DILEMMAS OF NUCLEAR STRATEGY
DILEMMAS OF NUCLEAR STRATEGY Edited by
ROMAN KOLKOWICZ Professor of Political Science U...
121 downloads
1601 Views
861KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
DILEMMAS OF NUCLEAR STRATEGY
DILEMMAS OF NUCLEAR STRATEGY Edited by
ROMAN KOLKOWICZ Professor of Political Science University of California Los Angeles
FRANK CASS
First published 1987 in Great Britain by FRANK CASS AND COMPANY LIMITED Gainsborough House, 11 Gainsborough Road, London E11 1RS, England This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” and in the United States of America by FRANK CASS AND COMPANY LIMITED c/o Biblio Distribution Centre 81 Adams Drive, P.O. Box 327, Totowa, NJ 07511 Copyright © 1987 Frank Cass & Co. Ltd British Library Cataloguing in Publication Data Dilemmas of nuclear strategy. 1. Nuclear warfare 2. Deterrence (Strategy) I. Kolkowicz, Roman II. Journal of strategic studies 355′.0217 U263 ISBN 0-203-98857-4 Master e-book ISBN
ISBN 0-7146-3236-8 (Print Edition) Library of Congress Cataloging-in-Publication Data Dilemmas of nuclear strategy “First appeared in a special issue… Journal of strategic studies, vol. 9, no. 4, published by Frank Cass”—Verso t.p. 1. Nuclear warfare. I. Kolkowicz, Roman. U263.D53 1987 355′.0217 87–6394 ISBN 0-7146-3236-8 This group of studies first appeared in a Special Issue on Dilemmas of Nuclear Strategy, The Journal of Strategic Studies Vol. 9, No. 4 published by Frank Cass & Co. Ltd. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form, or by any means, electronic, mechanical, photocopying, recording, or other wise, without the prior permission of Frank Cass and Company Limited.
Contents
Notes on Contributors
vi
I. INTRODUCTION The Rise and Decline of Deterrence Doctrine Roman Kolkowicz
3
II. NUCLEAR DETERRENCE: CRITICAL ASSESSMENTS The Not-Quite-Absolute Weapon: Deterrence and the Legacy of Bernard Brodie Gregg Herken
15
Deterrence: The Problem—Not the Solution Michael MccGwire
25
Nuclear Strategy: Can There Be A Happy Ending? Fred Charles Ikle
43
III. EUROPEAN PERSPECTIVES Nuclear Deterrence and European Security: Towards A Not So Happy Ending? Pierre Lellouche The Reception of American Deterrence Theory in the Federal Republic of Germany and the German ‘Nuclear Debate’ of the 1950s Wolfgang Heisenberg After NATO’s Dual Track Decision of 1979: Where Do We Go From Here? Christian Hacke
57
69
83
IV: ARMS CONTROL AND NUCLEAR STRATEGY Deterrence and Arms Control George Rathjens
103
v
Arms Races and Instability Michael Intriligator and Dagobert Brito
113
Notes on Contributors
Roman Kolkowicz is a Professor of Political Science at the University of California, Los Angeles, where he was the Founding Director of the Center for International and Strategic Affairs, and is currently Director of the Project on Politics and War. Gregg Herken is a Professor of History at the University of California, Santa Cruz, where he is affiliated with the University of California Institute on Global Conflict and Cooperation. Michael MccGwire is a Research Staff Member of the Brookings Institution, Washington, DC. Fred Charles Ikle is US Under-Secretary of Defense for Policy, and former Director of the US Arms Control and Disarmament Agency. Pierre Lellouche is Associate Director of the Institut des Rélations Internationale in Paris, France. Wolfgang Heisenberg is the former Director of the Fritz Thyssen Stiftung in Bonn, West Germany. Christian Hacke is a Professor of International Politics at the Universität der Bundeswehr Hamburg, Institute for International Politics in Hamburg, West Germany. George Rathjens is a Professor of Political Science at the Massachusetts Institute of Technology. Michael Intriligator is a Professor of Economics and Political Science at the University of California, Los Angeles, where he is Director of the Center for International and Strategic Affairs. Dagobert Brito is a Professor of Economics at Tulane University.
I. Introduction
2
The Rise and Decline of Deterrence Doctrine Roman Kolkowicz
The massive proliferation of nuclear weapons in the United States, the Soviet Union, and elsewhere has radically challenged traditional approaches to war and strategy and the uses of military force for rational purposes. Efforts to create an acceptable and viable strategy for the management of this dangerous military power in an adversarial superpower relationship have been anything but easy and reassuring. In the past two to three decades, however, after an initial period of adaptation and accommodation to the radical implications of nuclear weapons, a consensus has evolved among Western academic and governmental experts on a theory for the management of national security, for the furtherance of international stability, and for the avoidance of nuclear war. This consensus on strategic deterrence policy has centered since the 1960s on the doctrine of Mutual Assured Destruction (MAD), which is characterized by an emphasis on rational decision-making by top national leaders who are entrusted with the management of national security in a very dangerous world. In this rational, unitary-actor conceptualization of strategic decision-making, nuclear capabilities are seen to have vital deterring, dissuasive, and diplomatic-political utilities, while their actual military employment is considered irrational and highly undesirable, since both sides are assumed to suffer certain and devastating destruction in the event of a first, or retaliatory, nuclear strike by roughly equal adversaries. Thus, the idea of victory, or of defeat of the adversary, was no longer thought to be a rational objective: nuclear weapons were to serve a single purpose, to deter an attack by one superpower against the other. Mutual deterrence based on mutual vulnerability became the foundation of modern nuclear strategy. The consensus on a doctrine of MAD has in recent years been challenged by experts in and outside the government. Some of these critics argue that the Mutual Assured Destruction doctrine of deterrence is no longer viable; they find its inherent aspects of ‘consensual vulnerability’1 to be immoral, since they imply that in the name of strategic stability ‘killing people is good’ (countervalue targeting required by the logic of assured retaliation in MAD) while ‘killing weapons is bad’ (counterforce targeting which was considered to undermine strategic stability and to be indicative of first-strike or pre-emptive intentions). Others point at the so-called inherent aggressiveness of Soviet leaders and their continued arms build-ups, seemingly unconstrained by the imperatives of Mutual Assured Destruction doctrine and by the requirements and logic of deterrence stability. Still others maintain that in order to make our deterrence policy more credible to the Soviet Union, we need to go beyond
4 DILEMMAS OF NUCLEAR STRATEGY
the ‘passive’ deterring position of Mutual Assured Destruction, actually to prepare for warfighting and to consider adopting ‘prevailing’ and ‘countervailing’ strategies of protracted nuclear war. Clearly the ‘golden age’ of deterrence has passed, a brief period of intellectual ferment and political credibility nostalgically located sometime in the mid-1950s/ mid-1960s decade, when the promise of harnessing and controlling the nuclear Moloch was high. Through a combination of deterrence, arms control and crisis management, the advocates of the new science of war had promised to safeguard national security, to provide a stable international environment, and to develop a rational decision-making process for the management of national interests in a hostile nuclear world: ‘In 1961, the promise was high’, observed a noted strategic analyst. The civilian strategists came to Washington to assume influential roles’ in the development of strategic and foreign policies in the Kennedy-McNamara team. However, by the 1970s, he observed, ‘It is fair to say that their performance had not lived up to their promise. And that’s putting it mildly.’2 It is a commonplace by now that the erstwhile promises and pretensions of the nuclear ‘whiz-kids’ and the ‘wizards of Armageddon’ have not lived up to expectations. Arms control has not curbed or slowed down the vast arms build-ups; crisis management has shown itself to be an imperfect and fallible policy instrument; and the centerpiece of this nuclear strategic science, the doctrine of Mutual Assured Destruction, is under assault from all quarters. Critics maintain that while the political and technological context of international politics and strategic relations has changed profoundly, our thinking about war and its prevention remains rooted in the past, ossified around a deterrence paradigm of over two decades ago. Among the reasons cited for this ‘ossification’ of the strategic theory are:
– The persistent dominance of economic models in the study of strategy: the assumption that international conflict can be analyzed in terms of rational economic/strategic decision-makers was useful in the development of strategic theory, but it has proven fatal to the relevance of theorists who have shifted from model-building to policy-prescription. – The transition from strategic ‘prophet to courtier’: virtually ‘all the basic ideas and philosophies about nuclear weapons and their use have been generated by a few civilians working quite independently of the military’. 3 These strategic theorists entered the corridors of power, and ‘the United States has been living off its strategic theoretical capital ever since’.4 – The bureaucratization of strategic science: the attempt to free strategy from the fetters of excessive controls and influence by military/bureaucratic patrons, and thus to encourage participation by academics in the development and improvement of strategy, led to the spawning of many think tanks. However, think-tanks ‘as exemplified by the Rand Corporation, began to manifest dual loyalty’ and, according to its critics, tended to produce ‘both irrelevant policy advice and poor scholarship’.5 It is with this in mind that I invited a group of distinguished American and European scholars and specialists in strategic and international studies to a
THE RISE AND DECLINE OF DETERRENCE DOCTRINE 5
conference in December 1985 at the Villa Serbelloni, in Bellagio, Italy, and asked them to consider the following questions:
– What is the historical validity, theoretical vitality, and policy relevance of nuclear deterrence theories and doctrines? – In what ways have technological and policy changes in recent decades affected the original concepts of nuclear war and deterrence strategy? – Are there alternate ways for thinking about strategy in the contemporary nuclear context that might be more useful? While the preparations for the Bellagio conference of scholars and specialists were underway, I was invited by former Presidents Jimmy Carter and Gerald Ford to assist them in an analogous enterprise concerned with a reexamination of the relevance of strategic deterrence and arms control policies in the management of superpower relations. Several international conferences, chaired by the two presidents, took place in 1984–85. These conferences, whose participants were mostly high public officials and political and military leaders from several administrations, as well as from the Soviet Union, China, and other countries, provided an excellent complementary perspective on these complex issues. It is noteworthy that the general conclusions of these conferences concerning the state of superpower political-strategic relations and the roles of existing strategic doctrines were strikingly similar to the conclusions reached by the participants of the Bellagio conference.6 It is fair to say that the guiding intellectual spirit of the Bellagio conference and its proceedings was that of Bernard Brodie, to whom this volume is dedicated. Brodie was a pioneer of modern strategic studies in the nuclear era whose work has powerfully influenced generations of strategists and decision-makers. He was the first to perceive and publicly articulate the revolutionary implications of nuclear weapons for war and peace and for the management of international politics in a dangerous world. At the very dawn of the nuclear era, in 1946, he defined the essential paradox and inescapable truth of the nuclear condition: ‘Thus far, the chief purpose of our military establishment has been to win wars. From now on, its chief purpose must be to avert them. It can have no other useful purpose.’7 In his lifelong concern for the intellectual integrity and policy relevance of strategic studies in the nuclear era, Brodie insisted that ‘we need people who will challenge and dissect the prevailing dogmas’ of strategic fashions and fads. Above all, reflecting the teachings of Clausewitz whom he greatly admired, Brodie urged the ‘need to stress the superior importance of the political side of strategy to the simply technical and technological’. 8 He considered ‘the most important single idea in [Clausewitz’s] On War…the one that makes it great, is the idea that war must never be an act of blind violence, but must be dedicated to achieving the supreme goal of statecraft’.9 He found Clausewitz’s metaphor ‘War has its own language, but not its own logic’ to be the ‘single most important idea in all strategy’ because the use of military force must at all times be determined by political decision and rational purpose. He asserted therefore the need to be aware of the ‘inevitable limitations and imperfections of scientific method in strategic analysis and decisionmaking’ and to understand that ‘the most basic issues of strategy often do not lend themselves to scientific analysis…because they are laden with value judgements and therefore tend
6 DILEMMAS OF NUCLEAR STRATEGY
to escape any kind of disciplined thought’. In other words, ‘war would only be senseless destruction if it were not in pursuit of some valid political objective’.10 Brodie, like Clausewitz before him, was concerned lest strategy became an abstract, excessively theoretical science of war and thus became detached from its guiding and constraining political context. Above all, Brodie asserted, before leaders consider the use of military force, they must ask ‘De quoi s’agit-il?’ Can nuclear weapons still be considered militarily usable and politically rational? He asked, ‘Should we abandon deterrence strategies in favor of war-winning strategies? But what does this mean?’11 The papers presented and discussed at Bellagio addressed these central issues from a wide range of perspectives. The discussion pointed out the resistance after the Second World War to Brodie’s revolutionary formulation of the nuclear condition and his advocacy of a deterrence-only, non-military use of nuclear weapons. Brodie’s views, which rejected the idea of meaningful victory in a nuclear confrontation once the Soviet Union obtained rough parity with the West, were then challenged by influential ‘insiders’ like William L.Borden and Paul Nitze. Even though they conceded that nuclear war would cause enormous devastation, and while they understood the potential deterrence value of these weapons, they nevertheless considered the use of such weapons thinkable and rational. Borden’s and Nitze’s positions, reflecting at that time the views of the political and military mainstream in the United States, advocated an American strategy that was similar to that of the prenuclear era, and considered victory rational and obtainable.12 For Brodie and others, the idea of actually using these weapons was not thinkable and he argued therefore that military victory in nuclear war was not possible. These views were accepted in the 1960s by the official defense community and decision-makers and became the foundation for American strategic policies. Robert McNamara expressed this acceptance of the idea that there can be no meaningful victory in nuclear war and advocated a deterrence-only use of these lethal weapons:
Having spent seven years as Secretary of Defense…I do not believe we can avoid serious and unacceptable risk of nuclear war until we recognize …that nuclear weapons serve no military purpose whatsoever. They are totally useless—except only to deter one’s opponent from using them. 13 By the 1970s, however, there were noticeable changes in American official and expert communities concerned with strategy and specifically, with the credibility of Mutual Assured Destruction doctrines for the management of superpower relations. These concerns were motivated by several developments: a steady and worrisome Soviet arms build-up that not only began to catch up with the United States, but which threatened to go beyond parity and the assumed requirements of strategic stability; a perception of renewed Soviet aggressiveness and military pressures in the various parts of the Third World; and a Soviet unwillingness to establish ‘linkage’ between superpower normalization of relations and accommodations as evidenced by successful arms control agreements and détente and their own aggressive policies elsewhere. There was a growing perception in the United States that the existing forces and strategies under the assured destruction doctrines were becoming inadequate to protect US interests and to safeguard its security in the face of a massively expanding Soviet military capability.
THE RISE AND DECLINE OF DETERRENCE DOCTRINE 7
Whatever the specific reasons, the official American departure from the policies of Mutual Assured Destruction was first signalled by then Secretary of Defense James Schlesinger in 1974, and was continued as ‘countervailing strategy’ under Presidents Carter and Reagan. US strategy became ‘heavily oriented to fighting a nuclear war, even a protracted nuclear war’ and the expanding list of strike targets, signalling this strategic change, now included not only a minimum of 200 major Soviet cities, but also enemy missiles in hardened silos. The Reagan administration further deemphasized assured destruction and placed greater emphasis on providing nuclear options during a protracted war. Under these new guidelines, US nuclear capabilities ‘must prevail even under the conditions of a prolonged war’, and must be able to ‘force the Soviet Union to seek earliest termination of hostilities on terms favorable to the United States’. The pendulum of American official and informed public opinion has thus swung toward the prospect of fighting a nuclear war, connecting US strategy with the earlier idea of winning a war in the traditional sense of the term. 14 What about the Soviet side of this strategic equation? Do the Soviet leaders accept and follow the logic and policies of Mutual Assured Destruction or do they pursue some other strategic doctrines and policies? Since the very core of deterrence and mutual assured destruction rests on the mutuality of adversarial perceptions and consequent behavior, and on their acceptance of the ‘rules of the game’ of deterrence, the question of Soviet position becomes critical. Given the protracted antagonism and distrust of the superpowers towards each other, only infrequently interrupted by brief spells of relaxation and détente, can we assume a Soviet willingness to abide by Western rules of the game? During the Cold War period there was little question in the West regarding Soviet aggressive behavior and intent, and Soviet belligerent declaratory policy and threatening behavior culminating in the Cuban missile crisis did little to change these Western views. The strategic community in the West broadly agreed that Soviet strategic doctrine and policy remained highly aggressive and offensive and that included provisions for pre-emptive nuclear strikes under certain conditions. Moreover, many Western strategists and public figures were strongly persuaded that the Soviet military and political leaders believed that nuclear war, ‘if the imperialists initiate it’ was still fightable and winnable, and that their strategic planning was based in these beliefs, regardless of their benign declaratory policy which supported arms control and détente. Recently, however, some Western analysts began to reexamine these assumptions about the Soviets and concluded that the evidence suggests Soviet acceptance of the doctrine of Mutual Assured Destruction and the rejection of the possibility of victory in a nuclear confrontation. Foremost among these ‘revisionists’ is Raymond Garthoff, although these views still represent a distinct minority in the US. Most specialists maintain, however, that while Soviet declaratory policy has in recent years been following the so-called ‘Tula Line’ announced in a speech by Brezhnev in 1977, actual Soviet military planning, weapons acquisition, and strategic doctrine do not reflect a significant change in Soviet behavior. They maintain that the accommodational ‘Tula Line’ which disavows Soviet preemption, first use, and possibility of victory, and which accepts the principle of strategic ‘equality’, is merely a cynical and self-serving Soviet manipulation of Western public opinion. In actuality, they conclude that Soviet strategic policy continues unchanged in the old aggressive, offensive, and preemptive modes. Official US defense and strategic policies clearly reflect such
8 DILEMMAS OF NUCLEAR STRATEGY
perceptions of unchanged Soviet strategic continuity despite the Tula Line, and plan accordingly, as evidenced by the massive arms build-up initiated under the Carter administration and accelerated under the Reagan leadership. Now, if deterrence and mutual assured destruction are considered inadequate to contain Soviet aggression and for the protection of our national security, how useful and credible are some of the proposals for change? How does one respond to persistent calls for a ‘renaissance’ in strategic thinking now that the ‘golden age’ of deterrence is gone and its legacy is considered by many to be largely irrelevant for dealing with the changed strategic-political circumstances of superpower relations? Clearly, the most dramatic proposals for changing the nuclear deterrence stalemate and going beyond the ‘immoral’ aspects of Mutual Assured Destruction are contained in the ‘Star Wars’ proposals of President Reagan. The announced Strategic Defense Initiative, which is to turn us away from a passive deterrence-only policy in which the total vulnerability of the American (and Soviet) population is to ensure their total security, and towards a policy of actively defending one’s country, fired imaginations, but did not meet with universal approval and acceptance. While Reagan administration spokesmen find SDI to be a ‘more morally and intellectually satisfying’ strategy than MAD because the former would ‘protect the populations rather than avenge them’,15 others express ‘serious reservations, to put it mildly about SDI, on technological, military-strategic and political grounds’.16 Secretary Schlesinger describes himself as ‘a skeptic’ and remains ‘highly doubtful about the Strategic Defense Initiative—its pace, its prospective military utility, its costs, and particularly, its more ethereal versions of strategic defense’.17 However, most people, including skeptics of SDI, tend to agree with its advocates on a key point, and that is that despite the caveats, ‘we can’t get the SDI genie back into the bottle’ and must therefore proceed with our SDI research program, particularly since the Soviets themselves have been spending large amounts of resources on strategic defense research programs. The enormous complications likely to result from partial or full deployment of such technologies in a defense/offense strategic posture are disturbing and no satisfactory proposals or solutions for overcoming these complications have thus far been provided by SDI advocates (save the less than credible suggestion that the US share SDI technology with the Soviet Union). The papers and discussions at the Bellagio conference clustered around several rather well-defined positions concerning the soundness and desirability of deterrence theory and its Mutual Assured Destruction policies:
– First, that there is no escape from the decades old strategy and policy of deterrence based on offense-domination inherent in the Mutual Assured Destruction doctrine. Technological innovations like Star Wars are in the nature of fools gold, raising false promises of stability and survivability while in reality likely to introduce dangerous destabilizing factors into superpower strategic and political relations. Deterrence seems to have worked for several decades, and while it is not an ideal solution for the achievement of international stability and national security, it is still the only viable and coherent strategy, given the dynamics and realities of the superpower
THE RISE AND DECLINE OF DETERRENCE DOCTRINE 9
adversary relationship. Deterrence must be made stronger, more stable, more cooperative, and less threatening to respective adversaries.18 – Second, that deterrence does not work well, that it primarily constrains the US while barely constraining the Soviet Union who reject the ideas of selfconstraint and self-deterrence implicit in the American practice of deterrence. Moreover, the offense-dominated deterrence of Mutual Assured Destruction is immoral and essentially not even a strategy in the historical sense, since it stops working when it would be most urgently needed, at the outbreak of hostilities. Star Wars is the most useful way at present for breaking out of this nuclear stalemate, since it would provide for mutual survivability rather than mutual insecurity and assured incineration; it would also reduce the offensive weapons inventories and emphasize defense. In any event, the SDI genie is out of the bottle and it is therefore necessary to proceed with the research and development programs while at the same time offering the Soviets incentives to negotiate and cooperate, to reduce offensive weapons, vulnerability, and threats, and possibly to de-emphasize strategic pre-emptive doctrines which are highly threatening to mutual stability.19 – Third, that deterrence does not work at all, it merely legitimates and encourages deterrence dogmas and ideologies which in effect demand ever higher levels of arms development and deployments in the name of vaunted strategic stability. Deterrence ideologies of Mutual Assured Destruction generate arms races and confrontational moods, without providing additional security or stability. Neither technological fixes nor doctrinal conceptual/ methodological ‘innovation’ or ‘fine-tuning’ seriously address the fundamental problem of nuclear vulnerability and insecurity, and the most sensible and rational way out of the present dead-end of deterrence is through political and diplomatic means.20 – A fourth approach to deterrence theories and policies focuses primarily on the internal dynamics of the respective nuclear powers rather than on the bi-polar strategic and political causative factors from the international environment. 21 In searching for explanations for the contemporary strategic and political stalemate, whereby both sides are arming feverishly without evidence of a coherent and rational strategic vision or concept, these analysts looked away from the international arena and instead examined certain factors within the respective superpower systems for a better understanding of what drives these processes. Eschewing the usual explanations of external threat as the central motivator of strategy and politics, they assert that national security decision-making, at least in the pluralistic open societies of the West, is not a depoliticized process left largely to strategic experts, and that the very nature of deterrence doctrines and their political implications remain contested, and, given this reality, politics will determine defense and strategic policies as much as objective analysis and prescriptions of experts. Others argue that the main motive force driving US and Soviet strategic theories and policies stems less from the ‘real world’ of superpower confrontations, than from their different historical, political, and socio-cultural traditions, and in the Soviet
10 DILEMMAS OF NUCLEAR STRATEGY
case, as having been powerfully influenced by their unique, war-torn history, traditions of fatalism, and ‘kto-kovo’ distrust of claims to political rationality and benign motives of adversaries in conflictual situations.22 Thus, decisions rooted in domestic or even partisan politics have a powerful impact on the international arena, and in the process, set into motion radical changes in the heretofore widely accepted and fundamental premises and policies of nuclear strategy and stability among the superpowers. A distinct perspective on the issue of superpower strategic relations in general, and the issues of MAD and SDI in particular, was introduced by the European scholars and experts at Bellagio. Their views were shaped by concerns on how these broader issues might affect their own national security interests, and on the whole, they were very critical of the perceived American abandonment of the strategies of Mutual Assured Destruction and its extended deterrence provisions. They were also critical of the promises and expectations contained in President Reagan’s Star Wars proposals, describing them as a ‘mirage of technological solutions which do not exist yet, and may very well never exist’.23 They were very concerned with the possibility that with future strategic defenses on both sides, Europe might become decoupled from the superpowers’ nuclear equation, thus leading to the possibility of the ‘traditional nightmare of a limited conventional or tactical nuclear war fought on [European] soil’. Moreover, future SDI programs and the American strategic decoupling from Europe ‘might lead to a “fortress-America mentality”, with an unprotected and unprotectable glacis in Western Europe’.24 The evident complexity of the strategic nuclear problems in a world of massively proliferating weapons and military technologies had a sobering effect on the discussions at Bellagio. A sense of pessimism at times pervaded the deliberations of the participants who searched for ways to subordinate these proliferating weapons to rational and coherent political and strategic purposes. In this, the Bellagio conference reflected the pessimism of the deliberations among political and military leaders at the Carter/Ford conferences, where the record of Soviet-American arms control negotiations was found to be disappointing, where the promise of technical fixes of Star Wars was thought to be a chimera, and where efforts to conciliate SovietAmerican differences on certain issues were met with the usual political parochialism and grandstanding, effectively each side blaming the other for the troubles of the world.25 The final session of the Bellagio conference considered the future of deterrence: is it likely to last out this century, or beyond? Most strategists have generally ignored this question and left it largely to anti-nuclear peace activists, whose wishful or apocalyptic approaches led to their easy dismissal, and thus reinforced the strategists’ tendency to avoid the subject. This has led to a theoretical impoverishment of the field and to excessive concentration on rational models of deterrence which tend to be static and ahistorical. In the process, strategists paid insufficient attention to situations in which rational choice becomes virtually meaningless, due to constraints brought about by unanticipated developments and circumstances, or factors other than those accounted for by rational choice models of deterrence.26 The views and positions expressed at the Bellagio conference were arrayed along a wide strategic and political continuum and reflected the uncertainties and absence of a strategic consensus of the ‘real world’, so unlike the ‘golden age’ of strategic certitude and optimism of an earlier era.
THE RISE AND DECLINE OF DETERRENCE DOCTRINE 11
In the final analysis, the deliberations of the experts at Bellagio did not yield any easy or quick solutions to the nuclear dilemma. While technological fixes or methodological innovations of crisis management scenarios were largely found to be of limited relevance, there was a sense that political solutions may contain the promise to lead us, if not into the promised land of eternal peace and absence of war, then at least into peaceful coexistence on this dangerous planet. Thus coming around full circle, we may yet wish to agree with Bernard Brodie who asserted that ‘it is really marvelous how modern is the ring of what Clausewitz had to say on the subject’ of the need for politics to guide and control the arms and strategies of war.27
NOTES Several foundations and institutions generously supported this conference. We are grateful to the Rockefeller Foundation for providing the splendid Villa Serbelloni for the use of the conference; the Alfred Sloan foundation, the Fritz Thyssen Stiftung, the Carter Center at Emory University, the Institute on Global Conflict and Cooperation of the University of California, and the Center for International and Strategic Affairs at UCLA have generously supported this endeavor. Special recognition is due Donna Beltz, the Conference Coordinator, who skillfully managed conference administration and publications. In addition to this special journal issue, papers from the conference will be published in two volumes: one to be published by Frank Cass & Co. Ltd., and a second, distinct, volume to be published by Allen & Unwin, Inc. 1. See Fred C.Ikle, ‘Nuclear Strategy: Can There Be A Happy Ending?’ in Foreign Affairs (Spring 1985). 2. Colin Gray, ‘What RAND Hath Wrought?’, Foreign Policy 4 (Fall 1971). 3. Bernard Brodie, ‘The Development of Nuclear Strategy’ in Bernard Brodie, Michael Intriligator and Roman Kolkowicz (eds.), National Security and International Stability (OGH Publishers, 1983), p. 7. 4. Gray, op. cit. 5. Ibid. 6. See Roman Kolkowicz and Ellen Mickiewicz (eds.), The Soviet Calculus of Nuclear War (Lexington, MA: Lexington Books, 1986); and Ellen Mickiewicz and Roman Kolkowicz, Arms Control and International Security (New York: Praeger, 1986). 7. Bernard Brodie (ed.), The Absolute Weapon (New York: Harcourt Brace, 1946), p. 76. 8. Bernard Brodie, ‘Why Were We So (Strategically) Wrong?’, Foreign Policy (Winter 1971), pp. 151–62. 9. Bernard Brodie, ‘On Clausewitz: A Passion for War’, World Politics (Jan. 1973), p. 306. 10. Brodie, ‘The Development of Nuclear Strategy’, op. cit., p. 12. 11. Ibid., p. 13. 12. See the chapters in this volume by Robert Jervis and George Quester. 13. Cited in The New York Times (15 Sept. 1983), p. A–27. 14. Bruce Blair, op. cit.
12 DILEMMAS OF NUCLEAR STRATEGY
15. Secretary of the Navy John Lehman, Jr., cited in Arms Control and International Security, op. cit., ‘Introduction’. 16. Former Secretary of Defense Harold Brown, cited in Arms Control and International Security, op. cit., ‘Introduction’. 17. Former Secretary of Defense James Schlesinger, cited in Arms Control and International Security, op. cit., ‘Introduction’. 18. See conference papers by Robert Jervis, George Quester, Leonard Freedman, Joseph Nye, Pierre Lellouche, and Christian Hacke. 19. See conference papers by Colin Gray and Benjamin Lambeth. 20. See conference papers by Ken Booth and George Rathjens. 21. See conference papers by Arthur Stein and Roman Kolkowicz. 22. See conference paper by Roman Kolkowicz. 23. See conference paper by Pierre Lellouche. 24. See conference paper by Christian Hacke. 25. See statements by Ambassador A.Dobrynin, Henry Kissinger, Helmut Sonnenfeldt, and William Hyland, among others, in Arms Control and International Security, op. cit. 26. See conference papers by Joseph Nye and Ken Booth. 27. Bernard Brodie, The Absolute Weapon, op. cit.
II. Nuclear Deterrence: Critical Assessments
14
The Not-Quite-Absolute Weapon Deterrence and the Legacy of Bernard Brodie Gregg Herken
Bernard Brodie has been called—rightly—the ‘dean’ of America’s civilian strategists. Upon Brodie’s death in 1978, another eminent member of the fraternity of nuclear strategists, Thomas Schelling, acknowledged Brodie as ‘first—both in time and in distinction’—among those whose profession has been to think about the unthinkable. What I intend to argue here is not only that this ranking of Bernard Brodie at the top of the list of civilian strategists is deserved and correct, but that Brodie’s work, and his subsequent reflections upon it, is itself a kind of personal embodiment of the crisis which has befallen nuclear strategy and strategists, as well as a measure of their current discontent.1 Contrary to a popular impression, Brodie did not invent either the concept or the term ‘deterrence’.2 As George Quester has pointed out, the concept in strategy dates back at least to the warring city-states of ancient Greece, and the term itself appears in the 1820 edition of the Oxford English Dictionary as meaning ‘to frighten from’.3 Brodie, however, is responsible for popularizing ‘deterrence’ as it applied particularly to nuclear weapons. He deserves credit for bringing to public attention the fact, as he wrote, that ‘what was distinctively new’ about deterrence in the atomic age ‘was the degree to which it was intolerable that it should fail’. Indeed, he can fairly be said to have written the book on the subject. Brodie’s premier standing in the rank of civilian strategists was guaranteed by the small volume of essays he and his colleagues at Yale wrote in the weeks just after Hiroshima—almost literally under the shadow of the mushroom cloud—and published early in 1946 under the title, The Absolute Weapon: Atomic Power and World Order.4 For the next 40 years, The Absolute Weapon would be the essential primer for those seeking to understand how the bomb had affected military strategy and international diplomacy. In one of the two essays he wrote for the book, Brodie spelled out the theory of nuclear deterrence with an eloquence and an economy of words that has not been improved on in two generations: ‘The writer…is not for the moment concerned about who will win the next war in which atomic bombs are used. Thus far the chief purpose of our military establishment has been to win wars. From now on its chief purpose must be to avert them. It can have almost no other useful purpose.’ 5 The problem with Brodie’s formulation—and, indeed, the problem with deterrence —is represented by the single qualifier in Brodie’s last sentence: ‘almost no other
16 DILEMMAS OF NUCLEAR STRATEGY
useful purpose’. It is over the other purposes to which a military establishment equipped with nuclear weapons might be put, subordinate to deterrence, that disagreement arises between strategists. It is the reason why the current world of nuclear strategy is riven by discord. Interestingly enough, Brodie neither anticipated nor was prepared for the dawn of the atomic age. He first learned about the bomb, as did most citizens, when he read the newspaper on Sunday morning, 7 August 1945. At the time of the initial atomic raid on Japan, Brodie was writing a tightly-reasoned essay on how the battleship was about to make a comeback in modern warfare. In the moment he became aware of the bomb, Brodie recognized that it made this essay—and the bulk of his work to date, which had mostly focused on naval strategy—instantly obsolete. But as Brodie himself had earlier observed with regard to the effect of technological innovation in warfare—‘when a change comes, it is best if it is unequivocal’. 6 The clarity and depth of Brodie’s vision in The Absolute Weapon is all the more remarkable considering the fact that the book was begun within weeks of the event it analyzes and attempts to put in perspective. In the first of his essays, titled ‘War in the Atomic Age’, Brodie made a series of observations about the effect the atomic bomb would have upon war and peace that are now seen to be particularly prescient. First was Brodie’s point that the bomb was in fact a revolutionary new weapon. ‘Everything about the atomic bomb is overshadowed by the twin facts that it exists and that its destructive power is fantastically great’, he wrote. 7 While few would quarrel today with Brodie’s contention that the bomb inaugurated a ‘wholly novel form of war’, it was a view widely contested at the time—particularly by some in the military services, who argued that the bomb simply made necessary larger conventional armies, navies, or air forces.8 Brodie was also quick to recognize that the prospect of defending civilian populations against nuclear weapons was ‘exceedingly remote’. In the book he dismissed as infeasible and too socially disruptive some ambitious proposals of the time to disperse population by breaking major US metropolitian areas into ‘ribbon’ or ‘cellular’ cities of legally-proscribed size. Though Brodie’s pessimism about civil defense is now being challenged, I would argue that the fact we are still searching for a way to defend cities some 40 years after Hiroshima proves that his original point has merit. Brodie also understood that the unique destructiveness of the bomb for its size might make it a truly decisive weapon in warfare—one that would finally make practical Douhet’s dream of a war fought and decided in an afternoon. At a time when even President Truman’s science adviser was expressing doubt about the feasibility of ocean-spanning rockets, Brodie forecast the coming of the nucleartipped ICBM. Brodie was as well perhaps the first to appreciate how, in the atomic age, not even military superiority could guarantee security; and that superiority in numbers was no longer necessarily synonymous with military advantage. Well before the term ‘sufficiency’ would be applied to strategy, Brodie showed he understood the concept behind it when he wrote in The Absolute Weapon: ‘If 2,000 bombs in the hands of either party is enough to destroy entirely the economy of the other, the fact that one side has 6,000 and the other 2,000 will be of relatively small significance.’ 9 Finally, though he recognized the unique destructiveness of nuclear weapons and despaired of defending against them, Brodie did not imbue the bomb with
DETERRENCE AND THE LEGACY OF BERNARD BRODIE 17
supernatural powers, or argue that the absolute weapon had changed the very nature of international politics, forevermore and absolutely. He discounted fears that the bomb would become the new weapon of choice for saboteurs or terrorists: a prediction that thus far has held true, if only narrowly, and one that Brodie himself had begun to question near the end of his life. Unlike many of that day—including key figures in the Truman administration— Brodie did not think that the relative scarcity of uranium would be a lasting inhibitor upon the number of bombs in the world. He correctly predictd that other countries— Russia included—would have their own nuclear weapons within five to ten years. In the second of the essays he wrote for the book—‘Implications for Military Policy’—Brodie elaborated upon the significance of the predictions he had made in the first. The most revolutionary thing about the bomb, he argued, was that it radically changed the cost-benefit analysis of warfare familiar to von Clausewitz and his disciples. No longer could a state rationally expect to benefit from a war once its enemies, too, were in possession of the absolute weapon and able to use it in retaliation against an attacker. Here Brodie was early—if not quite the first—to challenge what was perhaps the dominant view of the bomb, made popular by one of its inventors, Robert Oppenheimer, that nuclear weapons were inherently and necessarily ‘a weapon for aggression’. It was Brodie’s mentor at the University of Chicago, Jacob Viner, who pointed out in the fall of 1945 how the atomic bomb could be ‘a war deterrent, a peace-making force’. Building upon this point, Brodie recognized that integral to the theory of deterrence was the threat of retaliation, and that the key to both was in the realm of perceptions. Such a threat, he wrote, ‘does not have to be 100 per cent certain; it is sufficient if there is a good chance of it, or if there is belief that there is a good chance of it’. The prediction’, he noted, ‘is more important than the fact’.10 In his second essay, Brodie made a case for a secure deterrent some 13 years before Albert Wohlstetter’s Foreign Affairs article on ‘The Delicate Balance of Terror’. The ‘first and most vital step’, Brodie wrote, ‘is to take measures to guarantee to ourselves in case of attack the possibility of retaliation in kind’. One needed ‘to make as nearly certain as possible that the aggressor who uses the bomb will have it used against him’.11 It is thus fair to say, as Lawrence Freedman does in The Evolution of Nuclear Strategy, that virtually all the basic axioms of nuclear strategy—as well as the issues in the current nuclear debate—were articulated in the weeks and months just after Hiroshima.12 And one might add that many of them were expressed first and best by Bernard Brodie in The Absolute Weapon. It would be a mistake to conclude that Brodie’s notion of deterrence was unchallenged—or even widely accepted—in the wake of the bomb’s use and the war’s end. Jonathan Schell has pointed out how Albert Einstein was only one of many who rejected Brodie’s conclusion in The Absolute Weapon that it was both possible and necessary ‘to develop the habit of living with the bomb’. ‘Unless [world government] prevails, and unless by common struggle we are capable of new ways of thinking’, the famous physicist wrote in September 1945, ‘mankind is doomed’.13 If Einstein, as Schell claims, was an ‘idealist’ and Brodie a ‘realist’, then there was a third category of theorists who combined attributes of both schools of thought, but who came up with an entirely different conclusion about the bomb. This was the group of theorists who believed that a nuclear war could be fought and won. It was
18 DILEMMAS OF NUCLEAR STRATEGY
these strategists who seized upon the conditional ‘almost’ in Brodie’s theory of deterrence to argue that there was another purpose for the bomb if deterrence failed: victory in a nuclear war. Ironically, even as Brodie was writing The Absolute Weapon, across the street from his office in Yale’s Hall of Graduate Studies, a student at the law school was finishing his own book which would have a conclusion diametrically opposed to Brodie’s. In There Will Be No Time: The Revolution in Strategy, William Liscum Borden attacked what he called the ‘mutual deterrent fallacy’ at the heart of Brodie’s analysis. ‘The essential point is that an armed peace cannot persist indefinitely, that either war or voluntary federalism must resolve the truce’, Borden wrote.14 Whereas Einstein had concluded that this very choice left the country with no alternative but to pursue internationalism, Borden argued that the United States— failing the attainment of one world—must prepare to fight and win the inevitable nuclear war. ‘A full-scale atomic war will not be won on pulverizing cities and industry, but by destroying the enemy’s military power of retaliation’, Borden predicted. The war he envisioned would be a ‘one-dimensional aerial duel’ fought ‘between highly decentralized military systems’ where each side would—at least initially—refrain from hitting the enemy’s civilian population, holding it hostage to the outcome of the conflict instead. 15 Borden’s book appeared in the spring of 1946, within weeks of the publication of Brodie’s. Like The Absolute Weapon, parts of There Will Be No Time seem today almost hauntingly prescient. Borden predicted that, in order to avoid what he called ‘a rocket Pearl Harbor’, the United States would have to station its future nucleartipped intercontinental-range missiles in protected underground ‘hedgehogs’ located well away from cities, and ‘on undersea platforms scattered throughout the world’s oceans’. ‘Should the great-power foreign ministers assemble around a table in the year 1960’, he wrote in 1946, ‘perhaps the vital unspoken question to be asked of each will be this: “How many atomic bombs could your country activate on about thirty minutes’ notice!”’ 16 Borden himself later characterized his own way of thinking about the bomb as ‘relativist’, to distinguish it from what he called Brodie’s ‘absolutist’ views on deterrence. 17 But contemporary strategists would recognize Borden as an early proponent of nuclear war-fighting, and what he termed the ‘war-between-the-bases’ as perhaps the first exposition of the nuclear strategy of counterforce. As the months and years passed after 1946, Borden’s dire prognosis of war ‘certain and inevitable’ in the absence of world government seemd to be proven wrong. But Brodie, too, in this time began to admit that some of his own conclusions in The Absolute Weapon were either mistaken, or at least subject to revision. While contemptuous in the book of what he termed the military’s ‘preatomic thinking’, Brodie gradually realized that his own thinking had not taken into account all the ways that nuclear weapons would change strategy and the future of war. These unanticipated effects were due, in turn, to unpredictable technological advances in the weapons themselves—advances that drastically increased both the number and power of the bombs, and in the process undermined some of the basis assumptions behind The Absolute Weapon. When he left Yale in 1949 to become a civilian consultant to the Air Targets Division of Air Force Intelligence, Brodie learned firsthand how some of his early views had been made obsolete. Atomic tests conducted during the spring of the
DETERRENCE AND THE LEGACY OF BERNARD BRODIE 19
previous year had confirmed that it was possible to more than double the explosive yield of a weapon with less than half the plutonium used in early models of the bomb. 18 The shrinking size and cost of nuclear weapons, as well as the increasing ease with which they could be manufactured in quantity, thus invalidated what had been one of Brodie’s most confident assertions in The Absolute Weapon—that the mechanism behind the bomb ‘must be ingenious and elaborate in the extreme, and certainly not one which can be slipped into a suitcase’.19 The same technological progress that made nuclear weapons simultaneously more numerous and cheaper also acted to invalidate another and more important contention of Brodie’s book—that the bomb did not ‘lend itself to discriminate use’.20 Brodie’s argument in 1946 that ‘for some time to come’ the ‘primary targets for the atomic will be cities’ and that ‘the bomb is inevitably a weapon of indiscriminate destruction’ was based upon the assumption that nuclear weapons would remain too few and to expensive to use against any but the enemy’s most valuable and most vulnerable targets.21 But the advent of tactical-sized atomic weapons, the Soviet Union’s entry into the arms race, and the subsequent growth of the US nuclear arsenal—which quadrupled in size during the three years that Brodie worked for the Air Force—rendered invalid this view of a nuclear strategy dictated by scarcity. 22 Perhaps the biggest surprise for Brodie was the phenomenal increase in the destructiveness of nuclear weapons. While still working as a consultant for the Air Force on war planning, Brodie had seen a graphic portrayal of the steadily increasing power of the bomb as he watched the aiming point for the attack on Moscow gradually shift from the spires of the Kremlin, to a spot midway between the Kremlin and a secondary target in the city, and finally to the city’s center.23 The real shock had come with the advent of the hydrogen bomb in the early 1950s, followed closely behind by the sooner-than-expected appearance of the intercontinental ballistic missile. In 1946, Brodie had written that not even the atomic bomb was ‘so absolute a weapon that we can disregard the limits of its destructive power’. Six years later, by contrast, the power of the new super-bomb seemed without limit. Brodie’s colleagues at the RAND Corporation, the Air Force-sponsored thinktank where he went in 1952 after leaving the Pentagon, remembered how he was ‘unsettled’ and even ‘swamped’ by the hydrogen bomb. A friend of Brodie’s at RAND suggested that he had chosen the right title in The Absolute Weapon, but that the book had been written about the wrong bomb.24 In the wake of the H-bomb’s arrival, Brodie seemed willing to put less of an absolutist emphasis upon deterrence—perhaps because the prospect of deterrence’s failure was now even more horrific than he had imagined in 1946. But Brodie’s subsequent search at RAND throughout the early and mid-1950s for an alternative strategy to deterrence failed to produce a more hopeful result. An earlier idea Brodie had unsuccessfully proposed to the Air Force as part of his effort to introduce restraint into nuclear targeting—the notion of ‘sample attacks’, whereby the US would bomb certain governmental centers in Russia, but only after giving the civilian population advance warning and sufficient time to evacuate—he now recognized was no longer feasible, since the Soviets had now acquired nuclear weapons and could respond in kind to attacks on cities.25 For a brief period at RAND, Brodie even entertained William Borden’s idea of a counterforce, or war-fighting strategy as an alternative to an absolute reliance upon
20 DILEMMAS OF NUCLEAR STRATEGY
deterrence, but his own quick calculations showed that a supposedly selective attack upon a ‘separated target system’, in which only military targets were struck, would none the less kill an estimated two million Russian civilians. He also was convinced, from the experience of the previous war, that hostilities which began with an attack on military bases would in time escalate to attacks on cities.26 An article that Brodie wrote for the October 1955 issue of Harper’s—titled ‘Strategy Hits a Dead End’—suggests that he had by this time already abandoned his search for a preferred alternative to The Absolute Weapon’s reliance upon deterrence. Looking back in the article upon the first nuclear decade, Brodie acknowledged how ‘many of the interpretations’ offered at the dawn of the atomic age had since been shown to be ‘too conservative’. ‘In retrospect’, he wrote, ‘it is clear that many of them were wedded to presumptions soon to be disproved—for example, that the bomb was fated to remain scarce, extremely costly, bulky and therefore difficult to deliver, and limited to about the same power and spatial effectiveness as the Nagasaki bomb’.27 Even a year before the Harper’s article, Brodie had written to a friend explaining why he temporarily abandoned his work at RAND on the book he intended to be a sequel to The Absolute Weapon, which was eventually published as Strategy in the Missile Age—a work, incidentally, that Lawrence Freedman has properly described as ‘a gloomy book’. 28 As he progressed, Brodie admitted to the friend, ‘it became clear that “strategy” and “unlimited war” are simply incompatible concepts in a world of H-bombs’.29 Unwilling in the early 1960s to join his RAND colleagues in promoting counterforce to the Kennedy administration—but also unable to evince the same faith in deterrence he had had in 1946—Brodie in his subsequent writings proposed that the United States put a greater reliance upon tactical nuclear weapons, as ‘a second line of insurance’ between absolute deterrence and all-out thermonuclear war.30 Yet even here Brodie’s controversial case for limited nuclear war seemed more in the nature of an intellectual argument than a seriously proposed solution to the problem of strategy in the thermonuclear age. Two of Brodie’s associates at RAND—Herman Kahn and Sam Cohen—each believed that by the time of his death Brodie’s thinking on strategy had travelled full circle, arriving back at the place it began in The Absolute Weapon—with a nearly absolute reliance upon deterrence. ‘A plan and policy which offers a good promise of deterring war is…by orders of magnitude better in every way than one which depreciates the objective of deterrence in order to improve somewhat the chances of winning.’ Brodie wrote at the conclusion of Strategy in the Missile Age, published in 1959.31 In the last essay he wrote on the subject, which appeared in International Security just before his death, Brodie reaffirmed his belief in deterrence—though it was, seemingly, a faith grown rather more worried and perhaps even desperate at the end. 32 There is a certain irony to Brodie’s contribution to nuclear strategy when seen in historical perspective. Despite the fact The Absolute Weapon had become a kind of icon of the popular faith in deterrence, its central message has been largely disregarded by most members of the so-called strategic ‘priesthood’. Foremost in that message, and in Brodie’s legacy, is a comment on the value of restraint. It is thus significant—even if often overlooked—that Brodie was in effect fired from his job with the Air Force in the 1950s for counselling restraint in nuclear
DETERRENCE AND THE LEGACY OF BERNARD BRODIE 21
targeting. Brodie’s original interest in reducing the vulnerability of the US deterrent was not only to keep it from becoming an unwitting magnet for surprise attack, but also to allow the United States to exercise restraint in its retaliation to aggression. Brodie’s idea of sample attacks—and particularly his remarkable proposal in 1951 to deliberately reduce the rate at which nuclear weapons were being added to the American arsenal, so as to force a more serious consideration of US objectives in a war —ran directly counter to Air Force thinking of the time, when restraint was a foreign concept.33 Often ignored as well is the fact that Brodie’s advocacy of tactical nuclear weapons —which he promoted as an alternative to making deterrence reliant entirely upon the threat of superbombs—had made him virtually a pariah at RAND by the time he left the think-tank in the early 1970’s to teach at UCLA. Instead, the emphasis at RAND was upon meeting the threat of conventional force with a conventional defense. Contrary to popular belief, it is not Brodie’s formulation of deterence as the prevention of war that has guided the nation’s strategists and war planners since Hiroshima, but the other supposed attributes of nuclear weapons—especially the belief in the political utility of strategic superiority, and the related idea that one can ‘extend’ deterrence by developing the theoretical capability to fight and win a nuclear war. Perhaps the crowning irony of Brodie’s legacy as a strategist is the fact that while the peace we have had for the past 40 years is the one Brodie predicted, we have been planning all along for Borden’s war. Yet the most profound misunderstanding concerning Brodie and his legacy may be the failure to realize that the universal popularizer of deterrence did not believe deterrence was a permanent solution to the problem of strategy in the nuclear age. As Schell has pointed out, Brodie’s concern in 1946—and that of his colleagues in The Absolute Weapon—was with the immediate, not the long-term, nuclear threat. ‘The “short term”’, Schell notes, ‘occupies them completely’.34 At the end of his 1955 Harper’s article, Brodie was still looking for what he called ‘the new ideas and procedures necessary to carry us through the next two or three dangerous decades’. Brodie’s subsequent work shows that he consistently rejected both the careless hope of total disarmament and the deceptive allure of preventive war as a way out of the balance of terror.35 By the time of his death, as Kahn and Cohen have testified, it is even possible that Brodie had given up hope of finding an alternative to deterrence. But it is also evident in The Absolute Weapon that Brodie, from the outset, had the wisdom of knowing where not to look for a solution to the problem. ‘[O]ne can scarcely assume that the world will remain either long ignorant of or acquiescent in the accumulations of such vast stockpiles of atomic bombs’, he wrote at the conclusion of his second essay in that book.
If existing international organization should prove inadequate to cope with the problem of controlling bomb production…a run-away competition in such production would certainly bring new forces into the picture. In this chapter and in the preceding one, the writer has been under no illusions concerning the adequacy of a purely military solution.
22 DILEMMAS OF NUCLEAR STRATEGY
He intended the strategy of deterrence, Brodie made plain at the end of The Absolute Weapon, only as a expedient—a means, not the end—whereby the nation would be ‘better able to pursue actively that progressive improvement in world affairs by which alone it finds its true security’.36
NOTES 1. Thomas Schelling, ‘Bernard Brodie (1910–1978)’, International Security (Winter 1978), 2–3. For other works on Brodie and his contribution to strategy, see Barry Steiner, ‘Using the Absolute Weapon: Early Ideas of Bernard Brodie on Atomic Strategy’, ACIS Working Paper No. 44 (January 1984), Center for International and Strategic Affairs, UCLA, and ‘Bernard Brodie and the American Study of Nuclear Strategy’ (unpublished manuscript); Fred Kaplan, Wizards of Armageddon (New York: Simon & Schuster, 1983), pp. 9–50; Lawrence Freedman, The Evolution of Nuclear Strategy (New York St. Martin’s Press, 1981), pp. 131–6 and 403–40; and Colin Gray, Strategic Studies and Public Policy: The American Experience (Lexington, KY: University Press of Kentucky, 1982). 2. As Samuel Huntington has noted, Brodie got the concept right but the suffix wrong—‘determent’ is the term used in The Absolute Weapon. 3. George Quester, Deterrence Before Hiroshima: The Airpower Background of Modern Strategie (New York: Wiley & Sons, 1966). Regarding the other origins of the concept of deterrence, see, for example, Alexander George and George Smoke, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press, 1974); and Robert Jervis, ‘Deterrence and Perception’, International Security (Winter 1982/83). 4. Bernard Brodie (ed.), The Absolute Weapon: Atomic Power and World Order (New York: Harcourt, Brace, 1946). The other members of the ‘Yale group’, as it became known, were Frederick Dunn, Arnold Wolfers, Percy Corbett, and William T.R.Fox. 5. The Absolute Weapon, p. 76. Brodie’s first essay on the significance of nuclear weapons appeared less than two months after the destruction of Hiroshima. ‘The Atomic Bomb and American Security’, draft manuscript, 1 Nov. 1945, Specials Collections, Sterling Library, Yale University. 6. Brodie, Seapower in the Machine Age (Princeton, NJ: Princeton University Press, 1941), p. 448. 7. The Absolute Weapon, p. 52. 8. For example, air-power enthusiast Alexander de Seversky argued after Hiroshima that an atomic bomb was only marginally more destructive than a conventional, high-explosive bomb; a Navy commander claimed in fall 1945 that exploding an atomic bomb at one end of a runway would leave people at the other end unhurt. Concerning these and other early misestimates of the power of the bomb, see ‘Atom Bomb Hysteria’, Reader’s Digest (Feb. 1946), pp. 82–97. 9. The Absolute Weapon, p. 48. 10. J.Robert Oppenheimer, ‘Atomic Weapons and the Crisis in Science’, Saturday Review of Literature (24 Nov. 1945). Jacob Viner, ‘The Implications of the Atomic Bomb for International Relations’, Proceedings of the American Philosophical Society, Vol. 90, No. 1 (29 Jan. 1946), pp. 53ff. The Absolute Weapon, p. 74.
DETERRENCE AND THE LEGACY OF BERNARD BRODIE 23
11. 12. 13. 14. 15. 16. 17.
18.
19. 20. 21. 22.
23.
24. 25. 26.
27. 28. 29. 30.
31. 32. 33.
The Absolute Weapon, pp. 75–6. Freedman, The Evolution of Nuclear Strategy, p. 44. Einstein is quoted in Schell, The Abolition (New York: Knopf, 1984), p. 29. William Borden, There Will Be No Time: The Revolution in Strategy (New York: Macmillan, 1946), pp. 16, 225. There Will Be No Time, pp. 87, 113. There Will Be No Time, p. 175. Borden used the terms in a paper he gave at a West Point symposium in 1984 on nuclear strategy, ‘Springtime of the Nuclear Debate’, undated and unpublished manuscript. My thanks to Mr Borden for a copy of his paper. The technological progress made in the development of nuclear weapons at this time is recounted in Hans Bethe, ‘Comments on the History of the H-Bomb’, Los Alamos Science (Fall 1982), pp. 43–53; and David Rosenberg, ‘The Origins of Overkill: Nuclear Weapons and American Strategy, 1945–1960’, International Security (Spring 1983), pp. 19–20. The Absolute Weapon, pp. 50–51. The Absolute Weapon, p. 21. The Absolute Weapon, p. 47. The number of warheads in the US nuclear weapons stockpile evidently increased from 250 in 1949 to 1,000 in 1952. Thomas Cockran et al., Nuclear Weapons Databook: Volume I, U.S. Nuclear Forces and Capabilities (Cambridge, MA: Ballinger, 1984), p. 15. Brodie notes the changing plans for the attack on Moscow in a letter to David Rosenberg, 11 October 1977, ‘Pending’ folder, Box9, Bernard Brodie MSS, Special Collections, UCLA library. The effect the advent of the hydrogen bomb had upon Brodie is detailed in Gregg Herken, Counsels of War (New York: Knopf, 1985), pp. 34–8. Brodie’s early idea of ‘sample attacks’ is recounted in Steiner, ‘Bernard Brodie and the American Study of Nuclear Strategy’, pp. 68–72. Brodie’s qualms about two million civilian fatalities in a counterforce attack of the early 1950s reveals how the inflation brought about by bigger and more numerous bombs has changed war planning. ‘By today’s exaggerated standards, one or two million is a separated target system’, Herman Kahn observed in a 1981 interview with the author. Brodie, ‘Strategy Hits a Dead End’, Harper’s (Oct. 1955), p. 34. Freedman, The Evolution of Nuclear Strategy, p. 133. Letter, Brodie to S.Jones, 6 June 1954, ‘J’ folder, Box 1, Bernard Brodie MSS, UCLA. Brodie introduced the ‘second line of insurance’ idea of limited war in Strategy in the Missile Age (Princeton, NJ: Princeton University Press, 1959), and took it further—some thought too far—in a later book, Escalation and the Nuclear Option (Princeton, NJ: Princeton University Press, 1966). Strategy in the Missile Age, pp. 408–9. Brodie, ‘The Development of Nuclear Strategy’, International Security (Spring 1978). Brodie’s disagreement with the Air Force and its doctrine is noted in Herken, Counsels of War, pp. 34–8; and Steiner, ‘Bernard Brodie and the American Study of Nuclear Strategy’, pp. 55–90.
24 DILEMMAS OF NUCLEAR STRATEGY
34. Schell, The Abolition, p. 34. 35. Brodie’s doubt that nuclear weapons would become a catalyst for disarmament or would government was first expressed in The Absolute Weapon: ‘There is happily little disposition to believe that the atomic bomb by its mere existence and by the horror implicit in it “makes war impossible”. In the sense that war is something not to be endured if any reasonable alternative remains, it has long been “impossible”’ (p. 75). Similarly, in Strategy in the Missile Age, Brodie observed of the logic behind preventive war: ‘It is somewhat bizarre to argue that it would be wise to choose now an infinitely drastic and terrible course mostly because the problem that would allegedly be liquidated in that way is one which we or our heirs would be too stupid to handle properly later’ (p. 233). 36. The Absolute Weapon, p. 107.
Deterrence: The Problem—Not the Solution Michael MccGwire
The thesis of this article is that many of the problems which currently assail the West stem from the adoption in the early 1950s of ‘nuclear deterrence’ as the basis of defense and foreign policy. My complaint does not apply to generic deterrence, a simple concept that applies at most levels of human interaction, but to the body of theory that grew up around nuclear weapons in the 1950s. There was, of course, no one theory, doctrine, or policy of nuclear deterrence, but a whole series of them that differed between countries and between government departments within countries. There was also a wealth of difference between the sophisticated approach of the more thoughtful deterrence theorists and the bowdlerized and homogenized version of their theories which was passed on to the rest of us. Nevertheless, at the heart of this body of theory was a central dogma concerning the requirements of deterrence and, more importantly, a frame of mind that went with the dogma. It is this frame of mind that is at the root of the problem. I will label this body of attitudes and ideas ‘deterrence dogma’. It was this that was being taught at service colleges and universities throughout the 1950s and 1960s under the title of ‘nuclear deterrence theory’. The dogma was the basis of innumerable and repetitive articles in service journals, and underlay most of the discussion of defense and foreign policy matters in the daily press and in the weekly and monthly magazines. And it appeared continually in NATO policy statements and national defense white papers. This deterrence dogma came to crowd out other longestablished ideas, becoming at times a virtual substitute for both foreign and defense policy.1 The usual rebuttal to this type of criticism is to claim that ‘nuclear deterrence has kept the peace for thirty-five years’, ‘peace’ being defined as the avoidance of EastWest conflict, particularly in Europe. A more thoughtful rebuttal acknowledges that we cannot be sure that it was deterrence that preserved the peace, but argues that ‘it is better to be safe than sorry’. One implication of this caveat is that while nuclear deterrence may not have been responsible for keeping the peace, we did not suffer any harm through adopting such a policy. It is this comforting and widely-held assumption—that deterrence has caused no harm—that is the focus of my argument. Only when the costs of nuclear deterrence have been clearly established will it be
* Reprinted by kind permission of International Affairs, the quarterly journal of the Royal Institute of International Affairs, London.
26 DILEMMAS OF NUCLEAR STRATEGY
worth turning to refute the more extravagant claim, that this same deterrence has kept the peace for 35 years.
The Origins of Deterrence Dogma In the wake of the First World War, war was seen as too important to leave to generals, and it was taken over by the politicians. They, having done little better in the Second World War, left the field to academics, who moved in to address the new conundrums that came with the atomic bomb. There was peculiar quality to this subdiscipline of ‘strategic studies’ as it emerged in the 1950s. It focused almost entirely on the problems of nuclear weapons, deterrence doctrine, and, in due course, theories of limited nuclear war. With a few notable exceptions, such as Bernard Brodie and Michael Howard, the field came to be dominated by academics from axiomatic disciplines like mathematics, physics and economics, who took over from the military and diplomatic historians who had traditionally presided over the study of international politics. For all these theorists, the Soviet ‘threat’ was a basic assumption. It was assumed that the Soviet Union had a relentless drive for territorial conquest, and the reality of a monolithic communist bloc went unquestioned. The governing concept of deterrence took it for granted that the Soviet Union felt an urge, above all things, to seize Europe. This assumption provided the basis for most strategic theorizing. The field developed a new breed of self-styled ‘tough-minded’ strategic analysts, who liked to think through problems abstractly and in a political vacuum. To this new breed, the opponent was not ‘Soviet man’, not even ‘political man’, but an abstract ‘strategic man’, who thought, as they did, in game-theoretical terms. The original concept of nuclear deterrence derived from a belief that the only practical way of restraining Soviet aggression was to threaten unacceptable punishment. Initially, the theorists concentrated on how to keep the threat of US nuclear retaliation ‘credible’, which required that the United States be seen to have both the capability to inflict unacceptable punishment and the will to do so. Capability, in turn, required that the means of inflicting the punishment be made invulnerable to Soviet surprise attack. The possibility of technological breakthrough added a further dimension to this problem. The need for invulnerability introduced a new factor into the offense-defense calculus and gave full rein to an extreme form of worst-class analysis. Imaginative strategic theorists worked hard to discover chinks in the armor of assured retaliation which a determined opponent might hope to exploit with a bolt-from-the-blue attack or in some other way. The view that the Soviet Union could only be held in check by the threat of nuclear devastation persisted, but by the second half of the 1950s another danger had become apparent: the temptation to pre-empt, a by-product of the emerging Soviet capability to strike directly at the United States with nuclear weapons. The advantages of getting in the first blow were so overwhelming that a prudent leader could now be expected to launch a nuclear strike if it was suspected that the other side was contemplating war. The assumed pressure to pre-empt introduced a new concern into the debate, the ‘stability’ of the strategic balance, and the simple requirement for Soviet aggression to be deterred came to be qualified by the somewhat
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 27
contradictory requirement for the Soviet Union to be reassured2 that the United States would not initiate a nuclear war, in case the Soviet Union should be driven to launch a pre-emptive attack. Two main requirements had to be met in order to provide the reassurance that would achieve strategic stability. First, enough weapons should be able to survive a first strike to inflict unacceptable punishment on the initiator: a retaliatory capability called ‘an assured second strike’. And since, in theory at least, both sides had to meet this requirement, the second requirement was that each had to eschew weapon systems and offensive or defensive deployments which might deprive the other of such a second-strike capability. These requirements were the underpinnings of ‘assured destruction’, a doctrine which was adopted by the arms control community in the 1960s, although it never become official US policy. Both deterrence theory and reassurance theory took the Soviet threat for granted. Such disagreement as there was on this point centered on which element was thought more likely to prompt a Soviet attack: expansionist urge or reciprocal fear. Both theories had in common the requirement for an assured second-strike capability, but the theoretical requirements which had to be met to ensure stability were more demanding than those for credibility. For an assured response to be credible, and thus to maintain deterrence, it was only necessary to be able to launch on warning or under attack. For the stability requirement, even a launch-under-attack capability was insufficient. Stability required that there must be a sufficient number of weapons able to suffer and survive a deliberate surprise attack, and then inflict the unacceptable punishment required by deterrence doctrine. Deterrence theory and its reassurance offshoot evolved over time. The ramifications of both theories were spun out to their logical conclusion, and the theory of deterrence moved away from the concept of massive retaliation to greater discrimination in the use of nuclear weapons. While this development was meant to increase credibility, and to improve the range of options available to the president, it also provided persuasive reasons for increasing the size of weapons inventories, without reducing the overall scale of nuclear devastation.
The Effect of Deterrence Dogma Four features of the deterrence dogma must be described before the effect of the dogma can be fully understood. The first of these was the abstract style of reasoning and the axiomatic nature of the underlying theories. This led to a definition of rational behavior which, in political terms, was at best a-rational and more often irrational, and favored models like the ‘prisoners’ dilemma’ in preference to studying the political psychology of opponents and allies. The second was the absence of serious Sovietologists from the theoretical debate, even though theories of limited war and escalation depended on assumptions about the Soviet reaction under given circumstances. The abstract nature of the debate was partly to blame for this omission; but it was also partly due to the way Soviet intentions were taken as given. Capabilities alone were of analytical interest. These two aspects of the dogma, when combined with a consideration of the catastrophic consequences of nuclear surprise, encouraged an extreme form of worst-case analysis in which a course of enemy
28 DILEMMAS OF NUCLEAR STRATEGY
action need only be ‘conceivable’ for it to be included in the calculus of threat. Strategic studies thus came to be dominated by ‘tough-minded’ theorists. A third feature of the deterrence dogma was the fundamentally punitive tone of nuclear deterrence as formulated in the West. Deterrence was not just a passive matter of locking the door and barring the windows to avoid tempting a light-fingered neighbor. Rather, it was an active policy which asserted a hostile Soviet intent and threatened wholesale devastation. The policy relied on the capability and the will to inflict such punishment, and a prerequisite for making this posture credible was that the electorate should continue to believe in a high level of Soviet threat. Fourth, reassurance theory had stemmed from the same roots as nuclear deterrence doctrine, and shared most of its underlying assumptions. The concept of reassurance had emerged as a response to certain undesirable side-effects of deterrence, and its concern was a narrowly-defined concept of stability rather than limiting the build-up of arms. Bearing these salient characteristics in mind, it is possible to consider the effect of deterrence dogma on the policies of the West over the years. In order to sharpen the argument, the terms ‘deterrers’ and ‘reassurers’ will be used as if they were two clearly defined schools of thought. There was, of course, no such thing. Rather, there existed a spectrum of opinion with individuals emphasizing different aspects at different times. However, this heuristic device highlights the important role of the two theories and their shared assumptions in shaping defense and foreign policy. People of great intelligence and considerable goodwill formulated these theories as they sought to grapple with the complexities of the nuclear era, and the only hindsight available to them was the events of the 1930s and 1940s. Their assessments of Soviet intentions were shaped in the immediate post-war years by the Greek civil war, the Berlin blockade, the emergence of the People’s Republic of China, and the Korean War. They believed that the Third World War would come about through Soviet aggression, and they were determined not to repeat the mistakes of the 1930s; Soviet aggression therefore had to be deterred. They were unusually perspicacious in their appreciation that nuclear weapons were qualitatively different from any that had been used before, and they were convinced that the implications of this development needed to be thought through. This does not explain why much the same perceptions persist in the 1980s. Deterrence dogma does explain this persistence. It imposed a rigid framework on our thinking that prevented our view of the world from evolving along with new information and more relevant hindsight. It also became a kind of intellectual tranquilizer, its sophisticated logic imparting a sense of false certainty and inhibiting attempts to challenge its underlying assumptions. Soviet theory, which was ‘innocent of the higher calculus of deterrence’,3 did not ossify in the same way. From 1948 to 1953, the Soviet perception of threat was the mirror image of the West’s, with equally good reason. But by 1959, the Soviet Union had concluded that the primary threat to its well-being was nuclear war, from whatever cause, rather than deliberate aggression by the West.4 Its primary objective, therefore, was to avoid war rather than to deter aggression. The same dichotomy, we shall see, underlies the current debate in the West.
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 29
Deterrence Theory and American Foreign Policy The components of deterrence dogma affected foreign policy, arms control theory, and military doctrine differently. In some cases the effect was indirect, by providing a rationale for policies, attitudes, and actions towards which the American polity was already predisposed. And this raises the question of whether deterrence dogma shaped US policy, or did its adoption reflect the innate preferences of American society?5 That question cannot be answered. However, the dogma certainly did appeal to the American habit of seeing the world in black and white, and the tendency to believe that problems should (and can) be solved—rather than managed or even avoided. The overall effect of deterrence dogma on foreign policy has been to narrow the policy’s focus and to limit unnecessarily the range of its policy options. At the military-tactical—or ‘colonel’s’—level, the analysis of threat focuses on the enemy’s capabilities, and hostile intentions are taken for granted. After the Second World War, it was to be expected that the ‘colonel’s fallacy’ in threat analysis would initially prevail, but deterrence dogma perpetuated it. Worst-case analysis of this sort is appropriate to contingency planning; but it is wholly inappropriate at the politicalstrategic, or ‘ministerial’ level of analysis which should underlie foreign policy. At this higher level, the primary concern should be the most likely course of events, and the assessment must take four factors into account. The first, the opponent’s ‘legitimate requirements’ for military forces to ensure his security, and the second, his actual military capabilities, are compared in order to establish whether the kind of surplus exists which would indicate aggressive intentions. Such an assessment is essential in order to avoid relying solely on the untested assertion that ‘the Soviets have more than they need’, the persistent claim of the last 35 years.6 The other elements of threat analysis are, thirdly, interests, and fourthly, intentions. It is difficult to determine any country’s interests (even one’s own), but it is relatively easy to identify what is not in a nation’s interests. We are prone to assume that what is bad for us must be good for our opponent, and the concept of negative interests is a very important tool for avoiding this pitfall. Soviet intentions must be examined directly. To claim that they should be ignored because they can change overnight is to fall into the trap of the ‘colonel’s fallacy’ and to ascribe worst-case intentions without doing the analysis. At the national level, intentions are remarkably consistent; radical change results from political shifts of a kind which have not occurred in the Soviet Union since the 1919 revolution. The evidence of more than 65 years of Soviet action and pronouncements, when combined with an analysis of Soviet interests and set in the historical context of geography and social inertia, produces a fairly clear picture of Soviet intentions, particularly on such fundamental matters as peace and war. Just as deterrence dogma predisposed the threat analyst to focus on the worst rather than the most likely case, it predisposed the foreign policy specialist to focus on the containment of Soviet power instead of pursuing an objective of a higher order which would allow a wider range of policies in support of that objective. If the primary objective of foreign policy is containment, that policy will inevitably be negative. A broader objective such as ‘securing cooperative Soviet behavior’ would open up more constructive policy options. At such a higher level, containment continues to be an important supporting objective, but it can be flanked by positive objectives such as increasing trade interdependence, fostering consultation on matters
30 DILEMMAS OF NUCLEAR STRATEGY
of mutual interest, and even encouraging rising expectations by improving the Soviet standard of living. Such policies would be incompatible with containment as the principal objective. However, an objective of ‘securing cooperative Soviet behavior’ could even accommodate collaborative policies such as joining with the Soviet Union to stifle dangerous international developments before they become unmanageable.7 During the period of détente in the early 1970s, the United States did adopt a higher-level objective and this did allow a broader range of initiative. These brought benefits that extended beyond the Soviet-American relationship and contributed to the improvement in East-West relations in Europe. But the simple assumptions inherent in deterrence dogma about Soviet intentions are incompatible with those which underlie a policy of detente. The deterrent calculus was manipulated to expose a ‘window of vulnerability’, and was then used to justify a return to more restrictive policies. While deterrence dogma worked to narrow the focus of foreign policy and limit the range of policy options unnecessarily, it also encouraged a particular style of foreign policy. Central to the dogma is the question of credibility. The will to inflict punishment is as important as the capability to do so. This can be analyzed into two parts: willingness to see the enemy suffer what is (by definition) ‘unacceptable’ punishment, and willingness to endure whatever retaliation the enemy can muster. Even in the days of the US atomic monopoly, if the American electorate was to support a policy of nuclear deterrence, the public had to see the Soviet Union as an enemy which deserved such a fate. And as the United States was brought within range of Soviet nuclear delivery systems, its people had to be persuaded that they were ready to suffer massive devastation, perhaps commit national suicide. Bluster and bravery are inversely correlated. The more improbable such an American response became, the more important it was to reaffirm the will to take such action, and to paint the issues at stake in stark, moral terms. In other words, a policy of nuclear deterrence tends to encourage exaggerated, moralistic rhetoric directed at domestic constituents as well as the opponent, and it becomes hard to distinguish between the two. In a populist democracy such as the United States, it is easier to inflame hostility than assuage it, and an administration can become a prisoner of its own past rhetoric.8 Add to this the need to make such a threat credible to one’s opponent, and it becomes almost inevitable that a policy of deterrence will favor intransigence and discourage serious negotiations and the search for compromise.
Deterrence, Arms Control Policy, and Military Doctrine The effects of deterrence dogma on US arms control policy and on NATO military doctrine are intertwined. An assured second-strike capability is a premise common to deterrence and reassurance theories, and concern for that capability has always been at the center of US arms control policy. However, arms control policy has had to reconcile the differing perceptions of the ‘deterrers’ and the ‘reassurers’ on this ostensibly straightforward issue. The ‘deterrers’ concentrate on ensuring that the US second-strike capability does not become eroded. To this end, they hypothesize a malevolent Soviet Union anxious to seize any chance to destroy the United States; they assume an American president
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 31
easily blackmailed into inaction;9 they then consider every possible scenario, present and future, in order to assess what is needed to ensure an adequate US capability; and the answer is always ‘more’. The ‘reassurers’, on the other hand, are concerned that both sides should have an assured second-strike capability. Their assumptions about Soviet perceptions of Western interests and assessment of the chances for Soviet success and US inaction are more carefully grounded in reality, and in that respect their assessment of US requirements is more moderate. On the other hand, the reassurers’ definition of an assured second strike justifies a significantly larger weapons inventory, since ‘stability’ requires the US force to be able to survive an attack, whereas to be ‘credible’ it only needs to be able to launch under attack. That capability is related to warning time, and not to the size of the initial inventory. The deterrers do not dispute the reassurers’ more demanding definition of force requirements, since more is always better. An assured second-strike capability is, however, only one of the requirements of reassurance doctrine. It is also presumed that both sides will avoid weapon systems that might deprive the other of an assured second strike. However, there is no reason to suppose that the Soviet Union has imposed a self-denying law on itself, and evidence does not suggest that the United States is serious about such restraint. In the absence of a self-imposed, self-denying law, reassurance theory becomes a recipe for arms racing, as each side seeks to ensure that it can absorb a first strike and still retaliate. While mutual reassurance provides the basis of Western arms control theory, it does not in itself determine US arms control policy. An inherent tension exists between the requirements of ‘credibility’ and ‘stability’, and experience shows that, when push comes to shove, deterrence invariably wins out over reassurance. When one then adds the requirements derived from the reassurers’ second-strike criteria to those derived from the deterrers’ assumptions about the current Soviet cost-benefit calculus and possible future capabilities, the resultant military requirements become hard to distinguish from a specification for US superiority, including a first-strike capability. But such a superiority would even then cover only one aspect of deterrence, since the official criteria for ‘credibility’ have evolved far beyond the simple requirement for an assured second-strike capability. For almost a decade people have argued that Soviet military doctrine recognizes the need to be able to fight and win a nuclear war and that the only way to deter Soviet aggression is with a comparable capability.10 The argument is logically flawed, but that is not really important. The new official criteria have nothing to do with the original concept of being able to inflict unacceptable damage on the Soviet Union, a capability that is manifest, but everything to do with reluctance on the part of the United States to take such action in defense of Europe, knowing that to do so would invite the nuclear devastation of North America. The argument does, however, illustrate some of the problems under discussion. The Soviet Union has finally been brought into the deterrent calculus, but only as far as its military doctrine for the contingency of world war is concerned. To avoid war is a Soviet objective of the first order. However, if war proves inescapable, the Soviet Union hopes not to lose, and plans accordingly. So does the West—and this point tells us nothing about the West’s broader intentions towards the Soviet Union.
32 DILEMMAS OF NUCLEAR STRATEGY
However, when Western planners assess Soviet intentions, they take Soviet operational concepts at face value, and persist in the ‘colonel’s fallacy’. Nor does the West offer any explanation of how Soviet interests would be served by the military seizure of Europe. It sidesteps that issue, failing to note that deterrence can be directed only against certain kinds of action, and talks instead of deterring an activity—war—or an actor—the Soviet Union. ‘Deterring the Soviet Union’ is a marvelous slogan used to justify almost anything from a new weapons system to all kinds of military deployment. It is also almost meaningless, because it ignores the factor of ‘temptation’, which makes up the other half of the deterrence equation. For deterrence to have any meaning, somebody must be tempted to take the action being deterred. Temptation has two parts: opportunity and urge. Military capabilities tell us little about the second part. Evidence that the Soviet Union’s urge to take over Europe had increased was missing from the rationale advanced by the Carter administration in 1980 to support a more extensive US deterrent capability, and this was one of its logical flaws. But taking temptation for granted had more far-reaching consequences in the case of ‘flexible response’, a concept accepted in Washington in 1961 and formally adopted by NATO in 1967. In so far as it replaced the policy of automatic resort to nuclear weapons with one in which NATO would initially use conventional weapons to check a Soviet assault on Europe, ‘flexible response’ was osensibly intended to increase the credibility of the Western nuclear deterrent. The conventional phase would serve the double purpose of allowing NATO to be ceertain that it was facing a major attack while giving Soviet forces the opportunity to see the error of their ways, and withdraw. This argument only begins to make sense if it is firmly believed that the Soviet Union has an inherent urge to take over Western Europe—the assump-tion that underlies NATO’s defense plans. This assumption is now, however, supported by the evidence. The evidence indicates that, in the event of a world war, if the Soviet Union is to avoid ultimate defeat, it must take over NATO Europe and prevent it being used as a bridgehead by the United States.11 But the problem facing Soviet planners was how, if war was inescapable, to do this without precipitating nuclear war in Europe, which would probably escalate to an intercontinental exchange and result in the nuclear devastation of Russia. The phase of conventional war envisaged by flexible offered the Soviet Union a way out of this impasse. It provided an opportunity to neutralize NATO’s theatre nuclear delivery systems by conventional means, and thereby remove the first rung on the escalation ladder. It also opened up the possibility of a conventional blitzkrieg to knock out NATO forces in Europe, thereby putting the question of escalation into doubt. A combination of both operations would increase the chances of success. This kind of Soviet capability was subsequently seen emerging in the late 1970s. Two conclusions can be drawn from this. In the first place, a Soviet drive into Western Europe would be a response to a strategic imperative where war was inevitable, not a surrender to the temptation to go to war for the sake of territorial gain. A Soviet offensive would therefore be a by-product of the much more momentous decision that world war was unavoidable. It could not be prevented by simple threats of punishment, which would already have been taken into account. Second, the new possibility of finessing nuclear escalation in the European threatre would reduce the constraints on launching a pre-emptive conventional attack.
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 33
Thus, in the event of an East-West crisis which seemed inescapably headed for war, a NATO policy of imposing a conventional pause or forgoing the first use of nuclear weapons would make it just that much more likely that the Soviet Union would cross the threshold from peace to war. These paradoxical and unintended effects of ‘flexible response’ illustrate the limitations of a theory built on the 1950s assumption that the cause of the Third World War would be Soviet aggression. But the effect of that outdated assumption on contemporary foreign policy behavior is even more distorting. Foreign policy-makers drew on the experience of the 1930s and concluded that war could be prevented by deterring aggression with the threat of punishment. The shorthand for such a policy was ‘deterring war’, a term that has become a cornerstone of Western policy pronouncements and permeates our thinking on peace and war. Besides falling into the fallacy of claiming to deter an activity rather than an action, the pious overtones of this expression obscure the radical nature of the underlying principles. US policies are now based on the idea of deterring or preventing war by the threat of punishment, rather than the time-honored principle of averting or avoiding war through negotiation and diplomacy.12 While deterrence dogma worked to undermine diplomacy, its moralistic quality worked to distort sound strategic analysis. An example of this was the 1960 expectation that since it was the Soviets who were threatening world peace, they would therefore recognize the logic of deterrence theory and the non-threatening nature of the US nuclear build-up.13 Soviet persistence in thinking in the traditional terms of matching an opponent’s capabilities was interpreted in the West, most favorably, as an indication of strategic backwardness, and, on the whole, as confirmation of aggressive intentions. More dangerous was the pseudo-moral argument based on the theoretical distinction between counter-force (aimed at military targets) and counter-value (aimed at cities) targeting. This ignored practical realities such as the fact that military targets are often located in or near cities, and the effects of fall-out, but provided a ‘moral’ justification for moving away from reliance on mutual assured destruction as a deterrent to war, to the adoption of a war-fighting strategy. Once introduced into the debate, the ‘moral’ factor became a powerful argument, since the arms control community had laid itself open to criticism by talking of mutually assured destruction as a policy, rather than as a description of the objective situation. By 1979, the arms controllers had largely lost their political constituency on the left, who charged that high theory had resulted in higher force levels, and the situation was ripe for an attack from the right on established policies. The concept of mutual assured destruction was condemned as morally abhorrent, and a space-based defense system was offered as a panacea to the problems that beset us, and one that was moral to boot. The increasingly obvious flaws in accepted ‘strategic’ theory and its inability to solve these problems thus opened the way for an oversimplified concept. The Strategic Defense Initiative launched in March 1983 looks to some unspecified future where the United States and the Soviet Union sit snug beneath their space-based shields. A wish more than a theory, as a means of addressing the problem of nuclear weapons it is flawed more fundamentally than the doctrine it seeks to replace. But while it makes little sense in the terms in which it is being peddled to the public, it has quite different implications in the context of classical military strategy. Were the
34 DILEMMAS OF NUCLEAR STRATEGY
United States to be successful in developing its space-based defense system, it would have regained the position of relative military advantage that it enjoyed for 20 years after the Second World War. Experience suggests that the Soviet Union will make every effort to prevent this, implying another and even more dangerous round in the arms race. While deterrence dogma has provided the unnecessary rationale for exotic weapons programs, it has also undermined the potential for cooperation in arms control, which offered significant prospects for Soviet-American collaboration in reducing the danger of nuclear war. Traditionally, the objectives of arms control have been twofold. At a political level, arms control negotiations can be a means of making war less likely, the process being as important as the details. At a military level, the purpose of arms control is to focus on the possibility of war and to maximize one’s relative advantage. There is an obvious tension between these two levels. However, if the negotiating process is successful in reducing the danger of war, then the military objective becomes less important—and, of course, the reverse is true. The arms control policy that stemmed from deterrence dogma appeared to share its objectives, but there was a crucial difference in the approach to making war less likely. Deterrence dogma decreed that one prevented war by the threat of punishment, embodied in a second-strike capability. The main aim of the ‘deterrers’ was to ensure that the means of retribution would survive a surprise attack. The concern of the ‘reassurers’ was to counteract the pressure leading to pre-emption by taking away its advantage. Both focused on the dangers of sudden attack, with the result that US efforts to reduce the likelihood of war were concentrated on developing measures to prevent crises from getting out of control. The obvious objection to this approach is that by the time one is in a crisis serious enough to engender thoughts of nuclear pre-emption, it is almost certainly too late. This was a technical, theoretical approach to a fundamentally political problem. In the late 1960s and early 1970s, when the ‘reassurers’ were dominant in formulating policy, the US approach to arms control did manage to balance the technical with the political, and this equilibrium fostered fruitful negotiations and contributed to the relaxation in international tensions. But then ‘push came to shove’, the deterrers regained their natural dominance in national security matters, and technical factors became paramount. The United States concentrated on enhancing its military capability in order to prevent war, and arms control became a field of bitter confrontation. As President Reagan has said, ‘it takes two to tango’, and it is therefore relevant that the Soviet Union approaches arms control in a traditional way that is political and practical rather than technical and theoretical. Soviet planners consider war something to be avoided through negotiation and compromise rather than prevented by threat of dire punishment. A primary objective of the arms control process is therefore to help create a political climate that makes war less likely—and this was seen as one of the major accomplishments of the SALT negotiations. In Soviet eyes, the US obsession with the ‘pressure to pre-empt’ obscures the greater danger that comes from the dynamic of events over which there can be no control. Priority should therefore be given to avoiding situations that lead to crises rather than devising ways to manage crises that may well be inherently uncontrollable.14
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 35
The Cost of Deterrence The costs of a policy based on nuclear deterrence can be summarized under seven indictments. Some of these reflect deep-rooted tendencies in American society. However, at the very least deterrence dogma nurtured these attitudes, and often provided the moral and intellectual justification that allowed such policies to flourish. In the first place, the intellectual framework provided by the theories of nuclear deterrence was inadequate. At a very early stage the question of how to handle the awesome new capability got entangled with the question of how to handle the Russians. With hindsight it is possible to see that the broader problem was addressed at too low a level of generality, that of ‘nuclear weapons’ and ‘deterring the Soviets’ rather than the higher-order objective of ‘avoiding world war’—a mistake that Bernard Brodie appears to have sensed, if not articulated. Defining the Soviet threat in worst-case terms meant that concern for the innate dangers of nuclear weaponry took second place to the aim of deterring the Russians. With the Soviet threat taken for granted, intellectual effort concentrated on developing a technical solution for what was essentially a political problem. As the analysts’ scenarios were embroidered, the simple principle of deterrence was obscured by theoretical ramifications, and the strategic debate increasingly became the preserve of specialists with the time and inclination to master the minutiae of nuclear theology. These high priests acquired significant authority in Western counsels and could silence disagreement by noting the dissenter’s ability to grasp the finer points of ‘sophisticated’ theory. Deterrence dogma soon became a sort of intellectual tranquilizer. The underlying theory was so tortuous that it could be used to support opposite sides of the same argument. It could also be transmuted into simple objectives like ‘preserving peace’ and ‘deterring war’—as unobjectionable as motherhood and apple pie. A combination of its simple and complex forms provided the intellectual and moral authority of a state religion, and the effect was just as stultifying. Second, deterrence theory has had a lasting and damaging effect on domestic attitudes. This is the sphere in which its effects have probably been most pernicious. The persistent assumption of a high level of Soviet threat, despite a balance in military capabilities which for most of the post-war period favored the West, pandered to a visceral anti-communism that is already endemic in many sectors of the American community. The threat of punishment which was a central part of deterrence fostered the impression of America the magistrate and Russia the lawbreaker, reinforcing the belief that the problem was one of good and evil rather than conflicting ideologies and interests. The ‘credibility’ of deterrence required that political ‘will’ should be reasserted at frequent intervals, and for 40 years the American people have been subjected to propaganda as assiduously as their Soviet counterparts.15 Meanwhile the irrefutable slogans of ‘deterring war’ and ‘deterring the Soviets’ persuaded Congress to invest in massive inventories of strategic nuclear weapons, which the Soviet Union duly matched. However amicable the objective state of Soviet-American relations, the level of armaments appeared to deny the reality of the situation. This apparent contradiction could not be sustained and, following the dictates of cognitive
36 DILEMMAS OF NUCLEAR STRATEGY
dissonance, the perceptions of threat on both sides moved back into line with the weapons inventories. Third, deterrence theory has stunted the development of US foreign policy. Because deterrence and its corollary, containment, have remained the dominating concept in foreign and defense policy, perceptions of threat among the public at large have remained frozen in the mold of the 1947–53 period. The theory of deterrence has provided the intellectual casing that makes this world vies impervious to contradictory evidence and analyses, even for some of the decision-making elites. The emphasis on containment and deterrence had the result that policy towards the Soviet Union has been limited to a narrow range of predominantly negative objectives. The theory fosters an assertive, apolitical style which favors preventive and punitive instruments, is distrustful of negotiations, and sees compromise as weakness. The possibility that US interests might be served by cooperating with the Soviet Union in dealing with problems around the world is excluded from consideration, and long-term objectives designed to help shape future Soviet policies are simply not addressed. Fouth, deterrence theory has fueled the arms race, rather than checked it. The belief that a sudden build-up in US strategic weapons in the early 1960s would ‘improve the stability of deterrence’ fostered the unwarranted assumption that the Soviet Union would accept it, on the grounds that this was in the interests of peace. 16 The reassurers, meanwhile, were concerned with relative rather than aggregate levels of weapons and, in any case, large inventories were theoretically more stable than small ones. It was the deterrers who focused attention on numbers when they saw that the Soviet Union would not only match the US build-up but surpass it. Arms control policy moved to cap the number of Soviet missiles, while the deterrers tried to regain their original advantage by multiplying the number of US warheads.17 As the Soviet Union moved to catch up in that field too, the deterrers sought relative advantage in the form of added diversity. Their arguments could be summed up as ‘more is better’. They redefined the requirements for stability, and, by using carefully selected units of account, were able to exploit reassurance theory in such a way as to discredit arms control and to demonstrate an urgent requirements for new offensive weapons programs. And since it was the concept of mutual assured destruction that had justified the treaty banning antiballistic missiles (the 1972 ABM treaty), once that concept had been discredited, the theoretical constraints on ballistic missile defense systems were weakened. The temptation to open the Pandora’s box of space-based systems became irresistible to many. The fifth indictment is that military strategy itself was driven from the so-called strategic debate. It was replaced in part by more taugible concepts like ‘force exchange ratios’, ‘damage expectancies’ and ‘kill progabilities’, which were amenable to mathematical permutations. But the debate also focused on the theoretical elaboration of deterrence and reassurance, and pondered the rationality of irrationality and the potential for compellance. The mathematical models were excellent for generating hard numbers, and anchored the debate at the level where Americans are most comfortable: the technical and budgetary aspects of force requirements. The theoretical elaborations applied Cartesian logic to complex problems of political psychology, resulting in bad politics and bad strategy, as the United States found to its cost in Vietnam.
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 37
In Europe, the debate came to be dominated by questions of deterrence, escalation, firebreaks, coupling, and decoupling. Military strategy was largely a side issue. The NATO concept of ‘flexible response’ adopted in 1967 may have made sense in terms of deterrence theory, but it offered the Soviet Union a strategic opportunity which contributed to its adoption of a new set of military objectives, and justified the radical restructuring of the Soviet armed forces and the introduction of fundamentally new concepts of operations.18 The politically divisive deployment of Euromissiles, decided on in 1979 and initially implemented in 1983, was seen by the Europeans as strengthening the coupling of Europe to America, by the Soviet Union as dangerously destabilizing, and by the Americans as a test of European loyalty to the alliance. Sixth, deterrence dogma encouraged unwarranted complacency and diverted public attention and intellectual resources from more fundamental problems. By defining crisis stability as the critical aspect of the nuclear problem it diverted attention from the avoidance of crises and, hence, the central importance of the SovietAmerican political relationship as the potential source of global war. By focusing exclusively on the possibility of Soviet aggression, the theory obscured the fact that was itself was the greater danger, and could be made more likely by the measures being taken to deter aggression. By claiming to have solved the problems of nuclear weapons, deterrence dogma dissipated the sudden urgency that this devastating capability had brought to the search for new ways of managing interstate relations. The pressure for a new approach had been building up for 50 years and more, as the two world wars had demonstrated the inherent limits of the existing international system. But the steam was let out of the movement to devise a new international order by the false promise of deterrence, and the abstract style of reasoning and ‘tough-minded thinking’ basic to the dogma were not receptive to radical ideas about conflict and cooperation. Seventh, deterrence dogma provided a pseudo-ethical facade for policies which were inherently unethical. A fundamental distinction between Marxist-Leninist and Judaeo-Christian theory lies in their differing views on the relationship between ends and means. By depicting the Soviet Union as bent on territorial aggression, and insisting that war was inevitable without the threat of nuclear devastation, deterrence dogma led the West to adopt the Soviet tenet that ends justify means. The ‘means’ were the military capability and the political will to devastate the Soviet Union, and the logic of the dogma justified, even demanded, an inexorable growth in nuclear capability, while at the same time it allowed us to claim moral rectitude in choosing counter-force over counter-value, or the arming of space over mutual assured destruction. As nuclear inventories grew, so did the West’s explicit threat to the Soviet Union evolve into an implicit threat to the survival of the human race; but the frame of mind of deterrence dogma made it ‘moral’ to put the world at risk in our search for total security.
The Justification for Deterrence The scope of these indictments may be broader than some can accept; but the comforting belief that deterrence dogma has done no harm is clearly untenable. What, then, were the benefits which justified these significant costs? In particular, what of the claim that deterrence has kept the peace since the Second World War?
38 DILEMMAS OF NUCLEAR STRATEGY
This cannot be disproved, but neither can the counter-claim that nuclear deterrence was not responsible for keeping the peace. As the policy of nuclear deterrence has been seen to impose significant costs, the conclusion ‘better safe than sorry’ is clearly invalid. It is therefore necessary to dissect the two conflicting claims and assess their relative probability. The original purpose of the deterrence policy was based on the assumption that the Soviet Union was set on military aggression which would lead to world war unless it was deterred. There were, of course, a number of cases where communist aggression was not deterred by the US nuclear capability: the Korean war 1950–53, Vietnam 1961–73, Hungary in 1956, Czechoslovakia in 1968, and Afghanistan since 1979. The two cases in the 1950s when the nuclear threat was used to obtain Chinese compliance are irrelevant to this particular argument.19 One is left, then, with the narrow claim that without the nuclear deterrence, the Soviet Union would have invaded NATO Europe, precipitating the Third World War. For that claim to be true, it should be self-evident that nothing less than nuclear punishment could have deterred such Soviet aggression. Given the devastation of Russia and its people in the Second World War, is one to believe that the prize of a vanquished Europe would have been worth the cost of another conventional war? What of the effect on other Soviet objectives, such as the attempt to build up on the power and prosperity of the Russian homeland, the progress of the world communist movement, and Soviet standing in the competition for world influence? The assumptions which underlie the claim reflect the war-clouded perceptions of 1947–50 rather than a realistic appraisal of the costs and benefits of such action to the Soviet Union. The claim is a classic example of the colonel’s fallacy and assumes a simple desire on the part of the Soviet Union to achieve territorial conquest without explaining how such a conquest would serve Soviet interests. Such an objective is not supported by historical analysis of Soviet behavior in the last 60 years, or by Soviet doctrine. In the wake of the Second World War the Soviet Union withdrew from Arctic Norway, Finland, Czechoslovakia, Yugoslavia, Bulgaria, Azerbaijan and Manchuria. Roughly ten years later it withdrew from Austria and gave up bases at Porkala and Port Arthur. This is not to suggest that the Soviet Union is benevolently inclined. It certainly means to maintain control of its national security zone, as was shown in Hungary, Czechoslovakia, and Afghanistan. But, as these cases demonstrated—to say nothing of others when direct military intervention was not resorted to—the military instrument is not the first one it reached for, and the record argues that the Soviet Union is not possessed by a ‘Napoleonic’ or ‘Hitlerian’ urge for military expansion. Many people now accept that the threat of Soviet aggression has been exaggerated, but insist that there is still a role for nuclear deterrence in preventing war. They postulate a situation where the Soviet Union might find itself in a complex crisis in which it might seem a lesser evil to resort to war, unless there was a high probability that it would escalate to an intercontinental exchange. This argument cannot be refuted by calculations made in a narrow set of circumstances and isolated in space and time. But crises are not isolated in that way, and this argument exemplifies the point under attack in this article. By focusing on a hypothetical worst case where war would be a lesser evil, it ignores the possibility that different US policy might encourage a less restrictive range of Soviet options. The argument concentrates on managing a crisis, rather than creating the
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 39
circumstances that would make such crises less likely. Above all, it ignores the effect of a deterrence-based policy on Soviet-American relations and the likelihood of war. This point is the nub of the argument. The advocates of nuclear deterrence implicitly assume that war can only come about as the result of a Soviet initiative. Originally, this initiative was seen as deliberate military aggression, but now it is also seen as possibly involving a resort to war as the lesser evil. It also assumed that any such resort to war can be deterred by the threat of nuclear escalation. Those who do not share these assumptions argue that it is precisely this obsession with the remote possibility of extreme cases (which may in any case be inherently uncontrollable) that increases the danger of war. The steady build-up of nuclear weapons to deter such hypothetical Soviet initiatives actually makes war more likely through accident, miscalculation, or the increased tension and tougher political posturing that go with such policies. The advocates of deterrence talk of necessary insurance. Their opponents can counter with the analogy of the man who is obsessed with the danger of a meteor strike: he steadily increases the thickness of his roof until his house collapses, himself and his family inside.
A Fundamentally Different Frame of Mind The final defense of the deterrent dogmatist is to ask, ‘What would you put in its place?’ The short answer is, ‘nothing’. The immediate objective is to reduce the salience of deterrence dogma in the formulation of policy, and allow the excessive space it now occupies to be taken over by traditional approaches to foreign and defense policy. These policies will, of course, be enriched by a more subtle understanding of the classic concepts of deterrence and reassurance, and by a deeper awareness of human frailty under stress in crises. But we should get it quite clear that the achievement of the last 40 years is to have avoided war despite the negative effect of a deterrence-based policy, and that the grail of peace does not lie in that direction. However, the situation is more complex than that. The problem stems as much from the attitudes engendered by the dogma as from the dogma itself. This way of thinking now pervades the Western political-military establishment. What kind of ideas can take the place of these attitudes which have proved so counter-productive? It will not be easy to change a frame of mind which is congenial to our way of thinking, panders to our prejudices, and reflects the conventional wisdom of 40 years. But we might start with three general principles. In the first place, we should recognize that it is world war, however it starts, not Soviet aggression, that poses the greatest threat to all our people. We should therefore pay less attention to developing the military means to deter the onset of war, and concentrate more on developing the political means of averting those situations that make war more likely. And in devising ways to counter the Soviet Union’s offensive capability, we must be sensitive to whether these could increase the tendency to war, in the long run as well as the short. Second, we should acknowledge that for all except the most extreme circumstances (which are probably inherently uncontrollable) mutual deterrence is an objective fact. We should also recognize that as long as large weapons inventories
40 DILEMMAS OF NUCLEAR STRATEGY
remain in place, cognitive dissonance will inhibit improved Soviet-American relations. We should therefore approach arms control with the objective of significantly reducing the existing number of nuclear weapons and blocking new avenues for arms racing, rather than seeking to shape the arsenals to deter, or, if necessary, control nuclear war. Third, we should recognize that the danger of war stems from the adversarial nature of the Soviet-American relationship, not from their nuclear arsenals. Worstcase scenarios masquerading as analyses of Soviet intentions reinforce the tendencies to conflict in the relationship and stifle the cooperative ones, increasing the likelihood of war. The state of Soviet-American relations bears directly on the peace of the world, and it should be explicitly acknowledged that the relationship is too important to be manipulated for domestic political gain or tactical advantage in the superpower competition. In sum—repeated Western attempts to find military solutions to the central problem of East-West relations have only served to worsen the situation. The problem is primarily political—and so must be our solution.
NOTES I am grateful to Catherine Kelleher, Ray Garthoff, and Dick Betts for their helpful comments on this article, which was originally published in International Affairs (London). 1. This is based on my own observations during the 1950s and 1960s. They are supported by the later research of William Kincade and Stephen Sonnenberg for the ‘Psychology of Deterrence Project’, whose findings are currently being prepared for publication. 2. This term, coined in 1972, is not part of the established strategic lexicon. The concept of reassuring an adversary about one’s intentions is, however, at the root of Western arms control theory, which focuses largely on restraining the pressures leading to pre-emption in crisis situations. This concept of mutual reassurance between adversaries is quite distinct from the role of a deterrent capability in reassuring one’s allies. 3. Arthur A.Schlesinger, A Thousand Days: John F.Kennedy in the White House (London: Andre Deutsch, 1965), p. 301, describing a discussion between Walt Rostow and Deputy Foreign Minister Vasily Kuznetsov in December 1960. 4. Michael MccGwire, Soviet Military Objectives (Washington, DC: Brookings, forthcoming), Ch. 2. 5. This question was raised by Thomas L.Hughes at the ‘Psychology of Deterrence’ workshop in December 1984. 6. For a preliminary attempt, see Michael MccGwire, ‘Soviet Military Requirements’ in Congress Joint Economic Committee, Soviet Military Economic Relations: Proceedings of a Workshop on 7–8 July 1982 (Washington, DC: U.S. Government Printing Office, 1983). See also MccGwire, Soviet Military Objectives. The short answer is that the Soviet Union has enough to cause the West justifiable concern, but not ‘more than it needs’.
DETERRENCE: THE PROBLEM—NOT THE SOLUTION 41
7. A successful example was South Africa’s halting of its planned nuclear test in response to US diplomatic pressure using Soviet-source intelligence. A potential opportunity was the initial state of the Iran–Iraq war. 8. The difficulty in re-establishing relations with China for many years as an example of this phenomenon. 9. The claim that the United States would be faced by a ‘window of vulnerability’ in the 1980s was based on such assumptions. 10. This requirement was publicized in Presidential Directive No. 59 in June 1980, deliberately leaked by the Carter administration. It was emphasized that PD 59 was not a new departure, but an elaboration of the doctrine announced by Schlesinger in 1974, which in turn was a clarification of existing operational policy. 11. MccGwire, Soviet Military Objectives, Chs. 2 and 4. 12. This distinction was first drawn by Raymond Garthoff when comparing US and Soviet approaches to arms control. 13. The same way of thinking underlay the 1946 Baruch Plan for controlling atomic weapons by proposing an international body to oversee atomic and later nuclear developments. 14. See MccGwire, Soviet Military Objectives, Ch. 10. 15. Cora Bell talks of ‘the Cold War image of Russian society, constructed with some deliberation in the West’, in Cora Bell, Negotiating from Strength (New York: Knopf, 1963; London: Chatto & Windus, 1962), p. 239. 16. Schlesinger, A Thousand Days, citing Walt Rostow. See also Raymond L.Garthoff, Intelligence Assessment and Policymaking: A Decision Point in the Kennedy Administration (Washington, DC: Brookings, 1984), pp. 24–6. 17. From 4,000 in 1970 to 9,200 in 1979. 18. This process started in 1968 and was largely complete by 1975: see MccGwire, Soviet Military Objectives, Chs. 2 and 3. 19. Barry Blechman considers that the nuclear threat was probably effective in 1958 against China in the case of the Offshore Islands, and could possibly have played a role in breaking the impasse in the Korean armistice talks, although there are other plausible explanations. His research shows that the Soviet withdrawal from Iranian Aserbaijan in 1946 was not under nuclear duress. 20. For instance, Yugoslavia since 1949, Poland in 1956, Albania since 1960, Romania since 1963, Poland 1980–81. See also Ken Booth, The Military Instrument in Soviet Foreign Policy 1917–72 (London: Royal United Services Institute for Defence Studies, 1973).
42
Nuclear Strategy: Can There Be a Happy Ending? Fred Charles Ikle
Our policies for preventing nuclear war, mercifully, have never been tested to the point of failure. This cannot be said of our policies that have sought to reverse, or at least to halt, the expansion of nuclear arsenals threatening our annihilation. For more than two decades now, the United States, using both diplomacy and self-restraint, has tried—and failed—to halt the nuclear arms competition with the Soviet Union. The most important reason for seeking to manage and control the nuclear arms competition is to reduce the risk of nuclear war. The US pursuit of these linked objectives, however, has been under the spell of a particular theory from the mid-1960s until recently. This theory holds that ‘the two sides’ can limit or reduce their nuclear weaponry if, and only if, they both consent to leave their territory totally vulnerable to each other’s nuclear attack. Such consensual vulnerability constitutes ‘strategic stability’, according to the theory, and is the best, if not the only, way to prevent nuclear war for the indefinite future. In the real world, nuclear forces are built and managed not by two indistinguishable ‘sides’, but by very distinct governments and military organizations. These, in turn, are run by people, people who are ignorant of many facts, people who can be gripped by anger or fear, people who make mistakes—sometimes dreadful mistakes. The intellectual poverty of the consensual vulnerability theory accounts for much of the current opposition to strategic defense. Of course, like most theories than gain throngs of adherents, this theory contains a kernel of truth. Up to a point, the nuclear age has led to ‘two sides’ mutually vulnerable to each other’s retaliation. So any military expert may say some valid things about mutual nuclear vulnerability without considering how governments really function, or how people think and act. So too, any surveyor can measure a plot of land without considering the shape of the earth. But we cannot navigate to a distant haven if we assume the earth is flat. Even the purely technical dimension of the nuclear age is often over-simplified, as if to fit a flat-earth theory. Today’s opponents of strategic defense argue that their concept—‘two sides’ agreeing to a stable relationship of consensual vulnerability— reflects not a choice of doctrine but an inescapable technological fact. The mutual threat of assured destruction as the guarantor of peace, they say, is unchangeable because nuclear weapons cannot be disinvented.
* This article is reprinted by kind permission of Foreign Affairs (Spring 1985).
44 DILEMMAS OF NUCLEAR STRATEGY
Although nuclear weapons cannot be disinvented, other things are being invented. To say that the immense destructiveness of nuclear weapons makes the nuclear era unique is but a partial truth. As we look back in human history, to be sure, nuclear arms represent a unique change in the meaning of war and peace. For the future, however, we must expect other changes of perhaps comparable scope and consequence. Technology does not stand still, not stay put in a few countries. And so heavenly law operates in this world to constrict the results of technological advances to ‘two sides’. It is important, therefore, to separate technological facts from doctrinal fictions. It is a fact that modern technology provides means that could be used to destroy people, wealth and even parts of nature with a totally unprecedented speed, intensity and magnitude. It is fiction, however, that the nuclear arsenals of the two superpowers are the only such means we need to be concerned about. It is a fact that other devices for mass destruction, existing or yet to be invented, could be delivered to an intended target in ways that are difficult, or perhaps too costly, to prevent. To conclude, however, that the only practical way to protect against destruction is to threaten retaliation in kind reflects not a permanent fact of technology but a choice of doctrine. The sequence in which science presents governments with possibility for new devices of destruction or new means to counter them is a vagary of history. Imagine that early in this century the development of science had taken a different turn. Biology and genetic engineering might have made dramatic advances, instead of the great progress in physics that then occurred. As a result, a different ‘Manhattan Project’ in the Second World War could have produced a radically new biological weapon, exploiting the destructive potential of genetic engineering that people have begun to worry about today. Such a weapon could be far more efficient for causing mass destruction than the biological agents that the United States eliminated in the early 1970s, in compliance with the biological weapons treaty. That treaty, we recall, permits defense measures but prohibits testing and stockpiling of offensive biological weapons—the reverse logic of the Anti-Ballistic Missile (ABM) Treaty. Today’s opponents of ballistic missile defenses often argue that such defenses are useless, since the ‘other side’ could always smuggle nuclear bombs into our cities— the suitcase bomb. Hence, they conclude we must rely exclusively on our ability to retaliate. None of these opponents of defense, surely, would choose to ban defenses and rely only on retaliation in kind if the ‘other side’ developed powerful new biological weapons—whether or not these could be delivered in suitcases. Yes, ballistic missile defenses could be circumvented. There are many ways to sneak a nuclear explosive into a city. Some day such an act might be attempted by a terrorist organization, or by a nuclear power that pursues irrational terrorist ends. A US-Soviet relationship of consensual vulnerability would afford absolutely no protection against such a disaster. On the contrary, it might make it more likely or even aggravate its consequences. Thus, it is not the destructiveness of a particular technology, but vagaries of both technological development and of American strategic thought, that have led to the excessive emphasis on retaliation and to the neglect of defenses in the current strategic posture of the United States. The notion of a stable strategic relationship based on the acceptance of mutual vulnerability reflects not a state of nature, but a state of mind.
NUCLEAR STRATEGY: CAN THERE BE A HAPPY ENDING? 45
Whenever the proponents of a theory resist new facts while attacking all alternative explanations as heresies, that theory turns into dogma. At that point, the proponents often try to buttress their case by distorting history. Critics of President Reagan’s initiatives on strategic defense have charged that his proposed defenses would upset a stable relationship of mutual deterrence that has obtained throughout the nuclear era. But this stable equilibrium of ‘two sides’ defenselessly poised for mutual assured destruction has never existed. For the first five years of the nuclear era, there was no stable balance of nuclear forces at all—the United States had a monopoly. For the next 15 years, neither the United States nor the Soviet Union eschewed defenses against nuclear attack in pursuit of a mutual vulnerability theory. In the 1950s, the United States spent about twothirds as much on air defenses as on nuclear offensive forces, some $200 billion in ten years (in today’s dollars). In 1960, the United States actually allocated more to its defenses against Soviet nuclear attack than to its nuclear offensive forces. Only in the 1960s did the theorized ‘stable’ balance of consensual vulnerability become a partial reality, in the very limited sense of shaping US force planning and US arms control policy. Beholden to the ‘flat earth’ theory, we saw to it that we would be proven right in our own garden. For two decades we shrank our budget for nuclear offensive forces nearly every year. We reduced expenditures on our defenses against nuclear attack drastically, and after 1970, we cut them practically to zero. And, most dangerous of all, we permitted our intelligence projections for Soviet forces to become warped by our own dogma. In particular, from the mid-1960s to the early 1970s, we misled ourselves by the mistaken forecast that the Soviet Union, in light of our self-restraint, would not want to overtake us in nuclear offensive forces, much less seek a capability for destroying most of our deterrent strength. The 1972 treaty curbing ballistic missile defense became the apotheosis of this belief. In the American view, it was meant to seal a Faustian bargain with the devil of nuclear destructiveness. By the terms of this bargain, both sides had to guarantee their own devastation on a possible Judgement Day, when one side having been attacked by the other, would want to avenge this attack. Thus, Judgement Day would never dawn. In American eyes, the implied bargain consisted of two parts: both sides would abstain from defenses that could destroy the opponent’s revenge forces after they are launched; and both sides would abstain from the kind of of fensive capabilities that could destroy the opponent’s revenge forces before they are launched. In the jargon, the former are called ‘active defenses’ the latter, ‘counterforce’ capabilities. Only the United States kept one part of this two-part bargain. Only the United States gave up on active defenses, again aircraft as well as against ballistic missiles. The Soviet Union, by contrast, has spent slightly more on active defenses since the signing of the ABM Treaty than on its nuclear offensive forces. Some of this effort went into a vigorous program for ballistic missile defense (both deployment of the treatypermitted Moscow system, and the development of future systems); the bulk of it went into air defense. The United States initially intended to keep the counterforce part of the bargain. From 1970 to 1972, the US Congress voted against improving the accuracy of our missiles in order to deny ourselves the ability to destroy Soviet missiles in their silos. But the Soviet Union continued its expensive build-up of nuclear forces, including enhanced versions of the SS-18 and SS-19 intercontinental ballistic missiles, contrary
46 DILEMMAS OF NUCLEAR STRATEGY
to a policy of eschewing counterforce. So, by the end of the last decade, as the Soviet built-up in counterforce capability continued unabated, we felt compelled to respond in kind. As the President’s Commission on Strategic Forces (the Scowcroft commission) put it in 1983: ‘…we cannot safely permit a situation to continue wherein the Soviets have the capability promptly to destroy a range of hardened military targets and we do not’. Clearly, our arms control efforts had failed to halt competition in counterforce capabilities. Thus came to naught the leading concept for strategic arms control that reigned in the United States from the mid-1960s until recently. By now, this story has often been told. Yet many Americans and Europeans still pine for the Faustian bargain that they believe the ABM Treaty had sealed. In a triumph of hope over experience, they want to try again. It is important to understand the essence of these painful lessons. The strategic world does not consist of ‘two sides’ each willing to be locked into a symmetry of utter vulnerability. Even though we tried to conform to this dogma, the ‘other side’, that is to say, the real Soviet leaders and their real military advisers, did not. These men made very effort to reduce the vulnerability of their country—and especially of themselves. They ordered the construction of a great many hardened command posts, located well away from cities, for military commanders and officials at all levels of the government and Communist Party. They maintained a vigorous effort to improve their air defenses and develop missile defenses. And, above all, they deployed missiles increasingly capable of destroying our nuclear forces, thus equipping themselves to eliminate the cause, the very source, of their own vulnerability. Why did not the Soviet leaders act in accordance with our theory? Perhaps they did not want to entrust forever the future of Soviet communism and the Soviet empire to the rationality of American leadership. Perhaps they did not want to rely on the discipline and technology of the American military forever to prevent a cataclysmic accident. Perhaps, in their eyes, American policy seemed incoherent—and, hence, made them skeptical of our intent—since it combined the American doctrine of ‘stable’ mutual vulnerability with the NATO deterrence policy that includes the threat of using nuclear weapons first, in response to a massive conventional attack. But it is also possible that they saw in the absence of US ballistic missile defense an invitation to acquire the missilery capable of destroying the US land-based deterrent. Those who remain opposed to strategic defense argue that missile defenses would merely stimulate the ‘action-reaction cycle of the arms race’. The Soviet leaders, they maintain, would feel compelled to thwart our active defenses. Up to a point, to be sure, this may be a valid prediction. For a short stretch of our voyage, it is safe to assume the earth is flat. For a while, yes, the Soviet leaders undoubtedly would try to thwart our active defenses, much as they are trying to thwart passive defenses such as hardening and mobility that protect our deterrent today. By increasing the numbers, accuracy and destructive power of their missile silos; by expanding their air defenses, they try to counter the survivability of our bombers; by stepping up their antisubmarine program, they try to negate the protective mobility and quietness of our nuclear missile submarines. Here is a classic action-reaction cycle in armaments. Ought the United States to desist from improving these protective measures for its deterrent forces? Would anyone advocate such US self-restraint—say, a halt in the hardening of our missile silos or in maintaining the protective elusiveness of our submarines—‘to slow down
NUCLEAR STRATEGY: CAN THERE BE A HAPPY ENDING? 47
the arms rade’? In this country, most opponents of strategic defenses would not endorse such an idea. On the contrary, they postulate a stubborn determination on each side to protect its retaliatory forces. Yet their dogma excludes the possibility that one side might have, as well, a stubborn determination to protect its homeland. Would the Soviet leaders, in fact, augment their offensive missile capabilities in order to thwart our defenses? This will depend on the ability of our defensive system to impose costs and difficulties sufficient to dissuade them from doing so. Also, very importantly, it will depend on our presenting to the Soviet leaders alternatives for their nuclear policy that they can find acceptable. President Reagan’s initiative seeks to meet both these conditions. It offers the Soviet leaders a new strategic relationship with the United States that will better meet their fundamental security requirements. And it ushers in new technologies for defense that hold the promise of making counter-measures unattractively costly. A case in point here may be the history of the Soviet bomber program. When we invested heavily in air defenses in the late 1950s, the Soviet Union ceased building new bombers and instead shifted to ballistic missiles. But, when it became clear that we had given up on air defenses, the Soviet gain invested in the development and then the production of a new strategic bomber. They decided to build new vehicles to take advantage of the freeway we opened for their bombers. By analogy, if we began to close the freeway in space that we have preserved for ballistic missiles, the Soviet military might stop investing in new ballistic missiles and instead rely on new bombers and cruise missiles. That in itself would be a great gain. Bombers and cruise missiles, compared with ballistic missiles, are less suited for surprise attack because of their longer time of travel. And bombers are safer as deterrent forces since, by takine to the air, they can be made nearly invulnerable in an alert, yet could safely return to their bases should the warning be false. Also, using in part the same technologies and systems, air defenses could later complement our ballistic missile defenses. In sum, throughout the nuclear era, all the evidence shows that the US-Soviet strategic relationship never conformed to the theory of stable consensual vulnerability—either before or since the ABM Treaty. Furthermore, contrary to that theory, the Soviet government has not designed and financed its vast nuclear build-up just to maintain forces for destroying American society. Instead, by acquiring active defenses and counterforce capabilities, it has shown its determination to reduce Soviet vulnerability rather than depend exclusively on such a threat of revenge. Unable to achieve a stable equilibrium of vulnerability between us and the Soviet Union in the past, are we likely to achieve such a strategic order in the future? Compelling reasons make this a most improbable outcome. If we put the flesh and bones of real governments on the abstraction of ‘two sides’, and if we recognize that these governments must function in the real world of many powers—of allies, of captive nations, of friends and foes—then we can readily see why such a two-sided equilibrium of mutual vulnerability is a relationship that cannot remain stable. To begin with, the makers of Soviet foreign policy take a long-term view. Even if they were prepared to trust the US government always to play the role of the reliable partner in such an equilibrium, they would feel it essential to take account of other nuclear powers, present or future. The planners in Moscow surely recognize that, despite the astonishingly successful American-Soviet cooperation in curbing the
48 DILEMMAS OF NUCLEAR STRATEGY
spread of nuclear weapons, the capacity to manufacture nuclear explosives will continue to spread to additional countries. They must be aware, for example, that they once helped China to acquire a nuclear capability and are now facing a slowly expanding strategic nuclear force across the world’s longest border. But, quite apart from the fact that the arena of nuclear powers does not consist of only two sides, there is a more fundamental reason why we should not expect the Soviet leadership to settle for the strategic order of consensual vulnerability. Like oil and water, the two ingredients in this order will always separate: the accord on a stable equilibrium of mutual restraint is psychologically incompatible with the constant threat of reciprocal annihilation. The first ingredient of this mixture represents the best in international relations: a continued willingness to cooperate in restraining one’s own military power, coupled with a serene reliance on the opponent’s prudence and his common sense. The second ingredient of the mixture represents the worst in international relations: an endless effort to maintain forces that are constantly ready to annihilate the opponent, couples with an unremitting determination to deny him escape him this grip of terror. The believers in the dogma of a stable, mutually agreed vulnerability fail to appreciate the dynanmic of this incompatibility. Their vaunted strategic order is unstable at its very core. We cannot keep the balance of mutual threats of mass destruction from ceaselessly tilting. Yes, we can try to adjust this quivering balance, year after year, augmenting our offensive nuclear forces, substituting more effective or more survivable nuclear arms for older ones, in order to defeat the inevitable Soviet attempts to reduce the magnitude of the American nuclear threat to themselves. This has been our policy in past decades. Today’s critrics of strategic defense argue that the Soviet rulers would never accept a strategic order in which the United States and its allies were no longer vulnerable to nuclear mass destruction, even if these rulers could acquire a similarly effective defense of their own homeland. Yet the critics firmly believe that these same Soviet rulers, whom they view as hellbent on maintaining their threat of our destruction, will forever nestle down with us in an equilibrium of mutual vulnerability. By accepting such an equilibrium, the Soviet rulers would have to accept indefinitely a future where any American president, or (in their eyes) perhaps even ‘some American general’, could unleash the engines that would destroy the Soviet Union. To the men in the Kremlin, such a world cannot look like a place of stability. It cannot be the chosen destination of Soviet policy. As Soviet Foreign Minister Andrei Gromyko put it in 1962: ‘…to base the policy on states on a feeling of universal fear would be tantamount to keeping the world in a permanent state of feverish tension and a hysteria anticipating war. In such an atmosphere, each state would fear that the other side would lose its nerve and fire the first shot.’ Contrary to Gromyko’s protestations today, the 1972 ABM Treaty did not convert the Soviet government to a strategic order of consensual vulnerability; Soviet investments in offensive and defense systems speak louder than current denials. One cannot address the future of our nuclear strategy without considering the role of American nuclear weapons in NATO’s defense policy. Ever since the Yalta agreement 40 years ago, a major objective of the United States has been to prevent Soviet domination of Western Europe. This imperative has confronted a stubborn problem: for an attack on Western Europe, the Red Army could amass manpower and firepower that always seemed superior to the forces the Western European nations
NUCLEAR STRATEGY: CAN THERE BE A HAPPY ENDING? 49
could muster. The creation of the Atlantic alliance provided but a partial solution. Even the combined forces of the alliance, the allied governments have held, could not be kept at a strength sufficient to defeat a massive and sustained Soviet attack in a conventional war. Hence it has always been agreed within the alliance that nuclear weapons must play an essential role, whether in deterring conventional attack, or in coping with such an attack should deterrence fail. This role, in turn, has exerted a pervasive effect on the development of our nuclear forces, strategy and arms control policy—an effect far greater than is often realized. Deterring or responding to a conventional invasion of Western Europe was the primary mission of the US Strategic Air Command well into the 1950s, and throughout the 1950s the United States opposed negotiating limits on its nuclear forces without a reduction in the conventional threat to Europe. It has proven difficult, however, to agree on a single, coherent concept for this role of nuclear arms in protecting allies. The political leaders, defense officials and military staffs of the Atlantic alliance have shaped and reshaped, time after time, several strategic concepts. The initial concept exploited the Amerian advantage in nuclear weapons. During the short-lived American monopoly, the application of this concept, in theory seemed simple. In the event the Red Army invaded Western Europe, the US Air Force would drop atomic bombs on targets in the Soviet Union. And later, when the United States had acquired a nuclear arsenal and delivery systems able to destroy the Soviet Union as a military power, it could not be certain of escaping grievous damage at home from Soviet retaliation. The Eisenhower administration agonized long and hard over this dangerous dilemma. So a second concept was born, which envisioned using nuclear weapons on the battlefield, not to retaliate, but to repel the invading divisions. This concept held to a more traditional, a more ‘conventional’, application of military power. As Secretary of Defense Charles Wilson put it in 1957, ‘The smaller atomic weapons, the tactical weapons, in a sense have now become conventional weapons…’. As long as the United States enjoyed an effective advantage in nuclear forces, these two concepts could lend coherence to the overall strategy. In the jargon of the nuclear theorists, we enjoyed ‘escalation dominance’: we thought we had more, or more effective, tactical nuclear weapons, so we could expect to deter, or if necessary defeat, a Soviet invasion by using tactical nuclear weapons; and we had superior global nuclear forces, so we could expect to deter the Soviet Union from using its intercontinental nuclear forces against us. Today, this ‘escalation dominance’ has been overtaken by the massive changes in nuclear arsenals. Indeed, on both sides of the Atlantic, it is no longer judged to be a solution. The United States patently has no plans to regain its former advantage in global nuclear arms, or to restore the West’s superiority in tactical nuclear arms. A few years ago, in fact, political opinion in Europe opposed even the deployment of the so-called ‘neutron bomb’, a weapon capable of killing enemy soldiers in their tanks without destroying the nearby towns. Yet, the more these tactical nuclear weapons can destroy the intended military target without destroying what they would seek to defend, the greater would seem their credibility as a deterrent. Despite the massive changes since the end of the Second World War, there is wide agreement on both sides of the Atlantic that nuclear weapons continue to contribute to the deterrence of a conventional invasion of Western Europe. This consensus
50 DILEMMAS OF NUCLEAR STRATEGY
comes easy. However, the concepts guiding the use of nuclear weapons should such deterrence fail are more painful to spell out. Some of Europe’s political leaders and elected representatives understandably prefer to avoid the issue. Some assert that NATO’s nuclear weapons must never be used to defeat invading Soviet divisions, but only to preclude armed aggression by threatening massive nuclear revenge. Here we meet the old problem bedeviling the theory of deterrence: the credibility of threatening an irrational act. If a conventional war ever broke out and the invading divisions pressed forward, could the act of nuclear revenge serve the national interest of any member of the alliance? If not, will the threat be credible in time of crisis? Folly likes company. Many of those who would constrict NATO’s nuclear policy to such a threat of revenge also oppose missile defenses, clinging to the dogma of a consensual vulnerability. And, to boot, they often fail to support increased conventional strength, thus leaning even more heavily on their dubious nuclear threat. By rejecting missile defenses for Europe, they would leave many critical elements of NATO’s conventional and nuclear forces vulnerable to shorter-range Soviet missiles, which—even if conventionally armed—could be accurate enough a decade hence to destroy some of NATO’s key facilities. In essence, they would rely almost exclusively on the global nuclear forces of the United States to provide the comprehensive guarantee that protects Europe. In the long run, the force of this guarantee depends on its support by the American people. Some European opponents of missile defenses for the United States appear to believe that to assure the vulnerability of the United States somehow preserves this guarantee. This logic makes no sense. It implies that the larger the number of Americans who would be killed in a nuclear war, the more credible a decision by the US president to respond in a conventional invasion of Europe by initiating nuclear war. Now, it must be said that a free alliance of democratic nations is a very ecumenical association. Many contradictory beliefs can coexist—provided the alliance is at peace. It is in time of war, or even during a major crisis, that the coherence of the allied strategy would be put to the final test. Our democracies will have to defend themselves for the foreseeable future against the military might and political ambitions of the Soviet ruling class, still dominated by men imbued with Lenin’s totalitarian philosophy. And who can foretell whether other totalitarian powers might not emerge as major military threats to Western democracy? Against such opponents, democratic governments cannot keep up the necessary strength, decade after decade, without the continuing support of their people. But the men and women who work and vote in our democracies have never been tested against the searing edge of nuclear war, nor have the politicians, soldiers and scientists who must dedicate themselves to sustain our military strength. Only the effects of low-yield nuclear weapons on isolated military targets can be predicted with some confidence, having been tested realistically. Much remains unknown. Sometimes a particular horror is overlooked, sometimes it is exaggerated. Possibly critical effects that might result from large-scale nuclear attacks have been discovered only by accident. For example, in 1954, one of our thermonuclear tests in the Pacific accidentally disclosed the potential extent of radioactive fall-out; the medical histories of the injured people subsequently disclosed the human consequences of fall-out; a high
NUCLEAR STRATEGY: CAN THERE BE A HAPPY ENDING? 51
altitude test in 1958 led to the accidental discovery of electromagnetic blackout; in the early 1970s, totally unrelated studies revealed how high-yield nuclear explosions might damage the ozone layer; and in the early 1980s, new studies opened the possibility of ‘nuclear winner’. Naturally, such revelations awaken those deep anxieties about nuclear war that slumber in our minds. For many people they strengthen the belief that nuclear war would mean ‘the end’—not just of their own lives and those of their families, but of their country and all they cherish. In various ways, therefore, the people in democracies express a strong belief that nuclear weapons cannot be used as instruments of national policy, except somehow to preclude ‘the end’. Recent opinion polls shows that close to 80 per cent of the respondents both in Western Europe and in the United States oppose the use of nuclear weapons if a conventional invasion of our European allies could not be repelled by non-nuclear means. A surprising number of people have lately been saying they would even oppose using nuclear weapons in response to a Soviet first use; they thus betray perhaps a touch of confusion about the intended working of deterrence. Within Western alliances, citizens who so strongly reject almost any military use of nuclear arms can readily organize themselves to promote policies that would reduce— in their eyes—the risks of nuclear war. This is as it should be in free democracies. It is also not unexpected that the Soviet government tries to reinforce, and perhaps redirect towards its own goals these political stirrings in the West. What the Soviet government has recently done on this score is trifling, however, compared with what in a major crisis a Leninist or a Stalinist might choose to do. Someone in control of Soviet military power and schooled in the uses of terror for political ends could lay iron hands on the deepest emotions and fears of a great many people in the West. He could make them believe that the horrors of nuclear destruction were about to become real—unless they prevailed on their leaders to make major concessions. Even the current deployment of intermediate-range missiles in Western Europe came close to being prevented by the combined pressures of Soviet manipulation and political opinion in the West. Had one of the West European elections turned out differently, the Soviet government would have succeeded in depriving the alliance of one of the supports it sought for deterrence. Mark that this was in peacetime, when nuclear destruction seems a remote abstraction. Mark also that the pressures employed by the Soviet government were mild—merely diplomatic admonitions and some furtive propaganda operations in the West. Can we count on future Soviet leaders being so ‘cautious’ that they would not use the hammer and anvil of nuclear terror, even in a crisis? In the future, Stalin’s heirs, recalling the effectiveness of mass terror in the pursuit of power, may discover new ways to turn the threat of nuclear destruction into the ideal instrument for totalitarian expansion. It would serve no useful purpose here to speculate on how they might go about this. One is reminded, though, of an idea that surfaced many years ago: detonating a large nuclear weapon more or less harmlessly in the sky as a ‘demonstration’ of one’s seriousness. (Of course, at that time the shoe was on the other foot; the idea was that the Western democracies would frighten the heirs of Stalin.) For the future, the ‘balance of terror’ cannot favor the defense of a democratic alliance. Sooner or later, it will favor those most at ease with, those most experienced in, the systematic use of terror.
52 DILEMMAS OF NUCLEAR STRATEGY
The West has long relied heavily on the creativity and inventiveness of its scientists and engineers to compensate for the Soviet advantages in territory and mobilizable manpower. But our scientists and engineers, like any professional men and women, require moral satisfaction to work creatively. It can become a demoralizing task to invent and perfect, year after year, better means for inflicting nuclear destruction—not to defend one’s country, but to revenge its destruction. The prospect of achieving strategic defense provides a new, positive motivation for our scientists and engineers who work on national security. The danger of demoralization by relying on a strategic order of consensual vulnerability goes further and deeper. Upon an alliance of democracies, such a policy imposes a passive, almost cynical, resignation toward the possibility of an atrocity unsurpassed in human history. It offers a prospect of anxiety without relief, an intellectual legacy crippling the outlook of each new generation, a theme of desolate sadness. As Kenneth Clark observed, civilizations require confidence, ‘confidence in the society in which one lives, belief in its philosophy…’. To insist that only a rigorously preserved vulnerability will preserve us from annihilation is a credo that must corrode the confidence upon which civilizations are built. The conclusions fall into place. We need to accomplish a long-term transformation of our nuclear strategy, the armaments serving it, and our arms control policy. To begin with, we must disenthrall ourselves of the dogma of consensual mutual vulnerability—the notion that unrelieved vulnerability of the United States and the Soviet Union to each other’s nuclear forces is essential for halting the competition in offensive arms, and is the best guarantee against the outbreak of nuclear war. Such a relationship never obtained in the past and it is most unlikely to come into existence in the future. Moreover, an agreed balance of mutual vulnerability would be repugnant on moral grounds. And it could in no way reduce the consequences of an accidental nuclear attack.1 In the long run, reliance on a balance of mutual vulnerability would favor totalitarian regimes, with a demoralizing effect on democracies. The key now for the needed transformation is technological development to make effective defense systems possible for the United States and our allies. The priority requirement is non-nuclear missile defenses capable of negating the military utility of a Soviet missile attack and of diminishing its destructiveness. Depending on progress in arms reductions, the missile defenses might later be complemented with air defenses. As strategic defenses make it increasingly unlikely that Soviet offensive forces can accomplish their mission, the incentive for new Soviet investment in them is reduced. We thus enhance Soviet willingness to join us in deep reductions of offensive forces. To this end, we ought to take two complementary approaches. We should energetically seek Soviet cooperation, since it would greatly ease and speed the transformation. But we must also be prepared to persist on the harder road, where the Soviet Union would try as long as possible to overcome our defenses, and would resist meaningful reductions in offensive forces. The better prepared we are and the more capable of prevailing on the hard road, the more likely it is that the Soviet Union will join us on the easy road. Critics of missile defenses fear that defenses could not be made leak-proof against a massive attack, that a fraction of the missiles might penetrate, causing dreadful destruction, with whole cities lost. Usually this fear assumes that the Soviet leaders
NUCLEAR STRATEGY: CAN THERE BE A HAPPY ENDING? 53
would expend the bulk of their missile force to destroy our cities. Yet if only a fraction of the attacking force were to reach some unpredictable targets, while missing most of the militarily important targets, to launch such an attack would accomplish no rational purpose. It would serve no strategic objective, and would be akin to a wanton terrorist act. It is precisely against such irrational acts that an exclusive reliance on deterrence would offer no protection at all. We cannot know at this time whether the Soviet Union will join us and agree on major reductions in offensive nuclear arms, and whether it would abide by them. As one tries to promote arms control, Soviet failure to comply with past agreements is a bone in one’s throat. But if the Soviet leaders are willing to cooperate on a purpose both East and West can share, that purpose surely cannot be the perpetual hostility that is inherent in the unopposed forces constantly poised for mass destruction. It can however be a strategic order that will eventually eliminate Soviet vulnerability to massive nuclear destruction, even if this means in turn surrendering their capability to inflict mass destruction on Western Europe and the United States. Conversely, should the Soviet leaders choose to preserve their doomsday capability at any price, the prospects for arms control would be bleak indeed. The more the offensive armaments can be reduced by agreements, the easier and cheaper the job of providing effective defenses. Yet, to be realistic about Soviet motivations, we must seek to develop and deploy systems that can provide effective defenses even without such reductions. The United States is now pursuing new technologies that hold promise for success on the ‘hard road’ as well. Thus, we make it all the more probable that the Soviet leaders will join us some day on the easy road of cooperation. Our thinking on nuclear strategy must reach far into the future. It is not enough that our strategy serve to prevent nuclear attack in this decade or the next. Nuclear armaments and defensive systems take ten years or more to design, develop, and build; and once deployed they will last for a quarter-century or more. We are today constrained in the choices for our nuclear policy by strategic theories that 30 years ago began to influence the development of our present weapons systems. The time to start designing a safer nuclear strategy for the twenty-first century is now. For a few more years, to be sure, many arms control experts, military technicians and bureaucrats in our alliance will cling to the dogma of the allegedly stable consensual vulnerability. They will marshal all sorts of evidence ot prove, as an ‘iron law’, that the means for nuclear attack will always outpace the means for defense, keeping mass destruction an inescapable threat for all nations. One is reminded of another theory that was influential at the beginning of the last century: Thomas Robert Malthus marshalled a huge body of evidence to prove that the growth of population will always outpace the means of subsistence, keeping mass poverty an inescapable condition of mankind. Malthus’ theory became so influential because it combined a seductive simplicity and rigor with an important kernel of truth (a truth to be seen this day in many impoverished regions of the world). The Malthusian theory turned economics, as Thomas Carlyle put it, into ‘the dismal science’. Yet today all economists, regardless of differing views on population policy, agree that new factors have negated Malthus’ iron law. The nineteenthcentury marriage of capitalism and modern science brought forth such an immense creation of wealth—enhanced by the social and cultural benefits of modern democracy—that economics ceased to be a dismal science.
54 DILEMMAS OF NUCLEAR STRATEGY
Nuclear strategy, too, will cease to impose its dismal prospect on the future, if we now apply our creativity in science, technology and statesmanship to build a new order for preventing nuclear war.
NOTE 1. These points have been elaborated by this author in ‘Can Nuclear Deterrence Last Out the Century?’ Foreign Affairs (Jan. 1973), pp. 267–85.
III. European Perspectives
56
Nuclear Deterrence and European Security: Towards A Not So ‘Happy Ending’? Pierre Lellouche
Writing in the Spring 1985 issue of Foreign Affairs, Fred Ikle, Under-Secretary of Defense for Policy in the Regan administration, strongly advocated no less than a ‘happy ending’ to nuclear strategy. Calling the ‘dogma of allegedly stable consensual vulnerability’ both abysmally dangerous for humanity and ‘morally demobilizing’ for our democracies, the author recommended a ‘long term transformation of our nuclear strategy towards effective defensive systems for the United States and our allies’. 1 To those short-sighted Europeans who do not relish the prospect of such a ‘happy ending’ to what has been, after all, the cornerstone of the European security system for 40 years, the Under-Secretary threatened that rejecting missile defense would bring the Europeans to
rely almost exclusively on the global nuclear forces of the United States to provide the comprehensive guarantee that protects Europe. In the long run the force of this guarantee depends on its support by the American people. Some European opponents of missile defenses for the United States appear to believe that to assure the vulnerability of the United States somehow preserves this guarantee. This logic makes no sense. It implies that the larger the number of Americans who would be killed in a nuclear war, the more credible a decision by the US President to respond to a conventional invasion of Europe by initiating nuclear war. The message could not be clearer. The allies had better rally around the Reagan administration’s ‘Strategic Defense Initiative’ for (a) NATO can no longer count upon a US decision to actually escalate to its ‘global nuclear forces’ in the event of a war in Europe since (b) the American public cannot be expected to bear forever the risk of nuclear annihilation just because the Europeans are geographically more exposed to the Soviet threat. In short ‘take it (SDI) or we leave’. Notwithstanding its peculiarly undiplomatic style (to put it mildly), Fred Ikle’s message is an important one which ought to be thoroughly considered by all West Europeans. In the first place, Fred Ikle’s views (and when I refer to Fred Ikle, I mean of course the policies of the Reagan administration) actually mean that the nuclear debate which started in Europe in 1977–79 (first with Helmut Schmidt on the SS-20, and
58 DILEMMAS OF NUCLEAR STRATEGY
two years later by Kissinger’s address in Brussels) and culminated with the December 1979 ‘Euromissile’ decision is by no means over. Those who believed that the debate about the role of nuclear weapons in Europe would die out with the deployment of the Pershings and with the gradual decline of the European peace movement are now shown to be wrong. Ironically, while the Pershing battle was going on in Europe—finally ending with a successful deployment—another debate, of far greater implications for Europe, had started in the US itself and called into questin, in as much a radical fashion as the European left, the continued reliance by the West on nuclear deterrence. The difference with Europe was that in the US, the debate started from the ‘top’ that is, the strategic community, the government) rather than from the bottom (the man in the street). The results, however, are no less staggering. With the launching of the Strategic Defense Initiative under the banner of making nuclear weapons ‘impotent and obsolete’, it is the Reagan administration that is actually continuing the anti-nuclear crusade that began seven years ago in Europe. Admittedly the Reagan administration acts with different concerns, and with quite different objectives in mind than the European ‘peace’ activists, but there is a potential for a much more destabilizing impact, since the new ‘truth’ comes from the ‘leader of the West’. Seen from that perspective, SDI constitutes the American ‘reflection’ of the Pershing debate in Europe. In Europe, the Pershing affair revolved around the classic issues of maintaining Flexible Response in an era of strategic parity what weapons are needed to ensure a credible escalation; how can we spread nuclear risks among the allies— with the Europeans perceiving the Pershing, not as a weapon of ‘recoupling’, but as an instrument of a limited European nuclear war. In a sense, SDI is a continuation of these concerns, but from an American perspective. Through, SDI, Americans are rediscovering their own vulnerability (which they had either long forgotten or were never aware of, according to the polls), and with that discovery, the very old and sensitive issue of how much nuclear risk the US should take to protect Europe has come back to the fore. In short, the stage is being set for another and extremely divisive—‘great nuclear debate’ (as Raymond Aron would have put it), pitting once again the US and its European allies on opposite sides of the same old and inherent issues of extended deterrence: that is, (a) the role of nuclear arms versus other types of weapons in the defense of Europe; and (b) nuclear risk-sharing between the protector and the protected. Whereas in 1967 a visible US nuclear superiority made it possible to find a compromise solution (that is, Flexible Response)—though not without ten years of extremely tense debate—this escape route is unfortunately no longer available today. Indeed, it is precisely the awareness on the part of American strategists of their inescapable vulnerability in the deterrence age which has led to the search for ‘postdeterrence’ formulas in the form of space-based defense shields. Moreover, not only is there no longer an easy way out, but this time around the stand taken by the US government feeds upon, and even legitimizes, a deep-seated rejection of nuclear deterrence in Western public opinion as a whole. Short of a long-gone nuclear superiority, the United States seems to have invested its faith in the notion that new non-nuclear technologies (be it in conventional weaponry or in space) will permit a successful redefinition—a ‘fundamental shift’ as
NUCLEAR DETERRENCE AND EUROPEAN SECURITY 59
Ikle puts it—of Western defense posture away from deterrence and towards something else…which remains to be found. In the meantime the trend is toward technology that can squeeze out nuclear deterrence, particularly from the European theater, through the combined effect of ET at the ‘bottom’ and SDI at the ‘top’. The problem, however, is whether the Alliance, and Europe, will be able to withstand what the Reagan administration calls in a cute understatement, the ‘transition phase’. Taken together, these various trends—the policies of the Reagan adminis-tration, the anti-nuclear mood of the public and the strategic community alike and the promises of new technologies—constitute a quite formidable threat to the established European nuclear-based system. The implications for Europe if these trends continue are bound to be considerable indeed. Thus, before going into some investigation of these trends and of the related issues at hand, it may be useful to recall what stakes are involved, particularly as far as the fate of Europe is concerned.
I At the risk of sounding overly Gallic and ‘out of fashion’ by contrast to what seems to be an increasing anti-nuclear US and NATO rhetoric, the fact remains that nuclear deterrence has been for more than four decades the keystone of both the European security system and continued peace in Europe. However imperfect or ‘immoral’ this system may be, the fear of uncontrollable nuclear escalation has worked. It has prevented Soviet leaders from taking any military short cuts in the great historical contest in which they see themselves opposed to the West, and it has done so despite a military situation in the European theater which greatly favours Soviet armies. Indeed, there have been instances in the nuclear era in which military force would have been used between East and West in Europe, or elsewhere, had it not been for the fear of mutual annihilation. In addition, when one considers the role of nuclear weapons in post-war Europe, it is important not to lose sight of the wider geopolitical implications of nuclear deterrence which go well beyond the military equation in the theater. Aside from their strategic value, nuclear weapons in Europe also provide the basis of the political-territorial order in the Continent; and this may be just as important as their military role. That is why their location and their possession has always had a fundamental political significance. Where certain weapons are deployed or not deployed, who gets to fire them, where they are targeted (against Soviet territory or not) are questions that are often as crucial politically as they are militarily. This was clearly shown during the intense five-year Euromissile debate. These are issues that directly affect the status of the countries involved, the influence of the two superpowers on their respective allies, and therefore the continued viability of the two-alliance system. As long as nuclear deterrence continues to apply in Europe, the political and territorial status quo will remain essentially frozen, at least militarily. As long as the fear of uncontrolled nuclear escalation continues to prohibit any military ‘solutions’ (in the Clausewitzian definition) to the fundamental contest for domination of the West European peninsula, the name of the game in Europe will be competitive influence: with, on the one hand, the Soviets trying to exploit their
60 DILEMMAS OF NUCLEAR STRATEGY
regional military superiority with the imposition over the long term of their version of a ‘pan-European security system’ under dominance (and without a US presence in Europe), and on the other hand, Western democracies trying to use trade and political relations to produce a change in Central Europe and a gradual rapprochement between the two parts of the continent. To be sure, this is by no means an ideal or an easy situation, particularly for the Europeans who have had to pay for each day of continued peace the exhorbitant price of their continued division, and therefore of their continued weakness. German peace activists had a point when during the Euromissiles debate they claimed that ‘a Pershing palisade’ would only consolidate further the dividing line between the two Europes and the two Germanies. Similarly, those European statesmen who have tried to bring about a genuine European unity in Western Europe have discovered that it is very hard indeed to build Europe with only one half of Germany: and it is equally hard to give Europe a voice in world affairs as long as its safety continues to depend upon a faraway protector. Yet, imperfect and difficult as it may be, this is the security order which was created, and accepted over time, in Europe—nuclear deterrence on the one hand, with its dual implications of forced peace and continued division, and political dialogue with the East on the other hand, to permit gradual political change. Symbolically, it is precisely this delicate course which was charted by the Alliance in 1967 with the adoption of both Flexible Response on the military side and of the Harmel Report on the political side. In stressing this fundamental link between nuclear peace and the European politicalterritorial order, the basic point I wish to make at this stage is that fiddling around with ideas such as the ‘immorality of deterrence’ and the need to make nuclear weapons ‘impotent and obsolete’, or calling for a ‘happy ending’ to nuclear deterrence, carries fundamental implications for the future of Europe, not just in terms of preventing war, but also in terms of the geopolitical order that would necessarily come about in peacetime as a consequence of such a new strategy (even if one assumes that peace can be maintained without nuclear weapons). Unfortunately, I am not quite sure (to use a polite understatement) that these kinds of implications are even understood by those, in the US in particular, who peddle ‘visions’ of a glorious post-nuclear world.
II Let us now go to the current strategic debate, since ideas matter very much indeed in the nuclear era we still live in, not just because of their impact on our own public (without the support of which no strategy can be viable), but also because our discussions also affect our adversary’s calculus of our credibility. Two series of questions have to be asked if one is to understand where the West is headed. First, are we dealing with a radical new challenge to the postwar nuclearbased European system, or more simply are we being confronted with a continuation of the traditional NATO anxieties and unresolved questions about its strategy? Second, to the extent that we are indeed seeing a new departure, away from deterrence, are these themes equally shared on both sides of the Atlantic?
NUCLEAR DETERRENCE AND EUROPEAN SECURITY 61
To a certain extent, the American debate of the early 1980s is far from being new. Viewed in a historical perspective, many of the arguments now being put forward are merely a continuation of the same old and inherent issues of how to maintain a military alliance between democracies in the nuclear age, where risks are not quite equal and decision-making on nuclear weapons is not equal at all. Countless books and articles have been written in the last 30 years on this very subject and the record is known to everyone. The first doubts expressed against the strategy of nuclear escalation in response to a European war date back to the early 1950s (that is, from the moment, in fact, the Eisenhower administration dropped its insistence for conventional rearmament in Europe and chose instead (in 1953) to go for its ‘new-look’, all-nuclear, ‘massive retaliation’ strategy. As early as 1954, Kauffman, Brodie and Kissinger among others, concluded that relying on such threats would in the end turn out to be more of a ‘self-deterrent’ than a deterrent to the potential aggressor, an accurate enough forecast, even though it came several years before Sputnik and the beginning of the new era of American vulnerability. When de Gaulle spoke at his famous 1959 press conference of his conviction that no US President would ever exchange New York for Paris or London, the French President was merely stating the facts of life of this new era. When Kissinger, in his famous Brussels address of September 1979 warned the Europeans to stop asking the US for impossible nuclear guarantees, the American statesman was also stating, 20 years after de Gaulle, the very same facts of life. Indeed, what was so shocking to the European allies at the time (leaving aside the French, who found in Kissinger’s admission a most gratifying justification of their autonomous deterrent) was not so much the content of the speech (since many in Europe had long had doubts about the US guarantee), but the fact that it was an American—not some crazy Frenchman—who had long been in charge of that policy, who was now saying that the system could no longer possibly work as such. Similarly, when McNamara wrote in 1982 and 1983 that NATO should move away from the first use of nuclear weapons, he was saying more bluntly what he wrote more delicately in his 1962 Athens and Ann Arbor speeches some 20 years earlier. All in all, it had taken ten years for the Alliance to work out a rather painful transition from massive retaliation to Flexible Response (1957 to 1967), and not even ten years later the same issues were once more on the table (the trigger being this time the Soviet achievement of strategic parity and the beginning of the deployment in 1976 of the first SS-20s). When viewed against this historical background, the new American debate of the early 1980s, with its two-pronged drive to (a) reduce the role of nuclear weapons in NATO strategy (ET, FOFA, No early/No first use), and (b) reduce or eliminate American vulnerability to Soviet nuclear strikes (be it the vulnerability of US missiles or US cities) with SDI, is in fact the natural continuation of traditional American concerns vis-à-vis both the US commitment to NATO and the share of nuclear risks associated with it. Similarly, one finds as in the past, the same American fascination with technology and US technological superiority, as the solution to these permanent geostrategic problems. Yet it would be wrong to conclude from this rapid examination of the historical record that today’s debate about the future of nuclear deterrence is only a rehashed version of the past. Other than the technology, which is changing, there are a number
62 DILEMMAS OF NUCLEAR STRATEGY
of distinctly new and important features in today’s discussion about nuclear deterrence. For one thing, the debate launched in the late 1970s about the respective role of nuclear and conventional weapons in NATO strategy coincided with the rise of a strong grassroots anti-nuclear peace movement in the West triggered by both the neutron bomb and the Euromissile decisions (in 1977–79). Whereas in the past alliance strategy was solely debated among professional analysts and politicians, and whereas their discussion was generally limited to arguing the most effective means of ensuring deterrence (that is, what proportion of ‘graduated’ conventional response, what kinds of nuclear weapons would be necessary), the debate of the late 1970s and early 1980s revealed two distinctly new trends. First, nuclear strategy was ‘democratized’ to the extent that the debate was now conducted in the streets by ordinary people who challenged not just the effectiveness of certain particular policies and weapons but, beyond that, the very legitimacy of the notion of nuclear deterrence itself both on moral grounds (that is, the Bishops), and on ‘existential’ grounds: how long can the West pretend to base its security on the threat of a global holocaust for the whole of humanity; is it not ‘better to be red than dead’?). In short, concern for survival, including the ecological survival of the planet (the theme of the so-called ‘nuclear winter’), came to transcend heretofore established notions of legitimate selfdefense and freedom. Nuclear weapons became the true enemy, and the certainty of mutual annihilation erased ideological differences with the adversary’s totalitarian system. Indeed, for some, the adversary itself came to be seen as a fellow victim of the great nuclear folly (that is, the SPD concept of ‘security partnership’). Reinforcing that trend, the growing sophistication and accumulation of nuclear weapons changed the public perception of what these weapons were all about. Instead of guaranteeing ‘no war’ at all through ‘pure’ deterrence, new weapons (such as the Pershing) came to be seen as warfighting weapons, not deterrence weapons. In all NATO countries, (with the single exception of France, in part because it continued to stick to a ‘pure’ massive retaliation, ‘no-war’ doctrine), the inevitable transition towards more and more refined and selective weapons and warfighting doctrines brought about the collapse of the domestic consensus. In the past, Western democracies has accepted atomic weapons because they were cheap and because they were supposed to prevent any war. Now they were rejecting them, in part because they accurately understood that the weapons—as deterrence itself—had changed, and could actually be used in war. This led Western democracies to an impasse: to remain credible, Western strategy, especially in Europe, had to move forward to ever more sophisticated weapons and selective options and yet, in doing so, this strategy was increasingly scaring the very people it was trying to protect. In a nutshell, the credibility of a nuclear doctrine had become inversely proportional to its social acceptability. The second new element was that the Western strategic ‘elite’, and with them the politicians, instead of trying to resist the growing protest wave, actually began to ride it, and even to add ‘intellectual’ and political legitimacy to the arguments of the peace movement. The trend began with the liberals in the US (McNamara’s articles on no first use and minimal deterrence) and, for quite different motivations, with the Social Democrats in Europe (the theme of an ‘alternative’ non nuclear defense in the Federal Republic of Germany), but it was soon continued by the conservatives (notably in the US) who, again from a different perspective, and with different objectives in
NUCLEAR DETERRENCE AND EUROPEAN SECURITY 63
mind, also began to criticize openly NATO’s ‘over-reliance’ on nuclear weapons, as both incredible and much too dangerous in times of crisis, particularly for the United States. All of a sudden, in 1982–83, an unlikely alliance began to emerge composed of American liberals, American conservatives and warfighters, and, on the other side of the Atlantic, a number of anti-nuclear activists and left-wing politicians. Although they disagreed on just about any subject, they agreed on one, namely, the need to reduce NATO’s reliance on the use of nuclear weapons. Ironically, this evolution took shape at the very time when NATO governments were fighting an extremely difficult political battle to deploy, on schedule, and against an unprecedented Soviet propaganda offensive, the first Euromissiles decided upon in December 1979. It was then that the ultimate touch came, on 23 March 1983, when President Reagan announced his ‘vision’ of a world freed from the threat of nuclear weapons through a fundamental shift of strategy—from the prevention of war through fear of mutual destruction to a state of ‘mutual assured security’ by means of defensive shields over the US and its allies. Were it not for the high stakes of this debate, it would be tempting to speak ironically of this bizarre and, indeed, rather incredible coalition which runs from bishops to generals, and from Ronald Reagan to SPD leftists, not to mention, of course, the powerful lobby of the arms and high-tech industry, who apparently all agree that it is time to close the nuclear parenthesis in the history of humanity and move towards something else, even though that ‘something else’ remains rather vague: ‘conventional deterrence’, star wars, or unilateral disarmament? In this extraordinary confusion, it has become nearly impossible to know who defends what policy, and which argument belongs to what camp. From one extremely tense Euromissile debate fought in the name of deterrence, the Alliance has been thrown without any transition into another debate for the exactly opposite cause. As in a Kafkaesque play, the same actors have now changed sides and have even exchanged arguments. The immorality and ‘holocaust’ arguments which used to belong to the anti-nuclear left are now used by the warfighters (Wohlstetter) and by the star warriors. Meanwhile, the European left, now deprived of its anti-nuclear crusade, finds itself in the rather uncomfortable position of having to fight both offensive nuclear weapons and defensive weaponry.
III The dangers of this situation are obvious. Not only is the Alliance running the risk of rapidly losing track of its own strategic goals, but it may also be throwing away whatever remains of its strategy, and of the domestic consensus on that strategy, in exchange for the mirage of technological solutions which do not yet exist and which may very well never exist. Worst of all, in the course of squeezing out what is left of NATO’s nuclear posture through ET and SDI, the Alliance may finally convince the Soviet leaders that it is indeed moving away from nuclear deterrence, thereby removing the fundamental uncertainty that has been the key to stability in Europe so far. To add to the complexity of the current situation, it is important not to lose sight of the fact that the growing anti-nuclear (or post-nuclear) tone of the Western debate on
64 DILEMMAS OF NUCLEAR STRATEGY
strategy has by no means reconciled the deep differences of strategic interests between the two sides of the Atlantic. In fact, the apparent transatlantic consensus to move away from nuclear weapons is as artificial as the apparent similarity that seemed to exist a few years ago between the various peace movements across the Atlantic was superficial. To a certain extent, the drive towards ET and SDI in the United States expresses, like the earlier peace movement in that country, an everpresent aspiration towards nuclear isolationism, that is, a concern not to be drawn into a ‘foreign’ nuclear war. Instead, in Europe, the critique of nuclear deterrence and the arms race, and the call even for de-nuclearization (in some quarters of the left) expresses, as in the earlier peace campaign in Europe, a growing desire to be left out of the superpowers’ competition, be it on the earth or in space and, above all, not to serve as a battleground for the superpowers. When viewed from this perspective, the evolution of the Western strategic debate seems even more worrisome. What seems on the surface to be a consensus to move away from nuclear weapons translates in practice into continued divergences as to what should be done to construct a new strategy. For one thing, those in the US who had thought they would win the support of the European public and the European left by promoting either ET or SDI are in for a big disappointment. The most telling example here lies with the erman SPD. Its Essen document on defense (adopted in May 1984) is resolutely anti-nuclear in its inspiration, and therefore rather supportive in principle of conventional defense. Yet the same document is also highly critical of what it views as ‘aggressive’ new NATO ‘doctrines’ (such as Airland battle and FOFA), and proposes instead a ‘structurally defensive’ posture, meaning in fact that not a single D-Mark nor a single soldier would be added to its forces, and that the Bundeswehr would not be able to fight any battle anywhere (neither forward nor in the rear). The same observation holds true in respect to SDI: logically, the European left which fought its non-nuclear battle against the Pershings could not be expected to become anenthusiastic supporter of SDI. Instead, what is happening is just the reverse: not only is the SDP, the Labour Party, and the rest of Europe’s left condemning SDI as a vehicle for the ‘militarization of space’ but, quite ironically, some on the left are even beginning to rediscover the forgotten merits of nuclear deterrence, albeit in its ‘purest’ form of minimal deterrence à la française (hence Egon Bahr’s sudden but suspect affection for the Force de Frappe). A similar observation could be made in respect to the behaviour of European governments. Leaving France aside, most European governments now officially support the evolution of NATO’s strategy towards ‘conventionalization’ and the socalled ‘no early use’, as they equally support (though with little enthusiasm) President Reagan’s Strategic Defense Initiative. If one looks at the facts closely, however, one discovers once again an interesting gap between words and deeds. The ‘yes’ to conventional defense has yet to be sustained by any increase in military budgets or personnel: in fact, all indicators are pointing in the opposite direction (stagnant defense budgets, demographic crisis), with the result that NATO may have a FOFA strategy, but that strategy exists only on paper. As to SDI, the government-to-government agreements recently signed by London (and soon by Bonn) with Washington are primarily directed at permitting European industrial access to the SDI program under satisfactory conditions (a most
NUCLEAR DETERRENCE AND EUROPEAN SECURITY 65
naïve endeavour and are far from representing a genuine approval of Ronald Reagan’s new strategic concepts. Indeed, while public criticism against SDI from key British and German ministers has now been effectively silenced (!), discussions with British and German policy-makers show that there is in the UK and FRG at least as much concern as is found in France with respect to the strategic implications for Europe of an SDI world—and for good reason! Leaving aside the huge financial technological challenge which is posed to Europe by a research program of this magnitude, and leaving aside even the anti-nuclear rhetoric of the Reagan administration in selling SDI (which further undermines domestic consensus in Europe), the strategic implications of an offense/defense world are not good for Europe. Quite the opposite in fact. Such a world would mean the end of selective employment options by the US strategic forces, which means a further neutralization of the superpowers’ strategic arsenals, the use of which would simply become even more unthinkable than in today’s world in contingencies other than a response to direct attack by one superpower against the other. An offense/defense deterrence world, far from eliminating nuclear weapons or nuclear deterrence, would in fact throw both superpowers back to an equation of pure massive retaliation, the very opposite of what is required in the context of extended deterrence. With strategic defenses on both sides, Europe will be genuinely decoupled from the superpowers’ nuclear equation in so far as strategic intercontinental or theater weapons (like the Pershing) are concerned. The logic of such a world would be either a de facto denuclearization of NATO’s defensive posture in Europe, or at the most, a clear limitation of nuclear employment options to battlefield weapons with no capacity to reach the Soviet Union. In short, for the Europeans, the logic of an SDI world (that is, defenses on both sides) leads nowhere else but to the traditional nightmare of a limited conventional or tactical nuclear war fought on their soil. The trouble, however, is that from a strictly American point of view, a convincing and quite understandable case can be made in favour of strategic defenses, to the extent that even partial protection would reinforce uncertainty and therefore deterrence between the two superpowers themselves, by prohibiting ‘easy’ counterforce pre-emptive strikes on each other’s arsenals. It is precisely because of these fundamentally divergent strategic implications that SDI is, in fact, not simply a political or a technological issue, but a terribly sensitive and divisive strategic problem for the Alliance. So sensitive in fact that it may, if mismanaged, break the transatlantic security relationship. While the Europeans will become increasingly aware of the new ‘vulnerability gap’ with their protectors (which may lead to a new questioning of the value of the Alliance and perhaps also to temptations and find some accommodation with the East), the Americans, following Fred Ikle or Senator Nunn, will begin accusing the Europeans of imposing an unacceptable nuclear risk on the US by refusing to do more on conventional defense. A visiouc circle, the basic lines of which are already visible, may therefore develop, aided, of course, by Soviet propaganda and diplomacy. In this con-nection, it is important not to misunderstand what Soviet options and worries really are in respect to SDI. While the Soviets are undoubtedly worried about the technological challenge represented by SDI, they also understand that the strategic implications of an of fense/defense world may also turn out to be rather positive from their point of
66 DILEMMAS OF NUCLEAR STRATEGY
view—provided, of course, and there is little reason to doubt, that they will not lose the technological race. If one bears in mind the fact that Europe, and the geopolitical grand design the Soviets want to establish in Europe (that is, a Europe decoupled militarily from the US and placed under Soviet political influence), remains the one central objective of Soviet strategy in the global East-West confrontation, then an SDI world in the long run can only be in their favor. This is where arms control comes in: a separate INF deal leaving the Soviets with a clear-cut regional nuclear superiority and a potential handle on future French and British modernization would be the first element of that strategy, followed at a later stage by an agreed amendment to the ABM Treaty allowing more ABM stations to be deployed on both sides. In the end, it is Flexible Response that would fall victim to such an evolution.
IV Important, though by no means ‘happy’, implications follow from the above analysis. The first is that NATO, while taking care to keep formally its Flexible Response doctrine, is probably seeing the last useful years of that strategy. A phase of American nuclear disengagement from Europe has begun, the product of several complex historical trends (the evolution of the correlation of forces with the USSR, the public mood in the US, the debate among strategists, the race of new technologies). I see no way to stop that trend, no more than I see a way to stop either the US SDI or the Soviet ‘SDI’. This means that the Europeans are basically back to square one: ‘Lisbon 1952’, providing themselves with genuinely strong conventional forces, and ‘Paris 1952’, thinking again about intra-European defense cooperation including the field of nuclear weaponry. For, while it is perhaps inevitable that the US nuclear deterrent will eventually be squeezed out from Europe, the same will not be true for European nuclear forces, simply because the issue of nuclear risk-sharing is not the same when viewed from California as when viewed from Britanny. On the continent, the fact of the matter is that all countries bear the same risks, and that is beginning to sink in even in France. Less reliance on US nuclear deterrence does not have to mean necessarily the end of deterrence in Europe, which would also mean the end of Europe as we now know it. It is up to the Europeans to reconstitute among themselves the basis of an autonomous deterrent of their own on the basis of the existing and growing European nuclear forces. Clearly this is an awesome challenge, though more a political than a military one, but one which could be carried out, perhaps when the trends analyzed above become clearer to all Europeans involved. Second, while the Europeans should refrain from fighting useless and counterproductive battles against SDI, the least the US could do is to end once and for all its anti-nuclear rhetoric: deterrence is here to stay, although admittedly in a different form. Rather than selling impossible technological dreams to our people, with the risk of a backfire when the dreams are not delivered, it would be infinitely more useful to either spell out the facts or, preferably, keep quiet altogether. Finally, in embarking on a new round of arms control talks, one can hope that the US will finally understand the geopolitical stakes of this process—particularly as far as the future of Europe is concerned—rather than concentrate, as in the past, on simple
NUCLEAR DETERRENCE AND EUROPEAN SECURITY 67
counting of weaponry coupled with naïve illusions about ‘teaching’ the Soviet the virtue of the post-nuclear ‘vision’ dear to Ronald Reagan’s heart.
NOTE 1. ‘Nuclear Strategy: Can There Be A Happy Ending?’ Foreign Affairs (Spring 1985).
68
The Reception of American Deterrence Theory in the Federal Republic of Germany and the German ‘Nuclear Debate’ of the 1950s Wolfgang E.Heisenberg
I. The Political Importance of the Concept of Deterrence The Reagan administration has often been accused by its critics of having abandoned the traditional American policy of deterrence and of preparing instead actually to conduct a nuclear war. These critics maintain that, unlike its predecessors in Washington, the present administration assumes that such a war can be waged and won. In this respect, it is not pursuing a policy of deterrence but rather one of war fighting. As proof of their assertions, the critics refer to the statements of American Secretary of Defense Weinberger that, if deterrence fails, the USA must be in a position to prevail in a strategic nuclear war.1 Secretary of Defense Weinberger and other government officials have attempted repeatedly and vigorously to counter these accusations. They have pointed out that the strategic policy of the government still has as its aim the prevention of nuclear war through deterrence and they have stressed that the administration does not think that a strategic nuclear war could be won and, moreover, does not intend to wage a limited nuclear war in Europe.2 What is this debate actually about? One can hardly seriously accuse the American Secretary of Defense of wanting to involve his country in a nuclear war. On the other hand, Weinberger has not changed his view that, in the event of deterrence failing, the USA must be able to defend its interests, prevent the other side from achieving its war aims and end the war under favorable conditions. The wording of the statements cited by Weinberger’s critics was refined and made more precise in his subsequent remarks, but his basic position did not change. It therefore seems that only words, subtle nuances in the formulation of official declarations, which could be misinterpreted or cause concern among America’s allies, are involved here. However, the fact that the American administration, as well as its critics, obviously view as a serious accusation the assertion that it is abandoning the principle of deterrence indicates that more than mere words are involved. In June 1982, Weinberger took the unusual step of publicly replying to the criticism, and in later annual reports of the Department of Defense much space was devoted to proclaiming that the Reagan administration was primarily interested in deterrence. If it is politically important whether American strategy can be described as a deterrence or a war strategy, then the question arises as to what the differences
70 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
between the two strategies are and whether there are clear criteria which make it possible to distinguish between them. This question would still have to be answered if the administration were of the opinion that the effectiveness of deterrence is determined solely by the ability to wage war. Then, it is true, in the view of the administration only an imaginary problem would be involved, which would have a certain importance only because part of the public erroneously believes that deterrence can be separated from waging war. Acceptance of deterrence would then serve primarily to justify the policy of the administration in public and to show its peaceful intentions. But this would mean a recognition on the part of the administration that in the public mind differing political ideas and expectations are associated with deterrence and waging war which cannot simoply be ignored. The question must be asked which political goals and expectations are behind a specific deterrence strategy and how they are to be realized politically. Secretary of Defense Weinberger’s answer to the accusation that he is pursuing a war strategy in unsatisfactory in so far as it refers exclusively to the subjective intention of the government to maintain the fundamental principle of deterrence, but not to objective features which distinguish the American strategy from one of ‘waging war’. This could mean that the government sees the difference between a deterrence strategy and a warfighting strategy only in the declared intention to avoid a nuclear or any other kind of war. Then, however, it would be exposed to the same criticism the French philosopher Glucksmann aims at the German peace movement: since under present conditions no reasonable person can be for war, and certainly not for an atomic war, declaring one’s support of peace serves no purpose and can easily be misunderstand. General declarations in favor of peace have not prevented new conflicts from breaking out. Under existing circumstances it is probably not even possible to talk about a general peace. In reality those who stress their desire for peace are concerned only about their own peace.3 This criticism, however, is valid even if the declared intention in using military force is exclusively to prevent wars, especially as this intention of course can never be unconditional. If the American deterrence strategy only differs from a strategy for waging war in its intention to avoid a war, then it does not differ at all from the traditional forms of power politics. In this case, deterrence becomes a synonym for military or defense policy. For this reason it is argued in less official statements that American strategic policy is based on deterrence because the United States is not even in a position to wage a long, limited strategic nuclear war. The vulnerability of the American strategic command and communications system (C3I-System) and the lack of an adequate civil defense program are good reasons to believe these claims.4 It is even doubtful that it will be possible in the foreseeable future to develop the ability to wage a strategic nuclear war. Nevertheless it would be dangerous for the present American administration to rely on such arguments: American strategic armaments expenditures are defended to a considerable degree with the argument that, if deterrence fails, the USA must be able to realize its national goals and end the war under favorable conditions. Opponents of this policy could at least conclude that the USA is attempting to achieve the ability to wage a nuclear war.
DILEMMAS OF NUCLEAR STRATEGY 71
These difficulties in justifying the concept of deterrence will of course become greater if strategic defense options should be created in the future. It only shows the ambiguity of the term ‘deterrence’ that President Reagan’s announcement of the ‘Strategic Defense Initiative’ was at first considered a radical departure from that concept, whereas recently Paul Nitze and other government officials have argued that the creation of a missile defense system would not change deterrence in any way. In view of this somewhat confused situation, one could assume that detterence is nothing more than one of those empty political catchwords concealing the most varied kinds of often more emotional than rational motives which defy any attempt at clear analysis. That is not quite the way it is. Of course, in specialized literature repeated attempts have been made to define the concept of deterrence more precisely, and there is no lack of critical statements on deterrence. As a whole, however, they provide neither a uniform nor a clear explanation of what distinguishes a policy or strategy of deterrence from the traditional forms of military power politics of strategies for waging war. Most literature on the development of a nuclear deterrence strategy attempts first of all to answer the question of how nuclear weapons can be most effectively used, whereas the decisive question of how the actual aim of a deterrence policy, the prevention of war, should be achieved is treated only by implication or at best marginally. In one of the earliest and most famous articles on nuclear strategy, Bernard Brodie’s 1946 essay ‘War in the Atomic Age’, this disparity was expressed clearly. Brodie began by pointing out the historically unique and almost inconceivable destructive power which could be achieved by the use of nuclear weapons. After a brief reference to the fact that the horrors of war had never influenced the behavior of governments in any uniform way, Brodie proceeded directly to the question of how nuclear weapons could be used most effectively under existing technical conditions.5 Brodie’s demand for an adequate retaliatory capacity was based on the vulnerability of large population centers to nuclear attack, not on the concept of deterrence.6 By far the greatest part of his essay was devoted to military questions. Brodie’s often cited observation: ‘Thus far the chief purpose of a military establishment has been to win wars. From now on its chief purpose must be to avert them. It can have no other useful purpose’7 indicates the motives involved in the concept of deterrence. In fact Brodie’s statement was little more than the experession of a vague feeling that as a result of the development of the atom bomb the function and purpose of military power ‘must’ change. Whether all or only certain kinds of war had to be prevented; what was to be understood by ‘war’, and whether it was important who waged a war; all these questions remained unanswered. However, if one takes seriously Brodie’s demand that as a result of the development of the atom bomb military power should be used primarily to prevent wars, the questions first have to be answered as to what means, if any, are available to influence the readiness of a state to wage war and to what extent military power can serve this purpose. Brodie hardly mentioned these questions. The later development of American strategic doctrine shows, however, that his assertion that a secure relatiatory capacity could have the desired deterrence effect is by no means self-evident. The principle of retaliation was never fully accepted in American strategic policy, although it did dominate public statements in this area for a time. It was pushed into the background as soon as the development of weapons technology made possible an effective and
72 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
limited use of nuclear weapons against military goals and nuclear weapons available in sufficient numbers at a relatively low price. Of course this is not to disparage Brodie’s great contribution to the development of the theoretical basis of American strategic policy. For at least several decades it determined the direction of thinking about nuclear weapons. I merely want to point out how strong the the constraints were to take up and to continue the existing principles of military strategy, which themselves had originally been conceived only for conventional forms of war, and to extend them by extrapolation to the newly developed weapons in spite of the insight that it was necessary to devise a fundamentally new strategy, a strategy of deterrence. From our present perspective, the deterrence theory developed in this way can be considered at best only a partial success. Efforts to develop a theoretical foundation for American nuclear strategy have made no progress since the mid-1960s. The flaws and limitations of this theory, repeatedly emphasized by critics of deterrence, have not been eliminated. This may be due to the fact that its influence on military practice has always remained somewhat unclear. Since the 1960s, innovations in military strategy have been accompanied by a tiring, monotonous discussion of the same strategic questions without anything being learned in a practical sense, without any new insights being achieved and without a growing consensus in fundamental principles being reached. Above all, the effects of nuclear strategy on the relations between the superpowers in regard to security policy have not met expectations. In spite of their strong political conflicts and ideological confrontation, the two superpowers have indeed refrained from military force in settling their differences. In this respect it can be said that deterrence has worked, deterrence theory has proved its worth. However, it has always been doubtful that the self-restraint of the superpowers can be attributed to a specific deterrence strategy. At any rate, it seems to be remarkably uninfluenced by changes in the political climate between them or in their strategic doctrines. Moreover, doubts remain as to the effectiveness and durability of deterrence. If Clausewitz compared the importance of battles for a war with that of cash payment for trade, it seems plausible nowadays to understand war generally as Hobbes did— as military power politics. Deterrence also involves the direct use of military force, and even though it may be possible to avoid the use of armed force for a time, it is difficult to see how it will be possible to do without it in the long term. This danger could be accepted if it were possible to show that a strategic nuclear war could be waged and ended sensibly. In this respect it is only consistent for the American Secretary of Defense to demand that the USA must be in a position to end a strategic nuclear war under favorable conditions in the event deterrence fails. But the American government has until now not been able to show convincingly how this aim might be achieved under realistic conditions. In this regard the Strategic Defense Initiative has given rise to hopes among the population which cannot be fulfilled in the foreseeable future. As it has not been possible to noticeably slow or even stop the arms race of the superpowers, a situation has developed in which their existing armaments potential is increasingly felt to be a threat in itself. The hope of achieving an ability to wage a nuclear war can result in additional increases of the strategic potential of both sides. Different ways are conceivable of leading deterrence theory out of the dead end in which it now finds itself. One possibility would be to understand deterrence primarily
DILEMMAS OF NUCLEAR STRATEGY 73
as a political task in which the function of military power would indeed be to prevent a potential enemy from using his armed forces but which would not be limited to military planning and armaments. It would also include a complete spectrum of other measures capable of influencing the behavior of other states. This approach would change deterrence from a subject of military strategy into a subject of international relations theory, a theory which would not exhaust itself in the development of general statements about the finternational system but which would include the actual behavior of the superpowers. In reality the policy of military deterrence has always been part of a much more complex foreign policy, but the nonmilitary components of this policy have until now been examined less completely and systematically than its military preconditions, perhaps because foreign policy is still considered by many people to be more of an art than a rational, purposive activity. The adoption of American theories and plans of nuclear weapons strategy in the Federal Republic during the late 1950s and early 1960s seems interesting to me primarily because in this case the rationale of American nuclear weapons strategy was taken over practically without modification by a country whose political and strategic situation was fundamentally different from that of the United States. This resulted in some differences in the two countries’ understanding of deterrence and in their attitudes towards nuclear weapons, which did not become visible because they did not find expression in the various strategic theories themselves. An analysis and comparison of these differences will make it much easier to achieve a GermanAmerican understanding in nuclear weapons questions.
II. The Political Preconditions for the German Adoption of American Deterrence Theory in the Post-War Years The development of the atomic bomb during the last years of the Second World War, its use against Japan, and the build-up of a strategic nuclear weapons potential, at first in the United States and later in the Soviet Union, seemed to call into question previous theories about the conduct of war as well as traditional ideas and opinions about the character and nature of war and the function of military power. At first primarily civilians—natural, social and political scientists—in the United States and several other Western countries attempted to analyze the changes nuclear weapons had brought about and in this connection to rethink practically all strategic theory in the broadest sense of the term. At the end of the 1950s, when in the course of the introduction of so-called tactical nuclear weapons in Europe the Federal Republic was forced to concern itself politically with the questions resulting from the development of nuclear weapons, the development of American strategic nuclear weapons theory was already concluded. Significant changes were indeed made later in American strategic policy and doctrine, but the theoretical basis of American nuclear strategy remained almost unchanged. Until then the discussions on strategy resulting from the development of nuclear weapons had produced almost no reaction in Germany. The political factors which were responsible for this situation deserve close attention, for they determined the
74 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
way in which the strategic ideas of the Americans and the British were adopted in the Federal Republic, and they explain why the attitude of the Federal Republic in such questions differs noticeably from that of other Western countries. There are obvious reasons why the Germans did not participate in the beginning in the strategic thinking of the Americans, the British and the French. After the surrender and the collapse of Germany, there was at first no German foreign and security policy. There was therefore no practical reason for Germany to concern itself with new theoretical problems in this area. At that time, German politics undoubtedly centered around questions involved in the reconstruction of the state, political life and the economy. Of course universities and other academic institutions were able, with the support of the victors, to resume their work comparatively quickly, but, as in other countries, there was no established discipline in Germany which could have concerned itself with these questions. In Germany, strategic issues were almost exclusively the province of the military profession. During the reconstruction of German political life and of the state apparatus, the military area was at first ignored. Directly after the end of the war, the victorious powers placed great value on preventing the development of any new military structure in Germany. They assumed that Germany must not be allowed to become a military power again. One of the few common goals the victors were able to agree upon in Potsdam was that Germany should be prevented from ever again becoming a threat to world peace. This was to be achieved by a constant monitoring and supervision of Germany’s ability to wage war and by the re-education of the Germans themselves. According to the guidelines published by the Allied Control Council, Germans were forbidden for the time being to pursue any activity which could be considered even remotely of a military nature. For example, military schools, paramilitary sport clubs, and other organizations, as well as military exhibitions, military ceremonies, and all military-related research work were prohibited. There were, of course, military experts in Germany after the war who would have been able to participate in the discussions about strategic theory in the leading Western countries, but it was almost impossible to obtain detailed information about the new nuclear weapons and, above all, the political and institutional preconditions for German participation were lacking. Some disciplines, such as contemporary history, economics, and political science, were concerned only indirectly with strategic questions. Whereas in the USA, Great Britain, and also to some degree in France, scientists and scholars from various disciplines recognized the importance of the security problems created by the new developments in weapons technology, addressed these problems, and in this way practically created a new academic discipline, a comparable development in the Federal Republic was not possible. However, political conditions changed comparatively rapidly in the postwar years. With the division of Germany and the beginning integration of the developing German states into the hostile blocs confronting each other in Europe, the victors of the Second World War soon abandoned their anti-military policy in Germany. In this situation, the beginnings of an independent German foreign and security policy could develop, but for the time being Germany still largely abstained from participation in the discussions of the Western victors on nuclear strategy. This abstinence is understandable if one considers the questionable role of the military and military values in German history. After the war, the Allied victors
DILEMMAS OF NUCLEAR STRATEGY 75
assumed the existence of a specifically German militarism which had to be eliminated. Indeed, it cannot be denied that German society, above all in Imperial Germany but also much later, was influenced and shaped by military ideas and values in a way that now seems excessive. Although the high point of this kind of militarism in Germany was reached before the First World War, many of its features continued to exert an influence in German society in the period between the wars, and National Socialism incorporated several elements of this tradition. For many Germans, therefore, the experiences of the Second World War and the defeat of Germany were a disproving of the earlier social order strongly influenced by military values, and this tendency was reinforced by the Allied attempts at re-education. The regulations issued and measures undertaken by the Allied Control Council to eliminate German militarism were not only accepted but also supported by many Germans as norms confirmed by their own experience. From our present perspective the re-education efforts of the Allies and their singleminded concentration on German militarism seem questionable or at least misplaced. Military values and behavior also played an important role in the victorious Allied countries and in other European states. As long as military power and the threat of military force are considered legitimate instruments of international policy, it will be very difficult in general to distinguish legitimate from illegitimate forms of militarism. The concentration of the Allies on the elimination of German militarism was thus less the result of a persuasive political analysis than a common denominator on which they were able to agree relatively easily. This made it easy for the Germans to avoid having to concern themselves very intensively with the deeper political causes of the new situation. However, there remained a certain tension and fear of contact between the military and civilian parts of German society. Above all, broad areas of the German political spectrum were affected by an ambivalent attitude towards military power.
III. Nuclear Weapons Strategy in the Federal Republic and the Concept of Deterrence In the first studies commissioned by political groups of Germany’s military situation after the war, nuclear weapons and the idea of deterrence played almost no role at all. These studies were based on the idea of a conventionally waged war of movement over an extremely large area; clearly they were strongly influenced by operational thinking of the battles of movement in the Second World War. The military strength of the other side was generally estimated to be very high. For example, the best known of these studies, the so-called Himmeroder Memorandum, assumed that the Western armed forces were not adequate for a set defense along a front of over 800 kilometers. For this reason, it was planned to concentrate forces in the north and south of the Federal Republic; the weaker forces in the center were only intended to slow the advance of the enemy. From the areas of concentration in north and south the troops of the invading forces were then to be attacked and split into smaller groups. The enemy forces were thus to be ‘stopped or contained, threatened on their flanks and deprived of their ability to act in a coordinated fashion’.
76 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
Serious questions about the military effects of nuclear weapons were asked in the Federal Republic only when NATO began to equip its forces with so-called tactical nuclear weapons during the introduction of the American ‘New Look’. As far as any German ideas in this regard became known publicly, they evidently assumed that nuclear weapons, in the event of permission being given to use them on the battlefield, could be employed according to military requirements and without any significant political restrictions. The main questions were how the expected increase in firepower through nuclear weapons could be used militarily within the framework of an extended war of movement and how the troop structure and deployment principles could be adjusted to these conditions. It is therefore not completely mistaken to assert that in these plans nuclear weapons were at first treated as a stronger form of artillery or air support. Precisely because it was not at all certain that available and projected Western armed forces were or would be able to fight over a large area as had been done in the Second World War, the question arose as to whether the impressive increase in firepower resulting from the introduction of nuclear weapons could compensate for existing shortcomings in conventional forces, especially as mobility itself could be regarded as providing troops with a certain protection against the use of nuclear weapons. The study by Otto Heilbrunn can be considered typical of such views. Heilbrunn assumed that even a conventional defense had to consider the possibility of a use of nuclear weapons from the very beginning, as particularly defensive forces would face serious problems if they were forced to prepare for a nuclear war after the fighting had begun. He advocated therefore a war with small, highly mobile units operating over the entire field of battle and suggested deploying forces in the rear of the enemy at the beginning of the battle. There they would be relatively well protected against nuclear weapons and, together with the units in front, could destroy the enemy.8 The studies prepared at the beginning of the 1970s under the direction of Kielmannsegg for the foundation Science and Policy, which advocated an early and massive use of nuclear tactical weapons, continued these ideas in a more comprehensive way. A central problem in this regard became obvious in the summer of 1955 when during the NATO war game ‘carte blanche’ possibilities of using so-called tactical nuclear weapons in Europe were played through. It became clear that basing decisions to use nuclear weapons to repel a Soviet attack in Central Europe solely on military considerations would cause such devastation in the Federal Republic that it would call into question the whole purpose of the defense. Widely differing military conclusions were drawn from these facts. They extended from the militarily consistent but politically unrealistic demand of Colonel von Bonin to take the German armed forces out of the integrated NATO command structure and strive for a purely national, conventional defense, to ideas for reducing the danger involved for civilian populations in the use of nuclear weapons by employing certain kinds of nuclear arms, for example, either nuclear mines (General Trettner) or tactical nuclear weapons with limited explosive power (General Kielmannsegg). Certainly it would be an oversimplification if one wanted to reduce the views of German military experts on nuclear weapons strategy to the ideas mentioned here. The air force and navy began to concern themselves with the problems created by nuclear weapons. In these services the influence of American ideas was strong from the very beginning. Within the German armed forces there were clear differences of
DILEMMAS OF NUCLEAR STRATEGY 77
opinion among the individual service branches. For example, in 1962 the press spokesman of the German Ministry of Defense, Colonel Schmückle, complained about the gap separating the air force’s ideas of war from those of the army experts.9 Only in the area of the fighting of a land war can we speak of an independent contribution of German military experts in nuclear weapons strategy. The fact that these ideas had little effect on the development of nuclear weapons strategy as a whole, however, cannot be explained as a result of the differences of opinion among the service branches, as is occasionally claimed. They were not greater than those among the service branches of other alliance partners. The decisive factor was that the operational requirements of the German army and the German views, which in part reflected the special situation of the Federal Republic, were in many respects incompatible with the theories of the US Air Force, which were dominant in the alliance. From the point of view of NATO doctrine, the German analyses presented problems whose insolubility created difficulties only for the Federal Republic. Although the German studies dealt with specifically military subjects, their direct influence on the fundamental debate about military policy, which dominated political discussion in the Federal Republic in the second half of the 1950s, must not be underestimated. They constituted, as it were, the background and framework of this debate, which was a decisive factor in forming the orientation of the Federal Republic in security policy and the German position on deterrence and nuclear weapons. In so far as this debate was about nuclear weapons, two questions were most prominent. One involved their possession and, above all, their symbolic function as an expression of maximum military and political power. On the other hand, the special German vulnerability in the event of a tactical use of nuclear weapons in Central Europe and its political and military consequences had to be considered. The course of the debate on the first question is only understandable against the background of problems many leading politicians still had at the end of the 1950s in determining the proper role of the military and military power. The Federal Republic’s renunciation of the production and possession of nuclear weapons had been accepted practically without opposition in 1954. To be sure, there was no serious political alternative to this decision, but the lack of any opposition showed that the political preconditions for the possession of nuclear weapons simply did not exist in the Federal Republic at that time. The acquisition of nuclear weapons would have required that Germany, as an independent factor in military power politics, play a role at least similar to that of other great powers; but at the time the Federal Republic was not prepared, or at least not fully prepared, to accept such a role. Precisely the symbolic importance of nuclear weapons as an expression of special military and political power made it impossible in the Federal Republic to attempt to strive for the power to decide when or how they should be used. Even the so-called Göttingen Declaration of German atomic scientists against the equipping of the German armed forces with nuclear weapons, which contributed to politicizing questions of nuclear strategy in the Federal Republic, was only made possible by a consensus on this question. In other respects the political orientations of the scientists involved differed greatly. The public effect of this declaration was due to its pointing out the special dangers which would result for the Federal Republic
78 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
from the use of nuclear weapons and to the suspicion that the German government might be playing down those dangers to promote its political ambitions. After the German government had made clear that it was not interested in having the power to decide when and how nuclear weapons should be used and that the discussion was actually only about their carriers, the campaign of the SPD against the nuclear weapons plans of the government under the slogan ‘Fight against atomic death’ quickly ceased to be a major issue at the end of the 1950s. The reason for the SPD’s failure was that it had merely mobilized confused emotions without itself being able to offer a clear position in the problems of defense policy created by nuclear weapons. Only the politicization of defense policy questions in the course of the debate led to an intensive preoccupation with the theories and plans developed in the USA and Great Britain. The concentration on military-strategic factors, which dominated American thinking, pushed the questions posed by nuclear weapons into the background. The public discussion of defense policy questions at the beginning of the 1960s again came to a halt. Until the end of the 1970s, questions of nuclear strategy were again a matter for military and civilian experts. The problems resulting from the special political and strategic situation of the Federal Republic also ceased to attract great public interest for the time being, but they continued to play an important role in the adoption of American nuclear weapons strategy whenever positions and decisions were involved which were not clearly rationally deducible. This became especially evident in the German attitude towards deterrence. If one sees the essence of deterrence in Brodie’s view cited above, that in the future military power could only serve to prevent wars, the question remains open as to whether this was intended to include all forms of war, or which wars would have to be prevented under what conditions. In the USA, consensus was achieved only in regard to a type of war described rather abstractly as ‘all-out nuclear war’ or ‘central nuclear war’. The Korean War clearly demonstrated that conventional wars could not or should not be prevented under all circumstances. And the ‘brinkmanship’ policy of Eisenhower and Dulles played in a certain way with the idea of a limited nuclear war to make effective use of American nuclear strategic superiority in foreign policy. From a German point of view the idea of deterrence could only be useful if it were understood in a very broad sense. In German eyes, the prevention of a limited nuclear war, which people in the Federal Republic imagined as being limited primarily to Europe, had to have almost the same priority as the prevention of a nuclear war between the superpowers. Even a massive conventional war in Central Europe could result in destruction which seemed unacceptable to many Germans after their experiences in the Second World War. For this reason, the idea of ‘intra-war deterrence’ never played as large a role in the Federal Republic as it did in the nuclear strategy discussion in the United States, and in the Federal Republic the inclination was much stronger to distinguish clearly between deterrence and waging war. Moreover, the doctrine of massive retaliation as originally defined by the American air force was not adopted by any political group in the Federal Republic. However, by the time this doctrine was accepted by the Federal Republic it had already developed into a theory of ‘graduated deterrence’. The special preconditions for the specifically German understanding of deterrence and thus of nuclear weapons strategy generally had widely differing effects. This can
DILEMMAS OF NUCLEAR STRATEGY 79
be shown in the views of deterrence held by the three individuals who presumably contributed most to the adoption of American ideas of nuclear weapons strategy in the Federal Republic: Franz Josef Strauss, who encountered American nuclear weapons strategy first as minister for nuclear energy and research and, later, as minister of defense; the philosopher and physicist Carl Friedrich von Weizsäcker, who made the American views understandable for the first time to a broad public in a series of articles in the weekly Die Zeit and exerted considerable influence on the position of the Protestant church in Germany, and the defense minister and, later, chancellor Helmut Schmidt. As all three of these men had conscientiously studied the American theories, their ideas on nuclear strategy seemed to differ little from each other at first glance. A closer look, however, reveals differences as well as common features in their attitude toward deterrence typical of the specifically German position on this question. Strauss viewed the Soviet Union and communism, not unlike Eisenhower and Dulles, primarily in categories of morality and power politics. In his eyes only a policy which emphasized power offered a real chance to stop the further spread of communism. Strategic nuclear weapons were for Strauss less usable weapons than the ultimate symbols of political and military power. Tactical nuclear weapons were easy to use and therefore made the atomic component of military power politically more effective. Deterrence was for Strauss primarily a new expression for the political effects of strategic nuclear power. He considered such power a solution to the problems which a security policy with a purely military orientation would have created for the Federal Republic. He did not deny that the Federal Republic could suffer immense destruction in the event that nuclear weapons were ever used, but as he considered a superior power position of the West the only possible chance of avoiding a military conflict with the Soviet Union, he believed this risk could be accepted. Above all, however, the idea of deterrence had the additional advantage of possibly being able to neutralize the special aversion of the Federal Republic to military power politics. Nuclear weapons seemed to be the suitable means to cleanse a Realpolitik based on military power, at least in the Federal Republic itself, of the stigma of militarism and at the same time to push into the background questions about the problems which would arise if that power were ever used. Of the three persons discussed here, Strauss’s position was probably closest to the American doctrine of massive retaliation, which, however—and this reflected the development of the American strategy debate at the time—he understood as a strategy of graduated deterrence. The concept of deterrence had for him a political function and importance for which the political preconditions were largely lacking in the United States. Unlike Strauss, Weizsäcker’s attitude towards deterrence was ambivalent. For Weizsäcker the enormous destructive power of modern technology embodied in nuclear weapons made clear the incompatibility of a policy based on power alone with the conditions of life in a world increasingly dominated by technology. In his view, deterrence therefore offered, for a limited time, the opportunity to develop a foreign policy which renounced war as a political instrument. In his opinion the doubts about nuclear weapons and the long-term effectiveness of deterrence gave the latter a political significance which went far beyond its pragmatic political function in Strauss’s policy. The extreme vulnerability of the Federal Republic, even in the
80 THE GERMAN ‘NUCLEAR DEBATE’ OF THE 1950S
event of a tactical use of nuclear weapons in Europe, made deterrence even more dangerous. In his military views, Weizsäcker dissociated himself from the doctrine of massive retalliation, but his ideas were quite compatible with the doctrine of graduated deterrence. Typical of his understanding of deterrence was the close connection he saw between it and a strategy of nuclear retaliation as well as the clear separation of deterrence from defense strategy, which could be deduced from that connection. His idea of deterrence was similar to that of Brodie but had a political component which was incompatible with the Real politik orientation of American foreign policy. For Helmut Schmidt, on the other hand, deterrence was a traditional defensive strategic principle not bound to nuclear weapons or a strategy of retaliation. With regard to nuclear deterrence, Schmidt’s arguments were similar to those of Weizsäcker, to whom he occasionally referred. He, too, considered the ‘balance of terror’ unstable and the chances of limiting a nuclear war small. He concluded that nuclear deterrence, as far as possible, had to be complemented by a balance of forces at the conventional level. In contrast, however, Weizsäcker sometimes warned that stressing conventional forces could again make war seem feasible. The decisive difference between Weizsäcker’s and Schmidt’s views was, however, in their understanding of deterrence. For Schmidt deterrence was not associated with any long-range political expectations and functions, as it was for Strauss and Weizsäcker. This does not mean that Schmidt and his cabinet simply abandoned the political aims Weizsäcker combined with deterrence. Rather, the new concept of deterrence was only possible because Schmidt did not regard defense policy as an independent factor but merely as the foundation of his foreign policy which not only stressed defense but also attempted to underpin politically the security of the Federal Republic by contacts and agreements with the Eastern bloc. We should mention that Schmidt first presented his views on defense policy in public at a time when criticism of the concept of deterrence by the intellectual left had begun to develop in connection with the opposition of the SPD to a possible equipping of the German armed forces with nuclear weapons.10 The 1980s’ strategy discussion in the Western alliance, especially the discussion about the American SDI project goes beyond earlier debates in that it is concerned with the basic idea of deterrence. It must be mentioned that many of the political factors which contributed to the special attitude of the Federal Republic in this question are still important. In the Federal Republic, the concept of deterrence is more closely associated with political ideas than in other countries, but criticism of deterrence in the Federal Republic is also stronger. The different positions on deterrence outlined here still play a significant role in German policy discussions, although they do not cover the whole spectrum of German opinion in this question. In view of the political expectations and hopes associated in the Federal Republic with the concept of deterrence, it would presumably not be easy for the country to accept a return to a classic defence strategy, even if this meant ‘deterrence by denial’. The seriousness of the difficulties which could then be expected would probably depend decisively on the security possibilities such an arrangement could offer the Federal Republic in the area of foreign affairs.
DILEMMAS OF NUCLEAR STRATEGY 81
NOTES 1. In the USA a controversy arose on this question when the contents of an internal planning document, the ‘Fiscal Year 1984–1988 Defense Guidance’, were leaked to the public. In this document it was stated, among other things, that ‘United States nuclear capabilities must prevail even under the condition of a prolonged war’. 2. Secretary of Defense Weinberger’s later reports modified this demand that the USA must be in a position in the event of a failure of deterrence to retain the upper hand. For example, the annual report for 1984 stated: ‘Should deterrence fail, we must be able to halt the attack and to restore the peace. In employing military force to restore the peace, the Reagan administration seeks to limit the scope, duration and intensity of conflict. In seeking to limit the scope of the conflict, our objective would be to deny enemy war aims in the theater in which the attack occurred.’ 3. Andre Glücksmann, Philosophie der Abschreckung, German edition (Stuttgart, 1984), p. 77. 4. As early as 1981 John Steinbruner wrote that ‘nonetheless, even 50 nuclear weapons are probably sufficient to eliminate the ability to direct U.S. strategic forces to coherent purposes’. He concluded: ‘In effect, this situation means that the aspirations for survivable, highly flexible strategic capability, articulated in recent years as a requirement of deterrence under a countervailing strategy and officially mandated in 1980 by Presidential Directive (PD) 59, are so far from realization that they do not provide a realistic standard for managing current forces.’ 5. ‘Governments are of course ruled by considerations not wholly different from those which affect even enlightened individuals. That the atomic bomb is a weapon of incalculable horror will no doubt impress most of them deeply. But they have never yet responded to the horrific implications of war in a uniform way.’ Bernard Brodie, ‘War in the Atomic Age’ in The Absolute Weapon (New York: Harcourt, Brace, 1946), p. 21ff., here p. 22. 6. ‘Under the technical conditions apparently prevailing today, and presumably likely to continue for some time to come, the primary targets for the atom bomb will be cities.’ Brodie, op. cit., p. 46. 7. Brodie, op. cit. 8. In this last point Heilbrunn’s ideas differ in many respects from the similar views of Liddell Hart. Cf. Otto Heilbrun, Conventional Warfare in the Nuclear Age (London: Allen & Unwin, 1965). 9. Schmückle in Der Spiegel, Nr. 41 (10 Oct. ‘1962), p. 44. 10. Here we should mention, above all, the criticism of Dieter Senghaas.
82
After NATO’s Dual Track Decision of 1979: Where Do We Go From Here? Christian Hacke
I. The Impact of NATO’s Dual Track Decision of December 1979 on Western Security The new strategic parity between the superpowers changed European and Soviet perceptions of the credibility of an American nuclear response, even a limited one, to a Soviet nuclear attack on Europe. It was feared that the United States was moving towards a posture of minimum deterrence, in which the United States would be forced to concede to the Soviet Union the potential for a military and political victory if deterrence failed.1 These changes in strategic parity and in regional superiority in Europe also suggested that the USSR was moving towards an ability to dominate escalation at all levels in time of crisis. The old position of the Carter administration, that large-scale attacks on Western Europe by Soviet medium-range forces could be deterred or countered by US strategic central deterrence alone, especially sea-based forces that were committed to NATO, did not convince the Western Europeans. This suggested that the NATO strategy of flexible response, already vulnerable at the conventional level, was now becoming vulnerable throughout the entire nuclear spectrum. Therefore, European attitudes towards the dual track decision of December 1983 developed and emerged in a strange mix of arguments, prejudices, and fears. Western Europeans concentrated on their domestic troubles and on the little progress which European detente produced. They watched the American fall from military and political superiority with mixed feelings. On the one hand, they feared a negative military impact on Western European security; on the other hand, they realized that the old political dependency of Western Europe on American dominance had been eased. Against this background, one can see that the NATO decision of December 1979 was a ‘rolling’ decision, calling for future deployments and arms control negotiations.2 NATO tried to give a military and a political answer to the military problem which resulted from the vast Soviet military build-up in Eastern Europe.3 The military problem was not simply the SS-20, but the entire range of nuclear modernization of which the SS-20 was an important part. NATO’s deficiency was in the Long Range Theater Nuclear Forces (LRTNF) systems based on the continent and capable of striking Soviet territory. The alliance had almost no LRTNF, only 156 F-111s based in England, and 50 British Vulkan bombers, which are extremely vulnerable. On the other side, the Soviet modernization by SS-20, Backfire, and
84 AFTER NATO’S DUAL TRACK DECISION OF 1979
SS-21, -22, -23, is aimed at decoupling Europe from US strategic forces in the sense that the USSR could threaten attacks while the Atlantic Alliance would have no response other than the resort to US strategic forces. The new enhanced vulnerability of American ICBMs sharpened the distinction and Europeans’ mistrust in American deterrence guarantees. Would an American President authorize a nuclear strike in Europe after a Soviet conventional attack if the Soviet would respond to such action by destroying as many as possible land-based ICBMs on American soil? The classic American argument, that their Poseidon arsenals would be enough to deter the Soviets in Europe did not sound as convincing as before. It was clear that the Soviet build-up of military arsenals in Eastern Europe pointed toward escalation dominance and absolute security. Against this background, the NATO Dual track decision was based on the following central arguments:
(1) With regard to alliance policy: to strengthen the confidence in the American nuclear guarantee. (2) With regard to defense policy: to deter the nuclear and conventional Soviet arsenals. (3) With regard to military strategy: to implement and to strengthen the strategy of flexible response and to deny the Soviet Union escalation dominance. (4) With regard to arms control: to include the problem of grey area weapons and to give the Soviet Union a detente incentive. (5) With regard to détente: to revitalize and to demonstrate the ongoing interest and belief that political détente must gain positve results in military arms control talks. (6) With regard to domestic politics: to legitimize a decision for armaments by offering a parallel track for arms control. The second rationale for the dual track decision was embedded in the fear that if the alliance allowed Soviet nuclear potential to grow to a thousand systems without an adequate counterweight on the Western side, the American nuclear guarantee would lose its credibility as the US lost escalation dominance. Furthermore, by accepting further growth of Soviet nuclear superiority below the intercontinental level, a new potential for political threat or pressure would emerge, and the domestic impact within Western Europe wouild be highly negative. On the one hand, the anti-nuclear protest, the challenge of political neutralism and pacifism, would grow, but at the same time the majority of the population in Western Europe, who are for an adequate military security posture and for NATO, would be deeply disillusioned: the capability of democratic majorities to define and implement security policy in the future would be at stake if the governments of Western Europe did not respond to the Soviet buildup. One could say that with the NATO decision to deploy Pershings and cruise missiles in Western Europe, the deterrence identity between Western Europe and the United States was emphasized. The military rationale behind the dual track decision was sound and convincing. Pershing II and cruise missiles would provide additional and credible deterrence
DILEMMAS OF NUCLEAR STRATEGY 85
without creating an offensive option against the Soviet Union and the Warsaw Pact. Cruise missiles in particular are seen as a weapon which meets the demands of extended deterrence. These new forces are less vulnerable to Soviet attack and, therefore, add to the deterrent effect. The new characteristics of these nuclear forces include the ability to reach targets in the Soviet Union, and therefore to deter the Soviets because their territory is no longer excluded in case of conflict. American nuclear forces of this kind in Europe demonstrate the military and political link between the United States and Western Europe and the Western European unity in the decision to deploy these forces demonstrated the willingness of Western Europeans to stand undivided despite regional and other political differences. In addition, American nuclear forces in Western Europe demonstrated American credibility with regard to extended deterrence, while American strategic systems concentrated mainly on America’s credibility with regard to central deterrence. In a historical perspective, Chancellor Schmidt’s speech of October 1977 in London made a strong comment on American vagueness concerning Western European interest vis-à-vis a credible extended deterrence posture,4 but Western European attitudes in general remain ambivalent. Western European governments on the one hand ask for a more defined responsibility in America’s nuclear decisions, but on the other hand want to return to the convenient position of supporting US nuclear decisions without being made accountable for them.5 It was only the absence of any illuminative or creative Soviet response which overshadowed the many inconsistencies among these various goals, which could not otherwise have been pursued with any sense of agreement. Had the Soviet Union, for example, offered a moratorium in September 1979 and kept it, the dual track philosophy might have failed. Consequently, the dual track decision embodied the many problems of the decade of the so-called ‘great compromise’ between domestic, foreign, and security policy problems within the East-West and the West-West context.6 The negative impact of the great compromise was seen in domestic politics. Under the impact of a growing anti-nuclear theology in Western democracies, the political leaderships did not have the courage to initiate a discussion about security and nuclear strategy and about the necessity for nuclear weapons with regard to the growing conventional and nuclear arsenals of the Soviet Union in Eastern Europe. The decision to deploy Pershing II and Cruise Missiles was not a decision of single modernization as it was officially represented, but implied many more serious aspects of defense and deterrence. Unfortunately, the ‘countering the SS-20’ argument became the predominant theme in the political debacle which, since 1979, had thus tended to overlook the overall requirements of NATO and US strategy in the face of Soviet improvement in nuclear capability and ability to dominate theater nuclear escalation. Unfortunately, the governments and experts were not able to argue convincingly about these problems, and so pulled the miraculous and frightful SS-20 out of the hat. There remain some central questions: after a more careful and realistis assessment of Soviet nuclear thinking and doctrine, would the deployment of cruise missiles alone have been sufficient for an extended deterrence? Pershing IIs are actually perceived by the Soviet Union as a strategic threat while the cruise missile is seen as defensive.
86 AFTER NATO’S DUAL TRACK DECISION OF 1979
The dual track decision shows three dimensions of Western European and especially West German vulnerability. The quest for American-West European coupling in the 1970s grew out of fear of a superpower bilateralism which might have neglected Western European security interests in a strict military sense. The search for American-Western European decoupling in the 1980s grows out of fear that a new Cold War between the superpowers might also endanger European security issues in a broader political East-West framework. The dual track decision was for the Federal Republic an instrument to minimize this vulnerability which grows out of a clear dependency on the USA. The negative consequences of SALT II for the extended deterrence posture in Europe were to be softened or minimized. The geopolitical and strategic position of the Federal Republic, which would make it the first and utmost victim in an East-West confrontation, shows the FRG’s vulnerability with regard to the Soviet Union. Therefore, every step towards solving the Intermediate Nuclear Forces (INF) problem was first tested and introduced in an arms control approach vis-à-vis the Soviet Union. As the last 15 years have shown, the West German foreign policy framework was broader in a détente climate, while the new ideological superpower hostilities have narrowed it. Consequently neither any cooperative nor any confrontative move in the superpower relationship should take place at the expense of German and/or Western European security interests. Before 1980, the INF option was kept open with regard to SALT II. After 1980, INF served as an instrument to bring the Soviet and American governments back to the arms control negotiating table.7 Sometimes it was hard to determine whether the Germans and West Europeans were concerned with a real fear of a Soviet threat or were much more interested in the never ending longing for an American commitment to the security of Western Europe. At the beginning of the 1980s this reassurance was perceived differently in Western Europe. Suddenly, the American commitment to the defense of Western Europe no longer provided assurance, but rather it produced the exact opposite, namely, fear. The formulation of the strategic defense initiative of the Reagan administration demonstrates this.
II. Challenges of the Security Policy of the Reagan Administration: SDI During the first two years of the Reagan administration there were no hints that a massive change in military strategy might occur. In fact, the opposite conclusion could have been drawn. The massive research and development and modernization of the strategic forces seemed to indicate that the Reagan administration wanted to strengthen the doctrine of deterrence and flexible response. The strong support of the Reagan administration for the stationing of INF arsenals in Western Europe seemed to indicate that military strength was considered a priority.8 It was no secret that Reagan personally was less than well disposed towards arms control in principle and the ABM treaty in particular. Long before Ronald Reagan became President, his aversion to the doctrine of mutual assured destruction as the basis for American security was well known. Reagan considers the strategy of retaliation to be immoral and unconscionable, echoing the central argument of the
DILEMMAS OF NUCLEAR STRATEGY 87
American and Western European peace movements which he condemned. It would not have been a surprise, therefore, had the President abrogated the ABM treaty at the time of its review in 1982. 9 Although revolutionary at the outset, Reagan’s speech of 23 March, in which he supported a reliable and total non-nuclear defensive system which would protect the soil and the lives of the US and its citizens and thus totally eliminate the threat of nuclear weapons, represented a logical conclusion to his previous thinking. In his second inaugural speech in 1985, President Reagan stated even the further goal of totally eliminating the threat of nuclear weapons through the development of SDI, although key advisers and parts of the foreign policy bureaucracy in the Reagan administration have now shifted away from this maximum goal, because ‘a perfect astrodome defense is not a realistic thing’.10 This reflects the most significant shift of argument within the Reagan administration and certainly contradicts the original proposals of the President himself. Much has been said about the strategic defense initiative as a means to protect the United States and maybe even Western Europe from a nuclear attack by the Soviet Union: but the vast majority of scientists and politicians have come to the conclusion that the combination of limitations imposed by scientific, technological, and system engineering concerns, as well as the high cost and the potential for countermeasures by the Soviets make a perfect defense illusory.11 The Reagan administration, therefore, has settled for a second line of defense, an SDI which concentrates on point defense. SDI’s purpose is no longer to abolish deterrence but to enhance it by giving additional protection against small attacks. This might refer to cities or to land-based missile sites and command and control centers. By reducing the total number of incoming warheads, the damage to civilian life and military property might be reduced substantially. SDI might even offer the prevention of nuclear winter and climactic catastrophe. Furthermore, the Reagan administration argues, SDI might force the diversion of Soviet resources in that they would be forced to improve their offensive countermeasures and thus a variety of preferred strategic options would be denied to the Soviets. None of these objectives however provide a way out of the fundamental dilemma of the present condition of the strategy of mutual assured deterrence. Even the acknowledged shift from total to point defense offers no better alternative. On the contrary, it broadens the dilemma, as L.Freedman has shown.12 A new, even more dangerous, mix of offensive and defensive measures comes into focus. Officials of the Reagan administration acknowledge that offensive weapons still have the advantage in all security considerations. Some defenders of SDI argue, therefore, that SDI can deny the Soviet a first-strike capability: but as the Scowcroft Commission has observed, the Soviets do not possess a first-strike capability. In the absence of a threat to the American retaliatory capability, this vision of a limited defense system seems to be not only useless, but dangerous. It triggers Soviet countermeasures, and an increase in offensive missiles. The result is an acceleration and widening of the arms race in both offensive and defensive terms. Furthermore, the Soviet Union could be tempted to pre-empt with anti-satellite weapons against the space-based defense systems; or even to pre-empt with nuclear weapons against the offensive forces of the other side to gain greater stability and success for their own defense system. It is precisely this new dangerous mix, which the Soviet Union fears:13 it is
88 AFTER NATO’S DUAL TRACK DECISION OF 1979
no secret that these systems, although originally planned as defensive, might in the future open new offensive options for the United States. Much has been said about the enormous financial burden, estimated as $10 billion, which would be imposed by even an ABM defense for ICBMs. For a damage limitation strategy within a deterrence strategy, the cost might be in the $100 billion range, a cost which would reduce the financial options for strategic nuclear and conventional forces. Imperfect strategic defense, therefore, might not reduce the danger of the present state of MAD (Mutual Assured Destruction), nor could it reduce the potential damage to the United States if deterrence failed. The ABM treaty prohibits any SDI-type defense; and the US should not abrogate or modify the ABM treaty despite new technological developments. This treaty is central to the effort to limit the strategic arms race by keeping at least the so-called defensive weapons out of it.14 Also, an ABM System seems to be ill-suited as a bargaining chip for arms control. The opposite is more likely. A certain state of offensive forces on the American side may serve as an incentive for arms control or reduction, but not ABM. The effectiveness of real strategic defense will for the foreseeable future rest on offensive weapons, on their constraints as well as on their technological innovations. Mutual assured stability, as the only central objective which is shared by both critics and supporters of SDI, will be reached more directly and more sensibly by arms control negotiations which concentrate directly on offensive limitations. Disarmament is cheaper and makes more sense. The problems of verification are minor in comparison to a very dangerous and never-ending transition phase of a new mix of offensive and defensive systems. Arms control is conducted under human control and logic, and is not dependent on a highly complex and vulnerable systems of millions of computer decisions where just one failure is one too much.15 The SDI decision bears some similarities to Tirpitz’s decision to accelerate the build-up of the German fleet at the beginning of this century. He closed his eyes to the fact that the British would develop countermeasures. The consequences of a system failure in the nuclear era would be murderous: after building a nationwide umbrella or a point defense system, far more warheads than would have been deployed without this initiative would be ready to be launched against American soil. Many critics see a fundamental ambiguity around SDI: is it really designed to restructure the superpower relationship on a safer and more equal basis? There might be more than a coincidental connection between SDI and Reagan’s indictment of the Soviet Union as an evil empire. The strong anti-Soviet and anti-Communist rhetoric and action the Reagan administration has produced so far seem to reflect a different mood, a search for a new vision of absolute security; but President Reagan offers a treacherous prospect of invulnerability. There is no way to escape from the harsh realities of the nuclear age through quick technological fixes. What was launched as an epochal beginning, might be seen in a historic perspective as an episode that illserved the personal reputation of President Reagan as a responsible and far-sighted leader. While the implementation of the second part of the dual track decision in December 1983 strengthened the political cohesion and the strategic basis of the Atlantic alliance, the SDI initiative might just trigger an opposite development.
DILEMMAS OF NUCLEAR STRATEGY 89
III. SDI and Western Europe With regard to military deterrence and political reassurance, the gulf between the Reagan government and Western European governments has deepened as a function of the different experiences America and Western Europe went through in the 1970s. Against the background of a loss of American power worldwide, the Reagan administration is playing the East-West game principally with two balls—defense and deterrence; while Western Europeans play constantly and with limited success with three balls: defense, deterrence, and détente. The Western Europeans try to work with the Soviets in building up a net of détente and rest more on deterrence than on defense with the aim of keeping up the shaky political compromise between Eastern and Western Europe which developed during the period of détente. From this perspective, the Soviets are a powerful but calculable adversary who can be kept within European limits by a shrewd mix of containment using military, political, economic, and diplomatic tools. The Reagan administration, on the other hand, seems in European eyes to react with different means to a differently perceived challenge. The Reagan administration perceives the Soviet Union as an aggressive power which seeks global advantages and successes either directly or by proxy. This view is probably shared by many Western European governments, but the gulf within the Western alliance deepened because the Reagan administration initiated a loose and ideologized rhetoric against the Soviet Union. This was paralleled by military programs which indicated substantially that the Reagan administration had adopted an offensive posture towards the Soviet Union.16 The first three years of the Reagan administration produced a massive build-up and modernization of strategic forces, the MX, B-1, and Trident II, which consumed $25 billion in the fiscal year of 1983, and will total over $200 billion in the subsequent five years. This effort to close the window of military vulnerability might be partly understandable with regard to the devastating experience of the US in the latter half of the 1970s, but it also opened new windows of political vulnerability. The Rapid Deployment Force (RDF) and other capabilities for projecting American military power into local and regional crisis spots were designed to establish a credible presence which would put the US in front in the event of a Soviet or Soviet-proxy challenge. Not the military build-up, but the political context of the Reagan administration, its dogmatic, anti-communist rhetoric, stands in stark contrast to the basically relaxed Western European attitude towards the Soviet Union. While Western Europeans deal with the Soviets as an adversary who should be deterred, contained, and confronted, the Reagan administration seems to confront first, contain second, and deter the Soviet Union under an entirely different priority. The American government started to refurbish its own strategic power before resuming START (Strategic Arms Reduction Treaty) talks. The intention was clear: a massive build-up in military strength was seen as an essential prerequisite for the future arms control talks. Beyond that, the Reagan administration attempted to organize a global response to the Soviet arms build-up and intervention directly or by proxy through efforts such as the so-called strategic consensus in the Middle East. It was not so much the American arms build-up, but the new political approach which disturbed many Western Europeans. With the ‘strategic consensus’, the Reagan
90 AFTER NATO’S DUAL TRACK DECISION OF 1979
administration put global pressure on allies and friends to work within and join a network of treaties which were to strengthen bilateral ties of various countries with the United States, especially in the Middle Easst and in Central America, in an attempt to strengthen the regional anti-communist and anti-Soviet forces all over the world. Instead of focusing on the individual, local and historic aspects of regional conflicts, the Reagan administration chose to press them all into a single anti-Soviet maneuver. Reagan’s neglect of the Camp David framework for peace in the Middle East,17 and for European détente interests and initiatives demonstrates this. Reagan’s bipolar re-ideologized view is also not always congruent with the complex realities in Africa and Latin America. When, in August 1981, the President authorized the production of neutron weapons without NATO consultation, ignoring the alliance controversy over the potential deployment of these weapons during the Carter administration, he put NATO under severe pressure. When in October 1981, Reagan said that he ‘could see where you could have the exchange of tactical nuclear weapons against troops in the field without bringing either one of the major powers to pushing the button’, he opened the way for speculation about a new doctrine of limited nuclear war in Europe which would preserve the homelands of the Soviet Union and the United States from nuclear attack. This loose rhetoric on nuclear war strategies stimulated the anti-nuclear movement in the United States, Western Europe, and worldwide. The strong anti-Soviet attitude and open discussion of the requirements for fighting a ‘protracted’ nuclear war in which American forces were supposed to ‘prevail’ and be able ‘to force the Soviet Union to seek earliest termination of hostilities on terms favorable to the U.S.’ and to ‘decapitate’ the Soviet military system by destroying its leadership and its central lines of communication, made even pro-American politicians, analysts, and citizens on both sides of the Atlantic wary.18 Michael Howard expressed this feeling:
When I read the flood of scenarios in strategic journals about first-strike capabilities, counterforce or countervailing strategies, flexible response, escalation dominance and the rest of the postulates of nuclear theologies, I ask myself in bewilderment: What is it about…What are we talking about? Has not the bulk of American thinking been exactly what Clausewitz described—something that, because it is divorced from any political context, is pointless and devoid of sense?19 Against this background the agreement to station Pershing II and cruise missiles was one of the rare decisions within the Western alliance that demonstrated a high degree of accord between the United States and Western European countries such as Germany, Britain, and Italy. When President Reagan presented SDI on 23 March 1983, Western European governments reacted instinctively with great reluctance and with a critical attitude because the great compromise in re-establishing extended deterrence by stationing Pershing II and cruise missiles was endangered. The longstanding and basic attitudes of Western Europeans on the one side and the US on the other came again into the open:
DILEMMAS OF NUCLEAR STRATEGY 91
(1) Despite many doubts, Europeans remain convinced that nuclear deterrence is essential, logic, basic, and desirable to their security, and that it can be based only on mutual assured destruction. Extended deterrence is the focus of Western European interests. (2) Western European détente interests are regional and political. They stick to the dual Harmel philosophy of military strength and political détente which must be pursued on parallel tracks. (3) The nuclear forces of France and Britain and the increasing European investment in the scientific and commercial aspects of space activities and interests of their own are basic. (4) Vulnerability is a natural state for Western Europeans. Extended deterrence based on mutual vulnerability corresponds to the historic European experience, but is new and just three decades old for Americans. SDI is the product of these different perceptions. It indicates the willingness of the United States to try to break out of the deterrence system which shaped Western security for the last four decades. Western Europeans, on the other hand, fear that SDI might also be conducive to a limited nuclear war option which the Reagan administration referred to earlier. They fear also that SDI might create two different security zones.
IV. Western Europe’s Reaction to SDI Evaluating the effect of SDI on European security includes the question of what effect SDI will have on the present East-West climate? How does SDI affect arms control negotiations and European perceptions of US interests and commitments visà-vis Europe, and what is the reaction from the public at large? In December 1985, Britain became the first European nation to sign a formal agreement with the US government guiding British participation in the strategic defense research program. The memorandum of understanding provides a framework within which British companies, University centers, and other research establishments will conduct SDI research. It paved the way for a step-by-step diplomacy by the Reagan administration through which the Americans have managed to deal with European partners unilaterally. Britain’s participation is not without qualifications, however: first and foremost, it is contingent upon the program remaining within the parameters of the 1972 ABM treaty and upon a US agreement not to deploy defensive systems automatically following a successful research effort. On the other side, the British government could not reach its goal to get an American guarantee that British industries would get SDI contracts for $1.5 billion. In May 1985, French President Mitterrand declared that France would not join the American SDI research program, and instead launched the initiative EUREKA. This French-initiated research program has as one of its objectives the creation of a single integrated European market that would push Western Europe on to the global stage as a leader in the development and production of new, sophisticated technologies, competitive with both Japan and the United States. Addressed to the 12 members of
92 AFTER NATO’S DUAL TRACK DECISION OF 1979
the European Community, as well as Austria, Switzerland, Sweden, Finland, Norway, and Turkey, the EUREKA concept reflects Western Europe’s ambition to create an alternative to US space research. On 5 and 6 November 1985, at a meeting in Hanover in the Federal Republic of Germany, the Western Europeans agreed to a charter for the EUREKA program and approved ten mid-term projects, including development of a standard microcomputer for educational and private uses, a new silicon computer chip, a high speed computer, several laser projects, and advanced optical electronics, to be funded primarily by the participating companies. While the EUREKA concept presumes a transnational financing effort, national funding will likely be at modest levels and for projects which are already under way or have been envisaged by the respective European governments in the area of space research. Thus far, only France and West Germany have pledged governmental funding to EUREKA. In contrast, the British government refuses to allocate new governmental resources to the project, stating instead that Britain’s financial contribution will be secured from a combination of sources such as industry’s own bank account, financial institutions, and existing statebacked research schemes, for example, the Department of Trade and Industry’s support for innovative programs. EUREKA has been characterized by some observers as a defensive reaction to the American Strategic Defense Initiative. Charles Hernu, the former French minister of defence, has denied this characterization of EUREKA, stating that even if SDI had not existed, EUREKA would have been necessary. While the official position of the French government is that EUREKA is a civilian program, its potential for military applications must be taken into account. EUREKA on the whole, however, represents a program in the development of civilian technologies. Among government circles in Europe and America there is discussion of the socalled European Defense Initiative (EDI), which might be designed as a supplement to SDI, although so far there is not official program for this, only supporters who represent European interests in the creation of such a system. It is based on the premise that missile defense technologies can be credibly developed, but that the US SDI is not fully responsive to the security requirements of Western European nations. Such a program would require the development of political solutions to military issues such as a common American-West European view about the nature of the Soviet threat to Western Europe, about the nature of US-European cooperation and division of labor, and, of course, about transatlantic collaboration and technological transfer.
V. The Position of the German Government Towards SDI In addition to Britain, the Federal Republic has reluctantly begun discussions of SDI matters with the Reagan administration. On 27 March 1986, the Secretary for Economics Bangemann and the American Secretary of Defense Weinberger signed two agreements. Their purpose is to establish a framework for the participation of German industries in SDI research, and to make sure that the technological transfer would take place on a fair give-and-take basis. The two accords contained two main points:
DILEMMAS OF NUCLEAR STRATEGY 93
(1) direct participation by West German companies in research on SDI projects; (2) a general agreement under which the two countries will share technology in commercial and other fields that might develop from the research. German participation is expected to be centered on optical systems, which could be key elements in devices intended to track enemy missiles, and would be crucial to mirrors which would be used to bounce laser beams through space until they strike their targets. However, the disclosure to the press of the secret accords of the American-West German SDI deal have made it clear that the general stipulations on secrecy, the transfer of technology, the rights of ownership, and other legal and commercial aspects have been treated rather loftily. The Export Administration Act, approved by Congress in 1985, allows no excepts for SDI to the American position towards technology transfer. Despite the fact that the financial scope of German participation on SDI will be DM50 million annually over the next five years (about $26 million), which is very small compared to overall German industrial investments (DM40–50 billion/annually or $20–25 billion), those two agreements have come under heavy domestic attack in West Germany. Furthermore, one must take into account that views about SDI had to be coordinated within the Kohl government. Basically, the majority of the Christian Democratic Union is in favor of SDI because it might lead to strategic innovations and to technological spin-offs, which might trigger new initiatives in high technology. The Social Democratic Party is strongly against SDI, as are parts of the Free Democratic party which, at the moment, constitutes together with the CDU the government in Bonn. The Free Democratic Party interprets the SDI agreements as a low-profile commitment by West Germany to the research program of SDI. Their concern is that its strategic implications might erode NATO’s policy of nuclear deterrence and sour West Germany relations with the Soviet Union and the GDR. From this perspective, the agreements are seen as an essentially commercial enterprise. The Free Democratic Party and even more so the parliamentary opposition, the SPD, and, of course, the Greens, express skepticism about the wisdom of joining a project like SDI that could unleash a new arms race in space, undermine the strategic balance of nuclear forces, and damage the Federal Republic’s relations with the countries of central and eastern Europe. By stressing the commercial aspects and playing down the West German endorsement of SDI’s controversial political aims, the FDP and of course the parliamentary opposition hopes to deflect criticism from the Soviet Union and East Germany, which have both warned that West Germany’s role in SDI would harm relations with them. The CSU, led by Franz-Josef Strauss, and parts of the CDU want a stronger commitment by the Federal Republic to SDI. Strauss’s interpretation is in accord with the desire of the Reagan administration to get a clear German endorsement of SDI. The Reagan government concentrates on the political aspect of West Germany participation as the most important ally in Western Europe. Public opinion in West Germany is divided about the issue, but one could say that a slight majority might be against SDI, because it fears the resulting political costs
94 AFTER NATO’S DUAL TRACK DECISION OF 1979
for the foreign policy of the Federal Republic, especially within the East-West context. Many Germans feel that by signing the SDI agreements, the Federal Republic has taken a political responsibility for SDI, which even Chancellor Kohl does not want to carry. The West German government therefore stresses that participation is restricted to private companies and that the government accepts no political obligation. The Bonn government views participation in the SDI plan as a commercial undertaking for German companies to stay abreast of new technologies. The United States, however, sees the program as a military venture and wants primarily political links established with the Bonn government. By playing down the military aspects of participation in SDI, the West German government hopes to limit hostile political retribution from East Germany. East Germany’s communist authorities, as well as the Soviet’s, have warned since the agreements that Bonn’s support for SDI will have a negative impact on relations between the Federal Republic and the GDR or the Soviet Union.
VI. Conclusion Despite the two agreements between the United States and Britain and the Federal Republic of Germany, it is clear that Western Europeans on a governmental level, but even more in the broad spectrum of the European party system and more still within the public, are very critical of the SDI initiative. The reasons are political, military and economic. First of all, Western Europeans fear that SDI may endanger the strategy of flexible response. The fundament of deterrence by mutual assured destruction might be abandoned for an illusion of mutual assured security. SDI is considered useful only as a bargaining chip for the arms control negotiations in Geneva, but American Secretary of Defense Weinberger and President Reagan have again and again stressed that strategic defense is one of the very highest priorities of the American government and of the President, and that it is not a bargaining chip. The overall concerns of Western Europeans concerning SDI concentrate on six points:
(1) Even an imperfect defensive system would lead to an acceleration of the offensive arms race. (2) Deployment of defensive systems for Europe would provide no additional security for Western Europe. There would be no effective defense against Soviet artillery, cruise missiles, and modern aircraft, and probably none against short-range ballistic missiles. (3) Politically seen, SDI nullifies the efforts of Western European over the last 15 years concerning central detente and security problems in Europe. The stationing of Pershing II and cruise missiles was seen as a symbol of an increased solidarity of risk within the alliance precisely because the American weapons can reach targets in the Soviet Union from Western European soil, thus further exposing the United States to Soviet retaliation. SDI indicates an American attempt to break out of this shared risk.
DILEMMAS OF NUCLEAR STRATEGY 95
(4) SDI introduces the fear to Western Europeans that a strategic decoupling of America’s security from that of Western Europe might lead to a ‘fortressAmerica mentality’ with an unprotected and unprotectable glacis in Western Europe. (5) Furthermore, Western Europeans fear that the preoccupation of the Reagan administration with strategic defense, could in the long run weaken the existing defense in Europe which rests on conventional defense. (6) The specter of a new strategic arms race could make the cost of defense more burdensome not only in conventional terms, but in terms of scientific and industrial resources as well. The European reaction to SDI has been in a way ambivalent. While hostile in substance, governments took mostly a ‘wait and see’ attitude. No European government wishes to confront alone the strategic options chosen by an American president. For the first time, Western Europe found itself in a position that the alliance was created to avoid: a Soviet superiority in nuclear and conventional military strength which could not be offset by the threat of US strategic involvement in an European war. There is deep distrust with regard to the vision of a 100 per cent proven defense system that would remove the threat of nuclear missiles. History has shown the Western Europeans over the last 40 years that the most densely weaponized region in the world can avoid conventional and nuclear war through a mutual balance of destruction. This is the paradox Western Europeans are prepared to live with in the future. They distrust the rapid solutions envisioned by the American president. Unfortunately, two negative trends are reinforcing each other: the more Europeans show an aversion to nuclear weapons, the more the United States will avoid committing its security to the uncertainties of European politics. It was the main goal of the Soviet Union to undermine the unity of the West. Thus far, Western cohesion has remained strong, but President Reagan’s strategic defense initiative, ironically, could change this. The Reagan administration would like to make the world safe against nuclear attack, but might end up undermining the cohesion of the West. Western Europeans see this initiative as a classic case of good intentions with bad results. President Reagan is longing for a dream, for the world in which he grew up: a world without nuclear weapons. But false hope is a bad guide to security. Furthermore, it seems that SDI reflects not only a technological hubris in the face of the nature of nuclear weapons, but also a complete misreading of the relationship between threat and response in the nuclear calculus of the superpower. Even a 95 per cent kill rate would be insufficient to save either society. Western Europeans see clearly that neither of the two superpowers can tolerate the notion of impotence in the face of the opponent’s arsenal. Consequently, an improved American defense will stimulate energetic Soviet efforts to ensure the continued ability of Soviet warheads to hit American targets. Western Europeans ask themselves whether an American monopoly of strategic defense would strengthen the American engagement for Western Europe. On the other hand, would a Soviet monopoly broaded Soviet foreign policy possibilities? Because nobody wants to give the other superpower a monopoly in SDI, a genuine dynamic in the arms race is the consequence. If Western Europe is not included in the strategic defense system, there will be two levels of
96 AFTER NATO’S DUAL TRACK DECISION OF 1979
security: one for the superpowers, and a second for Western Europe which would remain unshielded. Furthermore, Western Europe would still be vulnerable to nonballistic nuclear systems. British and French efforts to modernize their nuclear forces would be in vain. Given the very short time it takes Soviet missiles to reach targets in Western Europe, SDI could not provide any defense for Western Europe, even if the US chose to expand the SDI umbrella. The security dilemma of Europe can probably never be solved. Europeans, and especially Germans, are afraid, on the one hand, that the United States will not use its nuclear arsenal, and on the other hand, they are afraid that they will. Europeans feared the dwindling credibility of American deterrence, and they now begin to fear American attempts to restore it. They worry about too much arms control, which might sacrifice German interests on the altar of superpower community, and they worry about too little arms control, which might endanger German and European interests on the battlefield of superpower confrontation. This dilemma now has become the norm, and there are no technological fixes. Mutual assured destruction as deterrence will always be paralleled by uncertainties, not only within the East-West relationship, but also within the Western alliance. The strategic defense initiative would create only a false idea of clarity and security. The elimination of the fear of a nuclear war may, as paradoxical as it may sound, not give more security. Europe would still be vulnerable to conventional war or limited nuclear war. This has been Europe’s nightmare, one which extended nuclear deterrence was to insure against. Under an SDI umbrella, Americans would no longer have a need for joint security. All of this shows that the influence of SDI in Europe is potentially divisive. The debate in Western has sharpened on whether to take up the American initiative in an attempt to reduce its negative effect, or to support SDI as a whole. Basically, SDI has shed light on the wish for an independent European policy. EUREKA must be seen as an attempt to reduce the divisive influence that SDI has had on inter-European relations. Internal differences within the European countries will resurface: and we have already seen the beginning of this in the French fears of German and Italian participation in the SDI research program. But the central problem of SDI is the potential split between the USA and Western Europe on the question of how to perceive and how to meet the Soviet threat. The different perceptions of the United States and Western Europe concerning the Soviet threat did not emerge in the 1970s or even as late as the 1980s, but has been consistent from the 1950s on: the peoples of Western Europe saw less danger of a direct Soviet attack and therefore wanted defense on the cheap. They remained reassured, though whether this reassurance came from shrewdness or from selfdelusion, from confidence in American nuclear supremacy or basic disbelief in the reality of any Soviet threat, it would probably be impossible to say. In any case, throughout the 1950s and the 1960s, deterrence and reassurance both worked.20 Nuclear weapons are a frightening as well as reassuring element of the postwar world: on the one side, they symbolize the willingness jof the US to risk her own security for the sake of Europe, but on the other, they are frightening in an existential sense and—from a West German perspective—they imply that Western Europe would be the prime and foremost target and theater for nuclear war. Despite changes within the last two decades, Western Europe remains the central area of East-West superpower conflict which could be encapsuled by deterrence. The experience of nearly four decades has shown that nuclear arms in Central Europe strengthen the
DILEMMAS OF NUCLEAR STRATEGY 97
defensive deterrence posture of the West. Extended deterrence by the threat of first nuclear use serves as a rationale the Soviets have had to reckon with. Against the background of about 150 wars outside Europe since the Second World War, it seems correct to assume that the state of peace in Europe might be the result of the existence of nuclear weapons on both sides in a rough balance of terror and power. The Janus-headed impact of nuclear weapons on the security of Western Europe is essential for an understanding of deterrence: ‘For Europe the atomic weapon is both a much needed hope of effective defense and a terrible immediate peril greater than ever for this country.’21 It is this nuclear Janus, one head pointed toward the enormity of destruction, the other toward the undeniable possibility of military use, that we face. We stand in the middle, not knowing whether deterrence is strengthened by maximizing the enormity of destruction through countervalue threats or by maximizing the credibility of use through more ‘rational’ counterforce strategies. SDI might bring out a scenario that would offer more fear, insecurity, and political anxiety as a function of the superpowers attaining a defensive shield to protect themselves. The possibilities for Europeans to influence the direction of the strategic development between East and West is very small. The increase of European independence in political terms may further decrease European possibilities for influencing the East-West debate. There is not a third way, and there is no such thing as regional European security in an insecure world. Western Europe may put more pressure on the US by rejecting the SDI offer, although this would be difficult given the history and present circumstances of the Atlantic alliance. The of fensivedefensive arms race has not yet started. Europe has a responsibility to act, but this responsibility if mired in the reality of a fundamental dilemma. Neither dependence nor independence will assure Western European stability in a political or strategic sense, and the chances are slender that either the United States or the Soviets will play the game on European terms. Eastern European governments are forced to keep silent when the Soviets move, but Western European governments are not. If they do not stand up and argue with the United States for a security policy and strategy that fulfils the need for all the big Ds—Deterrence, Defense, and Détente—the public will react. SDI, in contrast to the dual track decision of 1979, does not meet the needs for all three Ds, and thus signals trouble ahead in the Atlantic alliance.
NOTES 1. See Paul Nitze, ‘Assuring Strategic Stability’, in Foreign Affairs (January 1976), p. 227. 2. See Stanley Hoffman, ‘NATO and Nuclear Weapons’, in Foreign Affairs (Winter 1981/82), p. 331. 3. See James A.Thomson, ‘The LRTNF Decision: Evolution of US Theater Nuclear Policy 1975–1979’, in International Affairs, Vol. 60 (1984), pp. 601ff. 4. Helmut Schmidt, ‘The 1977 Alastair Buchan Memorial Lecture’, in Survival (Jan./Feb. 1978), pp. 2–10. 5. Christoph Bertram, ‘The Implications of Theater Nuclear Weapons in Europe’, in Foreign Affairs (Winter 1981/82), p. 313. 6. The security and détente policy of the 1970s can be characterized as a great compromise in many ways:
98 AFTER NATO’S DUAL TRACK DECISION OF 1979
1. Détente policy between East and West, in which three variations of détente developed: the American, Soviet and Western European version, the last developed its most significant character in German Ostpolitik; 2. a great compromise within the Western Alliance, which was based on the Harmel philosophy to interpret security by military and preserve it by political means in a balanced manner, which certainly must not endanger the fundamental military security requirements; 3. a great compromise with regard to the relationship between governments and population in Western democracies. See also Christian Hacke, Von Kennedy bis Reagan, Grundzüge der amerikanischen Auβenpolitik 1960–1984 (Stuttgart: Verlag Klett-Cotta, 1984), p. 212ff. 7. Ulrich Weisser, ‘Perspectives of German Security Policy’, in Naval War College Review (Nov./Dec. 1984), p. 86ff. 8. ‘The development of cruise and ballistic missiles to Europe does not move NATO away from its strategy of flexible response. Rather, the deployment decision is essential to sustaining NATO strategy. In particular, it will link more firmly the US strategic deterrent to the defense of Europe.’ Richard Burt, in ‘Overview of Nuclear Arms Control and Defense Strategy in NATO’, Hearings, Sub-Committees on International Security and Scientific Affairs and on Europe and the Middle East, Committee on Foreign Affairs, House of Representatives, 97. Congress, 23 Feb., US Government Printing Office, Washington, 1982. 9. William Broad, ‘Reagan’s “Star War’s” Bid: Many Ideas Converging’, New York Times (NYT), 4 March 1985. 10. Lieutenant-General James Abrahamson, Director of the SDI Office in Science (10 August 1984). 11. See Harold Brown, ‘The Strategic Defense Initiative: Defensive Systems and the Strategic Debate’, in Survival (March/April 1985), p. 59. 12. Lawrence Freedman, ‘The Star Wars Debate: The Western Alliance and Strategic Defense’, in Adelphi Papers No. 191, ‘New Technology and Western Security Policy’, Part III, pp. 34–50. 13. See ‘Space-Based Defenses: A Soviet Study, Committee of Soviet Scientists for Peace against Nuclear Threat’, Institute of Space Research, USSR Academy of Sciences, 1984, in Survival (March/April 1985), pp. 83–90. 14. Thomas K.Longstreth, John E.Pike, and John B.Rhinelander, ‘The Impact of US and Soviet Ballistic Missile Defense Programs on the ABM Treaty’, A Report for the National Compaign to Save the ABM Treaty, March 1985. 15. See also Lawrence Freedman, ‘The Star Wars Debate…’, p. 45. 16. ‘We must not pursue a defense strategy that anticipates a point by point response to Soviet aggressive actions, but rather one which permits us to take full advantage of Soviet vulnerabilities.’ Caspar Weinberger, Statement before the Senate Armed Services Committee, 4 March 1981, Department of Defense, 1981.
DILEMMAS OF NUCLEAR STRATEGY 99
17. See Christian Hacke, ‘Amerikanische Nahost-Politik, Kontinuität und Wandel von Nixon bis Reagan’ (München: Verlag Oldenbourg, 1985), p. 204 f. 18. See Seyom Brown, The Faces of Power, Constancy and Change in United States Foreign Policy from Truman to Reagan (New York: Columbia University Press, 1983), p. 607. 19. Michael Howard, The Causes of War (London: Allen & Unwin, 1984), p. 141. 20. See Michael Howard, ‘Reassurance and Deterrence’, in Foreign Affairs (Winter 1982/83), pp. 309–24. 21. J.Robert Oppenheimer, ‘Atomic Weapons and American Policy’, in Foreign Affairs (July 1953), p. 529.
100
IV: Arms Control and Nuclear Strategy
102
Deterrence and Arms Control George W.Rathjens
In an essay on deterrence one should probably distinguish right from the outset between, on the one hand, punitive deterrence, that is, the deterrence through denial, that is, through having war-fighting capabilities sufficient so that an attacker can be denied a high probability of achieving his objective(s). Either might, in principle, be accomplished with either nuclear or non-nuclear arms. In the remainder of this article I shall be concerned mainly with punitive deterrence based on nuclear capabilities, and unless otherwise qualified, this will be what I have in mind when I use the term ‘deterrence’ without qualification. Basing security on such deterrence is a questionable proposition. There is severe criticism on both moral and pragmatic grounds from both the left and the right. From both quarters, there is also criticism of arms control; not only of the failures to accomplish much, but also of the approaches that have been used and, indeed, of the concept. Advocates of both deterrence and arms control contend that there can be a constructive linkage. Assuming defense policy grounded on the bedrock of deterrence, one can envisage arms control efforts contributing to security in important ways: by constraining or even reducing the costs of maintaining adequate deterrent capabilities, and by contributing to crisis stability and arms race stability, that is, by making it less likely that deterrence will fail in a crisis and by dampening any impetus for the nations involved to develop or build forces in the hope of achieving military advantage. I will discuss deterrence and arms control and their interaction, concluding that the two concepts, either singly or in combination are, at best, weak reeds on which to rely in trying to escape from the threat of nuclear war.
Deterrence as Seen From the Perspective of Two Actors With all that has been written about deterrence, the basic problems and arguments will be common knowledge to most readers, but some summation may be in order. Against the threat of an unconstrained attack by either superpower against the other, or against any other nation or nations, there can be little possibility of greatly mitigating damage through defense, and none of damage denial, President Reagan’s Star Wars initiative notwithstanding. Hence the emphasis of the last 20 or so years on punitive deterrence to deal with that threat. But against lesser threats, there may be
104 DILEMMAS OF NUCLEAR STRATEGY
more of a choice. Defense, that is, preventing the realization by an adversary of his objectives, may be feasible but costly; punitive deterrence may be less costly but also perhaps less credible. Credibility may be enhanced by threatening a limited response: one perhaps roughly proportionate to the event, but this may be excessively risky; there may be a higher-than-wanted likelihood of escalation occurring, and this may be the case whether or not one has, or thinks one has, a capacity for ‘escalation dominance’. Alternatively, credibility can be made acceptably high by posing the threat of a disproportionate response and by manipulating the probability of such a response through the deployment of weapons and delegation of authority and capacity to use them so that the likelihood of their use will seem persuasively high, even though use, after the fact of provocation, would be irrational. Note in this connection that there will always be some uncertainty about deterrence. Schelling has referred to ‘the threat that leaves something to chance’. At the one extreme, we can never really speak sensibly of ‘assured destruction’ except perhaps with referrence to physical capacity. There will always be some possibility of no retaliation after a devastating attack: after all, what purpose would retaliation then serve? At the other extreme, that is, even if the provocation is slight or moderate, a highly disproportionate response can never be excluded as totally incredible. We will have what McGeorge Bundy has called ‘existential deterrence’ for as long as nuclear weapons, or even the knowledge of how to make them, exists. There will always be some chance that even very low-scale conflict will escalate catastrophically. A policy based on deterrence is criticized—for example, in the American Catholic bishops’ pastoral letter—on the grounds that the killing of innocents—those who may have had little or no voice in the event(s) that may trigger their destruction—can not be condoned, nor can disproportionate destruction. But these moral objections may be overstated. As noted above, deterrence will always be a matter of probabilities, and will necessarily depend for its credibility in many circumstances on a victim of attack’s having command and control arrangements such that he will be unable to prevent a response that, after the fact of provocation, will seem disproportionate. If this be true, the just war disproportionality argument may be vitiated, or at least weakened. The probability of response can, in principle, be set sufficiently low that the weighted damage (that is, the expected damage, given retaliation, multiplied by the probability of retaliation)* will be commensurate with the threat to be deterred (but, of course, if the adversary’s estimate of weighted damage is low—and there will always be enormous uncertainty associated with such estimates—he may not be deterred. See below). The other moral objection to deterrence, that flowing from the proscription against killing innocents, may also be called in to question if it becomes technically feasible to base deterrence on threatening centers of political and military control in ways such that the threat to innocents will be severely limited. With improvements in accuracy of weapons delivery, and with concomitant reduction in weapons yields (from an average of 2.3 MT to 0.35 MT for US strategic weapons in a period of 20 years), this is, indeed, becoming increasingly feasible. Moreover, increasingly it is believed—at least, apparently, in the Reagan administration—that targetting the political-military control centers and assets of the Soviet Union is not
*
Recognizing that different levels of damage, Di, might be expected with differing probabilities, Pi, the weighted damage will, of course, be Σi Pi Di.
DETERRENCE AND ARMS CONTROL 105
just preferable to threatening population and industry as a means of deterrence, but necessary. But morality aside, there is the pragmatic argument against deterrence: it may fail. However great deterrent capabilities, their use may be judged not credible and so unwanted conflict may occur. Thus, we have seen conflict in Vietnam, over the Falkland Islands and between Israel and its neighbors, notwithstanding the existence of American, British and presumed Israeli nuclear capabilities. Or, nuclear weapons use may be triggered by events below the intended threshold of deterrence. In trying to avoid failures of either kind, efforts may be made, as suggested above, to adjust the threshold, but this will be an inherently chancy proposition. A threshold high enough to be safe may not deter; one low enough to deter may carry with it too large a risk of inadvertent catastrophe. Moreover, ‘thresholds’ will almost inevitably be poorly defined. Thus, we see widely disparate opinions about the feasibility of using nuclear, or for that matter other, weapons in selected and limited ways, particularly in Europe, a problem that seems destined to become more complex as it becomes increasingly possible to neutralize nuclear delivery systems with conventional ordnance and as the ‘firebreak’ between conventional and nuclear conflict erodes accordingly. With this background, I identify six positions on the role of nuclear weapons for American security policy.
(1) Rejection of all nuclear options. In this view, the actual use of nuclear weapons and deterrence based on them is held to be morally objectionable, ineffective, too risky, or some combination of the three. For those so believing, the only solutions to the problem of a nuclear threat would seem to lie in a perfect defense, as in Star Wars, or in the prevention of all conflict that might escalate to the point where others might use existing nuclear weapons or ones that might be built. (I subscribe to this view, rejecting that possibility of a perfect defense on technical grounds.) (2) Minimum deterrence. Those of this school hold that having weapons adequate to inflict ‘unacceptable damage’ on an adversary and command and control arrangements that would make use of such weapons credible in the event of a deliberate devastating attack is morally acceptable and likely to be effective against a rational adversary. Some would argue that a capability to deliver a single weapon might be sufficient for this purpose; few would suggest that the number need exceed a few hundred. With such capabilities, some degree of ‘existential deterrence’ will always exist with respect to lesser threats, but this is not likely to be very effective unless the mechanisms that would trigger response are such as to imply a veritable ‘doomsday machine’ (for which case, see the discussion below). Many of those favoring a minimum deterrence policy concede that an increase in conventional strength may be a correlative requirement; fewer seem as willing to accept as equaly logical the diffusion of minimum deterrent capacities to other nations.
106 DILEMMAS OF NUCLEAR STRATEGY
(3) A, B, C, D Extended deterrence and nuclear war-fighting. In consideration of extending the utility of nuclear weapons to cope with lesser threats, be they limited attacks against the United States or its forces or against third parties, there are four conceptual possibilities of interest, depending on how one comes down on the questions of first use of nuclear weapons and of war-fighting vs. coupling. The table below describes the taxonomy.
A and B describe the present situation. I distinguish between the two cases, A and B, because there are differences in thinking about the use of nuclear weapons, particualrly with respect to the NATO region. Most Europeans, most political figures, European and American, and most American moderates and liberals seem to subscribe to the belief that it is unlikely that a nuclear war can be limited. Believing this, they accept the view that the main purpose of having nuclear weapons in Europe and allocated for targets there is in coupling, that is, to enhance the likelihood that major conflict between the east and west will escalate to the point of nuclear exchange involving the ‘strategic’ forces of both sides. The hope is then, to deter conflict of any kind through punitive deterrence. Military planners and many American conser-vatives, in contrast, seem to believe that it might well be possible to use nuclear weapons in limited and selective ways to achieve military purposes, that is, to deny an adversary the realization of his military objectives. One’s policy preference, and to an extent one’s belief about the feasibility of nuclear war-fighting, ought logically to influence greatly decisions about weapons deployment, command and control, and arms control. If one believes that nuclear war-fighting is feasible and that posing the threat of limited nuclear conflict is preferable to posing a threat of unconstrained escalation, one will want to work hard to ensure that one can maintain effective control over one’s own nuclear capabilities during conflict. It will be just as important to ensure that any failures in adversary command and control will not lead to escalation—a most demanding requirement, considering that the adversary may wish the opposite, that is, to arrange his command and control system so that failures will make escalation likely. On the other hand, if one believes that the preferred—maybe, the only feasible—role for nuclear weapons is in coupling, one will want command and control arrangements that are such that there will be a perception on the part of the adversary that it will be unlikely that control can be maintained once some threshold of conflict is crossed. If it is felt that it is desirable to increase or decrease either the threshold or the probability of response, it ought not to matter much whether this is achieved through changes in one’s own force dispositions, command and control arrangements, etc. or in those of the adversary, or through a combination of both. Thus, if one wishes to see escalation to the use of nuclear weapons made more likely by having ‘use-them-or-lose-them’ type delivery systems exposed to conventional attack, the objective may be furthered
DETERRENCE AND ARMS CONTROL 107
either by more seriously threatening such adversary systems or by making one’s own more vulnerable, the latter suggestion one that would surely be seen as anathema to nuclear war-fighters. Unless accompanied by an increase in NATO conventional strength, a move to a no-first-use policy, cases C and D, would presumably result in some increase in the likelihood of conventional attack by Warsaw Pact forces against those of NATO (but see below for a mitigating argument), and this is the principal objection of opponents to a change in policy. The effect of such a change on the probability of nuclear war occurring is, however, not as clear inasmuch as that probability will be a compound one: the probability of conventional war, multiplied by the probability of nuclear use, given conventional conflict. With the first probability increasing with a change in policy and the contingent probability decreasing, it is not at all clear what will be the net effect.
Third Party Actions as a Casus Billi Virtually all of the discussion above, and, for that matter, US policy with respect to nuclear weapons, including arms control policy, has been predicated on the assumption that the object of deterrence is to prevent military attack against the US and/or its allies, or coercion, generally by the Soviet Union or its surrogates. Very similar arguments would apply in looking at the inverse possibilities. But increasingly, there is a belief abroad in the land that if the Soviet Union and the Union States are to become involved in direct conflict, it is more likely that the critical instigating event will be an action by a third party—very likely, by a party over which neither superpower may have much control, for example, by a client state in the Middle East, or the PLO, or Solidarity, or other dissidents in Eastern Europe— than a calculated decision by one or the other of the superpowers that military attack may be carried out with advantage. Thus, it may be that the greatest opportunity— even, the last clear chance—for avoiding catastrophe may lie in affecting decisions by a third party; and for this, the nuclear postures and policies of the great powers may be all but irrelevant. This is not to suggest that, once conflict has been initiated, decision by the superpowers (or other nuclear weapons states) to intervene or to increase the scale of conflict will be taken without regard to the global military balance. Certainly, such decisions by the United States and the Soviet Union would be influenced by perceptions by each of what the other could and might do; hence, deterrence may be ‘operative’; but it does mean that deterrence can not be regarded as the key to avoiding nuclear war. Indeed, it may well be only a minor factor.
Arms Control All of this suggests that the arms control efforts of the last 25 years ought to be doubly questioned. There is the issue of whether they have contributed to deterrence, or might do so in the future; and, even if the answer is affirmative, whether the likelihood of nuclear war has been, or can be, thereby reduced significantly. The record is dismal and the lessons unencouraging.
108 DILEMMAS OF NUCLEAR STRATEGY
The Limited Test Ban Treaty (LTBT) of 1963 and the ABM Treaty of 1972 are generally believed to be the most important arms control agreements of the period and the SALT I Interim Offensive Agreement of 1972 and the unratified SALT II Treaty next in significance. The LTBT has had only a slight effect on weapons effects testing and virtually none at all on weapons development. It has been effective mainly as an environmental protection measure. The ABM Treaty has probably served to allay, to some extent, concerns about the possibility of ABM deployment, and so may have reduced slightly the impetus for both superpowers, and also for France, the UK, and China, to invest in improvements in offensive systems. It is doubtful, however, that it has actually had much, if any, effect on ABM deployment. Its conclusion was a reflection of the fact that neither superpower saw much hope of effective defense anyway, and it is likely that either would be prepared to scrap the Treaty should it develop ABM technology to the point where effective deployment seemed possible. Spokesmen for the Reagan administration suggest that it may well be prepared to do so even earlier if the Treaty stands in the way of carrying out some of its SDI R & D program. However, there would be a price in abrogation, and so the Treaty may have some effect in slowing R & D on ballistic missile defense. Indeed, it may already have done so in a minor way. Presumably, there are some scientists and engineers who might otherwise be interested in working on BMD problems who have been deterred from doing so because they have believed that with the Treaty, deployment and some kinds of R & D will be unlikely for some years. The SALT offensive arms agreements were defended before the Congress by administration spokesmen at the time of their conclusion on the grounds that they would not inhibit any programs in which the US was really interested, and this appears to have been almost literally true. It was not until this fall—just months before the expiration date of the SALT II Treaty—that the US was forced, in order to stay within the SALT limits, to give up something which it probably would not have otherwise given up: a Poseidon submarine to make way for a new Trident submarine. While American proponents of SALT have claimed that the agreements have imposed important constraints on the Soviet Union, it is impossible to know whether the Soviets have scrapped some of the weapons they have in order to stay within SALT limits, or whether they have done so because of obsolescence. American critics of SALT have attacked the agreements on the grounds that they have not prevented the Soviet Union from doing much, if anything, that it wanted. It can certainly be said that the agreements have not yet contributed much to enhance deterrence or strategic stability, although to do so has been an important consideration in US approach to strategic arms limitations going back to the end of the Eisenhower administration. Throughout SALT, and in START as well, there has in fact, been a tendency to treaty improving strategic stability as virtually synonomous with mitigating the ICBM vulnerability problem, and to try to use arms control efforts to affect strategic force postures, especially that of the Soviet Union, so as to deal with the problem. Thus, there has been continuing interest in trying to limit as much as possible the numbers of ‘heavy’ Soviet ICBMs because of the fear that they might carry very large numbers of MIRVs; and to get the Soviets ‘out ot sea’, that is, to get them to emphasize SLBM forces at the expense of ICBMs, in the belief that the former would pose less of a threat to American ICBMs. Several particulars are noteworthy:
DETERRENCE AND ARMS CONTROL 109
(1) the ‘freedom to mix’ conditions in SALT I, the Vladivostok agreement, the Carter ‘comprehensive’ proposal of 1977, and SALT II, were designed to permit the replacement of ICBMs by SLBMs but not vice versa; (2) test limits were proposed on ICBMs but not on SLBMs in the 1977 proposals; (3) the ‘counting rules’ for SALT II were developed to limit the numbers of MIRVs per launcher; and (4) in ‘build-down’ proposals of the last couple of years single warheads have generally been treated preferentially as compared with MIRVs, and MIRVs on SLBMs preferentially as compared with those on ICBMs. But it must be noted that neither the United States nor the Soviet Union was very eager to seize what was surely the greatest opportunity of the last two decades to deal with the ICBM vulnerability problem; that is, to agree in SALT I to limit the development and deployment of MIRVs. That failure is perhaps the most dramatic evidence that neither power seems prepared to forego weapons systems that are technically attractive, whatever their impact on stability (and in the MIRV case, it was easily anticipated and widely discussed). The reasons for all of this emphasis on strategic arms in general and on the ICBM vulnerability problem in particular are fairly obvious. ‘Strategic’ weapons have commanded highest priority in American arms control efforts because
(1) they are the ones that have the potential for inflicting the greatest damage; (2) they are most directly threatening to Americans; and (3) they are the ones most easily seen and counted; hence, compliance with agreements relating to them is most easily verifiable. The ICBM vulnerability problem gets particular empahsis because
(1) the counterforce disarming attack seems to pose the most demanding requirements for strategic forces and because of historical memories, a legacy of Pearl Harbor; and (2) counterforce exchanges are peculiarly susceptible to simple (some would say, simplistic) quantitative modelling. But if a ‘first strike’ by either superpower is exceedingly improbable, and that is very widely accepted; if it seems unlikely, as it does, that changes in strategic posture of a scale that can be envisaged as a result of arms control efforts are otherwise unlikely to influence the behavior of American and Soviet leaders in crisis, and perhaps more importantly, other leaders who might be instrumental in causing or preventing crises; and if the opportunity for greatest leverage in averting a nuclear holocaust lies in affecting other elements of force posture—nuclear and conventional weapons in Europe—one has to wonder if all the emphasis in arms control on strategic weapons
110 DILEMMAS OF NUCLEAR STRATEGY
and on the window-of-vulnerability problem has not been misplaced. In fact, the lessons of the last 25 years—indeed, of the last 40 if one counts the AchesonLellenthal-Baruch efforts—suggest that arms control—at least, arms control based on negotiated agreements—may have little to offer in enhancing deterrence. As noted, the record of accomplishment is not impressive; and concerns about verification have resulted in an impetus to destabilizing deployments. Thus, there has been an emphasis, notably in SALT I, on limiting number of launchers rather than number of missiles or warheads, thereby providing incentives to develop and deploy very large missiles with many warheads; in effect, putting as many eggs as possible in one basket. And, notwithstanding growing sensitivity to that problem, and the development of fractionation limits to deal with it, the emphasis on limiting launchers has carried over to some extent, even into START, a point noted with muted disapproval in the Scowcroft Report. Moreover, again because of concern about verification, there have been proscriptions on the deployment of mobile ICBMs, notably in SALT II and in the latest Administration proposals, notwithstanding that mobility is an obvious, and quite possibly the preferred, way of getting around the ICBM vulnerability problem. Worse yet, there are troublesome conclusions to be drawn about the process of negotiation. The most obvious is that relating to bargaining chips. Time and again support for weapons programs has been solicited and obtained by arguing that they are needed for bargaining purposes; never mind whether they would contribute to stability or would have otherwise made sense on technical or military grounds. We have seen this notably in the case of the ABM debate of 1969–70, when the system proposed was obviously neither needed nor well matched to the task for which it was to be deployed; in the case of the Pershing and GLMCs deployments in Europe, when deployment would otherwise almost certainly have been prevented because of domestic oppostiion; and in the case of all of the strategic weapons programs advocated by the Reagan administration, most particularly in the case of MX, the deployment of which will be doubly unfortunate from the perspective of strategic stability: it will be an impressive counterforce weapon and an attractive target for adversary counterforce weapons. While, on occasion systems that have been rationalized as bargaining chips may be bargained away—the only actual example to date is the ABM Safeguard system—more often, they are likely, with time, to acquire constituencies for their support, and will be developed and deployed. Thus, we will end up with more arms than we would have, but for the negotiations. There is another less noted but also pernicious effect of the process. In order to secure support for arms control measures, it has been necessary—or at least Presidents Kennedy and Nixon judged it so—to accede to demands for support of systems or programs that might otherwise not have commanded Executive support or perhaps that of a majority of the Congress. Thus, we had the ‘safeguards’ associated with the Limited Test Ban Treaty and the ‘assurances’ associated with SALT I. And, who knows what has happened on the Soviet side, or what might happen on either side, to get support for weapons not covered by agreements but which some minority favors? There is a third effect of the negotiating process, in my view, more troublesome than either of the others. Debate about strategic arms and reportage about them has created an exaggerated impression, totally without justification, in my view, of the seriousness of the ICBm vulnerability problem and of the importance of modest
DETERRENCE AND ARMS CONTROL 111
differences in numbers and performance parameters of the weapons in question; indeed, of the general utility of nuclear weapons and of them as a source of instability. All of this is illustrated forcefully and in detail in such accounts as those of SALT I by John Newhouse (Cold Dawn), Gerard Smith (Doubletalk) and Thomas Wolfe (The Salt Experience) and Strobe Talbott’s books on SALT II (Endgame) and on the INF and START negotiations (Deadly Gambits). Thus, in a large sense the negotiating process has been, and may well continue to be unfortunately—even, dangerously—diversionary: drawing attention to scenarios for war initiation that are inherently most implausible and suggesting that by tinkering with weapons we can really make a difference in reducing the likelihood of nuclear war when, in fact, the greatest leverage probably lies not in ‘technical fires’ to strengthen deterrence but rather in getting at the sources of conflict: in the settlement of political differences, in the establishment of codes of behavior, and in dealing with regional problems. In brief, I suggest the following propositions:
(1) that nuclear deterrence as a cornerstone of security policy has to be questioned on both moral and pragmatic grounds; (2) that arms control negotiations, at least of the kind we have known in the last decades, must be questioned not only because they have been in large measure based on an acceptance of deterrence but also because (a) they have not, and are not likely, to contribute much, even to deterrence, (b) they have served as much to legitimize as to constrain the arms race, (c) they, like the SDI, have served to divert attnetion from more realistic and hopeful ways of dealing with conflict in a world of nuclear weapons; (3) that to the extent we do focus on arms, (a) our priorities ought probably to be changed, with more attention given to conventional arms, including notably those that can threaten nuclear capabilities, and less to the inetcontinental nuclear systems, (b) we should probably play down negotiations and focus much more attention on making sensible unilateral decisions without reference to negotiations possibilities.
112
Arms Races and Instability Michael D.Intriligator and Dagobert L.Brito
I. Introduction and Principal Themes The concepts of ‘stability’ and ‘instability’ are often used in discussions of arms races and the outbreak of war. Frequently, it is assumed that arms race instability is tantamount to instability against the outbreak of war. But there is a fundamental difference between these two concepts of instability. ‘Arms race instability’ refers to a situation in which there is a quantitative or qualitative arms race between two or more nations for which the levels of weapons or their capabilities spiral to higher and higher levels without reaching an equilibrium. By contrast, ‘instability against the outbreak of war’ refers to a high or growing likelihood of war outbreak. These are very different concepts, and a major purpose of this article is to show that, using a formal model of an arms race, it is possible to have arms race instability but, at the same time, stability against the outbreak of war. In fact, the instability of the arms race implies, by reinforcing mutual deterrence, a reduced likelihood of war outbreak. In that sense the arms race becomes itself a contribution to arms control, which is interpreted broadly as any initiative that reduces the likelihood of war. In the formal model of an arms race each side attempts to deter the other and, at the same time, each believes that the other is attempting to acquire arms superiority or planning a pre-emptive attack. An implication is that there is arms race instability as each side acquires more and more weapons. At the same time, however, there is stability against the outbreak of war as each side has more than enough weapons to deter the other. Each side, none the less, seeks to acquire additional weapons, reinforcing the suspicions of the other and prompting even further weapons acquisitions. The resulting large and growing weapons stocks, however, reinforce deterrence and make pre-emptive strikes impossible, thereby contributing to stability against the outbreak of war. Aspects of this formal model apply to the superpower nuclear arms race between the United States and the Soviet Union. This arms race, which has both quantitative and qualitative dimensons, continues to move numbers of weapons or their effectiveness to higher and higher levels. At the same time, however, these arms acquisitions have resulted in a situation of mutual deterrence, in which there is great stability against the outbreak of war due to the fear of potential retaliation. While mutual deterrence could exist at significantly lower levels of weapons on both
114 DILEMMAS OF NUCLEAR STRATEGY
sides, the additional weapons add to fears of a potential war, thereby reinforcing deterrence. They also provide insurance against technological breakthroughs that could reduce the effectiveness of weapons. Additions to weapons stocks, stemming from the instability of the superpower arms race, thus provide both additional deterrence and insurance against destabilizing technological breakthroughs, thereby reducing the likelihood of war. In that sense, arms races involving nuclear weapons ironically provide a way to avoid nuclear war. Nuclear weapons are part of the solution, as well as part of the problem, of avoiding nuclear war. A policy conclusion of this analysis is that the most dangerous aspect of the current arms situation is neither the presence of nor the increasing levels of weapons, which, in fact, create the stability of mutual deterrence, but rather the dangers of accidental initiation, miscalculation, or irrational action, for example, via technical mishap, delegation to undependable subordinates, or via third-party actions. Thus there is scope for arms control initiatives, and some of the most important items on the arms control agenda should be initiatives to prevent such problems. These initiatives, which can be unilateral, bilateral, or multilateral, can be part of a new agenda for arms control which emphasizes not the number of weapons but rather how weapons might be controlled to minimize the chance of their inadvertent use.
II. Military Strategy In order to understand the implications of arms races for stability against the outbreak of war; it is necessary to understand how arms would be used in the event of war; that is, it is necessary to understand military strategy. Both the sources of the arms race and its implications for stability or instability are based on how weapons would be used. Thus the first step in the analysis must be a study of military strategy. In our analytic work we have developed a formal model of a missile war in order to understand strategic choices. This model can be considered a representation not of an actual war but rather of a war simulation conducted explicitly or implicitly by defense planners, for example, in the Pentagon and in the Kremlin. The model describes the evolution over time of a war between two countries A and B as a result of the initial number of weapons on both sides, the strategic decisions made by both countries, and the effectiveness of weapons against both enemy weapons (counterforce targets) and enemy cities (countervalue targets).1 The strategic decisions made by both countries over time in the course of the war are summarized by two key choices, which, taken together, represent grand strategies. First, there is a choice of targets, between counterforce targets of enemy weapons and countervalue targets of enemy cities and industrial capacity. (While colocation, fall-out, etc. make pure counterforce or pure countervalue targeting impossible there is a choice between primarily counterforce and primarily countervalue targets). Second, there is a choice of rate of fire, between, at the extreme values, a maximum rate, that is, firing all weapons as rapidly as possible, and a zero rate, that is, holding weapons in reserve for later use. The results of our analysis of the choice of grand strategies are that there are four possible grand strategies:
ARMS RACES AND INSTABILITY 115
First strike: Counterforce targets/maximum rate of fire. Massive retaliation: Countervalue targets/maximum rate of fire. Limited strategic war: Counterforce target/zero rate of fire. War of nerves: Countervalue targets/zero rate of fire. The choice of grand strategy can change over time during the course of the war, and the optimal grand strategy, that which maximizes a pay-off function dependent on terminal weapons and casualties, leads to a choice of three successive grand strategies. One of the countries, say A, starts the war using a first-strike strategy, targeting enemy weapons and firing at the maximum rate to reduce the capabilities of the opponent. Country A then switches the targets to enemy cities, entailing a massive retaliation strategy, or, alternatively, depending on the evolution of the war, switches the rate to the zero rate, entailing a limited strategic war strategy. Finally, country A switches the rate in the first case or the targets in the second case, leading in either case to the last grand strategy, a war of nerves. In this final stage of the war, country A holds its weapons in reserve to threaten enemy cities in order to obtain the best conditions for war termination. With this optimal switching grand strategy, the only situation in which country B suffers significant casualties is the massive retaliation stage, in which A retaliates against B cities. The war could, however, proceed via limited strategic war in which country B never inflicts significant casualties on A since by the time B switches to countervalue targets B has already switched to the zero rate. In this case, enemy cities are held hostage rather than actually hit, since weapons targeted at enemy cities are held in reserve.
III. Deterring and Initiating Regions The optimal military strategies during a potential war, as developed in section II, can be used to obtain deterring and initiating regions in the weapons plane. We stress that these regions are based on a potential war as it might be simulated or analyzed by defense planners.2 Country A deters country B if, given a first strike by B (with B choosing counterforce targets and the maximum rate of fire), A can absorb this strike and still have enough weapons left to inflict an unacceptable level of casualties on B in a retaliatory strike (with A using a massive retaliation strategy, choosing countervalue targets at the maximum rate). Solving the formal model for the number of weapons A needs to have this capability to deter B implies a linear dependence of the (minimum) number of A weapons needed to deter B on the number of B weapons. This linear dependence is illustrated in Figure 1 by the line marked ‘A deters’ in the weapons plane, consisting of alternative combinations of weapons held by both A and B, shown as (MA, MB), where MA is the number of A weapons (‘missiles’) and MB is the number of B weapons (‘missiles’). Note that a certain number of A weapons is required for deterrence even if B has no weapons (that is, even if B has no weapons to strike A, it is still necessary for A to have a certain number of weapons to inflict unacceptable casualties on B). Also note that as the number of B weapons increases the minimum number of A weapons needed to deter B also increases (that is, as B has more
116 DILEMMAS OF NUCLEAR STRATEGY
Figure 1: Deterring and Initiating Regions in the Weapons Plane
weapons to use on a first strike A needs that many more weapons to have enough capabilities to inflict an unacceptable level of of casualties on B after absorbing B’s first strike). Figure 1 also shows the line marked ‘B deters’, which is the minimum number of B weapons required for B to deter A and which increases with the number of A weapons. Note that the line marked ‘A deters’ is interpreted as the minimum number of A weapons needed to deter B as a function of the number of B weapons, while the line marked ‘B deters’ is interpreted as the minimum number of B weapons needed to deter A as a function of the number of A weapons. In addition to deterring the opponent, either or both countries could consider attacking the opponent. Country A attacks country B if it can launch a first strike on B (choosing counterforce targets and the maximum rate of fire) and reduce the number of B weapons sufficiently that B does not have enough weapons left to inflict unacceptable casualties in A in a massive retaliation strike (with B choosing
ARMS RACES AND INSTABILITY 117
countervalue targets and the maximum rate). Again solving the formal model for the number of weapons A needs to have this capability to attack B implies a linear dependence of this (minimum) number of A weapons needs to attack B on the number of B weapons. This linear dependence is illustrated in Figure 1 by the line marked ‘A attacks’. As the number of B weapons increases A needs more and more weapons to attack B. Note that the line ‘A attacks’ is parallel to the line ‘B deters’ since both cases involve the same grand strategies: first strike for A, followed by massive retaliation for B. The intercepts are different, however, since the one for ‘A attacks’ involves the maximum acceptable number of casualties in A (as perceived by A) while the one for ‘B deters’ involves the minimum unacceptable number of casualties in A (as perceived by B). Furthermore the line ‘B deters’ is solved for the minimum number of B weapons, while the line ‘A attacks’ is solved for the minimum number of A weapons. Finally, symmetrically, the line marked ‘B attacks’ shows the minimum number of B weapons needed to attack A, and it is parallel to the ‘A deters’ line. The four lines in the weapons plane in Figure 1 for ‘A deters’, ‘B deters’, ‘A attacks’, and ‘B attacks’ define deterring and initiating regions in this plane. The upper shaded cone bounded by the ‘A deters’ and ‘B deters’ lines is one of mutual deterrence, in which each country deters the other. This region is one of stability against war outbreak since in it each country has the ability to inflict unacceptable damage on the other in a retaliatory attack. This region, the cone of mutual deterrence, is one of mutual assured destruction in which each country has enough missiles to deter the other and neither country has enough missiles to attack the other. The lower shaded sawtooth-shaped area is the region of initiation. In it neither country deters the other, and one or both can attack the other. In the middle portion of this area, the cone between the origin and the point I, A can attack B and B can attack A while neither can deter the other. Thus this portion is one of virtually forced pre-emption in which there is a great advantage to initiate, rather than retaliate. The ‘reciprocal fear of surprise attack’ based on the tremendous advantage in striking first forces both sides to initiate, each trying to pre-empt the attack of the other. The other portions of this area are ones of asymmetry; in one (on the lower right) A can attack B, but B cannot attack A, while in the other (upper left) B can attack A, but A cannot attack B. Thus, in these two asymmetric regions stability relies on the intentions of one side. The entire sawtooth-shaped area, however, is one of initiation. This region exhibits instability against war outbreak in that neither side has an adequate retaliatory capability to ensure deterrence, and one or both countries has enough capabilities, relative to the missiles held by the opponent, to attack the other. Figure 1 illustrates the irony of deterrence. It shows that weapons are needed to ensure peace, and, conversely, that low levels of weapons entail the dangers of an inadequate deterrence and an ability to attack, a highly unstable situation. The disarmed state at the origin is also a highly dangerous situation. Even though neither side has weapons to threaten the other, at this point either or both could acquire a small number of weapons and then be forced to pre-empt. Furthermore there may be incentives for each to acquire weapons from the disarmed state in order to attain a dominant position. Reducing weapons down to the disarmed state may be safe, for example, if the process of disarmament generates goodwill. At the disarmed state, however, if there is a desire for dominance or some misunderstanding or
118 DILEMMAS OF NUCLEAR STRATEGY
misperception either or both sides would start to acquire weapons, leading to movement through points in the region of initiation in a crisis atmosphere, a sure recipe for disaster. The deterring and attacking lines for both countries that define the upper cone of mutual deterrence and the lower sawtooth-shaped area of initiation in Figure 1 are themselves defined in terms of:
(i) the effectiveness of missiles in destroying enemy missiles (which depends on missile accuracy and yield, enemy active missile defense, and enemy missile dispersion and hardness against attack); (ii) the effectiveness of missiles in inflicting enemy casualties (which depends on missile accuracy and yield and enemy active and passive civil defense and dispersion); (iii) the (maximum) rates of fire of missiles (which depend on missile characteristics and the command and control structure); (iv) the time intervals during which one country can attack the other without a response (which depend on detection capabilities and command and control); (v) the time intervals for a retaliatory strike (which depend on missile characteristics and command and control); (vi) the minimum unacceptable level of casualties which would deter each country, as estimated by the other country; and (vii) the maximum acceptable level of casualties which each country could accept in a retaliatory strike, as estimated by itself. To give a numerical example, if it takes two missiles to destroy an enemy missile, each missile can inflict 250,000 casualties, the maximum rate of fire is ten per cent per minute, the first-strike interval is 15 minutes, the retaliatory strike interval is ten minutes, the minimum unacceptable level of casualties is 40 million, and the maximum acceptable level of casualties is five million, then point D in Figure 1, the minimum number of missiles required for mutual deterrence, is 414 missiles on each side, while point I in this figure, the maximum number of missiles in the area of forced preemption (where each can attack the other), is 52 missiles. Thus in this symmetrical example if each side has more than 414 missiles there is the stability of mutual deterrence while if each side has fewer than 52 missiles there is the instability of forced pre-emption. Obviously all the numbers used in these calculations are subject to uncertainty, particularly the estimates of the minimum unacceptable/maximum acceptable level of casualties. Thus it would be more appropriate to refer not to a cone of mutual deterrence and a region of initiation but rather to equal probability contours of war outbreak, showing relatively high probability of war outbreak in the region of initiation and relatively low probabilities of war outbreak in the cone of mutual deterrence. Alternatively the lines of deterrence and attack in Figure 1 can be replaced by bands, reflecting the uncertainty in the values of the parameters defining deterring and attacking regions. For purposes of exposition, however, the lines, the
ARMS RACES AND INSTABILITY 119
cone of mutual deterrence, and the region of initiation will be used in the following sections to analyze arms races and war initiation, arms race stability, and arms control.
IV. Arms Races and War Initiation An arms race refers to the interactive acquisition of weapons by two or more nations. Assuming only a single weapon and only two nations in a bipolar world, for example, the superpowers of the United States and the Soviet Union, an arms race can be represented by a trajectory in the weapons plane, such as that of Figure 1. Figure 2 illustrates several such trajectories, each of which represents a particular type of arms race, where the regions of deterrence and initiation of Figure 1 are also shown. The effects of the arms race on the initiation of war depend on the nature of the arms race in relation to the regions of deterrence and initiation, as illustrated by the five trajectories shown in the figure. Trajectory 1 starts from the totally disarmed state of the origin (or, more generally, at low levels of weapons on both sides), and, as a result of an arms race, both countries move to a highly unstable situation of forced initiation. This is an example of the case in which an arms race leads to war, which is the conventional wisdom. Trajectory 2 involves an arms race that, as a result of a weapons build-up in both countries, moves the situation from one in which A could attack B to one in which each side deters the other. This is a situation in which an unstable situation is stabilized as a result of an arms race, that is, ‘if you seek peace, prepare for war’. Trajectory 3 is one in which country A reduces its arms from an extremely high level while country B simultaneously increases its arms from a relatively low level. The result is a reduced chance of war outbreak, moving from a situation in which A can attack B to one in which each deters the other. Trajectory 4 is one in which both sides disarm with the result that they move from a stable situation in the cone of mutual deterrence to an unstable one in which each can attack the other. This trajectory illustrates the potential dangers of bilateral disarmament that is carried too far. Trajectory 5 is one in which, like the previous case, both sides reduce their level of armaments, but, unlike the previous case, they stay within the cone of mutual deterrence. Such selective reductions in arms need not cause instability as long as the situation remains in the cone of mutual deterrence. Thus, the relation of an arms race to the outbreak of war is not a simple one. The conventional wisdom that arms races lead to war is neither right, nor wrong; it depends on the circumstances. In certain situations an arms race can lead to war (as in Trajectory 1) while in other situations an arms race can lead to peace (as in Trajectory 2). Conversely a dssarming race can result in war (as in Trajectory 4) or can preserve peace (as in Trajectory 5). In fact, the contrast between Trajectories 4 and 5 illustrates the differences between disarmament, which, despite its lofty goals, can, as already noted, be highly dangerous in leading to unstable situations, and arms control which, by selectively reducing weapons but retaining mutual deterrence can promote greater stability, for example, in eliminating obsolete or unstable weapons systems but retaining enough weapons for deterrence.
120 DILEMMAS OF NUCLEAR STRATEGY
Figure 2: Weapons Trajectories and Regions of Deterrence and War Initiation
To relate this analysis to the post-war situation for the United States and the Soviet Union, perhaps the trajectory that comes closest to describing the history of superpower arms since 1945 is Trajectory 2. Before the Soviet Union had a significant weapons capability the situation was an asymmetric one in which the US, country A in the figure, could attack the Soviet Union, but the Soviet Union could not attack the United States. In this asymmetrical situation, the peace was preserved because of the commitment by the United States not to initiate a war, although it should be noted that the Soviet Union had considerably more conventional forces than the United States in this period. The 1960s and 1970s witnessed a major buildup of weapons on both sides, leading to the present situation of mutual deterrence, as illustrated in Trajectory 2, a situation of stability against war outbreak. In fact the greatest instability was probably in the late 1950s and early 1960s, particularly the 1962 Cuban missile crisis.
ARMS RACES AND INSTABILITY 121
The stability of mutual deterrence is currently fundamental to strategy and arms control. It has worked, and it is reasonable to predict that it will continue to work into the foreseeable future. Conversely, no viable alternative to deterrence has been yet identified. Ironically deterrence requires weapons of mass destruction, such as nuclear weapons, to deter the initiation of major war. Returning to the current superpower situation, the result of the arms race of the 1960s and 1970s led, as noted, to the present situation of mutual deterrence, as illustrated by Trajectory 2 in Figure 2 ending in the cone of mutual deterrence. But the superpowers are extremely ‘high up’ in this cone, with considerably more weapons than needed at the minimal deterrent point D. If the numerical example of the last section is at all close to the current situation, all that would be needed for mutual deterrence is 400–500 missiles (calculated as 414 in the example, but allowing for some uncertainty in the numerical estimates). But both superpowers have many times this number of strategic warheads. The historical reasons for having more than the minimal deterrent have involved misperceptions, bureaucratic inertia, overbuilding due to use of worst-case analysis, and other factors. Leaving aside the historical reasons for having reached a point high up in the cone of mutual deterrence, what are the implications for stability of being at such a point? Conventional wisdom suggests that the presence of these large stockpiles means that both sides are preparing for war and that they make war inevitable. But deterrence implies just the opposite: the fact that the superpowers both have substantial levels of weapons places them well in the cone of mutual deterrence, each side being able to deter the other and thus reducing the chance of war. Conversely, if either side significantly reduces its level of weapons then it would not be able to deter the other and further reductions would eventually allow the other side to attack. For example, in Figure 2, reductions by A from a point in the cone of mutual deterrence would imply a trajectory involving horizontal leftward movement, which would eventually cross the ‘A deters’ line, at which point A no longer deters B and, if continued, would ultimately cross the ‘B attacks’ line, at which point B could attack A. Similarly, bilateral reductions in weapons, as illustrated in Trajectory 4 in the figure, would generally lead to one country, then both, losing its ability to deter the other and, if continued, one country and then both would be able to attack the other. In fact, there are advantages in being high up in the cone of mutual deterrence. First, the factors that determine the cone are not known with certainty so the cone may be significantly smaller than shown, implying that the superpowers may not be that far above the point of minimal mutual deterrence (labelled D in Figure 2). Second, the nature of the arms race has changed from a quantitative to a qualitative race, in which weapons are replaced by newer types. For example, for the United States Minuteman land-based strategic missiles have been replaced by Minuteman III, which, in turn, may be augmented by the MX, while the Polaris submarine-based strategic missiles have been replaced by Poseidon, which are being replaced by Trident. As these new weapons are introduced they change the factors determining all four lines in Figures 1 and 2 and thus also change the cone of mutual deterrence.3 The newer generations of missiles have greater accuracy and yield, thus increasing the effectiveness of missiles in destroying enemy missiles and thereby shrinking the cone of mutual deterrence by moving point D upward in Figure 2. Building upon the numerical example of the last section, if country A improves the accuracy and yield of its missiles so that the number of B missiles destroyed by one counterforce A
122 DILEMMAS OF NUCLEAR STRATEGY
missile increases from 0.5 (in which it takes two missiles to destroy an enemy missile) to 0.8, the point D shifts up from 414 missiles on both sides to 463 A missiles and 541 B missiles.4 Thus, given the presence of a qualitative arms race and the possibility of significant upward shifts in the point of minimum mutual deterrence, there is considerable value in being well above this point, as is the case currently. If weapons levels in either or both countries were reduced to near this minimum level there is a possibility that new weapons could alter the situation in such a way that one or both countries would not be able to deter the other. Thus, the earlier quantitative arms race, by resulting in a point far up in the cone of mutual deterrence, has created a type of insurance against potentially destabilizing weapons improvements, such as improvements in accuracy and yield. Whether it was planned that way or not, the current situation is one of great stability not only at present, but also for most foreseeable changes. The superpowers have built a system with insurance against qualitative changes that is somewhat like an engineer designing a bridge so that it will stand despite very adverse conditions. The present system of mutual deterrence will continue despite significant potential qualitative changes in weapons on both sides.
V. Arms Race Stability or Instability An analysis of the arms race and its stability requires an understanding of its dynamics as determined by decision-making rules on both sides. Decisions on military procurement, however, are often based on certain targets or goals which may be heuristic.5 To the extent that defense decision-making institutions are large complex bureaucratic organizations, they tend to rely neither on passive mechanical responses to the other side nor on explicit optimization rules. Rather they tend to rely on rules of thumb or heuristic decision rules for arms acquisitions, based on history, attitudes of decision-makers, inter-service rivalry, and other aspects of the defense bureaucracy. For example, British naval policy in the period preceding the First World War, when the Royal Navy was the principal component for projecting force worldwide, called for a navy capable of defeating the combined fleets of the next two larger naval powers. As another example, recent US policy on conventional force capabilities calls for sufficient forces to be able to fight one-and-a-half wars at the same time, that is, a major conflict plus a separate local conflict. In both cases, the decision rule is a heuristic rule of thumb based on institutions, history, etc., rather than one which is either mechanistic or explicitly optimizing. Consider now a heuristic decision rule for the superpower nuclear arms race. Suppose each of the countries use a heuristic decision rule which calls for it to acquire enough weapons to deter the opponent. Then, in Figure 1, if MA (the missiles on country A) lies to the left of the ‘A deters’ line then the (horizontal) gap between MA and this line indicates the desired weapons acquisition for A. If both countries use such a decision rule then there will be a stable equilibrium reached at D, the point of minimal mutual deterrence. If, by contrast, each country uses a heuristic decision rule which calls for it to acquire enouch weapons to attack the opponent, then, in Figure 1, if MA lies to the left of the ‘A attacks’ line, then the (horizontal) gap between MA and this line becomes the desired weapons acquisition. In this case there
ARMS RACES AND INSTABILITY 123
is no equilibrium but rather an unstable arms race moving to higher and higher levels of weapons on both sides. The set of heuristic decision rules which probably best captures the nature of the superpower nuclear arms race is one in which each side attempts to deter the other, while, at the same time, each believes that the other is attempting to acquire enough weapons to attack it. Both the United States and the Soviet Union certainly profess this policy and belief, each professing to seek arms only to deter the other but each stating its belief that the other is, or could be, seeking enough weapons to attack it. The result in this case is an unstable arms race, moving to higher and higher levels of weapons on both sides in the cone of mutual deterrence and thus entailing stability against the outbreak of war. As each side seeks deterrence, given its assumption that the other side is an attacker, each side builds up its level of weapons, reinforcing the suspicions of the other. The result is an unstable arms race, with levels of weapons (or their capabilities) spiralling to higher and higher levels, as has been the case over the last 20 years, but with the system moving further and further away from the region of institution, thus reducing the chances of war outbreak due to reinforced mutual deterrence. The result is therefore arms race instability coupled with crisis stability, as each side has considerably more than enough weapons to deter the other but continues, nevertheless, to acquire yet even more weapons, adding to mutual deterrence.
VI. Policy Conclusions: Arms Control While the present system is one of stability against the initiation of war now and for the foreseeable future, given possible weapons improvements, there is nevertheless an important role for arms control. We use the terms ‘arms control’ to refer to any initiative that would reduce the chance of a major war, particularly nuclear war, or, secondarily, would limit the damage in case there were war or would reduce the costs of weapons. These initiatives may be unilateral, bilateral, or multilateral; they may be qualitative or quantitative; and they may be explicit or implicit. Bilateral negotiated explicit quantitative reductions in strategic arms, as in the SALT/START process, represent one type but not the only type of arms control. For example, unilateral steps taken to protect and to harden missile silos, explicit and implicit US-Soviet agreements stemming from the 1962 Cuban missile crisis, and the multilateral Nuclear Nonproliferation Treaty also represent arms control steps. It is a mistake to identify, as some people have, arms control only with SALT/ START agreements, which has made some people skeptical about arms control. It is also a mistake to identify arms control with disarmament or even with a step toward disarmament. Arms control and disarmament are, in a certain sense, opposing notions as illustrated in Figure 2, where complete disarmament is highly unstable and where an arms race from the disarmed point would lead to movement in the region of initiation. Indeed, the best arms control initiative from low levels of weapons on both sides (as at a point in the region of initiation) is a substantial increase in weapons to reach the cone of mutual deterrence, as in Trajectory 2 (which was what happened in the superpower arms race of the 1960s and 1970s). This increase in weapons represents an initiative that meets
124 DILEMMAS OF NUCLEAR STRATEGY
the primary purpose of arms control, namely, that of lowering the probability of nuclear war.6 A major conclusion of the earlier analysis is that while the superpower arms race may exhibit instability, it need not imply instability against the outbreak of war. Indeed, the arms race may reinforce mutual deterrence and thus enhance stability against the outbreak of war. Given this conclusion, where is the potential for arms control, particularly in achieving its primary goal of reducing the probability of war? We believe that there is a valid role for arms control in achieving this primary goal. The problem is not one of incentives to pre-empt, which have totally disappeared due to the overwhelming ability of each side to deter the other. Our analysis implies that there is substantial stability against premeditated attacks, such as a ‘bolt from the blue’, or even escalation from regional conflict or from crisis situations. Given the nature of mutual deterrence, an attack by either superpower on the other, whether in a normal or in a crisis situation, would be equivalent to an attack on itself and thus suicidal. But there still remain the problems of accidental initiation, miscalculation, or irrational action, for example, via technical mishap, delegation to undependable subordinates, or through the action of third parties. These are the major problems given high levels of weapons on both sides, as with such high levels there is the stability of mutual deterrence. These problems should be at the forefront of arms control initiatives, whether unilateral, bilateral, or multilateral, rather than the present emphasis on bilateral negotiated reductions, as in the START process, or unilateral defensive policies, as in the Strategic Defense Initiative ‘Star Wars’ approach.7
NOTES 1. For a presentation of the formal model of a missile war see Appendix A. See also M.D. Intriligator, ‘Strategy in a Missile War: Targets and Rates of Fire’, Security Studies Paper No. 10 (1967), University of California, Los Angeles, Security Studies Project; M.D. Intriligator, ‘The Debate Over Missile Strategy: Targets and Rates of Fire’, Orbis 11 (1968), 1138–59; M.D.Intriligator, ‘Strategic Considerations in the Richardson Model of Arms Races’, Journal of Political Economy 83 (1975), 339–53; and D.L.Brito and M.D.Intriligator, ‘Some Applications of the Maximum Principle to the Problem of an Armaments Race’, Modeling and Simulation 4 (1973), 140–4; D.L.Brito and M.D.Intriligator, ‘Uncertainty and the Stability of the Armaments Race’, Annals of Economic and Social Measurement, 3 (1974), 279–92; D.L.Brito and M.D.Intriligator, ‘Arms Races: Behavioral and Economic Dimensions’, in J.A.Gillespie and D.A.Zinnes (eds.), Missing Elements in Political Inquiry: Logic and Levels of Analysis (Beverley Hills: Sage Publications, 1982). For an overview of formal models of conflict, see M.D.Intriligator, ‘Research on Conflict Theory: Analytic Approaches and Areas of Application’, Journal of Conflict Resolution 26 (1982), 307–27. 2. For a derivation of deterring and initiating regions using the formal model see Appendix B. See also Intriligator, ‘Strategic Considerations…’, op. cit. and M.D.Intriligator and D.L. Brito, ‘Strategy, Arms Races, and Arms Control’, in J.V.Gillespie and D.A.Zinnes (eds.), Mathematical Systems in International Relations Research (New York: Praeger Publishers, 1977); Brito and Intriligator,
ARMS RACES AND INSTABILITY 125
3.
4.
5.
6.
7.
‘Arms Races…’, op. cit.; M.D.Intriligator and D.L.Brito, ‘Non-Armageddon Solutions to the Arms Race’, Arms Control 6 (1985), 41–57. The cone of mutual deterrence will continue to exist as long as a certain condition discussed in Appendix B is met. A sufficient (but not necessary) condition for this stability condition to be met is that, on average, it takes more than one missile to destroy an enemy missile. Other changes will also affect the point of minimum mutual deterrence D. For example, civilian defense via active (for example, ABM) or passive (for example, shelters) defenses also increases D. See M.D.Intriligator and D.L.Brito, ‘Heuristic Decision Rules, the Dynamics of the Arms Race, and War Initiation’, in U.Luterbacher and M.Wards (eds.), Dynamic Models of International Conflict (Boulder, CO: Lynne Reinner Publishers, 1985). Note that while the superpower arms race meets the primary goal of arms control of reducing the probability of nuclear war it fails to meet the secondary goals of reducing damage if there were a war and reducing the cost of weapons, illustrating the fact that few if any initiatives can satisfy all three goals. There are tradeoffs among these goals such that an initiative involving a ‘small’ reduction in the probability of war might be rejected if it involved a ‘large’ increase in the amount of damage if a war occurs. Nevertheless, the goal of avoiding a war is clearly a primary one. Agreements on limiting strategic arms, as in SALT and START, run two risks. One is that of reducing the weapons level below the point required for current and future mutual deterrence. There is a danger that later weapons developments in terms of accuracy and yield could make the negotiated limits incompatible with mutual deterrence. Thus any limitations are likely to be small, as has been the case historically. The second risk, analyzed formally in D.L.Brito and M.D.Intriligator, ‘Strategic Arms Limitation Treaties and Innovations in Weapons Technology’, Public Choice 37 (1981), 41–59, is that quantitative limits on weapons could lead to even further weapons innovation or to the development of new weapons as ‘bargaining chips’, either of which could be destabilizing. As to the SDI ‘Star Wars’ approach, while a complete strategic defense could undermine the stability of mutual deterrence, a limited strategic defense system could provide valuable protection against a small number of accidentally launched missiles, a decapitation strike, or threats of a small nuclear power. See M.D. Intriligator, ‘What “Star Wars” is Intended to Prevent’, New York Times (14 March 1985).
APPENDIX A A Formal Model of a Missile War Our model of a missile war analyzes the causes of changes over time in the number of missiles and casualties in two countries during a possible war. (See Intriligator, ‘Strategy in a Missile War’, 1967; The Debate Over Missile Strategy’, 1968; ‘Strategic Considerations in the Richardson Model of Arms Races’, 1975; Brito and Intriligator, ‘Some Applications…’, 1973; ‘Uncertainty and the Stability…’, 1974; Intriligator and Brito, ‘Strategy, Arms Races…’, 1977.) The model consists of a system of four differential equations with initial boundary conditions, and it determines the evolution over time of the war as a result of the initial numbers of
126 DILEMMAS OF NUCLEAR STRATEGY
weapons on both sides, the strategic decisions made by both countries during the war, and the effectiveness of weapons against both counterforce targets (enemy weapons) and countervalue targets (enemy cities). The variables of the model are MA(t) and MB(t), which represent the missiles in country A and B respectively at time t. The model consists of the following differential equations for changes in each of these four variables and accompanying boundary conditions:
(1) (2) (3) (4) The war starts at time t=0, at which point, as shown in the boundary conditions, missiles, country B has missiles, and there are no casualties on country A has either side. Country A launches its missiles at rate α(t), so −αMA in (1) represents the reduction in A missiles due to decisions by A to launch missiles. Similarly, −βMB in (2) represents the reduction in B missiles due to B decisions to launch its missiles at the rate β(t). (The dependence of MA, MB, α, β, and other variables on time is omitted in equations (1)–(4) for convenience.) Missiles can be targeted counterforce at enemy missiles or countervalue at enemy cities. If A uses the counterforce proportion α′(t) at time t, then, of the αMA missiles launched at this time, α′αMA are launched at B missiles while (1−α′)αMA are launched at B cities. If fA(t) is the counterforce effectiveness of A missiles at time t, that is, the number of B missiles destroyed per A counterforce missile, then α′αMAfA represents the B missiles destroyed by A counterforce missiles, as shown in equation (2). Similarly, β′βMBfB in equation (1) represents the A missiles destroyed by B counterforce missiles, where fB(t) is the counterforce effectiveness of B missiles. If vA (t) is the countervalue effectiveness of A missiles, that is, the number of B casualties inflicted per A countervalue missile, then (1−α′)αMAvA represents the B casualties inflicted by A countervalue missiles, as shown in equation (4). Similarly, (1–β′) βMBvB in equation (3) represents the A casualties inflicted by B countervalue missiles, where vB(t) is the countervalue effectiveness of B missiles. The evolution of the war over time, as summarized by the dynamic model (1)–(4), ; strategic decisions on both sides thus depends on the initial levels of missiles regarding rates of fire, α(t), β(t); strategic decisions on both sides regarding targeting α′(t), β′(t); the effectiveness of missiles of both sides against enemy missiles, fA(t), fB (t); and the effectiveness of missiles of both sides against enemy cities, vA(t), vB(t). From the viewpoint of either one of the countries, the problem of grand strategy is that of choosing both a rate of fire and a targeting strategy. For country A, the rate of fire α can range between zero and some maximum rate , determined on the basis of technical characteristics of weapons. Similarly, the counterforce proportion α′ can range between 0 and 1, where α′=1 is pure counterforce targeting (no cities) and α′=0 is pure countervalue targeting (only cities). Omitting intermediate values (which we
ARMS RACES AND INSTABILITY 127
have proved would never be used), the two extreme values for each of the two strategic variables of country A yield four alternatives for grand strategy for A:
(i) Maximum rate/counterforce, where A fires its missiles at the maximum ) and targets only B missiles (α′=1), destroying as many enemy rate ( missiles as possible—a first-strike strategy (ii) Maximum rate/countervalue, where A fires its missiles at the maximum ) and targets only B cities (α′=0), inflicting as many enemy rate ( casualties as possible—a massive retaliation strategy (iii) Zero rate/counterforce, where A holds its missiles in reserve (α=0) and targets only B missiles (α′=1), threatening to strike enemy missiles—a limited strategic war strategy (iv) Zero rate/countervalue, where A holds its missiles in reserve (α=0) and targets only B cities (α′=0), threatening to strike enemy cities—a war of nerves strategy. Assuming that country A treats the country B strategy as fixed, rather than trying to influence it, and that A has the goal of maximizing a pay-off function
(5)
which depends on missiles and casualties in both countries at the end of the war, time T, we have shown (see Intriligator, ‘Strategy in a Missile War’, 1967) that country A will select at any one time in the war t (where 0
128 DILEMMAS OF NUCLEAR STRATEGY
it can be one of a limited strategic war strategy (α=0 and α′=1) if the switching time for the rate precedes that for targets. Country A thus starts the war using only counterforce targeting at the maximum rate of fire so as to destroy as many as possible of the enemy missiles; it ends the war by holding enemy cities as hostages, placing itself in the best possible position for extracting concessions or a desired settlement of the war; and, depending upon which switching time occurs first, during the middle stage of the war it either inflicts massive casualties in country B or it holds its missiles in reserve, threatening enemy missiles.
APPENDIX B A Formal Model of Deterrence and War Initiation To use our model presented in Appendix A to analyze deterrence and war initiation it is reinterpreted in an essential way. It is treated not as a model of an actual war but rather as a model of a hypothetical or potential war, that is, a war that might break out at any time. Thus the model is treated not as a representation of actual missiles, casualties, strategic choices, etc. but rather as a representation of plans for a war, including expected missiles, casualties, strategic choices, etc. that are foreseen as possible situations by defense planners. The model is thus a representation of war simulations or war scenarios used by defense analysts, for example, those in the Pentagon or the Kremlin. From the vantage point of defense analysts in country A the model can be used to simulate various possible scenarios. Of particular importance are two cases. The first case is that in which B attacks A, in which A simulates the effect of the B attack and its own retaliation. In this case defense analysts in A would seek to have enough missiles to deter B by threatening it with unacceptable levels of casualties in its retaliatory strike. The second case is that in which A contemplates a first strike attack on B, in which its defense analysts would seek to destroy enough B missiles to disarm it and render a retaliatory strike ineffectual. These two cases are those of country A as a deterrer and country A as an attacker respectively. In the first case, of country A as a deterrer, its defense analysts should anticipate, as in the last section, that B will strike first using a first-strike strategy over the time interval from 0 to θ B, before A can retaliate. They would then retaliate using a massive retaliation strategy inflicting casualties in B over the retaliation time interval from θ B to θ B+ψ A, which is consistent with the switching strategy discussed in Appendix A. If defense planners in A simulate the result of this scenario they can solve the differential equations (1)–(4) using these strategic values if in this simulated war outbreak fB is assumed to be constant over the first strike inverval θ B and VA is assumed to be constant over the retaliation interval ψ A. These are reasonable assumptions if both intervals are relatively short, for example, measured in minutes. The solution for the casualties in country B at the end of the retaliatory interval is then
(6)
ARMS RACES AND INSTABILITY 129
showing explicitly the dependence of simulated casualties on initial number of ; rates of fire , ; missile effectiveness ratios fB, VA; and the time missiles intervals θ B, ψ A. If this expected number of casualties is sufficiently large then A would deter B. In particular, if the minimum unacceptable number of casualties in B is then, solving
(7) the number of A missiles required to deter B (by threatening an unacceptable for level of casualties) is given as
(8) and have been replaced by MA and MB respectively since the model is where one of a simulated war which can start at any time. The inequality shows the number of missiles required for A to deter B as an explicit function of the number of B missiles (and also the technical parameters , , fB, VA; timing parameters θ B, ψ A; and the minimum unacceptable number of B casualties ). Geometrically this inequality is the area to the right of the line marked ‘A deters’ in the weapons plane in Figure 1, with intercept on the MA axis and with slope . In the second case, of country A as an attacker, defense analysts in A anticipate , α′=1) over the time interval from 0 to θ attacking B using a first strike strategy ( A before B can retaliate. They then have to consider the effects of B retaliating using a massive retaliation strategy , β≤0) over the retaliatory time interval from θ A to θ A +ψ B. Solving the differential equations once again assuming the time intervals are sufficiently short that fA can be treated as constant over the first-strike interval and VB can be treated as constant over the retaliation time interval, the casualties defense planners in A anticipate in this simulated war outbreak is CA(θ A+ψ B) which is similar to (6) interchanging the roles of A and B. If defense planners in A regard as a maximum acceptable level of casualties then, solving
(9) for
yields
(10)
This inequality shows the number of A missiles required for A to attack B as an explicit function of the number of B missiles (and also the technical parameters , , fA, VB; timing parameters θ A, ψ B; and the maximum acceptable number of A
130 DILEMMAS OF NUCLEAR STRATEGY
casualties ). Geometrically this inequality is the area to the right of the line marked ‘A attacks’ in the weapons plane in Figure 1, with intercept on the MB. axis and with slope Figure 1 also shows ‘B deters’ and ‘B attacks’ lines, indicating regions in which defense analysts in country B, on the basis of their simulated war outbreak using the same model of a missile war, have enough missiles, respectively, to deter A from attacking (by threatening unacceptable casualties) and to attack A (suffering an acceptable level of casualties). If the technical and timing parameters are estimated to be the same in both countries then the ‘B deters’ line is parallel to the ‘A attacks’ line and the ‘B attacks’ line is parallel to the ‘A deters’ line. For example, the common . If, however, defense slope of the ‘B deters’ and ‘A attacks’ line is planners in the two countries have different assessments of these parameters (fA, fB, , , θ A, θ B) then the lines need not be parallel. This model of war initiation leads to regions of the weapons plane in Figure 1 which can be identified as ones of deterrence and of war initiation: The upper shaded cone bounded by the ‘A deters’ and ‘B deters’ lines is one of mutual deterrence, in which each country deters the other. This region above the point D of minimal mutual deterrence is one of stability against war outbreak since in it each country has the ability to inflict unacceptable damage on the other in a retaliatory attack. In this region each country has enough missiles to deter the other, and neither country has enough missiles to attack the other. It exists providing which is always satisfied if fA, bB<1, that is, it takes more than one missile to destroy an enemy missile. The lower shaded sawtooth-shaped area is one of forced initiation. In it neither country deters the other, and one or both can attack the other. In the middle portion of this area the cone between the origin and the point of initiation I, A can attack B and B can attack A, while neither can deter the other. Thus this portion is one of virtually forced pre-emption in which there is a great advantage to initiate, rather than retaliate. The ‘reciprocal fear of surprise attack’ based on the tremendous advantage in striking first virtually forces both sides to initiate, each trying to preempt the attack of the other. The other portions of this area are ones of asymmetry; in one, on the lower right, A can attack B but B cannot attack A, while in the other, on the upper left, B can attack A but A cannot attack B. In these two asymmetric regions stability relies on the intentions of one side. The entire sawtooth-shaped area, however, is one of initiation. This region exhibits instability against war outbreak in that neither side has an adequate retaliality capability to ensure deterrence and one or both countries has enough capabilities, relative to the missiles held by the opponent, to attack the other. The deterring and attacking lines for both countries that define the upper cone of mutual deterrence and the lower sawtooth-shaped area of initiation are themselves defined in (8) and (10) in terms of:
ARMS RACES AND INSTABILITY 131
fA, fB:
the effectiveness of missiles in destroying enemy missiles, which depends on missile accuracy and yield and enemy missile dispersion and hardness against attack; VA, VB: the effectiveness of missiles in inflicting enemy casualties, which depends on missile accuracy and yield and enemy civil defense and dispersion; the (maximum) rates of fire of missiles, which depend on missile characteristics and the command and control structure; θ A, θ B: the time intervals during which one country can attack the other without a response from the other, which depend on detection capabilities and command and control; ψ A, ψ B: the time intervals for a retaliatory strike, which depend on missile characteristics and command and control; the minimum unacceptable level of casualties which would deter each country, as estimated by the other country; and the maximum acceptable level of casualties which each country would accept in a retaliatory strike, as estimated by itself.
132 DILEMMAS OF NUCLEAR STRATEGY
Publications of the Project on Politics and War, University of California: The Soviet Calculus of Nuclear War, Roman Kolkowicz and Ellen Mickiewicz (eds.) (jointly with the Carter Center, Emory University) Lexington Books, 1986 The Logic of Nuclear Terror, Roman Kolkowicz (ed.), Allen & Unwin, 1987 The Roots of Soviet Power: Domestic Sources of Foreign and Defense Policies, Roman Kolkowicz (ed.) (forthcoming 1987) Communism, Militarism, Imperialism: Soviet Military Politics After Stalin, Roman Kolkowicz, Westview Press (forthcoming 1987) Arms Control and Arms Races: Lethal Logic, Dangerous Illusions (forthcoming 1987) The Soviet Military and the Communist Party, Westview Press Encore edition, 1986 International Security and Arms Control, Ellen Mickiewicz and Roman Kolkowicz (eds.), Praeger Publishers, 1986