History and Public Policy vs.

History of Public Policy

The Comparative History of Public Policy by Francis G. Castles
Review by: Douglas E. Ashford
Public Administration Review, Vol. 51, No. 4 (Jul. - Aug., 1991), pp. 358-363
Published by: Wiley on behalf of the American Society for Public Administration
Stable URL: http://www.jstor.org/stable/976751 .
Accessed: 14/06/2014 23:35

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.

Wiley and American Society for Public Administration are collaborating with JSTOR to digitize, preserve and
extend access to Public Administration Review.

http://www.jstor.org

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
 B O O K R E V I E WS I Patricia Editor
WIngraham,

andPublicPolicy
of thewelfarestatecame
vs.
earlyhistory
History fromexactlythesame reasoning
ford,1986). The languageof pseudo-
(Ash-

sciencecontinues withthenotation that

ofPublicPolicy
History thesubjectofthisbook is "underdeter-
mined,"a pointthatone hopes would
be the reasonfordoing comparative
analysis.Thereare otherslightly abra-
E. Ashford,
Douglas ofPittsburgh
University sive promises,but phrases such as
social scientistsare all "askingwrong
questions"(p. 7) triggers the concern
that theauthors of this book maythink
FrancisG. Castles,ed., The Comparative to social scienceis discussed. Castles
taskand it they have privileged access to truth.
HistoryofPublic Policy(New York: Oxford has by farthemostdifficult
When we learn (p. 10) that their intel-
UniversityPress, 1989) 340 pp., $45.00. is perhapsindicativeof thisdifficulty
chap- lectualtorpedo is no more than critical
thatthebook has no concluding
to
ter. The agreed-upon pointof depar- case studies,an approachtheyfail
T
note was widely discussed twenty years
tureis thataggregate nationalanalysis
1975;Lijphart, 1975),the
his is an important and accom- cannotdeal wellwithmuchpolicymak- ago (Eckstein,
plishedcomparative studythat theeditorfailsto reader's mind might understandably,
ingbehavior[though
shouldbe welcomedby histori- mistakenly, takea criti-
note thatseveralrecentcomparisons butin thiscase
callymindedsocialscientists and policy cal stancetoward what is to follow.
which startfromaggregatedata do
analysts. Whetherit will be equally
or eventaken makeimportant structuralandhistorical Thoughit will be treatedin more
welcomedby historians,
by themas a seriouschallengeto his- qualifications(See Floraand Heiden- detailbelow,theintroduction also con-
is anotherquestion.
toricalscholarship, heimer, eds., 1981;Hague,Hanneman, veysa limitedfamiliarity withhistorio-
Ifthisadmirable effortfails,itis forthe and Gargan,1989;Jallade,ed., 1988)]. graphical debatesthatmight alarmcon-
admirablereasonthatit is too ambi- Liketheoutwornargument thatvulgar ventionalhistoriansand mighteven
tious. This book setsout to compare Marxismexplainslittleabout modern offend manyhistorians who havebeen
thehistoricalexperienceof eightcoun- politicaleconomyor modernwelfare writingexcellenthistoriesof public
tries(Australia,Japan,Israel,Britain, states,thediscovery of historyalways policies and public policymaking for
West Germany,Sweden,the Nether- seemsto conveya mildstateofembar- decades (Harris,1972; Lowe, 1986;
lands,and the UnitedStates)around rassment among scientific model Lambert1963). An historiographical
twocontroversial polesofpolicyanaly- builders. Even so, it seems unfairto red flagappears(p. 11) whenwe are
sis: the developmentof social policy labeltheattempts at aggregateanalysis toldthathistory is "learning fromcatas-
and politicalpartiessincethelatenine- as tripsup the "primrose path"(p. 5), trophes."Ifhistory is to makea conge-
teenthcentury.Whilethe book con- whenyoucouldeasilyfindmanyother nial and productiveunionwithcom-
tainsnumerousself-effacing comments social scientistswho have done the parativepublicpolicy,policyanalysts
aboutthe pitfalls general-
of historical same. will need to be a bit moreimpartial
ization,thesewarnings have been suc- A secondproblemthatwillalert,if thanthis. To be sure,thereareAmeri-
cessfullyignoredby mostof the con- not alarm,policyanalystswill be the can and Frenchhistorical traditions that
tributorsand somewhat painfully brashpromise(p. 6) thatintricate his- emphasizecrisesbut theseare histori-
violatedbya fewofthecontributors. toricaland politicalquestionssuch as cal generalizations inthemselves and as
These comments mayseem unduly interestintermediation, partystructure, such are subjectto the same critical
harsh froma reviewerwho shares etc.,areto be reducedto "standardized analysisas anysocialsciencetheory.It
manyofthebasicaimsofthebook. In tests"similar to naturalscience. Social is not thatone rejectsthis implicit
some ways the substantivecontribu- sciencehas alwaysbeenmoresensitive alliancewithone school of historical
tionsare moreeasilydigestedthanis to Popper'sconcernover historicism thought,but thatthe association is
wherethe trou-
Castles'introduction, thanwiththeDarwinian menaceofsci- madewithout weighing thereasonsfor
blesomeproblemofhowto linkhistory entism.Manyofthewrongturnsinthe andagainstsuchan alliance.

358 PublicAdministration 1991,Vol.51,No.4

Review+ July/August

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
LaborMovements
and expectedlinkto policyprocesses,out- American workers weresomehowshut
comesand outputsneeds to be speci- out of the expansionof theAmerican
PublicPolicy fied. Ifthisis notdone,theshiftfrom welfare state.
A virtueof thebook is to pressstu- macro-level concerns to micro-level The difficult cases are of coursethe
dentsof publicpolicyto take certain realitiesmaybe highlyselectiveand countries(Australia,Israel,and Swe-
socialmovements moreseriously.For possiblyverymisleading (Ashford, den) where therehave been strong
thisgroupof authorsthe labormove- 1991). The case of the labor move- labormovements and weak or, in the
mentis clearlythe favorite child,but mentsis instructive.In two of the Swedishcase, delayed expansionof
forcountrieswithmore complicated examinedcountries, Britainand Ger- social programs. The macro-argument
social historiesthe persistenceof the many, strong labor movementsare seemsleastpersuasivein cases where
aristocracy, patternsof land ownership closelyalignedwithpartiesof theleft. laborleadershad good reasonsnotto
and the means of capital formation Undera strongly centralizedsystem the pressformoresocial spending. In a
arrayof social policies is complete, revealing discussionof Israel'sinternal
mightas easily enterinto historical
thoughnot overlygenerous in the religiousand social schisms,Shalev
accountsofpolicymaking.Despitethe British case. But as ManfredSchmidt shows how Labor's dominanceuntil
frequent disavowalof neo-Marxist pre- pointsout forGermany, thiscombina- 1977 dependedon institutionalizing
dispositions, the authorsclearlyshare tionof benefitshas been achievedin a
theviewthatthefailure oflabormove- Germany by placing the semi- dual market forsocialbenefits, a curi-
mentsto graspthereinsofpoweris,in ous analog to the dual labor market
autonomous bankingand fiscalinstitu- theory. The Israelilabor
some sense,a democratic failure.The union,the
tionsat the nationallevel in constant
causal inference is nevermadeentirely tensionwithan intricate Histadrut, enjoyed a monopoly of free
system of fed- healthinsurance (reminiscent of theLO
clear,noris themorebasicquestionof eralchecksand by reinforcing national being given the administrationof
whyand how democratic government aimswithan elaboratearrayof labor
might-oreven shouldembrace-the market trainingpolicies (Federal unemployment insurance in the1930s),
labormovement.The authorsare too helped the Labor government channel
Employment 1980). British workersavingsintoproductive
Institute,
intelligentto ignorethe factthathow uses,
unionshave been no less preoccupied
labor movementsalignedthemselves withtheirpolicymaking and essentially endorsed a three-class
autonomy, but system ofsocialbenefits forthesettlers,
withpoliticalparties,and specifically theirgoals remainprimarily collective theOriental
with parties of the left,is basic to Jews and the Jewishpoor.
bargaining and incomespolicy,so that
Therborn has moredifficulty explaining
understanding thepolicyoutcomes;but fewof thestabilizing institutionsfound theSwedishcase,butthere
is nowsim-
thereis, however,littleexplicitconsid- inGermany haveBritishparallels. ilar evidence that in the 1930s the
erationof how diversethesehistorical Thereare threecases whereweak Swedishleftchose fiscalstability over
experiencesactuallywere. It seems labor movementsappear associated social spending,
gave the labormove-
strange to be stilldebatingan issuethat withweak welfarestates: Japan,the mentspecialprivileges thatwere con-
many Marxistsfound perplexingin Netherlands (untilthe 1960s) and the solidatedduringthe war and delayed
1910. Therearenumerous studiessug- UnitedStates. Puttingaside the fact socialspendingafterthewar untilthe
gestingthatlaborwas neverespecially thatthe confirming cases of the more 1960s(Rothstein, 1985,1991).
enamoredby improvedsocial benefits general hypothesisare all negative
(Ashford, 1986; Baldwin,1989; Flana- (theydo notexplainwhystrong move- Thus,the"underdetermined" histori-
gan et al., 1983) even where they say mentsfail),in each of thesecases the cal analysesare readilyconverted into
theirbestinterests wereservedbymak- authorsdo a finejob revealingthe quite sensible,if limited,social science
ing allianceswithpartiesof the left. intervening variables. Pempelpoints hypotheses aboutthe relationof labor
JohnL. Lewis,ErnestBevinand Jules out that since the war Japan has to thegrowth of socialspending.The
Guesdehad a greatdeal incommon. enjoyedsteadyhighemployment, rela- problem is not thatthisis an unworthy
tivelyhighlevelsof equal servicepro- workinghypothesis but that,in their
How labor should be involvedin visionin education, health,and safety, enthusiasm forsocial democracy,we
definingthe public interesthas of and perhapsmostsurprising forthis are deprivedof thepossibility of inad-
coursebeen studiedin depthforsome capitalist convert,relative income vertentor even unintendedsocial
time. In thisrespect, muchof thepoli- equality.Therborn does a magnificent democracy.Castlesmakesno secretof
cycomparison inspired,as is thisbook, job showinghow theCatholic-Calvinist"wageearnerwelfare"sufficing where
by the new politicaleconomywriting (unholy?)allianceactuallydiminished a country can sustainfullemployment
may accomplish no more than to socialspendingin the1930sand coop- and steadygrowth(p. 21). Dunleavy
restate the ambiguous results of a erated(ratherliketheSwedishSAPand (p. 261) agreeswithnumerous histori-
decade or moreof inquiry intothecor- LO) to delayunemployment insurance cal studiesin notingthe"shallowpene-
poratistfoundationsof the capitalist untilthe1960s. Although theAmerican tration ofthelabormovement" insocial
state. Thereare mostcertainly reason labor movementsharesmanyof the policy which was, incidentally,an
whythesepolicymaking underpinnings same problems,Amentaand Skocpol important reasonwhyThatchercould
shouldbe analyzedbutin doingso the continuetheirearlierargumentthat so easilybegin to dismantleBritain's

BookReviews
359

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
"autonomous"(nationalized) social Therborn's dedication to Swedishsocial (Ashford, 1982). The neglected dimen-
policies. The brashclaimof Amenta democracymakesa similarcomplica- sions of partypoliticssuggestswhy
and Skocpol concerning the tionforSwedenawkward(he retreats generalizing frompolicydifferences to
exceptionalism"
"quintessential (p. 292) to differentialchancesfor"exit"in the the socio-economic foundationsof
of the Americanwelfarestateseems Netherlands and Swedento explaindif- modernstatesmayeasilyassumethat
less impressivewhenwe are also told ferentoutcomes), but thereis little policydifferences meanthesamething.
how Israelis delayed unemployment doubtthatthe "Red-Green" allianceof In a paradoxicalway,the authorsfall
insurancebecause the early settlers the early1930s transcended a funda- into the same logical trapthatthey
feared that the orientalJews were mentalpolitical cleavageof accuseothercomparativists
and historic and aggre-
"workshy" (Shalev,p. 115); the Dutch earlyindustrializers. Thoughnotmen- gatedata analystsof making.Method-
buried unemployment insurance tioned,the hightechnology demands ologically, thereare interesting parallels
because it threatened theirfragilesys- of Dutch agricultureand intensive betweenuncritical relianceon macro-
temof confessional politics(Therbom, farming methodsin a country reclaim- level generalizations and on national
p. 230); or thatthe institutional frag- ingitslandfromthesea mayhavecon- level accountingstatistics when inter-
mentationof labor marketpolicy tributedto a similar readinessto forego preting policymaking. The mostobvi-
makesitvirtually impossible to engage urban-rural cleavagesinDutchpolitics. ous blunderis thecontinued insistence
social policiesin Ger-
in redistributive (See Quadagno in Weiret al., 1988)
Perhapsthe moststriking omission
many(Schmidt, p. 79). The problem of thatthe federalstructure of American
in thebook's analysisof partypolitics
probinghistoryis thateverything can socialsecurity was, first,an intentional
and socialpolicyis how undemocratic
becomeexceptional.Because histori- urban-rural deviceto isolatethepoorin theAmeri-
distinctionsweredifferently
ans mostoftenstudyonlyone country, institutionalized can welfarestate and second, that
and, in turn,affected
and thenonlyone periodwithinthat somehowthe UnitedStatesis a pecu-
how partiesmightlaterinfluence social
liar "laggard"forhavinga two-level
country'shistory, theyare less likelyto
policies.Forexample,theIsraeliLabor structure fortheadministration ofsocial
translatehistoricaltruthsintosocialsci-
Party,and in a less obviousway,the policy. Schmidt's analysisof German
ence hypotheses.Thereis no reason
SwedishSAPwerealwaysvulnerable to "cooperativefederalism" (pp. 77-78)
why such an exerciseshould not be
clashesbetweenurbanand ruralinter- and Shalev'smoredevastating account
undertaken,and as Kousser (1982)
ests. Likud'svictory in 1977is in part of Israel'sclass-basedsystemof social
notes,thereluctance ofhistorians to do
the bankruptcy of a politicalsystem benefits oughtto qualifysimplegener-
thismeans thatsocial scientistsmay
that championed settlementbuton dis- alizations aboutAmerican socialsecuri-
need to revisehistory.Butin doingso
lines. Thoughdifferently ty. But even an acknowledgment
it is important thattheirreasons for criminatory of
in Britishand to a
selectingcertainhistoricalconstructions constituted Israel, agrarian interestsand thecomplexity of
lesser extent the United Statesconser- partisanpoliticsstillrequirescareful
are clearand theiruse of historical evi-
vativeshad theirrevengein the 1980s. interpretation atthepolicymaking level.
dence(mostofwhichtheydo notgath-
The failureoftheSwedishcompromise The political outcry over Lloyd
er themselves)is impartial.
loomedlargeinSwedishvotingin 1976 George's"People'sBudget"in 1909and
butwas onlyrecognized by socialsci- the ensuingconstitutional crisiswith
Partiesand ence around 1984 (Martin, 1984). theHouse of Lordswerenottheprod-
PublicPolicy ThoughPempeldoes notmentionthe uctofToryconservatism in themodern
alliancebetweentheJapaneseLiberal sensebutthelastgaspof an aristocrat-
Iftheneo-Marxist assumption about Partyand the farmers,the policies ic, land owningclass withinthe Con-
labor movements seems shaky, it towardrice and beef productionare Party whowereoutraged over
should also be recognized thatthe well known. WhileDunleavyis per- servative
control and theminor landtax included in thebud-
directlinkbetweenparty fectlyawareof the"conservative hege- get in orderto financeBritain'sfirst
socialpolicyand socialexpenditures is mony"withinBritish politicshe failsto
no less so. In the threecountriesof socialinsurance programs.
note thatthiseffecthas been recog-
thissample withgovernments of the nized forsome time(Beattie,1974) This same predispositionto treat
leftover extendedperiods,the out- and,moreimportant, thatit is derived partisan alignments as politicallyequiv-
comes divergedbeyondeasy general- from,if no longerdependenton, the alent over timeleads the authorsto
ization. Shalev'sanalysisof Israelis a landedaristocracy and itsmonumental neglecttransformations in the political
tellingindictment of thefailings of the to land taxesoverthenine- guidance of socialpolicy. Becausethey
resistance
IsraeliMapai,foundedin 1930 as the teenth century. tend to assume that the political leftis
politicalinstrument of the Histadrut, theonlysalientinnovator in socialpoli-
whichwas itselffoundedin 1920 to Differences in theinstitutionalizationcymaking, theauthors obfuscate one of
defendJewish settlers in rural areas. of urban-rural differences,and the the mostinteresting partisandimen-
The fusionof Zionistand socialistide- closelyalliedproblem ofterritorial
poli- sions of modernwelfarestates.The
als (Shalev,p. 109) in a ruralsetting tics,had important implicationsforthe expansionof national-level socialpoli-
makesIsraelan odd comparison to the construction of welfarestatesand per- cymakingwas not alwayscaused by
industrialstates of the prewarera. sistenteffectson social policymaking popularacceptanceof socialgoals. As

360 Review
PublicAdministration 1991,Vol.51,No.4
* July/August

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
we have seen in thisand otherbooks, Thoughtheauthorsnevertreattheir historical omissions.Who could com-
therearemajordifferences in howgov- partisaninterpretations in a compara- press eightnationalhistoriesand a
ernments movedfromone policyarea tive way, the mix is inconclusivein multitude ofsocialpoliciesintoa single
to another,a pointskillfully made by bothabsoluteand relative terms.Israel volume?Floraand hiscolleaguesneed-
Dunleavy(p. 255) in his discussionof and Sweden,withearlystronggovern- ed fourheftyvolumesjustto give us
the"ladderofintervention," (Whyother mentsof theleft,wentoppositedirec- the bare outlines(Flora,1986; 1987).
statesdidnothavepersonssimilar to T. tions. In Australiaand Germany, Nor are myreservations reallyabout
H. Marshalland RichardTitmusswho where governments of the leftwere thefaulty logicoflookingat policyfor-
helped formulate thisBritishtheoryis latein takingpower,bothgovernments mationthroughthe lenses of macro-
one of the fascinating questionsthe were alreadyencasedin social institu- economic and macro-socialchange,
authorsignore.) Thoughit tookvery tionsthatthe lefthad greatdifficulty even thoughthisroutebringsthese
differentformand oftenhad verydif- reshaping(Michalsky, 1985; Swenson, authorsperilously close to thelogicof
ferentmeaning, a less recognizedcom- 1989). In Japan,the lefthas never statistical inference whichtheyreject.
mon partisanelementwas thedecline come to power,thoughthereis inter- The problemis thattheynevercon-
ofclassicalliberalthoughtand thecon- estingevidencethatit does influence front theempirical possibility thatpoli-
fusionofthetraditional,liberalparties. intergovernmental social policymaking cymakingthroughtimeis an activity
UnderGladstone,the late Victorian (Muramatsu and Nakamura in Ashford, thatis bothobjectively and subjectively
LiberalPartybroughtthe latentdivi- 1991). In Britain,the leftfailedfor different fromhypothetical modelsof
sions of radicaland traditional Whigs manyreasonswhichDunleavysumma- social and economic change. That
intotheopenwiththeIrishLandActof rizes(pp. 263-264)as therefusal of the thereshouldbe a connection is hardly
1881,the firstinroadintoaristocratic TUC and CBI to become enthusiastic surprising. Socialscientists themselves
preservesand an earlyindication that corporatists. The politicalomissionis live throughtime.Describingpolicy-
the Poor Law responsibilities of local notthattheydidnotwantto do so, but makinghistorically and applyingsocial
governmentwere inadequatelyper- thattheyneverneededto do so. The hypotheses through time,however,are
formedbecause of the limitson local real concentration of economicpower fundamentally different endeavors.
taxation (Dangerfield,1961; Offer, in Britainwas alwaysthe city,which The immensedangerof focusingon
1981). In Germany,Bismarckused guardeditspoliticalboundaries justas the "underdetermination" of historical
socialreform to furtherisolatetheliber- skillfully as Germany'sBundesbank analysisis just that;historybecomes
als who never regained theirearly manipulated theambiguities ofGerman whatever we wishitto be.
influencein Germanpolitics. French, policymaking. As fortheunions,Bevin My pointis not thattheseauthors
Dutchand Germanliberalswere split refusedto joinKeynesand othersin a are in some sense irresponsible.
intoconfessional and secularelements. prestigious, thoughprobablyimpotent, Schmidt's oftheGermanSon-
summary
Swedish liberalsneverprosperedin Councilof EconomicAdvisorsin the
derwegis a modelof historical summa-
part because the early "Red-Green" depression. He neverconcealed his tion.
Therborn's account of the
alliancedeprivedurbanproperty own- disdainofpseudo-theorists oftheLabor Catholic-Calvinist alliancein Dutchpol-
ersof agrarian support.The successful Party,such as Herbert Morrison, or of icymaking is masterful,
even ifhis his-
advanceof social democracyand the intellectualmodel-buildersof new torical account of Swedish social
concentration of capitalmadetheCon- socialorders,suchas ErnestBeveridge.
democracyis, perhaps necessarily,
servativesthe obvious,if ineffective, The dismayof Amenta,Skocpol and
abbreviated.Castle'simpatience with
politicalopposition (Nilssonin Koblick, otherswiththe "failure" of the leftin the willfulliberalismof Australian
ed., 1985). Althoughthe crudecate- Americamay be no morethantheir
workers is similar to Amenta and
goriesof "left"and "right" maymake own refusal to acceptan historical situ- Skocpol's
reluctantrecognitionthat
some statisticalsense at the aggregate ationwhereliberalpartieshavedivided
American workers usuallyvotedto the
level,as partisandispositions are treat- socialchores. The American history of rightuntil1936(theydid notturnout
ed in this book, the policymaking partypoliticsand social policyis no for Roosevelt in large numbersin
nuancesof theleftand rightare much morebaffling thanthatof mostdemoc- 1932).
Pempelmaybe a bit conde-
more complex.Indeed, it is perhaps racies.
scendingin suggesting thatwe should
onlyin thisqualifiedsense,rather than notthinkofToyotamagnates as samu-
in the simplesocial policysense,that HistoricalContextand rai (p. 149),buthe has a balancedgrip
the UnitedStatesis as exceptionalas
Skocpol and her colleagues would PoliticalChoice on postwarJapanesehistory.Shalev
actuallyprovidessome social policy-
have us think.Throughhistory, both If thisreviewhas been overlycriti- making historyofIsraelthatis probably
Republicansand Democratsembody cal, itis becausethehistorical interpre- unknown to many Europeans and
someelementsof socialreform, butin tationofpoliticaland socialpolicymak- Americans.None hide the historical
an unwieldly setting of federalpolitics, ing is an importantstep toward contradictions of theirsocial science
the curiositiesof agrarianradicalism, enlargingthe vision of comparative models. The distortion is not factual
and thesheerscale of socialchangein policystudies. The doubtsexpressed but analytical.Moreover, it is analyti-
theUnitedStates. overthisbook are notso muchforits callybiasedin a veryparticular sense.

BookReviews 361

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
 In additionto such peculiarly pro- methodologicaldisputeis centralin ing a debateabouttheparameters of
fessionalcontroversies over whether decidinghow to relatepolicymaking therevolution, notabouttheobjective
historiansshould be writingsocial, behaviorto politics,sociologyor eco- meaning oftherevolution. Theontolo-
political,or diplomatic history,histori- nomics. Can the policy process be gy of normalsocial science provides
ans have theirown methodological directly translated intomacro-level con- the"object"of inquiry in a waythatis
concernsabout readinghistoryback- cepts?Another wayof stating thisis to impossibleto do in historicalinterpre-
wards,demarcating timeperiods,and ask whetheran abstractstatement tationor, moreprecisely, can onlybe
clarifyingtheirhistorical interpretationsaboutpolicymaking builton macro-the- by treating
done historically social sci-
withwhatsocial sciencecalls "reitera- oreticaldesignsof anyparticular social ence hypotheses as "real"forsomesit-
tion" and historycalls narrative.A science may not be a reificationof uationor some momentin time. It
wonderful illustrationis theformidable "policymaking" (Ashford, ed., 1991). Is does notseemcoincidental thatan his-
teamof radicalEnglishhistorians who whatwe call "policymaking" onlythe torianconcerned withcomparison enti-
spentmostoftheirlivesreworking crit- allocationof resources,the balancing tledhisbooktheChanging Boundaries
ical periodsin British social historyto of inflationand employment, theplay- ofthePolitical(Maier,ed., 1987).
tryto showthatMarxist ideas are con- ingof legislative games,thereconcilia-
For the same reason,in the policy
sistent throughtime (Himmelfarb, tionof class differences, the need for
1985). Thereis a sense in whichall political
consensus, etc.,or is ita pecu- and politicsserieseditedby Katzen-
historiansare archivaland so historians liarformof behaviorin boththesocial stein,Pempel,and myself the"contex-
mustmaketheirideologicaland evalu- science and historicalsenses? There tual"sectionprovideda themewhich
ative judgmentsexplicit. If theydo are good reasonsto thinkthatthecae- each authorused to clarifyhistorical
not,theyarelikelyto fallintowhatwas terisparibusassumption, controversial and situationalelementsof the policy
called "Whiggishhistory"(Ashford, even forordinaryhypothesesabout process. We neversatisfied ourselves
1989). Again,the pointis not thata humanbehavior(Moon,1975),cannot that "contexts"mightbe compared
discussionabout historicalsociology be indiscriminately applied,inadver- acrosscountries.Forsomecountries, it
(Skocpol,ed., 1984) cannotbe useful, tently as seemsthecase withthisstudy, evenproveddifficult to specify context
but thatsocio-economicanalysesof incomparing policymaking histories. throughtime. The Sonderweg,for
historicalanalysesmaylead to distor- example,suggestsa rangeand variety
The epistemological foundations of
tions and even logical reifications. thisargument ofbehavior thatis farmorebewildering
are compelling.When
Examplesfromthebook underexami- socialscientists thantheprimacy ofparliament in mod-
comparesocialclasses,
nation are, first,the deceptiveease em British politicalhistory.Thus,the
partisanidentitiesand riskbehavior,
withwhichthe particular historiesof to situationswith
effortof theseauthorsto specifythe
theyare referring
the leftbecomegeneralized comments known historicalcontextof policymaking is
parameters.It is, of course,
about the "left"and, second,the ten- possible to arguethatbecause these admirable.Butifcontexts are unique,
dencyto treatsuch metamorphasized parameters are arbitrarilydefinedeven thefitwithsocio-economic hypotheses
historical
generalizations likesocialsci- normalsocialscienceis a reification of needs explicittreatment. Good social
encevariables. reality. However,comparingpolicy- scienceand good history are hardto
The specificationof variablesby makingbehavioracrosscountries and mix. Untilwe knowhow to do this,
socialscientistsis an essentialstepin a through timeshatters theassumption of comparative historical analysisof poli-
particular formof reasoning.Castle's commonparameters.An interesting cymaking willremainhistory and poli-
recognition (p. 10) of Przeworski's cri- historicalexampleis the historical dis- cy ratherthanhistory ofpolicy.In dis-
tique of statisticalinference raisesthe puteovertheoriginsof theToryparty cussingthesamepointsomeyearsago,
same methodological point. Actually, (Clark,1985;Innes,1987). The dispute Stedman Jones(1976,p. 295),himself a
along withTeune (1971), Przeworski is notabouttheconsequencesof con- sociologistof the left,thoughthistory
(1985)once strongly arguedthereverse servative behavior, buthow "conserva- too importantto leaveto sociology.In
and his majorconcernsincehas been tive"intentions and motivesrelatedto seekinga "shortcut to theoreticalsalva-
to refinetheconceptof class so thatit the interpretation of theEnglishRevo- tion,"we may,as he noted,get the
will not fallintothe same trap. The lutionof 1688. The historians are hav- worstofboth.

References
and Con-
DouglasE., ed., 1991.History
Ashford, 7 1982. BritishDogmatismand French Beattie,A. J., 1974. "The Two-PartySystem:
textin ComparativePublic Policy. Pittsburgh: Pragmatism: Center-Local Relations in the RoomforSceptism?" In S. Finer,eds.,Adver-
ofPittsburgh
University Press. State.Londonand Boston:Allen&
Welfare sary Politics and Electoral Reform.London:
of
, 1989. "The WhigInterpretation Unwin. Anthony Wigram.
the Welfare State."Journal of Policy History, Baldwin, Peter, 1989. The Politics of Social Soli- Clark, J. C. D., 1985. English Society-1688-
vol. 1 pp. 24-43. darity: Class Bases of the European Welfare 1832; Ideology,Social Structureand Political
, 1986. The Emergenceof the Welfare State, 1874-1975. Cambridge: Cambridge Practice During the Ancient Regime. Cam-
States.Oxfordand NewYork:Blackwell. Press.
University Press.
University
bridge:Cambridge

362 PublicAdministration 1991,Vol.51,No.4

Review. July/August

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions
Dangerfield, George,1961. TheStrangeDeath of Jallade, Jean-Pierre, ed., 1988. TheCrisisofRedis- ence,1750-1970. Minneapolis: University of
LiberalEngland.NewYork:Capricorn. tributionin Welfare States. Stoke-on-Trent: Minnesota Press.
Eckstein,Harry,1975."Case Studyand Theoryin Trentham Books. Offer,Avner,1981. Property and Politics,1870-
PoliticalScience"in F. Greenstein and N. Pols- Kouser,J. Morgan,1982. "Restoring Politicsto 1914. Cambridge: Cambridge University
Press.
by,eds., HandbookofPoliticalScience,vol. 7 PoliticalHistory." JournalofInterdisciplinaryPrzeworski, Adam,1985. Capitalismand Social
Reading,PA:Addison-Wesley Press. History, vol. 12,pp. 569-595. Democracy. Cambridge: Cambridge University
FederalEmployment Institute,1980. EmploymentLambert,Rayston,1963. SirJohnSimon,1816- Press.
Policyin Germany-Challenges and Concepts 1904,and English SocialAdministration. Lon- Quadagno,Jill,1988. "FromOld-AgeAssistance to
for the 1980s. Nurnberg:Bundesanstaltfur don:MacGibbon andKee. Supplemental Security Income: The Political
Arbeit. Lijphart, Arendt,1975. "The ComparableCase Economyof Reliefin theSouth,1935-1972" in
Flanagan,RobertJ.,David W. Soskice,and Lloyd Strategy in Comparative Research."Compara- M. Weir,A. Orlott,and T. Skorpol,eds., The
Ulman,1983. Unionism, EconomicStabiliza- tivePoliticalStudies,
vol.8, pp. 158-177. PoliticsofSocial Policyin the UnitedStates.
tion,and IncomePolicies: EuropeanExperi- Lowe,Rodney,1986. Adjusting toAuthority: The Princeton: PrincetonUniversityPress.
ences.Washington: The Brookings Institution. RoleoftheMinistry ofLabourin British Poli- Rothstein,Bo, 1991. "Marxismand Institutional
Flora,Peter,1986,1987. Growth to Limits:The tics,1916-1939.Oxford: Clarendon Press. Analysis: Working Class Strengthand Welfare
Western EuropeanWelfare StatesSince World Maier,CharlesS., 1987. ChangingBoundariesof Statein Sweden"in D. Ashford, ed., History
WarII, vol. 1 and vol. 2 (1986);vol. 4 (1987), thePolitical.Cambridge: Cambridge Universi- and Contextin ComparativePublic Policy.
Berlinand NewYork:Walterde Gruyter. tyPress. Pittsburgh,UniversityofPittsburghPress.
Flora,Peterand ArnoldHeidenheimer, eds., 1981. Martin,Andrew,1984. "TradeUnionsin Sweden: 1985. "Managing the WelfareState:
TheDevelopment Welfare Statesin Europeand StrategicResponsesto Changeand Crisis"in P. LessonsFromGustavM6ller."Scandinavian
America. New Brunswick,NJ:Transaction Gourevitch, etal., Unionsand EconomicCrisis, PoliticalStudies,
vol.8, pp. 151-170.
Books. London:Allen& Unwin. Skocpol,Theda,ed., 1984. Visionand Methodin
Hage,Jerald,RobertHanneman,and EdwardT. Michalsky, Helga, 1985. "The Politicsof Social HistoricalSociology.Cambridge: Cambridge
Gargan,1989. StateResponsiveness and State Policy"in K. von Beymeand M.G. Schmidt, UniversityPress.
Activitism. London:Allen& Unwin. eds.,Politics and Policyin theFederalRepublic StedmanJones,Gareth,1976. "FromHistorical
Jose,1972. Unemployment
Harris, and Politics:A ofGermany.NewYork:St.Martin's Press. Sociologyto Theoretical History."BritishJour-
StudyofEnglishSocial Policy1886 - 1914. Moon, Donald, 1975. "The Logic of Political nal ofSociology,vol.27,pp. 295-305.
Oxford:Clarendon Press. Inquiry:A Synthesis ofOpposedPerspectives" Swenson,Peter,1989. Fair Shares: Unions, Pay,
Himmelfarb, Gertrude,1985. TheIdea ofHistory: in F. Greenand N. Polsby,eds.,Handbookof and Politicsin Swedenand Germany. Ithaca:
Englandand theEarlyIndustrialAge. New PoliticalScience,vol. 1, Reading,PA:Addison- CornellUniversity Press.
York:Knopf. WesleyPress,pp. 131-228. Teune,Henryand AdamPrzeworski, 1971. The
Innes,Joanne,1987."Jonathan Clark,SocialHisto- NilssonGoranB., 1985. "SwedishLiberalism in Logic of ComparativeSocial Inquiry. New
ryand England's'AncientRegime."Past and Mid-Nineteenth Century" in S. Koblick,ed., York: Wiley.
Present,vol. 115,pp. 165-200. Sweden'sDevelopment fromPoverty to Influ-

Comparative ativestudythematically
a country-by-country
ratherthanon em Europe:1660-1930. The chapters
basis. Insteadof bore such titlesas "Taxation," "Social
Government
and individualchapterson France,Italy, Services,"
and so forth,
and so forth.In
"Education,"
his chapterheadingsfol- each of these chaptersBarkertraced
Administration low the patternone usuallyfindsin thehistorical
development of thetopic
books devotedto one countryalone: in questionin Prussia,England,and
byjohn A. Rohr politicalparties,
interestgroups,courts, France. AlthoughBarker'sover-all
VirginiaPolytechnicInstituteand State the legislature,the bureaucracy, etc. approachwas thematic,neverthelesshe
University Happily,the book's mostdistinctive examinedeach themeon a country-by-
featureis also its greateststrength. country basis. Forexample,Barker's
Yves MWny, Government and Politicsin Menyhas a masterful graspoftheform chapteron educationwas subdivided
Western Europe:Britain,France,Italy,and and spiritof governmental institutions. intothreesectionsdealing,predictably
WestGermany, translated
byJanetLloyd Deftlyhe movesback and forth across enough,withFrance,Prussia,and Eng-
(NewYork: OxfordUniversity Press,1990), the institutional
landscapeof Western land. ThusBarker's thematicapproach
367 pp.; $55.00hardcover;paperbackedi- Europe,offering thereadera judicious ultimately collapsedintoa country-by-
tionavailable. blendof structure and function on the country analysis.
one handand of panoramaand detail
Notso withMWny.His chapteron
on theother.Becauseof thismethod-
Y cal Science at the
ves Meny is Professorof Politi-
Institut
d'Etudes Politiques de Paris.
ologicalinnovation,
parativegovernment
"Localand CentralBureaucracies,"
studentsof com- example,is dividedinto
and administration "Administrative Structures,"
for
sectionson
"The Civil
will ignorethis book at theirperil. Service,"and
He has taughtin the UnitedStatesand "Controlling
the Bureau-
in Italy. Meny'sapproachis notentirely without crats."Eachofthesesectionsis further
precedent. During World War II, subdivided;
thus,"AdministrativeStruc-
The most distinctivefeatureof this ErnestBarkerwrotea book titledThe tures"includes"Central Bureaucracies,"
book is thatMeny groundshis compar- Development ofPublicServices in West- "Peripheral Authorities,"
"Specialized

BookReviews
363

This content downloaded from 185.44.78.143 on Sat, 14 Jun 2014 23:35:54 PM

All use subject to JSTOR Terms and Conditions