Page 1
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 1/11
On Different Types of International Relations ScholarshipAuthor(s): Steve ChanSource: Journal of Peace Research, Vol. 39, No. 6 (Nov., 2002), pp. 747-756Published by: Sage Publications, Ltd.
Stable URL: http://www.jstor.org/stable/1555257 .
Accessed: 05/06/2013 05:21
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 [email protected] .
.
Sage Publications, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Journal of Peace
Research.
http://www.jstor.org
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 2
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 2/11
RE ? A R C i~~~~1f
? 2002 Journal fPeaceResearch,vol.39, no.6, 2002,pp. 747-756
SagePublicationsLondon,ThousandOaks,
CAand NewDelhi)
[0022-3433(200211)39:6,747-756; 030086]
On DifferentTypesof InternationalRelations
Scholarship*
STEVE CHAN
Departmentof Political Science, Universityof Colorado
Internationalrelations eaturesdifferenttypesof scholarship.These typesaremetaphor,history,theory,
engineering,and patternrecognition.This essaydiscussesthe natureand contribution of each type as
research haracteristic f these differentundertakings hows substantialovertimecontinuity in attract-
ing varying attention from identifiable communities. It also raises severalconcerns about the extant
literature.These concerns addressthe problems of endogeneity, selection effects, concept stretching,over-determination and indeterminacy,multiple conjunctural causality,and falsifiableproposition.
Althoughhardlyrepresentingnoveldiscoveries, hey presentgenericandseemingly persistentchallengesto valid inference.While not irrelevant o the conduct of social inquiryin general,this reviewaddresses
specifically nternationalrelationsscholarship.
ScholarshipTypes
Internationalelations s a pluralisticield.
Itsvariety
ofscholarship
attests to its
vibrancy nd its pre-paradigmatictate(asdistinctrom hepractice f 'normal'cience
describedby Kuhn, 1962). Colleaguescontinue o argueabout ts historiography,
epistemology, nd methodology.Not sur-
prisingly,ummaries f the field'spastand
present are often subject to objection.Reviews f internationalelations avebeen
alternativelyresentedn the formof 'greatdebates', battleof paradigms',methodo-
logical contests, and scholarlytraditionswith distinctivenationalorigins (e.g. the
'EnglishSchool','continental' R theory).These characterizationseflect deliberatechoicesbyindividual eviewersoemphasize
* This essayaddresses everalrecentsurveysof the disci-
pline. Geller & Singer (1998), Midlarsky 2000), and
Vasquez 2000a) attendto the warphenomenon,whereas
Carlsnaes,Risse & Simmons(2002) offer a moregeneralfield review.The author can be reachedat [email protected] .
certainscholarlydivisionsandde-emphasizeothers.
Fielddescriptions an becontroversial. or
instance,how accuratewould it be to depictthe field'shistoricaldevelopmentasa seriesof
ostensibledebatessuch as that between 'ide-
alism' and 'realism'(Schmidt, 2002), and
would it be warranted o draw a sharpdis-
tinction betweenquantitativeand qualitative
methodologiesthat overlooks heir common
logic of inquiry (King, Keohane & Verba,
1994)?Do rationalistand constructivistper-spectivesrepresenta validdichotomyimply-
ing mutuallyexclusiveexplanations(Fearon& Wendt, 2002)? And, to what extent have
representations f the field been dominated
by the views and worksof US academics o
the relativeneglectof contributionsby Euro-
pean colleagues (Jorgensen,2000), not to
mention those fromthe non-Westernworld?
The propagationof research gendasand the
training of successorgenerationshave not
been unaffected by career incentives, aca-
demic structures,and government funding
747
REVIEW
ESSAY
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 3
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 3/11
748 journal ofPEACE RESEARCH
priorities hatvaryacrossnationalsettings.A
field reviewcould just as reasonably eek to
accountfor the traditionof 'peacestudies' n
Scandinavia,he continuedcentralityof sov-
ereignty n Chinesediscourse,andthe role of
the Defense Advanced Research Projects
Agency in shapingresearchundertakingsn
the United States.There is surelymore than
one way to present he field,and none seems
entirelyunproblematic.Instead of relying on some well-known
perspectives uch as realism, iberalism,and
constructivismas a basis for this review,I
choose to emphasize the different analyticpurposes hatmotivate research.With minor
modification, this essay follows Bobrow's
(1972) classificationof different types of
internationalrelationsscholarship.The per-
tinent categories are metaphor, history,
theory,engineering,andpatternrecognition.
Eachtype makes a worthwhilecontribution
to knowledgeand/orpractice n its own way.
'None is betteror worse than another;they
are differentand a good job on one is verydifferent from a good job on another.
Different criteria are appropriateto each'
(Bobrow, 1972: 204-205). These consider-
ations areappliedto severalrecent antholo-
gies of the field or of partsof it (Carlsnaes,
Risse & Simmons, 2002; Geller & Singer,
1998; Midlarsky,2000; Vasquez, 2000a).
Selectiveexamplesare drawnfor illustration;
space limitation does not permit extensive
citationsor full discussion of all the relevant
studies.
Metaphorsare invitations to reason by
analogy.An imaginaryconstruct is used to
accentuatecertain aspects of reality.Thus,
one may use the prisoners' dilemma to
illuminatethe logic of interstate ompetition
(e.g. Axelrod, 1984). Alternatively, ne may
treat governments as if they are rational
unitaryactors,bureaucraciesollowingstan-
dardoperatingprocedures, r an analogueof
the human nervous system (e.g. Allison,
1971; Deutsch, 1966), and one may warn
about an impending ecological disasterby
conjuringup imagesof collapsingcommons
and survival on a lifeboat (Hardin, 1968,
1974). The powerof metaphors ies in their
ability to orient, evoke, and even provoke
and mobilize. Metaphors are suggestive
rather hanpredictive.Because heirpurposeis to offer a perspective, hey arejudged by
the extent to which they illuminate a situ-
ation orproblem.The perspectivetself is not
subjectto falsification.
Historyis concerned with the study of
eventsor charactersn a particular ime and
place. Its subjectmatter can in principlebein the futureas well as the past,althoughin
practice historical scholarship typicallyinvolves retrospective analysis. Historians
customarily eekto reconstructa sequenceof
events or a setting in order to account for
some known outcome. This reconstruction
yieldsa richtapestrywith a plot line. It does
not, however,rule out other plausibleplots
(thatis, alternative ausalchainswhich could
havealsoled to the sameoutcome). Nor is itusuallyclear in distinguishingamong neces-
sary,sufficient, or irrelevantconditions. Its
practitioners renot interested n generaliza-tion beyond the particularepisode or per-
sonality on hand. Not all case studies or
analyses using information from the past
constitutehistory in the sense definedhere.
A studyofWorldWarI maybehistory n one
case (e.g. Tuchman, 1962) but not so in
another case(e.g.
Choucri&North, 1975).What mattersis the differentanalyticpur-
poses motivatingthe research suchas when
a past episode is used to test or construct a
theory),and not the resortto evidencefrom
the past. Contemporary interpretativist
scholarship represents an example of
historicalwork as it eschewsgeneralizationsin favorof a contextualunderstandingof a
particular ituation and the relevantagent'sconstructionof meaning (e.g. Geertz,1973).
Historical analyses can be persuasive or
unpersuasive.However, as for metaphors,
volume 9 / number / november002
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 4
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 4/11
Steve Chan INTERNATIONAL RELATIONS SCHOLARSHIP
their conclusionsare not subjectto falsifica-
tion.
Theorys ahistorical,as it pursuesgeneral-
izationsgermane
to entire classes of entities
or phenomena.It is uninterested n the par-
ticulars;propernouns of people, places, or
time periodsmatteronly to the extent that
they represent pecificinstancesor members
of a relevantcategory.Theory consists of a
system of statements from which one can
deduce observable,verifiable implications.
Its core elements are concepts, axioms,
assumptions, and scope conditions. Parsi-
mony and'consilience' revirtues hat enable
theorists to account for a varietyof occur-
rences,includingnon-obvious and unantici-
patedones, on the basisof a few abstractions.
Workon collectiveaction (e.g. Olson, 1965,
1982) is suggestiveof this type of scholarly
enterprise. Some rationalist explanations,
suchasthose basedon the theoryof expected
utility (e.g.Buenode Mesquita,1981;Bueno
de Mesquita & Lalman, 1992), are also
exemplary.The explicitness of theoretical
formulationsoffers a huge advantage n thedetectionof possible logical inconsistency n
inferring testable hypotheses from the
original premise,in assessing he correspon-
dence betweenconceptsand indicators,and
in determiningwhether the test subjectsfall
within the realmof the theory'sscope con-
ditions (e.g. Simowitz & Price, 1990).
Whether nomothetic, deductive theory is
realistic or appropriate for international
relationsis subjectto debate (e.g. Bernsteinet al., 2000).
Engineering s about policy design and
intervention. It seeks to inform policy by
monitoring the relevant environment and
projectingits future course. It also seeks to
specifyactions or conditions that are neces-
saryand/orsufficient o bringaboutadesired
end. It belongs to what Simon (1969)
describedas the sciencesof the artificial.To
the extent that they made people (not justgovernment officials, but also informed
citizens)awareof the currentstatesof global
resourcesand demographicsand their likely
future evolution if the underlyingdynamics
is allowedto continue,thepioneering
studies
of the Club of Rome servedone 'engineering'
purpose(Meadowset al., 1974;Mesarovic&
Pestel, 1974). Engineering is further
advancedwhen one is promptedto ask 'what
if' questions and to engage in sensitivity
analysisfor determininghow much of X is
requiredto achieve a stipulated level of Y
within Z amount of time. Simulations of
alternativeecological futurescome to mind
as an example (e.g. Hughes, 1999). Engi-
neering, or studies with explicit policy
relevance, does not depend on fancy
methodologyor statisticalaggregation. anis
(1982) and George, Hall & Simons (1971)
undertook houghtfulcasestudies n order o
understand, respectively, past instances of
'groupthink' nd coercivediplomacy,and on
the basisof theseanalyses,proposeheuristics
intended to avert future mistakes. Paren-
thetically,a theoreticalunderstandings pre-
ferredthough not essentialfor engineeringsinceprescriptions ften relyupon intuition,
experience,and unsystematic rial and error.
Studiesthatofferthoroughchecksof efficacy
claims also perform a valuableengineering
function (e.g. Schrodt, 1990).
Works of pattern recognitionseek to
identify persistent regularities.Many 'styl-
ized facts' about international relations
derivefrom such large-N researchbasedon
systematically ollecteddata. We learnfromthis research hat strong tendenciessuch as
geographic contiguity, territorial disputes,
competitive armament, and alliance for-
mation increase he dangerof war (Geller&
Singer, 1998). These tendencies sometimes
challengereceivedwisdom, as in the case of
a positive association between balance of
power and occurrenceof war, a discovery
that contradictsrealism.Sometimes a recog-
nition of strong patterns provides theimpetusto search or theirexplanations.The
749
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 5
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 5/11
750 journal of PEACE RESEARCH
discovery hat democraciesrarely, f ever,goto waragainsteach otherencourages cholars
to look for the causesof this phenomenon
(e.g. Maoz & Russett, 1993). Thus, patternanalysescan be used to test or inform an
existingtheoryor to stimulate the construc-
tion of a new one. They are interestedin
making generic statements rather than
describing ndividualnuances. Their results
acquiremeaning in reference o substantive
expectations imported from other types of
scholarship.That stateswith a largenumber
of memberships of international organiz-
ations are more peaceful, that incomeequality is related curvilinearlywith econ-
omic development,and that formerBritish
colonies are more successfulin establishing
democratic governance are generalizations
which gain significance n the light of some
causal ogic external o the statisticalroutines
applied. In the absence of guidance from
such logic, the dataexercisesdegenerate nto
mindlessfishingexpeditionsand are vulner-
ableto spurious nterpretations.
Recent surveys of the field leave little
doubt about the distribution of interestsin
and activitiesrelatedto these differenttypesof scholarship.Studies fitting our descrip-
tions of metaphor, history, and pattern
recognitionare much more numerousthan
those characteristic f theory and engineer-
ing. Collective attention tends to go to the
formulation of alternativeperspectives,to
the demonstrationof the plausibility n prin-
ciple of particular hypotheses, and to the
multivariateanalysisof the relativestrengthof statistical associations between several
explanatoryvariables and an outcome of
interest. With the major exception of gametheoreticstudies (e.g. Morrow,2000), there
is a relative paucity of efforts to develop
simplified logical formulations that offer
testable deductions about international
relations.The difficultiesof workingout the
rules of macro-micro translation,rules thatenable one to infer social phenomenafrom
attributions o individuals,hamperthis line
of work (Coleman, 1991). Although one
often encounters a professed interest in
policyrelevance, ngineering tudiesare rela-tively rare. Among other reasons, policy
relevancesufferswhen the variables tudied
by international relationsscholars,such as
territory, national capability, and regime
characteristic,are not (at least in the short
term) manipulable by officials. These
remarks uggestthat the pursuitof different
typesof scholarshiphascontinuedalong the
lines observedby Bobrow30 yearsago.
One may speculate about the reasonsbehind this continuity. Perhapsmany col-
leagues are skeptical about the simplistic
assumptions of deductive theory which
abstractaway many importantdetails about
their subject matter (although some may
arguethat assumptionsshould be judgedby
their usefulness, not their correspondence
with observations;Friedman,1953). Others
are more inclined to engagein an inductive
approach,preferring o search for empirical
patternsin aggregateanalyses.These latter
studies haveundoubtedlybeen facilitatedby
the collection of largedatasets,such as the
Correlatesof War Project and the Polity
Project. It is therefore not surprisingthat
certaintopics (suchas those addressing tate
attributes, war/dispute involvement, event
interactions)have been more associatedwith
quantitative scholarship aimed at pattern
recognition.Qualitativeapproaches uch as
those emphasizing the construction of
meaning,the importanceof context-specific
understanding,and the role of social and
politicaldiscoursehave beenmorepopular n
the studyof othertopics.They have received
widerapplicationamong colleagues tudyinginternational aw and regimes,transnational
actorsand issues,and feminist and cultural
perspectives on international relations.
Scholarly orientation appears moreover
to be associated with the national contextof professional training and socialization.
volume 9 / number 1 november002
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 6
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 6/11
Steve Chan INTERNATIONAL RELATIONS SCHOLARSHIP
Methodological positivism is more widely
practicedin the USA than Europe (except
perhaps in Scandinavia),especially among
students of conflict behavior,as shown byGeller & Singer (1998), Midlarsky(2000),
andVasquez(2000a).
Ongoing Concerns
Thoughtful debate, critical introspection,
and systematic documentation characterize
the volumes just mentioned as well as the
handbook on international relations co-
editedby Carlsnaes,
Risse & Simmons
(2002). They demonstrate he contributions
to knowledge by the different types of
scholarshipand point to the challengesthat
still remain. Despite important progressto
date, it behooves us to recognize some
importantissuesrevealedby these reviews.
Concern for endogeneityraisesthe ques-
tion of whether a putative causal variable
shoulditselfbe explained.Thus, forinstance,
althoughthe formationof rival allianceshas
tended to be a precursor o the outbreakof
war, the causal attribution implied by this
connectionmaybe questionedbecausestates
can form alliances n anticipationof hostili-
ties. Likewise, although there is a general
agreement that democracies tend to be
peaceful, ess attentionhasbeen given to the
propositionthat peacehasbeen a conducive
factor in the historical consolidation of
democracies(Thompson, 1996). As a third
example, whereas shared membership ininternationalorganizationsand active trade
arestronglyrelatedto bilateralpeace, peace-ful statesmaybe moredisposedto join inter-
nationalorganizations ndengage n trade n
the firstplace (e.g. Russett& Oneal, 2001).
Thus, rather han takingan ostensiblecausal
variableas a given,one mayprofitably prob-lematize'it and 'endogenize' t as part of a
larger nquiry.
Selectionbiaspresentsanotherconcern.Ifdemocraciesareknown to fight less often, is
it becausethese regimesare more peaceful?
Or could it be that realizingdemocraciesare
more likely to win wars,autocracies end to
be more reluctant o challengethese regimesin the first place and, when faced with
demands from democracies,morewilling to
make concessions to them?In other words,
can the observedpeacefulness f democracies
be explained by their having fewer oppor-
tunities to fight ratherthan their inherent
pacificdisposition(Schultz,1999)?Likewise,
could the tendency for democraciesto win
more wars than autocraciesbe due to the
formerbeing
more selectiveinfightingonlythe more winnable wars as a result of their
needforpublicconsent and concernforelec-
toralpenalty (Reiter& Stam, 2002)? Selec-
tion effects similarly affect other common
researchquestions.Forinstance,contrary o
conventionalwisdom,shouldone not expect
the credibilityof immediate deterrence o be
inversely related to the success of such
attempts (e.g. Fearon,2002)? As additional
examples, states which decide to imposeeconomic sanctions will presumablychoose
easytargetspromising he greatest hancefor
their coercive policy to succeed, and states
which initiate wars against membersof an
alliance should pick victims which in their
view are the least likely to receivesupportfrom powerful friends. Accordingly, it is
difficultto judgethe relative uccessof sanc-
tion policiesorthe reliability f alliancecom-
mitments from knowncasesof sanctionsand
wars.Those occasionswhen sanctionor warwasforegoneshould alsobe partof thisjudg-ment.
A tendencyto dwellupon events thathave
happened is understandablebecause it is
inherently difficult to analyze non-occur-
rences.Yet, the study of non-events - the
proverbial dog that did not bark - can
improveknowledge.Those exceptionswhere
the expected did not materializehelp to
identifycriticalmissingfactors n our under-standing. Why did the power transition
751
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 7
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 7/11
752 journal of PEACE RESEARCH
betweenthe USA and UK not producewar?
Why did Moscow not resort o arms to avert
the disintegrationof its empire?Why did the
Westnot intervene n Rwanda o stop geno-cide?To makea differentpoint thoughin the
same vein, patternrecognitioncan comple-
ment in-depth analysisof particularcases.
While the formerare concernedwith central
tendencies,the latter can be broughtto bear
on the outliers.What can account for those
deviant cases that departfrom the norm?A
recent study by Siverson & Ward (2002)
underscoresanother helpful point: whereas
the field has soughtprimarily o explainthe
occurrenceof war,an alternative ocuson the
persistence of peace offers consequential
lessonsfor theoryand policy.
Aggregateanalysisnaturallyassumesthat
the casesbeingstudiedarehomogeneousand
independentunits. One seeks to increase he
number of these cases acrossspaceand over
time in orderto maximizeone'ssamplesize
for statisticalexaminationand the generaliz-
abilityof one'sconclusions.Yet,as Biersteker
(2002) persuasively rgued,conceptssuch as
state, sovereignty,and territoryhave vastly
differentmeaningacross paceand over time.
Their naturehashardlybeen fixed.The gain
in sample size thus often comes at the
expense of conceptual precision, as Sartori
(1970) warned some time ago. Qualitative
differencesn kindaresometimesoverlooked
due to an eagernessto assess quantitativedifferencesn degree.Mattersof classification
(such as in establishingdifferent types ofinterstatewar)should logicallyprecedecon-
cernsfor measurement suchas in determin-
ing the severityor magnitudeof particular
wars),sincequestionsabout 'how much'can
only apply to items belonging to the same
type (that is, only after one has settled the
issueof 'whatkind').
As forcaseindependence, arge-N pattern
recognition efforts typically assume that
people somehow fail to learn from the past.By treating, say, the Korean War, the
VietnamWar,the Gulf War,and the Warin
Kosovo each as a separate ase,one is in fact
asserting hat the US behavior n these con-
flicts is governedbya fixedsetof factors,andthat the behaviorof Washington's dversaries
is not influencedby perceived essons drawn
from the priorconflicts.Learningand adap-
tation should be especially relevant in
repeated encounters such as those among
enduring rivals (Goertz & Diehl, 2000).
Leng(2000), for instance,reportedevidence
of temporaldependencesuch that partiesto
a dispute tend to undertakereciprocalesca-
lation from one crisis to the next, leading
eventuallyto war.Axelrod's 1984) 'shadow
of future' offers another helpful reminder.
Suchtreatments,however,arerelatively are.
Richardson (1960: 12) remarkedthat his
classic formulation of arms race represents
'merelya descriptionof what people would
do if they did not stop to think'. Many
studies continue to take this posture, with
assumptionsof independence of cases and
invarianceof structurethat implicitly deny
the relevance of experience and people's
capacity o learn(e.g. Bernsteinet al., 2000).
Case studies especially face other
additionalconcerns.A few salient issues and
episodeswith relativelyeasy data access are
over-studied, offering slim assurance that
they constitute a viable basis for making
more general nferences.As Mitchell (2002:
512) put it in reference o the studyof inter-
national environmentalpolitics, 'we need to
examine more than the ozone depletion,climatechangeand acid rain cases that have
been the empirical testbeds for too manytheories'.The explanations or the successof
international environmental regimes pre-sented by these case studies tend to be over-
determined as they point to a variety of
causal actors uchasissuearea,actor charac-
teristics,and international ontext. The same
situation applies to the subfield of inter-
national law and compliance, where themultitudeof causal actorsproposedmakes t
volume 9 / number / november002
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 8
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 8/11
Steve Chan INTERNATIONAL RELATIONS SCHOLARSHIP
difficult to disentangle and weigh their
relativeimportance(Raustiala& Slaughter,
2002: 545). Even beforeone begins to con-
sider the interactive effects among thesefactors,one is at risk of exhausting he avail-
abledegreesof freedom.This typeof scholar-
ship thereforeengenders he classicproblem
of 'many variables, few cases' raised by
Lijphart 1971) over three decadesago.
Selection on the dependent variable,
unclear scope conditions, and over-
determinedexplanationsnaturally ead to a
sense thatdespitesignificantprogresso date,
there s still 'the lack of confidencewe havein
the ability to draw strong inferences from
much of the research o date' (Simmons &
Martin, 2002: 205). Indeed, considerable
skepticism n this abilityis warranted o the
extent that, for instance, 'much of the evi-
dence of high compliancewith international
law is merelyindicativeof the shallowness
of manyinternational greements nd should
not be generalizedo moredemandingcases'
(Raustiala& Slaughter,2002: 543). Com-
mentingon humanrightsregimes,Donnelly
(1986) had similarlycautionedagainstextra-
polationfrom the 'easy'developmentphaseof
such regimesthus far. It is difficult to judgefrom this experience whether states are
willing to make a qualitativencrease n their
commitment to these regimes,and whether
strongstateswill submit themselves o these
regimes,as so faronly weak states have been
challenged to conform to them. The US
refusal o supportthe InternationalCriminalCourt vindicates his premonition.
Insteadof over-determination,ndetermi-
nacyis a concernfor some scholarlyperspec-tives of a metatheoreticor metaphoricsort.
They are ndeterminaten the sensethatthey
provideconvergent expectations.As Fearon
& Wendt (2002) remarked,rationalist and
constructivistexplanationsare not mutually
exclusive.Echoing the same views, Choi &
Caporaso 2002: 490) concludedthat the so-called logic of appropriateness nd logic of
consequences pose too simple a dichotomy
becausepeople seldom act strictlyin one or
the other of these sharplydrawncapacities.
As a third example, preferencesand insti-tutions often predictsimilartradepractices
by states.Moreover, hey tend to be recipro-
callycausal(Milner,2002). Abstractdebates
about the relative merits of these grand
formulations are likely to leave both sides
exhaustedratherthan satisfied.At the same
time, studies that deliberately eek to assess
the 'excess empirical content' (Kugler &
Lemke, 2000: 156) of one or the other
perspectiveareratherrare.As Choi &
Capo-raso(2002: 490) argued,one shouldask how
well explanationsofferedby one metatheory
can performoverand abovethose presented
by another,noting that, for instance, 'vari-
ables central to constructivism, such as
norms and institutions, will be judged
increasinglyn terms of theirvalue-added o
otherexplanations'.
Outcomes such as war often have
multiple causes (Levy,2000: 325; Vasquez,2000b: 381). Thus, mono-causal expla-
nations are likely to disappoint, and
different combinations of causalfactors can
produce the same outcome. The searchfor
'nice laws'aboutempiricalregularitiesunder
given conditions and the identification of
'multiple conjuncturalcausality'have been
recommendedby Most & Starr(1989) and
Ragin (1987). These colleagues emphasize
that different factors can combine to
producethe same outcomes, and the effectsof a factor can be contingent upon the pres-ence or absence of other factors(that is, the
same factor can have different effectsunder
differentcircumstances).They offeracogent
logic of inquiry and a useful methodology
(in the case of Ragin, Boolean algebra) hat
integrate quantitative pattern recognitionand qualitativehistoricalanalysis.They seek
to strike a balance between the temptation
to make sweeping (often deterministic)generalizationsand the inclination to dwell
753
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 9
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 9/11
754 journal of PEACE RESEARCH
upon the excruciatingdetails of individual
cases.
Severalcontributors o the volumesbeing
reviewed recommended the pursuit ofmultiplemethodsfor the purposeof conver-
gent validation(e.g. Crenshaw,2000; Levy,
2002; Mitchell, 2002). This is certainly
sound advice,because each method and the
evidenceexaminedby it have theirparticular
weaknesses.In combination, however, one
may hope that the different methods and
data can offset each other's limitations and
offergreater nalyticconfidencebyproviding
cross-validation.Geller(2000) and Geller&Singer(1998) exemplifyanotherapproach o
validation.They followedthe adviceof King,Keohane& Verba 1994: 30-3 1) to look for
theoretical confirmation by checking the
available videnceat severaldifferent evelsof
analysis.Forinstance,is there corroboration
betweenevidenceat the monadic level (say,
the associationbetween a state'snumber of
neighborsand its incidence of war involve-
ment)and the
dyadiclevel
(the physicaldis-
tanceseparatingpairsof statesandtheirjoint
warproneness)?
Conclusion
The differenttypesof scholarshipdescribed
in this essayarenot mutuallyexclusive.They
make different contributions that can be
mutually beneficial, as when historical
studies isolate immediate causes that act as
catalysts or the general endencies dentified
in aggregateanalyses.Metaphorscan offera
basis for developing formal theoretical
formulations.Engineeringcan drawspecific
insightsfrom history,and generalheuristics
from patternrecognition.The variousconcernsraised n this review
are not peculiar to any particulartype of
scholarship.They are generic to the chal-
lenges of valid inference. Although some
issuessuch as selection effects and the endo-geneity problem have only recentlygained
the attention that they deserve,other issues
have been recognizedfor some time, if not
consistently acted on. For instance, in
describing a fictitious accident causing adrunkard's eath,Richardson 1960: xix-xx)
enjoinedconsiderationof interactive actors
and conjunctural causality.Therefore, one
would not necessarilyexpect 'eureka'reac-
tions to the problems identified here,
although their very persistence seems to
warrant more insistent caveats. That the
relative allocation of scholarlyattention to
the different types of research has been
marked by substantial continuity sinceBobrow's original analysis also perhaps
speaks to the persistenceof certain disci-
plinaryincentives and the reproductivepro-
clivities of different subfields. Such a
sociological analysis of the field, however,
will have to wait for anotheroccasion.
References
Allison,GrahamT., 1971. Essencef
Decision:
Explaininghe CubanMissileCrisis.Boston,MA:Little,Brown.
Axelrod,Robert,1984. TheEvolutionf Cooper-ation.New York:Basic.
Bernstein, teven;RichardN. Lebow, aniceG.
Stein & StevenWeber,2000. 'God Gave
Physics he EasyProblems: daptingSocial
Science o anUnpredictable orld',EuropeanJournal fInternationalelations(1):43-76.
Biersteker,homasJ., 2002. 'State,SovereigntyandTerritory',nCarlsnaes, isse&Simmons
(157-176).
Bobrow,DavisB., 1972.'TheRelevance oten-
tialof Different roducts',nRaymond anter& RichardH. Ullman, ds,TheoryndPolicyin InternationalRelations.Princeton,NJ:PrincetonUniversity ress204-228).
Buenode Mesquita, ruce,1981. TheWarTrap.NewHaven,CT:YaleUniversity ress.
Bueno de Mesquita,Bruce& David Lalman,1992. Warand Reason: omestic nd Inter-
national mperatives.ew Haven,CT: Yale
University ress.Carlsnaes,Walter;ThomasRisse & Beth A.
volume 9 / number / november002
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 10
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 10/11
Steve Chan INTERNATIONAL RELATIONS SCHOLARSHIP
Simmons, eds, 2002. Handbook of Inter-
nationalRelations.London:Sage.
Choi, YoungJong & JamesA. Caporaso,2002.
'ComparativeRegional Integration', n Carl-
snaes,Risse & Simmons (480-499).
Choucri, Nazli N. & Robert C. North, 1975.
Nations n Conflict:NationalGrowth ndInter-
nationalConflict.SanFrancisco,CA:Freeman.
Coleman,JamesS., 1991. 'Microfoundationsand
Macrosocial Behavior', in Joan Huber, ed.,
The Micro-Macro Linkages in Sociology.
Newbury Park,CA: Sage (153-173).
Crenshaw,Martha, 2000. 'Terrorismand Inter-
nationalViolence', in Midlarsky 3-24).
Deutsch, KarlW, 1966. TheNervesof Govern-
ment:Modelsof Political Communication nd
Control.New York:Free Press.
Donnelly, Jack, 1986. 'International Human
Rights: A Regime Analysis', International
Organization 0(3): 599-642.
Fearon,James D., 2002. 'Selection Effects and
Deterrence', InternationalInteractions28(1):
5-29.
Fearon, James D. & AlexanderWendt, 2002.
'Rationalism v. Constructivism: A Skeptical
View', in Carlsnaes, Risse & Simmons
(52-72).
Friedman,Milton, 1953. 'The Methodology of
PositiveEconomics', n Milton Friedman,ed.,
Essaysn PositiveEconomics.Chicago, IL: Uni-
versityof ChicagoPress(3-43).
Geertz, Clifford, 1973. The Interpretationof
Cultures.New York:Basic.
Geller,Daniel S., 2000. 'ExaminingWar:Empiri-
cal Patternsand TheoreticalMechanisms', n
Midlarsky 407-449).
Geller,Daniel S. & J. David Singer,1998. Nations
at War:A ScientificStudyofInternationalCon-
flict. Cambridge:CambridgeUniversityPress.
George,AlexanderL.; David K. Hall & William
E. Simons, 1971. TheLimitsofCoerciveDiplo-
macy.Boston, MA: Little, Brown.
Goertz,Gary& PaulF.Diehl, 2000. '(Enduring)
Rivalries',n Midlarsky 222-267).
Hardin, Garrett, 1968. 'The Tragedy of
Commons', Science 162 (December):
1243-1248.
Hardin, Garrett, 1974. 'Living on a Lifeboat',
BioScience 4 (October):561-568.Hughes, Barry B., 1999. InternationalFutures:
Chaos in the Face of Uncertainty,3rd edn.
Boulder,CO: Westview.
Janis, Irving L., 1982. Groupthink:Psychological
Studiesof PolicyDecisions and Fiascoes,2nd
edn. Boston, MA: Houghton Mifflin.
J0rgensen, Knud Erik, 2000. 'Continental IR
Theory: The Best Kept Secret', European
Journalof InternationalRelations (1): 9-42.
King,Gary;Robert 0. Keohane& SidneyVerba,
1994. DesigningSocialInquiry:Scientific nfer-
ence in Qualitative Research.Princeton, NJ:PrincetonUniversityPress.
Kugler, Jacek & Douglas Lemke, 2000. 'The
PowerTransitionResearchProgram:Assessing
Theoreticaland EmpiricalAdvances', n Mid-
larsky(129-163).
Kuhn, Thomas, 1962. The Structure f ScientificRevolution.Chicago,IL:Universityof Chicago
Press.
Leng, RussellJ., 2000. 'Escalation:Crisis Behav-
ior andWar',in Vasquez,2000a (235-258).
Levy,JackS., 2000. 'Reflectionson the Scientific
Studyof War',in Vasquez,2000a (319-327).
Levy,JackS., 2002. 'WarandPeace', n Carlsnaes,
Risse & Simmons (350-368).
Lijphart,Arend, 1971. 'ComparativePoliticsand
the ComparativeMethod',AmericanPolitical
ScienceReview65(3): 682-693.
Maoz, Zeev & BruceM. Russett,1993. 'Norma-
tive and Structural Causes of Democratic
Peace, 1946-1986', AmericanPoliticalScience
Review87(3): 624-638.
Meadows, Donnela H.; Dennis L. Meadows,
Jorgen Randers & William W. Behrens III,
1974. The Limits to Growth,2nd edn. New
York:New AmericanLibrary.
Mesarovic,Mihajlo D. & EduardPestel, 1974.
Mankind at the TurningPoint. New York:Dutton.
Midlarsky,Manus I., ed., 2000. Handbook ofWarStudies I. Ann Arbor,MI: Universityof
Michigan Press.
Milner,Helen V., 2002. 'InternationalTrade', n
Carlsnaes,Risse & Simmons (448-461).
Mitchell, Ronald B., 2002. 'International
Environment', n Carlsnaes,Risse& Simmons
(500-516).
Morrow,JamesD., 2000. 'The Ongoing Game-
Theoretic Revolution', in Midlarsky(164-192).
755
This content downloaded from 202.92.128.28 on Wed, 5 Jun 2013 05:21:43 AMAll use subject to JSTOR Terms and Conditions
Page 11
8/13/2019 2002 IR Metatheory
http://slidepdf.com/reader/full/2002-ir-metatheory 11/11
756 journal of PEACE RESEARCH
Most, Benjamin A. & Harvey Starr, 1989.
Inquiry, Logic and International Politics.
Columbia, SC: Universityof South Carolina
Press.Olson, Mancur,Jr., 1965. TheLogicof Collective
Action:Public Goods nd theTheory f Groups.
Cambridge,MA: HarvardUniversityPress.
Olson, Mancur,Jr.,1982. The Riseand Declineof
Nations: Economic Growth, Stagflation,and
Social Rigidities.New Haven, CT: Yale Uni-
versityPress.
Ragin, Charles C., 1987. The Comparative
Method:MovingBeyondQualitative nd Quan-
titativeStrategies.Berkeley,CA: Universityof
CaliforniaPress.Raustiala,Kal & Anne-Marie Slaughter,2002.
'International Law, International Relations
and Compliance', in Carlsnaes, Risse &
Simmons (538-558).
Reiter,Dan & Allan C. Stam,2002. Democracies
at War.Princeton, NJ: Princeton University
Press.
Richardson,LewisF, 1960. Armsand Insecurity.
PacificGrove,CA: Boxwood.
Russett,BruceM. & John R. Oneal, 2001. Tri-
angulatingPeace:
Democracy,nterdependence,and InternationalOrganizations.New York:
Norton.
Sartori,Giovanni, 1970. 'ConceptMisformation
in Comparative Politics', American Political
ScienceReview64(4): 1033-1053.
Schmidt, Brian C., 2002. 'On the History and
Historiographyof InternationalRelations', n
Carlsnaes,Risse & Simmons (3-22).
Schrodt, Philip A., 1990. 'A Methodological
Critiqueof a Test of the Effectsof the Mahar-
ishi Technologyof the United Field',Journal
of ConflictResolution 4(4): 745-755.
Schultz, Kenneth A., 1999. 'Do Democratic
InstitutionsConstrainor Inform?Contrasting
Two InstitutionalPerspectives n Democracy
and War', InternationalOrganization53(2):133-162.
Simmons,BethA. & LisaL.Martin,2002. 'Inter-
national Organizationsand Institutions', in
Carlsnaes,Risse & Simmons (192-21 1).
Simon, HerbertA., 1969. TheSciences ftheArti-
ficial. Cambridge,MA: MIT Press.
Simowitz, Roslyn & BarryL. Price, 1990. 'The
ExpectedUtility Theory of Conflict: Measur-
ing Theoretical Progress',American Political
ScienceReview84(2): 439-460.
Siverson, Randolph M. & Michael D. Ward,2002. 'The Long Peace:A Reconsideration',
InternationalOrganization 6(3): 679-691.
Thompson, William R., 1996. 'Democracyand
Peace: Putting the Cart Before the Horse',
InternationalOrganization 0(1): 141-174.
Tuchman, Barbara.1962. The Guns of August.
New York:Macmillan.
Vasquez,JohnA., ed., 2000a. WhatDo WeKnow
About War?Lanham,MD: Rowman & Little-
field.
Vasquez,JohnA., 2000b.
'Reexamininghe
Stepsto War: New Evidence and Theoretical
Insights', n Midlarsky 371-406).
STEVE CHAN, b. 1949, PhD in Political
Science(Universityof Minnesota, 1976); Pro-
fessor, University of Colorado, Boulder
(1984- ). Most recent book (co-edited with
JamesR. Scarritt),Copingwith Globalization
(FrankCass,forthcoming).
volume39 / number6/ november2002