Nuclear Myths and Political Realities

Author(s): Kenneth N. Waltz

Source: The American Political Science Review, Vol. 84, No. 3 (Sep., 1990), pp. 731-745
Published by: American Political Science Association
Stable URL: http://www.jstor.org/stable/1962764 .
Accessed: 18/07/2013 17:13

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to
The American Political Science Review.

http://www.jstor.org

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 NUCLEARMYTHSAND
REALITIES
POLITICAL
KENNETHN. WALTZ
Universityof California
Berkeley

wo pervasive beliefs have given nuclearweapons a bad name:

that nucleardeterrenceis highlyproblematic,and thata breakdownin deterrencewould
mean Armageddon.Both beliefs are misguidedand suggest that nearly half a century
afterHiroshima,scholarsand policy makershave yet to grasp the full strategicimplica-
tions of nuclearweaponry.I contrastthe logic of conventionaland nuclearweaponryto
show how nuclearweaponsare in fact a tremendousforce for peace and affordnations
that possess them the possibilityof securityat reasonablecost.

N unclearweapons course.Nevertheless,preparationto carry

have been given a bad name not just by the policy through proceeded apace. In
the Left, as one might have expected,but 1982 Secretaryof Defense CasparWein-
by the Centerand Rightas well. Through- bergersignedthe five-yearDefenseGuid-
out the long life of NATO, calls for ance Plan, which was to provide the
strengtheningconventional forces have meansof sustaininga nuclearwar; and in
been recurrentlyheard,reflectingand fur- Marchof that year an elaboratewar game
theringdebate about the wisdom of rely- dubbedIvy League"showed"that it could
ing on nuclear deterrence.Doubts were be done (Pringleand Arkin 1983, 22-40).
spread more widely when McGeorge Finally, in. March of 1983 President
Bundy, George Kennan, Robert Reagan offered his vision of a world in
McNamara,and Gerald Smith published which defensive systems would render
their argument for adopting a NATO nuclearweapons obsolete.
policy of "no first use" (Bundy et al. With their immensedestructivepower,
1982). From the Right came glib talk nuclearweapons are bound to make peo-
about the need to be preparedto fight a ple uneasy. Decades of fuzzy thinkingin
protractednuclearwar in orderto "deter" high placesabout what deterrenceis, how
the Soviet Union and proclaiming the it works, and what it can and cannot do
possibilityof doing so. BrigadierGeneral have deepenedthe nuclearmalaise.Forty-
LouisGuifridda,when he was directorof some years after the first atomic bombs
the Federal Emergency Management fell on Japan,we have yet to come to grips
Agency, well described the Reagan ad- with the strategicimplicationsof nuclear
ministration's intended nuclear stance: weapons. I apply nuclear reasoning to
"Theadministration,"he said, "categori- military policy and in doing so contrast
cally rejectedthe short war. We'retrying the logic of conventional and nuclear
to inject a long-war mentality" (Dowd weapons.
1984). Suchstatements,whichscaredpeo- Uneasiness over nuclear weapons and
ple at home and abroadout of theirwits, the searchfor alternativemeans of secur-
quickly disappeared from public dis- ity stem in largemeasurefromwidespread

AMERICAN POLITICAL SCIENCE REVIEW

VOLUME 84 NO. 3 SEPTEMBER1990

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 American Political Science Review Vol. 84

failure to understandthe nature and re- increasedthe ease with which destructive
quirementsof deterrence.Not unexpect- blows can be delivered, the distinction
edly, the language of strategic discourse between deterrenceand defense began to
has deterioratedover the decades. This blur. Earlyin PresidentKennedy'sadmin-
happens whenever discussion enters the istration, SecretaryMcNamarabegan to
political arena, where words take on promote a strategyof FlexibleResponse,
meanings and colorations reflecting the which was halfheartedly adopted by
preferencesof their users. Early in the NATO in 1967. FlexibleResponsecalls for
nuclearera deterrencecarriedits diction- the ability to meet threats at all levels
ary definition, dissuadingsomeone from from irregular warfare to conventional
an action by frighteningthat person with warfareto nuclearwarfare. In the 1970s
the consequencesof the action. To deter and 1980s more and more emphasiswas
an adversary from attacking one need placed on the need to fight and defendat
have only a force that can survive a first all levels in orderto "deter."The melding
strikeand strikeback hardenoughto out- of defense, war-fighting, and deterrence
weigh any gain the aggressorhad hoped overlooks a simple truth about nuclear
to realize. Deterrencein its pure form en- weapons proclaimedin the book title The
tailsno abilityto defend;a deterrentstrat- Absolute Weapon(Brodie1946). Nuclear
egy promisesnot to fend off an aggressor weaponscan carryout theirdeterrenttask
but to damage or destroy things the no matterwhat other countriesdo. If one
aggressor holds dear. Both defense land nuclear power were able to destroy
deterrenceare strategiesthat a status quo almost all of another'sstrategicwarheads
countrymay follow, hoping to dissuadea with practicalcertaintyor defend against
state from attacking. They are different all but a few strategicwarheads coming
strategiesdesignedto accomplisha com- in, nuclear weapons would not be abso-
mon end in differentways, usingdifferent lute. But because so much explosive
weapons differentlydeployed. Wars can power comes in such small packages, the
be prevented, as they can be caused, in invulnerabilityof a sufficientnumber of
various ways. warheadsis easy to achieveand the deliv-
Deterrenceantedatesnuclearweapons, ery of fairly large numbersof warheads
but in a conventional world deterrent impossibleto thwart,both now and as far
threatsare problematic.Stanley Baldwin into the futureas anyone can see. The ab-
warnedin the middle 1930s when he was solute quality of nuclearweaponssharply
prime minister of England that the sets a nuclear world off from a conven-
bomber would always get through, a tional one.
thought that helped to demoralize
England.It provedseriouslymisleadingin What Deters?
the war that soon followed. Bombers
have to make theirway past fighterplanes Most discussions of deterrence are
and through ground fire before finding based on the belief that deterrenceis diffi-
their targets and hitting them quite cult to achieve. In the Eisenhoweryears
squarely. Nuclear weapons purify deter- "massive retaliation" was the phrase
rent strategiesby removing elements of popularly used to describe the response
defense and war-fighting. Nuclear war- we would supposedly make to a Soviet
heads eliminate the necessity of fighting Union attack. Deterrencemustbe difficult
and remove the possibility of defending, if the threat of massive retaliationis re-
becauseonly a small numberof warheads quiredto achieveit. As the Soviet Union's
need to reach their targets. arsenal grew, MAD (mutual assured
Ironically,as multiplicationof missiles destruction) became the acronym of

732

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
Nuclear Myths and Political Realities

choice,actions
Military have to be the
thus preserving related
notionto an that
objective. Because of the awesome
deterrencedepends on being willing and power
nuclear
of able weapons,
to destroy the pressure
much, if not most, to use of a
them in ways that achieve the objective at
country. mini-
handThat whileone doing
must and able to adestroy
besuffering a
Itno-
mum of destruction would be
country in order to deter it is an odd immense.
that if a Soviet at-
tion, thoughto of
is preposterous thinkdistinguished lineage.
tack broke through NATO's
Duringthe 1950semphasiswas put on the defenses, the
United
massive States would strike
in massive thousands
retaliation. of
Beginning
Soviet military targets
in the 1960s the emphasis or hundreds
was put ofthe
on
Soviet cities. Doing so would
assured destruction in the doctrine serve no of
purpose. Who would want
MAD. Thus viewed, deterrencebecomes to make a bad
situation
a monstrous by launching
worsepolicy, as innumerable wantonly critics
can
have charged. One quotationthat
destructive attacks on a country can stand
strike back with
for many others. comparable
In a warning force,
to NATOor,
that matter,
fordefense on a country that could
ministers that became famous,
do so? In the event, we might strikea
notHenry Kissingercounseled the European
or industrial-
alliesorto
target nottwo-military
keep casualties low.to If
to keep "asking us multiply
the
chosen
strategic assurances that we cannot pos-
Soviet Union had run the preposterous
sibly mean or if
risk of attacking the center we do mean, of we
Europeshould
not want
believing it to execute
could because
escape if we execute,
retaliation, we
we risk the destruction
would thus show them that they of civilization"
were
(1981,
wrong 240).
while The notion
conveying thethat
ideathe thatfailure
more of
would follow if they persisted. suicide
deterrence would lead to national Among
or to mutual
countries annihilation
with abundant nuclearbetrays
weapons, a mis-
understanding of both
none can gain an advantage by strikingpolitical behavior
and
first. nuclear
The purposerealities.
of demonstrationshots
Introducing
is simply to remind the Eisenhower administra-
everyone-should
tion's New Look policy in threat-of
January
anyone forget-that catastrophe
1954, John Foster Dulles gave the impres-
ens. Some people purport to believe
that if a few warheadsgo off, many would
sion that aggression anywhere will
elicit heavy
follow. nuclear
This would retaliation.
seem to be the Just three
least
likely of all the unlikelypossibilities.Thatthe
months later, he sensibly amended
countryNuclear
nopolicy. gains bydeterrence, Dulles and
destroyinganother's
many others quickly
cities and then seeing a comparable came to num-
realize,
works not against minor aggression
ber of its own destroyedin returnis obvi- at the
periphery,
ous to everyone. but only against major aggres-
sion at the
Despite center, of
widespread international
beliefs to the con- poli-
tics. Moreover, to deter
trary, deterrence does not depend on major aggression,
Dulles now
destroying "theprobable
said,Deterrence
cities. hurt"on
depends need
only "outbalance the
what one can do, not on what one will do.probable gain"
(1954,
What deters359).
is the LikefactBrodie
that webeforecan do him, as
Dulles based deterrence
much damageto them as we choose, on the principle
and
of proportionality:
they to us. The country "Let the punishment
suffering the
fit the crime."
retaliatoryattackcannotlimit the damage
doneWhat would
to it; only we expect
the retaliator canthe United
do that.
States to do if the Soviet
With nuclear weapons, countriesneed Union launched
a major
threaten toconventional
use only a small attackamountagainst vital
of
U.S. interests-say, in Western Europe?
733

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 American Political Science Review Vol. 84

force. This is so becauseonce the willing- truly strategic weapons while we had
ness to use a little force is shown, the something like 2,000 in missiles and
adversaryknows how easily more can be bombs." But, he added, "withsome pro-
added. This is not true with conventional portionof Soviet deliveryvehiclessurviv-
weapons. Therefore,it is often useful for ing, the Soviet Union could do horren-
a countryto threatento use greatforce if dous damageto the United States"(Kiss-
conflict should lead to war. The stance inger 1979, 18). In other words, we could
may be intendedas a deterrentone, but not be sure that our two thousandweap-
the ability to carry the threat throughis ons would destroyalmostall of theirsixty
problematic. With conventional weap- or seventy. Even with numbersimmense-
ons, countriestend to emphasizethe first ly disproportionate,a smallforcestrongly
phase of war. Strikinghard to achieve a inhibitsthe use of a large one.
quick victory may decrease the cost of The catastrophepromised by nuclear
war. With nuclear weapons, political war contrasts sharply with the extreme
leadersworry not about what may hap- difficulty of predictingoutcomes among
pen in the firstphaseof fightingbut about conventional competitors. This makes
what may happenin the end. As Clause- one wonder about the claimed depen-
witz wrote, if war should ever approach dence of deterrenceon perceptionsand
the absolute, it would become "impera- the allegedproblemof credibility.In con-
tive ... not to take the first step without ventional competitions, the comparative
consideringwhat may be the last" (1976, qualitiesof troops, weaponry, strategies,
584). and leadersare difficultto gauge. So com-
Since war now approaches the abso- plex is the fighting of wars with conven-
lute, it is hardly surprisingthat President tional weapons that their outcomes have
Kennedy echoed Clausewitz'words dur- been extremelydifficult to predict. Wars
ing the Cuban Missile Crisis of 1962. liIt startmoreeasily becausethe uncertainties
isn't the first step that concernsme," he of their outcomes make it easier for the
said, "but both sides escalating to the leaders of states to entertainillusions of
fourth and fifth step-and we don't go to victory at supportablecost. In contrast,
the sixthbecausethereis no one aroundto contemplating war when the use of
do so" (R. Kennedy1969, 98). In conven- nuclearweapons is possible focuses one's
tional crises, leaders may sensibly seek attentionnot on the probabilityof victory
one advantage or another. They may but on the possibility of annihilation.
bluff by threateningescalatorysteps they Becausecatastrophicoutcomesof nuclear
are in fact unwilling to take. They may exchangesare easy to imagine,leadersof
try one stratagemor anotherand run con- stateswill shrinkin horrorfrom initiating
siderablerisks. Since none of the parties them. With nuclear weapons, stability
to the strugglecan predictwhat the out- and peace rest on easy calculations of
come will be, they may have good reason what one countrycan do to another.Any-
to prolongcrises, even crisesentailingthe one-political leaderor man in the street
risk of war. A conventionalcountry en- -can see that catastrophelurks if events
joying military superiorityis tempted to spiral out of control and nuclear war-
use it beforeothercountriesrightthe mili- heads begin to fly. The problem of the
tary balance.A nuclearcountryenjoying credibilityof deterrence,a big worry in
superiorityis reluctantto use it because a conventional world, disappears in a
no one can promise the full success of a nuclearone.
disarmingfirst strike. As HenryKissinger Yet the credibility of deterrencehas
retrospectivelysaid of the Cuban Missile been a constantU.S. worry. The worry is
Crisis, the Soviet Union had only "60-70 a hangoverfrom the 1930s. Concernover

734

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

credibility,and the relatedeffortsto show enable the Soviet Union to limit the
resolve in crises or wars where only casualtiesresultingfrom our retaliationto
peripheralinterestsare at stakewere rein- 3 % or 4 % of theirpopulation.Theirhope
forced because the formativeexperiences for such a "happy"outcome would pre-
of most of the policy makersof the 1950s sumablyrest on the confidencethat their
and 1960s took place in the 1930s. In re- first strike would be well timed and ac-
arming Germany, in reoccupying the curateand that their intelligenceagencies
Rhineland, in annexing Austria, and in would have revealedthe exact location of
dismantlingCzechoslovakia,Hitler went almost all of their intended targets. In
to the brink and won. "We must not let short, theirleaderswould have to believe
that happen again" was the lesson that all would go well in a huge, unre-
learned,but in a nuclearworld the lesson hearsed missile barrage, that the United
no longer applies. Despite rhetoricto the States would fail to launch on warning,
contrary, practice accords with nuclear and that if by chance they had failed to
logic because its persuasive force is so "deterour deterrent,"they would still be
strong, and the possible consequencesof able to limit casualtiesto only ten million
ignoring it so grave. Thus, John Foster people or so.1 But how could they enter-
Dulles, who proclaimedthat maintaining tain sucha hope when by Nitze'sown esti-
peacerequiresthe courageto go the brink mate their first strike would have left us
of war, shrankfrom the precipiceduring with two thousandwarheadsin our sub-
the Hungarianuprisingof 1956. And so it marineforce in additionto warheadscar-
has been every time that events even re- ried by survivingbombers?
motely threatenedto get out of hand at Nitze'sfear restedon the distinctionbe-
the centerof internationalpolitics. tween counterforcestrikes and counter-
Still, strategists' and commentators' value strikes-strikes aimed at weapons
minds prove to be impressively fertile. and strikes aimed at cities. Because the
The imagined difficulties of deterrence Soviet Union's first strike would be
multiply apace. One example will do: counterforce,any U.S. president would
Paul Nitze arguedin the late 1970s that, seemingly have good reason to refrain
given a certainbalanceof strategicforces, from retaliation,thus avoiding the loss of
the Soviet Union'ssupposedgoal of world cities still held hostage by the Soviet
domination,and its presumedwillingness Union's remaining strategic forces. But
to run greatrisks, the Soviet Union might this thought overlooks the fact that once
launch a first strike against our land- strategic missiles numbered in the low
based missiles, our bombers on the hundreds are fired, the counterforce-
ground, and our strategicsubmarinesin countervalue distinction blurs. One
port. The Soviet Union'sstrikewould tilt would no longerknow what the attacker's
the balance of strategic forces sharply intended targets might be. The Soviet
against us. Rather than retaliate, our Union'scounterforcestrikewould require
presidentmight decide to acquiesce;that that thousands, not hundreds, of war-
is, we might be self-deterred (1988, heads be fired. Moreover, the extent of
357-60). Nitze'sscenariois based on faul- theircasualties,shouldwe decideto retal-
ty assumptions, unfounded distinctions, iate, would depend on how many of our
and preposterousnotions about how gov- warheadswe chose to fire, what targets
ernmentsbehave. Soviet leaders, accord- we aimedat, and whetherwe usedground
ing to him, may have concludedfrom the burststo increasefallout. Severalhundred
trend in the balance of nuclear forces in warheadscould destroy either the United
the middle1970s that our relativelysmall States or the Soviet Union as ongoing
warheads and their civil defense would societies.The assumptionsmadein the ef-

735

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 American Political Science Review Vol. 84

fort to make a Soviet first strike appear this again reflectsconventional thinking.
possible are ridiculous. How could the In the absenceof a dominantweapon, the
Soviet Union-or any country, for that vulnerabilityof one weapon or another
matter-somehow bring itself to run stu- may be a big problem. If the means of
pendous risks in the presence of nuclear protecting sea-lanes of communications
weapons? What objectives might its were destroyed, for example, we would
leadersseek that could justifythe risksen- be unable to deploy and support troops
tailed? Answering these questions sensi- abroad. The problem disappears in a
bly leads one to concludethat deterrence nuclear world. Destroying a portion of
is deeplystable. Those who favor increas- one's strategicforce means little if suffi-
ing the strength of our strategic forces, cient weapons for deterrencesurvive.
however, shift to a different question. Thinking about deterrence is often
"The crucial question," according to faulted for being abstractand deductive,
Nitze, "is whethera futureU.S. president for not beinggroundedin experience.The
should be left with only the option of criticism is an odd one, since all state-
deciding within minutes, or at most ments about the military implicationsof
within two or three hours, to retaliate nuclear weapons are inferredfrom their
after a counterforceattack in a manner characteristics.Deterrersfrom Brodieon-
certain to result not only in military de- ward have drawn conclusions from the
feat for the United States but in wholly all-but-unimaginableincrease in easily
disproportionateand truly irremediable deliveredfirepowerthat nuclearwarheads
destruction to the American people" embody. Those who in the nuclearera ap-
(1988, 357). One of the marvels of the ply lessons learned in conventional war-
nuclearage is how easily those who write fare make the more problematic claim
about the unreliabilityof deterrencefocus that despiteprofoundchangesin military
on the retaliator'spossibleinhibitionsand technology the classic principlesof war-
play down the attacker'sobvious risks. fare endure (Rose 1980, 102-106). We
Doing so makesdeterrenceseem hardand all, happily, lack the benefit of experi-
leads to argumentsfor increasingour mili- ence. Moreover, just as deterrentlogic is
tary spendingin order"to deny the Soviet abstract and deductive, so too are the
Union the possibility of a successfulwar- weaknesses attributed to it. Scenarios
fighting capability" (1988, 360), a stra- showinghow deterrencemightfail arenot
tegic capabilitythat the Soviet Union has only abstractbut also far-fetched.Deter-
never remotely approached. rence rests on simple propositions and
We do not need ever-largerforces to relies on forces obviously sufficient for
deter. Smallerforces, so long as they are their purpose.
invulnerable, would be quite sufficient.
Yet the vulnerabilityof fixed, land-based
missiles has proved worrisome. Those Deterringthe Soviet Union
who do the worryingdwell on the vulner-
ability of one class of weapon. The mili- Underlyingmuch of the concernabout
tarilyimportantquestion,however, is not the reliabilityof nucleardeterrenceis the
about the vulnerability of one class of conviction that the Soviet Union is espe-
weapon but about the vulnerabilityof a cially hard to deter. Three main reasons
whole strategic-weaponssystem. Subma- are given for believing this. First, the
rine-launchedmissiles make land-based Soviet Union's ambitions are said to be
missiles invulnerable since destroying unlimited. In 1984 Secretaryof Defense
only the latter would leave thousandsof CasparWeinberger,when asked why the
strategic warheads intact. To overlook Soviet Union armed itself so heavily,

736

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

answered the question bluntly, 'World our countervailingcapabilitiesso that the

domination, it's that simple" (Rosenthal Soviets will clearly understandthat we
1989). Second, her military doctrine will never allow them to use theirnuclear
seemed to contemplatethe possibility of forces to achieve any aggressiveaim at an
fighting and winning combined conven- acceptablecost" (1980, 83).
tional and nuclearwar, while rejectingthe The belief that the Soviet Union'shav-
doctrine of deterrence.Third, the Soviet ing an aggressivemilitarydoctrinemakes
Unionhas appearedto many peoplein the her especially hard to deter is another
West to be striving for military super- hangover from conventional days. Ger-
iority. many and Japanin the 1930swere hardto
These three points make a surprisingly deter, but then the instrumentsfor deter-
weak case, even thoughit has been widely rence were not available. We can fairly
accepted. Ambitions aside, looking at say that their leaders were less averse to
the Soviet Union's behavior one is im- running risks than most political leaders
pressed with its caution when acting are. But that is no warrantfor believing
where conflict might lead to the use of that had they been confronted with
major force. Leadersof the Soviet Union second-strikenuclear forces, they would
may hope that they can one day turn the have been so foolhardyas to risk the sud-
world to communism. Although the den destruction of their countries. The
Soviet Union'sintentionsmay be extraor- decision to challengethe vital interestsof
dinary, her behaviorhas not been. Every- a nuclearstate, whetherby majorconven-
one agreesthat except in the militarysec- tional assault or by nuclear first strike,
tor, the Soviet Union is the laggingcom- would be a collective decisioninvolving a
petitor in a two-party race. The Soviet numberof political and military leaders.
Union has been opportunisticand disrup- One would have to believe that a whole
tive, but one expects the laggingparty to set of leaders might suddenly go mad.
score a point or two wheneverit can. The Rulerslike to continue to rule. Exceptin
Soviet Union has not scored many. Her the relatively few countries of settled
limitedinternationalsuccessesshould not democraticinstitutions, one is struck by
obscure the fact that what the Soviet how tenaciouslyrulerscling to power. We
Union has done mostly since 1948 is lose. have no reason to expect Russianleaders
The second point rests on basic mis- to be any different.The notion that Rus-
understandingsabout deterrence.It has sian leadersmight risk losing even a small
often been arguedthat we could not rely numberof cities by questingmilitarilyfor
on deterrencewhen the Soviet Union was uncertaingains is fanciful.Malenkovand
rejecting'thedoctrine. One of the draw- Khrushchevlost their positions for lesser
backs of the "theory"of assureddestruc- failures.
tion, according to Henry Kissinger,was With conventional weapons a status
that "theSoviets did not believe it" (1981, quo country must ask itself how much
238). The efficacy of nuclear deterrence, power it must harness to its policy in
however, does not depend on anyone's orderto dissuadean especiallyaggressive
acceptingit. Secretariesof defensenever- state from striking. Countries willing to
theless continue to worry that Soviet run high risks are hard to dissuade. The
values, perceptions,and calculationsmay varied qualities of governmentsand the
be differentfrom ours. Thus, Secretaryof temperamentsof leadershave to be care-
Defense Harold Brown, worried by the fully weighed. In a nuclear world any
Soviets' emphasis "on the acquisition of state will be deterredby another state's
war-winningcapabilities,"concludedthat second-strike forces. One need not
we must "continueto adapt and update become preoccupiedwith the character-

737

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 American Political Science Review Vol. 84

istics of the state that is to be deterredor of nuclearweapons is yet to be appreci-

scrutinizeits leaders. ated. So long as two or more countries
The third worry remains: the Soviet have second-strike forces, to compare
Union's seeming aspiration for military them is pointless. If no state can launcha
superiority. One might think that the disarming attack with high confidence,
worry should have run the other way force comparisons become irrelevant.
through most of the years of the Cold For deterrence one asks how much is
War. In the nuclear business the United enough, and enough is defined as "hav-
Statesmoved from monopolyto superior- ing a second-strikecapability."This does
ity. In the late fifties Khrushchevdeeply not imply that a deterrentforce deters
cut conventional arms, and the Soviet everything,but ratherthat beyond a cer-
Union failed to produce strategic war- tain level of capability, additionalforces
heads and missiles as rapidly as we had provide no additional coverage for one
expected. Nevertheless,the Kennedyad- party and pose no additional threat to
ministrationundertookthe largestpeace- others. The United States and the Soviet
time military buildup the world had yet Union have long had second-strikeforces,
seen in both nuclear and conventional with neither able to launch a disarming
weaponry. We forged ahead so far stra- first strike against the other. Two coun-
tegically that RobertMcNamarathought tries with second-strikeforces have the
the Soviet Union would not even try to same amount of strategicpower-since,
catch up. "Thereis," he said, "no indica- short of attaininga first-strikecapability,
tion that the Soviets are seeking to addingmoreweaponsdoes not changethe
develop a strategicnuclearforce as large effective militarybalance.
as ours" (Interview, U.S. News and
World Report, 12 April 1976, 52). The
Soviet Union'sgiving up would have been Why Nuclear Weapons
historically unprecedented.Instead, the Dominate Strategy
Soviet Union did try to compete, but to
catch up with the United Stateswas diffi- Deterrence is easier to contrive than
cult. In the 1970s, the decadein which we most strategistshave believed. With con-
are told the Soviet Union moved toward ventionalweapons, a numberof strategies
superiority (or, according to President are available, strategies combining and
Reagan,achievedit), the United Statesin deployingforcesin differentways. Strate-
fact added more nuclearwarheadsto its gies may do more than weapons to deter-
arsenalthan the Soviet Union did. mine the outcomes of wars. Nuclear
We have exaggeratedthe strengthof the weapons are different; they dominate
Soviet Union; and they, no doubt, have strategies. As Brodie clearly saw, the
exaggerated ours. One may wonder effects of nuclear weapons derive not
whetherthe Soviet Union ever thoughtit- from any particulardesign for their em-
self superioror believed it could become ployment in war but simply from their
so. Americans easily forget that the presence(1973,412). Indeed,in an impor-
Soviet Union has the strategicweaponsof tant sense, nuclear weapons eliminate
the United States, Britain, France, and strategy. If one thinks of strategies as
China pointed at it and sees itself threat- being designedfor defendingnationalob-
ened from the East as well as the West. jectives or for gaining them by military
More fundamentally,continuedpreoccu- force and as implyinga choice about how
pation with denying "superiority"to the majorwars will be fought, nuclearweap-
Soviet Union, if not seekingit ourselves, ons make strategyobsolete. Nevertheless,
suggeststhat a basic strategicimplication the convictionthat the only reliabledeter-

738

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

rent force is one able to win a war or at portant to keep the sea lanes open"as an
least end up in a betterposition than the exampleof the "quaintideas"still held by
Soviet Union is widespread. Linton F. the military (1960, 38). Conventional
Brooks,while a captainin the U.S. Navy, forceshave only a narrowrole in any con-
wrote that "waris the ultimatetest of any frontation between nuclear states over
strategy;a strategyuseless in war cannot vital interests, since fighting beyond the
deter"(1988, 580; see also Howard 1981, trip wire level serves no useful purpose.
15). Enlargingconventional capabilities does
NATO policy well illustratesthe futil- nothing to strengthen deterrence. Stra-
ity of trying to transcenddeterrenceby tegic stalematedoes shift militarycompe-
fashioning war-fighting strategies. The tition to the tactical level. But one must
supposed difficulties of extending deter- add what is usuallyomitted:nuclearstale-
renceto cover majorallieshas led some to mate limits the use of conventionalforces
arguethat we requirenuclearsuperiority, and reduces the extent of the gains one
thatwe neednuclearwar-fightingcapabil- can seek without riskingdevastation.For
ities, and that we must build up our con- decades U.S. policy has nevertheless
ventional forces. Once the Soviet Union aimed at raising the nuclearthresholdin
achievednuclearparity,confidencein our Europe. Stronger conventional forces
extendeddeterrentdeclined in the West. would presumablyenable NATO to sus-
One wonders whether it did in the East. tain a longer war in Europe at higher
Denis Healeyonce said that one chancein levels of violence. At some moment in a
a hundredthat a country will retaliateis major war, however, one side or the
enough to deter an adversary, although other-or perhaps both-would believe
not enough to reassure an ally. Many itself to be losing. The temptationto in-
have repeatedhis statement;but none, I troduce nuclear weapons might then
believe, has addedthat reassuringallies is prove irresistible,and they would be fired
unnecessarymilitarilyand unwise politi- in the chaos of defeatwith little chanceof
cally. Politically,allies who are unsureof limited and discriminantuse. Early use
one another's support have reason to would promise surer control and closer
work harder for the sake of their own limitationof damage.In a nuclearworld a
security. Militarily, deterrence requires conventionalwar-fightingstrategywould
only that conventional forces be able to appearto be the worst possibleone, more
defend long enough to determinethat an dangerousthan a strategy of relying on
attack is a major one and not merely a deterrence.
foray. For this, a trip wire force as envis- Attemptsto gain escalationdominance,
ioned in the 1950s, with perhaps fifty like efforts to raise the nuclearthreshold,
thousandU.S. troopsin Europe,would be betraya failureto appreciatethe strategic
sufficient. Beyond that, deterrence re- implicationsof nuclearweapons. Escala-
quires only that forces be invulnerable tion dominance, so it is said, requiresa
and that the area protectedbe of mani- "seamless web of capabilities"up and
festly vital interest.West Europeancoun- down "the escalation ladder." Earlier,it
triescan be countedon to maintainforces had been thought that the credibilityof
of trip wire capability. deterrencewould be greaterif some rungs
Nuclear weapons strip conventional of the escalationladderwere missing.The
forces of most of theirfunctions.Bernard inability to fight at some levels would
Brodie pointed out that in "a total war" make the threat to use higher levels of
the army "mighthave no function at all" force easy to credit. Butagain, since cred-
(1957, 115). Herman Kahn cited "the ibility is not a problem,this scarcelymat-
claimthat in a thermonuclearwar it is im- ters militarily.Fillingin the missingrungs

739

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 American Political Science Review Vol. 84

neitherhelps nor hurts. Escalationdomi- drously high. Effortsto fashion a defen-

nance is useful for countries contending sive and war-fightingstrategyfor NATO
with conventional weapons only. Domi- are pointless because deterrenceprevails
nance, however, is difficult to achieve in and futile because strategy cannot tran-
the absenceof a decisiveweapon. Among scend the militaryconditionsthat nuclear
nuclearadversariesthe question of domi- weapons create.
nance is pointless because one second-
strike force cannot dominate another. Nuclear Arms and Disarmament
Since strategic nuclear weapons will
always prevail, the game of escalation The probability of major war among
dominance cannot be played. Everyone states having nuclear weapons ap-
knows that anyone can quickly move to proacheszero. But the "realwar"may, as
the top rung of the ladder. Because WilliamJamesclaimed,lie in the prepara-
anyone can do so, all of the parties in a tion for wagingit. The logic of deterrence,
seriouscrisishave an overridingincentive if followed, circumscribesthe causes of
to ask themselvesone question:How can "realwars" (1968, 23). Nuclear weapons
we get out of this mess without nuclear make it possiblefor a state to limit the size
warheads' exploding? The presence of of its strategic forces as long as other
nuclearweaponsforces themto figureout states are unable to achieve disarming
how to deescalate,not how to escalate. first-strikecapabilitiesby improvingtheir
To gain escalation dominance, if that forces.
were imaginable,would requirethe abil- Withinvery wide ranges,a nuclearbal-
ity to fight nuclear wars. War-fighting ance is insensitiveto variationin numbers
strategiesimply that nuclearweapons are and size of warheads. This has occasion-
not absolute but relative, so that the ally been seen by responsible officials.
country with more and better nuclear Harold Brown, when he was secretaryof
weapons could in some unspecifiedway defense, said that purely deterrentforces
prevail. No one, however, has shown "can be relatively modest, and their size
how such a war could be fought. Indeed, can perhaps be made substantially,
Desmond Ball has argued that a nuclear though not completely, insensitive to
war could not be sustained beyond the changes in the posture of an opponent."
exchangeof strategicwarheadsnumbered Somehow, he nevertheless managed to
not in the hundredsbut in the tens (1981, argue that we need "to design our forces
9). After a small numberof exchangesno on the basis of essential equivalents"
one would know what was going on or be (1979, 75-76). Typically, over the past
able to maintain control. Yet nuclear three decades secretariesof defense have
weapons save us from our folly: fanciful sought, albeit vainly, the superiority
strategies are irrelevantbecause no one that would supposedly give us a war-
will runthe appallingrisk of testingthem. fightingcapability.But they have failedto
Deterrencehas been faultedfor its lack explainwhat we can do with twelve thou-
of credibility, its dependenceon percep- sand strategicnuclear warheads that we
tions, its destructiveimplications,and its could not do with two thousand or an
inability to cover interests abroad. The even smaller number. What difference
trouble with deterrence, however, lies does it make if we have two thousand
elsewhere:the trouble with deterrenceis strategic weapons and the Soviet Union
that it can be implementedcheaply. The has four thousand?We thoughtour deter-
claim that we need a seamlessweb of cap- rent did not detervery much and did not
abilities in order to deter does serve one work with sufficientreliabilityjust as we
purpose: it keeps military budgets won- were reachinga peak of numericalsuperi-

740

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

ority in the mid-1960s.Flexibleresponse, the balance of forces if agreementsare

with emphasison conventionalarms,was reached to reduce them. In a world of
a policy produced in our era of nuclear second-strike nuclear forces, alliances
plenty. "Superiority"and "parity"have have little effect on the strategicbalance.
had the same effect on our policy. The Soviet Union's failure to insist that
Many who urge us to build ever more British, French, and Chinese forces be
strategic weapons in effect admit the counted in strategic arms negotiations
military irrelevanceof additional forces may reflectits appreciationof this point.
when, as so often, they give political Finally, conventional powers have to
ratherthan militaryreasonsfor doing so: compareweapons of uncertaineffective-
spending less, it is said, would signal ness. Arms agreements are difficult to
weaknessof will. Yet militarily,only one reach because their provisions may bear
perceptioncounts, namely,the perception directly on the prospects for victory or
that a country has second-strikeforces. defeat. Becausein a nuclearworldpeaceis
Nuclear weapons make it possible for maintainedby the presence of deterrent
statesto escapethe dynamicsof armsrac- forces, strategicarms agreementsdo not
ing; yet the United States and the Soviet have militarybut economicand political,
Union have multipliedtheirweaponryfar significance. They can benefit countries
beyond the requirementsof deterrence. economically and help to improve their
Each has obsessively measured its stra- relations.
tegic forces against the other's. The arms A minority of U.S. military analysts
competition between them has arisen have understoodthe folly of maintaining
fromfailureto appreciatethe implications morenuclearweaponsthan deterrencere-
of nuclearweapons for military strategy quires.In the Soviet Union, MikhailGor-
and, no doubt, from internalmilitaryand bachev and some others have put forth
political pressuresin both countries. the notion of "reasonablesufficiency,"
Many of the obstacles to arms reduc- definedas havinga strategicforceroughly
tion among conventional powers disap- equal to ours and able to inflictunaccept-
pear or dwindle among nuclear nations. able damage in retaliation. Edward
Forthe former,the carefulcomparisonof Warner points out that some civilian
the quantitiesand qualitiesof forcesis im- analysts have gone further, "suggesting
portant. Because this is not so with that as long as the USSR had a secure
nuclearweapons, the problem of verify- second-strikecapabilitythat could inflict
ing agreementslargelydisappears.Provi- unacceptabledamage, it would not have
sions for verificationmay be necessaryin to be concerned about maintaining ap-
order to persuadethe Senate to ratify an proximate numerical parity with U.S.
agreement, but the possibility of non- strategic nuclear forces" (1989, 21). If
complianceis not very worrisome.Agree- leaders in both countriescome to accept
ments that reduce one category of con- the minorityview-and also realizethat a
ventional weapons may shift competition deterrent force greatly reduces conven-
to other types of weapons and lead to in- tional requirementson central fronts
creasesin their numbersand capabilities. both countriescan enjoy securityat much
Becausewith nuclearweapons sufficiency lower cost.
is easily defined, there is no military
reasonfor reductionsin some weapons to
result in increasesin others. Convention- StrategicDefense
ally, multiparty agreementsare hard to
arrive at because each party has to con- Strategic defenses would radically
sider how shifting alignmentsmay alter change the propositions advanced here.

741

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 AmericanPoliticalScienceReview Vol. 84

The Strategic Defense Initiative, in he was secretaryof defense, Robert

Reagan'svision, was to providean area McNamaraarguedthat the appropriate
defense that would protect the entire responseto a Soviet defensivedeploy-
populationof the UnitedStates.Strategic ment would be to expandour deterrent
defenseswereto posean absolutedefense force.Morerecently,CasparWeinberger
against what have been absolute andMikhailGorbachevhavemadesimni-
weapons,thus renderingthem obsolete. lar statements.Any countrydeployinga
Theconsequences thatwouldfollowfrom defenseeffectivefor a momentcannot
mountingsucha defensebogglethemind. expectit to remainso. Theeaseof deliver-
That a perfectdefenseagainstnuclear ingnuclearwarheadsandthedestructive-
weaponscouldbe deployedandsustained ness of smallnumbersof themmakethe
is inconceivable.This is so for two durabilityof defenseshighlysuspect.
reasons:(1) it is impossible,and (2) if it Thelogicof strategicdefenseis thelogic
werepossible,it wouldntIlast of conventionalweaponry.Conventional
Nuclearweaponsare smalland light; strategiespit weaponsagainstweapons.
theyareeasy to move, easy to hide,and That is exactlywhat a strategicdefense
easyto deliverin a varietyof ways. Even woulddo, therebyrecreating the tempta-
an unimaginably perfectdefenseagainst tions and instabilitiesthat have plagued
ballistic missiles would fail to negate countriesarmedonly with conventional
nuclearweapons.Sucha defensewould weapons. If the United States and the
insteadput a premiumon the otherside's Soviet Union deploy defensivesystems,
abilityto delivernuclearweaponsin dif- eachwill worry-no doubtexcessively-
ferentways: firingmissileson depressed aboutthebalanceof offensiveanddefen-
trajectories,
carryingbombsin suitcases, sive capabilities.Eachwill fear that the
placingnuclearwarheadson freightersto othermayscorean offensiveor defensive
be anchoredinAmericanharborsIndeed, breakthrough. If one sideshoulddo so, it
someonehas suggestedthat the Soviet mightbe temptedto strikein orderto ex-
Unioncanalwayshidewarheadsin bales ploitits temporary advantage.Thedread-
of marijuana,knowingwe cannotkeep ed specterof the hair triggerwould re-
themfromcrossingourborders.To have appear. Under such circumstancesa
even modestly effective defenses we defensivesystemwouldserveas theshield
would, among other things, have to that makesthe sworduseful.An offen-
becomea policestate.We wouldhaveto sive-defensive racewouldintroducemany
go to extraordinary lengthsto policeour uncertainties.A country enjoying a
bordersandexercisecontrolwithinthem. momentarydefensiveadvantagewould
Presumably, the SovietUniondoesthese be temptedto strikein the forlornhope
thingsbetterthanwe do. It is impossible thatits defenseswouldbe ableto handlea
to imaginethatan areadefensecan be a raggedand reducedresponseto its first
successbecausethereareso manywaysto strike.Both countrieswould prepareto
thwartit. In no way can we preventthe launch on warning while obsessively
SovietUnionfromexplodingnuclearwar- weighingthe balancebetweenoffensive
headson or in the UnitedStatesif it is anddefensiveforces.
determined to do so. Finally,let us imaginewhatis mostun-
Second,let us imaginefor a moment imaginableof all-that bothsidesdeploy
thatan airtightdefense,howeverdefined, defensesthat are impregnableand dur-
is aboutto be deployedby onecountryor able. Such defenses would make the
theother.Thecloseronecountrycameto world safe for World War Ill-fought
deployingsucha defense,the harderthe presumably in the mannerof WorldWar
otherwouldworkto overcomeit. When II but with conventionalweapons of

742

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

much greaterdestructivepower. world, states going to war can at once

Still, some have argued that even if believe that they may win and that,
some Americancities remainvulnerable, should they lose, the price of defeat will
defensesare very good for the cities they be bearable. World Wars I and II called
do cover. The claim is spurious. In re- the latter belief into question before
sponse to the Soviet Union's deploying atomic bombs were ever dropped. If
antiballisticmissiles to protect Moscow, the United States and the Soviet Union
we multiplied the number of missiles were now armed only with conven-
aimedat that city. We expectto overcome tional weapons, the lesson of those
their defenses and still deliver the "re- wars would be strongly remembered-
quired"number of warheads. The result especially by Russia, since she has suf-
of defendingcities may be that more war- fered more in war than we have. If the
heads strike them. This is especially so atom had never been split, the United
because both they and we, working on States and the Soviet Union would still
worst-case assumptions, are likely to have much to fear from each other. The
overestimatethe numberof missiles that starkopposition of countriesof continen-
the other country'ssystem will be able to tal size armedwith ever-more-destructive
shoot down. Strategicdefensesare likely conventional weapons would strongly
to increasethe damage done. constrain them. Yet in a conventional
Most knowledgeable people believe world even forceful and tragic lessons
that an almost leak-proofdefense cannot have provedto be exceedinglydifficultfor
be built. Many, however, believe that if states to learn. Recurrentlyin modern
improvedhard-pointdefensesresultfrom history one great power or another has
the SDI program,they will have justified looked as thoughit mightbecomedanger-
its price. Defense of missilesand of com- ously strong (LouisXIV'sand Napolean's
mand, control, and communicationsin- France, Wilhelm II's and Hitler's Ger-
stallations will strengthendeterrence,so many). Each time, an opposing coalition
the argumentgoes. That would be a solu- formed, if belatedly, and turned the ex-
tion, all right;but we lack a problemto go pansive state back. The lesson would
with it: deterrenceis vibrantlyhealthy. If seem to be clear:in internationalpolitics,
the Soviet Union believes that even one success leads to failure. The excessiveac-
Tridentsubmarinewould survive its first cumulation of power by one state or
strike, it will be deterred.3Since we do coalition of states elicits the oppositionof
not need hard-pointdefenses, we should others. The leadersof expansioniststates
not buy them. The deployment of such have neverthelessbeen able to persuade
defensesby one side would be seen by the themselves that skillful diplomacy and
other as the first step in deployingan area clever strategymight enablethem to tran-
defense. Strategic considerationsshould scend the normalprocessesof balance-of-
dominate technical ones. In a nuclear power politics. The Schlieffen Plan, for
world defensive systems are predictably example, seemed to offer a strategy that
destabilizing.It would be folly to move would enableGermanyto engageenemies
from a condition of stable deterrenceto on two fronts, serially: Germany would
one of unstabledefense. defeat Francebefore Russia could mobil-
ize fully and move westward in force.
Conclusion Later, Hitler, while denouncing the
"boobs"of WilhelmineGermanyfor get-
Nuclear weapons dissuade states from ting themselvesinto a war on two fronts,
going to war more surely than conven- reenactedtheir errors.
tional weapons do. In a conventional How can we perpetuatepeace without

743

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 AmericanPolitical Science Review Vol. 84

solving the problem of war? This is the References

questionthat stateswith nuclearweapons
must constantly answer. Nuclear states Ball, Desmond. 1981. "Counterforce Targeting:
How New? How Viable?" Arms Control Today
continueto competemilitarily.With each 11:1-9.
state tending to its security interests as Brodie, Bernard, ed. 1946. The Absolute Weapon.
best it can, war is constantlypossible. Al- New York: Harcourt, Brace.
though the possibility of war remains, Brodie, Bernard. 1957. "More about Limited War."
nuclearweaponshave drasticallyreduced World Politics 10:112-22.
Brodie, Bernard. 1973. War and Politics. New York:
the probabilityof its being fought by the Macmillan.
states that have them. Wars that might Brooks, Linton F. 1988. "Naval Power and National
bringnuclearweapons into play have be- Security." In The Use of Force, 3d ed., ed. Robert
come extraordinarilyhard to start. Over J. Art and Kenneth N. Waltz. Lanham, MD: Uni-
versity Press of America.
the centuries great powers have fought Brown, Harold. 1979. Annual Report, Fiscal Year
more wars, and lesser states have fought 1980. Washington: U.S. Department of Defense.
fewer:the frequencyof war has correlated Brown, Harold. 1980. Annual Report, Fiscal Year
less closely with the attributesof states 1981. Washington: U.S. Department of Defense.
Bundy, McGeorge, George F. Kennan, Robert S.
thanwith theirinternationalstanding.Yet McNamara, and Gerard Smith. 1982. "Nuclear
becauseof a profoundchangein military Weapons and the Atlantic Alliance." Foreign Af-
technology, waging war has more and fairs 60:753-68.
more become the privilege of poor and Clausewitz, Carl von. 1976. On War. Trans.
weak states. Nuclear weapons have re- Michael Howard and Peter Paret. Princeton:
Princeton University Press.
versedthe fates of strongand weak states. Dowd, Maureen. 1984. "Ferraro Suggests Reagan
Never since the Treaty of Westphaliain Go Home." New York Times 22 September.
1648, which conventionallymarksthe be- Dulles, John Foster. 1954. "Policy for Security and
ginning of modern history, have great Peace." Foreign Affairs 32:353-64.
powers enjoyed a longer period of peace Howard, Michael. 1981. "On Fighting a Nuclear
War." International Security 5:3-17.
than we have known since the Second James, William. 1968. "The Moral Equivalent of
WorldWar. One can scarcelybelieve that War." In War: Studies from Psychology, Sociol-
the presenceof nuclearweapons does not ogy, Anthropology, 2d ed., ed. Leon Bramson
greatly help to explainthis happy condi- and George W. Goethals. New York: Basic
tion. Books.
Kahn, Herman. 1960. On Thermonuclear War.
Princeton: Princeton University Press.
Kennedy, Robert F. 1969. Thirteen Days. New
York: Norton.
Kissinger, Henry. 1979. "Kissinger's Critique."
Notes Economist 3 February.
Kissinger, Henry. 1981. For the Record: Selected
This essay was presentedas the presidentialad- Statements, 1977-1980. Boston: Little, Brown.
dressat the annualmeetingof the AmericanPolitical Lewis, Paul. 1938. "Soviet Official Says Talks on
Science Association in Washington, 1988. David Arms Should Emphasize Cuts in Navies." New
Schleicher'spenetratingcriticismsand constructive York Times 19 October.
suggestionshelpedme greatlyin writingit. Nitze, Paul H. 1988. "Deterring Our Deterrent." In
1. Nitze blandly adds that if we do launch on The Use of Force, 3d ed., ed. Robert J. Art and
warning, "the estimatesin the Soviet civil defense Kenneth N. Waltz. Lanham, MD: University
manualsare overoptimisticfrom the Soviet view- Press of America.
point' (1988, 357). Pringle, Peter, and William Arkin. 1983. S.I.O.P.:
2. Quaintideasdie hard.In the fall of 1989NATO The Secret of U.S. Plan for Nuclear War. New
resisted discussing naval disarmamentwith the York: Norton.
SovietUnionbecauseof the needfor forcesto guard Rose, John P. 1980. The Evolution of U.S. Army
the sea-lanesto Europe(Lewis1988). Nuclear Doctrine, 1945-1980. Boulder: West-
3. An Ohio-class Trident submarine carries view.
twenty-fourmissiles,each with eight warheads. Rosenthal, Andrew. 1989. "Pentagon Report Softens
Soviet Menace." New York Times 28 September.

744

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions
 Nuclear Myths and Political Realities

Warner, Edward L. III. 1989. "New Thinking and

Old Realities in Soviet Defense Policy." Survival
31:13-33.

KennethN. Waltzis Professorof PoliticalScience,Universityof California,Berkeley,

CA 94720.

745

This content downloaded from 152.14.136.96 on Thu, 18 Jul 2013 17:13:15 PM

All use subject to JSTOR Terms and Conditions