Sei sulla pagina 1di 14

Taking Temporality Seriously: Modeling History and the Use of Narratives as Evidence

Author(s): Tim Büthe


Source: The American Political Science Review, Vol. 96, No. 3 (Sep., 2002), pp. 481-493
Published by: American Political Science Association
Stable URL: http://www.jstor.org/stable/3117924 .
Accessed: 16/02/2015 17:20

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

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

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

http://www.jstor.org

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3 September 2002

Taking Temporality Seriously: Modeling History and the Use


of Narratives as Evidence
TIM BUTHE Columbia University
ocial scientistsinterestedin explaininghistoricalprocessescan, indeedshould, refusethe choice
betweenmodelingcausalrelationshipsandstudyinghistory.Identifyingtemporalityas thedefining
characteristicof processesthatcan be meaningfullydistinguishedas "history," I show thatmod-
eling such phenomenaengendersparticulardifficultiesbut is bothpossible and fruitful.Narratives,as
a way of presentingempiricalinformation,have distinctivestrengthsthatmake themespeciallysuited
for historicalscholarship,and structuringthe narrativesbased on the model allows us to treatthemas
dataon whichto testthe model.At thesame time,thisuse of narrativesraisesmethodologicalproblems
not identifiedin recentdebates.I specify theseproblems,analyze theirimplications,and suggestways
of solving or minimizingthem. Thereis no inherentincompatibilitybetween-but muchpotentialgain
from-modeling historyand usinghistoricalnarrativesas data.

cross the empiricalsubfields of political sci- To advance and extend the methodologicaldebate
ence, we find in recent years a renewed and on this generalquestion,I examinetwo sets of specific
growinginterestin "historicalmacro-analysis" issues.First,what defines "history"as a distinctobject
(Katznelson1997,82), whichseeks to understandand of study? What are the implicationsof such a con-
causally explain processes with an importanttempo- ception of history for developing explicit theoretical
ral dimension,such as the formationand evolution of models? Is there an inherentincompatibilitybetween
formal and informalsocial institutions.This trend has modelingand the quest for explanationand narration
givenriseto "historicalinstitutionalism"(foroverviews of "history,"as some observers suggest (e.g., Elster
see Hall and Taylor1996;Immergut1998;Robertson 2000)? Based on an inclusiveconceptionof modeling
1993;Thelen 1999), which manifestsitself in much of and an explicit conception of "history"as processes
the newer literatureon the welfarestate and state for- rendereddistinctiveby the importanceof temporality,
mation in comparativepolitics and in the literature I arguethatmodelingsuchprocessesis particularlydif-
on American political development. In international ficultbut,nonetheless,possibleanddesirable.Farfrom
relations,manifestationsof this trend toward histor- being inherentlyfutile, modelinghistory is extremely
ical scholarshiprange from the interest in domestic useful, not least because models, by emphasizingthe
and internationalinstitutionsto the postmortemde- general,help us clarifywhatis historicallyand contex-
bate over the nature of the Cold War.Does this in- tually specificwhen we examinethe historicalrecord.
terestin historicalprocessesmerelyexpandthe subject Consequently,the "historicturnin the humansciences"
matter of political science or does it raise particular (McDonald 1996a)need not lead us away from what
methodological problems that require a distinct ap- scholarsof very differentpersuasionshave identified
proachto theorizingandto the presentationof empiri- as the particularstrength and source of progress of
cal informationto test the plausibilityor validityof our Americanpoliticalscience:the explicitmodelingof the
explanations?1 politicalphenomenawe seek to explain,so as to facil-
itate scrutinyof the deductivelogic of the explanation
Tim Blithe is the 2001/2002 Lindt Fellow of Columbia University, (see, e.g., Milner1998,and Waever1998).
completing a Ph.D. in the Department of Political Science on the Second,becausemost historicalworkin politicalsci-
impact of partisan politics on business confidence in the context of ence is narrativein form, I examine the relationship
globalization and variations in political and economic institutions. In between models and narrativesand, more generally,
2002/2003, he will be a James Bryant Conant Fellow at the Center for
European Studies at Harvard University, 27 Kirkland Street, Cam- the strengthsandweaknessesof narrativesas a type of
bridge, MA 02138 (tim.buthe@columbia.edu, http://www.buthe.net). "presentationof the results"of our analysis(Skocpol
For fruitful discussions and helpful comments on various ear-
lier versions of this paper I thank Jim Alt, Ruth Ben-Artzi,
1995, 44). Here, my focus differs from much recent
Charles Cameron, Derek Chollet, Lewis Edinger, Peter Hall, Robert
work that concentrateson narrativesas a source of
Keohane, Robert Lieberman, Sarah Lindemann, Solomon Major, empiricalinformationfor the analyst,either broadly,
Walter Mattli, Paul McDonald, Helen Milner, Andrew Nathan, conceptualizingall empiricalinformationas "text"to
Matt Nelson, Dan Nexon, Jamie O'Connell, Jack Snyder, Hendrik be interpreted(Ricoeur[1971]1979),ormorenarrowly,
Spruyt, Barry Weingast, the members of the Institutions Workshop
at Columbia University, and, especially, Ira Katznelson. I also thank concentratingon specificoralandwritten"histories"in
three anonymous reviewers for exceptionally thoughtful comments the formof narrativesas constitutiveelementsof ideas
and constructive criticisms. I have benefited from financial support and norms (e.g., Anderson [1983] 1991; Finnemore
from the Graduate School of Arts and Sciences and the Depart- 1996;FinnemoreandSikkink1998)withinterestingim-
ment of Political Science at Columbia University and, during the
final stages of this research while an Exchange Scholar at Stanford
plicationsfor policy (e.g., Van Evera 1994,36f). These
works have been joined by predominantlymethod-
University, from the hospitality of Stanford's Department of Political
Science. ologicalcontributions(e.g., Heise 1993;Lustick1996),
1 Throughout this paper I subsume
epistemological issues under which often seek to raise social scientists'awareness
"methodological" ones. of historians'carefullydeveloped methods to discern

481

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

internalvaliditythroughthe "criticaluse of sources." MODELING


All of those worksare concernedwithnarrativesas the
rawmaterial-as (primaryor secondary)sources-for Any attemptto answerthe questionof whethermod-
the analysis.Incontrast,I focuson narrativeswrittenby eling and the studyof historyare incompatibleshould
scholarsto presentthe resultsof their empiricalanal- startwith explicitdefinitionsof modelingandhistory.I
take a model to be a schematicstatementof a theoreti-
ysis, providinginformationabout actors,institutions, cal argument,a hypothesizedparsimoniousabstraction
events, and relationshipsin "a single coherent story,
albeit with subplots"(Stone 1987,74) to provide em- or simplificationof "reality"that depictsa deductively
piricalsupportfor a theoreticalargument.Insofaras sound, systematic,regularrelationshipbetween spec-
ified aspects of reality and helps to explain that rela-
they are independentof the informationused to con-
structthe model, these narrativescan serve as data to tionship.Modelingto providecausalexplanations,we
test the model or as "evidence"to supportthe model's customarilystart by designatingthe explanandumas
the "dependentvariable"and its hypothesizedcauses
plausibility.2 as the "independentvariables."3Modeling,however,
Using narrativesto provide empiricalsupport for mustgo beyondidentifyingregularitiesin a Hempelian
one's model has a number of benefits, especially
when we seek to examine explanationsfor historical fashion by also specifying the causal mechanisms
processeswith an importanttemporaldimension,but throughwhichthe elementsarelinkedin a regularway
it also raises methodologicalquestions that warrant (Coase [1981]1994;Elster 1989,3-10; Tilly 1997).The
closer attention. Specifically,how do we delineate a resultingmodels allow us, with the help of specified
sequence of events so as to justify the impositionof assumptions,to derive specific hypotheses about the
a narrativebeginningand end onto a continuousem- explanandum,given particular"values"for the inde-
piricalrecord?How does the impositionof a narrative pendent variables-though the specificitymay be re-
closure affect the generalityof our conclusions?What stricted to predictingthe direction of change in the
is a narrative's"truthclaim?"How usefularenarratives explanandum.4 Whetheremployingthe logic of instru-
for the assessmentof alternativeexplanations?These mental rationalityor not, whether assumingperfect
questions-and how we answer them-matters not strategicand computationalcapabilitiesor more lim-
ited ones, whether workingquantitativelyor qualita-
only for scholars interested in historical work. The
evidentiarystatusthat we attributeto a narrativeas a tively,we model to emphasizewhatwe considerto be
the mostimportant,systematic,andin thatsensegener-
consequenceof our conclusionsaboutthe relationship alizableelementsof the phenomenawe seek to explain.
between author-scholarand narrative,for instance,
affectsour reading(and writing)of all work that uses Insofaras the natureof the explanandumpermits,a
more formalexpositionof the model tends to facilitate
prose to presentits empiricalresults.
I analyzethe problemsidentifiedby these questions checking"a systemof causalgeneralizations.., for in-
and suggest ways of minimizingtheir impact on the ternalconsistencyandfor conclusionsthatarenot intu-
validityof ourfindingsandthe confidencewe canhave itivelyobviousat the outset"(McClelland1975,114),as
in them. I show how the theoreticalmodel can help Robert Powell (1999,99-102) illustratesin his spirited
structurenarrativesin waysthatmake themmore suit- defenseof the promiseandpracticeof formalmodeling
able as tests and how the use of multiple narratives in securitystudies.The languageof mathematicsis par-
can increaseour confidencein the model, illustrating ticularlysuitedfor scrutinizingthe deductivelogicof an
my argumentswithexamplesdrawnfromthe literature argument-and usingit does not requiremakingratio-
on the state andregimechange.I concludethat a care- nal choice assumptions,as BarryO'Neill's (e.g., 2001)
ful combinationof modelsandnarrativesyieldsclearer work illustrates-but it also imposes costs on author
and reader, which may or may not have correspond-
insightswithgreaterconfidencethana returnto unthe-
orized historywritingwith its often vast hiddencausal ing benefitsfor particularquestionsandindividuals.In
assumptionsandclaims.It thusallowsthe integrationof short,formalizationhas some advantages,but a model
need not be formal.Rather,the definingcharacteristic
rigoroustheorizingandthe studyof historicalprocesses of a model is that it provides an explicit, deductively
in a mannerthat is attentiveto, and respectfulof, the
historicalrecord(cf. Evans 1995,3f). soundstatementof the theoreticalargument,separate
from a particular empirical context.
2
By forcing us to specify the elements of our argu-
The historical narratives discussed here are therefore in many re-
ments, such as actors and their goals, modeling should
spects similar to case studies, discussed in a number of books and
articles on methodology in recent years (e.g., Achen and Snidal 1989;
Bennett 1999; Collier 1995; Fearon 1991; Geddes 1990; Jervis 1990; 3 To be sure, not all political science and certainly not all histori-
King, Keohane, and Verba 1994; Przeworski 1995; Rogowski 1995; cal scholarship seeks to provide causal explanations (Calhoun 1998,
Tetlock and Belkin 1996). However, these narratives are not used as 863ff; Latin 1995, 455; Somers 1998, 745ff). But much of contem-
"interpretative" (Lijphart 1971, 692) or "disciplined-configurative" porary political science scholarship does, and as Max Weber ([1906]
(Eckstein 1975, 99ff) case studies; rather, historical narratives are 1978) noted long ago, almost all explanations of political phenomena
intended to provide empirical support for original theoretical propo- make at least implicit causal claims. I therefore concentrate on the
sitions. Moreover, historical narratives differ from typical case studies methodological problems involved when seeking causal explanations
in that they trace historical processes over long periods of time to for historical processes.
test a theoretical argument with an important temporal dimension-- 4 Specificity decreases, for instance, when Jean-Laurent Rosenthal,
which raises methodological problems not discussed in the case study (1998, 89) relaxes the assumption of a unitary elite in his model of
literature. early modern regime change.

482

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

increase clarity and explicitness,reducing ambiguity probablybe suppliedinductively.5Moreover,induction


and vagueness.Thatsaid, in practice,models often are often has a furtherrole in the modeling process, es-
not explicitabout all elements of the analysis.Particu- pecially when empiricaldisconfirmationof our initial
larlywhenone mode of analysisdominatesa discipline, model leads us to the inductivediscoveryof additional
its core assumptionsare often dropped from discus- explanatoryfactorsthatcan accountfor the anomalies.
sion. These assumptionsremain a central part of the Integratingthese factorssystematicallyinto the model
analysis,but they are no longermentionedand maybe is surelypreferableto havingthemremainextraneous
"smuggledin" and thus become "immuneto rational and theoreticallyirrelevant(cf. Smelser 1967,33, 35),
criticism"(Barry 1985, 282). Yet, althoughmodeling as long as we keep in mind that the inductiveaddition
can be a rhetoricaldevice to hide one's values or other of "auxiliaryhypotheses"withoutsubsequentseparate
assumptions(McCloskey 1998), it need not be; and tests leads only to hypotheses,not conclusions(King,
explicitmodelingsurelymakes it easier to detect hid- Keohane, and Verba 1994, 21f, 101ff;Lakatos [1970]
den assumptionsthanleavingthe theoreticalargument 1974, 117f).6That said, even works that are entirely
implicit. inductivewouldbenefitin two waysfrommakingtheir
Modeling as such, then, does not require any par- argumentexplicitandseparatingit as a model fromthe
ticularontological,epistemological,or substantiveas- particularcase on whichit is based.First,sucha separa-
sumptions,thoughanyspecificmodelwillhaveto make tion forces the authorto distinguishbetween the con-
them and shouldmake them explicitto facilitatetheir ceptuallyabstractelementsof causalrelationshipsand
empiricalas well as their analyticalscrutiny.By defini- theirparticularmanifestations.Thisdistinctionis a pre-
tion, assumptionsmustbe "unrealistic"or "inaccurate" requisitefor assessingvalidity(see Adcock and Collier
in the sense of beingincomplete.But if assumptionsare 2001):Investigatingwhetherwe areindeed "measuring
manifestlyempiricallywrong-as in MiltonFriedman's whatwe thinkwe aremeasuring"(King,Keohane,and
([1953]1979,30) (in)famousexampleof the leaf-growth Verba 1994, 25) is nearly impossibleif the argument
patternon a tree-they causetwo majorproblems.First is presentedexclusivelyin operationalizedform. Sec-
is the familiarproblemof the joint hypothesistest: It ond, such a separationinduces the author to specify
will be impossible to tell whether empiricaldiscon- whichpartsof the explanationare strictlyhistorically-
firmationof a hypothesisderived from a model with contextually specific and to differentiatethem from
such assumptionsis due to a flaw or omission in the the parts that constitute a "potentiallygeneralizable
model or due to the erroneous assumptions.Second, model" (Sewell 1996, 270) in the sense of capturing
when predictionsthat are based on manifestlywrong insightsthat should,undercertainconditions,be appli-
assumptionsare confirmedempirically,such findings cable more broadly.This differentiation,in turn,facil-
are not very useful,becausethe models based on such itates the identificationof other cases,if any,on which
assumptionsprovide little insight into how the out- the argumentcould subsequentlybe tested.7
comes came about (Coase [1981]1994,17). Policythat
seeks to achieve or avoid these outcomes,if based on
such models, is likely to be ineffective and might be
5 For instance, in a paper on popular support for European Union
outrightdangerous. membership in the 1994 Austrian referendum (Biithe, Copelovitch,
The finalpoint to be made before turningto a defi- and Phelan 2002), we start deductively from general theories of
nition of "history"concernsthe roles of deductionand European integration and political economy to model support as a
inductionin modeling.Deduction is of centralimpor- function of, among other factors, the economic interests of industrial
sectors. We then work inductively from a qualitative analysis of the
tancefor the ex anteidentificationof the elements-the
public debate before the referendum to specify those industrial sec-
types of (f)actors-that constitutethe buildingblocks tors whose interests had salience for the public and might therefore
of a possible model. It is also the centralheuristictool be expected to have affected the outcome of the referendum. We
for formulatinglogically sound relationshipsamong test the resulting specified model quantitatively, using referendum
these elements, consistent with the model's core on- results from Austria's 121 districts as the dependent variable.
6 The implied oversight of this caveat renders problematic the fol-
tologicalassumptionsand consistentwith the assump- lowing methodological advice by Bates et al. (1998, 16): "The authors
tions drivingthe causalmechanisms,whichthemselves derive [testable hypotheses] from theory; but when the case materials
are usuallyderived from very general theories of the do not confirm their expectations, they do not respond by rejecting
their models. Rather, they respond by reformulating them and by
constraints, motivations, and cognitive processes em-
altering the way in which they think about the problem."
ployed in decision making and thus shaping human 7 This is not to say that historical works that do not use models cannot
agency. But a model based solely on deduction from be very insightful. Barrington Moore's (1996) classic, Social Origins
assumptions is a shell consisting of "empty theoreti- of Dictatorship and Democracy, for instance, presents an argument
cal boxes" (Bonnell 1980, 162; see also Smelser 1967, that is so highly contextualized that Victoria Bonnell (1980, 170)
finds that "his generalizations.., cannot be reduced to ... models."
22ff, 32); it needs empirical content in order to make And yet, it has provided insights and inspiration to several strands
predictions about what we should observe in a particu- of research on regime change, class formation, and processes of so-
lar instance. To be sure, this content could be supplied cial transformation. Similarly, the theoretical model remains largely
by further assumptions that are only subsequently sub- implicit in Philip Nord's (1995) The Republican Moment. Yet its rich
and vivid account of the transformation of Parisian civil society from
jected to empirical scrutiny (Kiser and Hechter 1998, the 1860s to the 1880s, focusing on the emergence of democratic cul-
799ff, 802ff), but if we truly had no empirical knowl- tures and norms of conflict resolution in autonomous spheres of civil
edge at all, what would be the basis for such conjec- society prior to the successful political democratization of 1871, is
tures? In practice, therefore, the empirical content will surely very insightful for theorists and practioners of democratization

483

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

Whensuccessbreedssuccess,whenvariablesfeed backinto whichever specific conception of temporality we adopt,


themselves,we have an excitingstory to tell, but unlesswe it calls our attention to the issue of sequence, and it
knowits metaphors[its model]... we haveno wayto tellit. injects a dynamic element. Dynamics complicate the
McCloskey(1991,36) modeling task; sequence enables it.

MODELSAND HISTORY:THEPROBLEM Dynamics


OF TEMPORALITY
The institutions within which actors interact are social
Several scholars have suggested in recent works that constructs, as are the aggregate actors that populate
the social scientific study of "history" raises particular so many of our models in political science. Due to
methodological problems (e.g., Bates et al. 1998; Elster factors such as uneven growth, increasing or dimin-
2000; Lustick 1996; Pierson 2000). Yet none of them ishing marginal utility, and accumulation or ratcheting
defines what he means by history, not even Terrence effects (e.g., Fearon 1998; Pierson 2000; Thelen 1999,
McDonald (1996b), in his programmatic essay, "What 392ff) as well as the tendency of actors to attempt to
We Talk about When We Talk about History." So what
manipulate or escape constraints (Almond and Genco
is it that makes history distinctive? 1977, 493, 518f), the passage of time makes it, ceteris
Conceptualizing history as "what historians do" paribus, more likely that institutions, actors themselves,
seems not very fruitful. Notwithstanding some real dif- and their preferences may change. Recognizing this
ferences in stance and approach to evidence separat- dynamic aspect of temporality does not mean that ev-
ing political scientists and political historians (Ingram erything is constantly in flux. In fact, institutions and
1997), what historians study is often studied with similar aggregate actors can be extremely stable for a long pe-
methods by political scientists, sociologists, economists, riod of time (Sewell 1996, 264). But the possibility of
and others. Whether or not the study of history raises
change implies that explanations of temporally large
particular methodological problems for modeling or processes must allow for change in the constitution
the presentation of empirical information would then of actors (e.g., Ertman 1997; Spruyt 1994; Tilly 1998,
become a matter of contested disciplinary conventions 7ff) as well as for change in their preferences. In the
rather than of characteristics of the subject matter (Tilly sense of the inherent dynamism of temporality, then,
1990). history is "the study of changes of things that change"
A conception of history as the study of "things of the
(Herbert Butterfield, as quoted in Schroeder 1997, 67).
past" is also not very promising. Studying past events Models of "history" must explain stability rather than
may-but need not-require the researcher to work assume it.9
with sources that call for particularmethodological, his- Because changes in preferences are all too easily in-
toriographic tools (McNeill 1998, 4f). Moreover, every- voked to explain changes in behavior and to claim the
thing that we can study empirically must already have unsuitability of models that hold preferences constant,
happened and, in that sense, is a "thing of the past." If it is important to specify what I mean by preferences.10
"history" is distinctive, then it must be conceptualized I follow Jeffry Frieden in defining preferences as those
as a set of phenomena with distinct characteristics. interests of a given actor that determine how the actor
I submit that history-as an object of study that may rank-orders the possible outcomes (1999, 42). Prefer-
require a distinct approach to theorizing and to the ences thus must be differentiated from the actions that
presentation of empirical information-must consist the actor may undertake to achieve any particular out-
of macroprocesses that cover an extensive temporal come ("strategies" in game theory parlance), but they
space. How does this definition raise particularmethod- should also be sufficiently specific to the situation that
ological issues? If the process itself is our explanandum, they can unambiguously yield a rank-ordering of the
then isolating events from the historical process within outcomes that would result from the conceivable ac-
which they occur risks depriving us of understanding tions in that situation. Consequently, "what are consid-
because any one event in such a temporal sequence, ered preferences in one [context] might be strategies in
far from being an independent observation, is mean- another" (Frieden 1999, 41).
ingful only if seen as part of the larger process (Elias A change in preferences becomes more likely over
[1937] 1997, 80ff, [1939] 1997, 390; Mink [1966] 1987, time for two reasons. First, new ideas arise over time.
64f, 67, 80, 82f). Temporality thus becomes the defining While Mark Blyth (1997) rightly warns against a facile
characteristic of "historical" explananda. There are, to attribution of causal force to ideas, ideas can indeed
be sure, several ways of conceptualizing temporality
(Aminzade 1992; Griffin 1992; Sewell 1996), not least
since time itself can be understood as a social construct; Hintze [1906] 1975; Tilly 1975, 1992) will surely differ from those that
and the significance of the passing of time depends might be relevant when we are concerned with the decision-making
process that leads to a particular policy decision.
upon the level of abstraction at which we work.8 But This is one of the strengths of Barry Weingast's chapter in Analytic
Narratives: His model of U.S. federal institutions allows him to ex-
plain why policy remained stable over several decades despite major
alike. I submit, however, that those insights would be clearer and economic and demographic changes that significantly increased the
more accessible if the authors had also presented an explicit theoret- numerical strength of groups opposed to the status quo and changed
ical model. some of their preferences (Weingast 1998, esp. 161ff, 184f, 188).
8 The conception and increments of time that are relevant when the 10 I thank one of the anonymous reviewers for thoughtful suggestions
process we want to explain is a vast sweep of state formation (e.g., for clarifying this issue.

484

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

cause a change in preferencesby making altogether Gibbons 1992, 55ff, 173ff), and the extensive form
new outcomes available,which forces a reorderingof used to depict and analyze such dynamic games in-
the outcomes.The rise of Keynesianismprovides an deed "take[s]sequenceinto account"(Bateset al. 1998,
illustrationof this phenomenon.At the outset of the 14)-although it is based on a truncatedconceptionof
Great Depression,certainsocial actors failed to rank temporalityin that, withinsuch game models, "actual
firstamongtheirpreferencesa Keynesiandemand-side chronologyis importantonly insofar as it influences
stimuluspolicyandfailedto influencepolicymakersac- what one player knows about the actions of the sec-
cordingly,even thoughwith hindsightthe adoptionof ond" (Kreps 1990, 18). Models of historicalprocesses,
such a policy would have been the outcome that max- in contrast,need to derivethe constitutionof the actors
imized their material utility.Later, similarlysituated and explanatoryvariablessuch as actors'preferences
actorsindeed preferredsuch a policyover alternatives withinthe model in orderto allow for change (cf., e.g.,
and employedvariousstrategiesto achieve the adop- JacksonandNexon 1999,302ff).At the sametime,such
tion of Keynesianpoliciesby governments.Thisfinding endogenizationof explanatoryvariablesto capturedy-
does not suggest that, at the earlierpoint in time, the namicchange(and explainstasiswhereit occurs)does
actorswere less instrumentallyrationalin rankingthe not requirea fundamentallydifferentapproachto the-
outcomes and selecting their strategiesaccordingly- orizing.Some recent work in economics,for instance,
the "outcome"of Keynesianismjust had not been for- relies on only minor modificationsof rationalchoice
mulatedyet (Hall 1989). assumptionsto develop dynamicmodels of preference
Second, ideas and especially norms, being social formation(e.g., Becker 1996),which could be used as
constructs,can change (Ball, Farr,and Hanson 1989; a buildingblock of a largermodel of an historicalpro-
Ruggie1983;Wendt1999,esp. 113ff).An examplefrom cess. Evolutionarymodels developed in biology seem
the literatureon regimechangeillustratesthe resulting to be well suitedto being adaptedto explainthe socio-
changein preferencesnicely:Assume that policies are political processes of persistence and change in the
at least in part a functionof regime type, and assume knowledge,values,andhabitsto whichwe customarily
furtherthat regime type has no effects on the citizens refer as "culture"(Boyd and Richerson1985).In sum,
otherthanthroughpolicy(thereareno normativecom- endogenizingexplanatoryvariablesdoes not require
mitments to a particularregime type). Social actors a fundamentallynew approachbut can be achieved
(individualsor groups) may, under these conditions, throughbuildingon, or adapting,variousexistingtypes
be expectedto have preferencesover (i.e., rank-order) of models.
the regime types based on which regime type affords Endogenizing explanatory variables, however,
them the greatestimpacton policy(e.g.,Bracher[1955] comes at the expense of parsimonyor worse:Scholars
1971;Hallgarten1952;Lepsius1978;Przeworski1991, who seek causalexplanationsusuallyfrownuponendo-
passim, esp. 51ff). If actors now develop over time a genization because when the dependent variable is
normativecommitmentto a democraticform of gov- not only explained by, but also (partly) explains the
ernment, they will probablyrank-orderthe possible independentvariables,we run the risk of circularrea-
outcomes of a regime change differentlybecause the soning.Canwe avoidthisproblem?Sequenceprovides
utilitythatthey assignto the outcome"democracy"has the answer.
increasedrelativeto all other outcomes (e.g., Bermeo
1992;Di Palma 1990;Weingast1997). The actorsmay
stillrank-ordertheirstrategiesas beforeandmaythere- Sequence
fore choose the same action(for instance,acquiescence Sequenceallowsus to endogenizethe explanatoryvari-
to the rule of the currentnondemocraticregimeif the ables without havingto abandonmodelingand scien-
perceived risks and costs associated with doing any- tific aspirationsbecauseit enables us to avoid circular
thingelse renderotherpossiblestrategiesprohibitive), reasoning.Endogenizationinvolvesincorporatinginto
but theirpreferenceshave changed.1" the modelsomevariationof causalfeedbackloops from
The dynamic quality of temporality suggests that the explanandumto the explanatoryvariables.In a
models based on assumptionsof stable institutional static model, such feedbackloops make the argument
contexts, stable preferences, and constant units for circular.Determiningcausalitythen becomesimpossi-
which we record variable, independent attributes at ble. The sequential element of temporality, however,
any given point in time would be unsuited if we are gets us around this problem, because it allows us to
concerned with explaining history, understood as a have causal feedback loops from the explanandum at
macroprocess. Yet taking temporality seriously does one point in time to the explanatory variables at a later
not require abandoning modeling as such. In fact, stan- point in time only.'12
dard game theoretic models can incorporate dynamic The enabling effect of sequence is nicely illustrated
elements (for an introduction, see Brams 1994, and by an example of a causal feedback loop in the chapter
on education in Abram de Swaan's (1988) In Care of
11 Higher-level preferences surely remain the same (in this case,
the State, where the author seeks to explain the histor-
for instance, the preference for longer life or safer possessions,
which may be threatened by the current regime's sanctions against
ical process by which elementary education, once seen
prodemocracy activists). But-and this is the key to the differentia-
tion between preferences and strategies used here-those unchang-
12 Note that
ing higher-level preferences are insufficient by themselves to explain anticipated reactions can undermine the assumption that
the rank-ordering of outcomes in the situation at hand. events at time t are independent of events at time t + 1.

485

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

as an individual and local responsibility, came to be quantitative or qualitative methodology). Where on


provided through nationwide collective and compul- this continuum a given work falls does not affect its
sory arrangements, administered or at least regulated usefulness or insightfulness but should affect our confi-
by the state. In de Swaan's model, an initial, partial dence in its conclusion, keeping in mind that any single
success in increasing the scope and raising the qual- work usually exists in the context of larger theoreti-
ity of public education increases the opportunities for cal and empirical literatures. At any given point along
cross-regional commercial activities and the effective- the continuum, multiple analytical techniques are at
ness of the state bureaucracies by raising the uniformity our disposal. From the viewpoint of the presentation
of language and knowledge. Over time, both of these of empirical information, we can differentiate broadly
consequences of increased and improved education between a primarily quantitative presentation, based
should be expected to swell the ranks of the "metropoli- on statistical analysis, and narratives, based on qualita-
tan elites" (supraregionally trading entrepreneurs and tive techniques such as process-tracing. In this section,
central state bureaucrats) who favor the provision of I spell out what makes narratives a particularly suit-
elementary education as a public good. A change in able form of presenting empirical information when
the explanandum, education, thus leads to a change we want to test models about historical processes, and
in the relative power of the actors in the conflict over I suggest ways of structuring narratives to attain this
the scope and quality of education at a later stage. It benefit.
strengthens the proponents of widened and improved Why narratives? Practical constraints, such as too
elementary education at the expense of the opponents, few instances of a given macrohistorical process, may
such as local elites and clergy, and in turn should lead to inhibit the use of statistical techniques. Yet time series
a further increase in the scope and quality of education. analysts have long used lagged variables, and in recent
Time itself thus becomes an element of the causal years some scholars have begun to develop more so-
explanation, a factor in the model. But time does not phisticated procedures to control for time dependence
function as a standard explanatory variable that di- in political phenomena (Beck and Katz 1996; Beck,
rectly affects the explanandum-otherwise, its effect Katz, and Tucker 1998). In short, we have very suit-
could easily be expressed through a linear differential able statistical tools for the empirical analysis of models
equation (McCloskey 1991, 22ff). Rather, it operates in with dynamic and sequential elements, although if the
the background to affect several explanatory variables nature of the explanandum indeed is such that all stages
in a variety of ways. Models that seek to help us explain within each instance of the historical process (the ex-
historical processes qua processes therefore must ex- planandum) are interrelated, one would need to incor-
plicitly incorporate a temporal dimension and consider porate into the statistical model a lag for each variable
carefully how each explanatory factor is affected by at each stage of the process prior to the final stage,
the passage of time in the process that we are trying to with a corresponding rapid decrease in the degrees of
explain. freedom.13 Moreover, various elements of the model
In sum, the importance of temporality is the distinc- and additional implications may be separately tested
tive characteristic of "historical"phenomena as objects using statistical techniques. Statistical methods can thus
of study, which raises particular methodological issues. be used in the empirical analysis of historical processes
The dynamic element of temporality complicates the and may be particularly valuable when complementing
modeling task by demanding that our models allow for and reinforcing insights gained from qualitative tech-
the possibility of change (and hence explain rather than niques (cf. Beck, Katz, and Tucker 1998;Berg-Schlosser
assume stability), but the incorporation of a sequen- and Quenter 1996; Mahoney 1999).
tial element enables us to do this without running the A preference for narratives, then, is due not to
risk of the circular reasoning often associated with en- the unavailability of analytical techniques that lead to
dogenizing the independent variables. Yet temporality other forms of presenting our results, but to particular
raises additional problems when we move to empirical strengths of the narrative form. The most important
testing. of these strengths is that narratives, in addition to pre-
senting information about correlations at every step
Model buildingis importantfor workingout the internal of the causal process, can contextualize these steps in
logic of a chosenset of assumptionsand relationships.But ways that make the entire process visible rather than
rigorousempiricalanalysisis neededto ensuretherelevance leaving it fragmented into analytical stages. Moreover,
of thoseassumptionsand relationships. narratives allow for the incorporation of nuanced detail
Lazonick(1991,303, emphasisadded) and sensitivity to unique events, which may be neces-
sary to understand the particular manifestation of an
element of the model but which are beyond the model.
CONFRONTING MODELS WITH DATA:
HISTORICAL NARRATIVES
13 The
problem here is neither a lack of information, since infor-
Ways of confronting a model with data range from mation about the pertinent variables at each stage of the process is
showing the consistency of the theoretical argument equally required for qualitative-narrative techniques such as process-
with one empirical observation of the explanandum tracing, nor quantification per se, which, at least at the basic level of
more versus less, is usually possible. Collaborative database projects
(Eckstein's [1975, 108ff] "plausibility probe") to a test show that such complex information can be gathered for many cases.
on a large sample that meets the criteria of statis- All of these techniques of course require carefully specifying the
tical analysis (irrespective of whether we employ a functional equivalents across different instances of the process.

486

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

In de Swaan'snarrativeof the introductionof state- that leads to the ad hoc-nessof manyinductivehistor-


provided educationin The Netherlands,for instance, ical explanations.
he notes how the Napoleonicinvasionmadethe central
metropolitanelite suddenlyexceptionallyindependent HISTORICAL
NARRATIVESAS DATA:FOUR
of local elites and clergy,thus strengtheningthem in
the societal group conflict over the expansion of the PROBLEMSAND (PARTIAL)SOLUTIONS
state,includingthe expansionof publiceducation.This Notwithstandingthe above strengths of narratives,
uniqueevent affectedthe speed of changeandthusthe their use as databringsto lightfour problemsand lim-
particularmanifestationof the process,but it did not itations not identifiedin recent methodologicalwork
changethe generaldynamicof the processas captured and debates on history,narratives,and social science
by the model. theory.16The first two problemsare particularto the
At the same time, narrativesmust not revertto un- use of narrativesas datain conjunctionwith models of
theorizedhistoricalaccounts,invokingextraneousfac- historicalprocesses;the thirdand fourth applyto any
torsin an ad hoc fashion,becausesuchaccountsarenot use of narrativesas data.
useful as a test of the causalpropositions.How can we The first problem concerns definingor delineating
avoid this problem?The model itself can help us write historicalsequencesas distinct,or whatBearman,Faris,
narrativesthat are useful as a test of the argument. and Moody call "casing"(1999;see also Mink [1966]
The theoreticalmodel can be used to structurethe 1987).If we conceiveof historyas a continuousstream
narratives.As we know from Arthur Danto's (1965, of interrelatedevents,then at the logicalextremethere
149ff) thoughtexperiments,even the imagined"ideal is no beginningand no end. A narrative,however,in-
chronicler"who recordsevery actionand event in per- herentlyimposes a beginningand end onto the histor-
fect chronologicalorder cannot provide a complete ical record. To be sure, all empiricalwork-whether
history,let alone a causal explanation.Any historical large-N statisticalor small-N case studywork-needs
narrativethereforemustsimplify"reality"by designat- to justify the boundariesof its units of analysis,espe-
ing some elements as salient and omittingmany more ciallywhen proceedingon the assumptionof indepen-
as not significant(McClelland1975, 75ff). The model dence of observations(cf. King, Keohane, and Verba
can help by providingthe criteriafor what is salient: 1994, 222). But the emphasison the interrelatedness
The actorsidentifiedin the model constitutethe actors of events acrosstime makes this problemparticularly
of the narrative,which traces their goals, beliefs, and acute for the empiricaltesting of models of history,
actions.Withineachnarrativewe thusemployprocess- whilethe literaryand"aesthetic"(Topolsky1998)qual-
tracing (Bennett 1999; George and McKeown 1985). ities of the narrativetend to obscureit.
The influenceof other elements of the model, such as The model can providea deductive,albeit only par-
how temporalprogressionaffects the actorsand their tial, solution to this problem. By specifying the ex-
preferences, should be systematically described.14 Be- planandumin generaltermsandtheorizingtemporality
yond the elements identifiedin the model, however, explicitly,the modeldelineatesa sequenceas distinct-
additionalcontext-specificinformationshouldbe min- based on our research objectives-in the potentially
imized. Informationthat is extraneousto the model infinitespaceof time.Note, however,thatthe explanan-
should be provided only insofar as it affects salient dumitselfprovidesajustificationforchoosingthe start-
elements and is needed either to understandthe re- ing and endingpoints of the narrativeonly insofaras
lationshipbetween these elementsor to appreciatethe the historicalprocess we seek to explain can plausi-
contingenciesof a particularhistoricalprocess.If the bly be said to have had a clear startingpoint (e.g.,
narrativecannot be written in terms of the model, an exogenous shock) and to have "run its course."
somethingis wrongwith the model. A numberof sociologistshave in recent years devel-
Using the modelin thiswayto disciplinethenarrative oped alternativeinductiveprocedures,often grounded
ensures equivalencein the sense that each narrative in the boundary specificationapproach of network
contains the same (or at least functionally equiva- theory, for delineating an event sequence as a dis-
lent) elements.Each narrativethus becomes a unit or tinct process and hence justifyinga distinctnarrative
"observation"--a"plausibilityprobe"to test the causal withclearend points:"eventstructureanalysis"(Heise
argument, though we may not be able to assume inde- 1993), "abstraction and generalization of interactions"
pendence for testing purposes (Sewell 1996, 258f).15 (Abell 1993), "interaction process analysis" (Kosaka
And most importantly, the model allows the analyst 1993), and "bicomponent analysis" of event popula-
to overcome the problem of deciding what matters for tions (Bearman, Faris, and Moody 1999). However,
the narrative (cf. Mink [1978] 1987, 187f)-the problem questions remain about the logic underpinning these
procedures (see Abell et al. 1993), and I doubt their
general applicability to macrosocial phenomena. This
14
Carefully tracing the positions taken by each of the actors identi- leaves us with a clear specification of the explanandum
fied in his model, de Swaan shows, for instance, that over time (for as the only generalizable, methodological justification
reasons consistent with the model, such as a growing cohesion of a
group-social actor), the demands of working-class parents intensified,
whereas "time" reversed the preferences of industrialists. 16 For
prominent contributions to those debates see Abell et al. 1993,
15 I address the issue of independence below, when discussing the use Elman et al. 1997, King, Keohane, and Verba 1994, Kohli et al. 1995,
of multiple narratives as a remedy for what I identify as the "third Laitin et al. 1995, Mahoney 1999, Ragin, Berg-Schlosser, and Meur
problem" of historical narratives as data. 1996, Schroeder 1994, and Somers et al. 1998.

487

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

for making choices about where to begin and end a Third, what is the status of any narrative'struth
narrative. claim?As Mink ([1978]1987, 199) put this "dilemma
De Swaan(1988),whose use of historicalnarratives of the historicalnarrative":
is exemplary,achievessucha delineationof a temporal
... As historicalit claims to represent,throughits form,
sequence by defininghis explanandumas the process part of the real complexityof the past, but as narrative
bywhichelementaryeducationwastransformedfroma it is a productof imaginativeconstruction,whichcannot
privateandlocal affairinto a responsibilityof the state, defend its claim to truth by any accepted procedureof
providedat a basiclevel throughoutthe territorywithin argumentor authentication.
its reach,accordingto certainstandards(e.g.,a uniform
language)andfinancedthroughcompulsorymeasures. The problem here, as Andrew Norman (1991, 131)
This definitionof the explanandumsuggests that the pointsout, is not that"adiscursiverepresentationhas a
narrativemust startwith the initial moves (ideational structurethatthatwhichit representsdoes not."There
or practical)awayfrom the previouslocal and private is no necessarylink between discursivestructureand
systemof schoolingbutneed not be concernedwiththat misrepresentation.Moreover,some objectivecriteria
system'spriorhistory.And the narrativecan end when for assessinga narrative'struthclaimexist, suchas the
uniform,state-financedelementaryeducationhasbeen extent to which an authoris able to providefrom the
establishedin the countryin question,withoutneeding historicalrecordevidence that, afterhavingbeen sub-
to concernitself with the subsequentevolution of the jectedto standardhistoriographicprocedures,supports
educationalsystemor otherrelatedaspectsof the wel- the author's"story."It is therefore hardlynecessary
fare state. The specificationof the explanandumthus to equate historianswith novelists or fiction writers
provides the criteriafor choosing the beginningand (Gaddis 1992/93,56). Rather,the problemis that, be-
end of the narrative. cause facts never speakfor themselves(Lustick1996),
The second problem concerns the need to thereis an interpretativeelementthatcannotbe evalu-
"conclude"the narrativewhile the processmay be on- ated fromwithinthe narrativewithoutcircularreason-
going, which restrictsthe "generality"of our conclu- ing. To assess this aspect of the narrative,the reader
sions (King,Keohane,and Verba1994,137). If we use will have to drawon knowledgefrom sourcesexternal
feedbackloops in our models,throughwhicha change to the narrativeat hand;i.e., the reader will need to
in the value of the dependentvariableis hypothesized know somethingabout the historicalperiod in ques-
to change the value of some or all of the explanatory tion from other sources-sources that themselvesare
variablesat a later point in time, time itself acquires bound to have narrativequalities.Consequently,it is
an explanatoryrole (in interactionwith the original more meaningfulto endorse good narrativework as
explanatoryvariables).As long as only the explanatory "plausible,""persuasive,"or "compelling"-as seems
variables,and not the explanandumitself, are affected to be the practiceamonghistorians-ratherthan"true"
by time in this way,truncatingtime by setting an end or "right,"thoughwe certainlymayfindsome narrative
point for the narrativedoes not introducethe selec- workthatis poor andeven plain"wrong,"suchas when
tion bias that is caused by truncatingthe range of the its interpretationis marredby logical inconsistencies
dependentvariable(King, Keohane,and Verba 1994, or makesincorrectassertionsabout the chronologyof
128ff).But more than with conventionalindependent events.
variables,of whichwe may restrictthe range,we have Moreover,that the truthclaimof a narrativecannot
to be verycarefulaboutassumingconsistentcontinuity be assessed from within the narrativeitself also has
of the relationshipbetween the explanatoryvariables importantimplicationsfor the utility of narrativesas
and the explanandumbeyond the investigatedperiod. data on whichto test hypothesesderivedfrommodels.
When time itself becomes a factor in the model, as Presumably,no one will consciouslypublisha model
discussed above, we have to consider carefully how with empiricalinformationthat directlycontradictsit.
each explanatoryfactor is affectedby the progression Except for discardedalternativeexplanations(see be-
of the processthat we are tryingto explain.The extent low) and possibleoversights,then, the vulnerabilityof
to which we can expect the effect of time on the ex- the model to empiricaldisconfirmationis ultimately
planandum(throughthe other independentvariables) also externalto the narrative.
to continue as observed during the time period covered However, the use of multiple narratives may increase
by the narrative depends on the tenability, beyond this our confidence that the model indeed captures the key
time period, of the assumptions that we had to make dynamics of the process. Using multiple narratives is
to model the effect of time. This limitation of historical appropriate because models should be applicable to
narratives affects the confidence we can have in the more than a single instance if they have the benefits
generalizability of the insights we gain from them. The
solution, however, lies not in modestly claiming that our
([1906] 1975), for instance, long ago provided a fascinatingly simple
conclusions cannot be generalized but, rather, in pay- example of what we would today call a "second image reversed"
ing careful attention to temporality in both the model (Gourevitch 1978) model of state formation, capturing an essen-
and the narratives and specifying the implications for tial element of a causal mechanism that might explain not only the
various transformations of the "organization of the state" over the
generalization accordingly.17
thousand-year stretch of European history to which he applies it, but
also phenomena that occurred many decades after the formulation
17 The applicability of a historical model by no means needs to end of his model, such as European integration and the later parts of the
with the time period on which it was empirically tested. Otto Hintze "third wave of democratization" (Huntington 1991, esp. 85ff).

488

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

of capturingwhatis generalizable.Havingassuredthat accountsof the abolitionof "buyingone's way out of


each narrativecontainsthe same elements,we can em- militaryservice"(Levi1998,109),in France,the United
ploy each narrativeas a unit or "observation"to test States, and Prussia,shows the same, separatelytheo-
the causal argument.To be sure, we are unlikely to rizedprocessat work.Finally,de Swaanuses five sepa-
be able to furnishsufficientlymany historicalnarra- rate narratives(for the United States,France,Britain,
tives to performstatisticaltests, and by virtueof treat- Germany,andTheNetherlands)to test his-separately
ing each historicalnarrativeas a self-containedunit theorizedthough not fully explicit-model of the his-
so as to incorporatetemporal progressioninto each toricalprocessbywhichelementaryeducationbecomes
"observation,"we have to reject any attemptto make the domain of the state. In all of these works, what
"manyobservationsfrom few" (King, Keohane, and persuadesthe readerof the validityand usefulnessof
Verba 1994,217ff). Moreover,there is the problemof the model is that for each of the countries,as Sewell
independence.18As WilliamSewell and others point (1996, 262) puts it with respect to Skocpol, the spe-
out, if separate historicalinstances of a process are cifichistoricalprocess"canbe narratedconvincinglyin
treated like ideal-typicalscientific experiments,such terms of the operationof analogouscausalprocesses,
treatmentassumesthat they are independentof each which...[thus] make sense of numerous details that
other. This assumptionis problematicbecause,unless otherwisewouldseempurelyaccidental."In this sense,
thereis a perfectinformationalseparationbetweenthe providingmultiple"plausibilityprobes,"farfrombeing
instances,actorsin one (later or perhapscontempora- "fruitless[ly]repetiti[ous]"(Skocpoland Somers1980,
neous) instance will have knowledge of the constitu- 191), should enhance confidencein the explanation.
tive actions and the outcome of the other instance(s) This way of combiningmodels with narrativesallows
(Sewell 1996, 258f). Such knowledge will violate the us to providea "scientific"causalexplanationof histor-
assumptionof independence-for purposesof testing ical processes,withoutdeprivingthem of their process
a model-only if it changeselements of the processas character.
modeled (e.g., the preferencesof actors), which is by Fourth,due to the limitedtruthclaimsof narratives,
no means a necessary,though a possible and indeed those who use historicalnarrativesas empiricalevi-
quite likelyconsequenceof havingknowledgeof prior dence for a causalexplanationwill probablyfail to as-
or other instances.19 Whereempiricalanalysissuggests sess alternativeexplanationsand,if they try,will fail to
such an effect, we thereforecannot rely upon the sta- convinceskeptics.Tryingto assessalternativeexplana-
tistical logic of the traditionalcomparativemethod tions would entailprovidinga second set of narratives,
(Lijphart1971). But as "plausibilityprobes"multiple similarlystructuredby the underlyingmodel of an al-
narrativesare very useful. Increasingthe number of ternative explanationfor the same phenomenon.To
confirmingnarrativesdoes not in any way "prove"the be sure, such alternativenarrativescould conceivably
model (Mohr1996,118ff),butin lightof the temptation be suppliedby the same authorjust as those working
of inductivistmodificationsof a givenmodel,the ability quantitativelyoften operationalizeand test alternative
of a model to withstandthe difficulttest of application explanations.In fact,GrahamAllison (1971)employed
to differentoccurrencesof the explanandumwithout threeseparatenarrativesto test three alternative,com-
ad hoc alterationsmakesmoreplausiblethatit hascap- peting models of foreign policymakingin Essence of
tured the central,generalizabledynamicsratherthan Decision.20But alternativenarrativesby the same au-
uniqueelements of a particularcase. thor serve primarilyas a rhetoricaldevice in support
In fact,it is the use of multiplenarrativesthat makes of the primary,favored explanation.The interpreta-
Theda Skocpol's(1979) Statesand Social Revolutions, tive freedom of the authormakes it unlikelythat less
MargaretLevi's (1998) "Conscription:The Price of convincingalternativenarrativeswouldbe acceptedas
Citizenship"(1998), and, especially,de Swaan's(1998, sound evidence of the failureof the alternativeexpla-
52-117) "The ElementaryCurriculumas a National nations. This, however, is not a serious problem.As
CommunicationCode" so convincing.To be sure,the MorrisFiorinanotes, meaningfulalternativeexplana-
extensivemethodologicaldebateaboutSkocpol'sbook tions are much more likely to be advancedby others
has produced"littleconsensus"aboutthe statusof her whose "perspectivesand commitments"allow them to
narratives(Mahoney1999, 1156), and her consciously argueas stronglyas possible in supportof those alter-
inductive approach makes it questionable to treat her natives. All such explanations must then be subjected
narratives as data to test a model, but the book derives to the collective assessment of the scholarly community
most of its persuasive power precisely from her ability at large (Fiorina 1995, 92).
to narrate three: instances of social revolutions in the
same terms. Multiple narratives work even more effec-
tively in Levi's "Conscription." Though it is sometimes
CONCLUSION
difficult to see how the various parts of her complex I have pursued two objectives: (1) I have sought to
model structure the three narratives, each of her three clarify, based on a broad conception of modeling and
an explicit definition of "history," the difficulties-
18 I thank one of the:anonymous reviewers for thoughtful comments
and possibilities-of modeling historical processes; and
on this issue.
19 Statisticians therefore
speak of seeking "independence of the error
terms,"which is a purely empirical matter, rather than "independence 20
I thank Steve Solnick for reminding me of Allison's use of alter-
of observations." native narratives.

489

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

(2) I have soughtto make the case for narratives,iden- studyof historydoes notrequirea fundamentallydiffer-
tify four overlookedproblemswith theiruse as datato ent approach.Notwithstandingsome remaininglimita-
test modelsabouthistoricalprocesses,andofferat least tions, historicalnarrativescan providestrongsupport
partialsolutionsto these problems. for the plausibilityof a historicalmodel,maintaininga
I have arguedthat modelingis renderedmore dif- basic commitmentto a scientificapproachwhile over-
ficult when dealing with "history"due to the impor- coming several of the problemsof the compatibility
tance of the dynamicand sequentialaspects of tem- of historicalnarrativeand social science identifiedby
porality for processes that can be methodologically critics of the "protoscientific"ambitionsof historical
meaningfullydistinguishedas "historical."But social analyses.And it may be that at the level of broadhis-
scientistsinterestedin the studyof such"history"need torical phenomena,where the temporaldimensionis
not returnto narrativesthat, detached from any ex- cruciallyimportantto any explanation,providingsuch
plicit causal model of the historicalprocess they seek supportfor the plausibilityof the argumentis the best
to capture,implicitlymake-often sweeping--causal we can hope for, at least fromany one scholar.Science
claims based on hidden assumptionsand unspecified is, afterall, a collectiveenterprise(Weber[1919]1946).
causalmechanisms.Thereis no inherentincompatibil-
ity betweenthe deductivemodelingof causalrelation-
shipsandthe studyof history.Quiteto the contrary,for REFERENCES
social scientists-and historians-who seek causalex-
Abell, Peter. 1993. "Some Aspects of Narrative Method." Journal of
planationsof historicalprocesses,makingtheirmodels Mathematical Sociology 18 (2-3): 93-134.
explicitandsubjectingthemto separatetheoreticalex- Abell, Peter, et al. 1993. "Symposium on Narrative Method." Journal
aminationhave severaladvantages.Most importantly, of Mathematical Sociology 18 (2-3): 93-266.
the greater transparencyof a model (compared to Achen, Christopher H., and Duncan Snidal. 1989. "Rational Deter-
a purely "operationalized"argument)facilitates the rence Theory and Comparative Case Studies." World Politics 41
(January): 143-69.
scrutinyof the argument'sassumptionsand internal Adcock, Robert, and David Collier. 2001. "Measurement Validity:
logic,whichis particularlydesirablefor inherentlycom- A Shared Standard for Qualitative and Quantitative Research"
plex historicalphenomena.Further,models help clar- American Political Science Review 95 (September): 529-46.
ify what insightsthe authorsuggeststo the reader,by Allison, Graham. 1971. Essence of Decision. New York: Harper
Collins.
inducingthe authorto specify which elements of the Almond, Gabriel A., and Stephen Genco. 1977. "Clouds, Clocks, and
argumentare,in the author'smind,uniquelyspecificto the Study of Politics." World Politics 29 (July): 489-522.
the historicalsetting and whichpartsof the argument Aminzade, Ronald. 1992. "Historical Sociology and Time." Sociolog-
are potentiallygeneralizable.In fact,the model should ical Methods & Research 20 (May): 456-80.
be epistomologicallypriorto, and independentof, the Anderson, Benedict. [1983] 1991. Imagined Communities. New York:
Verso.
narrative(s)used to "test" it, in order to enable the Ball, Terence, James Farr, and Russell Hanson, eds. 1989. Political
model-rather than the narrative-to do the explana- Innovation and Conceptual Change. New York: Cambridge Uni-
tory work (cf. Somers1996,79ff). versity Press.
Regardingthe benefits of using narratives,I have Barry,Brian. 1985. "Does Democracy Cause Inflation? Political Ideas
of Some Economists." In The Politics of Inflation and Economic
emphasizedthe ability of narrativesto contextualize Stagnation: TheoreticalApproaches and International Case Studies,
any given stage in the process,whichis essentialwhen eds. Leon N. Lindberg and Charles S. Maier. Washington, DC:
we seek to captureempiricallythe process for which Brookings Institutions, 280-317.
the model stipulatesan explanation.To be useful as a Bates, Robert H., et al. 1998. Analytic Narratives. Princeton, NJ:
test of a deductivelysoundmodel,a narrativeshouldbe Princeton University Press.
structuredby the model in thatthe presentationof em- Bearman, Peter, Robert Faris, and James Moody. 1999. "Blocking
the Future: New Solutions for Old Problems in Historical Social
piricalinformationfollowsthe model'sidentificationof Science." Social Science History 23 (Winter): 501-33.
actors,their preferences,etc., so as to minimizethe ad Beck, Nathaniel, and Jonathan N. Katz. 1996. "Nuisance vs. Sub-
hoc characterof the empiricalaccount. stance: Specifying and Estimating Time-Series-Cross-Section
Models." Political Analysis (Ann Arbor) 6 (1996): 1-36.
At the same time, using narrativesas data to test a
Beck, Nathaniel, Jonathan N. Katz, and Richard Tucker. 1998.
model thatseeks to capturean entirehistoricalprocess "Taking Time Seriously: Time-Series-Cross-Section Analysis with
raisesfour problemsnot identifiedin recentdebates.I a Binary Dependent Variable." American Journal of Political
have specified these problems and analyzed their impli- Science 42 (October): 1260-88.
cations for the usefulness of narratives as data. I have Becker, Gary S. 1996. Accounting for Tastes. Cambridge, MA:
Harvard University Press.
suggested ways of solving or minimizing the problems, Bennett, Andrew. 1999. "Causal Inference in Case Studies:
including the use of multiple narratives to increase our From Mill's Methods to Causal Mechanisms." Paper presented
confidence that the model indeed captures the key ele- at APSA, Atlanta, 1999. <http://www.georgetown.edu/faculty/
ments of a causal explanation of the processes that we bennetta/APSA99.html> (March 22, 2000).
are seeking to explain. Berg-Schlosser, Dirk, and Sven Quenter. 1996. "Makro-quantitative
vs. makro-qualitative Methoden in der Politikwissenschaft:
So does the study of history, understood as the anal- Vorztige und Maingelkomparativer Verfahrensweisen am Beispiel
ysis of processes that span a large temporal space, der Sozialstaatstheorie." Politische Vierteljahresschrift37 (Mdirz):
require a distinct approach to theorizing and to the 100-18.
Bermeo, Nancy. 1992. "Democracy and the Lessons of Dictatorship."
presentation of empirical information? I conclude that Comparative Politics 24 (April): 273-91.
it indeed engenders particular problems and requires Blyth, Mark M. 1997. "Any More Bright Ideas?' The Ideational
us to be attentive to issues that may not arise in analyses Turn of Comparative Political Economy." Comparative Politics 29
of phenomena where temporality plays no role. But the (January):229-50.

490

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

Bonnell, Victoria E. 1980. "The Uses of Theory, Concepts and Com- Geddes, Barbara. 1990. "How the Cases You Choose Affect the An-
parison in Historical Sociology." Comparative Studies in Society swers You Get: Selection Bias in Comparative Politics." Political
and History 22 (April): 156-73. Analysis (Ann Arbor) 2 (1990): 131-150.
Boyd, Robert, and Peter J. Richerson. 1985. Culture and the Evolu- George, Alexander L., and Timothy J.McKeown. 1985. "Case Studies
tionary Process. Chicago: University of Chicago Press. and Theories of Organizational Decision Making." Advances in
Bracher, Karl Dietrich. [1955] 1971. Die Auflisung der Weimarer Information Processing in Organizations 2 (1985): 21-58.
Republik. 5th rev ed. Villingen: Ring Verlag. Gibbons, Robert. 1992. Game Theory for Applied Economists.
Brams, Steven J. 1994. Theory of Moves. New York: Cambridge Princeton, NJ: Princeton University Press.
University Press. Gourevitch, Peter. 1978. "The Second Image Reversed: The Inter-
Blithe, Tim, Mark Copelovitch, and William Phelan. 2002. "The national Sources of Domestic Politics."International Organization
Domestic Politics of European Integration: Public Opinion, Ref- 32 (Autumn): 881-911.
erenda, and EU Membership." Paper presented at the Annual Griffin, Larry J. 1992. "Temporality, Events, and Explanation in
Convention of the International Studies Association (ISA), New Historical Sociology: An Introduction." Sociological Methods &
Orleans, March 2002. Research 20 (May): 403-27.
Calhoun, Craig. 1998. "Explanation in Historical Sociology: Narra- Hall, Peter A., ed. 1989. The Political Power of Economic Ideas:
tive, General Theory, and Historically Specific Theory."American Keynesianism Across Nations. Princeton, NJ: Princeton University
Journal of Sociology 104 (November): 846-71. Press.
Coase, R. H. [1981] 1994. "How Should Economists Choose?" In Hall, Peter A., and Rosemary C. R. Taylor. 1996. "Political Science
Essays on Economics and Economists. Chicago: University of and the Three New Institutionalisms." Political Studies (Oxford)
Chicago Press, 15-33. 44 (December): 936-57.
Collier, David. 1995. "TranslatingQuantitative Methods for Qualita- Hallgarten, George W. E 1952. "Adolf Hitler and German Heavy
tive Researchers: The Case of Selection Bias." American Political Industry."Journal of Economic History 12 (Summer): 222-46.
Science Review 89 (June): 461-66. Heise, David R. 1993. "Narratives Without Meaning?" Journal of
Danto, Arthur C. 1965. Analytical Philosophy of History. Cambridge: Mathematical Sociology 18 (2-3): 183-89.
Cambridge University Press. Hintze, Otto. [1906] 1975. "Military Organization and the Organiza-
de Swaan, Abram. 1988. In Care of the State:Health Care, Education tion of the State." In The Historical Essays of Otto Hintze, ed. Felix
and Welfarein Europe and the USA in the Modern Era. New York: Gilbert. New York: Oxford University Press, 178-215.
Oxford University Press. Huntington, Samuel P. 1991. The Third Wave:Democratization in the
Di Palma, Guiseppe. 1990. To CraftDemocracies: An Essay on Demo- Late TwentiethCentury. Norman: Oklahoma University Press.
cratic Transitions. Berkeley: University of California Press. Immergut, Ellen M. 1998. "The Theoretical Core of the New Insti-
Eckstein, Harry. 1975. "Case Study and Theory in Political Science." tutionalism." Politics & Society 26 (March): 5-34.
In Handbook of Political Science, eds. Fred I. Greenstein and Ingram, Edward. 1997. "The Wonderland of the Political Scientist."
Nelson W. Polsby. Reading, MA: Addison-Wesley, 79-137. International Security 22 (Summer): 53-63.
Elias, Norbert. [1937] 1997. Wandlungen des Verhaltens in den Jackson, Patrick Thaddeus, and Daniel H. Nexon. 1999. "Relations
weltlichen Oberschichtendes Abendlandes. Vol. 1, Uberden Prozepf Before States: Substance, Process, and the Study of World Politics."
der Zivilisation. 20th ed. Frankfurt am Main: Suhrkamp. European Journal of International Relations 5 (September): 291-
Elias, Norbert. [1939] 1997. Wandlungender Gesellschaft:Entwurfzu 332.
einer Theorie der Zivilisation. Vol. 2, Uber den Prozep der Zivili- Jervis, Robert. 1990. "Models and Cases in the Study of Interna-
sation. 20th ed. Frankfurt am Main: Suhrkamp. tional Conflict." Journal of International Affairs (New York) 44
Elman, Colin, et al. 1997. "Symposium on Diplomatic History (Spring/Summer): 81-101.
and International Relations Theory." International Security 22 Katznelson, Ira. 1997. "Structure and Configuration in Comparative
(Summer): 5-85. Politics." In Comparative Politics: Rationality, Culture and Struc-
Elster, Jon. 1989. Nuts and Bolts for the Social Sciences. New York: ture,eds. Mark Irving Lichbach and Alan S. Zuckerman. New York:
Cambridge University Press. Cambridge University Press, 81-112.
Elster, Jon. 2000. "Rational Choice History: A Case of Excessive King, Gary, Robert O. Keohane, and Sidney Verba. 1994. Designing
Ambition." American Political Science Review 94 (September): Social Inquiry: Scientific Inference in Qualitative Research. Prince-
685-95. ton, NJ: Princeton University Press.
Ertman, Thomas. 1997. Birth of the Leviathan: Building States and Kiser, Edgar, and Michael Hechter. 1998. "The Debate on Histori-
Regimes in Medieval and Early Modern Europe. New York: cal Sociology: Rational Choice Theory and Its Critics."American
Cambridge University Press. Journal of Sociology 104 (November): 785-816.
Evans, Peter. 1995. "The Role of Theory in Comparative Politics." Kohli, Atul, et al. 1995. "The Role of Theory in Comparative Politics:
World Politics 48 (October): 2-10. A Symposium." World Politics 48 (October): 1-49.
Fearon, James D. 1991. "Counterfactuals and Hypothesis Testing in Kosaka, Kenji. 1993. "Towards a Further Analysis of Narratives."
Political Science." World Politics 43 (January): 169-95. Journal of Mathematical Sociology 18 (2-3): 141-51.
Fearon, James D. 1998. "Bargaining, Enforcement, and International Kreps, David. 1990. Game Theory and Economic Modeling. New
Cooperation." International Organization 52 (Spring): 269-305. York: Oxford University Press.
Finnemore, Martha. 1996. "Constructing Norms of Humanitarian In- Laitin, David D. 1995. "Disciplining Political Science." American
tervention." In The Culture of National Security:Norms and Iden- Political Science Review 89 (June): 454-56.
tity in WorldPolitics, ed. Peter J. Katzenstein. New York: Columbia Laitin, David D., et al. 1995. "The Qualitative-Quantitative Disputa-
University Press, 153-85. tion: Gary King, Robert O. Keohane, and Sidney Verba's Design-
Finnemore, Martha, and Kathryn Sikkink. 1998. "International Norm ing Social Inquiry: Scientific Inference in Qualitative Research."
Dynamics and Political Change." International Organization 52 American Political Science Review 89 (June): 454-81.
(Autumn): 887-917. Lakatos, Imre. [1970] 1974. "Falsification and the Methodology of
Fiorina, Morris P. 1995. "Rational Choice, Empirical Contributions, Scientific Research Programmes." In Criticism and the Growth
and the Scientific Enterprise." Critical Review (New York) 9 of Knowledge, eds. Imre Lakatos and Alan Musgrave. London:
(Winter-Spring): 85-94. Cambridge University Press, 91-196.
Frieden, Jeffry A. 1999. "Actors and Preferences in International Re- Lazonick, William. 1991. "Rigor and Relevance in Economics: His-
lations." In Strategic Choice and International Relations, eds. David torical Analysis as Rigorous Economics." In Business Organiza-
A. Lake and Robert Powell. Princeton, NJ: Princeton University tion and the Myth of the Market Economy. Cambridge: Cambridge
Press, 39-76. University Press, 303-49.
Friedman, Milton. [1953] 1979. "The Methodology of Positive Eco- Lepsius, Rainer. 1978. "From Fragmented Party Democracy to Gov-
nomics." In Philosophy and Economic Theory, eds. Frank Hahn ernment by Emergency Decree and National Socialist Takeover:
and Martin Hollis. Oxford: Oxford University Press, 18-35. Germany." In The Breakdown of Democratic Regimes: Europe,
Gaddis, John Lewis. 1992/93. "International Relations Theory and eds. Juan J. Linz and Alfred Stepan. Baltimore: Johns Hopkins
the End of the Cold War."International Security 17 (Winter): 5-58. University Press, 34-79.

491

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
TakingTemporalitySeriously September 2002

Levi, Margaret. 1998. "Conscription: The Price of Citizenship." In Rosenthal, Jean-Laurent. 1998. "The Political Economy of Abso-
Analytic Narratives, eds. Robert H. Bates, et al. Princeton, NJ: lutism Reconsidered." In Analytic Narratives, eds. Robert H.
Princeton University Press, 109-47. Bates, et al. Princeton, NJ: Princeton University Press, 64-
Lijphart, Arend. 1971. "Comparative Politics and the Comparative 108.
Method." American Political Science Review 65 (September): 682- Ruggie, John Gerard. 1983. "International Regimes, Transactions,
93. and Change: Embedded Liberalism in the Postwar Economic Or-
Lustick, Ian S. 1996. "History, Historiography, and Political Sci- der." In International Regimes, ed. Stephen D. Krasner. Ithaca, NY:
ence: Multiple Historical Records and the Problem of Selec- Cornell University Press, 195-231.
tion Bias." American Political Science Review 90 (September): Schroeder, Paul W. 1994. "Historical Reality vs. Neo-realist Theory."
605-18. International Security 19 (Summer): 108-48.
Mahoney, James. 1999. "Nominal, Ordinal, and Narrative Appraisal Schroeder, Paul W. 1997. "History and International Relations The-
in Macrocausal Analysis." American Journal of Sociology 104 ory: Not Use or Abuse but Fit or Misfit."International Security 22
(January): 1154-96. (Summer): 64-74.
McClelland, Peter D. 1975. Causal Explanation and Model Building Sewell, William H., Jr. 1996. "Three Temporalities:Toward an Event-
in History, Economics, and the New Economic History. Ithaca, NY: ful Sociology." In The Historic Turn in the Human Sciences, ed.
Cornell University Press. Terence J. McDonald. Ann Arbor: University of Michigan Press,
McCloskey, Donald N. 1991. "History, Differential Equations, and 245-80.
the Problem of Narration." History and Theory 30 (February): Skocpol, Theda. 1979. State and Social Revolutions: A Comparative
21-36. Analysis of France, Russia, and China. Cambridge: Cambridge
McCloskey, Deirdre N. 1998. The Rhetoric of Economics. 2nd ed. University Press.
Madison: University of Wisconsin Press. Skocpol, Theda. 1995. "The Role of Theory in Comparative Politics."
McDonald, Terence J., ed. 1996a. The Historic Turn in the Human World Politics 48 (October): 37-46.
Sciences. Ann Arbor: University of Michigan Press. Skocpol, Theda, and Margaret Somers. 1980. "The Uses of Compar-
McDonald, Terence J. 1996b. "What We Talk about When We Talk ative History in Macrosociological Inquiry."Comparative Studies
about History: The Conversations of History and Sociology." In in Society and History 22 (April): 174-97.
The Historic Turnin the Human Sciences, ed. Terence J. McDonald. Smelser, Neil J. 1967. "Sociological History: The Industrial Revo-
Ann Arbor: University of Michigan Press, 91-118. lution and the British Working-Class Family." Journal of Social
McNeill, William H. 1998. "History and the Scientific Worldview." History 1 (Fall): 17-35.
History and Theory 37 (February): 1-13. Somers, Margaret R. 1996. "Where Is Sociology after the His-
Milner, Helen V. 1998. "Rationalizing Politics: The Emerging toric Turn? Knowledge Cultures, Narrativity, and Historical Epis-
Synthesis of International, American, and Comparative Politics." temologies." In The Historic Turn in the Human Sciences, ed.
International Organization 52 (Autumn): 759-86. Terence J. McDonald. Ann Arbor: University of Michigan Press,
Mink, Louis 0. [1966] 1987. "The Autonomy of Historical Under- 53-89.
standing." In Louis O. Mink: Historical Understanding, eds. by Somers, Margaret R. 1998. " 'We're No Angels': Realism, Rational
Brian Fay, Eugene O. Golob, and Richard T. Vann. Ithaca, NY: Choice, and Relationality in Social Science." American Journal of
Cornell University Press, 61-88. Sociology 104 (November): 722-84.
Mink, Louis 0. [1978] 1987. "Narrative Form as a Cognitive Instru- Somers, Margaret R., et al. 1998. "Symposium on Historical Sociol-
ment." In Louis O. Mink: Historical Understanding, eds. Brian ogy and Rational Choice Theory."American Journal of Sociology
Fay, Eugene O. Golob, and Richard T Vann. Ithaca, NY: Cornell 104 (November): 722-871.
University Press, 182-203. Spruyt, Hendrik. 1994. The Sovereign State and Its Competitors.
Mohr, Lawrence B. 1996. The Causes of Human Behavior: Implica- Princeton, NJ: Princeton University Press.
tions for Theory and Method in the Social Sciences. Ann Arbor: Stone, Lawrence. 1987. "The Revival of Narrative: Reflections on a
University of Michigan Press. New Old History." In The Past and the Present Revisited. London:
Moore, Barrington. 1966. Social Origins of Dictatorship and Democ- Routledge & Kegan Paul, 74-96.
racy. Boston: Beacon Press. Tetlock, Philip E., and Aaron Belkin, eds. 1996. Counterfactual
Nord, Philip. 1995. The Republican Moment: Strugglesfor Democracy Thought Experiments in World Politics. Princeton, NJ: Princeton
in Nineteenth-Century France. Cambridge, MA: Harvard Univer- University Press.
sity Press. Thelen, Kathleen. 1999. "Historical Institutionalism in Comparative
Norman, Andrew P 1991. "Telling It Like It Was: Historical Narra- Perspective." Annual Review of Political Science 2 (1999): 369-
tives on Their Own Terms."History and Theory 30 (May): 119-35. 404.
O'Neill, Barry.2001. "Risk Aversion in International Relations The- Tilly, Charles, ed. 1975. The Formation of National States in Western
ory." International Studies Quarterly 45 (December): 617-40. Europe. Princeton, NJ: Princeton University Press.
Pierson, Paul. 2000. "Increasing Returns, Path Dependence, and the Tilly, Charles. 1990. "How (and What) Are Historians Doing?"
Study of Politics." American Political Science Review 94 (June): American Behavioral Scientist 33 (July/August): 685-711.
251-67. Tilly, Charles. 1992. Coercion, Capital, and European States, AD
Powell, Robert. 1999. "The Modeling Enterprise and Security Stud- 990-1992. Rev paperback ed. Cambridge, MA: Blackwell.
ies." International Security 24 (Fall): 97-106. Tilly, Charles. 1997. "Means and Ends of Comparison in Macrosoci-
Przeworski, Adam. 1991. Democracy and the Market. New York: ology." Comparative Social Research 16 (1997): 43-53.
Cambridge University Press. Tilly, Charles. 1998. "Armed Force, Regimes, Contention, and De-
Przeworski, Adam. 1995. "The Role of Theory in Comparative Poli- mocratization in Europe Since 1650." No. 19. Irvine: Center for
tics." World Politics 48 (October): 16-21. the Study of Democracy Research Monographs, University of
Ragin, Charles C., Dirk Berg-Schlosser, and Gisele de Meur. 1996. California. <www.democ.uci.edu/democ> (September 2000).
"Political Methodology: Qualitative Methods." In A New Hand- Topolsky, Jerzy. 1998. "The Role of Logic and Aesthetics in Con-
book of Political Science, eds. Robert Goodin and Hans-Dieter structing Narrative Wholes in Historiography."History and Theory
Klingemann. Oxford: Oxford University Press, 749-68. 38 (May): 198-210.
Ricoeur, Paul. [1971] 1979. "The Model of the Text: Meaningful Ac- Van Evera, Stephen. 1994. "Hypotheses on Nationalism and War."
tion Considered as a Text."In InterpretiveSocial Science:A Reader, International Security 18 (Spring): 5-39.
eds. Paul Rabinow and William M. Sullivan. Berkeley: University Waever, Ole. 1998. "The Sociology of a Not So International
of California Press, 73-101. Discipline: American and European Developments in Interna-
Robertson, David Brian. 1993. "The Return to History and the New tional Relations." International Organization 52 (Autumn): 687-
Institutionalism in American Political Science." Social Science His- 727.
tory 17 (Spring): 1-36. Weber, Max. [1906] 1978. "The Logic of Historical Explanation." In
Rogowski, Ronald. 1995. "The Role of Theory and Anomaly in Max Weber:Selections in Translation, ed. W. G. Runciman. New
Social-Scientific Inference." American Political Science Review 89 York: Cambridge University Press, 111-31.
(June): 467-70. Weber, Max. [1919] 1946. "Science as a Vocation." In From

492

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions
American Political Science Review Vol. 96, No. 3

Max Weber: Essays in Sociology. transl. and eds. Hans H. Weingast, Barry R. 1998. "Political Stability and Civil War: Insti-
Gerth and C. Wright Mills. New York: Oxford University Press, tutions, Commitment, and American Democracy." In Analytic
129-56. Narratives, eds. Robert H. Bates, et al. Princeton, NJ: Princeton
Weingast, Barry R. 1997. "The Political Foundations of Democracy University Press, 148-93.
and the Rule of Law."American Political Science Review 91 (June): Wendt, Alexander E. 1999. Social Theory of International Politics.
245-63. New York: Cambridge University Press.

493

This content downloaded from 200.52.255.53 on Mon, 16 Feb 2015 17:20:51 PM


All use subject to JSTOR Terms and Conditions