Wesleyan University

Comparison and Beyond

Author(s): Jrgen Kocka
Reviewed work(s):
Source: History and Theory, Vol. 42, No. 1 (Feb., 2003), pp. 39-44
Published by: Blackwell Publishing for Wesleyan University
Stable URL: http://www.jstor.org/stable/3590801 .
Accessed: 19/12/2011 04:04

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.

Blackwell Publishing and Wesleyan University are collaborating with JSTOR to digitize, preserve and extend
access to History and Theory.

http://www.jstor.org
HistoryandTheory42 (February
2003),39-44 ? WesleyanUniversity2003ISSN:0018-2656

COMPARISONAND BEYOND

JURGENKOCKA

ABSTRACT

The meritsof the comparative approachto historyareundeniable. Comparison helpsto

identifyquestions,andto clarifyprofilesof single cases. It is indispensablefor causal
explanationsandtheircriticism.Comparison helps to makethe "climate"of historical
researchless provincial.Still, comparativehistoriansremainin a minority.Manycher-
ishedprinciplesof thehistoricaldiscipline-proximity to thesources,context,andconti-
nuity-are sometimesin tensionwith the comparativeapproach.More recently,new
transnationalapproaches--entangled histories,histoirecroisie--challenge comparative
and histoire croise'ecan be
historiansin a new and interestingway. But histoire comparede
compatibleandneedeachother.

This comment1first underlinesthe great importanceof comparisonfor gaining

historical insight by discussing majorfunctionscomparativeapproachesfulfill in
historical studies. It then tries to answerthe question why, nevertheless,compar-
ison has usually been a minority phenomenon among historians. Third, it will
draw attentionto a relatively new challenge thatcomparativehistory faces today,
and that may well lead to putting comparisoninto a new context. A few conclu-
sions are offered at the end. For the purposes of this comment I want to stress
that comparing in history means to discuss two or more historical phenomena
systematicallywith respect to their similaritiesand differences in orderto reach
certain intellectualaims.2
Which aims? What are, methodologically speaking, the purposes and func-
tions of comparisonin historical researchand presentation?I propose to distin-
guish among heuristic,descriptive,analytical,and paradigmaticpurposes.

1. Presentedto the panel "Problemsof ComparativeExplanation"at the FourthEuropeanSocial

Science History Conferencein The Hague, March2, 2002.
2. For other aspects and surveys of the literaturecf. Jiirgen Kocka, "The Uses of Comparative
History," in Societies Made up of History: Essays in Historiography, Intellectual History,
Professionalisation, Historical Social Theory,& Proto-Industrialisation,ed. RagnarBjork and Karl
Molin (Edsbruk,Sweden:AkademitryckAB, 1996), 197-209; Geschichteund Vergleich:Ansdtzeund
Ergebnisse internationalvergleichenderGeschichtsschreibung,ed. Heinz-GerhardHaupt and Jtirgen
Kocka (Frankfurtand New York:Campus, 1996); JtirgenKocka, "Storiacomparata,"in Enciclopedia
delle scienze sociali (Rome: Istituto della Enciclopedia italiana, 1998), vol. 8, S. 389-396; Heinz-
GerhardHaupt, "ComparativeHistory,"in InternationalEncyclopediaof the Social and Behavioral
Sciences (Amsterdamand New York:Elsevier, 2001), vol. 4, 2397-2403; HartmutKaelble, Der his-
torische Vergleich:Eine Einfiihrungzum 19. und 20. Jahrhundert(Frankfurtand New York:Campus
Verlag, 1999).
40 JURGENKOCKA

Heuristically,the comparativeapproachallows one to identify questions and

problemsthatone might miss, neglect, or just not inventotherwise.For this Marc
Bloch gave an example from his own research.As an agrarianhistorianhe had
studiedthe English enclosures of the sixteenth to the nineteenthcenturies.From
that he developed the assumptionthat something analogous should have taken
place in France,albeityet undiscoveredby local research.Startingwith this ques-
tion Bloch revealed for fifteenth-, sixteenth- and seventeenth-centuryProvence
correspondingthoughnot identicalchanges in the structureof landownershipand
in this way contributedto a far-reachingrevision of the historyof the region.3This
was an act of intellectual transfer, based on assumptions about similarities
between Englandand France,a productiveinsight madepossible by comparison.
Descriptively,historicalcomparisonhelps to clarify the profilesof single cases,
frequentlyjust of one single case, by contrastingthem (or it) with others. Many
examples come to mind: for example, all characterizationsof historicalphenom-
ena as "first"or "belated";or claims for particularity,like the notion of a "German
Sonderweg"or "AmericanExceptionalism";and many other examples--from a
typology of regional industrializationprocesses in WesternEurope to the idea of
a distinctive path of Western modernizationcompared with other parts of the
world. Comparisonin this sense is ubiquitous,and even plays a role in historical
works that one would not classify as comparativein the full sense of the word. It
should be addedthat comparisondoes not only help to supportnotions of partic-
ularity,but is also indispensablefor challengingand modifying such notions.4
Analytically, the comparative approach is indispensable for asking and
answering causal questions. This point has been made frequently, in method-
ological detail and with many examples.5Nowadays, global history is a field that
offers itself for comparativeapproacheswith causal aims, be it with respect to
the rise of science in differentcivilizations over the centuries,with respectto the
different paths of economic change and growth, or with respect to other prob-
lems.6Max Weberpioneeredthis type of ambitiouscomparison.Sewell and oth-
ers have stressed that comparison can play the role of an indirect experiment
facilitating the "testing of hypotheses."While one may be skeptical about this
claim (since the ceteris paribus clause can rarely be fulfilled in historical stud-
ies), it is beyond doubt that comparisonis indispensablefor historianswho like
to ask causal questionsand providecausal answers.Along the same line it should
be stressed that the necessary criticism of given explanations, including the

3. Marc Bloch, "Pourune histoire compar6edes soci6t6s europ6ennes"(1928), in Milanges his-

toriques,ed. Marc Bloch (Paris:Ecole des HautesEtudes en Sciences Sociales, 1983), vol. 1, 16-40.
4. The debate and researchabout the questions of a "GermanSonderweg"can serve as an exam-
ple. Cf. JiirgenKocka, "AsymmetricalHistoricalComparison:The Case of the GermanSonderweg,"
History and Theory38 (1999), 40-51.
5. Cf. William H. Sewell, "MarcBloch andthe Logic of ComparativeHistory,"History and Theory
6 (1967), 208-218; A. A. van den Braembussche,"HistoricalExplanationand ComparativeMethod:
Towardsa Theory of the History of Society,"History and Theory28 (1989), 2-24.
6. For a recent overview of these debates, see Gale Stokes, "The Fates of Human Societies: A
Review of Recent Macrohistories,"AmericanHistorical Review 106 (2001), 508-525.
 COMPARISONAND BEYOND 41

rebuttalof "pseudo-explanations"both of the local and of the generalizing type,

needs comparisonas well.7
Finally,just a word on the paradigmaticfunctionof comparison.In this respect
comparisonhelps to distance oneself a bit from the case one knows best, from
"one's own history."Verfremdung is the Germanword. In the light of observable
alternatives one's own development loses the self-evidence it may have had
before. One discovers the case with which one is most familiaras just one pos-
sibility among others. Frequentlyhistorians are relatively concentratedon the
history of their country or region. Because of this, comparisoncan have a de-
provincializing, a liberating,an eye-opening effect, with consequences for the
atmosphereand style of the profession. This is a contributionof comparisonthat
should not be underestimated,even today.
These points should suffice in orderto remindus of the many advantagescom-
parisonhas. Why is it neverthelessthe case that comparativehistory has had a
minoritystatusfor a very long time. and continuesto have thatstatuseven today?
There are many practicalreasons as well as reasons related to the culturaland
nationalfunctionsthe disciplinehas had over the centuries.Afterall, as a mass dis-
cipline history emerged in close interconnectionwith the rise of the nation-state,
at least in the West. I am not dealing with these impedimentsof comparativestud-
ies now. Rather,I want to discuss three seriousmethodologicalreasonsthat make
comparisondifficult,threecharacteristicsthatconstitutea certaintension between
the comparativeapproachand the classical traditionof historyas a discipline.
1. The more cases a comparative study includes, the more dependent it
becomes on secondaryliterature,and the more difficult it becomes to get nearto
the sourcesand readthem in theiroriginallanguage.But proximityto the sources
and command of their language has developed as a major principle of modern
historicalscholarshipas it has emerged since the late eighteenthcentury,for very
good reasons.
2. The comparativeapproachpresupposesthat the units of comparisoncan be
separatedfrom each other.It is neitherthe continuitybetweentwo phenomenanor
the mutualinfluencesbetween them thatconstitutethem as cases for comparison.
Ratherthey are seen as independentcases thatarebroughttogetheranalyticallyby
asking for similaritiesand differencesbetween them. In otherwords,the compar-
ison breakscontinuities,cuts entanglements,and interruptsthe flow of narration.
But the reconstructionof continuities,the emphasison interdependenceas well as
narrativeforms of presentation,are classical elements of historyas a discipline.
3. One cannot comparetotalities, in the sense of fully developed individuali-
ties. Rather,one compares in certainrespects. One has to decide with respect to
which viewpoints, questions, or Erkenntnisinteressenone wants to comparetwo
or more cases. The more cases one includes, the more importantbecomes this
selective decision about the viewpoints, questions, and problemswith respect to
which one wants to compare. In other words, comparison implies selection,
abstraction,and de-contextualizationto some degree. One realizes this right
7. Again Bloch has given examples. Cf. note 3 above.
42 JURGEN KOCKA

away if one thinks of multi-case comparisons.Whoever tries to compare, let us

say, twenty regional industrializationcases or demographic patterns in forty
Frenchcities in the middle of the nineteenthcentury,has to isolate the objects of
comparison,the relevant"variables"from theircontext to a large degree. But the
emphasis on context, on embeddedness,on Zusammenhangis dearand centralto
history as a discipline. Again there is a tension between the comparative
approachand some of the much cherishedand worthwhile principles of histori-
cal studies, at least in the West.
These are the major methodological reasons why comparative approaches
were traditionallynot in the centerbut at the peripheryof history as a discipline.
This also explains why comparativeapproachesbecame much more popularand
much more central once history became more social-science oriented in the
1970s and 1980s.8
Most recently the wind that blows into the face of comparativehistorianshas
become even stronger.In additionto the moretraditionaland conventionalobjec-
tions historiansmay have against too much and too rigorous comparison,there
are new ones, new reservationsagainst clear-cut comparativeapproaches,this
time on the side of the youngest, in an interesting way. After the end of the
East-West conflict around 1990 both the acceleratedprocesses of international-
ization and the renewed debates on globalization startedto change the way in
which we define historical questions and explore historical problems.As a con-
sequence, thereis a new stress on "entangledhistories,"on "histoirecroisle," on
"Verflechtungsgeschichte" or "Beziehungsgeschichte"which I find in some ten-
sion with basic principlesof comparativehistory.9
There is, fortunately,much interest now in transnationalapproachesto histo-
ry. The different currents of global or world history are cases in point.
Comparativeapproaches, internationaland interculturalcomparisons, are just
one way for realizing this rising transnationalcommitment.There are otherways
as well, for example, studies and interpretationsusing postcolonial theories.'0
According to this view one is much less interestedin similaritiesand differences
between, let us say, Europe and the Arab world, but ratherin the processes of
mutual influencing, in reciprocal or asymmetric perceptions, in entangled
processes of constitutingone another.In a way, the history of both sides is taken
as one instead of being considered as two units for comparison.One speaks of
entanglements;is interestedin travelling ideas, migrating people, and transna-
tional commerce;mutuallyholds images of "theother";and one talks aboutmen-

8. On the basis of a comprehensivesurvey HartmutKaelble identifies the 1980s as, quantitatively

seen, the breakthrough phase of comparative social history in Europe. Hartmut Kaelble,
"VergleichendeSozialgeschichtedes 19. und 20. Jahrhunderts:ForschungeneuropiiischerHistoriker,"
in Hauptand Kocka, eds. Geschichteund Vergleich,97.
9. Cf. Johannes Paulmann, "InternationalerVergleich und interkultureller Transfer: Zwei
Forschungsansitzezur europiiischenGeschichte des 18. bis 20. Jahrhunderts,"Historische Zeitschrift
267 (1998), 649-685; Globalization in WorldHistory, ed. Anthony G. Hopkins (London: Pimlico,
2002).
10. Cf. Robert J. C. Young, Postcolonialism: An Historical Introduction(Oxford and Malden,
Mass.: Blackwell Publishers,2001).
 COMPARISONAND BEYOND 43

tal mapping,including aspects of power, subordination,and dominance.Cultural

dimensions are usually centralto such an approach.Europe and non-European
partsof the world, the West and non-Westerncivilizations are the most preferred
topics for such approaches."Entangledhistories"has become a key phrase, for
instanceadvocatedby the sociologist-ethnologistShalini Randeria.Anothervari-
ation of this type of approachhas been called "histoirecroisele,"such as a his-
toire croisee between Germanyand France in the nineteenth century as propa-
gated by Michael Werner,BdnddicteZimmermann,and SandrineKott."
These are highly interesting and promising developments. But this type of
transnationalapproach goes beyond comparison. Or does it fall back behind
comparison?At any rate, from an entangled-historypoint of view, comparison
appearsa bit too mechanistic,a bit too analyticalin that it separatesreality into
differentpieces in orderto analyze,that is, to comparethe pieces as units of com-
parison,whereas it would be necessary to see them as one, as one web of entan-
glements, one "Zusammenhang"of Verflechtungenand relations. In fact,
EspagneandWerner,who came from the study of literatureandculture,and who
have pioneered this approach with respect to developments in Germany and
France, very early and effectively criticized the comparativeapproach.12In the
meantimethey have many sympathizers,particularlyamong culturalhistorians.
Certainly,neitherthe built-instrengthsof the historicalmethod nor the recent
interestin transculturaland transnationalentanglementsshould be taken as jus-
tifications for withdrawingfrom comparativehistory. Proximity to the sources
and control of languages are importantimperativesof historical research. But
they must not be taken as excuses for professional over-specialization, nor
should they prevent the broad perspectives and comprehensive interpretations
historiansmust be able to offer in this global age. The stress on continuity and
context are indispensable for and characteristicof historians'work. But on the
other hand continuity is just one guiding principle of historical reconstruction
among others, and while historianshave to take context seriously,their intellec-
tual operationsare always selective, viewpoint-relatedand, in this sense, analyt-
ical; they never reconstruct totalities in full. Consequently, comparative
approachesonly emphasize and make particularlymanifest what is implicit in
any kind of historical work: a strong selective and constructive component.
Comparativehistory compels its practitionersto explicitly reflect upon these

11. Sanjay Subrahmanyam,"Connected Histories: Notes towards a Reconfigurationof Early

ModernEurasia,"ModernAsian Studies 31 (1997), 735-762; Shalini Randeria,"GeteilteGeschichte
und verwobene Moderne," in Zukunftsentwiirfe:Ideen fiir eine Kultur der Verdinderung, ed. Jmrn
ROisenet al. (Frankfurt:Campus, 1999), 87-96; Le travail et la nation: Histoire croisde de la France
et de l'Allemagne,ed. BdnddicteZimmermannet al. (Paris:Maison des sciences de l'homme, 1999);
Jtirgen Osterhammel, "TransnationaleGesellschaftsgeschichte: Erweiterung oder Alternative?,"
Geschichte und Gesellschaft 27 (2001), 464-479; Sebastian Conrad, "Doppelte Marginalisierung:
Plidoyer flir eine transnationale Perspektive auf die deutsche Geschichte," Geschichte und
Gesellschaft 28 (2002), 145-169; Emma Rothschild, "Globalizationand the Return of History,"
Foreign Policy (Summer 1999), 106-116.
12. Transferts:les relations interculturellesdans l'espacefranco-allemand (XVIIIeet XIXesiecle),
ed. Michel Espagne and Michael Werner(Paris:Editions Recherche sur les civilisations, 1988).
44 JURGENKOCKA

epistemological premises of their work, while these premises are frequentlyjust

implicit in other approaches.The new interest in transnationalentanglementsis
most welcome and promising. However, it must not lead away from but should
incorporaterigorous comparison, which remains particularlyindispensable for
historical studies with global reach if they do not want to become merely specu-
lative or feuilletonistic.
But comparativehistoriansshould react to the old caveats and the new chal-
lenges in a productiveway. Usually they will limit the numberof cases they com-
pare in orderto take contexts sufficiently into consideration.More importantly,
they can and should incorporateelements of the "entangledhistories"approach
into the comparativedesign of their research. Certainly,the act of comparison
presupposesthe analyticalseparationof the cases to be compared.But that does
not mean ignoring or neglecting the interrelationsbetween these cases (if and to
the extent thatthey existed). Rather,such interrelationsshould become partof the
comparativeframeworkby analyzingthem as factorsthat have led to similarities
or differences,convergence or divergence between the cases one compares.
This has been done before. TakeAlexander Gerschenkron'sclassical compar-
ison of Europeanindustrializationsas an example. He, in a way, took European
industrializationas a whole. At the same time, he comparedits partsor segments,
that is, industrializationprocesses within different countries. He gave much
weight to the interrelationsbetween them, for example, to the export and import
of capital,labor,methods, and ideas as well as to processes of perception,imita-
tion, transfer,and rejection between the industrializersin different European
countries.And he showed that some of these interrelationscontributedto more
similaritywhile others led to importantdifferences between nationalpatternsof
industrializationin Europe.13PhilippTher investigates the origins, programmes,
organizations,and public support of opera houses in nineteenth-centuryEast
CentralEuropeand Germany.While analyzing their differences and similarities,
he also shows how they perceived and influencedone another--all of them ele-
ments of a comprehensivecultureof CentralEurope.14
Many other examples could be given in orderto show that it is both possible
and desirable to treat historical phenomena as units of comparison and, at the
same time, as componentsof a largerwhole. Comparativehistory andthe "entan-
gled histories"approachare differentmodes of historical reconstruction.There
is a tension between them, but they are not incompatible.One can try to analyze
in comparativeterms and tell a story, nevertheless.It is not necessary to choose
between histoire compare'eand histoire croisde. The aim is to combine them.

Friedrich-Meinecke-Institut
Berlin

13. A. Gerschenkron,Economic Backwardness in Historical Perspective (Cambridge, Mass.:

Belknap Press of HarvardUniversity Press, 1962), 5-51, 353-364.
14. Philipp Ther, "Geschichteund Nation im MusiktheaterDeutschlandsund Ostmitteleuropas,"
Zeitschriftfiir Geschichtswissenschaft50 (2002), 119-140.