Luebbert-Social Foundations of Political Order in Interwar Europe

Trustees of Princeton University
Social Foundations of Political Order in Interwar Europe

Author(s): Gregory M. Luebbert
Source: World Politics, Vol. 39, No. 4 (Jul., 1987), pp. 449-478
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/2010288 .
Accessed: 21/07/2011 14:04
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=cup. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Cambridge University Press and Trustees of Princeton University are collaborating with JSTOR to digitize,
preserve and extend access to World Politics.
http://www.jstor.org
SOCIAL FOUNDATIONS OF
POLITICAL ORDER IN INTERWAR
EUROPE
By GREGORY M. LUEBBERT*
TOUR typesof regimesof historicimportanceappeared in Europe be-

tween the two world wars: pluralistdemocracy,social or corporatist
democracy,traditionaldictatorship,and fascism.The vast body of liter-
aturethathas grown up around themhas rarelycast thesepoliticalorders
as historicalalternativesto each other,however.When it has done so, it
has normallycast pluralistdemocracyas the alternativeto fascism.Most
commonly,this has taken the formof contrastsbetween Germany and
Britain,and has been accompanied by the question,why was Germany
not like Britain?' Yet, pluralistdemocracysuch as appeared in Britain
was actuallythe least relevantalternativebetween the wars, forthe pos-
sibilityof stabilizingit where it did not alreadyexisthad been foreclosed
by World War I. Where liberalpartieshad failedto establishresponsible
parliamentaryinstitutionsbeforethe war, it would prove impossibleto
stabilize a pluralist democracy afterward. Henceforth, stabilization
would require corporatism in eitheritsfascistor social democraticvar-
iant ratherthan pluralism.
The appearance of fascism,however,was not predeterminedby the ir-
relevance of pluralistdemocracy.Germany did not become fascistbe-
cause it could not become a pluralistdemocracysuch as Britain.It became
fascistbecause it could become neithera corporatistdemocracysuch as
Norway nor a traditionaldictatorshipsuch as appeared in much of east-
ern Europe. It was stillpossible to establisha stable social or corporatist
* I would like to thankVinod
Aggarwal,Gabriel Almond,Paul Buchanan,David Collier,
Giuseppe Di Palma, JosephFizman, Ernst Haas, Andrew Janos,Daniel Verdier,and John
Zysman fortheirthoughtfulcommentson the longermanuscriptfromwhich thisarticleis
taken.I am especiallyindebtedto Daniel Verdierforhis researchassistance.
Among the most influentialof theseworksare thoseby Ralf Dahrendorfand Karl Die-
trichBracher,both of whom postulatea liberalor Anglo-Saxontheoryof the stateand con-
trastit witha German Idealisttradition.See Dahrendorf,Societyand Democracyin Germany
(New York: Norton, i967); Bracher,Die Aufidsung der Weimarer
Republik[The Dissolution
of the Weimar Republic] (Stuttgart:Ring Verlag, I Two influentialbut more sociologi-
callygroundedanalysesthatmake use of thecontrastbetweenBritainand Germanyare Bar-
ringtonMoore,Social OriginsofDictatorshipand Democracy(Boston: Beacon Press,i966), and
AlexanderGerschenkron, Breadand Democracyin Germany(Berkeley:Universityof Califor-
nia Press, 1943).
450 WORLD POLITICS
democracy.This could be done if socialistscould forma coalition with

the countryside.Even when socialistscould not do this,it was stillpossi-
ble to avoid fascism if interwarliberals could make such a coalition.
Where theycould do so, the resultwould be a traditionaldictatorship.
Fascism, in otherwords, could only appear if pluralistdemocracy,social
democracy,and traditionaldictatorshipcould not. An understandingof
fascism,then,requiresan understandingof whythesealternativeregimes
were not possible in some countries.
To understandwhy theywere not,we muststudythe determinantsof
both stateand labor-marketinstitutions. The formthe interwarstateul-
timatelyassumed in the crises of the I920S and I930S cannot be under-
stood apart fromthe mannerin which workersbecame organized in the
labor market.For both stateand labor marketinstitutionstook shape si-
multaneouslyas two halves of the same responseto working-classmobi-
lization.
Trade unions could be organized along corporatistor pluralistlines,or
they could simply be repressedby the state. Corporatismbetween the
wars denoted a systemin which economic interestsin a privateeconomy
were becoming functionallyorganized in groups that were centralized,
comprehensive,and authoritative.In its democraticvariant social de-
mocracy membershipin theseorganizationswas voluntary;the organ-
izations mediated the relationshipamong workers,capitalists,sectorsof
the economy, and the state. In social democracies, corporatistlabor
unions have served to offsetthe otherwiseinherentadvantages of em-
ployersin collectivebargainingand in the polity,and have played a crit-
ical role in the making of social policy.Corporatismhas not been mainly
a systemof restraintand controlof the working class, but of inclusion.
Social democracies appeared between the wars in Norway, Sweden,
Czechoslovakia, and to a lesserdegree in Denmark.2
In its authoritarianvariant fascism corporatistorganizationswere
mainly mechanisms throughwhich the state controlledand mobilized
workers and capitalists. Here corporatismprovided the institutional
means to subordinatethe workingclass; inclusionwas mainlysymbolic.
Corporatisttrade unions served as the transmissionbeltsof the material
benefitsand coercionby which thecontrollingpartyattemptedto buy the
support or force the acquiescence of labor. Between the wars, such re-
2 The interwarCzechoslovakian experienceand its corporatistaspects can be found in

Harry Klepetar,Seit i9i8 ... Eine Geschichte
der Tschechoslowakischen
Republik[Since i9i8
... A Historyof the Czechoslovak Republik](Moravska Ostrava: Verlag JuliusKittlsNach-
folger,1937), esp. 283-375-
POLITICAL ORDER IN INTERWAR EUROPE 451
gimesappearedin Nazi Germany, FascistItalyand,toa lesserdegree,in

Franco'sSpain.
In pluralistlabor markets,tradeunionswere legallyfree-within
some basic rules-to organize,bargaincollectively, and agitatepoliti-
cally.In contrastto theircounterparts in socialdemocracies, unionsre-
mainedcomparatively weaklyorganizedand lesseffective in bargaining
withemployers and thestate.Theydid notexhibitthesamecomprehen-
sivenessorcentralization; leaderscouldnotspeakwithauthority fortheir
memberson a broadrangeofsocialand economicquestions.There de-
velopedno legal or culturalpresumption thatunionleadershad to be
consultedand theirpolicydemandsaccommodatedby the state.The
UnitedStates,Britain,France,Ireland,and Switzerlandare examplesof
pluralistdemocracies.
The fourthvarietyof politicalorderthatappearedbetweenthewars
was thetraditionaldictatorship. It combinedseverecurtailments ofpolit-
ical competition withrepressionin the labormarket.Parliamentwas
closedor existedonlyas a rumpafterimportant partieshad been out-
lawed. Trade unionswere eithergreatlyrestricted or outlawedalto-
gether,but theywerenot,as in fascistregimes,replacedby state-spon-
soredorganizations of control.Such regimesofferedno solutionto the
problemofworking-class integration;rather, theyattempted toignoreit.
To thatextent,theywereinherently unstable.Such traditional dictator-
in
shipsappeared Finland,Austria,Hungary, and much of easternEu-
rope.
The fourregimesare,to be sure,ideal types.No societyacquiredin-
stitutionsthatcorresponded perfectlyto theidealizedmodels,and some
institutionswerea mixture, especiallyof thepluralistand socialdemo-
craticmodels.Belgium,theNetherlands, and Denmark,forinstance, fell
betweenthe pluralistand corporatist models,withthe firsttwo more
closelyapproximating the pluralistmodel,and Denmarkmoreclosely
approximating thesocialdemocratic model.The intermediate positionof
thesecountriesreflects and can be explainedby themixedcharacterof
theirdevelopmental experience, whichcombinedaspectsofboththeplu-
ralistand corporatistexperiences.
The fourregimesweretheinterwarbyproducts of fourroutesfrom
preindustrialpoliticsto thecrisesoftheI920S and I930s. Each routewas
delineatedby a historicallegacyand by the urban-rural coalitionthat
formedin responseto theinterwar legacywas pro-
crises.The historical
videdbythedominance, orlackthereof, ofliberalpartiesbeforetheFirst
WorldWar. Nowherebeforethewar wereliberalsin unambiguouspo-
452 WORLD POLITICS
liticalcontrol.Rather,theyexercisedmore or less influence,sometimesin

an alliance with, and sometimes in competitionwith, rural interests.
Whether theywere the senioror junior partner(or antagonist),however,
was of decisive importance.Where theywere dominant-as in Britain
(where the Liberal Party and the liberalized Conservative Party alter-
nated in power until the war), Switzerland (where the Radicals con-
trolledeverygovernmentfromi848 to i919), and France (where Repub-
licans were dominantafteri877) workersenteredpoliticsbeforeWorld
War I; theydid so graduallyand in groups,ratherthan suddenlyand as
a cohesive class. As a result,these societies,all of which acquired demo-
craticpolities,also acquired pluralistlabor markets.
The situationwas dramaticallydifferent in countriesin which liberals
were not dominant.Here, the war was a turningpoint in the transition
to mass politics.Its traumaand the example of the BolshevikRevolution
servedto radicalize workersand polarize politicsin all societies.Workers
and peasants were now politicallyactive everywhere.The explosion of
political participationwas vastlymore difficultto accommodate where
workershad not been broughtinto politicsgraduallyunder liberal tute-
lage. Where theyhad not been, theyhad come to be organized in politics
and in the labor marketin a more explicitlyclass-basedmanner.Yet, the
politicaland economic citizenshipof thismore cohesivelyorganized class
remainedincomplete.In effect,modernpoliticalordershad yetto be cre-
ated. Ultimately,corporatismin eitheritssocial democraticor fascistvar-
iant,ratherthan pluralism,would be requiredto stabilizeclass politicsin
thesesocieties.
The formof the politicalorder was determinedby the coalitionof ur-
ban and ruralintereststhatemergedin responseto thecrises.In themak-
ing of these coalitions,the allegiance of the familypeasantryproved de-
cisive: when it sided with an urban socialist party,the outcome was a
social democraticregime; when it sided with the partyor partiesof tra-
ditionalliberals,the outcome was invariablya traditionaldictatorship.In
thisessay,I will attemptto identifythe conditionsthatproduced each of
these coalitions.To set the stage,we must firstconsiderthe importance
of liberal dominance in the creationof pluralistdemocracies,and why it
was thatthe politicaleconomiesestablishedafterWorld War I could not
become pluralistdemocracies.
PLURALIST DEMOCRACIES
The model of the pluralistpath, as exemplifiedby the Britishtransi-

tion, has long fixed attentionon the essentialfeaturethat distinguished
earlydemocratizers.3 In thesesocieties,thebourgeoisiegaineda domi-
nantposition,throughthevehicleofliberalparties,beforeworkersmo-
bilizedautonomously. In societiesthatwerebecomingpluralist, working
classunionsand partiesgainedaccessto thestatethroughcollaboration
with the liberalcommunity. Liberalsgraduallyaccommodatedthem-
selvesto demandsforan enlargedsuffrage, thesecretballot,theelimi-
nationofunequalelectoraldistricts, thelegalizationoftradeunions,col-
lectivebargaining,therighttostrike, andso forth.In largemeasure,these
long-term reforms came aboutbecausetheyservedtheshort-term elec-
toralneedsof liberalparties.For example,becauseFrenchRepublicans
needed the supportof workersin theirstruggleagainstConservatives,
theymet the threemostimportant demandsof Frenchworkermove-
mentsafteri877: theestablishment ofuniversal,freesecularpublicedu-
cation;thepardoningoftheCommunards;and thelegalizationoftrade
unions,collectivebargaining, and therightto strike.4
Becauseworkersin thesesocieties gainedtherighttovoteearlyon and
because parliamentary institutionshad substantiallegitimacyamong
workers, liberalpartiesneverfoundthemselves confronted bythework-
ingclassas a whole.Instead,theywereconfronted byscatteredgroupsof
graduallymobilizedworkers, usuallyorganizedon a localbasis.In con-
trolof thestate,theliberalscould definethetermson whichworking-
classpartiesweregivenaccesstopolitics;theycontrolled howvoteswere
counted,constituencies defined,partiesregistered, campaignsfinanced,
and legislators
paid.The liberalpartiescouldextendparticipation with-
out actuallychangingthebalanceofclasspowerverymuch.Indeed,the
Lib-Lab alliancesthatwereso characteristic ofthispatternofmobiliza-
tionactuallyservedto reinforce thebalanceofclasspower.To theextent
thattheLib-Lab strategy actuallybenefitedworkers(ormerelyappeared
to benefitthem),theliberalideologywas made moreplausible,and the
launchingof an autonomousworking-class movementwas made more
difficult.
Liberalism withitsemphasison classharmony, com-
individualism,
petition,achievement, and merit-was an ideologythatlegitimated the
pluralistorder.Itsextendedpre-eminence meantthatit,ratherthanthe
3 Early democratizersare definedas societiesthatacquired responsibleparliamentaryin-

stitutions,manhood suffrage,reasonablyequitable electorallaws, and tradeunion rightsbe-
foreWorld War I. The countriesthatmet all fourof the criteriawere Britain,Ireland (as a
colony),France, Switzerland,and the United States.All othersfailedon at least one count.
The classicalstatementof the pluralistpath is foundin T. H. Marshall,Class,Citizenshipand
Social Development(Garden City,NY: Doubleday, i965), chap. 4.
4 Val Lorwin, The FrenchLabor Movement (Cambridge: Harvard UniversityPress, 1954),
i8.
454 WORLD POLITICS
landedinterestsor working-classmovements, definedtheparameters of
legitimatepoliticalinstitutions
and debate.Working-class partiesthus
foundthemselves in thedifficult
positionofarguingagainstan ideology
and parliamentary institutions
thatappearedlegitimatetoa largepartof
theworkingclass.In I920, FernandLoriot,a leaderof thenew Com-
munistParty,lamentedtheholdoftheliberalideologyon Frenchwork-
ers'attitudes:
Whateverwe do, we cannotavoidthefactthatin France,over50 yearsof
bourgeoisdemocracyhas createdwithinthe massesamongstwhom we
work ... an entirelydistinctive
mentality.. .. People here are stillcon-
vincedthatwe havea patrimony oflibertiesto defend.The wholeofour
mentalformation is basedon a revolutionary[i.e.,Jacobin]
tradition,skill-
fullymanipulatedbyourbourgeoisdemocracy.5
Ironically,one indicatorof the successof liberalsin integratingthe
workingclassis thatin somerespects theyactuallyhad to do lessforthe
materialinterests
ofworkersthandid theleadersofauthoritarian states.
PeterFloraand JensAlberhaveshownthatsocialwelfaremeasureswere
most highly developed before I9I4 in those European societies that
lacked parliamentaryinstitutionsand politicalrightsforworkers.Flora
and Alber argue thatauthoritarianregimesturnedto thesemeasuresas a
substitutefor parliamentarismand the grantingof political rights to
workers.The greaterlegitimacybestowedby a parliamentaryorder and
liberal ideologyactuallymilitatedagainstthe earlyintroductionof social
welfareinitiativesthatwere of value to workers.6
Much thesame processof accommodationunderliberaldominanceoc-
curred in the labor market. In France, the Republicans legislated the
rightto formtrade unions, to bargain collectively,and to strikein i884.
This law and the manner of its enforcementillustratethe impact of lib-
eral dominance: unions were obliged to registertheirconstitutionsand
bylaws and the names of theirofficers.Registeredboursesdu travailwere
5Quoted in Roger Magraw,France,i8I5-I9I4: TheBourgeoisCentury(New York: Oxford

University Press), 373.
6 Flora and Alber use a less restrictivedefinitionof democraticregime(see fn.3 forthc
definitionemployedhere).Their "constitutional-dualist monarchies"are Sweden, Germany,
Denmark, and Austria.Their parliamentaryregimesare Britain,Switzerland,France, Bel-
gium, The Netherlands,Italy,and Norway. Under the more restricteddefinitionemployed
here,The Netherlandsand Belgium almostfitthe requirementsof earlydemocracies,falling
shortonlyon the extentof thesuffrage.Norway (electorallaws) and Italy(franchise,electoral
laws, trade union rights)clearlydo not. Despite the definitionaldifferences,the association
betweenliberalismand late social welfareinitiativesremains.Indeed, Flora and Alber's data
show that it is strengthenedif Italy is properlyclassifiedas undemocratic.See "Moderniza-
tion,Democratization,and the Developmentof WelfareStatesin WesternEurope," in Peter
Flora and Arnold Heidenheimer,eds., TheDevelopmentof WelfareStatesin Europeand Amer-
ica (New Brunswick,NJ: TransactionBooks, i98i), 37-80.
theneligibletoreceivestatesubsidiesand rent-free
officespace.Registra-
tionallowed municipalities to keep trackof radicals.The government
used subsidiesto discourageradicalismand, fora time,thecreationof
labororganizations thatstretchedacrossdifferent
trades.
The law of i884 meant that the state recognized the existence of
unions.It did notmeanthatemployerswererequiredtorecognizethem.
Indeed,fortwo generations manyFrenchemployerscontinued,often
brutally,to resistrecognition.
The reluctanceof manyemployersin
Franceand theUnitedStatesto recognizeunionsuntilcompelledto do
so in the I930S was itselfa manifestationof the weightof the liberal ide-
ology. The tellingeffectsof the i884 law, however,were to encourage
decentralized unions, to enhance the legitimacyof the Republic, and to
redefinesubstantiallythe battleas one between workersand employers
ratherthan between workersand the state.7The infrequencyof political
strikeswas indicativeof the successof French Republicans and theirlib-
eral counterpartselsewhere. While political strikesby French, British,
and American workers were exceptions,they were common in other
countries.In Belgium, forinstance,therewere fourgeneral strikesin the
yearsbeforeI9I4. Elsewhere,too, politicalstrikeswere a regularfeature
of labor movements,even where as in Spain, Germany,and Italy the
politicalrestraintson labor protestwere much more repressive.8
Because in liberal-dominatedsocietiesworkerswere allowed to organ-
ize more or less as theysoughtto, tradeunions became incorporatedinto
the labor market as individual interestgroups. They were oftenorgan-
ized locally and along craftlines, and developed traditionsof local or
plant bargaining.By the time national working-classpartiesand trade-
union movementsstartedto form,these locally organized unions had a
generationor more of leaders and memberswhose vestedinterestsin the
established organizations had to be overcome in order to create more
comprehensivestructures.9
Among the pluralistdemocracies,France appears somewhat excep-
tional because of the eventual emergence of a radical working-class
movementin the formof theCommunistParty.In the termsused above,
theexplanationseems straightforward enough,however:liberalshad less
7The earlyhistoryof theFrenchlabor movementis discussedin Lorwin (fn.4), chap. 2.
8 The absence of politicalstrikesby Frenchworkers-and the feebleness of syndicalismin
theFrenchlabor movementgenerally-has been carefullydocumentedin PeterStearns,Rev-
olutionarySyndicalism and FrenchLabor: A Cause WithoutRebels(New Brunswick,NJ: Rut-
gersUniversityPress, 1971), esp. chap. 2.
9 For a discussionof the obstaclespresentedby earlydecentralization forsubsequentuni-
ficationof French labor,see Lorwin (fn.4), chap. 2. For a paralleldiscussionof the Swiss ex-
perience,see Eduard Weckerle,The Trade Unionsin Switzerland(Bern: Swiss Federationof
Trade Unions, 1947), i6-33.
456 WORLD POLITICS
ofa needtosolicittheworkers'supportbecauseofthecombinedelectoral
weightofthemiddleclassand thepeasantry andbecausethecentrality of
left
conflict
clerical-anticlerical French workers withno to
alternative the
RadicalRepublicansuntiltheappearanceoftheunifiedSocialistPartyin
I905. The inconsistencyof thelate formation of theSocialistPartyand
theeventualappearanceoftheCommunistPartypointsto anotherdeci-
sivefeatureoftheFrenchexperience: Francewas distinguished notonly
bytheapparently lowerlevelofworkerintegration before war,but
the
byitsuniquelytraumatic experience in theFirstWorldWar. As a result,
theworking-class movementwas,by thestandardsof pluralistdemoc-
racies,comparativelyradicalized.
SOCIAL DEMOCRACY AND FASCISM
Societiesthatbecamesocialdemocracies, traditionaldictatorships,or
fascistdictatorshipsafterWorldWar I differed frompluralistdemocra-
cies in thecomparative weaknessof theirliberalpartiesbeforethewar.
There wereseveralreasonsforthisweakness.Mostobviously, theseso-
cietiesbeganto industrializeat a laterdate in all casesafteri85o, and
in manycasesnotuntilthebeginning ofthe20thcentury. The amountof
timein whichthesubtending class,thebourgeoisie,couldassertitselfbe-
forethewar,beforetheensuingdisruptions, and beforetheexampleof
theBolshevikRevolution, was briefer.IoAlso,becauselateindustrializa-
tionsometimes broughtabouta greaterroleforthestate,itoftenresulted
in a correspondingly greaterdependenceof thebourgeoisieon thestate
and a lessplausiblecase fortheliberalideology.
Late democratizersenteredtheFirstWorld War withmorehighly
mobilizedworkingclassesthanearlydemocratizers. The war had up-
rooted,alienated,disoriented and, therefore,made moresusceptible to
radicalappealsupontheirreturnhome(fromfactories orthefront),mil-
lionsofyoungmenwhohad previously beenbeyondthereachofradical
organizers. The impactofthewarwas hardlymilderin neutralist coun-
tries,traumatized as theywerebytheeconomicboom-and-bust cycleof
war and peace. In short,a significant betweensocietieslike
difference
theseandthosethatbecamepluralist democracieswas notsimplythatlib-
eralismwas weaker,but thatworkersbecameclass-conscious over a
muchbriefer periodoftime.One indication ofthisis providedbyTable
i, whichreports thepercentage ofthevotefortheleftin thelastelection
Io An importantstudyof the impact of the war on politicalmobilizationis provided in
Charles S. Maier, RecastingBourgeoisEurope (Princeton:PrincetonUniversityPress, 1975),
esp. 3-15-
TABLE I
LEFT VOTING AND LEVEL OF INDUSTRIALIZATION
Prewar Postwar Labor in

Left Votea Left Votea Non -primarySectorsb
France 16.8 23.0 56.9
Switzerland 20.1 23.5 80.0
Britain 6.4 29.7 91.5
Netherlands 22.9 25.0 77.0

Denmark 29.6 28.7 63.3
Belgium 30.3 36.6 68.0
Norway 26.3 31.6 57.9

Sweden 30.1 38.9c 56.3
Finland 43.1 38.0 37.0
Germany 34.8 45.5 69.3
Austriad 23.0 40.8 68.1
Italy 22.8 34.3 43.8
a Last electionbeforethe war and first electionafterthe war. For both elections,the left
vote is definedas the social democratic,labor,or socialistparty,and all partiesto theleftof it.
Data forbothelectionsare fromPeterFlora, ed.,State,Economyand Societyin Western Europe,
1815-1975.A Data Handbook. Vol. 1: The Growthof Mass Democracies(Chicago: St. James
Press,1983),89-153.
b The percentageof the active labor forceemployedoutside the primarysectorin about
1920.Source:B. R. Mitchell,
European
Historical
Statistics,
2d. rev.ed. (New York:Factson
File, 1980),151-66.
c This electiontook place in 1917.
d Prewar figureis an average forseveralelections.Source: Walter Korpi, The Democratic
ClassStruggle(London: Routledge& Kegan Paul, 1983),38.
beforethe war and in the firstelectionafterthe war, and the level of in-
dustrializationin twelve European countries.The table demonstrates
thatthe societiesthathad not become democratichad had conspicuously
higher levels of support for working-classpartieseven beforethe war.
Moreover,in almostall cases thelevel of industrializationwas lower,and
thereforethe size of the working class was smallerin these societies.In
effect,working-classparties were having distinctlymore success under
distinctlyless favorableconditions.The war raised the leftvote every-
where,but did not close thegap. In none of theearlydemocratizersis the
leftvote over 30 percentin the firstpostwarelection.It was greaterthan
30 percentin all but two of the nine late democratizers.
In some societies,foremostamong them Italy,Norway, and Sweden,
another generationof peace mighthave resultedin pluralistdemocratic
regimes.In Italy,theGiolittiancoalitionsoughtto bringtogetherreform-
458 WORLD POLITICS
ist partsof the bourgeoisieand what amounted to the Italian equivalent
of the "aristocracyof labor" in a reformmovementthatbore a remarka-
ble resemblanceto the earlyLib-Lab coalitionin Britain.In Norway and
Sweden, reformcoalitionsbetween partsof the bourgeoisieand leading
sectorsof the still fairlysmall industrialworking-classmovement had
come into being during the years immediatelybeforethe war. Even in
Germany probablyamong the countriesleast likelyto become a stable
pluralist democracy because of the high level of class-consciousness
haltingattemptsto create a reformistcoalitionof liberalsand labor had
appeared in I9I2.
Late industrializationand thewar were themostproximatedestabiliz-
ing factors.It was of more fundamentalimportancethatin societiesthat
were to become social democracies,traditionaldictatorships,or fascist,
the liberal communitywas deeplydivided. Norway and Czechoslovakia,
with sociological faultlines that divided bourgeois political movements
into linguistic,regional,culturaland religiousblocs, illustratethis well.
In Norway, liberals were never able on theirown to sponsor successful
reformmovementsbeforeWorld War I. As a result,ruralinterestsdom-
inated the Storting,the dominantvalues of the partysystemwere agrar-
ian ratherthan industrial,and economic liberalizationwas long delayed.
The shiftto parliamentarysovereigntyitselfwas mainly broughtabout
by rural representatives, and was aimed at controllingthe big spending
of urban bureaucrats." In Czechoslovakia, liberal movementswere sim-
ilarlyconstrainedby theirdivisionsand by subordinationto rural inter-
ests. Czech liberalismwas also constrainedby the peripheralpositionof
Bohemia and Moravia in the Habsburg Empire.12
These divisions within the liberal communitygenerallyreflectedthe
incompletenessof nation-stateformationbefore industrialization.The
basic questions of territorialidentity,national language, and religionre-
mained unresolved.These "preindustrialcleavages,"to borrowSeymour
MartinLipset and Stein Rokkan's phrase,continuedto divide the liberal
community;as a result,the latterwas never able to obtain the pre-emi-
nent positionthatits counterpartsacquired in countrieslike Switzerland
and Britain.I3 Switzerland'sexperienceindicatesthatthe decisivepointis
II On the weaknessof liberalismin Scandinavia,see FrancisCastles,The Social Democratic
ImageofSociety(London: Routledge& Kegan Paul, 1978).
On Czechoslovakia,see VictorMamatey,"The EstablishmentoftheRepublic,"in Victor
Mamateyand RadomfrLuza, eds.,A HistoryoftheCzechoslovak Republic,i918-I948 (Prince-
ton: PrincetonUniversityPress,1973), 3-39.
3 Sweden and Denmark are partialexceptions.The Swedish liberal communitywas di-
vided into separatepartiesby the high church-lowchurchconflictand by the temperance
issue.This is in contrastto France,forinstance,wherereligiousconflictsolidifiedratherthan
dividedtheliberalcommunity.The Danish liberalcommunitywas dividedbyconstitutional,
notthatsuchcleavagesexistin a partysystem, butthattheypreventlib-

eralsfromacquiringa positionofdominance.SwissRadicalsand Cath-
olicConservatives wereconstantly in conflictoverclericalquestions,but
because the former outnumbered the latter
by aboutfourto one in the
legislature,thepoliticaldominanceoftheRadicalswas neverchallenged.
Takinga longerviewofcausality, then,we cansaythatsocietiesin which
preindustrial cleavageswere sufficient to underminethedominanceof
liberalsat theonsetof workermobilizationcouldonlybe stabilizedby
corporatist in theI920S and I930s, whether
institutions democratic orfas-
cist,because of the debilitatingeffectof these cleavages on the political
claims of the liberal community.
Liberal interestsin thesesocietieswere attackedsimultaneouslyby tra-
ditionalurban and ruraloligarchs,peasants,and workers.Paradoxically,
the same divisions that inhibitedliberal partiesfromestablishingtheir
pre-eminencealso preventedthemfromexpandingtheirsupportoutside
the bourgeoisie.The vulnerabilityof the bourgeoisieapparentlymade it
more difficultto persuade bourgeoisvotersand activiststo accept conces-
sions to workers' interestsin exchange fortheirsupport.Thus, in coun-
tries like Italy, Sweden, and Germany, liberal parties appear never to
have succeeded in gaining the percentageof the working-classvote that
theircounterpartsreceivedin Britain,Switzerland,and France.'4
This inabilityto createsuccessfulreformist alliances of thebourgeoisie
and workersin the earlyperiod of workermobilizationhad momentous
implications.Above all, it meantthatworkerswere more likelyto be mo-
bilized as a class than as a seriesof locallyorganized interestgroups and
trade unions. In the face of limited opportunitiesto gain influence
through alliances with bourgeois interests,arguments for class-based
worker organizationswere made more compelling.Liberalism was em-
piricallyrefutedby the politicaland labor-marketexperiencesof work-
ers.
The outcome was a high level of class-consciousness,radicalization,
and polarization.Under the circumstances,the conflictbetween the two
urban classes became so intense that their representativescould not
jointlyfound a stable politicaleconomy.It was not forlack of effort:the
military, and foreignpolicyconflicts
as well as bytheusual socialreformquestions.The classic
analysis of cleavage developmentis SeymourMartin Lipset and Stein Rokkan, "Cleavage
Structures,PartySystemsand Voter Alignments:An Introduction,"in Lipset and Rokkan,
eds.,PartySystems and VoterAlignments(New York: Free Press,i967), I-64.
4 There is no directevidencethatliberalpartiesin thesesocietiesreceivedfewerworking-
class votes,but-given theless frequentand less successfulappearanceof Lib-Lab alliances-
it seems impossibleto inferotherwise.Such alliancesare discussedin SeymourMartinLipset,
"Radicalism or Reformism:The Sources of Working-ClassPolitics,"AmericanPoliticalSci-
enceReview77 (JanuaryI983), 1-19.
460 WORLD POLITICS
usual democratic coalitionofthe I9205 was a coalitionof thereforming

bourgeoisie and labor.Yet,in everycasethiscoalitioneithercouldnotbe
formedat all at thegovernmental level(as in Italy,Sweden,Denmark,
Finland,Norway,and Czechoslovakia)or soonfailedundertheweight
of its internalcontradictions(WeimarGermanyand Spain). In all of
thesecountries, thecoalitionwas eventually formedbetweenrepresenta-
tivesofone oftheurbanclassesand themiddlepeasants.
THE COALITION WITH THE COUNTRYSIDE
The coalitionhad to be withmiddle peasantsbecause,withinthe

agrarianpopulation, onlymiddlepeasantscombineda massbase,a fairly
highand stablerateofpoliticalparticipation, a distinctpoliticalagenda,
and theabilitytodestabilizethepolityifthisagendawas notrealized.By
contrast, landedelitesalmosteverywhere werea spentpoliticalforcein
the I920S and I930s; theywereno longerin controlofthemiddlepeas-
antryand theruralproletariat.Althoughtheruralproletariat haditsown
agenda and, in some countries, thepotentialto providea mass base,it
was,byitself, thepostwarequivalentofMarx's"sackofpotatoes."Inca-
pable of leadingitselfor creatingitsown organizations, it could be or-
ganizedas a durableforceonlybygroupsthatwerenotactuallyofit:so-
cialists,reform-minded and Christian
liberals, democrats. Hence,itcould
be ignoredifthesourcesofexternalleadershipwereneutralized.
Middlepeasants,bycontrast, usuallyhad autonomousinterest groups
or a powerfulvoicein largeragrariangroups.In Norway,Sweden,Den-
mark,and Finland,theyhad theirownagrarianparties.In CatholicGer-
many,Austria,Czechoslovakia,Spain,and partsofnorthern Italy,they
had a decisivepositionin less homogeneousparties.In Protestant Ger-
manyand otherpartsofnorthern Italy,theyhad theabilityto abandon
swiftly traditional
representativeswhomtheyjudgedtohaveservedthem
poorly.
A socialistmovementthatsoughtto mobilizethe ruralproletariat
(landlesslaborersand smallholders dependent on thelabormarket)con-
tainedtheseedsofitsown failure, becauseitwouldbecomeentrappedin
ruralclassconflictand therebyalienatemiddlepeasants.A coalitionwith
middlepeasantsrequiredthatsocialistsaddressthreekindsof conflict:
ruralclassconflict,consumer-producer conflict,
and,sometimes, clerical-
anticlericaland regionalconflict.Mycontention is thatdefusingthesec-
ond setofconflicts was essentialto thecreationofa worker-peasant coa-
lition,butthattheseconflicts could notbe defusedif thesocialistparty
founditselfin confrontation withmiddlepeasantson ruralclass-struc-
tureissues. If the consumer-producerissuescould be defused,theless im-

portantclerical-anticlericaland regionalissues could also be defused.
Class conflictwas as centralto the politicallives of peasantsas it was to
the political lives of urban workersand the bourgeoisie.Where socialist
partiessoughtto mobilize the ruralproletariat,theybecame enmeshed in
thisconflictand threatenedmiddle peasantsin severalways.The greatest
threatarose fromcampaigns forland reform.Redistributionwas a cen-
tralissue of rural politicsin Spain, Italy,Finland, Denmark, and Czech-
oslovakia. The fact that redistributionwould have occurred at the ex-
pense of large landowners seems not to have reduced the threatfeltby
the middle peasants, especially in areas such as northernItaly, where
there was a substantialrural proletariatand littlesurplus land to redis-
tribute.When the assertionthattheirland was secure came fromparties
bornof a collectivistambition,it appeared especiallyincredible.For peas-
ants,the securityof theassetaround which theirentireexistencerevolved
had to be beyond question. Hence, the Spanish Socialists' campaign for
land reform in the south of Spain antagonized peasants even in the
north.'5
Socialist mobilization of agrarian workersalso threatenedthe middle
peasants'productiveprocess.Middle peasants,who, bydefinition,did not
rely heavily on hired labor, nonethelesshad an employer'sperspective.
For, to the extentthat theydid hire a workeror two, socialisteffortsto
unionize agrarian workers,to establish centralized hiring halls, to fix
wages, and to extend social welfarebenefitsto agrarian workersall im-
pinged on the peasants' costsof productionand theirfreedomto manage
theirland.
The cumulativeeffectwas to threatenmiddle peasantsin a thirdway.
At stake, ultimately,was theirsocial status.In this context,it is easy to
understandwhy socialists'protestationsthattheywere committedto the
interestsof both agrarian workersand middle peasants sounded as im-
plausible to middle peasants as the reformistliberals' claims that they
were committedto the interestsof both workersand the bourgeoisiedid
to urban workers. And just as the socialists'effortson behalf of urban
workersaroused intenseresentmentamong the petitebourgeoisie,even
when it was not directlyand materiallythreatened,theireffortson behalf
of agrarian workerswere bound to generateintenseresentmentamong
middle peasants.Indeed, resentmentand a senseofneglectwere thecom-
mon featuresof all antisocialistpeasant movements.
When Socialist partiesdid succeed in foundingregimes,it was not be-
I5See Edward Malefakis,AgrarianReformand PeasantRevolutionin Spain: Originsof the
Civil War(New Haven: Yale UniversityPress,1970), 219 and passim.
462 WORLD POLITICS
cause theyhad a superiorgrasp of the strategicrequirementsof the mo-

ment,but because theydid not attemptto organize the rural proletariat.
The level of industrializationor theclass structureof thecountrysidehad
littlebearingon this,as can be seen by examiningfirstthe size of the pri-
marylabor forcein comparisonto the totallabor force,and thenthe size
of the proletarianlabor forcein comparison to the total agrarian labor
force.These data, fortheearly I920S, appear in Table 2. Of thethreeleast
agrariancountries,one became fascist,one became a traditionaldictator-
ship (Austria,under the Dollfuss-Schuschniggregime),and one became
a social democracy.Of thethreemostagrariancountries,two became fas-
cist,and the third(Finland) became semi-authoritarian.
TABLE 2
RURAL POPULATION AND REGIME OUTCOME
Primary LaborForce Agrarian Proletariat

(as a percentage (as a percentage
oftotallabor oftotalprimary Regime
force) a laborforce)a Outcome
Germany 30.7 26 Fascism
Austria 31.9 26 Dictatorship
Denmark 37.7 43 Socialdemocracy
Czechoslovakia 40.3 35 Socialdemocracy

Norway 42.1 15 Socialdemocracy
Sweden 43.7 26 Socialdemocracy
Spain 56.0 35-40 Fascism

Italy 56.2 35-40 Fascism
Finland 63.0 30 Dictatorship
a Data in the first
two columnsare fromabout 1920.The agrarianproletariatis definedas
landless workersor smallholdersdependenton the agrarianlabor marketforsome part of
theirincome.
Sources: For thefirstcolumn,Mitchell(Table 1, fn.b), 151-66.For the secondcolumn,the
figuresare estimatesbased on data reportedin Frieda Wunderlich,FarmLabor in Germany,
1810-1945(Princeton:PrincetonUniversityPress,1961),15(Germany);Statistisches Handbuch
far denBundesstaatOsterreich, vol. 26 (Vienna: StatisticalOffice,1936),43 (Austria);Danmarks
Statistik. Meddelelser.
Statistiske Landbrugets Arbejdeskraft.1904og 1935,series4, vol. 102,no. 3
(Copenhagen: StatisticalDepartment,1937), 15 (Denmark); Manuel Statistique,1931, vol. 3
(Prague: StatisticalBureau of the Czechoslovak Republic,1931),300 (Czechoslovakia); Kare
Lunden, "The Growthof Co-operativesAmong the Norwegian Dairy FarmersDuring the
Period 1856-1905,"CahiersInternationaux d'HistoireEconomiqueetSociale,vol. 6 (Geneva: Li-
brairieDroz, 1973),-342(Norway); Statistisk ArsbokfdrSverige:1932 (Stockholm:Statistical
Bureau, 1932),24-25 (Sweden); Statistisk4rsbokfdr Finland: 1931 new series(Helsinki: Cen-
tral Bureau of Statistics,1931), 50-52 (Finland); Edward Malefakis,"Peasants,Politicsand
Civil War in Spain, 1931-1939,"in RobertBezucha, ed.,ModernEuropeanSocial History(Lex-
ington,MA: Heath, 1972),196,passim (Spain); J.C. Adams, "Italy,"in WalterGalenson,ed.,
Comparative Labor Movements(New York: Russell& Russell,1968),413-14(Italy).
The size of the agrarianproletariatis also unrelatedto the regimeout-

come. Spain and Italy,both of which became fascist,had huge agrarian
proletariats.Germanyand Austria,one of which became fascistand one
of which became a traditional dictatorship,had comparativelysmall
agrarianproletariats.We have fascistoutcomes,then,withbothlarge and
small agriculturalproletariats.The lack of a patterncontinues: Den-
mark, with a large proletariat,produced a democraticoutcome. Sweden,
withan agrarianproletariatsimilarto thatof Germanyand Austria,pro-
duced a democraticoutcome. Norway, witha verysmall proletariat,and
Czechoslovakia, with a ratherlarge agrarian proletariat,also produced
democraticoutcomes.
The outcomes were determinedby politicsratherthan by rural class
structure.Socialistssucceeded in makinga coalitionwithmiddle peasants
wherevertheagrarianproletariathad been mobilized byothersbeforeso-
cialistshad an opportunityto do so. So pre-empted,thelattercould freely
negotiatethe issues thatdivided rural producersand urban workers.By
contrast,where the rural proletariatwas available as in Spain, Italy,
Finland, and Germany it presentedsocialistswith a reservoirof poten-
tial supporttoo appealing to ignore.In thesecases,thelogic ofdemocratic
competitionand the lure of immediatepower underminedthe socialists'
abilityto acquire long-termpower because the acquisition of a constitu-
ency among the rural proletariatinvariablyrequired commitmentsthat
alienated the middle peasants.
Where socialistpartieswere able to avoid thisalienation,it was possi-
ble forthem to finda way out of the regimecrisesof the early I93os by
making a bargain with middle peasants.The typicalagreementinvolved
some combination of price supports,interest-ratereductions,debt re-
structuring,protectivetariffs,and tax reliefin exchange for increased
public works, deficitspending, and expanded unemploymentbenefits.
Agreements of this sort were negotiated in Denmark, Sweden, and
Czechoslovakia in I933, and in Norway in I935.i6
By enteringa coalition that satisfiedthe interestsof middle peasants
and made the latterdependenton the social democraticstate,social dem-
ocrats attained severalobjectives.Above all, theycreated a stable power
base.I7 On this theycould build the corporatistinstitutionsthat consoli-
i6
On Czechoslovakia,see Klepetair(fn.2), 283-375;on Scandinavia,see Sven Nilsson et al.,
eds., Kriseroch krispolitik
i Nordenundermellan rigstiden [Crisesand crisispoliticsin the in-
terwarNordic countries](Uppsala: Nordic HistoryAssociation,I974).
17 For fascism,too,thecoalitionwithruralinterests providedthesocial foundationsforthe
stabilizationof a new order.Once in power,fascistscould obviouslyrelyon coercion(rather
than inducement)to a vastlygreaterdegree than could social democrats.Yet, fascistsby no
means entirelyignoredthe materialneeds of theirconstituents; a plausiblecase can be made
464 WORLD POLITICS
dated theirhold on the stateand the workingclass.'8From thisposition

ofdominance,theycould, in turn,induce thecollaborationof thebusiness
community.'9And theycould carrythroughthe social reformsthat the
interestsof theirworking-classconstituentsrequired. In Denmark, for
example, in the years immediatelyfollowing the Kanslergade Agree-
ment,the Social Democratic governmentsucceeded in gainingpassage of
legislation establishingor improving workers' compensation,invalid,
unemployment,and health insurance,a pension plan, and public assist-
ance.20
NORWAY, SWEDEN, DENMARK, AND CZECHOSLOVAKIA
The criticaldistinctionbetweencases of socialistsuccess and failureis

to be found in the marginal number of agrarian workers in successful
movements.Because thereare no standardizeddata thatallow us to make
easy crossnationalcomparisonson this score, I have relied on data on
trade-unionenrollments,collectiveagreements,intergenerationalvoter
surveys,and election and census reports.Taken together,these figures
are revealingin spiteof theirlack of strictcomparability.
We will begin with Norway, Sweden, and Denmark, threecountries
in which socialistssuccessfullyaligned themselveswith middle peasants.
Norway provides the simplestcase. Because of earlier land reformsand
the exodus of much of the rural proletariatto North America, the rural
social structurehad become remarkablyegalitarianbytheinterwaryears.
As early as I900, 97 percentof farmersowned theirland and employed
littleor no hired labor.2,Accordingto an inter-generational surveymade
by the late Stein Rokkan, agrarianvoters(middle peasants,smallholders,
and day laborers) did not account formore than I3 to I5 percentof the
total Labor Partyvote in the I920S and I93OS.22 Even if we assume that
thattheirregimes,too,acquired widespreadlegitimacy.See A. JamesGregor,ItalianFascism

and DevelopmentalDictatorship(Princeton:PrincetonUniversityPress,I979), chaps. 5-9.
i8 The close associationbetweensocial democraticpower,democratic corporatism,and the
worker-peasantalliance in Scandinavia has been extensively discussed.See G0sta Esping-An-
dersen,PoliticsAgainstMarkets(Princeton:PrincetonUniversityPress,i985), chap. 5; Walter
Korpi, The DemocraticClass Struggle(London: Routledge& Kegan Paul, i983), chap. i; Phi-
lippe C. Schmitter,"Modes of InterestIntermediationand Models of Societal Change in
WesternEurope," Comparative PoliticalStudiesio (October I978), I-20.
19See Korpi (fn. i8), 48 and passim.
20
HarryHaue et al.,Det nyDanmark,i890-I978, Udviklingslinier og tendens[The new Den-
mark, i890-I978, lines of developmentand tendencies](Copenhagen: Munksgaard, i980),
i68.
21 Kare Lunden, "The Growth of Co-operativesAmong the Norwegian Dairy Farmers
During the Period i856-I905," CahiersInternationaux d'HistoireEconomiqueet Sociale 6 (Ge-
neva: LibrarieDroz, I973), 342.
22 Stein Rokkan, "Geography,Religion and Social Class: CrosscuttingCleavages in Nor-
wegian Politics,"in Lipset and Rokkan (fn. I3), Table 26, 429.
two-thirdsof these voterswere at least partlydependenton the agrarian
labor marketforsome income,we findthatat most io percentof the par-
ty'sconstituencycould have consistedof agrarianworkers.Finally,it ap-
pears thatwithinthe Norwegian tradeunion movement,agrarianwork-
ers accounted forno more than 5 percentof the membership.23
Sweden, with a considerablyless equal distributionof land, providesa
moredemandingtestofthehypothesis. In the early I930S, 26 percentof
the agrarian work force about 250,000 men and women-were day la-
borers(see Table 2). Since almost all of thesewere adult males, we may
assume that,includingwives and otheradult dependents,at least 400,000
votes were containedin thislabor force.To put thisin perspective,recall
thatthe Socialist electorateaveraged about 750,000, and the totalelector-
ate about 2,200,000, duringtheseyears.24
Yet, the Socialist Partyand the unions almostentirelyignoredthispo-
tentialconstituency.As Sven Anders Soderpalm has noted:
The Swedishagricultural workers'movement was weakand dividedfora
longtime.... In themiddleofthetwenties, noteven5 percentofa totalof
250,000 agriculturalworkerswereorganized,and onlyin thelargeestate
areason Skaone, Ostergdtland and theMilar provinceswas thereanygreat
supportforthemovement. The declineinpartresultedfromtheadvantage
theagricultural butit was also connectedto the
crisisgave to employers,
lack of interestwhichfora long timehad distinguished thelabormove-
ment'sattitudetowardtheruralproletarians' efforts.25
The Social Democrats did not seriouslydebate appealing to agrarian
workersuntil I930; by then,these workershad already been mobilized
by the Liberal and Agrarian Parties.Liberals and Agrarianshad in fact,
succeeded in mobilizing theseworkersagainstthe Social Democrats. "In
most of the electiondistricts,"Soderpalm reports,"the majorityof farm
workersvoted forthe bourgeoisparties."26
Although we cannot estimateaccuratelythe actual amount of support
theSocial Democratic Partyreceivedfromagrarianlaborersand theirde-
pendents,we can estimateratherpreciselythemaximum amount of sup-
port it could have receivedunder the most favorablecircumstances.As-
23 Two sourcesindicatethatwithinthe Norwegian trade union movement(LO) agricul-
turaland forestry workerstogetherconstitutedonlyio% ofthemembershipin I93i, and that
forestryworkerswere a majority.See Hans FredrikDahl, Norgemelloi krigene.Det norske
samfunnet i kriseog konflikt
[Norwaybetweenthe wars. The Norwegian societyin crisisand
conflict]
(Oslo: Pax Forlag,1971), 91, andBergeFurre,Norsk I905-I40 [Norwegian
historic,
history,
I905-I940] (Oslo:Det NorskeSamlaget,
I971), 200.
24 Election data forSweden are fromStig Hadenius et al., Sverigeefter 90oo[Sweden after
i900] (Stockholm:Bonniers,I978), Table I.C.,306-7.
25 Soderpalm,"The CrisisAgreementand theSwedish Social DemocraticRoad to Power,"
in StevenKoblik, ed., Sweden'sDevelopmentfrom PovertytoAffluence, I750-I970 (Minneapo-
lis: Universityof MinnesotaPress,I975), 263-64.
26 Ibid., 264.
466 WORLD POLITICS
suming that the rate of participationwas slightlyless for agrarian

workersthan forthe populationas a whole (50 percentas against6o per-
cent),we have an electorateof about 200,000.27 We know thatthreepar-
ties competed forthesevotes: the Social Democrats,the Liberals, and the
Agrarians. If we assume that in the early I930S one-halfof these votes
went to the Social Democrats an assumptionfavorableto the latter
thisconstituencycould not have representedmore than I3 or I4 percent
of the totalSocial Democratic vote.
Denmark, with the largestday-laborpopulation of the threeScandi-
navian countries,provides an even more demanding test. Of the total
agrarian adult labor forceof 48i,000 in I934, 2I0,000 (43 percent)were
hired laborers (see Table 2). Because, again, theywere overwhelmingly
male, we may suppose that,includingwives and otheradult dependents
(parents, grandparents),they representa voting population of about
400,000. To putthisin perspective,
we shouldnotethatin I932 theSocial
Democratic electoratewas just over 66o,ooo and the totalelectoratewas
just over i.6 million. In principle,thisday-laborpopulation should have
presentedthe Social Democrats with an extraordinarily temptingtarget.
In fact,it did not because thispopulationhad alreadybeen heavilymo-
bilized byanotherparty.The Radical Liberals had come intoexistencein
i9o6 afterbreaking away from the Liberal Party. The Liberals were
quintessentiallythe partyof the middle peasants,and the Radicals came
into existenceto mobilize agrarian workersagainstthe Liberals. From
19o6 untiltheintroduction ofproportionalrepresentation in i9i8, the So-
cial Democrats agreed not to competeagainstthe Radicals (a partyled by
bourgeois reformers)in rural Jutland.In exchange,the Radicals prom-
ised not to challenge the Social Democrats in a numberof urban constit-
uencies. This gave the Radicals an open field in the districtsin which
agrarianlaborerswere concentrated.As a result,by thetimeproportional
representationwas introduced,the Radicals had captured this constitu-
ency.
If we assume thatthe rate of participationforthispotentialelectorate
of 400,000 was slightlyless than forthe entirepopulation (6o percentas
against70 percentin the earlyI930s), we have an actualelectorateof
240,000.28 We will further assume thatthe partythathad historicallyor-
ganized thiselectoratereceivedhalfof itsvote-or I20,000 votes which
27 The rate of participationequals the share of the valid vote cast as a percentageof the
populationovertwentyyearsofage. See PeterFlora, ed.,State,Economyand Societyin Western
Europe, i8I5-I975. A Data Handbook.Vol. i: The GrowthofMass Democracies(Chicago: St.
JamesPress, i983), I4I-44.
28 The rate of participationequals the share of the valid vote cast as a percentageof the
populationover 20 yearsof age. Ibid., I05-6.
in factapproximateswhat the Radicals receivedin ruraldistrictsin I935.

We will also assumed thatthe remainderwas distributedamong the So-
cial Democrats and all otherpartiesat a ratioof fourto one. Under the
foregoinggenerous assumptions,90,ooo of thesevoteswent to the Social
Democrats.29With a totalSocial Democratic vote of 760,000 in I935, this
representsabout I3 percentof the Social Democratic electorate.Finally,
we should note that the participationof agrarian workers in Socialist
trade unions during theseyearswas virtuallynil.
In sum, in Norway and especiallyin Sweden and Denmark, the mo-
bilization of agrarian workersby otherpartieswas decisive. The objec-
tioncan be made thatthe real distinctionbetweenthesecountriesand the
countriesthatbecame fascist-especially Italy and Spain-is that,in the
former,agrarian workers were mobilized by other parties because the
transitionto mass politicswas more gradual. With a more sudden tran-
sition,thispopulationwould nothave been captured,and socialistswould
have stepped in. In thisconnection,the case of Czechoslovakia, in which
the transitionwas extremelyabrupt,becomes significant.
In Czechoslovakia, the working-classmovementwas divided ethni-
cally, regionally,and ideologically.The German Social Democrats, be-
cause of theirethnicbase, organized almostexclusivelyindustrialwork-
ers. The [National] Socialists,limitedto Bohemia and Moravia, were also
exclusivelyurban. Bohemian and Moravian agriculturalworkers-there
were not many of them-were in factinitiallyorganized by the Social
Democratic Partyduringthefirstfewyearsof theRepublic. But, because
these were heavily industrializedareas, agriculturalworkers were in a
less dependent position.They routinelymoved back and forthbetween
industrialand agriculturalemployment.In I92I, only28.4 percentof the
labor forceof Bohemia was in agriculture.The percentagesforthe other
provinceswere: Moravia and Silesia, 33.8; Slovakia,58.7; and Ruthenia,
62.4.30 The great majorityof agrarian workerswere thus located in Slo-
vakia and Ruthenia. A conservativeestimatewould be that,in Slovakia,
40 percentof the agrarianlabor forceconsistedof proletarians;in Ruthe-
nia it was probablyclose to half.In these provincestherewas almost no
industry,and large surplus rural populations were dependent on the
agrarian labor market.
The Social-Democratic Party never had an opportunityto organize
theseworkers.In the schismbetweensocial democracyand communism,
the Communistscapturedthe partyorganizationin Slovakia and Ruthe-
29 For electiondata forDenmark, see ibid.
3?Vaclav L. Beneg,"Czechoslovak Democracyand Its Problems,i9i8-I920," in Mamatey
and Luza (fn. I2), Table 2, p. 4I.
468 WORLD POLITICS
nia, while Social Democrats prevailedin the industrialareas of Bohemia

and Moravia. The resultwas that the Social Democratic Party became
mainly a Bohemian and Moravian partyand mainly one of industrial
workers.Thus, in the firstpost-schismelection,in 1925, the Social Dem-
ocraticPartywas wiped out in Slovakia and Ruthenia,receiving4.3 per-
centand less than2 percentof thevote,respectively. In contrast,theCom-
munistPartyreceived I3 percentof the Slovakian vote; with30.8 percent
of the Ruthenianvote,it became the largestpartyin Ruthenia.3'
The Communist Partycapturedthe social democraticorganizationin
these regions,but it did not succeed in mobilizing agrarian workers.32
These workerswere mainlyorganized by the regionalCatholic Populist
partiesand the secular Agrarianparties.The agrarians,who were mainly
parties of middle peasants, were especially successfulin capturingthis
constituencybecause of theirleading role in the land reformprogram.
They could unite middle peasantsand day laborersbecause land reform,
which consolidatedmainlythe positionof the former,was aimed specif-
icallyat Magyar and German landowners;it consequentlyexploitedeth-
nic cleavages in the countryside.33 The net effectof the schism and the
land reformwas thatthe Social Democrats neverdeveloped a significant
involvementin rural class conflict.The democraticworking-classmove-
ment (Social Democrats, [National] Socialists,and German Social Dem-
ocrats) remained urban and industrialin its orientation,and freeto col-
laborate with the representativesof the middle peasants (the secular
Agrarian partiesand the Catholic Populist parties).
The case of Czechoslovakia also demonstratesthatthe many overlap-
ping cleavages were not sufficientto preventa stable democraticcoali-
tion; neither was the absence of a coherentpeasant party.In this, the
Czech example stands as a useful correctiveto the propositiondrawn
fromthe Spanish, Italian,and German cases: thattheabsence of coherent
peasant representationand themultiplicity of cleavages in thosecountries
enfeebleddemocracy.
It was still possible forthe Socialists to make a coalitionwith middle
peasants to overcometheirconsumer-producer, religious,and regional
conflicts-even when peasants were themselvesdivided by these cleav-
ages. The socialistswere divided among themselvesby ethnicityand re-
gion (autonomist German Social Democrats versus centralizingBohe-
31Victor S. Mamatey, "The Development of Czechoslovak Democracy, I920-I938," in
Mamateyand Luza (fn. I2), I29.
32 The ruralcommunistvote came mainlyfromethnicminorities.
See RichardV. Burkes,
The Dynamicsof Communismin EasternEurope (Westport,CT: Greenwood Press, I976), 42
and passim.
33Czech land reformis discussedin Klepetair(fn.2).
mian-Moravian Social Democrats versus centralizingSlovakian Social

Democrats). The peasantpartieswere divided fromtheSocial Democrats
by region (centralizingSocial Democrats versus autonomist Slovakian
Populists) and by religion(secular Social Democrats versusCatholic Bo-
hemian-Moravian Populists versus Catholic Slovakian Populists). And
the peasant partieswere themselvesdivided by religionand region (sec-
ular, centralizingBohemian-Moravian Agrarians versus Catholic, cen-
tralizingBohemian-MoravianPopulists versussecular,centralizingSlo-
vakian AgrariansversusCatholic autonomistSlovakian Populists).These
cleavages provoked conflictsbitterenough to be definingfeaturesof the
partysystem;but so long as the coalitionservedthe otherinterestsof the
participants,theywere willingto avoid a showdown on theseissues.
SPAIN, ITALY, AND GERMANY
The positionof agrarianlaborersin thesocialistmovementlooks quite

differentwhen we turnour attentionto Spain, Italy,and Germany. Al-
thoughit is difficultto estimatethe votingpreferencesof agrarianwork-
ers, data on the composition of trade unions make it clear that these
workersexercisedenormous leveragein the socialistmovementsof Italy
and Spain; in Germany theirinfluencewas less, but still substantial.In
these countries agrarian labor had remained largely unmobilized by
other political groups; therefore,it presentedsocialistpartiesand trade
unions with a vast reservoirof potentialsupport.
The decisive class conflictin the Italian countrysideoccurred in the
northand center;it pittedday laborers(braccianti)againsta mixed group
of sharecroppers,tenants,and small and large owners.The two most im-
portantpointsof contentionwere the local monopolyover hiring,which
the sharecroppers,as periodic employers,resentedno less than tenants
and peasant owners, and the ownership of harvestingequipment, the
sharecropperscaringlittleforthesocialistvisionofcollectiveownership.34
The socialistcause was pressed in the countrysideby the Syndicalists
(USI) and Federterra.The latterwas theagrariansectionof thenormally
reformist,socialistConfederazione Generale del Lavoro (CGL); it pro-
vided 33 percentof the CGL's membershipin 1920.35 Federterrasought
to bridge the conflictbetweenlaborersand sharecroppers,but itsmassive
postwar mobilizationof laborersand its collectivistambitionsmade this
impossible. As its membershiprose fromabout93,000 in i9i8 to over
34 Seton-Watson,ItalyfromLiberalismto Fascism(London: Methuen,i967), 303-4.

C.
35Daniel L. Horowitz, The Italian Labor Movement(Cambridge: Harvard University
Press, i963), I24.
470 WORLD POLITICS
300,000 in i9I9, Federterraleaders lost controlof theirmembersand felt

compelled to act more militantlythan theirpast experiencewould other-
wise have recommended.36
The strategyof jointlyorganizingsharecroppersand laborersbecame
untenableas well, because the end of the war saw a dramaticincreasein
the size of the middle peasantryas laborersbecame sharecroppersand
landowners: the percentageof owners rose from i8.3 of the males en-
gaged in agriculturein I9II to 32.4 in I921.37 Federterraleaders feared
thattheirconstituencywould shiftits politicalloyaltyas its class position
changed.
As a solution to this, Federterra advocated collective ownership of
equipment among sharecroppers.This was an attemptto bind sharecrop-
pers to the Federterra organization. Instead, this new emphasis drove
sharecroppersinto the arms of local Catholic unions (CIL), thus trans-
formingFederterra: by I920, agrarian laborersprovided virtuallyall of
Federterra'smembership,but theyaccounted foronly io percentof the
agrarianmembershipof the CIL. The remainderconsistedof sharecrop-
pers,tenants(78.4 percent),and small owners (io percent).38
Attentionto theFederterraalone providesan incompletepicture.The
CGL, or at least itsurban,industrialwing,had always followeda reform-
ist course, and this did not change substantiallyafterthe war. But the
CGL was not the only importantsocialistorganizer of day laborers.In
I920, the three syndicalistunions (USI, UIL and the independentrail-
waymen s union) organized about one-thirdof the socialistlabor force.39
The agrariancomponent-overwhelminglylaborers,mainlyin Emilia
probablyaccounted forabout half of the syndicalistmembership.4o The
effectof the syndicalistunions was to furtherincreasethe dependence of
the socialistmovementon day laborers:once theseworkershad been po-
liticallymobilized by a part of the left,it became imperativethatthe So-
cialistPartysolicittheirsupport.
The role of the day laborers,however,did not stopeven there,because
of the special place theyoccupied in bolsteringthe positionof the intran-
sigentsin the Socialist Party.Intransigentswere dominant in the party
36 Maurice Neufeld,Italy:SchoolforAwakeningCountries: The Italian Labor Movementin

Its Political,Social and EconomicSettingfrom i8oo to 1960 (Ithaca, NY: Cornell University
School forIndustrialand Labor Relations,i961), 367.
37Maier (fn. I 0), 3 I I .
38 Horowitz (fn.35), I24-
39 See Neufeld (fn.36), 347. Also see Michael F. Hembree, "The Politicsof Intransigence:
CostantinoLazzari and the Italian SocialistLeft,i882-I9I9," Ph.D. diss. (Florida State Uni-
versity,I98I), 232.
4?The agrarianmembershipis an estimate.For the USI, it was assumed to be equal to the

membershiptaken away fromthe CGL: 8o,ooo.See Horowitz (fn.35), 75.
leadershipand in the syndicalistunions. Accommodationistswere dom-

inantin the parliamentarypartyand in the CGL. Since the intransigents
could not relyon the supportof the CGL, theywere inclined to accen-
tuate the claims of the syndicalistunions and, therefore,
of the agrarians.
So much so, it turnedout,thateven when thefascistthreathad convinced
the leaders of the CGL of the need to reach an accommodationwith the
Catholic unions in organizing the sharecroppersand small landowners,
the Socialist Partyleadershipresisted.4'
The lines of class conflictin the Spanish countrysidedifferedfrom
those in Italy. In Spain, class conflictpitted day laborers,tenants,and
marginal owners (all of whom were dependent on the labor market)
against middle and large owners; some of the latterwere directlyin-
volved in cultivationand some were membersof the urban bourgeoisie.
Two unions had traditionallymobilized farmlabor: theanarchists(in the
CNT, Confederacion Nacional del Trabajo) and the socialists (in the
UGT, Union General de Trabajadores). Until Primo de Rivera's dicta-
torship,theCNT had been byfarthemore importantorganizerof agrar-
ian workers. By the I930s, however,the CNT was a spent forcein the
countryside,mainly because of the heavy hand of the dictatorship.The
UGT, by contrast,had been favored by the dictatorshipand, with its
well-establishedpresence among industrialvoters,was ready to move
forcefullyin the agriculturalareas. The Federacion Nacional de Traba-
jadores de la Tierra (FNTT), the agrarian wing of the UGT, increased
its membershipfroma mere 36,639 in 1930 to 451,337 in 1933, thereby
accounting for about 40 percentof the total UGT membership.42The
proportionof the FNTT membershipthatwas dependentmainlyon the
labor market forits income was veryhigh,probablyin excess of 8o per-
cent.
One indicationof thisis, accordingto Edward Malefakis,the close re-
lationshipbetween the SocialistParty(PSOE) vote and the size of the la-
boringpopulation.Malefakisobservedthedisparityin thepatternof land
tenure between the north(small familyowners), the centerand littoral
(small and intermediateproperties),and the southwest (large estates
worked by day laborers).He demonstratedthat this distributiontrans-
lated into a steadyvotingpattern:rightistdeputies were elected almost
entirelyby the small peasant proprietorsof Old Castile and Navarre; the
Socialists were the largestpartyin southwesternSpain; and the political
centerwas elected in the geographiccenter.43
The PSOE leftno doubt about the kind of reformit wanted; land had
41Seton-Watson(fn 34),599. 42 Malefakis(fn. I5), 292.
43Ibid.,206.
472 WORLD POLITICS
tobe redistributedtothepoorerruralworkers, especiallythelandlessday

laborers.The partyboreprimaryresponsibility forthe I932 legislation
thatthreatened theland titlesnotonlyoflargelandownersin thesouth-
west,butalso of smallownerswho leasedout theirland in Old Castile
and Navarre.In I93I, thePSOE and theotherSan Sebastianpartiesen-
acteda seriesofemergency decreesthatforbadeexpulsionof smallten-
ants;forbadeownersfromwithdrawing theirland fromcultivation; es-
tablishedrulesthatfavoredcollectivesin therenting oflargeproperties;
establishedan eight-hour day and a defactowage increase;established
whatamountedto a closedshopin thelabormarket,thereby vastlyin-
creasingthepoweroftheFNTT; and established arbitrationboardsthat
generally reinforcedtheleadingpositionoftheFNTT.
Emboldenedordrivenbyitsgrowingsupportamongtheagrarianpro-
letariat,theFNTT was thesocialistorganization thatinitiatedtheturn
towardradicalization.This radicalizationalienatedthePSOE and UGT
fromtheCatholicmiddlepeasantsof thenorthand center.Indeed,this
radicalizationbeganwithan agricultural issue.In JuneI934, theFNTT
calleda nationwidestrikein responsetotherepealoftheI93I Municipal
Labor Act,theact thathad createdthearbitration boardson whichso
muchoftheFNTT's influence depended.The strikereceivedonlyluke-
warmsupportfromtheurbanwingof theUGT and eventually failed,
causingthelossoftensofthousandsofunionmembers.In itsaftermath,
and especiallyin thewake ofthefailedOctoberRevolution, thegulfbe-
tweenthetwowingsofthesocialistmovement, and betweenthemove-
mentand the middlepeasants,becameevengreater.The polarization
came to a head in February I936, when middle peasantsin thecenterand
northvotedoverwhelmingly fortheCatholicConfederacion de Derechas
Autonomas(CEDA) and agrarianlaborersin thecenterand southwest
votedoverwhelmingly forthe socialists.44
This electoraloutcomemir-
roredthealignments thatformedon behalfoffascismand theRepublic
a fewmonthslater.45
AlthoughGermanypresentsa morecomplicated picture,theconclu-
sionis thesame.Notwithstanding theirconventional imagesas quintes-
sentiallyurbanand working-class, theSocial DemocraticPartyand the
freetradeunions(ADGB) werebothdependenton theagrarianprole-
tariat.Let us note,to beginwith,thatin I928 purelyruraldistrictsgave
no less than25 percentof theirvoteto the socialistparties(SPD and
KPD) and 2I.5 percentof theirvoteto theSPD alone.46 These data be-
Ibid.,214.
44 45 Ibid., 2I9.
Derek Urwin and Frank Aarebrot,"Socio-geographicCorrelatesof LeftVotingin Wei-
46
mar Germany,I924-I932," Structures

in Per Torsvik,ed., Mobilization,Center-Periphery and
come more tellingwhen we considerthe strengthof the ADGB-affiliated
agrarian workers'union, Deutscher Landarbeiterverband(DLV).
Until i9i8, Germany'sagrarianworkers about 3,000,000 in all had
remainedpoliticallyunorganized.Agrariantradeunionswere illegal un-
til the collapse of the empire. This segmentof the proletariatmust have
seemed an especially temptingtargetfor the Social Democrats and the
free trade unions. Indeed, in the SPD's vision of the political and eco-
nomic developmentof the country,these workerswere assigned a vital
role. Politically,it was necessaryto weaken the positionof the Junkers.
Land reformwas not a plausible solutionbecause in early i919, when the
SPD mighthave had thepoliticalpower to introduceit,therewere severe
food shortages.Moreover,it was unsatisfactory because it conflictedwith
the SPD's axiom that the "economic development of the countryde-
pended on large-scale production in both industryand agriculture."47
Land reformwould have increased the number of allegedly inefficient
small farmsas well as the size of a middle peasantrythat"showed no in-
clinationtoward a socialismthatconsideredthemfatedto lose theirhold-
ings."48Like theirSpanish and Italian counterparts,the German Social
Democrats faced the dilemma of mobilizing agrarianworkerswhile not
raisingthemout of the proletariat.
The German solutionwas to apply an industrialmodel to agriculture.
Agriculturalworkerswould be collectivelyorganized and would sharein
all of the labor marketand social policy rightsthatapplied to industrial
workers.This solutionaddressed all of the concernsof the Social Dem-
ocrats.Increases in wages and taxes would weaken the Junkers,preserve
large units of production,preventfood shortages,and would not create
more middle peasants.At thesame time,workers,"at leasta considerable
number of them, could be won for the Social Democratic Party and
unionized....549
At the end of i9i8, there were only 20,000 unionized agricultural
workersin all of Germany.By I920, theirnumberhad risento about one
million.Membershipthendeclined and stabilizedat about 200,000 forthe
remainingperiod of the Republic. This is in contrastto about 95 million
industrial workers in the ADGB in I923, and a stabilized industrial
membershipof about 4.9 millionthereafter. It would appear thatsocialist
unions in the countrysidehad no more than a residual role. At the peak
of unionization, about one in ten ADGB members was an agrarian
Nation-Building(Oslo: Universitetsforlaget,
198I), 259-60.
47 Frieda Wunderlich,FarmLabor in Germany, i8io-I945 (Princeton:PrincetonUniversity
Press, 196I), 35.
48 Ibid. 49Ibid., 35-36.
474 WORLD POLITICS
worker; during the last half of the Republic, only about one in twenty-
fivewas.50
These figuresseverelyunderstatethe impact of the agrarian unions,
however. This becomes clear when we examine data on the industrial
and agrarian workers who were covered by collective contracts.The
number of industrialworkerscovered by collectiveagreements union
members or not was I3,I35,000 in I924, I2,276,000 in I 929, and
II,950,000 in I93I.5I The equivalent figures for agrarian workers were
2,37I,7I9 in I924, 2,749,398 in I926, and 2,I23,IIO in I928.52 Since the to-
tal population of agriculturallaborers was only about 3,000,000, this
means thattwo-thirdsof themor more were coveredby collectiveagree-
ments.Twenty percentof the totallabor forcecovered by ADGB agree-
mentswere in agriculture.We now begin to see the potentialimportance
of thisconstituencyto theSPD and theADGB. It seemslikelythata large
part of the 25 percentof the rural vote thatwent to the SPD and KPD
came fromthisconstituency.
We can see even more clearlythemannerin which thispowerfulunion
presence-and the largerpoliticaland economicstrategyof which it was
a part-was bound to antagonize the middle peasants.For the corollary
of that strategywas that the SPD and the ADGB took a stronginterest
in the living standardsof agrarian laborers.Indeed, in the firstyears of
the Republic, the SPD succeeded in winning passage of legislationthat
extended all of the social policy rightsof industrialworkersto agrarian
workers,guaranteed them the rightto unionize, and extended to agri-
culture the same systemof compulsoryarbitrationthat was used in in-
dustrialwage disputes.All agrarianworkersbecame coveredby compul-
sory health, pension, unemployment,and disabilityinsurance schemes.
The costs of these programswere covered in the same manner as those
forindustrialworkers: by joint employee and employercontributions.53
German agriculturalworkerswere surelythemostcomprehensivelypro-
tectedin Europe in the I920S.
These benefitswere being extendedto agrarianworkersduring years
when the middle peasants' income was collapsing.The net share of na-
tional income going to agriculturefell from I3.0 percentin I9I3 to 7.2
percentin I929.54 Several studiesprovide plausible evidence thatthe in-
5?Gerhard Bry, Wages in Germany,i87I-I945 (Princeton: PrincetonUniversityPress,

i960), 32.
51Ibid., 42.
52 Wunderlich(fn.47), 84, n. I4.
53Ibid.,I 26-59.
54Paul Jostock,"The Long-Term Growth of National Income in Germany,"in Simon
Kuznets,ed., Incomeand Wealth,Vol. V (London: Bowes and Bowes, I955), I09.
come of a peasant actuallyfellbelow thatof a worker in the late I 920S.55

In thiscontext,thepeasants'almosthystericalattackson Sozialpolitik, and
theirantipathy to theSPD, becomecomprehensible.
On balance,then,it appearsthatthe GermanSocial Democrats-
throughtheirefforts on behalfof thesocialrightsof agrarianworkers,
legislationthatimprovedthemarketpositionof theseworkers,and the
generalizationofcollectiveagreements also becameentangledin rural
classconflict.
THE TRADITIONAL DICTATORSHIPS
One questionremains:Why,in thecountries discussedso far,did lib-

eralsnotenterintoa coalitionwithmiddlepeasants?The factis thattra-
ditionalliberalswere involvedin such coalitionsin Finland,Austria,
Hungary,and muchofeasternEurope.In thesecountries, theoutcome
was a rathertraditionalauthoritarian regimethatwas distinguished,
aboveall,byitslackofinstitutionalinnovation.
Finlandexperienced parliamentary
a semi-competitive orderthatwas
premisedon theharassment and outlawingoftheCommunistParty,the
disablingoftradeunions,and theenfeeblement oftheSocialDemocratic
Party.Withparticipation and dissentthusconstrained,it was possibleto
preservea circumscribedparliamentarism.The Dollfuss-Schuschnigg re-
gimein Austriawas a morefullyformeddictatorship; itlastedfromthe
briefcivilwar in I934 untiltheAnschlussin I938. Under thedictatorship,
theparliamentwas closed,theCommunistand Social Democratic parties
and tradeunions were outlawed,and democraticfreedomswere circum-
scribed.Even so, the interestsincluded in the rulingcoalitionwere given
ratherbroad latitude for dissent,and the working-classorganizations
thatwere formallysuppressedwere allowed to carryon underground.In
Hungary, politicalcompetitionwas also severelycircumscribed,and the
Social Democratic Partyand trade unions were able to operate only un-
der veryrestrictedconditions.These regimeswere thus typicalof tradi-
tional authoritarianismin two ways: in theirtoleranceof limiteddissent
and in theirdisinterestin creatingsubstitutesforthe working-classinsti-
tutionstheysuppressed.56
55 Adolph Mtinzinger,"Eine bduerlicheBetriebserhebung in Wiirttemberg"[A surveyof

peasant productionin Wiirttemberg];Der Arbeitsertrag der bduerlichenFamilienwirtschaft,
Vol. II (Berlin I929), 873; Constantinvon Dietze, "Die Lage der deutschenLandwirtschaft"
[The position of German agriculture], fir Nationalakonomieund Statistik(Jena:
Jahrbficher
I929), Vol. I30, p. 3; Vol- 75, pp. 659-60.
56 See Risto Alapuro and Erik Allardt,"The Lapua Movement:The Threat of a Rightist
Takeover in Finland, 1930-32," and Walter Simon, "Democracy in the Shadow of Imposed
476 WORLD POLITICS
They were certainlynot fascist.That much is made clear not only by

theabsence of fascistinstitutions,but by the factthatin Finland and Aus-
triathe regimescame about in partas a reactionto growingfascistmove-
ments-the Lapua Movementin Finland and the Nazi Partyin Austria.
Indeed, the Austriangovernmentput down an attemptedNazi coup in
I934 and, then,in the face of the impendingAnschluss in I938, called a
referendumon Austrianindependenceand invitedSocial Democrats to
participatein a cabinetof national unity.
What distinguishedthese traditionalliberalsand allowed them to in-
stall thesedictatorshipswith the aid of agrarians?The answer appears to
be that in these countriesliberals were more cohesive. In Finland, the
bourgeoisie was divided into a Swedish Liberal Party,a Finnish Liberal
Party,and a ConservativeParty.These divisionsremained in the party
system,but were largelyovershadowedby theexperienceof thecivil war,
which forcedthebourgeoisieto uniteor perish,and bytheconstantthreat
posed by the existenceof a Bolshevik regime in neighboringRussia. In
Austria,the bourgeoisiewas less divided. A substantialpart of it was al-
ready organized by the clericalChristianSocial Party,which itselfwas a
coalition of bourgeois and middle peasant interests.Along with some
anticlericalGerman nationals,the Christian Social Party provided the
core of the authoritarianVaterland Front. What is strikingabout Fin-
land, Austria, and Hungary is that all threeexperiencedcivil wars that
were foughton explicitlyclass lines and thatrequired the bourgeoisieto
overcome its historicaldivisionsin orderto survive.57
Finland and Austria were also distinguishedby the exceptional co-
hesivenessof the middle peasantry.In each of these countries,the peas-
ants were organized in one party the Agrarian in Finland and Chris-
tian Social in Austria. Because thesewere the largestnonsocialistparties
in each countryand because thebourgeoisiewas less cohesive,it was pos-
sible forthem to dominate any coalition.In Hungary the peasants were
not as cohesivelyorganized, but theynonethelessprovided a mainstayof
the Horthyregime.
Elsewhere, traditionalliberals failed; the liberal communitywas di-
Sovereignty:The FirstRepublic of Austria,"bothin JuanLinz and AlfredStepan,eds., The

BreakdownofDemocraticRegimes:Europe (Baltimore:The JohnsHopkins UniversityPress,
I978). On Hungary, see Andrew Janos,The Politicsof Backwardness in Hungary,i825-I945
(Princeton:PrincetonUniversityPress,i982), chaps. 4-6.
57 Lithuania,Latvia, and Estonia providetextbookcases of coalitionsbetweenliberalsand
middle peasants leading to traditionaldictatorshipsbetweenthe wars. In each case, a brief
class-basedcivil war followedimmediatelyon World War I, and in each case, socialistswere
activein organizingagrarianlabor in the period beforethe instaurationof the dictatorship.
See Georg von Rauch, Die Geschichte der baltischenStaaten[The historyof the Baltic states]
W. Kohlhammer,
(Stuttgart: 1970).
videdintoa multiplicity of parties,noneof whichwas largeenoughto

claimplausiblytospeakforthebourgeoisie. The result,on theone hand,
was thatnoneofthemcouldoffer middlepeasantsa valuableallyand,on
theotherhand,thatnoneof them,individually, had an incentive
to be-
comesuchan ally.In thesecases,a bargainwas struckwithmiddlepeas-
antsbyeithersocialists(in Norway,Sweden,Denmark,Czechoslovakia)
or byentirely new fascistmovements (in Germany,Italy,Spain),which
suffered neitherthehistorical cleavagesthatdividedand paralyzedlib-
eralpartiesnortheinvolvement in ruralclassstruggle thathandicapped
socialistparties.In short,thepreconditions of fascistsuccesswerea di-
videdliberalcommunity and a working-class movement engagedina de-
fenseoftheruralproletariat.
CONCLUSIONS
In developingthisproposition,
we haveseenthatexplanationsofinter-
war politicscouchedin termsof a multiplicity
of overlappingcleavages
or theabsenceof a singlecoherentpeasantpartyfailto accountforthe
regime outcomes.58The foregoingdiscussion has also placed in a new
lightseveralcommoninterpretations offascismand democratic instabil-
itybetweenthe wars.Barrington Mooreand AlexanderGerschenkron
havearguedthata fascist outcomewascausedbytheexistence ofa landed
elitewhichadded the supportor acquiescenceof a subordinated rural
mass to a coalitionwiththe bourgeoisie.59 In thisview,what distin-
guishedsocietieswas thepresenceofa landedelitein controloftherural
masses:England and France becameliberaldemocraciesbecausethey
lackeda politically
decisivelandedelite;Germanybecamefascist because
itpreserved one.
When appliedto theinterwar years,Moore'sand Gerschenkron's ar-
gumentis questionableforseveralreasons.First,as we have observed,
thereis no correlation
betweentheruralsocialstructure (inparticularthe
size ofthedependentlaborforce),and theregimeoutcome.Second,even
ifa landedeliteexercisedeconomiccontrolofa ruralmass,itdid notnec-
essarilycontrolthe latter'spoliticalbehavior.In Spain, for example,
southernday laborersvoted overwhelmingly for the SocialistParty
throughout thelifeoftheRepublic.Third,an authoritarian outcomedid
notrequirethata landedelitehavepoliticalcontrolofa ruralmass.When
an independentpeasantry sided withthebourgeoisie, an authoritarian
outcomeensuedevenin theabsenceof an important landedelite.Fin-
58 See the discussionof the Czechoslovak case.
59 Moore (fn. i); Gerschenkron(fn. i).
478 WORLD POLITICS
land'ssmalland mediumpeasants,nevernotedfortheirsubordination,
alignedwiththebourgeoisie, firstto winthecivilwarand thento main-
tainan authoritarian regime.In Spain,thecriticalsupportfora fascist
solutioncamenotfromsouthern landedelites,who had alwaysbeenau-
thoritarianbutunabletocontroltheirworkers, butfrommiddlepeasants
in thenorthand center.
A finalexplanationfortheweaknessofinterwar democracy pointsto
a lack of cooperationamongdemocrats, engenderedbyacutepolariza-
tion.60
However,polarization did notso muchcausespecific outcomesas
reflectthe inabilityto findany stablesolution.Some countriesthat
adoptedauthoritarian regimes-Spainand Austriain particular-were
in fact conspicuously depolarized in 1931-1933 and 1919-1923 respec-
Some countriesthatbecamedemocratic-particularly
tively.6' Norway
and Sweden werein factamongthemostpolarizedin the 1920S and
had someofthemostradicalizedlaborpartiesin Europe.62 Wheredem-
ocratsfailed,it was not becausetheirsocietieshad becomepolarized;
rather,thesocietiesbecamepolarizedbecausedemocrats had failed.
In fact,itmattered littlein theendwhether socialistleaderswerecom-
mittedto radicalismor reformism.63 The unqualified commitment tothe
reformist, electoralroad to power,whichwas pursuedby theGerman
and,initially,theSpanishsocialists,couldbe as self-defeatingas themore
radicalposturetakenby theirItaliancounterparts. Indeed,theSpanish
experiencesuggeststhatsucha strategy was thecauseof radicalization.
Moderationand democratic requiredthatsocialists
stability ignorea sub-
stantialpartoftheworkingclass,and no socialistleadersconsciously de-
cidedto do so. Wheretheydid,it was becausetheyhad no choice.And
wheretheyhad no choice-whethertheywereinitially radicalafterthe
war,as in Norwayand Sweden,or reformist, as in Denmarkand Czech-
oslovakia-the outcomewas stabledemocracy.
60 This is the centralthemeof the essaysin thevolume editedby Linz and Stepan (fn.56).
6,
See Juan Linz, "From Great Hopes to Civil War: The Breakdown of Democracy in
Spain," and Walter Simon, "Democracy in the Shadow of Imposed Sovereignty:The First
Republic of Austria,"in Linz and Stepan (fn.56).
62 The levels of Norway's and Sweden's strikeactivities were among the highestin Europe
duringthe I920S. See WalterKorpi and Michael Shalev,"Strikes,Power and Politicsin West-
ern Nations, I900-I976," PoliticalPowerand Social Theoryi (i980), 30I-34; Douglas Hibbs,
"On thePoliticalEconomyofLong-Run Trends in StrikeActivity," British
JournalofPolitical
Science2 (April I978), 26-43.
63 Among themostimportant effortsto accountfortheoriginsofradicalismand reformism
in labor movementsis Lipset (fn. I4), esp. pp. I4-i6. Lipset contendsthatradicalismand re-
formismwere one among severalimportantvariablesin accountingforregimeoutcomesbe-
tweenthe wars.

Luebbert-Social Foundations of Political Order in Interwar Europe

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Luebbert-Social Foundations of Political Order in Interwar Europe

Caricato da

Copyright:

Formati disponibili

Trustees of Princeton University

Social Foundations of Political Order in Interwar Europe

TOUR typesof regimesof historicimportanceappeared in Europe be-

democracy.This could be done if socialistscould forma coalition with

2 The interwarCzechoslovakian experienceand its corporatistaspects can be found in

gimesappearedin Nazi Germany, FascistItalyand,toa lesserdegree,in

liticalcontrol.Rather,theyexercisedmore or less influence,sometimesin

The model of the pluralistpath, as exemplifiedby the Britishtransi-

3 Early democratizersare definedas societiesthatacquired responsibleparliamentaryin-

5Quoted in Roger Magraw,France,i8I5-I9I4: TheBourgeoisCentury(New York: Oxford

SOCIAL DEMOCRACY AND FASCISM

Prewar Postwar Labor in

Netherlands 22.9 25.0 77.0

Norway 26.3 31.6 57.9

ClassStruggle(London: Routledge& Kegan Paul, 1983),38.

notthatsuchcleavagesexistin a partysystem, butthattheypreventlib-

usual democratic coalitionofthe I9205 was a coalitionof thereforming

THE COALITION WITH THE COUNTRYSIDE

The coalitionhad to be withmiddle peasantsbecause,withinthe

tureissues. If the consumer-producerissuescould be defused,theless im-

cause theyhad a superiorgrasp of the strategicrequirementsof the mo-

Primary LaborForce Agrarian Proletariat

Czechoslovakia 40.3 35 Socialdemocracy

Spain 56.0 35-40 Fascism

The size of the agrarianproletariatis also unrelatedto the regimeout-

dated theirhold on the stateand the workingclass.'8From thisposition

NORWAY, SWEDEN, DENMARK, AND CZECHOSLOVAKIA

The criticaldistinctionbetweencases of socialistsuccess and failureis

thattheirregimes,too,acquired widespreadlegitimacy.See A. JamesGregor,ItalianFascism

suming that the rate of participationwas slightlyless for agrarian

in factapproximateswhat the Radicals receivedin ruraldistrictsin I935.

nia, while Social Democrats prevailedin the industrialareas of Bohemia

mian-Moravian Social Democrats versus centralizingSlovakian Social

SPAIN, ITALY, AND GERMANY

The positionof agrarianlaborersin thesocialistmovementlooks quite

34 Seton-Watson,ItalyfromLiberalismto Fascism(London: Methuen,i967), 303-4.

300,000 in i9I9, Federterraleaders lost controlof theirmembersand felt

36 Maurice Neufeld,Italy:SchoolforAwakeningCountries: The Italian Labor Movementin

4?The agrarianmembershipis an estimate.For the USI, it was assumed to be equal to the

leadershipand in the syndicalistunions. Accommodationistswere dom-

tobe redistributedtothepoorerruralworkers, especiallythelandlessday

mar Germany,I924-I932," Structures

5?Gerhard Bry, Wages in Germany,i87I-I945 (Princeton: PrincetonUniversityPress,

come of a peasant actuallyfellbelow thatof a worker in the late I 920S.55

THE TRADITIONAL DICTATORSHIPS

One questionremains:Why,in thecountries discussedso far,did lib-

55 Adolph Mtinzinger,"Eine bduerlicheBetriebserhebung in Wiirttemberg"[A surveyof

They were certainlynot fascist.That much is made clear not only by

Sovereignty:The FirstRepublic of Austria,"bothin JuanLinz and AlfredStepan,eds., The

videdintoa multiplicity of parties,noneof whichwas largeenoughto

Potrebbero piacerti anche