vary from situation to situation. The lower it is (for instance, because the
two states have compatible ideologies, are similar ethnically, have a common
culture, or because the citizens of the losing state expect economic benefits),
the less the impact of the security dilemma; the greater the costs, the greater
the impact of the dilemma. Here is another reason why extreme differences
in values and ideologies exacerbate international conflict.
It is through the lowering of the costs of CD that the proposed Rhodesian
"safety net" - guaranteeing that whites who leave the country will receive
fair payment for their property -would have the paradoxical effect of mak-
ing it more likely that the whites will stay. This is less puzzling when we see
that the whites are in a multi-person Prisoner's Dilemma with each other.
Assume that all whites are willing to stay if most of the others stay; but, in
the absence of guarantees, if there is going to be a mass exodus, all want to
be among the first to leave (because late-leavers will get less for their prop-
erty and will have more trouble finding a country to take them in). Then the
problem is to avoid a self-fulfilling prophecy in which each person rushes to
defect because he fears others are going to. In narrowing the gap between
the payoff for leaving first (DC) and leaving last (CD) by reducing the cost
of the latter, the guarantees make it easier for the whites to cooperate among
themselves and stay.
Subjective Security Demands. Decision makers act in terms of the vul-
nerability they feel, which can differ from the actual situation; we must
therefore examine the decision makers' subjective security requirements."'
Two dimensions are involved. First, even if they agree about the objective
situation, people can differ about how much security they desire - or, to put
it more precisely, about the price they are willing to pay to gain increments
of security. The more states value their security above all else (that is, see a
prohibitively high cost in CD), the more they are likely to be sensitive to
even minimal threats, and to demand high levels of arms. And if arms are
positively valued because of pressures from a military-industrial complex, it
will be especially hard for status-quo powers to cooperate. By contrast, the
security dilemma will not operate as strongly when pressing domestic con-
cerns increase the opportunity costs of armaments. In this case, the net
advantage of exploiting the other (DC) will be less, and the costs of arms
races (that is, one aspect of DD) will be greater; therefore the state will
behave as though it were relatively invulnerable.
The second aspect of subjective security is the perception of threat (that
is, the estimate of whether the other will cooperate)." A state that is predis-
posed to see either a specific other state as an adversary, or others in general
as a menace, will react more strongly and more quickly than a state that sees
its environment as benign. Indeed, when a state believes that another not
only is not likely to be an adversary, but has sufficient interests in common
with it to be an ally, then it will actually welcome an increase in the other's
power.
British and French foreign policies in the interwar years illustrate these
points. After the rise of Hitler, Britain and France felt that increases in each
other's arms increased rather than decreased their own security. The differing
policies that these states followed toward Germany can be explained by their
differences on both dimensions of the variable of subjective security."
Throughout the period, France perceived Germany as more of a threat than
England did. The British were more optimistic and argued that conciliation
could turn Germany into a supporter of the status quo. Furthermore, in the
years immediately following World War I, France had been more willing to
forego other values in order to increase her security and had therefore fol-
lowed a more belligerent policy than England, maintaining a larger army and
moving quickly to counter German assertiveness. As this example shows, one
cannot easily say how much subjective security a state should seek. High
security requirements make it very difficult to capitalize on a common inter-
est and run the danger of setting off spirals of arms races and hostility. The
French may have paid this price in the 1920's. Low security requirements
avoid this trap, but run the risk of having too few arms and of trying to con-
ciliate an aggressor.
One aspect of subjective security related to the predisposition to perceive
threat is the state's view of how many enemies it must be prepared to fight.
A state can be relaxed about increases in another's arms if it believes that
there is a functioning collective security system. The chances of peace are
increased in a world in which the prevailing international system is valued
in its own right, not only because most states restrain their ambitions and
those who d o not are deterred (these are the usual claims for a Concert sys-
tem), but also because of the decreased chances that the status-quo states
will engage in unnecessary conflict out of the quest for security. Indeed, if
there were complete faith in collective security, n o state would want an
army. By contrast, the security dilemma is insoluble when each state fears
It I \ I Security Dilemma 137
that many others, far from coming to its aid, are likely to join in any attack.
Winston Churchill, as First Lord of the Admiralty, was setting a high secur-
ity requirement when he noted:
Besides the Great Powers, there are many small states who are buying or
building great ships of war and whose vessels may by purchase, by some
diplomatic combination, or by duress, be brought into the line against
us. None of these powers need, like us, navies to defend their actual
safety of independence. They build them so as to play a part in world
affairs. It is sport to them. It is death to us."
It takes great effort for any one state to be able to protect itself alone
against an attack by several neighbors. More importantly, it is next to
impossible for all states in the system to have this capability. Thus, a state's
expectation that allies will be available and that only a few others will be
able to join against it is almost a necessary condition for security require-
ments to be compatible.
The main costs of a policy of reacting quickly and severely to increases in the
other's arms are not the price of one's own arms, but rather the sacrifice of the
potential gains from cooperation (CC)and the increase in the dangers of need-
less arms races and wars (DD). The greater these costs, the greater the incen-
tives to try cooperation and wait for fairly unambiguous evidence before
assuming that the other must be checked by force. Wars would be much more
frequent - even if the first choice of all states was the status quo - if they were
less risky and costly, and if peaceful intercourse did not provide rich benefits.
Ethiopia recently asked for guarantees that the Territory of Afars and Issas
would not join a hostile alliance against it when it gained independence.
A spokesman for the Territory replied that this was not necessary: Ethiopia
"already had the best possible guarantee in the railroad" that links the two
countries and provides indispensable revenue for the Territory." l 4
The basic points are well known and so we can move to elaboration.
First, most statesmen know that to enter a war is to set off a chain of unpre-
dictable and uncontrollable events. Even if everything they see points to a
quick victory, they are likely to hesitate before all the uncertainties. And i f
the battlefield often produces startling results, so do the council chambers.
The state may be deserted by allies or attacked by neutrals. O r the postwar
alignment may rob it of the fruits of victory, as happened to Japan in 1895.
Second, the domestic costs of wars must be weighed. Even strong states can
be undermined by dissatisfaction with the way the war is run and by the
necessary mobilization of men and ideas. Memories of such disruptions
were one of the main reasons for the era of relative peace that followed the
Napoleonic Wars. Liberal statesmen feared that large armies would lead to
despotism; conservative leaders feared that wars would lead to revolution.
(The other side of this coin is that when there are domestic consequences of
foreign conflict that are positively valued, the net cost of conflict is lowered
and cooperation becomes more difficult.)Third - turning to the advantages
of cooperation - for states with large and diverse economies the gains from
economic exchange are rarely if ever sufficient to prevent war. Norman
Angel1 was wrong about World War I being impossible because of economic
ties among the powers; and before World War 11, the U.S. was Japan's most
important trading partner. Fourth, the gains from cooperation can be
increased, not only if each side gets more of the traditional values such as
wealth, but also if each comes to value the other's well-being positively.
Mutual cooperation will then have a double payoff: in addition to the direct
gains, there will be the satisfaction of seeing the other prosper.'"
While high costs of war and gains from cooperation will ameliorate the
impact of the security dilemma, they can create a different problem. If the
costs are high enough so that DD is the last choice for both sides, the game
will shift to "Chicken." This game differs from the Stag Hunt in that each
actor seeks to exploit the other; it differs from Prisoner's Dilemma in that
both actors share an interest in avoiding mutual non-cooperation. In
Chicken, if you think the other side is going to defect, you have to cooper-
ate because, although being exploited (CD) is bad, it is not as bad as a total
breakdown (DD). As the familiar logic of deterrence shows, the actor must
then try to convince his adversary that he is going to stand firm (defect)and
that the only way the other can avoid disaster is to back down (cooperate).
Commitment, the rationality of irrationality, manipulating the communica-
tions system, and pretending not to understand the situation, are among the
tactics used to reach this goal. The same logic applies when both sides are
enjoying great benefits from cooperation. The side that can credibly
threaten to disrupt the relationship unless its demands are met can exploit
the other. This situation may not be stable, since the frequent use of threats
may be incompatible with the maintenance of a cooperative relationship.
Still, de Gaulle's successful threats to break up the Common Market unless
his partners acceded to his wishes remind us that the shared benefits of
cooperation as well as the shared costs of defection can provide the basis
for exploitation. Similarly, one reason for the collapse of the Franco-British
entente more than a hundred years earlier was that decision makers on both
sides felt confident that their own country could safely pursue a policy that
was against the other's interest because the other could not afford to
destroy the highly valued relationship.I6 Because statesmen realize that the
growth of positive interdependence can provide others with new levers of
influence over them, they may resist such developments more than would
be expected from the theories that stress the advantages of cooperation.
Gains from Exploitation fDCI
Defecting not only avoids the danger that a state will be exploited (CD),but
brings positive advantages by exploiting the other (DC). The lower these
lcri i i Security Dilemma 139
possible gains, the greater the chances of cooperation. Even a relatively sat-
isfied state can be tempted to expand by the hope of gaining major values.
The temptation will be less when the state sees other ways of reaching its
goals, and/or places a low value on what exploitation could bring. The
gains may be low either because the immediate advantage provided by DC
(for example, having more arms than the other side) cannot be translated
into a political advantage (for example, gains in territory), or because the
political advantage itself is not highly valued. For instance, a state may not
seek to annex additional territory because the latter lacks raw materials, is
inhabited by people of a different ethnic group, would be costly to garrison,
or would be hard to assimilate without disturbing domestic politics and
values. A state can reduce the incentives that another state has to attack it,
by not being a threat to the latter and by providing goods and services that
would be lost were the other to attempt exploitation.
Even where the direct advantages of D C are great, other considerations
can reduce the net gain. Victory as well as defeat can set off undesired
domestic changes- within the state. Exploitation has at times been frowned
upon by the international community, thus reducing the prestige of a state
that engages in it. O r others might in the future be quicker to see the state
as a menace to them, making them more likely to arm, and to oppose it
later. Thus, Bismarck's attempts to get other powers to cooperate with him
in maintaining the status quo after 1871 were made more difficult by the
widely-held mistrust of him that grew out of his earlier aggressions.'-
The variables discussed so far influence the payoffs for each of the four pos-
sible outcomes. To decide what to do, the state has to go further and calcu-
late the expected value of cooperating or defecting. Because such
calculations involve estimating the probability that the other will cooperate,
the state will have to judge how the variables discussed so far act on the
other. To encourage the other to cooperate, a state may try to manipulate
these variables. It can lower the other's incentives to defect by decreasing
what it could gain by exploiting the state (DC) - the details would be simi-
lar to those discussed in the previous paragraph - and it can raise the costs
of deadlock (DD). But if the state cannot make DD the worst outcome for
the other, coercion is likely to be ineffective in the short run because the
other can respond by refusing to cooperate, and dangerous in the long run
because the other is likely to become convinced that the state is aggressive.
So the state will have to concentrate on making cooperation more attractive.
One way to do this is to decrease the costs the other will pay if it cooperates
and the state defects (0). Thus, the state could try to make the other less
vulnerable. It was for this reason that in the late 1950's and early 1960's
some American defense analysts argued that it would be good for both sides
if the Russians developed hardened missiles. Of course, decreasing the
other's vulnerability also decreases the state's ability to coerce it, and opens
the possibility that the other will use this protection as a shield behind which
to engage in actions inimical to the state. But by sacrificing some ability to
harm the other, the state can increase the chances of mutually beneficial
cooperation.
The state can also try to increase the gains that will accrue to the other
from mutual cooperation (CC). Although the state will of course gain if it
receives a share of any new benefits, even an increment that accrues entirely
t o the other will aid the state by increasing the likelihood that the other will
cooperate. l 8
This line of argument can be continued through the infinite regressions
that game theory has made familiar. If the other is ready to cooperate when
it thinks the state will, the state can increase the chances of CC by showing
that it is planning t o cooperate. Thus the state should understate the gains
it would make if it exploited the other (DC) and the costs it would pay if
the other exploited it (CD), and stress or exaggerate the gains it would
make under mutual cooperation (CC) and the costs it would pay if there is
deadlock (DD). The state will also want to convince the other that it thinks
that the other is likely t o cooperate. If the other believes these things, it will
see that the state has strong incentives to cooperate, and so it will cooper-
ate in turn. One point should be emphasized. Because the other, like the
state, may be driven t o defect by the fear that it will be exploited if it does
not, the state should try to reassure it that this will not happen. Thus, when
Khrushchev indicated his willingness to withdraw his missiles from Cuba,
he simultaneously stressed to Kennedy that "we are of sound mind and
understand perfectly well" that Russia could not launch a successful attack
against the U.S., and therefore that there was no reason for the U.S. to con-
template a defensive, pre-emptive strike of its own.''
There is, however, a danger. If the other thinks that the state has little
choice but to cooperate, it can credibly threaten to defect unless the state
provides it with additional benefits. Great advantages of mutual coopera-
tion, like high costs of war, provide a lever for competitive bargaining.
Furthermore, for a state t o stress how much it gains from cooperation may
be to imply that it is gaining much more than the other and t o suggest that
the benefits should be distributed more equitably.
When each side is ready to cooperate if it expects the other to, inspec-
tion devices can ameliorate the security dilemma. Of course, even a perfect
inspection system cannot guarantee that the other will not later develop
aggressive intentions and the military means to act on them. But by reliev-
ing immediate worries and providing warning of coming dangers, inspec-
tion can meet a significant part of the felt need to protect oneself against
future threats, and so make current cooperation more feasible. Similar func-
tions are served by breaking up one large transaction into a series of smaller
ones.20 At each transaction each can see whether the other has cooperated;
and its losses, if the other defects, will be small. And since what either side
would gain by one defection is slight compared to the benefits of continued
cooperation, the prospects of cooperation are high. Conflicts and wars
I I Security Dilemma 14 1
among status-quo powers would be much more common were it not for the
fact that international politics is usually a series of small transactions.
How a statesman interprets the other's past behavior and how he projects
it into the future is influenced by his understanding of the security dilemma
and his ability to place himself in the other's shoes. The dilemma will operate
much more strongly if statesmen do not understand it, and do not see that
their arms - sought only to secure the status quo - may alarm others and that
others may arm, not because they are contemplating aggression, but because
they fear attack from the first state. These two failures of empathy are linked.
A state which thinks that the other knows that it wants only to preserve the
status q ~ and~ othat its arms are meant only for self-preservation will conclude
that the other side will react to its arms by increasing its own capability only
if it is aggressive itself. Since the other side is not menaced, there is no legitim-
ate reason for it to object to the first state's arms; therefore, objection proves
that the other is aggressive. Thus, the following exchange between Senator
Tom Connally and Secretary of State Acheson concerning the ratification of
the NATO treaty:
Secretary Acheson: [The treaty] is aimed solely at armed aggression.

Senator Connally: In other words, unless a nation ... contemplates, medi-
tates, or makes plans looking toward aggression or armed attack on
another nation, it has no cause to fear this treaty.
Secretary Acheson: That is correct, Senator Connally, and it seems to
me that any nation which claims that this treaty is directed against it
should be reminded of the Biblical admonition that 'The guilty flee when
no man pursueth.'
Senator Connally: That is a very apt illustration.
What I had in mind was, when a State or Nation passes a criminal
act, for instance, against burglary, nobody but those who are burglars or
getting ready to be burglars need have any fear of the Burglary Act. Is
that not true?
Secretary Acheson: The only effect [the law] would have [on an inno-
cent person] would be for his protection, perhaps, by deterring someone
else. He wouldn't worry about the imposition of the penalties on himself."
The other side of this coin is that part of the explanation for detente is that
most American decision makers now realize that it is at least possible that
Russia may fear American aggression; many think that this fear accounts for
a range of Soviet actions previously seen as indicating Russian aggressiveness.
Indeed, even 36 percent of military officers consider the Soviet Union's motiv-
ations to be primarily defensive. Less than twenty years earlier, officers had
been divided over whether Russia sought world conquest or only expansion."
Statesmen who do not understand the security dilemma will think that the
money spent is the only cost of building up their arms. This belief removes
one important restraint on arms spending. Furthermore, it is also likely to
lead states to set their security requirements too high. Since they do not
understand that trying to increase one's security can actually decrease it, they
will overestimate the amount of security that is attainable; they will think that
when in doubt they can "play it safe" by increasing their arms. Thus it is very
likely that two states which support the status quo but do not understand the
security dilemma will end up, if not in a war, then at least in a relationship of
higher conflict than is required by the objective situation.
The belief that an increase in military strength always leads to an increase
in security is often linked to the belief that the only route to security is through
military strength. As a consequence, a whole range of meliorative policies will
be downgraded. Decision makers who do not believe that adopting a more
conciliatory posture, meeting the other's legitimate grievances, or developing
mutual gains from cooperation can increase their state's security, will not
devote much attention or effort to these possibilities.
On the other hand, a heightened sensitivity to the security dilemma makes
it more likely that the state will treat an aggressor as though it were an inse-
cure defender of the status quo. Partly because of their views about the causes
of World War I, the British were predisposed to believe that Hitler sought only
the rectification of legitimate and limited grievances and that security could
best be gained by constructing an equitable international system. As a result
they pursued a policy which, although well designed to avoid the danger of
creating unnecessary conflict with a status-quo Germany, helped destroy
Europe.
Geography, Commitments, Beliefs, and Security Through Expansion
A final consideration does not easily fit in the matrix we have been using,
although it can be seen as an aspect of vulnerability and of the costs of CD.
Situations vary in the ease or difficulty with which all states can simultan-
eously achieve a high degree of security, The influence of military technology
on this variable is the subject of the next section. Here we want to treat the
impact of beliefs, geography, and commitments (many of which can be con-
sidered to be modifications of geography, since they bind states to defend
areas outside their homelands). In the crowded continent of Europe, secur-
ity requirements were hard to mesh. Being surrounded by powerful states,
Germany's problem - or the problem created by Germany - was always
great and was even worse when her relations with both France and Russia
were bad, such as before World War I. In that case, even a status-quo
Germany, if she could not change the political situation, would almost have
been forced to adopt something like the Schlieffen Plan. Because she could
not hold off both of her enemies, she had to be prepared to defeat one
quickly and then deal with the other in a more leisurely fashion. If France
or Russia stayed out of a war between the other state and Germany, they
would allow Germany to dominate the Continent (even if that was not
Germany's aim). They therefore had to deny Germany this ability, thus
making Germany less secure. Although Germany's arrogant and erratic
behavior, coupled with the desire for an unreasonably high level of security
IPILI Security Dilemma 143
(which amounted t o the desire to escape from her geographic plight), com-
pounded the problem, even wise German statesmen would have been hard
put t o gain a high degree of security without alarming their neighbors.
A similar situation arose for France after World War I. She was committed
to protecting her allies in Eastern Europe, a commitment she could meet only
by taking the offensive against Germany. But since there was no way to guar-
antee that France might not later seek expansion, a France that could success-
fully launch an attack in response to a German move into Eastern Europe
would constitute a potential danger to German core values. Similarly, a United
States credibly able to threaten retaliation with strategic nuclear weapons if
the Soviet Union attacks Western Europe also constitutes a menace, albeit a
reduced one, to the Soviet ability to maintain the status quo. The incompati-
bility of these security requirements is not complete. Herman Kahn is correct
in arguing that the United States could have Type I1 deterrence (the ability to
deter a major Soviet provocation) without gaining first-strike capability
because the expected Soviet retaliation following an American strike could be
great
- enough to deter the U.S. from attacking unless the U.S. believed it would
suffer enormous deprivation (for instance, the loss of Europe) if it did not
strike." Similarly, the Franco-German military balance could have been such
that France could successfully attack Germany if the latter's armies were
embroiled in Eastern Europe, but could not defeat a Germany that was free to
devote all her resources to defending herself. But this delicate balance is very
hard to achieve, especially because states usually calculate conservatively.
Therefore, such a solution is not likely to be available.
For the United States, the problem posed by the need to protect Europe is
an exception. Throughout most of its history, this country has been in a much
more favorable position: relatively self-sufficient and secure from invasion, it
has not only been able to get security relatively cheaply, but by doing so, did
not menace others.14 But ambitions and commitments have changed this situ-
ation. After the American conquest of the Philippines, "neither the United
States nor Japan could assure protection for their territories by military and
naval means without compromising the defenses of the other. This problem
would plague American and Japanese statesmen down to 1941 ."" Further-
more, to the extent that Japan could protect herself, she could resist American
threats to go t o war if Japan did not respect China's independence. These
complications were minor compared t o those that followed World War 11.
A world power cannot help but have the ability t o harm many others that is
out of proportion to the others' ability to harm it.
Britain had been able t o gain security without menacing others to a
greater degree than the Continental powers, though t o a lesser one than the
United States. But the acquisition of colonies and a dependence on foreign
trade sacrificed her relative invulnerability of being an island. Once she
took India, she had to consider Russia as a neighbor; the latter was expand-
ing in Central Asia, thus making it much more difficult for both countries
t o feel secure. The need t o maintain reliable sea lanes t o India meant that
no state could be allowed t o menace South Africa and, later, Egypt. But the
need t o protect these two areas brought new fears, new obligations, and
new security requirements that conflicted with those of other European
nations. Furthermore, once Britain needed a flow of imports during both
peace and wartime, she required a navy that could prevent a blockade. A
navy sufficient for that task could not help but be a threat to any other state
that had valuable trade.
A related problem is raised by the fact that defending the status quo often
means protecting more than territory. Nonterritorial interests, norms, and the
structure of the international system must be maintained. If all status-quo
powers agree on these values and interpret them in compatible ways, prob-
lems will be minimized. But the potential for conflict is great, and the policies
followed are likely to exacerbate the security dilemma. The greater the range
of interests that have to be protected, the more likely it is that national efforts
to maintain the status quo will clash. As a French spokesman put it in 1930:
"Security! The term signifies more indeed than the maintenance of a people's
homeland, or even of their territories beyond the seas. It also means the main-
tenance of the world's respect for them, the maintenance of their economic
interests, everything in a word, which goes to make up the grandeur, the life
itself, of the nation."26 When security is thought of in this sense, it almost
automatically has a competitive connotation. It involves asserting one state's
will over others, showing a high degree of leadership if not dominance, and
displaying a prickly demeanor. The resulting behavior will almost surely clash
with that of others who define their security in the same way.
The problem will be almost insoluble if statesmen believe that their secur-
ity requires the threatening or attacking of others. "That which stops growing
begins to rot," declared a minister to Catherine the Great."" More common
is the belief that if the other is secure, it will be emboldened to act against
one's own state's interests, and the belief that in a war it will not be enough
for the state to protect itself: it must be able to take the war to the other's
homeland. These convictions make it very difficult for status-quo states to
develop compatible security policies, for they lead the state to conclude that
its security requires that others be rendered insecure.
In other cases, "A country engaged in a war of defense might be obliged
for strategic reasons to assume the offensive," as a French delegate to an inter-
war disarmament conference put it.28 That was the case for France in 1799:
The Directory's political objectives were essentially defensive, for the

French wanted only to protect the Republic from invasion and preserve
the security and territory of the satellite regimes in Holland, Switzerland,
and Italy. French leaders sought no new conquests; they wanted only to
preserve the earlier gains of the Revolution. The Directory believed,
however, that only a military offensive could enable the nation to
achieve its defensive political objective. By inflicting rapid and decisive
defeats upon one or more members of the coalition, the directors hoped
to rupture allied unity and force individual powers to seek a separate,
peace.29
1, r Security Dilemma 145
It did not matter to the surrounding states that France was not attacking
because she was greedy, but hecause she wanted to be left In peace. Unless
there was some way her neighbors could provide France with an alternate
route to her goal, France had to go to war.
Ill. Offense, Defense, a n d the Security Dilemma
Another approach starts with the central point of the security dilemma - that
an increase in one state's security decreases the security of others - and exam-
ines the conditions under which this proposition holds. Two crucial variables
are involved: whether defensive weapons and policies can be distinguished
from offensive ones, and whether the defense or the offense has the advan-
tage. The definitions are not always clear, and many cases are difficult to
judge, but these two variables shed a great deal of light on the question of
whether status-quo powers will adopt compatible security policies. All the
variables discussed so far leave the heart of the problem untouched. Rut when
defensive weapons differ from offensive ones, it is possible for a state to make
itself more secure without making others less secure. And when the defense
has the advantage over the offense, a large increase in one state's security only
slightly decreases the security of the others, and status-quo powers can all
enjoy a high level of security and largely escape from the state of nature.
When we say that the offense has the advantage, we simply mean that it is
easier to destroy the other's army and take its territory than it is to defend
one's own. When the defense has the advantage, it is easier to protect and
to hold than it is to move forward, destroy, and take. If effective defenses
can be erected quickly, an attacker may be able to keep territory he has taken
in an initial victory. Thus, the dominance of the defense made it very hard
for Britain and France to push Germany out of France in World War I. But
when superior defenses are difficult for an aggressor to improvise on the
battlefield and must be constructed during peacetime, they provide no
direct assistance to him.
The security dilemma is at its most vicious when commitments, strategy, or
technology dictate that the only route to security lies through expansion.
Status-quo powers must then act like aggressors; the fact that they would
gladly agree to forego the opportunity for expansion in return for guarantees
for their security has no implications for their behavior. Even if expansion is
not sought as a goal in itself, there will be quick and drastic changes in the dis-
tribution of territory and influence. Conversely, when the defense has the
advantage, status-quo states can make themselves more secure without gravely
endangering others."' Indeed, if the defense has enough of an advantage and
if the states are of roughly equal size, not only will the security dilemma cease
to inhibit status-quo states from cooperating, but aggression will be next to
I46 The Cold War a n d Nuclear Deterrence
impossible, thus rendering international anarchy relatively unimportant. If

states cannot conquer each other, then the lack of sovereignty, although it pres-
ents problems of collective goods in a number of areas, no longer forces states
to devote their primary attention to self-preservation. Although, if force were
not usable, there would be fewer restraints on the use of nonmilitary instru-
ments, these are rarely powerful enough to threaten the vital interests of a
major state.
Two questions of the offense-defense balance can be separated. First, does
the state have to spend more or less than one dollar on defensive forces to off-
set each dollar spent by the other side on forces that could be used to attack?
If the state has one dollar to spend on increasing its security, should it put it
into offensive or defensive forces? Second, with a given inventory of forces, is
it better to attack or to defend? Is there an incentive to strike first or to absorb
the other's blow? These two aspects are often linked: if each dollar spent on
offense can overcome each dollar spent on defense, and if both sides have the
same defense budgets, then both are likely to build offensive forces and find it
attractive to attack rather than to wait for the adversary to strike.
These aspects affect the security dilemma in different ways. The first has
its greatest impact on arms races. If the defense has the advantage, and if the
status-quo powers have reasonable subjective security requirements, they can
probably avoid an arms race. Although an increase in one side's arms and
security will still decrease the other's security, the former's increase will be
larger than the latter's decrease. So if one side increases its arms, the other can
bring its security back up to its previous level by adding a smaller amount to
its forces. And if the first side reacts to this change, its increase will also be
smaller than the stimulus that produced it. Thus a stable equilibrium will be
reached. Shifting from dynamics to statics, each side can be quite secure with
forces roughly equal to those of the other. Indeed, if the defense is much more
potent than the offense, each side can be willing to have forces much smaller
than the other's, and can be indifferent to a wide range of the other's defense
policies.
The second aspect - whether it is better to attack or t o defend - influences
short-run stability. When the offense has the advantage, a state's reaction to
international tension will increase the chances of war. The incentives for pre-
emption and the "reciprocal fear of surprise attack" in this situation have
been made clear by analyses of the dangers that exist when two countries
have first-strike c a p a b i l i t i e ~ There
. ~ ~ is no way for the state to increase its
security without menacing, or even attacking, the other. Even Bismarck, who
once called preventive war "committing suicide from fear of death," said that
"no government, if it regards war as inevitable even if it does not want it,
would be so foolish as to leave to the enemy the choice of time and occasion
and to wait for the moment which is most convenient for the enemy."32 In
another arena, the same dilemma applies to the policeman in a dark alley
confronting a suspected criminal who appears to be holding a weapon.
Though racism may indeed be present, the security dilemma can account for
many of the tragic shootings of innocent people in the ghettos.
IixtiI - Security Dilemma 147
Beliefs about the course of a war in which the offense has the advantage
further deepen the security dilemma. When there are incentives to strike
first, a successful attack will usually so weaken the other side that victory
will be relatively quick, bloodless, and decisive. It is in these periods when
conquest is possible and attractive that states consolidate power internally -
for instance, by destroying the feudal barons - and expand externally. There
are several consequences that decrease the chance of cooperation among
status-quo states. First, war will be profitable for the winner. The costs will
be low and the benefits high. Of course, losers will suffer; the fear of losing
could induce states to try to form stable cooperative arrangements, but the
temptation of victory will make this particularly difficult. Second, because
wars are expected to be both frequent and short, there will be incentives for
high levels of arms, and quick and strong reaction to the other's increases
in arms. The state cannot afford to wait until there is unambiguous
evidence that the other is building new weapons. Even large states that have
faith in their economic strength cannot wait, because the war will be over
before their products can reach the army. Third, when wars are quick,
,,
states will have to recruit allies in advance." Without the opportunity for
bargaining and re-alignments during the opening stages of hostilities, peace-
time diplomacy loses a degree of the fluidity that facilitates halance-of-
power policies. Because alliances must be secured during peacetime, the
international system is more likely to become bipolar. It is hard to say
whether war therefore becomes more or less likely, but this hipolarity
increases tension between the two camps and makes it harder for status-quo
states to gain the benefits of cooperation. Fourth, if wars are frequent,
statesmen's perceptual thresholds will be adj~istedaccordingly and they will
be quick to perceive ambiguous evidence as indicating that others are
aggressive. Thus, there will be more cases of status-quo powers arming
against each other in the incorrect belief that the other is hostile.
When the defense has the advantage, all the foregoing is reversed. The
state that fears attack does not p r e e m p t - since that would be a wasteful
use of its military resources - but rather prepares to receive an attack.
Doing so does not decrease the security of others, and several states can do
it simultaneously; the situation will therefore be stable, and status-quo
powers will be able to cooperate. When Herman Kahn argues that ulti-
matums "are vastly too dangerous to give because ... they are quite likely
to touch off a pre-emptive strike,"'%e incorrectly assumes that it is always
advantageous to strike first.
More is involved than short-run dynamics. When the defense is dominant,
wars are likely to become stalemates and can be won only at enormous cost.
Relatively small and weak states can hold off larger and stronger ones, or can
deter attack by raising the costs of conquest to an unacceptable level. States
then approach equality in what they can do to each other. Like the .45-caliber
pistol in the American West, fortifications were the "great equalizer" in some
periods. Changes in the status quo are less frequent and cooperation is more
common wherever the security dilemma is thereby reduced.
I48 The Cold War a n d Nuclear Deterrence
Many of these arguments can be illustrated by the major powers' pol-

icies in the periods preceding the two world wars. Bismarck's wars surprised
statesmen by showing that the offense had the advantage, and by being
quick, relatively cheap, and quite decisive. Falling into a common error,
observers projected this pattern into the f ~ t u r e . The
~ ' resulting expectations
had several effects. First, states sought semi-permanent allies. In the early
stages of the Franco-Prussian War, Napoleon 111 had thought that there
would be plenty of time to recruit Austria to his side. Now, others were not
going t o repeat this mistake. Second, defense budgets were high and reacted
quite sharply to increases on the other side. It is not surprising that
Richardson's theory of arms races fits this period well. Third, most decision
makers thought that the next European war would not cost much blood
and treasure.36That is one reason why war was generally seen as inevitable
and why mass opinion was so bellicose. Fourth, once war seemed likely,
there were strong pressures to pre-empt. Both sides believed that whoever
moved first could penetrate the other deep enough to disrupt mobilization
and thus gain an insurmountable advantage. (There was no such belief
about the use of naval forces. Although Churchill made an ill-advised
speech saying that if German ships "do not come out and fight in time of
war they will be dug out like rats in a hole,"37 everyone knew that sub-
marines, mines, and coastal fortifications made this impossible. So at the
start of the war each navy prepared to defend itself rather than attack, and
the short-run destabilizing forces that launched the armies toward each
other did not operate.)38Furthermore, each side knew that the other saw
the situation the same way, thus increasing the perceived danger that the
other would attack, and giving each added reasons to precipitate a war if
conditions seemed favorable. In the long and the short run, there were thus
both offensive and defensive incentives to strike. This situation casts light
on the common question about German motives in 1914: "Did Germany
unleash the war deliberately to become a world power or did she support
Austria merely to defend a weakening ally," thereby protecting her own
position?39To some extent, this question is misleading. Because of the per-
ceived advantage of the offense, war was seen as the best route both to gain-
ing expansion and to avoiding drastic loss of influence. There seemed to be
no way for Germany merely to retain and safeguard her existing position.
Of course the war showed these beliefs to have been wrong on all points.
Trenches and machine guns gave the defense an overwhelming advantage. The
fighting became deadlocked and produced horrendous casualties. It made no
sense for the combatants to bleed themselves to death. If they had known the
power of the defense beforehand, they would have rushed for their own
trenches rather than for the enemy's territory. Each side could have done this
without increasing the other's incentives to strike. War might have broken out
anyway, just as DD is a possible outcome of Chicken, but at least the pressures
of time and the fear of allowing the other to get the first blow would-not have
contributed to this end. And, had both sides known the costs of the war, they
would have negotiated much more seriously. The obvious question is why the
rt.t; I Security Dilemma 149
states did not seek a negotiated settlement as soon as the shape of the war
became clear. Schlieffen had said that if his plan failed, peace should be
sought."' The answer is complex, uncertain, and largely outside of the scope of
our concerns. But part of the reason was the hope and sometimes the expect-
ation that breakthroughs could be made and the dominance of the offensive
restored. Without that hope, the political and psychological pressures to fight
to a decisive victory might have been overcome.
The politics of the interwar period were shaped by the memories of the
previous conflict and the belief that any future war would resemble it.
Political and military lessons reinforced each other in ameliorating the secur-
ity dilemma. Because it was believed that the First World War had been a mis-
take that could have been avoided by skillful conciliation, both Britain and,
to a lesser extent, France were highly sensitive to the possibility that interwar
Germany was not a real threat to peace, and alert to the danger that reacting
quickly and strongly to her arms could create unnecessary conflict. And
because Britain and France expected the defense to continue to dominate,
they concluded that it was safe to adopt a more relaxed and nonthreatening
military posture."" Britain also felt less need to maintain tight alliance
bonds. The Allies' military posture then constituted only a slight danger to
Germany; had the latter been content with the status quo, it would have been
easy for both sides to have felt secure behind their lines of fortifications. Of
course the Germans were not content, so it is not surprising that they devoted
their money and attention to finding ways out of a defense-dominated stale-
mate. Blitzkrieg tactics were necessary if they were to use force to change the
status quo.
The initial stages of the war on the Western Front also contrasted with
the First World War. Only with the new air arm were there any incentives
to strike first, and these forces were too weak to carry out the grandiose
plans that had been both dreamed and feared. The armies, still the main
instrument, rushed to defensive positions. Perhaps the allies could have suc-
cessfully attacked while the Germans were occupied in Poland.4L But belief
in the defense was so great that this was never seriously contemplated.
Three months after the start of the war, the French Prime Minister summed
up the view held by almost everyone but Hitler: on the Western Front there
is "deadlock. Two Forces of equal strength and the one that attacks seeing
such enormous casualties that it cannot move without endangering the con-
tinuation of the war or of the aftermath."43 The Allies were caught in a
dilemma they never fully recognized, let alone solved. O n the one hand,
they had very high war aims; although unconditional surrender had not yet
been adopted, the British had decided from the start that the removal of
Hitler was a necessary condition for p e a ~ e . ~ Wthe n other hand, there were
no realistic plans or instruments for allowing the Allies to impose their will
on the other side. The British Chief of the Imperial General Staff noted,
"The French have no intention of carrying out an offensive for years, if at
all"; the British were only slightly b ~ l d e r . ~So
' the Allies looked to a long
war that would wear the Germans down, cause civilian suffering through
I50 T h e Cold War a n d Nuclear D e t e r r e n c e
shortages, and eventually undermine Hitler. There was little analysis to sup-
port this view - and indeed it probably was not supportable - but as long
as the defense was dominant and the numbers on each side relatively equal,
what else could the Allies do?
To summarize, the security dilemma was much less powerful after World
War I than it had been before. In the later period, the expected power of the
defense allowed status-quo states to pursue compatible security policies and
avoid arms races. Furthermore, high tension and fear of war did not set off
short-run dynamics by which each state, trying to increase its security, inad-
vertently acted to make war more likely. The expected high costs of war,
however, led the Allies t o believe that no sane German leader would run the
risks entailed in an attempt to dominate the Continent, and discouraged
them from risking war themselves.
Technology and Geography. Technology and geography are the two main
factors that determine whether the offense or the defense has the advantage.
As Brodie notes, "On the tactical level, as a rule, few physical factors favor
the attacker but many favor the defender. The defender usually has the advan-
tage of cover. He characteristically fires from behind some form of shelter
while his opponent crosses open ground."46 Anything that increases the
amount of ground the attacker has to cross, or impedes his progress across it,
or makes him more vulnerable while crossing, increases the advantage accru-
ing to the defense. When states are separated by barriers that produce these
effects, the security dilemma is eased, since both can have forces adequate for
defense without being able to attack. Impenetrable barriers would actually
prevent war; in reality, decision makers have to settle for a good deal less.
Buffer zones slow the attacker's progress; they thereby give the defender time
to prepare, increase problems of logistics, and reduce the number of soldiers
available for the final assault. At the end of the 19th century, Arthur Balfour
noted Afghanistan's "non-conducting" qualities. "So long as it possesses few
roads, and no railroads, it will be impossible for Russia to make effective use
of her great numerical superiority at any point immediately vital to the
Empire." The Russians valued buffers for the same reasons; it is not surpris-
ing that when Persia was being divided into Russian and British spheres of
influence some years later, the Russians sought assurances that the British
would refrain from building potentially menacing railroads in their sphere.
Indeed, since railroad construction radically altered the abilities of countries to
defend themselves and to attack others, many diplomatic notes and much
intelligence activity in the late 19th century centered on this ~ubject.~'
Oceans, large rivers, and mountain ranges serve the same function as
buffer zones. Being hard to cross, they allow defense against superior num-
bers. The defender has merely to stay on his side of the barrier and so can
utilize all the men he can bring up to it. The attacker's men, however, can
cross only a few at a time, and they are very vulnerable when doing so. If all
states were self-sufficient islands, anarchy would be much less of a problem.
A small investment in shore defenses and a small army would be sufficient
to repel invasion. Only very weak states would be vulnerable, and only very
Jt.rc~h Security Dilemma 151
large ones could menace others. As noted above, the United States, and to a
lesser extent Great Britain, have partly been able to escape from the state of
nature because their geographical positions approximated this ideal.
Although geography cannot be changed to conform to borders, borders
can and d o change to conform to geography. Borders across which an
attack is easy tend to be unstable. States living within them are likely to
expand or be absorbed. Frequent wars are almost inevitable since attacking
will often seem the best way to protect what one has. This process will stop,
or a t least slow down, when the state's borders reach - by expansion or con-
traction - a line of natural obstacles. Security without attack will then be
possible. Furthermore, these lines constitute salient solutions to bargaining
problems and, to the extent that they are barriers to migration, are likely to
divide ethnic groups, thereby raising the costs and lowering the incentives
for conquest.
Attachment to one's state and its land reinforce one quasi-geographical
aid to the defense. Conquest usually becomes more difficult the deeper the
attacker pushes into the other's territory. Nationalism spurs the defenders
to fight harder; advancing not only lengthens the attacker's supply lines, but
takes him through unfamiliar and often devastated lands that require
troops for garrison duty. These stabilizing dynamics will not operate, how-
ever, if the defender's war materiel is situated near its borders, or if the
people d o not care about their state, but only about being on the winning
side. In such cases, positive feedback will be at work and initial defeats will
be i n ~ u r m o u n t a b l e . ~ "
Imitating geography, men have tried to create barriers. Treaties may pro-
vide for demilitarized zones on both sides of the border, although such zones
will rarely be deep enough to provide more than warning. Even this was not
possible in Europe, but the Russians adopted a gauge for their railroads that
was broader than that of the neighboring states, thereby complicating the
logistics problems of any attacker - including Russia.
Perhaps the most ambitious and at least temporarily successful attempts
to construct a system that would aid the defenses of both sides were the
interwar naval treaties, as they affected Japanese-American relations. As
mentioned earlier, the problem was that the United States could not defend
the Philippines without denying Japan the ability to protect her home
island^.^' (In 1941 this dilemma became insoluble when Japan sought to
extend her control to Malaya and the Dutch East Indies. If the Philippines
had been invulnerable, they could have provided a secure base from which
the U.S. could interdict Japanese shipping between the homeland and the
areas she was trying to conquer.) In the 1920's and early 1930's each side
would have been willing to grant the other security for its possessions in
return for a reciprocal grant, and the Washington Naval Conference agree-
ments were designed to approach this goal. As a Japanese diplomat later
put it, their country's "fundamental principle" was to have "a strength
insufficient for attack and adequate for defense.""' Thus, Japan agreed in
1922 to accept a navy only three-fifths as large as that of the United States,
and the U.S. agreed not to fortify its Pacific islands." (Japan had earlier
been forced to agree not to fortify the islands she had taken from Germany
in World War I.) Japan's navy would not be large enough to defeat
America's anywhere other than close to the home islands. Although the
Japanese could still take the Philippines, not only would they be unable to
move farther, but they might be weakened enough by their efforts to be vul-
nerable to counterattack. Japan, however, gained security. An American
attack was rendered more difficult because the American bases were unpro-
tected and because, until 1930, Japan was allowed unlimited numbers of
cruisers, destroyers, and submarines that could weaken the American fleet
as it made its way across the ocean.""
The other major determinant of the offense-defense balance is technol-
ogy. When weapons are highly vulnerable, they must be employed before
they are attacked. Others can remain quite invulnerable in their bases. The
former characteristics are embodied in unprotected missiles and many kinds
of bombers. (It should be noted that it is not vulnerability per se that is cru-
cial, but the location of the vulnerability. Bombers and missiles that are easy
to destroy only after having been launched toward their targets d o not cre-
ate destabilizing dynamics.) Incentives to strike first are usually absent for
naval forces that are threatened by a naval attack. Like missiles in hardened
silos, they are usually well protected when in their bases. Both sides can
then simultaneously be prepared to defend themselves successfully.
In ground warfare under some conditions, forts, trenches, and small
groups of men in prepared positions can hold off large numbers of attack-
ers. Less frequently, a few attackers can storm the defenses. By and large, it
is a contest between fortifications and supporting light weapons on the one
hand, and mobility and heavier weapons that clear the way for the attack on
the other. As the erroneous views held before the two world wars show, there
is no simple way to determine which is dominant. "[Tlhese oscillations are
not smooth and predictable like those of a swinging pendulum. They are
uneven in both extent and time. Some occur in the course of a single battle
or campaign, others in the course of a war, still others during a series of
wars." Longer-term oscillations can also be detected:
The early Gothic age, from the twelfth to the late thirteenth century, with
its wonderful cathedrals and fortified places, was a period during which
the attackers in Europe generally met serious and increasing difficulties,
because the improvement in the strength of fortresses outran the advance in
the power of destruction. Later, with the spread of firearms at the end of
the fifteenth century, old fortresses lost their power to resist. An age ensued
during which the offense possessed, apart from short-term setbacks, new
advantages. Then, during the seventeenth century, especially after about
1660, and until at least the outbreak of the War of the Austrian Succession
in 1740, the defense regained much of the ground it had lost since the
great medieval fortresses had proved unable to meet the bombardment of
the new and more numerous artillery.53
icZ;\ I* Security Dilemma 1 53
Another scholar has continued the argument: "The offensive gained an advan-
tage with new forms of heavy mobile artillery in the nineteenth century, but
the stalemate of World War I created the impression that the defense again had
an advantage; the German invasion in World War 11, however, indicated the
offensive superiority of highly mechanized armies in the field."'J
The situation today with respect to conventional weapons is unclear. Until
recently it was believed that tanks and tactical air power gave the attacker an
advantage. The initial analyses of the 1973 Arab-Israeli war indicated that
new anti-tank and anti-aircraft weapons have restored the primacy of the
defense. These weapons are cheap, easy to use, and can destroy a high pro-
portion of the attacking vehicles and planes that are sighted. It then would
make sense for a status-quo power to buy lots of $20,000 missiles rather than
buy a few half-million dollar tanks and multi-million dollar fighter-bombers.
Defense would be possible even against a large and well-equipped force;
states that care primarily about self-protection would not need to engage in
arms races. But further examinations of the new technologies and the history
of the October War cast doubt on these optimistic conclusions and leave us
unable to render any firm judgment."'
Concerning nuclear weapons, it is generally agreed that defense is impos-
sible - a triumph not of the offense, but of deterrence. Attack makes no sense,
not because it can be beaten off, but because the attacker will be destroyed in
turn. In terms of the questions under consideration here, the result is the equiva-
lent of the primacy of the defense. First, security is relatively cheap. Less than
one percent of the G.N.P. is devoted to deterring a direct attack on the United
States; most of it is spent on acquiring redundant systems to provide a lot of
insurance against the worst conceivable contingencies. Second, both sides can
simultaneously gain security in the form of second-strike capability. Third, and
related to the foregoing, second-strike capability can be maintained in the face
of wide variations in the other side's military posture. There is no purely mili-
tary reason why each side has to react quickly and strongly to the other's
increases in arms. Any spending that the other devotes to trying to achieve
first-strike capability can be neutralized by the state's spending much smaller
sums on protecting its second-strike capability. Fourth, there are no incentives
to strike first in a crisis.
Important problems remain, of course. Both sides have interests that go
well beyond defense of the homeland. The protection of these interests cre-
ates conflicts even if neither side desires expansion. Furthermore, the shift
from defense to deterrence has greatly increased the importance and per-
ceptions of resolve. Security now rests on each side's belief that the other
would prefer to run high risks of total destruction rather than sacrifice its
vital interests. Aspects of the security dilemma thus appear in a new form.
Are weapons procurements used as an index of resolve? Must they be so
used ? If one side fails to respond to the other's buildup, will it appear weak
and thereby invite predation? Can both sides simultaneously have images of
high resolve or is there a zero-sum element involved? Although these prob-
lems are real, they are not as severe as those in the prenuclear era: there are
many indices of resolve, and states d o not so much judge images of resolve
in the abstract as ask how likely it is that the other will stand firm in a par-
ticular dispute. Since states are most likely to stand firm on matters which
concern them most, it is quite possible for both to demonstrate their resolve
to protect their own security simultaneously.
Offense-Defense Differentiation
The other major variable that affects how strongly the security dilemma
operates is whether weapons and policies that protect the state also provide
the capability for attack. If they d o not, the basic postulate of the security
dilemma no longer applies. A state can increase its own security without
decreasing that of others. The advantage of the defense can only ameliorate
the security dilemma. A differentiation between offensive and defensive
stances comes close to abolishing it. Such differentiation does not mean,
however, that all security problems will be abolished. If the offense has the
advantage, conquest and aggression will still be possible. And if the offense's
advantage is great enough, status-quo powers may find it too expensive to
protect themselves by defensive forces and decide to procure offensive
weapons even though this will menace others. Furthermore, states will
still have to worry that even if the other's military posture shows that it is
peaceful now, it may develop aggressive intentions in the future.
Assuming that the defense is at least as potent as the offense, the differen-
tiation between them allows status-quo states to behave in ways that are
clearly different from those of aggressors. Three beneficial consequences fol-
low. First, status-quo powers can identify each other, thus laying the founda-
tions for cooperation. Conflicts growing out of the mistaken belief that the
other side is expansionist will be less frequent. Second, status-quo states will
obtain advance warning when others plan aggression. Before a state can
attack, it has to develop and deploy offensive weapons. If procurement of
these weapons cannot be disguised and takes a fair amount of time, as it
almost always does, a status-quo state will have the time to take counter-
measures. It need not maintain a high level of defensive arms as long as its
potential adversaries are adopting a peaceful posture. (Although being so
armed should not, with the one important exception noted below, alarm
other status-quo powers.) States do, in fact, pay special attention to actions
that they believe would not be taken by a status-quo state because they feel
that states exhibiting such behavior are aggressive. Thus the seizure or devel-
opment of transportation facilities will alarm others more if these facilities
have no commercial value, and therefore can only be wanted for military rea-
sons. In 1906, the British rejected a Russian protest about their activities in a
district of Persia by claiming that this area was "only of [strategic] import-
ance [to the Russians] if they wished to attack the Indian frontier, or to put
pressure upon us by making us think that they intend to attack it."j6
The same inferences are drawn when a state acquires more weapons
than observers feel are needed for defense. Thus, the Japanese spokesman
l ~ r ~ iSecurity
i Dilemma 155
at the 1930 London naval conference said that his country was alarmed by
the American refusal to give Japan a 70 percent ratio (in place of a 60 per-
cent ratio) in heavy cruisers: "As long as America held that ten percent
advantage, it was possible for her to attack. So when America insisted on
sixty percent instead of seventy percent, the idea would exist that they were
trying to keep that possibility, and the Japanese people could not accept
that."" Similarly, when Mussolini told Chamberlain in January 1939 that
Hitler's arms program was motivated by defensive considerations, the
Prime Minister replied that "German military forces were now so strong as
to make it impossible for any Power or combination of Powers to attack her
successfully. She could not want any further armaments for defensive pur-
poses; what then did she want them for?"'8
Of course these inferences can be wrong - as they are especially likely to
be because states underestimate the degree to which they menace others."
And when they are wrong, the security dilemma is deepened. Because the
state thinks it has received notice that the other is aggressive, its own arms
building will be less restrained and the chances of cooperation will he
decreased. But the dangers of incorrect inferences should not obscure the
main point: when offensive and defensive postures are different, much of
the ~~ncertainty about the other's intentions that contributes to the security
dilemma is removed.
The third beneficial consequence of a difference between offensive and
defensive weapons is that if all states support the status quo, an obvious
arms control agreement is a ban on weapons that are useful for attacking.
As President Roosevelt put it in his message to the Geneva Disarmament
Conference in 1933: "If all nations will agree wholly to eliminate from pos-
session and use the weapons which make possihle a successful attack,
defenses auton~atically will become impregnable, and the frontiers and
independence of every nation will become s e ~ u r e . " ~The " fact that such
treaties have been rare - the Washington naval agreements discussed above
and the anti-ABM treaty can be cited as examples - shows either that states
are not always willing to guarantee the security of others, or that it is hard
to distinguish offensive from defensive weapons.
Is such a distinction possible? Salvador de Madariaga, the Spanish states-
man active in the disarmament negotiations of the interwar years, thought
not: "A weapon is either offensive or defensive according to which end of it
YOLI art. looking at." The French Foreign Minister agreed (although French
~ o l i c ydid not always follow this view): "Every arm can be employed offen-
sively or defensively in turn. . . . The only way to discover whether arms are
intended for purely defensive purposes or are held in a spirit of aggression is
in all cases to enquire into the intentions of the country concerned." Some
evidence for the validity of this argument is provided by the fact that much
time in these unsuccessful negotiations was devoted to separating offensive
from defensive weapons. Indeed, no simple and unambiguous definition is
possible and in many cases no judgment can be reached. Before the American
entry into World War I, Woodrow Wilson wanted to arm merchantmen only
I56 T h e Cold War a n d Nuclear Deterrence
with guns in the back of the ship so they could not initiate a fight, but this
expedient cannot be applied to more common forms of armament^.^'
There are several problems. Even when a differentiation is possible, a
status-quo power will want offensive arms under any of three conditions,
(1)If the offense has a great advantage over the defense, protection through
defensive forces will be too expensive. (2) Status-quo states may need offen-
sive weapons to regain territory lost in the opening stages of a war. It might
be possible, however, for a state to wait to procure these weapons until war
seems likely, and they might be needed only in relatively small numbers,
unless the aggressor was able to construct strong defenses quickly in the
occupied areas. (3) The state may feel that it must be prepared to take the
offensive either because the other side will make peace only if it loses terri-
tory or because the state has commitments to attack if the other makes war
on a third party. As noted above, status-quo states with extensive commit-
ments are often forced to behave like aggressors. Even when they lack such
commitments, status-quo states must worry about the possibility that if
they are able to hold off an attack, they will still not be able to end the war
unless they move into the other's territory to damage its military forces and
inflict pain. Many American naval officers after the Civil War, for example,
believed that "only by destroying the commerce of the opponent could the
United States bring him to terms."62
A further complication is introduced by the fact that aggressors as well
as status-quo powers require defensive forces as a prelude to acquiring offen-
sive ones, to protect one frontier while attacking another, or for insurance in
case the war goes badly. Criminals as well as policemen can use bulletproof
vests. Hitler as well as Maginot built a line of forts. Indeed, Churchill reports
that in 1936 the German Foreign Minister said: "As soon as our fortifica-
tions are constructed [on our western borders] and the countries in Central
Europe realize that France cannot enter German territory, all these countries
will begin to feel very differently about their foreign policies, and a new con-
stellation will develop."63 So a state may not necessarily be reassured if its
neighbor constructs strong defenses.
More central difficulties are created by the fact that whether a weapon
is offensive or defensive often depends on the particular situation - for
instance, the geographical setting and the way in which the weapon is used.
"Tanks ... spearheaded the fateful German thrust through the Ardennes in
1940, but if the French had disposed of a properly concentrated armored
reserve, it would have provided the best means for their cutting off the pene-
tration and turning into a disaster for the Germans what became instead
an overwhelming victory."64 Anti-aircraft weapons seem obviously defen-
sive - to be used, they must wait for the other side t o come t o them. But the
Egyptian attack on Israel in 1973 would have been impossible without
effective air defenses that covered the battlefield. Nevertheless, some dis-
tinctions are possible. Sir John Simon, then the British Foreign Secretary, in
response to the views cited earlier, stated that just because a fine line could
not be drawn, "that was no reason for saying that there were not stretches
/{'IL I< Security Dilemma 157
of territory on either side which all practical men and women knew to be
well o n this or that side of the line." Although there are almost n o weapons
and strategies that are useful only for attacking, there are some that are
almost exclusively defensive. Aggressors could want them for protection,
hut a state that relied mostly on them could not menace others. More fre-
quently, we cannot "determine the absolute character of a weapon, but [we
can] make a comparison ... [and] discover whether or not the offensive
potentialities predominate, whether a weapon is more useful in attack or in
defense.""
The essence of defense is keeping the other side out of your territory.
A purely defensive weapon is one that can do this without being able to pene-
trate the enemy's land. Thus a committee of military experts in an interwar
disarmament conference declared that armaments "incapable of mobility by
means of self-contained power," or movable only after long delay, were
"only capable of being used for the defense of a State's territory.""" The most
obvious examples are fortifications. They can shelter attacking forces, espe-
cially when they are built right along the f r ~ n t i e r , ~but
; they cannot occupy
enemy territory. A state with only a strong line of forts, fixed guns, and a
small army to man them would not be much of a menace. Anything else that
can serve only as a barrier against attacking troops is similarly defensive. In
this category are systems that provide warning of an attack, the Russian's
adoption of a different railroad gauge, and nuclear land mines that can seal
off invasion routes.
If total immobility clearly defines a system that is defensive only, limited
mobility is unfortunately ambiguous. As noted above, shortrange fighter
aircraft and anti-aircraft missiles can be used to cover an attack. And,
unlike forts, they can advance with the troops. Still, their inability to reach
deep into enemy territory does make them more useful for the defense than
for the offense. Thus, the United States and Israel would have been more
alarmed in the early 1970's had the Russians provided the Egyptians with
long-range instead of short-range aircraft. Naval forces are particularly dif-
ficult to classify in these terms, but those that are very short-legged can be
used only for coastal defense.
Any forces that for various reasons fight well only when on their own
soil in effect lack mobility and therefore are defensive. The most extreme
example would be passive resistance. Noncooperation can thwart an
aggressor, but it is very hard for large numbers of people to cross the bor-
der and stage a sit-in o n another's territory. Morocco's recent march on the
Spanish Sahara approached this tactic, but its success depended on special
circumstances. Similarly, guerrilla warfare is defensive to the extent to
which it requires civilian support that is likely to be forthcoming only in
opposition to a foreign invasion. Indeed, if guerrilla warfare were easily
exportable and if it took ten defenders to destroy each guerrilla, then this
weapon would not only be one which could be used as easily to attack the
other's territory as to defend one's own, but one in which the offense had
the advantage: s o the security dilemma would operate especially strongly.
If guerrillas are unable to fight on foreign soil, other kinds of armies may
be unwilling to do so. An army imbued with the idea that only defensive wars
were just would fight less effectively, if at all, if the goal were conquest.
Citizen militias may lack both the ability and the will for aggression. The
weapons employed, the short term of service, the time required for mobiliza-
tion, and the spirit of repelling attacks on the homeland, all lend themselves
much more to defense than to attacks on foreign territory.hx
Less idealistic motives can produce the same result. A leading student of
medieval warfare has described the armies of that period as follows:
"Assembled with difficulty, insubordinate, unable to maneuver, ready to melt
away from its standard the moment that its short period of service was over,
a feudal force presented an assemblage of unsoldierlike qualities such as have
seldom been known to coexist. Primarily intended to defend its own borders
from the Magyar, the Northman, or the Saracen ..., the institution was utterly
unadapted to take the o f f e n ~ i v e . "Some
~ ~ political groupings can be similarly
described. International coalitions are more readily held together by fear than
by hope of gain. Thus Castlereagh was not being entirely self-serving when in
1816 he argued that the Quadruple Alliance "could only have owed its origin
to a sense of common danger; in its very nature it must be conservative; it can-
not threaten either the security or the liberties of other States."'O It is no acci-
dent that most of the major campaigns of expansion have been waged by
one dominant nation (for example, Napoleon's France and Hitler's Germany),
and that coalitions among relative equals are usually found defending the
status quo. Most gains from conquest are too uncertain and raise too many
questions of future squabbles among the victors to hold an alliance together
for long. Although defensive coalitions are by no means easy to maintain -
conflicting national objectives and the free-rider problem partly explain why
three of them dissolved before Napoleon was defeated - the common interest
of seeing that no state dominates provides a strong incentive for solidarity.
Weapons that are particularly effective in reducing fortifications and
barriers are of great value to the offense. This is not to deny that a defen-
sive power will want some of those weapons if the other side has them:
Brodie is certainly correct to argue that while their tanks allowed the
Germans to conquer France, properly used French tanks could have halted
the attack. But France would not have needed these weapons if Germany
had not acquired them, whereas even if France had no tanks, Germany
could not have foregone them since they provided the only chance of break-
ing through the French lines. Mobile heavy artillery is, similarly, especially
useful in destroying fortifications. The defender, while needing artillery to
fight off attacking troops or t o counterattack, can usually use lighter guns
since they d o not need to penetrate such massive obstacles. So it is not sur-
prising that one of the few things that most nations at the interwar disarma-
ment conferences were able to agree on was that heavy tanks and mobile
heavy guns were particularly valuable t o a state planning an attack."
Weapons and strategies that depend for their effectiveness on surprise are
almost always offensive. That fact was recognized by some of the delegates to
I C I L I ~ Security Dilemma 159
the interwar disarmament conferences and is the principle behind the common
national ban on concealed weapons. An earlier representative of this wide-
spread view was the mid-19th-century Philadelphia newspaper that argued:
"As a measure of defense, knives, dirks, and sword canes are entirely useless.
They are fit only for attack, and all such attacks are of murderous character.
Whoever carries such a weapon has prepared himself for homicide.""
I t is, of course, not always possible to distinguish between forces that are
most effective for holding territory and forces optimally designed for taking
it. Such a distinction could not have been made for the strategies and
weapons in Europe during most of the period between the Franco-Prussian
War and World War I. Neither naval forces nor tactical air forces can be read-
ily classified in these terms. But the point here is that when such a distinction
is possible, the central characteristic of the security dilemma no longer holds,
and one of the most troublesome consequences of anarchy is removed.
Offinse-Defense Differentiation and Strategic Nuclear Weapons. In the
interwar period, most statesmen held the reasonable position that weapons
that threatened civilians were o f f e n s i ~ e . 'But
~ when neither side can protect
its civilians, a counter-city posture is defensive because the state can cred-
ibly threaten to retaliate only in response to an attack on itself or its closest
allies. The costs of this strike are so high that the state could not threaten
to use it for the less-than-vital interest of compelling the other to abandon
an established position.
In the context of deterrence, offensive weapons are those that provide
defense. In the now familiar reversal of common sense, the state that could
take its population out of hostage, either by active or passive defense or by
destroying the other's strategic weapons on the ground, would be able to alter
the status quo. The desire to prevent such a situation was one of the ration-
ales for the anti-ABM agreements; it explains why some arms controllers
opposed building ABM's to protect cities, but favored sites that covered
ICBM fields. Similarly, many analysts want to limit warhead accuracy and
favor n~ultiplere-entry vehicles (MRV's), but oppose multiple independently
targetable re-entry vehicles (MIRV's). The former are more useful than single
warheads for penetrating city defenses, and ensure that the state has a second-
strike capability. MIRV's enhance counterforce capabilities. Some arms con-
trollers argue that this is also true of cruise missiles, and therefore do not want
them to be deployed either. There is some evidence that the Russians are not
satisfied with deterrence and are seeking to regain the capability for defense.
Such an effort, even if not inspired by aggressive designs, would create a severe
security dilemma.
What is most important for the argument here is that land-based ICBM's
are both offensive and defensive, but when both sides rely on Polaris-type
systems (SLBM's), offense and defense use different weapons. ICBM's can be
used either to destroy the other's cities in retaliation or to initiate hostilities
by attacking the other's strategic missiles. Some measures - for instance,
hardening of missile sites and warning systems - are purely defensive, since
they do not make a first strike easier. Others are predominantly offensive - for
instance, passive or active city defenses, and highly accurate warheads.

But ICBM's themselves are useful for both purposes. And because states seek
a high level of insurance, the desire for protection as well as the contempla-
tion of a counterforce strike can explain the acquisition of extremely large
numbers of missiles. So it is very difficult to infer the other's intentions from
its military posture. Each side's efforts to increase its own security by procur-
ing more missiles decreases, to an extent determined by the relative efficacy
of the offense and the defense, the other side's security. That is not the case
when both sides use SLBM's. The point is not that sea-based systems are less
vulnerable than land-based ones (this bears on the offense-defense ratio) but
that SLBM's are defensive, retaliatory weapons. First, they are probably not
accurate enough to destroy many military target^.'^ Second, and more
important, SLBM's are not the main instrument of attack against other
SLBM's. The hardest problem confronting a state that wants to take its cities
out of hostage is to locate the other's SLBM's, a job that requires not SLBM's
but anti-submarine weapons. A state might use SLBM's to attack the other's
submarines (although other weapons would probably be more efficient), but
without anti-submarine warfare (ASW) capability the task cannot be per-
formed. A status-quo state that wanted to forego offensive capability could
simply forego ASW research and procurement.
There are two difficulties with this argument, however. First, since the
state's SLBM's are potentially threatened by the other's ASW capabilities, the
state may want to pursue ASW research in order to know what the other
might be able to do and to design defenses. Unless it does this, it cannot be
confident that its submarines are safe. Second, because some submarines are
designed to attack surface ships, not launch missiles, ASW forces have mis-
sions other than taking cities out of hostage. Some U.S. officials plan for a
long war in Europe which would require keeping the sea lanes open against
Russian submarines. Designing an ASW force and strategy that would meet
this threat without endangering Soviet SLBM's would be difficult but not
impossible, since the two missions are somewhat different.'"urthermore,
the Russians do not need ASW forces to combat submarines carrying out
conventional missions; it might be in America's interest to sacrifice the abil-
ity to meet a threat that is not likely to materialize in order to reassure the
Russians that we are not menacing their retaliatory capability.
When both sides rely on ICBM's, one side's missiles can attack the other's,
and so the state cannot be indifferent to the other's building program. But
because one side's SLBM's do not menace the other's, each side can build as
many as it wants and the other need not respond. Each side's decision on the
size of its force depends on technical questions, its judgment about how much
destruction is enough to deter, and the amount of insurance it is willing to pay
for - and these considerations are independent of the size of the other's stra-
tegic force. Thus the crucial nexus in the arms race is severed.
Here two objections not only can be raised but have been, by those who
feel that even if American second-strike capability is in no danger, the
United States must respond to a Soviet buildup. First, the relative numbers of
missiles and warheads may be used as an index of each side's power and will.
It I: I. Security Dilemma 16 1
Even i f there is no military need to increase American arms as the Russians

increase theirs, a failure to respond may lead third parties to think that the
U.S. has abandoned the competition with the U.S.S.R. and is no longer will-
ing to pay the price of world leadership. Furthermore, if either side believes
that nuclear "superiority" matters, then, through the bargaining logic, it
will matter. The side with "superiority" will be more likely to stand firm
in a confrontation if it thinks its "stronger" military position helps it, or if it
thinks that the other thinks its own "weaker" military position is a handicap.
To allow the other side to have more SLBM's - even if one's own second-
strike capability is unimpaired - will give the other an advantage that can
be translated into political gains.
The second objection is that superiority does matter, and not only because
of mistaken beliefs. If nuclear weapons are used in an all-ornone fashion, then
all that is needed is second-strike capability. But limited, gradual, and con-
trolled strikes are possible. If the other side has superiority, it can reduce the
state's forces by a slow-motion war of attrition. For the state to strike at the
other's cities would invite retaliation; for it to reply with a limited counter-
force attack would further deplete its supply of missiles. Alternatively, the
other could employ demonstration attacks - such as taking out an isolated
military base or exploding a warhead high over a city - in order to demon-
strate its resolve. In either of these scenarios, the state will suffer unless it
matches the other's arms p o ~ t u r e . ' ~
These two objections, if valid, mean that even with SLBM's one cannot
distinguish offensive from defensive strategic nuclear weapons. Compellence
may be more difficult than deterrence," but if decision makers believe that
numbers of missiles or of warheads influence outcomes, or if these weapons
can be used in limited manner, then the posture and policy that would be
needed for self-protection is similar to that useful for aggression. If the
second objection has merit, security would require the ability to hit selected
targets on the other side, enough ammunition to wage a controlled counter-
force war, and the willingness to absorb limited countervalue strikes. Secretary
Schlesinger was correct in arguing that this capability would not constitute a
first-strike capability. But because the "Schlesinger Doctrine" could be used
not only to cope with a parallel Russian policy, but also to support an
American attempt to change the status quo, the new American stance would
decrease Russian security. Even if the U.S.S.R. were reassured that the pres-
ent U.S. Government lacked the desire or courage to do this, there could be
no guarantee that future governments would not use the new instruments for
expansion. Once we move away from the simple idea that nuclear weapons
can only be used for all-out strikes, half the advantage of having both sides
rely on a sea-based force would disappear because of the lack of an offensive-
defensive differentiation. To the extent that military policy affects political
relations, it would be harder for the United States and the Soviet Union to
cooperate even if both supported the status quo.
Although a full exploration of these questions is beyond the scope of
this paper, it should be noted that the objections rest on decision makers'
beliefs - beliefs, furthermore, that can be strongly influenced by American
policy and American statements. The perceptions of third nations of whether

the details of the nuclear balance affect political conflicts - and, to a lesser
extent, Russian beliefs about whether superiority is meaningful - are largely
derived from the American strategic debate. If most American spokesmen
were t o take the position that a secure second-strike capability was sufficient
and that increments over that (short of a first-strike capability) would only be
a waste of money, it is doubtful whether America's allies or the neutrals
would judge the superpowers' useful military might or political will by the
size of their stockpiles. Although the Russians stress war-fighting ability, they
have not contended that marginal increases in strategic forces bring political
gains; any attempt to do so could be rendered less effective by an American
assertion that this is nonsense. The bargaining advantages of possessing
nuclear "superiority" work best when both sides acknowledge them. If the
"weaker" side convinces the other that it does not believe there is any mean-
ingful difference in strength, then the "stronger" side cannot safely stand
firm because there is no increased chance that the other will back down.
This kind of argument applies at least as strongly to the second objection.
Neither side can employ limited nuclear options unless it is quite confident
that the other accepts the rules of the game. For if the other believes that
nuclear war cannot be controlled, it will either refrain from responding - which
would be fine - or launch all-out retaliation. Although a state might be ready
to engage in limited nuclear war without acknowledging this possibility - and
indeed, that would be a reasonable policy for the United States - it is not
likely that the other would have sufficient faith in that prospect to initiate
limited strikes unless the state had openly avowed its willingness to fight
this kind of war. So the United States, by patiently and consistently explain-
ing that it considers such ideas to be mad and that any nuclear wars will
inevitably get out of control, could gain a large measure of protection
against the danger that the Soviet Union might seek to employ a "Schlesinger
Doctrine" against an America that lacked the military ability or political
will to respond in kind. Such a position is made more convincing by the
inherent implausibility of the arguments for the possibility of a limited
nuclear war.
In summary, as long as states believe that all that is needed is second-
strike capability, then the differentiation between offensive and defensive
forces that is provided by reliance on SLBM's allows each side to increase
its security without menacing the other, permits some inferences about
intentions to be drawn from military posture, and removes the main
incentive for status-quo powers to engage in arms races.
IV. Four W o r l d s
The two variables we have been discussing - whether the offense or the
defense has the advantage, and whether offensive postures can be distin-
guished from defensive ones - can be combined to yield four possible worlds.
Ierwi Security Dilemma 163
OFFENSE HAS L)EFENSE H A S

T H E ADVANTAGE T H E ADVANTAGE
OFFENSIVE POSI'IJKE Security d~lcmrna,h u t

N O T IIISTINGIJ1SHABI.E security requirements
F R O M DEFENSIVE ONE n i a y hr cornpat~hle.
3 4
N o securlty d~lernrna,hut
OFFENSIVE POSTURE aggression poss~hle.
DISTINGUISHABLE Status-quo states can follow Doubly st<~ble
FROM LIEFENSIVF. O N E different policy than
aggressors.
W'lrning given.
The first world is the worst for status-quo states. There is no way to get
security without menacing others, and security through defense is terribly
difficult to obtain. Because offensive and defensive postures are the same,
status-quo states acquire the same kind of arms that are sought by aggres-
sors. And because the offense has the advantage over the defense, attacking
is the best route to protecting what you have; status-quo states will there-
fore hehave like aggressors. The situation will be unstable. Arms races are
likely. Incentives to strike first will turn crises into wars. Decisive victories
and conquests will be common. States will grow and shrink rapidly, and it
will be hard for any state to maintain its size and influence without trying
to increase them. Cooperation among status-quo powers will be extremely
hard to achieve.
There are no cases that totally fit this picture, but it bears more than a
passing resemblance to Europe before World War I. Britain and Germany,
although in many respects natural allies, ended up as enemies. Of course
much of the explanation lies in Germany's ill-chosen policy. And from the
perspective of our theory, the powers' ability to avoid war in a series of earl-
ier crises cannot be easily explained. Nevertheless, much of the behavior in
this period was the product of technology and beliefs that magnified the
security dilemma. Decision makers thought that the offense had a big
advantage and saw little difference between offensive and defensive military
postures. The era was characterized by arms races. And once war seemed
likely, mobilization races created powerful incentives to strike first.
In the nuclear era, the first world would be one in which each side relied
on vulnerable weapons that were aimed at similar forces and each side
understood the situation. In this case, the incentives to strike first would
be very high - so high that status-quo powers as well as aggressors would
be sorely tempted to pre-empt. And since the forces could be used to change
the status quo as well as to preserve it, there would be no way for both sides
to increase their security simultaneously. Now the familiar logic of deter-
rence leads both sides to see the dangers in this world. Indeed, the new
understanding of this situation was one reason why vulnerable bombers
and missiles were replaced. Ironically, the 1950's would have been more
hazardous if the decision makers had been aware of the dangers of their
posture and had therefore felt greater pressure to strike first. This situation
could be recreated if both sides were to rely on MIRVed ICBM's.
In the second world, the security dilemma operates because offensive and
defensive postures cannot be distinguished; but it does not operate as strongly
as in the first world because the defense has the advantage, and so an
increment in one side's strength increases its security more than it decreases
the other's. So, if both sides have reasonable subjective security requirements,
are of roughly equal power, and the variables discussed earlier are favorable,
it is quite likely that status-quo states can adopt compatible security
policies. Although a state will not be able to judge the other's intentions
from the kinds of weapons it procures, the level of arms spending will give
important evidence. Of course a state that seeks a high level of arms might
be not an aggressor but merely an insecure state, which if conciliated will
reduce its arms, and if confronted will reply in kind. To assume that the
apparently excessive level of arms indicates aggressiveness could therefore
lead to a response that would deepen the dilemma and create needless
conflict. But empathy and skillful statesmanship can reduce this danger.
Furthermore, the advantageous position of the defense means that a status-
quo state can often maintain a high degree of security with a level of arms
lower than that of its expected adversary. Such a state demonstrates that it
lacks the ability or desire to alter the status quo, at least at the present time.
The strength of the defense also allows states to react slowly and with
restraint when they fear that others are menacing them. So, although status-
quo powers will to some extent be threatening to others, that extent will be
limited.
This world is the one that comes closest to matching most periods in his-
tory. Attacking is usually harder than defending because of the strength of
fortifications and obstacles. But purely defensive postures are rarely pos-
sible because fortifications are usually supplemented by armies and mobile
guns which can support an attack. In the nuclear era, this world would be
one in which both sides relied on relatively invulnerable ICBM's and
believed that limited nuclear war was impossible. Assuming no MIRV's, it
would take more than one attacking missile to destroy one of the adver-
sary's. Pre-emption is therefore unattractive. If both sides have large inven-
tories, they can ignore all but drastic increases on the other side. A world
of either ICBM's or SLBM's in which both sides adopted the "Schlesinger
Doctrine" would probably fit in this category too. The means of preserving
the status quo would also be the means of changing it, as we discussed ear-
lier. And the defense usually would have the advantage, because compellence
is more difficult than deterrence. Although a state might succeed in changing
the status quo on issues that matter much more to it than to others, status-quo
powers could deter major provocations under most circumstances.
i c L ti 5 Security Dilemma I65
In the third world there may be no security dilemma, but there are security
problems. Because states can procure defensive systems that do not threaten
others, the dilemma need not operate. But because the offense has the advan-
tage, aggression is possible, and perhaps easy. If the offense has enough of an
advantage, even a status-quo state may take the initiative rather than risk
being attacked and defeated. If the offense has less of an advantage, stability
and cooperation are likely because the status-quo states will procure defen-
sive forces. They need not react to others who are similarly armed, but can
wait for the warning they would receive if others started to deploy offensive
weapons. But each state will have to watch the others carefully, and there is
room for false suspicions. The costliness of the defense and the allure of the
offense can lead to unnecessary mistrust, hostility, and war, unless some of the
variables discussed earlier are operating to restrain defection.
A hypothetical nuclear world that would fit this description would be
one in which both sides relied on SLBM's, but in which ASW techniques
were very effective. Offense and defense would be different, but the former
would have the advantage. This situation is not likely to occur; but if it did,
a status-quo state could show its lack of desire to exploit the other by
refraining from threatening its submarines. The desire to have more pro-
tecting you than merely the other side's fear of retaliation is a strong one,
however, and a state that knows that it would not expand even if its cities
were safe is likely to believe that the other would not feel threatened by its
ASW program. It is easy to see how such a world could become unstable,
and how spirals of tensions and conflict could develop.
The fourth world is doubly safe. The differentiation between offensive and
defensive systems permits a way out of the security dilemma; the advantage
of the defense disposes of the problems discussed in the previous paragraphs.
There is no reason for a status-quo power to be tempted to procure offensive
forces, and aggressors give notice of their intentions by the posture they
adopt. Indeed, if the advantage of the defense is great enough, there are n o
security problems. The loss of the ultimate form of the power to alter the sta-
tus quo would allow greater scope for the exercise of nonmilitary means and
probably would tend to freeze the distribution of values.
This world would have existed in the first decade of the 20th century if
the decision makers had understood the available technology. In that case,
the European powers would have followed different policies both in the
long run and in the summer of 1914. Even Germany, facing powerful en-
emies on both sides, could have made herself secure by developing strong
defenses. France could also have made her frontier almost impregnable.
Furthermore, when crises arose, no one would have had incentives to strike
first. There would have been no competitive mobilization races reducing the
time available for negotiations.
In the nuclear era, this world would be one in which the superpowers
relied on SL.BM's, ASW technology was not up to its task, and limited
nuclear options were not taken seriously. We have discussed this situation
earlier; here we need only add that, even if our analysis is correct and even
if the policies and postures of both sides were to move in this direction, the
problem of violence below the nuclear threshold would remain. On issues
other than defense of the homeland, there would still be security dilemmas
and security problems. But the world would nevertheless be safer than it has
usually been.
Author's N o t e
I am grateful to Robert Art, Bernard Brodie, and Glenn Snyder for comments, and to the
Committee on Research of the UCLA Academic Senate for financial support. An earlier ver-
sion of this essay appeared as Working Paper No. 5, UCLA Program in Arms Control and
International Security.
Notes
1. This kind of rank-ordering is not entirely an analyst's invention, as is shown by the fol-
lowing section of a British army memo of 1903 deal~ngwith British and Russian railroad con-
struction near the Persia-Afghanistan border:
The conditions of the problem may ... be briefly summarized as follows:
a ) IF we make a railway t o Seistan while Russia remains inactive, we gain a considerable
defensive advantage at considerable financial cost;
b) If Russia makes a railway to Seistan, while we remain inact~ve,she gains a considerable
offensive advantage at considerable financial cost;
c) If both we and Russia make railways to Seistan, the defens~veand offensive advantages
may be held t o neutralize each other; in other words, we shall have spent a good deal of money
and be n o better off than we are at present. O n the other hand, we shall be no worse off,
whereas under alternative ( b ) we shall be much worse off. Consequently, the theoretical bal-
ance of advantage lies with the proposed radway extension from Quetta to Selstan.
W.G. Nicholson, "Memorandum on Seistan and Other Points Raised in the Discussion on the
Defence of India," (Committee of Imperial Defence, March 20, 1903). It should be noted that
the possibility of neither side building railways was nor mentioned, thus strongly biasing the
analysis.
2. Paul Schroeder, Metternlch's D~plomacya t Its Zenith, 1820-182.3 (Westport, Conn.:
Greenwood Press 1969), 126.
3. Quoted in Michael Howard, The Continental Commrtment (Harmondsworth, England:
Penguin 19741, 67.
4. Quoted in Gerald Wheeler, Prelude to Pearl Harbor (Columbia: University of M m o u r i
Press 1963), 167.
5. Quoted in Leonard Wainstein, "The Dreadnought Gap," in Robert Art and Kenneth
Waltz, eds., The Use of Force (Boston: Little, Brown 1971), 155; Raymond Sontag, European
Diplomatic History, 1871-1932 (New York: Appleton-Century-Crofts 1933), 147. The
French had made a similar argument 50 years earlier; see James Phinney Baxter 111, The
Introduction of the Ironclad Warship (Cambridge: Harvard University Press 1933), 149. For
a more detailed discussion of the security ddemma, see Jervis, Perception and Misperception
in International Politics (Princeton: Princeton University Press 19761, 62-76.
6. Experimental evidence for this proposition is summarized in James Tedeschi, Barry
Schlenker, and Thomas Bonoma, Conflict, Power, and Games (Chicago: Aldine 1973), 135-41.
l c i v ~ s Security Dilemma I67
7. T h e results of Prisoner's Dilemma games played in the laboratory support this argu-
ment. See Anarol Kapoport a n d Albert C h a m m a h , Prisoner's Dilemma (Ann Arbor: University
of Michigan I'ress I 9 6 5 ) , 33-50. Also see Kohert Axelrod, Conflict of lnterest (Chicago:
M a r k h a m l 9 7 0 ) , 60-70.
8. Q u o t e d in Bernard Brodie, Strategy in the Missile Age (Princeton: Princeton University
I'ress 195Y), 6 .
9. Herbert York, The Aduisors: Oppenhernrer, Teller, and the Superbonrb (San Franc~sco:
Freeman 19 7 6 ) , 56-60.
10. For the development of the concept of suhject~vesecurity, see Arnold Wolfers, Drscorti
illid Collirboration (Baltimore: Johns Hopkins Press 1962), chap. 10. In the present section we
assume that the state believes that ~ t security s can he hest served by increasing its arms; later
we will d ~ s c u s ssome o f the conditions under which this s s u m p t l o n does not hold.
11. T h e question of when a n actor will see another as a threat is important m d under-
studied. For a valuahle treatment (although o n e marred by serious methodologic,~l fl,~ws),
see Raymond Cohen, "Threat Perception in International Kelations," Ph.1). d ~ s s .( H e l m \ \
University 1974). Among the important factors, touched o n below, are the lessons from the
previous war.
12. Still the hest treatment is Arnold Wolfers, Hrztairr and France Bettiwn TILVJ Wlrs (New
York: Harcourt, Br,~ce1 9 4 0 ) .
13. Quoted in I'eter Gretton, Former N a l d Person (London: Cassell 1968), 15 1.
14. Michael Kaufman, "Tension Increases in French Colon!;" New York Tntrrs. July I I, 1976.
15. Experimental support f o r thls argument is summarized in M o r t o n L)eut\cii, T h ,
Kcsr~lrrtroriof Conf1ic.t ( N e w Haven: Yale University Press 1973). 18 1-95,
16. R o g u Bullen, Pn/nierstotr Guizot, and tlw Collapsr of the E r r t ~ n t(:ordral~ ~ (1.01idon:
Athlonc 1'1-e\s 19741, 81, 88, 9.3, 2 12. For 3 different vlew of this case, w e Stanley Mellon,
"Fnrenre, Il~plomacy,and tanrasy," Rez~irwsin European Hzstory, 1 1 (September 19761, 7 6 - 8 0 .
17. S~tnll.lrly, a Ft-ench diplomat h,ts argued that "the worst result o f Loui\ XIV's ,lh,~n-
donrnenr of o u r t r a d ~ t i o n a lpolicy w a s rhe distrust it aroused towards us ahro,ld." Jules
L l m b o n , "The Perni,~nenr B ~ s c sof French Foreign Pol~cp," F i ~ r e i pAffinrs. Vlll ( J a n u a r y
I Y 30), 1-9.
18. Thi\ a\\umec, however, that these benefits t o the other w ~ l nor l s o Improve the othet-'\
power posltlon that ~t will be more able t o menace the state in the future.
19. W',ilter I.aFeher, ed., T l ~ c1)ynanrics of World Power; A llocarnrrrrtn~Hrstory o/
Clrirtcd StLrtesFor~lgllPolrry 194 5-1 97.3, 11: Eastern Ertrope and t l ~ cSorwt IJn~on( N C U York:
Chelsea House in association w ~ t hhlcGraw-Hill 197.3), 700.
2 0 . T h o m a s Schelling, T h e Strategy of C:onflrct ( N e w York: O x t o r d U r i ~ v c r \ ~ t Press
y
IL)h3), 134-35.
2 I. (1.5. (:ongres\, Senate, Committee o n Foreign Relations, Hearin,qs, k'r)rt/~Atl~7trtli
lkwt): 8 l st Cong., I at sess. ( 19491, 17.
22. B r ~ ~ cKussett
e a n d Elizabeth Hanson, Interest m d Idrolog)! (San Fra~iclsco:F r e e m m
I 9 7 i ) , 260; Morris l a n o w i t r , The Professional Soldrrr ( N e w York: Free Press 1 9 6 0 ) , chap. 13.
2.1. tiahti, O n T/~erwronrt~lear Wilr (Princeton: P r ~ n c e t o nUniversity Press I9hO), 138-60.
It should he noted t h ~ the t French example is largely hypothetical because Fr'lnce had n o intell-
tlon of tulhlling her obligations once Germany became strong.
24. Wolfers (fn. 9 ) , chap. 1.5; C. Vann Woodward, "The Age o f Keinterpret,ir~ot~,"
A111~rict771 Historical Rer&ri? Vol. 6 7 (October 1960), 1-1 9.
2.5. W ~ l l i a ~ Braisted,
n The United States N a l y rtr the P'~cific. 1897-1909 (Austin:
Ilnivers~tyof Teuas Press 19581, 240.
26. Carnhon (fn. 171, 18.5.
27. Quoted in Adam IJlatn, Exparrsrorr and Co-/-krstence ( N e w h r k : I'r,ieger I %8), 5. In
1920 the US. Navy's General Board similarly declared "A nation must advance o r retrocede
I I I world position." Quoted in W ~ l l i m iBraisted, T ~ J Unrtrd C States N ' Z ~InI ~thr I'~7crfic-.
1909-1922 (Austin: University o f Texas I'ress 1971), 488.
28. Quoted in M a r ~ o nBoggs, Attempts to Define and Lnnlt "Aggressi~v"Arnrnnrrrit 171
Diplowrilcy mzd Strntcgy (Columbia: [Jniversity of Missouri Studies, XVI, No. 1, 194 1 ) 4 1 .
29. Steven Ross, European Diplomatic History, 1789-181.5 (Garden City, N.Y.:
Doubleday 1969), 194.
30. Thus, when Wolfers (fn. l o ) , 126, argues that a status-quo state that settles for rough
equality of power with its adversary, rather than seeking preponderance, may be able to con-
vince the other t o reciprocate by showing rhat it wants only to protect itself, not menace the
other, he assumes that the defense has an advantage.
31. Schelling (fn. 20), chap. 9.
32. Quoted in First Fischer, War of illusions (New York: Norton 19751, 377, 461.
33. George Quester, Offense and Defense in the International System (New York: John
Wiley 1977), 105-06; Sontag (fn. 5), 4-5.
34. Kahn (fn. 23), 21 1 (also see 144).
35. For a general discussion of such mistaken learning from the past, see Jervis (fn. 5 ) ,
chap. 6. The important and still not completely understood question of why this belief formed
and was maintained throughout the war is examined in Bernard Brodie, War and Polrtrcs
(New York: Macmillan 1973), 262-70; Brodie, "Technological Change, Strategic Doctrme,
and Political Outcomes," in Klaus Knorr, ed., Historical Dimensions of National Security
Problems (Lawrence: University Press of Kansas 1976), 290-92; and Douglas Porch, "The
French Army and the Spirit of the Offensive, 1900-14," in Brian Bond and Ian Roy, eds., War
and Society (New York: Holmes & Meier 197.5), 117-43.
36. Some were not so optimistic. Gray's remark is well-known: "The lamps are going out
all over Europe; we shall not see them lit again in our life-time." The German Prime Minister,
Bethmann Hollweg, also feared the consequences of the war. But the controlling view was rhat
it would certainly pay for the winner.
37. Quoted in Martin Gilbert, Winston S. Churchill, 111, The Challenge of War, 1916-1 9 16
(Boston: Houghton Mifflin 1971), 84.
38. Quester (fn. 33), 98-99. Robert Art, The Influence of Foreign Policy on Seapouw, I1
(Beverly Hills: Sage Professional Papers in International Studies Series, 1973), 14-1 8, 26-28.
39. Konard Jarausch, "The ILlusion of Limited War: Chancellor Bethmann Hollweg's
Calculated Risk, July 1914," Central European History, 11 (March 1969), 50.
40. Brodie (fn. 8), 58.
41. President Roosevelt and the American delegates to the League of Nat~onsDisarmament
Conference maintained that the tank and mobile heavy artillery had re-established the domi-
nance of the offensive, thus making disarmament more urgent (Boggs, fn. 28, pp. 3 1, 108), but
this was a minority position and may not even have been believed by the Americans. The reduced
prestige and influence of the military, and the high pressures to cut government spending
throughout this period also contributed to the lowering of defense budgets.
42. Jon Kimche, The Unfought Battle (New York: Stein 1968); Nicholas William Bethell, The
War Hitler Won: The Fall of Poland, September 1939 (New York: Holt 1972); Alan Alexandroff
and Richard Rosecrance, "Deterrence in 1939," World Politics, XXIX (April 1977), 4 0 4 2 4 .
43. Roderick Macleod and Denis Kelly, eds., Time Unguarded: The Ironside Diaries,
1937-1940 (New York: McKay 1962), 173.
44. For a short time, as France was falling, the British Cabinet d ~ discuss
d reaching a nego-
tiated peace with Hitler. The officlal history ignores this, but it 1s covered in P.M.H. Bell, A
Certain Eventuality (Farnborough, England: Saxon House 1974), 4 0 4 8 .
45. Macleod and Kelly (fn. 43), 174. In flat contrad~ctionto common sense and almost
everything they believed about modern warfare, the Allies planned an expedition to
Scandinavia to cut the supply of iron ore to Germany and to aid F~nlandagainst the Russians.
But the dominant mood was the one described above.
46. Brodie (fn. 8), 179.
47. Arthur Balfour, "Memorandum," Committee on Imperial Defence, April 30, 1903,
pp. 2-3; see the telegrams by Sir Arthur Nicolson, in G.P. Gooch and Harold Temperley, eds.,
British Documents on the Origins of the War, Vol. 4 (London: H.M.S.O. 1929), 429, 524.
These barriers d o not prevent the passage of long-range aircraft; but even in the air, distance
usually aids the defender.
Icn I . Security Dilemma I69
48. See for example, the discussion of warfare among Ch~nesewarlords in Hsi-Sheng Chi,
"The Chinese Warlord System as a n International System," in Morton Kaplan, ed., N e w
Apprr~dchest o Internatrond Relat~ons(New York: St. Martin's 1968), 405-25.
49. Some American decision makers, including military officers, thought that the best way
out of the dilemma was t o abandon the Philippines.
SO. Quoted in Eking Morrison, Turmod and Tradition: A Study of the Lrfe awd Tinzes of
Henry 1.. Stirnson (Boston: Houghton Mifflin 1960), 326.
5 1. The U S . "refused to consider limitations on Hawaiian defenses, rmce these works
posed n o threat to Japan." Braisted (fn. 27), 612.
52. That is part of the reason why the Japanese admirals strongly objected when the civ~l-
Ian leaders decided to accept J seven-to-ten ratio in lighter craft in 1930. Stephen I'elz, Rnc-e
ro Pearl Harbor (Cambridge: Harvard Univers~tyPress 1974), 3.
53. John Nef, War a i ~ dHuman Progress (New York: Norton 1963), 185. Also see ihid.,
2.17, 242-43, and 323; C. W. Oman, T / JArt ~ of War in the Middle Ages (Ithaca, N.Y.: Cornell
LJnivers~tyPress 19.531, 70-72; J o h n Beeler, Warfare in Feudal Europe, 730-1200 (Ithaca,
N.Y.: Cornell IJniversity Press 1971), 212-14; Michael Howard, War in ~k~cropcwn History
(1.ondon: Oxford University Press 1976), 33-37.
54. Q ~ ~ i cWright,
y A Study of War (abridged ed.; Chicago: University of Chicago Press
1964), 142. Also see 63-70, 74-75. There are important exceptions to these general~zations-
the American Civil War, for instance, falls in the middle of the period Wright says is dominated
by the offense.
55. Geoffrey Kemp, Robert Pfaltzgraff, and Uri Ka'anan, eds., T h e Othcr Arms Race
(i.exington, Mass.: D. C. Heath 1975); James Foster, "The Future of Conventional Arnis
Control," Policy Scrcnces, No. 8 (Spring 1977), 1-19.
56. Richard Challener, Admirals, Generals, and American Foreign Polrcy, 1898-2 914
(Princeton: Pr~ncetonUniversity Press 1973), 273; Grey to Nicolson, in Gooch and Ternperley
( t n . 47), 4 14.
57. Quoted in James Crowley, Japan's Quest for Autononzy (Princeton: Princeton
University Press 1966), 49. American naval officers agreed with the Japanese that a ten-to-six
r,ltlo would endmger Japan's supremxy in her home waters.
58. E.1.. Woodward and R. Butler, eds., Doc~tmentson Brrtish Foreign Polrry, I9IC)-l9?9,
T h ~ r dseries, 111 (London: H.M.S.O. 1950), 526.
59. Jervis (fn. 5 ) , 69-72, 3.52-55.
60. Quoted in Merze Tatc, T h e United States m d A r n ~ a n ~ e n t(Cambridge:
s Harvard
Univcrs~tyPress 1948), 108.
6 1 . Koggs (fn. 28), 1.5, 40.
62. Kenneth Hagan, American Gunhoat lliplonzacy rrnd the Old Navy 1877-1889
(Westport, Conn.: Greenwood Press 1973), 20.
6.3. Winston Churchill, T h e G a t l ~ e r ~ nStorm
g (Boston: Houghron 19481, 206.
64. Brodie, War and Politics (fn. 35), 325.
65. Boggs (fn. 28), 42, 83. For a good argument about the possible different~~lt~on hetween
offens~veand defensive weapons in the 1930's. see Basil Liddell Hart, "Aggression and the
Prohle~nof Weapons," English Review, Vol. 5.5 (July 1932). 71-78.
66. Quoted in Boggs (fn. 28), 39.
67. On these grounds, the Germans claimed in 1932 that the French forts were offensive
(ibid.. 49). Si~nilarly,fortified forward naval bases can he necess'lry for 1,lunching a n attack;
see Rrai\ted (fn. 27), 643.
68. The French made this argument 111 the interwar period; see Rlchard Challener, Thc
k r m c h Theory of the Nation rn Arms (New York: C;olumbia IJniversity Press 1955), 181-82.
The Germ.lns disagreed; see Boggs (fn. 281, 44-45.
69. Oman (fn. 531, 57-58.
70. Quoted in Charles Wehster, Foreign Policy of Castlerc~zgh.1 I, 18 1.F-1822 [L.ondon:
G. Bell and Sons 196.31, .5 50.
71. Koggs (in. 28), 14-15, 4 7 4 8 , 60.
72. Quoted In Philip Jordan, Frontter Law and Order (Lmcoln: Un~versityof Nebraska
Press 1970), 7; also see 16-17.
73. Boggs (fn. 28), 20, 28.
74. See, however, Desmond Ball, "The Counterforce Potential of American SLBM
Systems," Journal of Peace Research, XIV (No. 1, 1977), 23-40.
75. Richard Garwin, "Anti-Submarine Warfare and National Security," Scientific
American, Vol. 2 2 7 (July 1972), 14-25.
76. The latter scenario, however, does not require that the state closely match the number
of missiles the other deploys.
77. Thomas Schelling, Arms and Influence (New Haven: Yale University Press 1966),
69-78. Schelling's arguments are not entirely convincing, however. For further d~scussion,see
Jervis, "Deterrence Theory Re-Visited," Working Paper No. 14, UCLA Program in Arms
Control and International Security.
Rationality at the Brink: The Role of Cognitive
Processes in Failures o f Deterrence
Jack L. Snyder
I . Introduction: Rationality in a Two-Value Came
A
ccording to the scenarios imagined by most strategists, nuclear
confrontation is a game involving a trade-off between two values.
First, there is the value associated with the immediate issue of con-
tention: for instance, in the Cuban missile crisis, maintaining U.S. prestige
in the world arena; in the Berlin crisis, maintaining the credibility of U.S.
commitments. Second, there is the value of minimizing the possibility that
an unwanted general war could result from this superpower confrontation.
Assuming a strong commitment on the part of both adversaries, any policy
that tries to attain the value associated with the immediate point of con-
tention will tend to increase the likelihood of general war. Conversely, any
policy that seeks to maximize the avoidance of war will jeopardize the pro-
tection of the first value.'
Trying to understand the dynamics of this two-value game has been a
chief preoccupation of strategists since the advent of nuclear weaponry. I t
is important, first of all, to understand how this game should be played, so
that the decision maker will be able to recognize the strategies that give him
the best chance of optimizing his interests and preparing an appropriate
force posture, military doctrine, and diplomatic strategy. At the same time,
it is also important to understand how the game will be played by flesh-and-
blood decision makers. This understanding is crucial for two reasons:
(1)Since each player's optimal strategy depends on the strategy adopted by
his opponent, such an understanding may help the player to estimate his
opponent's probable responses and to adjust his own strategy accordingly;
(2) If the non-optimizing strategies adopted by human decision makers tend
to occur in regular patterns, then a knowledge of these patterns may help
each player to monitor his own strategies.
Source: World l'ol~trcs,3 0 ( 3 )(1 978): 34\5-65.
Most discussions of the dynamics of nuclear confrontations have tended to

telescope these two questions - "which strategy should logically be adopted?"
and "which strategy will in fact be adopted?" - into a single issue. Either
implicitly or explicitly, it is assumed that players will tend to employ strategies
which, in a rough way, optimize their values. Take, for example, Thomas
Schelling's method for overcoming the shortage of empirical evidence regard-
ing the behavior of decision makers in a nuclear confrontation: "You can sit
in your armchair and try to predict how people will behave by asking how you
would behave if you had your wits about you. You get, free of charge, a lot of
vicarious, empirical beha~ior."~
According to Schelling, the very foundation of strategic theory is "the
assumption of rational behavior - not just of intelligent behavior, but of
behavior motivated by a conscious calculation of a d ~ a n t a ~ e s The
. " ~ theory
of deterrence, which receives its most difficult test in the type of superpower
confrontation that is under discussion here, rests on such premises. One of its
key assumptions is that the deterree will make an estimate of probable costs
and probable gains and, on that basis, will be deterred from pursuing a pol-
icy that is likely to lead to nuclear war, since no possible gains could outweigh
the costs of a nuclear exchange. The theory of deterrence explicitly assumes
that the decision maker will recognize the trade-off between the two values
and that he will employ a strategy which effects the optimal trade-off between
those values, as determined by the decision maker's indifference curve.
Similarly, the strategy of "compellence," as outlined by Schelling in Arms
and I n f l ~ e n c eassumes
,~ that the compellee will recognize tradeoffs and make
the appropriate cost-benefit calculations. That is, once the compeller has
"rigged the incentives so that the other party must choose in [the compeller's]
favor," it is assumed that the compellee will weigh those incentives correctly
and recognize his obligation to c a p i t ~ l a t e . ~
Strategic research has concerned itself primarily with the internal logic
of deterrence theory. To be sure, many strategists include caveats regarding
the fragility of rationality in crisis situations. For example, Bernard Brodie
cautions his readers not to let the seemingly ~ o w e r f u llogic of deterrence
make them sanguine about the improbability of nuclear attack. Brodie
points out that it may not be realistic to assume a dispassionate calculation
of costs and benefits under conditions of great uncertainty, especially when
an attack is ~erceivedas imminent.6 Still, the implications of such insights
have remained largely unpursued.
There does exist, however, a considerable body of research into nonrational
influences on the decision process which might profitably be applied to the
problem of deterrence in time of crisis. Most of this research approaches the
problem in terms of constraints on an essentially rational decision process -
constraints which, nevertheless, may be so great as to completely undermine
the rationality of that process. Such constraints may include:
(1) Organizational dynamics, e.g., the organizational processes and

bureaucratic politics outlined in Graham Allison's Essence of Decision.;
\ n \ tier Rationality at the Brink 1 73
Idiosyncratic psychopathologies that may prevent value optimization,

as illustrated in Alexander and Juliette George's Woodrow Wilson and
Colonel H o ~ s e . ~
Non-idiosyncratic cognitive processes that evaluate information and
options according to nonrational principles. Phenomena included in
this area would be, for example, ( a ) the tendency to establish, on the
basis of inadequate information, a stereotyped image of the adversary
and then to maintain that image tenaciously by means of unconscious,
selective processing of information,' and ( b ) the alleged tendency of
decision makers to advocate riskier policies when responsibility for the
decision is shared with a group than when the individual is solely
responsible. l o
4 more ambitious approach to nonrationality in decision making can be

found in John Steinbruner's The Cybernetic Theory of Decision.' Steinbruner
does not so much describe mere constraints on rational decision making as
an entirely independent, nonrational process by which decisions are made.
The operation of this process is based on the model of a computer's feedback
loops, supplemented by the general principles that govern the operation of
non-idiosyncratic cognitive processes - principles which have been established
by experimental psychology.
Steinbruner's model posits rules of decision that reject the rational-
analytic method. In his model, values that are in a trade-off relationship are
not integrated, hut pursued separately. Indifference curves are not con-
structed, even implicitly; p o b a b l e outcomes are not estimated; no attempt
is made to optimize values. Under conditions of uncertainty, decisions are
structured not by rational-analytic procedures, but by non-rational rules of
cognitive operations. For example, the decision maker will tend to concep-
tualize his decision environment so as to avoid recognizing trade-off rela-
tionships between his values. Trade-offs violate the principle of cognitive
consistency. Hence, when the environment is sufficiently unstructured to
permit some interpretive latitude, the decision maker will suppress the
trade-offs by conceptualizing his world in such a way that the values d o not
appear to conflict.
Furthermore, Leon Festinger points out that decision makers can solve
their cognitive dilemma by conceiving their decision as being wholly deter-
mined by the course of external events. "It is possible ... to reduce or even
eliminate the dissonance by revoking the decision psychologically. This
would consist of ... insisting that really no choice had been made for which
the person had any re~ponsibility."'~
In sum, Steinbruner's model holds that there are strong cognitive forces
in operation under conditions of uncertainty which predispose decision
makers to deny the existence of trade-offs, to deny choice, and to impute
unwarranted certainty to this view of their situation.
If Steinbruner's model is an accurate description of decision making under
uncertainty, it places the stability of deterrence in a new and disconcerting light.
Whereas deterrence theory requires at least an implicit recognition of trade-

offs, Steinbruner's model suggests that human decision makers are under
cognitive pressure to conceptualize their decision environment in such a
way as t o deny the existence of trade-offs.I3
Compellence strategies - generally viewed as rather reckless even by con-
ventional analysts - appear even more troublesome in this light. Adopting a
compellence strategy permits the compeller to forswear choice and avoid his
trade-off, so decision makers under uncertainty - as in a nuclear confronta-
tion - should be predisposed to adopt this type of tactic. Furthermore, a
"cybernetic decision maker" who is in the position of the compellee may not
accurately perceive his incentive to capitulate - especially if he too is locked
into a no-choice compellence strategy.
Despite this gloomy picture, there exists a possible mitigating factor
even within Steinbruner's scheme: the so-called "reality principle." Only
when uncertainty is great - in unstructured situations - d o non-rational
cognitive principles have full rein. When a decision environment is highly
structured - that is, when uncertainty is low and the trade-off is unavoid-
ably self-evident - the reality principle may not permit the decision maker
to avoid the realistic calculations and tough choices required by the
rational-analytic process of decision. Hence, the relevant question becomes:
is the specter of nuclear destruction a sufficiently palpable constraint so that
(1)it imposes an undeniable structure on the decision environment, and
(2) it forces the decision maker to recognize the trade-offs inherent in his
situation, despite the cognitive costs this entails?
In order to answer this question and, more generally, in order to test the
plausibility of Steinbruner's cybernetic model of decision, it will be helpful
to examine a historical case, the stakes and structure of which resemble
those of the hypothetical nuclear confrontations imagined by strategic ana-
lysts. American decision making during the Cuban missile crisis has these
characteristics. The "objective" structure of this case (as opposed to its per-
ceived structure) involved a trade-off between two values - the avoidance of
an unwanted war of monumental proportions and the securing of a web of
interests related to the maintenance of political prestige and military power.
With regard to the Cuban case, one may ask three questions: How did
U.S. decision makers deal with their value trade-off problem? How did they
perceive their options? And, what are the implications of their behavior for
our ideas about deterrence?
Before proceeding with a fuller explanation of the relevant aspects of
Steinbruner's theory and with the analysis of the case study, some of the
assumptions and limitations of this analysis should be made explicit.
(1) The psychological principles on which Steinbruner's theory rests will be

accepted as representing a rough consensus based on the discipline's
experimental work. In any event, the test for these principles will be in
the plausibility of the explanation they suggest for the case study.14
Sii) d c ~ r Rationality at the Brink 1 75
(2) This analysis offers no rigorous technique for conclusively demonstrat-

ing that 3 particular set of policy options was structured by non-
rational cognitive pressures rather than by an analytic approach. If a
player decides that circumstances leave him only one option, how can
the objectivity of that view be disproved? Certainly, there may be situ-
ations in which even a rational decision maker is left with the single
option o f initiating an unwanted war - or at least risking its immediate
provocation. Psychologist Jack Brehm points out that, even in experi-
mental conditions, it may be impossible to determine whether an effect
is caused by cognitive dissonance or whether it is simply a case in which
rational rules of decision produce the same results as nonrational cog-
nitive rules." This analysis does not mean to refute the view that a
model based on constrained rationality can explain the Kennedys' con-
ceptualization of their problem. However, circumstantial evidence will
cast doubt on such an explanation. In any case, it is not the goal of this
analysis to design 3 classical experiment, pitting the cybernetic model
against the rational-analytic one, but rather to test the rough plausibil-
ity of Steinbruner's approach and to determine whether it might suggest
insights that have interesting ramifications for strategic theory.
( 3 ) It is dangerous to generalize on the basis of one case study - or, more
accurately, on the basis of only one side of one case. It is doubly danger-
ous when one of the unexplored aspects - the Soviet process of decision -
offers a possible disconfirmation of the main hypothesis. That is, despite
uncertainty, the Soviets did not lock themselves into a no-choice policy.
Still, in the absence of hard information regarding the Soviet process of
decision, one can only speculate about the respective roles of analytical and
cognitive factors in shaping the Politburo's conception of the crisis and its
choice of options. That speculation will be reserved for the conclusion.
11. T h e Analytic versus t h e CyberneticKognitive Model of Decision
In a two-value game such as a nuclear confrontation is likely to represent,

the rational-analytic approach to decision making is characterized by
( 1 ) the "integration" of the conflicting values by means of an explicit or
implicit indifference curve; (2)cost-benefit calculations; and (3)the estima-
tion of the expected outcomes of alternative policies in order to determine
the best means of achieving an optimal trade-off of values. Uncertainty is
treated as a statistical problem. The decision maker is sensitive to all rele-
vant information."
In his cybernetic theory of decision, John Steinbruner has attempted to
formulate a decision paradigm that is equal in scope to the analytic para-
digm, but entirely different in its operating principles. It is organized around
two concepts: "short-cycle information feedback" and the elimination of
uncertainty."
T h e Cybernetic M o d e l
Steinbruner takes as his model simple, cybernetic decision mechanisms - such

as the thermostat - which effectively solve problems without the complex
calculations required by the analytical model. Since the cybernetic mech-
anism makes no calculations, it is unaffected by uncertainties which may
stem from a lack of relevant information, and which would stymie an ana-
lytic decision maker.
Despite its simplicity, such a mechanism can produce highly adaptive
behavior:
Roughly speaking, the mechanism of decision advanced by the cybernetic
paradigm is one which works on the principle of the recipe. The deci-
sionmaker has a repertory of operations which he performs in sequence
while monitoring a few feedback variables. ... The cook, in this model,
does not construct the relative preference for sweetness or tartness for an
average range of customers in baking pies. Rather, he follows established
recipes and watches attendance at the restaurant and the rate at which
his pies disappear.18
This is essentially a satisfying method rather than an optimizing one.

Sequential attention is given to alternatives until an adequate one is found."
"Value Disaggregation"
A cybernetic mechanism cannot deal with the problems of conflicting values.

It must disaggregate them and attend to them sequentially. O r it must assign
them to separate parts of the decision mechanism, such as different agencies
within the government. Of course, this disaggregation of conflicting values
is likely to result in a decrease in the efficiency of the mechanism, as some of
its policies work at cross-purposes with other policies. Steinbruner cites the
following example:
... separate entities of the government construct river projects to control

floods, on one hand, and to provide disaster relief to pay for flood dam-
age on the other. Though the separate problems jointly affect private
investment in flood plain areas, they are operated separately and deci-
sions about them are made separately. Jointly over the years, they have
produced uneconomic investment in flood-plain areas so that the more
flood control projects that have been constructed, the greater the national
flood losses have become. Since the decision process treated the programs
as separate issues, no one noticed the inherent problems until the invest-
ment had been made and the paradoxical flood losses began to occur.1°
Sometimes, conflicting values cannot be disaggregated by sequential

attention or by assignment to different bureaucratic pigeonhole^.^' The value
Srr\ t i t * Rationality a t t h e Brink 177
conflicts inherent in crisis diplomacy are of this kind. Any decision maker
knows he must consider the closely related problems of war avoidance and
diplomatic success simultaneously, not sequentially. He cannot tell one bureau-
cracy to worry about avoiding war and another to concern itself only with
winning the confrontation. If the value conflict is to be eliminated, it must be
done by psychological means, not bureaucratic ones.
The human mind, of course, can and will deal with such trade-offs
analytically i f compelled to d o so by a highly structured decision environ-
ment. However, trade-offs violate the experimentally established principle
of cognitive consistency. Therefore, if the decision environment is suffi-
ciently unstructured and entails sufficient uncertainty so that there is leeway
for interpretation, the decision maker will tend to conceptualize the prob-
lem in such a way that the trade-off can be denied. In other words, if the
point of view that there is no trade-off relationship can be taken (that is, if
it is not precluded by the "reality principle"), then it will be taken. Thus,
the problem is cast in a form with which the cybernetic decision maker can
deal."
Once such a conceptualization of a problem is established, cognitive
principles work to impute certainty to the correctness of that view by selective
processing of incoming information which depreciates the value of discon-
firming evidence and by "categorical inferences" of certainty or impossibil-
ity." With regard to the latter point, Steinbruner discusses John Kennedy's
inference that he would be impeached if the missiles were not removed from
Cuba:
That Kennedy might have taken his impeachment quite seriously as the
outcome of his following a conciliatory course in the crisis is [harder] to
imagine within the analytic paradigm. It would appear as a rare limiting
case (all other outcomes each assigned a probability of zero).
By contrast, cognitive theory readily accounts for the existence of
firm, categorical, nonprobabilistic beliefs in the presence of intense
uncertainty. The cognitive processing mechanisms of the mind provide a
number of ways in which beliefs become established, independent of the
weight of objective evidence. ... To the cognitive theorist it becomes
quite readily conceivable that Kennedy meant exactly what he said
about his impeachment - as he said it. As a general matter, cognitive the-
ory makes the assumption that structure will be imposed on certain situ-
ations, and uncertainty thereby resolved, not by probabilistic judgments
but by categorical inferences.14
ognitive Dynamics and Decision Making during the Cu

Missile Crisis
The following case study of U.S. decision making during the Cuban missile cri-
sis will stress the role of the cognitive principle of management of inconsistency
in structuring the decision makers' conception of their environment under

conditions of uncertainty. Specifically, it will point out:
(1) the tendency toward unwarranted assumptions of certainty regarding

opponents' intentions and the correctness of one's chosen policy;
(2) the tendency to see the two principal values at stake in the crisis not as
conflicting but as consonant;
(3) the tendency to adopt a strategy of compellence (which entails no trade-
offs for the compeller), rather than a strategy of negotiation (which
necessitates "value integration" - i.e., the recognition of a trade-off
between conflicting values).
111. T h e Cuban Missile Crisis: A Fourth Cut
Graham Allison has pointed out that the nature of the decision environment
in the Cuban missile crisis makes it particularly well suited for "Model I"
(rational actor) analysis: "In the context of ultimate danger to the nation, a
small group of men, unhitched from the bureaucracy, weighed the options
and decided."2s The decision makers were highly conscious of the need to
approach their dilemma in a rational-analytic manner. Furthermore, their
chosen course of action was highly successful in achieving their goals. Hence,
in such a case, Allison argues, alternative models "are forced to compete on
Model 1's home ground. The dimensions and factors uncovered by Model I1
[bureaucratic processes] and Model 111 [bureaucratic politics] in this case will
therefore be particularly s ~ g g e s t i v e . "This
~ ~ is also true for an explanation
based on nonrational cognitive processes of decision.
The present analysis will focus on the overall shaping of the decision mak-
ers' attitudes toward their options and the process by which those options
were weighed, rather than on nonrational constraints on the decision.
Avoiding the T r a d e - o f f
Kennedy and most of his advisors conceptualized the decision in a way that
avoided placing their two relevant values (war avoidance and the mainten-
ance of prestige in the international arena) in conflict. They achieved this by
conceiving the problem in terms of "risking war now" versus "running an
even greater risk of war later." If Kennedy did not act to save U.S. inter-
national prestige nozu, the loss of that prestige would contribute to an
increased chance of war later. Kennedy attributed virtual certainty to the
view that the Russians would be encouraged to push for greater and greater
concessions in Berlin and elsewhere unless the missiles were unconditionally
removed. Viewed in this light, Kennedy's choice of avoiding a diplomatic
solution (such as trading Cuban missiles for Jupiters in Turkey) in favor of
an essentially unconditional ultimatum must have seemed reasonable to
Snvdei Rationality at the Brink 179
him despite his estimate that his course of action entailed a probability
"between one out of three and even" of nuclear war." It would be worth
taking such horrendous risks if the diplomatic or do-nothing options
entailed even greater long-run risks of war. In sum, Kennedy's "war now
versus war later" formulation of the problem permitted him to deny the
trade-off relationship which seemingly existed between the values of avoiding
war and maintaining prestige. In fact, it permitted him to view the values as
mutually reinforcing: standing firm in Cuba would demonstrate America's
resolve and, hence, reduce the long-run likelihood of war.
Clearly, Kennedy's formulation of the problem is not prohibited by the
rational model simply because it corresponds closely to the formulation
predicted by the cognitive model. It is Kennedy's imputation of certainty
to a highly uncertain situation which most strongly suggests the operation
of the cognitive model, rather than the "no trade-off" character of his for-
mulation per se. Kennedy's decision environment was highly "underdeter-
mined." Many and diverse interpretations could be and have been given
to the Soviets' motivations - and to their future intentions, had the missile
gambit been successful. Some of these interpretations would hardly have
justified a one-in-three risk of nuclear war as the price of removing the
missiles, especially when their removal could have been secured with less
risk (albeit at a somewhat higher price) diplomatically. The rational paradigm
offers no guidance as to how such vast uncertainties can be resolved. The cog-
nitive paradigm, however, explains unambiguously, in terms of cognitive
principles and pressures, why Kennedy and the ExCom decided as they did.
The Subjective Structuring of the Decision Environment
In order to support the cognitive view of Kennedy's decision process, it will

be necessary to look more closely at the seeming trade-off which Kennedy
faced, and at his manner of dealing with it.
To repeat, there were two values involved in the missile crisis. The first
consisted of a web of interests including the preservation of the military stn-
tus quo ante, the maintenance of America's international prestige, and the
need to demonstrate to the Soviets that destabilizing faits accornplis and
"salami tactics" would not be tolerated - with all of these interests requir-
ing an unconditional withdrawal of the missiles from Cuba. The second
was to avoid a nuclear war which might be precipitated by measures
designed to achieve an unconditional with-drawal.
Aside from the question of whether, objectively, a trade-off relationship
exists, it is undeniable that the two issues are, in Steinbruner's jargon,
"highly interactive." That is, policies that affect one value must intimately
affect the other. Thus, reality constraints are too strong to achieve a stable
disaggregation of the values by the simplest and most routine cybernetic
means - for instance, letting one bureaucracy handle value A and another
handle value B, as in the example of flood control.
In such situations, the decision maker may be forced to deal with the
values analytically; but, since uncertainty is great in this case, one might
expect subconscious, nonrational, cognitive processes to structure the deci-
sion in such a way that the values d o not conflict.
Intuitively, it is not hard to imagine the cognitive stress which the sacrifice
of either value would have entailed for President Kennedy, especially when his
own prestige as well as that of the United States was on the line. The Bay of
Pigs, his disastrous confrontation with Khrushchev in Vienna, and the domes-
tic allegation that he was long on profile but short on courage, all combined
to make the unconditional withdrawal of the missiles an important value
indeed, perhaps tantamount to the avoidance of nuclear war itself.
With two such vital values at stake, it is not hard to imagine that Kennedy
was under strong cognitive pressure to view his situation in a way that would
permit him to adopt a strategy which held out the possibility of winning big
with respect to both values - even if it meant running the risk of losing big as
well. An analytic, trade-off-oriented formulation of the problem could not
achieve this for Kennedy; a cognitive, "no trade-off" formulation could.
"A M i s s i l e is a M i s s i l e "
Conceptually, there were two possible ways to avoid the trade-off. The first,
and simplest, would have been to disclaim the significance of the introduc-
tion of the missiles into Cuba. By telling themselves that the missiles did
not, in any appreciable way, affect the military balance or undermine the
prestige of the United States, the members of the ExCom could have acqui-
esced in the installation of the missiles and incurred no cognitive costs. Both
values - prestige maintenance and war avoidance - could have been viewed
as essentially irrelevant t o the missile issue.
In fact, a t the beginning of the ExCom's deliberations, Secretary of
Defense McNamara argued for exactly this view: "A missile is a missile. It
makes no great difference whether you are killed by a missile fired from the
Soviet Union or from Cuba."28 Ted Sorensen reports: "As some (but not all)
Pentagon advisers pointed out to the President, we had long lived within
range of Soviet missiles, we expected Khruschev to live with our missiles
nearby, and by taking this addition calmly we would prevent him from
inflating its i m p ~ r t a n c e . " ~ ~
Objective arguments on this point were mixed. McNamara's view was
bolstered by the fact that the vulnerable, soft-site missiles could be useful
only in a first strike, and that the Soviets were far from a credible first-strike
capability even with the additional deliverable warheads provided by the
Cuban emplacements." Meanwhile, the armed services pointed to the
reduced warning time for getting American bombers off the ground, and "
the diplomats and politicians stressed the importance of "appearances,"
independent of strictly military considerations." In any case, uncertainty
regarding the significance of the Russians' gambit was great enough to permit
McNamara to rationalize his dissonance-avoiding formulation.
5nt tiel Rationality a t the Brink 18 1
President Kennedy, however, rejected the "do nothing" course of action

from the outset. "We'll have to d o something quickly," he told the ExCom.
"I suppose the alternatives are to go in by air and wipe them out, or to take
other steps to render the weapons inoperable."" Kennedy was operating
under cognitive pressures different from those that influenced McNamara.
In the wake of Vienna, Berlin, and the Bay of Pigs, members of both parties
in Congress were urging Kennedy to stand firm against Khrushchev's
encroachments. Even before the revelation about the emplacement of
medium- and intermediate-range missiles, Republican candidates were
making Kennedy's do-nothing policy in Cuba a central issue of the election
campaign. After the revelation, Kennedy saw his impeachment as the likely
consequence unless they were removed. To quote Roger Hilsman, "The
United States might not be in danger, but the Administration most certainly
was."'4 Whereas the formulation "a missile is a missile" may have dimin-
ished the cognitive pressures on McNamara (who was chronically insensi-
tive to political concerns), it could only heighten Kennedy's troubles and
exacerbate the trade-offs he faced.
The other possible means for avoiding the trade-off between the avoidance of
war and the maintenance of prestige was the formulation, "risk war now to
avoid certain war later." This, in fact, was the conceptualization adopted by
Kennedy. In his speech of October 22 announcing the blockade, Kennedy
said: "Aggressive conduct, if allowed to grow unchecked and unchallenged,
ultimately leads to war. This nation is opposed to war. We are also true to our
word. Our unswerving objective, therefore, must be to prevent the use of
these missiles against this or any other country and to secure their withdrawal
or elimination from the Western HemisPhere."j5
Robert Kennedy recalled that "we all agreed in the end that if the Russians
were ready to go to nuclear war over Cuba, they were ready to go to nuclear
war and that was that. So we might as well have the showdown then as six
months later." j h
In the ExCom, Secretary of State Rusk had concluded his case for an air
strike with a similar sentiment: "If we don't d o this, we go down with a
whimper. Maybe it's better to go down with a bang.""'
Presidential biographers Schlesinger and Sorensen summed up Kennedy's
and, generally speaking, the ExCom's attitude, as follows:
In a general sense, the decision [to introduce rnissilesl obviously repre-

sented the supreme Soviet probe of American intentions. N o doubt a
"total victory" faction in Moscow had long been denouncing the gov-
ernment's "no win" policy and arguing that the Soviet Union could
safely use the utmost nuclear pressure against the United States because
the Americans were too rich or too soft or too liberal to fight. Now
Khrushchev was prepared to give the argument its crucial test.'x
The Soviets would move [in Berlin], [Kennedy] expected, but they
probably would whatever we did; and perhaps this show of strength
would make them think twice about it.39
This view per se was neither necessarily wrong nor necessarily incom-
patible with the analytic model. However, the unswerving nature of the
President's commitment t o this view under conditions of great uncertainty
is more reminiscent of the cognitive than of the analytic paradigm.
C o m p e l l e n c e a n d Certainty
Kennedy's steadfast commitment t o the "war now versus war later" view is
directly reflected in the strategy he adopted to force the Russians to remove
their missiles. That strategy was essentially one of compellence: a strategy
based on the renunciation of choice.
The President and his brother assured each other that they had no choice
but to compel the Russians to dismantle the missiles unconditionally. "It
looks really mean, doesn't it? But then, really there was no other choice. If
they get this mean on this one in our part of the world, what will they d o
on the next?"40
The President then tried t o convince the Russians that he was locked
into a no-choice situation. If a blockade did not get the missiles out, an air
strike would, Kennedy told Ambassador Dobrynin via his brother.41 In
effect, he told the Russians that he was not interested in compromise solu-
tions; he had no choice but to insist on the unconditional removal of the
missiles. Therefore, the Soviets would have only two options: submit t o his
demands or provoke an escalation of the conflict. However, since Kennedy
had read The Guns of August and was attuned to the Soviets' need to avoid
h ~ m i l i a t i o n he
, ~ ~would permit them the option of capitulating gracefully
before he unilaterally imposed a military solution. But if the Soviets were
really intent on pushing him, Kennedy told them in effect, there was noth-
ing he could d o to avert escalation.
Despite Albert and Roberta Wohlstetter's protestations about controlling
the risks in Cuba, the fact remains that "all [the Kennedys'] skill would have
been to n o avail if in the end [Khrushchev] had preferred his prestige, as they
preferred theirs, to the danger of a world war."43 And despite Sorensen's and
Schlesinger's description of the ExCom's decision process as painstakingly
open and rational, the fact remains that the President and most of his ad-
visors perceived only one real option: an uncompromising policy based on
compellence, rejecting any trade-off of values. Such policies are the hallmark
of the cognitive decision maker.
T h e Failure t o N e g o t i a t e
An analytical decision maker, on the other hand, would have characteris-

tically adopted a strategy based on negotiation, compromise, and explicit
5 n ~ d t . 1 Rationality at the Brink 183
value trade-offs. Kennedy did have an opportunity to strike a bargain in

which both the U.S.S.R. and the U.S. would have given up some prestige in
order to eliminate the immediate danger of war. The Soviets offered to
remove their missiles from Cuba if the Americans would remove theirs from
Turkey." Such a trade had already been proposed in the American press as
a face-saving, war-averting compromise. Walter 1,ippmann had argued that
"the two bases could be dismantled without altering the world balance of
power."" Inside the ExCom, Adlai Stevenson also argued for the comprom-
ise. As Ambassador to the United Nations, he realized that the emplacement
of missiles in Turkey was in fact quite comparable to the emplacement in
Cuba, and that he w o u l d find it-hard to construct a tenable argument
against the equity of the proposed trade.
In addition, Stevenson had hoped to link this exchange to a broader settle-
ment of some contentious issues between the two superpowers. Schlesinger
reports, however: "The President ... rightly regarded any political program
as premature. He wanted t o concentrate on a single issue - the enormity of
the introduction of the missiles and the absolute necessity for their removal.
Stevenson's negotiating program was accordingly r e ~ e c t e d . " ~ ~
Although Kennedy had previously ordered the dismantling of the mis-
siles in Turkey (an order which had never been carried out), he perceived
any compromise affecting Turkey as undermining a U.S. commitment. Still,
he did ask Roswell Gilpatric to prepare a scenario for the removal of the
missiles in Turkey and Italy.4' O n the one hand, Kennedy remained firm in
his cognitively reinforced view that the long-term likelihood of nuclear war
would increase greatly if he appeared to be giving in to Soviet pressure.
On the other hand, he seems to have balked at the thought that a very marginal
diminution of the American commitment to Turkey could have ramifications
significant enough to warrant a one-in-three risk of nuclear war.
Kennedy found a way to avoid the trade. As the cybernetic model would
predict, he seized upon his brother's idea of rejecting the most recent and
most official offer to trade Cuban for Turkish missiles and accepting instead
Khrushchev's informal offer to remove the missiles simply in exchange for
a pledge not to invade Cuba. Although the informal offer (1) was only
inferred from an ambiguous and emotional letter from Khrushchev and trans-
ferred more explicitly through a highly unorthodox, unofficial contact48 and
(2)had apparently been superseded in the Kremlin by a more recent and more
stringent one, Kennedy chose to "accept" the Khrushchev "offer." He then
locked himself into that policy by informing Dobrynin that an air strike would
soon follow if his "acceptance" were not agreed to by the Politburo.
Although Robert Kennedy privately assured Dobrynin that in the long
run "there would be no problem about the missiles" in Turkey,"'that was
hardly the point. It was prestige and appearances that mattered, not the
outmoded Jupiter missiles. Thus, President Kennedy had violated his own
cardinal rule of crisis management: "Above all, while defending our own
vital interests, nuclear powers must avert those confrontations which bring
an adversary to the choice of either a humiliating retreat or a nuclear war.""'
In sum, Kennedy's failure to negotiate, his adoption of the so-called

Trollope ploy of accepting a non-offer, his desire to "concentrate on a single
issue," and, in general, his insistence on a compellence strategy in which only
the adversary must deal with trade-offs, are all indicators of a non-analytic,
cognitive decision process.
By the time Kennedy announced the blockade, his perception of the situ-
ation and his analysis of the options were firmly established. In his mind,
he had structured his environment in a way which subjectively disposed of
dissonance-causing trade-offs. As predicted by the cognitive paradigm, he
had seen certainty in an inherently uncertain situation; he was confident
that his chosen policy was not merely the right one, but the only one he
could possibly adopt under the circumstances.
O n the day he announced the blockade, according to Schlesinger's report,
"Kennedy himself was never more c o m p ~ s e d . " ~ ~
It is fortunate that the Soviet process of decision was not ruled by the same
cognitive pressures as Kennedy's. Khrushchev's apparent willingness to opt
for peace rather than prestige led to the settlement whereby an unwanted war
was avoided." However, decision makers who use compellence strategies
cannot always count on their adversaries to weigh the costs and benefits "cor-
rectly" and to opt for peace. The origins of World War I, for example, illus-
trate the tragic outcome that can result when both sides view their dispute as
a "no trade-offs" game, and adopt strategies based on compellence.
IV. Cognitive Processes a n d t h e Failure of Deterrence
Most scenarios suggest that World War 111 is likely to ensue from a two-
value game between the superpowers. The case study discussed in this
analysis illustrates the tendency of human decision makers to deal with such
situations by avoiding the recognition of the trade-off relationship that exists
between a player's own values. According to experimentally supported cogni-
tive theories, this avoidance reduces "cognitive dissonance" and re-establishes
cognitive consistency.
In the case study, the simplest method of ignoring the trade-off relation-
ship consists of the view that there is no contradiction between values because
one of them is not really threatened. Secretary of Defense McNamara
employed this method of trade-off avoidance when he argued that a missile
is a missile, whether launched from Cuba or the Soviet Union.
However, sometimes reality constraints (for instance, obvious diplomatic
costs of a do-nothing solution) or political constraints (for instance,
Republican and Congressional attitudes during the Cuban crisis) can pre-
clude this direct means of avoiding a trade-off. In that case, the decision
maker will tend to sidestep the trade-off between war avoidance and, say,
prestige maintenance by conceptualizing his dilemma according to a "risk
war now or incur destruction later" formula. The decision maker thereby
i ~ ~ \ : i s > r Rationality at the Brink 185
allows himself to argue that only by running some risk of war over the
immediate issue of contention can he demonstrate resolve to his adversaries
and, thus, avoid an inevitable war in the future. This formulation makes the
two values consonant and extricates the d e c ~ s ~ omaker
n from the d~ssonance-
produc~ngtrade-off. However, ~t 1s l~kelyto produce war ~f the adversary
also ddopts it.
5trateg1sts have recogn~zedthat the "better war now than war later"
concept presents real problems for the theory of deterrence, even aslde from
the lessons of cognitive theory. To quote Warner Schillmg:
The level of destruction that would attend a nuclear war becomes less
relevant if the critical choices should be made through reference to
relative, rather than absolute, costs (better World War I11 now than
later).
... there will be many opportunities for statesmen to conclude -
accurately or inaccurately - that ... the intentions of their opponent
make the costs of war unavoidable."
If those with conventional views of nuclear strategy and crisis rnanage-

ment, based largely on the rational-analytic paradigm, interpret such remarks
as a challenge to the logic of deterrence, the lessons of cognitive theory must
underscore and redouble this concern. Whereas conventional strateg' w t s see
the "better war now than war later" formulation as a possible result of
objective calculations of interest, the cognitive model suggests that states-
men are even more likely to fall back on such a formulation than objective
calculations would warrant, since that formula solves one of the prevalent,
subconscious problems of decision makers under uncertainty.
Not only will the cognitive decision maker tend to seek out this formula,
but he will tend to lock himself into this conception of his environment and
the adversaries' intentions. Cognitive theory argues that the mind craves
certainty and will work to establish it even when it is unwarranted by object-
ive conditions. Selective processing and recall of information, in accordance
with the principles of reinforcement and cognitive consistency, can create
such certainty.
As a result, the decision maker is likely to become locked into a strategy
based on compellence. As the case study has shown, it is a short step from
the formulation "better risk war now than face the certainty of incurring it
later" to a strategy based on closed options and no choice. Ole Holsti has
pointed out that "when [cognitive?]stress increases, problem solving tends
to become more rigid," because the ability to "resist the pull of closure" is
r e d ~ c e d . 'In
~ addition, the case study suggests that the cognitive decision
maker is highly unlikely to adopt a policy based on negotiation. Negotiation
entails compromise and represents the quintessence of explicit recognition of
value trade-offs. The cognitive costs involved will be avoided by the decision
maker under uncertainty if at all possible.
Compellence is a most dangerous game if both players are locked into

that strategy by their trade-off-avoiding conceptualizations. Steinbruner
remarks:
Consider, for example, the game of Chicken, long a favorite of theorists

of bargaining as a simple model of political conflict. In one of its more
dramatic forms, the game consists of two players each of whom drives
directly a t the other a t 60 mph down the center of a deserted high-way
with an audience of peers looking on. The first player to swerve to avoid
collision loses, is labeled a chicken, and suffers the contempt of his peers.
The game poses a classic value trade-off problem - survival on the one
hand, preservation of honor on the other. Thomas Schelling has pro-
vided an analysis of the game under the assumption that both players are
analytic decisionmakers. The first player to establish clearly an irre-
versible commitment to the center of the road (such as by tying the wheel
and climbing in the back seat) will win the game. The other player, still
retaining control, faces a certainty of death as against a finite loss of
honor, and everyone knows how an analytic actor will resolve that
choice. The scenario, which Schelling labels "compellence," is played out
daily, usually for lesser stakes, on street intersections throughout the
nation.
One's sense of this game changes drastically if a cognitive decision-
maker is inserted into the scenario. There are at least two good reasons
why such a decisionmaker might not yield to a cleverly established com-
mitment by the opposing player: first, while focusing on other things, he
may not notice the commitment; second, he may simply fail to engage in
a value trade-off while carrying out his prior intention. Rather than
compellence, with such a player involved, one readily imagines disaster.
Viewed from the assumptions of the cognitive paradigm, moreover, it is
not a disaster which emerges from an error in calculation, but rather it
is the consequence of the normally functioning decision p r o c e ~ s . ~ '
In sum, the analysis of the Cuban case study in light of cognitive theory
has re-emphasized the dangers of a compellence strategy. It should also
make us more circumspect about the tendency to regard deterrence as a
deus ex machina for avoiding nuclear war.
The present analysis suggests that, in situations structured along the lines
of a probable nuclear confrontation, there are "regularities of human
thought" that tend to lead decision makers away from seeing the trade-offs
which must be seen if deterrence is to work. At the same time, it is clear that
confrontations for high stakes between superpowers d o not inevitably result
in nuclear destruction. O n the contrary, there appear to be several mitigating
factors:
(1) A world of mutual assured destruction may impose reality constraints on

one or both players, which may lead to a stronger tendency to recognize
Sn)cicr Rationality at the Brink 187
trade-offs than in a first-strike world. Whereas the Cuban crisis occurred in

a world where "mutual assured destruction" tended to limit the advantages
of "mobilizing" first, Kaiser Wilhelm and Tsar Nicholas confronted each
other in a situation that encouraged a pre-emptive first strike. According to
the doctrine of the day, the realities of warfare in the railroad era put a pre-
mium on rapid mobilization and gave strong incentives against being slow
to respond to an opponent's preparation for war. This situation, in which
both players employed compellence strategies in a first-strike world, proved
to be a deadly combination.
(2) In a sense, the argument that the risks in Cuba were controlled may
be right: Perhaps a moderate, well-thought-out compellence strategy
(like the blockade) does involve fewer risks than a hysterical approach
(like the Wilhelm-Nicholas "I can't stop" correspondence). Because the
reality factor does seem to matter, a decision maker faced with a calm
(and even semi-accommodating) ultimatum may be less able to ration-
alize a "war now versus war later" formulation, and more likely to
evaluate his trade-offs analytically than if confronted with hysterical
threats.
(3) Khrushchev's capitulation gives rise to the suspicion that, even at the
brink, not all decision makers are subject to the pressures of dissonance
which prohibit value integration. What explains this fact? It might be
speculated that Kennedy's dissonance was high because both values -
war avoidance and prestige maintenance - were extremely precious and
were subjected to an extreme challenge by the Cuban confrontation.
For Khrushchev, the dissonance may not have been as great if he did not
view Soviet (and his personal) prestige as vitally challenged by the
Cuban issue. The prospects of sacrificing a lesser value for a greater one
may not cause sufficient dissonance to force the subconscious to avoid
the recognition of that trade-off.
From the standpoint of policy, it would be idle to warn decision makers

about dangerous cognitive tendencies toward "value disaggregation" in a
crisis. However, insights from cognitive theory may be instructive for peace-
time discussions of weapons procurement policy. In a world of perfectly
invulnerable strategic forces, even a cybernetic-cognitive decision maker
might find it difficult to rationalize a "war now versus war later" formula-
tion of his dilemma. As the case of World War I shows, in a situation of per-
ceived vulnerability to a first strike, decision makers may be strongly
inclined toward this dangerous formulation. Thus, while strategic stability
has been recognized as important by strategists working in the rational-
analytic framework, it becomes doubly important when viewed in the light
of cognitive theory.
Strategic stability based on the mutual survivability of retaliatory forces
has been a proclaimed goal of American policies of strategic procurement and
arms limitation. However, it has had to compete with other goals which often
conflict with the requirements of stability. The deployment of large numbers
of high-accuracy MIRV's demonstrates that considerations of stability are not

always at the top of the list when these competing goals are considered.
Cognitive theory would seem to support the case that strategic stability should
be given a higher priority in such deliberations.
Author's N o t e
An earlier draft of this paper was prepared for Professor Warner Schilling's Colloquium on
Military Technology and International Relations at Columb~aUniversity, and appeared in the
Rand Corporation paper series (P-5740, October 1976). The earlier vers~on~ncludesa more
extensive discussion of the 1914 case.
Notes
I . This trade-off relationship 1s discussed in Alexander L. George, David K. Hall, and

William R. Simons, The Limits of Coercive Diplomacy: Laos, Cuba, Vietnam (Boston: L~ttle,
Brown 1971), 232-38.
2. From Schelling's article in Kathleen Archibald, ed., Strategrc Interaction and Conflict
(Berkeley: University of California Press 1966), 150, as quoted in Graham Alllson, Essence of
Decision (Boston: Little, Brown 1971), 19.
3. Thomas Schelling, The Strategy of Conflict (Cambridge: Harvard University Press 1960;
citation from New York: Oxford University Press 1963), 4.
4. Thomas Schelling, Arms and Influence (1971 ed., New Haven: Yale University Press
1966), 69ff.
5. Schelling (fn. 3), 37; emphasis added.
6. Brodie, Strategy in the Missile Age (Princeton: Princeton Univers~tyPress 1965), 280.
7. Allison (fn. 2).
8. George and George, Woodrow W ~ I s o nand Colonel House (New York: Dover
Publications 1956).
9. John D. Steinbruner, The Cybernetic Theory of Decision: N e w Dimensions of Political
Analysis (Princeton: Princeton University Press 1974), 101.
10. See Irving L. Janis, Victims of Groupthink: A Psychological Study of Forergn-Policy
Decisions and Fiascoes (Boston: Houghton Mifflin 1972), esp. 236-37. This shift-to-risk hypoth-
esis has come under criticism in recent years, however. See W.L. Tullar and D.F. Johnson, Group
decision-making and the risky shift: a transnational perspectiue (Technical Report 48, Rochester,
N.Y.: Management Research Center, University of Rochester 1972).
11. Steinbruner (fn. 9).
12. Leon Festinger, A Theory o f Cognitive Dissonance (Evanston, Ill.: Row, Peterson
1957), 43-44.
13. Steinbruner has discussed the implications of his theory for deterrence strategies in
"Beyond Rational Deterrence: The Struggle for New Conceptions," World Politics, XXVIII
(January 1976). The discussion presented here takes a different direction and should not
be construed as reflecting Professor Steinbruner's views on the implications of his model for
deterrence theory.
14. These principles and some of their experimental underpmnings are presented in Steinbruner
(fn. 9); Festinger (fn. 12); Janis (fn. 10); Jack W. Brehm and Arthur R. Cohen, Explorations in
Cognitive Dissonance (New York: Wiley 1962); and Joseph H. deRivera with James N. Rosenau,
The Psychological Dimension of Foreign Policy (Columbus, Ohio: Merrill 1968).
15. Brehm and Cohen (fn. 14), 71.
16. For fuller explanation and illustration of the rational-analytic paradigm, see Steinbruner
(fn. 9), and Paul Samuelson, Economics (New York: McGraw-Hill 1973).
'.,ni i f Rationality a t t h e Brink I89
17. Ste~nbruner(fn. 9), 5 1 .

18. lbzd., 5.5.
19. Ihid., 62.
20. lbrd., 73.
21. Ihid., 88.
22. IOid., esp. 103-09.
2.3. Experimental evidence suggests that when the objective probabil~tyof a des~rahle
occurrence I S relatively high, people rend to view that probab~lityas approachmg certatnty;
conversely, imposs~hility is otten irnputed to events w h ~ c hare only moderately unl~kely.
DeRiver,l (fn. 14), 109.
24. Steinbruner (fn. Y), 110. The principal \tudy which Steinbruner uses to ~ l l u s t r ~ ht e~ s
theory is the debate on the M.I..F. (multilateral force). His discussions of the Cuhan crisis are
limited to this examination of Kennedy's cerralnty regardmg his impeachment ~ n tdo one of
Kennedy's certainty in his inferences about Soviet intent~onsand the consequences of a no-
reyponse pol~c):89.
i L ) , 8-9.
25. A l l ~ s o ~(in.
26. Ihzd., 9.
27. Theodore Sorensen, Krtzncdy (New York: Harper & Row 196.5). 705.
28. E l ~ eAbel, The Missile Crrsrs (New York: Lippincort 1966), 43.
29. Sorensen (fn. 271, 682-83.
30. See 1.E Stone, "The Continuing Debate," in Robert A. D~vine,The Cuban M m d e
Crrsrs (Chicago: Quadrangle 1971), 163; and Arnold L. Horelick, "The Cuban Missile Cris~s:
An Analysis of Soviet Calculation\ and Behavior," World Politics, XVI (April 1964).
3 1. Abel (fn. 28), 52.
2 . Sorensen (fn. 27), esp. 678.
3 1. Ahcl ( f n . 28), 49.
34. Hilsman, To M o l v A N a t ~ o n(Garden City, N.Y.: Lhuhleday 1967), 197.
35. Robert F. Kennedy, Thrrtcetr Days (New York: Norton 19691, 166-67.
36. Robert Kennedy, quoted in Stewart Alsop and Charles Bartlert, "In Time of Crisis,"
S ~ l t ~ ~ tr. ~d ~~eqn ~Post
n f i (December 8, 1962), 16.
37. Ahcl (fn. 28). 70.
38. Arthur Schlesinger, Jr., A Thousand Days (Boston: Houghton Mifflin 1965). 796.
39. Sorensen (fn. 271, 694.
40. I'resident Kenned!; quoted in Kennedy (fn. 35), 67.
41. Ahel (fn. ZX), 199.
42. Kennedy (fn. 35), 127.
43. Wohlstetter m d Wohlstetter, "Controlling the Risks In Cuba," In Robert J. Art 'lnd
Kenneth N. Waltz, eds., T l ~ cUse of Force (Lloston: I.ittle, Brown 1971), 2.34. The quotation i \
from Stone (fn. 3 0 ) , 164.
44. In addition, each s ~ d ew ~ to s agree not to violate the sovereignty of the other's cl~ent.
It should he noted that the f ~ n a lagreement did entail one elenient of "compromise." 'Thrr
Ilnited Statrs agreed to torswear a n invasion of Cuba as a condition for the r e n i o v ~ ~o lf the
missiles. That was not much of a concession, and Kennedy cert'l~nly lost n o prcsrigc by agree-
ing to it. In fact, he had to warn his subordinates not to be excessively gleeful in public print
d h o i ~ tthe settlement.
45. Stone (fn. 301, 22 I.
46. Schlesinger ( f t . 377, 741; emphasis added.
47. Stone (fn. 301, 222.
48. Ahel (fn. 28), 17.5-77.
49. Stone (fn. 3 0 ) , 222.
50. Kennedy, speech at American University, June 10, 1963, quoted in Abel (tn. 28), Y 1.
5 1. Schlesinger (fn. 37), 742.
52. This docs not necessarily indicate that Khrushchev's decision was analytical In the
sense of weighing tr~de-offs,making ~ndifferencecalculations, etc. It 1s possible that the cog-
nitive pressures on Khr~ishchevand h ~ colleagues
s were structured differently from those o n
Kennedy. Only detailed information on the Soviets' decision process, which is currently
unavailable, could resolve this question.
53. Schilling, "Technology and International Relations," The International Encyclopedia
of the Social Sciences (New York: Macmillan and Free Press 1968), 593.
54. Holsti, "Crisis, Stress, and Decision-making," International Social Science Journal,
XXIII, No. I (1971), 61-62, citing Sheldon J. Korchin and Seymour Levine, "Anxiety and
Verbal Learning," ]ournal of Abnormal and Social Psychology, Vol. 5 4 (March 1957), 238.
55. Steinbruner (fn. 9), 147.
Contrasts in American and Soviet Strategic Thought
Fritz W. Ermarth
W
e are having trouble with Soviet strategic doctrine. Soviet thinking
about strategy and nuclear war differs in significant ways from our
own. To the extent one should care about this - and that extent
is a matter of debate - we d o not like the way the Soviets seem to think.
Before 1972, appreciation of differences between Soviet and American
strategic thinking was limited to a small number of specialists. Those who
held it a matter of high concern for policy were fewer still. Since that time,
concern about the nature, origins, and consequences of these differences is
considerably more widespread, in large measure as a result of worry about
the Soviet strategic arms buildup and the continued frustrations of achiev-
ing a real breakthrough in SALT.
Heightened attention to the way the other side thinks about strategic
nuclear power is timely and proper. The nature of the Soviet buildup and
some of our own previous choices have locked us out of pure "hardware
solutions" to our emerging strategic security problems that are independent
of the other side's values and perceptions. Whatever one thinks about the
wisdom or folly of the manner in which we have pursued SALT so far, it is
desirable that management of the U.S.-Soviet strategic relationship have a
place for an explicit dialogue. That dialogue should include more attention
to strategic concepts than we have seen in past SALT negotiations.
Moreover, whatever the role of SALT in the future, the existence of "rough
parity" or worse almost by definition means that we cannot limit strategic
policy to contending merely with the opponent's forces. In the cause of
deterrence, crisis management, and, if need be, war, we must thwart his
strategy. That requires understanding that opponent better.
The Need to Understand Strategic Doctrine
Let us define "strategic doctrine" as a set of operative beliefs, values, and

assertions that in a significant way guide official behavior with respect to
Source: Intrrnationul Srr~irity,3 ( 2 )( 1978): 1.38-55.

strategic research and development (R&D), weapons choice, forces, opera-

tional plans, arms control, etc. The essence of U.S. "doctrine" is to deter
central nuclear war at relatively low levels of arms effort ("arms race
stability") and strategic anxiety ("crisis stability") through the credible
threat of catastrophic damage to the enemy should deterrence fail. In that
event, this doctrine says it should be the aim and ability of U.S. power to
inflict maximum misery on the enemy in his homeland. Making the world
following the outbreak of nuclear war more tolerable for the United States
is, at best, a lesser concern. Soviet strategic doctrine stipulates that Soviet
strategic forces and plans should strive in all available ways to enhance the
prospect that the Soviet Union could survive as a nation and, in some polit-
ically and militarily meaningful way, defeat the main enemy should deter-
rence fail - and by this striving help deter or prevent nuclear war, along
with the attainment of other strategic and foreign policy goals.
These characterizations of U.S. and Soviet strategic doctrine and the dif-
ferences between them are valid and important. Had U.S. strategic policy
been more sensitive over the last ten years to the asymmetry they express,
we might not find ourselves in so awkward a present situation. We would
have been less sanguuine than we were about prospects that the Soviets
would settle for an easily defined, non-threatening form of strategic parity.
We would not have believed as uncritically as we did that the SALT process
was progressing toward a common explication of already tacitly accepted
norms of strategic stability.
It is, if anything, even more important that these asymmetries be fully
appreciated today. They are a crucial starting point for strategic diagnosis and
therapy. But they are only a starting point. The constellations of thought, value,
and action that we call, respectively, U.S. and Soviet strategic doctrine or pol-
icy are much more complicated, qualified, and contradictory than the above
characterizations admit by themselves. To be aware of these other ramifications
without fully understanding them could lead to dangerous discounting, on one
hand, or distorting, on the other, the real differences between U.S. and Soviet
strategic thinking.
Comparative Strategic Doctrine
The following discussion is intended only t o suggest some of the contrasts

that exist between U.S. and Soviet strategic thinking. The issues raised are
not treated exhaustively, and the list itself is not exhaustive. Our appreci-
ation of these matters is not adequate to the critical times in the U.S.-Soviet
strategic relationship we are facing. It would be highly desirable to develop
the intellectual discipline of comparative military doctrine, especially in
the strategic sphere. Systematic comparative studies of strategic doctrine
could serve t o clarify what we think and how we ourselves differ on
these matters, as well as to organize what we know about Soviet strategic
thinking.
I II ,I i 1 American and Soviet Strategic Thought 19 3
Although many have and express views on how both the United States
and the Soviet Union deal with strategic problems, there is in fact little sys-
tematic comparison of the conceptual and behavioral foundations of our
respective strategic activity. In this area, more than other comparative
inquiries into communist and non-communist politics, there are the obs-
tacles of secrecy in the path of research. Perhaps as vital, neither government
nor academic institutions appear to have cultivated many people with the
necessary interdisciplinary skills and experience.
The most influential factor that has inhibited lucid comparisons of U.S.
and Soviet strategic thinking has been the uncritically held assumption that
they had to be very similar, or a t least converging with time. Many of 11s have
been quite insensitive to the possibility that two very different political
systems could deal very differently with what is, in some respects, a common
problem. We understood the problem of keeping the strategic peace on equit-
able and economical terms - or so we thought. As reasonable men the Soviets,
too, would come to understand it our way.
Explaining this particular expression of our cultural self-centeredness is
itself a fascinating field for speculation. I think it goes beyond the American
habit of value projection. It may result from the fact that post-war devel-
opments in U S . strategy were an institutional and intellectual offspring of
the natural sciences that spawned modern weapons. Scientific truth is trans-
national, not culturally determined. But, unfortunately, strategy is more like
politics than like science.
The next five to ten years of the U.S.-Soviet strategic relationship could
well be characterized by mounting U.S. anxieties about the adequacy of our
deterrent forces and our strategic doctrine. There seems to be little real
prospect that the SALT process, as we have been conducting it, will sub-
stantially alleviate these anxieties. Even if a more promising state of affairs
emerges, however, it is hard t o see us managing it with calm and confidence
unless we develop a more thorough appreciation of the differences between
U.S. and Soviet strategic thinking. Things have progressed beyond the point
where it is useful t o have the three familiar schools of thought on Soviet
doctrine arguing past each other: one saying "Whatever they say, they think
as we do;" the second insisting, "Whatever they say, it does not matter;"
and the third contending, "They think what they say, and are therefore out
for superiority over us."
Comparative strategic doctrine studies should address systematically a
series of questions:
What are the central decisions about strategy, force posture, and force
employment or operations that doctrine is supposed t o resolve for the sides
examined?
What are the prevailing categories, concepts, beliefs, and assertions that
appear to constitute the body of strategic thought and doctrine in question?
What are the hedges and qualifications introduced t o modify the main
theses of official thinking?
What are the "non-strategic," e.g., propagandistic, purposes that might

motivate doctrinal pronouncements? Does the doctrinal system recog-
nize a distinction between what ideally ought to be, and what practically
is (a serious problem in the Soviet case)?
In what actions, e.g., force posture, does apparent doctrine have prac-
tical effect? Where does it lie dormant?
To what extent are doctrinal pronouncements the subject of or the guise
for policy dispute?
What perceptions does one side entertain as to the doctrinal system of
the other side? With what effect?
Answering these questions for both the United States and the Soviet
Union is admittedly n o easy matter, especially in a highly politicized envir-
onment in which many participants have already made up their minds how
they want the answers to come out with respect to assumed impact on U.S.
strategic policy. But we have the data to d o a good deal better than we have
to date.
U.S.a n d Soviet Doctrine Contrasted
What is U.S. strategic doctrine and policy? What is Soviet strategic doctrine
and policy? The Soviets provide definitions of doctrine (doktrina) and policy
(politika)that state they are official principles, guidance, and instructions from
the highest governing authorities to provide for the building of the armed
forces and for their employment in war.
The most useful thing about these definitions is that they remind us - or
should -that we d o not have direct and literal access to Soviet strategic doc-
trine and policy through the most commonly available sources, i.e., Soviet
military literature and various pronouncements of authoritative political
and military figures. Our insight into Soviet strategic policy is derived by
inference from such sources along with inferences from observed R&D and
force procurement behavior, what we manage to learn about peacetime
force operations and exercises, and occasional direct statements in more
privileged settings, such as SALT, by varyingly persuasive spokesmen.
The value of all these sources is constrained by the limitations of our
perceptive apparatus, technical and intellectual, and the fact that Soviet
communications on strategic subjects serve many purposes other than con-
veying official policy, such as foreign and domestic propaganda. For all
that, we have gained over the years a substantial degree of understanding
of the content of Soviet strategic thinking, of the values, standards, object-
ives, and calculations that underlie Soviet decisions. It is this total body of
thinking and its bearing on action that are of concern here.
Where lack of access complicates understanding of Soviet strategic doc-
trine, a n overabundance of data confuses understanding of the American
side, a point that Soviets make with some justice when berated with the evils
i I rndrt h American and Soviet Strategic Thought 195
of Soviet secrecy. If, in the case of the United States, one is concerned about
the body of thinking that underlies strategic action it is clearly insufficient to
rely on official statements or documents a t any level of classification or
authority. Such sources may, for one reason o r another, not tell the whole
story or paper over serious differences of purpose behind some action.
One of the difficulties in determining the concepts or beliefs that underlie
U.S. strategic action is that strategic policy is a composite of behavior taking
place in at least three distinguishable, but overlapping arenas. The smallest,
most secretive, and least significant over the long-term, assuming deterrence
does not fail, is the arena of operational or war planning. The second arena
is that of system and force acquisition; it is much larger and more complex
than the first. The most disorganized and largest, but most important for the
longer-term course of U.S. strategic behavior is the arena of largely public
debate over basic strategic principles and objectives. Its participants range
from the most highly placed executive authorities to influential private elites,
and occasionally the public a t large. Strategy-making is a relatively demo-
cratic process in the United States.
To be sure, may areas of public policymaking can be assessed in terms
of these overlapping circles of players and constituents. But the realm of
U.S. strategic policy may be unusual in the degree to which different rules,
data, concerns, and participants dominate the different arenas. These dif-
ferences make it difficult t o state with authority what U.S. strategic policy
is on an issue that cuts across the arenas. For example, public U.S. policy
may state a clear desire t o avoid counter-silo capabilities on stability
grounds. The weapons acquisition community may, for a variety of reasons,
simultaneously be seeking a weapons characteristic vital t o counter-silo
capability, improved ballistic missile accuracy. As best they can with
weapons available, meanwhile, force operators may be required by the logic
of their task t o target enemy missile silos as a high priority.
Despite these complexities, however, it is ~ o s s i b l eto generalize a body of
policy concepts and values that govern U.S. strategic behavior. There are
strong tendencies that dominate U.S. strategic behavior in the areas of declara-
tory policy, force acquisition, and arms control policy. Again, the case of US.
counter-silo capabilities may be cited. Today, the United States lacks high con-
fidence capabilities against Soviet missile silos; it may continue to lack them
for some time or indefinitely. This is in part the result of technological choice,
the early selection of small ICBMs and the deployment of low-yield MIRV
weapons. It is also the result of Soviet efforts to improve silo hardness. But the
main reason for this lack is that we have abided by a conscious judgment - that
a serious counter-silo capability, because it threatens strategic stability, is a bad
thing for the United States to possess.
The situation seems more straightforward, if secretive, on the Soviet side.
Soviet strategic policymaking takes place in a far more vertical and closed
system. Expertise is monopolized by the military and a subset of the top
political leadership. Although elites external t o this group can bid for its
scarce resources to some extent, they cannot seriously challenge its values
and judgments. Matters of doctrine, force acquisition, and war planning are
much more intimately connected within this decision group than in the
United States. Policy arguments are indeed possible. Public evidence suggests
a series of major Soviet debates on nuclear strategy from the mid-1950s to
the late 1960s, although identification of issues, alternatives, and parameters
in these debates must be somewhat speculative.
These considerations make difficult, but not impossible, the comparative
treatment of U.S. and Soviet strategic belief systems and concepts. One may
describe with some confidence how the two very different decision systems
deal with certain concerns central t o the strategic nuclear predicament of
both sides. Much about U.S. and Soviet strategic belief systems can be cap-
tured by exploring how they treat five central issues: (1) the consequences of an
all-out strategic nuclear war, (2) the phenomenon of deterrence, (3) stability,
(4) distinctions and relationships between intercontinental and regional
strategic security concerns, and ( 5 )strategic conflict limitation.
C o n s e q u e n c e s of N u c l e a r W a r
For a generation, the relevant elites of both the United States and the Soviet
Union have agreed that an unlimited strategic nuclear war would be a
sociopolitical disaster of immense proportions. Knowing the experiences of
the peoples of the Soviet Union with warfare in this century and with
nuclear inferiority since 1945, one sometimes suspects that the human
dimensions of such a catastrophe are more real to Russians, high and low,
than to Americans, for whom the prospect is vague and unreal, if certainly
- -
forbidding.
For many years the prevailing U.S. concept of nuclear war's consequences
has been such as to preclude belief in any military or politically meaningful
form of victory. Serious effort on the part of the state to enhance the
prospect for national survival seemed quixotic, even dangerous. Hence
stems our relative disinterest in air defenses and civil defenses over the last
fifteen years, and our genuine fear that ballistic missile defenses would be
severely destabilizing. Growth of Soviet nuclear power has certainly clinched
this view of nuclear conflict among critical elements of the U.S. elite. But
even when the United States enjoyed massive superiority, when the Soviet
Union could inflict much less societal damage on the United States, and then
only in a first strike (through the early 1960s), the awesome destructiveness
of nuclear weapons had deprived actual war with these weapons of much
of its strategic meaning for the United States.
The Soviet system has, however, in the worst of times, clung tenaciously
to the belief that nuclear war cannot - indeed, must not - be deprived of
strategic meaning, i.e., some rational relationship to the interests of the state.
It has insisted that, however awful, nuclear war must be survivable and some
kind of meaningful victory attainable. As most are aware, this issue was
debated in various ways at the beginning and end of the Khrushchev era,
with Khrushchev on both sides of the issue. But the system decided it had to
I I 111,ii t i I American and Soviet Strategic Thought 197
believe in survival and victory of some form. Not so to believe would mean
that the most basic processes of history, on which Soviet ideology and polit-
ical legitimacy are founded, could be derailed by the technological works of
man and the caprice of an historically doomed opponent. Moreover, as the
defenders of doctrinal rectitude continued to point out, failure to believe in
the "manageability" of nuclear disaster would lead to pacificism, defeatism,
and lassitude in the Soviet military effort. This should not be read as the tri-
umph of ideological will over objective science and practical reason. From
the Soviet point of view, nuclear war with a powerful and hostile America
was a real danger. Could the state merely give up on its traditional respon-
sibilities to defend itself and survive in that event? Their negative answer
hardly strikes one as unreasonable. Their puzzlement, alternating between
contemptuous and suspicious, over U.S. insistence on a positive answer is
not surprising.
In recent years the changing strategic balance has had the effect of
strengthening rather than weakening the asymmetry of the two sides' con-
victions on this matter. Dubious when the United States enjoyed relative
advantage, strategic victory and survival in nuclear conflict have become
the more incredible to the United States as the strategic power o f the
Russians has grown. For the Soviets, however, the progress of arms and
war-survival programs has transformed what was in large measure an ide-
ological imperative into a more plausible strategic potential. For reasons to
be examined below, Soviet leaders possibly believe that, under favorable
operational conditions, the Soviet Union could win a central strategic war
today. Notwithstanding strategic parity or essential equivalence of force,
they may also believe they could lose such a conflict under some conditions.
Deterrence
The concept of deterrence early became a central element of both U.S. and
Soviet strategic belief systems. For both sides the concept had extended or
regional dimensions, and a good deal of political content. There has, in
short, been some functional symmetry between the deterrence thinking of
the two sides: restraint of hostile action across a spectrum of violence by the
threat of punishing consequences in war. Over time and with shifts in the
overall military balance, latent asymmetries of thinking have become more
pronounced. For the United States, strategic deterrence has tended to
become the only meaningful objective of strategic policy, and it has become
progressively decoupled from regional security. For the Soviets, deterrence -
or war prevention - was the first, but not the only and not the last objec-
tive of strategy. Deterrence also meant the protection of a foreign policy
that had both offensive and defensive goals. And it was never counterposed
against the ultimate objective of being able to manage a nuclear war suc-
cessfully should deterrence fail. The Soviet concept of deterrence has
evolved as the strategic balance has improved for the Soviet Union from
primary emphasis on defensive themes of war prevention and protection of
prior political gains to more emphasis on themes that include the protection
of dynamic processes favoring Soviet international interests. Repetition of
the refrain that detente is a product of Soviet strategic power, among other
things, displays this evolution.
Stability
Strategic stability is a concept that is very difficult to treat in a comparative

manner because it is so vital to U.S. strategic thinking, but hardly identifiable
in Soviet strategic writings. In U.S. thinking, strategic stability has meant a
condition in which incentives inherent in the arms balance to initiate the use
of strategic nuclear forces and, closely related, to acquire new or additional
forces are weak or absent. In an environment dominated by powerful offen-
sive capabilities and comparatively vulnerable ultimate values, i.e., societies,
stability was thought to be achievable on the basis of a contract of mutually
vulnerable societies and survivable offensive forces. Emphasis on force surviv-
ability followed, as did relative uninterest in counterforce, active, and passive
defenses.
Soviet failure to embrace these notions is sufficiently evident not to
require much elaboration. One may argue about Soviet ability to overturn
stability in U.S. terms, but not about Soviet disinclination to accept the idea
as a governing principle of strategic behavior. Soviet acceptance of the ABM
agreement in 1972 is still frequently cited as testimony to some acceptance
of this principle. It is much more probable, however, that the agreement was
attractive to Moscow because superior U.S. ABM technology plus superior
U.S. ABM penetrating technology would have given the United States a
major advantage during the mid- to late 1970s. In a unilateral sense, the
Soviets saw the ABM agreement as stabilizing a process of strategic catch-
up against a serious risk of reversal. But it did not mean acceptance of the
U.S. stability principle.
The United States has always been relatively sensitive to the potential of
technology to jeopardize specific formulae for achieving stability, although
it has been relatively slow t o perceive the pace and extent to which com-
parative advantage has shifted from passive survivability to counterforce
technologies. The Soviets have also been sensitive to destabilizing technol-
ogies. But they have tended to accept the destabilizing dynamism of tech-
nology as an intrinsic aspect of the strategic dialectic, the underlying engine
of which is a political competition not susceptible to stabilization. For the
Soviets, arms control negotiations are part of this competitive process. Such
negotiation can help keep risks within bounds and also, by working on the
U.S. political process, restrain U.S. competitiveness.
Soviet failure to embrace U.S. strategic stability notions as strategic norms
does not mean, as a practical matter, that the Soviets fail to see certain con-
stellations of weapons technology and forces as having an intrinsic stability, in
that they make the acquisition of major advantages very difficult. What they
reject is the notion that, in the political and technical world as they see it, those
f I m,ir ti1 American and Soviet Strategic Thought 199
constellations can be frozen and the strategic competition dimension thereby

factored out of the East-West struggle permanently or for long periods.
Intercontinental a n d Regional
Defining the boundary line between strategic and non-strategic forces has
been a troubling feature of SALT from the beginning. It is one of diplo-
macy's minor ironies that forward capabilities the United States has long
regarded as part of the general purpose forces we have been hard pressed
t o keep out of the negotiations. But peripheral strike forces the Soviets have
systematically defined and managed as strategic seem very difficult t o bring
into the picture.
Geography imparted an intercontinental meaning t o the term strategic
for the United States. The same geography dictated that, for the Soviet
Union, strategic concern began a t the doorstep. Soviet concern about the
military capabilities in the hands of and on the territory of its neighbors is
genuine, although Soviet arguments for getting the United States t o legit-
irnatize and pay for those concerns a t SALT in terms of its own central force
allowances have been a bit contrived. They are tantamount t o penalizing
the United States for having friends, while rewarding the Soviet Union for
conducting itself in a manner that has left it mostly vassals and opponents
on its borders.
Underlying these definitional problems are more fundamental differ-
ences between U.S. and Soviet doctrines on what is generally called "coup-
ling." It has long been U.S. policy t o assure that U.S. strategic nuclear
forces are seen by the Soviets and our NATO allies as tightly coupled to
European security. Along with conventional and theater nuclear forces, U.S.
strategic nuclear forces constitute an element of the NATO "triad." The
good health of the alliance politically and the viability of deterrence in
Europe have been seen t o require a very credible threat to engage U.S.
strategic nuclear forces once nuclear weapons come into play above the level
of quite limited use. For more than twenty years NATO's official policy has
had t o struggle against doubts that this coupling could be credible in the
absence of clear U.S. strategic superiority. Yet the vocabularly we commonly
employ itself tends t o strain this linkage in that theater nuclear forces
are distinguished from strategic. Ironically, the struggle to keep so-called
Forward Based Systems out of SALT, because we could not find a good way
to bring in comparable Soviet systems, tended t o underline the distinction.
In o u r thinking about the actual prosecution of a strategic conflict, once
conflict a t that level begins we tend t o forget about what might be the local
outcome of the regional conflict that probably precipitated the strategic
exchange.
The Soviets, on the other hand, appear to take a more comprehensive
view of strategy and the strategic balance. Both in peacetime political com-
petition and in the ultimate test of a central conflict, they tend to see all force
elements as contributing to a unified strategic purpose, national survival and
the elimination or containment of enemies on their periphery. The U.S.S.R.

tends t o see intercontinental forces, and strategic forces more generally, as a
means to help it win an all-out conflict in its most crucial theater, Europe.
Both institutionally and operationally, Soviet intercontinental strike forces
are an outgrowth and extension of forces initially developed to cover periph-
eral targets. Land combat forces, including conventional forces, are carefully
trained and equipped to fight in nuclear conditions. In the last decade, the
emergence of a hostile and potentially powerful China has more firmly
riveted the "rimland" of Eurasia into the Soviet strategic perspective.
Whatever the consequence of a central U S S o v i e t nuclear conflict for
their respective homelaids, it could well have the effect of eliminating U.S.
power and influence on the Eurasian landmass for a long time. If, by virtue
of its active and passive damage-limitation measures, the Soviet Union suf-
fered measurably less damage than did the United States, and it managed to
intimidate China or destroy Chinese military power, the resultant Soviet dom-
ination of Eurasia could represent a crucial element of "strategic victory" in
Soviet eyes. In any case, regional conflict outcomes seem not to lose their
significance in Soviet strategy once strategic nuclear conflict begins.
Conflict Limitation
Nuclear conflict limitation is a theme on which influential American opinion

is divided. After much thought and argument, the previous administration
adopted a more explicit endorsement of limited strategic nuclear options as a
hedge against the failings of a strategy solely reliant on all-out war plans for
deterrence or response in the event of deterrent failure. The present adminis-
tration has appeared more doubtful about the value of limited nuclear
options because it appears generally to doubt the viability of nuclear conflict
limitations. It may also share the fear of some critics that limited options
could seem to make nuclear use more tolerable and therefore detract from
deterrence.
Theories of nuclear conflict limitation entertained in the United States
tend t o rest on concepts of risk management and bargaining with the oppon-
ent. We are interested in limited options because they are more credible
than unlimited ones in response to limited provocation. Whether or not
they can be controlled is uncertain; hence their credible presence enhances
the risk faced by the initiator of conflict. Should conflict come about, then
limited options might be used to change the risk, cost, and benefit calculus
of the opponent in the direction of some more or less tolerable war termin-
ation. This would not be a sure thing, but better have the limited options
than not.
How the Soviets view the matter of nuclear conflict limitations is obscure.
The least one can say is that they d o not see it in the manner described above.
From the early 1960s, after McNamara's famed Ann Arbor speech, Soviet
propagandists have denounced limited nuclear war concepts as U.S. con-
trivances to make nuclear weapons use more "acceptable" and to rationalize
3 r I American and Soviet Strategic Thought 201
the quest for counterforce advantages. They have replayed the criticism that
such concepts weaken deterrence and cannot prevent nuclear war from
becoming unlimited.
To some degree, Soviet propaganda on this theme is suspect for being
aimed at undermining U.S. strategy innovations that detract from the political
benefits of Soviet strategic force improvement. Given differences of view in the
United States o n this subject, moreover, the Soviets could hardly resist the
temptation to fuel the U.S. argument. There are several reasons why Soviet
public pronouncements should not be taken as entirely reflecting the content
of operative Soviet strategic thinking and planning regarding limited nuclear
use. For one thing, qualified acceptance in doctrine and posture of a non-
nuclear scenario, or at least a non-nuclear phase, in theater conflict displays
some Soviet willingness to embrace conflict limitation notions previously
rejected. Soviet strategic nuclear force growth and modernization, in addition,
have given Soviet operational planners a broader array of employment options
than they had in the 1960s and may have imparted some confidence in Soviet
ability to enforce conflict limitations. It would not be surprising, therefore, to
find some Soviet contingency planning for various kinds of limited nuclear
options at the theater and, perhaps, at the strategic level.
One may seriously doubt, however, whether Soviet planners would
approach the problem of contingency planning for limited nuclear options
with the conceptual baggage the U.S. system carries. It would seem contrary
t o the style of Soviet doctrinal thinking to emphasize bargaining and risk
management. Rather the presence of limited options planning in the Soviet
system would seem likely t o rest on more traditional military concepts of
economizing on force use, controlling actions and their consequences,
reserving options, and leaving time t o learn what is possible in the course
of a campaign. The Soviet limited options planner would seem likely t o
approach his task with a more strictly unilateral set of concerns than his
American counterpart.
Methods of Assessing the Strategic Balance
Comparative study of U.S. and Soviet strategic doctrine should give atten-
tion t o a closely related matter: how we perceive and measure force bal-
ances. Allusion has already been made to asymmetries between U.S. and
Soviet definitions of strategic forces, what should be counted in SALT, etc.
This is by no means the heart of the matter. U.S. and Soviet methodologies
for measuring military strength appear to differ significantly.
Many rather amateurish and misleading beliefs about the way the
Soviets measure and value military strength prevail; for example, that the
Soviets have some atavistic devotion t o mass and size. Mass they d o believe
in because both experience and analysis show that mass counts. They can
be quite choosey about size, however, as a look a t their tank and fighter
designs reveals. Within the limits of their technological potential, they have
been quite sensitive and in no way primitive in their thinking about quality/
quantity tradeoffs.
Another widespread notion is that the Soviets have an unusual propensity
for worst-case planning or military overinsurance. This is hard to demonstrate
convincingly in Soviet behavior. The Soviet theory of war in central Europe,
for example, is daring, not conservative. Despite much rhetoric on the danger
of surprise and the need for high combat readiness, Soviet strategic planning
has not accorded nearly the importance to "bolt-from-the-blue" surprise
attack that the United States has. This does not look like overinsurance.
The problem of measuring strength goes more deeply to differing appre-
ciations of the processes of conflict and how they bear on force measure-
ment. U.S. measures of the overall strategic balance tend to be of two general
types. First come the so-called static measures of delivery vehicles, weapons,
megatonage and equivalent megatonage, throwweight, and, perhaps, some
measure of hard-target kill (such as weapon numbers times a
scaled yield factor divided by the square of Circular Error Probability).
Comparisons of this type can display some interesting things about differing
forces. But they say very little about how those forces, much less the nations
that employ them, will fare in war. By themselves, static measures can be
dangerously misleading.
We then move on to the second, or quasi-dynamic, class of measures. Here
the analyst is out to capture the essential features of a "real war" in terms
general enough to allow parametric application, frequent reiteration of the
analysis with varying assumptions, and easy swamping of operational and
technical details which he may not be able to quantify or of which he may be
ignorant. Typically, certain gross attributes of the war "scenario" will be
determined, e.g., levels of alert, who goes first, and very general targeting pri-
orities. Then specified "planning factor" performance characteristics are
attributed to weapons. Because it is relatively easy (and fun), a more or less
elaborate version of the ICBM duel is frequently conducted. The much more
subtle and complicated, but crucial, engagement of air and sea-based forces
is usually handled by gross assumption, e.g., n percent of bomber weapons
get to target, all SSBNs at sea survive. Regional conflicts and forces are typ-
ically ignored. Of course, all command/control/communications systems are
assumed to work as planned - otherwise the forces, and even worse, the ana-
lyst would be out of business. Finally, "residuals" of surviving forces, fatality
levels, and industrial damage are totaled up. A popular variant is to run a
countermilitary war in these terms and then see whether residual forces are
sufficient to inflict "unacceptable damage" on cities. If so, then deterrence is
intact according to some. Others point to grossly asymmetric levels of sur-
viving forces to document an emerging strategic imbalance.
Most specialists agree and explicitly admit that this kind of analysis does
not capture the known, much less the unknown complexities, uncertainties,
and fortuities of a real strategic nuclear conflict of any dimension. Such litur-
gical admissions are usually offered to gain absolution from their obvious con-
sequences, namely that the analysis in question could be, not illuminating, but
I :rri,ir~h American and Soviet Strategic Thought 203
quite wrong. However, more heroic analytic attempts at capturing the real
conlplexity and operational detail of a major nuclear exchange are usually not
made because they are: a ) usually beyond the expertise of single analysts or
small groups, b) not readily susceptible t o varied and parametric application,
and c) still laden by manifold uncertainties and unknowns that are very hard
to quantify. Hence they are very hard t o apply to the tasks of assessing stra-
tegic force balances or the value of this or that force improvement. The more
simplistic analysis is more convenient. The analyst can conduct it many times,
and talk over his results with other analysts who d o the same thing. The whole
methodology thereby acquires a reality and persuasiveness of its own.
The influence of this kind of analysis in our strategic decision system has
many explanations. It has sociological origins in the dominance of econo-
mists and engineers over soldiers in the conduct of our strategic affairs. It
conforms with the needs of a flat and argumentative policy process in which
there are many and varied participants, from generals to graduate students.
They need a common idiom that does not soak up too much computer time
and can be unclassified. And finally, in part because of the first explanation
cited, when it comes to nuclear strategy, we d o not believe much in "real"
nuclear war anyway. We are after a standard of sufficiency that is adequate
and persuasive in a peacetime setting.
Two things about this style of strategic analysis merit staring in the context
of this paper. First, on the face of it, the value of simplistic, operationally-
insensitive methodologies is assuredly less in the present strategic environment
than it was when the United States enjoyed massive superiority. Not only are
weapons, force mixes, and scenarios more complicated than these methodo-
logies can properly illuminate, but the relative equality of the two sides going
into the conflict makes the subleties, complexities, and uncertainties all the
more important for how they come out. Second, the Soviets d o not appear to
do their balance measuring in this manner.
One can gain a fair insight into the manner of Soviet force balance analy-
sis from public sources, particularly Soviet military literature. Additional
inferences can be drawn from the organization and professional composition
of the Soviet defense decision system, and from some of the results of Soviet
decisions. O n the whole it appears that Soviet planners and force balance
assessors are much more sensitive than we are to the subtleties and uncer-
tainties - what we sometimes call "scenario dependencies" - of strategic
conflict seen from a very operational perspective. The timing and scale of
attack initiation, tactical deception and surprise, uncertainties about weapons
effects, the actual character of operational plans and targeting, timely
adjustment of plans to new information, and, most important, the continued
viability of command and control - these factors appear to loom large in
Soviet calculations of conflict outcomes.
The important point, however, is a conceptual one: Unlike the typical
U.S. planner, the Soviet planner does not appear t o see the force balance
prior t o conflict as a kind of physical reification of the war outcome and
therefore as a measure of strategic strength by itself. Rather he seems to see
the force balance, the "correlation of military forces," as one input to a

complex combat process in which other factors of great significance will
play, and the chief aim of which is a new, more favorable balance of forces.
The sum of these factors is strategy, and strategy is a significant variable to
the Soviet planner.
As a generalization, then, the Soviet planner is very sensitive to oper-
ational details and uncertainties. Because these factors can swing widely,
even wildly, in different directions, a second generalization about Soviet
force analysis emerges: a given force balance in peacetime can yield widely
varying outcomes to war depending on the details and uncertainties of com-
bat. Some of those outcomes could be relatively good for the Soviet Union,
others relatively bad. The planner's task is to improve the going-in force
balance, t o be sure. But it is also to develop and pursue ways of waging war
that tend to push the outcome in favorable directions.
This kind of thinking occasions two very unpleasant features in Soviet
military doctrine: a strong tendency to preempt and a determination to sup-
press the enemy's command and control system at all costs. The Soviets tend
to see any decision t o go to nuclear war as being imposed on them by a
course of events that tells them "war is coming," a situation they bungled
memorably in June 1941. It makes no difference whose misbehavior started
events on that course. Should they find themselves on it, their operational
perspective on the factors that drive war outcomes places a high premium on
seizing the initiative and imposing the maximum disruptive effects on the
enemy's forces and war plans. By going first, and especially disrupting com-
mand and control, the highest likelihood of limiting damage and coming out
of the war with intact forces and a surviving nation is achieved, virtually
independent of the force balance.
This leads to a final generalization. We tend rather casually to assume that,
when we talk about parity and "essential equivalence" and the Soviets about
"equal security," we are talking about the same thing: functional strategic
stability. We are not. The Soviets are talking about a going-in force balance in
which they have an equal or better chance of winning a central war, if they can
orchestrate the right scenario and take advantage of lucky breaks. It is the job
of the high command to see that they can. If it fails to do so, the Soviet Union
could possibly lose the war. This is not stability in our terms.
Again, this is not to argue that the Soviets d o not foresee appalling
destruction as the result of any strategic exchange under the best of condi-
tions. In a crisis, Soviet leaders would probably take any tolerable and even
some not-very-tolerable exits from the risk of such a war. But their image
of strategic crisis is one in which these exits are closing up, and the "war is
coming". They see the ultimate task of strategy to be the provision of forces
and options for preempting that situation. This then leads them to choose
strategies that, from a U.S. point of view, seem not particularly helpful in
keeping the exits open, and even likely to close them off.
It is frequently argued - more frequently as we become more anxious
about the emerging force balance -that the Soviets could not have confidence
4 American and Soviet Strategic Thought 205
in launching a strategic attack and achieving the specific objectives that the-
oretical analysis might suggest to be possible, such as destruction of
Minuteman. Particularly because they are highly sensitive to operational
uncertainties they would not, in one of the more noteworthy phrases of the
latest Defense Department posture statement, gamble national survival on a
"single cosmic throw of the dice." This construction of the problem obscures
the high likelihood that decisions to go to strategic war will be made under
great pressure and in the face of severe perceived penalty if the decision is not
made and the war comes anyway. They are not likely to come about in a situ-
ation in which the choice is an uncertain war or a comfortable peace. It also
obscures the fact that the heavy weight o f uncertainty will also rest o n the
shoulders of U.S. decisionmakers in a crisis.
Dangers of Misunderstanding
In sum, there are fundamental differences between U.S. and Soviet strategic
thinking, both at the level of value and at the level of method. The existence
of these differences and, even more, our failure to recognize them have had
dangerous consequences for the U.S.-Soviet strategic relationship.
One such might be called the "hawk's lament." Failing to appreciate the
character of Soviet strategic thinking in relation to our own views, we have
underestimated the competitiveness of Soviet strategic policy and the need
for con~petitiveresponsiveness on our part. This is evident in both our SAI,T
and our strategic force modernization behavior.
A second negative effect might be termed the "dove's lament." By pro-
jecting our views onto the Soviets, and failing to appreciate their real
motives and perceptions, we have underestimated the difficulties of achiev-
ing genuine strategic stability through SALT and over-sold the value of what
we have achieved. This has, in turn, set us up for profound, perhaps even
hysterical, disillusionment in the years ahead, in which the very idea of
negotiated arms control could be politically discredited. If present strategic
trends continue, it is not hard to imagine a future political environment in
which it would be difficult to argue for arms control negotations even of a
very hard-nosed sort.
The third and most dangerous consequence of our misunderstanding of
Soviet strategy involves excessive confidence in strategic stability. U.S. stra-
tegic behavior, in its broadest sense, has helped to ease the Soviet Union onto
a course of more assertive international action. This has, in turn, increased
the probability of a major East-West confrontation, arising not necessarily
by Soviet design, in which the United States must forcefully resist a Soviet
advance or face collapse of its global position, while the Soviet Union can-
not easily retreat or compromise because it has newly acquired global power
status to defend and the matter at issue could be vital. In such conditions, it
is all too easy to imagine a "war is coming" situation in which the abstract
technical factors on which we rest our confidence in stability, such as expected
force survival levels and "unacceptable damage," could crumble away. The
strategic case for "waiting to see what happens," for conceding the opera-
tional initiative to the other side - which is what crisis stability is all about -
could look very weak. Each side could see the great operational virtues of
preemption, be convinced that the other side sees them too, and be hourly
more determined that the other side not have them. This, in any case, could
be the Soviet way of perceiving things. Given the relative translucence of U.S.
versus Soviet strategic decision processes, however, our actual ability to pre-
empt is likely to be less than the Soviets', quite apart from the character of
the force balance. Add to that the problem of a vulnerable Minuteman
ICBM force and you have a potentially very nasty situation.
What we know about the nature of our own strategic thinking and that
of the Soviet Union is not at all comforting at this juncture. The Soviets
approach the problem of managing strategic nuclear power with highly
competitive and combative instincts. Some have argued that these instincts
are largely fearful and defensive, others that they are avaricious and confi-
dent. My own reading of Russian and Soviet history is that they are both,
and, for that, the more difficult to handle.
The United States and the Soviet Union share two awesome problems in
common, the creation of viable industrial societies and the management of
nuclear weapons. Despite much that is superficially common t o our heri-
tages, however, these two societies have fundamentally different political
cultures that determine how they handle these problems. The stamp of a
legal, commercial, and democratic society is clearly seen in the way the
United States has approached the task of managing nuclear security. Soviet
styles of managing this problem bear the stamp of an imperial, bureau-
cratic, and autocratic political tradition. While the United States is willing
t o see safety in a compact of "live and let live" under admittedly unpleas-
ant conditions, the Soviet Union operates from a political tradition that sus-
pects the viability of such deals, and expects them, at best, to mark the
progress of historically ordained forces to ascendancy.
It is not going t o be easy to stabilize the strategic competition on this
foundation of political traditions. But if we understand the situation clearly,
there should be no grounds for fatalism. Along with a very uncomfortable
degree of competitiveness, Soviet strategic policy contains a strong element
of professionalism and military rationalism with which we can do business
in the interest of a common safety if we enhance those qualities in ourselves.
The Soviets respect military power and they take warfare very seriously.
When the propaganda and polemics are pared away, they sometimes won-
der if we do. We can make a healthy contribution t o our own future, and
theirs, by rectifying this uncertainty.
Why Nuclear Superiority Doesn't Matter
Robert Jervis
Assured Destruction and Flexible Response
P
roponents of AD argue that the vulnerability of population centers in
both the United States and the Soviet Union that conies with mutual
second-strike capability has transformed strategy. Because a military
advantage n o longer assures a decisive victory, old ways of thinking are no
longer appropriate. The healthy fear of devastation, which cannot be exor-
cised short of the attainment of a first-strike capability, makes deterrence
relatively easy. Furthermore, because cities cannot he taken out o f hostage,
the perceived danger of total destruction is crucial at all points in the threat,
display, or use of force.
Four implications follow. First, because gaining the upper hand in purely
military terms cannot protect one's country, various moves in a limited war -
such as using large armies, employing tactical nuclear weapons, or even
engaging in limited strategic strikes - are less important for influencing the
course of the battle than for showing the other side that a continuation of
the conflict raises an ~lnacceptabledanger that things will get o ~ i of t hand.
New weapons are introduced not to gain a few miles of territory, but to
engage in what Schelling has called competition in risk taking.' Escalation
dominance - the ability to prevail at every level of military conflict below
that of all-out war - is thus neither necessary nor sufficient to reach one's
goals, be they to preserve the status quo or to change it. Being able to win
on the battlefield does not guarantee winning one's objectives, since the risk
of escalation may be too great to justify the expected benefits.
Second, it does not matter which side has more nuclear weapons. In the
past, having a larger army than one's neighbor allowed one to conquer it
and protect one's own population. Having a larger nuclear stockpile yields
no such gains. Deterrence comes from having enough weapons to destroy
the other's cities; this capability is an absolute, not a relative, one.'
Source: /'olitic~z/S&WCC Q ~ ~ a r t t ' r94(4)
~ y , ( 1979-80): 6 17-33,
Third, if national security is provided by one's capability to destroy the

opponent, not by the possession of a more effective military machine than
the other side, then the force that drives the security dilemma is sapped. The
security dilemma is created by the fact that in the prenuclear era weapons
and policies that made one country secure made others insecure. An army
large enough t o protect the state was usually large enough to threaten a
neighbor with invasion, even if the state did not intend such a threat. But
when security comes from the absolute capability to annihilate one's enemy,
then each side can gain it simultaneously. Neither side need acquire more
than a second-strike capability and, if either does, the other need not respond
since its security is not threatened.3
A fourth aspect of the AD position is that nuclear war is very unlikely
because to initiate it a statesman would have to be willing to run the risk
that his country's population centers would be destroyed. Not only is "the
balance of terror ... decidedly not d e l i ~ a t e , "but,
~ because statesmen know
that imprudent action could lead to all-out war, the resulting deterrence
covers a lot more than attacks on one's homeland. To take any major offen-
sive action is to run an intolerably high risk of escalation. The United States
and the Soviet Union may engage in fierce rhetorical battles and even use
force in such peripheral areas as Africa and Asia, but there are sharp limits
to how far they can push each other. The chance that such attempts would
lead t o total destruction is simply too great. (And it can be too great even
though it is very low. That is, even a very small probability of escalation is
sufficient to deter serious encroachments.)
The Flexible Response position is different on all counts. Its logic is best
seen in terms of what Glenn Snyder has called the stability-instability para-
dox."ecause the balance is so stable at the level of all-out nuclear war, each
side is relatively free to engage in provocations and military actions at lower
levels of violence. The most obvious application of this argument is that if
NATO lacks the ability to defend Europe with conventional weapons, it
faces the danger of having to fight such a war: thus the Soviet second-strike
capability would "deter our deterrent" (to paraphrase the title from a n
article by Paul Nitze)."he same argument can be applied to more bizarre
situations. To secure some highly valued goal the Russians might destroy
most of the American Minuteman force. Since its cities were still in hostage,
the United States would be deterred from striking back at Soviet cities.
For the advocates of FR, the United States must be prepared to fight a
war - or rather a variety of wars - in order to gain a better chance of deter-
ring the Soviets from making any military moves, to deter them from escal-
ating if they d o move, and to secure as favorable an outcome as possible at
any level of violence. In contrast to the AD view, FR argues that in the
nuclear era, as in earlier times, the absolute amount of armaments on each
side is less important than the relative amounts because each nation's mili-
tary forces as well as its population centers are potential targets. As decision
makers stop thinking that any war must be total and realize that the stability-
instability paradox allows a wider range of contingencies of controlled and
I Nuclear Superiority 209
less self-defeating strikes, the importance of the details of the strategic balance
becomes clear.'
Proponents of FR thus disagree with the AD position that the inherent
riskiness of any major provocation in the nuclear era means that a second-
strike capability protects against much more than an unrestrained assault
o n the country's homeland. Secretary of Defense Brown a r g ~ ~ ethat s "we
now recognize that the strategic nuclear forces can deter only a relatively
narrow range of contingencies, much smaller in range than was foreseen
only 20 or 30 years ago.""imilarly, Brown, like Schlesinger before him,
claims that "only if we have the capability to respond realistically and effect-
ively to an attack at a variety of levels can we ... have the confidence neces-
sary to a credible deterrent."' But, the proponents of AD would reply, this
argument advocating something approximating escalation dominance
misses the point. No state can respond "effectively" in the sense of being
able to take its population centers out of hostage; thus, it is the willingness
to run risks and the perceptions of this willingness that will determine
whether a response is "realistic" and a threat is credible.
Much of the difference between the two schools of thought turns on differ-
ing ideas about stability. Both groups agree on the overwhelming importance
of preserving one's cities. But for the proponents of FR, the common interest
in avoiding a mutually disastrous outcome can be used as a lever to extract
competitive concessions. Either side can take provocative actions because the
other cannot credibly threaten to respond by all-out war. Proponents of AD,
on the other hand, see stability as broader, and deterrence as covering
a wider set of interests, since it follows from the reasonable fear that any
challenge to an opponent's vital interest could escalate. Paradoxically, stability
is in part the product of the belief that the world is not entirely stable, that
things could somehow get out of control.
There are two elements that influence beliefs about the extent to which
the risks of escalation could be kept limited and controlled, and it is not
surprising that advocates of AD and FR disagree about both. The first eleni-
ent is the American reaction and the Soviet anticipation of it. Advocates of
FR fear that the Russians might be certain enough that the United States
would not use nuclear weapons in response to a major provocation to make
such a provocation worth taking. Those who support a policy of AD deny
this, noting that the United States has behaved too unpredictably for any state
to be sure what it will do. Part of the reason for the disagreement on this
point is that proponents of both AD and FR project their views onto the gov-
ernments of the United States and the Soviet Union. The latter believe, and
the former deny, that a large Russian arms build-up would intimidate the
United States.
The other element in the belief about whether the risks would seem
controllable is a judgment about the inherent limits of manipulation and
2 10 The Cold War and Nuclear Deterrence
prediction in human affairs. While these factors are rarely discussed expli-
citly, the tone of much of the FR writings implies that men can make fine,
complex, and accurate calculations. Friction, uncertainty, failures of imple-
mentation, and the fog of battle d o not play a major role. Men see clearly,
their subordinates are able to carry out intricate instructions, and the other
side gets the desired message. Thus, Secretary Brown recently argued that "if
we try bluffing [the Russians with a threat of massive retaliation], ways can
be found by others to test our bluffs without undue risk to them."I0 Such
attempts would involve reasonable risks only if the situation were under
complete control and seen by the Soviets as relatively safe, and then only if
they believed this to be the case. (But many proponents of FR also believe
that the United States cannot rely on tactical nuclear weapons to defend
Europe because their use could too easily lead to all-out war. This fits oddly
with the belief that the superpowers could fight a limited strategic war.)
For the advocates of AD, this is a dream world. War plans can be drawn
up on this basis, but reality will not conform. Furthermore, decision makers,
having experienced the multiple ways in which predictions prove incorrect
and situations get out of control, d o not commit the fallacy of believing that
escalation could be carefully manipulated and thus would not place any
faith in the precise options of limited nuclear warfare. FR advocates see the
need for a policy they consider to be prudential in the sense of being able
to cope with unlikely but dangerous contingencies because they d o not
think decision makers can be counted on to avoid terrible risks; proponents
of AD d o not think American policy has to cover such remote possibilities
because they are confident that statesmen are at least minimally prudent.
This difference in beliefs - or perhaps I should say in intuitions - goes
far to explain why some of the proponents of FR see a much greater dan-
ger of a Russian first strike than d o advocates of AD. One would not expect
any difference of opinion here since the question seems entirely technical.
But it is not. To launch a first strike in the belief that one could destroy most
of an opponent's strategic forces is to accept a set of complex and uncertain
calculations: the weapons have never been tested under fully operational
conditions; accuracies are estimated from performances over test ranges,
which may be different when the missiles are fired over different parts of
the earth; the vulnerability of the other side's silos (and one's own) can
never be known with certainty before the war; and the effects on the envir-
onment of huge nuclear explosions can only be guessed at. The same orien-
tation that leads one to believe that statesmen could be sufficiently
confident of their ability to prevent escalation to allow them to engage in
major provocations also fits with the conclusion that statesmen might place
sufficient confidence in their estimates to launch a disarming strike.
If differences in beliefs about the risks inherent in major provocations
are one source of the dispute between AD and FR, another is a difference
in perceptions of the risks that the Russians are willing to run. Most pro-
ponents of AD argue that while the desire to expand is not completely
absent, the Russians are not so strongly motivated in this regard as to be
It...\ it. Nuclear Superiority 211
willing to endanger what they have already gained. Proponents of FR argue

not that the Russians want war, but that they care enough about increasing
their influence to run significant risks to reach that goal. And by acquiring
massive military might, the Russians could hope to be better able to expand
without courting dangerous confrontations. The proponents of A D would
reply that almost no decision maker in the world's history would embark
on a course of expansion while his cities were held hostage. The sort of
leaders the proponents of FR posit are very rare - even Hitler probably was
not an example, since he knew that if he could militarily defeat the Allies
he could protect his own country.
Because the advocates of AD believe the Russians to be less strongly
motivated than d o those who call for FR, they believe that much less deter-
rence, both in terms of the damage that the United States needs to inflict
and the probability that it will be inflicted, will be sufficient. Thus there is
a disagreement over "how much credibility is enough": two policy analysts
therefore might agree on how likely the Russians thought it was that a
limited war would escalate and disagree over whether they would be
deterred. ''
Situations Calling for Flexible Response
The basic concern of the proponents of FR is that the threat to attack Soviet
population centers is not credible when the Russians can respond in kind.
In a crisis the United States must "have a wider choice than humiliation or
all-out nuclear action," to use President Kennedy's terms." The danger that
the proponents of FR see was expressed well by Secretary of Defense
Schlesinger in 1975: "If one side should remove the other's capability for
flexible and controlled responses, he might find ways of exercising coercion
and extracting concessions without triggering the final holocaust .... No
opponent should think that he could fire at some of our Minuteman or SAC
[Strategic Air Command] bases without being subjected to, at the very least,
a response in kind. N o opponent
.. should believe that he could attack other
U.S. targets of military or economic value without finding similar or other
appropriate targets in his own homeland under attack .... Above all, no
opponent should entertain the thought that we will permit him to remove
our capability for flexible strategic responses.""
We can examine the problem more clearly by seeing that Schlesinger and
other proponents of FR blur the distinction between two kinds of wars. The
first involves demonstration attacks. Since they d o not require large numbers
of missiles, neither the size of each side's force nor its vulnerability is import-
ant. The second is a counterforce war of attrition in which the Russians
would launch the first nuclear strike, trying to destroy as much of the
American strategic force as possible, either in one blow or by moving more
slowly and taking out the opposing forces in a series of strikes. Although the
United States would still be able to attack the Soviet Union's cities, the only
result of such a strike would be to have U.S. cities blown up thirty minutes
later. If the U.S. strategic force is vulnerable, the Russians can destroy much
of it without using a similar proportion of their force; if the U.S. force
cannot hit protected targets, it will not be able to reduce the Russian force.
But, and this is crucial, it is only in counterforce wars of attrition that the
comparison of each side's counterforce capabilities matters.
Examining a number of contexts in which defense problems arise, one
can see that the distinction between attacks that have an effect by demon-
strating resolve and those that aim at reducing an opponent's capability
recurs and is closely tied to the basic difference between the AD and FR
positions. If the AD position is correct and counterforce wars of attrition
are not a real possibility in the nuclear era, then the United States does not
have to worry that its Minuteman force is vulnerable or that the Russians
have a greater ability to destroy hard targets than the United States does.
To evaluate the arguments, it is useful to examine the potentially critical
situations.
Protecting Europe
One major fear is that the Soviets could launch a large-scale conventional
attack that would conquer Europe unless the United States escalated. If
the United States tried to stave off defeat by employing tactical nuclear
weapons, the Soviets could reply in kind, nullifying any advantage the West
may have gained. One FR remedy would be to develop the means to defend
against an attack at any level of violence. Thus the West would deploy con-
ventional forces to contain a conventional attack and tactical nuclear
weapons to cope with a like attack. This alluring argument is not correct.
An aggressor could attack in the face of escalation dominance if he believed
that the defender would not pay the price of resisting, a price that includes
a probability that the fighting will spread to each side's population centers.
The other side of this coin is that a state that could be confident of winning
a military victory in Europe could be deterred from attacking or deterred
from defending against an attack by the fear that the war might spread to
its homeland. Only if the risk of such escalation could be reduced to zero
would this element disappear and purely military considerations be deter-
minative. The advocates of FR thus overstate the efficacy of their policy.
Of course if the United States lacks escalation dominance it would have
to take the initiative of increasing the level of violence and risk in the event
of a Soviet attack on Europe. But the onus of undertaking the original move
would still remain with the aggressor. And since the level of risk is shared
equally by both sides, what is likely to be more important than the inhib-
ition against having to take the initiative is the willingness or unwillingness
to approach the brink rather than concede defeat, a factor not linked to escal-
ation dominance. Furthermore, some practical considerations reinforce this
conclusion. As Bernard Brodie argued, it is hard to imagine that the Soviets
would launch a conventional attack in the face of NATO's tactical nuclear
, Nuclear Superiority 2 13
weapons. Such an attack would require massed troops that would be an

inviting target for NATO's tactical nuclear weapons. The Soviets could not
be sufficiently confident that their strategic or tactical nuclear forces would
deter such a NATO response to leave their armies so vulnerable.14 And for
the Russians to initiate a tactical nuclear war would raise two difficulties.
First, the uncertainties about how such a war would be fought are so great
that it would be hard for any country to be confident that it would win.
Second, a war of this level of violence would be especially likely to trigger
the American strategic force.
An alternative FR policy is for the United States to develop large enough
strategic forces to threaten, and carry out if need be, a counterforce strike
with some of its forces, even though doing so would not leave the Soviet
Union totally disarmed. While the Soviet Union could retaliate against
American population centers, it would not d o so because its own cities were
still in hostage. Thus the United States could launch its strike "secure in the
knowledge that the United States had a residual ICBM force that could
deter attack upon itself."" This notion of security is an odd one, resting as
it does on the confident prediction that the Russians would calnily absorb
a counterforce first strike. This is especially odd because while the propon-
ents of FR tell us that we should pay close attention to Soviet military doc-
trine, o n this point they blithely disregard these texts which stress preemption
and deny that limited nuclear wars are possible.
A similar error is embodied in Secretary of Defense Schlesinger's defense of
limited nuclear options on the grounds that because the United States has com-
mitments to allies, "we require a nuclear capability that has an implemen-
table threat and which is perceived to have an implementable threat. Unless,
in the event o f certain hostile acts, we have a threat that we can implement,
the existence of the American force structure does not contribute logically
to deterrence." If, on the other hand, the United States has the ability
to launch limited nuclear strikes, he continues, it "will not be self-deterred
from responding to ... an act of aggression."lh But the concept of "self-
deterrence" is not useful and the argument cannot be sustained. The United
States is being deterred by the fear of Soviet retaliation. This danger is present
as long as the Soviets have second-strike capability; thus, it is a consequence
not of Soviet "superiority" but of parity. Even if the United States reached
Schlesinger's goal of preserving "an essential strategic equilibrium with the
USSR both in capabilities and in targeting options,"" the costs and risks of
employing the options would remain.
The argument that the side that had better counterforce capability could
safely launch such an attack even though the other side would not be disarmed
pertains only in wars of attrition in which each side tries to reduce the
other's strategic capability and spares the other's cities. The claim that the
United States can employ this option to protect Europe is the opposite
side of the coin of the claim that if the Soviet Union had a large margin of
counterforce superiority, it could use it to coerce the West. The validity
of this claim turns o n whether a war of attrition is a serious possibility or
whether the danger that such a conflict would escalate to attacks on popu-
lation centers would dominate decision makers' calculations.
Preemption
Some proponents of FR think it most unlikely that the Russians would

launch an attack on Europe, but fear that if the Soviet strategic force was
much more effective than the American one and if a significant proportion
of the American strategic force were vulnerable, the Russians might make a
preemptive strike in a grave crisis, perhaps one they had not sought, if they
thought that war was very likely. The ability to hit missile silos and com-
mand and control facilities that the proponents of FR call for, however,
increases this danger, since it enables the United States to destroy a large
proportion of USSR's land-based missiles (and most of the Soviet strategic
force is land based) if the United States were to strike first. It is a bit disin-
genuous t o argue that the United States needs a new type of missile
to decrease the chance that the Soviet Union would attack without also
acknowledging that some of the incentive the Russians would have to
attack those missiles comes from the very accuracy that is supposedly
needed in order to fight a counterforce war.18
More importantly, preemption makes sense only if being struck first is
much worse than getting the first blow in. A state whose leaders believe that
war will lead to total devastation will have no incentive to preempt even if
many of their missiles are vulnerable. Here, as at other points, the propon-
ents of FR make the crucial mistake of concentrating on purely military fac-
tors - the numbers and characteristics of weapons on both sides - and
ignoring the role of military moves as generators of risk. The FR argument
is that deterrence requires a sufficient number and kind of forces so that if
the other side struck first, it would be militarily worse off than if it had not.
Thus it would be dangerous if the Russians were able to use, for example,
200 missiles with 2,000 warheads and knock out most of the American
ICBMs. As Secretary of Defense Brown has put it: "we must ensure that no
adversary could see himself better off after a limited exchange than before
it. We cannot permit an enemy to believe that he could create any kind
of military or psychological asymmetry that he could then exploit to his
advantage."19 But the fact that the Russians would have gained a more
favorable ratio of missiles does not mean that they would be closer to any
meaningful goal or even that they would be closer t o it than they would
have been if the United States struck first and the ratio of missiles available
was less favorable to them. The only meaningful goals would be to preserve
their cities and, if possible, prevail in the dispute. But gains in purely mili-
tary terms do not accomplish these objectives in wartime any more than
they d o in peacetime. As long as each side retains the ability to destroy the
other's society, having more warheads than an opponent is an advantage
only if it makes the opponent back down, and the proponents of FR have
not shown how it will make such a contribution. The military advantages
I( rc 15 Nuclear Superiority 2 15
of striking first can only be translated into political gains if the war remains
counterforce and the state with the most missiles left after a series of
exchanges prevails without losing its population centers."' The FR fallacy
here is parallel to that involved in the claim that escalation dominance is
necessary or sufficient for deterring or prevailing in a conflict in Europe.
Competition in risk taking, rather than con~petitionin military capability,
dominates.
ars of Attrition and t h e
in a counterforce war of attrition the numbers and characteristics of the

weapons would matter a great deal. As in the prenudear era, what would
be crucial would not be absolute capability, but the relative strengths of the
opponents. The basic argument of t h e AD school is undercut bkcause the
primary targets of the warheads are not population centers but other
weapons. Is the likelihood of counterforce wars of attrition sufficient to
warrant the necessary preparations? Could there be a nuclear war in which
population centers were spared and the outcome determined by which state
is able to do the better job of reducing the other's military forces? Even if the
Russians had the ability to win such a war, they would have to be desperate
or willing to run terribly high risks to place sufficient faith in American self-
restraint to order an attack. Even if the United States could win such a war,
its threat to initiate it would not be credible (for example, in response to a
Soviet attack on Europe) unless the Russians believed that the United States
thought that control would be maintained throughout its course.
This control would have to be maintained, furthermore, although unprec-
edented numbers of civilians would be killed; a large Russian counterforce
strike could not be limited to destroying only military targets. Although it
would be obvious to the president that most American population centers
were still held hostage, sufficient damage would have been done to raise
sharply the danger of an all-out response. The chance of such a re?- ' LtlOll -
'
which would be present even if the United States said it would not react
in this way - would have to weigh very heavily on the Soviet decision makers.
But the existence of tight control would not ensure the s ~ ~ c c eof
s s a strat-
egy of attrition. If the Russians launched a counterforce strike and the
United States did not retaliate against Soviet cities, it might nullify a Russian
war-fighting strategy by not responding at all. This may seem as bizarre as
a counterattack on population centers, but on closer examination it makes
some sense. Why should the United States retaliate? What would the
Russians have gained by destroying a significant portion of the U.S. strategic
force? Why would they be in a better position to work their will after a
strike than before it? If the United States acts as though it is weakened, it
will be in a worse bargaining position, but this is within American control.
To withhold a response, while maintaining the ability to destroy Russian
cities later, could as easily be taken as a sign of high resolve as of low. The
United States would forego hitting many Russian military targets, but this
would not sacrifice much of value since attacking them would not limit the
Soviet ability to destroy the United States. Only if a war in Europe were
being fought at the same time, and thus a failure to respond created or mag-
nified an imbalance of land forces, would withholding a return counterforce
strike give up something of value. But for the Soviets to attack American
strategic forces (and NATO tactical nuclear forces) in conjunction with
fighting a war in Europe would be to run a very high risk of an American
counterattack on Soviet population centers.
The possibility of not responding to a Soviet counterforce strike points to
the odd nature of a nuclear war of attrition. The benefit of the efforts to
reduce an opponent's strategic forces comes only near the end, when the
state is able to take its society out of hostage. Unless and until that point is
reached, the side that is "losing" the counterforce war of attrition can d o
nearly as much damage to the side that is "winning" as it could before the
war started. Military efforts can succeed only if the "loser" allows them to
by sparing the "winner's" cities. Of course it will be costly for the "loser" to
initiate counterstrikes against population centers, since the "winner" will
presumably retaliate. But this is true regardless of the details of the strategic
balance.
If the ultimate threat, even during a war of attrition, is that of destroy-
ing cities, it is clear that such wars are more competition in risk taking than
they are attempts to gain an advantage on the battlefield. To concentrate on
the military advantages that accrue to one side or the other by counterforce
attacks is to ignore the fact that in any nuclear war the element of threat of
escalation will loom very large.21This general point is missed by Secretary
of Defense Brown when he says that the ability to hit a wide range of mili-
tary targets "permits us t o respond credibly to threats or actions by a
nuclear opponent."22 But what is crucial is less the capability than the will-
ingness to use it. Even if the United States had the ability to match the Soviets
round for round, target for target, it might not d o so - and the Russians
might move in the belief that the United States would not respond - because
the costs and risks were felt to be too great. And even if the United States
lacked such a capability, the Soviet fear of an all-out response could lead it
to expect that any provocation would be prohibitively costly. Since what
matters in limited strategic wars, even if they involve targets that are
predominantly military, is each side's willingness to run high risks, it is the
"balance of resolve" rather than the "balance of military power" that will
most strongly influence their o u t ~ o m e s . ~ ~ xammunition
tra cannot com-
pensate for weakness in will or a refusal - perhaps a sensible refusal - to
run the risk of destruction.
The importance of competition in risk taking implies that demonstration
attacks would be more useful than attempts t o reduce an opponent's mili-
tary capabilities. Such attacks could be aimed at a military installation, an
isolated element of an opponent's strategic forces, a command and control
facility, or a city. The purpose of such an action would be to inflict pain,
show resolve, and raise the risks of all-out war to a level that an opponent
i~ I \ Nuclear Superiority 2 17
would find intolerable. Such risks, of course, weigh on both sides, but only by
willingly accepting high risks can a state prevail. In addition to high resolve,
in order to engage in nuclear demonstrations a state needs to be able to carry
out a certain number of limited options. But the ammunition requirements are
nowhere near as high as they are for a counterforce war of attrition (and both
sides can simultaneously have the capability for demonstrations).
Demonstration strikes would exert pressure in three ways. First, they
would exact some degree of punishment on the other side. But the immedi-
ate pain inflicted would probably be less important than the underlying
motivation of these strikes - the implied threat to d o more harm unless the
opponent complies with the attacking state's demands. This threat gains its
credibility because the attacking state has shown that it is willing to engage
in very risky actions that have increased the chance that targets in its own
country would be struck. When both sides have second-strike capability,
one side prevails in a crisis, not by showing that it can inflict pain on the
other (for this is obvious and true for the both sides), but by demonstrating
that it feels so strongly about the issue at stake that it is willing to be hurt
in return rather than suffer a defeat. Third, any nuclear attack increases
the chance that uncontrolled escalation will occur. It is this specter that
exerts so much pressure on statesmen not to use nuclear weapons in the first
place or to make concessions in any conflict in which they are used. Even if
one side launched a counterforce strike, the war would almost surely end
before either had run out of ammunition. Resolve, not capability, would be
the limiting factor. When Secretary Brown claims that "fully effective deter-
rence requires forces of sufficient size and flexibility to attack selectively
a range of military and other targets"14 and argues that to do this the
United States needs an invulnerable ICBM, he is either thinking in terms of
a war of attrition or overstating the number of warheads the United States
needs.
Possible Objections
Before drawing the conclusions that are implicit in the previous analysis, I
should note three obvious objections. First, it can be argued that if I am
right, and the strategic balance is quite stable, an increase in American arms
will not have dire consequen~es.~' Since all the United States can lose hy
additional deployment is money, argue the critics, it is better to play it safe
and buy the extra systems. Moreover, how can anyone be sure that a war
of attrition will not occur? But surely there must be some judgments about
plausibility, some concern for costs, and some consideration of the chance
that the United States might teach others lessons that are both incorrect and
dangerous. The new weapons cost a lot of money and avoiding waste is not
a goal to be scorned lightly.lh Furthermore, although there are no strong
and direct links between the adversaries' defense budgets or between the
budgets and the degree of superpower conflict, it is hard to keep the military
and political tracks entirely separate. A final line of rebuttal is the most
important: t o develop a posture based on the assumption that limited
nuclear wars are possible is to increase the chance that they will occur. If
the Russians already believe in the possibility that such wars could be kept
limited, U.S. acceptance of this position would increase the likelihood of
their occurrence. O n the other hand, if the Russians now find these kinds
of war incomprehensible, they might learn t o accept them if the United
States talked about them long and persuasively enough. This could decrease
the chance that a nuclear war would immediately involve the mass destruc-
tion of population centers, but at the cost of increasing the chance of more
limited nuclear wars - which then could escalate. Such a trade-off is highly
likely, and even Schlesinger acknowledged that adoption of his doctrine
might increase the chance of limited nuclear strike^.^'
The second objection is that my analysis ignores the fact that the Russians
d o not accept the notion that mutual assured destruction creates stability.
Soviet military doctrine is an arcane field that cannot be treated in detail
here, although the bulk of the evidence indicates that the Soviet view of
strategy is very different from the A m e r i ~ a n .They
~ ~ appear to take war
more seriously. Indeed, much of Soviet military doctrine is pure military
doctrine - that is, the ideas are not particularly Russian or particularly
Marxist but simply those one would expect from people charged with pro-
tecting society and winning wars. Many statements by Soviet generals are
similar t o statements by American generals when the latter are not influ-
enced by the ideas or constrained by the power of the civilian leadership;
many American military officials seek the same program that the Russians
are following. Thus one cannot draw from the fact that Russians probably
buy more than is needed for deterrence the inference that they are willing to
run high risks to try to expand. The American generals who call for higher
spending are not necessarily more bellicose than those who disagree with
them.29 Both the Russian and the U.S. generals may want to prepare for the
worst and get ready to fight if a war is forced on them. The Russians may be
buying what they think is insurance, and we do not ordinarily think that
someone who buys a lot of insurance for his car is planning to drive recklessly.
While there is considerable evidence that the Russians want military
forces that would provide as good an outcome as possible should war be
forced on them, there is very little evidence that they think that such forces
could be used to coerce the West. It has yet to be shown that they think that
a superior ability to destroy military targets provides a shield behind which
they can make political advances or that Soviet military doctrine measures
American deterrence in terms of the United States' ability to match their
posture. The Russians may not accept the idea that mutual vulnerability is a
desirable state of affairs, but they seem to understand very well the potency
of the American threat to destroy their society. Indeed their outlook is
uncongenial to a counterforce war of attrition. While the Russians probably
would attack U.S. strategic forces in the event of war, they have not talked
about sparing the opponent's cities. Instead, they seem to be planning to hit
as many targets as they can if war breaks out.
l e ~ v i s Nuclear Superiority 2 19
Even if the Russians were to say that they believed a war of attrition was
possible, the United States would not have to adopt such a view. While it
takes the agreement of both sides to fight a counterforce war, this is not true
for AD. If one side denies that counterforce wars could be kept limited and
convinces the other side that it believes this, the other cannot safely act on
its doctrine. The Russians understood this in the periods when McNamara
and Schlesinger were enunciating their doctrines, and American statesmen
took their professions of disbelief seriously. Even if the Russians were to
reverse their position, they would have to take American denials seriously
also.
A third objection is that although the Soviet superior ability to destroy
strategic forces and the related existence of Minuteman vulnerability is not
a strategic problem, it is a political problem. Accordingly, because other
nations are influenced by indicators of nuclear superiority, the United States
must engage in this competition. (This argument loses some credibility since
most people who make it also claim that superiority is meaningful apart
from these perceptions.) There are several lines of rebuttal. First, there is lit-
tle evidence that European or Third World leaders pay much attention to
the details of the strategic balance. Second, the United States provides most
of the information and conceptual framework that underpins third-party dis-
cussions of the balance. The United States might be able to persuade others
that it would behave differently because the Russians could wipe out much
of the American capability to destroy Soviet missiles. But it would proba-
bly be easier to convince them that this was not true. Few world leaders
expect the United States and the Soviet Union to fight a war of attrition.
Moreover, if the Russians believe that superiority matters and thus may be
somewhat emboldened, the bargaining advantages they will gain will be
slight if the United States holds to the position that this is nonsense. If the
United States convinces the Soviet Union that it does not see a meaningful
difference in strength, the USSR cannot safely stand firm in crisis bargain-
ing because it will not have any reason to think that the United States is
more likely to retreat.
Conclusions
We can draw several conclusions. The question of which side has greater
ability to destroy the other's strategic forces matters only in a war of attri-
tion. Such a war seems unlikely enough so that it is not worth spending large
sums and running considerable dangers to prepare for it. Because either side
can use its nuclear weapons to destroy its opponent's population centers, the
danger of escalation would play a very large role in any war and could not
be controlled by having more missiles, more accurate missiles, and more
invulnerable missiles than the other side. The nuclear revolution cannot he
undone. As we have seen, many of the arguments about the supposed dan-
gers following from Soviet superiority in fact are consequences of parity. The
American deterrent is deterred by the fact that its cities are vulnerable, not
by the fact that the Russians have some supposed military advantage. Since
neither the United States nor the Soviet Union can take its cities out of
hostage, the state that is willing to run the greatest risks will prevail. Many
of those who call for the United States to match or surpass the Soviet's
nuclear arsenal are trying t o have the United States compensate for what
they feel is a weakness of resolve by an excess in weaponry. But such a defi-
ciency, if it exists, cannot be compensated. A wider range of options will
merely give the Russians more ways, and safer ways, of coercing the West.
If the balance of resolve is so important, is the United States at a disad-
vantage compared to the Soviet Union? Some would argue that the United
States has shown in Vietnam that it will not fight to defend its interests and
those of its allies. But few dominoes fell after April 1975; other states may
have been less impressed by the final American withdrawal than they were
by its willingness to spend so much blood and treasure on an unimportant
country. Furthermore, resolve is not so much an overall characteristic of an
actor as it is a factor that varies with the situation because it reflects the
strength of the state's motivation to prevail on a given issue. The state
defending the status quo has the advantage in most conflicts in which the
balance of resolve is crucial because it usually values the issue or territory
at stake more than its opponent does." It is easier for a state to convince
the other side that it will fight to hold what it has than it is to make a cred-
ible threat to fight rather than forego expansion. A world in which resolve
matters so much may not be so bad for the United States.
Even if both sides recognize the greater determination of the side defend-
ing the status quo, accidents and miscalculations are still possible, especially
in situations growing out of a crisis in a third area. To rely solely on AD
may be too dangerous. Some degree of insurance can be purchased by a
continuation of the present American posture, which includes the availabil-
ity of limited nuclear options. But these should be demonstrations, keyed to
competition in risk taking, not attempts to wage a war of attrition; thus, the
United States would not have t o match the Soviets on any of the standard
measures of nuclear power. It does not take a superior or even an equal mili-
tary force to show by limited use that one is willing to take extreme
measures rather than suffer a defeat. Such costs and risks are the trading
chips of bargaining in the nuclear era; even if the United States had the
weapons and doctrine for an FR policy, it could not avoid relying on them.
Although the United States should be able to conduct limited nuclear
demonstrations, it should not stress this part of its policy. At this point there
is n o reason t o think that such fantastic measures will ever be necessary, and
they should be looked on as something to be done only in the most dire
emergency, not as a tool of statecraft. Too much discussion of the possibil-
ity of such strikes might lead either or both sides to believe that the risks of
a limited exchange were manageable, that escalation would remain under
tight control. At best, the United States would therefore create a world in
which limited nuclear wars were more likely to occur. At worst, these beliefs
would be tested and proven to be incorrect.
]imI - Nuclear Superiority 22 1
Of course a policy of AD supplemented by the ability to conduct demon-

stration attacks may not succeed, The specter of all-out war is probably
compelling enough to make both sides so cautious as to render forcible
changes of the status quo on important issues too dangerous to be attempted.
But miscalculations are possible, even in situations that seem very clear in
retrospect, and states are sometimes willing to take what others think are
exorbitant risks to try to reach highly valued ,goals. Both a cautionary tale
and reminder that superior military capability does not guarantee deter-
rence is provided by the Japanese decision to go to war in 1941. Japan
struck because her leaders saw the alternative not as the foregoing of gains,
but as losing "her very existence."" They were thus very highly motivated
- much more so than American decision makers thought. Furthermore, they
knew perfectly well that they could not win an all-out war. But they were
not expecting to have to fight such a war; they thought that the war would
be limited as the United States would prefer to concede dominance in East
Asia rather than engage in a long and costly struggle. It is always possible
that the Russians might similarly believe that a nuclear war could be kept
limited because the United States would rather concede than move closer
to the abyss. The penalty for miscalculation would be much greater for
Russia than it was for Japan, and so their caution should be much greater.
The danger remains, however, and it cannot be met by building more
weapons.
Acknowledgement
I would like t o thank 1)esniond Ball, Richard Berts, Thomas Brown, James Dighy, James King,
George Quester, Michael Mandelbaum, Stanley Sierkiewicz, Dennls Ross, and Glenn Snyder
for comments, and the Solomon Guggenheim Fund for financial support.
Notes
1. The phrase is attributed to Thomas Schell~ng in Herman Kahn, O n t x a l a t ~ ) n

(Baltimore, Md.: Penguin, 1968), p. 3. The topic is discussed in Schelling, Arms and Influence
( N e w Haven, Conn.: Yale University Press, 1966), pp. 92-125.
2. Thus there is a second meaning in the title of the book of brilliant essays written In 1946
by Bernard Brodie, Arnold Wolfers et al., The Absolzrte Weapon ( N e w York: Harcourt, Brace,
1946).
3. Many proponents ot AD also argue that the Soviet U n ~ o nwould feel threatened by
increases in American strategic forces, however, and this fear is In some tension with the belief
described here.
4. bernard Brodie, War and Polrtics ( N e w York: Macmillan, 19733, p. 380.
5 . Glenn Snyder, "The Balance of Power and the Balance of Terror," in The Kalnnre of
I'osuer, ed. Paul Seahury (San Francisco, Calif.: Chandler, 196.5). This paradox w ~ seen s by
Snyder as explaining why mutual second-strike capabil~tycould lead t o conventional wars; the
proponents of FR take this one step further by arguing that the overall strategic stability also
allows for l~mitednuclear wars.
6. Paul Nitze, "Deterrmg O u r Deterrent," Forergn Policy 25 (Winter 1976-77): 19.5-2 10.
7. U.S., Congress, Senate, Committee on Armed Services, Hearings on Fiscal Year 1975
Authorization, 93rd Cong., 2d sess., 1974, p. 51.
8. Department of Defense, Annual Report, E Y 1980 (Washington, D.C.: Government
Printing Office, 1979), p. 76. Brown's posture statement is a combination of FR and AD and
so is more honest, but less coherent, than many of the previous statements.
9. Department of Defense, Annual Report, El! 1979 (Washington, D.C.: Government
Printing Office, 1978), p. 54.
10. Department of Defense, Annual Report, El! 1980, p. 75.
11. This also partially explains why many of the proponents of FR think that the threat to
destroy Russian cities would be an insufficient deterrent and that the United States should
develop a targeting policy aimed at convincing the Soviet leaders that their regime would not
be able to maintain control of the country after a war.
12. "Radio and Television Report to the American People o n the Berlin Crisis, July 25,
1961," in Public Papers of the Presidents of the United States: ]ohn E Kennedy, 1961
(Washington, D.C.: Government Printing Office, 1962), p. 535.
13. Department of Defense, Annual Report, El! 1976 a n d El! 1 9 7 T (Washington, D.C.:
Government Printing Office, 1975), pp. 11-4-11-5.
14. Bernard Brodie, Escalation a n d the Nuclear Option (Princeton, N.J.: Princeton
University Press, 1966).
15. Colin Gray, "The Scope and Limits of SALT," Foreign Affarrs 56 (July 1978): 788.
16. U.S., Congress, House, Committee on Armed Services, Hearings on Military Posture
and H.R. 12564, Department of Defense Authorization for Appropriations for Fiscal Year
1975, 93rd Cong., 2d sess., 1974, part 1, pp. 47,49.
17. Ibid., p. 29.
18. The United States might get around this dilemma by building missiles that were invul-
nerable, but that lacked accurate MIRVs. It is interesting to note that the U.S. Air Force has
done a much better job of developing a and accurate missile than it has in making
that missile able to survive a Russian attack.
19. Department of Defense, Annual Report, E l! 1979, p. 56.
20. This point is overlooked by Paul Nitze in "Assuring Strategic Stability in an Era of
Detente," Foreign Affairs 5 4 (January 1976): 226-30, and in "Deterring our Deterrent," p. 210.
21. Indeed the incentives for the state that is behind in a counterforce war to escalate
increase as its military situation worsens. If it fears it may soon lose its second-strike capabil-
ity, the losing state may will feel greater pressure t o up the ante while it still can.
22. Department of Defense, Annual Report, E Y 1980, p. 78.
23. For further discussion of this point see Robert Jerv~s,"Deterrence Theory Revisited,"
World Politics 3 1 (January 1979): 314-22, The Soviet stress on the importance of the "correl-
ation of forces" is not inconsistent with this notion.
24. Speech before the Council on Foreign Relations and the Foreign Pol~cyAssociation, New
York, 5 April 1979, p. 3. Also see Department of Defense, Annual Report, E Y 1980, pp. 77-78.
25. There is a similar contradiction in McCeorge Bundy's claim that although nuclear
superiority is meaningless, we need arms control agreements "To Cap the Volcano" (Foreign
Affairs 4 8 [October 19691: 1-20).
26. Bernard Brodie argues that the strategic balance is so stable that saving money should
be the main goal of arms control ("On the Objectives of Nuclear Arms Control," International
Security 1 [Summer 19761: 17-36). His position is further developed in "The Development of
Nuclear Strategy," ibid. 2 (Spring 1978): 65-83. I am greatly indebted to these articles.
27. House Armed Serv~cesCommittee, Hearings on Military Posture, p. 50.
28. For a dissenting view, see Raymond Garthoff, "Mutual Deterrence and Strategic Arms
Limitation in Soviet Policy," International Security 3 (Summer 1978): 112-47.
29. A study of postwar situations reveals that the U.S. milltary often advised against foreign
military adventures. See Richard Betts, Soldiers, Statesmen, and Cold War Crises (Cambridge:
Harvard University Press, 1977).
30. Jervis, "Deterrence Theory Revisited," p. 318.
31. Robert Butow, Tojo a n d the Coming of the War (Princeton, N.J.: Princeton University
Press, 1961), p. 203.
Strategic Stability Reconsidered
Colin S. Gray
I
n an important article published in 1978, John Steinbruner claimed: "As
the United States force posture has evolved over the past 1 5 years, the
idea of stability has emerged as the central strategic objective, and the
asserted conceptual consensus seems to be organized around that object-
ive."' This essay will focus on whether the theories of stability most widely
held in the West may not be gravely deficient and whether the integrity of
the concept of strategic stability may not itself be questionable.
Discussion of stability and its possible requirements is in fact a discus-
sion of deterrence theory, which in reality is a debate about the operational
merits of different postures and doctrines. No useful, objective, doctrine-
neutral exploration of the idea of stability is possible. The discussion that
follows makes no pretense of neutrality; instead, it endeavors, first, to
explain the roots, meaning, and deficiencies of the still dominant theories of
stability and, second, to suggest a theory that has much greater internal and
external integrity.
It is important to recognize that, for all its popularity, there is no useful con-
sensus about the meaning of stability. Most commentators, and certainly the
government of the United States (and NATO), acknowledge the value in the
twin concepts of arms race stability and crisis stability. As arms race stability
is commonly understood, it is the condition wherein neither party to an arms
competition will press military developments or deployments in quest of
major advantage, because such advantage is judged to be unattainable, how-
ever desirable. Crisis stability is a quality of strategic relations: during periods
of acute crisis, instruments of war (mechanical, electronic, organizational)
should not be the immediate cause of war. At this level of generality, these
concepts have been widely understood and approved (in the West) since at
least 19h0.2 Consensus breaks down, however, over the particular policy
Source: Dadalz~s:Iouynal of the American Academy of Arts and Sciences, 109(4) (1980):
135-54.
implications. From an operational perspective, how is arms race stability to

be achieved and maintained, and how is crisis stability enforced, vis-a-vis a
Soviet adversary?
This quest seeks a theory of stability that will work "well enough,"
given the full dimensions of Western strategic security problems in the con-
text of the military consequences of the unique "cultural thoughtways"bf
a particular major adversary. As a working hypothesis, I believe that the
ideas of arms race and crisis stability, and the theory of deterrence to which
they usually make (often implicit) reference, have (mis)led Western policy-
makers into neglecting the operational dimensions of strategy. Indeed,
many politicians, officials, and analysts seem to believe that nuclear strat-
egy cannot really have any operational dimensions. As its first priority, an
adequate theory of deterrence must determine the military (and relevant
civilian) requirements of war itself. Deterrence theory, the leitmotiv of
Western strategic preparation, is fully consistent with a strategic force pos-
ture that has n o credibility as a threat, because it would not be intelligently
usable in practice. It is essential to recognize that both Western ideas of
stability and the Soviet approach to the determination of principles that
will guide their defense preparation and war planning have deep cultural
roots - they are not accidents of history.
For much of this essay, "stable deterrence theory" will refer to the propos-
ition that stability in arms competition and in time of crisis is maximized
when each side is unambiguously vulnerable at home, and also confident that
a large number of its strategic offensive weapons are invulnerable prior to
launch and during mission e x e c ~ t i o nThis
. ~ condition of mutual assured vul-
nerability has been identified for many years as a mutual assured destruction
(MAD) posture. Orthodox stability theory in the West even today is based in
good part on the assumption that mutual societal vulnerability is desirable. It
must be observed, however, that a MAD posture, in principle, is compatible
with a wide variety of strategic targeting plans.
By way of providing an initial point of doctrinal reference for this essay, it
is my belief that the strategic balance would be stable if Western governments
were allowed to enjoy not-implausible prospects of both defeating their enemy
(on his own terms) and ensuring Western political-social survival and recov-
ery. This admittedly close parallel to the known Soviet approach to defense
planning is already U.S. policy with respect to the requirement for the defeat
of the enemy? and is not incompatible with the more familiar connotation of
(arms race and crisis) stability. It rests on the proposition that forces that do
not lend themselves to politically intelligent employment in war are probably
insufficient to deter - at least at those rare times when an adversary may seek
a military solution to his problems.6 The costs of major war today are antici-
pated to be so high, and so many of the weapon systems on both sides lack
realistic field tests,' that the call for a war-survival capability would hardly be
likely to encourage Western governments down the path of military adventure.
In this essay I will argue that the West requires a concept of stability
appropriate for giving a theoretical underpinning to the determination of
8 I ,i Strategic Stability Reconsidered 225
the military requirements that will enable it to defend its vital interests. The
stability theory dominant in the 1960s and 1970s was addressed at root to
a relationship between two supposedly like-minded and ultimately (after
detente had done its work) like-intending adversary-partners. Although a
sea-change is evident in this new decade in official U.S. (and NATO) appre-
ciation of Soviet habits and motives, the burden of obsolescent strategic the-
ories o f stability remains.
Concepts and Culture
There is much to recommend the working hypothesis that Soviet and

American strategic concepts reflect the character and content of their diver-
gent and distinctive "strategic cultures." Jack Snyder has defined strategic
culture "as the sum total of ideas, conditioned emotional responses, and
patterns of habitual behavior that members of a national strategic commu-
nity have acquired through instruction or imitation and share with each
other with regard to nuclear strategy."# The insertion of "nuclear" precludes
too much. John Erickson wrote recently on Soviet "style," stressing the per-
vasiveness of all arms thinking in Soviet military preparation.' Recognition,
if not uncritical acceptance, of this working hypothesis is important for the
following reasons. First, it might help to explain how and why the concept
of strategic stability took such firm root in the U.S. defense and arms control
community. Second, it should facilitate more accurate comprehension of
Soviet deeds and words. Third, it should help U.S. policymakers to identify
programs and doctrines that, while broadly compatible with American values,
are adequately responsive to Soviet developments.
T h e United Stares
The concept of strategic stability took firm hold as an American strategic

desideratum long before it was substantially field tested in the Soviet-
American military competition or before there was a directly relevant for-
mal arms control process. American theoreticians reasoned that the
multitier arms competition between East and West could be stabilized
through cooperative management effected through tacit agreement or for-
mal bargaining. Moreover, the literature of the early and mid-1960s shows
clearly that the U.S. defense community knew what strategic stability was
and how it could best be forwarded. A gifted Israeli commentator o n the
U.S. arms debate wrote in 1964 that
stability has become a fundamental concept in nuclear strategy, and a

magic formula. Strategic situations are measured by the degree of their
stability. . . . Once a situation of stability has been achieved, the initiation
of war by surprise no longer assures any gain or advantage. A situation
is stable, therefore, when there is no temptation to force the issue; it is a
situation of mutual neutralization in which both the householder and

the burglar know that even if one slays the other, the latter will manage
to retaliate posthumously.10
The U.S. defense community, with very few exceptions, decided that a
stable military balance would mean a safer world with fewer resources
expended on defense than would be the case with an unstable military bal-
ance; should be compatible with the support of U.S. foreign policy interests
(though it is unclear whether very careful analysis was performed on this
subject); and should eventually find favor with the Soviet Union, because of
its technological inevitability and near-evident desirability. In a book that
perhaps merits description as the fullest and most mature statement of
1960s-style stability (through mutual vulnerability) theory, Jerome Kahan
wrote: "A mutual stability approach, in the broadest sense, rests on the
premise that the United States is benefited if the Soviet Union maintains a
strategic deterrent capability comparable in overall strength to our own; it
is an acceptance of both the mutual assured destruction relationship and
numerical parity." A little earlier he had written, "If, then, the USSR's stra-
tegic doctrine is largely understandable and somewhat comparable to ours, it
is possible to establish a relatively effective U.S. policy of mutual stability.""
The United States seemed to know what it wanted and to believe that
what was good for the United States would come to be seen by the Soviet
Union as good for it also. A stable military balance, in American perspec-
tive, was one in which each side's military forces looked roughly compar-
able and neither side believed it could gain a significant military advantage
by striking first, because neither side would be able to protect its domestic
assets against retaliation. These elements of stability derived initially, in
good part, from discouraging analyses of the future promise of damage-
limiting strategies. Military-technological prediction - that future societal
vulnerability will be a fact, not a matter for choice - was transformed into
normative terms. Far from being a problem, mutual vulnerability was seen
as an opportunity to establish more stable Soviet-American strategic rela-
tions. The Soviet Union might prefer to compete for "useful advantage" so
long as that was believed to be attainable, but technology surely has a logic
that the Soviets must, and will, respect. In 1970 Roman Kolkowicz
expressed the then popular, and perhaps even plausible, view that
Soviet strategic doctrine and capabilities appear to have lagged behind

those of the United States by about five years ... , modern defense tech-
nology determines to a large extent the kind of strategic doctrines and
policies that will be adopted by the superpowers. Thus, technology
seems to have a levelling effect which subsumes political, ideological and
social differences in various political systems.12
The convergence of strategic ideas hoped for in the late 1960s and early
1970s - that is, a Soviet convergence with the American concept of a stable
Cra) Strategic Stability Reconsidered 227
military balance - did not occur. By 1979-80 most American commentators

o n Soviet-American strategic policy issues accepted as a very probable fact
the existence of a "conceptual gap" between Soviet and American thinking
on strategic issues that appeared t o be enduring, because each side's thought
was rooted in what has come to be termed "strategic culture.""
The important difference between 1980 and 1970 (or 1960) is that what
was then plainly recognized as a possibility - that the Soviet Union would
not wish to engage in genuinely "reciprocal measures for arms stabiliza-
tion" - has now become plausible fact. Indeed, a major question that
should be posed is whether, or perhaps how, the United States can conduct
serious arms control business with the Soviet Union, which shows no evi-
dence of endorsing a recognizable or attractive concept of strategic stabil-
ity. Through the 1960s and at least part of the 1970s such a troublesome
question could be, though should not have been, ignored or deferred. In the
late 1950s and early 1960s, when American arms control theory was being
forged, uncooperative Soviet ideas and practices could plausibly be inter-
preted as reflections of a relatively backward technology or of a policy1
intellectual "lag." By the early 1970s the SALT process appeared to carry
promise for the cooperative management of strategic relations. It was recog-
nized that the Soviet Union still had to catch up in some important military
respects and that program momentum reflecting pre-SALT I thinking and
practices would take some time to be amended to be compatible with the new
relationship.
Today, the U.S. defense community has to grapple with the implications
of the hypothesis that Soviet military ideas and activities are deeply rooted
in local soil and hence are very likely to endure; that the Soviet General
Staff is extremely well acquainted with Western ideas on stability - Soviet
military thinking is not crude and "uneducated"; and that there are no
important apparent strains between the policy preferences of the Soviet mili-
tary and the Soviet political leadership.I4
It would be difficult to exaggerate the importance of this very wide-
spread, if somewhat belated, Western recognition of the strategic cultural
distinctiveness of the Soviet Union. The distinctiveness blurs markedly, of
course, as Robert Jervis has observed, if Soviet military thinking is com-
pared with American professional military thinking." The American mili-
tary establishment prepares to fight and, if possible, to win wars, and
probably would prefer to support a military doctrine as traditional in its
concerns as that espoused by the Soviet Union.lh Yet it is only in the Soviet
Union that those preferences are fully expressed in postural terms.
Soviet thought on the military dimensions of statecraft, what loosely can
be called "strategic theory," is distinguished by its rarity." Soviet writings
tend to focus on efficient force preparation and implementation - generically
operational matters - or on grand-strategic, highly politicized topics. There
are no functional Soviet equivalents to the Western theories of deterrence,
limited war, and arms control, just as the key Western concepts spawned by
those theories - stability, escalation control, bargaining, sufficiencyladequacy,
and the rest - appear to play no identifiable role in guiding Soviet military
planning.ls In the half-light of the growing appreciation of the alien charac-
ter of Soviet strategic culture, U.S. policymakers have to reassess the relevance,
and prudence, of the strategic ideas that have held intellectual and declara-
tory (policy, if not always war-planning) sway for the past fifteen years.
Despite the evidence accumulating on RussianISoviet strategic culture
and the military-program momentum implications of that culture, Western
commentators continue to deny, implicitly, that stability is a condition
describing a military-political relationship. "Stability," as it pervades much
of American theorizing about deterrence questions, is essentially static and
absolute. It tends to lack a sense of competition. Using this logic, the United
States has a deterrence problem of finite physical dimensions. The complex
military balance is stable if the Soviet urban-industrial target set is ad-
equately covered and if the United States looks, and preferably is, resolute
in its willingness to retaliate.
What kind of damage would most likely be judged acceptable by a Soviet
leadership? This question has been posed, and even answered of lateI9 - with
conclusions that cast grave doubts on the merits of the society destruction
bedrock of the theory that identifies stability with mutual vulnerability - but
the covering of the urban-industrial target set still is accorded pride of place
in official U.S. stable deterrence prose.20
It may be that this society punishment oriented theory can provide a
robust basis for a stable military balance, even in an adversary relationship
with an alien Soviet strategic culture. It is possible that the Soviet military
(and political) establishment is seeking the unattainable in its evident pur-
suit of a war-waginglwar-winning capability, and that the United States
would be ill-advised to compete very vigorously with military programs
designed to improve war-waging performance. However, now that it is gen-
erally recognized that the Soviet military effort marches to the beat of a dis-
tinctly non-American d r ~ m m e r , ~and ' as the Soviet military competitive
position continues to improve across the board, there should be no serious
resistance to consideration of the possibility that the consequences of main-
stream Western stability theory may lead to under-recognized dangers.
The ideas that comprise the concept of a stable military balance reflect
fairly faithfully the world view, values, and pertinent education of those
commentators, policymakers, and theorists who have articulated American
strategic culture.22 The United States is a satisfied world power with a fun-
damentally defensive strategic mission as its international responsibility.
From 1946, when The Absolute Weapon was p ~ b l i s h e d , ~ " the ~ present
day, American strategic theorists have tended to argue, explicitly or impli-
citly, that the development of nuclear weapons has imposed a "techno-
logical peace." The mainstream concept of stability speaks eloquently to the
long-recognized U.S. tendency to define conditions as problems to be solved.
The existence of very large and diverse strategic nuclear arsenals solves
the problem of possible premeditated war between nuclear-armed states,
because the war initiator will know that he cannot deny the enemy the
:,r, I\ Strategic Stability Reconsidered 229
capability of destroying his society in retaliation. Moreover, this ability to

destroy a society in a second strike can deter not only attacks on the U.S.
homeland, but also - with only a modest loss of credibility - attacks on at
least some of the overseas vital interests of the United States. The balance
of terror is thus massively indelicate. As Soviet strategic capabilities
improved relative to those of the United States over the decade 1965-7.5,
the United States sought to retain o r restore the credibility of strategic deter-
rence through the advertisement of more flexible targeting designs (the so-
called Schlesinger doctrine).24However, it is important to note that 1970s-style
strategic flexibility was, at bottom, an endeavor to retain the credibility of
the ultimate sanction of the very large countersociety strike. What beliefs,
attitudes, and perspectives are reflected in this simple theory?
First, it reflects a belief that nuclear war would mean the end of history.
The assumed certainty of unrestrained escalation and mutual destruction
leads easily to the conclusion that there can be n o intelligent way of prepar-
ing for or waging nuclear war.'"
Even if some stable balance theorists are prepared to admit that nuclear
war could have quite a wide range of outcomes, they tend to reject the pos-
sible policy implication - that the United States should design a policy and
posture to minimize the prospective damage in war. U.S. political culture,
unlike Soviet political culture, does not take an instrumental view of the
value of the lives, and quality of life, of its citizens. But U.S. foreign policy,
in its potential need for military support, rests heavily on nuclear threat.
However, no operational nuclear strategy is compatible with U.S. societal
values. An important reason US. strategic commentators have focused so
heavily on deterrence, as opposed to military operational questions, is that
they realized American society is profoundly unwilling to contemplate, or
debate coolly, the prospect of losing tens of millions of people.
For the better part of two decades, the United States has been highly
dependent on latent nuclear threat, but American society and even the
defense community have shown little inclination to think beyond prewar
deterrence, let alone to invest large resources in a capability to prevail in,
survive, and recover from a nuclear war. Michael Howard was close to the
mark when he wrote recently:
But such credibility [of nuclear response] depends not simply on a per-
ceived balance, or imbalance, of weapons systems, but on perceptions of
the nature of the society whose leaders are threatening such retaliation.
Peoples who are not prepared to make the effort necessary for opera-
tional defense are even less likely to support a decision to initiate a
nuclear exchange from which they will themselves suffer almost incon-
ceivable destruction, even if that decision is taken at the lowest possible
level of nuclear escalation."
Second, stable balance theory reflects a conviction that an enduring East-

West political modus vivendi is possible, if only because nuclear arsenals mean
that neither superpower would dare intrude into regions well understood to
be of vital interest to the other. The relationship between intense arms compe-
tition, its associated first-strike alarms, and political tension remains ill under-
stood, but a plateau of stable deterrence resting on total societal vulnerability
and sufficient weapon invulnerability, should - so the argument goes - calm
many of the anxieties that the arms competition can foster.
The more reasonable supporters of SALT I tended t o avoid asserting that
the Soviet political leadership and General Staff had been educated to
accept American-style stable deterrence thinking. They assumed instead
that American strategic vigilance would deny the Soviet Union any militar-
ily meaningful future advantage and that Soviet leaders would rein in their
programs appropriately. In addition, it was widely assumed that the five-
year "Interim Agreement" on strategic offensive arms would be superceded
by a permanent treaty regime that would greatly assist stability, owing to
the survivability for offensive forces and the predictability for defense plan-
ning it would provide.27 Stability could be enforced through expensive com-
petitive effort, but the case for attempting t o encourage stability through
negotiated joint management of the strategic balance had to be attractive.
In short, stable balance theory was believed t o reflect inescapable tech-
nological those truths were t o be codified, at least in part, via the
SALT process; and the SALT process was t o be both the centerpiece and the
beneficiary of a multichannel and increasingly entangling dttente venture.
Third, stable deterrence theory indicated "how much is enough."29 U.S.
strategic culture is oriented toward problem-solving. The U.S. defense and
arms control community has extreme difficulty accommodating the idea
that it is condemned to an endless competition with the Soviet Union. Stable
deterrence plus the parity principle appeared to reduce the stress and strain
of unwelcome and unfamiliar strategic thought to a fairly simple problem
of efficient management.
Fourth, stable deterrence, with its logical implication of a finite need for
weapons, appeals to the Western belief that peacetime defense preparation
has an almost wholly negative social impact. An insular strategic culture,
such as that of the United States, tends generally to view as inherently
wasteful the allocation of scarce resources for defense functions. It supports
substantial armed forces in peacetime only because such forces are, at best,
regrettable necessities. Major defense program initiatives often are taken
belatedly and reluctantly, and have to be justified very specifically in terms
of identifiable or very plausible threats.
Even on its own terms it is legitimate t o question the validity of main-
stream U.S. stable deterrence theory. For example: as Henry Kissinger has
argued forcefully, in policy practice it constitutes "a revolution in NATO doc-
trine"" (which was not noticed, or was simply disregarded, by its propon-
ents); it has nothing to say on the problem of self-deterrence (which is not
a trivial deficiency, because it is likely that the United States would be under
the most pressure to lead an escalation process); and it is not responsive to
the fact that deterrent calculations are not always relevant in the sequence
Cray Strategic Stability Reconsidered 23 1
of events that lead to war. Leaving such reservations aside,31the most trouble-
some aspect of main-stream stable mutual deterrence theory is that it does
not speak to Soviet reality.
T h e Soviet Union
Soviet thinking on the preferred character of the complex East-West military

balance is easily identified as a product of the lessons perceived in Russian
and Soviet history, the nature and rationale of the Soviet state, and what may
best be termed "strategic logic." The Soviet Union cannot endorse a Western-
style concept of military stability. The legitimacy of Communist party rule in
the Soviet imperium resides in its claim to be the sole authoritative interpreter
of the scientifically correct theory of historical change - and the peoples and
the physical resources of that imperium, allied to "progressive forces" every-
where, are the instruments for effecting that process of historical change. Save
as a tactical ploy, the Soviet Union cannot endorse a concept of stability in
the relations between socialist and nonsocialist states. Richard Pipes almost
certainly is correct when he argues that Marxism-Leninism became the state
ideology in Russia because the grosser features of that ideology, and the prac-
tices that they legitimized, fit so well a Russian national political character
marked by cunning, brutality, and submissiveness." Soviet military thinking
today, by this argument, is influenced by, and expresses, a strategic culture
that is, at root, Russian rather than Marxist-Leninist."' The important point
is that obligatory Soviet ideology and Russian historical impulses both drive
Soviet military thinking in the same direction.
The commitment to permanent struggle, the need for eternal vigilance,
the militarized character of society, the fundamental distrust of independent
power centers (domestic and foreign) - all are enduring features of Russian1
Soviet strategic culture. "The revolution in military affairs," as evidenced in
Soviet military programs and discussed in detail in the Penkovskiy "Special
Collection," was dramatically different from the revolution in strategic think-
ing caused by nuclear weapons in the West.
The Western nonoperational focus on deterrence as opposed to defense
- -
is totally alien to Soviet strategic culture and, indeed, is viewed as danger-

ous, irresponsible, and scientifically incorrect. Since 1956 the Soviet Union
has rejected Lenin's "inevitability of war" thesis, but it has continued to
believe that war is possible, that the outcomes could range between victory
and defeat, and that more military power cannot fail to pay political divi-
d e n d ~ . "The
~ notion of having enough military power is alien to Soviet
thought and appears to be contrary to the Soviet reading of their history and
the histories of other states." The Soviet Union recognizes that equivalence
and parity are the necessary basis for East-West security relations, but that
the necessary basis is not, and cannot be, accepted as being sufficient.
Quite aside from any ideological imperative, Soviet geopolitics - like
Russian geopolitics in times past - is the story of near-continuous struggle
against actual or potential enemies who posed, or might pose, a threat to the
(mu1ti)national existence.36 Russian and Soviet history teach the lesson that
"those who fall behind, get beaten."37 The Soviet Union is improving its
security condition by increasing its control over its external environment.
Even discarding cultural and geopolitical explanations, the detail of
Soviet military activity drives one to recognition of the deeper imperatives
that have molded Soviet strategic c u l t ~ r e . 'There
~ is an enormous inertia
behind the Soviet military establishment, some of which can be explained
in Western military-rational ways. But much of it reflects a mindless move-
ment that flows from habitual "safe-siding" through minimal decision-
making, eschewing of potentially dangerous initiatives, and generally
focusing on doing what one knows one can d o - in a society that is near-
obsessed with the fear of disruptive change and that seeks to avoid risks.
Of course, innovation is possible in the Soviet Union, though it gener-
ally is ordered and even organized from above. The Soviet military buildup
and modernization programs of the past fifteen years (in particular) speak
to forces very deep within the character of the Soviet system. Alarmed
Western observers see clear evidence that the Soviet Union is building more
military power than it needs for defense (a totally alien formulation in
Soviet perspective) and rejecting the Western concept of a stable military
balance. What we see, rather, is the cumulative product of a bureaucratic-
industrial system that rarely changes a course once set (there is no evidence
to suggest any official Soviet desire to change military direction) and that is
steadily providing the military means needed for preponderance - the Soviet
vision of a desirable military relationship with potential enemies.
The strategic cultural attitudes that flow from its history make it unlikely
that the Soviet Union will join the United States in managing a stable military
balance. The Soviet commitment to compete for relative advantage (real or
illusory) is so fundamental, and so rational in Soviet terms, that stability
can only be enforced.
The implications of this strategic cultural theme could be very grim for
Western security. By dint of fairly steady effort, and moved by a prudence
that has expansive military requirement implications, the Soviet Union
could come to believe that in East-West crises it will be the United States
that will back down. The ideas and military program details associated with
the dominant Western concept of stability amount to a posture, military
and civilian, that is not serious about the actual conduct of war. To itemize:
the United States has a very limited hard-target counterforce and counter-
control capability; it lacks survivable command, control, communications,
and intelligence (C31)assets; it has no homeland defense; it has no real plans
for timely industrial mobilization or for postwar recovery; it has no vision
of how all parts of the military posture should cooperate in a global war; it
has made only the most feeble preparation for strategic-force reconstitu-
tion; and it has no convincing story to tell vis-a-vis war aims and the polit-
ical character of a postwar international order.j9
These criticisms are leveled in the context of attempts to provide
adequately in those areas by a Soviet adversary.40That some weapons and
s , I~ r i Strategic Stability Reconsidered 233
operating practices promote stability and that others promote instability is

an idea that is alien to Soviet strategic ~ u l t u r e . For
~ ' many years the Soviet
Union has attended, in great detail, to what in Western perspective might
be called the unilateral crisis stability of her military posture: missile silos
have been superhardened; some missiles are truly mobile (SS-20s and stock-
piled SS-16s); political and military command and control facilities have
been proliferated and superhardened, and so on. Whether for reasons of
political-cultural insensitivity or cold military calculation, the Soviet Union
seems either unwilling or unable to take a systemic approach to what
Western analysts identify as stability problems. Judging by the evidence of
Soviet deeds, and to employ Western terminology, it is stabilizing, in Soviet
perspective, for their strategic and combined armed forces to threaten suc-
cessful surprise attacks against U.S. strategic systems and NATO's posture in
Europe." Similarly, Soviet military thinkers see nothing unstable about a
strategic context wherein Soviet society is afforded some useful measure of
protection via civil defense and air defense and American society has none.
There is a distinct possibility that a future U.S. administration could
believe the Soviet Union to be deterred by the assumed short fuse from
provocative (Soviet) military action to nuclear holocaust - a belief that pro-
jects stable deterrence reasoning onto Soviet decision-making processes -
while a Soviet leadership could believe it had a very good prospect o f win-
ning a war and that the United States should appreciate its own weak polit-
ical position and back down. In short, both sides might falsely project the
perspective of their strategic culture onto the other, with very dangerous
consequences.
Stability Dissected
The concept of stability is used in a wide variety of senses, among which two
merit individual analytic attention: arms race stability and crisis stability.
The idea of arms race stability holds that the basic engine of competition is
the first-strike fear encouraged by defense programs designed to threaten at
least part of the opponent's ability to wreak massive societal damage in a
second strike." Thus a stable arms competition is one wherein neither side
invests in programs that the other would view as a challenge to its assured
destruction capability, and hence would be motivated to offset. This kind of
logic was elaborated in some detail in the late 1960s. It was argued that the
arms race was driven not so much by the reality of first-strike danger as by
the fears that flowed from anticipation of such danger.44
"Sympathetic parallelism" in armament programs was the logical corollary
of the arms race "spiral" theory. Just as the superpowers could stimulate each
other to build increasingly capable weapons, so should they be able, through
deliberate restraint and perhaps explicit cooperative management, to remove

much of the anxiety that drives reactive armament programs.
The concept of arms race stability carries with it the stable deterrence ideas
that incorporate the desideratum of mutual assured destruction capabilities,
but such linkage is not inevitable. Arms race stability could obtain where one
side maintains a permanent, variably substantial lead and is in a political,
financial, and industrial position to deter most arms race challenges. With
some qualifications, this kind of arms race stability characterized Great
Britain's naval relations with her actual and potential rivals from the 1840s
until 1914.
Stability can also obtain when there is a rapid change in technological gen-
erations and considerable unpredictability concerning the building programs
of rivals, yet where a tolerable balance of military power is maintained -
albeit almost exclusively through competition. Indeed, as Bernard Brodie
observed in assessing the complex naval competition of the later decades of
the nineteenth century, there were periods in strategic history when stability,
by any reasonable definition, was best maintained through unrestrained
omp petition.^^ Arms control processes are as likely to constrain the wrong
(i.e., ultimately "stabilizing") as the right (i.e., ultimately "destabilizing")
defense technologies - given human frailty in strategic prediction.
In its loosest, though most easily defensible, sense, arms race stability
could pertain simply to the pace and degree of rival postural change,
regardless of the character of that change. An unusually rapid succession of
deployed weapon generations on both sides would appear to many people
to constitute an unstable situation. However, such rapid change may reflect
a particularly fecund period of parallel defense research activity rather than
unusual hostility, and may be fully compatible with some important defin-
itions of a stable situation. Nonetheless, rapid postural change might very
likely breed fears abroad that militarily significant, if transitory, break-
throughs were a distinct possibility, breakthroughs that might facilitate or
enable disarming first strikes to be planned with some confidence.
Perhaps the major problem with the concept of arms race stability is that
it rests on a theory of arms race dynamics that is easily ~hallenged.~"he
stability theory of the 1960s posited an abstract and very simple model of
arms competition. The banner-carriers for arms race stability in the late
1960s leapt from abstract propositions to defense policy claims and arms
control proposals (e.g., d o not deploy BMD or MIRV because they will be
destabilizing). The arms race (and crisis) instability claims used t o challenge
BMD and MIRV (and, later, the M k 12A RV and MX, and so on) were, by
and large, interesting and internally consistent. But were they true, and how
might they be validated or invalidated? This argument applies, of course, to
all sides in the ongoing U.S. defense debate.
It is possible that Soviet offensive force deployments in the 1970s would
have been greater than actually was the case had the United States, in the
absence of the ABM treaty, deployed Safeguard or Site Defense BMD.
L I ! , ~ Strategic Stability Reconsidered 235
However, in the presence of U.S. BMD deployment, opponents would very

likely have attributed the pace and much of the character of the Soviet
ICBM and SLBM programs to alleged Soviet BMD-offset motivations. The
kind of Soviet offensive-force deployments that lack a strong strategic
rationale in the absence of U.S. BMD, assuming there is a mutual assured
destruction framework to Soviet thinking, have occurred anyway.
On the basis of the often ambiguous and incomplete evidence available,
the U.S. defense and arms control community should now consider the
proposition that Soviet arms programs are driven not by a determination to
(over)compensate for American programs that could threaten Soviet main-
tenance of an adequate capability to destroy American society, but, rather, by
some combination of a doctrinal imperative to improve Soviet war-waging/
war-winning ability and bureaucratic defense-industrial momentum. This
proposition suggests that for many years our arms control physicians may
have diagnosed falsely (and hence sought to cure inappropriately) the causes
of the arms race disease.
Many people who debate arms race stability/instability charges are really
concerned that continuous competitive military-technological innovation might
open temporary windows of opportunity for possible exploitation. "Gaps"
might occur between some elements of superpower postures, but they should
not be such as to call into serious question the overall quality of deterrent effect
purchased by the United States through its military investment.
Deterrence stability is compatible with a formidable rate of change in coni-
peting postures. For example, the charge that the MX will be destabilizing can
be sustained only if arms race instability is equated with a significant change
in posture that may provide a substantial incentive for postural change on the
Soviet side. A crisis instability charge is fragile in that the very survivability
of a system such as the MX should remove the Soviet incentive to go first in
a "use them or lose them" spirit.
The concept of crisis stability refers to a strategic condition wherein the very
character, readiness, and mobilization procedures of armed forces in co11-
frontation should not themselves comprise the proximate cause of war. Very
often crisis stability/instability is deemed to inhere in particular kinds of
weapons. However, as Thomas Schelling has argued persuasively, to focus
on weapon technology is to miss a good part of the potential problem.
To impute this influence to "weaponry" is t o focus too narrowly on

technology. It is weapons, organizations, plans, geography, communica-
tions, warning systems, intelligence, and even beliefs and doctrines about
the conduct of war that together have this influence. The point is that
this complex of military factors is not neutral in the process by which
war may come about.4'
Schelling draws a particularly valuable distinction between the static and

dynamic dimensions of (crisis) stability:
The static dimension reflects the expected outcome, at any given moment,
if either side launches war. The dynamic dimension reflects what happens
to that calculation if either side or both sides should move in the direction
of war, by alert, mobilization, demonstration, and other actions that
unfold over time.48
It is easy to slip into self-congratulation concerning the apparent stability of

the military standoff in Europe and the central nuclear relationship. However,
the stability of those balances is not tested day-by-day, nor even by the kinds
of crises such as those over Berlin, Hungary, Cuba, and Czechoslovakia. In
none of those cases is it very plausible to argue that either the Soviet Union or
NATO was strongly motivated to launch a theater or general war. The real test
for crisis stability would be the one occasion in forty or fifty years when nearly
everything appeared to be at stake and one or both leadership groups could
not see any nonmilitary solutions to its, or their, problems.
A force posture and strategic doctrine that is good enough for one crisis
may not be for another. Anyone inclined to believe that U.S. and NATO
forces are broadly resilient to crisis stress should ask himself what it might
take t o dissuade a very desperate Soviet leadership.
Robert Jervis has argued, quite rightly, that rival schools of thought over
the requirements of deterrence differ over how much deterring it is prudent
to assume the Soviet Union might need: "Thus there is a disagreement over
'how much credibility is enough': two policy analysts therefore might agree
on how likely the Russians thought it was that a limited war would escal-
ate and disagree over whether they could be deterred."49
Taken to its logical extreme, the more pessimistic argument might lead to
the conclusion that at some point in the future the Soviet Union might be so
desperate as to be "beyond deterrence" - meaning, again logically, that con-
siderations of crisis stability, however rigorous, would be irrelevant. The
only question remaining would be: How well would the West fare in the
war? Analysts may agree on the general characteristics of a crisis-stable mili-
tary balance and even on the character of Soviet strategic culture, yet dis-
agree on whether particular U.S. military postures are sufficiently crisis-stable.
The reason for the disagreement lies in the range of political crises that each
is willing to consider as relevant to the sizing and nature of the U.S. defense
effort. Some interpretations of the military implications of crisis stability
have a potentially dangerous managerial, as opposed to a strategic, perspec-
tive on security issues. Crisis stability is fully compatible with a U.S. strategic
force posture that could take the initiative, compete for escalation dominance,
and, if need be, fight the war through to a military decision. However, crisis
stability very often is considered narrowly in the context either of a rigid appli-
cation of mutual assured destruction reasoning or, beyond that in sophistica-
tion, in the minor addendum of flexible targeting. Typically, any capability
i Strategic Stability Reconsidered 237
that threatens Soviet strategic forces, prelaunch or during mission execution,

is held to be an affront to crisis stability. Crisis stability properly understood
does not lend its conceptual authority to such judgments. For reason of
extended deterrence duties, the United States cannot afford a crisis stability
that precludes first use of strategic nuclear weapons. Nicholas Spykman
writes: "There is no possibility of action if one's strength is fully checked;
there is a chance for a positive foreign policy only if there is a margin of force
which can be freely used.""'
Jerome Kahan has written that "in order to establish a mutual stability
policy, it is necessary to classify strategic systems as either stabilizing or
destabilizing and to avoid the latter."" Following classical mutual vulner-
ability theory, Kahan claims that weapons threatening to the (countervalue)
mission performance of strategic offensive forces are destabilizing, "since
they can directly negate an opponent's deterrent capability."" Examples of
"stabilizing" weapons include SLBMs; MRVs, or inaccurate MIRVs; long-
range cruise missiles; manned bombers; and missile site (or bomber base)
BMD. "Destabilizing" weapons include accurate MIRVs, strategic ASW
systems, area BMD, and area air defense.
This simple classification is only as useful as are its doctrinal premises.
If, for example, the Soviet Union does not equate the quality of its deterrent
with its ability to devastate urban-industrial America, defense of the urban-
industrial U.S. homeland would not threaten the Soviet deterrent. Moreover,
it can be argued, as noted above, that overall stability in the East-West military-
political relationship requires that the United States be able to initiate
strategic nuclear use in defense of forward-located allies, and that such cen-
tral war initiation, no matter how selective, cannot be credible in the event,
unless a president of the United States was confident that damage to the
U.S. homeland could be limited severely.
Given the Soviet traditional military approach to nuclear war planning,
strategies and tactics that in the West tend to be judged as destabilizing
almost certainly have no such implications in Soviet thinking. Soviet political
and military thinkers would he most unlikely, for example, to view programs
intended to provide active and passive defense of the American homeland as
anything other than common sense.'' To the extent that those programs
threatened the success of Soviet plans for the military conduct of the war, they
would be candidates for some Soviet response. However, the mechanistic
ying-yang envisaged in some simpleminded defense-offenselaction-reaction
theories of the arms race is the stuff of the (American) seminar room, not of
the real world of Soviet defense decision-making.
The small strategic theory community has p a ~ dvery l~ttleattentlon to the

place, let alone the details, of C z l . L ~ k epeace and security, everyone, from
every school of thought, 1s f o v good C31.Understandably, it would be d ~ f f ~ c u l t
to generate a debate over whether the Un~tedStates requlres h~gh-qual~ty C'I.
Noncontrovers~alsubjects tend to escape attentlon, John Ste~nbrunerhas
written that "the most severe problems with the concept of stability result
from the fact that its technical definition has not included a critical dimension
of strategic capability: namely, the physical and organizational arrangements
for exercising deliberate command of strategic forces."s4 Steinbruner argues
that when the concept of stability is expanded to accommodate C31 desider-
ata, the preferred force structure (given classic stability themes) might alter
markedly. For example: "The submarine-based strategic force which is clearly
the most stable under the conventional definition is just as clearly the worst
in terms of command stability."""
Those theorists who believe that deterrence is a function of mutual soci-
etal vulnerability should be concerned lest command instability results in
unintended armed conflict or in essentially uncontrolled escalation in the
course of a war. Those theorists who believe that deterrence flows from the
promise of proficient military conduct should be concerned lest command
instability denies the U.S. armed forces the ability to wage war in a militarily
intelligent fashion.
A good fraction of the strategic debate of recent years rested on quite
unrealistic assumptions concerning the quality and survivability of U.S.
(and NATO) C31 assets. There was much weaving of interesting strategic
targeting tapestries in the 1970s, but I suspect that most of the targeting
schemes that envisaged the protracted and progressive unfolding of a delib-
erate design of destruction (for carefully calculated military and political
effect) failed to take adequate note of likely or ~ossiblecommand instabil-
ity phenomena (U.S. and Soviet).
There is ample evidence suggesting that classic stability theory, which
encourages the belief that nuclear war would be the end of history, pro-
moted a relaxed climate concerning the many details of actually managing
a central war campaign. A dominant belief that nuclear forces have failed
if they are ever used is hardly likely to stimulate officials to think very real-
istically about command stability problems in a nuclear war environment.
Steinbruner's persuasive advocacy of the need to place command stability at
the center of nuclear (and other) planning concerns fails to recognize that
the relative neglect of command stability issues flows in good part from the
widespread acceptance of a classic stability theory (based on the assump-
tion of the desirability of mutual societal vulnerability) that he approves.
Stability and U.S. Strategy
John Newhouse, the privileged chronicler of the NSC perspective on the

SALT I negotiations, asserted that stability was "a truly divine goal."'6 Today
it is apparent that the theories of arms race and crisis stability that permeated
the U.S. approach to SALT I were either wrong or misleading. At a general
conceptual level, arms race and crisis stability are, of course, unexception-
able. N o one favors frenetic arms race activity per se or military postures that
could precipitate war: so much is well-nigh axiomatic. Where the mainstream
Ci r d) Strategic Stability Reconsidered 239
of U.S. strategic theorizing erred was in tying the multifold concept of stability
to a particular theory of deterrence that did not match the burgeoning evidence.
That theory held that each superpower had an assured destruction (counter-
value) requirement vis-a-vis the other and that an enduring stable deterrence
relationship could be constructed only on such a basis.
This theory of arms race stability was wrong - it could not explain the
course of the strategic arms competition in the 1970s (under the aegis of
SALT I or in the shadow of SALT 11). Whatever mix of motives and institu-
tional forces drove Soviet weapon procurement, the principle of sufficiency
supported by the idea of assured destruction (let alone mutual assured
destruction) clearly was not prominent among them. It is a matter of histor-
ical record that the Soviet Union since 1972 has worked hard to undermine
whatever degree of strategic stability (based on mutual societal vulnerability)
there may have been at that time. In their ICBM, air defense, BMD (in
research and development), ASW, and civil defense programs the Soviets have
provided persuasive evidence that their systemic view of the arms competition
is dramatically different from the view adhered to by succeeding U.S. admin-
istrations. They have sought, and are continuing to seek, "useful advantage"
through whatever degree of preponderance the United States permits.'?
The "classical" theory of crisis stability may o r may not be correct; for-
tunately, the 1970s have not provided a field test. However, the Soviet per-
spective on strategic matters suggests that the explanatory power of the
theory may be poor. Richard Burt expressed this skepticism when he wrote:
"Central strategic war, according to Soviet literature, is not likely to stem
from mechanistic instabilities within the superpower military relationship,
but rather from real and enduring differences between competing political
systems and national interests." j8
In principle, certainly, it is sensible to argue that it would be undesirable
for the superpowers to deploy forces that lend themselves to first-strike
destruction. However, it is no less sensible to argue that "the reciprocal fear
of surprise attack"" as the principal proximate cause of war merits prob-
able identification as a U.S. "mechanistic" fantasy. This is not to endorse a
total indifference to Burt's "mechanistic instabilities," but to suggest that
the traditional theory of crisis stability - on the basis of which particular
weapons and doctrines are praised or vilified - needs considerable amend-
ment. It overemphasizes the roba able role of mechanistic instabilities in an
acute East-West crisis while taking a wholly apolitical approach to an inher-
ently political phenomenon; and it is inimical to the extended deterrence
requirement that the United States be capable and willing to take the stra-
tegic initiative.
Many of the elements of a new theory of strategic stability already have
been expressed in official prose and action over the past five years. However,
the theoretical revolution remains incomplete. What is missing, above ail
else, is a recognition of the pervasiveness and longevity of competition and
a positive approach to the functions of strategic nuclear forces. On this last
point, for example, Harold Brown treats both arms race and crisis stability
in negative terms. For arms race stability, the United States must ensure
"that the balance is not capable of being overturned by a sudden Soviet
technological break-through"; for crisis stability, it must ensure that neither
the United States nor the Soviet Union would feel itself under pressure to
initiate an exchange in a crisis.60
Brown's concerns are appropriate, but they do not approach the heart of
what stable deterrent ideas should indicate vis-a-vis U.S. force planning. An
adequate concept of stability has to be anchored in a prospectively effective
theory of deterrence a t the highest levels of violence. Crisis stability should
be approached in terms of the calculations of probable war-waging prowess
made by the parties involved. Concern about mechanistic, or technical, cri-
sis (in)stability would be policy-appropriate only in a condition of such
intense antipathy that overall central war campaign analyses would dom-
inate decision processes. The Soviet Union would not "go to war" because
a large fraction of its ICBM force was theoretically vulnerable to a U.S. first
trike,^' any more than would the United States should the situation be
reversed. Crisis stability, if possible, would flow from a Soviet belief that
any escalation of the military conflict would produce negative military and
ultimately political returns. The U.S. Department of Defense acknowledges
this but it does not recognize that the United States is most unlikely to
be able to enforce stability if damage to the U.S. homeland cannot be limited
severely.63
Strategic stability should not be equated with strategic stalemate. The
United States cannot afford a master strategic concept that implies thorough-
going mutual U.S.-Soviet strategic d e t e r r e n ~ eIf. ~strategic
~ stability is to retain
its preeminence as a U.S. policy goal, it should be redefined for compatibility
with the extended deterrent duties that the geopolitics of the Western Alliance
place on the U.S. strategic force posture. A stable strategic balance, in
U.S./NATO perspective, is one that would permit the United States to:
Initiate central strategic nuclear employment in expectation of gain (a

requirement of NATO strategy)
Seize and hold a position of "escalation dominance"
Deter Soviet escalation, or counterescalation, by a potent threat posed
to the most vital assets of the Soviet state and by the ability of the United
States to limit damage to itself6'
A Soviet Union confronting a United States with military and civilian

programs appropriately supportive of the above objectives would have very
little incentive either to effect a military "breakout" from a regional crisis
or to engage very persistently in a competition in risk-taking at very high
levels of violence. Crisis stability would be enforced through the Soviet per-
ception of the United States as a very tough wartime adversary indeed. It
might be objected that a U.S. president should not be entrusted with the
capabilities suggested above.66 However, even if such a concern is valid
(which is extremely dubious), it must be weighed against the greater danger
i I 1 Strategic Stability Reconsidered 24 1
of a president not having recourse to such capabilities. The concept of

strategic stability envisaged here is the only one that speaks persuasively to
Soviet strategic culture, and it is intended, of necessity, only to minimize
that self-deterrence element that is the most crippling deficiency in existing
U.S. official strategic thought. Self-deterrence cannot be removed altogether
because the United States would know that even under the aegis of a stable
military balance, as defined here, tens of millions of American casualties
would most likely result from central war. Nonetheless, the United States
would have a guiding concept from which military requirements could be
derived in support of militarily and politically intelligent strategic targeting
plans. This concept relates robustness in crisis regimes to anticipation of
success or defeat in war and to a judiciously competitive program of peace-
t ~ m earmament.
As stated earlier, the identifiable Soviet approach to arms competition is
the steady acquisition of a more and more formidable war-fightinglwar-
survival capability. It is highly improbable that the Soviet Union can be dis-
suaded trom pursuing this a p p r ~ a c h . ~The
' evidence of the 1970s suggests
that although in principle stability might be encouraged through negotiated
SALT restraints - whereby both sides agree to forgo those capabilities that
the mutual vulnerability theory of stability holds to be undesirable - it is far
more likely that stability has to be enforced through competition. It is vir-
tually self-evident that Soviet strategic culture precludes the negotiation
route to enhanced stability, except in the context of very vigorous U.S.
strategic effort. Moreover, there is growing agreement within the Western
defense community that stability cannot rest intelligently on the threat of
massive societal destruction (save, possibly, as an ultimate threat). Such
damage is unacceptable to the United States, while it may be insufficiently
unacceptable to Soviet politicians. If the U.S. concept of a stable military
balance in extremis makes more or less formal reference to the assured
destruction threat, the United States has a deterrence theory that probably
is fundamentally unsound. The "ultimate threat" posed by the United
States would not be credible because it would never be in the U.S. interest
actually to implement it. Execution of such a threat would be the negation
of strategy: in and of itself it would solve no military or political problems,
while it would near-guarantee a Soviet retaliation that would preclude a
U.S. recovery.
The strategic nuclear targeting review of the late 1970s has prepared the
way for serious discussion of the concept of stability suggested in this
essay." The U.S. government recognizes that Soviet military and political
assets should be the primary focus for U.S. strategic offensive attention;"
that "limited nuclear options" have little promise unless the United States
has a good theory of escalation dominance (and the forces to match); and
that Soviet economic "recovery" targets are both difficult to identify and
are probably of relatively little interest. However, Washington does not yet
recognize that crisis and intrawar stability cannot rest on intelligent stra-
tegic offensive ~ l a n n i n galone. The U.S. Single Integrated Operational Plan
(SIOP) can have integrity only in the context of active and passive defense.
Fortunately, there is good reason to believe that the technology of air and
missile defense for the late 1980s and b e y ~ n d , ' ~with substantial civil
defense assistance, could restore a much more even relationship between
offense and defense, and a useful meaning to the concept of stability.
Author's Note
This paper represents my views alone. It has had only limited circulation among the Hudson
Institute staff and n o formal review procedure. N o opinions, statements of fact, o r conclusions
can properly be attributed to the Institute, its staff, its Members, o r its contracting agencies.
I am very grateful to Keith Payne of the Hudson Institute professional staff for his valuable
assistance with this article.
References
1. John Steinbruner, "National Security and the Concept of Strategic Stability," The]ournal
o f Conflict Resolution, 22(3) (September 1978): 413.
2. For example, see Thomas C. Schelling, The Strategy of Conflict (Cambridge, Mass.:
Harvard University Press, 1960). "It is not the 'balance' - t h e sheer equality or symmetry
in the situation - that constitutes mutual deterrence; it is the stability of the balance. The
balance is stable only when neither, in striking first, can destroy the other's ability to strike
back" p. 232.
3. Ken Booth, Strategy and Ethnocentrism (London: Croom, Helm, 1979), p. 14.
4. "Hostages must remain unambiguously vulnerable and retaliatory forces must remain
unambiguously invulnerable." Ian Smart, Advanced Strategic Missiles: A Short Guide,
Adelpbi Papers no. 63 (London: International Institute for Strategic Studies, December
1969), p. 28.
5. See Harold Brown, Department of Defense Annual Report, Fiscal Year 1981 (Washington,
D.C.: U.S. Government Printing Office, January 29, 19801, pp. 65, 67.
6. See Colin S. Gray and Keith B. Payne, "Victory Is Possible," Foreign Policy, no. 39 (Summer
1980): 14-27.
7. See Stanley Sienkiewicz, "Observations on the Impact of Uncertainty in Strategic
Analysis," World Politics, 30, II(1) (October 1979): 90-1 10.
8. Jack Snyder, The Soviet Strategic Culture: Implications for Limited Nuclear Operatrons,
R-2154-AF (Santa Monica, Calif.: RAND, September 1977), p. 8.
9. John Erickson, "The Soviet Military System: Doctrine, Technology and 'Style,'" in S o ~ i e t
Military Power and Performance, Erickson and E. J. Feuchtwanger (eds.) (Hamden,
Conn.: Archon, 19791, pp. 1 8 4 3 .
10. Y. Harkabi, Nuclear War and Nuclear Peace (Jerusalem: Israeli Program for Scientific
Translations, 19661, p. 48.
11. Jerome Kahan, Security in the Nuclear Age: Developing U S . Strategic Arms Policy
(Washington, D.C.: Brookings Institution, 1975), p. 272.
12. Roman Kolkowicz et al., The Soviet Union and Arms Control - A Superpower Dilemma
(Baltimore: Johns Hopkins University Press, 1970), pp. 35-57.
13. See Fritz Ermarth, "Contrasts in American and Soviet Strategic Thought," International
Security, 3 ( 2 ) (Fall 1978): 138-55.
14. See William E. Odom, "Who Controls Whom In Moscow," Foreign Policy, no. 19 (Summer
1975): 109-23.
15. Robert Jervis, "Why Nuclear Superiority Doesn't Matter," Political Science Quarterly, 94(4)
(Winter 1979-80): 630.
Lid) Strategic Stability Reconsidered 243
See Kenjamin S. Lambeth, "The Political Potential of Soviet Equ~valence,"Intrrwationnl

Srr-rrvrty, 4 ( 2 ) (Fall 1979): 22-39.
An evcellent pre5entation of this important pomt is given by Robert 1.egvold In "Strategic
'Doctrine' m d SALT: Soviet and American Views," S u r v i ~ u l ,21 ( 1 ) (Jan~~arylFehruar!:~:~ry
1979): 8-1 3.
See 1-ambeth, "The Political Potential of Soviet Equivalence," passim.
The ~~clclass~fied literature o n U.S. nuclear targeting policy is almost pathetically t h ~ n M . y
recent contributions include "Nuclear Strategy: The Case for a Theory of Victory,"
Internationid Srmrity, 4 ( l ) (Summer 1979): 54-87; and "Targeting Problems for Centrill
War," Nau'11 War (:allege Reiww, 33(1) (J'inuary-iebruary 1980): .3-21. I b r a very d ~ f -
terent, hut usetul, perspective, see William H. Kincade, "A Strategy for All Seasons:
Targeting Doctrme and Strategic Arms Control," Biilletin of the Atomic Scientists, 3 4 ( 5 )
(May 1978): 14-20.
See Brown, Lkpnrtrnent of Defense Annual Report, FY 1981, p. 65.
Ihid., pp. 82-3.
See Booth: Strategy and Ethnocentrrsm; and "American Strategy: The Myths Revisited,"
in Aniericarr Thinking ahout Peace and Wilr, Booth and Moorhead Wright (eds.) (New
York: Harper and Row, 1978), pp. 1-35. Also ~ ~ s e f is u l M'itthew P. Gallagher and Karl
F. Spielmann, Sollret L)ecrsron-Makrng for Defense: A Criticlire of U.S. Perspcrtilm o n t h t
Arms Race (New York: Praeger, 1972).
Bernard Krodie, The Al~soluteWeapon ( N e w York: Harcourt, Brace, 1946).
See Henry S. Rowen, "Formulating Strategic Doctrine," in C;orntnission o n the O ~ g ~ ~ t i i z ~ t i o ~ i
fir t / ~ Conduct
e of Forergn Policy, vol. 4, Appendix K (Washington, I>.<:.: U.S. Government
Printing Oftice, 1975), pp. 219-34.
Leon Sigal, "Rethinking The Unthinkable," Forergn Policy, no. 3 4 (Spring 1979): 39.
Michael Howard, "The Forgotten Dimensions of Strategy," h r e ~ g nAffairs. 5 7 ( 5 ) (Summer
197')): 983.
Notwithstanding the retrospective wisdom that claims, w ~ t hstrict accuracy, that "they
1SAl.T I and SA1.T 111 did not create the problem ot Minutem'ln s u r v ~ v a b i l i tand ~ cannot
be expected to cure it" (Secretary Brown, in U S . Senate Committee o n Foreign Relations,
' T / ISA1,T
~ 11 Treirty, Hearings, Part I, 96th Congress, 2d session [Washington, D.C.: U.S.
<;ove~-nrnent Prmtlng Office, 19791, p. 159), the historical record shows quite ~ ~ n s r n h i p -
ously that SALT I w x advertised, officially, in 1972, as linpos~ngimportant constraints o n
Soviet "l~ght"missile programs. "The SS-19 problem," as it is known, entailed some mix
of 11.4. off~clalincompetence, dishonesty, and cowardice. See D,lvid S. Sulliv,ln, S ~ ~ i e t
SA1.7' I k e p t i o n (Boston: The Coal~tionfor Peace Through Strength, 1979), pp. 1-3.
See Wolfgang K.H. Panofsky, "The Mutual Hostage Relationship between A n ~ e r i c ~and i
Russ1.1," Foreign AffLirs. 52( I ) (October 1973): 109-1 8.
See Alain Enthoven and K. Wayne Smith, H o w Much Is E~zough?Shilping the L)efensc
Progrmr, 1961-1 960 (New York: Harper and Row, I97 I), chapters 5-6.
"Kissinger I.ooks at Future of NATO," Corrgressioniil Record. September 6 , 1979, p. E4292.
These ideas .ire developed In Gray and Payne, "Victory Is I'ossible."
Richdrd Pipes, "A Reply," to Wladislaw G. Kratnow, "Anti-Sowet or Ant]-Russ~an?"
F,nrounter, 5 4 ( 4 ) (April 1980): 67-72. Pipe's reply is o n pp. 72-75.
See Richard Pipes, "Why the Soviet Union Thinks It Could F ~ g h and t Win a Nuclear W'ir,"
Comnzentary, 64( 1 ) (July 1977): 21-34.
See I.amheth, "The Political Potential of Sov~etEquivalence"; and Joseph 1).Lhuglass, Jr.,
and Amoretta M. Hoeber, Sovret Strategy for N~tcleirr War (Stanford, Calif.: Hoover
l n s t ~ t u t ~ oPress,
n 1979). For a usefully skeptical view, see K x l F. S p i e h l n n , Tile P o h i i z l
Utility of Strategrc Superiorlty: A Preliminary Investigation into the Sovict Vieu~,IDA
Paper 1'1349 (Arlington, Va.: Institute for Defense Analyses, May 1979). Still more skep-
tical is Robert I.. Arnett, "Sovict Attitudes towards Nuclear War: D o They Really Think
They <:an Win?" The /ournal of Strategic Studies. 212) (September 1979): 172-9 I.
I a m 'ltrracted t o the r n e r ~ tof the follow~ngjudgment offered by Benjamin 1.amheth: "It
would probably not be overly facetious t o suggest that for Sov~etm~litaryplanners, the
favored measure of strategic sufficiency is the notion that 'too much I S not enough.'" H o u ~
To Think Ahout Souret Milrtary Doctrine, P-5939 (Santa M o n ~ c aCalif.:
, RAND, February
19781, p. 7.
36. See Richard Pipes, Russra under the Old Regrme (New York: Scribner's, 1974), chapter I ;
and Colin S. Gray, The Geopolitics of the Nuclear Era: Heirrtland, Rrnzlands, and the
Technological Reuolzrtion (New York: Crane, Russak lfor the N a t ~ o n a l Strategy
Information Center], 19771, chapter 3.
37. Joseph Stalin (speech in 1931), quoted in Arthur J. Alexander, Decrsron-Making in Sotiet
Weapons Procurement, Adelphi Papers nos. 147-148 (London: International Institute for
Strategic Studies, Winter 1978/9), p. 2.
38. Ibid., passim. See also Karl F. Spielmann, Analyzing Souret Strategic Arms Decisrons
(Boulder, Colo.: Westview, 1978).
39. These allegations are presented and defended in detail in Gray, "Nuclear Strategy: The
Case for a Theory of Victory"; and Gray and Payne, "Victory Is Possible."
40. See Douglass and Hoeber, Soviet Strategy for Nuclear War.
41. See Johan J. Hoist, "Strategic Arms Control and Stabil~ty:A Retrospective Look," in Why
ABM? Policy Issues in the Missile Defense Controversy, Hoist and Wllliam Schneider, Jr.
(eds.) (New York: Pergamon, 1969), chapter 12.
42. Such capabilities discourage adventure on the part of the imperialists.
43. See Jerome Wiesner, "The Cold War is Dead, but the Arms Race Rumbles On," Brrll~trn
of the Atomic Scientists, 2 3 ( 6 ) (June 1967): 6-9; and George W. Rathjens, "The Dynamics
of the Arms Race," Scientific American, 220(4) (Aprd 1969): 15-25.
44. See Lawrence Freedman, U S . Intelligence and the Soviet Strategrc Threat (London:
Macmillan 1977), passim.
45. Bernard Brodie, Sea Power in the Machine Age (Princeton: Princeton Universiry Press,
1941), pp. 252-46. This is a modest expansion of Brodieis point but is faithful to his plaln
meaning.
46. I have challenged that theory at some length in The Sovret-American Arms R x e
(Farmborough, Hampshire, England: Saxon House, 1976).
47. Thomas C. Schelling, Arms and Inpuence (New Haven: Yale University Press, 19661, p. 234.
48. Ibid., p. 236.
49. Robert Jervis, "Why Nuclear Superiority Doesn't ,Matter," p. 622.
50. Nicholar Spykman, America's Strategy in World Politics: The United States and the
Balance of Power (Hamden, Conn.: Archon, 1970, first pub. 1942), p. 21.
51. Kahan, Securrty in the Nuclear Age, p. 272.
52. Ibid., p. 273.
53. Note the judgment on this point in Howard, "The Forgotten Dimensions of Strategy,"
pp. 982-3, 985-6.
54. John Steinbruner, "National Security and the Concept of Strareg~cStability," p. 417.
55. Ibid., p. 422.
56. John Newhouse, Cold Dawn: The Story of SA1.T (New York: Holt, Rinehart and
Winston, 1973), p. 9.
57. See Erickson, "The Soviet Military System," pp. 28-29.
58. "Arms Control and Soviet Strategic Forces: The Risks of Ask~ngSA1.T to Do Too Much,"
The Washington Revrew of Strategic and International Studies, l ( 1 ) (January 1978): p. 22.
59. The title of chapter 9 in Schelling, The Strategy of Confiict.
60. Brown, Department of Defense Annual Report, FY 1981, p. 69.
61. Notwithstanding the enormous significance that the Soviets attach to the surprise disrupt~\.e/
disarming blow, their operational practices v l s - h i s t h e ~ rstrategic forces have never
approached the day-in, day-out instant readiness ethos of SAC and the U.S. SSBN force.
62. Department of Defense Annual Report, FY 1981, pp. 65-68, 85-88.
63. Damage limitation on a major scale is very far indeed from the desiderata of Harold Brown's
Department of Defense. See Brown, Department of Defense Annual Report, Fiscal Year
1979 (Washington, D.C.: U.S. Government Printing Office, February 2, 1978), p. 65. By
way of contrast, a very useful discussion of the Sower approach to damage limitation IS
Strategic Stability Reconsidered 245
I l a n ~ e l Goure and Gordon H. McCorm~c, "Soviet Stratesic Defense: The Neglected

D i ~ n e n s ~ oofn the IJ.S.-Sowet Bnlmcr," Orhis, 24( l ) (Spring 1980): 103-27.
64. For .I strong statement t o this effect from a highly cred~hlesourcr, see (:orrgrcssr~~rzill
Record. "K~ssingerLooks at Future of NATO."
6.5. 1 have explored t h ~ thesis s in some detail in "Ti~rgetlngProblems for Centr,ll W.lr."
66. Sonle US. coninientators believe rhat they should he evenhanded 111 theil- prew-iprions
herween the United States and the Soviet Union. V~ctorIJtgofi of the NSC statt, tor exnmplc,
in ,I speech before the American Institute of Aeronautics , ~ n d Astronautic In Monterc),
C ~ l ~ f o r n i Fa ,e b r ~ u r y1, 1978, argued to the effect t h ~ the
t IJn~tcdStates could not he trusted
not to ahuse a strategically superlor pos~rion.
67. See Pipes, "Why the Sov~erUnion Thinks It Could F g h t a n d Win a Nucle,lr W'lr."
6 8 . liseful background is provided in Henry C . Rowcn, "The Evolution of Strategic Nuclear
Iloctrine," in Str'rtcgic T~J~octght rtr t l ~ eNuclear Agr, I.~urenceMartin (ed.) ( h l t i m o r e :
l o h n s Hopkins Univers~ryPress, 1979), pp. 131-.56.
69. Drporttnctrt of Dcfetrse Anmtal Kcport, k Y 1981, pp. 66, 67, 86.
70. See Calm S. Gray, "Ballistic Missile Defense: A New llehate for n New Decxie," lrrtc'rr~atrot~~rl
.Sec.unty, forthcoming.
Managing Nuclear Multipolarity
JohnJ. Weltman
I
t appears probable that the next few decades will see the emergence of
many new nuclear weapons powers, especially in the third world. The
recent Israeli attack against Iraq's nuclear reactor demonstrated dramat-
ically the importance of attempting to manage an increasingly proliferated
world. Systematic analysis of the implications of this regime for the stabil-
ity of the central balance, and for stability in regional conflict situations,
has begun. What is required now is an assessment of the scope for arms
control measures designed to minimize the likelihood that such a process
will result in the weapon's being used. Without ceasing to think in terms of
measures t o discourage and retard decisions to acquire nuclear weapons
and to minimize the spread of capabilities to do so, we must think about
managing the effects of that measure of nuclear spread which it would be
imprudent t o assume we can prevent.
The Inevitable Spread of Nuclear Weapons
Why should we believe that nuclear proliferation is probable? What general

effects should we expect this process to have?' The significant spread of
nuclear weapons is probable over the next few decades because technological
and economic constraints against their acquisition have declined, and because
political and strategic incentives favoring the acquisition of nuclear weapons
have increased. There is furthermore little reason to expect that either of these
trends will not continue.
The spread of the civil atomic power industry means the spread of cap-
abilities and expertise necessary (but not sufficient) to the construction of
nuclear weapons. Since a good deal of the process of producing these
weapons can be carried on as a by-product of civil programs, the marginal
additional cost of the weapons declines as the civil industry expands. But
diversion of materials from a civil program is not the only route to a weapons
capability. There exists also a "direct" route, through the construction of
Source: International Security, 6 ( 3 )(1981-82): 182-94.
Wellmarl Managing Multipolarity 247
facilities designed expressly for a weapons program. New techniques for the
fabrication of weapons-grade material have made even this considerably
cheaper than in the decades immediately following the Second World War.
Recent reports concerning a weapons program in Pakistan which has reached
an advanced stage suggest that the "direct" route to the acquisition of nuclear
weapons no longer is beyond the economic capabilities of many states in the
third world.
Political trends in the contemporary world have implications parallel to
these technological and economic ones. One cannot but observe a relative
decline in the last few decades of the influence of superpower alliances as
determinants of the foreign policies of small states, as well as the emergence
of many new international actors in the wake of the post-war demise of
European colonial systems. It follows that, to an extent considerably greater
than in the immediate post-war years, foreign and defense policy in smaller
states is largely influenced by the idiosyncracies of their local or regional
situation. Decisions about the acquisition of weapons of any type are no dif-
ferent from other defense policy decisions. There is little reason to assume that
a "nuclear allergy" will be any more prevalent in the third world than it has
been among the governments of industrialized states.
To be sure, it is simplistic to merely divide states into those which are
nuclear weapons powers, and those which are not. To be useful in a military
sense, a nuclear device must be produced in some quantity; it must be so con-
structed as to be deliverable; and a state must possess the means to deliver it.
Furthermore, the military significance of a given nuclear capability depends
upon the military capabilities o f putative opponents. Finally, detonation of a
nuclear device may have little other than symbolic significance as an indicator
of military capability. A state may explode a device which cannot be delivered
against an opponent. Conversely a state may possess a usable military nuclear
capability without the necessity of first physically testing a device. Knowledge
of weapons technology is now sufficiently advanced to allow governments to
rely confidently upon computerized simulation techniques to indicate the like-
lihood that the products of a nuclear weapons construction program would in
fact explode when desired.
These considerations, however, do not affect the general point that the
economic, technological, and personnel requirements for the creation of a
nuclear weapons program with military significance in various regional con-
texts are widespread. Having a capability does not of course mean that a
state will or must employ it; but it does suggest that it would be imprudent
to assume that a government would normally deny itself that military where-
withal which it concludes might be useful to it in achieving its local foreign
policy objectives.
If capability and interest in a state coalesce to produce a favorable deci-
sion regarding the acquisition of nuclear weapons, the elaborate physical
safeguards structures that exist under both the NPT regime and bilateral
agreements should not be regarded as more than retardants. Superpower
sanctions - unilateral or collective - would no doubt be considerably more
effective, if undertaken. But to assume that the application of such sanc-

tions can occur in other than exceptional cases is to assume that the super-
powers would normally engage in self-denying behavior which presumes a
community of interest and action. Such behavior is not impossible. But history
suggests it is rare and will remain so, short of a radical change in the decen-
tralized nature of the system of states. Perhaps all may change if future pro-
liferation should result in some weapons' use in war for the first time since
1945. Such a transformation of behavior cannot be ruled out. But prudence
dictates that such presumptions should be treated as articles of faith, rather
than as conclusions derived from political analysis.
S u p e r p o w e r Foreign Policy Objectives vs. S t e m m i n g t h e Nuclear T i d e
There is a recent illustration of the difficulties which lie in the way of continu-
ing and effective superpower sanctions against the emergence of new nuclear
weapons powers. The superpowers apparently combined in August 1977
to dissuade the government of South Africa from proceeding with a test of
a nuclear device, after facilities for such a test were discovered by satellite
reconnaissance. Is this the harbinger of a superpower anti-nuclear condo-
minium? I think not. The exceptional nature of the South African experience
must be borne in mind. The superpowers were able t o act jointly in this case
because neither had cross-cutting political interests. South Africa is politically
isolated. The African policy of each superpower, independently arrived at, is
based on the attempt to maintain good relations with South Africa's black
neighbors. Neither saw any interest in jeopardizing those relations by sup-
porting South Africa. If, in fact, South Africa was preparing a test, and if
indeed she was dissuaded by a collective superpower demarche, the success
of the dissuasion was based upon a random coalescence of policy between the
superpowers, a coalescence that both antedated the proliferation question
and was far broader in scope.
In South Asia on the other hand, no such coalescence exists. Indeed, since
the Second World War, each of the superpowers has sought influence and
alliances in the region by supporting one side or another in regional conflicts.
Thus the superpowers have conflicting policies in the region and each is disin-
clined to pursue policies which might jeopardize its regional alignments. When
the question of nuclear proliferation arose in the region at the time of the
Indian "peaceful" nuclear explosion in 1974, these pre-existing patterns of
superpower conflict prevailed over any latent inclination the two might
have had to meet the problem of nuclear proliferation by common action.
Furthermore the individual efforts of each superpower were subverted by
the competitive condition which prevails between them. Each superpower
is constrained against imposing sanctions against a n ally, or allowing the
other superpower to so act, for fear of precipitating a realignment, or at
least of giving relative advantage to its superpower opponent.
Neither India nor Pakistan is diplomatically isolated; each possesses a meas-
ure of diplomatic flexibility which affords it leverage against its superpower
I I Managing Multipolarity 249
patron. For the superpowers to make anti-proliferation policy their first pri-
ority would require them to ignore the imperatives of their relationship with
each other. To complicate matters still further, what is required for the con-
trol of the spread of conventional arms often conflicts with what is required
to control nuclear spread.
Thus the initial American reaction to the Indian explosion was straight-
forward: a threat to cut off all nuclear supplies unless India accepted full-
scope safeguards. But it was soon apparent that such pressure would only
drive India further in the direction of an autarchic nuclear industry in which
she would pay even less deference to the NPT regime than she had paid hith-
erto, and that it would have the effect of making India's general relationship
with the Soviet Union even closer. While the strategic symbolism of the
Indian explosion may have been directed at China, the explosion must have
appeared a threat more concrete than symbolic from the Pakistani perspective.
Certainly that sense of threat, and the determination to develop a nuclear
capability in response, has been steady and uninterrupted in Islamabad, sur-
viving unchanged despite dramatic and violent domestic political upheaval.
The American response to intelligence reports that Pakistan was surrepti-
tiously accumulating the components of a uranium enrichment facility was
in the form of the carrot and the stick. Major conventional arms transfers,
if Islamabad would desist from her nuclear enterprise, was the carrot; a cut-off
of existing military assistance was the stick. But a carrot for Pakistan must
he perceived in New Delhi as a stick - albeit a small stick - with which to beat
India.
In any event, carrot and stick both have become submerged in the re-
energized "great game" between the superpowers, catalyzed by the events in
Afghanistan. The image now is of Washington arduously pressing conven-
tional arms upon a reluctant Islamabad. Far from imposing conditions, the
American concern now is to sweeten the deal sufficiently for it to be accepted.
Details will differ in other regions and in other years. But on balance we
should expect outcomes not dissimilar from these. Prevention of the spread
of nuclear arms will tend to conflict with prevention of conventional arms
proliferation. Both will give way to what Washington and Moscow each per-
ceive to be indicated in order to maintain relatively stable standing with what
Raymond Aron calls its "enemy-partner" in the central balance. Perhaps the
superpowers are mistaken in believing that their relative influence in the
third world significantly affects that balance. Nevertheless they continue to
believe it and to act on that belief. There is certainly no sign that this belief
grows weaker.
Who will Have t h e Bomb? To What Effect?
What of the effects to be anticipated from nuclear spread? Should we expect

that nuclear weapons will be used in war? H o w will nuclear spread affect
the strategic balance between the United States and the Soviet Union?
As for the last of these questions, much depends upon the attributes of
the new nuclear weapons states. For at least until the end of this century, the
stability of the central strategic balance will remain substantially unaffected
by any force which could be deployed by most of those states commonly
included in lists of potential "Nth countries." There are a few (East or West
Germany, Japan, a United Europe, and possibly Brazil) whose acquisition of
nuclear weapons could affect the stability of the central balance. But if any
of these states should in fact deploy what it is capable of producing, the
effect upon the central balance will still be a function of the political context
within which that deployment occurred, as much as it will be a function of
the nuclear "hardware" taken by itself.
With these exceptions, nuclear acquisition by others can represent no
more than strategic "pinpricks" to the superpowers. No force which any of
the remaining states would be capable of building during the foreseeable
future could threaten the survivability of more than a small fraction of super-
power strategic forces. Against any of these "Nth" powers, a superpower
could threaten an overwhelming and certain countervalue response, which
could be mounted by a fragment of its forces small enough not to affect
materially its strategic posture against the other superpower. There is no
reason, furthermore, not to expect that the superpowers will continue
to retain a wide qualitative lead over these states in all matters related to
strategic technology.
This is not to suggest that the proliferation of small and relatively primi-
tive nuclear forces will create no dangers for superpower governments and
for their populaces. These dangers do not, however, include the gravest:
instability which might make central war more probable. Catalytic war is
not a major danger, but government-sponsored terrorism or blackmail might
be. Insofar as the latter might become a problem, its resolution is a matter
of enhanced physical security and improved intelligence procedures, activ-
ities which are beyond the scope of this essay. Moreover, acts of terrorism or
blackmail involving complicity (of omission or commission) by governments
or territorially based quasi-governmental organizations, can be effectively
countered or deterred by threat of retaliation if the responsible agent can
be identified. Thus the dangers represented by these acts in a nuclearized
international system should not be overestimated.
This guarded optimism must be severely qualified if we turn to the impli-
cations of nuclear weapons acquisition by the "exceptions" mentioned above.
These "middle nuclear powers" are chosen because of their substantial eco-
nomic capabilities, and their potential to engage in "state of the art" research,
development, and production. If one or more of these powers should deploy
a force which approaches its capabilities, such a force could no longer be con-
sidered merely a "pinprick," from the superpower perspective. In such a con-
text the problem of superpower vulnerability might grow somewhat more
difficult with resultant complications for the management of the strategic
relationship between the superpowers. Third states by themselves would
not be able to threaten the survival of superpower forces; but they might
LVeIt r-ii,iii Managing Multipolarity 25 1
present a threat to a sufficient fraction of those forces to adversely affect

their survivability in a material fashion in the event of a subsequent exchange
between the superpowers themselves. To deal with this possibility, super-
power planners might feel compelled to request additional deployments.
"Survivability" as a criterion for strategic nuclear force acquisition - and
potentially as a mutually acceptable and salient benchmark for arms control -
would grow increasingly ambiguous.
The precise extent of this ambiguity would depend upon the identity and
intentions of the Nth powers, and the character of their forces. It is clear,
however, that some increased incalculability in a number of areas would
follow. Planners in each superpower would experience increased difficulty
in determining the material requirements of a deterrent posture and in
distinguishing between deterrent and other effects of deployments. Even if
we make the large assumption that deterrence is, and will continue to be,
the principal doctrinal premise of superpower force posture decisions, the
scope for honest disagreement about its requirements will be enlarged. It
will become more difficult to fix upon formulae from which the details of
arms control agreements, or informal understandings, might be deduced,
even if there is a will on both sides to come to such understandings. In add-
ition to these problems of bilateral agreement between the superpowers in
such an environment, it goes without saying that attempts to reach multi-
lateral accommodation become more difficult the greater the number of
parties who must be involved. Finally crises will become less predictable,
and thus more difficult to manage, as the number of actors whose behavior
must be anticipated rises.
It would be superficial, however, to attribute this incalculability primarily
to the effects of the spread of the weapons themselves. The root causes of the
incalculability we should expect to be a product of such a multipolar nuclear
environment are political, rather than technological. Whatever possihilities
may exist for minimizing this incalculability must therefore lie in the political
realm also.
Middle-power nuclear forces are, after all, not novelties. Nuclear multi-
polarity in this sense has characterized the international system since the
emergence o f the "independent" British deterrent two decades ago, followed
of course by the French and, more recently, by the Chinese. Yet these forces
have not in any demonstrable way adversely affected the stability of the super-
power strategic relationship. One may attribute this to the smallness of scale
of the middle-power forces actually deployed, as well as to the qualitative
margin that the superpowers have maintained over them.
Proliferation anio
To argue in this manner, however, is to ignore the crucial importance of the

political context within which these forces were acquired and, since acqui-
sition, have been deployed. Beneath the rhetoric of "omnidimensionality,"
there has been little significant doubt about the direction in which these
weapons were and are pointed. Nuclear weapons were introduced into a pre-
existing pattern of alignments and conflicts, a pattern which the weapons'
introduction did not change. In order to predict the effects of middle-power
nuclear weapons acquisition in the future, we must similarly examine political
context, and not merely technological capability.
What developments are likely to induce further middle powers to go
nuclear? Brief examination is sufficient to suggest that further proliferation
at this level is likely only as a response t o prior, and major, change in the
pattern of alignments that has characterized the international system since
the Second World War. Thus any German resort to nuclear weapons must first
assume the breakdown in some fashion or other of the division of Europe laid
down at Yalta and Potsdam. Similarly, a nuclear Japan implies the prior col-
lapse of a security system in the Pacific based upon American hegemony.
If such dramatic changes were to occur, the incalculability which would
result would render formal or informal attempts at arms control futile, until
such time as new and firm political patterns might emerge. It does not fol-
low, however, that the likelihood of major war must increase, unless we
assume a technology which must produce a world of universal vulnerability.
Far more likely is a situation in which volatility in arms races is accom-
panied by political caution. To the extent that policy is capable of advancing
or retarding the emergence of such an environment it must be directed
toward preserving the utility of existing alignment patterns in the eyes of
potential nuclear powers.
Proliferation in the Third World
The effects and potential manageability of nuclear acquisition by states

below the middle-power level are problems of greater immediacy, yet of
lesser danger and greater potential susceptibility to manipulation through
prudent policymaking. That this is an immediate problem the newspapers
demonstrate each morning. What the dangers to be anticipated from this
mode of nuclear spread are, and what they are not, should be made clear.
Instability in the central strategic balance is not a danger here, nor is increased
danger of local conflicts' escalating to nuclear conflict between superpowers
drawn in by their smaller allies. What level of potential there may be for
this sort of escalation has been present for some time, is recognized by
superpower planners, and is indeed commonly manipulated by them as a
means of gaining relative advantage. It is not obvious why the presence of
local nuclear weapons should make this potential greater. The increased dan-
gers to be anticipated as a function of nuclear proliferation are local dangers
for local actors, namely the danger that these weapons might actually be
employed in indigenous conflicts (or might otherwise be the cause of violence).
Even a t this level we should not easily assume that the future holds noth-
ing but a series of local catastrophes. To assume this is to assume that third
world states will prove unable in their own interests t o deduce from their
I 1 Managing Multipolarity 253
situations the requirements of prudent behavior, as have those states in the

ind~lstrializedworld which have existed for many years now in a strategic
environment involving nuclear weapons.
From the broadest perspective, one may thus argue that nuclearization of
regional conflict patterns in the third world will tend to be self-regulating.
Given enough time, mutual deterrent postures will tend to develop. A state
which appears to contemplate the acquisition of nuclear weapons will find
itself quickly emulated by its local adversaries. The concern of each to
preserve the survjvahility of his forces will follow. This will result eventu-
ally in a situation in which neither adversary can have sufficient certainty
of its capacity to annihilate opposition
.
~
forces to balance even the residual
risk of the horrendous consequences which would follow unless total suc-
cess is achieved in a first strike. No doubt there are some regions where short
distances, weather, and topography combine with economic and industrial
constraints to suggest that local powers will never be able to achieve levels
of survivability relative to one another approaching those achieved by the
superpowers. These conditions would certainly appear to apply, for example,
in the Middle East. But high population densities in small areas also char-
acterize most countries in the region. Failure to eliminate even a single deliver-
able weapon would thus be to risk catastrophe; and short distances mean
that no great sophistication in means of delivery is required for a successful
countervalue response.
A prudent power in a local conflict situation may well attempt to put off
for a while the day when it might be faced by a local nuclear-armed oppon-
ent, if it can do so successfully by conventional means. This appears, for
example, to he Israel's policy. Elimination - even for a few years - of an
opponent's capability to wreak havoc upon one's society carries greater cer-
tainty than does reliance upon attempts to deter him from using an existing
capability, as long as the former can be successfully accomplished. But it
would be unwise to assume that such efforts to eliminate potential nuclear
capabilities can succeed indefinitely. Measures of defense or concealment
that an opponent may undertake may allow him to develop his nuclear
capability past the point where it could be eliminated without risk of major
radioactive contamination. Furthermore, political constraints against repe-
titions of actions like the Israeli raid on the Osirak reactor are likely to be
intense, in spite of the protestations of those imposing the constraints that
prevention of further nuclear weapons proliferation remains their goal.
While elimination may be the preferred policy option for such a power,
reliance on deterrence may thus come in time to be the only viable option.
But what if nuclear weapons get into the hands of a government led by a
madman? This is a common objection to any argument that further nuclear
proliferation need not lead inexorably to universal disaster. In answer to it
one must take care to consider the character of the attitudes and behavior we
choose to call "n~adness."There is a common list of figures - contemporary
and historical - who function for us as examples of "madmen" in politics.
But we must understand that their madness consists in the extremity of their
intentions and rhetoric, not in any lack of capacity for prudence in the choice
of means. At the very end, Hitler may have been determined to involve his
nation in his personal suicide. But prior to the war there is every evidence to
suggest he was prepared to moderate his actions in the event other powers
should threaten him with dire consequences.
It is the somber virtue of nuclear weapons that they reduce consequences
to a starkly simple level. We should not expect nuclear weapons to moderate
the intentions of fanatical individuals with their hands upon the instruments
of mass destruction; but equally we should not expect them to use those means
to achieve their ends unless they conclude that they can use them witho l t dire
consequence to themselves.
It would be ingenuous, however, to suggest that an "invisible hanc" will
infallibly operate to prevent the use of nuclear weapons in local wars. If we
are optimistic enough to assume that political analysis is capable of making
predictions at all, such predictions must take the form of broad statements
of tendencies, rather than precise forecasts of particular events. All we can
say with certainty here is that over time, the pacific trends described he -e will
tend to operate. But even one event which runs counter to the trend will be
a disaster beyond measure for those involved. Hostilities involving nuclear
weapons may occur prior to the lapse of enough time for a mutually stable
weapons posture to develop. The problem for policy thus becomes one of
minimizing the probability that such adverse events might occur.
Managing Regional Nuclear Balances
What measures might be employed to reduce the likelihood that n ~ ~ c l e a r

weapons might be employed in regional conflict, once they have been intro-
duced into an area, or once their introduction appears imminent? We s ~ o u l d
be aware of the difficulties under which such a discussion must labor. To a
considerable extent regional
- conflict patterns evolve in an autonomous man-
ner, responding to indigenous influences rather than superpower pressures
and interests. Thus the capacity for external agents to influence evenl-s, no
matter how well conceived their policy, must be limited. Furthermore, regional
political and strategic patterns differ considerably from one another, and the
precise circumstances under which nuclear weapons introduction might , ~ c c u r
are legion. Prudent policy must be responsive to detailed circumstances more
than to general notions deduced in advance; but some broad guidelins do
suggest themselves.
Before surveying the measures which outsiders, including superpowers,
might consider with a view to minimizing nuclear instability in regional con-
flicts, we should explore a little further the character of the possible dargers.
Exact simultaneity in the acquisition of nuclear weapons is unlikely, as is
qualitative symmetry once acquired. Thus a period of time will exist in
which one side can observe that it could attack with impunity. The problem
LL eltni,in Managing Multipolarity 255
is to reduce the likelihood that such an observation will be followed by

nuclear weapons use.
At first glance, a number of measures seem to lie at the disposal of outside
agents with a view to achieving this end. The superpowers, individually and
collectively, might attempt to bridge the period of greatest danger by threat-
ening to retaliate with nuclear weapons against any local first user of nuclear
weapons. Local arms control agreements might be pressed, perhaps under
some outside aegis. Finally outside agents might favor the control of local
nuclear arms by facilitating the emergence of a local weapons environment
which itself minimizes propensity for first use.
If credible, superpower threats of retaliation against first use would
doubtless be effective; but can in fact such threats be credibly made? Both
individual and collective action of this sort would be subject to the same dif-
ficulties we have already observed regarding superpower action to prevent
proliferation in the first place. Such action might be successful in excep-
tional cases against states already diplomatically isolated. In most cases,
however, we should not expect the superpowers to consider action of this
sort unless we first assume a radical devaluation by them of their alignment
and influence relationships in the third world.
Formal regional nuclear arms control agreements might in time come
about. N o doubt the assistance of outside agents might facilitate the process.
However, even under the most optimistic assumptions about the capacity of
intensely hostile parties to agree (and to surmount the technical difficulties
attendant upon such agreements, whatever the attitudes of the parties), for-
mal arms control agreements are unlikely to be concluded until long after the
onset of the period of greatest danger. The possibility of such agreements
should not be ignored. Attention should be given to problems of concluding
them and to the details of their content. These agreements cannot, however,
take priority among measures to ensure regional nuclear stability. Arms con-
trol negotiations - successful or not - have never been anything but time-
consuming and tortuous. In the circumstances we are considering here, they
will inevitably come too late to meet immediate and acute difficulties.
What measures might the superpowers or other concerned states em-
ploy to encourage the quick emergence of stable regional military nuclear
balances? These outside agents would be governments already possessing
nuclear weapons, principally the superpowers. The greatest danger that
nuclear weapons might be employed in wars flows from situations in which
that use could be contemplated without risk of retaliation. It follows that the
task for agents outside a region in which nuclear weapons are about to be
introduced is to prevent such a situation from emerging or to reduce as
much as possible the time period during which it might prevail. The means
to accomplish this aim is the timely transfer of strategic technology.
Depending upon individual circumstances, this technology might include
survivable command and control systems, delivery vehicles and warheads,
or information about these systems. The object is to make such transfers as
will ensure that no power in the region might expect that it could employ
nuclear weapons without a substantial risk of a response in kind from

another power in its locale.
Such measures involve serious risk, and make substantial requirements on
decision-making processes if they are to be effective. The risk is simply that the
known disposition of the superpowers to engage in such transfers might
encourage some states which might not otherwise d o so, or which might
not do so as quickly, to acquire nuclear weapons. The important economic
constraints which now apply in many states against acquisition would in these
circumstances be vitiated. A state might no longer feel that a decision to
acquire the weapons must involve the creation of a costly, effective, and sur-
vivable force. It might now conclude that the acquisition - or appearance of
acquisition - of even the most primitive and vulnerable device and delivery
system, would suffice to trigger the outside supply of a more sophisticated
system, gratis.
Yet this risk must be balanced against the high risk of local nuclear hos-
tilities flowing from indigenously motivated nuclear weapons acquisition.
There is of course no uniform answer in advance to the question whether
technology transfers or restraint in a given situation would be the less risky
course. That choice can only be made in the light of detailed circumstances.
But an effective choice requires the capacity to be both timely and flexible
in the execution of policy. It also requires an intelligence capability of great
accuracy and acuity, directed both to the capabilities and to the intentions
of potential new nuclear weapons states.
Conclusion
Substantial further spread of nuclear weapons is an eventuality which

should be anticipated in spite of existing efforts to prevent it. The dangers
to be expected from this are not of the horrendous character they have been
painted to be by some. But dangers nevertheless do exist and measures to
control the effects of the spread of nuclear arms in the third world should
now be considered, bearing in mind the limits on the effectiveness of any
such measures.
What lessons should we have learned from the efforts made to control
nuclear arms over the last few decades? We should have learned that dis-
armament and arms control are not necessarily congruent with one another,
and may often be inconsistent objectives. We should also have learned that
the character of the weapons themselves may be more important than for-
mal international agreement in facilitating the control of the effects of arms.
Weapons that constrain governments to restraint in their use as a matter of
self-interest have proven more effective than treaties that attempt to create
common interests in reducing the likelihood of weapons use and the war-
bearing potential of crises. Why should we not assume that similar lessons
will apply to the problem of managing the effects of the spread of nuclear
arms in the third world?
it 11 i Managing Multipolarity 257
Author's N o t e
An earl) version of t h ~ essay

s was prepared while the author was a guest of the Lkpartmenr of
Political Sc~ence,T h e Colorado College. H e is grateful t o the rnembers mid staff of the
Lkplrt~nerrt,a n d for t h e ~ rsupport a n d generous hospitality. Special thanks are d u e t o Roherr
1). L.oev), Timothy Fuller, William Stivers, a n d Helen 1-ynch.
Note
I. I have examined these questions at length elsewhere a n d here shall only deal with them
In summary form. See the author's "Nuclear D e v o l u t ~ o na n d World Order," World l'olztics,
Vol. XXXII, N o . 2 (janu,lry 19801, pp. 169-1 93.
Common Security: A Programme for Disarmament -
The Report of the Independent Commission
on Disarmament and Security Issues
Chapter 1 - Common Survival
L
ess than two generations after the carnage of the Second World War,
the world seems to be marching towards the brink of a new abyss,
towards conflicts whose consequences would exceed experience and
defy imagination. Having survived the tragedies of two global wars in this
century, wars that touched virtually all nations, leaving tens of millions
dead, hundreds of millions wounded or homeless, and a whole continent in
shambles, mankind might have embraced new means of organizing the
international community, means that could prevent such catastrophes in the
future. Indeed, important efforts have been made towards this end, but in
1982, nearly four decades after the Second World War, the inescapable con-
clusion is that these efforts have not yet succeeded.
Humanity has made only limited progress towards the limitation of
nuclear and conventional weapons and has not taken even halting steps
towards disarmament. Arms races between the great nuclear powers and
between rivals in particular regions have continued for decades and now
seem to be accelerating. Every year has brought advances in the technology
of warfare; developments which mean that future wars would be more
destructive and inhumane. Every year has witnessed the spread of advanced
military technologies to more nations. Every year has seen new examples of
the suffering such weapons can cause; new demonstrations of man's appar-
ently limitless capability to inflict pain and destruction on his neighbours,
even his countrymen. And, most chilling, every year has uncovered new evi-
dence that humanity may eventually confront the greatest danger of all -
worldwide nuclear war.
It is long past the time for men and women to halt these trends. The dan-
gers are far too great to be ignored. Decisive action must be taken now to
Source: Common Security: A Programme for Disarmament - The Report of the Independent
Commission on Disarmament and Security Issues (London: Pan Books, 1982), Chapter 1,
'Common Survival' pp. 1-13, Chapter 6 , 'Recommendations and Proposals' pp. 138-76.
Disarmament and Security Issues 259
halt and reverse the spiral of the arms race and the deterioration of political
relations. and to reduce the risks of conventional and nuclear wars.
Arms a n d Insecurity
Nuclear weapons are awesome instruments of war. Modern technology has

radically transformed both the likely character and the potential stakes of
modern warfare. Weapons with intercontinental ranges, with flight-times
measured in minutes, and previously unimagined explosive power, can
destroy in seconds what it has taken centuries to create. Both the United
States and the Soviet Union possess thousands of warheads in their strategic
and intermediate-range nuclear forces, every one of which is more power-
ful than the atomic bombs dropped on Hiroshima and Nagasaki. Even
these thousands of weapons do not begin to exhaust the nuclear arsenals of
the two sides: additional thousands of shorter-range nuclear weapons, so-
called tactical battlefield systems, are deployed and ready to be used.
Together with the two great powers, three more states - China, France,
and the United Kingdom - maintain smaller but, by traditional standards,
powerful nuclear arsenals. As many as ten additional nations may be in a
position to acquire nuclear weapons relatively quickly should they choose
to d o so; one or two may already have covert stores of nuclear explosives.
Nor has the technological revolution ignored the non-nuclear, or so-
called 'conventional', weapons of war. Technology has greatly augmented
the lethal and destructive potential of all military operations - large and
small, regardless of whether they involve the great powers or not. Today,
modern jet fighters armed with air-to-air missiles are nearly as common in
Africa and Asia as in North America and Europe. Patrol boats with anti-
ship missiles are seen in the Gulf of Iran and the Caribbean as well as the
Norwegian Sea and the Mediterranean. And modern tanks in huge numbers
already have fought in the sands of the Middle East and North Africa.
Together, the nations of the world spend the equivalent of about $650
billion on their armed forces each year, more than one twentieth of their
total annual incomes. Three quarters of this huge sum is accounted for by
the industrial countries, but military expenditure by developing states is far
from trivial, and is growing rapidly.
The persistence of wars and armaments, the dreadful spiral of political and
military tensions, and the danger of nuclear holocaust all reflect the weak-
nesses and limitations of the international political system in which we live.
The hopes expressed in 1945 for a world order in which the United
Nations would be the guarantor of international peace and act as protector
of states against aggression recede further by the year. Instead, we live in a
milieu in which each state feels obliged to display its willingness to wage war
in defence of what it regards as vital national interests. Military strength is
seen as a symbol of this resolve, but the continuing expansion of national
arsenals is in turn interpreted by other nations as evidence of hostile intent,
a cycle which undermines the security of the international community as a

whole.
This is the international scene that the developing nations, mostly newly
independent countries, have entered in the postwar era. For the greater part
they have absorbed its environment, adopted its style and, in the process,
strengthened its imprint on human affairs. But there are important respects
in which the contradictions, dilemmas, and paradoxes are even greater for
the Third World than for the industrialized countries.
The sacrifice which militarization imposes on the Third World is of a
qualitatively different order from that which falls on richer countries. In a
developing nation the decision to add a battalion or buy a warship consti-
tutes more than a mere budgetary choice; it often results in increased
human deprivation for the poorer members of that society. In this situation
the problem for most of the developing countries is not so much one of dis-
armament as one of avoiding total absorption into the prevailing military
culture and of finding security through other means, in particular by con-
tributing t o an effective system of international security in which the bur-
den of making the world safe for all will be shared by all. For these nations
as for the rest of the international community, a return to the vision of the
UN Charter is not remote idealism but an urgent practical necessity.
The problems of peace and disarmament are thus also the problems of
international order. As long as the community of nations lacks a structure
of laws backed by a central authority with power and legitimacy to enforce
these laws, then nations are likely to continue to arm, in most cases for
legitimate reasons of self-defence, but in others to gain unilateral advantage.
Armaments are not the only cause of international conflict, and are often
its symptom. Nevertheless, frequently armaments are acquired because of
the erroneous assumption that security can somehow be achieved at the
expense of others.
What is National Security?
Traditionally, the concept of national security has been taken to refer to

both physical and psychological security, which may be subject to threats
from both internal and external sources. Clearly, a secure nation is one that
is free from both the fact and the threat of military attack and occupation,
that preserves the health and safety of its citizens, and generally advances
their economic well-being. There are also less tangible dimensions to secur-
ity. Citizens of all nations want to be able to remain true to the principles
and ideals upon which their country was founded, free to chart futures in a
manner of their own choosing. National security also has an international
dimension. It means that the international system must be capable of peace-
ful and orderly change, and open for the exchange of ideas, trade, travel,
and intercultural experience.
As we have noted, the perceived requirements of national security dictate
that nations maintain military forces adequate to the dangers posed to their
Disarmament and Security Issues 26 1
security - dangers from within and without. But the realities are such that
military strength alone cannot provide real security. By every index of mili-
tary strength it is evident that most nations have become more powerful over
the years. Yet, judged by the increasingly strident tone of international and
domestic debates about these issues, it is also clear that greater national
military might has not led to a greater sense of national security.
The growth of the peace and anti-nuclear movements in Europe and
North America is instructive. These movements gathered strength at pre-
cisely the time when many governments were stressing the need for security
through expanded nuclear weapon programmes.
The Impact of Technology
Technology has changed the world in which we live, but understanding of

its impact on international relations has not kept pace. National boundaries
are no longer, if they were ever, impervious shields, the penetration of which
could be prevented by military forces. Populations cannot huddle behind
national borders, build up armed forces, and cut themselves off from the
rest of the world in order to live securely. In part, this is because of the great
economic interdependence of the international community and the ways
that modern comn~unicationsand transportation are binding us together as
a global audience to all events. More pointedly our interdependence reflects
the crucial technological fact of the contemporary age: there are no effect-
ive defences against missiles armed with nuclear warheads; none exist n o w
and none are likely to he developed in the foreseeable future.
N o matter how many weapons a nation adds to its arsenal, it cannot
directly diminish its vulnerability. N o known technology, provides, even
potentially, a means for the effective and reliable defence of a people from
nuclear attack. Thus, one central irony that must be faced is that no matter
what ~ ~ n i l a t e rchoices
al a nation makes in pursuit of security, it will remain
vulnerable to nuclear attack and thus ultimately insecure.
Technology imposes other costs as well. The advanced technologies
incorporated in modern weapons mean that the domestic burdens of arma-
ments are great - not just the use of enormous sums of money, but the diver-
sion of scarce resources, particularly highly skilled individuals and also
materials, from solving social problems. Thus, a second irony is that the
more we strive for security from external threats by building up armed
forces, the more vulnerable we become to the internal threats of economic
failure and social disruption.
Both paradoxes suggest that neither physical nor psychological security
can be achieved without the development of an international system which
would outlaw war and seek the elimination of armaments through their
gradual but substantial reduction. This does not mean an international
order wedded to the status quo. Progress towards economic and social
development, the alleviation of political injustices, and the furtherance of
human rights must continue. But when nations resort to arms, international
society must isolate the conflict and resolve it by peaceful means. Only in
such a world will people be able to feel a true sense of national security.
Consequently, if the world is to approach even the possibility of achieving
true security - ending the danger of nuclear war, reducing the frequency and
destructiveness of conventional conflicts, easing the social and economic bur-
dens of armaments - important changes are necessary in the way that nations
look at questions of armaments and security. Most important, countries must
recognize that in the nuclear age, nations cannot achieve security at each
other's expense. Only through cooperative efforts and policies of interlocking
national restraint will all the world's citizens be able to live without fear of
war and devastation, and with the hope of a secure and prosperous future for
their children and later generations.
Achieving Common Security
All nations would be united in destruction if nuclear war were to occur.

Recognition of this interdependence means that nations must begin to organ-
ize their security policies in cooperation with one another. Obviously, this will
not happen overnight. But a political process can be started which - if care-
fully managed and consistently pursued - can develop sufficient momentum
to outrun the effects of past failures.
In view of the current global distribution of economic resources and techno-
logical potential, to say nothing of military capabilities, implementation of a
worldwide policy of common security must begin with relations between the
United States and the Soviet Union, and between the two major alliances -
NATO and the Warsaw Pact. But the developing world is neither immune to
the consequences of East-West conflict nor is it without fault as a contribu-
tor to the risk of war. Increasingly, political tensions between East and West
affect the developing world, aggravating conflicts between local nations in
particular regions. But in some instances, developing nations have played a
less passive role, seeking out the political and diplomatic support of one of
the great powers, or aiming to gain its economic or military assistance.
The costs and dangers of this involvement are familiar. Competitive
arms purchases by developing nations result in the diversion of scarce
resources from the requirements of economic development to the military
sector. In turn, the contrast between popular expectations for economic
growth and improvement in the quality of everyday lives, and the reality of
the slow pace of economic development, feeds dissatisfaction, resulting at
times in domestic upheaval, and at other times in pressure to divert internal
unrest and criticism to external enemies. Moreover, involvement of the
great powers on opposing sides of these regional conflicts can sometimes
result in a dangerous escalation, the end result of which is unpredictable.
The avoidance of war, particularly nuclear war, is thus a common respon-
sibility. The security - even the existence - of the nations of the world is inter-
dependent. For both East and West, the avoidance of nuclear catastrophe
depends on mutual recognition of the need for peaceful relations, national

restraint, and amelioration of the armaments competition. But, if East-West
relations are to be stabilized and sustained, then regional conflicts in the
developing world also must be resolved - or at least their eruption into open
conflict avoided - and the opportunities for competitive great-power involve-
ment thus reduced. In a deeper sense, international security also depends on
the easing of the present sharp differences in the basic conditions of life in the
different parts of the world.
In their quest for security, nations must strive for objectives more ambi-
tious than stability, the goal of the present system in which security is based
on armaments. For stability based on armaments cannot be sustained indef-
initely. There is always the danger that the fragile stability of an inter-
national system based on armaments will suddenly crumble, and that
nuclear confrontation will take its place. A more effective way to ensure
security is to create positive processes that can lead to peace and disarma-
ment. It is essential to create an irreversible process, with a momentum such
that all nations cooperate for their common survival.
Acceptance of common security as the organizing principle for efforts to
reduce the risk of war, limit arms, and move towards disarmament means,
in principle, that cooperation will replace confrontation in resolving con-
flicts of interest. This is not to say that differences among nations should be
expected to disappear - given the ideological differences between East and
West no meaningful convergence can be expected. Similarly, the problems
between North and South, rooted in years of oppression and the stark dif-
ferences in the economic circumstances of the two hemispheres, cannot be
expected to be solved overnight; nor can the many regional and intra-
national conflicts through the world. The task is only to ensure that these
conflicts do not come to be expressed in acts of war, or in preparations for
war. It means that nations must come to understand that the maintenance
of world peace must he given a higher priority than the assertion of their
own ideological or political positions.
Principles of Common Security
To accompl~shthese objectives, all countries should adopt the following

principles as the basis for their security policies.
egitimate R i g h t
A secure existence, free from physical and psychological threats to life and
limb, is one of the most elementary desires of humanity. It is the funda-
mental reason why human beings choose to organize nation states, sacrific-
ing certain individual freedoms for the common good - security. It is a right
shared by all - regardless of where they live, regardless of their ideological
or political convictions.
Military Force is not a Legitimate Instrument for

Resolving Disputes between Nations
The adage that violence begets violence is as true for relations between
countries as it is for relations between individuals. Historically, the use of
force as an instrument of national policy has only rarely been effective over
the long run. In the nuclear age, it raises risks which are disproportionate
to any conceivable gain. Too often, the use of force is claimed to be in self-
defence. Prevailing definitions of self-defence must be tightened and nar-
rowed. Renewed renunciation of force as an instrument of national policy
is an important element in a policy of common security. Nevertheless, all
states must retain the right t o use force in their own defence and, in accord
with the conditions and procedures specified in the Charter of the United
Nations, in collective defence of victims of aggression.
Restraint is Necessary in Expressions of National Policy
The urge of nations to win advantage over others, to gain security at each
other's expense, is the engine that drives the competitive acquisition of
armament and pushes the world towards nuclear war. It reflects the false
premise that security can somehow be gained unilaterally. Consequently,
policies which seek advantage - either through the accumulation of arma-
ments, or by bargaining in negotiations for unilateral gain, or, most dan-
gerously, by the exercise of military power - should be renounced. Restraint
should be the watchword of all states: restraint, out of respect for the right
of others t o security, but also in selfish recognition that security can be
attained only by common action.
Security cannot b e Attained through Military Superiority
The renunciation of unilateral advantage includes acceptance that any suc-

cessful effort to reduce armaments and the risk of war would have to be
based on the renunciation of military superiority and, more generally, of
threatening military postures. This would include the objective of estab-
lishing parity between the major military blocs, as well as establishing it as
a guiding principle for several pairs of rivals, or groups of rivals, in other
specific regions on a flexible basis. Parity must take into account geographic
and strategic circumstances and allow for the disparate histories and mili-
tary traditions that lead nations to place varying emphases on different
kinds of military force; adversaries should not be expected to have armed
forces that mirror one another in all aspects. It must also be recognized that
parity is as much a perceptual as an objective phenomenon. The basic aim
must be t o establish security at the lowest possible level of armaments.
Negotiations could aid greatly in the establishment of these conditions and
could help to avoid the suspicion that one side or the other might threaten
to ignore parity once it had been established.
With parity and the absence of threats established as guiding principles for
military relationships, it is equally important that the nations of the world
act in concert to reduce armaments substantially. In making such reduc-
tions, particular attention should be paid to those types of weapons which
raise the greatest concern on either side, as these carry the greatest danger
of leading to war. The larger military powers must ass~lmethe major
responsibility for initiating and sustaining efforts to reduce armaments, but
all nations should share in, and would benefit from, significant progress
towards this end. The benefits of reducing armaments in terms of allevi-
ating the economic and social burdens of the arms race are obvious. Of even
greater importance would be the creation of a political atmosphere in which
peaceful relations among nations could flourish, and in which there would
be a lesser risk of war.
Disarmament efforts d o not move forward in a political vacuum. They must

reflect political interests and the political order and are thus an integral part
of international relations. However, it is important not to construct, as a
matter of deliberate policy, linkages between particular negotiations to limit
specific aspects of the arms race and international behaviour in general. The
task of diplomacy is to split and subdivide conflicts rather than ,generalize
and aggregate them. Linking them into broader issues tends to limit, rather
than broaden, the scope for diplomatic manoeuvre. Progress in arms nego-
tiations is not a reward for either negotiating partner; it is a means for both
to capitalize on their common interest in security and survival.
At the same time it must be recognized that significant movement towards
disarmament will proceed only with difficulty in the absence of broader polit-
ical accomn~odation.The two interact and must move together. They can aid
one another in facilitating progress, but neither can proceed very far without
progress in the other. Just as arms negotiations would fail in the absence of
political accommodation, so too would movement towards more cooperative
political and economic relations come to an end without concurrent progress
towards stabilization of the military balance and reductions in the size of
armed forces.
Third World T e n s i o n s
The Third World has been the scene of most of the world's violence since
1945. The cost of this upheaval and destruction has been tremendous.
There are many causes of Third World conflict. For most of the postwar
266 T h e Cold War a n d N u c l e a r D e t e r r e n c e
period, turmoil in developing regions was the result of the struggle for
independence. But even now, when there are virtually no colonies left, many
sources of tension and potential conflict remain.
In the absence of a natural basis for the borders for many Third World
nations, territorial claims and pressures for the fragmentation of national
societies have been frequent and sometimes intense. In many developing
nations, historic animosities, religious and racial hatreds and battles for
political influence and privilege among disparate elements of society all lead
to violent conflict.
Last, but far from least, pressures stemming from economic underdevel-
opment and the maldistribution of resources and wealth produce stresses
and strains both within and between nations. Hunger, malnutrition,
poverty and ill-health on a massive scale all work to spur political change,
sometimes through violent means.
The developing regions are fragmented and torn by a variety of indigen-
ous conflicts, but many of them have been complicated by the superimpos-
ition of East-West tensions. As these tensions rise there is increased risk of
their being transferred to Third World regions where indigenous conflicts
provide opportunities for them to flourish. Conversely, regional conflicts
can themselves lead to wider escalation of tension involving the danger of
great power confrontation. The Third World has a deep and continuing
interest in de'tente, in curbing the arms race, and in improved relations
between the great powers.
Finally, we should note the broader tension between the industrialized
nations and the developing world. Politically, ideologically and econom-
ically, the North-South dialogue is frozen. The growing economic and social
disparities between North and South have been catalogued frequently, most
recently in the report of the Brandt Commission: North-South: A pro-
gramme for survival. A failure to rectify these trends could lead in time to
worldwide chaos and international conflict. For the present, North-South
tensions are mainly of an economic nature, damaging the development
prospects of the Third World and also making it impossible to implement
long-term economic arrangements that could provide greater prosperity for
all. But potentially much more is at stake.
Over the long term, a decline in North-South relations can have the
most serious impact on the psychological atmosphere in which we all must
live, on the basic fabric of international politics, and on the risk of war.
Common Dangers a n d Common Security
In the modern age, security cannot be obtained unilaterally. Economically,

politically, culturally, and - most important - militarily, we live in an increas-
ingly interdependent world. The security of one nation cannot be bought at
the expense of others. The danger of nuclear war alone assures the validity of
this proposition. But the obvious economic and political inter-relationships
between different nations and different parts of the world strongly reinforce
the point. Peace cannot be obtained through military confrontation. It must
be sought through a tireless process of negotiation, rapprochement, and nor-
malization, with the goal of removing mutual suspicion and fear. We face
common dangers and thus must also promote our security in common.
The destructive power of modern nuclear and conventional weapons,
both in quantity and quality, has totally outrun traditional concepts of war
and defence. In the event of a major world war, which would escalate inex-
orably to the use of nuclear weapons, all nations would suffer devastation
to a degree that would make 'victory' a meaningless word. The only realis-
tic way to avoid such a catastrophe is to quickly develop a process by which
progress towards disarmament is made rapidly, and to establish a system of
political and economic cooperation among nations such that all gain an
Important and equitable stake in its contmuance.
In sense, the truth of these statements seems already to have been rec-
o g n l ~ e dby people throughout the world. We are greatly encouraged that ,is
the C o m m ~ s s ~ ohas
n met and worked there has been a vlrtual explosion of
popular sentiment In favour of peace and d~sdrmament.It 1s long past tlme
for all governments to respond to the popular urge for true security. If they
fail to meet these expectations, we will all be the victlms of t h e ~ rfolly.
Chapter 6 - Recommendations a n d Proposals
A New Departure
We are deeply concerned about trends in the development, deployment, and

proliferation of armaments. They are exacerbated by the deterioration in
political relations. Unless states manage to reverse them, the world may be
heading for catastrophe. Preventive action is therefore needed urgently. The
problems we confront are man-made problems. Humanity has it within its
power to contain the dangers and embark upon a programme for the reduc-
tion and eventual abolition of the forces of destruction. The efforts so far
have been too feeble and their results too meagre for this Commission to
recommend merely renewed commitment and enhanced endeavour. More
of the same will not do. We recognize the constraints which apply, the com-
peting interests and mutual suspicions which permeate international rela-
tions. We see the need for a new beginning in the peaceful struggle against
war and destruction.
Principles for Action
All states have a right to security. In the absence of a world authority with
the right and power to police international relations, states have to protect
themselves. Unless they show mutual restraint and proper appreciation of

the realities of the nuclear age, however, the pursuit of security can cause
intensified competition and more tense political relations and, at the end of
the day, a reduction in security for all concerned.
Nuclear weapons have changed not only the scale of warfare but the
very concept of war itself. In the nuclear age war cannot be an instrument
of policy, only an engine for unprecedented destruction. States can no
longer seek security at each other's expense; it can be attained only through
cooperative undertakings. Security in the nuclear age means common secu-
rity. Even ideological opponents and political rivals have a shared interest
in survival. There must be partnership in the struggle against war itself. The
search for arms control and disarmament is the pursuit of common gains,
not unilateral advantage. A doctrine of common security must replace the
present expedient of deterrence through armaments. international peace
must rest on a commitment to joint survival rather than a threat of mutual
destruction.
General and Complete Disarmament
In its final document, the first special session of the General Assembly
devoted to disarmament charged the Committee on Disarmament with
elaborating a comprehensive programme leading to general and complete
disarmament. The Committee completed its task in April 1982. The
Commission strongly supports the goal of general and complete disarma-
ment. We recognize that this objective will not be realized in the near future.
But the ideal of a world in which international relations are based on the
rule of law, cooperation, and the peaceful pursuit of political ends must be
held high. This is the goal as well as the measure of efforts to reach inter-
national agreements on arms limitation and disarmament. To make progress
in that direction it is necessary to develop a concrete and comprehensive
programme of action reflecting the complex interrelationships of the many
critical elements in the present situation. It is necessary to break the impasse
and start a downward spiral.
Economic Pressures and Common Security
The economic and social costs of military competition constitute strong rea-
sons for countries to seek disarmament. The costs of military spending are
especially onerous in the difficult economic circumstances of the 1980s. These
costs, of course, are different for different countries. But some are common to
almost everyone: use of government revenues; diversion of scarce scientific
and technical skills from social pursuits; denial of investments which could
otherwise increase economic growth. The journey towards reversing the arms
race will follow a different path for each country. But for all countries the
economic prize will be great.
The Commission is disturbed by the current international situation.

Dialogue and moderation appear to be breaking down; tensions are accel-
erating. While weapons are not the only source of conflict among states,
competitions in arms exacerbate existing conficts and assume a dangerous
and ~ e l f - ~ r o ~ e l l momentum.
ing
Deliberate efforts to establish links between specific negotiations for the
limitation and reduction of arms and the general international behaviour
of one's opponent are inconsistent with our notion of common security.
Negotiations for the limitation and reduction of arms require a high degree
of continuity and stability. They are not gifts to an adversary or rewards for
his good behaviour, but rather a means of pursuing common security and
profiting from shared interests. The task of diplomacy is to limit, split, and
subdivide conflicts, not to generalize and aggregate them. Prior political
agreement cannot be made a precondition for negotiations about arms lim-
itation. Indeed, agreements on arms limitation and disarmament could make
it easier to resolve outstanding issues. The Commission considers the notion
of political linkage an umound principle which should be abandoned.
The prospects for arms limitation and disarmament will to some extent
always depend o n the general political climate. However, all states share an
interest in preventing the arms race from dominating their relations and
driving them towards armed conflict. Negotiations about, and agreements
to Iin~itand reduce, arms can provide an engine for improving relations and
restoring confidence. When tensions occur the need for communication and
negotiation is particularly strong.
Elements of a Programme for Arms Control a n d Disarmament
The Commission's recommendations, taken together, constitute a broad

programme for substantial progress towards arms limitation and disarma-
ment. The recommendations fall into six categories: (1) the nuclear chai-
lenge and East-West relations; (2)curbing the qualitative arms competition;
(3)ass~lringconfidence among states; (4)strengthening the UN security sys-
tem; (5)regional approaches to security; and (6) economic security.
There will be no winner in a nuclear war. The use of nuclear weapons

would result in devastation and suffering of a magnitude which would ren-
der meaningless any notion of victory. The size of existing nuclear stock-
piles and the near certainty of devastating retaliation make it futile and
dangerous to consider nuclear war an instrument of national policy.
Nuclear war would amount to an unprecedented catastrophe for humanity
and suicide for those who resorted to it.
No Victors in a Nuclear War
Were they ever to cross the nuclear threshold, nations would be set on a
course which does not lend itself to prediction. The very process of destruc-
tion would render prior calculations and attempts to exercise control fruit-
less. We reject any notion of 'windows of opportunity' for nuclear war. Any
doctrine based on the belief that it may be possible to wage a victorious
nuclear war is a dangerous challenge to the prudence and responsibility
which must inspire all approaches to international peace and security in the
nuclear age. We conclude that it is impossible to win a nuclear war and dan-
gerous for states to pursue policies or strategies based on the fallacious
assumption that a nuclear war might be won.
No Limited Nuclear War
The idea of fighting a limited nuclear war is dangerous. Nuclear weapons

are not war-fighting weapons. Once the nuclear threshold had been crossed
the dynamics of escalation would inexorably propel events towards catas-
trophe. Doctrines and strategies of limited nuclear war thus carry danger-
ous connotations. Their acceptance would diminish the fears and perceived
risks of nuclear war and blur the distinction between nuclear and 'conven-
tional' armed conflict, thus lowering the nuclear threshold.
Even if it is understood that nuclear war cannot be controlled, nations
would feel compelled to attempt to limit war should it begin. Paradoxically,
preparations for such contingencies, manifested in the acquisition of certain
weapons and control systems, can be dangerous to the extent that they may
be interpreted as suggesting the possibility of fighting a limited nuclear war
as a matter of deliberate policy.
Deterrence cannot be made foolproof. It could collapse in many differ-
ent ways: because of a technical accident, a human error or miscalculation,
the snowballing effect of a local conflict, among others. Nations must guard
against these possibilities through cooperative agreements for emergency
communications. But they also must abandon doctrines and preparations
for fighting limited nuclear war as a matter of deliberate policy.
Nuclear deterrence cannot provide the long-term basis for peace, stabil-
ity, and equity in international society. It must be replaced by the concept
of common security.
The conclusion is therefore inevitably that nuclear weapons must be
eliminated. We are fully aware, however, that this can only be achieved
through a gradual process which must be initiated by concrete steps.
1.1 Reductions and qualitative limitations of nuclear forces: Nuclear weapons

are part of the established reality. The nuclear arms race continues. In a very
real sense a nuclear shadow hangs over all political and armed conflicts in the
contemporary age. Most disturbing is the development and deployment of
weapons which may lead to a lowering of the nuclear threshold with the
attendant increased risk of nuclear war. The greatest danger would be for
people anywhere to become so used to an open-ended nuclear arms race that
they become complacent about the danger involved, or lose faith in their
capacity to turn the tide. But nations are not condemned to live by the ugly
dictates of nuclear weapons. They have the choice and indeed the responsibil-
ity to c ~ i r band eliminate the horrendous forces of destruction which nuclear
weapons represent.
We believe that there is an urgent need for agreements specifying major
reductions of nuclear weapons and restraints on their qualitative improve-
ments, with a view to maintaining parity at the lowest possible level of
forces. Stabilizing the nuclear arms race in this way could create a basis for
further steps in the direction of stopping the production of nuclear weapons
a n d reaching agreen~enton their eventual elimination. There is a need to
create a downward momentum. Nations cannot confine their efforts to
managing the existing high levels of armaments. Major reductions and con-
straints on qualitative 'improvements' must be a dominant theme in future
negotiations and agreements.
1.2 Reductions and qualitative limitations in US and USSR strategic forces:

Nuclear deterrence can be but a temporary expedient. It provides no per-
manent solution to international security. The consequences of failure are
too terrifying to leave the system unchanged. The world must break with a
system which equates the maintenance of peace with holding millions of
human beings and the fruits of their labour as hostages for the good behav-
iour of the governments of the nuclear weapon states.
The process of strategic arms limitation therefore is indispensable. It is
important, too, because it has become a key factor in the relations between
the United States and the Soviet Union, affecting the very framework and
climate of international relations. The 1972 and 1979 SALT agreements
constitute an important beginning; they must be preserved and the process
continued to provide a downward spiral in nuclear arms.
Negotiations must be resumed without precondition and further delay.
The objective of the negotiations must be twofold: First, the parties
should reaffzrm the important limitations and restraints that the SALT 11
Treaty provides, and agree on any necessary clarifications or adjustments
of the Treaty in that connection. Second, the parties should seek a follow-
on treaty providing for major reductions and qualitative limitations
resulting in essential parity at substantially lower and more stable levels
of forces. Particular emphasis should be accorded t o reductions and qual-
itative limitations that would reduce fears of a 'first strike', an attempt t o
disarm an opponent o r t o forestall a possible attack by a preemptive sur-
prise attack. A n y new agreement should also contain provisions necessary
to assure adequate uerification of these reductions and qualitative limitu-
tions, and should prohibit deployment of weapon systems which would cir-
cumuent agreed limitations and reductiorzs or reader verification impossible.
Successive agreements should point to the eventual elimination o f
strategic nuclear arms through interim stages that restrict the arsenals of the
nuclear weapon states to small, secure strategic forces in consonance with

the principle of equal security.
1.3 The anti-ballistic missile treaty must be upheld: The 1972 Treaty Limiting
Anti-Ballistic Missile Systems is an important agreement designed to lessen the
chance of nuclear war and to constrain the strategic arms race from escalating
into broader dimensions. It does not suggest that international peace and secur-
ity should be based on the ability of the great powers to inflict unacceptable
destruction on each other. It does reflect the fact that for the foreseeable future
there are no effective means of defending against ballistic missiles. States must
coexist, therefore, in a condition of mutual vulnerability, making the pursuit
of common security a matter of survival for humanity.
The Anti-Ballistic Missile Treaty is a substantial and necessary building
block in a viable system of common security. Abrogation of the Treaty would
undermine the whole strategic arms limitation and reduction process. The fail-
ure t o uphold the Anti-Ballistic Missile Treaty could lead to a destabilization
of the international situation and a greater risk of nuclear war. W e urge that
the treaty be upheld.
1.4 Parity in conventional forces in Europe should be established at lower

levels: The major military confrontation between East and West is in Europe
and takes place between NATO and the Warsaw Pact. The concentration of
military power assembled in Europe is the greatest in history. The Commission
recognizes the complex interrelationships which exist among the various elem-
ents of the armed forces on both sides, nuclear and conventional, as well as
between the force postures of the two alliances. A fair appraisal of the East-
West balance of forces on the continent of Europe is extremely complicated. So
many aspects of economy, geography, technology, traditions, military organiza-
tion, and threat perceptions are involved. A comprehensive approach to arms
limitation and reductions must be adopted in order to assure approximate mili-
tary parity at substantially reduced levels and to reduce the risk of nuclear war.
We are convinced that a large-scale conventional war in densely populated
Europe would be enormously destructive and in all likelihood would escalate
to the nuclear level. It would affect not only the nuclear weapon states or
allied states, but also neutral and non-aligned countries. War is not an accept-
able option for the resolution of political conflict in the nuclear age. The
armies which are poised against each other in Europe today are much larger
than would be necessitated by realistic appraisals of basic security needs.
Common security would be enhanced by drastic mutual reductions.
Since 1973 the two alliances in Europe have been negotiating in Vienna
about an agreement on mutual force reductions in Central Europe. They
have reached consensus on most of the basic principles that would govern
an agreement. It would provide for reductions in two phases leading to equal
collective ceilings of 900,000 men, a subceiling of 700,000 for land-force
personnel in the reduction area, and associated measures designed to ensure
compliance with the provisions of the agreement and to enhance both sides'
confidence. The parties have still to agree on what is the number of troops in
the reduction area at the present time, the details of the linkage between the
two phases of reductions, and the scope of the associated measures. The
Commission considers that the outstanding differences could be resolved sat-
isfactorily provided there were the political will to do so. Continued stalemate
will seriously diminish public confidence in negotiations for arms reductions.
We urge that the participating states conuene a meeting of Fowign Ministers
to resolve the differences and conclude an agreement before the end of 1982.
An agreement specifying parity and reduction of conventional forces in
Central Europe should be accompanied by commitments to abstain from
moving arms and troops to areas where they would diminish the security of
other countries in Europe. Agreement in Vienna on conventional forces in
Central Europe would provide a basis for, and facilitate the negotiation of,
agreements on withdrawal and reduction of nuclear weapons in Europe.
A subsequent agreement on parity of conventional forces in Europe a t sub-
stantially reduced levels could facilitate more far-reaching agreements on
the withdrawal and reduction of nuclear weapons. Such agreements would
be more likely if in the negotiations for conventional force reductions the
parties were to emphasize reducing those elements of the two sides' military
postures which the parties consider the most threatening.
1.5 Reducing the nuclear threat in Europe: The nuclear arsenals in Europe
are awesome. Furthermore, the Commission is deeply concerned about
those nuclear postures and doctrines which dangerously and erroneously
suggest that it may be possible to fight and 'win' a limited nuclear war. In
the event of a crisis their effect could be to drive the contending forces
across the threshold of a nuclear war. The Commission is convinced that
there must be substantial reductions in the nuclear stockpile leading to
denuclearization in Europe and eventually to a world free of nuclear
weapons. A necessary precondition is a negotiated agreement on substan-
tial mutual force reductions establishing and guaranteeing an approximate
parity of conventional forces between the two major alliances.
Therefore, the Commission supports a negotiated agreement for approxi-
mate parity in conventional forces between the two alliances. Such an agree-
ment zuould facilitate reductions in nuclear weapons and a reordering of tile
priority now accorded to nuclear arms in military contingency planning.
The Commission has devoted much time and effort to examining various
alternative ways for bringing these changes about.' Among the alternatives
studied was nuclear-weapon-free zones, which are dealt with in Section 5.3
concerning regional security arrangements. It should be remembered in this
connection that some countries in Europe do not belong to any of the military
alliances and have renounced the acquisition of nuclear arms.
Here we propose a functional approach concentrating on specific weapons
and classes of weapon. Our proposal for the gradual removal of the nu-
clear threat posed to Europe includes establishment of a battlefield-nuclear-
weapon-free zone and measures to strengthen the nuclear threshold and
reduce pressures for the early use of nuclear weapons, and substantial reduc-
tions in all categories of intermediate- (medium-) and shorter-range nuclear
weapons which threaten Europe.
(a) A battlefield-nuclear-weapon-free zone in Europe. We call special
attention to the dangers posed by those nuclear weapons whose delivery
systems are deployed in considerable numbers to forward positions in
Europe. These are known as 'battlefield' nuclear weapons. A large portion
of NATO's and the Warsaw Pact's nuclear munitions in Europe are of this
type. The weapons are designed and deployed to provide support to ground
forces in direct contact with the forces of the opponent. Their delivery sys-
tems have ranges up to 150 kilometres, and are primarily short-range rockets,
mines, and artillery. Most of the delivery systems are dual-capable, i.e. they
can fire either conventional munitions or nuclear munitions.
Because of their deployment in forward areas battlefield nuclear weapons
run the risk of being overrun early in an armed conflict. Maintaining
command and control over such weapons in 'the fog of war' would be dif-
ficult. Pressures for delegation of authority to use nuclear weapons to local
commanders and for their early use would be strong. The danger of cross-
ing the nuclear threshold and of further escalation could become acute. It
should be remembered in this connection that the areas close to the
East-West border in Central Europe are densely populated and contain large
industrial concentrations.
The Commission recommends the establishment of a battlefield-nuclear-
weapon-free zone, starting with Central Europe and extending ultimately
from the northern to the southern flanks of the two alliances. This scheme
would be implemented in the context of an agreement on parity and mutual
force reductions in Central Europe. No nuclear munitions would be per-
mitted in the zone.2 Storage sites for nuclear munitions also would be pro-
hibited. Manoeuvres simulating nuclear operations would not be allowed in
the zone. Preparations for the emplacement of atomic demolition munitions
and storage of such weapons would be prohibited.
There also should be rules governing the presence in the zone of artillery
and short-range missiles that could be adapted for both nuclear and con-
ventional use. The geographic definition of the zone should be determined
through negotiations, taking into account the relevant circumstances in the
areas involved, but for illustrative purposes, a width of 150 kilometres on
both sides may be suggested. Provisions for verifying compliance with these
prohibitions would be negotiated. They would have to include a limited
number of on-site inspections in the zone on a challenge basis.
The Commission recognizes that nuclear munitions may be brought
back to the forward areas in wartime, and that nuclear weapons may be
delivered by aircraft and other longer range systems. However, we consider
the establishment of the ~ r o p o s e dzone an important confidence-building
measure which would raise the nuclear threshold and reduce some of the
pressures for early use of nuclear weapons. It is consistent with our rejec-
tion of limited nuclear war as a matter of deliberate policy.
The agreement for withdrawal of 'battlefield' nuclear weapons from the

forward zone should be followed by substantial reductions in the number
of nuclear munitions in Europe with adequate measures of verification.
(b) Maintain a clear nuclear threshold. To contain and reduce the dan-
ger of nuclear confrontation in Europe it is important to maintain a clear
distinction between nuclear and conventional weapons. We urge the nuclear
weapon states to abstain from deploying weapons which blur the distinc-
tion by appearing to be more 'useable'. The so-called 'mini-nukes' and
enhanced radiation (neutron) weapons both fall into this category.'
(c) Reduction of intermediate- (medium-) range nuclear weapon Systems.
The Commission welcomes the opening of negotiations between the United
States and the Soviet Union on intermediate-range nuclear weapons and urges
the parties to give the search for agreement the highest priority. The com-
petitive deployment of these weapons constitutes a serious blow to political
and military stability between East and West, particularly in Europe. Nego-
tiations should reduce the number o f all such weapons to essential parity at
the lowest possible level, preferably at a level which would mean that N A T O
would forgo the introduction of a new generation of intermediate-range
missiles in Europe. Furthermore, we call on the parties also to agree to a
ban on deployment of new short-range nuclear weapon systems to areas
from which they could threaten the same targets which are threatened by
intermediate- (medium-) range nuclear weapons.
In addition to an accord on intermediate-range nuclear weapons in
Europe the parties should commit themselves to continue negotiations to
limit-all other nuclear forces which threaten Europe, including sea-based
cruise missiles. All nuclear weapons which are deployed in or against Europe,
including French and British forces, should be taken into ons side ration.^
1.6 A chemical-weapon-free zone in Europe: The world may be on the

brink of a major new arms race in chemical weaponry. The Commission
considers chemical weapons particularly abhorrent, and condemns any use
of such inhumane weapons.
Chemical weapons (including contact gasses and nerve agents) fall between,
and share some of the characteristics of, both conventional and nuclear war-
fare. They may be dispensed from munitions adaptable to most types of con-
ventional weapon system. They have indiscriminate and unpredictable effects
due to weather. Some can persist, poisoning the environment for a long time.
It has been estimated that if chemical weapons were used in densely populated
Europe, the ratio of non-combatant to combatant casualties could be as high
as twenty to one. Moreover, the use of chemical weapons would blur the dis-
tinction between conventional and nuclear warfare. This would increase the
danger of one sliding into the other.
Chemical weapon stockpiles include both bulk storage containers for
chemical agents from which munitions can be charged, and such munitions
as artillery shells, rocket warheads, aerial bombs, and mines already loaded
with chemical agents. Since they are highly toxic, special safety precautions
are needed during storage and handling. This is why it is generally assumed
that chemical weapons are stored in a small number of central depots in
Europe. Information about the possible distribution of chemical weapons t o
troops in the field is both uncertain and contradictory. The development of
so-called 'binary' munitions, however, could facilitate their distribution.
These munitions are filled with two less toxic chemicals which are com-
bined t o create a lethal nerve gas only after the munition has been fired.
The Commission calls for the establishment of a chemical-weapon-frecj
zone in Europe, beginning with Central Europe. The agreement would
include a declaration of the whereabouts of existing depots and stockpiles
in Europe, adequate means to verify their destruction, and procedures for
monitoring compliance on a continuing basis, including a few on-site
inspections on a challenge basis. The training of troops in the offensive use
of chemical weapons also would be prohibited.
1.7 Confidence- and security-building measures in Europe: The Commission

considers the Final Act of the 197.5 Helsinki Conference on Security and
Cooperation in Europe and the follow-up process important t o the evolution
of the security arrangements in Europe. It points beyond confrontation t o
cooperation and the pursuit of common security. A system of confidence-
building measures relating to military manoeuvres has been instituted and
adhered t o by the participating states. In the follow-up meeting in Madrid,
which will reconvene in November 1982, the participating states are nego-
tiating the mandate for a Conference on Disarmament and Confidence and
Security Building Measures. The first phase would be devoted to negotiat-
ing agreement on Confidence and Security Building Measttres which would
apply to all of Europe, contribute to military security, be verifiable, and
constitute n binding and lasting commitment. The Commission considers
this effort nn important contribution to the growth o f a system and prac-
tice of commoH security in Europe. The second phase should comprise
negotiations for substantial disarmament in Europe.
Competitions in armaments focus as much on the characteristics of the

weapons being acquired as on their number. Contrary to the principles of
common security, states have sought to guarantee their survival and enhance
their influence by developing or purchasing weapons that are more effective
and lethal. The nuclear-weapon states continue to develop new kinds of
nuclear weapons and new means of delivering them. They are searching for
new means of warfare in space and other frontiers of human exploration. At
the same time, a growing number of other states are increasing their potential
to develop nuclear weapons at some future time.
All these developments aggravate existing political tensions among nations
and make more difficult the avoidance and resolution of conflicts. The appear-
ance of new types of military capabilities, no less than the appearance of
greater numbers of weapons, can contribute to regional instability, raise fears

of war and suspicion of hostile intentions. If nations are to live in common
security, qualitative aspects of the arms race, like its quantitative features, must
be constrained.
Advances in military capabilities begin in the human mind, proceed in
different strands in numerous offices and laboratories. Only when they are
near completion do they coalesce in the concrete form of a new weapon.
New applications of technology can sometimes be stabilizing, but more
often they generate new instabilities and competitions. Large research and
development establishments represent vested interests which generate pres-
sures for further research and increased effort.
It is difficult to identify points in the process of military research and
development at which nations could agree to exercise restraint and at which
compliance with such agreements could be verified. The notable exception
is the point at which prototypes of weapons are tested in the field. The pos-
sibility of restricting the development of new or improved weapon systems
at this critical point should be utilized on a much more extensive scale.
Indeed, agreements already have been reached which restrict qualitative
aspects of arms competitions at the testing stage; the 1963 Partial Test Ban
Treaty, the 1974 Threshold Test Ban Treaty, the 1972 Anti-Ballistic Missile
Treaty, and the 1979 SALT I1 Treaty.
Moreover, just as in the case of quantitative limitations, nations are
unlikely to be willing to exercise unreciprocated unilateral restraint for sub-
stantial periods. The major nuclear-weapon states have a special responsi-
bility, but all nations must seek qualitative restraints. Steps have t o be taken
in common by nuclear- and non-nuclear-weapon states, by arms exporters
and arms purchasers, by East and West, b y great powers and small states.
2.1 A comprehensive test ban treaty: The conclusion of a treaty banning all
nuclear tests would make the introduction of new weapon designs into the
armories of the nuclear-weapon states much more difficult. It would be a
major constraint on the qualitative development of more sophisticated
nuclear weapons. It also could be an important contribution to limiting the
improvement of the present stocks of nuclear weapons. Hence it would
enhance the acceptability and credibility of the Non-Proliferation Treaty,
which works to limit the spread of nuclear weapons.
The Commission considers that efforts should be concentrated o n the
negotiation of a treaty banning all nuclear tests. Such a treaty is needed in
order to forestall a new round of nuclear weapon developments which
could exacerbate East-West relations, reduce stability, and weaken the
Non-Proliferation Treaty.
The Commission welcomes the decision of the Committee on Disarma-
ment in April 1982 to establish an ad hoc working group on a nuclear test
ban. The Commission trusts that it will soon be possible to negotiate and con-
clude the Comprehensive Test Ban Treaty which for more than a quarter of a
century has been awaited in vain by peoples the world over.
In further support thereof, we urge that the trilateral negotiations

between the United States, the Soviet Union, and the United Kingdom on a
comprehensive test ban be resumed immediately in order to settle the still
unresolved issues, including the question of verification. Political will is
needed in order to transcend the remaining obstacles. The Commission is of
the view that it is possible to establish an effective system of verification and
confidence building by arrangements involving the International Seismic
Data Exchange, agreed procedures for consultation and on-site inspection,
and a network of national seismic stations.
During the period between completion of negotiations and formal mti-
fication of the test ban treaty, all the nuclear powers should participate in a
uoluntary moratorium on all nuclear tests.
2.2 A ban on anti-satellite systems: Outer space has become an important

part of the military competition between East and West. The military
machines of the major powers have become increasingly dependent on space-
based support. Satellite systems have opened up a wide range of possibilities
for verification and warning, for command, control, and communications. If
these satellites were threatened, it could result in a substantial expansion o f
the strategic arms race into outer space, as each side sought to protect its own
system.
Between 1977 and 1979 the Soviet Union and the United States discussed
a ban on anti-satellite weapons. Time is running out. The Commission rec-
ommends that these negotiations he reopened and that priority be given to
iz suspension and prohibition of the testing of anti-satellite zoeapons. I t is
essential that such a ban go into effect before irreversible technological

'progress' hils been made. Negations also should aim at reaching izgreement
that would ban the deployment of anti-satellite weaporzs and require the
dismantling of existing systems.
Further bans on weapons and activities in outer space will undoubtedly
be needed. The exploitation of outer space raises a number of complex tech-
nical questions and judgements. The Commission urges the major ir~dustrial
powers to detlelop a dialogue with the aim of identifying and preventing
military uses of outer space that might constitute threats to international
peace and security. This dialogue should lead to negotiated bans and limits
on specific weapon systpms or entire areas of activity.
2.3 A chemical weapon disarmament treaty: The existing chemical and

biological arms control and disarmament agreements are among the few
safeguards against the dangers of an expanded arms race. Use in war of both
chemical and biological weapons is prohibited by the 3 925 Geneva Protocol
and its associated body of customary international law. Possession of bio-
logical weapons, including toxin weapons, is outlawed by the 1975 Biological
Weapons Convention. But the possession of chemical weapons is not pro-
hibited and a number of states have reserved the right to use them i t they are
attacked with chemical weapons. The majority of states are parties to these
agreements and have under the 1975 Convention committed themselves to

continue negotiations 'in good faith' to prohibit possession of chemical
weapons.
Since the First World War, chemical weapons have only been used in
conflicts in the developing world. All reports of alleged use also are limited
to Third World countries. Thus a new arms race in chemical weapons poses
worldwide dangers, in particular for the developing world.
Pressures to build up stocks of chemical weapons are in danger of sub-
verting the existing accords. It is vital to accelerate negotiations aimed at
extending and strengthening existing agreements by the introduction of a
comprehensive chemical weapon disarmament treaty banning such
weapons altogether. This requires resumption of the stalled bilateral talks
between the United States and the Soviet Union. When these negotiations
last were convened, in July 1980, there was agreement in principle o n the
use of on-site inspection as a verification technique. Bilateral talks d o not,
of course, substitute for renewed efforts within the Committee on Disarma-
ment to negotiate agreement on a comprehensive chemical weapon disar-
mament treaty, but would strengthen those efforts.
The negotiations involve complex technical matters and sensitive polit-
ical issues and will require time to conclude successfully. Therefore we call in
addition for agreement on consultative procedures so that problems arising
under the Geneva Protocol and the Biological Warfare Convention can be
resolved through international cooperation. Such procedures could include
the option of consultative meetings being convened at the expert level under
the auspices of the United Nations that would be open to all states.
A chemical weapon disarmament treaty should contain provisions for a
permanent consultative commission composed of all the parties to the treaty
and served by a small technical staff. The commission should ensure imple-
mentation of the treaty and thereafter monitor continued compliance. It could
also be charged with the establishment of an effective complaint procedure.
Appropriate verification must be agreed for each stage of implementation
of a treaty on chemical weapon disarmament. Both the declaration and
destruction of stockpiles and production facilities and subsequent monitor-
ing of compliance with provisions for non-production of chemical weapons
must be verified under adequate international control. Verification measures
should include a combination of voluntary confidence-building measures,
national verification measures, and agreed international means.
Developing countries have a special interest in ensuring compliance with
a treaty banning stockpiles and production of chemical weapons. Since very
few developing states have the technology to develop adequate national
means of verification, international means are necessary also in order to
protect their interest.
Over the past fifteen years scientific understanding of the molecular and
cellular processes of life has grown enormously. So far there is no evidence
of military exploitation of this knowledge. Should the biological sciences be
tapped for military purposes, however, hideous new weapons could emerge.
Our well-being and economic and social development could be drastically

retarded. The Commission calls for an international convention which
would prohibit any secret development or experimentation in the military
applications of molecular biology and its associated disciplines.
2.4 Universal adherence to the Non-Proliferation Treaty: Preventing the

spread of nuclear weapons is a critical element in any international effort to
halt and reverse the nuclear arms race and ensure the maintenance of inter-
national peace and security. Progress in this direction demands obligations
and responsibilities on the part of both nuclear-weapon states and non-
nuclear-weapon states.
The problem of proliferation has fallen into two sections, popularly
termed 'vertical' and 'horizontal' proliferation. Vertical proliferation refers
to the growth of the stockpiles of nuclear weapons held by existing nuclear-
weapon states. Horizontal proliferation refers to the spreading of nuclear
weaponry to new countries. Efforts to stop both kinds of proliferation
resulted in the conclusion of the Non-Proliferation Treaty in 1970, which
committed non-nuclear-weapon states to refrain from acquiring such weapons
and the nuclear-weapon states to halting and reversing their processes of
qualitative and quantitative growth of nuclear weapons.
The Non-Proliferation Treaty is the centerpiece of the widespread inter-
national interest in maintaining the presumption against proliferation. One
hundred and eighteen states are now parties to the treaty. However, France
and China, which are nuclear-weapon states, as well as a number of import-
ant countries on the threshold of being able to build nuclear weapons, have
so far failed to sign and ratify the treaty. The Cotnmission urges all states
to adhere t o the Non-Prolifeuation Treaty.
Some opponents of the treaty point to its discriminatory nature, accepting
n~iclearweapons for those five countries which already have them but for-
bidding others to develop similar capabilities. But, by its very nature, non-
proliferation involves a degree of discrimination. The key issue is how this
fact of life is handled. The Commission recognizes that the failure of the
nuclear-weapon states to make progress towards nuclear disarmament, as
promised in Article VI of the Non-Proliferation Treaty, affects the attitudes
and commitments of others. The proposals we have made for a complete
nuclear test ban, for the reduction and withdrawal of nuclear arms in Europe
and in the Soviet and American stockpiles, are a reflection of our concern
to strengthen the treaty's appeal. Failure to stop vertical proliferation will
cornpromise the integrity of the Non-Proliferation Treaty.
2.5 Safeguarding the nuclear fuel cycle: International cooperation is needed

order to reduce the danger that the development and application of peace-
ful uses of nuclear energy may lead to diversion of nuclear materials for mili-
tary purposes. IJarticularly sensitive parts of the nuclear fuel cycle should be
placed under international authority. This could include the establishment
of international fupl banks, an international plutonium storage scheme, and
internationally managed sites for spent fuel storage. Regional organizations

can contribute significantly to such international arrangements, which should
be drawn together by the International Atomic Energy Agency through its
Committee on Assurance of Supply.
Participants in the 1977-80 International Fuel Cycle Evaluation
acknowledged that fuels usable in weapons require special procedures. The
Committee on Assurance of Supply of the International Atomic Energy
Agency may be developed into a central negotiating and management
forum comprising both suppliers and recipient countries. Such cooperation
would conform with Article IV of the Non-Proliferation Treaty which
underlines the need for equitable cooperation in the use of nuclear energy
for peaceful purposes.
2.6 The need t o limit conventional arms transfers: The volume of arms trans-
fers has more than doubled during the past decade. Deliveries are now close
to $30 billion per annum and orders are substantially higher. More than three
quarters of all arms transfers go to the countries of the developing world.
In our view, there is an urgent need for a concerted effort to develop a
fair system of guidelines and restraints covering arms exports, based on
cooperation among recipient and supplier states.
Supplier states should open talks aimed a t establishing criteria by which
they could regulate arms transfers on an equitable basis. Restraints need to be
defined in terms of quantities and qualities, geography and military circum-
stances. The guidelines for arms transfer should include such principles as
No significant increase in the quantity of weapons which are transferred

to a region.
No first introduction of advanced weapon systems into a region which
create new or significantly higher levels of combat capability.
Special restrictions on the transfer of lethal weapons to warring parties,
taking into account the inherent right of individual or collective self-
defence.
Adherence to the implementation of UN resolutions and sanctions.
0 No transfer of particularly inhumane and indiscriminate weapons.
Special precautions to be taken when transferring weapons, such as hand-
held anti-aircraft weapons, which, if they fell into the hands of individuals
or sub-national groups, would be especially dangerous.
The United States and the Soviet Union held Conventional Arms Transfer
talks in 1977-80. The Commission endorses the resumption of such talks
which should include also France, the United Kingdom, and other major
supplier states. Another need is for talks between supplier states and recipi-
ents in regions where tensions are particularly severe. There is a need for
multilateral restraints.
Recipient states should similarly undertake to develop guidelines and
codes of conduct designed to curb the flow of arms and avoid arms races.
Disarmament and Security issues 283
An important beginning was made by eight Andean states in the Declaration

of Ayacucho in 1974 in which they pledged to 'create conditions which per-
mit effective limitation of armaments and put an end to their acquisition
for offensive warlike purposes in order to dedicate all possible resources to
economic and social development'. Regrettably, the discussion of specific
restraints broke down. However, at a meeting in Mexico City in 1978,
twenty 1.atin American and Caribbean states agreed to exchange informa-
tion on weapon purchases and work towards a regime of restraints on arms
transfers.
Recipient states may wish to bar or limit certain types of weapon. They
may consider that if those weapons were used in their part of the world they
would enhance offensive capacities and introduce incentives for rapid action in
a crisis. They may wish, too, to outlaw weapons which are starkly inhumane
in their effects. The 'rules of the game' will need to be tailored to the specific
circumstances of the area in question. Regional Conferences on Security and
Cooperation could discuss general principles. States which are participating in
zones of peace or similar groupings could decide on more specific guidelines.
The latter would have to be adhered to also by the supplier states.
Adequate verification is an important part of any agreement on arms l i m -

tation or reduction. States are loath to enter into such agreements on the
basis of good faith alone. The development of the so-called national tech-
nical means gave the parties to arms control agreements confidence that
they could monitor compliance with the provisions of treaties adequately.
Technologies used to observe and monitor military activities have advanced
impressively. Military secrecy still exists, however.
Consequently, monitoring compliance with treaty proscriptions remains
an issue in negotiations. There should be a close link between the scope and
design of treaties and the means prescribed for their verification. There are
no all-purpose forms of verification. Requirements have to be determined
in each specific instance. Verification requires cooperative arrangements
and, in some instances, on-site inspections.
While the purpose of verification is to provide for timely detection of any
illegal, surreptitious activity, it could hopefilly also lead to improued confi-
dence among treaty parties and promote compliance with treaty norms.
3.1 Confidence-building measures relating to military expenditure, research

and development: Satellites can only detect forces in being or information.
However, it takes from seven to fifteen years for a modern weapon system
to move through the various stages of research, development, testing and
deployment. States hedge against the possible future results of decisions
other states may have made today, assuming the worst about the decisions
of the adversary. Confidence building is necessary if the spirals of suspicion
and fear are to be broken.
The greater sharing of information about budgetary expenditure, for

example, could enhance confidence. A standardized reporting system has
been developed and tried out under the auspices of the United Nations. The
35th General Assembly of the United Nations urged all states to report
information about their expenditure for military purposes in accordance
with the system. The Commission urges all states to comply with the reso-
lution of the General A ~ s e m b l y . ~ ~
In view of the momentum and vested interests which affect the process of
military research and development, the Commission urges the major indus-
trial powers to conduct a dialogue about questions relating to research and
development of all types of military forces. This would provide an oppor-
tunity t o voice concerns about the implications of actual and possible pro-
grammes, so that the response could be taken into account prior to national
decisions about procurement and deployment. The danger of unintended
destabilization and aggravated competition could thereby be reduced.
Having outlined a programme for arms control and disarmament in rela-

tion primarily t o the competition and conflicts among the industrialized
countries, we focus on the need to promote international security in a
global context with emphasis on the developing world.
4 Strengthening the United Nations Security System
We are convinced of the need to strengthen the security role of the United
Nations. A new conceptual approach must be developed in order to pro-
mote common security in the world at large.
4.1 More effective use of the Security Council and the Secretary General:
Within the UN, primary responsibility for maintaining international peace
and security rests with the Security Council. Regrettably, states have tended
only t o turn t o the Council as a last resort when conflict has already, or is
on the verge of breaking out. If they are to be persuaded to shed this atti-
tude, the Security Council itself must enhance its capacity to preempt con-
flicts. The permanent members, in particular, should seek to foster a close
understanding and collaboration among themselves and encourage a mutu-
ally supportive partnership with the Secretary General to facilitate initiatives
under Article 99 of the Charter.
Article 99 specifically authorizes the Secretary General 'to bring to the
attention of the Security Council any matter which in his opinion may
threaten the maintenance of international peace and security'. The Security
Council should adopt an initiating resolution explicitly calling upon the
Secretary General to bring to its immediate attention potential threats to
the peace. In addition, we recommend that the Secretary General should
report t o the Council on a regular basis throughout the year. There should
be a special annual 'state of the international community' message to be
delivered in person by the Secretary General to a meeting of the Security
Council with the Foreign Ministers in attendance. This message should be

delivered at a public session so that all states become aware of the Secretary
General's assessment. It should be followed by a private discussion of its
implications by the Foreign Ministers of Security Council members. They
should attempt to identify specific measures which the Council might take
to head off possible conflicts.
To help assert the UN's primacy in international peace and security and
to enhance the role of the Security Council we believe that it would be use-
ful for the Council to hold occasional meetings outside UN headquarters.
This would provide the opportunity for a more focused discussion and con-
sultation on the problems of a particular region.
4.2 Collective security - a first step: A key proposal in our recommendations

is the implementation of a modified version of the U N Charter2 concept o f
collective security. Its basis would be political agreement and partnership
hetween the pernravzerrt rncwzbers of the Security Council and Third World
countries. Its scope would be limited to Third World conflicts arising out of
border disputes or threats to territorial integrity cazised by other factors. Its
purpose would be to prevent the conflicts from being settled by armed
force, and not to pronounce on the substantive issues in dispute. It would
be underpinned by an understanding - 'concordat' - among the permanent
members of the Security Council to support collective security action, at
least, to the extent possible, of not voting against it. The cooperation of the
permanent members of the Security Council is particularly important. Their
consent is a prerequisite for the effective functioning of the United Nations
in maintaining international peace and security.
As distinct from peacekeeping operations, collective security procedures
would have anticipatory, preventive, and enforcement elements. They would
all be integrally linked, each reinforcing the other.
At the anticipatory and preventive levels three phases of UN action
would be necessary:
(i) O n being alerted by at least one of the disputing parties to the danger
of a possible conflict, the Secretary General would constitute a fact-
finding mission to advise him on the situation.
(ii) If circumstances warrant, and with the consent of at least one of the
disputing parties, the Secretary General would seek the authorization of
the Security Council to send a military observer team to the requesting
state to assess the situation in military terms and to demonstrate the
Council's serious concern.
(iii) In the light of circumstances and the report of the military observers,
the Security Council would authorize the induction of an appropriate
ON military force at the request of one of the disputing states with a
view to preventing conflict. This force would be deployed within the
likely zone of hostilities, in the territory of the requesting state, thereby
providing a visible deterrent to a potential aggressor.
All three phases would be covered by the political concordat among the
permanent members of the Security Council whereby they would commit
themselves to support particular types of collective security action, and thereby
placed on an assured basis.
The introduction of substantial UN forces before the outbreak of hostil-
ities would, in most cases, prevent violations of territory from occurring at all.
Nevertheless, there could be situations where violation of territory might still
take place with an attack so sudden as to preempt the possibility of effective
preventive measures. In such circumstances limited enforcement measures
would become necessary. The first objective would be to establish a negot-
iated ceasefire. The Council would call on the warring parties to cease hos-
tilities and notify them of the dispatch of collective security forces to establish
and maintain an effective ceasefire. The parties would be asked to cooperate
fully in the achievement of this objective, it being clearly understood that UN
forces would have the right of self-defence if attacked by either of the two
warring parties.
Full-scale collective security enforcement action would, of course, imply
restoration of the status quo ante through military means. This is the ultim-
ate deterrent enshrined in Chapter VII of the Charter. Although not real-
izable in the immediate future, it must remain a goal towards which the
international community works.
For the present, other means could be used to ensure that aggression does
not prevail. The introduction of a ceasefire should be accompanied by an
appeal by the Security Council to the aggressor state to withdraw its troops
to its original borders. In the event of a refusal to comply, the Council would
immediately consider ways of enforcing its will through the other provisions
of Chapter VII, including the imposition of mandatory economic sanctions.
4.3 Process of implementation: We identify the following key components

for implementing our approach to collective security:
(i) Third World support

The Non-Aligned Movement has long been an advocate of a strength-
ened UN role in international security. Its support would be critical in
facilitating the proposed concordat among the permanent members of
the Security Council.
(ii) A political concordat among the veto powers
The scope of this concordat would be limited, in both procedural and
operational terms. The permanent members of the Security Council
would be committed to supporting collective security action in the man-
ner described, and, at least to the extent possible, to not vote against it.
(iii) An operational structure for UN standby forces
Article 43 envisages agreements between the UN and member govern-
ments on the provision of military standby forces. The Military Staff
Committee should be reactivated and strengthened for this purpose.
Furthermore, the respective roles of the Secretary General and the
Military Staff Committee would need to be carefully considered so as to

ensure that enforcement action by the UN is not allowed to become, or
perceived by Third World countries to be, a vehicle for great power inter-
ference. Standby forces should not be recruited exclusively or over-
whelmingly from the forces of the permanent members of the Security
Council. We consider it particularly important that a greater number of
Third World countries should become potential contributors of standby
forces. This objective could be accomplished most readily on a regional
basis. Where states of the region deem it suitable, regional or sub-
regional cooperation for the establishment, equipping and training of
standby forces along the lines that have already been successfully devel-
oped by the Nordic states should be actively encouraged.
The presence of standby forces in a particular region where it was thought

that enforcement action might be required would mean that they could be
rapidly deployed to the scene of the conflict, either to be stationed on the
border as a deterrent to aggression or to establish a ceasefire as soon as pos-
sible after a violation of territory has taken place. In the case of Africa,
arrangements establishing standby forces within the region, moreover, would
provide the necessary military infrastructure to enable the Organization of
African Unity to effectively contribute to peacekeeping operations which it
may have itself initiated, even though the necessary funding and specialized
technical support might still have to be provided under UN auspices.
Specifically, in connection with the proposal for establishing a UN col-
lective security system, we envisage that regional organizations could play
a vital role in alerting the Security Council and the UN Secretary General
to the danger of an imminent threat to the peace and in supplementing UN
efforts to maintain peace.
4.4 Improved capability for peacekeeping: Since our proposal on collective

security will not apply to all conflict situations, there will be continuing need
for UN peacekeeping operations. We recommend that a small complement of
professional military personnel be included in the staff of the Under-Secretary
General for Special Political Affairs who is responsible to the Secretary
General for the coordination and management of all peacekeeping operations.
Participation in peacekeeping operations is not compulsory but voluntary
and only a small number of countries have responded to the UN's call in the
past. We believe steps should be taken to encourage wider participation in
peacekeeping through:
( a ) A General Assembly resolution requesting states to incorporate train-

ing for peacekeeping as part o f their armies' basic training course,
assisted by a standard training manual issued b y the U N Secreturcat.
( b ) A joint undertaking between states with experience in peace-keeping
a ~ l dan appropriate UN agency to assist in the training and equiping of
troops from Third World countries.
(c) Regional arrangements to promote units for peacekeeping duties on a

standby basis.
(d) The stockpiling of certain types o f equipment and supplies which are
always necessary. This would improve the capacity o f the U N to under-
take peacekeeping operations at short notice. The major powers should
be asked to contribute transportation aircraft and special units for logis-
tic and signals support; other states should be asked to earmark units
for medical services, including field hospitals. Contribution of special
units would also improve capabilities for disaster relief operations.
The UN also must be prepared to respond to new kinds of challenges to

international peace and security. For example, the emergence of extensive
piracy in the areas off South East Asia might suggest the creation of a small
UN naval patrol force based on the voluntary assignment of naval vessels and
crews to UN duty by member states, and the consent of the littoral states.
4.5 An appropriate funding mechanism with built-in automaticity: The UN

has experienced great difficulty in eliciting the financial contributions neces-
sary to pay for peacekeeping operations from some member states, including
one or two members of the Security Council. We believe that collective secur-
ity operations and, for other purposes, peacekeeping ones as well, need to be
financed through an independent source of revenue.
We underline the importance of adopting a means o f automatic financ-
ing that spreads the burden widely and fairly throughout the international
community. All will benefit, all should contribute.
Pending agreement on automatic funding from an independent source of
revenue, we recommend that the General Assembly should agree on a spe-
cified percentage surcharge to be added to the assessed contributions of all
member countries to the regular budget. These moneys would be placed in
a special reserve fund earmarked for implementing all aspects of collective
security operations. Current peacekeeping operations, too, would benefit
from a similar approach.
5 Regional Approaches to Security
The Commission's recommendations for strengthening the UN's security

system stem from the conviction that there is no alternative to preserving
and enhancing the primacy of its role in maintaining international peace.
Although Third World countries in recent years have increasingly sought to
handle their own conflicts outside the UN, in many of the conflicts neigh-
bowing countries take opposing sides. This demonstrates that a regional
approach can often prove inadequate or counter-productive. There are
some situations in which a regional forum could provide a more appropri-
ate framework than the UN for arriving at a ~oliticalsettlement, but even
in such cases financial and operational limitations at the regional level
sometimes work against effective security solutions.
Regional approaches should, therefore, be viewed not as substitutes for

UN action, but as a means of complementing and strengthening it. There is
a need to develop an operational connection between regional security ini-
tiatives and the U N security system. This kind of link, moreover, would be
fully in accord with Chapter VIII of the Charter which explicitly anticipates
that regions might wish to establish their own arrangements for dealing
with matters relating to international peace and security. It makes only two
provisos: that these arrangements and bodies must be 'consistent with the
purposes and principles of the United Nations', and that 'the Security
Council shall at all times be kept fully informed of activities undertaken or
in contemplation under regional arrangements or by regional agencies for
the maintenance of international peace and security'.
There is a great unexplored potential at the regional level not only to
meet and resolve actual conflict situations as they arise, but also to promote
a general sense of security through cooperative measures with the aim of
facilitating disarmament, encouraging policies of mutual restraint and
improving the economic welfare of member states. In making the recom-
mendations set out below, however, the Commission has been conscious
that the various regions and sub-regions differ widely in respect both of
indigenous rivalries and the degree of involvement by the major powers. We
fully appreciate that any initiative for regional cooperation will require
regional consensus, but we are convinced that consensus can in turn be con-
solidated and expanded through cooperation.
5.1 Regional conferences on security and cooperation: The commission rec-

ommends that the countries making up the various regions, and in some
instances sub-regions, of the Third World consider the convocation of peri-
odic or ad hoc Regional Conferences on Security and Cooperation similar
to the one launched in Helsinki for Europe in 1975. Regional Conferences
on Security and Cooperation could add new substance to the concept of
common security. The priorities must be developed by the countries con-
cerned and reflect the circumstances in the individual regions both with
respect to agenda and participation. The Secretary General of the United
Nations should he invited to participate.
It is envisaged that the Regional Conferences could provide an overall
framework for cooperation not only on matters directly relating to security,
but in the economic, social, and cultural spheres as well.
In the area of security, the Conferences could consider such matters as
adoption of codes of conduct and confidence-building measures, establish-
ment of zones of peace and nuclear-weapon-free zones, and agreements on
arms limitations and reductions. Subsidiary bodies could be set up to deal
with aspects of implementing the Conferences' decisions or to carry out any
further studies that might be required. Depending on the character of their
membership, Regional Conferences might consider it useful, for instance, to
establish a Boundaries Commission to investigate and make recommendations
on solutions for border disputes or a similar body to look into difficulties
arising from the demarcation of territorial waters and exclusive economic

zones. Regional study institutes could be created to analyse security issues of
direct relevance to the particular region and to formulate recommendations
for the consideration of the Conference; such institutes should be funded by
governments and possibly receive a financial input from the UN as well, but
should be allowed to operate independently of government direction.
The Regional Conferences would also be the appropriate bodies for
launching any regional peacekeeping or peacemaking initiative to meet a
given crisis situation. It would, however, be essential for them to keep the
Security Council fully informed about any specific security arrangements
contemplated. We further recommend that general working procedures for
tying regional security arrangements into the UN security system should be
formulated. These should preferably be established soon after the Regional
Conference is constituted so as to create a standby operational framework
for activating cooperation with the UN to cope with conflict situations
when it is needed.
In our opinion, the concept of regional security will be unlikely to take root
unless it is sustained by programmes for economic cooperation to encourage
countries to see themselves as having a national stake in actively working to
achieve regional harmony. An important focus of the Regional Conferences
must therefore be the establishment of joint projects that are designed to bene-
fit all participating states. The UN's regional economic commissions could
have an important part to play in this connection - the Economic Commission
for Europe, for example, has performed a valuable function in assisting the
development of the Conference on Security and Cooperation in Europe.
Involvement of these Commissions would moreover ensure a UN contribution
of funds and technical assistance for security-building through economic
cooperation. This would provide an effective infrastructure for the link
between regional security initiatives and the UN security system.
The Regional Conferences could also consider schemes for regional
cooperation on the peaceful exploitation of nuclear energy in a manner
which would strengthen an equitable non-proliferation regime. Regional
cooperation could comprise regional fuel banks, plutonium storage schemes
and arrangements for spent fuel management. It could provide structure
and substance to general international projects which should be drawn
together by the International Atomic Energy Agency.
5.2 Zones of peace: The creation of zones of peace has been proposed most
notably for the Indian Ocean and South East Asian a r e a d Within the zone,
peace should be maintained by the countries themselves through the peace-
ful resolution of disputes in a context of political and economic coopera-
tion, as well as mutual military restraint. An essential factor in ensuring its
viability, however, is agreement by outside powers to respect its purposes
and specific provisions.
Zones of peace would be a flexible mechanism for developing coopera-
tion at the sub-regional level, while the proposed Regional Conferences on
Security and Cooperation could provide a general framework for consider-

ing objectives and experiences of the different zones within their region and
for establishing links between them. States within the zones could cooper-
ate on developing a code of conduct and confidence-building measures as
well as on an agreement to limit arms competition. Some important sug-
gestions along these lines were put forward by the President of Mexico in
February 1982 as part of a proposal to further a relaxation of tensions in
Central America. The main elements encompass renunciation of all threats
or use of force, balanced reduction of military troops in the area, and a sys-
tem of non-aggression pacts.
It is important to note that the Kuala Lumpur Declaration of 1971 on
the establishment of South East Asia as a Zone of Peace, Freedom and
Neutrality was issued by a grouping of countries which had already put sig-
nificant emphasis on economic, social and cultural cooperation and had
formed themselves into the Association of South East Asian Nations to fur-
ther this objective. Similarly, the Economic Community of West African
States started its existence in 1975 as a purely economic grouping and in
1981 its sixteen West African member states adopted a Protocol on Mutual
Assistance in Defence Matters. The Gulf Cooperation Council, established
in 1981 with the ultimate aim of achieving unity of their six member coun-
tries, has likewise stressed the need to build 'coordination, integration and
cooperation in all fields'.
The Commission considers that the concept o f zones of peace could be
an important contribution to the maintenance of international peace and
security. Political difficulties that might seem to militate against is realization
in the immediate future should not, in our view, inhibit g r ~ u p sof countries
from continuing their work towards the establishment of such zones as a
long-term objective.
5.3 Nuclear-weapon-free zones: The Commission believes that the establish-

ment of nuclear-weai)on-free zones on the basis of arrangements freely arrived
at among the states of the region or sub-region concerned, constitutes an
important step towards nun-proliferation, common security and disarma-
ment. They could provide mutual reassurance to states preferring not to
acquire or allow deployment o f nuclear weapons as long as neighhouring
states exercise similar restraint. This would improve the chances for the region
not to become enveloped in the competition of the nuclear-weapon states.
The nuclear-weapon states would have to undertake a binding commitment
to respect the status of the zone, and not to use or threaten to use nuclear
weapons against the states of the zone.
The Treaty of Tlatelolco, prohibiting nuclear weapons in Latin America,
is a path-breaking regional arrangement in this field. A party to it is not
bound, though, until all the signatories have completed ratification, unless it
waives this condition. Brazil and Chile have not done so. At present the treaty
is in force for twenty-two Latin American states. Argentina has signed but not
ratified the treaty. Cuba has neither signed nor ratified. The Commission
strongly urges all states concerned to adopt all relevant measures to ensure
the full application of the treaty.
Proposals for creating nuclear-weapon-free zones in Africa, the South
Pacific, South Asia and the Middle East have been put forward in the
United Nations and have received support in the General Assembly. The
process of establishing nuclear-weapon-free zones in different parts of the
world should be encouraged with the ultimate objective of achieving a
world entirely free of nuclear weapons.
Should it prove impossible to agree on legally defined nuclear-weapon-
free zones, states could, as an interim measure, pledge themselves not to
become the first to introduce nuclear weapons in the region. The nuclear-
weapon states would have t o guarantee the countries concerned that they
would not be threatened or attacked with such weapons.
6 Economic Security
The present condition of the world economy threatens the security of every
country. The Commission believes that just as countries cannot achieve
security at each other's expense, so too they cannot achieve security through
military strength alone. Common security requires that people live in dig-
nity and peace, that they have enough to eat and are able to find work and
live in a world without poverty and destitution.
6.1 The costs of military spending: Military competition reduces both mili-
tary and economic security. Military spending is part of the problem, not
part of the solution. The human cost of military effort has long been appar-
ent in a world where more than 1,000 million men, women and children
have n o chance t o learn t o read and write, and more than 600 million are
hungry or starving.
But the economic problems of the 1970s and early 1980s make the waste
of human effort even more intolerable. The presumed economic benefits of
military spending are a dangerous illusion. Increased military spending would
make our economic problems worse, not better. Military expenditure is likely
to create less employment than other forms of public expenditure, with greater
risks for inflation and for future economic growth. These dangers are exacer-
bated by the peculiar character of the modern military effort, with its increas-
ing emphasis in both developed and developing countries alike on expensive,
technologically sophisticated armaments. All but a very few countries now
face the most troubling choices in deciding how to spend their limited gov-
ernment revenues - on health programmes or on improving the lives of old
people, on unemployment benefits or on investment in economic growth and
development, on education or on foreign aid. The costs of military spending
must be counted in terms of these other opportunities forgone.
6.2 Disarmament and development: The link between disarmament and devel-
opment, in the new economic context of the 1980s, is close and compelling.
The 'crisis' in the world economy described by the Brandt Commission in

1980 has become even more serious. The military tensions analysed in the
present report have been a major contributory factor in making this crisis
worse. But the process of building common security could help to resolve it.
In the first place, for several developing countries, military expenditure, par-
ticularly on sophisticated imported weapons, threatens the economic develop-
ment which is the only basis for lasting security. In the second place, revenues
now used on the military could constitute a major source for increasing devel-
opment assistance by developed and capital-surplus countries. Some govern-
ments argue that they cannot increase or even maintain their foreign aid
because of competing domestic claims on government resources. These claims
are real and urgent. But even a tiny share of the expenditure currently going
to military purposes - about $650 billion a year - would go a long way
towards resolving the Third World's pressing needs. Third, reductions in mili-
tary spending would increase the prospects for resumed growth in the world
economy, and thus for worldwide economic security. Developing countries
need to import the goods and services that developed countries need to export.
Resources saved from the military could finance this expansion. We share the
view that such economic recovery is an essential investment in future security.
L.imiting military competition would have immense benefits for the secul--
ity of all countries; it would have economic benefits as well. Reductions in
military spending will provide resources to reduce poverty and increase
social wellbeing even in the richest military powers. They should also pro-
vide resources for development.
Schemes for linking disarmament and development will be different in
different countries and regions. In countries with large military expenditures,
they should take the form of releasing resources from defence budgets for
foreign development assistance. The main military powers spend from four
to over one hundred times as much on defence as on foreign economic aid.
A ten per cent cut in procurement by the nuclear powers alone would be
more than enough to double total foreign aid and other financial flows to
the thirty-one least-developed countries. Such rather mechanical calculations
would probably not lead to appropriate targets, although there is certainly
need for international cooperation in discussing the various possibilities for
verifying the switching of resources from the military to development.
It might be possible, instead, to devise targets described in physical terms;
countries might announce that they would use funds from their defence
budget to build a fertilizer factory, for example, or to contribute the services
of a hundred paramedical workers. It is up to the imagination of people in
each country to find ways to participate in such 'peace competition'.
6.3 Regional conferences on disarmament and economic security: It 1s essen-

tial that people and governments in all regions should participate In findmg
new resources for development. The Commission urges that one of the flrst
t o p ~ c sfor the Regional Conferences descr~bedin recommendation 5.1,
lncludlng the Conference on Security and Cooperanon in Europe, should be
disarmament and economic security. Countries should consider convening

a high-level conference to discuss common problems of economic security,
and their common interest in reducing the regional costs of military spend-
ing. Such a conference could provide an opportunity to inform people and
governments about the economic costs of military competition; to initiate
cooperation in providing information and analysis about military spending;
to initiate common efforts to achieve more security at less cost.
The Commission urges that the Regional Conferences launch major
campaigns t o increase public awareness o f the dangers of military competi-
tion, including the dangers for economic security. Such campaigns should
be an initial step in a continuing long-term public-education effort. Their
cost could be met with a small fraction of one per cent of regional military
expenditure. The United Nations should coordinate the efforts of regional
conferences and participate actively in the information campaigns.
The Commission finds it unacceptable that a substantial share of the
world's scientific potential be devoted to ever more refined forms of destruc-
tion, while our countries urgently need research into preventing and curing
disease, into new methods of food production, into alleviating the problems
of old people, and into preserving the physical environment. The Regional
Conferences should consider ways of converting to civilian uses the scien-
tific and technical resources now consumed for military purposes: from
research and development workers and facilities in developed and certain
developing countries to technicians with scarce industrial skills throughout
the world. The real social costs of devoting resources to military spending
vary greatly in different regions, and should accordingly be discussed at a
regional level. The Regional Conferences should propose detailed pro-
grammes to use military skills for urgent civilian needs in the particular
region. Such schemes should include national plans t o convert specific mili-
tary facilities - research establishments or other military installations - to
civilian purposes.
6.4 Common security and common prosperity: We share the conviction of

the Brandt Commission that the South and the North, the East and the West
have 'mutual interests' in economic progress. No country can resolve its
problems alone. A reduction in the present high levels of military spending
would therefore be in the economic interests of all countries, even those
who spend relatively little on their own military efforts.
The principle of common security asserts that countries can only find
security in cooperation with their competitors, not against them.
No country can hope to win military advantage by out-running its com-
petitor in an economically costly arms race. All countries are hurt by the
economic difficulties of the major economies. Common security is not only
a matter of freedom from military fear. Its objective is not only to avoid
being killed in a nuclear apocalypse, or in a border dispute, or by a machine
gun in one's own village. Its objective, in the end, is to live a better life: in
common security and common prosperity.
Notes
1. See Annexe Two: Comment by Egon Bahr.

2. Giorgi Arbarov expressed doubts about the arms control value of this proposal as
nuclear murutions could he quickly reintroduced Into the pro-scribed area. Such an agreement
which is o f small rn~litarysignificance would he difficult to negotiate, ,lnd could create an
unfounded impression o f enhanced security. In his opinion, other more effective measures are
needed - radical reductions up to a complete ban of all medium-range and t a c t ~ c dnuclear
weapons. This would amount t o a genuine zero-option for Europe.
3. Rohert Ford, David Owen, and Cyrus Vance comment :IS follows o n the Commission's
recommendations on enhanced rad~ationwcapons: We d o not advocate the deployment of
such weapons at this time. We consider, however, that both in their asserted benetits for rnili-
wry effect~venessand in their alleged adverse impact on the risk of nuclear war, any incre-
mental etfects of enhanced radiation weapons are relatively m n o r as compared t o the b a s ~ c
prohlems raised hy any nuclear weapon. A decision to initiate nuclear war, the most mornen-
tous deciswn any polit~calleader would ever confront, would not be made more easily or morc
quickly because enhanced radiation weapons, rather than nuclear wcapons of older design,
were available for uie.
4. J o o p deli Uyl endorses the proposal o f the Comrniss~onfor the gradudl removal of the
nuclear threat t o Europe. He maintains his conviction that an overall balance o t nuclear arms
does not require precise parity of nuclear weapons on every level and for every class of
weapon. Hc re'lffirnis his opposition to the stationing of new n ~ ~ c l e aweapon
r spterns in
NATO and Warsaw I'acr countries.
F. See F1tz01 Docrc~neizto f t / ~ Tenth
c Special Srssron o f the General AsseinOly, New York,
CJnitrd Nations, 1 972, AIRESIS-10 I 2 pp. 14-1 5; and Study or1 All t l ~ cAspects o/ Rqro11~11
Dmzrnmnent. New York, Un~tedNations, 198 1, A13514 I h pp. 15- 19.
Redefining Security
Richard H. Ullman
S
ince the onset of the Cold War in the late 1940s, every administration
in Washington has defined American national security in excessively
narrow and excessively military terms. Politicians have found it easier
to focus the attention of an inattentive public on military dangers, real or
imagined, than on nonmilitary ones; political leaders have found it easier
to build a consensus on military solutions t o foreign policy problems than
t o get agreement on the use (and, therefore, the adequate funding) of the
other means of influence that the United States can bring to bear beyond its
frontiers.
Even the Carter Administration, which set out self-consciously to depart
from this pattern, found in its later years that the easiest way to deflect its
most potent domestic critics was to emphasize those aspects of the dilemmas
it faced that seemed susceptible to military solutions and to downplay those
that did not. Jimmy Carter's failure to win reelection may suggest not that his
instincts in these respects were faulty but merely that his conversion
was neither early nor ardent enough.
Just as politicians have not found it electorally rewarding to put forward
conceptions of security that take account of nonmilitary dangers, analysts
have not found it intellectually easy. They have found it especially difficult
to compare one type of threat with others, and to measure the relative con-
tributions toward national security of the various ways in which govern-
ments might use the resources at their disposal.
The purpose of this paper is to begin to chip away at some of these ana-
lytical problems. It proceeds from the assumption that defining national
security merely (or even primarily) in military terms conveys a profoundly
false image of reality. That false image is doubly misleading and therefore
doubly dangerous. First, it causes states to concentrate on military threats
and to ignore other and perhaps even more harmful dangers. Thus it reduces
their total security. And second, it contributes to a pervasive militarization
of international relations that in the long run can only increase global
insecurity.
Source: International Security, 8(1) (1983): 129-53.
I I I I Redefining Security 297
Security versus What?
O n e way of moving toward a more comprehensive definition of security

may be to ask: what should we be willing t o give u p in order t o obtain more
security? how d o we assess the tradeoffs between security and other values?
The question is apposite because, of all the "goods" a state can provide,
none is more fundamental than security. Without it, as the 17th-century
philosopher Thomas Hobbes observed in a passage often cited but endlessly
worth recalling:
there 17 n o place for Industry, because the f r u ~ thereof

t is uncertam: and
consequently n o Culture of the barth, n o N a v ~ g a t ~ o nnor , use of the
commod~tlesthat may be Imported by Sea; n o c o m m o d ~ o u sB u ~ l d ~ n 110 g;
Instruments of movlng a n d removmg such t h ~ n g sas requlre much force,
n o Knowledge of the face of the Farth; n o account of T ~ m e n; o Arts; n o
Letters; n o Soc~ety;and w h ~ c h1s worst of all, contmuall feare, and dan-
ger of v ~ o l e n death;
t And the l ~ f eof man, ~ o l ~ t a r poore,
y, nasty, b r u ~ t ~ s h ,
and short.'
For Hobbes it did not much matter whether threats t o security came from
within o r outside one's sown nation. A victim is just as dead if the bullet that
kills him is fired by a neighbor attempting to seize his property as if it comes
from an invading army. A citizen looks t o the state, therefore, for protec-
tion against both types of threat.
Security, for Hobbes, was an absolute value. In exchange for providing it
the state can rightfully ask anything from a citizen save that he sacrifice his
own life, for preservation of life is the essence of security. In this respect,
Hobbes was extreme. For most of us, security is not a n absolute value. We
balance security against other values. Citizens of the United States and other
liberal democratic societies routinely balance security against liberty. Without
security, of course, liberty - except for the strongest - is a sham, as Hobbes
recognized. But we are willing t o trade some perceptible increments of secur-
ity for the advantages of liberty. Were we willing to make a Hobbesian
choice, our streets would be somewhat safer, a n d conscription would swell
the ranks of our armed forces. But our society would be - and we would
ourselves feel - very much more regimented.
The tradeoff between liberty and security is one of the crucial issues of our
era. In virtually every society, individuals and groups seek security against the
state, just as they ask the state to protect them against harm from other states.
Human rights and state security are thus intimately related. State authorities
frequently assume - sometimes with justification - that their foreign enemies
receive aid and sustenance from their domestic opponents, and vice versa. They
often find it convenient, in any case, to justify the suppression of rivals at home
hy citing their links to enemies abroad.
The most profound of all the choices relating t o national security is, there-
fore, the tracieoff with liberty, for a t conflict are t w o quite distinct values, each
essential t o human development. At its starkest, this choice presents itself

as: how far must states go, in order to protect themselves against adver-
saries that they regard as totalitarian, toward adopting totalitarian-like con-
straints on their own citizens? In the United States it is a tension that arises
every day in the pulling and hauling between police and intelligence agencies
and the Constitution. At a practical level, the choices become: what powers
d o we concede to local police? to the F.B.I.? to the C.I.A. and the other arms
of the "intelligence community"?
Other security choices may seem equally vexing if they are not equally
profound. One is the familiar choice between cure and prevention. Should
the U.S. spend a (large) sum of money on preparations for military inter-
vention in the Persian Gulf in order to assure the continued flow of oil from
fragile states like Saudi Arabia, or should it be spent instead on nonmilitary
measures - conservation, alternate energy sources, etc. - that promise sub-
stantially (although not rapidly) to reduce American dependence upon Persian
Gulf oil? A second choice involves collaboration with regimes whose values
are antithetic to America's own. Should the United States government forge
a relationship of greater military cooperation with the Republic of South
Africa, and risk racial conflict in its cities at home? O r should it continue to
treat South Africa as an international outlaw and perhaps enhance domestic
racial harmony - an important characteristic of a secure society - at the cost
of enabling the Soviet navy to pose a greater potential challenge to the safety
of the sea lanes around Africa upon which so much vital cargo flows? A third
choice involves military versus economic assistance to poor countries. Should
U.S. policy aim at strengthening Third World governments against the mili-
tary threats that they assert they perceive to come from the Soviet Union and
its allies, or a t helping their citizens develop greater self-reliance so as, per-
haps, ultimately to produce more healthful societies with lower rates of birth
and thus relieve the rising pressure on global resources? Finally, many
choices juxtapose international and domestic priorities. If a stretched national
budget cannot afford both increased outlays for military forces and for a
more effective criminal justice system at home, programs that create work
opportunities for poor inner-city teenagers, or measures to improve the qual-
ity of the air we breathe and the water we drink, which expenditures enhance
"security" more?
The tradeoffs implied in these and many other, similar questions are not
as profound as that-between security and liberty. But they-are nevertheless
capable of generating conflicts of values - between alternate ways of view-
ing national security and its relationship t o what might be called global
security.
There is, in fact, no necessary conflict between the goal of maintaining a
large and powerful military establishment and other goals such as developing
independence from Persian Gulf oil, promoting self-sustaining development
in poor countries, minimizing military reliance on repressive governments,
and promoting greater public tranquility and a more healthful environment
at home. All these objectives could be achieved if the American people chose
11 Iiin,ii: Redefining Security 299
to allocate national resources to do so. But it is scarcely likely that they - or

their Congressional representatives -will choose to make all the perceived sac-
rifices that such large governmental programs entail.
Indeed, the present Administration, supported by Congressional major-
ities, has embarked upon a substantial buildup of military spending while
at the same time reducing outlays - and perceptible concern - for the other
ohiectives listed here. Such policies are not merely neglectful of what some
writers have called the "other dimensions" of security. They sometimes cre-
ate conditions - increased worldwide arms expenditures, heightened intra-
regional confrontations, and greater fragility rather than resilience in Third
World governments - that make the world a more dangerous rather than a
safer lace. To use an image from the theory of games, there is a real dan-
ger that the policy choices of present and future U.S. administrations will
place us on a square on the game board in which all the players are worse
United States and some other nations more secure, or richer, while yet others are
left less well off. Instead, it might he "negative-sum," making all the nations
perceptibly less secure, with fewer disposable assets to spend on welfare
rather than on military forces.
To make this point is not to argue that a well-armed Soviet Union
increasingly confident of its abilities to project military power at long dis-
tances poses no potential threat to American security. Clearly it does. Nor is
it necessarily to argue (although I would do so) that much of what appears
threatening about recent Soviet behavior has its origins in Soviet responses
to American policies and force deployments. That is a topic for a separate
discussion.' But it is to argue that the present U.S. Administration - and, to
a substantial degree, its predecessor - has defined national security in an
excessively narrow way It happens also (as will be suggested later) to be a
politically quite expedient way.
A Redefinition of Threats
In addition to examining security tradeoffs, it is necessary to recognize that

security may be defined not merely as a goal but as a consequence - this
means that we may not realize what it is or how important it is until we are
threatened with losing it. In some sense, therefore, seci~rityis defined and
valorized by the threats which challenge it.
We arc, o f course, accustomed to thinking of national security in terms of
military threats arising from beyond the borders of one's own country. But
that emphasis is douhly misleading. It draws attention away from the non-
military threats that promise to undermine the stability of many nations during
the years ahead. And it presupposes that threats arising from outside a state are
somehow more dangerous to its security than threats that arise within it.
A more useful (although certainly not conventional) definition might
be: a threat to national security is an action or sequence of events that
(1)threatens drastically and over a relatively brief span of time to degrade

the quality of life for the inhabitants of a state, or (2)threatens significantly to
narrow the range of policy choices available to the government of a state or t o
private, nongovernmental entities (persons, groups, corporations) within the
state. Within the first category might come the spectrum of disturbances
and disruptions ranging from external wars to internal rebellions, from
blockades and boycotts to raw material shortages and devastating "nat-
ural" disasters such as decimating epidemics, catastrophic floods, or massive
and pervasive droughts. These are for the most part fairly obvious: in their
presence any observer would recognize that the well-being of a society had
been drastically impaired.
The second category is perhaps less obviously apposite. In considering it, it
may be helpful to reflect on the way in which the threat from Nazi Germany
to the United States was discussed in the years immediately preceding
American entry into World War I1 - or, indeed, the way the threat from the
Soviet Union has been viewed throughout most of the postwar era. Death and
physical destruction are, of course, one realization of the threat. They repre-
sent "degradation of the quality of life" in its most extreme form, and they
would be an inevitable result of war - even a war from which the United States
emerged victorious.
But suppose war had not come. Suppose Hitler's Germany or Stalin's
Russia had asserted domination over Western Europe and, perhaps, other
parts of the globe as well. The conquerors would have organized those soci-
eties in a manner that almost certainly would substantially have closed
them to the United States. That, of course, would have meant fewer oppor-
tunities for American traders and investors. But so, also, would there have
been fewer opportunities for unfettered intellectual, cultural, and scientific
exchange. And the extinction of civil and political liberty in countries which
shared our devotion to those values would have made it more difficult to
assure their preservation in an isolated and even besieged United States. In
a very large number of ways, the range of options open to the United States
government, and to persons and groups within American society, would
have been importantly diminished.
It is easy to think of degradation of the quality of life or a diminution of
the range of policy choices as "national security" problems when the source
of these undesirable conditions is a large, powerful, antagonistic state such
as Nazi Germany or Stalin's U.S.S.R. And it is even (relatively)easy to organ-
ize responses to such clear and present dangers. But it is much more difficult
to portray as threats to national security, or to organize effective action
against, the myriads of other phenomena, some originating within a national
society, many coming from outside it, which also kill, injure, or impoverish
persons, or substantially reduce opportunities for autonomous action, but
do so on a smaller scale and come from sources less generally perceived as
evil incarnate. Interruptions in the flow of critically needed resources or,
indeed, a dwindling of the available global supply; terrorist attacks or
restrictions on the liberty of citizens in order to combat terrorism; a drastic
1 111 .I I Redefining Security 301
deterioration of environmental quality caused by sources from either within

or outside a territorial state; continuing violence in a major Third World
state chronically unable to meet the basic human needs of large numbers of
its citizens; urban conflict at home perhaps (or perhaps not) fomented by the
presence of large numbers of poor immigrants from poor nations - all these
either degrade the quality of life and/or reduce the range o f policy options
available to governments and private persons.
For a leader trying to instill the political will necessary for a national
society to respond effectively to a threat to its security, a military threat is
especially convenient. The "public good" is much more easily defined; sac-
rifice can not only be asked but expected; particular interests are more easily
coopted or, failing that, overriden; it is easier to demonstrate that "business
as usual" must give way to extraordinary measures; dissent is more readily
swept aside in the name of forging a national consensus. A convenient char-
acteristic of military threats to national security is that their possible conse-
quences are relatively apparent and, if made actual, they work their harm
rapidly. Therefore, they are relatively noncontroversial.'
The less apparent a security threat may be - whether military or non-
military - the more that preparations to meet it are likely to be the subject of
political controversy. The American and the Soviet military establishments
are symbiotically allied in the effort to coax resources from their respective
political chiefs. Each regularly dramatizes (and surely exaggerates) the
threat posed by the other. The effects of such arguments within the Kremlin
are not easy to docunlent, but the evidence suggests that they are often per-
suasive. So are they generally persuasive for American Congressmen anx-
ious to demonstrate to their constituents that they are "pro-" national security.
The contrast with the generally unenthusiastic reception given to programs
aimed at aiding poor countries, ameliorating the disaffection of poor persons
at home, halting environmental degradation, stockpiling strategically import-
ant materials, or other such measures is striking but scarcely surprising.
Proponents of such programs in fact frequently do justify them on the ground
that they promote national security. But because their connection to security
is often not immediately apparent, opponents find it easy to reject or simply
ignore such argun~ents,if not to refute them.4
Preparing for Catastrophe
A comparison between American society's preparations for two events, each

carrying relatively low risks but each posing the threat of catastrophically
high costs, is instructive. One is nuclear war between the United States and
the Soviet Union. The other is a large earthquake along the San Andreas fault
that runs much of the length of the state of California. Nuclear war would
undoubtedly result in many more casualties and much greater damage, but a
major earthquake along the San Andreas fault, and the gigantic tidal wave
that would likely follow it, might well kill or seriously injure hundreds of
thousands of persons and cause billions of dollars of damage to property.

Certainly it would be devastating to regional, if not national, security.
Seismologists say that the probability of such an earthquake occurring within
half a century is relatively high, from 2 to 5 percent in any one year.5 The
odds that large-scale nuclear war will occur cannot be so confidently calcu-
lated, but they are surely much smaller.
Every year the United States government spends many billions of dollars
to build up nuclear forces whose purpose, at least according to strategic
theory, is t o make nuclear war between the U.S. and the U.S.S.R. less likely.
Americans regard that as a proper function of government. So, also, d o
most Americans probably regard the construction of shelters and other
facilities that might reduce the damage caused by nuclear war should it
occur. But administrations in Washington or in likely target states and
municipalities habitually spend very much less - indeed, quite small sums -
on such measures, and they spend even less on measures that might reduce
the damage from a catastrophic e a r t h q ~ a k e . ~
H o w can we explain these discrepancies? Regarding so-called "passive"
defenses against nuclear weapons (shelters and the like, as distinguished
from "active" defenses such as missiles to shoot down missiles), one explan-
ation is that the task seems too daunting, a quixotic effort given the size of
the attack the Soviet Union could launch. When scores of millions might be
killed, the prospect of saving tens of millions - as, indeed, a large-scale
effort at civil defense might make possible - seems heartening only to the
most zealous student of what has come to be called "comparative recovery
rates" between the U.S. and the U.S.S.R. And the cost of such a shelter pro-
gram would be enormous, very expensive insurance against a catastrophic
but very unlikely risk. Yet there is little doubt that it could (within these
macabre limits) be made effective.'
Against earthquakes, of course, shelters can offer little protection. The
danger to life and property along the San Andreas fault comes because
many hundreds of thousands of California residents have individually made
decisions to locate their homes and businesses there. In their view, the advan-
tages of cost or location outweigh the disadvantages of exposure to the risk
of major catastrophe. They might increase their own and their families'
chances for survival by strengthening existing buildings or replacing them with
more resistant structures. But the probability is that, owing to the geologic
properties of the San Andreas fault, an earthquake there would be so severe
that for many structures such measures would be ineffective. In such a situ-
ation governmental authorities can do little but monitor, warn, and make sure
that emergency facilities are on hand for the moment when a devastating
quake occurs. Alas, while federal and state agencies currently monitor seismic
events, they have done relatively little actually to prepare for the predicted
disaster. Yet there is no doubt that, should it occur, the consequences would
be extraordinarily dire.'
It scarcely needs stating that there are vast differences between
the threats to "national security" posed by nuclear weapons and those
1 il'rw,in Redefining Security 303
posed by catastrophic natural disasters. Nuclear wars, after all, originate

in human minds: other minds may therefore initiate actions t o affect the
adversary's calculations of costs and benefit, of risks and reward. Behind
earthquakes and floods are n o minds. They cannot be deterred. Rut their
potential damage can be substantially reduced by the application of fore-
sight and the expenditure of resources. Indeed, the probability that an incre-
mental expenditure on protection against earthquakes or floods will be
effective is surely very much greater than the probability that a compara-
ble incremental expenditure will enhance deterrence against nuclear war.
Yet Americans and their elected representatives are prepared t o acqui-
esce in - indeed, in some instances they show enthusiasm for - vast pro-
grams of weapons acquisition which, in the name of forestalling nuclear
war, have given the United States enough nuclear weapons t o exterminate
the world's population several times over. But the polity is ill-equipped
to make resource allocations that, dollar for dollar, would contribute a t
least as much to "security" as would the acquisition of the additional nuclear
weapons upon which the present Administration seeks t o spend many Idlions
of dollars.
The example of protection against earthquakes raises other interesting
points of comparison. While some community measures are useful, risk
aversion against such disasters is very largely in the hands of individuals.
Individuals can also affect a t least to some limited extent the degree t o
which they will be a t risk in the event of nuclear war. They can choose not
t o live in the vicinity of likely nuclear targets, and householders can provide
themselves with substantial protection against fallout and at least some pro-
tection against blast effects. Rut the pattern of a Soviet nuclear attack - and,
therefore, the location of likely danger - is very much more difficult t o pre-
dict than the danger zone of a major earthquake. And the opportunity costs
t o a citizen of choosing t o live in a place so remote that injury from nuclear
weapons effects are likely to be minimal are very much greater than the
costs of choosing not to live near the San Andreas fault or another area
of similarly great seismic instability, whose locations are all well known. In
addition, protection against nuclear weapons effects is much Inore a com-
munity matter than is protection against earthquakes. Particularly is this
true for residents of multiple-family urban dwellings. Only communities
can afford t o construct the deep, strong shelters that would offer city resi-
dents even a remote chance of surviving a nearby nuclear explosion."
The other nonmilitary security measures discussed thus far in this paper are
almost all considerably farther than protection against earthquakes toward
the community end of a spectrum running from the individual to the national
community."' Economic assistance to poor countries, programs to reduce
dependence upon Persian Gulf oil, military relations with repressive regimes,
efforts to combat air and water pollution, stockpiling of scarce resources, all
require either governmental allocation of resources or governmentally framed
policies and regulations. Like the acquisition and deployn~entof military
forces, they all depend upon organization to be effective; in a polity like the
United States, the impetus for such organization must come from government,
the ultimate wielder of carrots and sticks.
Indirect Threats: Conflicts over Territory a n d Resources
At the root of most of the violent conflicts in history has been competition for
territory and resources. The coming decades are likely to see a diminution in
the incidence of overt conflict over territory: the enshrinement of the principle
of national self-determination has made the conquest of peoples distinctly
unfashionable. But conflict over resources is likely to grow more intense as
demand for some essential commodities increases and supplies appear more
precarious. These conflicts will also have their territorial aspects, of course,
but the territory in contention is likely either to be unpopulated or only
sparsely populated. Much of it will be under water - oil-rich portions of the
continental shelves. Those parts above water will be the ostensible prizes,
often isolated or barren islands whose titles carry with them exclusive rights
to exploit the riches in and under the surrounding seas.
Such struggles over resources will often take the form of overt military con-
frontations whose violent phases will more likely be short, sharp shocks rather
than protracted wars. In most instances they will involve neighboring states -
Chile and Argentina, Iraq and Iran, Greece and Turkey, Morocco and Algeria,
China and Vietnam, and many others. Most will be in the Third World. None
is likely to involve the United States, although American firms - oil companies
and other resource-extracting enterprises - may well be caught up on either
side of a particular dispute. Thus, if national security is defined in conven-
tional ways this country's national security is not likely to be directly affected
by such disputes.' '
Their indirect impact upon American national security is likely to be
large, however. Supplies of essential commodities will be at least temporar-
ily disrupted. Local regimes may fall, their places taken by successors often
less friendly t o the United States. Outside powers hostile to American inter-
ests, such as the Soviet Union or Cuba, may intervene to support local
clients, placing pressure on Washington to launch (or at least organize)
counter-interventions. In some quite plausible scenarios Washington might
intervene to protect local clients whether or not Moscow or Havana were
involved. Those circumstances that might lead to a direct confrontation of
Soviet and American forces are, of course, the ones most dangerous to U.S.
national security. Luckily, they are also the least likely.
"Resource wars" (as some call them) have figured prominently in dooms-
day forecasts for more than a decade. But they are only one way - and not
the most important way - in which resource issues will impinge upon
national security in coming years. It will not require violent conflict for
resource scarcities t o affect the well-being - and the security - of nations on
every rung of the development ladder. In considering ways in which such
scarcities might affect national security, analysts should distinguish those
I I 118 Redefining Security 305
that arise from expansion of demand from those arising from restrictions
on supply.
Behind expanding deniand, of course, lies the continuing rapid growth in the
world's population. Specialists note that the rate of population growth has
not yet overtaken that of the globe's capacity to feed, house, and care for its
people." But that capacity is sorely strained. Moreover, global mechanisms
for distributing or for managing resources are not effective enough to pre-
vent local catastrophic failures or to prevent the consumption of some cru-
cial renewable resources at greater-than-replacement rates. Those resources
include tropical forests and other sources of fuelwood, fish stocks, the ozone
layer surrounding the earth, and the global supply of clean air and water.
Moreover, these problems are interconnected. Here is but one example: As
Third World villagers cut down more and more forests in their search for
fuelwood, the denuded land left behind is prey to erosion. Rains carry topsoil
away, making the land unfit for cultivation. The topsoil, in turn, silts up
streams in its path. Meanwhile, the fuel-short villagers substitute dung (which
otherwise they would use for fertilizer) for the wood they can no longer
obtain, further robbing the soil of nutrients and bringing on crop failures.
Unahle to sustain themselves on the land, many join the worldwide migration
from the countryside into the cities.''
That migration - caused by many factors - has given rise to an explo-
sive growth in the population of most Third World cities. Many are ringed
by shantytowns containing millions of squatters, a high proportion of them
unemployed, malnourished, and living in squalor. Under the weight of these
enormous numbers municipal services break down and the quality of life
for all but the very rich suffers drastically. Such cities are forcing grounds
for criminality and violence. Some suffer a breakdown of governmental
authority and become virtually unmanageable. Others are governable only
by increasingly repressive means that lead, in turn, to a decline in the per-
ceived legitimacy of the regime in power. Especially is this the case in
nations that are marked by ethnic or religious divisions. When the resources
of a nation are severely strained, those at the bottom of a social hierarchy
are quick to imagine - often with justification - that those who govern dis-
tribute the benefits at their disposal in ways that favor some groups at the
expense of others.
There is a widespread assumption that these are the circumstances from
which revolutions are born. In fact, there is little evidence that any recent
revolution except perhaps the one in Iran has had urban roots. Although
rapid population growth and its attendant miseries have certainly given rise
to conflicts, particularly along communal lines, the governing authorities in
most Third World countries have been able to contain them. Rather than
forging links among urban (and rural) dispossessed persons, recent arrivals
in Third World cities have tended to be overwhelmingly preoccupied with
retaining (and, if possible, expanding) whatever economic niches they have

been able to carve for themselves. They have thus far provided few recruits
for those who would organize revolutions, nor much in the way of troubled
waters in which outside powers might fish.14
First World governments and peoples might be advised not to take too
much comfort from this record. Although the consequences of explosive
Third World population growth and rapid urbanization have not yet been
felt much beyond their countries of origin, the strains on fragile political
structures will not ease before the end of the century, if then: the would-be
workers who will seek employment in the swollen cities of the Third World
during the 1990s have already been born. Even if these strains d o not give
rise to revolutions (and, perhaps, to foreign interventions), they are likely to
make Third World governments more militantly confrontational in their
relations with the advanced, industrialized states. And they will produce
multifold other pressures on the rich nations. For the United States, the most
directly felt pressure is that of would-be immigrants, some coming through
lawful channels, most coming illegally. The pressure is especially severe -
and probably increasing - from Mexico, but it comes from all over the
Caribbean and Central American region and from other continents as well. As
population growth in the poor countries hobbles economic development, the
gap in living standards between them and the rich countries is likely to con-
tinue to widen, and resentment of the rich - rich nations and rich persons -
will continue to grow. So will pressures for immigration. The image of
islands of affluence amidst a sea of poverty is not inaccurate. This image has
given rise to doomsday scenarios in which, several decades from now, the
poor will threaten the rich with nuclear war unless the rich agree to a mas-
sive redistribution of wealth." But even if these scenarios d o not eventuate
(and the superior destructive capabilities of the rich make such denouements
unlikely), the pressure engendered by population growth in the Third World
is bound to degrade the quality of life, and diminish the range of options
available, to governments and persons in the rich countries.
This paper is not the place for detailed discussion of ways to slow popu-
lation growth in the Third World, to help Third World countries absorb their
multitudes of new citizens, and to introduce order into their processes of
urban development. It is sufficient to say that most such ways involve trans-
fers of resources and expertise to Third World countries. The record of the
United States in these areas is generally abysmal: among the O.E.C.D. nations
it is near the bottom of the league tables with regard to official development
aid calculated on a per capita basis. Only in population programs has the U.S.
made a respectable effort.16 But U.S. programs to assist other nations to solve
their population problems are increasingly coming under attack from the
"right-to-life" movement in this country, many of whose supporters are in the
forefront of those pressing for large increases in military spending. They,
and the opponents of economic assistance in general, may someday pay a
significant price for their arbitrarily narrow definition of national security.
li I irii,ir~ Redefining Security 307
hreats from the Su
Population growth dominates the problem of rising worldwide demand tor

resources. Moreover, overall demand is rising even more rapidly than popu-
lation growth figures alone would indicate. Many developing countries
contain growing "modern" sectors, enclaves of affluence and higher living
standards that enjoy the same wasteful consumption patterns of the indus-
trialized world. That imposes yet additional strains on world resources. By
contrast, no single factor dominates the problem of constraints on resource
supplies. A crucial distinction is whether the resource in question is re-
newable, like forests or fish stocks or feedgrains, or nonrenewable, like
(preeminently) oil. A second crucial distinction is whether the resource is
becoming increasingly scarce through "normal" depletion or through efforts
by governments (or, indeed, private persons) artificially to restrict supplies
by means of boycotts, embargoes, cartel agreements, recovery limitations,
and the like. Supply constraints are most injurious when they are sudden.
For virtually every raw material there are substitutes with properties suffi-
ciently similar so that replacement is possible. But whether or not replacement
can take place without painful disruption depends upon whether the shortage
in supply of the original item was foreseen adequately far in advance to make
possible smooth adjustment.
The United States is in a particularly fortunate position. Study after
study in recent years has concluded that oil is the only comn~oditywhose
sudden cutoff would have a drastic effect on national welfare or on eco-
nomic activity. Indeed, the same applies in large measure to all of the
advanced industrialized market-economy states. Since most produce a con-
siderably smaller proportion of their domestic oil consumption than the
United States, most would find an oil cutoff even more disruptive.'' But
other essential imported materials for them, as for the United States, either
come from highly reliable suppliers - like-minded states - or from a suffi-
ciently diverse range of suppliers so that a boycott by one or more would
not impose really serious harm.Ix Regarding foodstuffs, the O.E.C.D. coun-
tries are for the most part well provided for. Collectively they produce large
agricultural surplusses." Individual O.E.C.D. states that import a high pro-
portion of their domestic food consumption - Japan is the most important -
need not worry about major disruptions of supply because their purchasing
power will give them first claim on world markets.
The problem is much more serious for Third World states. Many are not
able to feed then~selvesand find it difficult to pay for imported foodstuffs,
a difficulty compounded since 1973 by the rising cost of the oil they also
must import."' Food is indeed a weapon that can be wielded against them -
although the industrialized states are most unlikely to employ it. The much
more serious danger they face is their acute vulnerability to natural disas-
ters that may cripple their own food production or substantially reduce the
supply (and therefore raise the price) of foodstuffs on the world market.
As population growth brings more mouths t o feed, the situations of many

Third World states are likely to grow more and more precarious.
Demand and supply are always related, of course. One approach to the
resource problem is slowing the growth of demand by slowing the growth
of population. But supply-side measures are equally - . necessary. When the
too-rapid exploitation of renewable resources is viewed as a supply prob-
lem, the solution seems to lie in creating mechanisms for effective regula-
tion of the rate at which fish are caught, forests are cut, seed crops are
harvested for food, and effluents are released into streams and emissions
into the atmosphere. Sometimes the nation-state is the appropriate arena
for such regulatory activity. In other instances, international mechanisms
("regimes," in the current academic jargon) are required. Such measures are
likely t o be really effective, however, only when they are combined with
efforts to slow the growth of demand. Moreover, as noted earlier, increas-
ing demand for many commodities is a product not merely of population
growth, but of rising affluence. And rising affluence is often not accom-
panied by rising sensitivity to the need for resource management, and the
appropriate technical and political skills t o make management possible.
As indicated above, one way to cope with depleting supplies of any
commodity is to find substitutes for it. That applies even to some renew-
able resources - although not, of course, to clean air and water. It applies
more obviously to nonrenewable resources. For minerals and fuels, a sensible
strategy is to create stockpiles that make it possible to cope with short-run
interruptions of supply while developing substitutes to cope with long-run
inevitable depletion.
These are scarcely difficult principles t o grasp. What is difficult is to per-
suade governments to allocate funds to put the principles into practice.
Especially for powerful countries like the United States that are used to get-
ting their way in the world, it seems easier to arouse the political will to
respond to a supply disruption with military means than to forestall the dis-
ruption in the first place by fostering alternate sources of supply, or by
developing substitutes for the resource whose supply is threatened.
Assessing Vulnerability
In every sphere of policy and action, security increases as vulnerability

decrease^.^^ At the most basic level of individual survival, this is a law of
nature, seemingly as well understood by animals as by humans. At that level
it is a reflexive response. Reducing vulnerability becomes a matter of policy,
rather than of reflex action, when it seems necessary to calculate the costs
and benefits involved. How much security d o we buy when we expend a
given increment of resources t o reduce vulnerability? That is a difficult ques-
tion even in relatively simple situations, such as a householder stockpiling a
commodity against the possibility of a disruption in accustomed channels of
supply. At the level of the community, rather than the individual, it becomes
I I Redefining Security 309
very much more difficult: different members assess risks differently, and they
may well be differently damaged by a disrupting event. An investment in
redundancy that seems worthwhile to one family may seem excessively costly
to another. Neither will know which is correct unless the crunch actually
comes. And even then they might disagree. They might experience distress
different1y.
At the level discussed in this paper, where states are the communities
involved and where the problems are for the most part considerably more
complicated than a simple disruption in an accustomed channel of supply, the
relationship between decreased vulnerability and increased security is tormid-
ably difficult to measure. Consider even the relatively simple measure of
adding crude oil to the US. Strategic Petroleum Reserve, the (for the most
part) underground stockpile whose purpose is to make it possible for the
nation to ride out a cutoff in deliveries from one or more major foreign oil
suppliers. We know, of course, the cost of buying and storing a given incre-
ment of crude oil. But until mid-1981 the government of Saudi Arabia (the
world's major exporter of oil) took the position that U.S. stockpiling of oil was
an unfriendly act. It claimed that it maintained high levels of oil production to
provide immediate benefits - "moderate" prices - t o Western (and other) con-
sumers, not to make it possible for Washington to buy insurance against the
day when the S a ~ ~ leadersh~p
di might want to cut production so as, say, to
influence U.S. policy toward Israel. Successive administrations in Washington
have regarded the retention of Saudi good will as something close to a vital
American interest, on both economic and strategic grounds. They therefore
dragged their feet on filling the Strategic Petroleum Reserve."
Who can say with assurance that those administrations were wrong?
Who could measure - before the event - the effects of putting Saudi noses
out of joint? It may well have been that even so seemingly modest a meas-
ure as adding to the oil stockpile would ripple through Saudi and Middle
Eastern politics in such a manner as ultimately to bring about just that
calamity against which the stockpile is intended to offer insulation, that is,
a production cutback. Moreover, being finite in size, the stockpile may not
offer sufficient insulation against a protracted deep cutback. But, by the
same token, who can be sure that even if the reserve remains unfilled (its
level is still far below the total originally planned)," and even if the United
States takes other additional measures to mollify the Saudis, an event will
not occur that will trigger a supply disruption in any case? If that occurs,
the nation would clearly be better off if it possessed a healthy reserve of
stored oil, even one insufficient to cushion the entire emergency.
Ever since the OPEC embargoes of 1973-74, Western governments have
been extremely sensitive to any hint of a further cutoff of oil or, for that
matter, of other, less critically needed resources. It is not surprising that
many analysts both in Washington and in other NATO capitals interpreted
the Soviet invasion of Afghanistan at the end of 3 979 not simply as Moscow's
ruthless effort to handle a local political dilemma but as the start of a Soviet
march toward the Persian Gulf. Since then, both the Carter and the Reagan
Administrations have regarded raising a robust combined-arms military

force earmarked for Gulf contingencies - the so-called Rapid Deployment
Force - as the most appropriate and, not so coincidentally, also the politically
most saleable response to the threat of instability in the Gulf.
Yet there is wide agreement among specialists that additional overt Soviet
border-crossing aggression in the Middle East is an unlikely contingency. Far
more likely is the coming to power in a major oil-producing state like Saudi
Arabia of a militantly anti-Western regime that might restrict production.
Against such an eventuality the Rapid Deployment Force offers little insurance,
for there would be great resistance in Congress and in the public at large to
any Presidential use of American forces for intervention in the turbulent
internal politics of the region.
It requires a long and more relaxed view to deemphasize military inter-
vention as an instrument of policy, however. And a longer view is much
more possible under conditions of reduced vulnerability. Then the occupant
of the Oval Office would be more likely to feel that he really has the option
of allowing the politics of regions like the Middle East to run their course.
Were the United States less vulnerable to interruptions in the supply of the
region's oil, administrations might find they had a wider range of options
for pursuing other interests, such as protecting communication routes or
the independence of Israel. Communications routes, for instance, can be
protected at many points. And the American commitment to Israel would
cost less if the U.S. were not simultaneously supplying some of Israel's ene-
mies with the most potent weapons in its inventory and then giving the
Israelis additional weapons to offset them.
As this paper has suggested, many of the conditions that may most affect
U.S. security have their origins in circumstances that have little or nothing
to d o with the rivalry between the United States and the Soviet Union. Yet
many of them, if not managed, have the potential to give rise to crises
between the superpowers as one or the other intervenes to secure resources
or to support its clients in a domestic or regional conflict in the Third
World. For crisis prevention, if for no other reasons, political leaders in
Washington - and in Moscow, too - should pay heed to these condition^.'^
There are, of course, other reasons. To the extent that the quality of life
in the United States is degraded by resource scarcities and by the deterior-
ation in the quality of life beyond its borders, Americans should be con-
cerned. That is but the counsel of prudence. Focussing attention on these
"other dimensions of security" will require political leadership of the high-
est order, however. Morever, it will require far greater consensus than now
exists regarding what is to be done.
The absence of consensus is, indeed, a formidable obstacle. There is
no agreement within the American policy community regarding ways of
coping with resource scarcities or with the problems of poverty and explo-
sive population g o w t h in the Third World. The Administration currently
in Washington is ideologically committed to market solutions in virtually
every sphere of policy. Thus, rather than develop government stockpiles of
tIliiri,iii Redefining Security 311
oil and other scarce resources it prefers to leave the task to private entities.
Indeed, so opposed is the Reagan Administration to governmentally directed
resource management that it has even encouraged the depletion of the largest
oil stockpile it itself owns, the oilfields set aside as so-called Naval Petroleum
reserve^.^'
The same is true for investments in alternate energy sources. The Adrnin-
istration has drastically reduced federal allocations for energy research and
development of all sorts. Nuclear fusion, solar energy, unconventional oils -
all have had their appropriations sliced. (Only the Clinch River hreeder
reactor, a project in the home state of the Republican Senate majority
leader, has been spared.)lh Not surprisingly, in an economic climate marked
by both recession and high interest rates, the private sector shows few signs
of acting upon the Administration's preferences, ideologically congenial
though they may be. Despite bargain prices, there has been little stockpiling
of commodities. And, with a worldwide oil glut, the private sector has shown
n o inclination to invest in energy alternatives.
Opponents of the Administration's position assert that, regardless of the
economic climate, the marketplace is incapable of adequately discounting
scarcity. Therefore, they argue, the intervention of a single, authoritative
actor - by definition, the federal government - is required to build up stock-
piles and to fund research and development activities that are not likely to
pay off within commercially acceptable timeframes."
Measuring Security
That intervention will necessarily give rise to what appear to be inefficien-

cies. They will appear so because it will be possible to compare the costs of
resources stockpiled, or developed by new production techniques, with the
costs for the same or similar commodities bought on the market. Usually -
unless there has been an intervention of a different sort, such as an embargo
by suppliers - the costs of stockpiles or substitutes will be higher. It is easy
to quantify these so-called inefficiencies. And once quantified, they are easy
to decry. On the other hand, it is much more difficult to assign a weight to
the security that the community may have purchased by sustaining them.
It is at least as difficult, however, to assign a weight to the quantity of secur-
ity that the community purchases by a given investment in military hardware
or in manpower. A missile or a tank or an infantry battalion that never enter
combat are like commodities p~~rchased for a stockpile. They also are ineffi-
ciencies. Yet we less often look at military purchases that way. We do, of course,
incessantly decry "waste and inefficiency" in the armed services and in the
defense industries. But we usually mean that better management could have
purchased comparable military capability for less money. Rarely do we ask
whether the possession of that particular capablity is in itself "efficient."
That is not to say that we do not often compare military with nonmilitary
expenditures. Indeed, such comparisons are a staple of political discourse.
Someone points out that for the price of, say, one Navy F-14 fighter it would
be possible to build a certain number of daycare centers or black-lung clin-
ics for the mining towns of Appalachia. And we know that, unlike the F-14,
the centers or clinics would be "used" (indeed, we hope the F-14 will never
enter combat). Moreover, we know quite precisely how much welfare we
purchase with a childcare center or a clinic. We can quantify it in terms of
children attending (and mothers working) or patients treated. But at that
point the comparison between guns and butter ends. We can weigh American
forces against Soviet forces, and we can compare the capabilities of one
weapons system against another. But we cannot really quantify the security
we buy with the funds we spend on an F-14 or, indeed, on an entire carrier
task group. We assume that the task group will deter hostile actions by
unfriendly nations. But it may be that a smaller American Navy will deter
them equally well, and a carrier air wing minus one F-14 may be fully capable
of meeting all the threats that ever come against it.18
This discussion has sought to show that we generally think about - and,
as a polity, dispose of - resource allocations for military and for nonmilitary
dimensions of security in quite different ways. Regarding military forces,
although analysts and interest groups may have their own ideas about such
issues as the appropriate size of the American fleet or the composition of its
air wings, there is general agreement on the principle that there must in the
end be a single, authoritative determination, and that such a determination
can come only from the central government of the polity. Because we
acknowledge that there is no marketplace in which we can purchase mili-
tary security (as distinguished from some of its components), we would not
look to private individuals or firms or legislators or regional governments
to make such a determination, even though we might disagree with the
determination that the federal government makes.
By contrast, as indicated above, there is no consensus about the need for
a single, authoritative determination regarding the nonmilitary dimensions
of security. The polity as a whole is therefore much more responsive to alle-
gations that a given investment in, say, a commodity stockpile is "inefficient"
than it is responsive to the same allegation regarding a given investment in mili-
tary forces. Moreover, the alleged inefficiency is far more easily demon-
strated. The situation is similar regarding measures for coping with the other
problems mentioned in this paper: rapid population growth, explosive urban-
ization, deforestation, and the like. Here, also, the current American
Administration - and much of the public - is committed to "efficient7' market-
place solutions rather than to solutions involving international regimes or
governmentally sponsored transfers of resources.
Changing t h e Consensus
Because of these preconceptions regarding the appropriate role of govern-

mental authority both in defining problems and in proposing solutions, the
II I Redefining Security 3 13
tendency of American political leaders to define security problems and their

solutions in military terms is deeply ingrained. The image of the President
as Commander in Chief is powerful. When in this role he requests add-
itional funds for American military forces the Congress and the public are
reluctant t o gainsay him. When he requests funds for economic assistance
to Third World governments, he is much more likely t o be disputed even
though he may contend that such expenditures also provide the United
States with security.
Altering that pattern will require a sustained effort at public education. It
is not an effort that administrations themselves are likely to undertake with
any real commitment, particularly in times when the economy is straightened
and when they find it difficult enough to find funds for the military goals they
have set for themselves. The agents for any change in public attitudes are
therefore likely to be nongovernmental.
Over the past decade or so a vast array of public interest organizations
have begun to put forward alternate conceptions of national security. Nearly
all are devoted t o particular issues - limiting population growth, enhancing
environmental quality, eradicating world hunger, protecting human rights,
and the like. Some are overt lobbies expressly seeking t o alter political out-
comes. Others devote themselves to research and educational activities, but
are equally concerned with changing governmental behavior. Jointly they have
succeeded in substantially raising public awareness of the vulnerability of the
society to a variety of harms nonmilitary in nature, and of the limitations of
military instruments for coping with many types of political problems.
One should not overestimate the achievements of these nongovernmental
organizations, however. Awareness on the part of a substantial informed
minority is one thing. Embodying it in public policy is a very much larger step.
A society's consciousness changes only gradually - usually with the change of
generations. The likelihood is that for the foreseeable future the American
polity will continue to be much more willing to expend scarce resources on
military forces than on measures to prevent or ameliorate the myriad pro-
foundly dislocating effects of global demographic change. Yet those effects are
likely to intensify with the passage of time. Problems that are manageable
today may prove far less tractable in the future. And while political will and
energy are focussed predominately on military solutions to the problems of
national security, the nonmilitary tasks are likely to grow ever more difficult
to accomplish and dangerous to neglect.
Notes
1. Thc Leviuthaw ( 16.5I ) , Part I, Ch. XIII.

2. There is n o hetrer to hegin that discussion t h a n Robert Jervis, Pcrccptmn nrrd
Mlspcrc-eptron rn lnterrz~ztionnl Polltics (Princecon, N.J.: Princeton IJniversity Press, 19761,
chapter 3.
3. Thi\ is not t o say that there a r e n o t recrinimativns following wars o r nr~l~r,iry
crises.
Indeed, the govcr~imentst h ~ leadt nations when w a r is thrust upon them - o r when they
initiate war themselves - are often subject to pillory. It may be alleged that their complacence
allowed their nations' defenses to atrophy to a point where their military forces no longer
deterred attack. O r they may be accused of recklessness that brought on a needless and expen-
sive war. But while the war is still in prospect, or while it is actually underway, there are too
seldom any questions of leaders' abilities to command the requisite resources from their per-
ceptibly threatened countrymen.
4. The same is true, it should be noted, about some "ordinary" foreign threats. In 1975
a majority of Senators and members of Congress did not believe that the presence of Soviet-
supported Cuban troops in Angola posed a significant threat t o U.S. security, and legislated
limits on potential American involvement. Three years earlier they imposed a cutoff on U.S.
bombing of targets in Cambodia and North Vietnam on the supposition that continued bomb-
ing would no longer (if it ever did) promote US. security. For a discussion of these
Congressional curbs on the President's ability to commit American military resources, see
Thomas M . Franck and Edward Weisband, Foreign Policy By Congress (New York: Oxford
University Press, 1979), esp. pp. 13-23 and 46-57.
5. For a recent authoritative study, see An Assessment of the Consequences and Preparations
for a Catastrophic California Earthquake: Findings and Actions Taken (Washington: Federal
Emergency Management Agency, 1980). For a summary of current estimates, see Richard
A. Kerr, "California's Shaking Next Time," Science, Vol. 215 (January 22, 1982), pp. 385-387.
6. The Federal Emergency Management Agency's (FEMA) fiscal year 1983 appropriation
for civil defense was $147,407,000; for "comprehensive emergency preparedness planning"
for earthquakes it was $3,120,000. California's total budgeted expenditure for earthquake
safety for fiscal year 1983 was $13,391,000. For a detailed breakdown, see State of California,
Seismic Safety Commission, Annual Report to the Governor a n d the Legislature for July
1981-June 1982 (Sacramento: August 1982), pp. 16-21.
7. The "classic" appeal for a large U.S. civil defense program, based upon hypothes~zed
comparative U.S. and Soviet recovery rates, is T.K. Jones and W. Scott Thompson, "Central War
and Civil Defense," Orbis, Vol. 22, No. 3 (Fall 1978), pp. 681-712. For a more recent dis-
cussion, see Robert Scheer, With Enough Shovels: Reagan, Bush a n d Nuclear War (New York:
Random House, 1982), pp. 104-1 19.
The enormous cost is one principal argument against a large-scale U.S. civil defense pro-
gram. But another relates to strategic doctrine. A civd defense program that promises to offer
effective protection might in a crisis inv~tean enemy first-str~keattack. The adversary, so thls
reasoning runs, would read large-scale civil defenses as indicating that we ourselves were pre-
pared t o initiate nuclear war. It would therefore strike at the first sign that we were beginning
to move our population into shelters, as we surely would during a severe international crisis.
Thus we enhance stability by not opting for civil defenses: the other side knows that since our
population is exposed, we would not be likely to initiate nuclear war, and the incentives for
them to strike preemptively are thereby reduced.
8. The FEMA study cited above (note 5 ) estimates that the likely damage from the most
probable (but far from the most destructive) major earthquake on the San Andreas fault might
be $17 billion, but it indicates that the figure might be low by a factor as high as three (p. 22).
9. The most authoritative generally available projection of the effects of a variety of types
of Soviet nuclear attacks on the United States is The Effects of Nuclear War (Washington,
D.C.: Congress of the United States, Office of Technology Assessment, 1979).
10. It should be noted that the currently preferred mode of avoiding nuclear war
(as distinguished from diminishing the likely effects of nuclear war) is at the far end of this
spectrum: the maintenance of a deterrent nuclear striking force is preeminently a national
responsibility - one, incidentally, beyond the grasp of all but the wealthiest nation-states.
Other modes of avoiding war, such as negotiation and d~sarmament,are also endeavors w h ~ c h
only duly legitimate national authorities, as distinguished from sub-national groupings or private
individuals, can undertake. Earthquakes differ from nuclear war in that thev cannot be either
deterred or forestalled. But societies can protect against their effects. That is why, despite obvious
differences, the comparison with nuclear war as a threat to soc~etalsecurity seems instructive.
IJllrn,iti Redefining Security 3 15
11. For a discussion of the kinds and scope of dlsputes that are likely to arise, see Ruth
W. Arad arid Uzi 8. Arad, "Scarce Natural Resources and Potential Conflict," in Arad et a/.,
Shurrn~G k ) h d Resourcas, 1980s Project/Council on Foreign Relarmns (New York: R.lcC;raw-
Hill, 1979), pp. 2.5-104.
12. See the tables in the statistical annexes to Roger D. Hansen et al., U.S. Foreign Policy
(2nd the Thrrd World: Agenda 1982, Overseas Development <:ouncil (New York: I'raeger
Publishers. 19821., esn. , tables U-8 and C-I.
13. For a discussion that brings out the seamless nature of this problem, see Lester K. Brown,
"World Population Growth, Soil Erosion, and Food Security," Science, Vol. 2 14 (November 27,
19811, pp. 995-1002.
14. For a thorough survey o f extant social science research on Third World urban growth and
its relationsh~pto political ~nstabillty,see the unpublished paper by Henry Bienen, "Urbanization
and Third World Stability," Research Program in Development Studies, Woodrow Wilson
School, I'r~nceton University, December 1982.
IS. For a prototypical example, see Robert L. Hellbroner, An Inquiry into the Hrlmar~
Conditron (New York: W.W. Norton, 1975), esp. pp. 4 2 4 . 5 . For a provocative var~ation,see
McCeorge Rundy, "After the Ileluge, the Covenant," Saturday Revieu~lWorltf,August 24,
1975, pp. 18-20, 1 12-1 14.
16. For the 0.E.C.D. rankings, see Hansen, Agenda 1982, tahle F-8 and figure F-18. For
population programs, see D,ina Lewison, "Sources of Population and Family Plannmg
Assistance," Populi~tionReports. VoI. I I, No. 1 (January-February 1983).
17. See David A. Deese and Joseph S. Nye, eds., t~rrergyand Security (Cambridge, Mass.:
liall~nger,I98 1 ), csp. pp. 13 1-228 and appendix R, "Worldwide Product~onand lJse of Crude
Oil."
18. See the well-documented discussion In Arad, "Scarce Natural Reso~~rce\," pp. 32-59.
For 3 widely cited earlier statement, see Stephen D. Kraner, "Oil is the Except~on,"Foreigtz
Polic-): No. 14 (Spring 19741, pp. 68-84. John E. Tilton, The trit~rreof N o t r f ~ dMin~rals
(Washungton: The Brookings Institution, 1977), reaches the same conclus~ons.
19. A concise survey of global patterns of food production m d consumptloll is in Paul
R. Ehl-lich, Anne H. Ehrlicb, and J o h n I? Holdren, .k~screncc~: Popdi~tion,Reso~irccs,Eruiron-
rirent (S.ln Franc~sco:W.H. Freeman, 1977), pp. 284-297. For a current accounting hy a IJ.S.
A g r ~ c ~ ~ l t uDepartment
rc ofticul, see Terry N. h r r , "?'he World Food Sit~lat~on , ~ n dC;lotxiI
Grain I'rospects," Scrence, Vol. 2 14 (November 27, 198 I), pp. 1087-1095.
20. See Deew and Nye, Ener<yyund Security, pp. 229-58.
2 1. Some ni~ghtargue that this 1s not the case in the strategic nuclear re1,itionship between
the Unitcd States and the Sowet Union, and that ~t is the knowledge w i t h ~ neach government
r h ~ ~t t ssociety 1s highly vulnerable to nuclear attacks by the other that keeps it from ever
I.iunchlng such a n attack itself. Security 15 thus a product of v~ilticr,lhil~ty.'This argument ha\
cons~der,ibleforce as a logiccd construct. Yet, not surprisingly, neither superpower is content
to act upon ~ t As . technological developments seem to make possible the I~mitattonof damage
from 'it le,ist some torms of nuclear attack, each pursues them for fear th'lt the other will
secure ;I momentary adv,lntage. We are therefore faced with the worst of situations, In whlch
o ~ i cor the other may be unduly optimistic regarding the degree to which it might I~mitdarn,lge
to its obvn society if it were to srr~kefirst. Decreased vulnerahilit) rzcct~ratelyassessed m'iy well
enhance security even in strategic nuclear relations; misleadingly assessed it may bring disaster.
22. See, e.g., W'llter S. Mossherg, "Kowtowing on the Oil Reserve," T / J ~VIrill/ Street
/oitrnal, May 14, 1980, p. 20, and Sheilah Kast, "Filling Our Strategic Oil Reserve,"
Washrngton S t u , February 9, 198 1, the latter quoting Secretary-of-State-ciesign'lte Alexander
M. F-la~g,Jr., as calling the Saudi positwn "oil hlackma~l."
23. The Energy Infornlation Adrnin~stratlon'sMonthly Emrg?~Rel~ieiu(Washington: IJ.S.
Department of Energy) presents a running tally of the slzc o f the Strategic Petroleum Reserve.
For a technical account of how the reserve is maintained, see Kuth M. Davis, "Nation,ll
Strategic Petroleum Reserve," Sc-iencc,Vol. 2 13 (August 7, l 9 8 l ) , pp. 6 18-622. See also Deese
and Nye, Energy anti Security, pp. 326-328, 399-40.3.
24. For an excellent discussion of the genesis and prevention of superpower crises, see
Alexander L. George, Managing U.S.-Soviet Rivalry: Problems of Crlsis Pre~nention(Boulder,
Colo.: Westview Press, 1983).
25. Richard Corrigan, "Three Bowls of Oil," National~ournal,December 5, 198 1, p. 2 167.
26. See these articles by Richard Corrigan, the National !ournalk energy correspondent:
"The Next Energy Crisis: A Job for the Government o r the Free Market?." June 20, 1981,
pp. 1106-1 109; "On Energy Policy, the Administration Prefers to Duck, Defer and Deliberate,"
July 18, 1981, pp. 1280-1283; and "Down for the Count," May 22, 1982, p. 919.
27. Corrigan, "Energy Policy," National!ournal, July 18, 1981, p. 1283.
28. Part of the difficulty of comparing guns and butter may arise from the fact that pollties
demand different orders of satisfaction from the evaluation of the two. Regarding daycare cen-
ters or clinics, officials often feel satisfied when they can certify that services of a given quality
have in fact been delivered. They seldom feel it necessary to ask whether their delivery has really
enhanced the welfare of the community, the nation, or the world: they regard the question as
either self-evident or as impossible to answer. But publics have come t o demand more of
accountings for military expenditures. After Israel's sweeping victories in Lebanon in 1982 it
was not enough to ascertain that the American-armed Israeli forces had decisively defeated the
Soviet-armed Syrians and Palestmians, nor even that the campaign had vastly enhanced Israel's
short-run security. Observers asked - and regarded the question as entirely appropriate -
whether it had really enhanced Israel's long-run security.
For a discussion of assessing the benefits of welfare programs, see Alice M. Rivlin, Syste?natir
Thinking for Social Action (Washington: The Brookings Institution, 1971), pp. 46-63.
Security in the Third World: The Worm About to Turn?
Mohammed Ayoob
I
11 the literature on international relations, the term 'security' has trad-
itionally been defined t o mean immunity ( t o varying degrees) of a state
o r nation t o threats emanating from outside its boundaries. In the words
of Walter Lippmann, 'a nation is secure t o the extent t o which it is not in
danger of having to sacrifice core values, if it wishes to avoid war, and is
able, if challenged, to maintain them by such victory in such a war'.'
According to Arnold Wolfers, Lippmann's definition 'implies that security
rises and falls with the ability of a nation to deter an attack, o r t o defeat it.
This is in accord with the common usage of the term." Expanding on the
concept o f security as protection of core values, in the context of small
Third World states, Talukder Maniruzzaman has stated that: 'By security
we mean the protection and preservation of the minimum core values of
any nation: political independence and territorial integrity." This position
is one that we can readily identify as the realist position in the Western and
Western-influenced literature on international relations.
However, some authors have differed significantly from this exclusively
state-centric realist perspective. They have viewed the problem of security
from the perspective of the international system and have focussed on what
has, of late, come to be called international security. By adopting this system-
centred perspective they have tried t o mitigate some of the more Hobbesian
characteristics of the realist position. They have taken their cue from views
such as those expressed by Martin Wight, who had argued that 'if there is
an international society, then there is an order of some kind t o be main-
tained, o r even developed. It is not fallacious to speak of a collective inter-
est, and security acquires a broad meaning: it can be enjoyed o r pursued in
common. Foreign policy will take some account of the common interest. It
becomes possible t o transfer to international politics some of the categories
of constit~tionalisrn.'~
Proponents of the system-oriented approaches to security have taken the
society of states ('anarchical' though it may be, t o use Hedley Bull's phrase)'
as a relevant object of security. They have argued that the security of the
Source: Intcrnatlonal A f f n ~ r s ,60( 1 ) ( 1984): 4 1-5 1

parts of the system is inextricably intertwined with that of the whole.

Indeed, the earliest of modern systemic analysts of security - the idealists of
the inter-war period - refused to distinguish the security of the parts from
that of the whole system. The post-Second World War breed of system-
centred scholars has been more discriminating than its predecessors. These
have argued from the assumption that the various segments of the inter-
national system are so far interlinked that they are 'interde~endent'.~ While
much of the initial impetus for this line of argument came from the eco-
nomic problems facing the Western industrialized states from the early 1970s
onwards, the increasingly awesome concentration of weaponry (particularly
nuclear weaponry) in the hands of the two superpowers gave further strength
to the 'interdependence' argument by overlaying it with such emotion-
charged slogans as 'cooperation or extinction', thus making the physical as
well as the economic security of the international system the object of this
intellectual exercise.
However, what is most interesting for our purpose is the fact that all
these contending schools of thought tended to define the concept of 'secur-
ity' in external or outward-directed terms - that is, external to the com-
monly accepted unit of analysis in international relations: the state. This
definition and the process by which it was reached were both understand-
able and ethnocentric in character. They were understandable because they
were the products of a particular intellectual tradition which faithfully
reflected a particular process of historical and political development which
could be traced back at least to the Peace of Westphalia if not to an earlier
period. Moreover, between 1648 and 1945 the natural development of the
'system of states' symbolized by, if not established at, Westphalia, and its
interaction with the domestic political processess of the major European
powers, led to the legitimation both of the system and of the individual
participants therein. These two trends - of interaction among sovereign
states on the one hand and greater identification of individuals with their
respective states on the other, the latter helped along tremendously by the
increasing correspondence between national and state boundaries and by
the increasingly representative character of the governments concerned -
therefore strengthened each other and in doing so firmly laid the founda-
tions of the intellectual tradition in which, at least in terms of political
analysis, the security of individuals and of groups came to be totally sub-
sumed within the category of state security. Therefore, while there has been
much debate about the compatibility or incompatibility of state security
and the security of the international system and on the measures that could
be taken to reconcile the two demands, the security of units below the level
of the state has rarely, if ever, been an important point at issue in Western
discussions and analyses of the concept of security.
Developments since 1945 have strengthened these Western notions of
what security is all about. In dividing the Western world into two halves and
in stabilizing that division by means of a mutual balance of terror, the cold
war (and its latter-day manifestation, detente) has frozen the predominant
1 t Security in the Third World: The Worm About to Turn? 3 19
Western connotation of 'security' in a hi-polar mould. The concept of 'alliance

security' - whether of the Atlantic alliance or of the Warsaw Treaty Organ-
ization - has therefore been superimposed on the concept of national or state
security, while its essential externally-directed thrust remains unchanged.
Moreover, by making the security of the major Western states, whether cap-
italist or socialist, the central concern of the security of the international sys-
tem as a whole (by the mutual assured destruction doctrine and because the
security of their European allies is considered by both the United States and
the Soviet Union as indispensable to their own security) the dominant strand
in Western strategic thinking has increasingly obliterated even the distinction
between the 'realist' (state-centric) and 'idealist' (system-centric) approaches
to the study of international security, indeed to the study of international rela-
tions as a whole.'
Third World 'Insecurity': The Worm Within
To turn to the Third World, however, is to find a different situation al-

t ~ g e t h e rThe
. ~ three major characteristics of the concept of state or national
security in Western states - namely, its external orientation, its strong link-
age with systemic security and its binding ties with the security of the two
major alliance blocs - are, in the Third World, if not totally absent, so thor-
oughly diluted as to be hardly recognizable. The primary aim of this paper
is to analyse how and why they are radically different in the context of the
Third World and what are the implications for the international system as
a whole that follow from these differences.
To take the first and, in one sense, the fundamental attribute of the Western
concept of security (in that it is a corollary of the doctrine of state sovereignty
in its pure and pristine form), namely, external directedness, one can immedi-
ately see that, despite the rhetoric of many Third World leaders, the sense
of insecurity from which these states - and, more particularly, their regimes -
suffer, emanates to a substantial extent from within their boundaries rather
than from outside. This does not mean that external threats are totally absent,
for they are not. But the 'mix' of internal and external sources of threat to
these state structures, and particularly to their regimes, is quite often heavily
weighted in favour of internal sources. Moreover, external threats quite often
augment the problems of insecurity that exist within state boundaries and, in
many cases, would be quite ineffective if internal threats and domestic fissures
did not exist within Third World societies. Any perceptive observer of the
South Asian scene in 1970-1 would have realized that the Indian 'threat' to
Pakistan was very secondary to that posed by East Bengali nationalists; also
that the Indo-Pakistani war of 1971 would either not have been fought, or, if
fought, would have had a very different outcome if the bulk of the East
Bengali population had not been disenchanted with the then existing structure
of the Pakistani state.~imilarly,Iran's ideological and political 'threat' to Iraq
after the Islamic revolution of 1979, which led Saddam Hussein to launch his
invasion of Iran, would not have reached the proportions it did in the Iraqi
regime's perception had the government in Baghdad been more representative
of the majority of its population and had it not been as narrowly based as it
is today. Its invasion of Iran was largely designed to pre-empt an anticipated
popular movement against it from within - a movement that would have
owed much to the exemplary success of the Iranian revol~tion.'~
On occasion, a regime - and I make no distinction here between so-called
'right-wing' and 'left-wing' regimes - will 'externalize' threats directed at it,
in order both to portray such threats as 'illegitimate' (in the sense that they
emanate from abroad and violate the norm of state sovereignty and its corol-
lary of non-intervention by other states) and to portray its repressive actions -
often repressive in the extreme, as any student of Latin America or Africa,
or even of Syria and Iraq, would immediately recognize - as 'legitimate'.
By turning a political (and quite often a social and economic) problem into
a military one, and by presenting the military threat as coming from exter-
nal sources, regimes in the Third World quite often try to choose an arena
of confrontation with domestic dissidents that is favourable to themselves,
namely, the military arena. While this strategy might work well in the initial
stages of such confrontations, it usually leads to much bigger conflagrations
within a decade or two of the initial, usually unorganized, outbursts of polit-
ical dissent.
This characterization of the security problem as faced by the Third
World states and its differences from the pattern of security issues faced by
developed Western states has dealt so far only with what one may call the
'symptomatic' level of the question, the above-mentioned differences being
merely the symptoms of a much deeper divergence in the respective experi-
ences of Western and Third World states. These differences are related to
two major variables: (a) the history of state formation in the Third World
as compared to its counterpart in the West, and (b) the pattern of elite
recruitment and regime establishment and maintenance in the Third World
as compared to the same processes in the developed states. These major
variables have their own corollaries which will be analysed as the discus-
sion proceeds, but it is essentially the differences in these two broad inter-
related areas between Western and Third World states - differences not so
much in absolute and culture-based terms as in relative and time-based ones -
that determine the differences in the primary security orientations of the
two sets of states.
As a result of a centuries-old process of development, modern states in
the industrialized Western world (which we will call the European world, for
short, and which includes North America and Australasia) have reached a
position which can be referred to as one of 'unconditional legitimacy'."
Moreover, not only are the prevailing state-structures in the European world
legitimate, they are also strong and cohesive. In fact, the two attributes com-
plement each other. Western states are, therefore, strong states (although all
of them may not be strong powers they are strong in terms of their state
structures). By contrast, state structures in the Third World in the present
Security in t h e Third World: T h e Worm About t o Turn? 32 1
form d o not enjoy 'unconditional legitimacy' and are weak as states (once
again one must be careful to distinguish weak states from weak powers)."
There are more reasons for this major difference in respective strengths. The
first is related to the time factor. Most states in the Third World are only
recent participants in the modern system of states, which is European in ori-
gin and in its defining characteristics. Until a few decades ago they were
mere 'objects' rather than 'subjects' in international relations. Even after the
conclusion of the decolonization process, because of the enormous time gap
between the development of modern state structures in the Third World and
the development of the same structures in Europe, their capacity to act effect-
ively in a system which is defined primarily by its state-centric character is
low. The economic gulf between the developed Western and developing
Southern states, with limited and partial exceptions, adds to the latters' inef-
fectiveness as participants in the system. One can, therefore, even in the
1980s, speak with some justification of two types of actors in the inter-
national system: the primary actors (the original European members of the sys-
tem and their offshoots in North America and Australasia) and the secondary
actors (the late-comers, the bulk of the Third World).
This late development of modern state structures has meant they still
lack legitimacy (in their present form) in large parts of the Third World.
Defined, as they have been, primarily by boundaries drawn by the colonial
powers for the sake of administrative convenience or in some form of trade-
off with colonial competitors, these structures have not yet developed the
capacity to ensure the habitual identification of their populations with their
respective states and the regimes that preside over these post-colonial struc-
tures within colonially-dictated boundaries. It is no wonder, therefore, that
internal dissent and lack of identification pose such a strong threat to the
structures of Third World states.
The problem is drastically compounded when one looks at the level of
consensus on fundamental social and political issues within Third World
societies. Western societies reached the current high level of consensus on
such issues of social and political organization after centuries of internal
conflict and upheaval (the Cromwellian episode in Britain and the French
Revolution are two outstanding examples). Having gone through these tur-
moils, and having had the luxury of doing so over centuries - a luxury no
longer available to Third World states - the present high level o f consensus
has been attained. For, despite the (sometimes extreme) rhetoric of the vari-
ous political groupings in the West, their positions on vital political and
social issues vary only marginally from each other. This is why, no matter
how heated the debate and how 'real' the differences among the political
parties in Britain, the Tory victory in the June 1983 elections has not led to
any of the opposition leaders going underground to launch a campaign of
subversion and insurgency against the 'Thatcher regime'. However, with
certain limited exceptions (maverick Third World states like India for ex-
ample, and one is not sure for how long they will continue to be exceptions)
this is a norm that does not prevail in the Third World. Since fundamental
issues of political, social and economic organization are involved in political

contests they become, literally, life and death issues for the contestants.
With so much at stake the ferocity of their attitudes towards one another
also becomes understandable.'Qut from the point of view of the security
of Third World societies this has extremely negative consequences. The divi-
sions, both vertical and horizontal, within these societies are exacerbated
and the level and intensity of internal threats to state structures (which are
identified with the ruling groups) escalate. This is one of the major reasons
why in most Third World societies open political contests are banned.
This brings us to another major difference between Western and Third
World states. In the absence of a consensus on fundamental issues and of
open political debate and contest, most of these states are ruled by regimes
with narrow support bases - both politically and socially - which usually
came to power by means of a coup d'etat and which hang on so tenaciously
to office that they have to be, more often than not, physically liquidated to
pave the way for any form of political transition.14 (Once again, there is no
major difference between 'rightist' and 'leftist' regimes on this score.) Since
it is these regimes, and their bureaucratic and intellectual hangers-on, who
define the threats to the security of their respective states, it is no wonder
that they define it primarily in terms of regime security rather than the secur-
ity of the society as a whole. Security, as stated above, has traditionally been
defined as the protection and preservation of core values. However, in the
case of most Third World states, the core values of the regime - with self-
preservation at the core of the core - are often at extreme variance with the
core values cherished by large segments of the population over which they
rule.'Wnce again, given these discrepancies in the definition of core values
and, indeed, of security itself, it is no wonder that major threats to the secur-
ity of these regimes emanate from within their own societies.
Uneven economic development, great, growing and glaring disparities in
wealth and income, communal and ethnic tensions - all these factors con-
tribute to the lack of societal consensus on fundamental issues, and to the
unrepresentative and repressive character of most Third World regimes and,
therefore, to the internal threats to their security and to the security of the
state structures over which they preside. One cannot deny that some of
these schisms exist in developed societies as well, for example, in racial ten-
sions in the United States, denominational antagonisms in Northern
Ireland, uneven economic development as between northern and southern
Italy and ethno-linguistic differences in Canada. However, the scale and
intensity of these fissures when measured against the overall identification
of a state's population with the concept of that particular state and its
unconditional acceptance of the legitimacy of the particular government in
power are minimal, and certainly manageable. This is a state of affairs that
does not prevail in most Third World states and therefore the proportions
of the problem, in terms of scale and intensity, are different.
In addition to the sources of friction and conflict that exists within Third
World states, the economic and social dimensions of nation-building have
\ o Security in the Third World: The Worm About to Turn? 323
been made more difficult by the workings of an international economy

(reflected in the operation of national economies in the Third World) which
increases economic disparities in both absolute and relative terms between
the developed and underdeveloped states and between the rich and the poor
within Third World states.lh This makes elite - mass identification within
Third World countries a task virtually impossible of achievement, and cor-
respondingly increases the alienation of large segments of society both from
their ruling elites and, quite often, from the state structures over which
these elites preside. This, in turn, creates a double source of insecurity for
those who claim to represent states in the Third World, for both their legit-
imacy and the legitimacy of the state structures are open to challenge. The
dimensions of the security problem, and of the concept of security itself, in
the Third World are, therefore, very different from those applied to, and
common in the literature of, the developed West.
T h e International Context: War by Proxy?
This, however, is not the end of the story. There is also a remarkable dif-
ference between the respective relationships between the security concerns
of the Third World states on the one hand, and those of the developed coun-
tries on the other, to the security and stability of the international system as
a whole. This has grave adverse effects on Third World states as far as the
systemic inputs into their security problems are concerned. I have argued
earlier that the developed states' security concerns are so firmly interlinked
with those of the international system as to make them virtually indistin-
guishable. Any major threat to the security of a developed state immediately
takes on the character of a crisis for the whole system, particularly since it
has the potential to destabilize the dominant global balance of power
between the two superpowers - a balance which in turn forms the under-
pinning of the stability and security of the international system as it is cur-
rently organized.
This is far from the case when one considers the security problems, exter-
nal or internal, of Third World states. (The possible exception to this con-
cerns some of the major oil exporters, though even in their case the linkages
between their security concerns and those of the system or of the dominant
powers appear to be of an ephemeral and very limited character. There is no
commitment in the developed world to the security of these states as states:
the only commitment seems to be to the security of access to their oil
resources.) Even in the case of the most important and the largest countries
of the Third World, the link between their security and that of the system as
a whole ( a link that is defined, whether one likes it or not, by the dominant
powers in the system) is very fragile, if not totally non-existent.
This means that conflicts, whether internal or intra-regional, or even,
sometimes, cross-regional, within and among states in the Third World are
considered permissible by the dominant powers, as long as they do not
threaten to draw the latter into direct confrontation with each other. The
only conflict in the Third World with this sort of potential is the Arab-Israeli
conflict, particularly if the balance of power in the Middle East were to
change dramatically to Israel's disadvantage. But this is not a typical Third
World conflict, for the simple reason that Israel, in terms of its ideological
origins, its pattern of colonization in Palestine, the organization of its soci-
ety and polity, the composition of its elite (even under Likud), its links with
strong and important European and American constituencies (both Jewish
and Gentile) and the intensity of one superpower's commitment to its exter-
nal security (some would argue its expansion), is a European state. Israel
may be physically located in the Third World but it is not of the Third
World. Therefore Israeli security is linked directly to issues of systemic secur-
ity, which is not the case with any other African or Asian country.'-
Thus, while setting limits to Third World conflicts (the limits being
determined by the superpowers' refusal to enter into direct confrontation
over what are considered peripheral issues), the stability of the central bal-
ance not only permits local (intra-state or intra-regional) conflicts in large
parts of the Third World, but may in fact even encourage the eruption of
such conflicts, partially as a way of letting off steam t o help cool the tem-
perature around those core issues which are considered directly relevant
and vital to the central balance and, therefore, to the international system.lx
The fragility of political institutions and state structures in the Third
World permits such encouragement, because it allows internal issues t o take
on international dimensions. Fragile polities are by definition easily perme-
able. Therefore, internal issues in Third World societies not only get trans-
formed into inter-state issues quite readily, they also lend themselves easily
to political and military intervention by the superpowers. Traditional inter-
state rivalries, compounded by the complex disputes introduced as a result
of colonially imposed boundaries, when added to the internal political
fragilities of Third World states provide very fertile ground for superpower
rivalry to be played out in the relatively 'safe' areas of the globe without the
imminent threat of direct confrontation between Washington and Moscow.
In the process, however, they exacerbate the problem of security - no mat-
ter whether it is defined in maximalist or in minimalist terms - for vast
parts of the Third World.19
There are a number of secondary factors which lead the superpowers to
tolerate and, quite often, encourage conflicts in the Third World; for ex-
ample, these conflicts (a) keep the arms industry of the developed world in
business, pay for a substantial proportion of R&D investments, and help
recycle petrodollars into developed economies; ( b ) provide convenient test-
ing grounds for new weapons systems which can than be improved upon in
the light of combat experience; (c) provide a relatively safe instrument for
testing the limits of the adversary's tolerance and a rough and ready meas-
ure of its 'will' to resist political and military encroachments; ( d ) provide
opportunities for 'linkage' between issues, thereby allowing a superpower
which finds itself in a disadvantageous position in one context to choose
i Security in t h e Third World: T h e Worm About t o Turn? 325
another point, where it is more favourably placed, to put pressure on its

adversary; (e) provide the superpowers with the opportunity to demonstrate
their 'credibility' to those allies (in Europe and the Pacific) which are con-
sidered vital to the superpowers' own security; and ( f ) provide one way of
ensuring access to strategic raw materials, such as oil and minerals, which
are considered essential for the security of the superpowers and their vital
allies.
In the process of achieving these aims, the superpowers - and therefore
the systemic variables as a whole - impinge upon the security of Third
World states in a fashion extremely deleterious to those states. This is exactly
the reverse of what the systemic factors achieve in the case of developed
polities, when they tend to augment the security of individual states by
making it virtually indivisible from the security of the whole system. In the
case of Third World states, the systemic inputs add to the already destabil-
izing internal dynamics of Third World polities to render these states and
their regimes extremely insecure. The two processes, in fact, feed upon each
other to create a vicious cycle of insecurity which has become nearly impos-
sible to break.
The effects of weak state structures, weak domestic political institutions,
lack of societal consensus, distorted economic development and lack of
regime legitimacy on the one hand, and the adverse way in which systemic
variables impinge on the security problems of the Third World states on the
other, create an environment of insecurity and instability in which inter-
state rivalries, encouraged as they are by the policies and actions of exter-
nal forces, are relatively easily transformed into overt military hostilities.
Inter-state tensions and conflicts in various regions of the Third World, the
obverse side of the regional security coin, are different in character from
inter-state conflicts in the Western world. The former are much more com-
plex and directly related to domestic divisions within societies than is the
case with the latter. Moreover, given the weak linkage between the security
of Third World states and central issues of global security in the nuclear age,
war as an instrument of policy, whether for domestic or external reasons, is
still a realistic option for many regimes in the Third World. Again, given the
stalemate in the central balance, wars by proxy (within defined limits) in the
Third World appear to be a realistic option for superpower decision-makers.
A corollary of the abovementioned factors is that both superpowers
have developed a vested interest in maintaining insecure regimes, which
lack legitimacy and are founded o n narrow social bases, in power as their
clients in many parts of the Third World. However, their commitment to
the maintenance of these regimes and the systems over which they. preside ~
is far more limited than their commitment to maintain the political systems and
governments in power in what can be called their 'core' allies. Washington's
commitment to Somoza or the Shah was qualitatively different from its com-
mitment to Western Europe from the days of the Marshall Plan. Similarly, des-
pite the expansion of Soviet capabilities in the 1970s, Moscow's commitment
to Mengistu in Ethiopia or the MPLA in Angola is qualitatively different
from its commitment to the ruling parties in Czechoslovakia and Poland.

Afghanistan is the exception that proves the rule and its contiguity to the
USSR might have made the critical difference. But even in Afghanistan, the
commitment of Soviet resources is paltry as compared to what Moscow
would have been willing to commit had such a situation arisen for example
in Bulgaria, not to speak of East Germany. This paradox of massive political
and military investment (the latter largely in the form of arms transfers and
military advisers) in insecure regimes on the part of the superpowers, which
nevertheless stops short of a final commitment to save them when the chips
are down, adds substantially to the problems of insecurity that many Third
World regimes face; for it makes these regimes more reckless, more repressive
and less flexible on the mistaken notion that they would be bailed out by their
superpower patrons when in dire straits.20 Explanations of this phenomenon,
pointing to the lack of superpower capability in specific situations even if the
will existed, as well as to the autonomous and massive nature of the revolts
against client regimes (as in Iran), while valid, d o not detract from the super-
powers' responsibility for creating an atmosphere and expectations that lead,
in the final analysis, to increasing insecurity at the individual, regime and
regional levels.
Implications of a Shifting Balance
Such policies on the part of the superpowers have the potential to create
an even more dangerous situation. So far, the limited success that both
Washington and Moscow have achieved in winning friends and influencing
regimes in the Third World through their parallel policies of investing in
insecure regimes has resulted from the insulation, as argued above, of the
central balance against the negative effects of Third World conflicts and
proliferating insecurities. However, this insulation has been a function of
the relative stability of the central strategic balance, as a result of which the
relationship between the superpowers on the one hand and the Third World
on the other has so far been, by and large, a one-way street, in which the
superpowers, jointly or separately, have dictated the limits to which regional
insecurities could or could not affect their bilateral relationship.
However, the stability of the central balance itself is largely dependent
upon technological factors. For the last three decades the balance has not
tilted dangerously against either one of the contestants because the techno-
logical equilibrium between the superpowers has been, by and large, main-
tained. The overall American advantage in technological achievement has
been neutralized in the military - nuclear field by greater Soviet investment
in military-related technologies a t the expense of other sectors of society.
One cannot, however, pretend that this situation will last for ever. Given the
inherently escalatory nature of modern technology - a phenomenon we
have witnessed since the Industrial Revolution - one cannot rule out serious
disequilibrium in the military-technological balance in the near future.
-\voch Security in the Third World: The Worm About to Turn? 327
Destabilizing technological escalation may take one of two forms, and the
two scenarios reflect the apprehensions of the hard-liners in Washington
and Moscow respectively. Either the Soviet Union may, by a combination of
increasing investment in miltary R&D (of course, at greater domestic cost)
and greater political use of its conventional superiority in Europe, attain a
position whereby it is able to make serious political, if not military, inroads
into Western Europe; or the United States, by the sheer momentum of its
overall technological lead, may reach a stage where it is not only able to
counterbalance Soviet conventional superiority in Central Europe but also,
and more important, effectively to neutralize the Soviet Union's strategic
capability to strike at the North American continent. In both cases the sta-
bility of the central balance will be seriously compromised and all the cur-
rent wisdom about nuclear deterrence would become irrelevant.
Even if the final stages of either scenario are not reached, growing evi-
dence of a trend towards either conclusion would introduce an element of
grave uncertainty into superpower calculations. In fact, in this case, as in
others in international relations, the perceptions of a situation by the parties
concerned are more important than the actual situation itself. It would not
be wrong in this context, therefore, to note that there are increasing signs,
i f one looks at the recent rhetoric coming out of the White House and the
Kremlin, that perceptions both in Washington and Moscow tend towards
the more alarmist scenarios propounded in the United States and the Soviet
Union. These alarmist perceptions, in my view, are directly linked to the
uncertainties in the minds of superpower policy-makers and the relevant
foreign policy elites regarding the long-term effect of technological escala-
tion in the strategic arms race. The entire arms control exercise has been pri-
marily related to minimizing if not eliminating these uncertainties. However,
given the current state of mind prevalent in both Washington and Moscow
it is apparent that these attempts at arms control and limitation have not
had significant, long-term psychological effect. The failure of the US Senate
to ratify SALT 11 and the general deterioration in Soviet-American relations
since then, particularly the periodic breakdowns in bilateral communication,
have more than neutralized any benevolent effect that SAI>TI may have had
on the psychological atmosphere surrounding superpower relations.
Given this context, one can readily imagine that in a period of equilibrium
transition (which would itself be a function of perceived instability in the cen-
tral balance), that is, before the superpower strategic balance stabilizes at a
new and higher level of technological equilibrium, the two global powers,
increasingly apprehensive of the final outcome, would suffer from something
of a 'state of nerves'. This is a period which we may be entering in the second
half of the 1980s, if we have not done so already. In a state of great nervous
tension, superpower policy-makers (like all human beings) are bound to exag-
gerate out of all proportion the significance of occurrences - like conflicts and
insecurities in the Third World - which in more normal circumstances they
would tend to treat as routine in character. It is through the medium of these
quite often distorted perceptions that Third World conflicts and insecurities
could find a convenient entry point from which they might be able to affect
the course of the dominant bilateral r e l a t i ~ n s h i p . ~
This
' would not only give
them a degree of autonomy vis-a-vis the superpower relationship, it could
introduce complications into that central equation which, at a time of uncer-
tainty in the strategic balance, could prove extremely dangerous. Third World
insecurities, therefore, may in the next decade or less become a significant
form of input into the dominant global relationship, in some ways paying the
superpowers back in their own coin. The two types of instabilities (in all their
political and military, but primarily psychological, manifestations) could feed
upon each other, thus compounding the security problems of the global pow-
ers and the developed countries as well as the Third World. If, or rather when,
this happens, one could start speaking about the convergence of Western
(including Soviet) and Third World perceptions of 'security'. However, by
that time we would all be living in a much more insecure world.
Notes
1. Walter Lippmann, US foreign policy: shield of the Republic (Boston: Little, Brown, 1943),
p. 51.
2. Arnold Wolfers, Discord & collaboration: essays on international politics (Baltimore:
Johns Hopkins University Press, 1962), p. 150.
3. Talukder Maniruzzaman, The security of small states in the Third World, Canberra
Papers on Strategy and Defence No. 25 (Canberra: Strategic and Defence Studies Centre,
Australian National University, 1982), p. 15.
4. Martin Wight, 'Western values in international relations' in Herbert Butterfield and
Martin Wight, eds, Diplomatic investigations (London: Allen & Unwin, 1966), p. 103.
5. Hedley Bull, The anarchic01 society (London: Macmillan, 1977).
6. For example, see Robert 0 . Keohane and Joseph S. Nye, Power and rnterdependence
(Boston: Little, Brown, 1977).
7. The coincidence of the two approaches is probably best reflected in Leonard Beaton,
The reform of power: a proposal for an international security system (Inndon: Chatto cYc
Windus, 1972).
8. The term 'Third World' is used in this article in a generic sense, and deliberately so. It
is undoubtedly true that there are diverse elements within the Third World; it is also true that
there are intramural problems, conflicts and antagonisms within it. However, these countries
share enough in terms of their colonial past and their unequal encounter w ~ t hthe European
powers following the Industrial Revolution to set them apart from the European states which
have traditionally formed the 'core' of the modern system of states. They also share attributes
of economic undetdevelopment and social dislocation, which are at least partially artriburable
to their encounter with the West (and which have continued even after the formal process of
decolonization has been completed). They are still inadequately linked to the issue areas which
dominate the international system as it is constituted today, despite all the rhetoric that sur-
rounds such debates as those regarding the New International Economic Order. Some of these
arguments are foreshadowed in my chapter, 'Autonomy and intervention: superpowers and the
Third World' in Robert O'Neill and D.M. Horner, eds, New directions in strategic thinking
(London: Allen & Unwin, 1981), pp. 104-16.
9. For a systematic account of growing East Bengali disenchantment w ~ t hthe concept of
Pakistan, see Rounaq Jahan, Pakistan: failure in national integration (New York: Columbia
Universitv Press. 19721.
10. 1 have argued this at some length and related it to the history of state formation and
the pattern of elite recruitment in both Iraq and Iran in my chapter, 'Regime security or
Security in t h e Third World: T h e Worm About t o Turn? 329
regional srcur~t!,Z pet-spectlves from tlie Gulf' III L h n a l d H. M.icmillen, Asimz pers/wrtir,crs rjn
intcrn~7ti0ti~~l s(,~xrity(1.ondon: h h c n ~ i l h forthcoming).
,
I I. For a most valuahle account of the orlglns ~ r l dgrowth, both In geogr'1phica1 ~ n d
chronolog~calterms, of the modern s!stem of sr,ltes centred o n the tluropc,in stares, see Martin
Wight. Systcws of st'rtrs (1.eicesrer: Leicesrer Universit) Press, 19771, chs 4 ,ind 5 . Also, tor dit-
k r e n t perspectives, ree ~ o r e p l iK. Str'iyer, O n thc nietfielwl o r ~ g i n sof t h r r71oticrn stirtc.
(Princeton, Nj: Pruiccton University Press, 1970); 1.eonarri Tivey, ed., Tlw wri!iorr-stritc: tho
fi1rni~7tionof modern politrcs (Oxford: ~ M a r r ~Robertson, n 198 I ); a n d Ralph I'ettn~~ln, St~rtc'
rind c.1~755(L.ondon: C,room Helm, 1 9 7 9 ) .
11. For ,I d ~ s c u s \ ~ oofn the concepts of w e ~ ka n d strong st'ltes '1s opposed t o weak ' ~ n d
Strong powers, sce B'lrry Bumn, I'eoplr, StL7tesanti fcdr (Krtghton, Susseu: tl,irvcstcr; 198.3),
pp, 65-9.
1.3. K.ingl,tdesh, 8tnfr.l. (:yprus, Irr~n,Nic'iragua, El S:~lv.idot-'ind G ~ ~ , ~ t e t n,1,1~1 ltestify a to
this p'ltt"".
14. Again, the examples are numerous, ranging from Iraq under tlie t l a s h e m i t c t o
Nic,iragu,i under Son~o/,i.
1.5. S o ~ n c t ~ m eben
e s the ~nainten~unce of e x ~ s t i n gsr'lre h o u n d ~ r i e sIS not c o n s ~ d e r e d,i core
v ~ l u chy Important segments of the state's population.
I h . :\mong other sources on this subject. see the t w o reports published by the Krandt
C o m m ~ s s ~ o North-.Soutl~:
n: LI p r ~ g r ~ r nf i r~ s~n r ~ ~ 1 ,(~1980)
1 1 and C:o~nniotzcrisis. North-Sorith:
c ~ ~ o p ~ ~f01. 7 tuwrld
i ~ ~ nr c ~ - o ~ v (r 1y9 8 1 , (both London, Sydney: 1',1n).
17. O n e could '11-gue chat Afghan~sranon the one hand , ~ n dSouth Afrtc.~o n the other
prolmbly have o r could h,iw similar degrees of Irnporr,ince t o the USSR 'ind the U n ~ t e dStcites
respectively. However, the Soviet refusal t o commit more than ~ h o u t100,000 troops 111
Afgh,int\r.in despite ~ t cso n t ~ g u i t yt o the Soviet U n ~ o na n d the contlnulng in\urgenc) I \ ,In ~ n d i -
cLltlon t h ~ tth ~ smay n o t he the c.lse In t h ~ instance. \ Morcovet; when the Sowet l l n i o n moved
IT\ troops into i-\fgli,~n~sr;~n, it w a s sure that tt W J S not r ~ s k i n ga direct milltar) confrontation
w ~ t hthe United States. In fact. the very d e c ~ s ~ otno d o so s i g n ~ f ~ ethat d the conflict 111
~\fgh,~nist,lnwas n o t directly related t o the central concerns of the superpower relcltionshlp.
South Africa does not threaten as yet t o hecome an 'trea of direct superpower c o n t r o n t a t ~ o n .
l lowever, when the nationaltsr struggle in South A f r ~ c ne\calatcs to \vh,it .il-e con\idcrcd d'in-
gerous proportions, the IJn~redSt'ltes m a ) d e c ~ d et o Incorporate ~ t securlry s into the US vlslon
of s)\ttviitC s e c ~ ~ r i t It ) . would d o this not s o much because of the c o u n t r ) ' ~rniner.11 resources
o r str,~regicloc,ition 3s hecduse the currently doriiinant elite s h , ~ r ern.iny ch,ir:ictcr~st~csw ~ t h
~ t 1st-nel~
r counterp,lrt which make hoth of them h , ~ s i c ~ l l'European' y s t ~ t e sL ~ n provide
d thetn
w ~ t hpowerful const~tuenciesin the Western world. It is ~nrerestingt o note in this context t h ~ t
despite the Amcric,~nc o t n n i ~ t ~ n e nofr over 500,000 troops in South Vietnam, S.iigon could he
w e a i l y ,~b,lndoned.'l.he only s,irt\factory explanation is related t o the lack ot a societal con-
scmsus in the United States reg'lrding the security of South Vietn'lm ~ n its d ruling elite c ~ ~the id
r e , i l i ~ a t ~ othat,
n ' ~ f t e rnll, what happened in V~etnarnJ l d not directly ~ f f c c the t superpower h.~l-
, ~ n c eIn the long term. This consensus IS certainly and strongly present in the United Swtes I ~ I
I\rael'\ c'ire, and one suspects in South Africa's also.
18. This argunienr h r ~ sbeen developed In Sisir Gupra, 'Great Power relat~ons,wol-Id order
~ n the d Thlrd World', F o r ~ r pAffizirs Reports, Jul./Aug. 1977. Vol. 17. Nos. 7-8.
19. t o r details o f this argument , ~ n d c'ise s t u d ~ e s d e ~ n o n s t r ~ i t i n gits v;~l~ciiry,w e
h l o h n m n ~ e dA)ooh, ed., Conflict a n d iritcrtlcwtrori in t l ~ cT l ~ i r d\vorld ( L o n d o n : L r o o m ticlni.
1980).
70. T h e late S h ~ hof Iran's hirrerncss towards the United States after his overthrow b v ~ s
the result of the talse ~ s s ~ ~ n l pthat t ~ o itn would protect him (paticularly slncc it had put him 111
pomer in 1953) irrespect~vco f the massive nrlture of the opposition t o 111s rule.
L I . In ,I curIoLIs \vrly, Henry Ktss~nger.hy p r o p o u n d ~ n gh ~ 'link,~ge' s t h e s ~ sregarding Soviet
heh,ivto~lr,might habe paved the w.iy for \uch a n occurrence; tor it is hut .l short step from
. ~ p p l > i n gthe 'I~nkage' formul,i t o Soviet polley toward specif~c~ssueslconflictsIn the T l i ~ r t l
Vl'orld t o a p p l y ~ n git generally t o Third World conflicts, particu,~rly\vhet~~t I S , ~ s s ~ l r n eth,it d
the S o ~ ~ Union et might h a \ e ,I finger in almost every 'Third World ple.
Nuclear W a r and Climatic Catastrophe:
Some Policy Implications
Carl Sagan
It is not even impossible to imagine that the effects of an atomic war fought
with greatly perfected weapons and pushed by the utmost determination
will endanger the survival of man.
- Edward Teller
Bulletin of the Atomic Scientists, February 1947
The extreme danger to mankind inherent in the proposal by [Edward

Teller and others to develop thermonuclear weapons] wholly outweights
any military advantage.
- J. Robert Oppenheimer, et al.
Report of the General Advisory Committee, AEC
October 1949
The fact that no limits exist to the destructiveness of this weapon makes
its very existence and the knowledge of its construction a danger to
humanity. ... It is ... an evil thing.
- Enrico Fermi and 1.1. Rabi
Addendum, ibid.
A very large nuclear war would be a clamity of indescribable propor-

tions and absolutely unpredictable consequences, with the uncertainities
tending toward the worse. ... All-out nuclear war would mean the
destruction of contemporary civilization, throw man back centuries,
cause the deaths of hundreds of millions or billions of people, and, with
a certain degree of probability, would cause man to be destroyed as a
biological species ...
- Andrei Sakharov
Foreign Affairs, Summer 1983
Source: Foreign Affairs, 62(2) ( 1 983-84): 257-92.

-, W,I Nuclear War and Climatic Catastrophe 331
A pocalyptic predictions require, to be taken seriously, higher standards

of evidence than d o assertions on other matters where the stakes
are not as great. Since the immediate effects of even a single thermo-
nuclear weapon explosion are so devastating, it is natural to assume - even
without considering detailed mechanisms - that the more or less simultaneous
explosion of ten thousand such weapons all over the Northern Hemisphere
might have ~lnpredictableand catastrophic consequences.
And yet, while it is widely accepted that a full nuclear war might mean the
end of civilization at least in the Northern Hemisphere, claims that nuclear
war might imply a reversion of the human population to prehistoric levels, or
even the extinction of the human species, have, among some policymakers at
least, been dismissed as alarmist or, worse, irrelevant. Popular works that
stress this theme, such as Nevil Shute's O n the Beach, and Jonathan Schell's
The Fate of the Earth, have been labeled disreputable. The apocalyptic claims
arc rejected as unproved and unlikely, and it is judged unwise to frighten the
public with doomsday talk when nuclear weapons are needed, we are told, to
preserve the peace. But, as the above quotations illustrate, comparably dire
warnings have been made by respectable scientists with diverse political inclin-
ations, including many of the American and Soviet physicists who conceived,
devised and constructed the world nuclear arsenals.
Part of the resistance to serious consideration of such apocalyptic pro-
nouncements is their necessarily theoretical basis. linderstanding the long-
term consequences of nuclear war is not a problem amenable to experimental
verification - at least not more than once. Another part of the resistance
is psychological. Most people - recognizing nuclear war as a grave and ter-
rifying prospect, and nuclear policy as immersed in technical complexities,
official secrecy and bureaucratic inertia - tend t o practice what psych-
iatrists call denial: putting the agonizing problem out of our heads, since
there seems nothing we can d o about it. Even policymakers must feel this
temptation from time t o time. But for policymakers there is another con-
cern: if it turns out that nuclear war could end our civilization o r our
species, such a finding might be considered a retroactive rebuke to those
responsible, actively or passively, in the past or in the present, for the global
nuclear arms race.
The stakes are too high for us t o permit any such factors to influence our
assessment of the consequences of nuclear war. If nuclear war now seems
significantly more catastrophic than has generally been believed in the mili-
tary and policy communities, then serious consideration of the resulting
implications is urgently called for.
It is in that spirit that this article seeks, first, t o present a short summary,
in lay terms, o f the climatic and hiological consequences o f nuclear war that
emerge from extensive scientific studies conducted over the past two years,
the essential conclusions of which have now been endorsed hy a large num-
ber of scientists. These findings were presented in detail a t a special confer-
ence in C~imhridge,Mass., involving almost 100 scientists on April 22-26,
1983, and were publicly announced at a conference in Washington, D.C., on
October 3 1 and November 1, 1983. They have been reported in summary

form in the press, and a detailed statement of the findings and their bases
will be published in Science.' The present summary is designed particularly
for the lay reader.
Following this summary, I explore the possible strategic and policy
implications of the new findings. They point to one apparently inescapable
conclusion: the necessity of moving as rapidly as possible to reduce the
global nuclear arsenals below levels that could conceivably cause the kind
of climatic catastrophe and cascading biological devastation predicted by
the new studies. Such a reduction would have to be to a small percentage
of the present global strategic arsenals.
The central point of the new findings is that the long-term consequences of
a nuclear war could constitute a global climatic catastrophe.
The immediate consequences of a single thermonuclear weapon explosion
are well known and well documented - fireball radiation, prompt neutrons
and gamma rays, blast, and fires.' The Hiroshima bomb that killed between
200,000 and 200,000 people was a fission device of about 12 kilotons yield
(the explosive equivalent of 12,000 tons of TNT). A modern thermonuclear
warhead uses a device something like the Hiroshima bomb as the trigger -
the "match" to light the fusion reaction. A typical thermonuclear weapon
now has a yield of about 500 kilotons (or 0.5 megatons, a megaton being
the explosive equivalent of a million tons of TNT). There are many weapons
in the 9 t o 20 megaton range in the strategic arsenals of the United States
and the Soviet Union today. The highest-yield weapon ever exploded is
58 megatons.
Strategic nuclear weapons are those designed for delivery by ground-
based or submarine-launched missiles, or by bombers, to targets in the
adversary's homeland. Many weapons with yields roughly equal to that of
the Hiroshima bomb are today assigned to "tactical" or "theater" military
missions, or are designated "munitions" and relegated to ground-to-air and
air-to-air missiles, torpedoes, depth charges and artillery. While strategic
weapons often have higher yields than tactical weapons, this is not always
the case.3 Modern tactical or theater missiles (e.g., Pershing 11, SS-20) and
air support weapons (e.g., those carried by F-15 or MiG-23 aircraft) have
sufficient range to make the distinction between "strategic" and "tactical"
or "theater" weapons increasingly artificial. Both categories of weapons
can be delivered by land-based missiles, sea-based missiles, and aircraft;
and by intermediate-range as well as intercontinental delivery systems.
Nevertheless, by the usual accounting, there are around 18,000 strategic
thermonuclear weapons (warheads) and the equivalent number of fission
triggers in the American and Soviet strategic arsenals, with an aggregate
yield of about 10,000 megatons.
Nuclear War and Climatic Catastrophe 333
The total number of nuclear weapons (strategic plus theater and tactical)
in the arsenals of the two nations is close to 50,000, with an aggregate yield
near 15,000 megatons. For convenience, we here collapse the distinction
between strategic and theater weapons, and adopt, under the rubric "stra-
tegic," an aggregate yield of 13,000 megatons. The nuclear weapons of the
rest of the world - mainly Britain, France and China - amount to many hun-
dred warheads and a few hundred megatons of additional aggregate yield.
N o one knows, of course, how many warheads with what aggregate yield
would be detonated in a nuclear war. Because of attacks on strategic aircraft
and missiles, and because of technological failures, it is clear that less than the
entire world arsenal would be detonated. On the other hand, it is generally
accepted, even among no st military planners, that a "small" nuclear war
would be almost impossible to contain before it escalated to include much of
the world arsenak4 (Precipitating factors include comsnand and control
malfunctions, communications failures, the necessity for instantaneous deci-
sions o n the fates of millions, fear, panic and other aspects of real nuclear war
fought by real people.) For this reason alone, any serious attempt to examine
the possible consequences of nuclear war must place major emphasis on
large-scale exchanges in the five-to-seven-thousand-megaton range, and
many studies have done Many of the effects described below, however,
can be triggered by much smaller wars.
The adversary's strategic airfields, missile silos, naval bases, submarines
at sea, weapons manufacturing and storage locales, civilian and military
command and control centers, attack assessment and early warning facil-
ities, and the like are probable targets ("counterforce attack"). While it is
often stated that cities are not targeted "per se," many of the above targets
are very near or colocated with cities, especially in Europe. In addition,
there is an industrial targeting category ("countervalue attack"). Modern
nuclear doctrines require that "war-supporting" facilities be attacked. Many
of these facilities are necessarily industrial in nature and engage a work
force of considerable size. They are almost always situated near major trans-
portation centers, so that raw materials and finished products can be effi-
ciently transported to other industrial sectors, or to forces in the field. Thus,
such facilities are, almost by definition, cities, or near or within cities. Other
"war-supporting" targets may include the transportation systems themselves
(roads, canals, rivers, railways, civilian airfields, etc.), petroleum refineries,
storage sites and pipelines, hydroelectric plants, radio and television trans-
mitters and the like. A major countervalue attack therefore might involve
almost all large cities in the United States and the Soviet Union, and pos-
sibly most o f the large cities in the Northern Hemisphere.' There are fewer
than 2,500 cities in the world with populations over 100,000 inhabitants,
so the devastation of all such cities is well within the means of the world
nuclear arsenals.
Recent estimates of the immediate deaths from blast, prompt radiation,
and fires in a major exchange in which cities were targeted range from sev-
eral hundred million to 1.1 billion people - the latter estimate is in a World
Health Organization study in which targets were assumed not to be

restricted entirely to NATO and Warsaw Pact countries.' Serious injuries
requiring immediate medical attention (which would be largely unavailable)
would be suffered by a comparably large number of people, perhaps an
additional 1.1 b i l l i ~ n Thus
. ~ it is possible that something approaching half
the human population on the planet would be killed or seriously injured by
the direct effects of the nuclear war. Social disruption; the unavailability of
electricity, fuel, transportation, food deliveries, communications and other
civil services; the absence of medical care; the decline in sanitation meas-
ures; rampant disease and severe psychiatric disorders would doubtless
collectively claim a significant number of further victims. But a range of
additional effects - some unexpected, some inadequately treated in earlier
studies, some uncovered only recently - now make the picture much more
somber still.
Because of current limitations on missile accuracy, the destruction of
missile silos, command and control facilities, and other hardened sites
requires nuclear weapons of fairly high yield exploded as groundbursts or
as low airbursts. High-yield groundbursts will vaporize, melt and pulverize
the surface a t the target area and propel large quantities of condensates and
fine dust into the upper troposphere and stratosphere. The particles are
chiefly entrained in the rising fireball; some ride up the stem of the mush-
room cloud. Most military targets, however, are not very hard. The destruc-
tion of cities can be accomplished, as demonstrated at Hiroshima and
Nagasaki, by lower-yield explosions less than a kilometer above the surface.
Low-yield airbursts over cities or near forests will tend to produce massive
fires, some of them over areas of 100,000 square kilometers or more. City
fires generate enormous quantities of black oily smoke which rise at least
into the upper part of the lower atmosphere, or troposphere. If firestorms
occur, the smoke column rises vigorously, like the draft in a fireplace, and
may carry some of the soot into the lower part of the upper atmosphere, or
stratosphere. The smoke from forest and grassland fires would initially be
restricted to the lower troposphere.
The fission of the (generally plutonium) trigger in every thermonuclear
weapon and the reactions in the (generally uranium-238) casing added as a
fission yield "booster" produce a witch's brew of radioactive products,
which are also entrained in the cloud. Each such product, or radioisotope,
has a characteristic "half-life" (defined as the time to decay to half its ori-
ginal level of radioactivity). Most of the radioisotopes have very short half-
lives and decay in hours to days. Particles injected into the stratosphere,
mainly by high-yield explosions, fall out very slowly - characteristically in
about a year, by which time most of the fission products, even when con-
centrated, will have decayed t o much safer levels. Particles injected into the
troposphere by l ~ w - ~ i e explosions
ld and fires fall out more rapidly - by
gravitational settling, rainout, convection, and other processes - before the
radioactivity has decayed to moderately safe levels. Thus rapid fallout of
tropospheric radioactive debris tends to produce larger doses of ionizing
.8 )ti Nuclear War and Climatic Catastrophe 335
radiation than does the slower fallout of radioactive particles from the
stratosphere.
N ~ ~ c l e explosions
ar of more than one-megaton yield generate a radiant
fireball that rises through the troposphere into the stratosphere. The fire-
halls from weapons with yields between 100 kilotons and one megaton will
partially extend into the stratosphere. The high temperatures in the fireball
chemically ignite some of the nitrogen in the air, producing oxides of nitro-
gen, which in turn chemically attack and destroy the gas ozone in the mid-
dle stratosphere. But ozone absorbs the biologically dangerous ultraviolet
radiation from the Sun. Thus the partial depletion of the stratospheric ozone
layer, or "ozonosphere," by high-yield nuclear explosions will increase the
flux of solar ultraviolet radiation at the surface of the Earth (after the soot
and dust have settled out). After a nuclear war in which thousands of high-
yield weapons are detonated, the increase in biologically dangerous ultra-
violet light might be several hundred percent. In the more dangerous shorter
wavelengths, larger increases would occur. Nucleic acids and proteins, the
fundamental molecules for lite on Earth, are especially sensitive to ultra-
violet radiation. Thus, an increase of the solar ultraviolet flux at the surface
of the Earth is potentially dangerous for life.
These four effects - obscuring smoke in the troposphere, obscuring dust
in the stratosphere, the fallout of radioactive debris, and the partial destruc-
tion of the ozone layer - constitute the four known principal adverse envi-
ronmental consequences that occur after a nuclear war is "over." There
may be others about which we are still ignorant. The dust and, especially,
the dark soot absorb ordinary visible light from the Sun, heating the atmos-
phere and cooling the Earth's surface.
All four o f these effects have been treated in our recent scientific inves-
tigation."he study, known from the initials of its authors as TTAPS, for the
first time demonstrates that severe and prolonged low temperatures would
follow a nuclear war. (The study also explains the fact that no such climatic
effects were detected after the detonation of hundreds of megatons during
the period of U.S.-Soviet atmospheric testing of nuclear weapons, ended by
treaty in 1963: the explosions were sequential over many years, not virtu-
ally simultaneous; and, occurring over scrub desert, coral atolls, tundra and
wasteland, they set no fires.) The new results have been subjected to
detailed scrutiny, and half a dozen confirmatory calculations have now been
made. A special panel appointed by the National Academy of Sciences to
examine this problem has come to similar conclusions."'
Unlike many previous studies, the effects d o not seem to be restricted to
northern mid-latitudes, where the nuclear exchange would mainly take
place. There is now substantial evidence that the heating by sunlight of
atmospheric dust and soot over northern mid-latitude targets would pro-
foundly change the global circulation. Fine particles would be transported
across the equator in weeks, bringing the cold and the dark to the Southern
Hemisphere. (In addition, some studies suggest that over 100 megatons
would he dedicated to equatorial and Southern Hemisphere targets, thus
generating fine particles locally.)11While it would be less cold and less dark
at the ground in the Southern Hemisphere than in the Northern, massive
climatic and environmental disruptions may be triggered there as well.
In our studies, several dozen different scenarios were chosen, covering a
wide range of possible wars, and the range of uncertainty in each key param-
eter was considered (e.g., to describe how many fine particles are injected into
the atmosphere). Five representative cases are shown in Table 1, below,
ranging from a small low-yield attack exclusively on cities, utilizing, in yield,
only 0.8 percent of the world strategic arsenals, to a massive exchange
involving 75 percent of the world arsenals. "Nominal" cases assume the most
probable parameter choices; "severe" cases assume more adverse parameter
choices, but still in the plausible range.
Predicted continental temperatures in the Northern Hemisphere vary
after the nuclear war according t o the curves shown in Figure 1 on the fol-
lowing page. The high heat-retention capacity of water guarantees that
oceanic temperatures will fall at most by a few degrees. Because tempera-
tures are moderated by the adjacent oceans, temperature effects in coastal
regions will be less extreme than in continental interiors. The temperatures
shown in Figure 1 are average values for Northern Hemisphere land areas.
Table 1: Nuclear Exchange Scenarios
?6 Yield Warhead
% Yield Urban or Yield Total
Total Surface Industrial Range Number of
Case Yield ( M T ) Bursts Targets (MT) Explosions
1. Baseline Case, 5,000 57 20 0.1-10 10,400
countervalue and
co~nterforce~~~
11. 3,000 MT nominal, 3,000 50 0 1-10 2,250
counterforce only'"'
14. lOOMT nominal, 100 0 100 0.1 1,000
countervalue only(''
16. 5000 MT "severe," 5,000 100 0 5-1 0 700
counterforce only'"- *'
17. 10,000 MT "severe," 10,000 63 15 0.1-10 16,160
countervalue and
c~unterforce'~,
a. In the Baseline Case, 12,000 square kilometers of inner cities are burned; on every square
centimeter an average of 10 grams of combust~blesare burned, and I . 1 % of the burned
material rises as smoke. Also, 230,000 square kilometers of suburban areas burn, with
1.5 grams consumed at each square centimeter and 3.6% rising as smoke.
b. In this highly conservative case, it is assumed that no smoke emission occurs, that not a
blade of grass is burned. Only 25,000 tons of the fine dust is raised into the upper atmos-
phere for every megaton exploded.
c. In contrast to the Baseline Case, only inner cities burn, but with 10 grams per square centimeter
consumed and 3.3% rising as smoke into the high atmosphere.
d. Here, the fine (submicron)dust raised into the upper atmosphere IS 150,000 tons per megaton
exploded.
1 Nuclear War and Climatic Catastrophe 337
Figure 1 : T e m p e r a t u r e Effects of N u c l e a r W a r C a s e s
C a s e I 1: 3000 h1T n o m ~ n d c, o u n t e r f o r c e o n l )
Case 14: I00 MT n o m ~ n a l c, i n e s onl!
TFMPCRATUKt>
---------------
FREEZINC, P O I N T OF P U R E WATER
c ~ t ~ aensd c o u n t e r t o r i e
C a s e 17: 1 0 0 0 0 MT " s e v e r e " ,

~ ~ r a ~n de ~ so u n t e r f o r c e
T I M E POST-DETONATION (days)
Note: In this Figure, the average temperature of Northern Hemisphere land arc,l\ ( ~ w iron1 ~ y
coastlines) 1s shown varylng w ~ t htime after the hve Cases of nuclear war defined in Tahle 1.
T h e "ambient" temperature is the average in the Northern Hemi\phere o \ e r all latitude\ ,und
seasons: thus, normal winte1- temperatures a t north temperature latitudes are lower th,ln is
shown, a n d normal tropical remper,ltures ,Ire higher tli,ln shown. Cases described as "noniinal"
Jssume the m o \ t likely values of parameters (such as dust part& size o r the trecluency ot
firestorms) that a r e imperfectly known. Cases marked "severe" represent ,idverse hut not
~tnpl~iusihle values of these parameters. In Case 14 the curve ends when the ten1per;itures come
w ~ t h l na degree of the ambient values. For the four other Cases the curve5 are \ho\vn enduig
'~trer 300 d ~ y s hut
, t h ~ sis s ~ m p l yhecause the calculations were not extended further. In these
tour <:a\cs the curve\ will continue t o the directions thev are headed.
Even much smaller temperature declines are known to have serious con-
sequences. The explosion of the Tambora volcano in Indonesia in 1815 led
to an average global temperature decline of only 1°C, due to the obscuration
of sunlight by the fine dust propelled into the stratosphere; yet the hard
freezes the following year were so severe that 1816 has been known in
Europe and America as "the year without a summer." A 1°C cooling would
nearly eliminate wheat growing in Canada.12 In the last thousand years, the
maximum global or Northern Hemisphere temperature deviations have been
around 1°C. In an Ice Age, a typical long-term temperature decline from
preexisting conditions is about 10°C. Even the most modest of the cases
illustrated in Figure 1 give temporary temperature declines of this order.
The Baseline Case is much more adverse. Unlike the situation in an Ice Age,
however, the global temperatures after nuclear war plunge rapidly and take
only months to a few years to recover, rather than thousands of years. N o
new Ice Age is likely to be induced by a Nuclear Winter.
Because of the obscuration of the Sun, the daytime light levels can fall to
a twilit gloom or worse. For more than a week in the northern mid-latitude
target zone, it might be much too dark to see, even at midday. In Cases 1 and
1 4 (Table I ) , hemispherically averaged light levels fall to a few percent of
normal values, comparable to those at the bottom of a dense overcast. At
this illumination, many plants are close to what is called the compensation
point, the light level at which photosynthesis can barely keep pace with plant
metabolism. In Case 17, illumination, averaged over the entire Northern
Hemisphere, falls in daytime to about 0.1 percent of normal, a light level at
which plants will not photosynthesize at all. For Cases 1 and especially 17,
full recovery to ordinary daylight takes a year or more (Figure 1).
As the fine particles fall out of the atmosphere, carrying radioactivity to
the ground, the light levels increase and the surface warms. The depleted
ozone layer now permits ultraviolet light to reach the Earth's surface in
increased proportions. The relative timing of the multitude of adverse con-
sequences of a nuclear war is shown in Table 2, on the following page.
Perhaps the most striking and unexpected consequence of our study is that
even a comparatively small nuclear war can have devastating climatic conse-
quences, provided cities are targeted (see Case 14 in Figure 1; here, the centers
of 100 major NATO and Warsaw Pact cities are burning). There is an indication
of a very rough threshold at which severe climatic consequences are triggered -
around a few hundred nuclear explosions over cities, for smoke generation, or
around 2,000 to 3,000 high-yield surface bursts at, eg., missile silos, for dust
generation and ancillary fires. Fine particles can be injected into the atmosphere
at increasing rates with only minor effects until these thresholds are crossed.
Thereafter, the effects rapidly increase in severity."
As in all calculations of this complexity, there are uncertainties. Some fac-
tors tend to work towards more severe or more prolonged effects; others
tend to ameliorate the effects.14 The detailed TTAPS calculations described
here are one-dimensional; that is, they assume the fine particles to move
vertically by all the appropriate laws of physics, but neglect the spreading in
IV- 1 1 7
IN VY
w H 1
W VI I/t'
H H H
H w-1
w PY T/V
H-IN H H
1 1 bV
-1 7 7
w 7
w-7 H 1
H-IAI IN 7
H-w H 7
latitude and longitude. When soot or dust is moved away from the reference
locale, things get better there and worse elsewhere. In addition, fine particles
can be transported by weather systems to other locales, where they are car-
ried more rapidly down to the surface. That would ameliorate obscuration
not just locally but globally. It is just this transport away from the northern
mid-latitudes that involves the equatorial zone and the Southern Hemisphere
in the effects of the nuclear war. It would be helpful to perform an accurate
three-dimensional calculation on the general atmospheric circulation fol-
lowing a nuclear war. Preliminary estimates suggest that circulation might
moderate the low temperatures in the Northern Hemisphere predicted in our
calculations by some 30 percent, lessening somewhat the severity of the
effects, but still leaving them at catastrophic levels (e.g., a 30°C rather than
a 40°C temperature drop). To provide a small margin of safety, we neglect
this correction in our subsequent discussion.
There are also effects that tend to make the results much worse: for
example, in our calculations we assumed that rainout of fine particles
occurred through the entire troposphere. But under realistic circumstances,
at least the upper troposphere may be very dry, and any dust or soot car-
ried there initially may take much longer to fall out. There is also a very sig-
nificant effect deriving from the drastically altered structure of the
atmosphere, brought about by the heating of the clouds and the cooling of
the surface. This produces a region in which the temperature is approxi-
mately constant with altitude in the lower atmosphere and topped by a
massive temperature inversion. Particles throughout the atmosphere would
then be transported vertically very slowly - as in the present stratosphere.
This is a second reason why the lifetime of the clouds of soot and dust may
be much longer than we have calculated. If so, the worst of the cold and the
dark might be prolonged for considerable periods of time, conceivably for
more than a year. We also neglect this effect in subsequent discussion.
Nuclear war scenarios are possible that are much worse than the ones
we have presented. For example, if command and control capabilities are
lost early in the war - by, say, "decapitation" (an early surprise attack on
civilian and military headquarters and communications facilities) - then the
war conceivably could be extended for weeks as local commanders make
separate and uncoordinated decisions. At least some of the delayed missile
launches could be retaliatory strikes against any remaining adversary cities.
Generation of an additional smoke pall over a period of weeks or longer
following the initiation of the war would extend the magnitude, but espe-
cially the duration of the climatic consequences. O r it is possible that more
cities and forests would be ignited than we have assumed, or that smoke
emissions would be larger, or that a greater fraction of the world arsenals
would be committed. Less severe cases are of course possible as well.
These calculations therefore are not, and cannot be, assured prognostica-
tions of the full consequences of a nuclear war. Many refinements in them are
possible and are being pursued. But there is general agreement on the overall
conclusions: in the wake of a nuclear war there is likely to be a period, lasting
Nuclear War and Climatic Catastrophe 34 1
at least for months, o f extreme cold in a radioactive gloom, followed - after

the soot and dust fall out - by an extended period of increased ultraviolet light
reaching the surface."
We now explore the biological impact of such a n assault on the global
environment.
The immediate human consequences of nuclear explosions range from

vaporization of populations near the hypocenter, to blast-generated trauma
(from flying glass, falling beams, collapsing skyscrapers and the like), t o
burns, radiation sickness, shock and severe psychiatric disorders. Rut o u r
concern here is with longer-term effects.
It is now a commonplace that in the burning of modern tali buildings,
more people succumb to toxic gases than to fire. Ignition of many varieties
of building materials, insulation and fabrics generates large amounts of
such pyrotoxins, including carbon monoxide, cyanides, vinyl chlorides,
oxides of nitrogen, ozone, dioxins, and furans. Because of differing prac-
tices in the use of such synthetics, the burning of cities in North America
and Western Europe will probably generate more pyrotoxins than cities in
the Soviet Union, and cities with substantial recent construction more than
older, unreconstructed cities. In nuclear war scenarios in which a great
many cities are burning, a significant pyrotoxin smog might persist for
months. The magnitude of this danger is unknown.
The pyrotoxins, low light levels, radioactive fallout, subsequent ultra-
violet light, and especially the cold are together likely to destroy almost all of
Northern Hemisphere agriculture, even for the more modest Cases I 1 and 14.
A 12" to 15°C temperature reduction by itself would eliminate wheat and
corn production in the United States, even if all civil systems and agricultural
technology were intact.'' With unavoidable societal disruption, and with the
other environmental stresses just mentioned, even a 3,000-megaton "pure"
counterforce attack (Case 11) might suffice. Realistically, many fires would
be set even in such an attack (see below), and a 3,000-megaton war is likely
to wipe out U.S. grain production. This would represent by itself an unprece-
dented global catastrophe: North American grain is the principal reliable
source of export food on the planet, as well as an essential component of
U.S. prosperity. Wars just before harvesting of grain and other staples would
he incrementally worse than wars after harvesting. For many scenarios, the
effects will extend (see Figure 2) into t w o o r more growing seasons.
Widespread fires and subsequent runoff of topsoil are among the many addi-
tional deleterious consequences extending for years after the war.
Something like three-quarters o f the U.S. population lives in o r near cities.
In the cities themselves there is, on average, only about one week's supply of
food. After a nuclear war it is conceivable that enough of present grain stor-
age might survive to maintain, on some level, the present population for more
than a year. But with the breakdown of civil order and transportation systems
in the cold, the dark and the fallout, these stores would become largely inac-
cessible. Vast numbers of survivors would soon starve to death.
In addition, the sub-freezing temperatures imply, in many cases, the
unavailability of fresh water. The ground will tend to be frozen to a depth of
about a meter - incidentally making it unlikely that the hundreds of millions
of dead bodies would be buried, even if the civil organization to do so existed.
Fuel stores to melt snow and ice would be in short supply, and ice surfaces
and freshly fallen snow would tend to be contaminated by radioactivity and
pyrotoxins.
In the presence of excellent medical care, the average value of the acute
lethal dose of ionizing radiation for healthy adults is about 450 rads. (As
with many other effects, children, the infirm and the elderly tend to be more
vulnerable.) Combined with the other assaults on survivors in the postwar
environment, and in the probable absence of any significant medical care,
the mean lethal acute dose is likely to decline to 350 rads or even lower. For
many outdoor scenarios, doses within the fallout plumes that drift hun-
dreds of kilometers downwind of targets are greater than the mean lethal
dose. (For a 10,000-megaton war, this is true for more than 30 percent of
northern mid-latitude land areas.) Far from targets, intermediate-timescale
chronic doses from delayed radioactive fallout may be in excess of 100 rads
for the baseline case. These calculations assume no detonations on nuclear
reactors or fuel-reprocessing plants, which would increase the dose.
Thus, the combination of acute doses from prompt radioactive fallout,
chronic doses from the delayed intermediate-timescale fallout, and internal
doses from food and drink are together likely to kill many more by radi-
ation sickness. Because of acute damage to bone marrow, survivors would
have significantly increased vulnerability to infectious diseases. Most
infants exposed to 100 rads as fetuses in the first two trimesters of preg-
nancy would suffer mental retardation andlor other serious birth defects.
Radiation and some pyrotoxins would later produce neoplastic diseases and
genetic damage. Livestock and domesticated animals, with fewer resources,
vanishing food supplies and in many cases with greater sensitivity to
the stresses of nuclear war than human beings, would also perish in large
numbers.
These devastating consequences for humans and for agriculture would
not be restricted to the locales in which the war would principally be
"fought," but would extend throughout northern mid-latitudes and, with
reduced but still significant severity, probably to the tropics and the Southern
Hemisphere. The bulk of the world's grain exports originate in northern
mid-latitudes. Many nations in the developing as well as the developed
world depend on the import of food. Japan, for example, imports 75 percent
of its food (and 99 percent of its fuel). Thus, even if there were no climatic
and radiation stresses on tropical and Southern Hemisphere societies - many
of them already at subsistence levels of nutrition - large numbers of people
there would die of starvation.
Nuclear War and Climatic Catastrophe
~l~iildi) 343
As agriculture breaks down worldwide (possible initial exceptions might

include Argentina, Australia and South Africa if the climatic impact on the
Southern Hemisphere proved t o be minimal), there will be increasing reliance
on natural ecosystems - fruits, tubers, roots, nuts, etc. But wild foodstuffs
will also have suffered from the effects of the war. At just the moment that
surviving humans turn t o the natural environment for the basis of life, that
environment would be experiencing a devastation unprecedented in recent
geological history.
Two-thirds of all species of plants, animals, and microorganisms on the Earth
live within 25" of the equator. Because temperatures tend to vary with the sea-
sons only minimally at tropical latitudes, species there are especially vulnerable
to rapid temperature declines. In past major extinction events in the paleonto-
logical record, there has been a marked tendency for tropical organisms to show
greater vulnerability than organisms living at more temperate latitudes.
The darkness alone may cause a collapse in the aquatic food chain in
which sunlight is harvested by phytoplankton, phytoplankton by zooplankton,
zooplankton by small fish, small fish by large fish, and, occasionally, large
fish by humans. In many nuclear war scenarios, this food chain is likely to
collapse a t its base for at least a year and is significantly more imperiled in
tropical waters. The increase in ultraviolet light available a t the surface of
the earth approximately a year after the war provides an additional major
environn~entalstress that by itself has been described as having "profound
consequences" for aquatic, terrestrial and other ecosystems.'-
The global ecosystem can be considered an intricately woven fabric com-
posed of threads contributed by the millions of separate species that inhabit
the planet and interact with the air, the water and the soil. The system has
developed considerable resiliency, so that pulling a single thread is unlikely to
unravel the entire fabric. Thus, most ordinary assaults on the biosphere are
unlikely to have catastrophic consequences. For example, because of natural
small changes in stratospheric ozone abundance, organisms have probably
experienced, in the fairly recent geologic past, ten percent fluctuations in the
solar near-ultraviolet flux (but not fluctuations by factors of two or more).
Similarly, major continental temperature changes of the magnitude and extent
addressed here may not have been experienced for tens of thousands and pos-
sibly not for n~illionsof years. We have no experimental information, even for
aquaria or terraria, on the simultaneous effects of cold, dark, pyrotoxins, ion-
izing radiation, and ultraviolet light as predicted in the TTAI'S study.
Each of these factors, taken separately, may carry serious consequences
tor the global ecosystem: their interactions may be much more dire still.
Extremely worrisome is the possibility of poorly understood or as yet
entirely uncontemplated synergisms (where the net consequences of two or
more assaults on the environment are much more than the sum of the com-
ponent parts). For example, more than 100 rads (and possibly more than
200 rads) of external and ingested ionizing radiation is likely to be delivered
in a very large nuclear war to all plants, animals and unprotected humans in
densely populated regions of northern mid-latitudes. After the soot and dust
clear, there can, for such wars, be a 200 to 400 percent increment in the solar
ultraviolet flux that reaches the ground, with an increase of many orders of
magnitude in the more dangerous shorter-wavelength radiation. Together,
these radiation assaults are likely to suppress the immune systems of humans
and other species, making them more vulnerable to disease. At the same
time, the high ambient-radiation fluxes are likely to produce, through muta-
tion, new varieties of microorganisms, some of which might become patho-
genic. The preferential radiation sensitivity of birds and other insect
predators would enhance the proliferation of herbivorous and pathogen-
carrying insects. Carried by vectors with high radiation tolerance, it seems
possible that epidemics and global pandemics would propagate with no
hope of effective mitigation by medical care, even with reduced population
sizes and greatly restricted human mobility. Plants, weakened by low tem-
peratures and low light levels, and other animals would likewise be vulner-
able to preexisting and newly arisen pathogens.
There are many other conceivable synergisms, all of them still poorly
understood because of the complexity of the global ecosystem. Every syn-
ergism represents an additional assault, of unknown magnitude, on the
global ecosystem and its support functions for humans. What the world
would look like after a nuclear war depends in part upon the unknown syn-
ergistic interaction of these various adverse effects.
We d o not and cannot know that the worst would happen after a
nuclear war. Perhaps there is some as yet undiscovered compensating effect
or saving grace - although in the past, the overlooked effects in studies of
nuclear war have almost always tended toward the worst. But in an uncer-
tain matter of such gravity, it is wise to contemplate the worst, especially
when its ~ r o b a b i l i is
t ~not extremely small. The summary of the findings of
the group of 40 distinguished biologists who met in April 1983 to assess the
TTAPS conclusions is worthy of careful consideration:I8
Species extinction could be expected for most tropical plants and ani-
mals, and for most terrestrial vertebrates of north temperate regions, a
large number of plants, and numerous freshwater and some marine
organisms. ... Whether any people would be able to persist for long in
the face of highly modified biological communities; novel climates; high
levels of radiation; shattered agricultural, social, and economic systems;
extraordinary psychological stresses; and a host of other difficulties is
open to question. It is clear that the ecosystem effects alone resulting
from a large-scale thermonuclear war could be enough to destroy the
current civilization in at least the Northern Hemisphere. Coupled with
the direct casualties of perhaps two billion people, the combined inter-
mediate and long-term effects of nuclear war suggest that eventually
there might be no human survivors in the Northern Hemisphere.
Furthermore, the scenario described here is by no means the most severe that
could be imagined with present world nuclear arsenals and those contemplated
for the near future. In almost any realistic case involving nuclear exchanges
between the superpowers, global environmental changes sufficient to cause an
extinction event equal to or more severe than that at the close of the Cretaceous
when the dinosaurs and many other species died out are likely. In that event,
the possibility of the extinction of Homo sapicns cannot be excluded.
The foregoing probable consequences of various nuclear war scenarios have

implications for doctrine and policy. Some have argued that the difference
between the deaths of several hundred million people in a nuclear war (as
has been thought until recently to be a reasonable upper limit) and the
death of every person on Earth (as now seems possible) is only a matter of
one order of magnitude. For me, the difference is considerably greater.
Restricting our attention only to those who die as a consequence of the war
conceals its full impact.
If we are required to calibrate extinction in numerical terms, I would be
sure to include the number of people in future generations who would not
be born. A nuclear war imperils all of our descendants, for as long as there
will he humans. Even if the population remains static, with an average life-
time of the order of 100 years, over a typical time period for the biological
evolution of a successful species (roughly ten million years), we are talking
about some 500 trillion people yet to come. By this criterion, the stakes are
one million times greater for extinction than for the more modest nuclear
wars that kill "only" hundreds of millions of people.
There are many other possible measures of the potential loss - including
culture and science, the evolutionary history of the planet, and the signifi-
cance of the lives of all of our ancestors who contributed to the future of
their descendants. Extinction is the undoing of the human enterprise.
For me, the new results on climatic catastrophe raise the stakes of nuclear
war enormously. But I recognize that there are those, including some policy-
makers, who feel that the increased level o f fatalities has little impact on
policy, but who nevertheless acknowledge that the newly emerging conse-
quences of nuclear war may require changes in specific points of strategic
doctrine. I here set down what seem to me some of the more apparent such
implications, within the context of present nuclear stockpiles. The idea of a
crude threshold, very roughly around 500 to 2,000 warheads, for trigger-
ing the climatic catastrophe will be central to some of these considerations.
(Such a threshold applies only to something like the present distribution of
yields in the strategic arsenals. Drastic conversion to very low-yield arsenals -
see below - changes some of the picture dramatically.) I hope others will
constructively examine these preliminary thoughts and explore additional
implications of the TTAPS results.
1. First Strike. The MIRVing of missiles (the introduction of multiple war-
heads ), improvements in accuracy, and other developments have increased
the perceived temptation to launch a devastating first strike against land

targets - even though both sides retain a powerful retaliatory force in air-
borne bombers and submarines at sea. Much current concern and national
rhetoric is addressed to the first-strike capability of extant or proposed
weapons systems. The mere capability of a first strike creates incentives for
a preemptive attack. Launch-on-warning and simultaneous release of all
strategic weapons are two of several ominous and destabilizing innovations
contrived in response to the fear of a first strike.
The number of U.S. land-based strategic missiles is about 1,050; for the
Soviet Union, about 1,400. In addition, each side has at least several dozen
dedicated and alternative strategic bomber bases and airstrips, as well as
command and control facilities, submarine ports and other prime strategic
targets on land. Each target requires - for high probability of its destruction -
two or perhaps three attacking warheads. Thus, a convincing first strike
against land targets requires at least 2,200 and perhaps as many as 4,500
attacking warheads. Some - for example, to disable bombers that succeed in
becoming airborne just before the first strike - would detonate as airbursts.
While many missile silos, especially in the United States, are surrounded by
farmland and brush, other strategic targets, especially in Europe and Asia, are
sufficiently near forests or urban areas for major conflagrations to be set, even
in a "pure" counterforce attack. Accordingly, a major first strike would be
clearly in the vicinity of, and perhaps well over, the climatic threshold.
A counterforce first strike is unlikely to be completely effective. Perhaps
10 to 40 percent of the adversary's silos and most of its airborne bombers
and submarines at sea will survive, and its response may not be against
silos, but against cities. Ten percent of a 5,000-warhead strategic arsenal is
500 warheads: distributed over cities, this seems by itself enough to trigger
a major climatic catastrophe.
Such a first strike scenario, in which the danger to the aggressor nation
depends upon the unpredictable response of the attacked nation, seems risky
enough. (The hope for the aggressor nation is that its retained second-strike
force, including strategic submarines and unlaunched land-based missiles,
will intimidate the adversary into surrender rather than provoke it into retali-
ation.) But the decision to launch a first strike that is tantamount to national
suicide for the aggressor - even if the attacked nation does not lift a finger
to retaliate - is a different circumstance altogether. If a first strike gains no
more than a pyrrhic victory of ten days' duration before the prevailing winds
carry the nuclear winter to the aggressor nation, the "attractiveness" of the
first strike would seem to be diminished significantly,
A Doomsday Machine is useless if the potential adversary is ignorant of
its presence.I9 But since many distinguished scientists, both American and
Soviet, have participated vigorously in recent studies of the climatic conse-
quences of nuclear war, since there appears to be no significant disagreement
in the conclusions, and since policymakers will doubtless be apprised of
these new results, it would appear that a decision to launch a major first
strike is now much less rational, and therefore, perhaps, much less probable.
I , , ~ ~ < I I I Nuclear War and Climatic Catastrophe 347
The better political leaders understand the nuclear winter, the more secure
are such conclusions.
If true, this should have cascading consequences for specific weapons sys-
tems. Further, the perceived vulnerability to a first strike has been a major
source of stress and fear, and thereby a major spur to the nuclear arms race.
Knowledge that a first strike is now less probable might make at least some
small contribution to dissipating the poisonous atmosphere of mistrust that
currently characterizes Soviet-American relations.
2. Sub-threshold War. Devastating nuclear wars that are nevertheless
significantly below the threshold for severe climatic consequences certainly
seem possible - for example, the destruction of 1 0 or 20 cities, or 100 silos
of a particularly destabilizing missile system. Nevertheless, might some
nation be tempted to initiate or engage in a much larger, but still reliably
sub-threshold nuclear war? The hope might be that the attacked adversary
would be reluctant to retaliate for fear of crossing the threshold.
This is not very different from the hope that a counterforce first strike
would not he followed by a retaliatory strike, because of the aggressor's
retention of an invulnerable (for example, submarine-based) second-strike
force adequate to destroy populations and national economies. It suffers the
same deficiency - profound uncertainty about the likely response.
The strategic forces of the United States or the Soviet Union - even if they
were all at fixed sites - could not be destroyed in a reliably sub-threshold war:
there are too many essential targets. Thus, a sub-threshold first strike power-
fully provokes the attacked nation and leaves much of its retaliatory force
untouched. It is easy to imagine a nation, having contemplated becoming the
object of a sub-threshold first strike, planning to respond in kind, because it
judges that failure to do so would itself invite attack. Retaliation could occur
immediately against a few key cities - if national leaders were restrained and
command and control facilities intact - or massively, months later, after
much of the dust and smoke have fallen out, extending the duration but
ameliorating the severity of the net climatic effects.
This, however, may not be the case for such nations as Britain, France
or China. Because of the marked contiguity of strategic targets and urban
areas in Europe, the climatic threshold for attacks on European nuclear
powers may be significantly less than for the United States or the Soviet
Union. Provided it could be accomplished without triggering a U.S.-Soviet
nuclear war, first strikes against all the fixed-site strategic forces of one of
these nations might not trigger the climatic catastrophe. Nevertheless, the
invulnerable retaliatory capability of these nations - especially the ballistic-
missile submarines of Britain and France - makes such a first strike unlikely.
3. Treaties o n Yields and Targeting. I would not include this possibility,
except that it has been mentioned publicly by a leading American nuclear
strategist. The proposal has two parts. The first is to ban by treaty all
nuclear warheads with yields in excess of 300 or 400 kilotons. The fireballs
from warheads of higher yields mainly penetrate into the stratosphere and
work to deplete the ozonosphere.
The reconversion of nuclear warheads to lower individual yields would

reduce (although not remove) the threat of significantly enhanced ultraviolet
radiation at the surface of the Earth, but would in itself have no bearing on
the issue of climatic catastrophe, and would increase the intermediate-
timescale radioactive fallout. Within the present strategic arsenals, there is no
mix of yields that simultaneously minimizes ionizing radiation from fallout
and ultraviolet radiation from the Sun.
As delivery system accuracy has progressively improved, there has been a
corresponding tendency toward the deployment of lower-yield warheads,
although not through any concern about the integrity of the ozonosphere.
There is also a trend toward higher fission fractions, implying more radio-
active fallout. Limitations on the sizes and therefore, to some extent, on the
yields of new warheads are part of recent U.S. arms control proposals. With
the bulk of Soviet strategic warheads having yields larger than their U.S.
counterparts, however, treaties limiting high yields place greater demands on
Soviet than on U.S. compliance. Moreover, to enforce a categorical yield
ceiling seems to imply verification problems of some difficulty.
The second part of the proposal is to guarantee by treaty that cities would
not be targeted. Then the worst of the climatic effects might be avoided,
although the climatic consequences of "pure" counterforce exchanges can
still be extremely serious (Figure 1).The encoding of targeting coordinates,
however, is in principle done remotely, and involves different coordinates for
each warhead. Even if we could imagine international inspection teams
descending unannounced on Soviet or American missile silos to inspect the
targeting coordinates, an hour later the coordinates could be returned to
those appropriate for cities.
Targeting policy is among the most sensitive aspects of nuclear strategy,
and maintaining uncertainty about targeting policy is thought to be an
essential component of U.S. deterrence. The proposal is unlikely to be
received warmly by the U.S. Joint Strategic Targeting Staff or its Soviet
counterpart. It is also difficult to understand how those skeptical of the
verifiability by reconnaissance satellites of SALT I1 provisions on the deploy-
ment of missiles ten meters long can rest easy about verification of treaties
controlling what is encoded in a microchip one millimeter long. Neverthe-
less, a symbolic, unverifiable targeting treaty, entered into because both
sides recognize that it is not in their interest to target cities, might have
some merit.
4. Transition to Low-Yield High-Accuracy Arsenals. A conceivable re-
sponse to the prospect of climatic catastrophe might be to continue present
trends toward lower-yield and higher-accuracy missiles, perhaps accom-
panied by development of the technology for warheads to burrow sub-surface
before detonating. Payloads have been developed for the Pershing I1 missile
that use radar area-correlators for target recognition and terminal guidance;
the targeting probable error is said to be 40 meter^.^" It is evident that
a technology is gradually emerging that could permit delivery accuracies of
35 meters or better over intercontinental ranges.
.,I& Nuclear War and Climatic Catastrophe 349
It is evident as well that burrowing technology is also under rapid devel-

opment." A one-kiloton burst, two to three meters sub-surface, will excavate
a crater roughly 60 meters across.12 Clearly, high-accuracy penetrating war-
heads in the one-to-ten-kiloton range would be able, with high reliability, to
destroy even very hardened silos and underground command posts.
Low-yield sub-surface explosions of this sort cannot threaten the
ozonosphere. They minimize fires, soot, stratospheric dust and radioactive
fallout. Even several thousand simultaneous such detonations might not
trigger the nuclear winter. Similar technology might be used for pinpoint
attacks on military/industrial targets in urban areas. Thus, the TTAPS results
will probably lead to calls for further improvements in high-accuracy
earth-burrowing warheads.
There are, I think, a number of difficulties with this proposal, as attractive
as it seems in a strictly military context. A world in which the nuclear arsenals
were completely converted to a relatively small number of burrowing low-
yield warheads would be much safer in terms of the climatic catastrophe. Rut
such warheads are provocative. They are the perfect post- TTAPS first-strike
weapon. Their development might well be taken as a serious interest in mak-
ing a climatically safe but disabling first strike. Greatly expanded deployment
of anti-ballistic missiles might be one consequence of their buildup.
Retaliation from surviving silos, aircraft and especially submarines, as
discussed above, is likely, whatever the disposition of yields in a first strike.
Also, arsenals cannot be converted instantmeously. There would be a very
dangerous and protracted transition period in which enough newer
weapons are deployed to be destabilizing, and enough older weapons are
still in place to trigger the nuclear winter.
However, if the inventories of modern higher-yield (more than ten kiloton)
warheads were first brought below threshold, a coordinated U.S.-Soviet
deployment of low-yield burrowers might be accomplished in somewhat
greater safety. O n many launchers, each with a single warhead, they might
provide a useful reassurance to defense ministries a t some points in the tran-
sition process. At any rate, the dramatic reduction of arsenals neccssary to go
below threshold before large-scale burrower deployment is indistinguishable
from major arms reduction for its own sake (see below).
5. Consequences for the Developing World. Before the TTAPS calculations
were performed, it was possible to argue that the developing world would
be severely affected by secondary economic consequences, but not funda-
mentally destroyed by a northern mid-latitude nuclear war. Now it seems
more likely that nations having no part in the conflict - even nations entirely
neutral in the global confrontation between the United States and the Soviet
Union - might be reduced to prehistoric population levels and economies, or
worse. Nations between 70°N and 30°S, nations with marginal economies,
nations with large food imports or extensive malnutrition today, and nations
with their own strategic targets are particularly at risk.
Thus, the very survival of nations distant from any likely nuclear conflict
can now be seen to depend on the prudence and wisdom of the major
nuclear powers. India, Brazil, Nigeria or Saudi Arabia could collapse in a

nuclear war without a single bomb being dropped on their t e r r i t ~ r i e s . ~ ~
Quite apart from any concern about the deflection of world financial,
technical and intellectual resources to the nuclear arms race, the prospect of
nuclear war now clearly and visibly threatens every nation and every person
on the planet. The diplomatic and economic pressure accordingly placed on
the five nuclear powers by the other nations of the world, concerned about
their own survival, could be at least marginally significant.
6. Shelters. The usual sorts of shelters envisioned for civilian populations
are ineffective even for the nuclear war consequences known before the
TTAPS study. The more ambitious among them include food and water for a
week or two, modest heating capabilities, rudimentary sanitary and air fil-
tration facilities and no provisions for the psychological burdens of an
extended stay below ground with unknown climatic and ecological conse-
quences propagating overhead. The kinds of shelters suitable for prolonged
subfreezing temperatures, high radiation doses, and pyrotoxins would have
to be very much more elaborate - quite apart from the question of what
good it would be to emerge six or nine months later to an ultraviolet-bathed
and biologically depauperate surface, with insect pests proliferating, disease
rampant, and the basis of agriculture destroyed.
Appropriate shelters, able to service individual families or family groups
for months to a year, are too expensive for most families even in the afflu-
ent West. The construction of major government shelters for civilian popu-
lations would be enormously expensive as well as in itself potentially
destabilizing. The prospect of the climatic catastrophe also heightens the
perceived inequity between government leaders and (in some cases) their
families, provided elaborate shelters, and the bulk of the civilian popula-
tion, unable to afford even a minimally adequate shelter.
But even if it were possible to build perfectly effective shelters for the
entire populations of the United States and the Soviet Union, this would in
no way address the danger to which the rest of the world would be put.
Shelters for the combatant nations under circumstances in which only their
citizens are threatened are one thing. Shelters for the combatant nations
when gravely threatened noncombatant nations have only rudimentary or
nonexistent shelters are a very different matter.
7. Ballistic-Missile Defense Systems. It might be argued that the prospect
of a climatic catastrophe strengthens whatever arguments there may be for
ground-based or space-based ballistic missile defense (BMD) systems, as pro-
posed by President Reagan in his March 23, 1983 "Star Wars" speech. There
are grave technical, cost and policy difficulties with such proposals.24 Even
advocates do not envision it being fully operational in less than two or three
decades.
Optimistic informed estimates of porosity or "penetrance" (the fraction
of attacking missiles successfully detonating at their targets despite the BMD)
are no lower than 5 to 30 percent. The present world arsenal of strategic
warheads is so much greater than the threshold for climatic catastrophe that,
i,-iq,i.~ Nuclear War and Climatic Catastrophe 35 1
even if 5 to 30 percent of attacking missiles get through in something like a

full exchange, the catastrophe could be triggered. And most competent esti-
mates put the porosity - at least for the foreseeable future - at 5 0 percent to
99 percent. Further, one likely response to an adversary's anticipated deploy-
ment of BMD systems would be a proportionate increase in the stockpiles of
offensive warheads in compensation.
There are three phases in the trajectories of incoming missiles when they
might be attacked: boost phase, midcourse phase, and terminal phase.
Boost-phase and midcourse interception would, at best, require an untried
technology deployed at scales never before attempted. Only terminal-phase
KMDs exist at the present time (anti-ballistic missiles or ABMs), and even they,
ineffective as they are, may require ruinous capital investments before they
can provide meaningful levels of defense. Developments in terminal-phase
n~aneuverabilityof attacking warheads are likely to raise the price tag of an
effective BMD sharply again. Even in the best of circumstances, offense will
he more effective and less costly than defense.
Finally, terminal-phase interception, generally effective only for hard-
target defense, is characteristically designed to occur at very low altitudes.
There would be an advantage to the offense if it fused the incoming missiles
so they would explode if attacked ("sympathetic detonation"). In some
schemes, the BMD itself involves nuclear warheads exploded near the ground.
A fair fraction of hard targets, especially in Europe and the Soviet Union, are
within a few tens of kilometers of cities or forests. Thus, the most readily
deployable BMD suffers the disability, when it works at all, of generating fires
contributory to a climatic catastrophe, quite apart from its porosity.
8. Other Possibilities. There are a number of other conceivable responses
to the climatic catastrophe, some even more desperate than those discussed
above. For example, a nation might relocate its silos and mobile launchers
(the latter inviting barrage attack) to cities and forests to guarantee that
a barely adequate counterforce first strike by its adversary would trigger
a global climatic catastrophe with high confidence. Or nations with small
nuclear arsenals or marginal strategic capability might contemplate amassing
a threshold arsenal of some 500 to 2,000 deliverable warheads in order to be
taken seriously in "great power" politics.
But these and similar contrivances increase the probability of nuclear
war or the dangers attendant to nuclear war sufficiently that they are likely
to be rejected by the nation contemplating such moves or, failing that, by
other nations. Major relocations of strategic weapons systems or the
deployment of new strategic arsenals are readily detectable by national
technical means.
None of the foregoing possible strategic and policy responses to the

prospect of a nuclear war-triggered climatic catastrophe seem adequate
even for the security of the nuclear powers, much less for the rest of the
world. The prospect reinforces, in the short run, the standard arguments for
strategic confidence-building, especially between the United States and the
Soviet Union; for tempering puerile rhetoric; for resisting the temptation to
demonize the adversary; for reducing the likelihood of strategic confronta-
tions arising from accident or miscalculation; for stabilizing old and new
weapons systems - for example, by de-MIRVing missiles; for abandoning
nuclear-war-fighting strategies and mistrusting the possibility of "contain-
ment" of a tactical or limited nuclear war; for considering safe unilateral
steps, such as the retiring of some old weapons systems with very high-yield
warheads; for improving communications a t all levels, especially among
general staffs and between heads of governments; and for public declar-
ations of relevant policy changes. The United States might also contemplate
ratification of SALT I1 and of the 1948 U.N. Convention on the Prevention
and Punishment of the Crime of Genocide (ratified by 92 nations, includ-
ing the Soviet Union).
Both nations might consider abandoning apocalyptic threats and doctrines.
To the extent that these are not credible, they undermine deterrence; to the
extent that they are credible, they set in motion events that tend toward apoca-
lyptic conclusions.
In the long run, the prospect of climatic catastrophe raises real questions
about what is meant by national and international security. To me, it seems
clear that the species is in grave danger at least until the world arsenals are
reduced below the threshold for climatic catastrophe; the nations and the
global civilization would remain vulnerable even a t lower inventories. It
may even be that, now, the only credible arsenal is below threshold. George
Kennan's celebrated proposal2s to reduce the world arsenals initially to 50
percent of their current numbers is recognized as hard enough to imple-
ment. But it would be only the first step toward what is now clearly and
urgently needed - a more than 90-percent reduction (Kennan proposed an
ultimate reduction of more than 84 percent - adequate for strategic deter-
rence, if that is considered essential, but unlikely to trigger the nuclear win-
ter. Still further reductions could then be contemplated.
The detonation of weapons stockpiles near or above threshold would
be, we can now recognize, in contravention of the 1977 Geneva Convention
on The Hostile Use of Environmental Modification Techniques, signed by
48 nations and duly ratified by the Soviet Union and the United States.26
And Article 6 of the 1968 Nuclear Non-Proliferation Treaty requires the
United States and the Soviet Union, among other signatory states, "to pur-
sue negotiations in good faith on effective measures relating to cessation of
the nuclear arms race at an early date and to nuclear disarmament. ..." I d o
not imagine that these treaties can, by themselves, play a determining role
in producing major reductions in the world strategic arsenals, but they
establish some sense of international obligation and can at least expedite
urgent bilateral and multilateral consultations.
7 I 1 Nuclear War and Climatic Catastrophe 353
We have, by slow and imperceptible steps, been constructing a Doomsday

Machine. Until recently - and then, only by accident - no one even noticed.
And we have distributed its triggers all over the Northern Hemisphere. Every
American and Soviet leader since 1945 has made critical decisions regarding
nuclear war in total ignorance of the climatic catastrophe. Perhaps this
knowledge would have moderated the subsequent course of world events
and, especially, the nuclear arms race. Today, at least, we have no excuse for
failing to factor the catastrophe into long-term decisions on strategic policy.
Since it is the soot produced by urban fires that is the most sensitive trig-
ger of the climatic catastrophe, and since such fires can be ignited even by
low-yield strategic weapons, it appears that the most critical ready index of
the world nuclear arsenals, in terms of climatic change, may be the total
number of strategic warheads. (There is some dependence on yield, to be
sure, and future very low-yield, high-accuracy burrowing warheads could
destroy strategic targets without triggering the nuclear winter, as discussed
above.) For other purposes there are other indices - numbers of submarine-
launched warheads, throw-weight (net payload deliverable to target), total
megatonnage, etc. From different choices of such indices, different conclu-
sions about strategic parity can be drawn. In the total number of strategic
warheads, however, the United States is "ahead" of the Soviet Union and
always has been.
Very roughly, the level of the world strategic arsenals necessary to induce
the climatic catastrophe seems to be somewhere around 500 to 2,000
warheads - an estimate that may be somewhat high for airbursts over cities,
and somewhat low for high-yield ground-bursts. The intrinsic uncertainty
in this number is itself of strategic importance, and prudent policy would
assume a value below the low end of the plausible range.
National or global inventories above this rough threshold move the
world arsenals into a region that might be called the "Doomsday Zone." I f
the world arsenals were well below this rough threshold, no concatenation
of computer malfunction, carelessness, unauthorized acts, communications
failure, miscalculation and madness in high office could unleash the nuclear
winter. When global arsenals are above the threshold, such a catastrophe is
at least possible. The further above threshold we are, the more likely it is
that a major exchange would trigger the climatic catastrophe.
Traditional belief and childhood experience teach that more weapons buy
more security. But since the advent of nuclear weapons and the acquisition of
a capacity for "overkill," the possibility has arisen that, past a certain point,
more nuclear weapons do not increase national security. I wish here to sug-
gest that, beyond the climatic threshold, an increase in the number of stra-
tegic weapons leads to a pronounced decline in national (and global) security.
National security is not a zero-sum game. Strategic insecurity of one adver-
sary almost always means strategic insecurity for the other. Conventional
pre-1945 wisdom, no matter how deeply felt, is not an adequate guide in an

age of apocalyptic weapons.
If we are content with world inventories above the threshold, we are
saying that it is safe to trust the fate of our global civilization and perhaps
our species t o all leaders, civilian and military, of all present and future
major nuclear powers; and to the command and control efficiency and
technical reliability in those nations now and in the indefinite future. For
myself, I would far rather have a world in which the climatic catastrophe
cannot happen, independent of the vicissitudes of leaders, institutions and
machines. This seems to me elementary planetary hygiene, as well as ele-
mentary patriotism.
Something like a thousand warheads (or a few hundred megatons) is of the
same order as the arsenals that were publicly announced in the 1950s and
1960s as an unmistakable strategic deterrent, and as sufficient to destroy
either the United States or the Soviet Union "irrecoverably." Considerably
smaller arsenals would, with present improvements in accuracy and reliability,
probably suffice. Thus it is possible to contemplate a world in which the global
strategic arsenals are below threshold, where mutual deterrence is in effect to
discourage the use of those surviving warheads, and where, in the unhappy
event that some warheads are detonated, there is little likelihood of the
climatic ~atastrophe.~'
To achieve so dramatic a decline in the global arsenals will require not
only heroic measures by both the United States and the Soviet Union - it
will also require consistent action by Britain, France and China, especially
when the U.S. and Soviet arsenals are significantly reduced. Currently pro-
posed increments in the arsenals at least of France would bring that nation's
warhead inventory near or above threshold. I have already remarked on the
strategic instability, in the context of the climatic catastrophe only, of the
warhead inventories of these nations. But if major cuts in the U.S. and
Soviet arsenals were under way, it is not too much to hope that the other
major powers would, after negotiations, follow suit. These considerations
also underscore the danger of nuclear weapons proliferation to other
nations, especially when the major inventories are in steep decline.
Figure 2, on the following page, illustrates the growth of the American
and Soviet strategic inventories from 1945 to the present.28 To minimize con-
fusion in the Figure, the British, French and Chinese arsenals are not shown;
they are, however, as just mentioned, significant on the new scale of climat-
ically dangerous arsenals. We see from the Figure that the United States
passed the Doomsday Threshold around 1953, and the Soviet Union not
until about 1966. The largest disparity in the arsenals was in 1961
(a difference of some 6,000 warheads). At the present time the disparity is
less than it has been in any year since 1955. A published extrapolation of the
present strategic arsenals into 1985 is shown as dashed, nearly vertical lines,
accommodating new U.S. (Pershing 11, cruise, MX and Trident) and Soviet
(SS-21, -22, -23) strategic systems. If these extrapolations are valid, the
%gctt> Nuclear War and Climatic Catastrophe 355
Figure 2: Past a n d Future Nuclear Stockpiles
Climatic
Catdstrophe
Expected
1940 1950 1960 1970 1980 1990 2000 2010

YEAR
United States and the Soviet Union would have almost identical numbers of
inventories by the late 1980s.
The uppermost (dash-dot) curve in Figure 2 shows the total U.S. and
Soviet arsenals (essentially the world arsenals) climbing upward since about
1970 with a very steep slope, the slope steepening still more if the projection
is valid. Such exponential or near-exponential runaways are expected in
arms races where each side's rate of growth is proportional to its perception
of the adversary's weapons inventory; but it is likewise clear that such rapid
growth cannot continue indefinitely. In all natural and human systems, such
steep growth rates are eventually stopped, often catastrophically.
It is widely agreed - although different people have different justifica-
tions for this conclusion - that world arsenals must be reduced significantly.
There is also general agreement, with a few demurrers, that at least the early
and middle stages of a significant decline can be verified by national tech-
nical means and other procedures. The first stage of major arms reduction
will have to overcome a new source of reluctance, when almost all silos
could be reliably destroyed in a sub-threshold first strike. To overcome this
reluctance, both sides will have prudently maintained an invulnerable retali-
atory force, which itself would later move to sub-threshold levels. (It would
even be advantageous to each nation to provide certain assistance in the
development of such a force by the other.)
As arsenals are reduced still further, the fine tuning of the continuing
decline may have t o be worked out very carefully and with additional safe-
guards to guarantee continuing rough strategic parity. As threshold inventories
are approached, some verifiable upper limits on yields as well as numbers
would have to be worked out, to minimize the burning of cities if a nuclear
conflict erupted. O n the other hand, the deceleration of the arms race would
have an inertia of its own, as the acceleration does; and successful first steps
would create a climate conducive to subsequent steps.
There are three proposals now prominently discussed in the United States:
Nuclear Freeze, Build-Down, and Deep Cuts. Their possible effects are dia-
grammed in Figure 2. They are by no means mutually exclusive, nor d o they
exhaust the possible approaches. A negotiated Freeze would at least prevent
the continuing upward escalation in stockpiles, would forestall the deploy-
ment of more destabilizing systems, and would ~ r o b a b bel ~accompanied by
agreement on immediate annual phased reductions (the curved lines in the
middle to late 1980s in Figure 2). To reduce the perceived temptation for a
first strike, de-MIRVing of missiles during arms reduction may be essential.
The most commonly cited method of following the Freeze with reductions
is incorporated in the Kennedy-Hatfield Freeze Resolution: percentage reduc-
tions. Under this approach, the two sides would agree on a percentage - often
quoted as being between five percent and ten percent - and would agree to
decrease deployed warheads by that percentage annually. The percentage
reduction method was proposed to the Soviet Union by the United States
at the Vienna Summit in June 1979 and was to be applied to the limits and
sub-limits of the SALT I1 accords until these reached a reduction of 50 percent.
The Build-Down proposal is one in which modernization is permitted,

but each side must pay a price in additional reductions of warheads for each
warhead mounted on a modernized missile. In many current versions of the
proposal, it would also require both sides to decrease their total warhead
inventories by about five percent a year (again, the percentage annual reduc-
tion approach), to ensure that at least some reductions would take place
even if modernization did not. The rate of decline for Build-Down illustrated
in Figure 2 is essentially that of Representative Albert Gore (D.-Tenn.), in
which rough parity at 8,500 warheads each is adopted as a goal for
199 1-92, and the levels are reduced to 6,500 warheads each by 1997."
There is concern that the "modernization" of strategic systems that Build-
Down encourages might open the door to still more destabilizing weapons.
It is also by no means clear that all proponents of Build-Down envision
further reductions below the interim goal of about 5,000 warheads each for
the United States and Soviet Union. If this rate of Build-Down continued
indefinitely, the two nations would not cross back below threshold until
about the year 2020. As dramatic a change from the present circumstances
as this represents, in light of the present global crisis, it is, I think, too
leisurely a pace.
Deep Cuts, originally advocated by George Kennan and Noel Gayler"'
as an initial halving of the global arsenals in some relatively short period of
time, proposes the turning in of the fission triggers of thermonuclear
weapons, deployed or undeployed, to a binational or multinational author-
ity, with the triggers subsequently gainfully consumed in nuclear power
plants (the ultimate in beating swords into plowshares). A highly schematic
curve for something like Deep Cuts is also shown in Figure 2, starting from
Gore's assumption of parity by 1991-92. Halving of the present global
arsenals would then occur around 1995, and the global arsenals would
return to below the Doomsday Threshold by the year 2000.
The actual shape of these declining curves would very likely have kinks
and wiggles in them to accommodate the details of a bilaterally - and even-
tually multilaterally - agreed-upon plan to reduce the arsenals without
compromising the security of any of the nuclear powers. The Deep Cuts
curve shown has a rate of decline only about as steep as the rate of rise
beginning in 1970. Much steeper declines may be feasible and should be
considered.
No one contends it will be easy to reverse the nuclear arms race. It is
required at least for the same reasons that were used to justify the arms race
in the first place - the national security of the United States and the Soviet
Union. It is necessarily an enterprise of great magnitude. John Stuart Mill
said: "Against a great evil, a small remedy does not produce a small result. It
produces no result at all." But if the same technical ingenuity, dedication and
resources were devoted to the downward slopes in Figure 2 as to the upward
slopes, there is no reason to doubt that it could be negotiated safely.
In the deployment o f more stabilizing weapons systems, in the possible
development - especially in later stages of arms reductions - of novel means
of treaty verification, and (perhaps) in the augmentation of conventional

armaments, it will, of course, be expensive.
But, given the stakes, a prudent nuclear power should be willing to spend
more every year to defuse the arms race and prevent nuclear war than it does
on all military preparedness. For comparison, in the United States the annual
budget of the Department of Defense is about 10,000 times that of the Arms
Control and Disarmament Agency, quite apart from any questions about the
dedication and effectiveness of the ACDA. The equivalent disparity is even
greater in many other nations. I believe that the technical side of guarantee-
ing a major multilateral and strategically secure global arms reduction can
be devised and deployed for considerably less - perhaps even a factor of 100
less -than the planet's direct military expenditures of $540 billion per year.3'
Such figures give some feeling for the chasm that separates a prudent
policy in face of our present knowledge of nuclear war from the actual pres-
ent policies of the nuclear powers. Likewise, nations far removed from the
conflict, even nations with little or no investment in the quarrels among the
nuclear powers, stand to be destroyed in a nuclear war, rather than bene-
fiting from the mutual annihilation of the superpowers. They too, one
might think, would be wise to devote considerable resources to help ensure
that nuclear war does not break out.
In summary, cold, dark, radioactivity, pyrotoxins and ultraviolet light fol-

lowing a nuclear war - including some scenarios involving only a small
fraction of the world strategic arsenals - would imperil every survivor on
the planet. There is a real danger of the extinction of humanity. A thresh-
old exists at which the climatic catastrophe could be triggered, very
roughly around 500-2,000 strategic warheads. A major first strike may
be an act of national suicide, even if no retaliation occurs. Given the mag-
nitude of the potential loss, no policy declarations and no mechanical
safeguards can adequately guarantee the safety of the human species. No
national rivalry or ideological confrontation justifies putting the species at
risk. Accordingly, there is a critical need for safe and verifiable reductions
of the world strategic inventories to below threshold. At such levels, still ad-
equate for deterrence, at least the worst could not happen should a nuclear
war break out.
National security policies that seem prudent or even successful during a term
of office or a tour of duty may work to endanger national - and global - secur-
ity over longer periods of time. In many respects it is just such short-term think-
ing that is responsible for the present world crisis. The looming prospect of the
climatic catastrophe makes short-term thinking even more dangerous. The past
has been the enemy of the present, and the present the enemy of the future.
The problem cries out for an ecumenical perspective that rises above
cant, doctrine and mutual recrimination, however apparently justified, and
Sdg'tn Nuclear War and Climatic Catastrophe 359
that at least partly transcends parochial fealties in time and space. What
is urgently required is a coherent, mutually agreed upon, long-term policy
for dramatic reductions in nuclear armaments, and a deep commitment,
embracing decades, to carry it out.
O u r talent, while imperfect, to foresee the future consequences of our
present actions and to change our course appropriately is a hallmark of the
human species, and one of the chief reasons for our success over the past
million years. Our future depends entirely on how quickly and how broadly
we can refine this talent. We should plan for and cherish our fragile world
as we do our children and our grandchildren: there will be no other place
for them to live. It is nowhere ordained that we must remain in bondage to
nuclear weapons.
Acknowledgements
For stimulating discussions, and/or careful reviews of an earlier version of this a r t d e , I am

grateful t o Hans Rethe, McGeorge Bundy, Joan Chittester, Freeman Dyson, I'a~tl Ehrllch, Alton
trye, Richard Garwin, Noel Gayler, Jerome Grossman, Averell H'lrrimm, Mark Harwell, John
P. Holdren, Eric Jones, George F. Kenn.111, Robert S. McNamara, Carson Mark, P h i l ~ pMorrison,
l a y Orear, William Perry, David Plrncnrel, Theodore Postel, George Kathjens, Joseph Rotblat,
Herbert Scoville, Brent Scowcroft, J o h n Steinbruner, Jeremy Stone, Edward Teller, Br~,lnToon,
R ~ c h a r d'T'urco, I'JLII Warnke, Victor Weisskopt, Robert R. Wilson, and Albert Wohlstettcr. 'They
are howeber In n o wa! t o he held responsible tor the opinions stated o r the conclu\ions dr,~\vn.
I deeply appreciate the encouragement, suggestions a n d critical assessments provided by Lester
Grinspoon, Steven Soter and, especiall!: Ann Druyan, a n d the dedicated tran\criptions, through
ni,lny draft\, hy Mary Roth.
This 'irticle c\,ould not have been possible wlthout the h ~ g hscientif~ccompetence . ~ n d
J e d t c a t ~ o nof my co-authors 011 the TTAPS study, Rich'lrd I? Turco, Owen B. 'Ti)o11, Thomas
I? Ackerm.ln, and J a m e s B. Pollack, and my 19 coauthors of the accompan?rng s c ~ c n t ~ paper
t~c
on the long-term biological consequences of nuclear war. Finally, I \vish t o thank my Soviet col-
leagues, V.V. Alexandrov, E.1. Chazov, G.S. Gol~tsyn,and E.P. Velikhov among others, for organ-
t n n g independent confirmattons of the prolmble existence of 1' post-nuclear-war clirn,ltrc
catastrophe, , ~ n dtor helping t o generate a ditferent kind of cllrnate - o n e of mutual concern m c i
cooperation that is essential ~fw e are t o emerge sdfely from the tr'lp that our t w o n'itlons have
jo~nrlyset tor ourselves, our civil~zation,a n d o u r species.
Notes
1. R.I? Turco, O.B. Toon, T.P. Ackerrnan, J.B. Pollack and L t r l Sagan, ITTAPS] " G l o h ~ l
Attnospher~cConsequences of Nuclear War," Science, in press; P.R. Ehrlich, M.;\. Harwell,
Peter H. Raven, Carl Sagan, G.M. Woodwell, et al., "The Long-Term Biological Consequences
of Nucle.~rWar," Scirnce, in press.
2. Saniuel Glasstone and I'hil~pJ . Dolan, The Effects o f N L I C ~ PWar,
I I I . 3rd cd., \V,l\hington:
I)epartrnent of Defense, 1977.
3. T h e "tactical" Pershing I, for example, is listed as carrying warheads with yields a s high
as 400 kilotons, whlle the "strategic" Poseidon C-3 is listed with ,I yield of only 40 kilotons.
\Vorltf Arrnarnents and Drsi~nncment,SIPRl Yearbook 1982, Stockholm I n r e r n a t ~ o n ~Pc,~ce ll
Research Institute, L.ondon: Taylor a n d Francis, 1982; J . Record, U.S. N~rclenrWcrzpons in
Enropc. W ~ s h i n g t o n :Rrookings Institution, 1974.
4. See, e.g., D. Ball, Adelphi Paper 169, London: International Institute for Strategic
Studies, 1981; P. Bracken and M. Shubik, in Technology in Society, Vol. 4, 1982, p. 155.
5. National Academy of SciencesINational Research Council, Long-term Worldturde Effects
of Multiple Nuclear Weapons Detonations, Washington: National Academy o f Sciences, 1975;
Office of Technology Assessment, The Effects of Nuclear War, Washington, 1979; J. Peterson
(Ed.), Nuclear War: The Aftermath, special issue Ambio, Vol. 11, Nos. 2-3, Royal Swedish
Academy of Sciences, 1982; R.P. Turco, et a/., loc. cit. footnote 1; S. Bergstrom, et ul., Effects
of Nuclear War on Health and Health Services, Rome: World Health Organization, Publication
No. A36.12, 1983; National Academy of Sciences, new 1983 study in press.
6. See, e.g., J. Peterson, op. cit. footnote 5.
7. S. Bergstrom, up. cit. footnote 5.
8. [hid.
9. R. P. Turco, et al., loc. cit. footnote 1.
10. National Academy of Sciences, 1983, loc. cit. footnote 5.
11. J. Peterson, op. crt. footnote 6 .
12. National Academy of Sciences, 1975, op. cit. footnote 5.
13. The climatic threshold for smoke in the troposphere is about 100 million metric tons,
injected essentially all at once; for sub-micron fine dust in the stratosphere, about the same.
14. The slow warming of the Earth due to a CO, greenhouse effect attendant to the burn-
ing of fossil fuels should not be thought of as tempering the nuclear winter: the greenhouse
temperature increments are too small and too slow.
15. These results are dependent on important work by a large number of sc~entisrswho
have previously examined aspects of this subject; many of these workers are acknowledged in
the articles cited in footnote 1.
16. David Pimentel and Mark Sorrells, private communication, 1983.
17. C. H. Kruger, R. B. Setlow, et a/., Causes a n d Effects of Stratospheric Ozone Reduction:
An Update, Washington: National Academy of Sciences, 1982.
18. P. Ehrlich, et a[., loc. cit. footnote 1.
19. The term "Doomsday Machine" is due to Herman Kahn, Thinking About the
Unthinkable, New York: Horizon Press, 1962.
20. Aviation Week a n d Space Technology, May 15, 1978, p. 225.
21. Ihid.
22. S. Glasstone and P.J. D o h , op. cit. footnote 2.
23. The distribution of the coldest regions will vary with time and geography. In one recent
but still very crude three-dimensional simulation of the nuclear winter, the temperature has, by
4 0 days after the war, dropped by 15 to more than 4 0 centigrade degrees over much of the
globe, including a vast region extending from Chad to Novosibirsk, from the Caspian Sea to
Sri Lanka, embracing India, Pakistan and western China, and having its most severe effects in
Afghanistan, Iran and Saudi Arabia. V. V. Alexandrov and G. 1. Stenchikov, preprint,
Computing Center, U.S.S.R. Academy of Sciences, Moscow, 1983.
24. Richard Garwin, testimony before the Subcommittee on International Security and
Scientific Affairs of the House Committee on Foreign Affairs, U.S. Congress, November 10,
1983; Hans Bethe, manuscript in preparation.
25. George F. Kennan, "The Only Way Out of the Nuclear Nightmare," Manchester
Guardian Weekly, May 31, 1981. This is Kennan's acceptance speech for the Albert Einstein
Peace Prize on May 19, 1981, in Washington, D.C.
26. Article 1, paragraph 1, states: "Each State Party to this Convention undertakes not to
engage in military or any other hostile use of environmental mod~ficationtechniques having
widespread, long-lasting or severe effects as the means of destruction, damage, or injury to
another State Party." Paragraph 2 goes on: "Each State Parry to this Convention undertakes
not to assist, encourage or induce any State, group of States or international organization to
engage in actwities contrary to the provisions of paragraph 1 ..."
27. Since higher-yield tactical warheads can also be used t o burn cities, and might do so
inadvertently, especially in Europe, provision for their elimination should also eventually be
made. But initial attention should be directed to strategic warheads and their delivery systems.
Nuclear War and Climatic Catastrophe 36 1
28. T h e t o t ~ lwarheads calculated in Figure 2 include str'ltegtc a n d theater we'lpons, but

not t,lcrical weapons. N o t all p u h l ~ s h c dsources are in perfect agreement o n t h e w number\.
T h e p r ~ n c ~ psources d u\ed here ,Ire the Report of the Sccrctmy of Defcnsc [Harold Bror~vz/t o
the ( h g r c ~ s ort s the FY 1982 Brcilxet, FY 198 Arrtl~orrzntrorr Rcqrrc~stand t Y 1986 1l~~fi.rzsc
I'iograrns. W a s h ~ n g t o n : Dep,irrmenr o f Defense, 198 1; and Nutiom1 Dcfcnse Artdgc~
t~strr~zirt~s. FY 198 3 . Otficc of the Assi5tant Secretary of Defense, Comptroller, March 1982.
Ikyond 1983, projected increases in arsenals are shown tor U.S. lid Soviet ,trsen,~l$J \
ne:lrly vertical dashed lines. with the s u m of these arsenals a\ the line a t the top ot the Figure
t e r n ~ i n ~ l t ~Inn ga n arrowhead. T h e datll are from F r m k R x n a h ! In the speci,~li\cue of At~tbro
cited in footnote 5 , pp. 76-83. See also Cortrltrrforcc lssrtcs tor ~ I U.S. J ~ StrC7tcgic h'rrrlrL7r
tflrcrs, Congression,ll I3udget Office, January 1978.
Figure 2 shows three regions: a n upper region In which the n ~ ~ c l e winter ~ir could altno5t
certainly he triggered; a lower region a t which it could nor he tr~ggered:and a t r ' ~ n \ ~ t i o ;.one, n
shown shaded. 'The I ) o ~ ~ n d , ~ ot r i et ~h i ~transition zone are more uncert.lin than \ho\vn, and
depend among other t h ~ n g so n targeting strategy. Bur the threshold prob'ihly l ~ e shrr\veen se\.er.~l
hundred and a tew thousand conrempory strategic weapons.
2 9 . (:orzgrcssion~71Rrcord, August 4 , 1983, Vol. 129, No. 114.
3 0 . (;eorge t Kennan, / ( I ( - . cit. footnote 24; Noel Gayler, "How t o Break the Momentum
of the Nuclear Arms Race," T i ~ eN ~ York J Trntcs Mugrrrmc~.April 2.5, 1982.
3 I. Ruth I.eger Sivard, World Militury und Sociul Expcndrturi~s. 1.eesburg (VJ.): World
Prtorit~es, 198.3.
The OffensiveIDefensive Balance o f Military Technology:
A Theoretical and Historical Analysis
Jack S. Levy
T he literature on international relations and military history contains

numerous references to the offensive or defensive balance of military
technology and its impact on war. Historians often characterize a par-
ticular era as favoring the offense or the defense, and theorists often hypothe-
size that technology favoring the offense increases the likelihood of war or
contributes to empire-building. More generally, it has been suggested that the
history of warfare and weaponry can be viewed in terms of the interplay
between the offense and the defense (Snow, 1983: 83). These analyses are not
generally meaningful, however, because they are rarely guided by any explicit
definition of the key concept of the offensiveldefensive balance. The concept
itself has been defined in a variety of ways which are often contradictory and
which confuse the meaning of the hypotheses in question. Attempts to clas-
sify the balance historically are also inconsistent. These inconsistencies are
obscured by the failure of both the theoretical and historical literature to
acknowledge and build upon earlier scholarship and also by the absence of any
general review of the literature. As a result, little is known about the offensive1
defensive balance and its impact on war.
Much more work needs to be done if the concept of the offensiveldefensive
balance is to have any utility for explaining international conflict. This work
must begin with a critical assessment of the existing literature in order to
synthesize and evaluate the numerous but disparate and generally superfi-
cial treatments of the concept. Three aspects of the problem which are inte-
grally related but which are often treated separately in the literature must
be incorporated: hypothesis construction, concept definition, and historical
measurement.
The Hypotheses
One of the first attempts to generalize about the consequences of either

an offensive or defensive balance was that of Clausewitz (1976: 293), who
Source: lnternatzonal Studies Quarterly, 28(2) (1984):219-38.

Qc.t i T h e OffensiveDefensive Balance o f Military Technology 363
suggested that the superiority of the defense may leave both sides with no
incentive to attack and thus 'tame the elementary impetuosity of War.'
More explicit propositions were made by the proponents of the 'qualitative
principle' in the 1920s and 1930s. Their argument that offensive and defen-
sive weapons can be distinguished and that the former (and only the for-
mer) should be abolished was based on the explicit assumption that
offensive but not defensive weapons are conductive to war, an assumption
shared by most of the participants at the 1932 League of Nations Conference
for the Reduction and Limitation of Armaments (Wheeler-Bennett, 1935;
Boggs, 1941). Hart, for example, wrote that 'any strengthening of the de-
fensive at the expense of the offensive is a discouragement to aggression'
( 1932: 72).
A more systematic attempt to delineate the consequences of the offen-
siveldefensive balance of military technology was suggested by Wright in his
classic A Study of War. He argues that the superiority of the offense generally
results in the following: an increase in the probability of war; political expan-
sion, unification, and empire building; a decrease in the number of states
in the system; and shorter duration and lower costs of wars. Superiority of
the defense, on the other hand, results in a strengthening of loci1 areas jnd
thus facilitates revolts, the disintegration of empires, and the decentralization
of states; an increase in the number of states; decrease in the decisiveness of
wars and their importance for world politics; strategies of protracted stale-
mates and mutual attrition, which result in wars of longer duration and greater
destructiveness (Wright, 1965: 129, 292-93, 673, 797, 1520). Similarly,
Andreski (1968: 75-76) argues that
other things being equal, the predominance of the attack over defence
tends to diminish the number of independent governments within a
given area and to widen the areas under their control, and/or facilitates
the tightening of control over the areas already under their domination;
while the superiority of defence tends to produce opposite results.
One can fmd s l n ~ ~ l generallzatlons

ar In Quester's Offense and Defense ~n
the International System (1977). He wrltes that: 'Offenses produce war
and/or emplre; defenses support Independence and peace' (p. 208). When
the offense holds the advantage 'both srdes are p r ~ m e dto reap advantages
by push~ngInto each other's territory, (and) war may be extremely Ilkel)
whenever polit~calcrlsis erupts' (1977: 7). Offensive superiority is conducive
to emplre and a 'flnal pollt~cald e c ~ s ~ owhde
n' defens~vesuperlorlty leads to
p o l ~ t ~ c mdependence
al and prolonged wars ( 1977: 8, 3 1, 208). Quester rec-
ognizes the comphcatlons mtroduced by nuclear technology, and xgue4
that the c a p a b ~ l ~ tfor
y a counterforce offens~veencourages war whlle ,I
countervalue offensive capability promotes peace ( 1 977: 69).
Glpln also devotes attention to the relat~onsh~p between the offense and
the defense. He arguec that: ' M ~ l ~ t a rmnovatlons
v that tend to favor the
otfense over the defense st~mulateterrltorlal expanslon and the polltlcal con-
\olldatlon of lnternatlonal systems by emplres or great powers' ( 198 1 : 61 ).
Gilpin also argues that the offensiveldefensive balance affects the costs of
changing the status quo, and that the higher the costs the fewer incentives for
war (1981: 60-62). This is similar to Bean's (1973: 207) argument that defen-
sive superiority increases the costs of conquest and consequently reduces the
number of conquests, though those that d o occur take longer. The strong
implication is that defensive superiority reduces the likelihood of war.
Jervis provides the most systematic effort to trace the theoretical impact
of the offensiveldefensive balance on the likelihood of war. Using the con-
ceptual device of the security dilemma, Jervis (1978: 188-190) identifies a
number of related linkages between offensive superiority and war. Most
importantly, offensive superiority increases the benefits from striking first and
increases the costs of allowing the adversary to strike first. This increases in
turn the incentives to strike first and therefore the likelihood of war.
Defensive superiority reduces both the benefits to the attacker who initiates
a war and the costs to the defender who waits and absorbs the first blow,
leaves neither side with an incentive to strike first, and thus reduces the like-
lihood of war (Quester, 1977: 211). The likelihood of war is also increased
by the erroneous perception of offensive advantage.
Second, offensive superiority contributes to arms races, which are them-
selves assumed to lead to war. The expectation that war will be frequent and
short places a premium on high levels of existing armaments and on a quick
response to an adversary's increases in armaments. When the defense is super-
ior, however, inferior forces are sufficient for deterrence. Defensively superior
weapons may further provide a dampening effect on the arms race because in
such a situation security does not require the matching of the adversary arm
for arm. Third, Jervis argues that offensive superiority increases the incentive
to seek alliances in advance (Osgood, 1967: 81), which contributes to polar-
ization, tensions, and an increased probability of war.' It can also be argued
that all of the destabilizing dimensions of the spiral model, including the psy-
chological dynamics that reinforce them (Jervis, 1978: 67-66), work to the
fullest extent when the offense is superior.
There are numerous historical examples which are said to illustrate the
destabilizing consequences of offensive superiority.' Perhaps the most widely
cited is World War I. The perceived advantages of the offense (Hart, 1932: 72;
Farrar, 1973; Quester, 1977: 103; Jervis, 1978: 190-1 91) created enormous
pressures for early mobilizations, which were widely believed to make war
inevitable (Fay, 1928: 11, ch. X; Levy, 1983b). Israel's preemptive strike
against Egypt in 1967 may have been encouraged by the perceived advantages
of the offense. The stability of the nuclear balance is widely believed to derive
from the absence of incentives to strike first, incentives which are reduced
by the existence of invulnerable retaliatory capabilities and countervalue
potential.
While many of the hypotheses regarding the consequences of the offen-
siveldefensive balance are inherently plausible, there are critical analytical
problems which must be resolved before they can be accepted as meaningful or
valid. These problems have to do with the theoretical logic of the hypotheses,
T h e Offensive/Defensive B a l a n c e of Military Technology 365
the definition of the offensiveldefensive balance, and the empirical validity of

the hypotheses.
The hypothesis that the likelihood of war is increased when the military
technology favors the offense is theoretically plausible only on the basis of the
rather strong assunlption that decisionmakers correctly perceive the offen-
siveldefensive balance. However, it is perceptions of one's psychological envir-
onment that determine decisions, not the 'objective' operational environment
(Sprout and Sprout, 1965). The assumption of accurate perceptions is there-
fore open to question. The inherent difficulty of determining the offensive1
defensive balance and the alleged tendency of the military to prepare for the
last war rather than the next one may result in some profound misperceptions.
I t is widely agreed, for example, that in 1914 military technology favored thc
defense (Hart, 1932: 75; Fuller, 1961: ch. 8-9; Montgomery, 1983: 472) but
that most decisionmakers perceived that it favored the offense. It was not the
offensiveldefensive balance that intensified worst-case analysis and increased
the incentives for preemption, b ~ i tdecisionmakers' perceptions of that bal-
ance. If the offensiveldefensive balance is not defined in terms of the percep-
tions of decisionmakers (and in most conceptualizations it is not so defined),
thcn the hypothesis is technically misspecified. Hypothesis regarding the con-
sequences of war, on the other hand, are properly defined in terms of the
'objective' balance.
The second problem relates to the definition of the offensiveldefensive
balance. What does it mean to say that the offense is superior to the defense,
or vice r w s a ? This will be treated at length in the following section, but one
point should be made here. An hypothesis regarding the offensiveldefensive
balance has no explanatory power unless that concept can be nominally and
operationally defined independently of its hypothesized effects. For example,
Wright ( 1 965: 796-97) states that
... it is difficult to judge the relative power of the offensive and defensive
except by a historical audit to determine whether on the whole, in a
given state of military technology, military violence had or had not
proved a useful instrument of political change. ... During periods when
dissatisfied powers have, on the whole, gained their aims by a resort to
arms, it may he assumed, on the level of grand strategy, that the power
of the offensive has been greater. During the periods when they have not
been able to d o so, it may be assumed that the power of the grand stra-
tegic defensive has been greater.
It would be tautological to use this conception of the offensiveldefensive bal-

ance to predict to the military success of the aggressor, though it would he
legitinlate to predict to the frequency of war. Similarly, it is meaningless to
hypothesize that offensive superiority increases the incentive to strike first if
the offensiveldefensive balance is defined by the incentive to strike first. The
separation of hypothesis construction from concept definition and the absence
of rigorous definition has increased the dangers of tautological propositions.
The failure to subject these hypotheses to systematic empirical testing is

another major problem. Most attempts to identify the offensiveldefensive
advantage in various historical eras are not guided by an explicit definition
of the concept,bnd rarely is there a demonstration that a given balance had
an effect on the frequency of wars occurring or on the decisions for a par-
ticular war. The apparent a priori plausibility of a particular hypothesis
may derive more from its tautological construction than from its corres-
pondence with reality. In the absence of a more thorough analytic treatment
and a more systematic empirical analysis the validity of any of these hypoth-
eses cannot be accepted.
Definitions of the Offensive/Defensive Balance
Use of the concept of offensiveldefensive balance to refer to a variety of dif-

ferent things has led to a great deal of confusion. Theoretical propositions
which are meaningful or interesting for one use of the term may not be very
meaningful for another, and for this reason the various usages of the con-
cept must be identified and examined.
Concern here is with the offensiveldefensive distinction with respect to
military technology and perhaps tactics4 but not with respect to policy. The
question of whether national policy is offensive (aggressive) or defensive is
not unimportant, but is analytically distinct and not directly relevant to the
hypotheses surveyed earlier. These propositions all suggest that there is some-
thing about military technology itself that affects the likelihood or nature of
war, and that what is important is whether technology gives an advantage to
the offense or defense. This relative advantage may be one of several variables
affecting the likelihood of war by affecting policy, but itself is analytically dis-
tinct from policy.
The offensiveldefensive balance of military technology has been defined
primarily in terms of the ease of territorial conquest, the characteristics of
armaments, the resources needed by the offense in order to overcome the
defense, and the incentive to strike first.
Territorial C o n q u e s t
The most common use of the concept of the offensiveldefensive balance is

based on territorial conquest and the defeat of enemy forces. Quester (1977:
15) states that 'the territorial fixation then logically establishes our distinction
between offense and defense.' Jervis (1978: 187) argues that an offensive
advantage means that 'it is easier to destroy the other's army and take its terri-
tory than it is to defend one's own.' 'The essence of defense,' on the other
hand, 'is keeping the other side out of your territory. A purely defensive
weapon is one that can d o this without being able to penetrate the enemy's
land' (1978: 203). A defensive advantage means that 'it is easier to protect
and hold than it is to move forward, destroy, and take' (Jervis, 1978: 187).
I I \ , The Offensive/Defensive Balance of Military Technology 367
Wright (1965: 793) lncludes these notions of defeat of enemy forces and
territor~alselzure In h ~ rather
s complex definition:
O n a tactical level the offensive or defensive quality of a unit may be

estimated by considering its utility in an attack upon a n enemy unit like
itself or in an attack upon some other concrete enemy objective, such as
territory, commerce, or morale.'
A primary purpose of protecting territory, of course, is the protection of people

and property. What is perhaps implicit in the above definitions is made
explicit by Tarr (1983): 'Defense refers to techniques and actions, both active
and passive, to repel attack, to protect people and property, to hold territory,
and to minimize damage by the attacker.' This linkage of territorial conquest
to p o p ~ ~ l a t i odefense
n creates a problem, however. While territorial defense
was sufficient for the protection of people and property in the pre-nuclear era
(or at least in the era before strategic bombardment), that is no longer true.
As Schelling (1966: ch. 1 ) and others have noted, the uniqueness of the nuclear
age lies in the fact that the defeat of the adversary's military forces and terri-
torial penetration are no longer necessary for the destruction of his popula-
tion centers. The destruction of population and the coercive power that it
makes possible are no longer contingent upon military victory. For this rea-
son the protection of territory (from invasion) is analytically distinct from the
protection of population. The inclusion of both in a definition of the offen-
siveldefensive distinction only creates confusion (unless the use of that con-
cept is explicitly restricted to the pre-nuclear era), for the hypothesized effects
of a n 'offensive' advantage are precisely the opposite for the two concepts.
The likelihood of war presumably increases as territorial conquest becomes
easier, because the probability of victory increases while its expected costs
decrease. But the ability t o destroy enemy population and industrial centers
contributes to deterrence in the nuclear age, and therefore it decreases the
likelihood of war (or at least nuclear war).
It is because of the distinction between deterrence and defense (Snyder,
1961: 14-16) that the meaning of the offensiveldefensive balance may dif-
fer in the nuclear and pre-nuclear eras. Whereas in the pre-nuclear era both
deterrence and defense were based on the capacity to defeat the armed
forces of the enemy, that is only true for defense in the nuclear age, for
deterrence is ultimately based on countervalue punishment. The use of mili-
tary force for the purpose of defeating enemy armed forces is analytically
distinct from the use of force for coercion (Schelling, 1966: ch, 2). Conse-
quently, traditional hypotheses (Wright and others) regarding the effects of the
offensiveldefensive balance of military technology are not necessarily applic-
able for nuclear powers a t the strategic level. Neither the concepts nor the
hypotheses are interchangeable.'
Now let us return t o the territorial conception of the offensiveldefensive
balance. O u r earlier discussion leads t o the question of what, besides the
numbers of troops or weapons, contributes t o the defeat of enemy forces
and conquest of territory. One answer is provided at the tactical level, based
on movement towards the armed forces, possessions, or territory of the enemy.
A condition of relative passivity and immobility in waiting for the enemy
to attack defines the strategic and tactical defensive (Wright, 1965: 807).
Clausewitz (as quoted in Boggs, 1941: 68) states:
What is defense in conception? The warding off a blow. What is then its
characteristic sign? The state of expectancy (or of waiting for this blow) ...
by this sign alone can the defensive be distinguished from the offensive in
war. ...
Clausewitz also writes: 'In tactics every combat, great or small, is defensive
if we leave the initiative to the enemy, and wait for his appearance on our
front' (as quoted in Boggs, 1941: 68).
Both offensive and defensive modes of war on the tactical level are neces-
sary, of course, for the achievement of either offensive or defensive object-
ives. The pursuit of any offensive goal requires a supporting defense, and the
defense alone can never bring victory but only stalemate. Mahan refers to
'the fundamental principle of naval war, that defense is insured only by
offence' (Boggs, 1941: 70). Clausewitz writes that an absolute defense is an
'absurdity' which 'completely contradicts the idea of war' (Boggs, 194 1: 71 ).
At some point it is necessary to seize the tactical offensive in order to avoid
defeat.'Thus the familiar maxim: the best defense is a good offense. It is nec-
essary, however, to distinguish between the strategic and tactical levels.
A general fighting offensively in strategic terms needs only to invade and then
hold territory to enable him to adopt the tactical defensive (Strachan, 1983).
It may be strategically advantageous to maneuver the enemy into a position
in which he is forced to take the tactical offensive under unfavorable condi-
tions. As the elder Moltke stated in 1865, 'our strategy must be offensive,
our tactics defensive' (Dupuy, 1980: 200). In addition, military tactics may
be offensive in one theater and defensive in another. The Schlieffen Plan, for
example, required a holding action against Russia in the east in order to
move against France in the west. Nevertheless, with certain types of weapons
systems more movement and tactical mobility is possible than with others.
It is difficult to measure movement historically while controlling for non-
technological f a ~ t o r s however.
,~ This leads us to the question of whether the
offensive/defensive balance can be defined by the characteristics of weapons
systems themselves.
T h e Characteristics of Armaments
While nearly all weapons can be used for either the strategic or tactical offen-
sive or the strategic or tactical defensive, the question is whether there are
some weapons systems which contribute disproportionately more to one than
to the other. As stated by the Naval Commission of the League of Nations
T h e OffensiveDefensive Balance of Military Technology 369
Conference for the Reduction and Limitation of Armaments (1932-1936)

(Koggs, 1941: 82).
Supposing one state either a ) adopts a policy of armed aggression or

b) undertakes offensive operations against another state, what are the
weapons which, by reason of their specific character, and without preju-
dice to their defensive purposes, are most likely to enable that policy or
those operations to be brought rapidly to a successful conclusion?
Hart (1932: 73) argues that certain weapons 'alone make it possible under
modern conditions to make a decisive offensive against a neighboring coun-
try.' What are the characteristics of such weapons?
Both Fuller and Hart identify mobility, striking power, and protection as
the essential characteristics of an offensive weapon (Wright, 196.5: 808).
Striking power (the impact of the blow) is not alone sufficient. A mobile
gun contributes more to the tactical offensive than an immobile one, and its
penetrating power is further enhanced if it is protected. But protection is
even more important for the defense. Mobility and protection are inversely
related, for it is easier to protect immobile weapons and wait passively for
the enemy to attack. The offensive value of the medieval knight ultimately
was negated by the heavy armor which protected him but restricted his
mobility. Thus Dupuy and Eliot (1937: 103) give particular emphasis to the
offensive advantages of mobility and striking power, noting that they too
may be in conflict. Roggs (1941: 84-85) argues that 'the defense disposes
especially of striking power and protection, to a lesser degree of mobility,
while the offense possesses mobility and striking power, and protection to
a lesser degree.' He concludes that mobility is the central characteristic of an
offensive weapon and arg~lesthat 'armament which greatly facilitates the
forward movement of the attacker might be said tentatively to possess rela-
tively greater offensive power than weapons which contribute primarily to
the stability o f the defender' (Boggs, 1941: 85). Our later survey of attempts
by military historians to identify the offensiveldefensive balance in various
historical eras will show that tactical mobility is the primary criterion used
to identify an advantage to the offense.
In terms of the characteristics of armaments, then, tactical mobility and
movement toward enemy forces and territory are the primary determinants
of the offense, at least in land warfare; protection and holding power con-
tribute more to the defense. Other weapon characteristics such as striking
power, rapidity of fire, and the range of a weapons system do not contribute
disproportionately to either the offense or the defense.' Much more work
needs to be done here, however, because of the lack of precision of some of
these concepts.
The classification of weapons systems by their contribution to mobility
and tactical movement toward enemy forces and territory is much less useful
for naval warfare. This was evident from the proceedings of the League of
Nations Conference for the Reduction and Limitation of Armaments, where

the problems and disagreements confronting the Naval Commission were even
more serious than those confronting the Land Commission and where tech-
nical arguments were even more likely to follow the flag (Boggs, 1941: 50-60).
The United States, among others, declared that the qualitative distinction
could not be applied to navies. Hart, a proponent of the qualitative principle
in general, restricted it to the materials of land warfare (Boggs, 1941: 50, 81).
The main problem with attempts to apply these principles of mobility and
tactical movement to naval warfare is the absence of anything comparable
to the territorial standard occurring in land warfare. The command of the
seas, the ultimate objective of naval warfare (Mahan, 1957), can be served
by passive as well as aggressive action, for the neutralization of the enemy
fleet by a blockade may serve the same function as its defeat. Moreover,
aggressive action toward the enemy fleet does not always result in battle,
for an inferior navy can often avoid battle without sacrificing major terri-
torial objectives, unlike land warfare.
Application of the territorially-based criterion of tactical mobility to aer-
ial weapons systems raises the question of whether the offensive or defensive
character of these weapons is determined independently of land warfare or
by their contribution to the defeat of enemy ground forces and territorial con-
quest. Many aerial weapons systems do contribute to the tactical offensive on
the ground because of their striking power, mobility, and surprise (for example,
in the Nazi blitzkrieg). Yet air power also has an independent capability to
destroy the enemy's war making industrial capabilities, and hence contributes
to deterrence in the nuclear age. This deterrent effect of air power takes place
independently of its effect on the tactical offensive on the ground but cannot
easily be incorporated into a conception of the offensiveldefensive balance
based on tactical mobility.
Some armaments that traditionally have been considered as defensive and
therefore assumed to be 'stabilizing' (in the sense that they discourage aggres-
sion and reduce the likelihood of war) are often considered to be destabiliz-
ing in the nuclear age. Air defenses, anti-ballistic missile defenses, and even
civil defenses are considered under the prevailing strategic doctrine to be
destabilizing because by protecting populations they threaten to undermine
deterrence. This reinforces our earlier point that the hypothesized conse-
quences of a military technology favoring the offense (or the defense) may not
be interchangeable between the nuclear and pre-nuclear eras.
The definition of the offensiveldefensive balance, in terms of the charac-
teristics of armaments, raises other questions as well. One is whether it
is possible to define the offensiveldefensive character of a weapon by its
intrinsic performance characteristics alone, apart from the prevailing doc-
trine that determines its use. For example, essentially the same tank that
was used in much of World War I as protected fire support was used in World
War I1 as the organizing element of mobile offensive warfare (Fuller, 1945:
ch. VI). The offensive character of Napoleonic warfare was due far more to
the innovative tactics of Napoleon than to weapons systems themselves
1 \ T h e Offensive/Defensive Balance of Military Technology 371
(Howard, 1976: 75-76; Preston and Wise, 1979: 189-191)."' It must be

concluded that the offensive or defensive character of a weapons system
must be defined by both its intrinsic characteristics and the tactical doctrine
which determines its use.
What is important, of course, is not the characteristics of an individual
weapon, but rather the aggregate impact of all weapons systems in a given
arsenal. How is a given mixture of armaments, designed for different pur-
poses and deployed for use in different theaters on land, sea, and air, to be
aggregated so that their net effect on the offense and defense can be classi-
fied? This overall impact cannot be determined apart from the composition
of an enemy's weapons systems and the terrain where the combat takes place.
The offensive value of the tank, for example, was reduced by the develop-
ment of new anti-tank technologies in the early 1970s. To complicate matters
further, most hypotheses relating to the offensiveldefensive balance treat that
concept as a systemic-level attribute (the hypotheses that offensive superior-
ity contributes to an increased frequency of war and to empire-building, for
example). They suggest that at a given time the offensiveldefensive balance
can be characterized by a single value throughout the system. The balance
must be aggregated not only over all weapons, functional roles, and theaters
for a given state, but also over all states in the system. This is difficult given
differential levels of industrialization and military power, uneven rates of
technological diffusion, and doctrinal differences among various states in the
system." Some of these problems are minimized, however, if the focus is
restricted to the leading powers in the system, because they are often compar-
able in terms of power and technology.
Gilpin distinguishes between the offense and the defense in terms of an eco-
nomic cost-benefit framework. 'To speak of a shift in favor of the offense
means that fewer resources than before must be expended on the offense in
order to overcome the defense' (Gilpin, 1981: 62-63). Gilpin goes on to say
that 'the defense is said to be superior if the resources required to capture ter-
ritory are greater than the value of the territory itself; the offense is superior
if the cost of conquest is less than the value of the territory' (p. 63).
Clearly the second definition does not follow from the first. Whereas the
first uses the relative costs of overcoming the defense at two different times
and independently of the resulting benefits, the second definition introduces
an entirely new concept - the actual value of the territorial conquest itself.
The value of territorial conquest is undoubtedly an important variable lead-
ing to war but it is analytically distinct from military technology and ought
to be treated separately. Under Gilpin's second definition the hypothesis
becomes equivalent to the statement that a positive (expected) utility of ter-
ritorial conquest increases the likelihood of war. This may be true (Bueno de
Mesquita, 1981), but it is not the hypothesis under consideration here.
Moreover, the definition of the offensiveldefensive balance by the utdity of
territorial conquest reduces to a tautology the hypothesis that offensive

superiority increases the utility of territorial conquest.
One of the two conceptualizations of the offensiveldefensive balance sug-
gested by Jervis is more consistent with Gilpin's first formulation. Jervis (1978:
188) poses the question as follows: 'Does the state have to spend more or less
than one dollar on defensive forces to offset each dollar spent by the other side
on forces that could be used to attack?' That is, what is the relative marginal
utility of devoting military spending to the offense rather than to the defense?
This approach is potentially valuable, but it is incomplete. It defines what
it means to say that the offense (or the defense) has an advantage, but fails to
provide any criteria for specifying what constitutes the offense or the defense
in the first place. The marginal utilities cannot be compared until the offense
and defense are first defined, and until this is done the concept is not particu-
larly useful.
The definition of the offensiveldefensive balance by the relative resources
that must be expended on the offense in order to overcome the defense can
be conceptualized in another way and related to the conception based on ter-
ritorial conquest. This refers to attackldefense ratios rather than military
spending. What ratio of troops does an attacker need in order to overcome
an enemy defending fixed positions? This notion is mentioned but not de-
veloped by Quester (1977: 212): 'The significant impact of defensive or offen-
sive technology shows up in the minimum ratios of numerical superiority
required for such an offensive.' It follows the same logic as Foch's comment
regarding the power of the offensive prior to World War I: 'Formerly many
guns were necessary to produce an effect. Today a few suffice' (Montross,
1960: 686). The conventional wisdom is that the offense needs at least a three-
to-one advantage, but the point here is that this ratio varies as a function of
existing military technology and the tactical doctrine guiding its use. The
offensive1 defensive balance is then defined as being inversely proportional to
the minimum ratio of forces needed by the attacker in order to overcome an
adversary defending fixed positions.12 The greater the minimum ratio, the
greater the advantage of the defense.
It is important to note here that the minimum ratio of forces needed by the
attacker in a particular era is analytically distinct from the relative numbers of
forces actually possessed by two adversaries in a particular situation. The
probability of victory is a function of both. To say that the balance of military
technology (as a function of attackldefense ratios) favors the offense does not
mean the attacker is likely to win. That would be true only if the attacker actu-
ally possessed the requisite number of troops in a particular situation.
The problem arises as to what ratio should be used as a baseline, the
zero-point indicating the transition from a defensive advantage to an offen-
sive advantage. The most obvious ratio is one to one, but that is widely
regarded as favoring the defense.'" While it would not be technically incor-
rect to say that the balance always favors the defense because the attacker
always requires numerical superiority, this is neither interesting nor useful.
If the offensiveldefensive balance is defined as attackldefense ratios, it is
T h e OffensiveIDefensive Balance o f Military Technology 373
preferable to conceive of this in relative rather than absolute terms. It is use-

ful to speak of shifts in the balance and to compare the balance at different
times, but not to speak about the absolute state of the balance. Thus the
hypothesis should technically state that 'the higher the minimum ratio of
forces needed by the attacker in order to overcome an adversary defending
fixed positions, the lower the likelihood of war.'
This conception of the offensive/defensive balance is more useful than the
others surveyed above, at least tor land warfare. The attacktdefense ratio
could he measured in one of two ways. It could be determined empirically
from an analysis o f a variety of battles in a given era, with the force ratios and
results determined for each and some kind of average computed. The prob-
lem, of course, would be the need to control for asymmetries in geography,
troop quality, and doctrine. Alternatively, the ratio could be conceived in per-
ceptual terms and measured by the perceptions of military and political elites
of what ratio of forces is necessary for either attack or defense. While this
information is not readily available it might be inferred from an examination
of the war plans of the leading states. The methodological problems involved
in either of these approaches are quite serious, however.
One of the questions asked by Jervis (1978: 188) of the offensiveldefensive

balance is the following: 'With a given inventory of forces, it is better to
attack or to defend? Is there an incentive to strike first or to absorb the
other's blow?' This conceptualization is more flexible than earlier criteria
based on tactical mobility and characteristics of armaments because it can
incorporate considerations of deterrence and be applied to the nuclear age.
It creates some problems, however, which Jervis may recognize hut does
not fully develop. For one thing, the hypothesis that a military technology
favoring the offense increases the incentive to strike first is reduced to a tau-
tology and hence carries no explanatory power. In focusing attention on the
linkage between the incentive to strike first and war,'"t ignores the more
basic question of what conditions create an incentive to strike first. These
antecedent conditions possess the greatest explanatory power and operate
through the intervening variable of the incentive t o strike first. This leads to
a related problem: there are numerous factors besides technology and doc-
trine affecting the incentive to strike first, including geographic constraints
and diplomatic and domestic political considerations, factors which also have
an independent effect on war. If the offensiveldefensive balance is defined as
the incentive to strike first, then it becomes confounded with these other
variables and it becomes impossible to distinguish their independent effects.
The incentive to strike first is best conceptualized as an intervening variable
leading to war and as the product of several distinct variables, one of which
is military technology and doctrine. One key issue is to elaborate the aspects
of military technology or doctrine which affect the incentives to strike first,
but this cannot be fully analyzed here.
It is important to distinguish the incentive to strike first from other con-

cepts that have also been used to define the offensiveldefensive balance. The
incentive to strike first should not be confused with aggressive policy, which
is influenced by a wide range of variables. A state may have revisionist ambi-
tions but be constrained by a military technology favoring the tactical defense,
as well as by other variables. Or, a state with purely defensive ambitions may
rationally initiate war if it perceives that through a preemptive strike it can
minimize its losses against an assumed aggressor. The distinction between the
incentive to strike first and seizing territory is particularly likely to be con-
fused.'Vhese are clearly distinct for naval and air warfare (particularly in the
nuclear age) but the difference is more profound. One may simultaneously
have a policy of not striking first and a strategy of active defense and ter-
ritorial penetration in the event that war does break out. This was Bismarck's
policy in the 1870s and 1880s (Langer, 1964) and perhaps Israel's in 1973.
Germany's Schlieffen Plan called for passive defense (holding ground) in the
East and rapid territorial penetration in northern France regardless of who
initiated the war.
The failure to recognize these distinctions only creates confusion and
may result in the incorrect use of hypotheses designed for other purposes.
Hypotheses appropriate for a territorially-based definition of the offensive1
defensive balance of military technology may not be valid for a definition
based on the incentive to strike first. While the ease of territorial seizure
may shorten wars and lower their costs (Wright, 1965: 673), this may not
necessarily be true for the incentive t o strike first. Nor is an incentive to
strike first in the nuclear age likely t o have the same consequences as an
attackldefense ratio which favors the offense. Further, it is not clear that the
incentive to strike first itself has the same causes or consequences in the pre-
nuclear period as it does in the nuclear age, though this might be an inter-
esting area for future research.
The same types of problems arise with respect to the various other con-
ceptualizations of the offensiveldefensive balance examined above. The con-
cept has been defined in terms of the defeat of enemy armed forces, territorial
conquest, protection of population, tactical mobility, the characteristics of
armaments, attackldefense ratios, the relative resources expended on the
offense and the defense, and the incentive to strike first. These separate def-
initions are often not interchangeable, and hypotheses based on one defin-
ition are often either implausible or tautological for another definition. This
is particularly true for applications of the offensiveldefensive balance to the
nuclear age. Because the most advanced weapons of this era are used pri-
marily for coercive purposes and the weapons of earlier eras were used pri-
marily to engage enemy armed forces, the concept of the offensiveldefensive
balance of military technology may mean entirely different things in the two
different situations. Certainly one reason for the confusion and ambiguity
among these hypotheses is the fact that they are based on common concepts
such as 'offense', 'aggressor', and 'initiator' which have ordinary language
meanings apart from more precise technical meanings. This is all the more
I c\ L The OffensiveIDefensive Balance of Military Technology 375
reason why any attempt to use such a concept must first define it explicitly
and be very clear regarding precisely which hypotheses are relevant.
Classification of the OffensiveIDefensive Balance in History
The third section of this article surveys a variety of efforts to classify the
offensiveldefensive balance of military technology in the Western inter-
national system over the last eight centuries." This survey will be useful
because of the absence of any previous review of this body of literature and
because of the general failure of earlier studies to acknowledge or build upon
each other. More importantly, it may reveal whether or not the concept has
acquired an informal definition in its empirical application, in spite of the
conceptual ambiguity demonstrated above. While the concept of the offen-
siveldefensive balance of military technology has taken on a variety of mean-
ings, the question arises as to how that concept has been used in attempts to
classify the offensiveldefensive balance in past historical eras. If different
authorities have generally used the offensiveldefensive balance to mean the
same thing (even in the absence of any formal nominal or operational defin-
ition), and if they have generally agreed on the state of the balance in various
historical eras, then it can be concluded that the ambiguity of the concept has
not precluded its effective use in empirical analysis. Consistent usage and
agreement by various authorities on the state of the balance in different his-
torical eras would permit 'intercoder agreement'to be used as the basis for
accurate historical measurement (provided these measurements are inde-
pendent)." Lack of agreement on classification, however, would suggest that
the collective judgment of authorities cannot he used as the basis for measure-
ment. It would also support the earlier conclusion that the offensiveldefensive
concept needs to be defined much more explicitly and rigorously before it can
be used in historical analysis.
There is little dissent from the view that the late Middle Ages was charac-
terized by the ascendancy of the defense over the offense in war. The Crusades
had stimulated a revival in military architecture, and advances in the art of for-
tification outpaced increases in destructive power and improvements in siege
tactics (Fuller, 1945: 68; Montross, 1960: 161-163; Ropp, 1962: 20; Nef,
1963: 185; Wright, 1965: 795, 1525; Osgood, 1967: 43; Brodie and Brodie,
1973: 31; Bean, 1973: 207; Preston and Wise, 1979: 68-69, 78, 8 1; Gilpin,
1981: 62; Montgomery, 1983: 166-171). As a result, only a small percentage
of sieges were successful (Montgomery, 1983: 169). The defensive power of
the new concentric stone castles was reinforced by logistical considerations.
Armies could not be maintained in the field for long periods and invading
armies could not easily bypass the feudal castles (Bean, 1973: 218). In addi-
tion, the replacement of chain mail by plate armor to protect the knight greatly
reduced his mobility (Montross, 1960: 163; Preston and Wise, 1979: 85) and
the pike-phalanx system was becoming increasingly invulnerable to cavalry
charge (Bean, 1973: 206). It has also be argued that the success of the English
with the longbow increased the tactical superiority of the defense (Dupuy,
1980: 88), presumably because its range made it more difficult for the offense
to close. For these reasons, all of the above authorities accept Oman's (1953:
356) argument that 'by 1300 the defensive obtained an almost complete mas-
tery over the offensive. l 8
By the mid-15th century fire power had moved from an auxiliary role to
one where it was central and decisive (Howard, 2976: 33). Developments in
heavy artillery led to a sharp resurgence of offensive superiority. This was sym-
bolized by the seige of Constantinople in 1453, where the greatest of all
medieval fortifications was reduced by the Turks in less than two months.
Dupuy (1980: 107) argues that by the end of the 15th century artillery had
made medieval fortifications obsolete. In addition, greater mobility, and hence
greater offensive capability, of this artillery is evidenced by the use of horse-
drawn artillery and chains of 'wagon forts' as mobile fortifications employ-
ing bombards (Montross, 1960: 189; Brodie, 1973: 51; Quester, 1977: 4 7 4 8 ;
Dupuy, 1980: 100). Small firearms also began to have a significant effect on
battle at the end of the 15th century, dominating over the pike and leaving
the armored knight vulnerable and lessening his local defensive effectiveness
(Nef, 1963: 29; Quester, 1977: 48-49). Thus a drastic change in the offensive1
defensive balance is said to occur close to 1450 (Fuller, 1945: 81-87; Montross,
1960: 193-95; Nef, 1963: 185-186; Wright, 1965: 294; Bean, 1973: 207;
Quester, 1977: 47-49; Preston and Wise, 1979: 91-92; Dupuy, 1980: 99,
106-107; McNeill, 1982: 83; Montgomery, 1983: 224).
There is much debate regarding how long this period of offensive super-
iority lasted. Wright (1965: 294-295, 795) argues that it lasted for two cen-
turies until 1648, a view supported by Nef (1963: 185) and Quester (1977:
49). Wright (1965: 294-95) points to the increase in mobility of infantry gen-
erated by the gradual abandonment of medieval armor; the disappearance of
pikemen, halberdiers, and heavy cavalry; and the adoption throughout Europe
of Turkish Janissary tactics, with armies equipped with cutlass and longbow
and supported by light cavalry and artillery. Mobility was further increased in
the first half of the 17th century after Gustavus Adolphus reduced the weight
of weapons, introduced the light field gun and the concept of mobile massed
artillery fire, and adopted a more flexible tactical organization (Dupuy, 1980:
137-38; McNeill, 1982: 123).
These arguments are rejected by other authorities who instead argue
that the science of fortification soon overcame the new developments in
artillery, leading to a shift back to the defense by 1525 or so (Hale, 1957:
274; Montross, 1960: 211, 250-54; Bean, 1973: 208; Howard, 1976: 35;
Preston and Wise, 1979: 106; McNeill, 1982: 90; Montgomery, 1983: 224).
Dupuy and Dupuy (1977: 455) argues: 'A 16th-century fortress, if provided
with adequate stocks of food and ammunition, was as impregnable as the
13th-century castle had been in its day.' Thus there is no consensus as to
whether the balance of military technology throughout most of the 16th
century and the first half of the 17th century favored the offense or the
defense.
I , The Offensive/Defensive Balance of Military Technology 377
Authorities generally agree that for nearly a century after 16.50 the bal-
ance of military technology lay with the defense." This was largely due to
the development of a new science of fortifications by Vauban and other mili-
tary engineers in the late 17th century (Guerlac, 1969). These elaborate for-
tifications became increasingly invulnerable to artillery, and frontal assault
became nearly impossible. This was the age of geometric warfare, of pos-
ition and maneuver rather than pitched battle. Military operations were
centered around fixed fortifications and were restricted by poor logistical
systems and short supply lines, and guns were deficient in range, accuracy,
and penetrating power (Preston and Wise, 1979: 1 4 2 4 4 ; Dupuy, 1980: 144).
Vauban also developed the science of siegecraft with his system of approaches
by parallels, but such systems generally remained one step behind systems
of fortifications.
The balance did not turn against the defense until 1789 (Nef, 1963: 185;
Wright, 1965: 295; Osgood, 1967: 46; Howard, 1976: 55; Quester, 1977: 57;
Preston and Wise, 1979: 142-143). This view is inconsistent, however, with
the general characterization of the warfare of Frederick the Great as offen-
sive in nature, based on Frederick's emphasis on the decisiveness of the bat-
tle rather than static maneuver, his willingness to take risks, his use of the
oblique order as a tactical device, and his emphasis on mobility (Dupuy, 1980:
148-1.54). Preston and Wise (1979: 147-149) recognize this and say simply
that Frederick differed from the norm of 18th-century warfare. The hesi-
tancy to characterize the military balance during this period as offensive
probably derives from the fact that Frederick's innovations were primarily
tactical and strategic rather than technological, and because niost historians
describing the military balance focus on the latter rather than the former.
Still, it cannot be denied that Frederick demonstrated what was possible
given the technology of the time. The recognition by many that Frederick
constitutes an exception to the static character of 18th-century warfare sug-
gests that the characterization of the entire 18th-century military balance as
defensive is open to question.
The Napoleonic period presented a similar set of problems; Napoleonic
warfare was characterized by mobility and the tactical offensive hut this had
little to do with military technology itself (Howard, 1976: 76). Preston and Wise
(1979: 189) argue that in some respects Napoleon was an 'arch-reactionary
toward new weapons and technological progress in the material of war' and
that his successes came through a 'more efficient use of well-known weapons.'
The offensive character of Napoleonic warfare derived from the generalship of
Napoleon and his changes in military organization, strategy, and tactics,
including the democratization of war and mass mobilization. The divisional
formation, the employment of light infantry, the use of the column of attack
instead of the line, a more flexible use of artillery on the battlefield to gain
a superiority of fire at a given point, and the logistical advantages of living
off the country were particularly important in contributing to mobility
(Fuller, 1961; ch. 3; Hart, 1964: ch. 8; Ropp, 1962: 98-102; Howard, 1976;
Quester, 1977: ch. 7; Preston and Wise, 1979: ch. 12; Strachan, 1983: ch. 4).
These authorities often refer to the offensive character of Napoleonic warfare

but do not trace it to the balance of military technology itself.
Few judgments are made regarding the offensiveldefensive balance for
the first half of the 19th century, perhaps because of the relative absence of
European war during that period. By mid-century, or by 1870 at the latest, the
balance had shifted in favor of the defense, which continued through World
War I. Reference is made to the holding power of entrenchments, barbed wire,
the machine gun, the breech-loading rifle, the difficulty of frontal assualt and
closing with the enemy, and to the generally static nature of warfare as demon-
strated in the American Civil War, the Russo-Turkish and Russo-Japanese
Wars, and others. As summarized by Boggs (1941: 76-77), beginning with the
American Civil War and extending through World War I there was a trend
'toward enormous increase in the masses of men under arms, and in the range,
casualty-producing capacity, and rapidity of fire of infantry weapons, without
any counteracting growth in the means of advancing of this fire.' This con-
clusion, including the view that the 'objective' balance favored the defense in
1914, is supported by Hart (1932: 72, 75), Millis (1956: 167), Montross
(1960: 633, 649), Fuller (1961: chs. 8-9), Ropp (1962: 162), Wright (1965:
1525), Brodie and Brodie (1973: 131-56), Howard (1976: 103, 105), Preston
and Wise (1979: 266), Dupuy (1980: 195, 199), Gilpin (1981: 62), and
Montgomery (1983: 441,458, 472).
Some of these authorities concede, however, that the railroad and the
development of motorized transport and then the tank all contributed to mo-
bility and helped the offense, as demonstrated by the wars of Bismarck
(Wright, 1949: 186, Brodie and Brodie, 1973: 148-50). Quester (1977: ch. 8)
suggests that for this reason there was for a time a net advantage to the of-
fense. Montross (1960: 649) rejects this view and argues that the offensive pre-
vailed in 1870 because 'intelligent defensive tactics were seldom employed.'
A more serious problem is raised by the gap between the 'objective' bal-
ance of military technology (as judged retrospectively by historians) and the
balance as perceived by the military and political leaders of the time. This
gap widened shortly after the American Civil War, the defensive lessons of
which were resisted by professional soldiers (Montross, 1960: 633; Fuller,
1961: ch. VI; McNeill, 1982: 242). By the turn of the century Foch and the
French school (Possony and Mantoux, 1969) epitomized the viewpoint pre-
vailing at the time that military technology favored the offense, and this belief
structured the war plans and influenced the behavior of the great powers
(Montross, 1960: 685-88; Hart, 1973: 72; Quester, 1977: 80; Jervis, 1978:
190-91; Dupuy, 1980: 216; Montgomery, 1983: 441; Strachan, 1983: 105-6).
The French Army Field Regulations of 1913 stated that 'the French Army ...
admits no law but the offensive' (Tuchman, 1962: 151). Most authorities
currently argue that the 'objective' balance favored the defense. Vyvyan
(1968: 165) argues: 'Never has the dogma of the offensive been more preva-
lent; never, because of the lead of firepower over tactical mobility, has that
dogma been less applicable.' Because of this gap between perceptions and
reality and because of the ambiguity regarding the role of perceptions in
L cc \ The Offensive/Defensive Balance of Military Technology 379
definitions of the offensiveldefensive balance, the period prior to World War I

becomes difficult to classify in terms of the offensive1 defensive balance.
The interwar period presents a similar gap between the analysis of mili-
tary historians and the perceptions of statesmen. Most military historians
argue that by 1930 or so the military technology favored the offense (Hart,
1932: 76; Millis, 1956: 252-53; Montross, 1960: 774; Fuller, 196 1: ch. 12;
Wright, 1965: 300-301; Gilpin, 1981: 62).20The speed, mobility, and strik-
ing power of the armored division with tactical air support had a great
advantage over field defenses and minor fortifications. The new warfare was
characterized by fluidity and speed, deep penetrations, and broad encircle-
ments. The stalemate of World War I had been transformed into the blitzkrieg
of World War II. Most observers at the time, however, perceived that the
military technology favored the defense (Montross, 1960: 766-67; Wright,
1965: 795; Gibson, 1969; Alexandroff and Rosecrance, 1977; Quester, 1977:
ch. 11; Jervis, 1978: 192-93). This view was disputed by the leading propon-
ents of armored warfare (Fuller, de Gaulle, Guderian, and Tukhachevski) but
their view was of the minority.21Thus the balance of military technology in
the interwar period becomes difficult to classify.
It is seen that military historians and others have evaluated the offensive1
defensive balance of the last eight centuries of the Western international sys-
tem in terms of the contribution of weapons systems to tactical mobility and
territorial penetration. This is in spite of their lack of formal definition of
the concept and the variety of theoretical perspectives surveyed earlier. This
implicit agreement on the appropriate criterion no longer exists, given the de-
cline in the dominance of land warfare and the rise of deterrence based on
nuclear punishment. Because of the lack of consensus on the meaning of the
offensiveldefensive balance in the nuclear age, it would not be useful to survey
attempts to classify the balance during this period. Let us now summarize the
extent of 'intercoder agreement' regarding the offensiveldefensive balance of
military technology for the previous eight centuries.
There is unanimous agreement among the references cited that the period
from 1200 to 1450 was characterized by defensive superiority and that the
period from 1450 to 1525 was characterized by offensive superiority. The
authorities are split on the 1525 to 1650 period. There is complete agree-
ment that the defense was superior from 1650 to 1740. Some argue that this
defensive superiority continues until 1789, though Frederick's emphasis on
the tactical offensive leads some to assert the opposite and others to make
no specific evaluation. The 1789 to 1815 period is generally regarded to
favor the offense but because of innovations in tactics rather than arma-
ments. Little attention is given to the 1815-1850 period. The next hundred
years pose a problem because of the gap between the objective and per-
ceived balance and the uncertain conceptual status of the latter. These author-
ities generally agree that from 1850 to 1925 or so the balance favored the
defense but that nearly all statesmen perceived that it favored the offense from
1870 to 1914. Similarly, from 1930 to 194.5 the balance favored the offense
but that the actors themselves perceived that it favored the defense.
A rough calculation shows the following degree of consensus among our

authorities. Of the 450 years from 1495-1945," only two periods totaling
5 5 years claim a definite consensus of offensive superiority. Two periods total-
ing at most 130 years claim a consensus of defensive superiority. Four periods
constituting a minimum of 265 years are uncertain, either because of diverg-
ing views, or because of the diametric opposition of the evaluations of actors
and analysts and the ambiguous conceptual status of perceptions in definitions
of the balance.23The inescapable conclusion is that there exists considerable
divergence of opinion among leading authorities regarding the offensive1
defensive balance during the last five centuries of the modern era, and that
a method of 'intercoder agreement' cannot be used to provide a basis for clas-
sification during this period. This analysis suggests that the concept of the
offensiveldefensive balance of military technology needs more theoretical
attention and operational definition before it can be applied to systematic
empirical analysis.
Conclusions
The concept of the offensiveldefensive balance of military technology has

been defined in the literature in terms of the defeat of enemy armed forces,
the ease of territorial conquest, protection of population, tactical mobility,
the characteristics of armaments, the relative resources expended on the
offense and the defense, and the incentive to strike first. I have suggested an
alternative definition based on attackldefense ratios: the offensiveldefensive
balance is inversely proportional to the ratio of troops needed by an attacker
to overcome an enemy defending fixed positions.
While many of these individual concepts may be useful, they mean fun-
damentally different things and are not interchangeable. This is particularly
true for the nuclear era, where the end of the predominance of land warfare
and the emergence of deterrence based on nuclear punishment has clouded
the offensiveldefensive distinction. Since the dominant weapons in the pre-
nuclear era were used primarily for the defeat of adversary armed forces,
whereas the most advanced weapons in the nuclear era are used by the lead-
ing powers primarily for coercion and bargaining, definitions of the offensive1
defensive balance that might be useful in the nuclear age may not be useful
in the pre-nuclear era, and vice-versa. The concept of the incentive to strike
first, which is often used to define an offensive advantage today, is not the same
as tactical mobility, which is widely used to define an offensive advantage in
earlier times.
The inclusion of fundamentally different concepts under the umbrella of
the offensiveldefensive balance has created considerable confusion in the lit-
erature. Because these different concepts are not interchangeable, theor-
etical propositions regarding the causes or consequences of an offensive or
defensive advantage may be useful for one definition but implausible for
another. Weapons characteristics that are stabilizing (i.e., reduce the likeli-
hood of war) in one era may be destabilizing in another. Hypotheses designed
t The Offensive/Defensive Balance of Military Technology 381
to explain the consequences of a military technology favoring tactical

mobility are not necessarily applicable to a military technology which creates
an incentive to strike first.
The ambiguity of the concept of the offensiveldefensive balance is not just
a function of the nuclearlpre-nuclear distinction. Even concepts restricted to
land warfare in the pre-nuclear era have different meanings and may have
different theoretical consequences. O u r survey of attempts by military his-
torians and others to identify the offensiveldefensive balance in various his-
torical eras is relevant here. In spite of the common focus on land warfare,
there is remarkably little consensus among these authorities regarding the
state of the offensiveldefensive balance in most periods. This suggests that
far more conceptual clarification and rigorous operational definition are
necessary before the offensiveldefensive distinction can be useful in histor-
ical analysis.
To conclude, the concept of the offensiveldefensive balance is too vague
and encompassing to be useful in theoretical analysis.24Many of the individ-
ual variables that have been incorporated into the more general idea may
themselves be useful, however. Few would doubt the utility for deterrence
theory of the concept of the incentive to strike first, for example, and the con-
cept of attackldefense ratios suggested here deserves further exploration. Much
more conceptualization is necessary before these individual variables can he
effectively used in empirical analysis, however. There is already a body of
theory regarding the consequences of an incentive to strike first. What is
needed are comparable theories regarding the consequences of military tech-
nologies which contribute to tactical mobility or to the ease of territorial con-
quest, or which reduce the ratio of forces needed by the attacker to overcome
an adversary defending fixed positions. Interaction effects between these sep-
arate variables also need to be explored. Further theoretical development of
this kind is necessary, because in its absence there is little reason to believe
that these individual concepts have an important impact on war, and there-
fore little reason to use these concepts in empirical analysis.
Author's Note
This rcse,lrch w ~ supported

s by a Fellowship for Independent Study and Rese,lrch from the
N x i o n a l Endowment for the Humanities. I wish t o thank Harrison Wagner, Cl~ffMorgan, Roger
Beaumont, and the editorr of ISQ for their helpful comments on earlier verslons of this article.
Notes
1. T h e theoret~calliterature is divided o n the question o f whether all~ancescontribute t o

w.ir o r t o peace. T h e e n i p ~ r ~ c evidence
al IS also mixed (Bueno d e Mesquitn a n d Singer, 1973;
Levy, 198 I ) . N o r have the linkages herween arms races a n d wars been thoroughly established
(Wallace, 1979).
2. Here 'destabili~ing' is used t o mean an increase in the likelihood o r f r e q u e l ~ qof w,lr.
3. T h e most notable exceptlon here is Quester ( 1 9 7 7 ) , though the connection bctwccn his
nominal definitions and historical classifications is n o t , ~ l w , ~ ycle'~r.
s
4. For this reason I am not concerned here with Boggs' (1941: 63-72) distinction between
the offensive and defensive on the 'grand strategic level', which he sees as based on a 'political'
or Clausewitzian theory of war. Boggs suggests that on this level the offensiveldefensive dis-
tinction is based on the political objectives toward which military operations are conducted,
and that the concept is generally defined this way by military theorists: 'The difference between
offensive and defensive is a difference in objectives, not a difference in the means employed to
reach the objective.' (Boggs, 1941: 72). This conceptualization not only involves the analytical
problem of distinguishing in principle between 'offensive' and 'defensive' policies (which vary
from theater to theater and during the course of a war) and the enormously difficult methodo-
logical problems involved in determining a state's objectives o r intentions. It also confounds
the two important concepts of the motivations o f statesmen and the offensiveldefensive bal-
ance, and deprives the latter of any independent meaning.
5. Several aspects of Wright's definition are open to question. The offensive character of a
weapon system cannot be judged only by its effectiveness against an enemy unit like itself, as
examples of submarines o r antitank weapons clearly indicate. Wright's focus on commerce as
the object of naval warfare (p. 793) can be questioned on the grounds that the primary aim of
seapower is the defeat o f the adversary's naval forces (Mahan, 1957). Nor is the notion of an
attack on enemy morale very useful.
6. Similarly, it is not at all clear that territorial conquest and defeat of enemy forces should
both be included in a single definition, for they do not necessarily go hand in hand. One obvi-
ous problem concerns naval and air warfare, where the defeat of enemy forces may be directed
toward control of sea lanes o r control of the air, but certainly not territorial conquest per se.
Even in land warfare, however, one can conceive of a strategy of deep territorial penetration
that aims to bypass enemy military forces rather than defeat them (see, for example, Vlgor,
1983), o r a strategy aimed to defeat enemy forces without seizing territory. The latter is also
relevant to the nuclear age, for a counterforce strategy may be aimed at destroymg enemy
forces for its own sake and independently of both territorial control or even coercion.
7. When Clausewitz (1976: 114) argues repeatedly that the defense is the stronger form of
warfare, he conceives of defense not in static terms as the warding off of blows, but rather as
a dynamic conception of a holding action until conditions are ripe for a counter-offensive
(Boggs, 1941: 71; Howard, 1983: 54).
8. One objective indicator of such movement is the relative rates of advance of armies in
different periods. Record (1973), for example, has surveyed historical rates of armored advance
in thls century. These rates are affected as much by the numbers of troops on each side, geo-
graphical terrain, and political considerations as they are by the balance of military technology,
so that its use as an indicator of the offensiveldefensive balance is open to question.
9. Wright (1965: 808) notes that Fuller also includes holding power (the ability to hold
captured territory) as an important element of an offensive weapon. If newly acquired terri-
tory cannot be held there is little net gain for the offense, yet by definition it is even more
important for the defense to hold territory. Wright (1965: 808) also argues that rap~dityof fire
is another weapons characteristic that provides an advantage to the defense, but it is not clear
whether this generalization applies beyond the case of the machine gun and the static warfare
of the late 19th and early 20th centuries. In addition, Wright (1965: 808) argues that the range
of a weapon contributes a net advantage to the defense because it keeps the offense at a dis-
tance and restricts its mobility. But long-range weapons may contribute equally to the pene-
trating power of the offense by weakening defensive fortifications from a distance. Boggs
(1941: 86), for example, argues that while heavy mobile artillery contributes t o both the tac-
tical offensive and defensive, its striking power against enemy fortifications dominates, thereby
favoring the offense. More generally, Fuller (1945: 9 ) argues that range IS the dominant char-
acteristic of an offensive weapon.
10. A more esoteric example of tactical doctrine precluding the optimal use of available
military technology can be found in McNeill (1982: 9-11). The Asians' use of war chariots as
fighting platforms as well as for transport increased their mobility and firepower beginning in
the 18th century BC, but the Europeans lost these offensive advantages because of a doctrine
which led them to dismount from the chariots and fight o n foot as infantry.
I \ The Offensive/Defensive Balance of Military Technology 383
1 1. Note that the greater the extent t o which the offensiveldefensive halance is affected by
doctrtnnl considerat~ons,the less its utility as a systemic-level concept.
12. O n e obvious prohleni here is what is meant by 'overcoming an adversary'. Here 1 mean
the minimum ratio of forces necessary t o give the attacker a higher probability o f winning than
los~ng,recognlzlng that stalemate I S also a possible outcome. Other definitions are possible, o f
course ( a .iO% chance of victory, or of avoiding defeat, for example), but the specif~ccriterion
is less important than its consistent application. It is recogn~zed,of course, that we must make
the ceterrs parihrts assumption in order t o control for asymmetries in terrain, logistics, morale,
tra~ning,and leadership, whlch are also important.
13. The fact that a one-to-one ratio is widely regarded to he insufficient tor attack pro-
vides a basis for an interpretation of the cornmon argument that the advantage in war always
lies with the defense (Clausewitz, 1976: 114, 128; Machiavell~,Discourses, bk i, ch. XLV;
Dupuy, 1980: 326). To say this does not mean, however, that the extent of the advanmgc to
the defense 1s constant.
14. The theoret~calconsequences of an incentive t o strike first have heen thoroughly
explored by deterrence theorists (Ellsherg, 1960; Schelling, 1960; Wagner, 1983).
15. This point was emphasized to me by Harrison Wagner.
16. This survey begins in the late M ~ d d l eAges because references t o the offensive1
defensive balance in earlier times are few and scattered, as are references t o non-Western
systems.
17. This assmes that the authorities consulted reflect a representative sample of viewpoints
and they make independent evaluations of the offensiveldefensive balance over time. hdm~ttedly,
there may he some bias in the authorities consulted 111 the following survey, for a11 are
Anglophones who deal prl~narilywith land warfare and basically ignore naval and air consider-
attons. The land focus is n o t a problem, for most theoret~caltreatments of the offensiveldefen-
slve h;il,ince define it in these terms. Nor is there good reason .to believe that continental military
h~storianswould reach fundamentally different conclusions, particularly since they also would
tend to Ignore naval and alr considerations. More serious, however, is the questlon of ~ndepend-
cnt measurement. I'resumahl!: there is some reciprocal and cumulative relation\hip between
these authorities, so that their classifications of the offenslveldefensive balance are not truly inde-
pendent. This is particularly likely glven the absence of rigorous nominal or operational defin-
itions guiding their analyses. It is in this sense that a greater variety of sources rn~ghthe valuable.
The relative absence of expltcit references to others' work, however, and the incoris~stencyof
their conclusions, suggests that this problem is not too serious.
18. Gilpin ( 198 1 : 6 2 ) argues that the 14th century marked a resurgence o f otfensive cap-
a b i l ~ t ~ because
cs of the invention of gunpowder and artillery, hut most would r e g d his view
as premature by a century.
19. One exception here may be Montross (1960: 327), who implies that by 1670 there
may have heen 'advantages t o be gained by striking first'.
20. O n e exception is Quester (1977: ch. 11-12), who seems to suggest that the balmce
may have favored the defense.
21. There is considerable evidence contradicting the assertion that the entire German
General Staff recognized the superiority of the offens~ve(Howard, 1976: 13 1-33; Questcr,
1977: ch. 12).
22. The year 1495 marks the origins of the modern great power system (Lxvy. 198.33: ch. 2 ) .
The period before 1495 is excluded because the basic criter~onof a European sovereign state
system is not satisfied, so that earlier warfare cannot easily he compared with modern war
behavior.
2.1. For this analysis the period 1850-1890 is liberally credited to the detense and
1790-1815 is cred~tedto the offense, rather than being labelled uncertain. The 1495-1525
period is classified as offensive, and the 1650-1740 period is classified as defensive. The periods
1525-16.50, 1740-1850, and 1890-1945 have been classified 21s uncertain.
24. More technically, the offensiveldefensive balance is a multidimensional concept,
but theories should be based o n unidimensional concepts if at all possible (Shively, 1974:
ch. 3 ) .
References
Alexandroff, A. and R. Rosecrance (1977) 'Deterrence in 1939.' World Politics 29: 404-424.
Andreski, S. (1968) Military Organization and Society. London: Routledge and Kegan Paul.
Bean, R. (1973) 'War and the Birth of the Nation State.' lorrrnal of Economic History
33: 203-221.
Boggs, M.W. (1941) 'Attempts to Define and Limit "Aggressive" Armament in Diplomacy and
Strategy.' University of Missouri Studies, XVI, No. 1. Columbia, Missouri.
Brodie, B. and F. Brodie (1973) From Crossbow to H-Bomb. Bloomington: Indiana University
Press.
Bueno de Mesquita, B. (1981) The War Trap. New Haven: Yale University Press.
Bueno de Mesquita, B. and J.D. Singer (1973) 'Alliances, Capabilities and War: A Review and
Synthesis in C. Cotter (ed.) Political Science Annual, vol. 4. Indianapolis: Bobbs-Merrill.
Clausewitz, C. (1976) O n War. J.J. Graham (Trans.), A. Rapoport (ed.). Baltimore: Penguin.
Dupuy, R.E. and G.F. Eliot (1937) I f War Comes. New York: Macmillan.
Dupuy, R.E. and T.N. Dupuy (1977) The Encyclopedia of Mzlitary History, rev. ed. NY:
Harper and Row.
Dupuy, T.N. (1980) The Evolution of Weapons and Warfare. Bloomington: Indiana University
Press.
Ellsberg, D. (1960) The Crude Analysis of Strategic Choices. Rand P-2183. Santa Monica,
Rand Corp.
Farrar, L.L., Jr. (1973) The Short- War illusion. Santa Barbara, ABC-Clio.
Fay, S.B. (1928) The Origins of the World War. 2 vols. New York: Free Press.
Fuller, J.F.C. (1945) Armament and History. New York: Charles Scribner's Sons.
Fuller, J.F.C. (1961) The Conduct of War, 1789-1961. New Brunswick: Rutgers University Press.
Gibson, I.M. (1969) 'Maginot and Liddell Hart: The Doctrine of Defense,' in E. M. Earle (ed.),
Makers of Modern Strategy, ch. 15. New York: Atheneum.
Gilpin, R. (1981) War and Change in World Politics. Cambridge: Cambr~dgeUniversity Press.
Guerlac, H. (1969) 'Vauban: The Impact of Science on War,' in E.M. Earle (ed.), Makers of
Modem Strategy. New York: Antheneum.
Hale, J.R. (1957) 'International Relations in the West: Diplomacy and War,' in The New
Cambridge Modem History, vol. I, Cambridge: Cambridge University Press.
Hart, B.H.L. (1932) 'Aggression and the Problem of Weapons.' Englzsh Reliew 55: 71-78.
Hart, B.H.L. (1964) Strategy. New York: Praeger.
Howard, M. (1976) War in European History. New York: Oxford University Press.
Howard, M. (1983) Clausewitz. Oxford: Oxford University Press.
Jervis, R. (1978) 'Cooperation Under the Security Dilemma.' World Politics 30: 167-214.
Langer, W.L. (1964) European Alliances and Alignments, 1871-1890, 2nd ed., New York:
Vintage Books.
Levy, J.S. (1981) 'Alliance Formation and War Behavior.' Journal of Conflrct Resoltrtron 25:
581-614.
Levy, J.S. (1983a) War in the Modem Great Power System, 149T-197.T. Lexington: University
Press of Kentucky.
Levy, J.S. (1983b) 'Organizational Routines and the Causes of War,' Austin: The University of
Texas, mimeo.
Machiavelli, N. (1979) The Discourses, in P. Bonda and M. Musa (eds and translators). The
Portable Machiavelli. New York: Penguin.
Mahan, A.T. (1957) The Influences of Sea Power Upon Hlstory, 1660-1 783. New York: Hill
and Wang.
McNeill, W.H. (1982) The Pursuit o f Power. Chicago: University of Chicago Press.
Millis, W. (1956) Arms and Men. New York: New American LibraryIMentor.
Montgomery of Alamein, V. (1983) A History of Warfare. New York: William Morrow.
Montross, L. (1960) War Through the Ages, 3rd ed., rev. New York: Harper and Row.
Nef, J. (1963) War and Human Progress. Cambridge: Harvard University Press.
Ir T h e OffensiveIDefensive B a l a n c e o f Military Technology 385
Oman. C.W.C. (195.3) T!J~Art of War in the Middle Ages, J.H. Beeler (Rev. ed.) Ithaia, NY:
COI-nellUniversity I'ress.
Osgood, K.E. (1967) 'The Expansion of Force,' In R.E. Osgood and R.W. Tucker (eds.) korcc,,
O r d e ~ ;c~nd/usticr.Balt~more:Johns Hopkins liniversity Press.
Possony, S.T. and E. Mantoux (1969) 'Du Picq and Foch: The French School,' In t.h'l. Earlc
(ed.) Makers of Modent Strategy, ch. 9. New York: Atheneum.
Preston, R.A. and S.E Wise ( 1 9 7 9 ) Men in Arnts. New York: Holt, Rinehart, and Wtn5ton.
Quester, <;.H. ( 1977) Offense a n d Defense in the lntcrnationLzlSystrm. New York: Wiley.
I<ecord, 1. (1973) 'Armored Advance Rates: An Historical Inquiry.' M111tary Ke~ve~c, i3:
63-72,
Kopp, T. (1962) W L Tin~ the Modeni World. London: Collier-Maimillan.
Schellmg, T.C. (1960) The Strategy of Conflict. New York: Oxford Un~versityPress.
Schell~ng.T.C. (1966) Arnzs a n d Influence. New Haven: Yale University P r e s .
Sh~vely,W.P. ( 1 974) The Crizft of Political Research. Englewood Cl~ffs,New Jcrwy: Prenticc-
Hall.
Snow, I1.M. ( 1 983) The Nuclear Future. Alabama: University ot Alabama I'res,.
Snyder, G.H. (1961) Deterrence a n d Defense. Princeton: Pr~ncetonUn~versityPress.
Sprout, 14. and M. S p r o ~ ~(1965)
t The Ecological Perspective o n Humon Affirm. I'r~nceton:
Strachan, H. (1983) European Armies a n d the Conduct of War, London: George Allen 8(
Unwin.
Tars, D. ( 1 9 8 4 ) 'Defense as Strategy: A Conceptual Analysis,' in S.J. Cimhalri (ed.) N a t i o m l
Security Strategy. New York: Praeger.
T ~ ~ c h m a B.W.
n , (1962) The C ~ m of s August. New York: Dell.
Vigor, P.1-I. (1983) Sotwt Blitzkrieg T/~eory,New York: St. Martin's Press.
Vyvyan, 1.M.K. (1968) 'The Approach o f the War of 1914' in The N ~ I LCLztn1~rrcigc
, Modcnt
H~story,Vol. XII, 2nd ed., pp. 140-70. New York: Cambridge Univers~tyPress.
Wagner, R.H. ( 1983) 'The Theory of Games and the Problem o t International CooperL~t~orl.'
Anrerican Politicnl Srwnce Reurew 77: 330-346.
Wallace, M. (1979) 'The Role of Arms Races in the E ~ c ~ ~ l a t of
i o ~lSp~teS.']01f~t1il/
n o/(:onf/i~t
R ( d u t i o n 23: 3-1 6.
Wheeler-Bennett, J.W. (19.35) The Pipe Dream of Pecicc. New York: W ~ l l ~ , i m Morrow.
Wright, (2. (1949) 'Modern Technology and the World Order,' in W.t. Ogburn (ed.) Tc.cliizolog~
a n d Internatiorzd Relations. Chicago: University o f Chicago I'ress.
Wright, (2. ( 1965) A Stiidy of War, Rev. ed. Chicago: Chicago University I'ress.
Why Even Good Defenses may be Bad
Charles L. Glaser
nce again, the United States is in the midst of a debate over whether
t o deploy defenses designed to protect U.S. cities and population
from Soviet missile attack. This debate is, most immediately, the result
of President Reagan's "star wars" speech, in which he asked the rhetorical
question: "wouldn't it be better to save lives than to avenge them?" He offered
a future vision of "truly lasting stability" based upon the "ability to counter
the awesome Soviet missile threat with measures that are defensive."' Just six
months later a senior interagency group recommended to the President that
the "U.S. embark on early demonstrations of credible ballistic missile defense
technologies to its allies and the Soviet U n i ~ n . " ~
There is, in addition to this most recent catalyst, a deep-seated, endur-
ing reason why the possibility of defending the United States from Soviet
nuclear attack is a recurrent issue. Put most simply, it is quite natural for
the United States to want t o remove itself from a situation in which the
Soviet Union has the capability t o virtually destroy it. The United States
cannot, today, physically prevent the Soviet Union from wreaking such
destruction. U.S. security therefore depends upon its ability to deter Soviet
nuclear attack. If deterrence works, then the United States will be able to
avoid nuclear war with the Soviet Union. Unfortunately, the possibility that
deterrence could fail cannot be easily dismissed. Deterrence will have to
work for decades and centuries - that is, unless the current situation, in
which the United States is vulnerable to Soviet nuclear attack, is dramat-
ically altered. While one cannot specify with confidence the way in which
the superpowers' nuclear arsenals might come to be used, knowing that
deterrence could fail in a variety of ways is sufficient to create a feeling that,
given enough time, deterrence will fail. Consequently, as long as the United
States remains vulnerable to Soviet nuclear attack, the possibility of nuclear
attack will create an interest in defense against it.
The current debate over the deployment of ballistic missile defense
(BMD), like the one in the late 1960s, is highly polarized. Defense, accord-
ing to its opponents, is undesirable on all scores. They argue that defense
Source: International Security,
iJl,iic,c Why Even Good Defenses may be Bad 387
will not work effectively, will increase the probability of war, and will cause
arms races. Proponents, on the other hand, see few, if any, disadvantages
with defense. They argue that defense will reduce the damage the Soviet
Union could inflict on the United States, will not increase the probability of
war and might decrease it, and might even improve the prospects for achiev-
ing arms control agreements which limit offensive nuclear forces.'
T h e vast m a j o r i t y of t h e d e b a t e h a s pivoted on t h e technological feasibility
of effective BMD. The implicit assumption is that if effective BMD could be
developed and deployed, then the United States should pursue the BMD route
and the associated change in its nuclear s t r a t e g ~ .The
~ principal argument
against defenses is that they will not work. Opponents of defense, presumably
because they believe that effective defense is infeasible, tend not to examine
carefully either the advantages or the disadvantages of effective defense. As a
result, examination of a world in which the superpowers have deployed effect-
ive defense has been left to the advocates of defense, and a question of fun-
damental importance continues to be overlooked by the debate:
Could the deployment of effective defenses by both superpowers create a

nuclear situation preferable to our current one, in which both countries
maintain redundant assured destruction capabilities?
I am using the term "defense" to refer only to area defense, i.e., systems
designed to protect cities and other value targets. BMD that would protect
the United States by reducing the Soviet Union's ability to inflict damage is
an area defense. By contrast, a point defense is designed principally to pro-
tect nuclear force capabilities.'
By "effective defenses," 1 have in mind systems that are capable of deny-
ing one's adversary an assured destruction capability. Defenses which can-
not eliminate assured destruction capabilities are far less interesting because
they would not s~gnificantlyreduce the damage the United States would suf-
fer in an all-out nuclear war.h Another way, then, of stating the above ques-
tion is: Could the United States be more secure than it is today if, as a result
of mutual deployment of defenses, neither the United States nor the Soviet
Union had assured destruction capabilities?
My objective in this essay is to analyze this question.

A country with an assured destruction capability can inflict extremely
high levels of damage. Nuclear situations in which the Soviet Union lacks
an assured destruction capability, therefore, range from those in which the
United States is invulnerable to attack to those in which the Soviet Union
could destroy a sizable fraction of the U.S. population. The most interest-
ing alternatives to mutual assured destruction situations are, of course,
those in which the U.S. defense reduces the potential damage the Soviet
Union could inflict far below the level required by assured destruction.
Analytically, however, there are important similarities that cross the full range
of nuclear situations in which assured destruction capabilities d o not exist.

As a result, this analysis applies equally well to all such situations. In fact,
a distinguishing feature of this analysis is that it examines the requirements
of strategic nuclear deterrence in situations in which defenses have eliminated
assured destruction capabilities. In contrast, most analyses of strategic nuclear
deterrence require that the United States possess an assured destruction
capability, almost as if this were a prerequisite for deterrence.
There is general agreement that defenses capable of eliminating assured
destruction capabilities d o not exist today, and are extremely unlikely to be
developed in the foreseeable future.' However, to facilitate examination of
the issues that lie beyond the technical feasibility of BMD, I hypothesize in
this article that highly effective defenses are available. Although effective
defenses are, at best, a distant prospect, their presumed advantages have a
significant influence on the BMD debate. Assuming, for the sake of analy-
sis, that effective defenses are available makes possible a closer examination
of the desirability of defensive situations. (I will use the term "defensive situ-
ation" to refer to nuclear situations in which defenses have eliminated assured
destruction capabilities.)
The article focuses on situations in which both the United States and the
Soviet Union deploy defenses. This case is important because it is the most
probable outcome of U.S. deployment of defense. The Soviet Union is ex-
tremely likely to deploy defenses in response to a U.S. deployment. There is
little reason t o assume that in the long run the United States could maintain
a technological advantage that enabled only the United States to have effective
defense. Furthermore, the case of symmetric deployment is especially interest-
ing due to the intuitive appeal of reducing U.S. vulnerability to attack without
creating an advantage that threatens Soviet s e ~ u r i t y . ~
This analysis of how mutual deployment of effective defense would affect
U.S. security proceeds through a number of stages. I identify three features
of the nuclear situation that affect the United States' ability t o avoid nuclear
war with the Soviet Union: 1) the United States' ability to deter premedi-
tated Soviet attack; 2) the crisis stability of the nuclear situation; and 3 ) the
robustness of the U.S. deterrent to changes in Soviet forces. Next, I compare
the probability of nuclear war in defensive and assured destruction situations
by examining these three features for both types of nuclear situations. The
final stage of the analysis compares U.S. security in defensive and assured
destruction situations based upon expected costs. This requires considering
the damage that would result if nuclear war occurred as well as the prob-
ability of its occurrence.
The conclusion of this analysis is that defensive situations, even those in
which defenses were perfect, are not clearly preferable to assured destruction
situations. This conclusion is indeterminate because in defensive situations
the probability of certain types of wars would increase, but the damage of
other types of wars would decrease.
The indeterminacy of this conclusion should not obscure its policy
significance. The assumptions used in the analysis have made a best case
Why Even Good D e f e n s e s may b e Bad 389
for defense: effective defense is assumed to be technically achievable; the

enormous econon~iccosts required to deploy any effective defense are over-
looked; and the deployment of defenses is assumed to avoid the creation of
asymmetries in the superpowers' capabilities that could create incentives for
preventive attack and could encourage adventurous, crisis-provoking hehav-
ior. Even in this best case, defensive situations are not clearly preferable to
the current assured destruction situation. In addition, many o f the nuclear
situations that could result from starting down the BMD route are far less
desirable than our current mutual assured destruction situation. Without
the possibility of a best outcome that is clearly preferable to our current sit-
uation, there is now no good reason to invest enormous resources in strategic
defense and to risk creating a more dangerous world. The arguments for not
dramatically altering the nuclear status quo are much stronger than those
that call for U.S. deployment of an area defense.
Perfect Defense
It is important to begin with an examination of the strategic implications of

perfect defenses, however distant they may seem, because that is the goal
towards which many advocates of strategic defense, including President
Keagan, wish to move. Despite the widespread presumption that perfect
defenses are desirable if feasible, there are two major shortcomings of a
world of perfect defenses that draw into question whether it would be safer
than our current nuclear situation.'
First, there could he no guarantee that perfect defenses would remain
perfect. The technical challenge of developing and deploying a defense that
would make the U.S. invulnerable to nuclear attack is enormous. Such a
defense is commonly referred to as "perfect." The difficulty of maintaining
a perfect defense indefinitely is likely to be far greater than developing it in
the first place. Consequently, so-called perfect defenses should not he envis-
ioned as a permanent technological solution to the dangers posed by nuclear
weapons. The far more likely course of events is that a world of perfect
defenses would decay into a world of imperfect defenses."'
A nuclear situation in which both superpowers were invulnerable to
nuclear attack would be extremely sensitive to even small improvements in
the ability of one country's offense to penetrate the adversary's defense. For
example, the ability to penetrate the adversary's defense with ten warheads
would provide the potential for enormous destruction when compared to
no destruction. The country that first acquired even a small capability to
penetrate the adversary's defense would have attained an important coer-
cive advantage: nuclear attack could be threatened with impunity since
effective retaliation would be impossible given the adversary's inability to
penetrate one's own defense. Recognizing that the adversary is likely to
acquire a similar capability - that is, that one's defense will not remain
impenetrable - could create pressure to reap the benefits of the strategic
advantage quickly. This time pressure would be especially strong if one's

advantage could be used to prevent the adversary from acquiring the capa-
bility to penetrate one's defense.
By contrast, when both superpowers possess redundant assured destruc-
tion capabilities, as is the situation today, the addition of tens or hundreds
or even thousands of warheads would not significantly change the nuclear
situation. As a result, the probability of gaining a strategic advantage is
extremely low, especially when both superpowers are aware of and react to
changes in the other's nuclear force.
The dangers, in a world of impenetrable defenses, that result from this
sensitivity to small offensive improvements would be increased by the strong
incentives the superpowers would have to defeat each other's defense. Each
country could be expected to make the acquisition of a strategic advantage
a priority. Moreover, because there would be no guarantee that perfect
defenses would remain perfect, even a country that did not want to acquire
an advantage would feel compelled to acquire additional strategic capabil-
ities. Such a country would want to improve its defense to offset anticipated
improvements in the adversary's offense. In addition, there would probably
be a strong instinct to improve one's offense as well as a hedge against the
possibility of not being able to offset, with improvements in one's defense,
the adversary's enhanced offense. One's adversary, however, would not be
able to know with confidence that these strategic programs were intended
only to maintain a situation of equal capability. Consequently, even if both
countries preferred to remain in a world of perfect defense, an interactive
competition which threatened to reduce the effectiveness of the defenses
would be likely to ensue. (Nuclear situations would continue to be sensitive
to relatively small changes when the defenses were imperfect. This lack of
"robustness" to changes is examined in detail below.)
The second problem with perfect defenses is that they could increase the
probability of superpower conventional wars. Today's nuclear forces greatly
increase the ~ o t e n t i a costs
l of any direct U.S.-Soviet military confrontation.
As a result, nuclear weapons increase the risk of starting a conventional war,
and therefore contribute to the deterrence of conventional war. Impenetrable
defenses would eliminate this contribution. There is disagreement among
strategic analysts about which features of the superpowers' extensive surviv-
able strategic arsenals are most critical for deterrence of conventional war. Few,
if any, commentators however believe that the existing arsenals do not con-
tribute at all to the deterrence of conventional war."
Perfect defenses might be in the U.S. security interest despite the increased
probability of conventional war. That conventional war would be more likely
does mean, however, that there is an important trade-off to consider. As World
Wars I and I1 demonstrated, global conventional wars can be extremely
destructive. The net effect of increasing the probability of major conventional
war, while eliminating the possibility of more destructive but extremely
unlikely nuclear war, might not be positive. The evaluation of this trade-off
would involve many factors, including estimates of the probability of nuclear
Cxl,~sci Why Even Good Defenses may be Bad 391
and conventional wars with and without perfect defenses, estimates of the size
and costs of these wars, and the availability of options for reducing the prob-
ability and costs of conventional war. The objective of this short discussion
is to call attention to this trade-off, not to resolve it.
In short, then, what are commonly called perfect defenses would have
two shortcomings. First, they would probably not be truly perfect, but
instead only temporarily impenetrable. The undermining of one country's
defense would create a situation in which the incentives to initiate a nuclear
war would be greater than today. Second, even if the temporary nature of
impenetrable defenses is ignored, the net effect of both superpowers' deploy-
ing impenetrable defenses remains unclear because major conventional wars
could become more likely.
Imperfect Defense and the Probability of Nuclear War
Understanding security in a world of perfect defense is relatively easy

because as long as the defenses remain impenetrable, there is no possibility
of a strategic nuclear war.I2 Assessing security in a nuclear situation in which
imperfect defenses have been deployed is more difficult. Since, in this case,
the United States would be vulnerable to Soviet strategic nuclear attack, we
need to evaluate the United States' ability to reduce the probability of these
attacks.
The following analysis considers nuclear situations in which both coun-
tries have imperfect defenses, but each is capable of denying the other an
assured destruction capability. Implicit in this formulation is a relationship
between one country's offensive force and the adversary's defensive force.
When defenses are imperfect there will always be, at least in theory, an
offense which is sufficiently large to have an assured destruction capability.
Therefore, for one country's imperfect defense to deny the adversary an
assured destruction capability, either the size of the adversary's offense must
be limited or the defense must be able to expand and improve to offset
increases in the size of the offense. This analysis does not examine the feasi-
bility of achieving these conditions. It assumes the establishment of a nuclear
situation in which neither the United States nor the Soviet Union has assured
destruction capabilities.
The probability that the United States will avoid war with the Soviet
Union depends upon the following three features of the nuclear situation:
( 1 ) T h e United States' ability t o deter Soviet nuclear attack during peri-
ods' w h e n war does not appear t o he imminent, that is, w h e n there is n o t a
severe crisis. Deterrence of this type of attack requires that the Soviet Union
believe that the net effect of starting a nuclear war would be negative, that is,
that the Soviet Union would be worse off after the war than before it. I will
term these "premeditated attacks." Surprise attacks, including the infamous
"bolt from the blue," fall within this category.
( 2 ) The crisis stability o f the nuclear situation. In a crisis, one or both

superpowers might fear a nuclear attack by the other. If striking first is
believed to be preferable to being struck first, and if a country believes the
probability that the adversary will strike first is sufficiently high, then
launching a first strike would be preferable to taking a chance on being
struck first. This type of first strike is commonly termed a "preemptive
attack." Unlike the case of premeditated attack, the country launching a
preemptive attack would expect to be less well off after the war than before
it. The crisis stability of the nuclear situation is a measure of how severe a
crisis must be (or how high one's estimate that the adversary will strike first
must be) before striking first becomes one's best option.I3
( 3 )The robustness of the nuclear situation. The adequacy of U.S. forces
depends not only on their ability to reduce the probability of preemptive
and pre-meditated attacks, but also on how sensitive this ability is to poten-
tial changes in the Soviet forces. The more easily the Soviet Union could
build forces that either would make a premeditated attack attractive or
would significantly increase the incentives for preemptive attack, the greater
the probability of a nuclear war. The robustness of the U.S. nuclear force is
a measure of the difficulty the Soviet Union would encounter in trying to
reduce U.S. security.
These three measures of the quality of the nuclear situation (the United
States' ability to deter premeditated attacks, the degree of crisis stability,
and the robustness of U.S. forces to change) are frequently used to assess
the adequacy of U.S. nuclear forces. What distinguishes the following analysis
from standard analyses of the nuclear situation is the assumption that
assured destruction capabilities d o not exist. Past analyses have asked the
question: what capabilities are required to minimize the probability of war?
The answers all include the need for an assured destruction capability (or at
least a large retaliatory capability). This analysis, by examining the effect on
these three measures of the nuclear situation, explores how the elimination
of assured destruction capabilities by mutual deployment of defenses would
affect the probability of nuclear war.
Premeditated Attacks: Is Assured Destruction Necessary

for Deterrence?
Consider a nuclear situation in which Soviet defenses could deny the United
States an assured destruction capability. In this situation, the most basic and
generally accepted U.S. deterrent requirement (that is, possession of an assured
destruction capability) would not be satisfied. A natural conclusion is that the
U.S. deterrent would be inadequate. This belief fueled opposition to strategic
defense during the earlier BMD debate.I4 But closer examination of nuclear
situations in which both superpowers deploy defenses shows that U.S. deter-
rent requirements could be satisfied without U.S. possession of an assured
destruction capability.
Why Even Good Defenses may b e Bad 393
The requirement that the United States have an assured destruction cap-
ability implicitly assumes that the Soviet Union can annihilate the United
States: the standard argument is that t o deter a n annihilating attack, the
United States should be able t o threaten credibly to annihilate the Soviet
Union in retaliation. But if the United States could, by deploying defenses,
eliminate the Soviet Union's annihilation capability, then deterrence of this
attack would not be necessary. Furthermore, it is difficult to imagine any other
Soviet actions the deterrence of which requires the United States to threaten
the annihilation of the Soviet Union. So, if the Soviet Union did not have
the ability t o annihilate the United States, then the United States would not
need t o be able t o annihilate the Soviet Union in retaliation. Consequently,
a mutual deployment of defenses that eliminated both U.S. and Soviet annihi-
lation capabilities need not result in a n inadequate U.S. deterrent. The United
States would, of course, still need a nuclear retaliatory capability to deter
other Soviet nuclear attacks.
What capability would the United States need to deter attacks against its
homeland when defenses had denied the Soviet Union an annihilation
capability? Deterrence requires that the United States have the ability fol-
lowing any Soviet attack t o inflict costs greater than the benefits the Soviet
Union would achieve by attacking. To determine the U.S. retaliatory require-
ment, we must estimate the value the Soviet leaders would place on attacking
the United States. We need to consider why the Soviet Union might attack the
United States and what it would hope to gain by doing so. In the most general
terms, the Soviet Union could use its nuclear force to damage or weaken the
United States and to coerce the United States. The U.S. forces required to deter
these actions are examined briefly below.
For all of the concern about attacks against U.S. cities, it is not clear why
the Soviet Union would ever launch a n all-out countervalue attack. Still
such an attack is not impossible, s o we need t o estimate the value the Soviet
Union might place on attacking U.S. cities. O n e possible reason for attack-
ing U.S. cities would be to weaken the United States, thereby reducing the
U.S. ability t o oppose the Soviet Union's pursuit of its foreign policy object-
ives. Presumably people believe the Soviet Union is interested in annihilat-
ing the United States because this would make it the dominant world power.
The analogy, if U.S. defenses had eliminated the Soviet ability to annihilate
the United States, would be a countervalue attack designed t o weaken the
United States.
To deter this type of attack, the United States would need a retaliatory
capability that could weaken the Soviet Union as much as the Soviet counter-
value attack could weaken the United States. A countervalue capability
roughly equivalent to the Soviet countervalue capability should be sufficiently
large to satisfy this requirement. In fact, this is a very conservative requirement
because U.S. retaliation would not only deny the Soviet Union the desired
increase in relative world power, but would also inflict direct costs by destroy-
ing Soviet value targets. Because the Soviet Union could first attack U.S. forces,
and then attack U.S. cities, the United States should have forces that provide a
countervalue capability essentially equal to the Soviets' both before and after a
Soviet counterforce attack.15 I will call this an "equal countervalue capability."
The second way in which the Soviet Union might use its nuclear capa-
bility is to coerce the United States. While the benefits to the Soviet Union
of attacking U.S. cities can be questioned, the potential benefits of coercing
the United States are far more obvious. If the Soviet Union could inflict
enormous damage on the United States and the United States lacked the
ability to deter these attacks, then the Soviet Union might be able to com-
pel the United States to compromise its security and vital interests.
As in other cases, deterrence would require that the United States be able
to threaten the Soviet Union with expected costs greater than expected bene-
fits. In the case of coercion, however, the United States could deny the
Soviet Union any benefit simply by refusing to perform the action the Soviet
Union demanded. The costs threatened by the United States need not be
greater than the benefits the Soviet Union hopes to gain through its coercive
demand because any U.S. attack combined with refusal of the Soviet demand
would result in a net Soviet loss. If faced with a coercive threat, the United
States could refuse the Soviet demand and tell the Soviet Union that attacks
against value targets would be reciprocated. To adopt this strategy, the United
States would have to be confident that it could deter the Soviet Union. This
would require that the United States believe that the Soviet Union finds the
U.S. retaliatory threats credible.
A large disparity in U.S. and Soviet countervalue capabilities could
undermine U.S. credibility. So, a reasonable force requirement for denying
the Soviet Union the coercive use of its nuclear forces is that the Soviet
Union not have an advantage in countervalue capabilities: an advantage
should not exist in the deployed forces, nor should the Soviet Union be able
to gain a countervalue advantage in surviving forces by launching a counter-
force attack. Therefore, U.S. forces which satisfy the equal countervalue
requirement should be sufficient to deny the Soviet Union a capability which
enables it t o coerce the United States.16
In summary, a reasonable requirement for deterrence of Soviet attacks
on the United States is possession of an equal countervalue capability."
Requiring that the United States possess an equal countervalue capability is
significantly different from requiring an assured destruction capability. The
equal countervalue requirement explicitly couples U.S. and Soviet capabil-
ities t o inflict countervalue damage. The equal countervalue requirement
could be satisfied by both the United States and the Soviet Union at all levels
of vulnerability to attack. In contrast, the assured destruction requirement
demands that the United States have a retaliatory force capable of inflicting
a specific level of countervalue damage independent of the size of the Soviet
ability to inflict damage. According to the equal countervalue requirement,
if the United States can reduce the Soviet Union's ability to inflict counter-
value damage, then the United States can afford to have its ability to inflict
countervalue damage in retaliation reduced. Moreover, improvements in
:81+.e~ Why Even Good Defenses may b e Bad 395
Soviet defenses which reduce the damage the United States could inflict on
the Soviet Union could be compensated for by improvements in U.S. defenses.
The assured destruction requirement, on the other hand, demands that im-
provements in Soviet defenses be offset either by an increase in the size of the
U.S. offense or by an increase in the ability of the offense to penetrate the
Soviet defense.
risk Stability: W h a t woul be t h e Effect of Defense4 "
There is a common belief that defenses capable of eliminating an adver-

sary's assured destruction capability would decrease crisis stability: a coun-
try that can protect itself (that is, a country that can deny its adversary a
second strike annihilation capability) is more likely to strike preemptively
in a c r i s i s . ' T h e following analysis explores this proposition and identifies
the conditions under which it is correct.
Crisis stability depends upon the decision-maker's incentives to strike
preemptively in a crisis, that is, during times when there is reason to believe
one's adversary is likely t o launch a first strike. The decision t o preempt in
a crisis would depend upon how the costs of being struck first compare t o
the costs of being struck second." If the adversary has an assured destruction
capability, then there would be little if any incentive for a rational decision-
maker to preempt: a preemptive attack could not deny the adversary an anni-
hilating retaliatory capability, so there would be little difference between the
costs of being struck first and second. In an assured destruction situation, the
vulnerability of the adversary's forces does not create a preemptive incentive.
The adversary's force is sufficiently large and survivable that the fraction of
the force that would survive a counterforce attack would still be able to
inflict the damage required for annihilation.
For the same reason, the adversary would have little incentive to preempt
if one's own surviving force would be sufficiently large t o annihilate the
adversary. Since a leader's decision to preempt would be fueled by anticipa-
tion of the adversary's preemption, possession of an assured destruction cap-
ability by either country should be sufficient to create a highly crisis-stable
nuclear situation.
If one's own defense eliminates the adversary's assured destruction cap-
ability and if the adversary's retaliatory capability is partially vulnerable,
then preemption would reduce the damage from a n all-out countervalue
attack. As a result, if the decision-maker anticipates a countervalue first strike,
then there would be an incentive to preempt.20 Since without defenses there
would be virtually no incentive to preempt (because the adversary could main-
tain his assured destruction capability), deploying defenses that eliminate
assured destruction capabilities would decrease crisis stability.
But there is another important case, the one in which the adversary's
retaliatory capability is invulnerable to a first strike. In this case, there is
nothing to be gained by striking first because the magnitude of the adversary's
retaliatory strike would be no less than if he had struck first - that is, the costs
to one's own country of suffering an all-out countervalue first strike or second
strike would be equal. So, because the adversary's forces were invulnerable,
there would be no incentive to preempt. This would be true when defense had
not been deployed, even if the adversary's ability to inflict countervalue dam-
age were far below the annihilation level. Deploying defenses would reduce
the adversary's ability to inflict damage, but would not create an incentive to
preempt. This example shows, at least in principle, that defenses that reduce
the adversary's retaliatory capability below the annihilation level would not
always decrease crisis stability.
The practical significance of this observation should not be overesti-
mated. An invulnerable retaliatory capability requires not only that the
forces be invulnerable, but also that attacks against the command and con-
trol system would not reduce the size of the possible retaliatory attack.
These conditions might not be achievable. Submarines in port are vulner-
able t o attack and much of the command system is now highly vulnerable.
The combination of one's effective defenses with the adversary's offensive
force vulnerabilities would result in an incentive to preempt. Consequently,
while in theory defenses that eliminated assured destruction capabilities
need not decrease crisis stability, in practice they probably would.
The fundamental insight we can draw from this discussion is that defenses
d o not by themselves create incentives for preemption. The source of pre-
emptive incentives is offensive force vulnerabilities. Therefore, the effect of
defenses on crisis stability should not be evaluated without considering the
vulnerabilities of the offensive force to a counterforce attack. By reducing
retaliatory capabilities, defenses can increase the significance of offensive
vulnerabilities.
Because reducing the degree of offensive force vulnerability would
enhance crisis stability, one way to offset the decrease in crisis stability that
would result from deploying effective defenses would be to accompany the
deployment with programs to reduce offensive force vulnerabilities. One
approach for reducing force vulnerability is to protect offensive forces with
active defenses. Area defenses, although not designed specifically for this
mission, could increase force survivability. In addition, there are many other
ways to increase force survivability, including deploying point defenses. If
effective area defense were feasible, then defenses that could provide a high
degree of force survivability, including survivability of command and con-
trol, would also be feasible. In this case, the reduction in crisis stability that
would result from deploying effective defenses could be small.
In summary, effective defenses would be likely to decrease crisis stabil-
ity. It would probably be possible, however, to keep this negative effect of
defenses quite small. The source of preemptive incentive is offensive force
vulnerability. Therefore, if offensive forces could be made highly survivable,
then the effect on crisis stability of defenses that eliminate assured destruc-
tion capabilities would be small.
I I Why Even Good Defenses may be Bad 397
We do not live in a static world. Consequently, in addition to evaluating

U.S. security as if U.S. and Soviet forces could be held constant, we must
also examine the effect of possible changes in Soviet forces on U.S. security
and the probability of these changes. More specifically, we must evaluate
not only the United States' ability to deter premeditated Soviet attack and
the degree of crisis stability, but also the probability of changes in Soviet
forces that could reduce the United States' ability to deter premeditated
attacks or that could reduce crisis stability. The robustness of U.S. forces is
a measure of the difficulty the Soviet Union would encounter in trying to
reduce U.S. security."
All other things being equal, the more easily U.S. security could be jeop-
ardized by changes in Soviet forces, the less desirable the nuclear situation.
A nuclear situation which would be highly desirable when the two coun-
tries' forces could be held fixed, but which lacks robustness, might not be
preferable to one which is less desirable when the forces are held fixed, but
which is more robust.
I have already discussed the lack of robustness of nuclear situations in
which perfect defenses have been deployed. This section extends that analy-
sis by considering cases in which imperfect defenses have been deployed.
The conclusion remains the same: nuclear situations in which defenses sig-
nificantly reduce the vulnerability of value targets would lack robustness.
The following discussion compares the difficulty the Soviet Union would
have undermining U.S. deterrence of premeditated attacks in defensive and
in assured destruction situations. It assumes that the requirement for deter-
rence of premeditated attacks, that is, the equal countervalue requirement,
is satisfied in the initial nuclear situation. The United States' deterrence of
premeditated attacks could be undermined by two types of changes: improve-
ments in Soviet defenses that reduce the United States' ability to retaliate, and
improvements in the penetration capability of Soviet offenses that increase
the vulnerability of U.S. value targets to attack.
The robustness of U.S. nuclear forces to these changes depends upon
two interdependent factors. The first is the magnitude of the change in
potential countervalue damage required so that the Soviet Union would n o
longer be deterred from launching a premeditated attack. Specifically, how
much must the Soviet Union reduce the U.S. countervalue threat to gain a
strategic advantage! or how large an increase in Soviet countervalue capa-
bility is required to provide a significant advantage? The second factor is
the technical difficulty of changing the threat to value targets by this amount.
For example, assuming that in a specific nuclear situation the Soviet Union,
to gain an advantage, must increase its countervalue capability by 50 war-
heads, how difficult would it be for the Soviet Union to achieve this change?
The combination of these two factors determines the overall difficulty of
acquiring a strategic advantage.
The discussion of perfect defenses focused on the first factor, the magnitude
of the change, and argued that even small changes could have strategic sig-
nificance. Situations in which imperfect defenses had been deployed would
suffer, although less severely, from the same sensitivity. The following ex-
ample illustrates this observation. Imagine three nuclear situations, one in
which both superpowers have impenetrable defenses, one in which each
superpower can penetrate the other's defense with ten warheads, and one in
which both superpowers have assured destruction capabilities. Now con-
sider how a change in one country's nuclear force that enabled it to pene-
trate the adversary's defense with ten additional warheads would affect the
adversary's security in each situation. The addition of ten warheads of
countervalue capability to one country's force would be less significant
when added to the nuclear situation in which both countries started with
ten penetrating warheads than when added t o a situation in which both
countries had perfect defenses. The advantage in countervalue capability
would be harder to use coercively when the adversary would be able to
threaten retaliation against one's own value targets.
In contrast, the addition of ten penetrating warheads to one force when
both countries had assured destruction capabilities comprised of thousands
of warheads would be far less significant than when added to the nuclear
situation in which both countries had ten penetrating warheads. The addi-
tion to the mutual assured destruction situation might not even change the
country's ability to inflict damage; the addition in the ten warhead situ-
ation, while it might be difficult to use coercively, could result in a significant
difference in the two countries' ability to inflict damage.
The general conclusion t o be drawn from this specific example is that
the lower the vulnerability of value targets in a given nuclear situation, the
smaller the change in their vulnerability required t o gain an advantage. This
conclusion can be restated specifically in terms of defenses: the smaller the
number of warheads that could penetrate a country's defense, the more sen-
sitive that country's security would be to offensive changes that reduce the
effectiveness of its defense.
The second factor affecting robustness, the technical difficulty of chang-
ing countervalue capability to gain an advantage, depends upon the type,
size, and number of changes required to achieve a strategic advantage. The
type of change is determined by whether the status quo is an assured
destruction situation or a defensive situation. In assured destruction situ-
ations, it is the difficulty of reducing the adversary's offensive threat that
affects robustness. In defensive situations, on the other hand, both the dif-
ficulty of further reducing the adversary's offensive threat and the difficulty
of penetrating the adversary's defense would affect robustness.
Assessing the relative difficulty of penetrating a specific defensive sys-
tem with an offensive system or of defeating a specific offensive system with
a defensive system is beyond the scope of this paper. Moreover, such an
assessment would necessarily be highly speculative because effective defen-
sive systems have not yet been developed. Consequently, it is impossible to
( ILii r Why Even Good D e f e n s e s may b e Bad 399
compare the difficulty of defeating those defensive systems with the difficulty
of developing defensive systems to defeat today's offenses or the offenses of the
future. One fact that bears upon this issue should, however, be mentioned.
Even if defenses were developed that were perfect against currently deployed
offenses, experts believe that the task of developing offensive countermeasures
to defeat those defenses would be relatively easy." The defensive system \vould
be understood by its adversary, enabling the development of countermeasures
designed specifically with the defense in mind. The defense, by contrast, to
remain effective, would have to be able to overcome the full range of possible
countermeasures. This asymmetry means that defenses may always be at a dis-
advantage, that is, the development of effective defenses against a competitive
threat may always be more difficult than developing offenses that can pene-
trate defenses.
The size o f the change required to gain a strategic advantage affects the
technical difficulty of achieving the change. (This is why the two factors
affecting robustness are interdependent.) A defense which must reduce the
offensive threat by a large amount is harder to build than one that must
reduce the same offensive threat by a small amount. Similarly, a new offen-
sive system which must be able to penetrate the adversary's defense with
many weapons would be harder to build than one that had to penetrate the
same defense with only a few weapons. Even taking into account the likely
asymmetry hetween offense and defense mentioned above, it is not possible
to say with certainty whether the changes required to gain an advantage in
an assured destruction situation would be easier or harder to achieve than
in a defensive situation. As discussed above, however, the size of the requis-
ite change in assured destruction situations is larger than in defensive sitii-
ations. Due to this difference, gaining an advantage will tend to be more
difficult in assured destruction situations than in defensive situations.
The larger the number of changes in a country's forces required to gain
an advantage, all else being equal, the harder the advantage will be to
obtain. The number of force changes required to achieve an advantage
depends upon the diversity of the adversary's forces. In assured destruction
situations, ensuring one's ability to destroy large numbers of the adversary's
value targets is the strategic requirement. Diversification of one's offensive
force helps to ensure the continuing achievement of this objective by
increasing the number of defensive changes that are required before the
adversary could eliminate one's assured destruction capability. For example,
an offensive force which could annihilate the adversary with either an air-
breathing threat or a ballistic missile threat requires that the adversary
develop two types of highly effective defense. Obviously, this is a harder
task than developing an effective defense against a single threat.
This article has discussed defenses in general, not defenses against spe-
cific types of offensive threats. But when we think about the feasibility of
defense, it is crucial to keep in mind the potential diversity of offensive
threats. If BMD were technologically feasible, but defense against advanced
technology bombers or cruise missiles were impossible, then the strategic
significance of the BMD would be greatly reduced. The technological feasi-

bility of defenses that would reduce vulnerability t o attack is determined by
the difficulty of defending against all offensive threats.
By contrast, in a nuclear situation in which one's own defenses have sig-
nificantly reduced the vulnerability of value targets, the strategic require-
ment is the maintenance of a low level of vulnerability. In this case the
adversary's ability to diversify offensive forces makes maintaining low vul-
nerability more difficult. Each of these offensive threats must be defended
against, and the adversary's ability to defeat any of the defenses would be
sufficient to make maintenance of low vulnerability impossible.
To make defensive situations more resistant t o the adversary's offensive
improvements, a country could diversify its defenses against each offensive
threat. It would be more difficult, however, to diversify one's defenses to
require the same number of force changes as in an assured destruction situ-
ation. Because the number of required offensive changes would be deter-
mined by the offense against which the defense was the least diversified, it
would be necessary t o diversify one's defenses against each of the adver-
sary's offenses. So, for example, if the adversary had two offenses with dif-
ferent penetration modes, then to force him to have t o make two changes
would require a total of four defenses. By comparison, in an assured
destruction situation, two offenses are sufficient to require the adversary to
make two changes, that is, to deploy two effective defenses. Consequently,
the relative difficulty of diversification would tend to make it more difficult
to make a defensive situation resistant to the adversary's offensive changes
than to make an assured destruction situation resistant to the adversary's
defensive changes. For what could be termed "structural" reasons, defen-
sive situations could not be made as resistant to change as assured destruc-
tion situations. In addition, a defensive situation which is not diversified is
susceptible to catastrophic failure, because one offensive breakthrough by
the adversary could render one's country vulnerable to large attacks.
To summarize, nuclear situations in which defense significantly reduces
the vulnerability of value targets would lack robustness for two reasons: rela-
tively small changes in vulnerability could threaten one's security; and the
difficulty of achikving these changes would be relatively low due t o the
small change- requiredand to the greater
- difficulty of making- defensive situ-
-
ations resistant to offensive changes.
A lack of robustness would not be so dangerous if the United States and
the Soviet Union would not have incentives to try to change the equal
countervalue condition of the nuclear situation. If a political environment
could be created in which the superpowers chose not to attempt to gain a
strategic advantage, then the need t o make the nuclear situation resistant to
change would be reduced. Moreover, superpower cooperation in structur-
ing the nuclear situation could contribute to the situation's robustness.''
But, in a world of imperfect defenses, as described for the case of perfect
defense, countries would feel tremendous pressure to pursue, or at least to
Why Even Good Defenses may b e Bad 401
prepare to pursue, capabilities for defeating the adversary's defense. Even in

the unlikely event that a highly robust situation could be achieved (which
would require making one's defense highly resistant to the adversary's
offensive innovations), it would be hard to have high confidence in this
robustness: a country could not know with certainty that offensive threats
that would undermine the defense could not be developed. This uncertainty
could not be overlooked because the change required to gain an advantage
would still be small and the adversary's incentive to try to alter the nuclear
situation would be obvious.
These conditions would make establishing a political environment in
which cooperation was possible far harder under conditions of reduced vul-
nerability than under assured destruction.'%nd, given our limited success
in negotiating strategic arms control treaties when both countries have
redundant assured destruction capabilities, there is little reason to be opti-
mistic about the prospects for cooperation.
In conclusion, situations in which defenses have eliminated countries'
large retaliatory capabilities would suffer a lack of robustness. The danger
of this condition would be increased by the incentives and pressures that
both the United States and the Soviet Union would feel not to cooperate and
to increase the other's vulnerability.
Could D e f e n s e Create a Preferable Nuclear Situation?
The preceding discussion of imperfect defenses capable of eliminating both

superpowers' assured destruction capabilities compared the probability of
nuclear war in defensive and assured destruction situations. Specifically, it
compared the United States' ability to deter Soviet p-emeditated attack, the
crisis stability, and the robustness of defensive and assured destruction situ-
ations. This analysis of the probability of nuclear war is, however, not by
itself sufficient to determine in which type of nuclear situation the United
States would be more secure. This is because U.S. security depends upon the
cost if war were to occur, as well as the probability of war.
Comparison of nuclear situations requires a measure that combines
these U.S. security objectives, that is, to minimize the probability of war and
to minimize the costs if war occurs." These objectives should be evaluated
simultaneously. For example, if a change in the nuclear situation would
reduce the damage of a nuclear attack but would also increase the probabil-
ity of the attack, then the change might not increase U.S. security. Examining
only the probability of war or the costs if war were to occur is insufficient
to understand the net effect of the change. The correct measure of security
is the probability of the war multiplied by the costs if there were a war,
which is the expected cost.
It is crucial to keep these two aspects of U.S. security in mind when ana-
lyzing defenses. Much of the debate over BMD tends to ignore the need for
simultaneous evaluation. Proponents of defenses emphasize the reductions

in damage that defenses could provide. Opponents of BMD argued during
the 1960s that deployment would increase the probability of nuclear war
by undermining deterrence and decreasing crisis stability. Neither of these
arguments is sufficient to draw a conclusion about area defenses: each looks
at only one aspect of U.S. security.
A complete analysis of the expected costs in a specific nuclear situation
would describe the full spectrum of wars in which the United States might
become involved, would evaluate the probability that they would occur,
and would estimate the damage that would result if they did occur. This
spectrum of wars would include conventional wars and limited nuclear
wars in addition to all-out premeditated and preemptive wars. The preced-
ing sections of this article d o not focus on conventional and limited nuclear
wars.26 However, the analysis is sufficient to draw the fundamental conclu-
sion about nuclear situations in which effective defenses have been deployed:
the probability of nuclear war would be higher in these situations than
assured destruction situations, but the costs of certain types of nuclear wars
would be lower. Therefore deciding whether these defensive situations are
preferable to assured destruction situations requires making a trade-off.
This trade-off is highly subjective. Analysts are likely to disagree about the
level a t which reducing the size of the adversary's potential attack begins to
make a significant difference, about the relative probability of different
types of nuclear war, and about how to balance a believed increase in the
probability of war against a decrease in its potential costs. Because individ-
uals will make these judgments differently, defensive situations cannot be
said, in general, to be inferior or superior to assured destruction situations.
Before clarifying how the preceding analysis leads to this conclusion, I
want t o pause for a moment to stress its significance. Saying that defenses
are neither clearly desirable nor clearly undesirable might appear to provide
little policy insight. However, although indeterminate, this conclusion differs
markedly from the conventional wisdom that effective defenses would be
desirable, and weighs heavily against deploying defense to limit U.S. vulner-
ability to attack. Recall that the analysis has considered a best case for
defense. Even making these optimistic assumptions, a defensive situation
might not be preferable to our current assured destruction situation. More
realistic, less optimistic assumptions about defenses result in nuclear situ-
ations which would be more dangerous than today's. Advocates of strategic
defense are driven by the promise of a world far safer than the current one.
If defenses could create such a world, then taking a chance on ending up in
one of the more dangerous, and more likely, defensive situations might be
justified. But to run great risks and to spend enormous resources in the hope
of reaching a nuclear situation that might not be preferable to our current
assured destruction situation, and might be worse, make little sense.
The remainder of this section explains how this conclusion follows from
the preceding analysis. The analysis first identified three factors that influence
the probability that the U.S. will be able to avoid nuclear war with the Soviet
# It Why Even Good Defenses may b e Bad 403
Union: 1) the U.S. ability t o deter premeditated attacks; 2) the crisis stability
of the nuclear situation; and 3) the robustness of the nuclear situation. I then
evaluated how the deployment by both superpowers of effective defenses,
that is, defenses sufficiently capable to deny the adversary an assured destruc-
tion capability, would affect these factors. The findings of this evaluation are:
(1) Effective defenses need not undermine deterrence of premeditated attacks.

Assured destruction is not necessary for deterrence; an equal counter-
value capability (i.e., the possession of a countervalue capability equal to
the Soviets' both before and after a Soviet counterforce attack) is sufficient
for deterrence of premeditated attacks. The equal countervalue require-
ment could be satisfied when defenses of any level of effectiveness had
been deployed and at all levels of vulnerability of value targets to attack.
(2) Crisis stability would be likely t o decrease if assured destruction capa-
bilities were eliminated by defense. It might, however, be possible to
keep this negative effect of defense quite small. At least in principle, a
defensive situation could be made as crisis-stable as an assured destruc-
tion situation by deploying invulnerable retaliatory capabilities. While
making one's retaliatory capability entirely invulnerable might not be
possible, achieving a high level of invulnerability might be possible.
(3) The Achilles heel of defensive situations is the tremendous difficulty of
maintaining them. Defensive situations, unlike assured destruction situ-
ations, would be likely to lack robustness. This means that changes in
the adversary's forces that could undermine deterrence of premeditated
attack and create incentives for preemptive and preventive attack
would be far more likely in defensive situations. The lack of robustness
W O L I I ~likely result in tense superpower relations, making security cooper-
ation extremely difficult. Due to the difficulty of creating robust defensive
situations, the probability of nuclear war would be higher than in assured
destruction situations.
Overall, then, this evaluation concludes that the probability of nuclear

war in defensive situations would be higher than in assured destruction situ-
ations. This increase in the probability would not be due primarily to a
decrease in the United States' ability t o deter premeditated attacks or to
maintain crisis stability: if the superpowers could not change their forces,
then there might be defensive nuclear situations in which these wars would
not be more likely than in assured destruction situations. This constraint,
however, is unrealistic. The superpowers would be able to alter the status
quo, that is, t o change the offensive and defensive forces which are deployed
a t a given time. Due t o the lack of robustness in defensive situations, this
competitive armament would be more likely to result in a nuclear war in defen-
sive situations than in assured destruction situations.
So, if security depended only on the probability of nuclear war, then
defenses would unambiguously decrease security. But security depends also
on the damage that would result if there were a war. Because defenses
would decrease the damage of certain wars, they would thus have some
positive effects as well as negative ones.
To illustrate this in greater detail, I will examine briefly each path to
nuclear war. Consider first a nuclear situation in which effective defenses have
been deployed, the equal countervalue requirement is satisfied, and assume this
situation characterized by offensive and defensive forces of a given size and
capability is maintained, that is, neither country builds forces that increase its
countervalue capability. Deterrence could fail even when the equal counter-
value requirement was satisfied: this is true today when the assured destruc-
tion requirement is satisfied, and it would be true in a defensive situation in
which the equal countervalue requirement was satisfied. The damage in an
all-out countervalue war, however, would be lower in the nuclear situation
with defenses. Since the probability of premeditated nuclear war might not be
greater in the defensive situation, the expected costs along this path to nuclear
war would be lower than in an assured destruction situation.
In this defensive situation, the damage from a preemptive attack would
be lower than in an assured destruction situation. The probability of pre-
emptive war, and therefore the expected costs of a preemptive attack, would
depend upon the situation's crisis stability. If the defensive situation were
less crisis-stable than an assured destruction situation, then the probability
of preemptive attack would be higher. In this case, the defensive situation
would have a higher probability of preemptive attack and lower costs if the
preemptive attack were to occur. Therefore the comparison of expected
costs from preemptive attack in defensive and assured destruction situations
would be indeterminate. If, on the other hand, the defensive situation were
as crisis-stable as the assured destruction situation (which is possible if the
retaliatory capability is invulnerable), then the expected cost along the pre-
emptive path would be lower than in an assured destruction situation.
Now consider the case in which we drop the constraint that the status
quo defensive situation would be maintained, that is, we include the possi-
bility that countries might deploy forces that increase the adversary's vul-
nerability or decrease their own. Unlike the preceding case, in this case the
damage in defensive situations might not be lower than in assured destruc-
tion situations. The changes in the countries' forces might return the defen-
sive situation to an assured destruction situation. The potential damage
would not be determined by the status quo forces. If a country could build
its way out of the defensive situation (i.e., regain its assured destruction
capability), then the damage to the adversary could be the same as in a
mutual assured destruction situation. The expected costs of nuclear wars
resulting from changes in the status quo could be greater in defensive situ-
ations than in assured destruction situations: the damage in defensive situ-
ations is not constrained by the status quo forces and might not be lower
than in assured destruction situations; and, due to the lack of robustness in
defensive situations, the probability of nuclear war along this path is higher.
In summary, defenses could reduce the damage that could occur in cer-
tain types of wars. This positive effect would, however, tend to be offset by
Why Even Good D e f e n s e s may b e Bad 405
the increased probability of wars resulting from the lack of robustness of

defensive situations. The net result, therefore, of both superpowers' deploy-
ing effective defenses might not be to increase U.S. security.
As I stressed above, this indeterminancy is extremely significant. It
undermines the commonly held belief that defensive situations, and so-
called defense-dominance, would be preferable to assured destruction s i n -
ations. Even after making the best case for defenses (that is, ignoring
questions of technical and economic feasibility, the effect on the probability
of superpower conventional wars, and a number of other issues discussed
briefly at the end of this article), assured destruction nuclear situations
might be preferahle to those in which defenses drastically reduce the super-
powers' vulnerability to nuclear attack.'-
The inlplications of this conclusion for U.S. policy are obvious, and pro-
found. Whether the United States could be more secure in a world of highly
effective defenses should no longer be viewed primarily as a technological
issue. The United States should examine more completely the strategic and
political issues associated with defense against nuclear attack before niak-
ing a decision to pursue such a fundamental change in nuclear strategy.
Given the reservations about deploying defenses raised by the preceding
analysis, the recent enthusiasm for BMD must be judged imprudent.
Additional Problems with Strategic Defense
This examination of strategic defense has analyzed a best case for defense.
Effective defense was hypothesized to be technologically and economically
feasible. Even with these highly controversial assumptions, the decision to
pursue the deployment of defense and to make the associated fundamental
shift in nuclear strategy is found to have serious shortcomings. The case for
effective defense and for starting to deploy defenses in the foreseeable future
is further weakened by a number of factors:
( 1 ) Uncertainty. The effectiveness of U.S. defenses would be uncertain
and small uncertainties would be highly significant. In addition to the
uncertainties inherent in the operations of complex systems, the effective-
ness of defenses would be uncertain due to the severe limits on testing. The
defense could not be tested against a full scale attack or against Soviet offenses.
And while estimates could be made of effectiveness against deployed Soviet
offenses, there would always he reasonable questions about Soviet penetration
aids that could be quickly added to their offensive force.
Small uncertainties would be significant because, with the large offensive
forces which are currently deployed, a small difference in the percentage of
penetrating weapons would translate into a large difference in destructive
potential. The uncertainties involved with a defense which was in fact per-
fect would be likely to be large enough to leave the United States unsure
about whether it was vulnerable to an annihilating attack by the Soviet
Union.
The effect of uncertainty would affect U.S. policy in a number of ways.

The United States would never feel adequately defended. (Nor would the
Soviet Union.) Even without uncertainties, there would always be arguments
that the United States needed additional defense to improve its protection
against Soviet attacks and as a hedge against Soviet offensive breakthroughs.
The strength of these arguments would be greater than those made about
the inadequacy of today's offensive forces since defensive capability would
start to become redundant only once the defenses were perfect. The existence
of uncertainties would be likely to result in unrelenting requests for additional
defenses, yet fulfilling these requests would yield little satisfaction and add
little to the public's sense of security.
A second effect of uncertainty would be the creation of fears that the
Soviet Union had a superior defensive capability. Prudent military analysis
could require assessing uncertainties in favor of Soviet defense and against
U.S. defense. As a result, if the United States and the Soviet Union had com-
parable defensive capabilities, U.S. defenses would not provide confidence
that the United States was maintaining a strategic nuclear balance and
would likely be judged inferior. This conclusion would contribute to the
demands for improving defensive capabilities.
(2)Allies. Any comprehensive analysis of defensive situations must con-
sider the reaction of U.S. allies and the implications for their s e c ~ r i t y . 'One
~
issue of great importance to them has already been raised, that is, the effect
of defenses on the probability of conventional war. If strategic defense were
believed to increase the probability of conventional war, then tremendous
resistance from the European allies should be anticipated. Conventional
wars in Europe are expected to be so costly that they are barely less unaccept-
able than are nuclear wars to many Europeans. A second concern would focus
on the vulnerability of allies to nuclear attack. A policy that drastically reduces
the United States' vulnerability to nuclear attack while leaving its European
and other allies highly vulnerable cannot look good from their perspective.
A third concern would be the effect of defenses on the independent deterrent
capabilities of the French and British. A highly effective but imperfect Soviet
defense would leave the United States with a modest retaliatory capability, but
would probably eliminate the value of these independent European deterrents.
(3) Suitcase bombs. The ability to defend effectively against ballistic mis-
siles, cruise missiles, and bombers could greatly increase the importance of
clandestinely delivered nuclear weapons. Nuclear bombs could be placed
on Soviet ships and commercial airplanes, or could be carried into the
United States by Soviet agents. These alternative types of delivery are pos-
sible today, but are not of great importance due to the Soviets' large ballistic
missile and air-breathing threats.
These alternative forms of delivery would not necessarily render defense
useless: the Soviet ability t o deliver clandestinely weapons in a crisis might
be severely limited; hiding weapons before a crisis would be risky unless
early detection would be impossible; and the damage from clandestine
CI w.)! Why Even Good Defenses may b e Bad 407
attacks might be less extensive than is currently possible without defenses.

Still, the observation that defense against the delivery systems which are
most important today would not eliminate vulnerability to nuclear attack
raises basic issues about strategic defense: What threats must the United
States be able to defend against? H o w would a "partial defense," that is, a
defense against standard delivery systems, affect the nuclear threat? How
would highly effective or perfect defense against standard delivery systems
affect the political and military uses of nuclear weapons?
Conclusion
Strategic defense and the prospect of being invulnerable to nuclear attack

have undeniable appeal. But there is no excuse for being romantic or un-
realistic about the nature of a world in which the superpowers have built tens
of thousands of nuclear weapons and sophisticated delivery systems, and in
which the knowledge about these technologies cannot be destroyed. Strategic
defense cannot return us to a pre-nuclear world. Defensive situations have
not been studied as carefully or extensively as assured destruction situations.
There is, however, no reason to believe defensive situations would be either
less complex or easier to manage than assured destruction situations.
The hest of worlds in which both superpowers have effective defense
would not be so good and might not be preferable to today's redundant
assured destruction situation: in all but the case of perfect defense, the U.S.
would still depend upon deterrence for its security; the lack of robustness
in defensive situations would make them sensitive to small changes in forces
and would create strong incentives to pursue threatening improvements in
offensive forces; the acquisition of these forces would increase the prob-
ability of nuclear war; the probability of large conventional wars between
the superpowers and their allies might well increase; and, the threat posed
by clandestinely delivered nuclear weapons would be much more significant
than today.
Any serious policy for deploying defenses must address the dangers that
would result from the difficulty of maintaining the defensive situation. This
article has argued that no defensive situation could be highly robust. The
most robust defensive situations will require superpower cooperation. This
brings to the forefront the issue of U.S.-Soviet relations in a defensive
world. Recent statements by the President have suggested that effective
defenses would eliminate the need for offensive weapons.?' This outcome is
not in~possible,but is extremely unlikely. A more realistic assessment is that
deploying defenses would lead to an intense offensive and defensive nuclear
weapons competition between the superpowers and to tense, strained rela-
tions. We should expect that arms control agreements to limit or reduce offen-
sive nuclear forces would be difficult, if not impossible, to negotiate. Careful
thought should be given to whether, in a defensive situation, a cooperative
relationship between the superpowers would be possible, and to whether

the pressures for confrontation could be kept low. If these would not be
possible, and I believe they would not be, then the prospects for improving
security by shifting t o a world of effective defenses must be judged to be
especially gloomy.
The United States appears to be a t the beginning of a major shift in
nuclear weapons policy. There is no evidence that the decision t o pursue
highly effective defense was based upon a complete analysis of defensive
situations. Unfortunately, a world in which both superpowers deployed
effective defense is far less attractive than its proponents suggest: even after
making the most optimistic assumptions, defensive situations might not be
more secure than assured destruction situations; and the more likely out-
comes of deploying BMD would place the U.S. in a situation far less secure
than today's. Until a convincing argument is presented for this fundamental
change in U.S. nuclear weapons policy, the United States should cut back
severely on its enthusiasm and funding for strategic defense, attempt to
repair the damage that is likely to have occurred in Soviet understanding of
U.S. nuclear weapons policy, and pursue with renewed determination a pru-
dent policy of offensive weapons acquisition and arms control.
Author's Note
The author would like to thank Robert Art, Albert Carnesale, Lynn Eden, M~chaelNacht,
Thomas Schelling, Stephen Van Evera, Stephen Walt, and the members of the Avoiding Nuclear
War working group for their helpful comments on earher drafts of this article.
Notes
1 . The N e w York Times, March 24, 1983, p. 20.

2. Clarence A. Robinson, Jr., "Panel Urges Defense Technology Advances," Avration Week
and Space Technology, October 17, 1983, p. 16.
3. A ballistic missile defense is a system capable of destroying Sowet missiles (or warheads)
in flight. The terms "ballistic missile defense" (BMD) and "anti-ballistic missile" (ABMI are
usually used interchangeably. BMD programs which might contribute to the goals described in
President Reagan's so-called "star wars" speech are also referred to as the Strategic Defense
Initiative ISDI).
, ,
Influential arguments made in opposition to BMD in the earlier debate are found in: Abram
Chayes and Jerome Wiesner, eds., ABM: A n Evaluation of the Decrsron to Deploy Anti-Ballrstrc
Missile Systems (New York: Harper and Row, 1969); and Richard Garwin and Hans Bethe,
"Anti-Ballistic Missile Systems," Scientific American, March 1968, pp. 164-174. Arguments in
favor of BMD were presented in Johan J. Hoist and William Schneider, Jr., eds., W h y ABM?
Policy Issues in the Missile Defense Controversy (New York: Pergamon Press, 1969). A recent
book on the subject which does not promote any specific BMD policy is Ashton B. Carter and
David N. Schwartz, eds., Ballistic Missile Defense (Washington, D.C.: Brookings Institution, 1984).
Recent arguments against BMD are found in: Space-Based Missile Defense, A Report by the
Union of Concerned Scientists (Cambridge, Mass.: Union of Concerned Scientists, 1984); and
William E. Burrows, "Ballistic Missile Defense: The Illusion of Security," Foreign Affairs, Vol. 62,
No. 4 (Spring 1984), pp. 843-856. For current arguments in favor o f BMD, see Keith B. Payne
Why Even Good D e f e n s e s may be Bad 409
and Colin S. Gray, "Nucle,~rPolicy and the Detenswe Tr,lnsit~on,"Foreign Afhzirs, Vol. 62, No.
4 ( S p r ~ n gI 984), pp. 820-842.
4. For a n explicit st,itement of rhis helief from strong opponents of IIML), see S p a c r - R a s c ~ i
h / l i s ~ i 1)cfensc,
l~~ in w h ~ c hthe Llnion of Concerned Scientists states, "if it were possible t o put
In place o\'ernight a tnll! etfccrive, invulnerable defense against nuclear weapons, there could
lir~rdlybe serlous oblections t o doing this" ( p . 7 1).
5. This d~srinctionis important because these t w o types of detense h ~ v teu n d m i r n t ~ l l yd ~ f -
fercnt ~ t r ~ i t e g i c~ n ~ l ~ c a t i o1'n scountry's
: area defense, if suiiic~entlyeffective, could reii~ic-rthe
si/e o t the ~ ~ d v e r s a r ydeterrent
k threat; a country's pomt detense, hy incre'ls~ngthe s ~ / eot it\
offensive torce that cvould survive a counterforce attack, could rncreasc the size of the cori~try's
deterrent threat.
6. An assured destruction i a p a b ~ l i t yis generally understood t o he the capalxliry, f o l l o w ~ n g
' full sc,ile counterforce attack against one's forces, t o inflict a n extremely high h e 1 of d ~ n i a g e
1
upon one's adversary. T h e levels of potential damage which analysts believe assured desrruct~on
require5 are u s u ~ l l ysimilar t o those prescribed by Robert McNam'lra, U S . Secret~r! of Defense
from 1960 t o 1968. McNamara's criteria for assured dectrucrion, which were influenced b! the
J i m i n ~ s h i n grnarg~naldarnage potenti,ll of increasmg the size of the U S . force, rcqu~redthat the
U n ~ r e dSr'ltes be able t o destroy, In a retaliatory attack, .~pproximately2.5 percent of the S o v ~ c t
n 50 percent of Soviet industry. H e judged that such a level of destruction \voultl
p o p ~ ~ l a t i oand
he ~ntolerablet o the Soviet Umon and, therefore, thar the c,ipabiliry t o inflict this level of dani-
.igc would he suff~cientt o deter deliberate Soviet nuclear a r t ~ c k so n the United St,~tes.See A h i n
C:. Enthoven and K. W < ~ y n cSmith, H o u , Much IS t m m g h ? S h a p r t ~t ~l l~I ~ ~ f &PYO~IZIIII s~
1961-1960 ( N n v York: Hdrper and Kow, 1971), pp. 172-1 84, 207-210.
A rel,lted, but conceptually distinct, interpretation of assured destruction focuses o n the
r c l a t i o n s h ~between
~ the costs a decision-maker associates w ~ t hthe nuclear ~ t t a c k, ~ n dthe
dani'ige thar cvould result tram such a n attack. Assured dcstruct~on In this Interpretatton
req~liresthat the potential damage in one's retaliatory capab~lityshould be sufficiently high
that ~ncreasingthe potential d'lmage would not result in significantly higher costs t o the adver-
sary. In this article, assured destruction is intended t o have this second me,inlng. Cle,lrly, an!
evaluation of the costs associated w ~ t hsuch unprecedented damage is hrghlj. s ~ ~ h j e c t i vMan! r.
people believe t h a t the United States would have t o he able t o reduce damage t o itselt far helow
the Levels specificd by M c N a m a r a before it could significantly improve the outcome o f ,in all-
o u t war; others belleve that .iny r e d u c t ~ o nin damage, even if d:iniage renuineci well a h o \ e
these levels, would he sigmficant. T h e t w o d~fferentunderstmdings of assured destruction 'Ire
often n o t dlstinguished bemuse M c N a m a r a said that a n assured destruction ccipcil>~lity would
he suffic~entlv . ltlrrre
<>
t o a n m h ~ l a t eone's adversarv in retallation and because analvsrs rend t o
~ s s u ~ thdt n e costs t o one's adversary could not be increased if the advers,lry could '~lreadyhe
annih~lated.
T h e arguments in this a r t ~ c l ed o not depend upon ,I specific assessment of the Iwel of rer,ll~-
atory datiiage required tor ;~ssureddestruction. Instead, the arguments \ i e w the level of dam-
age required for assured destructioll as a n i~llpreciseboundary, above which ,~dditlon.lld,lmajie
does nor signif~caritlyincre,~sethe costs of a n attack, c ~ n helow d which reductions In damage
would signdicanrly reduce the costs. People disagree o n the loc'lriion of rhis houndar); but thr
argument\ apply in :]I1 c,lses.
7. For an authoritative an,llysis of the technical feas~hilityot RML), see Ashron K. Llrter,
Ilirei-teti t n e r g y Missile Dcfertse 112 Space (Washington, D.C.: 11.5. Government P r i n t ~ n gOffice,
1984). (:arrer judges 'is "extremely remote the prospect that directed-energy KhlD (in concert
with other layers it necessary) will succeed In reducing the vulnerahliry of 0.5. popu1,lrion c~rnd
wciety t o the neighborhood of 100 megatons o r less" (p. 68). See also: SPLITC-ht~s~d MISSIIC
Drfiwse; Spurgeon M. Keeny and Wolfgang K.H. Panofsky, " M A D Versus NUTS," I ~ ) r c r j i r ~
Affairs, VoI. 60, N o . 2 (Winter 1981-82). pp. 297-303; and Carter and S ~ h w ~ ~Bnllist~c rt~,
Mrssilc l k f i w s e , in which Carter concludes: "the prospect thar RML) will thw'lrr the rnutu.11
h o s r ~ g erelationsli~p- I I thrs is taken literall? t o mean the ab~liryof e'lch superpowel- t o d o
socinll! ~ n o r t a ldamage t o the other with nuclear weapons - is s o remote '1s t o he of n o prxtical
111terest" ( p . 1 I ) .
8. It should not go unmentioned that many of the advocates of BMD favor asymmetric
deployment - that is, situations in which the United States can gain a strategic advantage by
deploying BMD which is superior to Soviet BMD. See, for example, Colin S. Gray, "Nuclear
Strategy: The Case for a Theory of Victory," International Security, Vol. 4, No. 1 (Summer 1979),
pp. 54-87; and Colin S. Gray and Keith Payne, "Victory Is Possible," Foreign Policy, No. 39
(Summer 1980), pp. 14-27. In contrast to these earlier articles, in the recent "Nuclear Policy and
the Defensive Transition," Gray and Payne argue as though the defensive sltuatlon they advocate
would be symmetric. They do not explain the origins of this apparent inconsistency.
9. The following discussion assumes that both countries would know the effectiveness of both
their own defense and the adversary's defense. This is admittedly unrealistic, since there would
always be uncertainties about the effectiveness of the defenses, and because the implications of
these uncertainties could be significant. The reason for assuming that the effectiveness of the
defenses would be known, however, is to focus the examination of perfect defenses on other
issues. This assumption of certain information strengthens the arguments for perfect defense and
therefore reinforces the best case assumptions used in this analysis. Some of the comphcations thar
would likely result from uncertainties about effectiveness are discussed later in this articlc.
10. Many advocates of pursuing h~ghlyeffective defense argue that even ~f the prospects
for effective defense do not look extremely promising today, history suggests that major
technological changes should be expected. For example, Payne and Gray observe In "Nuclear
Policy and the Defensive Transition" that: "All of recorded history has shown swings in the
pendulum of technical advantage between offense and defense. For the strategic defense t o
achieve a very marked superiority . . . would he an extraordinary trend in the light of the last
30 years, but not of the last hundred o r thousand years. Military history is replete with ex-
amples of defensive technology and tactics dominating the offense" (p. 826). This argument
would, however, apply at least as well t o the maintenance of the defensive world they advo-
cate and points to the major problems that would exist in defensive sltuations.
11. For an insightful discussion of why large nuclear arsenals reduce the probability of
superpower conventional wars, even when neither superpower has an advantage in purely
military terms, see Robert Jervis, "Why Nuclear Superiority Doesn't Matter," Political Science
Quarterly, Vol. 94, No. 4 (Winter 1979-80), pp. 617-633. At the other end of the spectrum,
Gray and Payne in "Victory Is Possible" find the U.S. strategic force inadequate to meet its
extended deterrence commitments, but admit that U.S. strategic nuclear forces d o contribute
to deterrence of Soviet conventional attack in Europe (p. 16).
12. This assertion depends on the assumption made above that both countries know thar
the defenses are perfect. If defenses were not known to be perfect, although in fact they were,
then nuclear attack might be carried out (but would not result in damage) and nuclear threats
might be used coercively.
13. The probability of preemptive nuclear war depends on the probability of crises, as well
as the crisis stability of the nuclear situation. For example, a change in the nuclear s~tuation
which increases crisis stability but also increases the probability and severity of crises could
increase the probability of preemptive nuclear war. The following comparison of the prob-
ability of nuclear war in defensive and assured destruction s~tuationsdoes not take into con-
sideration the relative probability of crises. Because defensive situations are likely t o increase
tensions between the superpowers and because superpower cooperation will he more difficult
than in assured destruction situations, the effect of not including the probability of crises in
this analysis probably favors defensive situations. Therefore, this simplification tends to re-
inforce the best case which t h ~ analysis
s makes for defensive situations.
14. That deterrence requires assured destruction capabilities was rarely made as a separate
argument. It was, however, an integral part o f the argument that BMD would necessarily result
in an arm race. The inevitability of this arms race was based in part on the assertion that each
superpower, to maintain an effective deterrent, would have to possess an enormous retaliator!
capability. See, for example, Chayes and Wiesner, eds., ABM, pp. 49-54.
IS. A similar argument is made by Donald Brennan in "The Case for Population Defense," In
Hoist and Schneider, Why ABM?, pp. 100-106. An earlier version of this argument appeared in
Donald Brennan and Johan Hoist, Ballistic Missile Defeme: Two Vietc,s, Adelphi Paper KO.43
(London: International Institute for Strategic Studies, 1967), pp. 9-1 1.
\,l,i-i t Why Even Good D e f e n s e s may b e Bad 41 1
Including 111 this analysis uncertainty a n d ~mperfecrinformation a b o u t the level of vulner-

a b i l ~ t pt o countervalue attack would weaken t h ~ sargument. Reclundar~tassured destruction
capabilities are extremely large by a n y reasonable evaluation. There is little opportunity t o
~ n i s l ~ ~ dt hg ~e sd r s t r u c t ~ v epotenti.~l, a n d assessments o f damage are theretore not 5ensitlve t o
relat~velysmall d~fferencesin force sire. In contrast, 111 a defensive s ~ t u a t i o nin which each
country's ability t o inflict damage has been greatly reduced, relative force capahil~tiecwould
he harder t o evaluate, a n d uncertainties, m ~ s e v a l u a t i o ~ lasn, d ~ n ~ s p e r c c p t i o nwould
s he more
likely t o result in a perceived advantage t h a t could result in a fadure of deterrence.
16. 7111s does n o t mean, however, that the Soviet Union \vould necessar~lyhe ~ ~ n a b tl oe
coerce t h r United States. As In a s i t u a t ~ o nof m u t ~ l a lassured destruction capahilitic5, if the
Soviet I J n ~ o nwere able t o convmce the United States that it would carry o u t a threat t o attack
L1.S. cities, then the Soviet Union might be able t o coerce the United States. T h e U.S. posses-
sion of 311 equal c o ~ m t e r v a l u rcapability, by making possible a highly credible retaliatory threcir
comparable t o the Soviet threat, would make it difficult for the Soviet Union t o make 11s coer-
cive threat convinc~ng.If the Soviet Union were able t o coerce the United States, the key t o its
success would be greater recolve a n d willing~iesst o rake risks than the United St.~tes,a n d nor
a n a d v a ~ i t a g ein nuclear forces.
17. There is a third reason, n o t discussed in the text, w h v the Soviet Union might a t t x k
the United States: t o eliminate the U.S. abillry t o deter the Soviet Union from pursuing ~ t for- s
elgri policy object~ves.For example, c o n s ~ d e ra hypothetical case in which the United Stares
deters Soviet attack o n Western Europe entirely with threats of strategic nuclear retaliation. If
a Soviet counterforce attack c o ~ ~ suffic~ently ld reduce the potentla1 cost of U S . retdi:ltion, then
the Soviet Union could judge thar attacking the United States, incurring U.S. ret,ll~ation,a n d
~ n v a d ~ nagn d a c q u i r ~ n gWestern Europe could result in a net benefit. This rype of Soblet
nuclear attack o n the U.S. homeland, unlike the t w o d~scussedin the text, does not depend
upon the Soviet Union's ability t o i n f l ~ c tcountervalue damage o n the Unitrd States. 111 con-
trast, it i \ a purely milttary attack, motivated entirely by the desire t o reduce the damage the
United States could inflict o n the Soviet U n ~ o n T . h e United States could eliminate the Soviet
incentive for launching this rype of attack hy making its forces i n v u l n e r ~ b l e .However, even
with invulnerable forces, a n d with the equal countervalue requirement \atisfied, there might
he nuclear s i t u ~ t i o n sIn w h ~ c hthe U.S. countervalue threat would be ~ n s u f t i c ~ e n t llarge y to
deter S o v ~ e tattack o n Europe. This would he true for the same reason thar perfect defenses
could increase the probability of conventional war: U.S. escalation t o the nuclear level would
n o longer he suff~clentlycostly t o deter Soviet attack.
18. See, for example, Spnce-Based Mtssile Defense, pp. 79-80; Willian~C. Foster, "Str'lteg~c
Weapons: Prospects for Arms C:ontrol," Foreign Affairs, Vol. 47, No. 3 (April 1 %9), pp. 4 1 4 4 1 5 ;
and Robert 1.. Rothstein, "The ABM, Proliferation and International Stalillr);" borcqrz A f i l r s ,
Vol. 46, N o . 3 (April 1968), pp. 4 9 8 4 9 9 .
19. Extens~vediscussions of c r i s ~ sstabil~tyinclude: T h o m a s C. Schelling, The Str~ztegyof
Conflirt (Cambridge, Mass.: Harvard University Press, 1960), pp. 207-254, ,und Arms 1117d
Itrflz~crtrc( N e w Haven: Yale University Press, 1966), pp. 221-248; and Glenn H. Snyder,
Ilc~crrcnccand Defense ( P r ~ n c e t o nPrinceton : Univers~tyPress, 1961 ), pp. 97-1 14.
LO. T h e assumption rhat decision-makers would anticipate a c o u n t e r v d u e s t r ~ k eis implicit
in many discussions of c r ~ s i sstability. It underlies the logic that says if 1' counterforce attack
could reduce the adversary's countervalue then there will he a n incentive t o strike
tirst. A crisis, however, should provoke fears of a counterforce attack. If we assume the adver-
s'lry's first strike would he counterforce, then the nuclear s ~ t u a t i o nIS far more cris~s-stablethan
i f w e nssunie the attack would be countervalue. For a good discuss~onof this argument see
Snyder, Deterrcnre and Defense, pp. 104-109. If we assume that both countries a n t i c i p ~ t e
counterforce first strikes, then the effect of defenses o n crisis stability is likely t o he m ~ n i m a l .
(;iven this assumption, the lncentlves t o preempt would be small, o r nonexlsrent, with o r with-
o u t defenseq.
21. Arms r'ice stahilitv is the standard measure o f t h i s characteristic of the nuclear s ~ t u ~ ~ t i o n .
I have chosen t o use the term "robustness" t o avoid the confusion rhat surrounds the term
"arms r'lce stability." Arms race stability brlngs t o m m d a t least t w o issurs which are related t o
rohustne\s, but whlch are conceptually distinct. First, arms race stability I S often considered a n
4 12 T h e Cold War a n d Nuclear D e t e r r e n c e
indicator of the likelihood andlor intensity of arms races that will occur in a specific nuclear
situation. Arms races, however, can occur for a variety of reasons which are only peripherall!-
related to the effect of building nuclear forces on the adversary's security. Consequently, arms
races can occur In highly robust nuclear situations, as has occurred in our current highlv redundant
and diversified assured destruction situation.
Second, use of the term "arms race stability" can connote a belief that arms races cause
wars. Whether arms races actually cause wars is a theoretical issue on which there is substan-
tial disagreement. But one can assert that the probability of war depends upon the robusmess
of the nuclear situation without believing that, in general. arms races cause wars. Robustness
is a measure of how sensit~vea country's security would be to the adversary'c buildup of forces.
It does not imply that the process of competitive armament itself leads to w ~ r Rather,
. assum-
ing a force buildup takes place either competitively o r un~laterally,a war IS more likely when
the initial nuclear situation is less robust.
22. On the existence of countermeasures, see Carter, Directed Energy M~ssilrDefense 172
Space, pp. 69-70.
23. For a discussion of how arms control might help to Increase the robustness of deten-
sive situations see Charles L. Glaser, "The Implication of Reduced Vulnerabil~tyfor Security in
the Nuclear Age," Ph.D. dissertation, Harvard University, 1983, pp. 217-228.
24. For a discussion of the factors which affect the probab~litythat countries will be able
to cooperate, see Robert Jervis, "Cooperation Under the Security Dilemma," World Politrcs,
Vol. 30, No. 2 (January 1978), pp. 167-214.
25. This formulation of objectives assumes that the United States IS a strictly status quo
power. This may not he entirely accurate, but is a reasonable assumption for this discuss~onof
nuclear weapons policy. A further concern about this formulation is that it does not mclude
the objective of minimizing losses that could result from nuclear coercion. W h ~ l et h ~ iss clearly
an objective of U.S. policy, the nuclear capabilities required to achieve ~tclosely resemble those
required t o minimize the probability of nuclear war. Consequently, not explicitly ~ n c l u d ~ nthic
g
objective does not bias the analysis.
26. For a more complete analysis see Glaser, "The Irnpl~cationsof Reduced Vulnerahil~ty
for Security In the Nuclear Age."
27. The transition to a defensive situation deserves extensive analysis that is beyond the scope
of this paper. The preceding analysis can, however, be used to analyze U.S. securlty durmg a sym-
metric transition. A symmetric transition would likely be the safest possible transit~on.Therefore,
because a transition would probably not be symmetric, the significance o t the symmetric tr'in-
sition should not be overestimated. However, because the transition to a defense situation is
generally believed to be very dangerous, it is interesting to briefly consider this best case.
Secur~tyin transition states would be determined by the same factors as In the defensive
situations already studied. Applying the conclusions o f the analysis of defensive s~tuationsto
the transition, we find: 1 ) Premeditated attacks could be deterred equally well throughout the
transition; 2 ) Crisis stability m ~ g h tnot decrease during the transition; and 3 ) Robustness
would decrease during the transition. It follows that the probab~lityof nuclear war In the final
state of the transition is likely to be greater rhan in any of the transit~onstates.
The analysis of defensive situations presented in this art~cleturns the convenr~onalwisdom
about strategic defenses on its head. It finds that the desirability of effective defenses, which is
usually taken for granted, is at best highly questionable; and that the probability of war dur-
ing a symmetric transition, which is usually believed to be high, could he lower rhan in the
final defensive state.
28. For a discussion of the likely alliance reaction to extensive U.S. homeland defense, see
David S. Yost, "Ballistic Missile Defense and the Atlantic Alliance," International Security, Vol. 7,
No. 2 (Fall 1982), pp. 154-158.
29. The New York Times, March 30, 1983, p. 14.
Transarmament: From Offensive to Defensive Defense
Johan Galtung
1 . Reactions to a n Attack
T he word-pair 'offensiveldefensive' is problematic, hut also crucial. In

an effort to have a fresh look at the whole problem of security, a fig-
ure giving a spectrum of reactions to an attack on a country may be
useful (Figure I). The spectrum is one-dimensional which means that it is
simplistic, possibly too simplistic - hut it may nevertheless be useful.
At the bottom end there is no resistance at all in case of an attack; at the top
end total destruction - of oneself as well as of the attacker. In between are all
other forms of reactions - the spectrum includes all 'wavelengths', so to speak.
The basic thesis of this article is simply that almost all the current debate
concerning which reactions to make use of is focussed on two major cuts
along this dimension, between nuclear and conventional arms on the one
hand, and between violent and non-violent reactions on the other. The lat-
ter is the distinction around which not only pacifism but also large sections
of the peace movement are organized: the rejection not only of nuclear arms
and other weapons of mass destruction, but also of violence in general,
meaning all kinds of conventional military systems. Although most people
might agree that there is such a distinction, only relatively few would share
the optimism o f pacifists - when they point to such examples as Gandhian
actions in India against the British Empire - with regard to the efficacy of
non-military reactions everywhere, alone.' Hence, as is very well known, it
is the distinction between nuclear and conventional weapons that dom-
inates the political debate and action completely, and not only in the mili-
tary and political establishments (including the war establishments), but
also in the peace movement. The thesis, then, is that this is most ctnforttr-
nate, that it means cutting the dimension at points that certainly are import-
ant but have the distinct disadvantage that one of them is located too high
on the scale of destruction, and the other one too low. The cuts to the left
in Figure I are simply insufficient.
Source: / o ~ i r m z lof P r , i z ~ .K~ C S ~ L I T 2C l~(,2 ) ( 1984): 127-39.
Figure I: A Spectrum of Reactions to Attack
total
destruction
offensive
violent
1 1 1 ------------ ----
conventional
military
defense ( C M D )
defensive para-military
defense (PMD)
now non-military
violent defense (NMD)
I\
....................
no
resistance
Hence, the argumentation here is in favor of a third cut, that between

offensive and defensive reactions to a t t a ~ k In
. ~ trying to define this cut,
which like the other two by no means is a sharp one, one should first of all
emphasize that it refers to the objective capability of the reaction 'systems'
(the weapons being a part of that concept), not to the subjective motivations
that may be attached to them. In other words, it is not a question of whether
a reaction system is intended to be used for an attack; the whole issue is
whether it is capable of being used for an attack. Hence, the best judge as to
whether a weapon system is defensive or offensive is a possible target of the
system, the adversary, not the subjective mind behind it. Thoughts and
words come and go, actions depend on what is objectively possible, given by
the constraints of natural laws only. The adversary is the best judge; just as
we, in our self-defense, are the best judge of the adversary.
Hence, I would locate the definition of the offensiveldefensive distinction
in geographical space: can the weapon system be effectively used abroad, or
can it only be used at home? If it can be used abroad then it is offensive,
particularly if that 'abroad' includes countries with which one is in conflict.
If it can only be used at home then the system is defensive, being operational
only when an attack has taken place.
2. The Range and Impact Area of Weapons
Locating the definition in space makes it possible to formulate the problem

in terms of two variables: the range (of the weapons carriers) and the impact
i,,iltiing Transarmament 4 15
area (of the weapon itself, whether it is a classical impact weapon, an incen-
diary weapon, a high explosive or weapons of mass destruction - chemical1
toxic, biological, radiological, nuclear or geophysical). If we now divide 'range'
into immobile/short/long and 'impact area' into local/limited/ extended, then we
arrive at the nine combinations in Figure 11, four of them defensive according to
the approach taken above, five of them offensive.
Naturally, it all depends on where the borderline between 'short' and
'long' on the one hand, and between 'limited' and 'extensive' on the other, is
located. An indication is already given above: the effects of the reaction to an
attack should be within one's own country.' Of course, there may be coun-
tries so small that almost any weapon system wo~lldreach outside and/or
have an impact area that would also include adversary territory. In general
this would call for research into other types of weapon systems, for the use
of highly immobile systems with only local impact along the borders (border
fortifications are a classical answer in this connection), leaving the 'short'/
'limited' combination to core areas of the country. But even if some of this
should reach into some minor parts of the adversary's territory, this does not
in any major way affect the type of reasoning we are trying to develop here.
In order to discuss this more fully let us contrast the extremes in Figure 11.
On the one hand, in the upper right hand corner, are very long range
weapons systems with extensive impact areas: intercontinental ballistic mis-
siles, long range bombers and submarines, all of them with dual capability,
i.e. useahle for weapons of mass destruction. They would certainly be clas-
sified as offensive by anybody.
O n the other hand, in the bottom left hand corner would be such weapons
systems as land, sea, or air mines with local impact only, or a pipeline buried
underground that can easily be filled with an explosive, ignited and make
hundreds of kilometres unpassable for tanks. As mentioned, fortifications
also belong in this category, but some of them would have guns with an
impact area that would no longer be 'local', but 'limited'. Real long range
guns would be alien to the logic of purely defensive defense, however.
Figure 11: Offensive vs. Defensive Systems

iwzpact
C(YCU
A
extensive of fe n
2.
limited de fen
<
local si ve cc
> YLZiZRL'
immobile short long
Then there are all the in-between categories, and they are numerous.
However, they are not that difficult to handle from the point of view of the
present analysis. Long range weapons systems with local impact would clearly
be offensive: a Pershing I1 is still an offensive weapon when equipped with a
conventional warhead with a highly local impact: a very long range gun with
a nuclear warhead would be an offensive weapon even if stationed 'at home'.
More important is the 'short'/'limited' combination since that would bring
us to the borderline between offensive and defensive. The 'immobile'/'exten-
sive' combination, the nuclear mine or short range nuclear arms seem useless
both for defense and offense. They are 'thought-errors', leading mainly to self-
deterrence.
'Short range' means mobile, but mobility should not be useful for offen-
sive purposes. One would be thinking in terms of jeeps and similar vehicles
on land, motor torpedo boats on water, small submarines, and small air-
craft using roads as airstrips, possibly with vertical take-off and landing,
possibly helicopters. There would be nothing against these means of trans-
portation being very quick: the problem is not speed, but range. In speed
there is protection, and the possibility of coming quickly to the rescue where
defense against aggression is needed. Speed is certainly also important in
aggression, but only useful when combined with sufficient range t o reach
outside one's country.
Hence, one would be thinking in terms of highly mobile and small units
with limited range, on land, in the water, in the air. In order to compensate
for the limited range, they would have to be well dispersed all over the
national territory, but, because of the limited range, essentially with local or
district (sub-national) functions alone. If the length from one end of a coun-
try to another is so long that the range can also reach possible adversary ter-
ritory sideways - Chile, Norway and Sweden being obvious examples -
then one should renounce on weapons systems with ranges of that type, let-
ting the non-offensive character of the system take priority over the wish to
use all systems all over the national territory - and deploy systems with
shorter range, dispersed. However, if they are to operate in a dispersed and
essentially local manner, they also have to be relatively autonomous. This
does not mean that they are not under national command, only that they
are capable of operating even if that command should be seriously impaired
through adversary attack. And this, in fact, means that the whole C31sys-
tem - command, communication, control, and intelligence - also has to be
dispersed, less centralized, and that the country should not depend on out-
side suppliers for armament^.^
Having now established that they should have short range but possibly
be very quickly mobile, well dispersed, small, local and autonomous, we can
turn to the impact area of the weapons. This should be 'limited' for the very
simple reason that it is limited how much one wants to destroy of one's own
territory even if a more extensive impact area were to be more destructive to
adversary forces. This, then, would point in the direction of very efficient,
precision guided weapons with considerable destructive power but limited
impact area, an example being 'smart rockets'. They certainly exist today
I Transarmament 4 17
and are generally seen as very effective against tanks in the form of antitank
weapons, and against ships, but perhaps less so against aircraft, particularly
when they make use of the old trick of interposing themselves between
defensive forces and the sun. However, there would be ways of dealing also
with this problem. Let it only be added that such forces in addition would
have weapons with a highly local impact such as ordinary guns, thereby
completing the four cells in the defensive area of Figure 11.
3. The Gray Zone
Of course there is a gray zone in between. There is the famous case of the
anti-aircraft guns that are defensive when pointing upwards, yet can be
used as highly offensive weapons when mounted with a different angle for
targets on the ground on a carrier (a ship, for instance) with a long range.
This, however, is no argument at all against the distinction made. What has
happened in that and similar cases is that a new weapon system has been
created, from something immobile with limited or even local impact area to
something long range with limited impact area. That one major physical
component in two weapons systems could be the same, or the same with a
minor modification, is trivial. A country that wants to base its security on
defensive forms of defense would simply not undertake that type of trans-
formation of the weapons systems, and try to make them so that they can-
not be suspected of making such a move either. At the same time, however,
this serves as a warning not to be naive in believing that any component of
a weapon system is inhevently defensive or offensive; it depends on the total
system. It should not, however, depend on motivation. As motivations
change so may the objective character of the weapons system - hence it is
an engineering problem to make systems that are highly resistant, 'robust',
to such changes, retaining the defensive character over a vast range of trans-
formation of the components.
Going back to Figure 11, there are still a number of clarifications to be
made. More particularly, if we make use of all three cuts that have been
made on this single dimension, cutting the dimension in four regions, some
comments about each of the four regions might be appropriate to bring out
the issues.
Fivst, there are the weapons of mass destruction, with most of the public
debate and action concentrated on nuclear arms. They are classified here as
offensive, and that is not entirely unproblematic. The reasoning was indicated
above: weapons of mass destruction are so destructive that nobody in his right
mind would use them at home, at most against an adversary, and even then
only against a very much hated adversary. One reason for this is that the
weapons are not only destructive of the homosphere (human beings and their
settlements) but also of the biosphere, lithosphere, and the hydrosphere - in
other words of the whole environment (the atmosphere too for that matter,
but that effect will be dispersed unless there is a 'nuclear winter'. In other
words, nuclear weapons (and other weapons of mass destruction for that
4 I8 The Cold War and Nuclear Deterrence
matter) are simply not credible as defensive weapons - which, of course, is a

major reason why they are usually conceived of and discussed in connection
with long range weapons carriers like those found in the USINAT0 triad.
A country may have short range carriers (such as 1SSmm howitzers, self-
propelling, mounted on trucks or trains) with dual capability (e.g. for ERW,
'neutron grenades'); the question still remains whether they are credible for use
on own territory. Admittedly the answer is not a very clear-cut one, indicating
that the dimension in Figure I is not entirely one-dimensional - but the basic
thesis still remains: that weapons of mass destruction are essentially offensive
weapons in the sense made use of here. In fact, they are so aggressive that they
are for destruction rather than for conquest of enemy territory. It is not sur-
prising that short-range nuclear weapons are being withdrawn on a unilateral
basis by USINATO; they are thought-errors.
Then there is the second category which is a very important one: conven-
tional, offensive weapon systems. When the basic distinction is made only in
terms of nuclear vs. conventional, then it is easily forgotten how extremely
offensive conventional weapons can be. The second World War was an ex-
ample, so were the Korean and Indochina Wars and all or most of the other
local war after 1945 for that matter, such as recent wars in Lebanon and
Afghanistan. Of course, a major war today would not be fought with exactly
the same arms, but for instance with the missiles, bombers and submarines now
at the disposal of the super-power, but 'only' with conventional war-heads.
They are destructive, and also so offensive that although reduction or elimin-
ation of nuclear arms would be advantageous, much of what can be said about
the danger of war still remains valid, with conventional offensive weapon sys-
tems doing the job. And it is precisely because the third cut along the dimen-
sion in Figure I has not been made that it becomes possible for certain political
and military establishments to smuggle in conventional offensive armament as
a 'compensation' for a possible nuclear disarmament - riding on the fear of
nuclear arms, particularly in the likely war 'theater' countries in Europe.
Then there is the third category: conventional military defense. It has
been described in some detail above, so let us here only look at one more
point. If the units carrying the burden of conventional military defense
(CMD) are short range mobile, small, local, quick, dispersed, and autono-
mous, then they are very much like guerrilla forces. The only difference
between CMD and paramilitary defense (PMD) would be that the latter
would tend to be even more local, more embedded in the local human and
natural environment, and operate less in the open, although they would
probably wear some kind of uniform in conformity with the regulations of
the laws of war. The often-used term 'militia' also enters the ~ i c t u r ehere.
including some of its policing functions. In this connection it should be
pointed out that PMD probably has proven, after 1945, to be the most
effective form of reaction to an attack, whether that attack takes the form
of direct violence of military forces or the structural violence of excessive
exploitation within and/or between countries.
Gallung Transarmament 41 9
4. Non-Military Defense
Then there is the fourth category, non-military defense (NMD). Most models
of that type of defense would also operate on the assumption of small units,
local and autonomous, dispersed - in other words the same structure that
has already been argued for CMD and PMD. One might say that there are
two reasons underlying this: never to offer the adversary any targets with
such a high concentration of defense potential that it would be worthy of
a nuclear attack, and at the same time being able to resist an attack in all
corners of the country. (Of course, he may destroy the whole country, but
he may d o that anyhow.) For the case of non-military defense this obviously
means not only territorial defense in the sense of resistance in geographically
well defined units, but also social defense in the sense of all organizations
and associations in a country finding their own ways of resisting attack by
not producing goods or services for the adversary etc. Clearly this IS defen-
sive as it is only meaningful in one's own society. This becomes even more
clear when one looks at the following short list of 12 fundamental strategies
for NMD, organized in three groups with four strategies in each.'
I. Antagonist-oriented defense strategies

A. 'Attack should not pay'
I . Self-inflicted sabotage on objects of value to adversary
2. Noncooperation and civil disobedience, 'emptying' social structure
B. 'Incapacitation of the antagonist'
3. Creating empathy
( a ) Positive interaction before attack; helpfullness, assistance
( b ) Cooperation with the person; noncooperation with the
status - friendliness at the personal level
4. Creating sympathy through suffering inflicted by adversary
11. Defense strategies aimed at protecting oneself
5. Efficient communication inside one's own group
6 . Effectively hiding selected people and objects
7. Decreased vulnerability of the population through alternative
structure
8. Communication and enaction of one's own values
111. Defense strategies aimed at deterring the antagonist
9. Organization of N M D prepared in peace time
10. Communication of preparedness through maneuvers
11. Communication of commitment to NMD
12. High level of satisfaction in one's own group
Of course, it may be argued that N M D specialists can penetrate inter-

national frontiers and organize the population elsewhere in attempts to over-
turn their regimes, as can PMD volunteers. This is true, but in that case it
is a question of transfer of know-how; the real fighting will have to be done
by the local population against its own leaders. It is more like sending a book
across the border, or a teacher, less like sending a nuclear-tipped missile.
Looking at the total dimension, again, the case has now been made that
on the one hand there are offensive reactions to an attack, with weapons of
mass destruction (including nuclear weapons) and with conventional offen-
sive weapons. O n the other there are defensive reactions, and they are of
three types: conventional military defense, paramilitary defense, and non-
military defense. Just like an offensive reaction today is conceived of as
including both nuclear and conventional weapons (in Soviet/ WTO strategy
nuclear weapons are not for first use, whereas in USINATO strategy nuclear
weapons are possibly for first use), a defensive reaction could include all
three types, combining CMD, PMD and NMD. The problem of whether
they are combinable is an important one, just as it is for offensive strategies.
That problem, however, will be taken up below.
The basic point to be discussed here is not so much the structure of
offensive vs. defensive systems, as their function. The key difference is that
offensive systems can be used for attack. They are potentially aggressive,
and hence provocative. Whether they will be used for attack is another mat-
ter; the important point is that any possible adversary may have reasons to
suspect that they can be used for attack simply because what is possible may
also become reality. What is impossible may not; this is the whole point
underlying an objectively defensive posture.
At this point some comments about the ambiguities of the two import-
ant words 'defense' and 'deterrence' may be in order.
The word defense obviously has two meanings: any reaction to an
attack, in other words the use of any weapon system from any point on the
dimension of Figure I (including the bottom point which may also be some
kind of defense, perhaps in the longer run); and then the other meaning a
limited part of the spectrum only, what is here somewhat clumsily referred
to as 'defensive defense'. And this spills over into the double meaning given
to the word 'deterrence': deterring an attack through the threat of effective
retaliation (German: Vergeltung), or deterring attack through the promise
of effective resistance (German: Verteidigung, not including 'Vergeltung').
One may say that there is a broad use of the terms defenseldeterrence cov-
ering all points on the spectrum, and the narrow use limiting it to the
(purely) defensive systems only.
It is probably not possible to change the semantics since word usages are
so deeply ingrained at this juncture. But it is absolutely impermissible when
people participating in the debate d o not clarify what they mean. At any
point where the words 'defense' and 'deterrence' are used it should be made
clear whether the terms are limited to (purely) defensive systems, or also to
systems operating on the territory of the adversary.
Offensive defense is offensive, in both senses of that term: it can be used
to start an offensive in the sense of aggression, and it is offensive in the sense
of provoking the other side. It is not the manpower, capital, research, and
organizational work that go into a military system in general that provokes;
~ , , I Ii Ir t l _ Transarmament 421
it is the offensive component of all of this. Thus, it is misleading to analyse

arms races only in terms of the magnitude and rate of growth of all the
factors that have been put into the organization, or - better - the destructive
capability of the organization (the 'bads' and 'disservices' that can be ren-
dered). Only the offensive components should he counted. On the other
hand, it is quite clear from what has been said above that the border-line is
not a very sharp one (although sharper, it is argued here, than most people
believe) - hence the easy solution, is to count all military assets, disregard-
ing the offensiveldefensive distinction. The distinction that is made use of -
army, navy, and the air force - has approximately the same level of intellec-
tual depth as the corresponding division for the animal kingdom: animals on
land, in the water, and in the air. Zoology made a great step forward when
the distinction between vertebrates and avertebrates was made systematic
use of; it is high time that similar distinctions - actually very old in the field
of military science - become much more prominent in the debate.
Recently they have been surfacing. Let me compare the points made to
some major statements, starting with the Swiss official doctrine: An official
Swiss publication (Zentralstelle fiir Gesamtverteidigung, 1973), makes a num-
ber of important statements (the only point missing is nonmilitary defense):
This entails the prohibition of any recourse to indiscriminate conduct of

war against the population of the opponent, even though in the atomic
age, only the threat of the destruction of the opponent's population
seems to be an effective deterrent (p. 23).
Should the large units of the army cease to exist as effective fighting
formations, then there will be recourse to guerrilla warfare in order to
prevent the opponent from gaining complete control over the occupied
territory and to prepare for the liberation (p. 30).
Military cooperation with other states is inadmissible for Switzerland
in peace-time because of its status as a permanently neutral state. Should
Switzerland become involved in a war, then such a cooperation can be
considered with the opponent of the aggressor (p. 29).
In order to appreciate better the Swiss doctrine, the reader is also referred
to the quotes given in footnotes 3 and 4 and Fischer 1982. The latter, how-
ever, also points out how all is not well with the Swiss case: conscientious
objection is not admitted; a heavy military hierarchy paralleling and partly
doubling the civilian one; arms exports; gross cost overruns by the military;
no preparation for non-military defense; little or no peace and conflict
research.
The Swedish doctrine is somewhat weaker, based on the idea that
'Defense should not be perceived as a threat by anybody, and it should
be elaborated in such a way that the exclusively defensive purpose - to
defend one's own country - becomes absolutely clear' (Totalfiirsvarets
upplysningsnamd, p. 8 ) . What is missing here is a discussion of doctrine and
of objective capability, not of the subjective perception, with Swedish
recommendations as to what perception is legitimate and which ones are

not. The Swedish Peace and Arbitration Society, on the other hand, advo-
cates non-military defense only, in a very well argued case against military
defense as 'building on an improbable war scenario, based on a dubious
balance of power theory, preventing us from taking concrete initiatives for
disarmament, eroding our non-alignment, forcing us to contribute to world
militarization' (Svenska Fredsoch SkiljedomsForeningen, 1982). However,
I am afraid their view will remain that of a small minority for a long time
to come, at least in democratic countries - which does not mean the case
should not be made.
Per Hansen, a Norwegian colonel, in Galtung & Hansen 1981, has the
following eight points in his plan for the defense of Norway: make full use
of our terrain; dynamiting of roads to impede enemy progress; fortifications
and coastal artillery to stop the enemy; mobile forces to stop the enemy if
he passes fortifications; anti-aircraft guns to protect important areas and
mobile forces; supply security in the form of well stocked and decentralized
depots; preparation for guerrilla warfare in occupied territory in order to
tie down the enemy and maintain Norwegian jurisdiction; civilian resist-
ance to prevent the enemy from making use of Norwegian social resources
(pp. 19-20). As can be seen, again a case of convergence; all three elements
are present.
Spannocchi (1975) gives 'the Austrian answer' as Raumverteidigung,
defense in depth, in the Austrian space itself, as opposed to Vorneverteidigung,
defense at the border itself. One might note the psychological problem in the
Raumverteidigung: the population close to the border is so to speak given up
in the first round, possibly to occupying forces, while defense goes on else-
where. The question is, of course, what is worse: to be occupied or to be the
main battlefield because it is close to the border, under the age-old doctrine
of preserving every inch of one's own territory? Loser (1981) is in the same
tradition.
In Britain, the report of the Alternative Defense Commission (1983) is
the proposal that comes closest to what is proposed here. One very positive
point about the book is the concreteness of the policy proposals for Britain,
obviously destined for a Labour government. When are all countries in
Europe going to have such books/explorations published, and when are
they going to become major elements in the public debate? When, to put it
differently, is the peace movement going to take its task so seriously that
concrete alternatives are discussed, not only concrete criticism?
Particularly important in the German debate about defensive defense is
the book by Afheldt (1983) - with a number of technical explorations by
his cousin, Brigadegeneral Eckart Afheldt. In another important book in the
German debate (Komite fiir Grundrechte and Demokratie, 1981) five
options for alternatives to the present NATO policy are discussed: disen-
gagement, neutralism, strictly defensive defense, civilian defense, unilateral
and complete disarmament. What I miss in the book is a clearer way of
looking at these options as building blocks that can be combined in many
Ciait u r q Transarmament 423
ways, e.g. by combining nos. 3 and 4 (and also including guerrilla) in a

more total defense concept that does not provoke.
The Social Democratic Information Service also presents a number of alter-
natives, under alternative defense policies, arguing that such defensive defense
concepts as Raumverteidigung- -
and small, mobile units could and should
be combined, but does not argue the combination of this with social defense
(i.e. nonmilitary defense, although the concept also includes guerrilla defense)
(Sozialdemokratischer Informationsdienst 1982). However, some years ago
such concepts were totally unmentionable in such party documents - this type
of debate is now moving, and very rapidly.
Riiling (1978) makes the point that 'such a strategy of "inoffensive
deterrence" is the only strategy open to small states surrounded by might-
ier neighbors. Yugoslavia, with special respect to the Soviet Union, after the
events in 1968, adopted a policy of people's resistance against any invasion.
Next to its ordinary army, "the people" were organized to participate in the
resistance as a partisan-force which is forbidden to recognize any capitula-
tion, and which has to continue fighting after the defeat of the regular
Yugoslav forces. The prospect of continued fighting is supposed to have a
great deterrence effect' (p. 344). Roling also mentions that 'the Rumanian
law of 1972 is based on the same principle: territorial defense by the whole
population' (p. 347).
Hollins (1982) points out that 'a world in which the capacity for aggres-
sive warfare had been eliminated - and this could be accomplished by 1990 -
would by no means resolve all the world's critical problems. But it would be
a very different world from the one we have now' (p. 65). I would share that
view,. both parts of it.
Roberts (1982)makes the point that the whole debate about security has
been too much dominated by armamentldisarmament issues: 'The creation
of a national non-nuclear defense, and the various attempts to set limits to
international conflicts, deserve more place in the discussion than they get.
"Uni" versus "multi" is by no means the only issue worth debating' (p. 177).
A statement with which it is hard to disagree.
5. Defense and Common Security
Let us now have a closer look at this type of defense, a triad consisting of
conventional, paramilitary and non-military defense. The major arguments
in favor are obvious: a defense of this kind is not provocative since it can-
not be used for an attack, hence should not lead to any arms race. Within
this type of defense doctrine it would be entirely possible for both parties
to have not only a high level of security but also a relatively equal level of
security. In addition it is possible for them to cooperate: it would be in the
interest of either party to make the other party feel secure, which would
mean that there could even be an exchange between adversaries of tech-
niques of defensive defense (but not necessarily giving information about
424 T h e C o l d War a n d N u c l e a r D e t e r r e n c e
their exact location). This means that a setting is given for common security
(Galtung 1984, Section 3.4) and that is already something
With this approach there would still be arms and even armament as a
process, but with the distinct possibility that a stable plateau can be achieved,
in other words not only common security but a stable and common security.
Of course, in world history this mutually defensive posture has probably been
a normal state of affairs for most pairs of neighboring countries, the accu-
mulation of offensive arms and offensive arms races being an exception. But
in addition to this, a strong defensive defense should have a high deterrent
value, 'deterrence' then taken not in the sense of retaliation, but in the sense
of being able to stave off an attack. Nevertheless, should the attack come,
and that would be the third line of argument, then the level of destruction
would be lower since there would be no incentive (except for pure terrorism)
to use nuclear arms and other weapons of mass destruction. Not only the
defense system but also the social system itself (Galtung 1984, Section 5.4)
would be organized in such a way that no immediate target would present
itself as being worthy of a nuclear attack.
6. S o m e Critical Issues of Defensive Defense
This, however, does not mean that the type of defense advocated here is
unproblematic. The following is a short list of some basic and critical con-
siderations, for a debate that should now take place not only within the
peace movement but in our societies in general.
First, defensive defense presupposes a high level of national self-reliance
in defense matters. If weapon systems are not supposed to be quick, long
range, and mobile, then they cannot be transported from one country to
another in order to help that other country (rather than attacking it) either.
Under a doctrine of defensive defense military alliances based on high levels
of mobility are severely curtailed. This, of course, does not mean that there
cannot be all kinds of diplomatic, otherwise political, and economic sup-
port in case of an attack. World public opinion would still function, and
even more so than before because a country with purely defensive defense
cannot possibly be accused of having provoked an attacker. This may look
like a severe reduction of defensive capability, but could also be seen in
exactly the opposite manner. Clearly, a country which is used to relying on
allies, and particularly on a superpower ally, will not mobilize all its defense
resources. This is true in times of peace and even more so in times of war.
Military forces in a client country in an alliance, given the idea that 'I have
to fight for 24 hours till help comes from the superpower, possibly even
with superweapons', will certainly not exercise their defense potential to the
maximum. Rather, the strategy would be to put up a decent show but try-
ing to do so in such a way that national and personal honors are preserved,
yet one manages to survive till the major burden of the battle is taken over
by the superpower. A policy of national self-reliance would rule this out.
I Transarmament 425
If one really means what one says, that freedom is worth a fight, then that
fight has to be done by nobody else than oneself. The triad advocated above
(CMD, PMD, and MND) is so diverse, and on the other hand so dispersed
throughout the country, that it should serve exactly as a network capable of
mobilizing all kinds of defense potentials. Not only is it capable of mobil-
izing women, but also all or most of those who are dismissed from military
service for health reasons. Military 'defense' in today's offensive form is
simply incapable of mobilizing the population.
Second, a policy of defensive defense presupposes a high level of local
defense self-reliance. If the units are to be small, dispersed, and locally sup-
ported, very often also locally based, there has to be a high local capacity
to keep a fight going even if the national center has been rendered incapable
of doing so. Again the same reasoning applies: in a highly hierarchical
national defense system, itself possibly a replication of highly hierarchical
international defense systems, the local units might tend to wait for support
from the center and thus yield much less resistance than they otherwise
could do. If no such support is forthcoming they might give up, capitulate.
But if everything has been prepared in advance they might not only con-
tinue the struggle, but also, knowing that they have only themselves to rely
upon, d o more than otherwise could have been expected of them. Hence, it
is obvious that a policy of defensive defense presupposes not only a higher
level of national self-reliance, but also of local self-reliance. This type of
military doctrine, hence, is structurally compatible with a social structure
much more based on national and local self-reliance in general, just like the
vertical alliance pattern with hierarchical organization inside a country is
compatible with the social structure one finds, for instance, in transnational
organizations. Obviously the economic structure does not completely deter-
mine the military structure, but there is a relation between the two. A com-
plete change in defense structure would presuppose at least some change in
economic, political and social structure in general. It may be argued that
this is to ask for too much. It may also be argued that this type of change
probably will have to take place anyhow as a reaction to the general world
crisis, and that the change is not that fundamental.
Third, a defensive defense is vulnerable t o a n enemy w h o attacks the sys-
t e m with offensive arms from his o w n country. As a matter of fact, all the
adversary would have to do would be to set up a long range gun on his own
territory, capable of hitting targets in a systematic manner, and destroy
them from one end of the country to another. Aircraft would have to oper-
ate over the territory; a long range gun (or battery of missiles) would not.
Hence, it stands to reason that a defensive defense would have to be sup-
plemented by some element of interdiction capability. These are counterforce
weapons, for instance aircraft capable of hitting the gun just mentioned. And
then we are, of course, back to the problem; any interdiction capability would
also be an offensive capability, and hence could be provocative. Consequently
it is a question of having as little as possible, making weapons very counter-
force and not countervalue, building them into the military doctrine down to
the letter and verse of the instruction manuals, at all levels of the military
organizations, as interdiction weapons only. How much is necessary and how
much is sufficient is difficult to say; military experts close to the peace move-
ments would be the ideal persons to advise on this. Clearly, very soon an
offensive capability becomes too much, and becomes too provocative.
Fourth, a policy of defensive defense is not offensive against an outside
adversary, but could be highly offensive against an inside adversary. The
types of weapons that are described above as being defensive are defensive
because they cannot reach outside national borders in any significant man-
ner. But they can certainly hit inside those borders, otherwise they would
not have any capability at all. And they would not necessarily distinguish
between external and internal foes of the regime. As a matter of fact, they
are exactly the type of weapons that a repressive government might use
against insurgent forces, whether their claims are justifiable or not. They
are more adequate than offensive weapons: the case of the Iranian revolu-
tion showed rather convincingly how helpless the Shah was with his 'mod-
ern' weaponry designed for long range operations against a levee en masse
of the population. Clearly this is an important problem, and a typical ex-
ample of how a policy designed go solve one problem may not only not solve
another one but also aggravate it. The only solution I can imagine would
be to make the country less vulnerable, simply by reducing or even elimin-
ating major contradictions within the country (Galtung 1984, Section 5.4).
That would permit distributing the control over these means of destruction,
the weapons, in a more equitable manner in the population, not regarding
it as a total monopoly of the government. This does not necessarily mean
going so far as one does in Switzerland in the sense of people having arms
at home. That would be too similar to the US situation where a country-
wide dispersion of firearms has proved to increase the level of insecurity
considerably. What may not work in the US seems t o work in Switzerland
and Norway where militia arms are not used for private violence either.
Fifth, a defensive defense policy presupposes a higher level of readiness
for defense in the population. It clearly supposes a higher level of mobiliza-
tion: self-reliance at the national and local levels, and consensus. But this
does not necessarily mean militarization. I do not think that it can be said
a highly mobilized Norwegian population against nazi and quisling rule
was militarized because it wanted t o defend itself. Militarization would
have much more to do with excessive Military - Bureaucratic - Corporate -
Intelligentsia complexes, over-armament, offensive armament and such things.
Nor is it necessarily the case that this type of defense presupposes a con-
stant Feindbild. In times of peace a policy of non-alignment and even neu-
trality would serve to build down such Feindbild. In times of war it would
come about anyhow, only that the non-military component of defense
would try to see to it that it would be directed against the enemy as a soldier
and not the enemy as a person. But what is absolutely clear is that a defensive
policy, because it relies much more on popular participation, would pre-
suppose a high level of consensus. That, of course, has the major advantage
Galt ung Transarmament 427
that mobilization of the military potential cannot happen against the popular
will, as when forces are used offensively in total disregard and contempt of
what a population might feel, relying o n professional soldiery and a general
decoupling of military society from civilian society. International adventurism
would be impossible.
Sixth, a defensive defense with three different components presupposes
that the three components do not work at cross purposes. This is the famous
problem of the Mix between military and non-military types of defense.
Suffice it here only to say that the problem may be more important in theory
than in practice. In practice there are several possibilities. There is the Mix in
space: conventional defense along the borders and in thinly populated areas,
with PMD and N M D elsewhere. There is the Mix in time: conventional
defense first, then PMD and N M D as fall-back possibilities. There is the Mix
in what one might call functional space: conventional defense for geographic
and precise targets, PMD and N M D for more diffused and dispersed targets
such as the population as a whole, society as a whole, nature: then C M D and
PMD for more offensive purposes inside one's own territory, NMD for more
defensive purposes. There is the Mix of all these mixes - the question of
course being whether it becomes mixed-up. War time experience seems to
indicate that it does not, that the population is able to entertain different
types of defense at the same time, and that the adversary also makes a dis-
tinction between the three, perhaps behaving in the most aggressive way
against PMD, less so against C M D and much less so against NMD. It could
also be the other way round under certain circumstances. In any case, the task
remains that of making the country indigestible, like a hedgehog, not appe-
tizing to a wolf, and highly inoffensive to each other.
That, however, is a discussion that would lead far beyond the scope of
this article. Suffice it here only to say that the strength of this type of
defense is precisely its versatility, and that the enemy of it would be those
whose thinking has become so one-dimensional that they can only think in
terms of one foot of the tripod, not in terms of all three. In fact, the argu-
ment against it is probably not so much that it is ineffective as a deterrent
and as a defense. The argument might rather be the opposite; it is so effect-
ive that it could also be successfully turned against the countries' own gov-
ernments. In other words, the opposite of the argument above where the
emphasis was on conventional military defense as an instrument which in
the hands of the government could be used to crush a rebellion. Paramilitary
defense and non-military defense, meaningless unless they are in the hands
of the population itself, could also be used to topple a government. One
might say that this already constitutes a balance of power, that one type of
defensive defense may be the answer to the other in internal power struggles.
But the much better answer, of course, would be to say that a condition for
a purely defensive form of defense is that the country has come so far in rid-
ding itself of basic internal contradictions that neither a government, nor the
population, would use force in order to provoke some basic discontinuity in
the history of the country.
7. Conclusion: Towards a N e w Consensus?
In conclusion let us put this aspect of alternative security policies in per-

spective. Obviously, it is filled with contradictions, reflecting the fact that we
live in a perilous world, partly of our own making. Some of these contra-
dictions can only be softened, possibly overcome, if other aspects of alter-
native security policies are also enacted, very often referred to as 'political' -
in the present article in terms of 'non-alignment', 'inner strength' and 'outer
usefulness', dealt with elsewhere (Galtung 1984, Sections 3-5).
What should be emphasized here is only one point: the focus here is on
transarmament, not on disarmament. I have argued elsewhere that disarma-
ment of offensive forces is absolutely indispensable, to avoid major wars, and
that, the route via disarmament negotiations as practised so far is a blind alley.
The policy advocated here is a combination of disarmament and transarma-
ment, not the obviously fallacious policy of trying to obtain disarmament in
offensive weapons through armament in offensive weapons, but an effort to
obtain disarmament in offensive weapons through transarmament to defen-
sive weapons. Some of the latter are undoubtedly still highly violent. But
within the defensive defense concept argued here there are three different types
of defense: conventional, para-military and non-military. Over time this might
develop, if one can try to be optimistic in these troubled years, towards non-
military defense, which is more or less the way in which we handle conflicts
in civilized societies, with strikes, some civil disobedience, non-violent conflict
resolution mechanisms, and so on. Many would feel impatient, and ask why
not go straight to non-military defense, why not general and complete disarma-
ment and not this approach via conventional military defense! And my answer
would be along two lines: first, because the overwhelming majority of the pop-
ulation does not believe in non-military defense, only some of those who are
conscientious objectors, or pacifists in some other way, do. Second, because a
glance at world history in general and European history in particular should
convince anybody that we live in a dangerous world. Security does not come
automatically, there is a need for some kind of defense.
There is also a need for a new consensus in defense matters. Offensive
deterrence is only credible if the population in democratic countries believe in
it; they d o not. Non-defense is not believed in either. Defensive defense, with
the human right for everybody to choose the branch (conventional military,
para-military or non-military) which (s)he believes in, might be an answer.'
Author's N o t e
The ideas developed in this article should be seen in a broader context of alternative security
policies. My book on that topic, There Are Alternatives!, (Spokesman, Nottingham, 1984) from
which the present article is taken and references to current literature are added, is an effort to
develop a more comprehensive approach. This is important lest one IS led to believe that the
road out of the present highly dangerous situation is a question of new types of military hard-
ware only. I am indebted to Nils Perter Gleditsch and Jan Oberg for helpful editorial comments.
i 4,1i I !Iiic Transarmament 429
Notes
I. The experience in Norway during the 1940-45 occupatloll can be summarized as fol-
lows: c ~ v i l ~ a nnonviolent
, resistance was very effective agalnst n,izification of civilian soclety
hy the Quisling regrme, hut ineffect~veagainst the German occupation - a s was also the mili-
tary reslstmce. Militxy liberation from the outside was both necessary and sufficient.
2. O f course, the idea o f defensive defense is not a t all new, and played a consider.ihle role
111 the debates and cornn~iss~ons o f the L e a g ~ ~ofe N a t ~ o n sin the 1930s. See, for instance,
Gr~ffin( 1936).
.3. 'The O f f ~ c ~Swiss
al p u h l ~ c . ~ t ~(Ze~itralstelle
on fur Gesanirverteid~gung19731, is very clear
o n this point 'The army as ,I whole has a defensive rnissron and its preparations are made with
a view towards fighting o n l ) ~within its o w n territory' (p. 28).
The document also rnentlons the possibility of non-military defense as a last resort, hut
does not take up the sticky issue of whether this could be one way of making use of the defen-
sive potenti'il of conscientious oblectors.
4. Again, the SWISSmake thc point: '... we have t o guarantee a minrntol de,qiw o f self-
sttfficicnq w ~ t hregard t o armaments. Sufficient supplies must be kept in order t o malnt;iln the
combat-effectiveness of our army in case of war' (ihid. p. 29).
5.For a general review of the effects of nuclear war, see Galtung (19841, particularly sec-
tions 2.4, 3.4. and 4.4 For the effects of nuclear war on the climate, see Sagan (198,3184);~ l s o
N PS C ~ I P I I ~3I Novemher
S~, 1983.
6. See lohan Galtung (1978), chapters 12, 13, 15 for an analysis of nonm~litarydefense,
also Galtung ( 1980), sectlon 4.4.
7. @berg (1983) p. 167 points out how 'nonviolence as well ,is the philosophy of defensive
defense ha\ its origin In the Orient, not in our culture'. I agree, hut one should not underestimate
the importmce of defensive defense as the defense of the weaker parties, and the parties with les\
aggressive inclinations even if there is n o trace of Oriental thmking - like in the case of the Swiss.
References
Afheldt, Horst 1983. Uefewsicw Verteidigung. Keinhek be1 H a m h ~ t r g Rowohlt.

:
Alternative Defense Commission 1983. Defense without the Llornl~.I.ondo11 a n d New York:
1Bylor & Franc~s.
Fischer, Dietrich 1982. 'Inv~~lnerahilityw ~ t h o u t Threat: The SWISS( h n c e p t of C;encr,ll
Iktense'. ]oitrnizl of Prare Research, vol. 19, no. 3, pp. 205-225.
G ~ l t u n gJohan
, 1978. tssays in Peace R~~search,vol. 11. Copenhagen: Ejlers.
Galrung. Joh'ln 1980. The True Worlds. New York: The Free Press/Macmillan.
Giltung, Johan 1982. t.rtt~irovintrnt,Dc~wlopmnrtand Mrlitnry Ar-tiz~rtyOslo: Univers~trtsforhget.
Galrung, Johan 1983. 'Alternative Defense Strateg~es', W R I (War Resisters' Internat~onal)
Neir~sletter,April, pp. 5-7.
Galtung, J o h a n 1984. Es giht Alternative! Wieshaden: Westdeutscher Verlag. Ed~tions111
Engl~sh,Dutch, Spanish Spring 1984.
(;,lltung, Johan & Per Hansen 198 1. Totirlfors~w.Oslo: Univers~tetsforlaget.
Griffin 1. 1936. Altrrnnti~~c, to Keurtnament. London: Macmillan.
Hollins, Harry B. 1982. 'A Defensive Weapon System', 17 T / JBztlletin
~ of Atontic S c ~ ~ n t ~ vo1.
s t s , 38,
no. 6, pp. 63-65.
Komitee t i ~ rGrundrechte und Deniokratie 1981. Friedrn mrt ~znticrctr W~lffen.Re~nhckbei
Hamhurg: Rowohlr.
L.ticer, Jochen 198 1. 1 V d v rot norh tot. Miinchen: Olzog.
Roberts, Adam 1982. 'The Alternative, Non-nuclear Way t o Defend Ourselves', Nui, Society,
4 February, \wl. 59, no. 1003, pp. 17.5-177.
Roling, Bert 1978. 'Fexihility of Inoffensive Deterrence', Bulletin of Peizcc l'roposals. vol. 9,
n o . 4, pp. 339-347.
Sagan, Carl 1983184. 'Nuclear War and Climatic Catastrophe: Some Policy Implications,
Foreign Affairs, vol. 62, no. 2, pp. 257-292.
Sozialdemokratischer Informationsdienst 1982. 'Argumente fiir eine alternative
Verteidigungspolitik', Frau und Geseelschaft, September.
Spannocchi, Emil 1976. Verteidigung ohne Selbstzer storung. MiinchedWien: Hanser.
Svenska Fredsoch Skiljedomsforeningen 1982. Ett forsvarspolitiskt program for fred. Stockholm.
Totalforsvarets Upplysningsnemnd (no date). Sveriges Sakerhet. Stockholm.
Zentralstelle fiir Gesamtverteidigung 1973. General Defense. Bern.
Oberg, Jan 1983. At udvikle sikkerhed og sikre udvikling. Copenhagen: Vindrose.

(Professor Barry Buzan, Professor Lene Hansen) Int PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Professor Barry Buzan, Professor Lene Hansen) Int PDF

Uploaded by

Copyright:

Available Formats

SAGE L ~ B ~ A ROF---

Barry Buzan and Lene Hansen

First published 2007

Apart from any fair dealing for the purposes of research or

SAGE Publications Ltd

SAGE Publications Inc.

SAGE Publications India Pvt I,td

SAGE Publications Asia-Pacific Pte Ltd

British Library Cataloguing in Publication data

A catalogue record for this book is available

ISBN: 978-1-4129-2139-8 (set of four volumes)

Library of Congress Control Number: 2006938798

Typeset by Televijay Technologies (P) Limited, Chennai

19. The OffensiveIDefensive Balance of Military Technology:

22. Economic Structure and International Security: The Limits of

39. The Emerging Structure o f International Politics

57. Revisiting Copenhagen: Or, on the Creative Development of a

Grateful acknowledgement is made to the following sources for permis

1. 'The Sources of Sov~etConduct', X

9. 'Cooperation Under the Security Dilemma', Robert Jeruis

18. 'Nuclear War and Climatic Catastrophe: Some Policy Implications',

27. 'The Case Against Linking Environmental Degradation and National

O 1993 by the President and Fellows of Harvard College and the

0 The Johns Hopkins University Press. Reprinted with permission of

65. 'Human Security: Paradigm Shift or Hot Air?', Roland Paris

I. The Origins a n d Dynamics of International Security Studies

The Birth of International Security Studies

Second, while military capabilities were crucial to assessing national

T h e Driving Forces of International Security S t u d i e s

Although they were to make a comeback as the explicit conceptual debates

Great Power Politics

Great power politics subdivides into three elements: the distribution of

T h e internal Dynamic of Academic Debates

An idealized model of how academic knowledge is created would p e d i c t

economics or interpretative epistemologies from the humanities (sociology,

To identify institutionalization as a driving force is to further highlight that

for research. Successful graduate programs are characterized by the ability

11. T h e Evolution of International Security S t u d i e s

Developments in Soviet capabilities and doctrine was the overwhelming

Cold War Challengers to Mainstream Strategic Studies

for a revision of military-state-centric security. Barry Buzan's People, States

particular resolution to the questions of authority and security, but it was

underpinned Western thought and politics from classical Greece onwards,

After the Cold War: Traditional versus Wider Security

Agendas and Understandings

state discourse. This move from Conventional to Critical Constructivism

referent object usually located at the 'middle scale of limited collectivities',

predominantly as part of a CSS agenda or in an attack on neo-liberal eco-

terrorist threat and in particular in granting the US a privileged position in

~ ~ , M. and Ling, L.H.M. (2004) 'Power, Borders, Security, Wealth: Lessons

Recommended Books on ISS

T he political personality of Soviet power as we know it today is the

organized Socialist production at home, would rise against the remaining

So much for the historical background. What does it spell in t e r m of the

power by announcing that decisions of the Politburo were being taken

This is, of course, outstandingly the problem of the personal position of

mass of human beings among whom no independent organizational structure

Source: I'olrtlc-'11 Scicnce Q~rczrterly,I.XVII(4) (1952):48 1-502.

Today, the alternative t o a policy of the national interest to which people

order t o be either expedient - a rational means toward a n accepted end - o r

Hitler's wanton attack on his neighbors be allowed to qualify as such -

"regions of special interest". It is a well-known and portentous phenom-

Traditionally, the protection and preservation of national core values