Findley Economy Revolution

8/13/2019 Findley Economy Revolution

1/27

Economic Bases of Revolution and Repression in the Late Ottoman EmpireAuthor(s): Carter Vaughn FindleySource: Comparative Studies in Society and History, Vol. 28, No. 1 (Jan., 1986), pp. 81-106Published by: Cambridge University PressStable URL: http://www.jstor.org/stable/178681

Accessed: 29/03/2010 10:15

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 [email protected].

Cambridge University Pressis collaborating with JSTOR to digitize, preserve and extend access to

Comparative Studies in Society and History.

http://www.jstor.org
http://www.jstor.org/stable/178681?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=cuphttp://www.jstor.org/action/showPublisher?publisherCode=cuphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/178681?origin=JSTOR-pdf


2/27

Economic Bases of Revolution andRepressionin the Late OttomanEmpireCARTER VAUGHN FINDLEYOhio State University

Central o late Ottomanhistoryis a series of events that marksa milestone inthe emergenceof modem formsof political thoughtandrevolutionaryactionin the Islamic world. The sequence opened with the rise of the Young Ot-toman ideologues (1865) and the constitutional movement of the 1870s. Itcontinued with the repressionof these forces under Abdiilhamid11(1876-1909). It culminated with the resurgenceof opposition in the Young Turkmovement of 1889 and later, and especially with the revolutionof 1908.Studied so farmostly in politicaland intellectual erms,1the sequenceseemswell understood.The emergence of the Young Ottomans-the pioneers ofpoliticalideology, in anymodemsense, in the MiddleEast-appears to resultfromthe introductionof Western ideas and from stresses created within thebureaucracy y the political hegemonyof the Tanzimatelite (ca. 1839-71).2The repressionunderAbdiilhamid ollows fromthe turmoilof the late 1870s,the weaknessesof the constitutionof 1876, andthe craft of the new sultan increatinga palace-dominatedpolice state. The emergenceof the YoungTurksshows that terrorultimatelyfostered, rather hankilled, the opposition.Too,their eventualrevolutionary uccess shows how muchmore effective thanthe

I ErnestE. Ramsaur,Jr., The YoungTurks: Prelude to the Revolutionof 1908 (Princeton,1957);BernardLewis, TheEmergenceof ModernTurkey,2d ed. (London, 1968), 150-74, 194-230; Serif Mardin,The Genesis of YoungOttomanThought Princeton, 1962); idem,Jon Tiirk-lerin Siyasi Fikirleri [Young Turkpolitical ideas] (Ankara, 1964); idem, LibertarianMove-ments in the OttomanEmpire, 1878-1895, Middle East Journal, 16:2 (1962), 169-82; idem,Power, Civil Society, and Culture n the OttomanEmpire, ComparativeStudies nSocietyandHistory, 11:3 (1969), 258-81; Feroz Ahmad, The YoungTurks:The Committeeof Union andProgress in TurkishPolitics, 1908-1914 (Oxford, 1969); StanfordJ. Shaw andEzel K. Shaw,History of the OttomanEmpireand Modern Turkey(Cambridge, 1977), II, 263-67, 273ff.;CarterVaughnFindley, The Advent of Ideology in the Islamic Middle East, PartII, StudiaIslamica, 56 (1982), 147-66; Donald Quataert, The 1908 Young Turk Revolution: Old andNew Approaches, Middle East StudiesAssociationBulletin, 13:1 (1979), 22-29.2 Mardin,Genesis, 121-32; idem, Power, Civil Society, andCulture, 277; idem, SuperWesternization n Urban Life in the Ottoman Empire in the Last Quarterof the NineteenthCentury, in Turkey:Geographicand Political Perspectives, Peter Benedict et al., eds. (Leiden,1974), 403-46.0010-4175/86/1481-2342 $2.50 ? 1986 Society for ComparativeStudyof Society andHistory


3/27

82 CARTER VAUGHN FINDLEY

Young Ottomansthey were as political mobilizers.3 Finally, internationalpolitical forces played a part, as indicatedby the role of refugees from theRussianEmpire n thedevelopmentof Turkishnationalism,andby the excite-ment that the Russo-JapaneseWar and the Russianrevolutionof 1905 rousedin the OttomanEmpire,as elsewhere in Asia.4No doubt, this political-intellectual nterpretationovers manyof the mostimportantpoints. Yet, it remains to ask whetheranalysisof additional ink-ages between the sequence-ferment-repression-revolution-and its histor-ical context would not add significantlyto understanding f the sequence.This essay answers the questionpositivelyby showingthat the sequencewaslinked to economic, as well as intellectual and political, developments.5Infact, as 1908approached, he economic situationevolved into a variationon awell-knowntheorythat seeks the originsof revolution n a sharpreversalfollowing a prolonged period . . . of economic and social development. 6Evidence for the interpretation ffered here emerges from comparisonoftwo sets of quantitativedata: he salariesrecorded n thepersonneldossiers ofthe OttomanForeign Ministry,andthe commodity prices publishedin Istan-bul newspapersof the period 1851-1914. The methodof analysis is to pro-duce time series, of salarystatisticsin the one case, andcommodityprices inthe other,andthen, by comparing hese series, to arriveat conclusionsaboutchangesovertime in theeconomicpositionof a key sector of the bureaucraticintelligentsia.7

3 Since I have treatedproblems of ideology and political mobilization in anotherstudy,discussionof these topicsherewill be schematic. Interested eaders houldsee Findley, Adventof Ideology, Part II.4 NikkiKeddie, ReligionandIrreligion n EarlyIranianNationalism, ComparativeStudiesin Societyand History, 4:3 (1962), 265; L. S. Stavrianos,GlobalRift(New York, 1981), 388-90; David Kushner,The Rise of TurkishNationalism(London, 1977), 10-14.5 A similar argumentfor the years right around 1908 appearsin Donald Quataert, TheEconomicClimateof the 'YoungTurkRevolution'in 1908, Journalof ModernHistory, 51:3(1979), D1147-D 1161 (availablefromUniversityMicrofilms,Ann Arbor,Michigan,orderno.IJ-00049). See also idem, Commercialization f Agriculturen OttomanTurkey, 1800-1914,InternationalJournal of TurkishStudies, 1:2 (1980), 52-53.6 James C. Davies, Towarda Theory of Revolution, in WhenMen Revolt-and Why,James C. Davies, ed. (New York, 1971), 134-47. Some scholarsquestionwhetherthe termrevolutionis appropriateor the Young Turkcase of 1908. The view taken here is that it ismeaningful o speakof revolutionaryransformationf a political system-a fundamental,vio-lent, restructuring f the politicalgame-as distinctfrom more drasticrevolutions hat transformsocioeconomicrelations,andperhapsculture,as well. Revolutionaryransformationf a polityisalso distinguishable rom the less drasticcoup d'6tat, an irregularandusuallyviolent changeinthe identityof those who wield power, without necessarily any restructuring f the politicalprocess. The view of political revolutiontakenhere is congruentwith the ideas of Davies, andother theoristsof revolution. Consideringwhat the OttomanEmpirewas like before 1908, astrongcase can be madethat the Ottomanexperienceof 1908 was a politicalrevolutionas heredefined.7 Forassistance n this research,I am indebted o the late WilfordL'Esperance,and to CharlesIssawi, Mehmet Genq, Andreas Tietze, Russell Major, David Landes, Metin Heper, YilmazEsmer, LarsSandberg,DonaldQuataert, evket Pamuk,FerozAhmad,JustinMcCarthy,TomWhitney, andJim Wagner.


4/27

ECONOMIC BASES OF OTTOMAN REVOLUTION 83

While bureaucraticsalaries, particularlythose of a single agency, mayseem to have littlebearingon the subject ndicated n the title of thisessay, theForeign Ministry-aside from being the one Ottomangovernment agencywhose personnelrecordsare available to researchersas a discretecorpusofmanageablesize-was particularly mportantin the rise of the modernistintelligentsia.This was especially true in the generationof the Young Ot-tomans,almost all of whose leaderswere associatedwith this ministry.Whileless strongin the generationof the Young Turks, whose leaders came frombothmilitaryandcivil services, the link to the civil bureaucracywas a signifi-cantfactof thatperiod, too.8 The economic fortunesof less elite segmentsofOttoman ociety arenow becomingbetterknown,9and we shall make at leastsome comparative omments aboutthemin this discussion. Given limitedbutgrowingpoliticalmobilization n thisperiod,however, theeconomic fortunesof the elites surelyhad more to do with the originsof majorpolitical move-ments than did those of the masses, however important he latter were inresponding o opposition appeals.I. FOREIGN MINISTRY SALARIESIn 1877, the Ottomangovernmentbegan to keep official personnelrecords,includingregularmentionof salarychanges.Data collected from all recovera-ble files of career officials of the Foreign Ministryprovide the basis for astudyof bureaucratic alariesthroughout oughlythe secondhalf of the nine-teenthcentury. Analysis of the salariespresents problemspertaining o therepresentativeness f the data, the monetaryunit of payment, the relation ofthe salary figures in the records to the actual receipts of the officials, andvariationsover the years in the numbersand seniority of the officials forwhomthereare data. Once the observationsandadjustmentshatcan be madein responseto these problems,as set forth in the appendix,have been carriedout, we arein positionto opendiscussion of the salarystructureby presentingseries of adjustedmeans and medians. These figures are shown in Table 1,and are presented graphicallyin Figure 1. Considerationof the means and

8 Phrases ikedbureaucratic ntelligentsiaarejustifiedin speakingof the Ottomanelites in thesense that,historically, governmentservice was the predominant,almost the exclusive, way forintellectuals to earn their livelihoods. One of the best-chartedthemes of nineteenth-centuryOttomanhistoryis the link between reformand the creationof a Western-oriented modernistsegment within the bureaucraticntelligentsia. In the civil bureaucracy, he Foreign Ministryplayedthekey role in shapingthe modernist eadership hat dominated hegovernmentduring heTanzimat. The rise of the Young Ottomansin the 1860s representsthe emergence from themodernist ntelligentsia of a movementopposingthe leadingTanzimat tatesmen. About thesametime, bureaucratic ndliterarycareerswere also beginningto differentiate,a process linkedespeciallyto the rise of journalism Mardin,Genesis, 124-27; idem,Jon Turklerin,94 etpassim;Lewis, Emergence, 88-89, 147-50; CarterVaughn Findley, BureaucraticReformin the Ot-tomanEmpire(Princeton, 1980), 126-40, 209-17; idem, Ideology, PartII, 151-52).9 KorkutBoratav,A. G. Okgiin,and S. Pamuk, OttomanWages and the WorldEconomy,1839-1913, Review (publishedby the FernandBraudelCenter,StateUniversityof New York,Binghamton),forthcoming.


5/27


TABLE 1Monthly Salaries of Ottoman Foreign Ministry Officials Serving inIstanbul, Adjusted Means and Medians

(gold kurus per month)Means Medians Means Medians

186318641865186618671868186918701871187218731874187518761877187818791880188118821883188418851886

5,5364,1372,6912,6681,6552,6742,6002,9603,3863,3312,1482,7732,6802,2162,4611,9171,8871,6661,7531,8441,9161,9912,0702,151

4,5503,1001,5003,0501,5503,0173,0251,1501,1501,0441,0501,3391,0351,0111,2181,1741,2121,2151,2161,2011,2311,242

1887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911

2,0241,8891,7401,6681,6041,5221,5111,6041,5341,5991,7161,6861,7231,8211,9321,9101,9471,9641,9601,4731,5309661,4141,177

1,2561,2551,2311,2131,1881,1811,1441,1441,0391,1531,0621,1931,2011,2301,1691,2421,2181,2781,2181,3151,3471,2661,2121,169925

SOURCE:Salarystatisticsare computedfrom the salarynotationsin the personnelfiles of 366officials of the OttomanForeign Ministry.The sources andprocedures or the computationsareexplainedmorefully in the appendix.

medians leads on to analysis of long-term change in the salary distribution,and to an initial consideration of what the salaries shown in the table meant interms of living standards.

Perhaps the most conspicuous point in these salary statistics is that themeans are almost invariably higher than the medians. The greater the gap, thestronger the indication that the salary distribution was inegalitarian, withmany low salaries and a few higher ones. The greater the inequality, thegreater the extent to which the median-the midmost salary when all salariesare ranked by amount-excels the mean as an indicator of the fortunes ofmost members of the group studied. The mean-median gap does narrow over


6/27


10,000

C tMedan

1,000

Median

1860 1870 1880 1890 1900 1910FIGURE. MonthlySalariesof OttomanForeign MinistryOfficialsServingin Istanbul,AdjustedMeans and Medians

time, since themeans tend to fall, while the mediansremainnearlyon a level.The exceptionally high values in both curves for the early years may beanomalies due to computationfrom small numbers of cases. Or they maysignalthat salarieshad been muchhigherduringtheTanzimat ca. 1839-71),a political period when the civil bureaucracydominated the polity as neveragain.The ideal way to learn more about the salary distribution would be tocomputestatistical measures of dispersion,but this is not possible with themanipulateddata distributionthat emerges from the controls for seniority.One way to get around this problem, and incidentallyto learnmore aboutsalaries n the precedingperiod,is to compare he salaryof thehighestofficialof the ministry,the foreignminister,with the lowest salarypaidthere. SinceOttoman officials normally began their careers as unpaid apprentices,ver-ificationof the lowest salarypresentsno problem.Therewerealwaysofficialswitha salaryof zero. The salariesof the ministers, however, varied over timein ways that reinforce the implicationsof the mean-mediangap shown inFigure 1. Because indications of these salariesappearboth in the personnelrecordsand in othersources, extendingbackto the 1830s, the available dataon this pointcover a much longer spanof time than the statistics in Table 1.Examiningnotationsof the amountsof foreignministers'salaries,we findthat, of eight mentions for 1838-76, seven were in the rangeof 60,000 to75,000 kuru?per month;and the one anomaly, in 1872, was 50,000-still


7/27

86 CARTER VAUGHN FINDLEYmorethanfortytimes the median for thatyear, shown in Table 1.10Recentstudies indicate that an Ottoman laborer would have been fortunate o earn250 kurusper month in the early 1870s; thus the dimensions of Ottomanbureaucratic litism emerge clearly from these figures. IThe militaryandfiscal crises of the late 1870s lowered the minister'ssalaryto around40,000kurusper month.12For 1880-84, it fell again, by half, to an all-timelow of20,000 kuru. 13In 1885-95, the figurewas again30,000 permonth.14 From1896 through he Young Turkrevolution,the salarywaveredin the rangeof36-46,000 kuru?.In the purgesand salarycuts that followed the revolution,the minister'ssalaryfell to 25,000.15 Thatwas still abouttwenty times theadjustedmedianForeign Ministry salary shown in Table 1, while the suc-cessor of the lucky laborerwho earned 250 kuru?per monthin 1872 wouldperhapshave earned 350 by that date.16 World War I broughta cut of 50percent o all bureaucratic alaries. By 1916, inflation forceda restorationofthe amountcut and even the additionof cost-of-living supplements,althoughthese proved inadequate. 7Since the highest salary indicates the rangeof salaries, the lowest salary

10 Salaries mentionedfor 1838-76 (and data sources)are:75,000 kurusper month(Haus-,Hof-, and Staatsarchiv,Vienna, TurkeiVI/67, 18 April 1838);65,000 (BasbakanlikArchives,Istanbul(cited as BBA), MaliyedenMiidevver(cited as MM) 11738, p. 11, entryof 17 Safer1256/20 April 1840); 75,000 (BBA, Dahiliyesicill-i ahvaldefterleri citedas DSA)II, 218, entryof 1273/1856-57); 61,455 (BBA, MM 10531, p. 20, entry of 27 Mart 1277/8 April 1861);75,000 (BBA, MM 10529, pp. 11, 22, entries of 26-27 Mart and 10 Nisan 1279/7-8 and 22April 1863);50,000 (BBA, DSA II, 416, entryof 2 Cemaziyilahir1289/7 August 1872);75,000(BBA, DSA XXII, 37, entryof I Rebiyulahir1291/18 May 1874);75,000 (BBA, DSA II, 416,entryof 8 Zilhicce 1291/16 January1875, incumbent n office until May 1876).1 To computethis wage estimate, I have multipliedthe highest daily wage estimatesin theappendixof Boratavet al., OttomanWages, by twenty-six,the averagenumberof workdaysper monthassuminga six-day workweek and full employment. See also Charles Issawi, TheEconomic Historyof Turkey, 1800-1914 (Chicago, 1980), 37-43.12 Salariesfor 1876-79 (and data sources):39,000 kurus(BBA, DSA II, 416, entry of 25Cemaziyulevvel1293/18 June 1876); 40,000 (ibid., entryof 16 Safer 1295/19 February1878);39,000 (ibid., entryof 9 ?aban 1296/29 July 1879). Inall notesmentioningministerial alariesof1876 andlater,consistencywith the procedureoutlined in the appendixwouldrequiremultiply-ing the salariesby 0.975 to convert them into gold kurus.Since the highest salariesmay havebeen paid in gold (Findley, BureaucraticReform, 237)-one more inequityof the salarysys-tem-I have not done this here.13 BBA, DSA IV, 114, entries of 5evval 1297/September 1880, 14 Muharrem1300/25November 1882, 19 Cemaziyulahir1301/16 April 1884; BBA, DSA I, 576-77, entry of 20Cemaziyulahir1299/9 May 1882.14 BBA, DSA I, 576-77, entryof 15 Zilhicce 1302/25 September1885, incumbent n officethrough1895.15 Mentionsof salariesfor 1896-1908 begin with45,000 kurus(BBA, DSA I, 576-77, entryof 15 Receb 1313/1 January1896). Thefollowing arefromHariciyeArchives, Istanbul citedasHar.), Sicill-i Ahval collection (cited as SA) 429: 40,000 (entryof 18 Cemaziyulevvel1313/6November1895); 36,000 (9 Sevval 1314/24 March 1896);46,000 (4 Receb 1318/28 October,1900);25,000 (17 ?aban 1326/14 September 1908).16 Boratavet al., OttomanWages, appendix.17 Ahmed Emin [Yalman], Turkey n the World War(New Haven, 1930), 151-53; ZaferToprak,Turkiye'de Milli Iktisat, 1908-1918 [Economic nationalism in Turkey] (Ankara,1982), 334-35.


8/27


being zero, it is clear that the scope of inequalitywithin the ministrywasfantasticallywide. The elitism of this ministryandprobablyothers,especiallythat of the top officials, was also of grotesqueproportionswhen compared othe statusof humble folk. The rangeof Foreign Ministrysalariesdid fall overtime, however, especially with the government bankruptcy in the late1870s,18 he revolution n 1908, and WorldWarI. Whatfinally destroyed heeconomic elitism of the rulingclass was the dilemma, from 1914 on, of thepersonon fixed income faced with runawayinflation.This inflationarycatastropheraises a question so far neglected for earlierdates:what the officials' salaries shown in Table I meant in termsof theirliving standards.This questionis difficult to deal with, but thereare indica-tions of whatOttomanofficials regardedas a living wage. Forthe end of theperiod,there are also two systematic,butcontradictory, alculationsof livingcosts. Fromthese sources, some inferencesemerge.Analysis of the bureaucrats'estimatesof a living wage confrontsseveralproblems.Some of the estimatesrange quite high, tellingperhapsmoreaboutwhatbureaucratswantedthanabout what they neededto support heirfami-lies. 19Only infrequently,moreover,do specifics about the size of the familyto be supportedaccompanythe estimates. On the otherhand, the differencebetween nominal entitlementsand net receipts (discussed in the appendix)ceases to be a concern in analysis of these estimates, for the officials ob-viously knew that their salaries would not be paid regularlyand must havemade allowancefor this disparity.Fortunately, here are some estimatesof the 1890s thatyield a consistentand seemingly realistic indicationof what officials then thought it took tosupporta family. For example, one bureaucraticmemoirist recorded somecalculations,around he turnof this century,about how much he wouldneedto retire.He wrotethatsince he had a small family, he could get by on 5 lirasa month,or about540 kuru? n silver.20Sucha sum would not have sufficed,then, for a man with full family responsibilities.As if to support his point, apetitionerassertedin 1897 that he could not accepta post at a salaryof 600kuru?because he could not supporthis family on that.21 For officials withlarge families, or with bad habits such as alcoholism, the definition of aninadequatesalarycould be much higher.22

18 A. Du Velay, Essai sur l'histoirefinancierede la Turquie Paris, 1903), 316-461; DonaldBlaisdell,EuropeanFinancial Control n the OttomanEmpire(New York, 1929), 74-107; ShawandShaw, History, II, 221-27; Issawi,An EconomicHistoryof the MiddleEastand NorthAfrica(New York, 1982), 64-65.19 E.g., Levant Herald, 4 August 1875, p. 276, circularfrom grand vezir to provincialgovernors,implyinga figureof 3,000 kurus.20 AsqidedeHalil Ibrahim,Hatiralar [Memoirs],R. E. Koqu,ed. (Istanbul,1960), 114-15.21 BBA, Bab-i Ali Evrak Odasi 6641, Hasan Tahsin to Grand Vezir, 20 Kanun-i Sani1312/1897.22 Har., SA 531, entry of Cemaziyulahir1314/1896: unable to manage on 920 kuru?permonth because of size of family and alcoholism;Har., SA 270, entryof 17 Nisan 1327/1911:2,000 kurusper monthinadequate or largefamily at Aleppo.


9/27

88 CARTER VAUGHN FINDLEYStill, it appears hatan official of the mid-1890s would have consideredasalaryof 1,000 kuru?permonthadequate o supporta family. Sayingthat shehadonly a very small pension, and that her son's salarywas only 250 kuru?per month, an official's widow petitionedin 1892 for the son's salaryto beraised to 1,000 kuru?.23Also referring o the 1890s, when he, too, was anofficial, the writer H. Z. Usakligil discussed the significancethata raise to1,000 kuru?had for him on accountof an importantdeath in the family.24On this basis, the adjusted Foreign Ministrymediansfor the 1890s appearcomfortable,althoughby definition half the officials made no morethanthemedian,andsome of those with lower salaries musthave had largefamilies.By theYoungTurkperiod,the mostnearlycomparable alaryestimatesarefifty to one hundredpercenthigher. In a reporton a proposedreorganizationof 1912, the director of the Foreign Press Office, a part of the ForeignMinistry,said thatsalariesof 1,500-2,000 permonthwould be necessarytoattractproperlyqualifiedofficials. Thedirectorgeneralof anotherdepartmentreported hat his staff should be reorganized n classes receiving800-1,000,1,000-1,200, 1,200-1,500, and 1,500-2,000 kuru?per month.25 It is amatterof judgmentto estimateat what age, on average, such an individualwouldbecomethe chief provider or his family, presumablyanextendedone.The assumption hatthis occurredat aboutage thirty-fivewouldagain proba-bly suggest a needed salaryof 1,500. The latest mediansshown in Table 1offer little confidence that most bureaucratsmadethat much in 1912.Systematic estimates of living costs computed from prices of a typicalshopping ist of goods, with thepricesweighted by likely amountsconsumed,are availableonly for the eve of World War I. Even then, there are widedisparities n theestimates. Oneplacesthebudgetof a mid-level official in1914 at the low figureof 235.25 kuru?.26Thebudgetis incomplete,however,since it omits suchpredictable xpendituresas tobacco, transportation,nd-

most important-housing. There is a more complete estimatefor 1914 thatsets thebudgetfora familyof middlestanding ortahalli), as reckonedby theIstanbulChamberof Commerce, at 945 kuru?.27Of this, 150 is for rent;tobacco and transportation gain do not appear.In the present state of re-search, it is not clear how the differencebetween these estimatesshouldbeexplained. The fact that the lower one comes from the European-controlledPublicDebt Administration,whichpaidits Ottomanemployeesregularlyand23 Har., Terciime Kalemi Evraki 1406, no. 226, petition of Enise Hanim, 20 Agustos1308/1892.24 HalidZiya U?akligil,KirkYil [Memoirs](Istanbul, 1969), 358.25 Har., Mutenevvi249, bothreportsenclosed in dossieron reorganization f ForeignMinis-try, ca. 1912.26 Toprak,Turkiye'de Milli Iktisat, 332-33.27 VedatEldem, OsmanliImparatorlugununktisadi5artlariHakknda bir Tetkik Economicconditionsin the OttomanEmpire](Ankara, 1970), 214-15.


10/27


TABLE 2Percentage Relatives, Ottoman Foreign Ministry Salaries (Istanbul),Adjusted Means and Medians(base period 1880-82)

Means Medians Means Medians186318641865186618671868186918701871187218731874187518761877187818791880188118821883188418851886

3162361531529415214816919319012215815312614010910895100105109114118123

3752561242521282492509595868711085831009710010010099102102

1887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911

11510899959187869187919896981041101091111121128487558167

104104102100989794948695889899101961021001051001081111041009676

SOURCE: These statisticswere computedfrom those in Table 1, using the procedureoutlined innote 30.so was able to get good service at modest rates, may have depressed thefigure.28 Since, to judge from workers' wages, it must have been possible fora family to subsist on the lower budget, the difference may stem from variantconcepts of what was appropriate for the ill-defined mid-levels of Ottomansociety. Given the medians in Table 1, however, the larger budget seemsmore realistic. An official of 1914 with a nominal salary of 1,500 kurus mightperhaps have taken home 945 a month on average. Because extrapolationfrom the latest medians in Table 1 suggests that most officials of 1914 did nothave a nominal salary of 1,500 kurus-yet did have one several times the

28 RobertG. Landen, The Emergence of the ModernMiddleEast (New York, 1970), 173.


11/27


lower of the two budgets-to accept the higherestimate fits the assertionofone qualifiedobserverthat official salariesbefore World WarI amountedtoscarcely . . . a living wage. 29Thesefiguresfrom 1914 take us as far as it is possibleto go withcurrentlyavailablecost-of-living estimates. We can, however, prepare or a differentapproachto the salary-pricecomparison by convertingthe adjusted salarystatistics shown in Table 1 into percentagerelatives, which we shall latercomparewith a grainprice average expressedin the same terms.30Table 2presentsthe salaryrelatives, analysis of which follows in the last section ofthis study.

II. ISTANBUL COMMODITY PRICESTo date, no scholar has computed-or even found the sources for-a timeseries based on a comprehensivemarket-basket alculation of Istanbulcon-sumerpricesforthe latenineteenthcentury.31Exceptfor 1914, when we havethe two contradictory ost-of-living figures, determination f living costs forthatperiodthereforedependson some proxyfor a systematiccalculation.Onerecentstudyuses Britishconsumerpricesas a crude indicator of Ottomanprices.32Growing integrationof the Ottomaneconomyinto the world marketarguesfor this approach.Yet, a look at the goods used to computethe 1914cost-of-livingestimates discussed in the previoussectionsuggests that 60-75percentof either budget went for local goods whose prices must have re-spondedto local factors:perishablefoodstuffs, from fruit to fish; firewoodand charcoal;housing.33Persons with discretional income, businesses, or

29 Ahmed Emin [Yalman],Turkey, 151.30 The pointof recomputing ime series as percentagerelativesis to convert values expressedin other terms intopercentages,with the value for an arbitrarilyelected baseperiodset equalto100. This techniquenot only facilitatesanalysisof changeover time in a single statisticalseries,but also-if the baseperiodused for all series is the same-permits the comparisonof differentseries. Forcomputationof percentagerelatives on a common base periodturnsdisparatevaluesand measures(salaryper month,price per unit) into comparablevaluesexpressedon a commonscale. To convertthe salariesstated n kuru?n Table 1 intopercentagerelatives,we use 1880-82as the baseperiod.For themeans,the divisor used to computethe relatives s thearithmeticmeanof the three annual means falling in the base period. Since medians are not mathematicallymanipulable, have takenas the divisor the middle value of the threesalarymedians for the baseperiod.To computethe relatives,the values in kuru? or each yearare dividedby the appropriatedivisor, and the result is expressed as a percentage. By this procedure,as Table 2 shows, theaverageof the means for the baseperiod(1880-82) equals 100. In the case of the medians, it isthe middle value-here appearing becauseof rounding n calculation)as a pairedvalue-for thesame years thatequals 100.31 Boratavet al., OttomanWages, secondsection;cf. Issawi,EconomicHistoryof Turkey,44-50, 332-36; Safi Yorulmaz, IstanbuldaToptanE?ya Fiyatlari(1884-1911 Yillannda)[Wholesalecommodity prices], Konjonktur 1946), 45-55; and Donald Quataert, OttomanReformandAgriculturen Anatolia (Ph.D. diss., Universityof California,Los Angeles, 1973),21-23, 366-70, which uses the same sources analyzedhere.32 Boratavet al., OttomanWages, second section.33 Toprak, Tiirkiye'de Milli Iktisat, 333; Eldem, OsmanlhImparatorlugunun ktisadi5artlarl, 214-15. Incomputingthe percentage or the estimate in Toprak,I added 100 kuru? or


12/27

ECONOMIC BASES OF OTTOMAN REVOLUTION 9I

governmentagencies may well have spentmore on imports.As a guideto thefortunesof the averageOttoman,however, Britishprices are not necessarilypreferable o even a narrowlybased indicatorfrom the local market.In this study, we shall use a compositeaverageof Istanbulgrain prices. Itcan be objectedthatgrainprices give a better idea aboutthe living standardsof the poor than about those of relativelyaffluentofficials, that most grainstraded n Istanbulwere imported,at least untilthemid-1890s, or thatthepriceof grainfell faster thanthat of other items duringthe pricedecline extendinginto the same decade.34Yet, movementsin grainpriceshelp us locate timesof economic distress that, as other sources show, did have an effect on theofficial class. The figureswill show, too, that Istanbulgrainprices respondedto local forces as well as to ones from farafield;andallowances can be madein discussionfor the exceptionalextent of the price decline.The prices analyzed here come from commodity quotationsof the years1851-1914.35 To reduce distortionfrom seasonal price fluctuations,thesehave been averagedon a quarterlybasis, with all prices statedin gold kuru?perokka, a unit of weight equal to 2.828 pounds(1.283 kilograms).36 nitialrent(cf. 150 in the largerEldemestimate). On integration nto the worldeconomy, see ?evketPamuk,Osmanli Ekonomisive Diinya Kapitalizmi(1820-1913) [Ottomaneconomy and worldcapitalism](Ankara, 1984), chs. 2, 7.34 On provenanceof grainstraded n Istanbul,see note 36. With the extensionof the railroadinto Anatolia, Ottoman-growngrains began to assume an important-in some years, prepon-derant-place in the Istanbul market, but starting only in the 1890s (Quataert, EconomicClimate, Dl 157;idem, Limited Revolution:The Impactof the AnatolianRailwayon TurkishTransportation nd the Provisioningof Istanbul, 1890-1908, Business HistoryReview, 51:2(1977), 151, 154-58). Boratavet al., OttomanWages, emphasizethe exceptionaldecline ingrainpricesafter 1873.35 I collected quotationsfrom newspapers publishedon, or soon after, I March, 1 June, 1September,and 1 December of each year. The publications(followed by abbreviations),withquarters ndyearsforwhichthey served, are:Journal de Constantinople,1851-2d qtr.of 1865(Ifoundno quotations or 1850);Levant Herald(LH), 3d 1865-2d 1875;La Turquie,3d 1875-3d1880; Constantinople Messenger (CM), 4th 1880-2d 1881; LH, 3d 1881-1st 1882; EasternExpress(EE), 2d 1882-1884; Journal de la Chambrede Commercede Constantinople JCCC),lst-3d 1885;EE, 4th 1885;LevantHerald and EasternExpress(LHEE), lst-2d 1886;JCCC, 3d1886-1887; LHEE, 1888-1891; JCCC, 1892-2d 1914;Moniteuroriental, I August 1914. Con-stantinopleMessenger, EasternExpress, and Levant Herald and EasternExpress are alternatenames, inspiredby the censor, for Levant Herald.36 I convertednewspaperquotations nto gold kurusper okkaby relying, whereverpossible,on notations,publishedwith theprices, on the value of themonetaryunits andmeasures n whichthe quotationswere given. In othercases, the best guidancecame from Poids et mesures enTurquie, JCCC, no. 456, 23 September1893, pp. 446-67; no. 460, 21 October1893, pp. 495-96; no. 466, 2 November 1893, pp. 567-68. Since, for every commodity, several varieties orprovenanceswerequoted,I computedcommodity averagesas averagesof varietalsubseries.Theprocedurewas to select for each varietyall years in which there werequotations or at least threequarters.From each subseries, all otheryears were excludedas offering insufficientcontrolforseasonal fluctuations.Averaging the quotationsfor the selected years producedvarietalpriceseries, which were thenaveraged o produce hecommodity averages nTable 3. To minimizetheimpactof pricedifferencesamongcommoditieson thecompositegrainpriceaverage,I computedit as an averageof percentagerelatives (Table 4). The computationof percentagerelatives isexplained n note 30. Again 1880-82-an interval ntermediate,n terms of both time andprice-


13/27


efforts to takeaccount of a largerrangeof commoditiesfounderedon the factthatmanyof the items quotedwere not consumergoods, or not directlyso,while varietiesand quality gradesof some consumergoods, such as tea andsugar, varied so widely over time as to be unintelligible.While concentrationon grainswas morea matterof necessity thanchoice,grain productswere and are extremely important n the Turkishdiet. Theprime exampleof this is bread,of which the averageOttomanat the turn ofthiscenturyreportedly onsumeda kilogramor morea day.37As a Levantinewomanof Istanbulonce explainedto anEnglishman, vous etes carnivore, esuis carnipain. 38The priceof bread was officially controlled n Istanbul,sothatthe effects of shifts in wholesale wheatand flourpriceswerepassedon totheconsumeronly intermittently.This does not, however, deprivewheat andflourprices of value as indicatorsof movements in the economy. Too, Ot-tomanbakershadways, at times, of disadjusting the official breadprice,forexample,by decreasing oaf size.39 And Ottomansconsumedother bakedgoods, various types of noodles, cracked wheat (bulgur), and other grainproducts.With this introduction,we may now examine the price tables. Table 3presentsannualaverageprices for four cereal items-hard and soft wheats,barley, and flour. The table states the kuru?prices in decimals;in fact, thesubdivisionof the kuruswas the para, at forty to the kuru?.40Of the four commodities, hard wheat was normally used for makinglevels, betweenthe highestand lowest prices recorded-served as the base period. For wantofindicationsof consumptionvolumes, the compositegrainaverageis necessarilyan unweightedone. This limitsits value froman economic pointof view, and couldbe taken to indicateuse of asingle commodityseries for comparisonwith salaries.Here, the unweightedaverageis preferredas havingfewer gaps.Varietalsubseriesfor each commodity,with the dates for whicheach is quoted, follow. HardWheat: Azov-Taganrog, 1850-90; Ismail-Bessarabia, 1850-90; Galatz-Danube-Constantza,1851-90; Rumelian(Balqik, Burgaz), 1850-96; Edime, Rodosto, 1862-1903; Anatolia-Ban-dirma, 1896-1906; AnatolianFirst, 1905-14; AnatolianSecond, 1907-14. Soft Wheat:Ruma-nian(Galatz,Galatz-Braila,Danube, Braila), 1850-88, 1893-96, 1906-13; Rumelian(Burgaz-Varna-Balqik,Varna-Balqik),1850-96; Burgaz-Plovdiv, Plovdiv-Zagora, Zagora, 1880-96;KonyaFirst, 1904-15; KonyaSecond, 1904-14; AnkaraFirst, 1904-12; AnkaraSecond, 1904-12. Barley: Braila, Danubian, 1851-1913; Rumelian, 1858-1908; Black Sea, Odessa, 1888-1900; Mersin, 1899-1914; Anatolian First, 1903-14; Anatolian Second, 1903-14. Flour:Odessa 2d, 000, 1, and successor grades, 1869-1914; Odessa 3d, 00, and successor grades,1868-1914; Danube2d, Braila3d, and successor middlinggrades, 1870-80, 1882-85, 1888,1890-1908, 1911-14; Danube3d, Braila00, 4th, andsuccessor lower grades, 1868-76, 1878-79, 1890-97, 1899-1902, 1907-08, 1911-14; Local, local Braila, local kirma, 1868-1914.37 Quataert, Economic Climate, Dl1154.38 Bulletinmensuelde la Chambrede commerce rancaise de Constantinople,no. 256 (31July 1908), 156.39 Quataert, Economic Climate, D1155, events of 1908.40 Gapsin the priceseries have two possibleexplanations.Some signify insufficientnumbersof quarterlyquotations.Others,especially longer gaps like that of 1896-1903 for soft wheat,orshortgaps acrossall series, signify thatno quotationswere being published.The reasonis neverstated,but sometimes (i.e., 1855, 1915-18) it was obviously war.


14/27

TABLE 3Annual Averages of Istanbul Grain Prices(in gold kurus per okka)

Hard Soft Hard SoftWheat Wheat Barley Flour Wheat Wheat Barley Flour1851 0.79 0.63 0.45 1883 1.12 0.79 1.721852 0.87 0.76 0.49 1884 0.91 0.96 0.70 1.451853 1.16 0.95 0.57 1885 0.94 0.95 0.59 1.491854 2.05 1.89 1886 1.01 1.04 1.761887 1.01 0.99 0.56 1.631856 1.81 1.35 0.84 1888 0.98 0.93 0.53 1.391857 1.51 1.05 0.60 1889 0.97 0.89 0.52 1.601858 1.20 0.98 0.68 1890 0.96 1.02 0.63 1.601859 1.29 1.03 0.66 1891 1.30 1.34 0.74 1.871860 1.43 1.19 0.70 1892 0.98 1.01 0.62 1.571861 1.33 1.17 0.68 1893 0.89 0.90 0.56 1.321894 0.59 0.69 0.41 1.021864 1.00 0.91 0.50 1895 0.61 0.76 0.47 1.041865 0.97 0.88 0.49 1896 0.69 0.51 1.131866 1.27 1.13 0.63 1897 0.90 0.51 1.501867 1.59 1.41 0.89 1898 1.07 0.62 1.591868 1.51 1.43 0.93 2.11 1899 0.98 0.67 1.421869 1.17 1.03 0.64 1.77 1900 0.87 0.65 1.331870 1.27 1.14 0.71 1.87 1901 0.81 0.55 1.331871 1.33 1.18 0.75 2.05 1902 0.80 0.60 1.261872 1.23 1.09 0.63 1.79 1903 0.88 0.61 1.311873 1.54 1.47 0.75 1904 0.90 0.94 0.58 1.361874 1.47 1.27 0.84 1905 0.94 0.96 0.63 1.381875 1.22 1.15 0.78 1.75 1906 0.98 0.95 0.70 1.371876 1.14 1.09 0.63 1.84 1907 1.14 1.10 0.87 1.641877 1.27 1.28 0.65 1.83 1908 1.27 0.85 1.821878 1.29 1.28 0.77 1.82 1909 1.37 1.26 0.89 1.721879 1.46 1.31 0.78 1.86 1910 1.17 1.13 0.78 1.511880 1.52 1.47 0.90 2.15 1911 1.13 1.03 0.80 1.541881 1.36 1.32 0.74 2.08 1912 1.15 1.19 0.97 1.651882 1.26 1.12 0.72 1.84 1913 1.22 1.21 0.93 1.671914 1.14 1.13 0.78 1.60SOURCES:Pricestatisticsarecomputedfromcommoditypricequotationspublished n contempo-rary stanbulnewspapers.The sourcesandprocedures or thecomputationsareexplainedinnotes35 and 36.


15/27

TABLE4Percentage Relatives, Annual Averages of Istanbul(base period 1880-82)CompositeGrainHard Soft Price HardWheat Wheat Barley Flour Average Wheat

1851 571852 631853 841854 1491856 1311857 1091858 871859 941860 1041861 96

495973145

576272

104 10681 7675 8679 8492 8990 86

546177147

1148983859591

18831884 661885 681886 731887 731888 711889 701890 701891 941892 711893 651894 43


16/27

70 6368 6287 80109 113110 11879 8188 9091 9584 80113 9598 10689 9984 8099 8299 98101 99113 114102 9486 91

696786112105 11088 8393 91102 9689 8510710487 9191 8491 9190 9592 99106 111103 9991 90

18951896189718981899190019011902190319041905190619071908190919101911191219131914

44506578716359586465687183998582838883

SOURCE: These tatistics recomputedrom hose n Table3, using heprocedureutlinednnote30.

1864 731865 701866 921867 1151868 1091869 851870 921871 961872 891873 1121874 1071875 881876 831877 921878 941879 1061880 1101881 991882 91


17/27


200Grain Price Average

50 -

i 30-20-

10 I i i i i1850 1860 1870 1880 1890 1900 1910FIGURE. GrainPrice Average, Istanbul(in percentagerelatives, base period 1880-82)

noodles. Soft wheatwas chiefly used formakingbread.Barley,while the sec-ond most widely cultivatedgrainin Turkeyafter wheat in the 1920s and thepreferredgrainfor breadmakingn some provincesat thattime,41was not amajorconstituentof thediet of Istanbuldwellers. InIstanbul,mostbarleywasconsumedby horses, a fact that affectedliving costs to the extent thatpeopleusedhorsesfortransportation.Barleywas also a raw material or the brewingindustry,which dates back to the late nineteenthcenturyat least, and whoseproductswere surelynot unknown o theofficial class. The mainjustificationfor including barley in the averageis that its record is the longest and mostnearlyuninterruptedf the commodityseries. Includingboth flour and softwheat in the averagewill have the effect, in years when prices for both areavailable,of reinforcing he representation f the commoditiesmost directlyrelatedto breadmaking.Since the main interest of this study is in the compositecereal price aver-age, we may proceed, without further comment on Table 3, to the pricerelatives, which provide the basis for computationof this average. Thesestatistics are listed in Table 4 and presented n graphicform in Figure2.The composite grain averagewill serve for comparisonwith the adjustedsalary means and medians. Before looking at this comparison, it will behelpfulto examinethe fluctuations n the compositeaverageand to comparethem withother informationabouteconomic conditions n Istanbulduring hisperiod.

41 G. Stratil-Sauer, Cereal Production n Turkey, EconomicGeography,9:4 (1933), 324,327-39; Quataert, LimitedRevolution, 149.


18/27


The most importanteatureof the compositegrain averageis no doubt theextent to which it parallelswell-knownfluctuations n othermajormarketsofthe period. In the midst of a downwardprice trend for grains and othercommodities that spanned almost the entire nineteeth century, we find acyclical rise into the 1870s, followed by a decline to depression evels in the1890s, and then an upturnthat lasted, on the world market,until 1920.42Superimposedon this pattern,however, are short-termpeaks, usually at-tributable o local or regionalcrises. The first and sharpestpeak in 1854-56coincides with the CrimeanWar. That of 1867-68 coincides not only with apricepeakin the international rainmarkets,43but also with the Cretancrisisof 1866-69.44 Occurringat a time when the financialpositionof the Ottomangovernmentwas steadily worsening,the crisis caused economic strain,whichthe YoungOttoman deologuesdid not fail to exploit.45In 1868, distress wasso great in Istanbul that the wives of officials serving in the provincesbe-siegedtheMinistryof Financein screamingmobs, demanding heirhusbands'salaries. The finance minister had to be assigned a special guard. Publicsecurity virtually ceased to exist on the outskirts of the city and declinedwithin it.46The 1870s brought graver troubles. The Anatolian famine of the early1870s had limited effect on the Istanbulmarketbecause little Anatolianpro-duce could reach the capitalas yet. Even so, priceswentup, for 1873markeda cyclical highon the worldgrainmarket.The firsthalfof the 1870swas alsoa time of troubles for the civil bureaucracy, n the sense that instabilityofpoliticalleadershipheighteneduncertaintyof tenure n office, the probabilityof nonpaymentof salaries, and even inefficiency in tax collection.47 Thesecond half of the 1870s was one of the lowest pointsof the entirenineteenthcentury for the Ottomans, a fact evidenced especially in the governmentbankruptcyand the Russo-TurkishWar, which broughtthe Russianarmytothe outskirtsof Istanbul,flooded the city with refugees, and left the Ottomangovernmentsaddled with a huge indemnity. One result was the reissue of

42 Brian K. Mitchell, with collaboration of Phyllis Deane, Abstract of British HistoricalStatistics(Cambridge,1962), 488-89; United StatesDepartmentof Agriculture USDA), Year-book 1921 (Washington,1922), 146;Quataert, OttomanReformandAgriculture, 188, 389-90; Issawi, EconomicHistoryof MiddleEast, 10; Pamuk,OsmanliEkonomisi, 131-36.43 Mitchell,Abstract, 488-89; USDA, Yearbook1921, 146.44 Shaw and Shaw, History, II, 151-52.45 EnverZiya Karal,Osmanli Tarihi (Ankara, 1977), VII, 37; Du Velay, Essai, 279.46 Politisches Archiv des AuswartigenAmtes (Bonn), Abt. A (856/3), 868, I.A.B.q 95,Brassiervon Saint-Simon o Bismarck,13February1868;Uebel to Bismarck,9 July 1868(T139,mf. roll 354, in the microfilmholdingsof the U.S. NationalArchives, Washington,D.C., whereI consulted the Germandiplomatic correspondence).47 Shaw andShaw,History, II, 156;Du Velay, Essai, 316-461; RodericH. Davison,Reformin theOttomanEmpire, 1856-1876 (Princeton,1963), 301-10; Archives des Affaires6trangeres(Paris),Turquie390, 24 October1871, Vogiiuto Remusat;Turquie391, 9 January1872, idem toidem;Turquie391, 27 March1872, idem to idem;Turquie404, 17 May 1876, fromBourgoing.


19/27


papermoneyfrom 1876 to 1883-the only time, other than 1863, duringtheperiodfor which we have salarystatistics when papermoney was in circula-tion. The papermoney soon sank as low as 1,300 kuru? o the gold lira.48Thereafter, n the 1880s, we see that the downwardtrend in prices reap-pears, through1894. Duringthis interval,there s only theprice peakof 1891,whichwas due to a cropfailurein Russia and a consequentsurgein Ottomanexports. Europeanand American grain prices showed small rises at thetime.49After 1894, when prices turnedup again, the price high of the late 1890scoincidedwith cropfailurein Anatolia, the Greco-TurkishWar of 1897, andthe Armenian crisis of the same period.50Thereafter,the high of 1907-9signaledan unusualcombinationof woes-international financialcrisis, do-mesticandinternationalropfailure,exceptionallywidespreaddistress-thatprovided he economic backdrop or the YoungTurkrevolutionof 1908.51Itis significantthatsimilarproblems,includingprolongeddrought(1905-10),foreshadowed he Mexicanrevolutionof 1910. There,the internationalinan-cial crisisof 1907, whichbeganin the UnitedStates, is recognizedas a majoreconomic factor.52That crisis should probablybe rankedwith the Russo-JapaneseWar and the Russian revolutionof 1905 as an element in inducingthe wave of revolutions and disturbances hat swept Asia and the colonialworld in the years precedingWorld WarI.53 Finally, the war-induced nfla-tionraisedthecost of living by a factorof morethantwentybetween 1914and1920.54To see what thesepricefluctuationsmeant n thelives of officials, we mustnow comparethe compositegrainprice averagewith the salarystatistics.

48 Shaw and Shaw, History, II, 182ff., 221ff.; Du Velay, Essai, 354-57; Issawi, EconomicHistory of Turkey,326-29, 361-65.49 Quataert, Ottoman Reform and Agriculture, 21-23, 211-12; cf. Mitchell, Abstract,489; USDA, Yearbook1921, 146.50 Robert Melson, A TheoreticalInquiryinto the Armenian Massacres of 1894-1896,ComparativeStudies in Society and History, 24:3 (July 1982), 481-509. Qualifiedobserversnoted at the time how the turmoil in eastern Anatolia affected the grain trade. See Hand-elsbericht ur das Jahr1897, Deutsches Handelsarchiv 1898), 511-16 (unfortunately, havetocite this sourcefrommemory,as I can no longerfind my photocopy;I have verified the citationfrom otherrecords).51 Quataert, Economic Climate, D1157, D1161; Pamuk, OsmanliEkonomisi, 135-36.52 Ram6nRuiz, The Great Rebellion:Mexico, 1905-24 (New York, 1980), 120-35. On thecrisis of 1907 in the UnitedStates, see Milton Friedmanand AnnaSchwartz,A MonetaryHistoryof the UnitedStates, 1867-1960 (Princeton, 1963), 156-68. I have not found a good discussionof the worldwideeconomic effects of the crisis; see commentsin William C. Schluter,ThePre-WarBusiness Cycle, 1907 to 1914 (New York, 1923), 13-34.53 Keddie, Religion and Irreligion, 265; Stavrianos,Global Rift, 367-427. Detailed re-search nto the economichistoryof other countries hatexperiencedrevolution ust before WorldWarI-for example, Iran(1905-11) and China(1911)-might well disclose commonfactors naddition o financialcrisis, such as the droughtor cropfailurethatappears n both the Ottomanand Mexican cases.54 Ahmed Emin [Yalman],Turkey,151; Toprak,Tiirkiye'de Milli Iktisat, 331.


20/27

ECONOMIC BASES OF OTTOMAN REVOLUTION 99III. CONCLUSION: LIVING STANDARDS AND SHIFTING LEVELSOF POLITICAL ACTIVISMFromthe salary-price omparisonemergethe conclusionsof this essay. Sincethere is no sign in Ottomansources that salariesvaried systematicallywithpricespriorto World WarI,55the best way to comparethe two is to assumeno statisticaldependenceof one serieson the other.We shall simplycomputetwo sets of ratios by dividing each year's relatives for the adjusted salarymeans and medians (Table 2) by the same year's relative for the compositegrainpriceaverage(Table 4). In the few yearsfor which the grain averageismissing, it will not be possible to computethese ratios. In any yearwhentherelative for the salary statisticis greaterthan that for the grain average, thevalue for the ratio will be greaterthan one. When the opposite relationshipoccurs, the ratio will be less than one. The lower the value of the ratio, thegreaterthe economic distress it implies. Table 5 presentsthese ratios.The salary-priceratios show that the economic situationof ForeignMinis-tryofficials variedwidely over time. The late 1860s and the 1870s witnesseda markederosion of living standards.Qualitative evidence confirms thispoint. Intheprevioussection, we noted riotsby the wives of officials in 1868,followed in the 1870s by problems stemming from administrative rreg-ularities,governmentbankruptcy,and the Russo-TurkishWar.After 1880, an improvement n official living standardsappearsto havebegun, as evidenced in the doubling of both of the Table 5 ratiosby 1894.Even if the fall in grainprices exceeded that in othergoods, therewas stillroomfor ratioscomputedon a broader-based rice indicator o show markedimprovementduringthis interval. While the Ottomaneconomy was predomi-nantly agricultural,and agriculturalprices were falling between 1880 and1894, an improvement n the living standardsof officials, or of othersocialgroups,is not implausible.The latenineteenthcenturywitnessed animportantrailroad-related xpansion in Ottomanagriculture,the effects of which insome ways outweighedthose of the pricedeclines. Expansion n agriculturaloutputandexportvalues did not translatedirectlyinto increase n governmentrevenuesor, by extension, into funds availableto the governmentfor salarypayments, as a growing number of Ottoman taxes passed under control offoreign creditorsin these years.56The significance of this fact for officialsalary payments is not clear, given the limited information now availableabout how salary paymentswere actuallymade. In any case, the qualitativeevidence for 1880-94 tacitly reinforcesthe evidence of Table 5 throughthe

55 Toprak,Tiirkiye'de'Milli Iktisat, 334, cost-of-living adjustmentof 1916.56 Quataert, Ottoman Reform andAgriculture, 15-17, 189-91, 352-54; idem, LimitedRevolution, 143, 159-60; idem, AgriculturalTrends and GovernmentPolicy in OttomanAnatolia, 1800-1914, Asian andAfricanStudies, 15:1(1981), 83; Issawi, EconomicHistory ofTurkey,353-55; idem, EconomicHistory of MiddleEast, 105.


21/27

IOO CARTER VAUGHN FINDLEY

TABLE 5Ratios of Percentage Relatives: Adjusted Salaries to Grain Price Averages

Ratio Ratio Ratio Ratiofor for for forMeans Medians Means Medians186418651866186718681869187018711872187318741875187618771878187918801881188218831884188518861887

3.42.31.80.81.41.81.92.02.21.11.51.71.51.51.11.10.91.01.21.21.51.61.51.5

5.63.01.12.31.52.72.61.10.90.81.01.30.90.91.00.91.01.11.11.31.41.31.4

188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911

1.61.41.20.91.11.31.81.61.61.41.21.31.51.71.71.61.61.51.11.00.60.80.8

1.51.41.31.01.31.41.91.61.71.31.31.31.41.51.61.51.51.41.41.21.11.01.10.9

SOURCE: These statistics were computed by dividing the salary relatives in Table 2 by thecompositegrainprice averagesin Table 4.

absence of reports of distress like that of the 1860s or 1870s, though there wassuffering from irregularity in salary payments.After 1894, things changed again, mainly thanks to the rise in prices (Table4). For this period, both series of salary-price ratios indicate worsening condi-tions, especially just before 1908. Our narrowly based price indicator mayagain misrepresent the magnitude of the change. But the witness of othersources, noted in the previous section, becomes eloquent as we approach the1908 revolution. The international financial crisis of 1907, together with otherfactors, reminds us that this was a time of economic and political disturbancearound the world. After 1908, the situation of Ottoman workers improved,57

57 Boratav t al., OttomanWages.


22/27

ECONOMIC BASES OF OTTOMAN REVOLUTION IOIbut the ratios of Table 5 show nothing of the sort for officials. Literarysources, already noted, indicate a drastic worsening of their plight duringWorld War I.What is most interestingis the correspondencebetween these economicfluctuationsand the political and intellectualhistoryof the period. Here wefind a parallelismclear enough to suggest a causal link between economicdistressandpoliticalagitation.A statisticalargument, ike thatcentralto thisarticle, cannot in itself prove the existence of such a link; but some of theevidence here considered, and some that emerges from other scholarship,does indicate a causal connection, which we must emphasize as our finalpoint.58

58 One way to appreciatehe strengthof the causalargumentn this case is to obey the dictatesof a strictconcern for method and consider the null hypothesis that there was no connectionbetween economic distress and political behavior. Aside from some contraryevidence alreadypresented,one of the best ways to assess this hypothesisis to examine how von Wangenheim,thenGerman hargeat Istanbul-he was ambassador here n 1914-dealt with the same idea in adispatchof 1901 (PolitischesArchiv des AuswartigenAmtes (Bonn), Tiirkei134, Bd. 18, to vonBulow, 26 August 1901, corresponding o T 139, roll 392, in the microfilm collection of theUnited States NationalArchives). Commentingon press reports hatOttomanofficials in IstanbulandSalonica hadpetitioned he sultan o have theirback salariespaid,on theground hattheyandtheirfamilies would starveotherwise, Wangenheimarguedthat it would be wrongto concludefrom this that there was any dangerof revolution.One of his argumentswas that officials andmilitaryofficers blamedtheirproblems,not on the sultan,but on theirsuperiors,with the resultthatcomplaintslike these served the sultan's efforts to maintainhis own political dominance.Anotherargument oncernedthe likelihood of starvation.The abilityof poorTurksto get by onverylittleexcelledeven whathe had seen in Spain,wroteWangenheim.To backupthepoint, hedescribedhow elderlyTurks would fish on the landingin front of the embassysummerresidenceat Tarabyaon the Bosphorus. Too, poor Turks carried the sharingof goods to the social-democraticideal, and shopkeepersalso would take mercy on the poor, so that the shockingindigence observable in other European capitals was unseen. Turks in office had the addedadvantageof enjoyinginfluence,which enabled themto extractbribesfromthepublic. Knowingthatthey did so enabled the sultan to accustom his officials to irregular alary payments. Salarypaymentshad thus reachedthe pointof being a specialact of graceby the ruler,announced nthenewspapers,andcelebratedalmost like a nationalholiday, notonly by the officials, butalsoby the tradesmenwho suppliedthem on credit betweenpaydays. Only ChristiansandforeignersinOttoman ervicesuffered,Wangenheimargued,as theylacked access to the Muslims'businessarrangements nd love for their fellows. Wangenheim'scomments on the sultan'smanipulationof salarypaymentsare probablyworthtaking seriously. Yet it is quite unclearwhy a ChristianOttomanofficial could find no supportamonghis coreligionists,at anyrate. Even moreperplex-ing is the ingenious way Wangenheim'sargument hanneled he grievancesof Ottomanofficialsintoa limbowheretheyhad neitherseverehumancosts for the officials norpoliticalcosts for theregime. Perhaps n gratificationat this conclusion, a pencilnote below Wangenheim'ssignature,probablyby von Billow, states: very well writtenandcorrectlyobserved. Infact, theargumentis a piece of orientalism n the sense of EdwardSaid. Why shouldthe behaviorof poorold Turks,fishingon a landing,haveprovidedanybetterguideto thepoliticalbehaviorof Ottomanofficialsandmilitaryofficers than that of peasantsdiggingpotatoesin Prussiawould have provided o thebehavior of Germandiplomatslike Wangenheim?Apartfrom having discussed these old menwithhis Montenegrindoorman,as he says, how well didWangenheimunderstand hem?It is notworthwhileto belabor such questions, since the kind of argumentWangenheimsoughtto makecould be updatedand strengthened.Yet the fact remains thatrevolution came only seven yearsafterhe wrote this dispatchand that Salonica and Istanbul,the sources of the news reportsonwhich he commented,were its most important enters.


23/27


To sum up the evidence for this causal link, we may start from the pointthat both periods of political and ideological ferment, those of the YoungOttomans and the Young Turks, were times of economic distress. In bothperiodsthe distressextended well beyondthe official milieu from which thesalary statisticsderive.59 And in both periods the ideologues and activistsdisplayed at least some responsivenessto economic problems. Among theYoung Ottoman eadership,almost all of whom came fromextremely priv-ileged backgrounds, he response may nothave gone muchbeyonddenuncia-tion of generalgrievances,such as foreigncommercialprivilegesor thepublicdebt.60Given the massive long-term problemsof the Ottomaneconomy, itwas perhapsnatural or the earliestideologues to respondmore to them thanto specific short-termproblems.Later,the YoungTurksappear o have cou-pled discussion of general issues with exploitationof specific crisis condi-tions. No doubt, the extent to which the scope of political mobilization hadbroadenedby then contributed o this result. Those whojoined or respondedto the Young Turk movement included not only civil bureaucrats nd largeelementsof themilitary,but also nonbureaucraticlites, membersof the non-Muslim communities, sometimes even workers. The connections betweenYoung Turkactivists and these variousgroups are not all well established.The rootsof the movementin the civil bureaucratic ndmilitaryelites are, ofcourse, best known.61In the case of the workers,to cite a less-knownexam-ple thatis particularly ignificantwherepolitical mobilization s concerned,Young Turk activists had forged alliances with aggrieved worker groupsbefore revolutionbroke out in 1908, and had apparently ed at least oneLudditedisturbance,which resultedin good partfromthe currenteconomiccrisis. Consideringhow little researchhas been done in Ottoman abor histo-ry, it is highly likely that more such examples of economically motivatedpolitical activism await discovery.62

59 Withparticulareference o agrariandifficulties of 1873-75 and 1907-8, Quataert, Com-mercializationof Agriculture, 52-53, makes much the same point by speakingof social andpolitical dislocations thatarose out of crises in agriculture.60 Mardin,Genesis, 166-68, 321-23, 354, 388. As noted in the precedingsection, there isevidence that the Young Ottomansdid seize upon the economic problemsthat surroundedheCretancrisis (1866-69), at least.61 Among the many sources thatcould be cited on this point, Shaw and Shaw, History, II,263-66; Mardin,Jon Tiirklerin, 11, 22-27, 32-33, 39-41, 225-26, et passim.62 Donald Quataert,Social Disintegrationand Popular Resistance in the OttomanEmpire,1881-1908 (New York, 1983), chs. 4, 5; idem, OttomanLudditesand the Changing CarpetIndustryn Usak, Anatolia, 1860-1914 (Paperpresentedat the ThirdInternationalCongressonthe Social and Economic History of Turkey, Princeton,24-26 August 1983); idem, personalcommunications,November-December 1983. Further n theworkingclass and on thebeginningof socialist influence, especially among non-Muslims, see Paul Dumont, Une organisationsocialisteottomane:LaFederationouvrierede Salonique, Etudesbalkaniques,11:1(1975), 78;idem, Sources in6ditespourl'histoire du mouvementouvrieret des courantssocialistesdansl'empireottomanau debutdu XXe siecle, in Social and EconomicHistory of Turkey 1071-1920), OsmanOkyarandHalil Inalcik,eds. (Ankara,1980), 383; idem, A proposde la 'classe


24/27


In considering the correspondence between economic and political-intellectualhistory, the intervalin the 1870s and 1880s between the YoungOttomanand Young Turkperiodsis also significant. Economically, it was atime of relief for the bureaucraticntelligentsia,andprobably or othersectorsof society, too. Politically, it was the periodwhen Abdtilhamidquashedtheoppositionand consolidatedhis palaceregime. Makingdue allowance for hisintelligenceandtheauthorityof his office, it seems-unless we supposetheseeconomic and political facts to be unrelated-that the last great flourishofOttomansultanism was madepossible in some degree by a detente, if it wasno more thanthat, in the generaleconomic decline of the empire. When thebenefits of this detente began to dwindle, political opposition re-emerged,sultan or no sultan. After the economic situationtook a sharpdownturn n1907, revolution broke out. Here we have anotherexample of James C.Davies's J-curve theoryof revolution,with the variation hat theperiodofimprovement in this case not so prolonged:ca. 1880-94) yieldedto a periodof gradualeconomic erosion (ca. 1894-1907), priorto the sharpreversalof 1907, which precipitated he crisis.63In the case of the bureaucraticntelligentsiaand, at key moments, othersectorsof Ottomansociety, economic patterns hus variedover time in waysthatdisplay a clearly intelligible relationship o the rise and fall of politicalactivism. Extensivedocumentationof the causal connection will require ur-ther researchof a kind different from the statisticalanalysis presentedhere.Yet, it is clear that the familiarpolitical and intellectualexplanationsof thesequenceof events thatran from YoungOttoman erment,throughHamidianrepression, to Young Turk revolution must, as comparativeor theoreticalanalysis suggests, expandto include the economic dimension.

APPENDIXCOMPUTATION OF SALARY STATISTICS

Sources and MethodsThemainsource orcalculation f thesalary tatistics s the collectionof personnelrecordsSicilli-iAhval,citedin the notes as SA) in the archivesof the OttomanForeignMinistryn IstanbulHariciye, ited as Har.).This collectionncludes771envelopes,eachcontaining ocumentationn one official. In addition, ecordsofsomehigh-rankingndividualsssociatedwith heForeignMinistry ppear nly n thePrimeMinisters'Archives(Basbakanllk rsivi, cited as BBA), Istanbul,n theDahiliye sicill-i ahval defterleri(cited as DSA), consistingof 196 largeregisters;seeAttilaCetin, Ba?bakanlikArtivi Kilavuzu[Guide to BBA] (Istanbul, 1979), 46; andCarterVaughn Findley, BureaucraticReformin the OttomanEmpire:The Sublimeouvriere'ottomane a la veille de la revolutionjeune-turque, Turcica, 9:1 (1977), 229-52;Georges Haupt and Paul Dumont, Osmanli ImparatorlugundaSosyalist Hareketler [Socialistmovements in the Ottomanempire] (Istanbul, 1977; not seen).63 Davies, Toward a Theoryof Revolution, 136.


25/27


Porte, 1789-1922 (Princeton, 1980), 267. Manyof the files in the ForeignMinistrypersonnelrecordsshow only briefcareerrecords,since young menoftenacceptedoneor two appointments,but did not continue (cf. Findley, BureaucraticReform, 234-39). To distinguishcareerists from noncareerists,I collected only files with servicerecordsspanning ifteen solaryears. Theresult was a groupof 366 individuals,not allin serviceat once. Inthe analysis, I recorded he salaryof each official, as of 1June oftheGregorian alendar,forevery yearof service. Since thepricedata come only fromthe Istanbulmarket,only salaries of officials serving in thatcity have been used incomputingthe statisticspresentedhere. As the programfor executing the computeranalysis of both salaries and prices, I used the StatisticalPackage for the SocialSciences. See NormanH. Nie et al., SPSS: StatisticalPackagefor the Social Sciences,2d ed. (New York, 1975).Problems of AnalysisCorrect nterpretationf the salarydata contained n these sourcesdependson under-standing-and, wherepossible, eliminatingdistortions hatresultfrom-certain prob-lems of either the documentationor the data collection method.These problems,andthe proceduresor assumptionsthat I have used in coping with them, are as follows.Representativeness f the Data. Two questionsof this type arise. Was the ForeignMinistryof elite status in relation to the rest of the civil bureaucracy?Was the civilbureaucracy f elite status in relationto the restof the populace?The only answer forthe firstquestionis the impressionisticone that while the Foreign Ministry may havebeen exceptionallyprivilegedduringthe Tanzimat,it was no longer so under Abd-ulhamid Findley,BureaucraticReform, 135-37, 153-55, 242-43, 255-57). On theelitism of the civil bureaucracyn relationto the populace, the text presentsevidenceof an income gap thatwas wide, but narrowedwith time.Monetary Units. Because the Ottoman monetary system was complex, and thepersonnelrecordsare not specific as to money of payment,care is neededto convertthe salariesstated n the records nto unitsof constantvalue. The Ottomancoinageofthisperiodwas officially bimetallic.The gold lira, nominallyconsistingof 100 kurus,stood throughoutthe period at essentially 1.11 to the pound sterling. The silvercoinage was mintedin unitsnominallyworth20 kuru? the silver mecidiye)and less,thegold-silverratiohavingbeen set in 1844 at 1:15.0909. There werealso small coinsof base metal. Finally, papermoney was in circulation,usually much depreciated,duringthe intervals1839-63, 1876-83, and 1914-22. See CharlesIssawi, TheEco-nomicHistory of Turkey,1800-1914 (Chicago, 1980), 326-31; idem, An EconomicHistory of theMiddleEast and NorthAfrica (New York, 1982), 186;GeorgeYoung,Corpsde droit ottoman(Oxford, 1905-6), V, 1; CarlAnton Schaefer, Geldwesenund Staatsbankfragen der Tiirkei, in Das TurkischeReich, Josef Hellauer, ed.(Berlin, 1918), 33; RodericH. Davison, Ka'ime, Encyclopedia of Islam, 2d ed.,IV, 460-61; I have also computedmoney marketstatisticsfromthe samenewspapersused to computecommodity prices (see note 35).Since gold was mostly hoarded,and papermoney circulatedfor only a few of theyearsfor whichstatisticsappear n the tablesin the text, the silvermecidiyemusthavebeen the primaryunitfor salary payment.Convertingsalaries assumed to be stated insilverinto unitsof constantvalue becomesa problemafterthe value of silverbegantodecline in the 1870s. By the early 1880s, it took 108 kurus n silver to buy a gold lira(equalto 100 kurus in gold). On the worldmarket,the value of silver declinedmuchfurtherhereafter. nIstanbul,however,the Ottoman ilvercoinageheldsteadyaround108 kurusto the gold lira through1914, accordingto the money rates publishedinlocal newspapers.For an explanationof this stabilization,see Salgur Kanqal, Ladualisationde l'espace monetaireottoman (PaperpresentedattheThirdInternational


26/27


Congresson the Social and Economic Historyof Turkey, Princeton,24-26 August1983).The Ottomangovernmentrespondedto the decline in silver by decreeingin 1879thatthe silver mecidiyewas worth 19 kurus n gold, rather han20, andproceedingtouse this rate in governmenttransactions. The capitalist commercial sector did notacceptthis rate. In fact, there was an error in it, for if it took 108 kuru? n silver toequal 100 kuru? n gold, 100 kurus n silver were worth 92.59 in gold. At one fifth ofthat,the value of the silvermecidiyeshould have been reduced,not to 19kurus,butto18.52. The new official ratefor the mecidiye representedan errorof 2.6 percentof18.52, or 2.5 percentof 19. The governmentdid not rectify this erroruntil August1909, when it reset the rate for the gold lira at 102.6 kurus n silver while holdingthemecidiyeat 19 (Schaefer, Geldwesen, 30). For the yearswhen governmentpolicywas based on this error,I have interpretedhe personnel-recordalaryfiguresto meanpaymentin mecidiyes assumedby the payer to be worth 19 kurusin gold. To statesalaries n gold kurus,I have compensatedfor the errorof 2.5 percent by multiplyingall salariesfor the appropriate ears by 0.975. Since the adjustment f the official ratewas a tardyresponseto the decline in silver, I have madethis adjustment rom 1876on, thatbeing the firstyear when the money rates in Istanbulnewspapers ndicate anappreciabledecline in silver.DisparitybetweenNominalSalaryEntitlementsand ActualReceipts. Ottomanoffi-cials did notget paidregularlyor fully. Thesalaryfiguresin thepersonnelrecords hussignify gross nominal entitlementsby way of salary, ratherthan net receipts. Thedifferenceconsistedpartlyof deductions or thingslike retirementunds. By 1913, the

deductionsamounted o 9 percent(Har., Mutenevvi 156, TurhanPasato Said HalimPasa,9 10bre[sic] 1913). A largerpartof the gross-netdifferenceconsisted of delaysof payment, some of which were never made up. Indeterminably arge gross-netdifferencesraisequestionsabout the value of the personnel-recordalarydata;yet, fora varietyof reasons, they remainworthyof study. Thereare, for example, no knownsources from which to verify the actualreceiptsof substantialnumbersof civil offi-cials. Similarly, despite some efforts (Findley, BureaucraticReform, 277-78, 331-32), the Ottomans never produced any comprehensivegradingof officials and theirsalaries.Thus, there is no convenientsource, like the barem(Frenchbareme)adoptedunder herepublic,fromwhich to verify salariesof officials of different ypes. Further,while theproblemsof the salary systemaresometimes takento meanthat salarieswereunimportanto officials, there is evidence, too voluminousto presenthere, that offi-cials were keenly concerned about their salaries, precisely because they were notregularlypaid (Findley,BureaucraticReform,236-39; I planto discuss this evidencemorefully in a book on the Social Historyof OttomanCivil Officialdom ). Reflec-tion shows, too, that the salarydata in the personnelrecordsdo tell several things.Theyindicate heprioritiesof those withdiscretion o assignsalaries.They indicate heupper imitof whatofficials legitimatelyreceivedby way of salary.If we can assumethat the gap between nominal and actual salaries did not vary materiallyover longperiods, the nominal salariesyield good relative indicatorsof long-termchanges inbureaucraticortunes. It is essentially as relativeindicators hat the salariesareusedhere.Variationover Time in the Numberof Officialsfor WhomSalary Data are Avail-able. I calculatedsalarystatistics for the period1850-1914, butfound that the numberof cases droppedoff, and the statisticsbecameerratic,atboth ends of thatperiod.Fortheearlieryears, the cause of the trouble s that the personnelrecordswerenotcreateduntil 1877. The recordscover the individual'slifespanfrombirthforward.Thus, theoldestfiles containdata foryears long before 1877;yet, thefurtherback one looks, thefewerthecases. The problem n the lateryearsarises from the purgesthatfollowed the


27/27

io6 CARTER VAUGHN FINDLEYYoungTurkrevolutionof 1908 (Findley,BureaucraticReform,296-98). Since I haddecided to collect records only of officials with at least fifteen years of recordedservice, the decline in the numberof cases on which I have databegins, not in 1908,but some fifteenyearsearlier.The result is a data distributionn which the numberofofficials servingin Istanbulwith known salariesbegins at 21 in 1850, rises to 203 in1891, then starts o fall, plungesfrom 125 to 53 between 1908 and 1909, and standsat8 for 1914. To cope with this, I have adoptedan arbitraryutoff point, and show nostatistics for years with fewer thanfifty cases.Apparent ncrease over Time in theSeniorityof the Officials or WhomSalaryDataareAvailable.Thefactorsthat caused the numberof cases to riseand fall also createda misleadingappearance f a long-term ncrease n seniority.If the oldest recordsdatefrom 1877, statisticsfor earlieryears will reflect a population hatis more and morejunior, the further back the dates of the statistics. After 1893, my rule excludingofficials with fewer than fifteenyearsof service, andthe decline in maintenanceof therecords n 1908, will create the appearanceof a population hatbecomes more seniorwith the passageof time. In the interveningperiod, 1877-93, the senioritydistribu-tion, though heavy withjuniorofficials, was notchanging.The increase n the degreeof seniority is thus an artificial productof the data collection procedureand thepropertiesof the records, not an attributeof the populationstudied. Given the like-lihood of a positive associationbetween seniorityand salary, this increase must bebroughtundercontrolstatisticallybefore the salarystatisticscan be regardedas reli-able indicators.For themeans,standardizationrovidesthe neededadjustment.See R. G. D. Allen,Statistics or Economists(London, 1968), 111-14. I have standardizedacross time,dividingthe salariesquotedfor each year into three bracketsdefinedaccordingto theseniorityof eachofficial at thetime, andrecomputing he salarymeanforeachyearasa weighted average of the means for each seniority bracket. The three senioritybracketswere definedas includingofficials whose recordedcareerscomprised spansof 0-14, 15-29, and 30 or more years. The weights used to recomputethe averagerepresent he proportionsamong the average number of cases falling into the threesenioritybracketsduringthe years 1880-82, which I selected to serve as the baseperiodfor the standardization.Thebaseperiodfalls in the interval1877-93, whenthesenioritydistribution pparentn the recordswas not distortedby the factors discussedabove. (For a differentreason, I shall also use the same base period in computingpercentagerelatives of salaries and commodityprices). During the base period, onaverage60 percentof the officials in service were in the 0-14 yearsenioritybracket,30 percentwere in the 15-29 year bracket,and 10 percentwere in the 30-or-morebracket.I thereforeused0.6, 0.3, and0.1 as weightingfactorsto recompute he meanfor each yearas a weightedaverageof the means for the three brackets.No statisticsare shownforyearswhentherewereno cases in one or moreof the senioritybrackets.Forthe median,as a positionalstatistic, thereis no counterpart f standardization.Because the distortionsof the senioritydistribution ie mostly in the lack of seniorofficials in the earlieryears, and of junior officials after 1893, the best alternativeseemedto be to computemediansfor only the middle senioritybracket,comprisingofficials with 15-29 yearsof service as of each yearof computation.Of course, thismethodfurtherreducesthe numberof cases used in calculation.Since the maximumnumberof cases in the 15-29 yearbracket n any yearis 84, I shall show no statisticsfor any year without at least 21 cases. Because of differences in the adjustmenttechniques or means and medians, the years for which these statisticsappear n thetablesdiffer slightly.

Findley Economy Revolution

Documents

Transcript of Findley Economy Revolution