Emerging Propety

The End of the Cold War: Predicting an Emergent Property
Author(s): Bruce Bueno de Mesquita

Source: The Journal of Conflict Resolution, Vol. 42, No. 2 (Apr., 1998), pp. 131-155
Published by: Sage Publications, Inc.
Stable URL: http://www.jstor.org/stable/174566 .
Accessed: 04/02/2014 11:38
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Sage Publications, Inc. is collaborating with JSTOR to digitize, preserve and extend access to The Journal of
Conflict Resolution.
http://www.jstor.org
This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

All use subject to JSTOR Terms and Conditions
The End of the Cold War
PREDICTING AN EMERGENT PROPERTY
BRUCE BUENO DE MESQUITA

Hoover Institution,StanfordUniversity
Gaddis claimed that internationalrelationstheory failed to predictthe Gulf War,the Soviet Union's
collapse, and the cold war's end. Subsequently,he acknowledgedthat the expected utility model captures
the logic behindcomplex adaptivesystems such as the cold warinternationalsystem. Thatmodel correctly
predicted two of the events to which Gaddis pointed. Here, that model is used to simulate alternative
scenarios to determinewhetherthe cold war's end could have been predictedbased only on information
availablein 1948. The simulationsshow a 68%to 78%probabilitythatthe UnitedStateswould win the cold
warpeacefullygiven the conditionsin 1948 andplausibleshifts in the attentivenessof each stateto security
concerns over time. The analysis demonstratesa rigorousmethod for testing counterfactualhistories and
shows thatthe pro-Americanend to the cold war was an emergentpropertyof the initialpost-WorldWarII
conditions.
The end of the cold war is a turningpoint not only for the world of affairsbut also
for the smaller world of internationalrelationsresearch.Decades of scholarshipby
areaspecialists, historians,philosophers,statisticians,andpunditsfailed to provide a
clear, explicit predictionthat the cold war was aboutto end. In light of this apparent
failure,Gaddis(1992) arguedthatinternationalrelationstheorywas a failure.Gaddis
challenged the core of internationalrelations theorizing, stimulating a lively and
sometimesheatedexchangeaboutwhetherinternationalrelationstheorywas a failure.
That debatechallengedthe very notion thatthe absenceof explicit predictionsabout
the end of the cold war was germaneto the questionshe raised.In light of a portion
of that debate,Gaddisrevised his assessment,not on the relevanceof predictionbut
ratheron the prospect of successful predictionthroughtheory building. In a recent
study, Ray and Russett (1996) reportedthat, in privatecorrespondence,Gaddis has
agreedthatthe so-called expected utility model "strikesme as an importantadvance
over earlier approachesto predictive modeling because it takes into account the
emergentpropertiesof complex adaptivesystems ... [and]there has been a sort of
AUTHOR'SNOTE:I am indebtedto JohnGaddisfor suggestingthis studyto me andfor his generosity

of spirit. James Morrow and Kiron Skinner provided helpful suggestions, and Gary Slack provided
invaluableprogrammingassistance.
JOURNALOF CONFLICTRESOLUTION,Vol. 42 No. 2, April 1998 131-155
? 1998 Sage Publications,Inc.
131

132 JOURNALOF CONFLICTRESOLUTION
Bueno de Mesquita-JohnLewis Gaddisconvergence"(takenfromGaddis'se-mail to

James Ray, February8, 1995, and quotedin Ray and Russett 1996, 1578).
The purposeof this studyis to elaborateon thatconvergenceby using the so-called
expected utility model to see whether the structureof relations among states after
WorldWarII containedan emergentproperty-the seeds for the unfoldingdevelop-
ment and end of the cold war. By this, I mean that the exercise undertakenhere is
concernedto see, in a probabilisticsense, what the likely evolution of international
relationslooked like given the characteristicsof the pre-coldwarbut post-WorldWar
II period.I will show thatone can demonstrate,only frominformationaboutthe world
in 1948, thatit was highly likely thatthe UnitedStateswould emergeas the victorover
the Soviet Union in a peaceful resolutionof the cold war.
National leaders are, in the approachI take, assumed to adapt strategicallyto
changing circumstancesby choosing the course of action they believe at any given
time is best for them (and,by extension,theirnation).Theirbehavior,then,is adaptive
and complex. It is adaptivein that behaviorrespondsto shifting circumstances.It is
complex in thateach state (or leader)has a strategicinterestin structuringthe way in
which the environmentchangesto encourageothersto adaptin a mannerfavorableto
the statein question.In particular,butnot exclusively,the UnitedStatesandthe Soviet
Union, as the main protagonistsin the cold war, each had an interestin shaping the
adaptivebehaviorof all otherstates.Each, of course, hoped to promotea situationin
which the internationalsystem evolved in its favor.
I will show that by simulatinga wide arrayof plausible counterfactualhistories,
we can see thatthe initialstartingpointof the cold warhadembeddedin it an emergent
property.Specifically, the probabilityof an American cold war victory was much
higherthanthe probabilityof a Soviet victorygiven the initialcold warconditionsand
rationallycomplex, adaptivebehavior.The resultsof my analysis supportthe conten-
tion that critical aspects of the cold war's end were predictableand that many of its
specific characteristicswere highly likely butnot inevitable.It suggests a reorientation
of discussionaboutpathdependencewhile highlightingthe importanceof some policy
choices over others.'It providesa means to view counterfactualhistoriesin a rigorous
way and suggests policies by which the cold war could have ended sooner or could
even have been avoided altogether.It also implies that we have a tool that could be
informativein simulatingthe evolution of, for instance,relationsbetween the United
States and China or the United States and the Islamic world over the next 50 years,
but these areissues best left for futurestudy.
This analysisis being doneretrospectively,andthereforesome commentis required
on the claim thatwhatis presentedhere serves as evidence thatthe end of the cold war
was predictable.Some might be skeptical, fearing that I have engaged in Monday
morningquarterbacking. Hindsight,being 20/20, would not be an interestingbasis for
trying to salvage model constructionand internationalrelations theorizing. Others
might be skeptical, fearing that the simulationspresented here mask the supreme
difficulty really predictinga specific event in advance.To allay these concerns,I
of
1. By pathdependence,I meanthe extentto which a particularoutcomecould only have arisenthrough

one uniquecombinationof historicalcircumstances.

Buenode Mesquita/END OF COLD WAR 133
first demonstratethat the expected utility model has predicted many specific and
difficultto anticipateevents well in advance,includingseveralspecific events Gaddis
pointed to as exemplars of the failure of internationalrelations theory. Then, after
explainingthe intuitionbehindthe model's logic, I go on to show thatno hindsightis
used in constructingthe simulationsof the end of the cold war discussed here.
Gaddis's(1992) articlepointedto the failureto predictthreeevents as evidence for
the inadequacyof currentinternationalrelationstheory.He wrote,
Theabruptendof theColdWar,an unanticipated hotwarin thePersianGulf,andthe

suddendisintegration of theSovietUnionastonished
almosteveryone.... Thefactthat
they aroseso unexpectedlysuggeststhatdeficienciespersistin the meansby which
contemporary princesandthesoothsayers theyemployseekto discernthefuturecourse
of worldaffairs.(P.5)
AlthoughI wholeheartedlyagreethatdeficienciespersist,I wish to establishthe prima

facie basis for my claim that reliable prediction of specific internationalevents is
possible by quotingfrom an interviewin which I said,
Applyingmytheoryon whynationsgo to war... I willgo outon a limbandciteplaces

whereI thinkwaris likely:... warbetweenIraqandSaudiArabiaorbetweenIraqand
otherstatesontheArabian Peninsula waris settled.("AConversation"
oncetheIran-Iraq
1982,30)
The Iran-IraqWarended in mid-1988, and of course the hot war in the PersianGulf
that Gaddis referredto as unanticipatedbegan in the summerof 1990 with the Iraqi
invasion of Kuwait.So, at least one of the threeevents to which Gaddisreferredwas
publicly predicted,includingthe timingof the event, well in advanceusing the model
appliedhere.
The prediction of the Gulf War was not a solitary, isolated, correct prediction.
Indeed,representativesfrom the U.S. governmentwho use an in-house version of the
expected utility model (called Factions) braggedto the Russians in 1995 about its
effectiveness, as reportedin Izvestiya.Thatarticlenotes,
Expertsengagingin studieswithintheframework of thissystemstatethaton thebasis

of longexperienceof usingit, it canbe saidwitha greatdegreeof confidencethatthe
Inparticular,
forecastsarehighlyaccurate. accordingtothem,the"Factions"methodwas
usedin May1991to predicttheAugustputsch.("RussiaIs Doomed"1995)
The putsch or coup was the crisis thatprecipitateda second of Gaddis'sexemplarsof

unpredictedcritical events that was anticipatedwith the expected utility model: the
collapse of the Soviet Union.
Since the introductionof this model, there have been dozens of applicationsto
real-time predictions in published articles. These have been readily accessible to
researchers,and some have been independentlyevaluated.For instance, an articlein
the Far EasternEconomicReview (McGurn1996) assessed the 12 predictionsmade
in Bueno de Mesquita,Newman, andRabushka(1985) and concluded,

Basedon theirfindings,theypredicted12 developments, of whichonly one proved

inaccurate(a technicalityabouthow landwouldbe valued).Theothersprovedpretty
muchonthemark:increased welfarespending,thegrowingimportanceof demonstrated
"loyalty"to Chinain the awardof key contracts,andan "ongoingreinterpretation"
of
theJointDeclaration'sprovisionslongaftertheyweresigned.(P.68)
The methodologyhas been explainedin detail in numerouspublicationsso thatits

workings are readily accessible. Although the model has been modified-and, one
hopes, improved-over time, its core assumptions, data requirements,and basic
computationshavenot changed.Rather,it has beenmadedynamic,whereasits original
incarnationprovided only one-shot, static predictions.The dynamic version of the
model also has receivedwide coveragein the academicliterature,includingtwo books
(Bueno de MesquitaandStokman1994;Bueno de Mesquita,Newman, andRabushka
1996) and numerousjournalarticles.Those seeking the least technicalexplanationof
this model shouldconsultchapter5 of Red Flag over Hong Kong (Bueno de Mesquita,
Newman, and Rabushka 1996), whereas those inclined to study a more technical
presentationmight examineEuropeanCommunityDecision Making(Bueno de Mes-
quita and Stokman 1994). For an independentassessment of the model's general
accuracyacross many applications,see Feder (1995).
AN INTUITIVE EXPLANATION OF THE MODEL
In this section, I providean intuitivesense of how the model works.Those readers

who are interestedin the technicalitiesof the model are stronglyurgedto consult the
referencedliterature.Given space limitations,it is not feasible to present a highly
detailed descriptionof the model here. Thereis, however, no desire to leave readers
with the sense that the model is a black box. The following discussion should help
form a view of how it works, and the more technicalpublicationsthat are referenced
provide all the detail necessary for those who want a thoroughunderstandingof the
model.
The model predictshow the policy choices of people, groups,and nationschange
over time and how those choices influence policy decisions. It is, then, an applied
model, reflectinga balancebetweenpureformalismand assumptionsthatensurethat
the model can be estimatedfor a wide varietyof real-worldproblems.This method-
ology representsan attemptto marrythe invaluableexpertiseandnuancedknowledge
of area specialists with the theoreticalinsights and tools developed by social choice
theorists and game theorists.It is a linkage between substantivespecializationand
expertise in how people make decisions in an attemptto model complex adaptive
systems and identify theirlikely evolution throughtime.
Policy makinginvolves interestscompetingto influencechoices. Political leaders,
military officials, bureaucrats,businessmen, foreign governments, ethnic groups,
otherspecial interests,andordinarypeople areamongthose who wantto shapepolicy.
These groups differ in the intensity and influence that they bring to decisions. They
also differin whattheywant.In the applicationin this study,theinterestedstakeholders

Buenode Mesquita/END OF COLDWAR 135
areidentifiedas nation-states,but the methodologyitself need not be restrictedto this

coarse a view of any polity. It is straightforward to breakopen each state and expand
the analysis to include any numberof domestic interests,as has been done in many
otherapplications.
Influencedependson controlover resources.Politicalinfluenceis used to alterthe
policy wishes of other groups. The object of each stakeholderis to build a strong
coalition in favor of the choice desired by that individual and his or her backers.
Building such a coalition can involve some problems.First, people might disagree
aboutwhatwould be the best decision.If these disagreementsexist, thenit can be hard
to find supporters.Second, some competitorsare betterat shapingdecisions than are
others.This is truebecausethey controlthe rightresourcesor mustermore supporters.
Third,everyone comparesthe costs and benefits of getting what he or she wants and
of being seen as a deal maker.Politics is the art of the practical,but not everyone
involved in politics is inclined to be practical.Some people believe so deeply in the
policy they favorthatthey arenot willing to compromise.They careonly aboutgetting
whatthey want. I call gettingwhat one wantspolicy satisfaction.Otherscare so much
about being deal makersthatthey will supportany decision that they think they can
win. These people value being partof the winning side as being more importantto
them than what they actually win. I call the desire to be part of the winning side
regardlessof outcome political satisfaction. Most people fall in between these ex-
tremes. They are willing to make concessions to help forge a deal, but they are not
willing to give up everythingthey believe in. Political bargainingdepends on how
willing people are to give up the policy they most preferto make a deal. The essence
of all politics is embodiedin the willingness of competitorsto tradebetween policy
satisfactionand political satisfaction(Lamborn1991).
Politics involves cooperationandcompetition.Groups,whetherformalor informal,
propose policy compromisesor try to coerce opponentsinto doing what they want.
When stakeholdersdecide on a course of action, they evaluate an elaboratearrayof
choices. Among these is the optionto do nothing.If theythinktheireffortsat changing
existing circumstanceswill fail, then they live with the currentpolicy or status quo.
The processis like a very complicatedcardgame.The decision to live with the current
policy is much like folding in the gameof pokerwhenone has been dealta particularly
poor hand.
In politics, it is as if each player is dealt a hand. The practicalvalue of the hand
when played depends on the influence, attentiveness,and policy objective of each
player.Strongerplayers (or those with strongbackingfrom others)drawbettercards
thando weakerplayers.Some playerspay closer attentionto theircardsthando others.
This influences theirperceptionsof the situation.Based on theirinfluence, attentive-
ness, and policy objectives,each decision makerformsperceptionsabouteach rival's
hand.With thatinformation,each playerdecides on a strategyof policy proposalsto
make to some or all of the other players. The content of each proposal is a specific
suggestion about a policy that a stakeholderis willing to support or for which a
stakeholderis seeking support.The proposedpolicy need not be the same as whatthe
proposercurrentlysupports.

After all the players submittheirsecretproposalsto one another,each reviews the

hand-the proposals-that holds. Some proposalsarebetterfor the recipientthanare
others.Indeed,some proposalsareperceivedas not credible;thatis, the recipientdoes
not believe that the proposercan mobilize a sufficiently powerful coalition to make
the recipientaccept the proposedpositional change if the recipientdoes not want to.
Otherproposals seem sensible but fall by the wayside because a superioroffer was
made by a different stakeholderwho has a strong enough coalition to fend off the
coalitions thatare backingrival proposals.Each player would like to choose the best
offer made to him or her, and each proposerenforces his or her bids throughthe use
of power to the extent that he or she can. Each actor selects the best offer he or she
receives-or the least bad one.
At the end of a round of proposal making, players learn new information.By
monitoring responses to their proposals, players learn how much leverage they
(togetherwith their coalition of supporters)can exert on otherdecision makers.If a
proposal is accepted, then a player learns that he or she made the best offer to the
recipientof the acceptedbid. If a proposalis rejected,then the decision makerlearns
that an alternativeproposalmustereda sufficientlypowerfulcoalition to fend off the
decision maker'sefforts.
When the players finish sorting out their choices, each shifts to the position
containedin theproposalhe or she accepted(if any). Of course,when a decision maker
agreesto a compromisewith someoneelse, thatplayerhopes thatthe otherplayeralso
will live up to his or her end of the bargain.Alas, politics involves promises that are
not binding.Proposalsare enforcedif a decision makerhas the strengthor supportto
make sure thatothersdo what they promisedto do. Decision makersin the model, as
in reality,are free to renege on a proposeddeal if they have a betteralternativeor so
long as someone else forces an agreementon them.
To predictwhat proposalsare made, which deals are accepted,and how decisions
are really made, the expected utility model focuses on trade-offs between policy
satisfaction and political satisfaction. Everyone cares about both, but to different
degrees. Remember,policy satisfactioninvolves getting a decision that is consistent
with the stakeholder'spolicy preference.Political satisfactioninvolves makingdeals
and being partof the winning team.Policy satisfactionoften comes at the expense of
political satisfaction, and political satisfaction generally involves giving up some
policy satisfaction.I model the trade-offbetween these two types of satisfactionby
estimating the shape of each player's indifferencecurve with respect to its current
political andpolicy satisfactionandthe location,relativeto thatindifferencecurve, of
proposed deals. Of course, deals that move players to higher indifferencecurves are
preferredto thosethatleave thembelow or on theircurrentindifferencecurve,although
at differentpoints on it.
Leaderspursuethe biggest benefits at the lowest cost. Certainly,they do not make
choices that are expected to leave them worse off than the statusquo. In evaluating
proposals, decision makers consider how much policy satisfaction and political
satisfaction each provides. Ideally, any decision maker would like to have infinite
policy satisfactionand infinite political satisfaction,but realistically,it is likely that
some trade-offhas to be madebetweenthe two. The problemis to figureout how much

of one the decision makerwill sacrifice in exchange for more of the other.I briefly
summarizethat decision process by sketchingthe basic intuitionbehindthe analytic
approach.Of course, the actualanalysis is quite a bit more complicated.The mathe-
maticaldetails of the model can be found in the works cited previously.
In the model, informationaboutresources,salience, andpolicy preferencesis used
to calculate the perceived relationshipbetween each stakeholderand every other
stakeholder.The model estimates what proposal,if any, is best to make to the other
participantsin the decision process. Sometimes, it is best to live with the status quo
and propose nothing.Othertimes, a compromisecan be struckthatmakes each party
to the deal betteroff. One stakeholdermight gain policy satisfactionat the expense of
political satisfaction,and anothermight gain political satisfactionat the expense of
policy satisfaction.So long as both perceive a net gain, the deal can be made. At still
other times, a stakeholderconcludes that it has the ability,either on its own or with
the help of others,to coerce an opponentinto acceptingits point of view. Sometimes
efforts at coercion result in the desired outcome, but other times coercion leads to a
fight, such as the Gulf War,in which both sides thinkthey are strongenough to secure
benefits from the otherside at acceptablecosts (Haselkorn1998).
In the model, as in real life, I do not assume that people always make the best
proposalsthey can. Rather,I assume thatpeople can make mistakesbut that they do
so while tryingto be as careful as possible with the limited informationthey have at
the time when choices aremade.In thatsense, I tryto takeinto accounthumanfrailties,
differences in perception,and shortsightedness,each of which can lead to errorsin
judgmentandto undesiredor unanticipatedoutcomes.I focus on the strategicchoices
stakeholdersmake before they know how things will turn out. I assume that each
stakeholdermakes what it believes is the best choice given what it knows at the time
a proposal or action must be taken. Hindsight cannot be allowed to color analytic
judgmentsif predictionsand explanationsaboutfuturedecisions are to be accurate.
APPLICATION OF THE MODEL TO

PREDICTING THE COLD WAR'S EVOLUTION
To apply the expected utility model retrospectivelyto the cold war withouttaking
advantageof ex post facto knowledge, I have undertakennumeroussimulations,all
based on the same initial data. The initial data, drawn from the Correlatesof War
project,include a list of the 36 most powerfulsovereigncountriesas of 1948. For each
country,there is an estimate of its 1948 nationalcapabilities and its position on a
securitydimensionscaled from -100 to 100.
The securitydimension is measuredin termsof the similarityin militaryalliance
portfoliosfor each pairof countriesas of 1948, using Correlatesof Waralliance data,
as explained in Bueno de Mesquita (1975, 1981); that is, each state can have any
possible mix of defense pacts, nonaggressionpacts (or neutralitypacts), ententes,and
no alliance relationshipwith each of the other states in the internationalsystem. A
security portfolio or military alliance portfolio is defined as the array of military
alliance agreementsa state has with all other states. The degree of shared security

interestsof any pairof statesis evaluatedas the correlation(using Kendall'stau b) of

their securityportfolioswith one another.For instance,by the height of the cold war,
the United States and the other members of NATO shared very similar security
portfolios with one another.They tendedto form mutualdefense pacts with the same
sets of states, and they also tendedto avoid alliances altogetherwith the same set of
states. Thus, if I were taking advantageof actualinformationabout the similarityin
alliancecommitmentsin, say, 1965, thenthemeasurementprocedureI use would show
a high degree of similarity of interests or correlationon the security dimension
between, for instance,the United States and the United Kingdomor the Netherlands
and between the Soviet Union and Bulgariabut would show a low level of similarity
or correlationbetweenBulgariaandthe UnitedKingdomor betweenthe UnitedStates
and the Soviet Union, and so forth.
The measure,as used here, reflects the focus in 1948 on security concerns while
capturingthe world before the clear emergence of a bipolar structurebetween the
friendsof the UnitedStatesandthe friendsof the Soviet Union. This measurehas been
widely used in numerousstudies in the literatureon securityand, althoughcertainly
imperfect,has beendemonstratedto varyin theoreticallypredictedways acrossa large
arrayof security-relatedconcerns.The initialsecuritypolicy score for each simulation
is calculatedby examining the differencein the similarityof alliance portfolios for
each countryvis-a-vis the UnitedStatesandthe Soviet Union. A scoreof 100 indicates
a securitypolicy most like thatof the UnitedStates.A scoreof -100 indicatesa security
policy most like thatof the Soviet Union. A score of 0 indicatescomplete neutrality
betweenthe two sides, with negativevalues beingincreasinglypro-Sovietandpositive
values being increasinglypro-American.This normalizedscale reflectspolicy prefer-
ences, not political or militarypower per se.
As of 1948, sides had not yet been clearly drawnin the cold war struggle so that,
as the datareflect,the centraltendencywas towardneutralityin choosing between the
United States and the Soviet Union. It is changes in these scores over time thatI am
interestedin predicting,especiallychanges in the score thatreflectthe policy position
at which power was balancedbetween the competingworldviews at any given time.
Of course,I am equallyinterestedin predictingchangesin the policy preferencescores
for the United States and the Soviet Union. The balanceof power policy position is
definedhereas the securitypolicy position(between-100 and 100) for which it is true
that as much aggregatepower falls above that securityposition as falls below it; that
is, the balanceof power securityposition is the position of the statethatis the median
power, whether we sum power from below or from above on the security policy
dimension.
Capabilitiesaremeasuredusing the Correlatesof Warcompositecapabilitiesindex.
This widely used indicatoris the average of six components:share of the system's
militaryexpenditures,shareof the system's militarypersonnel,urbanpopulation,total
population,steel production,andcommercialfuel consumption.Capabilityscores are
normalizedso thatthey sum to 100.2It shouldbe noted thatbecause the startingdate
is 1948, neitherJapannorEast or West Germanywere sovereignstatesand so arenot
2. They actuallysum to 99.692 due to rounding.

includedin the analysis.If anything,this introducesa pro-Sovietbias into the analysis

becauseJapanandWestGermanycertainlyemergedas more powerfulstatesthandid
East Germany.The capabilitiesdataarefurtherbiasedagainstthe findingherethatthe
United States was especially likely to win the cold warbecausethe Correlatesof War
compositecapabilitiesindex is knownto overweightboththe Soviet Union andChina.
This overweightingresultsfromthe fact thatironandsteel productionhaddiminished
in importanceas indicatorsof power in the post-WorldWarII era and also because
both had large but relativelypoor populations.
Salience for nationalsecuritypolicy is set equal to 100 for each countryin 1948.
This reflectsthe situationat thattime when the early signs of the cold warwere on us.
The year 1948 is chosen because it is a key dividing year.The WorldWarII alliance
among the United States, Britain,and the Soviet Union was clearly shatteredby the
conflict overIranin 1946, disagreementswith the SovietUnion overthe fate of Eastern
Europeand over the governanceof dividedGermany,andso forth,but neitherNATO
northeWarsawPactwould emergeto define the centralparametersof the international
order for the next 40 or so years. Of course, by 1948 there were bilateralalliances
between the Soviet Union and what becamethe membershipof the WarsawPact, but
no multilateralagreementswere yet in place.
To be sure thatthe resultsof my analysisdo not dependon the selection of 1948, I1
have replicatedthe process using 1946 and 1947 as base years andalso have analyzed
expandedandreducedsets of statesas the stakeholders.The resultsarethe same. The
model being used predictschanges in policy positions-that is, supportivenessof the
United States or the Soviet Union-in response to the pulls and tugs of competing
coalitions of interest.In the simulationsrun for this study,the model was allowed to
iterate over 25 periods (which we can think of as each being 2 years in duration,
although the model is not instructiveon calendartime), with the salience variable
permittedto changerandomlyfor eachcountryduringeachround.3Therandomization
of salience is intended to capturethe range of possible fluctuationsin the relative
importanceof securityissues comparedto otherissues including,for instance,domes-
tic political concerns, economic policy, humanitarianconcerns, and the like from
periodto periodandfromcountryto country.It allows the possibility to test the model
while controlling,in a sense, for the potentialimpactof exogenousrandomshocks that
altertherelativeimportanceof securityissues fromstateto stateandfromtime to time.
The introductionof randomsalience is an innovationfrom earlierapplicationsof
this model. One problem in doing long-termanalysis, whether it is postdiction or
real-timeprediction,is thatthe likelihoodof an exogenous randomshock to the initial
data increasesover time. A centralproblemin long-termassessmentsis anticipating
the seemingly unpredictable.For instance,the assassinationof PresidentJohn Ken-
nedy in 1963 certainlycould not have been foreseen in 1948. Nor could one readily
foresee, and so model, the effects of the launchof Sputnikin 1957, the suddendeath
of Prime Minister Lal BahadurShastri in India in 1965, Watergate,the Profumo
3. Given thatthe model does not attachcalendartime to its roundsor iterations,this is one place where
hindsightcan be said to have been used. It is a convenienceto speak of each roundas if it were of 2 years
duration,but the model is not informativeon this. Given the natureof the data,we can be confidentthata
roundreflects no less than 1 year but might reflect 2 or more years.

scandalin England,and so forth.Each of these events could potentiallyinfluence the

attentivenessof a governmenttowardsecurityor domestic affairs.Such shocks arise
more or less randomlyin time andspace.By randomizingsalienceacrossstakeholders
(space) and iterations(time), the model allows us to estimate a wide arrayof such
unanticipatedevents. In so doing, we can see whetherthere are strongpushes in the
general configurationof the data that generallyfavoredthe continuationof the cold
war or an American or a Soviet victory. Indeed, this innovation appearsto be an
importantstep in devising long-term counterfactualscenarios as a means to test
alternativestates of history,both past and future.
The methodology appliedhere does not attemptto replicatethe actualhistory of
the past 50 years.Thatwould be subjectto ex post facto biases basedon knowing how
things turnedout. Instead,the model simulates alternativehistories to see to what
extent the actualend of the cold war was strictlypath dependentand to what extent
the peacefulcollapse of the Soviet Unioncan be seen to have emergedfromthe starting
conditions in 1948 given an arrayof alternativehistoricalpaths from 1948 forward.
Eachrandomdistributionof salience scores, of course,representsa criticalelementin
a differentpath thathistorycould have taken.Indeed, we can see whetherany of the
100 simulationsI have run comes close to replicatingthe actualexperiencebetween
1948 and the present,and we can see whetherthe preponderanceof paths led to an
Americanvictory,a Soviet victory,or the continuationof the cold war.Note that the
objective is not to predictthe precise details of the actualpath of circumstancesand
choices thatled to the end of the cold warbut ratherto determinewhetherthatprecise
path was partof a preponderantset of paths favoringAmericanvictory. Specifically,
through the randomizationof salience, we can see whether the American victory
depended on a unique arrayof circumstances,a small bundle of possible circum-
stances, or a very broadrangeof feasible conditions.As the evidence makesclear,the
end of the cold war was one of a numerousandpreponderantset of prospectivepaths
that historycould have traversedand thatled to Americanvictory.The paths leading
to otheroutcomes arefar less numerousand, in that sense, less probable.
It is importantto recognize thatnone of these simulationstakes advantageof any
informationthatcould not, in principle,have been known in 1948. The datacertainly
were readily knowable. The model, of course, did not exist in 1948, but it makes no
use of information about internationalpolitics after that year. For instance, the
simulationsdo not updatethe nationalcapabilitieseach year because those changes
are partiallyproductsof the cold war ratherthanits antecedents.The model predicts
shifts in nationalpositions on the pro-U.S./pro-Sovietscale from iterationto iteration
based only on its internallogic, the 1948 data,and randomfluctuationsin the values
of the salience variable.The initial dataassumptionsaredisplayedin Table 1.
PREDICTING THE COLD WAR'S END
Before turningto the analysis, let me say a word about how the results are to be
interpreted.Eachof the 100 simulationsthatwererundevelops an alternativeplausible
scenarioor counterfactualhistoryof the post-1948 years. The methodI use does not

Buenode Mesquita END OF COLDWAR 141
TABLE 1
InitialData Conditions, 1948: Pro-Americanor Pro-SovietPolicy Preferences
Country Capabilities Position
Argentina 0.972 89.6

Australia 0.889 1.3
Belgium 1.182 2.8
Brazil 0.993 89.6
Bulgaria 0.345 -100.0
Canada 1.610 56.2
China 11.941 1.3
Czechoslovakia 1.401 -91.0
Denmark 0.240 1.3
Egypt 0.408 3.2
England 7.863 3.6
France 3.597 2.8
Greece 0.418 1.3
Hungary 0.450 -91.0
India 2.468 1.3
Iran 0.491 2.4
Iraq 0.157 3.7
Israel 0.125 1.3
Italy 2.426 1.3
Mexico 0.774 89.6
Norway 0.230 1.3
The Netherlands 0.836 2.8
Pakistan 1.485 1.3
The Philippines 0.408 1.3
Poland 3.273 -91.0
Romania 0.606 -91.0
SaudiArabia 0.125 2.8
SouthAfrica 0.680 1.3
Soviet Union 18.256 -100.0
Spain 1.683 1.3
Sweden 0.648 1.3
Syria 0.104 2.8
Thailand 0.414 1.3
Turkey 1.347 2.4
United States 29.956 100.0
Yugoslavia 0.891 -100.0
NOTE:Salience initiallyis set at 100 to reflect securityconcernsandthen varies randomlyfromcountryto

countryand from model roundto model round.
providean ex antefacto way to choose one scenarioover anotheror say thatthe chosen
scenario is the predictedpath of the cold war.Of course, we can, ex post facto, find
scenarios that look very much like actualhistory,and I will point to such scenarios
(andothersthatdo not look like actualhistory)as I go throughthe analysis.These are
interestingcases thatsuggest a varietyof alternativeways in which the cold warcould
have ended or been prolonged.They are instructiveaboutforeign policy making for
the future,but they are not centralto the predictionsof interest.

The centralpredictionsare aboutthe likelihood that the cold war would continue
or would end with anAmericanor a Soviet victory.Predictionsaboutthese likelihoods
can be made with confidence by examiningthe relative probabilityof each of these
threecore possible outcomes.If the distributionof simulationoutcomes thatsupport,
for instance,an Americanvictoryand a Soviet victoryareaboutequal,then the model
will have failed to predict the centralfeatureof the end of the cold war, that is, a
peaceful victoryby the UnitedStates.Such an approximatelyequaldistributionwould
falsify my claim that this model could predict the end of the cold war, as would a
preponderanceof predictionsfavoringa Soviet victory.If, however,the vast majority
of simulationssupportthe claim for a peacefulAmericanvictoryin the cold war,then
the model will have succeeded in showing that the outcome of the cold war was a
predictableemergentpropertyof the initial conditionsreflectedin the datain Table 1,
with the actual path taken reflecting "noise" or a small perturbationaround the
preponderantcentraltendencyof the predictions.
I begin the analysis, then, with some summarystatisticsthat speak to the model's
predictionsregardingthe likely evolution of the cold war.Then, I examine selected
simulationsto illustratethe characteristicpaths the model suggests for the evolution
of the cold war.These suggest the main alternativehistoriesthatthe model indicates
were likely. The 100 simulationsallow us to estimate the probabilityof alternative
outcomes.The simulationsalso allow us to see the extentto which the end of the cold
warwas attainedoptimallyormight have been achievedearlierwith the same or fewer
costs. The simulationsalso allow us to speak of counterfactualstrategiesfor ending
the cold war and to comparethem to the course actuallyfollowed.
The 1948 data reflect a system of nations highly centralizedaroundneutrality,
uncertainof whetherto invest theirfutureswith the Soviet Union or the UnitedStates.
The initial locus of power was centeredat 2.4 on the policy scale. This locus reflects
the position of greatestnationalsecurityin the sense that at position 2.4, power was
evenly balancedbetweenthose morepro-Americanand those morepro-Soviet.From
a structuralrealist perspective, this can be described as the policy that maximized
security;it reflects a decision to balance one side against the other.From a spatial
modeling perspective, 2.4 representsthe location of the median power unit on the
securityissue dimensionin 1948 andis analogousto Black's (1958) medianvoter.The
1948 structurereflectsa stablesystemin which neitherthe UnitedStatesnorthe Soviet
Union had enough supportto defeat the neutralcenterso long as everyone remained
focused exclusively on security.Thus, a neorealistview, in which securityremainsthe
focus, could not predictthe end of the cold war.The cold war's end dependedon at
least some states being diverted from security concerns by other issues, whether
domestic or foreign.
The randomizationof the salience variablein the 100 simulationswas used exactly
to see what happensto the powerful neutralcenter as the securityconcerns of states
waned and waxed in responseto fluctuatinghypotheticaldomestic or otherpressures.
By allowing salience to vary on the security dimension, my analysis departsfrom
neorealistor structuralrealistprecepts.In thatview of the world, which is tested as a
separatesimulation(the 101st), securityalways is the paramountconcernof states, so

that domestic, economic, humanitarian,or otherconcernscould never become more

salient than security.
We can see how the waningandwaxing of securityconcernsmighthave influenced
the unfoldingcold war by looking at changes in the locus of the balanceof power at
different times across the simulations.By the end of the fifth iterationof the 100
randomsimulationsof the model, for instance,the policy stance thatrepresentedthe
averagepoint of maximalsecurityagainstdefeatby eitherextremewas 19.6. Remem-
bering that positive values reflect a pro-Americanstance, this indicates a decidedly
more pro-Americanworld by about 1958 than was true on average in 1948. The
average, however, can be misleading. The 100 simulationsinclude examples at this
stage for which the balance of power position is as pro-Soviet as -91.0 and as
pro-Americanas 99.7; thatis, a broadrangeof systemevolutionsaway fromneutrality
was feasible,includingquickAmericanvictoryoverthe Soviet Union andquickSoviet
victory over the United States.
By the 15thiterationof the model,roughlyequivalentto the year 1978, the position
thatbalancedpower on averageacross all of the simulationswas located at 12.5, still
considerablymore pro-Americanthan at the outset but not quite so much as in the
simulatedworld of 1958. This fits nicely with the emergencein the 1970s of detente,
includingthe recognitionof the Soviet Union as a more-or-lessequal superpower.By
the hypothetical average world of 1978, the variation in possible outcomes had
decreased,both in the sense thatthe most pro-Sovietoutcomewas now -84.9 instead
of-9 1.0 andthe most pro-Americanoutcomewas 86.5 insteadof 99.7 andin the sense
that the standarddeviation aroundthe averagepolicy outcome had droppedfrom 44
to 33 and continuedto dropthereafter.
Table2 summarizesthe overallpatternof evolutionof the emergingcold war.The
columns define four generic outcomes: a strong pro-Soviet outcome, a weak pro-
Soviet outcome,a weakpro-Americanvictory,anda strongpro-Americanvictory.The
weak outcomesaredefinedto reflectbalancesof power within 10 points on the policy
scale of the originalsituation(i.e., a weakpro-Sovietoutcomerangesfrom-7.6 to 2.4,
and a weak pro-Americanoutcome ranges from 2.4 to 12.4), with strong outcomes
more than 10 points above or below the initial condition.
Table 2 makes clear several importantgeneralizationsabout the likelihood of
Americanvictoryin the cold war.It is evidentthat,althoughthe system startedout as
neutral(100%of firstroundbalanceof powerpositions areat 2.4 on the policy scale),
it stronglytips in the pro-Americanposition in the vast majorityof cases. More than
two thirdsof outcomesin the simulationsarefavorableto the United Statescompared
to the initial condition.The Soviet Union was engaged in an uphill strugglefrom the
outset. Furthermore,the simulationsindicatethat the United States had a very good
prospectof emerging victoriousfairly early on-much earlierthan it ultimatelydid.
This suggeststhatthepolicies followed duringthe cold warby the UnitedStatesmight
have slowed its end or that the policies followed by the Soviet Union might have
prolonged the Soviet quest while ultimately proving futile. Interestingly,Gaddis
(1997) recently came to a similar conclusion. He noted that the United States, by
focusing on presumed(and knowingly exaggerated)Soviet militarypower over, for

TABLE2
Distributionof Policy Outcomes:The EmergentEnd of the Cold War
Pro- Weakly Weakly Pro-

Year
Round/Approximate Soviet Union Pro-Soviet Union Pro-UnitedStates United
States
Round5/1958 14 15 29 42
Round 15/1978 24 5 20 51
instance, economic might, might have unintentionallyprolonged the cold war. The
long peace, then, might have been due to poor strategizingratherthanany stabilizing
effects of bipolarity.In this regard,it is interestingto note that alreadyby the fifth
iterationof the model, the United Statesis predictedto gain a strongvictory over the
Soviet Union in 42% of the cases. How these more rapidvictories might have been
achieved and how the Soviet Union might have won are topics discussed in the next
section.
Table2 depicts a coarse view of the distributionof likely outcomes of the cold war
as estimatedthroughsimulation.Now we may considera somewhatmorerefinedview,
particularlywith regard to defining what constitutes a U.S. victory. The model
simulationslead to five generic sets of predictions:a slow Americanvictory,a quick
Americanvictory,a continuationof the cold war, a quick Soviet victory, and a slow
Soviet victory.Victoryis definedfor these purposesas arisingwhen the policy stance
of the United States, the Soviet Union, and the majorityof the remainingpowers in
the world converge (to, of course, a pro-Americanor a pro-Sovietposition). The cold
war is said to continue if such convergence does not take place, with the two
superpowersremainingon oppositesides of the neutralpositionof zero on the security
policy scale. Another view of victory is that the preponderanceof global powers
endorsethe Americanor Soviet perspective.Here,the notionis thateven if the Soviet
Union and the United States failed to converge,if enough powers favoredone or the
otherviewpoint,then the remainingsuperpowercould be viewed as marginalizedand
isolated. Figure 1 depicts such a perspective. It shows the cumulative probability
distributionfor the policy stance of the balanceof power in the global setting across
the 100 simulations.Figure1 displaysthe likelihoodof a pro-Americanor apro-Soviet
world by the final iterationof each simulation.
Figure 1 shows quitedramaticallythatthe cumulativeprobabilityof a pro-American
worldview or of Americanhegemony was extremelyhigh. Only 22% of the simula-
tions lead to predictedpro-Sovietcold warresolutionsby this definition,whereas78%
yield a pro-Americanworld outlookby the end of the simulations.This is a key view
of the emergent propertyof American hegemony. It is ironic that the simulations
capturethis-which certainlyhas been echoed in the real world-whereas much of
the academiccommunityin the 1980s focused on the apparentdecline of American
hegemony (Gilpin 1981; Keohane 1984; Krasner1981; but for a view consistentwith
the finding here, see Russett 1985; Kuglerand Organski1989).

.9 -
, .8-'
._ .7-
.>
.4
E -3 -122% Pro-USSR, 78% Pro-US

?4 .2 -
'
5 ,^ \
:-/.
-
0 - i- ...........
.... . ....... .
vi ..
-100 -50 0 50 100
Pro-Soviet<
7.------>Pro-U SA
Figure 1- The Evolution of the Balance of Power
Another viewpoint can be attainedby looking at examples of each of the five

generic evolutions of the cold war. In one such generic outcome-the dominant
one-the policy stancesof the UnitedStates,the Soviet Union, andthe majorityof the
remainingpowers in the world rapidlyconverge (say in fewer than 10 model rounds
or about20 years)to a pro-Americanoutcome.Figure2 illustratesan exampleof such
a case, which I identifyas simulation8. Figure3 (simulation19) illustratesthe second
generic outcome, which most closely mirrorswhat actually happened.The United
States andthe Soviet Union remainedfar apartfor a long time, ultimatelyconverging,
along with much of the rest of the world, to a decidedlypro-Americanposture after
about 20 iterations or 40 years. This is the next most common outcome in the
simulations.Figures4 and5 reflectmirrorimage outcomesin which the United States
relatively quickly (simulation 38) or slowly (simulation 58) adopts a pro-Soviet
position along with the majority of the rest of the world. Each of these occurs
infrequently.Finally,Figure6 (simulation96) shows a decidedlypro-Americanworld
in which, nevertheless,the Soviet Union andmany of its predictedallies (essentially
the WarsawPact countries)remain aloof from the pro-Americansystem. Figure 6,
then, illustrates an example in which the cold war continues and reminds us that
Americanvictory was noatatall inevitable (see also Greenstein 1996; Zelikow and

1 00-s A US Position
75o
75J \
e? ^A /r- --c
EL 25 /
t
o l--. .7v
-3 - Balance of Power Position
Z -I
-25
> -50 - / Soviet Position
-75
-100 _-
-100
I-' . ..-. ...... i
. .. . ..
1948 1958 1968 1978 1988 1998
Simulated Year
Figure 2: A Quick American Victory: Simulation 8
Rice 1995). Thepredictedhistoriesundergirdingthesefiguresareexplainedin the next

section.
These five generic cases remind us that the balance of power could shift to a
pro-Americanor pro-Sovietposition without Soviet-Americanconvergence.This is
importantbecause the simulationsindicate that there were significantopportunities
for the Soviet Unionto hangon in not-so-splendidisolationeven aftertheUnitedStates
won the supportof the vast majorityof states and the vast majorityof global powers.
The simulationsinclude 16 instancesin which the Soviet position remainsbelow -50
on the scale even thoughthe forecastis for a pro-Americanbalance of power. There
are no instances in the simulationsin which the United States maintainsa position
above 50, and yet the predicted outcome is for a pro-Soviet balance of power.
Apparently,the Soviet Union was vastly more likely to hold out in ideological or
political isolation against all odds than was the United States. Put differently,the
UnitedStatesmanifestsmorepolicy flexibilitythandoes the Soviet Unionormanifests
a greaterpreparednessto accepteven anunfavorableresolutionto the cold warin those
instancesin which the Soviet Union remainsfirm.The United States,however,more
often remainsfirmthandoes the Soviet Union so thatthe numberof instancesin which
the United States might have faced the choice of giving in to Soviet preferencesor
remainingfirm is only a small minorityof cases. We must not lose sight of the fact
that68% of the simulationsend with the Soviet Union acceptingan Americanvictory
and 78% end with the vast majorityof the world'spowers sharingin a pro-American
victory,whetherthe Soviet Union participatedor not.

t I Ir . -
LI I . q-
100 - United States Position

'
0 f .1, ,
75 -
I
0
0
E 50 - '\
I
/
I
25 - / Balance of Power Position
I
/
I
0-
if
3
__J
-25 -
0 /
-50 -
0
cn
0~ Soviet Position /
-75 -
0a - - - --/
-100 - _ /
1948 1958 1968 1978 1988

1988 1998
Simulated Year
Figure 3: An American Victory: Simulation 19
100 j
r
75-
50 - Bs Position
a.,
rPosition
25-
0-
z -25 -
a)
-50 -
0
CO)
-75 -
Soviet Position /
I.. .
-100-
1948 1958 1968 1978 1988 1998
Simulated Year
Figure 4: A Quick Soviet Victory: Simulation 38

100 - United States Position
o 75-
.)
E 50-
6
IL 25 -
-- Balance of Power Position
-
Iu0 0-
-25 -
.1
o
2 -75 -
0-
Soviet Position /
-100 -
1948
148 1958 1968 1978 1988 1998
Simulated Year
Figure 5: A Soviet Victory: Simulation 58
100 - I
I
c i
(U
o 75- I
United States Position
C
U)
E 50-
0- 25- ,'<--- Balance of Power Position
0--
a)
-25-
U) Soviet Position
> -50 -
o
U)
2 -75-
0-
Soviet Position
-100 -
, I - I I I I
11948 1958 1968 1978 1988 1998

Simulated Year
Figure 6: Cold War Continues: Simulation 96

Buenode Mesquita/ END OF COLD WAR 149
So far, we have seen thatthe simulationsshow that the United States had a better
than two thirdschance of winning the cold war.They also highlightthe observation
that the actual path to victory was not the only possible route. Many policies could
have helped foster the end of the cold war and the establishmentof a pro-American
balanceof power.I turnnow to anexaminationof the five hypotheticalworldsdepicted
in the figures. In so doing, I provide additionaldetails of the politics andpolicies that
supportthe counterfactualhistoriesrepresentedby these simulations.These areuseful
to compareto the actualhistoryof events between 1948 andthe fall of the BerlinWall
in 1989 or the collapse of the Soviet Union in 1991.
COUNTERFACTUALHISTORIES
I begin withthe historythatproducesFigure3, thecircumstancemost like the actual

events of the cold war.ThenI turnto the historiesassociatedwith the otherfigures.A
comparisonbetween Figures 2 and 3 helps suggest alternativestrategiesfor ending
the cold war.
As notedearlier,NATOwas not yet in existence in 1948. In the sequenceof events
that makes up Figure 3 (simulation 19), we can see that NATO begins to take clear
shape by the third iteration(approximating1954). During that iteration,Italy and
Franceare predictedto shift positions from 3.6, or essentially from neutrality,to 90,
or to a stronglypro-Americanstance. GreatBritainalso moves in this roundfrom a
relativelyneutralposition to 53, adoptinga position clearlyfavoringthe United States
(currentlyat 94 on the scale) over the Soviet Union (which is at -84 at this juncture).
By the next iteration,NATObecomes quite solid, addingTurkey,Greece, Canada,
andSpain.Israel,Iran,Egypt,Iraq,Pakistan,Argentina,Mexico, andBrazilalso adopt
pro-Americanpolicies. Meanwhile,the model estimatesthatthe Soviet campincludes
Hungary,Romania,Poland,Czechoslovakia,Bulgaria,andYugoslavia.The system is
now clearly bipolar.The pro-U.S. nations control 56% of the power in the system,
whereas the pro-Soviet forces are favored by 44%. Many states continue to remain
neutral,with India being the most powerfulin thatcategory.
As the system continues to evolve, the bipolar characterbecomes emphasized
further.By the fifth iteration(roughly 1958), Greecemoves away fromits pro-Ameri-
can postureto a more neutralposition, whereasSweden, the Netherlands,Denmark,
the Philippines,and Australiaadoptan Americaviewpoint,furtherstrengtheningthe
American bloc. China, which had been pro-American-recall that the 1948 data
precede the Communist Chinese victory over the Kuomindang-moves now to a
decidedly more neutralstance.The Soviet bloc, looking much like the actualWarsaw
Pact, continuesto hold firmly at positions between-80 and-100.
During the next several iterationsin the simulation,the system becomes increas-
ingly polarized.The two main power blocs tightenwhile, in the neutralrange,China
becomes a majorplayer.By the 10th iteration(roughly 1968), however, the neutral
bloc begins to drift toward the Americanposition. China moves in this period to
position 35 on the scale. By round12 (approximating1972), the Soviet Union moves
from -84 (it shifted early on to -84 from -100) to -51. The Soviet Union still is

obviously stronglyanti-American,butthis movementindicatescracksin the solidarity

of the Soviet bloc. This Soviet shift occurs at a time when American salience for
security is ratherhigh (79), whereas the Soviet Union is more preoccupied with
domestic problems (its salience for security is 24). The shift in the Soviet position
takes place in an environmentof remarkablylow tension. The model estimateswhat
percentagesof dyadic relationshipsinvolve significanttension. Although a value of
about 30% is normal, at this point the model indicates that only about 2% of
relationshipsare sufficientlytense to impose real political costs; thatis, the softening
of the Soviet position (which is accompaniedby a smallersoftening in the American
policy stance)is roughlyequivalentto the policy of detentepursuedduringthe 1970s.
After the Soviet Union adjustsits position to -51, the system stabilizesfor the next
eightrounds(or about16 years).In the 20thiteration,however,correspondingroughly
to 1988, thereis a suddenanddramaticturnof events.The United Statesandits NATO
and other allies have occupied position 68 for eight or so rounds.The Soviet Union
has hovered around -51, whereas its allies have remained even more staunchly
anti-American,keeping themselves around-81. Now, however, the Soviet Union
suddenly altersits posture,moving to 68; thatis, it suddenlyadoptsa policy outlook
in line with thatof the United States.4
It is strikingto notethatthis simulationproducesanend to thecold war,culminating
in an Americanvictory,in 1988. It was just then, of course, thatMikhail Gorbachev
signaled his preparednessto allow the East Europeanstates to pursue their own
destinies. As MargaretThatcherobserved in 1988 during her final official visit to
Washington,D.C., duringthe Reagan years, "We'renot in a cold war now" (Shultz
1993, 1131).
Justpriorto Soviet acceptanceof America'spolicy posturein the simulation,all of
the world has subscribedto the Americanoutlookexcept for the Soviet Union and its
East Europeanallies. Shortlyafterthe Soviet Union signals the end of the Brezhnev
Doctrine-or, in the model's terms, shortly after the Soviet move to 68-the East
Europeanstates abandontheirstanceandjoin in the new world order.The cold waris
over.The Soviet collapse does not occurbecauseof a disparityin salience for security
concerns.Both the UnitedStatesandthe Soviet Union have salience scores in the low
30s range at this point in the simulation.What has happenedseems to be that the
preponderanceof power on the American side ultimately makes the Soviet Union
susceptibleto acceptinga deal at 68 ratherthanrisk an outcome even more like the
original Americanpolicy at 100. The Soviet leadershipseems to have run out of the
energyto resistthe growingconcentrationof powerin favorof the Americanperspec-
tive. In this instance, then, low salience for the security dimension (of about 33)
supports the liberal hypothesis that internalproblems (especially economic ones)
brokethe Soviet Union. The power concentrationbackingthe United States supports
the conservativeclaim thatReagan's militarybuildupand aggressive foreign policy
posturemade it clear to Gorbachevthatthe Soviet Union could not compete success-
fully. Of course, still otherscontendthatReagan'ssoft-line policies in his lateryears,
4. I againremindreadersthatthe 2-yearintervalassociatedwith each roundis ex post facto; the model
does notspecify how long a roundlasts in calendartime.Theuse of 2 yearsis a conveniencethatcorresponds
remarkablywell with actualevents.

coupled with the HelsinkiAccords, were the realkey to the Soviet collapse (Deudney
and Ikenberry1991; Risse-Kappen1991). The simulationreinforcesboth strandsof
currentargumentsfor the end of the cold war:the Soviet Union implodedbecause of
internalfailings, and it collapsed because of externalpressure.The simulation,then,
may help to reframe what seems to be a fruitless debate between advocates of
alternativepoints of view. Kegley (1994), for instance,cast the debatein a seemingly
either/orcontext while favoringthe liberal,internalimplosion explanation.Yet, there
is no compelling logical reason to view these as competing explanations.Quite the
contrary,with the SovietUnion facingtwo sourcesof high costs for continuingthe cold
war, the pressureto alterits behavioris greaterthanif it faced only one source. This
is a point reinforcedby analysesof others,includinga prescientperspectiveexpressed
by Zbigniew Brzezinskias early as 1963 (Brzezinski1963; George 1991; Greenstein
1996; Lepgold, Bueno de Mesquita,and Morrow 1996).
Simulation 19, of course, is not a replica of the actualevents that ended the cold
war.Yet, in many details, it provides a historyratherclose to actualevents. NATO's
formationis clearlyevidentearlyon, as is its growingstrengthas the systempolarizes.
The WarsawPact emerges as a resilient alliance thatis resistantto externalpressure
but ultimatelydoomed by the superiorcapabilitiesof the United States to attractand
hold allies andto isolate the Soviet bloc as well as by the prospectsof an internalcrisis
in the Soviet Union. The simulationproduces a history in which Soviet-American
relationsare distantbut stable for a long period, followed by a suddenand dramatic
collapse of the Soviet bloc, much as actuallyoccurred.In many otherdetails, simula-
tion 19 reproducesimportantfeaturesof the cold war,includingtheemergenceof India
as a leader of neutralstates and the large swings in policy position of China. Again,
rememberthatthe base year for the data, 1948, is before the Communistvictory over
the Nationalistforces in China,so thatthe model was not informedof actualchanges
in China.
Figure 2 (simulation 8) depicts a much faster transitionfrom the cold war to
Americansuccess. In this simulation,the Soviet Union acceptsthe Americanapproach
to policy within threeiterationsor, roughly,by the end of 1954. The key to the rapid
Americansuccess is thatthe Soviet Union earlyon is tornby domesticconsiderations
thatsuggest internalinstability.Forinstance,in round2 of this simulation,even before
a NATO-likeallianceforms, Czechoslovakia,Romania,and Hungaryshift from their
adamantlypro-Sovietposition to neutrality(1.3 on the scale). This happensbecause
the Soviet Union is not committedto holding its allies because it is distractedby
domestic politics or some issue otherthansecurity.The salience score for securityin
round 2 for the Soviet Union is only 13, leaving about 87% of its effort turned
elsewhere, such as to domestic issues. This contrastswith America's65 salience for
securityat this stage in the simulation.
Why might Soviet salience have been so low in the early 1950s, as reflected by
simulation8? Stalin's deathcould easily have precipitatedan internalpolitical crisis.
Indeed,in importantrespects,it did.Following his death,a power struggleensuedthat
took several years to resolve. During this period, the United States played a fairly
quiescentrole by not takingany bold steps to exacerbatethe internalproblemsin the

Soviet Union. Presumably,America'scautionwas out of fearof a warthatmightresult

or because the United States was preoccupiedwith its own problemsin Korea.
Had the United States done more at thattime to encourageinternalturmoilin the
Soviet Union and to supportseparatistmovements, of which there were plenty, in
WarsawPactcountries,in this simulation,atleast, warwould nothave resulted.Rather,
the simulation suggests that the Soviet system, weakened internally and not yet
recoveredfromWorldWarII, would have collapsedandthe cold warwould have been
nippedin the bud. This is broadlyconsistentwith views expressedby membersof the
Committee on the Present Danger in the early 1950s. Simulation 8 highlights the
benefits thatAmericanfirmnessmighthave producedearly on. It also remindsus that
althoughNATOgreatlyfacilitatesthe defeat of the Soviet Union by coordinatingthe
capabilitiesof a large and powerfulbloc, NATOwas not essential in the early days.
Certainly,this does not remain true by the late 1950s. No simulation that ends in
Americansuccess after10 roundsfails to producea NATO-likeEuropeanalliance.So,
we can infer that NATO was critical to the Soviet defeat once the cold war became
well entrenched but that its entrenchment might have been averted by a more
aggressive American foreign policy in the early 1950s even without resort to
NATO.The generalizationthatNATOwas instrumentalin bringingabouta successful,
pro-Americanend of the cold war once it was well under way holds true in the
simulations regardlessof whetherNATOjust coordinateda powerful bloc, NATO
deterredthe Soviet Union, or the leadersof the Soviet Union neverintendedto attack
the West (Vasquez1991).
The importantrole thatNATOplayedand the dangersthatmight have plaguedthe
United States without NATO are evident from Figure 4 (simulation 38). In this
simulation,the Europeanstates fail to band togetherto form a NATO-likealliance.
The United States seems to be pursuingan isolationistforeignpolicy.
By the sixth iterationof the model, roughlyequivalentto 1960, the United States
alone can no longerforestalla Soviet victory.AlthoughAmericansaliencefor security
is high, so is that of the Soviet Union and its WarsawPact allies. Furthermore,the
leaders of the Soviet Union perceive themselves in a highly conflict-prone,tense
environment.More than 96% of their relationshipsare seen as conflict prone by
iteration6. This suggests that a war was likely and that the war, without European
assistance for the United States, would have ended in victory for the Soviet bloc.
Failureto stop the Soviet Union earlyon andfailureto formNATOby the mid-1950s
apparentlycould have been a disastrousrecipe.The isolationistposturearguedfor by
some afterWorldWarII seems to have promisedto producea Soviet-dominatedworld
order,violently imposed.
Figure 5 (simulation58) tells a somewhatdifferentstory.NATOdoes form in this
case by the end of iteration3. In thisinstance,theNATOalliance,ratherthantightening
over time, unravelsduringround6 (roughlyequivalentto 1960), whereasthe Warsaw
Pact countriesremain staunchin their supportof the Soviet Union's foreign policy.
Despite the unravelingof NATO,the United States holds steadily at 78 on the scale,
indicatinga stronglyanti-Sovietviewpoint. The stresslevel is very high, with nearly
70% of American relationshipsbeing tense and strained,whereas 100% of Soviet
relationshipshave the same characteristic.

Despite the tension, no suddenshifts in policy occur.Rather,the system languishes

in a heatedup statefor anotherfourrounds,equivalentto about8 years.At thatpoint,
a small crack occurs in the WarsawPact as Poland and Czechoslovakia distance
themselvesabit fromthe Soviet Union,movingto -33 on the scale.WithNATOhaving
come apartand with tensions very high, the United States seeks new partnerships.It
links up with CzechoslovakiaandPolandin an effortto redirectthe system. However,
Americansalience then dropsmarkedly,suggestingan internalpolitical crisis around
1972 (round12). The internaldistraction,combinedwith the loss of NATO,leads the
United States down a gradualpath of rapprochementwith the Soviet Union on its
terms. By round 17 (approximating1982), the United States is in a new world order
dominatedby Soviet policy. The new balanceof power is located at -66 on the scale.
Of course,it is importantto rememberthatthe simulationsindicatethatsuch a scenario,
althoughpossible, was very unlikelyto arisein reality.Recall thatthe vast majorityof
randomlyselected scenariosyield a peaceful Americanvictory,not a Soviet one.
Figure 6 (simulation96) reinforcesthe view that the end of the cold war was not
inevitable.We can alreadysee NATOemergingby round2 in this simulation.By then,
Canada,the Netherlands,France,Britain,Turkey,Norway,and othershave shifted to
a strongly pro-Americanposition. The WarsawPact also is well formed. NATO
continues to consolidate, addingItaly in the thirditerationand Greece in the fourth.
Most Asian and Middle Easternstates also shift to a pro-Americanposture,whereas
the Soviet Union and its WarsawPact allies (and a few other states) remain anti-
Americanand pro-Soviet.Although smalljockeying back and forth takes place after
round 5, the system is essentially stabilized in a long-term, unchanging bipolar
structure.
Interestingly,there is anotherscenario in which the cold war, bipolar structure
persists. If the salience variable is not permittedto change but is fixed across all
iterationsat 100, indicating that security remains the paramountconcern, then the
model emulates the structural,neorealistview of the world. Underthose conditions,
the cold war persistsin a highly stableenvironment;thatis, the neorealistsimulation
fails to predict the end of the cold war. It is consistent with Gaddis's claim that
neorealismfailed to anticipatethis critical event. Indeed, neorealism'spredictionof
bipolar stability is borne out by the simulation,whereas reality showed us quite a
differentoutcome.Fluctuationsin the relativeimportanceof nonsecurityissues (such
as domesticpolitics andeconomic or humanitarianconcerns)andsecurityissues were
an essentialpartof the actualend of the cold war.
CONCLUSION
The end of the cold war did not depend on going down one uniquepath. Conse-
quently,it is not importantto establishthatany one simulationwas morelikely to arise
thanany other.Whatis importantis to see to whatthe preponderanceof possible paths
lead. Depending on how we define victory,between 67% and 78% of the simulated
histories or paths from 1948 forwardproducedpredictionsthat the cold war would
end on pro-Americanterms.The specific pathmighthave beenunique,butmanypaths

or mixes of policies yield the same conclusion. This means that the end of the cold
war certainly was predictable.Predicting the particulardetails of the history that
culminatedin the Americanvictoryin the cold war would, of course,requireconsid-
erablymore data.But, the model also has pointedout thatex antetherewere superior
historiesto the one thatactuallyplayed out;thatis, therewere alternativepolicies that
could have led to a quicker,peaceful end to the cold war on terms favorableto the
United States. The few cases of Soviet victory,by contrast,were considerablymore
likely to arise throughviolence than throughpeaceful transition.The counterfactual
historiessimulatedheregive us insightinto how we mightplanthe resolutionof future
policy disputes.
The end of WorldWarII appearsto haveplantedthe seeds of a cold wartermination
thattook time to germinate.The ultimateend of the cold war appearsto have been an
emergentpropertywhose essentialantecedentscould alreadybe detectedin the world
structureat least as of 1948.
REFERENCES
Black, D. 1958. The theoryof committeesand elections. Cambridge,UK: CambridgeUniversityPress.

Brzezinski, Z. 1963. After the test ban: the U.S. must take the initiative in Europe. The New Republic,
31 August, 18-21.
Bueno de Mesquita,B. 1975. Measuringsystemic polarity.Journalof ConflictResolution3:187-215.
. 1981. Thewar trap.New Haven, CT:Yale UniversityPress.
Bueno de Mesquita,B., D. Newman, and A. Rabushka.1985. Forecastingpolitical events: Thefuture of
Hong Kong. New Haven, CT:Yale UniversityPress.
. 1996. Redflag over Hong Kong. Chatham,NJ: ChathamHouse.
Bueno de Mesquita, B., and F. Stokman, eds. 1994. European Communitydecision making: Models,
applicationsand comparisons.New Haven, CT:YaleUniversityPress.
A conversationwith BruceBueno de Mesquita:Wherewaris likely in the next year or two. 1982. U.S.News
and WorldReport,3 May, 30.
Deudney, D., and J. Ikenberry.1991. The internationalsources of Soviet change. InternationalSecurity
16:74-118.
Feder, S. 1995. Factions and policon: New ways to analyze politics. In Inside CIA'sprivate world:
Declassified articlesfrom the agency'sinternaljournal, 1955-1992, editedby H. BradfordWesterfield,
274-92. New Haven, CT:Yale UniversityPress.
Gaddis,J. L. 1992. Internationalrelationstheoryandthe end of the cold war.InternationalSecurity17:5-58.
- . 1997. Weknownow: Rethinkingcold war history.Oxford,UK: Clarendon.
George, A. L. 1991. The transitionin U.S.-Soviet relations, 1985-1990: An interpretationfrom the
perspectiveof internationalrelationstheoryand political psychology.Political Psychology 12:469-87.
Gilpin, R. 1981. Warand change in worldpolitics. New York:CambridgeUniversityPress.
Greenstein,F. I. 1996. RonaldReagan,MikhailGorbachev,and the end of the cold war:What difference
did they make. In Witnessesto the end of the cold war, edited by W. C. Wohlforth.Baltimore:Johns
HopkinsUniversityPress.
Haselkorn,A. 1998. Thepoison arrow.New Haven, CT:Yale UniversityPress, forthcoming.
Kegley, C. 1994. How didthe cold wardie? Principlesfor an autopsy.MershonInternationalStudiesReview
38:11-41.
Keohane, R. 1984. After hegemony:Cooperationand discord in the worldpolitical economy. Princeton,
NJ: PrincetonUniversityPress.

Bueno de Mesquita/END OF COLDWAR 155
Krasner,S. 1981. Transforminginternationalregimes:Whatthe ThirdWorldwants and why.International

Studies Quarterly25:119-48.
Kugler,J. M. and A.F.K.Organski.1989. The end of hegemony?InternationalInteractions15:113-28.
Lamborn,A. 1991. Theprice of power. Boston: Unwin Hyman.
Lepgold,J., B. Buenode Mesquita,andJ. D. Morrow.1996. The strugglefor masteryin Europe,1985-1993.
InternationalInteractions22:41-66.
McGurn,W. 1996. We warnedyou. Far EasternEconomicReview, 13 June, 68.
Ray, J. L., and B. M. Russett. 1996. The future as arbiter of theoretical controversies:Predictions,
explanationsand the end of the cold war.BritishJournalof Political Science 25:1578.
Risse-Kappen,T. 1991. Did "peacethroughstrength"end the cold war?InternationalSecurity16:162-88.
Russett,B. 1985. The mysteriouscase of vanishinghegemony;or is MarkTwainreally dead?International
Organization39:207-31.
Russia is doomed to stability,consider Americanexperts. 1995. Izvestiya, 3 April, 1-4. (Translatedby
ForeignBroadcastInformationService)
Shultz, G. P. 1993. Turmoiland triumph:My years as secretaryof state. New York:Scribner.
Vasquez,J. 1991. The deterrencemyth: Nuclearweapons and the preventionof nuclearwar.In The long
postwarpeace, editedby C. Kegley. New York:Longman.
Zelikow, P., and C. Rice. 1995. Germanyunified and Europe transformed.Cambridge,MA: Harvard
UniversityPress.


Emerging Propety

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Emerging Propety

Hochgeladen von

Copyright:

Verfügbare Formate

The End of the Cold War: Predicting an Emergent Property

Author(s): Bruce Bueno de Mesquita

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

BRUCE BUENO DE MESQUITA

AUTHOR'SNOTE:I am indebtedto JohnGaddisfor suggestingthis studyto me andfor his generosity

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

Bueno de Mesquita-JohnLewis Gaddisconvergence"(takenfromGaddis'se-mail to

1. By pathdependence,I meanthe extentto which a particularoutcomecould only have arisenthrough

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

Theabruptendof theColdWar,an unanticipated hotwarin thePersianGulf,andthe

AlthoughI wholeheartedlyagreethatdeficienciespersist,I wish to establishthe prima

Applyingmytheoryon whynationsgo to war... I willgo outon a limbandciteplaces

Expertsengagingin studieswithintheframework of thissystemstatethaton thebasis

The putsch or coup was the crisis thatprecipitateda second of Gaddis'sexemplarsof

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

Basedon theirfindings,theypredicted12 developments, of whichonly one proved

The methodologyhas been explainedin detail in numerouspublicationsso thatits

AN INTUITIVE EXPLANATION OF THE MODEL

In this section, I providean intuitivesense of how the model works.Those readers

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

areidentifiedas nation-states,but the methodologyitself need not be restrictedto this

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

After all the players submittheirsecretproposalsto one another,each reviews the

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

APPLICATION OF THE MODEL TO

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

interestsof any pairof statesis evaluatedas the correlation(using Kendall'stau b) of

2. They actuallysum to 99.692 due to rounding.

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

includedin the analysis.If anything,this introducesa pro-Sovietbias into the analysis

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

scandalin England,and so forth.Each of these events could potentiallyinfluence the

PREDICTING THE COLD WAR'S END

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

Country Capabilities Position

Argentina 0.972 89.6

NOTE:Salience initiallyis set at 100 to reflect securityconcernsandthen varies randomlyfromcountryto

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

that domestic, economic, humanitarian,or otherconcernscould never become more

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

Pro- Weakly Weakly Pro-

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

E -3 -122% Pro-USSR, 78% Pro-US

Figure 1- The Evolution of the Balance of Power

Another viewpoint can be attainedby looking at examples of each of the five

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

> -50 - / Soviet Position

Figure 2: A Quick American Victory: Simulation 8

Rice 1995). Thepredictedhistoriesundergirdingthesefiguresareexplainedin the next

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

100 - United States Position

1948 1958 1968 1978 1988

Figure 3: An American Victory: Simulation 19

Figure 4: A Quick Soviet Victory: Simulation 38

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

100 - United States Position

Figure 5: A Soviet Victory: Simulation 58

0- 25- ,'<--- Balance of Power Position

11948 1958 1968 1978 1988 1998

Figure 6: Cold War Continues: Simulation 96

This content downloaded from 129.24.219.108 on Tue, 4 Feb 2014 11:38:20 AM

I begin withthe historythatproducesFigure3, thecircumstancemost like the actual