The Emergence of Cooperation National Ep

The Emergence of Cooperation: National Epistemic Communities and the International
Evolution of the Idea of Nuclear Arms Control

Author(s): Emanuel Adler
Source: International Organization, Vol. 46, No. 1, Knowledge, Power, and International Policy
Coordination (Winter, 1992), pp. 101-145
Published by: The MIT Press
Stable URL: http://www.jstor.org/stable/2706953 .
Accessed: 11/02/2011 12:24
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=mitpress. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
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.
The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to International
Organization.
http://www.jstor.org
The emergenceofcooperation:
nationalepistemiccommunitiesand
theinternationalevolutionoftheidea
ofnucleararms control
Emanuel Adler
An Americanepistemiccommunity played a keyrole in creatingthe interna-

tionalshared understanding and practiceof nucleararmscontrol,whichgave
meaningto and helped coordinateexpectationsof superpowercooperation
duringthe Cold War.' In thisstudy,I analyzehowthecommunity's theoretical
and practicalideas became politicalexpectations,were diffusedto the Soviet
Union, and were ultimatelyembodiedin the 1972 antiballisticmissile(ABM)
armscontroltreaty.
In the late 1950s,when the idea of nuclear arms controlwas introduced,
nucleardeterrencewas onlya conceptthatcould neitherbe takenforgranted
nor ruled out.2However,havingbecome aware of the vulnerability of U.S.
nuclearweapons and concernedabout the reciprocalfear of surpriseattack,
the strategistsand scientistsmakingup the U.S. epistemiccommunitypre-
dictedthatboth the nationalsecurityof the United States and the chances of
avoidingnuclearwarwouldbe enhancedifthe superpowerswouldcollaborate
For theircommentsand insights,I am gratefulto the membersof the reviewcommitteeof

InternationalOrganization;to theothercontributors to thisspecial issue,especiallyPeterHaas and
M. J. Peterson;to mycolleagues at the Center forScience and InternationalAffairs,especially
JosephNye; and to HaywardAlker,StephenGraubard,JosephGrieco,ErnstHaas, and Thomas
Schelling.Research fundswere providedby the Center for Science and InternationalAffairs,
HarvardUniversity, and by the Leonard Davis InstituteforInternationalRelations,the Hebrew
University of Jerusalem.An earlierversionof thisarticlewas presentedat the annual meetingof
theAmericanPoliticalScience Association,Washington,D.C., 1988.
1. An epistemiccommunity, as definedin thisissue ofIO, is a networkof individualsor groups
withan authoritativeclaim to policy-relevant knowledgewithintheirdomain of expertise.The
community membersshareknowledgeabout thecausationofsocial and physicalphenomenain an
area forwhichtheyhave a reputationforcompetence,and theyhave a commonset of normative
beliefsabout whatwillbenefithumanwelfarein such a domain.While membersare oftenfroma
number of differentprofessionsand disciplines,they adhere to the following:(1) shared
consummatory values and principledbeliefs;(2) sharedcausal beliefsor professionaljudgment;(3)
common notions of validitybased on intersubjective, internallydefinedcriteriafor validating
knowledge;and (4) a commonpolicyproject.
2. LawrenceFreedman,TheEvolutionofNuclearStrategy (New York: St. Martin'sPress,1983),
p. 191.
International 46, 1, Winter1992

Organization
? 1992 bythe WorldPeace Foundationand the MassachusettsInstituteof Technology
102 InternationalOrganization
to stabilize the nuclear balance througharms control.Energized by their

shared epistemiccriteriaabout the causes of war,the effectsof technological
change on the armsrace, and the need fornuclear adversariesto cooperate,
these strategistsand scientistsreached into the places where decisions are
made and intothe mindsof thepeople who made them,3therebyturningtheir
ideas into widespread national securitypolicyand practice.They also were
instrumental in diffusing to the SovietUnion. Indeed, after
thisunderstanding
a timethe Sovietsagreed to negotiatewiththe Americanson the basis of this
understanding, and it has formedthe foundationof U.S.-Soviet cooperation
overthelast thirty years.
The relevance of my studyof the arms controlepistemiccommunityfor
understandinginternationalcooperationlies in the notion that domestically
developed theoreticalexpectationswhichwere createdby a nationalgroupof
expertsand were selectedbytheU.S. government as thebasis fornegotiations
with the Soviets became the seed of the ABM partial securityregime.4
Althoughmanyof theseoriginalexpectationswere later"renegotiated"at the
bargainingtable and the Americanscame to followa more politicalapproach
to armscontrol,itwas theset selectedbytheU.S. government thatbecame the
regime'sconceptualbasis.5
Thus,theAmericansand Sovietssignedthe 1972 ABM treatyand createda
regimenotonlybecause thebalance ofpowerand technology had changed,nor
because of anydeep sharingof strategicculturalor politicalgoals,butbecause
theywereable to convergeon an Americanintellectualinnovationas thekeyto
advancingboth their irreconcilableinterestsand their shared interestof
3. See WesleyW. Posvar,"The New MeaningofArmsControl,"AirForceMagazine,June1963,

p. 38. For anotherstudyon intellectualsand nuclearweapons,see Roman Kolkowicz,"Intellectu-
als and the Nuclear DeterrenceSystem,"in Roman Kolkowicz,ed., The Logic of Nuclear Terror
(Boston: Allen & Unwin,1987),pp. 15-46.
4. Krasnerhas definedinternationalregimesas "sets of implicitor explicitprinciples,norms,
rulesand decision-making proceduresaroundwhichactors'expectationsconvergein a givenarea
ofinternational relations."Whethertheregimeconceptapplies to international security, however,
has been debated.On theone hand,Jervisand othershave arguedthattheanarchiccharacteristics
ofthisissue-areatendto lowerincentivesforcooperationand regimebuilding.On theotherhand,
Nyehas shownthatonce we taketheset of agreements, injunctions,and institutions
as forming not
just one comprehensivesecurityregimebut an incompletemosaic of partialsecurityregimes,the
notionof securityregimesmakes sense. These partialsecurityregimeshave led to the creationof
understandings aboutwhatit takesto negotiatesecurityagreements, whattypeof normsand rules
can be applied, and how. In some cases, theyhave helped to institutionalize rules of reciprocity,
limitcompetition,transfer information needed to complywiththe agreements,and enhancecrisis
stabilitybygeneratingstable expectations,includingthe expectationthatdiplomacyand negotia-
tionsshouldnotbe interrupted in the eventof international
crises.Taken together,and regardless
of theirvarious degrees of success, partial securityregimeshave amounted to a discreetyet
significanteffortto limitand controlautonomousactionin thesecurityarea. See StephenKrasner,
"StructuralCauses and Regime Consequences: Regimesas Intervening Variables,"in StephenD.
Krasner,ed., International Regimes(Ithaca, N.Y.: Cornell UniversityPress, 1987), p. 2; Robert
Jervis,"SecurityRegimes,"in Krasner,IntemationalRegimes,pp. 173-94; and JosephS. Nye,Jr.,
"Nuclear Learning,"International Organization 41 (Summer1987),pp. 371-402.
5. Robin Ranger,Armsand Politics,1958-1978:ArmsControlin a ChangingPoliticalContext
(Toronto:Macmillan,1979).
Nucleararmscontrol 103
avoidingnuclearwar.Once nucleararmscontrolbecame conventionaland was

routinizedin governmentpractices,however,the superpowerssaw it in their
interestto conformwitharmscontrolagreements.
The politicalselection,retention,and diffusion at nationaland international
levelsofnewconceptualunderstandings suggestan evolutionary approach.The
mutuallyreinforcing nationaland internationalarmscontrolgames-two-level
games,as it were-were structurednot onlyby fixedinterestsand powerbut
also bycommonunderstandings and practices.Such an evolutionary approach
is at odds with explanationsof internationalchange advanced by structural
realismand approachesbased on it.6
For example,Steve Weber has used a modifiedstructuralrealistanalysisto
shed lighton superpowercooperationduringthe Cold War. He argues that
"the conditionof nuclear deterrenceconstitutesa structuralchange in the
internationalpolitical system"and that,beginningin the early 1960s, the
superpowersbecame "socialized" to structuralchange and constraintsin
differentways.7Thus,in theABM case, a lackofsharedinterestsor compatible
visions of the long-termgoals to be achieved throughagreementled the
superpowersto learn different lessons,hence doomingthe detenteepisode of
in
the 1970s.Beginning the mid-1980s,however,forreasons thatWeber does
notfullyspecify,expectationsbegan to converge.
Weber's approachdiffers fromminein manyconceptualand practicalways.
Weber uses a conventionalstructuralanalysisto show how a new structural
organizingprinciple,mediatedbyideas, influencesconceptsof stateinterests.8
approachto showhow epistemiccommuni-
In contrast,I use a structurationist
tiesplaya role in establishinginterpretationsof interestsas practicesthathelp
6. See RobertD. Putnam,"Diplomacyand DomesticPolitics:The Logic ofTwo-LevelGames,"

InternationalOrganization42 (Summer 1988), p. 434. My approach is furtherdeveloped in
"CognitiveEvolution:A DynamicApproach forthe Studyof InternationalRelations and Their
Progress,"in Emanuel Adler and BeverlyCrawford,eds.,Progress inPostwarInternational
Relations
(New York: Columbia University Press, 1991), pp. 43-88. See also Emanuel Adler, ThePowerof
Ideology:The Questfor Technological Autonomyin Argentinaand Brazil (Berkeley:Universityof
CaliforniaPress, 1987). For otherapproachesdealingwiththe role of ideas in worldpolitics,see
JudithGoldstein,"Ideas, Institutions, and AmericanTrade Policy,"International Organization42
(Winter1988), pp. 179-217; JohnS. Odell, U.S. International Monetary Policy:Markets,Power,and
Ideas as Sourcesof Change (Princeton,N.J.: PrincetonUniversityPress, 1982); JohnG. Ruggie,
"International Regimes, Transactions,and Change: Embedded Liberalism in the Postwar
Economic Order," in Krasner,International Regimes,pp. 195-231; and Peter A. Hall, ed., The
Political Power of Economic Ideas: KeynesianismAcross Nations (Princeton,N.J.: Princeton
UniversityPress, 1990). For key structuralrealist studies, see Kenneth N. Waltz, Theoryof
InternationalPolitics(Reading, Mass.: Addison-Wesley,1979); and RobertGilpin,Warand Change
in WorldPolitics(New York: CambridgeUniversity Press,1981).
7. Steve Weber, "Realism, Detente and Nuclear Weapons," InternationalOrganization44
(Winter1990),p. 77.
8. The conventionalstructuralanalysisrefersto the approach outlinedby Waltz in Theoryof
InternationalPolitics.Accordingto Weber,the new structuralorganizingprinciple"followsfrom
jointcustodianship,a functionthatwas acquiredbytheUnited Statesand SovietUnion and which
fundamentally themfromotherstates."See Weber,"Realism,Detente and Nuclear
differentiates
Weapons," p. 77.
organize,structure, and coordinateinternational behavior.9 In Weber'stheoret-

ical world,structural realityconstrainsbehaviorand thenchallengesagentsto
coordinatetheirbehavior.In my theoreticalworld,agents coordinatetheir
behavioraccordingto commonpracticesthat structureand give meaningto
changinginternational reality.
The epistemiccommunity approachhas someclear"comparativeadvantages."
First,it allowsus to understandwhysuperpowercooperationwas conceptual-
ized via armscontrolin the firstplace. Second, it increasesour sensitivity to
domesticpoliticalfactors,especiallyto the notionthatwithineach national
actordifferent interpretations of the nationalinterestcompeteforthe shaping
of internationalagendas as well as internationalpractices.Third,in waysthat
allow forempiricalresearch,focusingon an epistemiccommunity drawsour
attentionto the impactof scientificknowledgeon international cooperation
processes.Fourth,theapproachshowsthatstatesbecomesocializednotonlyto
structural constraints butalso to each other'sunderstanding oftheworld.Fifth,
it helps us see that,in spite of or even because of superpower disagreement
overpoliticalinterestsand visions,thefactthatthe "Sovietsalso seem to have
understoodand sharedto some degreetheAmericanconcernsabout armsrace
and crisisinstabilitiesthatmightbe engenderedbyABM deployments"'0 was
notinconsequentialforpeacefulchange. The outcome of a lack of such shared
understanding mighthave been nuclearwar,ratherthanthetemporary demise
of detente.Sixth,commonepistemic understandings proved to be more lasting
thandisagreements over long-term goals. With the end of the Cold War, most
of the divergentlong-term goals are gone; what stillremains, though, are an
abundance ofweapons and the practice of arms control.
On the basis of my theoreticalapproach, I have devised an evolutionary
researchframework to describehow armscontrolideas were selectedfromthe
lot,carriedintothepowerstratum,and survivedto become realityin 1972.This
framework consistsoffivevariables:(1) unitsofvariation(the "geneticstuff," as
it were), consistingof tentativenew conceptualvariants,interpretations, and
meaningsbased on expectations,which circulatewithinthe academic and
politicalcommunities;(2) innovation,or the processes by whichintellectual
communitiespackage such unitsof variationand therebycreate a collective
understanding-as,in our case, aboutthenuclearpredicament;(3) selection, or
thepoliticalprocessesthatdeterminewhichpoliciesare effectively adopted by
the government; (4) diffusion,or the spread of expectations,values, and other
typesof ideas to othernations;and (5) unitsof effective modification, or the
9. Structurationtheory,as definedby Wendt,is "a relationalsolutionto the agent-structure

problem that conceptualizes agents and structuresas mutuallyconstitutedor co-determined
entities."See Alexander E. Wendt, "The Agent-Structure Problem in InternationalRelations
Theory,"InternationalOrganization41 (Summer1987), p. 350. See also AnthonyGiddens,Central
of CaliforniaPress,1979); and AnthonyGiddens,
Problemsin Social Theory(Berkeley:University
TheConstitutionofSociety:OutlineoftheTheoryofStructuration(Cambridge:PolityPress,1984).
10. Weber,"Realism,Detente and NuclearWeapons," p. 69.
patternednormativebehaviorof two or more statesthatresultsin part from

innovation,selection,and diffusion of expectations."In the followingsections
of myarticle,I relate the conceptof an epistemiccommunity to the issue of
nuclear strategyand offeran empiricaldescriptionof the variables. First,
however,let me statetheworkinghypothesesthatinformmyapproach:
(1) In a strategicrelationship,expectations are not derived in some
automaticand deterministic fashionfroma structuralconditionbut emerge
frommeaningsand understandings or "theories"thatshowa relationbetween
causes and effectsand createinterpretations of structure.
(2) When thereis no priorexperiencewiththe phenomenonat hand,such
as nuclear war, these theories are based on generalizable and abstract
propositionsand models.
(3) Because of the "scientific"and technicalnatureof these theories,they
are mostlikelyto be developedin academiccircles,givenvalidationthere,and
takento thepoliticalsystembyacademiccommunities.
(4) Throughdirectand indirectmeans,nationstransmitto each otherthe
contentoftheirtheories.
(5) This transferof meaningsand conceptsfromnation to nation allows
decision makers of differentnationalitiesand cultures to share historical
experience, epistemic criteria,and expectationsof proper action and to
rationallycalculatetheirchoicesaccordingto an intersubjective understanding
of thestructuralsituationand of each other'spayoffs.
(6) The sharingof strategicepistemiccriteriainduces decision makersto
behave accordingto thesecriteria,thushelpingto fulfill themin practice.
(7) Internationalcooperation emerges,changes, and decays along with
shared meanings and expectations and thus depends on whetheror not
decisionmakers willmake the rationalchoice to learn.
Knowledge,power,and nuclearstrategy
Epistemic communities
Both nationaland internationalepistemiccommunities mayplayrolesin the
evolutionof internationalcooperation in fields characterizedby technical
uncertaintyand complexity.But the political influence of transnational
epistemiccommunities,such as the Pugwashgroup in the securityfield,12is
mostlikelyto reston the transferfromthe internationalto the domesticscene
of the ideas that nationalscientistsand expertsraise at theirtransnational
meetings.Pugwash,forexample,can be best describedas whatJohnRuggie
calls a "switchboard"throughwhichconnectionsare "establishedand main-
11. This framework is partlyinspiredbyStephenToulmin'sdiscussioninHuman Understanding

(Princeton,N.J.:PrincetonUniversity Press,1972),pp. 122-23.
12. J.Rotblat,HistoryofthePugwashConferences (London: Taylor& Francis,1962).
tained,ratherthanbeinga depositoryof activity and authority."'3 The decisive

"customers,"then,frombothdomesticand international politicalperspectives,
nationalexperts,and ultimately,
are,first, nationalgovernments.
That is whywe need to pay moreattentionto the internationalinfluenceof
nationalepistemiccommunitiesin variousfields,includingarmscontrol.They
maybe able to affectinternational politicalprocessesand outcomesbybinding
presentand futuredecision makersto a set of concepts and meaningsthat
amountto a new interpretation of realityand also by becomingactorsin the
process of politicalselectionof theirown ideas. As internationalnegotiation
agendas are formulatedon the basis of these ideas and as negotiationand
diplomaticprocessesstartto takeplace, diplomatsact to advancenotonlya set
of policies but also a set of ideas. They "communicate,"as Michael Brenner
puts it,"to the leaders of otherstatestheir'theoretical'understanding about
the military-political characteristics of nuclear weapons in addition to signal-
lingtheirintenton theparticularissue at hand.... This exchangeofbeliefsand
imagesis especiallysignificant in the area of nuclearweaponswheretheissues
ofperception and deterrent psychology bulkso large."'14
The success of epistemic communities is historicallycontingent.Historical
contingency is afforded by the state of technology, the distribution of powerin
the international system, domestic political and administrative structures and
procedures, and political, economic, and military events. As the historical
contextchanges,theoriesor policy proposals that previouslydid not make
much sense to politiciansmay suddenlyacquire a political (perhaps even
urgent)meaning,thusbecomingpoliticallyviable.
The fieldof military strategy is propitiousforthe emergenceof an epistemic
community because, as WesleyPosvar has argued,strategyis "formulatedby
the cumulative action of subordinate and outlyingelements. Individual,
piecemeal decisions add togetherand build upon one another, and the
aggregatecomprisesthe strategicpostureof the nation."" Thus, althoughthe
governmentor state agencies are directlyin charge of developingnational
13. John G. Ruggie, "Changing Frameworksof InternationalCollective Behavior: On the

Complementarity Tendencies,"in Nazli Choucriand ThomasW. Robinson,eds.,
of Contradictory
Forecastingin InternationalRelations: Theory,Methods,Problems,Prospects(San Francisco:
Freeman,1978),p. 403.
14. Michael J. Brenner,"The Theoristas Actor,the Actor as Theorist:Strategyin the Nixon
Administration," StanfordJournalofInternational Studies7 (Spring1972),pp. 109-10.
15. See WesleyW. Posvar,"The ImpactofStrategyExpertiseon theNationalSecurityPolicyof
the United States,"PublicPolicy,vol. 13, 1964, p. 39. See also MargaretGowing,"An Old and
IntimateRelationship,"in Vernon Bogdanor,ed., Scienceand Politics(Oxford:ClarendonPress,
1984), p. 68. Accordingto Gowing,"The scientistsof the atomic era indeed became acutely
conscious of phenomenawhichrule politicallife: the conflictof desires and aims, the conflict
betweenthe interestsof different generations,the difficulty of calculatingconsequences. In the
years of their ascendencytheyproved that theywere not all-wise nor indeed all-wickedbut
human.Theycould changetheirmindswithdevastatingspeed. Theycould be bothwise
infinitely
and foolish,both myopicand far-sighted, both judicious and ridiculous,both clear-headed and
muddled.Theyturnedout to be, indeed remarkably likethepoliticians."
Nuclear armscontrol 107
mayalso be able to
outsidethe structureof government
strategies,institutions
performthisfunction.'6
The "imaginary" science of nuclear strategy

To prescribe an effectivecourse of action, a communityof strategists
requires a theorythat, as Charles Reynolds suggests,"show[s] a causal
relationshipbetween conditions,a governingprinciple,and a result. The
[political]actorthenhas thechoice,shouldhe so wish,to procuretheresultby
theconditions."'17For themostpart,strategists
fulfilling arriveat theirtheories
byinductiveprocesses,as theylook to thepast forinformation, understanding,
and inspiration.Butwhenthereis no priorexperience,as in thecase ofnuclear
war,strategicthinking mustdepend principally on theoriesthatseek to explain
human behavior on the basis of some generalizable propositions,such as
rationality,and on thebasis of abstractmodels,simulations,and games.Thus,
because the science of nuclearstrategyhas no empiricalreferencepointsand
data banks,itcannotbe falsifiedand is,in thissense,"imaginary.""8
This is especiallytrueof nuclear armscontrol,since theoryon thissubject
was developed in the absence of experiencewithnuclearwar and at a time
when therewas littleor no meaningfulexperiencewithnucleardisarmament
and armscontrol.Theorizingabout nucleararmscontrolrequiresassumptions
about how weapons would operate in various hypotheticalnuclear war
scenariosand whatmightor mightnotdeterconflicting powersfromlaunching
a surprisenuclear attack.These assumptionsmust rest partlyon a theoryof
internationalbehavior,arrivedat mainly on the basis of conjectures,assump-
tions,and nonscientific
expectations.
Armscontroltheory,therefore,cannotbe a priorivalid or true.Its validity
and poweras a conceptualbasis forinternationalcooperationwill depend on
the following:the temporaryexistenceamong the membersof an epistemic
communityof shared expectationsand of intersubjectiveand consensual
meanings,arrivedat via verbalcommunication; thedomesticpoliticalselection
of shared expectationsas practicesof governments, based on the fact that
expectationsmeet the decision makers'criteriaforadvancingnationalinter-
ests;and thefulfillmentoftheseexpectationsin practice,once theyare diffused
to other nations and become the epistemiccriteriaon which a strategic
relationshipbetween two or more nations is based. On all three levels-
epistemiccommunity, domesticpoliticalsystem,and international system-the
16. See Posvar,"The ImpactofStrategyExpertiseon theNationalSecurityPolicyoftheUnited

States,"p. 40. See also JohnGarnett,"StrategicStudiesand Its Assumptions,"in JohnBayliset al.,
eds., Contemporary (New York: Holmes & Meier, 1987).
Strategy
17. Charles Reynolds,ThePoliticsof War:A StudyoftheRationality of Violencein International
Relations(New York: St. Martin'sPress,1989),p. 28.
18. I owe thisinsightto HaywardAlker.On thenonscientific see Reynolds,The
basis ofstrategy,
Politicsof War; and Eugene B. Skolnikoff, Science, Technologyand AmericanForeignPolicy
(Cambridge,Mass.: MIT Press,1967),p. 110.
sharingof premisesand expectations,or "theories,"creates the "evidence"

thatconfirms thevalidityof norms.
Because the superpowersare engagedin a strategicsituationcharacterized
by the interdependenceof expectations,the sharingof deterrence,stability,
and arms controlexpectationsinduces policymakersto behave as if theyare
true,thusfulfilling thetheories'conditionsin practice.Progressin armscontrol
and the absence of war over timemaythenhelp reinforcethe beliefin stable
deterrenceand arms control expectations.In this manner,the science of
nuclearstrategy has an inputin creatingtherealityitis supposedto explainand
predict.
It also followsthatthe power of expectationsas an explanatoryvariableis
independentof the "instillation"of the expectationsin anysubjectivemind.If
armscontrolideas succeeded in transforming the practiceof deterrenceand
cooperation with the adversary,what matteredwas not that the personal
expectationsof thepeople involvedchangedin the courseof theircareers,nor
was it how preferenceswere firstproposed. Instead,what matteredwas how
the preferenceswere ultimatelydisposed throughthe presenceor absence of
social validation.19Furthermore, the realizationof communicableexpectations
and theories depends on whethertheir practical applications are readily
perceivedbypolicymakers. For example,Thomas Schelling'stheoryofinterde-
pendentdecision seems to lead to importantand strikingpoliticalproposals
and actions. And these proposals are strikingand importantnot merely
because of their contentbut also because they seem to be based on his
theories.20 Realitythusresultsfroma collectiveredefinition of problemsthat
carriesfirstthe clout of "scientificknowledge"and thenthe clout of political
and institutional power.
Knowledge relatingto arms controlcannot be separated fromvalues, for
whilevalues are backward-looking in theirfrequentappeal to past conductfor
justification,theyalso guide anticipatory and goal-directedbehaviorand thus
affectexpectations.Humanvalues affectaction byinfluencing our definitionof
a particularsituation and by directing our choice of relevant "facts" or
"interests."The interdependenceoffactsand values impliesa constantshifting
between empiricaland normativeelementsin decision making.21 Thus, arms
controlexpectationsbecame a politicalpractice,bothwithintheUnitedStates
and betweenthe superpowers,onlyafterarmscontrolacquired (1) domestic
politicalvalue; (2) foreignpolicyvalue (as a means of achievingforeignpolicy
goals); (3) instrumental internationalvalue (as a means of preventing nuclear
war); (4) intrinsicvalue (arisingfromthe reasoned assumptionsbehind the
19. Aaron Wildavsky,"ChoosingPreferencesbyConstructing Institutions:

A CulturalTheoryof
PreferenceFormation,"American PoliticalScienceReview81 (March 1987),p. 9.
20. Probably the most succinct and best expositionof Schelling's arms control theoryis
"Reciprocal Measures for Arms Stabilization,"in Donald G. Brennan, ed., Arms Control,
Disarmament, (New York: Brazillier,1961),pp. 167-86.
and NationalSecurity
21. Adler,"CognitiveEvolution,"p. 61.
theory);and (5) moral value (the consequentialistethical standardwherein

stable deterrenceand armscontrolare temporarily good foravoidingnuclear
war).
Unitsofvariation:arms controlexpectations
Since thedawnofthenuclearage, twointellectualcommunities and twosetsof

collectiveunderstandings, values, and visionshave had the crucial impacton
national securitypolicy making.22 Embedded in these two worldviewsare
differentexpectationsabout war,cooperationwiththe adversary, and technol-
ogy,themostimportant ofwhichis probablytheexpectationofnuclearwarand
ofitsoutcome.
Those who favoredarms controlshared a loose cause-and-effect mode of
reasoningwhich was sufficientto qualifythem as believers in a body of
"knowledge"thatwas distinctively "theirs."Because theyexpectedwar in the
and misperception,
nuclearage to breakout as a resultof crisisinstability as it
had in 1914,and predictedthatnuclearwar could neverbe won yetwould be
likelywithoutmeasuresto avoid it,theyplaced the greatestrelativevalue on
forcesand tacticsdesignedto preventa firststrike(ratherthanon an American
war-fightingcapability);put a premiumon cooperationwiththeadversary;and
promotedthe developmentof a highthresholdof nuclearweapons use. They
also predictedthattechnologywould not be able to create the "magicbullet"
withwhichto achieve superiority, but theyvalued technologicalchangesthat
mighthelp stabilizethenuclearbalance.23
On the basis of this particularinterpretationof war, cooperation,and
technology,the arms controllersdeveloped a distinctiveset of assumptions
22. These twocommunitieshave been the most,thoughcertainlynot the only,influential ones

froma policypointofviewin the nucleardebate. Also involvedwere communitiesthatstrovefor
nuclear abolition and total disarmamentand for solving the nuclear predicamentthrough
internationalinstitutionsand world government.See Robert A. Levine, The Arms Debate
(Cambridge,Mass.: HarvardUniversity Press,1963); ArthurHerzog,The War-PeaceEstablishment
(New York: Harper & Row, 1965); and Robert E. Osgood, The NuclearDilemma in American
StrategicThought(Boulder, Colo.: WestviewPress, 1988). For example, the peace movement,
institutionallyrepresentedbytheCommitteefora Sane NuclearPolicy(SANE), promoteda vision
of peace radicallydifferentfromthatof the armscontrollers.On some occasions,however,SANE
came to the help of armscontrol.And the peace movementalso playeda significant role in efforts
to set aside disarmamentideas and make roomforarmscontrolduringthe periodwhenscientists
who were generallyfavorableto disarmamentagreed neverthelessto supportarmscontrolas a
temporarymeasure. On SANE, see Milton S. Katz, Ban the Bomb: A Historyof SANE, the
Committee fora Sane NuclearPolicy,1957-1985 (New York: GreenwoodPress,1986).
23. Brodie augured this approach, arguing(for the wrongreasons, as it later emerged) that
nuclearweapons shouldbe used onlyto deterthe adversary.In whatis probablythe mostquoted
Brodie summarizedthe messageof his book: "Thus far,
sentencein the fieldof nationalsecurity,
the chiefpurpose of our militaryestablishmenthas been to win wars. From now on, its chief
purpose mustbe to avertthem.It can have no otherusefulpurpose." See BernardBrodie, The
AbsoluteWeapon(New York: HartcourtBrace, 1946), p. 76. See also Levine, TheArmsDebate,
p. 240.
about the reciprocalfearof surpriseattackand crisisstabilitythatbecame the

backboneof armscontrol.Interpreting the stateof theworldin 1960 as being
extremely dangerousbecause of the Cold War, theydoubtedthatthepolitical
and ideologicaldividebetweenthe superpowerswould be bridgedin the near
future,but theyneverthelessexpressedconfidencein theSovietabilityto learn
the secretsof deterrenceand armscontroland stressedthatconflicting powers
have commoninterests, whichprovidea basis forcooperation.The membersof
thiscommunity expectedgeneraldisarmamentto fail,althoughtheyreached
no consensusabout whetherdisarmamentmightbe an optionin the longrun.
Theyperceivedarmscontrolto be an integralpartof nationalsecuritypolicy,
believedthatarmscontrolcould includea varietyof unilateralmeasures,and
expectedthatin timearmscontrolmighthelp createa psychology ofpeace.
This set of viewswas challengedby an intellectualand politicalcommunity
thatexpectedwar to breakout because of a premeditatedattackbyan aspiring
worldhegemonicpower,as it had in 1939. The maincause-and-effect mode of
reasoningof thiscommunity was, accordingto Robert Jervis,thatwars "are
caused by statesfailingto develop the militarystrengthand crediblethreats
necessaryto dissuade othersfromchallengingthe statusquo. Furthermore,
threatsare most likelyto be believed when the state can carrythemout at
reasonablecost."24Thus,the community membersregardedtheuse of nuclear
weapons as quitepossibleand expectedthatiftherightmeasuresweretaken,a
nuclearwar could be won. Expressinga preferenceforcounterforce strategies,
they emphasized a less restrictedtype of deterrenceand maintainedthat
cooperationwiththe adversarywould lead to instability and was dangerous.
While theywere optimisticthatmilitarysuperiority and even victorycould be
achieved throughtechnologicalfixes,theyshared the view thattheirstrategy
would make nuclear war less likelyover the long run.25AlbertWohlstetter,
Herman Kahn, Richard Pipes, Eugene Rostow,Colin Gray,Fred Ikle, Keith
Payne,EdwardTeller,RichardPerle,and KennethAdelman,to mentionjust a
few,have,moreor less,held theabove set ofviews,whichalso was prevalentin
themilitary establishment.
Jervisis right in pointing out that the views which identifythe two
communities overlapand are partlycompatible.The overlapoverdeterrenceis
more apparent than real, however,because the two outlooks are based on
24. RobertJervis,"Arms Control,Stability,and Causes of War," Daedalus 120 (Winter1991),

p. 172.
25. William Borden's ThereWillBe No Time (New York: Macmillan,1946) also made early
referencesto theusabilityofnuclearweaponsin war,to theexpectationthatnuclearwarscould be
won, and to counterforcetargeting,active defenses, and intercontinentalballistic missiles
(ICBMs). Bordendid notexpectnuclearweapons to revolutionizestrategy; he expectedthemonly
to reinforcesome of the oldest and mostclassic elementsof strategy.Believingthatthe nuclear
adversarieswould spare each other'scitiesbecause of theirvulnerability,Borden expected that
nuclearweaponswouldbe used againstmilitary installations.See RobertJervis,"StrategicTheory:
What's New and What's True," in Kolkowicz,TheLogic ofNuclearTerror, p. 48; and Levine,The
ArmsDebate,p. 240.
differenttheoriesof war and therefore,in practice,theirpolicyprescriptions

contradicteach other."Arms control,"suggestsJervis,"stressesthe dangers
that arise when reassurancesand promises-especially the promise not to
strike-are either not made or are not believed; deterrencestresses the
dangersthatarise when threatsare absentor dismissed."26 It is thereforenot
surprisingthatthearmscontrolideas metwithchallengefromthoseadvocating
nuclearsuperiorityand thattheirchallengewas as manifestin the 1960s,when
theylargelyopposed a partial test ban treaty(PTBT) and supportedABM
deployment, as itwas in the 1980s,whentheyplaced theirprestigeon the line
in favor of the strategicdefense initiative(SDI). Only recently,with the
revolutionaryeventsin EasternEurope and the end of the Cold War, has this
challengebegunto weakenand theconceptualizationofthestrategicdebate to
change.
Intellectualinnovation
The arms control epistemic community
The arms control epistemic communitywas an informalassociation of
who forintellectual,ideological,and political
scientistsand civilianstrategists
reasonsadopted thearmscontrolapproach,in spiteofall theirdifferences over
national security
issues,includingarmscontrol itself.
Two subgroupsconstitutedthis community.One group of experts,whom
Robert Levine characterizedas "analyticalmiddle marginalists,"considered
the underlyingcause of internationalconflictto be the clash between the
interestsof nations as theypursue their separate goals. They stressedthe
ofdisarmament
futility and thedangersofmisperception and crisesthatgetout
of hand, and theyexpected that forthe foreseeablefuturethe worldwould
have to depend forstabilityon the possessionof nuclearweapons. The other
group,whom Levine called "moderate antiwarmarginalists,"believed that
armamentswere indeed a serious cause of internationaltension and that
thereforereducingweaponswouldreducetensions.But theyalso believedthat
the intensityof mutual grievancesas manifestedin the Cold War made a
transitionalperiod, wherein peace was guaranteed by nuclear deterrence,
unavoidable.27While they preferreddisarmamentto limited arms control
measures,thelatterwereseen as muchbetterthanan unlimitedand dangerous
nucleararmsrace.
These twogroupsconvergedintoan epistemiccommunity because, surpris-
ing as it mayseem, theywere in agreementabout the short-term advantages
and necessityof armscontroland therewas scarcelya memberof eithergroup
who did not concede the validityof the recommendationsof the other.As
and Causes ofWar," p. 173.

26. Jervis,"ArmsControl,Stability,
27. Levine,TheArmsDebate,pp. 61 and 89-90.
Posvarwroteat thetime,"One mightevenquestionwhethertheterm'schools'

as applied to these groups should be abandoned in favorof somethinglike
converging pointsofview."28
Certainlysome of the epistemiccommunity'smembersdid not get along
well, and sometimestherewere personal,career,and institutional conflicts.29
Manyofthearmscontrollers, havingmade originalintellectualcontributions in
theirown fieldsof expertiseand in nuclearstrategy, guardedtheirown ideas
and interpretations. But theirdiscussions,arguments,and mutualcriticisms
actually helped them in shaping a consensus over concepts, surmounting
interdisciplinarybarriers,and creatinga commonvocabulary.
Membersof thiscommunity kneweach otherwell: theyfrequently encoun-
tered each otheron televisionand in round-tableand debate performances,
oftenwere colleagues at the same or nearbyuniversities, and regularlymade
use ofeach other'swrittenand oral presentations. Thus,theylearnedfromone
another and togethergenerated the standardsby which theyverifiedthe
validityoftheirideas. In thisway,theycame to shareexpectationsthatsetthem
apartfromthe expertsand policymakers who had a strongfaithin technologi-
cal fixes,militarysuperiority,and "victory"in nuclearwar.Yet "admission"to
the armscontrolepistemiccommunity was based not onlyon the sharingof
epistemiccriteriabut also on an activededicationto "the cause," collectively
recognizedexpertise,and "the abilityto come up with new proposals and
arguments."30 The result,as one memberof thiscommunity put it,was a group
of people who had experiencesin commonand were supremelyconfidentin
theirabilityto deal rationallyand analyticallywithalmostanyproblem.3'
Several factors explain the abilityof these people to prevail in many
instances. To begin with, confidentin their abilityto use their scientific
knowledgeto solveproblems,32 armscontrollersused theirscientific prestigeto
gain legitimacyand authoritywithin the political system. They were one
community, yettheywere everywhere: dispersed among government bureaus,
research organizationsand laboratories,profitand nonprofitorganizations,
university researchcenters,and thinktanks. Such dispersionwas important
because theireffectiveness depended on theirrelativeautonomyfrompolitical
power,theirabilityto keep separatefromcurrentcriticalpressures,33 to retain
theirscientificintegrityand authority,and to continue to innovate. At the same
time,they were public figureswho required a certain power legitimation, and
thiswas achievedthroughpersonallinkswithpolicymakers or withindividuals
28. Posvar,"The New MeaningofArmsControl,"pp. 39-40.

29. Thomas Schelling,personalcommunication.
30. Herzog,The War-PeaceEstablishment, p. 4.
31. Donald F. Hornig,"Science and Governmentin theUSA," in HarveyBrooksand ChesterL.
Cooper, eds.,ScienceforPublicPolicy(New York: PergamonPress,1987),p. 20.
32. Ibid.
33. Posvar,"The Impact of StrategyExpertiseon the National SecurityPolicyof the United
States,"p. 49.
such as Paul Nitze,who linkedthe community withgovernment institutions,34

and throughthe factthattheirarmscontrolideas, afterbeing diffusedto the
politicalsystem,were in demandbyPresidentsEisenhowerand Kennedyand
theiradvisers.In otherwords,power legitimationarose fromthe creationby
the armscontrolepistemiccommunity of a politicallyviablealternativebothto
disarmamentand to military superiority.
A smallbut keygroupof civilianstrategists withinthe epistemiccommunity
had been affiliatedwiththe RAND Corporation,whichhelped turncivilian
strategyintoa profession.35 FromRAND, thestrategists absorbedan engineer-
ing approach and the methodologies,models, and assumptionsthat helped
them articulatetheir ideas on arms control.By 1957 or 1958, noted Fred
Kaplan, "a definitivestrategiccommunity had formedwithinRAND. It had
reached-by dint of small numbers,a commonoutlook,a [mostly]common
academicbackgroundin mathematicsand economics,and theforcefulness ofa
fewstrongpersonalities-a fairlytightconsensuson themajorissues,the most
solidlyheld of whichwas the not unlikelyprospectof a Sovietsurpriseattack
againstthe increasingly vulnerableStrategicAir Command."36 While to some
prominentRAND strategists the prospect of a Soviet surprise attack meanta
redoublingofefforts to achieve militarysuperiority through technological fixes,
to otherssuchas Thomas Schelling, Lewis Bohn, and Amrom Katz itmeant the
necessity of stabilizing mutual deterrence by means of arms control technical
measures.
Schellingspentpartsof 1957 and 1958 at RAND, wherehisworkinfluenced
and was influencedby theoristssuch as Bernard Brodie, Daniel Ellsberg,
Malcolm Hoag, Herman Kahn, WilliamKaufmann,and AlbertWohlstetter.37
Later, MortonHalperin,a Yale graduatestudent,attractedthe attentionof
Schelling,and they decided to collaborate. Donald Brennan, another key
memberof the community in its earlyyears,had workedfornineyearsat the
MIT Lincoln Laboratory,had come into direct contact with the RAND
and had developed a close relationshipwithKahn, withwhomhe
strategists,
went to workat the Hudson Institute.(By 1964, however,Brennan became
disenchantedwiththeidea ofbasingstrategicstability on controlling defensive
weaponsand crossedoverto the"other"community. The senseofbetrayalthat
was feltbyarmscontrollerssuggeststhestrength oftheirfeelingsofcommunal
cohesion.)
Schelling,Brennan,and othereconomistsand mathematiciansused game
theory-at that time a relativelynew methodologyinventedby John von
34. See StrobeTalbott,The Masterof theGame: Paul Nitzeand theNuclearPeace (New York:
AlfredA. Knopf,1988).
35. JenniferE. Sims, "The Developmentof AmericanArms ControlThought,1945-1960,"
Baltimore,Md., 1985,p. 284.
Ph.D. diss.,JohnsHopkinsUniversity,
36. Fred Kaplan, WizardsofArmageddon (New York: Simon& Schuster,1983),pp. 123-24.
Press,1960),p. vi.
ofConflict(London: OxfordUniversity
37. Thomas C. Schelling,TheStrategy
Neumann and Oscar Morgenstern38-tomake deductions and predictions

about deterrenceand armscontrol.Aided bythe rapid advance in computers,
game theoryallowedthestrategists to make all kindsof assumptions,construct
imaginary situationsand worlds,and deduce fromtheirmodelsthe answersto
the problemsposed by Soviet nuclear weapons.39"What I got out of game
theory,"said Schelling,"was more of a conceptual framework,a way of
organizingproblems.... It helps one to see ... whethersome outcomesare
betterthanothersforbothparties."40 This formalapproachcertainlyreflected
theascendanceofbehaviorismwithinacademia duringthe 1950s.
The rationalityassumption,the realistassumptionsabout the nature and
resolutionof conflicts,and the fearof communism, whichwas almostequal to
the fear of nuclearwar itself,were transmitted fromRAND to the political
structuresthatlaterformulatedarmscontrolpolicies. Strategists who worked
otherthanRAND and were trainedin a moreclassical and less
in institutions
behavioristapproach also made a contribution.Furthermore,intellectuals
fromvarious traditionshad become acquainted witharms controlideas and
methodsin the universities, in thinktanks,and in governmentalinstitutions,
and they played an active role, helping to train-through teaching and
publication-securityanalystswhojoined thecommunity lateron.41
The role ofthescientists whojoined therankswas as important as thatofthe
Theywere trainedmainlyin physicsand engineeringand had been
strategists.
involvedsince the 1940s in the makingof weapons and other technological
systems,such as air defenses.Theyhad participatedin government-sponsored
projects42and had become disenchantedabout the failureof disarmament
negotiationsas well as pessimisticabout technologicalsolutionsto the nuclear
weapons predicament.
The scientistswere the firstto participate in active discussions with
policymakerson arms control. As members of the Presidential Science
AdvisoryCommittee(PSAC), theyhad access to PresidentEisenhower,who
gave supportto "his" scientists.They also had firsthand experiencewitharms
control,as activeparticipantsin the testban treatytalks,whichwere used as a
38. See Johnvon Neumannand Oscar Morgenstern, Theoryof Games and EconomicBehavior,
2d ed. (Princeton,N.J.:PrincetonUniversityPress,1947).
39. The economiststended to treat strategicproblems in a formal,detached, and almost
apoliticalmanner.This approacheliciteda strongreactionand evenled somewritersto portraythe
nuclear strategistsas lesser human beings. See, for example, Anatol Rapoport, Strategyand
Conscience(New York: Harper & Row, 1964); Ralph Lapp, TheNewPriesthood:TheScientific Elite
and theUsesofPower(New York: Harper & Row, 1965); and IrvingL. Horowitz,The WarGame:
(New York: BallantineBooks, 1963).
StudiesoftheNew CivilianMilitarists
40. Thomas Schelling,citedbyHerzog in The War-PeaceEstablishment, p. 49.
41. Sims,"The DevelopmentofAmericanArmsControlThought,"p. 286.
42. Twentyor more physicistswere recruitedfromMIT to participatein ProjectCharles and
workon continentalair defense.The project'sproductwas a three-volume reportwhichconcluded
thata defenseof theUnited StatesagainstSovietbomberswas feasibleand shouldbe undertaken
promptly. ProjectVista dealtwiththenonnucleardefenseofEurope. See GreggHerken,Counsels
of War(New York: OxfordUniversity Press,1987),pp. 61, 63, and 65.
testinggroundfortheirideas-a paradigmaticcase, so to speak,thatcould be

applied to othercases lateron.43
Among the most prominentof these scientistswere Jerome Wiesner,
HerbertYork,Isador Rabi, JerroldZacharias,JamesFisk,BernardFeld, Paul
Doty, George Kistiakowsky, Hans Bethe, Eugene Rabinowitch,Jack Ruina,
George Rathjens,SpurgeonKeeny,WolfgangPanofsky,HarveyBrooks,and
JamesKillian,presidentof MIT. That manyof the scientistsand a majorityof
the strategists were fromeitherHarvardor MIT (and thusreferredto as the
"Charles Rivergang") was doublysignificant. First,it was easier forthemto
interacton a dailybasis, formally or informally exchangingideas. Second, the
institution withwhicha thinkerwas connectedhelped determinewhetherhis
or her ideas got a hearingwhere it matteredmost-at the WhiteHouse, the
Pentagon,or the State Department.44 And the factthatmanymembersof this
communityhad access to governmentsecrets-whether throughRAND,
PSAC, or Pentagonresearchagencies-was importantbecause it made them
feel like "insiders"and providedthemwithinformation theythoughtreliable.
The scientistsregardedtheirexperiencein handlingmajorsecurityprojects
as a model for organizingarms control.For instance,Wiesner, seeing no
incentiveforthe developmentof special seismicdetectorsbecause theywere
not needed in the developmentof nuclearweapons and because no bureau-
cracyor organizationexistedto createa politicalstakein them,spokeofa need
"to createa vestedinterestin armscontrol,to developa cadre ofpeople whose
full-timeoccupationis researchand developmenton means of arms control
and on theanalysisofthepoliticaland military problemsof armscontrol."45
Peace researchand conflictresolutionwere attainingacademic legitimacy
and were being fueled by a score of interdisciplinary programs,and many
academicians, strategists,and scientists found the Bulletin of the Atomic
Scientiststo be a perfect medium for disseminating the emerging ideas of
nucleararmscontrol.46
We shouldbe carefulnot to confusethe armscontrolepistemiccommunity
witha profession.47 The community cut across professions;its memberswere
involved armscontrolonlypart theirtime;theysharedresponsibility
in of for
theirdecisions with politicalactors; and theirethical standards did not arise
43. Experience in one of the major wartime laboratories,especially the MIT Radiation
Laboratoryand theLaboratoriesof theManhattanProject,or an apprenticeship withone or more
ofthemilitary "summerstudies"stillappears to be a usefulqualification advising.See
forscientific
Herken,Counselsof War,pp. 116-21; and Harold K. Jacobsonand Eric Stein,Diplomats,Scientists
and Politicians:The UnitedStatesand theNuclearTestBan Negotiations (Ann Arbor:University of
MichiganPress,1966).
44. BusinessWeek,Special Report,13 July1963,p. 75.
45. JeromeB. Wiesner,WhereScienceand PoliticsMeet(New York: McGraw-Hill,1965),p. 176.
46. Sims,"The DevelopmentofAmericanArmsControlThought,"pp. 303-4.
47. On the question of whethernuclear strategyis a profession,see E. Licklider,The Private
NuclearStrategists(Columbus:Ohio State University Press,1971),chap. 7. See also WesleyPosvar,
"StrategyExpertiseand National Security,"Ph.D. diss.,HarvardUniversity, Cambridge,Mass.,
1964.
froma professionalcode.48Indeed, this communitycan be described as a

and distinctcross-
functional,politicallydriven,ideologicallyself-contained,
sectionofthe"scientific estate."49
The innovationprocess
Between1955and 1960,a groupofciviliannuclearstrategists, someofwhom
were associated withthe RAND Corporation,50 gave a new meaningto the
conceptof war,based on the assumptionthatnucleardeterrencehad become
unstable and that a catastrophecould now occur against the wishes of the
adversarystates. These notionswere fueled by a stringof events,including
of U.S. strategicforces,51the
Wohlstetter'sinvestigationof the vulnerability
Killian Committee'spresentationof a report on "Meeting the Threat of
SurpriseAttack" in 1955,52the Soviet tests of an intercontinental ballistic
missile (ICBM) in August 1957, the launchingof the firstSoviet satellite
(SputnikI) intospace twomonthslater,and PresidentEisenhower'sestablish-
mentof the GaitherCommitteewhich,byrecommending an across-the-board
militarybuildup,53alarmedEisenhowerand made himmorereceptiveto arms
controlideas.
JamesKillianarguedthatSputnikI createda "crisisofconfidencethatswept
the countrylike a windblownforest fire. Overnightthere developed a
widespreadfear[unfounded,as it turnedout] thatthecountrylayat themercy
machineand thatour own government
of the Russian military and itsmilitary
armhad abruptlylostthepowerto defendthehomelanditself."54 Indeed, after
Sputnik,Wohlstetter'sstudies on the vulnerability of ICBMs caught the
attentionof some strategistsfrom RAND and elsewhere, most notably
Schelling,who opposed an indiscriminate
quest formilitary superiorityand the
48. Licklider,ThePrivateNuclearStrategists, pp. 119-22,130,and 135.

49. See Don K. Price, The ScientificEstate (Cambridge,Mass.: Belknap Press, 1965). See also
Robert Gilpin and ChristopherWright,eds., Scientistsand National PolicyMaking(New York:
ColumbiaUniversity Press,1964).
50. For a discussionof the group and a descriptionof the RAND Corporation,see Kaplan,
WizardsofArmageddon,especiallypp. 51-73. See also Paul Dickson, Think-Tanks(New York:
Atheneum,1971).
51. Wohlstetterwas a logician-mathematician at RAND. His studiespromptedtwo important
reports,R-266 and R-290, dealing with the vulnerability of bombers and the vulnerability of
ballisticmissiles,respectively.These studies expressedthe triumphof quantitativeeconomics-
orientedstudyat RAND. As Kaplan noted,"ThroughWohlstetter's ownpersonalinfluencewithin
RAND, vulnerability began to loom as the preoccupyingissue, the virtualobsession,of strategic
analysis."See Kaplan, WizardsofArmageddon, pp. 121-22.
52. Bundyrecentlycharacterizedthe 1955reportas "one ofthemostinfluential in thehistoryof
Americannuclearpolicy."See McGeorge Bundy,Dangerand Survival:ChoicesAbouttheBomb in
theFirstFiftyYears(New York: Random House, 1988),p. 325.
53. For a discussionof the GaitherCommitteereportand its influence,see MortonHalperin,
"The GaitherCommitteeand the PolicyProcess," WorldPolitics13 (April 1961), especiallypp.
382-83.
54. James R. Killian, Jr.,Sputnik,Scientistsand Eisenhower(Cambridge,Mass.: MIT Press,
1977), p. 7.
belief in the possibilityof winninga nuclear war on the ground that this
orientationmight,in fact,lead to such a war. So insteadof planninghow to
regain-by means of a massive technologicaland rearmamentleap-the
invulnerabilitythathad suddenlybeen lost,the strategists startedto concen-
trate on ideas about how to regain invulnerability by means of unilateral
stabilizingforcedeployments, as well as diplomacy.Thus, as noted by Strobe
Talbott,"They began layingthe conceptualfoundationsfornegotiationsthat
mightlimitthenumberofweaponswithwhichtheSovietUnion could carryout
a preemptiveattack.Thiswas theenterpriseofnucleararmscontrol."55
Some of the scientistswho had helped draftthe GaitherCommitteereport
also became disenchantedwith its recommendationsand with the trendin
U.S.-Soviet relations.Having become membersof PSAC,56theymade their
ideas knownto Eisenhower,who was receptiveand supportive.57 Spurgeon
Keeny,a memberof the armscontrolcommunity, remarked retrospectthat
in
the Gaither report represented"the high watermarkof the belief that a
technological solution could be found," a position the PSAC scientists
increasinglycame to see as unrealistic.58
Thus as the PSAC scientistsstartedto offerEisenhowerreliable technical
information withwhichhe could counterthoseopposinga testban treatywith
theSoviets,theyalso began to transmit to himand othergovernment a
officials
set of armscontrolassumptions,expectations,and values. They also proposed
the creationof a "peace agency" to embodyand empowerthem.President
Kennedy later created this institutionand called it the Arms Control and
DisarmamentAgency(ACDA). Accordingto SavilleDavis,
but a usefulone, to saythatthePresident[Eisen-
It is an oversimplification,
hower]now listenedprimarily to menwhose information and judgmentof
factindicatedthata safeguardedarms-control agreementwouldbe to the
advantageof the nationalinterestand securityof the United States,
whereasbeforethattimehe had listenedchieflyto menwho said such an
55. Talbott,TheMasteroftheGame,p. 70.

56. The membersof the firstPSAC were RobertBacher,WilliamBaker,Hoyd Berkner,Hans
Bethe, Detler Bronk,James Doolittle, James Fisk, Caryl Haskins, James R. Killian, George
Kistiakowsky,EdwinLand, Emanuel Piore,Edward Purcell,Isador Rabi, H. P. Robertson,Jerome
Wiesner,HerbertYork, and JerroldZacharias. On the PSAC, see Killian,Sputnik,Scientists and
Eisenhower, pp. 107-217.
57. In CounselsofWar,p. 116,Herkenquotes Eisenhower'sreactionto theidea ofa nuclearwar:
"You can't have thiskindofwar; therejust aren'tenoughbulldozersto scrape the bodies offthe
streets."
58. Spurgeon Keeny, cited by Herken in Counselsof War,p. 117. JeromeWiesner,another
communitymember,offeredthe followingview: "I firstbecame involvedin the disarmament
problemas a memberof thePresident'sScience AdvisoryCommittee.Priorto thatI had been the
StaffDirectorof the GaitherStudy.The conclusionsof thisstudyconvincedme thatit was not
reallyfeasibleto protecttheAmericanpeople ifa global nuclearwar occurred,and thatboththe
Russiansand ourselveswould sufferterribly. In fact,I became convincedthatas longas theSoviet
Union was prepared,as it seemed to be, to attemptto matchour military effort,therewas no help
of avoidingan enormousloss of lifein theeventof a majornuclearwar,regardless ofthemagnitude
ofourdefenseeffort" (emphasisadded). See Wiesner,WhereScienceand PoliticsMeet,p. 174.
agreementwould gravelydamage nationalsecurity.... Like mostshiftsin

policy,thisone willnotbe foundin documents.Policyis determinedbypo-
liticalmomentumsoperatingon the existingbalance offorcesin Washing-
ton.The arrivalofthenew groupof presidentialadvisersset up such a fresh
momentum.59
Manyof the strategists and scientistswho were drawnto armscontrolideas
met in 1958 at two conferencesin Geneva-one dealingwithsurpriseattack
and the otherwitha nucleartestban-to exchangeideas and tryto reach an
agreementwiththeirSovietcounterpartsovercooperativetechnicalmeasures
to avoid nuclearwar. Neitherconferenceproduced any such agreement.But
theSurpriseAttackConference,whichwas byfarthemostrelevantone froma
strategicarmscontrolperspective,togetherwitha preparatory conferenceheld
by Americansin Washingtonto formulatean AmericanpositionforGeneva
became a watershedin the consolidationof an emergentnucleararmscontrol
approach.60
The Surprise Attack Conference and its preparatoryconference,which
broughttogetherthe PSAC scientistsand the RAND strategiststo discuss
stable deterrenceand arms control,consolidatedthe ranksof the emerging
epistemiccommunity.It can even be suggestedthat at the SurpriseAttack
Conferencethe armscontrolepistemiccommunity was born.6"In anycase, at
that conferencean arms control seed was planted in the minds of many
True, the Sovietsreactedwith
reluctantSovietscientistsand politicalofficials.
dismayto the Americantechnicalapproach thatwas presented,arguingthat
deterrencewould only fuel the arms race. But the papers writtenfor the
conferencesuggestedto theAmericanand Sovietexpertshowa surpriseattack
could be preventedand how deterrencecould be stabilizedand managed by
means of armscontrol.Aftera week, accordingto JohanHolst, deliberation
changed into " 'cognitivenegotiations'aimed at exploringthe positionof the
adversary... and at conveyingtheWesternthoughtsand concerns."62 Thus,in
retrospect,one can agree withBernardBechhoeferthatthe talksservedas a
"catalystfor much of the serious rethinking of arms controland stabilized
deterrencewhichtookplace in theU.S. between1959 and 1961."63
59. Saville R. Davis, "Recent PolicyMaking in the United States Government,"in Brennan,
ArmsControl,Disarmament, and NationalSecurity, p. 385.
60. See JohanJ.Holst,"StrategicArmsControland Stability:A RetrospectiveLook," in Johan
J. Holst and WilliamSchneider,eds., Wy ABM. PolicyIssues in theMissileDefenseControversy
(New York: PergamonPress,1969),p. 282. On the testban conference,see Donald A. Strickland,
"Scientistsas Negotiators:The 1958 Geneva Conferenceof Experts,"MidwestJoumalofPolitical
Science13 (November1964),pp. 372-84.
62. See Holst, "StrategicArms Control and Stability,"p. 268. "The Westerners,"observed
Holst, "frequently voiced the expectationthattheyshould be able to convincethe Easternersby
logicalargument"(p. 263). RegardingtheSovietreactionto theAmericantechnicalapproach,see
p. 260.
63. See Bernard G. Bechhoefer,PostwarNegotiations forArms Control(Washington,D.C.:
BrookingsInstitution, pp. 261
1961),p. 475. See also Holst,"StrategicArmsControland Stability,"
and 282.
The ideas that the strategistsand scientiststook to the SurpriseAttack

Conferencewere a response to changes in technologyand weapons systems,
the balance of power between the superpowers,and American domestic
Yet theyalso were rigoroustheorieswhichhad been deduced froma
politics.64
set of hypothesesabout technologyand stabilityand which had evolved
togetherwiththeoriesabout strategicwar, limitedwar, and escalation,and
whose referencepointwas not past experiencebut onlyexpectationsof the
future. Being a disciplined creation derivingfrom artificialworlds and
speculationsin the strategists'minds,these theories could not have been
built-inor determinedonlybystructure.65
It would be naive to suggest,however,that the arms control epistemic
community creatednucleararmscontrolfromscratch.66 Indeed, a greatdeal of
thinking on armscontrolwas producedin the United States startingimmedi-
atelyafterWorld War II. (Leaving aside Hedley Bull's contribution to arms
controlideas,67it can be arguedthatarmscontrolwas an American invention,
as politicaleconomywas an inventionofBritishand Scottisheconomistsin the
eighteenthcentury.)
Veryfewpeople were as influentialin the intellectualdevelopmentof the
armscontrolapproachas Leo Szilard,whomNormanCousinsdescribedas "an
idea factory."68AlthoughSzilard remainedan outsiderto RAND and to the
halls of government,his indirectinfluencewas considerable because he
affectedthosewho had an impacton politicaldecisions.About a decade before
arms controlideas had gained prominence,Szilard anticipatedthe nuclear
stalemateand the use of mobileICBMs, called forintermediatestepsof force
reduction with differenttotals for differentsystems,considered that an
overwhelming was one of the
counterforcecapabilitywould cause instability,
firstpeople to oppose an ABM system,and pleaded fora no-first-usepolicyon
nuclearweapons. Some of Szilard's proposalswere unorthodoxand visionary
64. Sims,"The DevelopmentofAmericanArmsControlThought,"chap. 5.

65. In fact,themajorityofthe"brass" thoughtthesetheoriesto be quite"odd," since"it seemed
to follow[fromthe theories]thatSovietforcesshould perhapsnot even be targeted,and maybe
American cities should not be defended,even if a defense of populations some day became
ofour forcesmade us moretrigger-happy
feasible.For ifthevulnerability and was thusa dangerto
them,then by the same logic theirvulnerability was a danger to us: we should thereforenot
threatentheirstrategicforces,eitherdirectly,by targetingthem,or indirectly,by defendingour
neutralizingthem." See Marc Trachtenberg,"StrategicThought in
cities and thus effectively
America,1952-1966,"in Marc Trachtenberg, ed., TheDevelopment ofAmericanStrategicThought:
Writings 1961-1969and Retrospectives
on Strategy (New York: Garland,1988),p. 456.
66. See Freedman,TheEvolutionofNuclearStrategy, p. 197. The mostcomprehensivestudyto
date on the intellectualbasis of arms controlis Sims's "The Developmentof AmericanArms
ControlThought."This sectionbuildssubstantially on herstudy.
67. See HedleyBull, TheControloftheArmsRace (London: Weidenfield& Nicolson,1961); and
Hedley Bull,HedleyBull onArmsControl,selectedand introducedbyRobertO'Neill and David N.
Schwartz(New York: St. Martin'sPress,1987).
68. Norman Cousins, "Foreword," in Helen S. Hawkins,G. Allen Greb, and GertrudWeiss
Szilard, eds., Toward a Livable World:Leo Szilard and the Crusadefor NuclearArms Control
(Cambridge,Mass.: MIT Press,1987),p. xii.
and thusmade people thinkhardabout unorthodoxsolutions.For example,he

proposed the relocation of populations to eliminate large urban targets,
suggestedthatthe superpowersshould hold each other'scitieshostagewhile
forgoing war,and advancedthe idea of a nuclearfreezone in Europe. He also
envisionedan internationalcorps of scientistsand engineersto reportand
investigatenuclear violations,pioneered the idea of meetingsbetween U.S.
and Sovietscientists,and pushed the stilllargelyundefinedconceptof nuclear
deterrenceto thelimitbysuggesting to HermanKahn theidea ofa "doomsday
machine."69
Edward Shilsand WilliamFox, althoughto a lesserdegreethanSzilard,also
anticipatedthenucleararmscontrolapproachbya decade. Shilsmade a strong
case forintegrating bilateralarmscontrolnegotiationsand defenseplanning.
Fox linkedarmsreductionto the decrease of vulnerability to nuclear attack.
Concurrently,the Acheson-Lilienthalreport (1946) stressed the need for
unobtrusiveinspection,improvingcontinentalair defense,and, most impor-
tant, reducingincentivesfor surpriseattack. While Hans Bethe called for
superpowerbilateralnegotiations,a panel of consultantson disarmament, set
up in 1952 by Dean Acheson,called fornew waysof communicating withthe
leaders of the SovietUnion to discussthe armsrace.70At about the same time,
an emerging"realist"schoolof international relationschallengedtherational-
ist approach of the postwarscientists-who placed theirfaithin scientific
method, reason, and internationalorganizationand who expected world
disarmamentto occur once a world governmenthad been created7"-and
arguedinsteadthatthe nuclearpredicamenthad no moralsolutionand could
be mitigatedonlywiththehelp ofprudentialbehaviorand diplomacy.72
69. See BartonJ.Berenstein,"Introduction,"in Hawkins,Greb,and Szilard,Towarda Livable

World,pp. xvii-xxiv;and Leo Szilard, "Shall We Face the Facts? An Appeal fora Truce Not a
Peace," BulletinoftheAtomicScientists 5 (May 1949),pp. 269-73. See also Counselsof War,p. 206,
in whichHerkendiscussesthe doomsdaymachine,a fancifuldeviceto ensurepeace byblowingup
theworldas thepenaltyforaggression.
70. See Edward Shils, "AmericanPolicyand the Soviet PolicyRuling Group," Bulletinof the
AtomicScientists3 (September 1947), pp. 237-39; and WilliamT. R. Fox, "Atomic Energyand
InternationalControl,"in WilliamF. Ogburn,ed., Technology and Intemational Relations(Chicago:
Universityof Chicago Press, 1949), pp. 102-25. See also Sims, "The Developmentof American
Arms Control Thought," pp. 228 and 308. The Acheson-Lilienthalreport is cited as U.S.
DepartmentofState,Committeeon AtomicEnergy,A Reporton theIntemationalControlofAtomic
Energy,March 1946. In Danger and Survival,p. 159, Bundy describedthis reportas "the high
water-markof the Americaneffortto grapplewiththe issue of internationalcontrol."See also
McGeorge Bundy,"Early Thoughtson Controllingthe Nuclear Arms Race: A Report to the
SecretaryofState,January1953,"IntemationalSecurity 7 (Fall 1982),pp. 3-27.
71. See Robert Gilpin, American Scientistsand Nuclear Weapons Policy (Princeton,N.J.:
PrincetonUniversityPress, 1962), chap. 4. See also Bechhoefer,PostwarNegotiations forArms
Control,parts2 and 3.
72. See Hans J. Morgenthau,PoliticsAmongNations,5th ed. (New York: AlfredA. Knopf,
1978); Hans J. Morgenthau,Scientific Man VersusPowerPolitics(Chicago: University of Chicago
Press, 1946); Reinhold Niebuhr,"The Illusion of World Government,"Bulletinof theAtomic
Scientists5 (October 1949), p. 289; and Reinhold Niebuhr,The Structure ofNationsand Empires
(New York: Scribner's,1959).
Early work on arms controlintensifiedin the 1950s with the advent of

behaviorismand mathematicaltechniquesand the rise of the civilianstrate-
gists. For example, in 1951, David Inglis and Donald Flanders suggested
limitingthe nuclear capabilitiesof the adversaryto a particularstable level.
Three years later, James Newman suggestedthat inspectionneed not be
comprehensive, onlypractical.And soon after,assumingan expectedvulnera-
bilityto surpriseattack and arguingthat arms controlverification could be
undertakenby nationaltechnicalmeans, Inglisproposed a testban. In turn,
Hornell Hart showed that expected increases in Soviet offensivecapability
could counteranyexpectedimprovement in passiveand activedefenses.73
By 1956,some of the mostimportantdistinctions betweenarmscontroland
disarmamenthad been suggested.For example, Richard Meier had used game
theoryto deduce several propositionsabout arms control, includingthe
In
principleof a highnuclearthreshold. addition, a ballisticmissilebuilderfor
Convair had introducedthe idea of achievinginvulnerability by means of
second-strikeforces,an idea thatwas furtherdeveloped by Warren Amster,
who suggestedthe viabilityof mutual assured destruction. As Jennifer Sims
pointsout,Amster'ssuggestion was subsequently noted in an articleby C. W.
Sherwin,which was in turnquoted by Schelling.74
Beginningin 1957,thePugwashmeetingsthatbroughttogetherWesternand
East European scientiststo discuss disarmamentalso played an intellectual
role in the "invention"of the nucleararmscontrolapproach.75 The dominant
paradigmat the firstPugwashmeetingswas disarmament;armscontrolideas
were receivedwithskepticism.Said Schelling:"I was almostexpelled froma
Pugwash Conference[1960] because of the belief by the Soviets and some
Americansthat anyonewho thoughtabout armscontrolwasn't interestedin
disarmament."76 But discussionsbetweenWesternand Eastern scientistshad
some impacton armscontrolideas and policies. Accordingto J. Rotblat,"In
many instances the scientistsfrom the West received, for the firsttime,
reasoned objectionsto theirviews fromscientistsin the East and vice-versa.
This confrontation of ideas, of prejudices,and of causes of mistrust, was in
73. See David R. Inglis and Donald A. Flanders,"A Deal Before Midnight,"Bulletinof the
Atomic Scientists7 (October 1951), pp. 305-6 and 317; James R. Newman, "Toward Atomic
Agreement,"BulletinoftheAtomicScientists 10 (April 1954),pp. 121-22; David R. Inglis,"Ban the
H-Bomb and Favor theDefense,"BulletinoftheAtomicScientists 10 (November1954),pp. 353-56;
and Hornell Hart,"The Remedies Versus the Menace," Bulletinof theAtomicScientists10 (June
1954),pp. 197-205.
74. See R. L. Meier,"BeyondAtomicStalemate,"BulletinoftheAtomic 12 (May 1956),
Scientists
pp. 147-53; C. W. Sherwin,"SecuringPeace ThroughMilitaryTechnology,"BulletinoftheAtomic
Scientists12 (May 1956), pp. 159-64; WarrenAmster,"Design for Deterrence,"Bulletinof the
AtomicScientists12 (May 1956), pp. 164-65; and JenniferSims, "The American Approach to
NuclearArmsControl:A Retrospective," Daedalus 120 (Winter1991),p. 258.
75. See Rotblat,Historyof thePugwashConferences; JosephRotblat,"Movementsof Scientists
Against the Arms Race," in Joseph Rotblat, ed., Scientists,theArms Race and Disarmament
(London: Taylor& Francis,1982),pp. 115-57.
76. Thomas Schelling,citedbyHerzog in The War-PeaceEstablishment, p. 52.
itselfveryvaluable, as it gave an opportunity forbetterunderstanding of the

motivationof others and, in some cases, removed misunderstandings and
dispelledfears."77
The mainvalue ofthesediscussions,however,layin the factthatthe lessons
theygeneratedwere taken by Americanscientistsback to the U.S. political
system,where theybecame part of a collectiveunderstandingabout what
should be done to controlthe nuclear armsrace. For example,Walt Rostow
and JeromeWiesner,who played key roles in the Kennedy administration,
discussed mattersof internationalsecuritywith the Soviets at a Moscow
meeting of Pugwash in December 1960; and they came back from the
disarmament meetingwiththefeelingthattheSovietUnion mightbe readyfor
action in arms control.78 Pugwashmeetingsthat dealt witha test ban treaty
playeda similar,ifnotmoreimportant, role in developingsomeofthetechnical
bases forarmscontrol.
Severaldevelopmentscloserto thehallsofgovernment duringtheformative
yearsof nucleararmscontrolideas also had some influenceon theirevolution.
First,the Baruch plan of 1946 lefta legacythatwas not overlookedby arms
Second,as Simsnotes,theinconclusivedisarmament
controllers.79 negotiations
of the 1950s "had a great impact on the evolutionof ideas about weapons
control.The ups and downsofthenegotiationsthemselvesinspiredcontroversy
and commentary:the Open Skies proposal [1955] stimulatedthinkingabout
limitedagreementsand bilateral negotiations;preparationfor the Surprise
AttackConferencegeneratedstudies-particularlyat RAND-of the techni-
cal requirementsof strategicstability;'Atoms for Peace' [1953] inspired
considerationof controlledinformation sharingas a stabilizinginstrument."80
Third,thecenterofinterestin newideas ofarmscontrolshiftedfromtheState
Department,whereJohnFosterDulles was no enthusiastof armscontrol,8" to
the WhiteHouse; and PresidentEisenhower,withthe aid of his armscontrol
assistantHarold Stassenand ofNelson Rockefeller,"discovered"nucleararms
controlbeforetheacademicstrategists did.82And,fourth, theidea ofa testban
treatycontinuedto evolve throughoutthe 1950s, and negotiationswiththe
77. Rotblat,HistoryofthePugwashConferences, pp. 14-15.

78. ArthurM. Schlesinger,Jr.,A ThousandDays: JohnF. Kennedyin theWhiteHouse (Boston:
HoughtonMifflin, 1965),p. 301.
79. See Bechhoefer,PostwarNegotiations forArms Control,part 2; and Bundy,Danger and
Survival,chap. 4. In June 1946, Bernard Baruch, the U.S. negotiatorat the United Nations,
proposed the followingplan for the internationalcontrolof nuclear energy:the United States
would place its entireatomicweapons productionunder an internationalauthority,and other
nationswould be barredfromproducingnuclearweapons and would allow theirfacilitiesto be
placed under the internationalauthority.The plan also promotedthe peaceful use of nuclear
energy.
81. Dulles supportedarms controlonly on the conditionthat the price was rightand that
Americanprestigeabroad wouldbe enhanced.
82. By developingplans forair reconnaissance,a nuclearfreeze,nucleararmsreductions,and a
set ofobjectivesforarmscontrol,Stassenauguredthe "goldenera of armscontrol."
Sovietson sucha ban werein factinitiatedbeforethecrucialeventsof 1958-60.

Most of the policyproposals,however,were stillin the disarmamentmode,
withthepossibleexceptionof Open Skies.
Creativityor innovationdoes notinvolvenew ideas but newcombinationsof
ideas-for example,thatsurpriseattackand preventing nuclearaccidentsmust
be considered togetheror that arms controlshould not be separated from
militaryand defense policy. For the firsttime, arms control theorywas
crucial idea of the interdependenceof expecta-
articulatingthe strategically
tions.Schelling'snotion-"he thinkswe think... he'll attack;so he thinkswe
shall, so he will, so we must"83 was in the air at the Surprise Attack
Conference.Thus,as academiciansand policymakers graduallybegan to reach
a commonunderstanding aboutwar and peace, weapons and negotiations, and
conflictand cooperation,theirapproach to internationalnegotiationsshifted
frommeasures designed to remove nuclear weapons fromworld affairsto
measuresdesignedto make theirpresencemoretolerable.84
Gerald Holton,editorofDaedalus, thejournalof theAmericanAcademyof
Artsand Sciences,arguedthatby 1959 an "enormouslyrefinedartand science
of controllingwar" had become "critical." The academy thus convened a
summersession to deal with arms control,and this session resultedin the
participationofmorethanfifty individualsand thepublicationofseveralbooks
thatbecame landmarksin theintellectualhistory of armscontrol.85In 1960,the
academyalso initiatedtwo projectsto studyand formulatethe ideas of arms
control,therebycreatingan opportunity foracademicians,publicofficials, and
journalists-many of whom had come to the arms controlworldview-to
strengthen theirsharedunderstanding about armscontrol.
In thefallof 1960,Daedalus publisheda special issue thatdistilledthe main
insightsand currentsof thoughton armscontroland soon became knownas
"the Bible of armscontrol."Withthispublication,nucleararmscontrolcame
of age.86 Intellectual consolidation and refinementof the arms control
approach continued, however,within the frameworkof a Harvard-MIT
83. Schelling,TheStrategy p. 207.

ofConflict,
84. Freedman,TheEvolutionofNuclearStrategy, p. 199.
85. See Thomas Schellingand MortonHalperin,Strategy and ArmsControl,2d ed. (New York:
Pergamon-Brassey, 1985); Brennan,ArmsControl,Disarmament, and NationalSecurity; Bull, The
ControloftheArmsRace; Louis Henkin,ed.,ArmsControl:IssuesforthePublic(Englewood Cliffs,
N.J.:Prentice-Hall,1961); David H. Frisch,ed., ArmsReduction:Programand Issues (New York:
TwentiethCentury,1961); and ErnstW. Lefever,ed.,ArmsandArmsControl(New York: Praeger,
1962). See also Robin Ranger,"The Four Bibles ofArmsControl,"in Susan J.Shepard,ed.,Books
and thePursuitofAmericanForeignPolicy,special issue ofBook Forum,vol. 6, 1984,pp. 416-32.
86. See Gerald Holton,ed.,ArmsControl,special issue ofDaedalus, publishedin Fall 1960 and
issued as vol. 89, no. 4, of the Proceedingsof the AmericanAcademyof Artsand Sciences. The
impactof the special Daedalus issue on arms controlwas so great that a revisedand enlarged
editionwas rushedintoprintin early1961.Afterthisprinting of20,000copies was sold out,itwent
into a second printing.Such sales were unprecedentedfor a specialized workof thiskind. See
Ranger,"The Four Bibles ofArmsControl,"pp. 417-18.
seminaron arms controlthat met more or less continuouslythroughoutthe

1960s.
Reflectingon theentireperiod,Schellingwrote:
That 15-yearperiodfrom1957 to 1972 is a remarkablestoryof intellectual
achievementtransformed intopolicy.... Three books appeared in 1961
thatepitomizedan emergingconsensuson whatstrategicarmscontrol
shouldbe about. Each was a groupeffort, and each stimulateddiscussion
evenwhilebeingwritten.... A numberof participantsin theHarvard-MIT
seminartookpositionsin theKennedyWhiteHouse, Departmentof State
and Departmentof Defense; othersfromRAND and elsewhere,who had
been partof thisintellectualmovement,movedintothegovernment as well.
thatthoseideas became thebasis forU.S.
So it is notcompletelysurprising
policyand were ultimately implementedin theABM treaty.87
Politicalselection
A politicalselectionprocessdeterminedthe epistemiccommunity's successin
translatingits theoriesinto policies. The policymaker,in principleat least,
served as judge, jury, and, if necessary,executionerover the professional
output of strategictheories. Many, though not all, of the community's
aspirationswere satisfiedonlythroughpolicydecisions.88It was notnecessarily
the best-fittedideas thatwere selected and turnedintopolicies,however,but
thosewhichbest fitthe interestsof policymakers and whichpassed the testof
domesticpolitics.This is whythe epistemiccommunity had to persuade other
actorsin the systemof thevalidityof itsideas. The keywas not onlyinventing
new conceptsbutraisingthemto newheightsofpublicawareness.89
87. See Thomas C. Schelling,"What Went Wrong with Arms Control,"ForeignAffairs65

(Winter 1985-86), p. 223. Schellinghad made the followingstatementin 1969: "Whateverthe
prospectsforsuccessfulnegotiationswiththe SovietUnion duringthe comingmonthsand years,
on the subjectof strategicweapons,therecould notbe a greatercontrastbetweenthe seriousand
businesslikeprospectsfor realisticnegotiationsin 1969 and all the fantasyand pretenseabout
'generaland completedisarmament'thatcharacterizedthebeginningofour decade.... We think
differentlynow,partlybecause technologicalprogressobligesus to butpartlybecause we havebeen
thinkingand talkingand writingand holdinghearingsand preparingbudgetjustificationsand
negotiating withallies and enemiesduringthispast decade.... [The] concernwithvulnerability of
retaliatorysystems... became the primarycriterionforthe selectionof a weapons systemitself
[and] it has become also the primarycriterionforthe designof an armsagreement betweenthe
United States and the USSR.... The problemof 'accidentalwar' was recognizedto be primarily
one of information and decision ratherthan sheer mechanicalaccident,[and the] traditionof
non-use,the somewhatselfconfirming expectationsof non-use,growsstrongereveryyear." See
testimony of Thomas Schelling,in U.S. Congress,House Committeeon ForeignAffairs,Strategy
and Science: Towarda NationalSecurity Policyforthe1970s: HearingsBeforetheSubcommittee on
National SecurityPolicyand ScientificDevelopments,91st Congress,1st sess., March 1969, pp.
123-24.
88. Colin S. Gray, StrategicStudiesand Public Policy: The AmericanExperience(Lexington:
University PressofKentucky,1982),p. 26.
89. Accordingto Gray,"Contemporaryarms-control theorywas an inventionof the strategic
studiescommunity in theperiod 1958-60." See ibid.,p. 72.
The selectionprocess startedunderEisenhower,and by the timeKennedy

entered officecertainsignificant trendswere under way. First,because the
United States was activelyengaged in test ban negotiationswiththe Soviet
Union,Kennedyinheriteda framework fornegotiationsand a policyon which
to build. Second, because Eisenhowerhad been listeningsince 1958 mainlyto
hisPSAC scientists, he helpedlegitimizetheemergingarmscontrolconceptsas
thefocusofthepolicydebate. Kennedy,then,was awareofthepositionsofthe
various governmentagencies with an interestin arms controland inherited
personneland organizationalstructures to deal withit,includingthe scientific
adviser,PSAC, and an interagency groupcalled the Committeeof Principals.90
The Eisenhowerlegacyis insufficient, however,to explainwhyideas ofstable
deterrence and arms control were accepted so rapidly by the Kennedy
administration. "Indeed," quipped E. Licklider,"the ideas no soonerbecame
public than theyseemed to become governmentalpolicy."91Kennedy,who
enjoyedclose connectionsto manysecurityintellectualsand had an innovative
orientation,92 played an importantrole. Certainly,presidentialleadershipwas
crucialin thisarea because armscontrolideas were relatively new and seemed
counterintuitive. They were rejectedby some prominentstrategists, such as
Wohlstetter, and some prominentscientists, suchas Teller;had littlesupportin
the military;and got onlya mixedreview Congress.Moreover,because the
in
conceptsof disarmamentand armscontrolhad been used interchangeably, the
latterwas boundto be misrepresented and misunderstood as pacifist,
idealistic,
and disarmament-driven. And since armscontrolideas lacked provisionsfor
completeand effective theywere not likelyto be accepted readily
verification,
byparts of the bureaucracy,Congress, and theAmericanpublic.93
Yet thecontentand qualityoftheideas gave thembroad politicalappeal and
helped Kennedybuild a politicalcoalitionon theirbehalf.Indeed, disarma-
mentproponents,iftheyso wished,could regardarmscontrolideas as a first
step toward disarmament.And conservativescould be reassured that arms
controlcould meanmore,ratherthanfewer,weapons.Thus,in everyimportant
aspect, as Colin Gray observed,the armscontrolcommunity "foundthatthe
Kennedyadministration and itsbriefera offereda permissiveenvironment in
whichitcould exerciseinfluence."94
90. Firestone notes that the Committeeof Principalswas "a high-levelinteragencygroup

designedto coordinateand ultimatelyratifyarms controlpolicy.Founded in August 1958, the
committeewas initiallycomposed of the secretaryof state as chairman,the secretaryof defense,
the chairman of the AEC [Atomic Energy Commission],the directorof the CIA [Central
IntelligenceAgency],and thepresident'sspecial adviseron science and technology."See Bernard
JohnF. Kennedyand theSovietUnion(Westport,Conn.:
J.Firestone,TheQuestforNuclearStability:
GreenwoodPress,1982),p. 76.
p. 155.
91. Licklider,ThePrivateNuclearStrategists,
A ThousandDays, p. 104; and WilliamW. Kaufmann,TheMcNamara Strategy
92. Schlesinger,
(New York: Harper & Row, 1964),p. 1.
93. Firestone,TheQuestforNuclearStability,p. 153.
Studiesand PublicPolicy,p. 97.
94. Gray,Strategic
The creationof ACDA provideda home forthe armscontrolcommunity, a

politicalvoice foritstheoreticalideas, and a laboratory
wherepolicyideas were
firstdesignedand developed. From 1961 onward,the ACDA memberstook
partin mostofthepoliticaldeliberationson nationalsecurityand armscontrol
policy;still,theagencyremaineda weak bureaucraticplayer.95 Being entrusted
withthemissionof controlling the armsrace withoutundermining themilitary
balance, the agency'spower depended on the level of externalthreat,which
usuallywas high,and on theforcethatthePresidentpersonallyputbehindthe
armscontrolprocess,whichwas notconsistently high.96
More significant, however,was the factthat manyof the most prominent
membersof the arms control communityhad taken key positions in the
Kennedyadministration. At the WhiteHouse, McGeorge Bundy(formerly of
Harvard) became the adviser of the National SecurityCouncil (NSC) and
brought along Carl Kaysen and Walt Rostow. Jerome Wiesner became
scientificadviser,and James Killian, George Kistiakowsky, Paul Doty, and
HarveyBrooks became membersof PSAC. Abram Chayes went to the State
Department,whileHerbertYork,JackRuina, and George Rathjensworkedat
the Pentagon in research and weapons development(with Rathjens also
workingfor ACDA). Key positionsin the Pentagonwere filledby Defense
SecretaryRobertMcNamara's "whizkids": RoswellGilpatrick,HenryRowen,
and Charles Hitch. And the officeof assistant secretaryof defense for
internationalsecurityaffairs(ISA) was entrustedto Paul Nitze,who was then
onlyat thebeginningofa longcareeras the"masterofthearmscontrolgame."
These institutional and recruitment developmentshelped create a networkof
relationsbetweenpoliticalelitesand thearmscontrollers. The armscontrollers
also affectedpoliticalelites indirectlythroughop-ed articlesin TheNew York
Timesand Washington Post and articlesinForeignAffairs.
While manyof the community scientistsaffectedthe policymaking process
throughthe PSAC and ACDA, the strategistshad effectively taken over the
ISA office.Nitze offeredSchellinga job as his armscontroldeputy;Schelling
declined but recommendedthe appointmentof his Harvard colleague John
McNaughton, who was a professorof law. According to Fred Kaplan,
McNaughtontoldSchellingthathe knewnothingabout armscontrol,to which
Schellingrepliedthathe would teach himall therewas to know.97 Apparently
he did: McNaughtonwas appointedassistantsecretaryof defenseforISA in
95. See Steve Weber and SidneyDrell, "Attemptsto Regulate MilitaryActivitiesin Space," in
AlexanderL. George,PhilipJ.Farley,and AlexanderDallin, eds., U.S.-SovietSecurity
Cooperation:
Achievements,Failures,Lessons (New York: OxfordUniversity
Press,1988),p. 388. On ACDA, see
Paul F. Walker,"The U.S. Arms Controland DisarmamentAgency:Policy-Makingin Strategic
ArmsLimitations,"Ph.D. diss.,MIT, Cambridge,Mass., 1978.
96. HerbertF. York,MakingWeapons,TalkingPeace: A Physicist's Odyssey fromHiroshimato
Geneva(New York: Basic Books, 1987),p. 119.
97. Kaplan, WizardsofArmageddon, pp. 332-33.
1963,became McNamara's "generalcounselorand chiefaide on armscontrol,"

and had a hand in persuadingthePentagonofthemeritsofthe PTBT.98
The abilityof the epistemiccommunitymembersto persuade and forge
alliances withpolicymakers was crucialfortheirultimatesuccess. As soon as
Kennedy became president,Wiesner "started to educate Kennedy on the
limitationsof the ABM," and Ruina and York were influentialin persuading
Kennedyto bag the Nike-Zeus antiballisticmissile.99A committeeon the level
of forceswhichACDA had appointed and whichincluded Wiesner,Bethe,
Doty, and Henry Kissingerapparentlyparticipatedin preparingthe 1965
defensebudgetand helped McNamara place a limiton Minutemanforces.100
On the other hand, some policymakersused the arms control expertsto
rationalizeand explaintheiractions,especiallyaftertheCuban missilecrisisof
October1962 demonstratedtheneed forarmscontrol.
The hotlineagreementofJune1963-the agreementto installa teletypeline
betweenWashingtonand Moscow to serve as an emergencycommunication
linkin case of crisisand to preventunintendednuclearwar-illustrateshow
various persons,institutions, and factorsmentionedabove played a role in
determiningarms control outcomes. The idea of a hotline had firstbeen
suggestedbySchellingduringa meetingof a taskforceset up byJohnMcCloy,
Kennedy'sdisarmamentadviser.Schellingthentold HenryOwen of the State
Departmentabout it, and he in turnpassed the idea along to Gerard Smith,
head of the State Department'spolicyplanningstaff.When journalistJess
Gorkin,who editedParade,theBostonGlobe's Sundaymagazine,pickedup the
idea, he wrotearticlesabout it and sentopen lettersto Eisenhowerand Soviet
PremierNikitaKhrushchevoutlininghis proposals.He laterwas able to talk
brieflywith Khrushchevabout the matter and tried to sell the idea to
presidentialcandidates Kennedy and Nixon. He thereforecreated public
awarenessthatled otherindividuals,periodicals,and expertsto beginadvocat-
ingtheapproach.
But itwas Owen, workingwithingovernment circles,who led the American
governmentin April 1962 to finallyaccept the idea. The Soviets reacted
favorablyto the Americanproposalbut tied theiragreementto othergeneral
disarmamentproposals.The idea mighthave died of "linkagedisease" had it
notbeen thatthe Cuban missilecrisis,in a vividand practicalway,showedthe
superpowerleaderswhatSchellinghad in mind.Severalmonthslater,theidea
had turnedintoreality.
With characteristicmodesty,Schellinglater remarkedthat "it was not a
question of inventingthe hotline, but simply of realizing that such an
A ThousandDays, p. 494.
98. Schlesinger,
99. See York,MakingWeapons,TalkingPeace, pp. 222-26; and Kaplan, WizardsofArmageddon,
p. 345.
100. Desmond Ball, Politicsand Force Levels: The StrategicMissileProgramof the Kennedy
Administration of CaliforniaPress,1980),pp. 82-85.
(Berkeley:University
elementarymeans of communicationdid not alreadyexist."The hotlineidea

arose fromSchelling'sinterpretation of a structuralcondition,an interpreta-
tion that differedmarkedlyfromthe "hard-liner"view that "Washingtonis
ideologicallyclose enough to Moscow withoutmakingthe White House a
branchofficeoftheKremlin."101
The transitionfromdisarmamentto arms controlmeant that the bureau-
cracyhad to go througha processof adjustmentand conceptualevolution.In
1961,Americansand Sovietswerestillformally negotiating"totaland universal
disarmament"withinthe frameworkof the Eighteen-NationDisarmament
Committeein Geneva. In practice, however,they had already started to
negotiatearmscontrolbut chose to referto it as "first-stagedisarmament."'102
Also duringthistransitionperiod,Kennedyand the armscontrolcommunity
began to framePTBT negotiationsaroundthe conceptof nucleardeterrence
stabilityand around expectationsthata technicalagreementover testscould
amountto a firststepin thethawingoftheCold War.
By the end of the Kennedyyears,armscontrolhad become an irreversible
factorin the domesticpoliticalgame and a keyconsiderationas agreements
werenegotiatedand evenas newweapons systemswerecontemplated.Indeed,
thebureaucracyand the armscontrollerswho participatedin thebureaucratic
process were constantlyinvolvedin preparinga position for negotiation,
defendingan existingagreement,or carrying out bureaucraticguerrillawarfare
againstmilitary Thus,theinstitutionalization
programs.l03 ofarmscontrolideas
guaranteedthatas bureaucraticbattlesflaredup, institutions and individuals
who carriedthe armscontrolideas would throwtheirweightin favorof their
selection by the Presidentand his closest advisers.Domestic politics then
became the arena wherenationalsecurityand worldorderideas were raised,
legitimated,and selected as policychoices and where theywere testedonce
theybecame nationalpoliciesand had international effects.
"Educating"and persuadingMcNamarawerevitalstepsin themovementof
armscontrolideas frominnovationto politicalaction. Like Kennedy,McNa-
marawas wellsuitedforarmscontrolexpectationsand values.Withhisanalytic
commandofthenuclearweaponsproblemand hismanagerialand engineering
instinctto do somethingaboutan irrationalsituation,l04
McNamaracame to see
in armscontrolthe rationalalternativeto nuclearwar. He trustedthe RAND
strategists,whose techniquesand analyticstylehe shared,and was instrumen-
tal in protectingthe community'smembersfromthewrathof the JointChiefs
of Staff.
101. My discussionof the hotlineidea is based on WilliamL. Ury'sBeyondtheHotline:How

CrisisControlCan Prevent NuclearWar(Boston: HoughtonMifflin, 1985),pp. 142 and 144,and on
interviews.
102. See Talbott,TheMasteroftheGame, p. 79; and Schlesinger,
103. See Adler,ThePowerofIdeology.See also WilliamL. Hyland,"InstitutionalImpediments,"
in RichardBurt,ed.,ArmsControland Defenseinthe1980s (Boulder,Colo.: WestviewPress,1982),
p. 101.
104. Schlesinger,
Once McNamara was persuaded thattherewas no technicalsolutionto the

armsrace105and thatit had to be limited,he became a powerhouseforarms
controlideas-an epistemiccommunity's dream.He workedhardeveryyearat
budget time to prevent ABM deploymentand helped persuade Lyndon
Johnsonto tryarmscontrolwiththe SovietsbeforedeployingABMs. To make
sure Johnsonwould not change his mindon ABMs, McNamara assembledin
the Oval Officeall past and presentPSAC chairmen,who made clear their
oppositionto it.And in a speechin San Franciscoin 1967,he toldtheAmerican
people about the "mad momentumof the arms race." Indeed, Johnson's
electionin 1964 proveda victoryforthosewho opposed ABM systems,ifonly
because it ensured that McNamara would continueas defensesecretaryfor
fourmoreyears.106At the 1967Glassborosummitmeeting,McNamaralectured
Soviet PremierAleksei Kosyginabout "the action-reactionphenomenon,"
explainingthatshould the Sovietsproceed withthe deploymentof theirown
ABM system(nicknamedGalosh), all the United States had to do, really,was
to increaseitsoffensiveforces,thusneutralizingSovietdefensesbutfuelingthe
armsrace evenmore.107
Ideas about controllingABMs had startedin the late 1950s,roughlyat the
same timeDefense SecretaryNeil McElroy,undertheinfluenceofSputnikand
the GaitherCommitteereport,authorizedthe armyto develop an operational
ABM systemcalled Nike-Zeus.108 Also at thattime,McElroycreatedwithinthe
Pentagonthe AdvancedResearch ProjectsAgency(ARPA) and the Director-
ate of Defense Research and Engineering(DDR&E). The scientistsat these
institutionsplayedan important role in thedevelopmentofan AmericanABM
technicalcapability,but some of themwere among the firstto raise doubts
about the technicalcapabilityof ballisticmissile defenses (BMDs) to fulfill
theirmissionand thereforewere among the firstto promotetheircontrol.109
York reportsthat Ruina was the firstmemberof the community to seriously
105. JohnNewhouse,Cold Down: The Storyof SALT (New York: Holt, Rinehart& Winston,
1973),p. 69.
106. In Counselsof War,pp. 197-98,Herkenquotes some of McNamara's concerns:"There is a
kind of mad momentumintrinsicto the developmentof all nuclear weaponry.... If a system
works-and workswell-there is a strongpressurefromall directionsto procureand deploythe
weapon out of all proportionto the prudentlevel required." Herken points out that what
McNamara termed"an action-reaction phenomenon"dominatedand escalated thearmsrace.
107. In November1964, the Sovietsfirstparaded what appeared to be an ABM system.The
system,called Galosh, "was believed to be composed of a networkof radars and a two- or
three-stage,solid-fueledinterceptor missiledesignedforlong-range,ex-atmospheric interception
ofincomingICBMs." See ErnstJ.Yanarella, TheMissileDefenseControversy: Technology,
Strategy,
and Politics,1955-1972 (Lexington:University Pressof Kentucky,1977),p. 118.
108. On the politicsofABM controlup to 1972,see Yanarella, TheMissileDefenseControversy;
Benson D. Adams, Ballistic Missile Defense (New York: American Elsevier, 1971); Morton
Halperin,Bureaucratic 1974);
Politicsand ForeignPolicy(Washington,D.C.: BrookingsInstitution,
Newhouse,Cold Down; Gerard Smith,The Storyof SALT, 2d ed. (New York: Pergamon-Brassey,
1989); Kaplan, WizardsofArmageddon;and Herken,Counselsof War.
109. See David N. Schwartz,"Past and Present:The HistoricalLegacy," in AshtonB. Carter
and David N. Schwartz,eds., BallisticMissileDefense(Washington,D.C.: BrookingsInstitution,
1984),pp. 332-33.
studyan ABM moratorium and to getthePentagoninterestedin theidea.110 In

1962, Schelling,Charles Hertzfeld,Thomas Wolfe, and Daniel Fink partici-
pated in taskforcediscussionsabout controlling ABMs.
During 1964, as Americanslearned about Galosh, the battle positionsfor
and against ABMs were drawn and the epistemiccommunity was called to
defenditsturf.One of the firstshotsin the ideologicaland politicalbattlewas
firedby Wiesnerand York in 1964 in the pages of Scientific American,where
theyarguedthatitwas their"professionaljudgment"thatthenucleardilemma
had "no technicalsolution,"a clear referenceto the ABM system.111 Other
arms controllersstartedto work more quietlywithinACDA to formulate
practical arms control ideas about banning ABMs. Five years of intense
involvement bymembersoftheepistemiccommunity followed.This had less to
do with intellectualinnovationand scientificanalysis than with politics,
politicalalliance formation,and lobbyingand rallyingthe supportof bureau-
crats, Congress, and public opinion against BMDs. By 1967, a new "thin
defense" system,called Sentinel,whichwas aimed at protectingAmerican
cities against a Chinese attack and against an accidental Soviet attack,was
beingapprovedfordeploymentbytheJohnsonadministration.112 Nevertheless,
in 1968 the Soviets were persuaded to negotiate a strategicarms control
agreement,and ISA's MortonHalperin (who was coauthorwithSchellingof
Strategy andArmsControl)succeeded in producingan agenda fortheABM and
strategicarmslimitationtalks(SALT) and in gettingtheJointChiefsofStaffto
givetheirreluctantapprovalto it.
Halperin's success was due partlyto ISA, whichprovidedan institutional
home to armscontrolideas and theircarriers.ISA's powercame fromthefact
that policymakerssuch as Defense SecretaryClark Cliffordwere willingto
listento itspeople. And because Halperinwas the firstto generatea strategic
arms controlnegotiatingagenda, he could benefitfromthe bureaucracy's
uncertaintyand lack of experience.Knowingthat the Chiefs would never
accept an agenda developed at ACDA, Halperin took the action to the
Pentagon,wherehe involvedmilitary personnelin the agenda-makingprocess
in a skillfulmannerthat allowed him to retainhis politicalalliance withthe
State Department and ACDA but at the same time enabled him to put
pressureon the Chiefsto reacha quickdecision.113
Congresswas anotherbattlefront wherethe epistemiccommunity did well.
During the Eisenhower and Kennedy years, arms controllersmanaged to
engage SenatorHubertHumphreyand his DisarmamentSubcommitteein the
business of arms control.Hearings held by this subcommitteediffusedarms
controlideas to the public; Humphreyhimselfmade a contributionto the
110. York,MakingWeapons,TalkingPeace, pp. 222-23.

111. For a discussionofWiesnerand York's article,see Herken,Counselsof War,p. 193.Herken
notesthatPaul Nitzecalled the article"outrageous,an incitement, an exampleof dirtypool."
112. Sentinelwas a lightarea missiledefensesystemset to be deployedin fifteensites in the
continentalUnited States,one site in Hawaii, and one in Alaska. The systemconsistedofvarious
radarsand eithera Spartanmissileor a Sprintmissile,dependingon the site.See Adams,Ballistic
MissileDefense,p. 177.
113. Newhouse,Cold Down, pp. 50 and 115-16.
special 1960 Daedalus issue. But relationsbetweenthe epistemiccommunity

and Congresswere cementedmainlyduringthe 1968-69 congressionalABM
debate.
This debate was aided and even fueled by the interventionof arms
controllers.In a June1968 letterto TheNew YorkTimes,Wiesnerarguedthat
the ABM systemwas a waste of resources.One monthlater,Senator Eugene
McCarthyreleased a positionpaper thatwas writtenby Wiesnerand Kistia-
kowskyand recommendeda unilateralfreeze on Sentinel and on offensive
nuclearweapons deployment."'4 Indeed, at thattimean alliance betweenthe
armscontrolepistemiccommunity and powerfulsenatorsbegan to take shape.
When reporterspicked up the storyabout the technical controversyin
CongressregardingthedeploymentofBMDs in thevicinity ofmajorAmerican
cities,theirmedia coveragehelped close the ranksbetweenscientificassocia-
tionsand a varietyofpeace and grass-roots groupsthathad formedaroundthe
ABM issue. Some ofthesegroupsweredirectlymobilizedbythescientists, who
also organizedpopular committeesand publicralliesin majorU.S. cities.115 In
fact,the politicalenvironment-theVietnamWar, campus unrest,the excep-
tionallylow prestigeof the American military-was veryconducive to the
creationof an anti-ABMcoalition.
DuringtheABM debate,SenatorJ.WilliamFulbright was alertedto thefact
thatgovernment-based scientistswho opposed Sentinelweretheonlyscientists
beforethe ArmedServicesCommittee.As Sentinel'sfatewas
called to testify
beingdecided byNixon'snew Republican administration, Fulbright'sForeign
Relations Committee decided that Senator Albert Gore's Disarmament
Subcommitteewould begin holding educational hearings on ABMs and,
breakingwith an old tradition,would invitenongovernmental scientiststo
testify."The real purpose,"wroteBenson Adams, "was to providea means to
counter the influenceof the Armed Services Committee and to oppose
BMD."" 6 Arms controllerssuch as Wiesner, Ruina, York, Kistiakowsky,
Bethe, Rathjens,and Panofskywere summonedto these hearingsand, using
the ABM systemas theirshowcase,diffusedthe armscontrolparadigmto the
senators,themedia,and theAmericanpublic.So impressivewas theepistemic
community's anti-ABMpressurethatprominentscientistssuch as Wohlstetter
and Kahn, who favoredABMs, became involvedin the debate onlyto counter
it.117
114. Adams,BallisticMissileDefense,p. 186.

115. Yanarella, TheMissileDefenseControversy,pp. 144-47.
116. Adams,BallisticMissileDefense,p. 193.
117. Kahn said thatthe publicdebate had been one-sidedbecause about "ninetypercentofthe
scientistswho normallyspeak in public,or who consultpart-timeforthe government on defense
issues,as well as the vast preponderanceof the publicliteratureon the subject,opposed ABM."
See Herman Kahn, "The Missile Defense Debate in Perspective,"in Holst and Schneider,Why
ABM, p. 285. For a good source on the involvement of pro- and anti-ABMscientistsin the ABM
debate, see Anne Hessing Cahn, Eggheadsand Warheads:Scientistsand theABM (Cambridge,
Mass.: MIT CenterforInternationalStudies,1971). In the "battle of books," the counterpartto
WhyABM was the anti-ABMworkedited by Abram Chayes and JeromeB. Wiesner,ABM: An
EvaluationoftheDecisiontoDeployan Antiballistic MissileSystem(New York: Signet,1969).
Armscontrollerslost the ABM battlein Congressby one vote, and a new

systemcalled Safeguard,aimed at defendingAmerican land-based ICBMs
against a Soviet attack, was approved by Congress.118 Nevertheless,the
epistemiccommunity had won the war, because by educatingboth Congress
and thepublicon deterrenceand armscontrolithad managedto puttheNixon
administration, as well as theABM supportersin Congressand in thescientific
community, on thedefensive.119 Furthermore,thecommunity createda moreor
less permanent,broad political constituencyin favor of arms control by
concernabouttheissueamonga muchwiderpublicthanbefore-grass-
instilling
roots,labor,and religiousgroups,professionalassociations,and peace organi-
zations.
It is true thatthe success of armscontrolideas was directlyrelated to the
public insistencethat militaryself-restraint accompanypolicies of nuclear
rearmament,120 but the meaning,direction,and contentof self-restraint were
providedby the epistemiccommunity, whichhelped make the public aware
thatABM policywas one examplein whichself-restraint was required.By the
end ofthe 1960s,armscontrolhad become one ofthedominantinterpretations
ofnationalsecurity-so dominant,in fact,thatdecisionmakersand negotiators
in the Nixon administration, who had long believed in American military
superiority, were now endorsingnuclear parityand takingstable deterrence
and armscontrolforgranted.121National SecurityAdviserKissinger,however,
was no strangerto armscontrolideas; he had been one of the contributors to
the 1960 Daedalus issue and had kept close links to arms controllers in
Cambridgeall along.
The Nixon administrationappears to have signed the SALT and ABM
treatiesbecause of a diminishedfear of surpriseattackfollowingthe attain-
ment of invulnerability, as well as the awareness that a ceilingon offensive
weapons would constrain the Sovietsmore than it would the Americans(the
United States was not thenbuildingany new ICBMs, Polaris submarines,or
bombers)and would limitthe Soviet SS-9s, whichthe Americansfeared the
most. Soviet insistenceon a treatywas also instrumental, and certainlythe
public uproarover ABMs thatthe epistemiccommunity and its allies helped
118. Safeguard incorporatedboth area and terminaldefense capabilities,using the same

componentsas Sentinelbut deployingthese componentswiththe aim of defendingMinuteman
silos. For commandand controlreasons,Washington,D.C., would be defendedas well. Adams,
BallisticMissileDefense,p. 200.
119. Accordingto Brenner,"Each testimony[beforeCongress]delineated the technicaland
politicalaspects of the issue while assiduouslydrawingthe necessarydistinctionsbetweenthose
questionsamenableto scientific judgmentand thoserequiringsubjectiveestimates.By stipulating
thelogicalconnectionsbetweenacceptanceof ABM and itsmultipleconsequences,theseanalyses
heightenedawarenessof the issue's subtle interdependencies. They discreditedthe Administra-
argumentthatin the past had reliedsuccessfully
tion'scasual use of the syllogistic on faith(in the
simple equation that more arms means more security)and fear (of Soviet aggression)." See
Brenner,"The Theoristas Actor,"pp. 115-16.
120. Sims,"The DevelopmentofAmericanArmsControlThought,"pp. 13-14.
121. Firestone,TheQuestforNuclearStability, p. 150.
bringabout put the Nixonadministration on noticethatthe Americanpeople

did notwantABMs in theirbackyards.Questionsof linkageand detentealso
playeda role.
All of these factorswould not have had any meaning-indeed, would not
have been rationallyconsideredor even relevant-had not the ideas of stable
deterrenceand arms controlbecome a salientparadigmof nationalsecurity
and been diffusedto government wheretheywere instrumental
institutions, in
shapingan armscontrolagenda and a politicalcoalitionto carryit out. During
theABM debate process,the armscontrolepistemiccommunity and its allies
convincedtheAmericanpeople thatthe superpowershad a mutualinterestin
avoidingnuclearwar and thatthisinterestshould be symbolically, politically,
and practicallymanifestedin an ABM armscontroltreaty.
to the SovietUnion
Intellectualdiffusion
The diffusionof American arms control ideas to the Soviet Union was
necessaryforthe creationof the ABM regime.Helping to create an interna-
tionalnegotiationagenda and providetheepistemicframework fornegotiation
and agreement,theseideas structured notonlytheAmericandomesticbutalso
the internationalpoliticalgame. Accordingto Marshal Shulman,the transfer
of ideas had "a residualeducationaleffectthatyoucannotalwaysmeasurebut
whichmaybe terriblyimportant.There is a kindof diffusionof conceptions
that goes on, there is an educational process ... because we are just ...
beginningto have insightsintowhatmakesforstability."122
The international of nucleararmscontrolideas began in the 1950s
diffusion
and continuedthroughout the 1960s.Directmeanswerenegotiationproposals,
bargainingand negotiationpositions,summitmeetings,technicalconferences
(such as the Surprise Attack Conference), and scientificforums(such as
Pugwashand the "Doty," "Dartmouth,"and "Panofsky"groups).'23Indirect
means included Western statementsand strategicdebates, congressional
hearingsand debates,pressreports,and academicbooks and articles.
122. See testimonyof Marshal Shulman, in U.S. Congress,Senate Committeeon Foreign

Relations,The Strategic and ForeignPolicyImplicationsofAnti-Ballistic MissileSystems:Hearings
Beforethe Subcommitteeon InternationalOrganizationand DisarmamentAffairs,part 1, 91st
Congress, 1st sess., March 1969, p. 154. See also the testimonyof Vincent P. Rock, in U.S.
Congress,Strategy and Science:Hearings,p. 224,whichincludedthe followingargument:"In terms
of aid, in termsof weapons .. . thereis a greatdeal of copying,of actionand reaction,reciprocal
actionof a kind,betweenthe nationsof theworld.... As we know,all nationscollecteach other's
output.There is a tremendousinteraction
basic and applied scientific goingon as a resultofhaving
to read and cope withtheideas theotherfellowis puttingout."
123. The Soviet-AmericanDisarmamentStudiesGroup,referredto as theDotygroup,started
to meet in 1965 and met forten years.The firstconferenceof the Darmouthgrouptookplace in
1959. An officialcollaborationbetweentheAmericanand Sovietacademies of scienceshas taken
place undertheguidanceofW. Panofskyand S. Sagdeev.
When ideas of strategicnucleararmscontrolwere firstraisedbyAmericans

in the 1950s,theywere interpretedbythe Sovietsto mean inspectionwithout
disarmamentand evoked suspicionsof espionage and a capitalistplot. (In
Russian,kontrolmeans to count,audit,or inspectand does not sharewiththe
Englishconceptof controlthe meaningof regulationand management.124) At
the Surprise Attack Conference,the Soviets insisted that they could not
envisage a technical solution to the nuclear predicament.125 They argued
insteadthat"any technicaldevice mayfail,but a technicalfailuremaylead to
disaster only in a climate of artificiallyheightenedtensions." 126 But the
conferencewas neverthelessa turningpoint because even this formaland
ostensibly"unsuccessful"discussionhad a constructive effecton thesuperpow-
ers' continuousdialogueoverstrategicissues."'
By the early1960s,the Sovietshad begunto moveunilaterallyto make their
strategicweapons invulnerableand to recognizethe value of reconnaissance
satellites.Moreover,theirmilitarywritingsstartedto mentionthataccidents
could be a casus belliin timesof internationaltension,128and eventuallythey
adopted an Americandevice to preventaccidentalwar,the permissiveaction
links(PALs).129Theywentalong withthe idea of the hotlineand were finally
persuadedto signthePTBT. "Thus,by1963,"wroteRobin Ranger,"the Soviet
Union's adaptation of the concepts of arms control to meet its political
objectiveshad producedan implicittheoryof politicalarmscontrol."'130 Aware
thattheAmericanperceptionofstrategicproblemswas primarily technicaland
that,to engageAmericans,theywouldhave to discusspoliticalproblemswithin
theAmericantechnicalframework, the Sovietsalso recognizedthe benefitsof
negotiatingpolitical agreementswithoutappearing to do so. Accordingto
Ranger,one of the unintendedconsequencesof thisprocesswas thatas "the
bilateralSoviet-Americanrelationshipbecame more clearlydefinedthrough
124. WalterC. Clemens,Jr.,Can Russia Change?The USSR Confronts Global Interdependence

(Boston: UnwinHyman,1990),p. 67.
125. Holst,"StrategicArmsControland Stability,"pp. 258,264, and 268.
126. The SovietviewswerecitedbyBruce J.Allynin "Toward a CommonFramework:Avoiding
InadvertentWar and Crisis,"in GrahamT. Allisonand WilliamL. Ury(withBruce J.Allyn),eds.,
WindowsofOpportunity: FromCold WartoPeacefulCompetition in US-SovietRelations(Cambridge,
Mass.: Ballinger,1989),p. 188.
127. Holst,"StrategicArmsControland Stability,"p. 282.
128. See Allyn,"Toward a CommonFramework,"p. 188; and JohnL. Gaddis, TheLong Peace
(Cambridge:CambridgeUniversity Press,1987),p. 204.
129. See Peter Stein and Peter Feaver,AssuringControlofNuclear Weapons:The Evolutionof
Permissive ActionLinks(Cambridge,Mass.: HarvardCenterforScience and InternationalAffairs,
1987).
130. See Ranger,Armsand Politics,p. 7. Dinersteinand his colleagues at RAND noted in the
early1960sthatthe technicalarmscontrolapproachseemed to hold no interestforSovietmilitary
planners.Accordingto Dougherty,however,"Some changewas noticeableaftertheCuban missile
crisis.... Duringthepastdecade [1963-1973],therehavebeen signsthattheSovietshavebegunto
takemoreseriouslytheWesternideas of 'armscontrol.'" See HerbertS. Dinerstein,Leon Goure,
and Thomas W. Wolfe,"Introduction"to the Englishtranslationof SovietMilitary Strategy,ed. by
SovietmarshalV. D. Sokolovskiy(EnglewoodCliffs, N.J.:Prentice-Hall,1963),p. 77; and JamesE.
Dougherty,How to ThinkAboutArms Controland Disarmament(New York: Crane,Russak,1973),
p.71.
agreementsthatwerecouchedin termsofthe explicittheoreticalframework of

technicalarms control... the Soviet leadershipbecame increasingly explicit
Americans,conversely,
about itsapproachto strategicstability."'31 increasingly
became involvedin politicalarmscontrol.
The PTBT and hotlineagreementsreachedin theearly1960s,thespace and
nonproliferation treaties signed in the late 1960s, and three agreements
negotiatedand concludedconcurrently withSALT/ABM negotiationsin the
early 1970s-the seabed treatyof February 1971, the nuclear accidents
measuresand directcommunications linkagreementof September1971,and
thebiologicalweaponsconventionofApril1972-were important stepstoward
strategicagreementsbecause theyindicatedto Soviets and Americansalike
thatcooperationby means of arms controlagreementswas indeed possible.
Moreover,thesetreatiesbecame a testinggroundforsome of theprovisionsof
the SALT agreements,such as the use of "national technical means of
postagreementevaluationconferences,and provisionsforwith-
verification,"
drawalfromagreements.132 Both sides saw thatsome intermediategoals were
beingachievedand some progresswas beingmade. The Sovietscame to realize
that, as Arthur Schlesingerput it, "arms control mightbe a means of
approachingratherthanavoidinggeneraland completedisarmament."'133
The ABM treatywas theculminationofa decade-longprocessofdiffusion of
Americanarmscontrolideas to the Soviets.The patternusuallybeganwithan
American proposal, followedby a Soviet response,a new set of American
suggestionsbased on that response, and engagementin a new round of
negotiations."'Much of the officialdiscussionof strategicdoctrineand force
positionconsistedof such "talkingto Moscow at a distance,"includingthe
educationof the Sovietsin therequirements of a safe and securesecond-strike
forceduringtheKennedyand Johnsonyears.
Part of this education took place by means of direct contacts between
Americanand Sovietscientists, suchas in Pugwashmeetingsor in themeetings
of a committeeorganizedby Doty and the AmericanAcademy of Arts and
Sciences. These meetings,argued Frank von Hippel, "often provided an
opportunity to investigatenew experimentalideas thatgovernmentagencies
have been loathto exploreforfearofreducingpoliticalmaneuvering room."135
For example,duringthe twelfth Pugwashmeeting,in 1964,JackRuina toldhis
Sovietcounterpartsabout the idea of controlling ABMs. HerbertYork recalls
that"afterJack'spresentationthe head of the Soviet delegationapproached
himand said theremusthave been somethingwrongwiththe translation.He
131. Ranger,Armsand Politics,p. 209.

132. Samuel B. Payne,TheSovietUnionand SALT (Cambridge,Mass.: MIT Press,1980),p. 75.
133. Schlesinger,
134. See Michael Mandelbaum,"WesternInfluenceon the Soviet Union," in SewerynBialer
and Michael Mandelbaum,eds., Gorbachev'sRussia and AmericanForeignPolicy(Boulder, Colo.:
WestviewPress,1988),p. 364; Holst,"StrategicArmsControland Stability,"p. 245; and Yanarella,
TheMissileDefenseControversy,pp. 197-98.
135. Frankvon Hippel, "Arms ControlPhysics:The New Soviet Connection,"PhysicsToday,
November1989,p. 39.
explained that he actuallyheard the interpreter say Jackproposed to limit

defensiveweapons." Ruina and MurrayGell-Mannthendrafteda paper on the
subjectand submittedit to the conference.The Soviets"stillconsideredit a
strangenotionbut agreed to thinkmore about it." In lateryears,some of the
Soviet scientistswho had participatedin the meetingswithAmericanarms
controllershelped persuade Soviet policymakers.Von Hippel reportsthat
"Lev Artsimovich(who was head of the Soviet fusionprogram)and Mikhail
Millionshchikov (who was vice presidentforapplied physicsand mathematics
of the Soviet Academyof Sciences) subsequentlyhelped bringtheirgovern-
mentaround ... therebycontributing to the achievementin 1972 of theABM
Treaty."In turn,Dotyadds that"it is widelythoughtthatthewillingnessofthe
USSR to negotiatean ABM Treatyarose fromthe seminarsthatthisgroup
held."136
Before 1968,the Sovietleaders gave everyindicationthattheycould not or
did not want to understandwhy defenses were "bad." At the Glassboro
meeting,KosyginrejectedMcNamara's initiativeto ban defenses.Beneaththe
surface,however,not onlythe Sovietperceptionsbut also the Sovietpolitical-
strategicgame had startedto change. Referringto the relativelylong timeit
took the Soviets to react to McNamara's proposals,a Soviet diplomatsaid,
"Don't thinkwe weren'tstudyingthe problem.It was just too soon. We didn't
thinkwe were ready."137 In fact,as David Holloway pointed out, both "the
ambitionto attainsuperiority and the recognitionofmutualvulnerabilitywere
presentin Soviet thinkingin the 1960s. But a choice became necessaryonly
withthe attainmentof parity[and was] forcedby the practicalconsideration
that the pursuitof superioritymightprove extremelycostly,and ultimately
unsuccessful."138
Two importantnewpremisesthatthe Sovietshad alreadyadopted made the
acceptanceofstrategicarmscontrolpossible.Accordingto Michael MccGwire,
"One was thata war in Europe would not inevitablylead to massivestrikeson
Russia, exceptin retaliationforan attackon NorthAmerica.The otherwas
oftheU.S. strategicarsenalmeantthata preemptive
thatthesize and diversity
strikeon theUnitedStateswoulddo littleto limitthedevastationofRussia."139
In any event,Samuel Payne argues,in 1968 "an era ended for Soviet arms
136. See York,MakingWeapons,TalkingPeace, p. 223; J.P. Ruina and M. Gell-Mann,"Ballistic

Missile Defense and the ArmsRace," inProceedings oftheTwelfth PugwashConference on Science
and WorldAffairs,Udaipur,India, 27 Januaryto 1 February1964,pp. 232-35; Von Hippel, "Arms
ControlPhysics,"p. 39; and Paul Doty,"Arms Control:1960, 1990,2020," Daedalus 120 (Winter
1991),p. 40.
137. Regardingthe Sovietreactions,see Newhouse,Cold Down,p. 102.
138. David Holloway,The SovietUnionand theArmsRace, 2d ed. (New Haven, Conn.: Yale
UniversityPress,1989),p. 44.
139. See Michael MccGwire,"Whythe SovietsAre Serious About ArmsControl,"Brookings
Review,Spring1987,p. 11. In a reviewof MccGwire'sbook, MilitaryObjectivesin SovietForeign
Policy (Washington,D.C.: BrookingsInstitution,1987), Bluth argued that the changes which
MccGwiresaid occurredat the end of 1966 actuallystartedin 1964 and 1965. Bluth's pointwas
indeed provedby a flurry of Soviet articlesdiscussingthe possibilityof doctrinalchange in the
1964-65 period. See ChristopherBluth,"The Evolutionof SovietMilitaryDoctrine,"Survival30
(March-April1988),p. 149.
controlpolicyand a new era began. Before 1968 strategicarmslimitationwas

simplynoton theagenda fordiscussionin the SovietUnion."140
By 1968, a factionof the Soviet politicaland academic establishments had
alreadystartedto oppose the view of the Soviet military. Payne suggeststhat
those who firstraised ideas of strategicarmscontrolwere mainly"academic
specialistsand commentatorson foreignaffairswho writefor the scholarly
journalsin the fieldand also forthe centralpress."141 These Sovietswere well
awareofwhatthemembersoftheAmericanarmscontrolepistemiccommunity
were writingand sayingand shared some of their expectations.142Andrei
Kokoshinremarkedthat"at thebeginning, theAmericanshad a largerpool of
ideas of armscontroland we borrowedsome of them."143 In fact,notedPayne,
the majorityof the Soviet armscontrollers'attackson ABMs "were director
indirectquotationsfromstatementsmade byAmericanopponentsofABMs":
Iu. Arbatov,forexample,quoted George Rathjens as someone who believed
that"withthe presentrelationshipof forcesthe strategicpositionof the USA
basicallywould not change if the Soviet Union had twiceas large or half as
large a strategicforce."144V. V. Larionov,echoing an Americanstatement,
argued that "fromthe point of view of national securitythe effortto have
quantitativesuperiorityin rockets and bombers has lost its significance,
because at any realizable level the otherside, spendingsufficientenergyand
resources,can also reachthatlevel."1145
AlthoughSovietarmscontrollers werewell awareoftheideologicaldivisions
on nuclearissues in theUnited States,theydrewconfidencefromthefactthat
a stronggroupof armscontrollobbyistsexistedin the United States and used
thisfactto persuade reluctantSovietactors.146 Accordingto Payne,the Soviet
leadership
apparentlyaccepted the SALT I agreementsforsome of thesame reasons
thatthe [Soviet]armscontrollershad advancedforstrategicarmslimitation
overthepreviousseveralyears.As thenegotiationsgatheredmomentum
and, even more,afterthe agreementswere signed,membersofand spokes-
menforthe supremeleadershipincreasingly echoed armscontrollerargu-
ments.Ideas thathad previouslybeen aired in SShA4and Mirovaia
ekonomikai mezhdunarodnye otnosheniianowappeared in Brezhnev's
editorialsinPravda,Izvestia,and Kommu-
speeches and in authoritative
nist.
147
140. Payne,TheSovietUnionand SALT, p. 18.

141. Ibid.,p. 7.
142. GeorgiA. Arbatovand WilliamOltman,SovietViewpoint (New York: Dodd, Mead, 1981),
p. 130.
143. AndreiKokoshin,personalcommunication.
144. Payne,TheSovietUnionand SALT, pp. 46 and 59.
145. See ibid., p. 126. V. V. Larionov made the statementin "Transformatsiiakontseptsii
'strategicheskoi " SShA, November1971.Gromykolatermade thissame point,as
dostatochnosti,'
did the ministerof foreignaffairs.It was almostan officialstatementof the Sovietgovernment's
position.
146. Payne,TheSovietUnionand SALT, pp. 40-41, 47, and 76.
147. Ibid.,p. 76.
For example, in the summaryreport of the Central Committee to the

Twenty-Fourth PartyCongress,armscontrolstrategicnegotiationswere por-
trayedas aimingto preventa new round in the arms race while protecting
Soviet securityand releasing significant resourcesfor creativeobjectives.148
Schelling and Halperin could not have explained arms control'spurposes
better.Thus, even if the Soviet leadershipstartedthe negotiationswithouta
clear pictureof theirend resultor withouta definitionof the politicalbattle
between arms controlsupportersand opponents,the fact that it eventually
chose arms controlshows that the bureaucratswho opposed it suffereda
temporarydefeat.149 By placingthe ABM agenda at the highestlevels of the
Sovietgovernment-thus circumventing thebureaucracy-Americanpolicymak-
ers helped Sovietarmscontrolsupportersprevailin the Sovietpolicygame.At
the same time,once the top Sovietleaders threwtheirweightin favorof arms
control,theyput pressureon the Americanpolicygame,helpingto break the
politicalimpassebetweenAmericanABM supportersand opponents.'50
One cannotavoidnoticingthehegemonicqualityoftheprocessinvolving the
diffusion of U.S. armscontrolideas to the Soviets.Some lightis shed on this
phenomenonby Scott Jamesand David Lake's notionof the "three faces of
hegemony."'' The firstand second "faces" are onlytangentially relatedto the
diffusionprocess. Accordingto the firstface or process, the hegemonuses
positiveand negativesanctionsas a means of directlyinfluencing the policy
choices of other states. Thus, the United States was able to affectSoviet
behaviorby means of "linkagepolitics"and the "China card," but onlyafter
the superpowershad alreadybeen negotiatingarmscontrolforsome timeon
the basis of an Americanagenda. Accordingto the second face,the hegemon
pursues a sort of rational-choiceTrojan-horsestrategyin order to alter the
incentivesand the politicalinfluenceof societalactorsin foreigncountries.In
the case of strategicarms,the United States used its superiortechnological
powerto persuade Sovietactorsthata defensiveweapons race wouldnotbe in
the Soviet interest.Later, these actorsdid have some influenceon the Soviet
politicalgame. But thisface of hegemonyonlyexplainshow the United States
148. Ibid.,p. 66.

149. Otherreasonsthatled theSovietleaders to signa Soviet-U.S. strategicarmscontroltreaty
includedthe following:theyperceivedthatthe Americansheld a strongedge in the technological
race; theyrealizedthatmultipleindependently targetablereentryvehicles(MIRVs) were entering
into the strategicequation; theywanted to institutionalize
paritywiththe United States and, if
possible,improvetheirstrategicsituationin areas unrestricted by SALT; theywantedto project
power; theywished to strengthendetente with the West; and they hoped that the resulting
economicsavingscould be directedback to theciviliansector.
150. See RaymondL. Garthoff, "Mutual Deterrenceand StrategicArmsLimitationin Soviet
Policy,"IntemationalSecurity 2 (Summer1978), p. 126; and MortonHalperin,"The Decision to
Deploy the ABM: Bureaucraticand Domestic Politics in the JohnsonAdministration," World
Politics25 (October 1972),p. 95.
151. The facesreferto complementary and mutuallyreinforcing processes.Theyare ideal types,
and "the distinctionbetween'faces' tendsto break down at the margin."See ScottC. Jamesand
David A. Lake, "The Second Face of Hegemony:Britain'sRepeal of the Corn Laws and the
AmericanWalkerTariffof 1846,"IntemationalOrganization 43 (Winter1989),p. 4.
was able, in the bargainingprocess,to raise the cost for the Soviets of not
controllingdefenses; it does not explain the "sudden" Soviet interestin
controlling ABMs.
The thirdface of hegemonyis directlylinkedto diffusion processesand may
help us understandwhycertainideas "diffuse"betterthanothers.In thisface,
the hegemon "uses ideas and ideologyto structurepublic opinion and the
politicalagenda in othercountriesso as to determinewhatare legitimateand
illegitimatepolicies and formsof political behavior."'52On a closer look,
however,thisface too providesonlya limitedexplanation.For obviousreasons,
affecting public opinion was almost irrelevantin the Soviet case. Moreover,
Jamesand Lake's descriptionofthethirdface givesus fewclues as to howU.S.
ideas managed to controlthe Soviet agenda and helped to structurepolicy
preferencesin the SovietUnion.
There seems to be a fourthface of hegemonythatgoes a longwayto clarify
whattheotherthreefacesfailto explain.The UnitedStateswas able to diffuse
its ideas to the Soviets and "gently"impose its agenda on thembecause the
U.S. arms control epistemic communityhad undergone the process of
ramification, therebygainingadherentsin the SovietUnion. This expansionof
thecommunity's base allowed armscontrolunderstanding to flowto the Soviet
polity,thusbecomingan integralpartof the Sovietdomesticpoliticalgame. It
also endowed Soviet arms controllerswith a legitimateclaim to a new
interpretation whichbecame thebasis on which
of the Sovietnationalinterest,
politicalcoalitionswere created and, ultimately, policies were made. In the
fourthface ofhegemony,then,hegemonicideas structure notonlythepolitical
agendas but also the political games of other countries.They also play a
reflexive role byincreasingthepropensity ofothercountriesto learn.
Once strategicnegotiationswere underway,the United Statesexpressedits
willingnessto scrap ABMs entirelyif the Sovietswould limittheirSS-9s and
eliminatetheirABM systempoised aroundMoscow. Eventually,in 1972,both
sides settledforlimitingABMs to twosites,and a fewyearslatertheyagreed
on a one-sitelimit."The signingat Moscow in May 1972oftheSALT treaties,"
wroteGregg Herken,"seemed an occasion of barelyrestrainedjoy forthose
who had come to identifythemselves collectivelyand sometimes self-
consciouslyas 'the arms control community.'... The treaties seemed to
represent,therefore,a substantial-if not yet final-acceptance of the idea
thattherecould be no victorin a nucleararmsrace."153
In the longrun,the diffusion of armscontrolideas to the SovietUnion had
profoundeffects.Since thelate 1960s,theSovietpoliticalsystemhas carriedan
understanding embodied in politicaland academic institutions-andperhaps
even,recently, in militaryinstitutions-whereindefensesare seen as detrimen-
tal to,and armscontrolas beneficialfor,nationalsecurity.This understanding,
152. Ibid.
153. Herken,Counselsof War,p. 247.
similarto the U.S. case, helped balance and moderatepressuresarisingfrom

an opposingand also verypowerfulunderstanding, carriedmainlybymilitary
elites,thata protractednuclearwar could be foughtand won and thatarms
controlmightpreventa victoryon thebattlefield.
When MikhailGorbachevcame to power,he not onlyadopted some of the
classic arms controlassumptions,which by then some American elites had
forgottenabout or did not want to remember,but also pushed themfurther.
And he managed to affectAmerican and internationalpolitical games by
creatingfavorableconditionsand opportunities forarmscontrol.Helpingthis
trendwas a groupof Sovietcivilianstrategistswho had acquiredtheirdefense
expertisein the West withWesternstrategistsand were now enjoyingdirect
access to power. Even high Soviet militaryofficersadmittedthat theirnew
ideas of unacceptable damage could be traced back to McNamara."54Thus,
once arms control ideas became embodied in domestic and international
procedures and institutions,the domestic and internationalgames were
irrevocablychanged. Each new generationof leaders now had to make its
(rational) decisions on the basis of an inheritedintellectualcode of arms
controlideas which,withthepassage of time,was enlarged,refined,and taken
forgranted.
The political road for arms controlideas was, however,full of obstacles
arisingin part fromthe internationalgame but in even largerpart fromthe
domesticgame. Setbacksincludedthe SALT II treatyand the lack of support
forarmscontrolduringthe firstyearsof the Reagan administration. President
Reagan, aided by Edward Teller, Richard Perle, and othermembersof the
"deterrence" community, devised the SDI not as a complementbut as an
alternativeto armscontrol.Nevertheless,the practiceof and nationalinterest
in armscontrolnotonlysurvivedand keptthesuperpowersbusytalkingrather
thanfighting, butwithGorbachev'scomingto powerbecame,once more,a key
factorin superpowerrelations.And since the revolutionsof 1989 in Eastern
Europe, armscontrolhas also become a means forenablingthe transitionto a
newEuropean order.
and the
Conclusions:thearms controlepistemiccommunity
emergenceofprudentialregimes
The role playedbythe armscontrolepistemiccommunity in the emergenceof

nucleararmscontrolcooperationbetweenthesuperpowerswas significant and
multifaceted:
First, the communitycreated an intellectualclimate favorable to arms
control.Decision makersneed not have read Schellingto understandarms
154. Edward L. Warner III, "New Thinkingand Old Realities in Soviet Defence Policy,"
1989),pp. 18-20.
Survival31 (January-February
control;theideas were in theair as partof thevague buthistoricallyimportant

155
"spiritofthe times."
Second, the membersproduced the technicalknowledgerequiredto deal
with nuclear arms control.This knowledge,in turn,was used by the arms
controllersto gainpoliticallegitimationand authority.
Third, the communityfocused attentionon cooperativephenomena and
helpedprovidethesuperpowerswithreasonswhy-despite all theirideological
and political differencesand despite the fact that in the past disarmament
negotiationshad neverbeen takenveryseriously-itwas importantthatthey
cooperate.156
Fourth,it paved thewayforthe creationofvestedinterestsin armscontrol,
including governmentagencies such as ACDA and a large number of
nongovernment interestand pressuregroups.ACDA, evenifleadingwhatPaul
Walkerdescribedas "a ratherprecariousexistencewitha historyofverymixed
was importantbecause it institutionalized
success,"'157 armscontrolideas and
proceduresand providedthe bureaucracywithan institutional counterweight
to theJointChiefsof Staff.Because armscontrolideas were also institutional-
ized in the Pentagonand the WhiteHouse, theycould more easilyfindtheir
wayintopolicyagendas.The epistemiccommunity also helpedfosternewareas
of researchand developmentand otherarmscontrolactivities, withthe result
thatadditionalcommunitiessprangup around these activitiesat universities,
thinktanks,armscontrolassociations,academies of science,and the meetings
oftheAmericanAssociationfortheAdvancementof Science.
Fifth,the epistemiccommunity helped generatean awarenessabout arms
controlthat eventuallyled to public supportfor it. Actingto influencethe
media because of the opposition encounteredin some inner policymaking
circles,the communitywas able to conveyto people the national security
qualityand value of armscontrolideas. By suggestinghow armscontrolideas
mightbe relatedto U.S.-Soviet relationsand bycreatingin people's mindsan
almost instinctiveanalogy between arms controland avoidingnuclear war,
armscontrollers were able to shape publicattitudeswell intothefuture.
Sixth,armscontrollershelped persuadeCongressabout thevalue of specific
arms control agreements.Thus, duringthe ABM debate, they acted as a
155. RobertJervis,"Realism, Game Theory,and Cooperation,"WorldPolitics40 (April 1988),

p. 325.
156. This awarenessofthevalue ofcooperationfornationalsecurity was essential.Accordingto
Davis, "As the naive typeof unsafeguardedarmscontrolof the 1920'sbecame clearlyinappropri-
ate to the problemsof the nextthreedecades, theredeveloped a relativelyharmlesstraditionin
politicsof payingit lip service,so as not to offendthe gentlerelementsof publicopinion,and of
ignoringitin practice.This traditionofthewhitelie [was]carriedoverintothenuclearage.... For
severalcriticalyearsthe habitof pretendingto workfordisarmamentservedto maskthe factthat
the politicalleadershipof theUnited Statesdid notwantdisarmament. thosein
More specifically,
Washingtonwho consideredarmscontrolundesirableor impracticalclearlyhad theupperhand in
the processof makingand administering policy,withthe help of otherswho thoughtthe Russians
would never sign anyway,or would sign and cheat." See Davis, "Recent Policy Making in the
UnitedStates Government,"pp. 379-80.
157. Walker,"The U.S. ArmsControland DisarmamentAgency,"p. 13.
counterweightto governmentscientistswho advocated the deploymentof

defenses.
Seventh,memberswere able to propose a logicallycoherentarms control
negotiationagenda and helped thinkthroughthe bargainingpositionsto be
taken in the ABM negotiations.Armed with arms control theory,they
suggestedthe winningbargainingtacticsand called the negotiators'attention
to focalpointsof cooperationsuchas BMDs, space, thebottomof thesea, and
so on. They also pointed to the need for confidence-building measures to
preventaccidentalwars. And theyexplainedthe politicalconsequences that
wouldfollowfromtechnologicalchangesand fromvariousalternativebargain-
ingpositions.
Eighth, the communityhelped formulatespecificnorms and rules, re-
searched and proposed verification means, and suggestedposttreatyreviews
and conditionsforwithdrawalfromagreements.
Ninth,arms controllersin manycases became what Robert Gilpin called
"full partnerswith politicians,administration, and militaryofficersin the
formulation ofpolicy."'158
Finally,the community was instrumentalin transmittingarmscontrolideas
to theSovietUnion.
The factthatmanymembersof the armscontrolepistemiccommunity were
broughtinto the halls of government,where they persuaded and worked
togetherwith policymakersto institutionalizethe arms control paradigm,
explainsin partwhyarmscontrolexpectationswere politicallyselectedbythe
Americangovernment. The pluralisticnatureof theAmericanpoliticalsystem
and the relativelydecentralizedprocess by whichpolicyagendas are deter-
minedactuallyhelped theepistemiccommunity createan armscontrolagenda
withinthegovernment. Byprovidingcommunity memberswithseveralalterna-
tivesourcesofpoliticalpower,thepoliticalsystemhelped protectthemagainst
political,ideological,and personnelchanges at the top. When theywere not
able to counton thedirectsupportofthePresidentand hisimmediateadvisers,
they could turn to other governmentinstitutions,includingCongress. In
addition,government institutionssuch as ACDA, ISA, and, at times,the NSC
protectedarmscontrollersfromopposinginterestsand pointsofview.
The ideologicalaffinity betweenthe armscontrollersand theKennedycadre
smoothed the transitionof arms controlideas fromthe intellectualto the
politicalrealm.These ideas also had some inherentadvantages.In one stroke,
they addressed the two most importantconcerns of the time-enhancing
nationalsecurityand avoidingnuclearwar-and theyexpresseda middle-of-the-
road position,appealing to both prodisarmamentand conservativepolitical
groups who were nondogmatic.Thus, they produced a balance, indeed a
temporaryconsensus,betweencompetingtrendswithinthe governmentand
society.
and NuclearWeaponsPolicy,p. 299.

158. Gilpin,AmericanScientists
President Kennedy's backing of early arms control measures played an

importantrole in the political selection process. Even more important,
however,was the fact that he gave the arms controllersa chance to get
establishedwithingovernmentinstitutions, where they could spread their
influencethroughout thepoliticalsystem.
The educationof McNamara byarmscontrollersalso playeda criticalrole;
indeed, he was the national security"czar" formost of this crucial period.
Convincedthat therewas a mad momentumin the arms race that could be
mitigatedonlybyarmscontrol,McNamara fiercely opposed ABM deployment
and persuaded Johnsonto engage the Soviets withan arms controlagenda
before the momentumcould take another turn for the worse. Although
McNamara had leftthe governmentby 1968, the task of "selling" an ABM
treatyto thegovernment was continuedbythearmscontrollers.
By thattime,however,publicconsciousnesshad changedand strategicarms
controlhad gained wide supportamong the Americanpeople. This was not
only because at the beginningof the 1960s a group of expertswithinthe
governmenthad championed arms control ideas, which McNamara then
but also because these ideas were validatedin later
helped to institutionalize,
yearsby structuralchanges,such as the attainmentof strategicparityby the
SovietUnion. As thetwosuperpowersstartedto act bothindependently and in
coordinationon the basis of armscontrolideas, theygenerated domestic and
international tendencies that would induce future generations of leaders to
continuewiththe armscontrolprocess.
The Sovietpoliticalelites,fortheirpart,had been affectedbythediffusion of
armscontrolideas forovera decade and,fortheirownreasons,agreedto leave
aside rhetoricaldemandsfortotaldisarmamentand negotiatearmscontrolon
the basis of an American agenda. We should be careful,however,not to
conclude fromthis that the Soviet leadershipsimplysaw the light,dropped
classic Soviet militarydoctrine-best exemplifiedby V. D. Sokolovskiy's
writings"59-and adopted Americanstrategicdoctrinesand politicalbeliefsand
goals. But aftertheSovietsachievedstrategicparitywiththeUnitedStatesand
aftertheAmericanarmscontrolmovementgrewin size and power,theidea of
stabilizingthe armsrace throughtechnicalarmscontrolbegan to make more
sense to the Soviets,if only because arms controlnegotiationsand treaties
could be used to achieveSovietpoliticaland strategicobjectives.
Sharing with the United States the desire to avoid nuclear war and
encouraged to turn the achievementof parityinto political power, Soviet
leaders-not unliketheirAmericancounterparts-sawarmscontrolideas as
an obvious focal point for pursuingboth shared and divergentinterests.
Indeed, all the Sovietshad to be persuadedaboutwas thatarmscontrolwould
help detertheWest and limititsweapons,thathavingachievedparitywiththe
UnitedStatesand havingbuiltan invulnerablenuclearforceitwas in theSoviet
interestto keep the situationstable,and that arms controlcould be used to
159. See Holloway,TheSovietUnionand theArmsRace, p. 40.

But the ABM regimedepended also on

furtherSoviet political interests.160
Soviet willingnessto negotiateon the basis of the American arms control
paradigmand on the sharingof some meaningsand conceptsabout stability,
deterrence,theuse offorce,and cooperationwithadversaries.
TerryNardin'sdistinction between"purposive"and "practical"association
why,contrary
is usefulforillustrating to whatSteveWeber suggests,161 security
regimesneed notnecessarilydepend on theparties'learningthe same lessons,
adopting similarmilitarydoctrines,and sharingpolitical beliefs and goals.
Accordingto Nardin,purposiveassociationis "a relationshipamongthosewho
cooperate for the purpose of securingcertain shared beliefs,values, and
interests,who adopt certainpracticesas a means to thatend, and who regard
such practicesas worthyof respect only to the extentthat theyare useful
instruments of the commonpurpose." 162 An internationalregimebased on
purposive association-or what can be called instrumentalassociation-
assumesthattwoor more nation-stateshave indeed learned the same lessons
and developed commonpoliticalbeliefsand goals and are actingtogetherto
achieve those goals. The furtherwe get frompower politics,the higherthe
likelihoodfor the emergenceof instrumentalregimes.163 The ABM regime,
however,was clearlynot instrumental: power politicswas essential,and the
parties shared neitherpolitical beliefs and goals nor objectiveor scientific
knowledgeregardinghowto avoidwar.
Practicalassociation,on the otherhand,is "a relationshipamongthosewho
are engagedin thepursuitofdifferent and possiblyincompatiblepurposes,and
who are associatedwithone another. .. onlyin respectingcertainrestrictions
on how each maypursuehis own purposes."164 An internationalregimebased
on practicalassociation-or what can be called prudentialassociation-may
resultfromthe recognitionby two or more states thatit is in theirseparate
intereststo cooperate.In otherwords,thepartiesconvergeon a recognitionof
whathas to be preventedratherthanofwhathas to be mutuallyachieved;each
side constrainsitselfin order to constrainthe other. A prudentialregime
emerges,however,onlyaftergovernments sharesome epistemiccriteriaabout
whyand how theyshouldcooperate,how to startnegotiations, whatto include
in the agenda,and howto conceptualizenormsand rulesforparticulartasks.
Because thistypeof "knowledge"willmostlikelybe developedby national
institutionsand politicallylegitimizedby national governments, an interna-
tionalregimewillemergeonlyaftermeaningsand understandings are diffused
and, based on them,a negotiationagenda is created,agreed upon, and acted
upon. Writlarge,then,armscontrolpracticebecame an institutionalized way
160. RaymondL. Garthoff, "On Mutual Deterrence:A Reply to Donald Brennan,"Interna-

tionalSecurity3 (Spring1979),p. 198.
161. Weber,"Realism, Detente and NuclearWeapons," p. 72.
162. Terry Nardin, Law, Morality,and the Relationsof States (Princeton,N.J.: Princeton
University Press,1984),p. 14.
163. I owe thisinsightto CraigMurphy.See CraigN. Murphy,"Color It Mitrany:Two Patterns
of Progressin InternationalRelations,"workingpaper,WellesleyCollege,Wellesley,Mass., 1989.
164. Nardin,Law, Morality, and theRelationsofStates,p. 9.
to "know";thatis,itbecame a meansforgenerating and diffusing "information"

about a commoninterestin avoidingnuclearwar. Thus, an overlappingset of
epistemiccriteria,togetherwithconvergenceon a commonpractice,enabled
the superpowersto develop a coordinationgame and to discoverthe extentto
whichits symboliccontentssuggestedcompromises,limits,and regulations.I
cannot improve on Schelling's pertinentobservations:"The players must
bargaintheirwayto an outcome.... Theymustfindwaysof ... communicating
theirintentions.... The fundamental psychicand intellectualprocessis thatof
participatingin the creationof traditions, and the ingredientsout of which
traditionscan be created,or the materialsin whichpotentialtraditionscan be
perceivedand jointlyrecognized,are notat all coincidentwiththemathemati-
cal contentof thegame.'"165
The countriesat the receivingend of ideas, whichbecome the targetfor
strategicpersuasionduringthe prenegotiationand negotiationprocesses,will
allowthemselvesto be constrainedbymutualinjunctionsonlyto theextentthat
threeconditionsare met. First,the policyproposals and the normativeand
epistemicunderstandings being diffusedmustbe interpretedas advancinga
shared interestin avoidinga particularoutcome,such as a nuclearwar or an
environmental disaster.166Second, the proposalsmustcreateopportunities for
advancingothernational,political,military-strategic, and economicinterests.
The positiveexpectationof furthering all these interestswill tend to increase
the value of cooperation,affectthe calculationof risk,and, overall,induce
cooperation.Third,the parties mustbecome conscious of theirinterdepen-
dence and its implications.An awarenessof limitson independentbehavior
stemsnaturallyfromchangesin technology, thebalance ofpower,and political
and economic conditions.But it also resultsfromthe interpretations that
people giveto thesechanges.
Those, then,who develop the originalexpectations,who really"create" the
political intereststhat spur motivationtowardthe forgingof a regime,are
creatinga regimepotential.The expectationscreated by the arms control
epistemiccommunity were thusa necessarycondition,thoughcertainlynotthe
onlycondition,forthe forgoingof the ABM regime,and theypreceded rather
than followedthe units of effectivemodification-namely,the creation of
normativebehaviorpatternsand the formalcreationof the regime.167Indeed,
internationalnorms,rules,and decision-making proceduresexpressonlytacit
or explicitcollectiveunderstandings and the theoreticalexpectationsthatare
transformed into practicesof governmentand externalizedto other nation-
states.
pp. 106-7.
ofConflict,
165. Schelling,TheStrategy
166. See ArthurA. Stein,"Coordinationand Collaboration:Regimesin an AnarchicWorld,"in
Regimes,pp. 125-27.
Krasner,International
Regime:AnomaliesforContempo-
167. See Roger K. Smith,"Explainingthe Nonproliferation
Organization
raryInternationalRelationsTheory,"International 41 (Spring1987),pp. 253-81.

The Emergence of Cooperation National Ep

Uploaded by

Copyright:

Available Formats

The Emergence of Cooperation National Ep

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Emergence of Cooperation National Ep

Uploaded by

Copyright:

Available Formats

The Emergence of Cooperation: National Epistemic Communities and the International

Evolution of the Idea of Nuclear Arms Control

An Americanepistemiccommunity played a keyrole in creatingthe interna-

For theircommentsand insights,I am gratefulto the membersof the reviewcommitteeof

International 46, 1, Winter1992

to stabilize the nuclear balance througharms control.Energized by their

3. See WesleyW. Posvar,"The New MeaningofArmsControl,"AirForceMagazine,June1963,

avoidingnuclearwar.Once nucleararmscontrolbecame conventionaland was

6. See RobertD. Putnam,"Diplomacyand DomesticPolitics:The Logic ofTwo-LevelGames,"

organize,structure, and coordinateinternational behavior.9 In Weber'stheoret-

9. Structurationtheory,as definedby Wendt,is "a relationalsolutionto the agent-structure

patternednormativebehaviorof two or more statesthatresultsin part from

11. This framework is partlyinspiredbyStephenToulmin'sdiscussioninHuman Understanding

tained,ratherthanbeinga depositoryof activity and authority."'3 The decisive

13. John G. Ruggie, "Changing Frameworksof InternationalCollective Behavior: On the

The "imaginary" science of nuclear strategy

16. See Posvar,"The ImpactofStrategyExpertiseon theNationalSecurityPolicyoftheUnited

sharingof premisesand expectations,or "theories,"creates the "evidence"

19. Aaron Wildavsky,"ChoosingPreferencesbyConstructing Institutions:

theory);and (5) moral value (the consequentialistethical standardwherein

Since thedawnofthenuclearage, twointellectualcommunities and twosetsof

22. These twocommunitieshave been the most,thoughcertainlynot the only,influential ones

about the reciprocalfearof surpriseattackand crisisstabilitythatbecame the

24. RobertJervis,"Arms Control,Stability,and Causes of War," Daedalus 120 (Winter1991),

differenttheoriesof war and therefore,in practice,theirpolicyprescriptions

and Causes ofWar," p. 173.

Posvarwroteat thetime,"One mightevenquestionwhethertheterm'schools'

28. Posvar,"The New MeaningofArmsControl,"pp. 39-40.

such as Paul Nitze,who linkedthe community withgovernment institutions,34

Neumann and Oscar Morgenstern38-tomake deductions and predictions

testinggroundfortheirideas-a paradigmaticcase, so to speak,thatcould be

froma professionalcode.48Indeed, this communitycan be described as a

48. Licklider,ThePrivateNuclearStrategists, pp. 119-22,130,and 135.

55. Talbott,TheMasteroftheGame,p. 70.

agreementwould gravelydamage nationalsecurity.... Like mostshiftsin

The ideas that the strategistsand scientiststook to the SurpriseAttack

64. Sims,"The DevelopmentofAmericanArmsControlThought,"chap. 5.

and thusmade people thinkhardabout unorthodoxsolutions.For example,he

69. See BartonJ.Berenstein,"Introduction,"in Hawkins,Greb,and Szilard,Towarda Livable

Early work on arms controlintensifiedin the 1950s with the advent of

itselfveryvaluable, as it gave an opportunity forbetterunderstanding of the

77. Rotblat,HistoryofthePugwashConferences, pp. 14-15.

Sovietson sucha ban werein factinitiatedbeforethecrucialeventsof 1958-60.

83. Schelling,TheStrategy p. 207.

seminaron arms controlthat met more or less continuouslythroughoutthe

87. See Thomas C. Schelling,"What Went Wrong with Arms Control,"ForeignAffairs65

The selectionprocess startedunderEisenhower,and by the timeKennedy

90. Firestone notes that the Committeeof Principalswas "a high-levelinteragencygroup

The creationof ACDA provideda home forthe armscontrolcommunity, a

1963,became McNamara's "generalcounselorand chiefaide on armscontrol,"

elementarymeans of communicationdid not alreadyexist."The hotlineidea

101. My discussionof the hotlineidea is based on WilliamL. Ury'sBeyondtheHotline:How

Once McNamara was persuaded thattherewas no technicalsolutionto the

studyan ABM moratorium and to getthePentagoninterestedin theidea.110 In

110. York,MakingWeapons,TalkingPeace, pp. 222-23.

special 1960 Daedalus issue. But relationsbetweenthe epistemiccommunity

114. Adams,BallisticMissileDefense,p. 186.

Armscontrollerslost the ABM battlein Congressby one vote, and a new

118. Safeguard incorporatedboth area and terminaldefense capabilities,using the same