harper - 2008 - the greek census inscriptions of late antiquity

8/13/2019 Harper - 2008 - The Greek Census Inscriptions of Late Antiquity

1/37

The Greek Census InscriptionsofLate AntiquityKYLE HARPER

I INTRODUCTIONTrustworthy umbers, reliable glimpsesof rural life, nd documentary ources fromoutside Egypt are rare in ancient studies.A dossierof evidence,then,which recordsofficially easured quantities f land and labour in the astern countryside f theLateRomanMediterraneanshouldbeof tremendousnterest.uchevidence xists, ntheformofGreek census inscriptions rom leven cities Astypalaia, Chios, Cos, Hypaipa,Magnesia on theMeander, Miletos, Mylasa, Mytilene on Lesbos, Perissa on Thera,Samos, andTralles. Nevertheless, he ensus inscriptionsresent forbiddingumble fabbreviations, ractions,nd bureaucraticargon, factwhichhas rendered he tones flittle seoutside the pecialist iteraturenLateRoman fiscal ssessment.2. H. M. Jonesoffered he tandard nalysis f the ocuments n 953.His study as the irtue fmakingthe ry,technical nformationf the nscriptionseaningful or conomichistory,nd ithas held thefield or verhalfa century.3Despite its nduring alue, Jones sstudy sflawed nsmallbut significantays.Moreimportantly,herecent iscovery f a new fragmentrom hera radically hangestheeconomicprofile f theGreek countrysideffered y this ollection, ot tomentionthat

Iwould like to express my gratitude to those who have commented on this paper at various phases, especiallyRichard Duncan-Jones and Scott Johnson. I also wish to thank the Editor, the anonymous referees for JRS, andChris Wickham, whose responses have greatly improved the article. Above all I am indebted to Christopher Jones,Michael McCormick, and Brent Shaw, who have patiently helped me develop this paper from the time itwas achapter inmy dissertation on Late Roman slavery.1ASTYPALAIA: IG XII.3, nos 180-2; A. D?l?age, La capitation du Bas-Empire (1945), 190-4. CHIOS: D?l?age,op. cit., 182-6, pi. III. COS: R. Herzog, Koische Forschungen und Funde (1899, repr. 1983), no. 14;M. Segre,Iscrizioni di Cos (1993), ED 151. HYPAIPA: J.Keil and A. Premerstein, Bericht ?ber eine dritte Reise inLydien undden angrenzenden Gebieten Ioniens , Kaiserliche Akademie der Wissenschaften, philosophisch-historische Klasse,Denkschriften 57 (1914), nos 85-7; IGSK 17.2, R. Meri? et al., Die Inschriften von Ephesos 7.2 (1981), nos 3804-6.MAGNESIA: O. Kern, Die Inschriften vonMagnesia am Maeander (1900), no. 122.MILETOS: I.Milet 3.1389-90.MYLASA: IGSK 34.1,W. Bl?mel, Die Inschriften vonMylasa (1987), nos 271-81. MYTILENE: IG XII.2, nos 76-80;S. Charitonides, Hai Epigraphai tesLesvou: sympler?ma (1968), no. 17; E. Erxleben, Zur Katasterinschrift MytileneIG XII 2, 77 , Klio 51 (1969), 311-23; R. Parker and H. Williams, A fragment of a Diocletianic tax assessment fromMytilene , EMC 39 (1995), 267-73. SAMOS: IG XII.6, 2.980. THERA: G. Kiourtzian, Recueil des inscriptionsgrecques chr?tiennes des Cyclades (2000), nos i42a-g; E. Geroussi-Bendermacher, Propri?t? fonci?re et inventaired esclaves: Un texte in?dit de Perissa (Thera) tardo-antique , in V. Anastasiadis and P. Doukellis (eds), Esclavageantique et discriminations socio-culturelles (2005), 335-58. TRALLES: IGSK 36.1, F. Poljakov, Inschriften vonTralleis und Nysa (1989), no. 250; P. Thonemann, Estates and the land in Late Roman Asia Minor , Chiron 37(2007), 435-77, provides a new edition of the inscriptions from Tralles and Astypalaia, along with some newreadings forMagnesia, Thera, and Lesbos. I thank Dr Thonemann for generously making his work available tomein advance, and I have benefited from his careful study.2 The basic discussions are D?l?age, op. cit. (n. 1), 163-96; A. H. M. Jones, Census records of the Later RomanEmpire , JRS 43 (1953), 49-64; J.Karayannopulos, Das Finanzwesen des fr?hbyzantinischen Staates (1958), 43-53;A. Cerati, Caract?re annonaire et assiette de l imp?t foncier au Bas-Empire (1975), 244-60; W. Goffart, Caput andColonate: Towards a History of Late Roman Taxation (1974), 113-21; R. Duncan-Jones, Structure and Scale in theRoman Economy (1990), 199-210; G. Kiourtzian, Note sur l inscription cadastrale IG XII 3,N? 343 de Th?ra , inE. Magnou-Nortier (ed.), Aux sources de la gestion publique (1993), vol. 1, 35-44.3 Consequently, most historians have followed his presentation: e.g.M. Jameson et al., A Greek Countryside: TheSouthern Argolid from Prehistory to the Present Day (1994), 112; C. Wickham, Framing the Early Middle Ages:Europe and theMediterranean, 400-800 (2005), 277; W. Scheidel, The Roman slave supply , in K. Bradley and P.Cartledge (eds), The Cambridge World History of Slavery, 1: The Ancient Mediterranean World (forthcoming).Duncan-Jones, op. cit. (n. 2), 199-210, offered an under-appreciated dissent.

JRS 98 (zoo8), pp. 83-II9. (?World Copyright Reserved.Exclusive Licence to Publish: The Society for the Promotion of Roman Studies 2008


2/37

84 KYLE HARPER

our knowledgeof theLateRoman economyhas expanded since the I95OS.4 ll of thisjustifies reappraisal f the nscriptions.his article eginsbyreopening he uestionofthedate and contextof the inscriptions,hallenging heconventionalwisdomwhichascribesthem oDiocletian or his immediate uccessors nd arguingthat they elong tothe ater ourthentury Section I).Then the rticle onsiders he conomicdata preservedinthe nscriptions.he collection ffers aluabledocumentary nsightsnto the tructureofwealth (III)and thedeploymentf labour (iv) inLateAntiquity.A final ection (v)considers thedemography f a largeslavepopulation recorded n the remarkable ewfragment rom hera.The goal of this rticle s to reframehat the nscriptionsan and cannotsayabouteconomichistory.6he interpretationhich emerges tands ncontrast o that resentedbyJones,who saw inthe nscriptions structural risis f LateRoman agriculture.heinscriptionshould be seen, rather, s artefacts f a dynamic period in theEasternMediterranean,duringwhich thecomposition f thearistocracy as mutatingand thelabour ystem as complex. andedwealthwas stratifiedut fragmented,n rural ectorthat as deeply nfluenced,hough otutterly ominated yurban landowners.he practice f assigning ead-tax liabilitycapitatio)to the argest andowners as common,butthedossier suggests hat and leasing,free f fiscalties,remained prominent trategyamong smallerlandlords.Most importantly,hesedocumentsinsistthat slave labourplayed a vitalrolein griculturalroduction n elite-ownedand.The newfragmentromThera provides ncomparable vidence for slave-based state inLateAntiquity. emographic nalysis f this opulationsuggests eproductiveuccess ndmeaningful evels fmale manumission.The demographicfindings ave far-reachingmplications or thenature f estatemanagement nd the omplex haracter f social relationshipsntheruraleconomy fLateAntiquity.

II THE DATE AND CONTEXT OF THE INSCRIPTIONSThe inscriptionsome from levencities, ll of them nthe astern egean islands r themainland of far-westernsia Minor. The geography s interesting,othadministrativelyand economically. ine setsof inscriptionsome from woprovinces, sia and Insulae.7BothAsia and Insulae fellwithin theproconsulate fAsia, an administrativenit independentfrom henormaldiocesanhierarchynd underthe ontrolof a proconsulwhowas equal inranktothe icarofAsiana.8The onlycensus nscriptionsriginatingutsideof theproconsul s jurisdictionre those from iletos andMylasa, in theprovince ofCaria. The provincial overnmentfCaria was underthe uthorityf thediocesanvicarofAsiana thenormalpost-Diocletianicrganization. ore than momentaryegionalfad , the inscriptions hould be theproducts of an official dministrative ction that

4 To name only a few recent contributions on the rural economy of the Late Empire: Wickham, op. cit. (n. 3),especially 259-302; P. Sards, Economy and Society in the Age of Justinian (2006); A. Chavarria and T. Lewit,Archaeological research on the late antique countryside: a bibliographic essay , inW. Bowden, L. Lavan andC. Machado (eds), Recent Research on the Late Antique Countryside (2004), 3-51; J. Banaji, Agrarian Change inLate Antiquity: Gold, Labour, and Aristocratic Dominance (2001).5 This study is reliant upon the published editions. Thonemann, op. cit. (n. 1) has shown the value of closely reexamining the stones.6 The reflections of A. Bowman, Landholding in theHermopolite nome in the fourth century A.D. , JRS 75 (1985),137-63, on theHermopolite land registers could be applied in this case: the result may be a greater rather than alesser degree of uncertainty about many important issues... the picture ismore complex, the developments moresubtle and ambiguous than might once have appeared .7 For Insulae, see J.Marquardt, R?mische Staatsverwaltung (1881), vol. 1, 348-9; Hierocles, Synecdemus, 686-7(Ed. G. Parthey (1866), 26-7).8Notitia Dignitatum Or. XX.5-8 (Ed. O. Seeck (1876), 45-6); A. H. M. Jones, The Later Roman Empire,284-602: A Social, Economic, and Administrative Survey (1964), 375.


3/37

THE GREEK CENSUS INSCRIPTIONS OF LATE ANTIQUITY 85

included he rovinces nder the ontrol f the roconsul s well as those nder the owerof thevicar.9The inscriptionsre key documents f post-Diocletianic iscality,ith itsdual systemof liability n land and persons, ugatio nd capitatio.10t is important o recognize hatthese nscriptions ecordthetax liabilities f individual andowners; hey re not a complete countof soulsor thingsnthe enseof amoderncensus.Large landowners ould beliable for the apitation taxesof their egistered ependents, hether slaves, tenants, rpermanent mployees.1 fundamentalawofA.D.37i drewadistinction etween enantswith their wn property ho leasedextra land and coloniwho exclusively orked theproperty f an individual andlord.12he landlord as only responsible orthehead taxinthe atter nstance, hen theworkerhad no landof hisown; the xtent funregisteredtenancy s therebyendered nvisiblen these tones. he inscriptionsikewisemeasuredthe and nterms f itsfiscal urden assessed nunits farable,vines, lives, ndpasture).We thus eetherural andscapethrough particular rism, hefiscalssessment f specificlandowners ortheir egisteredabourers nd taxable and.The inscriptionsrenotuniformntheir ormatrmethod ofmeasurement. thasbeenarguedthatthere re twosubsets nthegroup, primary'nd 'secondary' egisters.13ocalledprimary egistersecorded uantities f landand persons inrawphysicalunits.14Secondaryregisters,y contrast, eflectedhe onversion f rawphysicaltotals ntofiscalunits, ugaand capita; the cheduleused to convert hysical ntofiscal nits isunknownand represents sourceofuncertainty.15he differencesf formatre said to reflecthefiscal rocess tdifferenttages fmotion,first he lainmeasurementnd then ts alculation ntotaxliability.16ut this ivision nto rimarynd secondary egistersstooneat.17No two citiesmemorializedthefiscal iabilitiesnexactlythe ameformr atexactly hesame instant n thecycle.This diversity ithin an envelopeof similarity oints to acombination f imperial timulus nd local control ver thefiscal rocess.The inscriptions ave eludeda precisedating.Jones laimed thattheywere robablyengraved nthe ate thirdr earlyfourthentury .D.,whenDiocletian and hiscolleaguesand successors reknown tohave beenactive incarryingut censuses'.'8On this ogic,9 The phrase isGoffart's, op. cit. (n. 2), 121. The province of Phrygia and Caria was separated fromAsia already

by the 250s A.D.: C. Rouech?, 'Rome, Asia, and Aphrodisias in the third century', JRS 71 (1981), 103-20. It isinteresting that possibly in the mid-fourth century, and certainly in the early fifth, the positions of proconsul andvicar were combined: D. Feissel, 'Vicaires et proconsuls d'Asie du IVe au VIe si?cle: remarques sur l'administrationdu dioc?se asianique au bas-empire', Antiquit? tardive 6 (1998), 91-104. The inscriptions could originate from amoment when the offices were combined, or they could tell us that combined jurisdiction reflected an underlyingadministrative coherence; see CT 7.6.3 (a.D. 377), cited below, which also insinuates joint financial administration.10J.-M. Carri?, 'Diocl?tien et la fiscalit?', Antiquit? Tardive 2 (1994), 33-64.11 See Section iv, for the status of the labourers. J.-M. Carri?, ' Colonato del Basso-Impero : la resistenza delmito', in E. Lo Cascio (ed.), Terre, proprietari e contadini delVlmpero romano (1997), 75-150; B. Sirks,'Reconsidering theRoman colonate', ZRG 110 (1993), 331-62.12CT 11.1.14; Carri?, op. cit. (n. 11), 100: 'un testo fondamentale'; C. Grey, 'Contextualizing colonatus: the origoof the Late Roman Empire', JRS 97 (2007), 155-75, at x^9- See O. Seeck, Regesten der Kaiser und P?pste f?r dieJahre 311 bis 476 n. Chr. (1919), 27, for the reasons itmust be A.D. 371. Iwill argue below that the inscriptions andthis law are connected to precisely the same census.13Karayannopulos, op. cit. (n. 2), 46-7; D?l?age, op. cit. (n. 1), 169, 181?2.14The primary registers are Thera and Lesbos, along with Hypaipa, Miletos, and Mylasa.15The secondary registers are from Astypalaia, Chios, Cos, Magnesia, Samos, and Tralles. A. H. M. Jones,'Capitatio and Iugatio', JRS 47 (1957), 88-94, is a lucid discussion of fiscal assessment. The size of the iugum isconsidered in Section in.16Erxleben, op. cit. (n. 1), 315.17Hypaipa uniquely records household declarations. Magnesia alone lists farms alphabetically by district. The

Thera inscription may include both primary and secondary elements. A block from Lesbos provides the onlyindication that land might be graded into first and second class, as suggested by the complex schedule outlined inthe Syro-Roman Lawbook, 106c: Ed. W. Selb, Das syrisch-r?mische Rechtsbuch, 3 vols (2002). Astypalaia includestwo levels of arithmetic within the fiscal units.18Jones, op. cit. (n. 2), 49.


4/37

86 KYLE HARPER

most have favoured Tetrarchic ate. 9Recently, honemannhasmade a strong ase thatthe nscriptionshouldnot date before boutA.D. 3 O, llowingtime or he ensus rocesstobe completed. e also claims that he tonesmust date before he3zos,arguing hat neof theproperties tMagnesia may have been in thepossessionof thegoddessArtemis.This reading f the tone snot certain, nd,moreover, there ere temple roperties ntheeastern editerraneanintothe ate fourthentury.20nstead, he ater ears f thereign fTheodosiusmark theterminusntequem for he nscriptions:he agan religious fficialsrecorded n theregistersre impossible o imagine fter he380s.21 he assignment o theTetrarchicperiod is so flimsyhat it isworth reconsideringhe entireperiod betweenC.A.D. 3io and 390 as a possiblecontext.Within thisrange, here snothing hich conclusively emandsa particular ate,but there re reasons toprefer laterfourth-centuryoriginfor he nscriptions.The inscriptionsan be analysed from heperspective f four ating criteria: cripts,onomastics, anguage, ndprosopography. nfortunately,anyof the ublished ditionsof thecensus inscriptionsre inadequate.The inscriptionsrom hera, however,havereceived arefulre-edition. he editornotes the ppearanceofcursive elta, a letter ormwhich isfirstttested nthe yclades around theturnf thefourth-to-fifthenturies,nthecatacombsofnearbyMelos.22He prefers odate the nscriptionsrom hera sometime nthe ourseof thefourthentury.any observers ave beenstruck y the nconsistencyfletter orms nd irregularitiesf spacing throughouthe nscriptions.23eil and Premerstein escribed the irregularndoften ndistinctettersf the ate imperial eriod intheHypaipa inscriptions, ontrasting harplywith another local inscription hatcan benarrowly ated toA.D.30I.24 cripts ithvariable letter orms ndpoor lineationould beconsistent ith an earlier r later ourth-centuryate, thoughn set fpublic inscriptionsfrom oman Asia these haracteristics itmore comfortably ith a laterdating.25 he

19D?l?age, op. cit. (n. i), 163, implying a Diocletianic date; Karayannopulos, op. cit. (n. 2), 45-6, for A.D. 289 or298; T. R. R. Broughton, Roman Asia Minor , in T. Frank (ed.), ESAR 4 (1938), 914-15. Erxleben, op. cit. (n. 1),314, argued for the years A.D. 307?313 on the basis of the Hypaipa inscriptions, which he argued show theregistration of the urban plebs; these seem to be village household declarations. Cerati, op. cit. (n. 2), 255, offeredthe firstdissent, arguing for a date in the late fifthor sixth century. Kiourtzian, op. cit. (n. 1), favours a date in thecourse of the fourth century.20Thonemann, op. cit. (n. 1), 438-9. Line a3 records a property described as [xco(piov) Apjx?ui?o? jtpo?Gl)vop(ioi?) novo7t?pyou HpaK?Axoi). Many toponyms throughout the census inscriptions are derived from thenames of the pagan gods; the chorion of Artemis names the place, not itsowner. It is unusual (but cf. di6-i8) thatthis entry did not list an individual declarant (though perhaps there was not room on the line after the verbosedescription of the farm s location), but this does not make the goddess the owner. Above all, temples owned landlate into the fourth century. See R. Delmaire, Largesses sacr?es et res privata: l aerarium imp?rial et sonadministration du IVe au Vie si?cle (1989), 641-5.21 e:g. Magnesia, line di3. The priests at Tralles and Magnesia do not imply an early fourth-century date; cf. thepagan priests among the civic dignitaries in the municipal album from Timgad, dated to the 360s A.D. (Ed.A. Chastagnol, Valbum municipal de Timgad (1978)). On the official suppression of paganism under Theodosius,see J. Curran, Pagan City and Christian Capital: Rome in the Fourth Century (2000), 209-17.22 Kiourtzian, op. cit. (n. 1), 215. The letter form is late also at Aphrodisias, C. Rouech?, Aphrodisias in LateAntiquity (1989), 332, with cautions about using late letter forms as a dating tool.23The Astypalaia inscriptions are said to have be written in letters of a late age , without reference to specificletter forms: IG XII.3, no. 180.Mylasa: IGSK 34.1, no. 271: mauvaise gravure, lignes irr?guli?res ... aspect cursif.Lesbos: Parker and Williams, op. cit. (n. 1), 268: crudely cut .24 Keil and Premerstein, op. cit. (n. 1), 67-8: unregelm?ssige und vielfach undeutliche Buchstaben der sp?tenKaiserzeit. The fragments of A.D. 301 have been shown to concern a trust belonging to an association of woolsellers: Th. Drew Bear, An act of foundation atHypaipa , Chiron 10 (1980), 509-36. See also G. Fagan, Bathing inPublic in the Roman World (1999), 344.25 cf. Rouech?, op. cit. (n. 22), xxii. On the other hand, the census inscriptions are unique expressions of publicepigraphy, leaving us without direct comparanda. Thonemann, op. cit. (n. 1), 444, notes that the inscriptions mayreflect documentary practices.


5/37


inscriptionslso use abbreviation ymbols hat re not attested pigraphically eforethefourth entury nd are unusual beforeA.D. 350.26The onomastic vidence s inconclusive utmay also suggest later ate. In addition toa scatter f individual ames, there re two arge ets f names: a groupof I52 slaves fromThera and about 65 landowners rom agnesia. Most of thenames are commonGreeknames, lbeitwith a late ntique flavour.27he admixture f Jewish nd Christiannamesis distinct.Among the slaves, Theodoulos, Eustathia, Theodote, and Sambatia areattested. landowner nAstypalaiawas also calledTheodoulos, a nameborne exclusively yChristians nd Jews.28he relationship etween nomastic hangeand religiousconversionn ate Antiquity scontroversialnd farfrom lear.Bagnall has uncoveredsurprisinglyuick pace ofonomastic hange n gypt, ut itwould appear that herate ftransformationnnaminghabitswas slower in thecore areas of theGreekMediterranean.29twould not be unreasonabletoquestionwhetherflagrantlyhristian namesshouldbe soprominentn a publicrecord f slavesduring periodofpersecution.30The language f the nscriptionss a highly bbreviated ialectofLate Roman bureaucratese. he technical erminologyfC1y67ndKcoaXai indicates period contemporarywith or later than Diocletian. As Cerati noted, the presence of a neologism likeUWyoKSC aXOvould argue that theterms fDiocletianic policy had evolved.31 he term,used atAstypalaia, does not appear inother documentsuntil a praetoriandecree ofA.D. 480 and then n imperial onstitution ssuedinA.D. 498.32 erati s imputation asthat theseinscriptionsre later,perhapsmuch later,than theDiocletianic reform.parallelcanbe drawnwith the seof the ord JtapoLKoLnthese nscriptions,term hichenteredofficialparlance rather ate.33 ts appearance in the census inscriptionslsopredates nyotherofficial oman usage,but therelative aucity f official ocuments nGreek from hefourthentury akes itdangeroustospeculate n thebasis of language.Prosopographyholds themost hope for dating.The sixty-fiveandowners n theMagnesia inscriptionnclude ixdecurions,five r sixmembersof thesenatorial rder,

26Namely, small superscript Greek letters as abbreviations. Thonemann, op. cit. (n. i), 443-4; A. Chaniotis, TheJews of Aphrodisias: new evidence and old problems , SCI 21 (2002), 209-42, at 215.27Geroussi-Bendermacher, op. cit. (n. 1), 345-9, has a good analysis of the names on Thera. Only some villagersat Hypaipa carried the name Aurelius, which is perhaps unsurprising for the sort of modest households attestedhere. Over the fourth centuryM. Aurelii disappear: cf. J. Reynolds and R. Tannenbaum, Jews and God-fearers atAphrodisias (1987), 20.28 IG XII.3, no. 182. Geroussi-Bendermacher, op. cit. (n. 1), 347.29R. Bagnall, Religious conversion and onomastic change in Early Byzantine Egypt , BASP 19 (1982), 105-24;E. Wipszycka, La valeur de l onomastique pour l histoire de la christianisation de l Egypte. A propos d une ?tudede R. S. Bagnall , ZPE 62 (1986), 173-81; R. Bagnall, Conversion and onomastics: a reply , ZPE 69 (1987), 243-50.The bank of names preserved from Late Roman Aphrodisias (see the name index ofRouech?, op. cit. (n. 22)), wouldalso suggest a slower pace of change throughout the fourth century.30 Compare, for instance, the name index from the second half of the third century in Phrygia available inE. Gibson, The Christians for Christians Inscriptions of Phrygia (1978), which shows a thoroughly normal set ofGreek names among a group of known Christians in the immediate pre-Tetrarchic period. It is also worth notingthe extreme frequency in the census inscriptions of names ending in -lo?, a form not unusual from the late secondcentury, but very popular from the fourth century. Geroussi-Bendermacher, op. cit. (n. 1), 349; Kiourtzian, op. cit.(n. 1), 215; B. Salway, What s in a name? A survey of Roman onomastic practice from c. 700 b.c. to a.d. 700 , JRS84 (1994), 124-45, at I3^- The names at Magnesia are more traditional, which might be expected among theconservative, landowning classes of a provincial town. Still, names ending in -lo? are well represented.31 If the abbreviation k? = K(?


6/37

88 KYLE HARPER

twoperfectissimi,nd an Asiarch.34 one of the individuals f senatorialrankcan beidentified,et the rominence f senators s tself he rucialdatum.35 round A.D. 300, theSenate claimed only c. 6oomembers, the bulk of themWesterners.Over the fourthcentury, henumberof senators xpanded into thethousands;the late350s-360s sawfeverish rowth.36he Easternprovinceswere themain recruiting roundsfor thenewSenate at Constantinople. The number of senators in the small but random sampleprovided ythe nscriptionss truly triking,specially ince uncan-Joneshas shownthatwe have atmost 8 per cent of theMagnesia register. This leads us to project as many as60-75 senators n thecomplete list.37 e are forcedto believe in one of twoscenarios.Eithera highpercentagef allEastern senators nder iocletianhappenedtoown landatMagnesia, or the inscription ostdates the expansion of theEastern Senate and thedevaluation f the enatorial itle. urely, he nscriptionsre anunrecognized rtefact fa crucial fourth-centuryransformation:hereconfigurationf the asternelite.38If the census inscriptionsate to sometime fter the350SA.D., it requiresus to reevaluate their riginal purpose.39 hough it has gone unremarked, ne of thecensusinscriptionsreserves herevealing etail that he mpetus ehindthe ct of inscribingasan imperial rder todo so.40 his is in itself nusual: referenceso census books andpages inLate Antiquity recommon,whereas there sno clearparallel to the ractice finscribingax liabilities n stone.44he very xistence f the ensus inscriptionsegs for34The senators atMagnesia are XauTtpoxaxoi, i.e. clarissimi: see Jones, op. cit. (n. 8), 528-9. Jones convincingly

argued that the .OG8?aaxoi should be decurions (though the word does not have this sense in theHigh Empire,in near-contemporary inscriptions the council ofMagnesia is called f| (|>iA,OG8?ao~xoc ?ouA,f|); otherwise the curialclass ismissing entirely. The size of their properties matches other curial-scale holdings.35The members of the senatorial order are Priscillianus, Capitolinus, Hermonactiane, Eutyches, Aristocleia. Asenator, Attalus, appeared in a patronymic on Thera. These should be Easterners, since the landholdings of even thewealthiest members of theWestern Senate in the fourth century were limited to theWestern provinces: see especiallyD. Vera, Simmaco e le sue propriet?: Struttura e funzionamento di un patrimonio aristocr?tico del quarto sec?lod.C, inAtti del Colloque Genevois sur Symmaque (1986), 231-70, at 243-5.36 For the creation of the Eastern Senate, see G. Dagron, Naissance d une capitale: Constantinople et sesinstitutions de 330 ? 451 (1974), 119-210; A. Chastagnol, Remarques sur les s?nateurs orientaux dans le IVe si?cle ,AAntHung 24 (1976), 341?56. See especially P. Heather, Senators and Senates , inA. Cameron and P. Garnsey (eds),CAH2, vol. 13 (1998), 184-210.37Duncan-Jones, op. cit. (n. 2), 137-8. The stone for farms starting with the letter beta has thirty-seven farms. Thisis the nearest-complete stone, although another fragment shows thatwe are missing some betas . Using theCIG asa database of place-names, Duncan-Jones estimated that beta should account for about 3.5 per cent of all placenames. Thus we have, atmost, 8 per cent of the original, possibly less. There were also six to eight representativesof the curial order listed. The size of town councils varied, but if the number visible in the extant fragments was 8per cent of the total, itwould imply the right order of magnitude: forwhich see B. Salway, Prefects, patroni, anddecurions , in A. Cooley (ed.), The Epigraphic Landscape of Roman Italy (2000), 115-71, at 127. Indeed, the smallnumber of attested decurions strongly reinforces the inference that the original number of senators was alsooriginally much larger.38The number of Greek senators grew throughout the second and especially the third centuries, as a coterie of oldmunicipal ?lites from cities like Ephesus entered the order: see H. Halfmann, Die Senatoren aus denKleinasiatischen Provinzen des r?mischen Reiches vom 1. bis 3. Jahrhundert , inEpigraf?a e ordine senatorio (1982)2, 603-50; H. Halfmann, Die Senatoren aus dem ?stlichen Teil des Imperium Romanum bis zum Ende des 2. Jh. n.Chr. (1979); B. Remy, Les Carri?res s?natoriales dans les provinces romaines d Anatolie au haut empire (31 avantJ.C.-284 apr?s J.C.) (1989). Even assuming that fully half of all 600 senators were Eastern by A.D. 300 (Themistiusclaimed that there were still only 300 Eastern senators as late as A.D. 357), and that the inscription originallyrecorded sixty senators, then one out of every five Eastern senators held land atMagnesia, a ratio which seemsimplausibly high.39On the importance of the material function and overall physical appearance of inscriptions: A. Cooley,Introduction , The Epigraphic Landscape of Roman Italy (2000), 1.Thonemann, op. cit. (n. 1), 439, argues that thecensus inscriptions were permanent memorials of the Diocletianic census. Itwould weaken the case ifDiocletianintended a regular census cycle (see the references below, n. 48), but Thonemann s account is the first to treat theproblem of the function of the stones seriously.40Mylasa: IGSK 34.1, no. 275: ?]vi Kaiaap[o? /oia]xayu?xco[v /^?0]oi) npOKofwrjaioi).41 Karayannopulos, op. cit. (n. 2), 47, with references to libri censuales, censuales paginae, diagrapha, polyptycha,KG??l?, etc. The Volcei inscription (CIL X.407 = Lit. 3.1, no. 17) is a possible parallel, but see the apt cautions ofGoffart, op. cit. (n. 2), 113-14: we do not know what the Volcei register is.


7/37


explanation, nd indeed there re administrativeatterns n the laterfourth entury,specificallyn thereign fValentinian andValens,whichmay offer solution.42 heirtenure asmarkedbyaggressive ffortsoestablish entral ontrol verthefiscal rocess,inpart by compiling lists,breviaries, nd registers f public obligations.43 otably,Themistius ndAmmianusbothobservedthatValensmanaged the mpireas though twere aprivate ousehold,with the bility o see its udget, ncome ndexpenses, rovinceby province.44hismetaphorpoints towards hefiscal eforms hichmay havecreatedpoliticalmotive and an administrativeontext or he reation nddisplay fmonumentaltaxrecords.Valentinian andValens implementedweeping nstitutionaleformsimedat centralizing ontrol ver thefiscal rocess. Imperial axationwas a delicatebalance of imperialstandards nd localaction.45 hile Julianraised urial utonomy o a principle fhisrule,Valentinian andValens dramatically eversed his olicy infavour f imperial versight.The emperorsmade an abortive ffort o substitute etiredmperial fficials orcurialagents ntheprocessof taxcollection.46he tension nherentnthis olicy isvisible,forexample, in an Italian inscription hich shows thattheemperorshad to order theirgovernors oaudit the ctual taxcharges gainstthe rchived iabilities.47heir centralizingpolicymade the intersectionf the tax archivesand thedistribution f annualcharges flashpointfdispute.Cyclical imperial ensuseswere rarely onducted n thefourthentury,o the ontrol f thetaxregistersave thetown ouncils realmeasureofinformalower.48 alentinianandValens intervenedt this ensitiveuncturenthetaxprocess, a move which provides a possible politicalmotive behind thecreation ofepigraphic rchives.49If thiscentralizing olicy provides thepoliticalbackground, tmay be possible toconnectthe toneswith a specific dministrativection.Converging vidence ointsto acensusinAsiaMinor intheyearA.D. 37I. Inan inscribed etter o the roconsul fAsia,Valens describeda seriesof imperial ctswhich occurredon thebasis ofprogressivelybetter iscal ata from sia. Initiallyalens had acted onthebasisofan assessment , ut42A good general account of their rule: N. Lenski, Failure of Empire: Valens and theRoman State in the FourthCentury A.D. (2002), 264-307. Jones, op. cit. (n. 8), 138-54. A. Giardina and F. Grelle, La tavola di Trinitapoli.Una nuova costituzione di Valentiniano I ,MEFRA 95 (1983), 249-303.43 Lenski, op. cit. (n. 42), 272-4.44Themistius, Or. 8.114 (Ed. H. Schenkl and G. Downey (1965), 174): 7tpoo p?v (bent p oiK?a? ua?? xf|? xoaauxr|??pxf|?, xi ju?v 7tp?CT?iaiv ?Kaaxou exou?, xi 5? ?va^o?xai. See P. Heather and J.Matthews, The Goths in theFourth Century (1991), 13-26. Cf. Ammianus Marcellinus, Res gestae 31.14.2: ut domum propriam .45 Curial role, see A. H. M. Jones, The Greek City from Alexander to Justinian (1940), 144; Jones, op. cit. (n. 8),456-7; Goffart, op. cit. (n. 2), 7-26 (but cf. the review of R. Duncan-Jones, JRS 67 (1977), 202-4, for a balanced

opinion of imperial oversight); J.H. W. G. Liebeschuetz, Antioch: City and Imperial Administration in the LaterRoman Empire (1972), 161-7. For regional variety: D?l?age, op. cit. (n. 1), 34-7, 224-5, 24?~5> R- Bagnall, Egypt inLate Antiquity (1993), 157-9. The imperial centre provided standards of tax assessment and channels of appeal,while town councils retained material power through control of the process: the maintenance of census books, thedistribution of annual charges, and the act of collection itself.When Symmachus needed to know who actuallypossessed a disputed property, he went to the town council ? not the governor ? and asked who had been payingtaxes on it. Symmachus, Rel. 28 (Ed. O. Seeck, MGH AA 6.1 (1883), 302-3). For the role of town councils indistributing charges, see, e.g., CT 8.15.5 (a.D. 366).46 CT 12.6.5 (a.D. 365); CT 12.6.7 (a.D. 364). The reformwas perhaps not carried out inAfrica and Egypt. Cf. CT12.6.9 (a.d. 365); CT 12.1.97 (a.d. 383); D?l?age, op. cit. (n. 1), 34-7.47Analysed inGiardina and Grelle, op. cit. (n. 42), especially 263-5. Cf. also CT n.4.1 (a.d. 372).48 For the absence of a regular census: Jones, op. cit. (n. 8), 454. The absence of evidence is striking, forwhen acensus did happen, it is audible (e.g. Lactantius, De mort. pers. 23.1-2 (Ed. J.Creed (1984), 36); cf. theCappadocianfathers below). Seeck thought there were regular five-year censuses conducted throughout the fourth century, butthe evidence he collected is unconvincing after the 320s A.D.: O. Seeck, Zur Entstehung des Indictioncyclus ,Deutsche Zeitschrift f?r Geschichtswissenschaft 12 (1894), 279-96; noted by T. D. Barnes, The New Empire ofDiocletian and Constantine (1982), 226-37.49A series of laws which mandated the careful recording of annual receipts against tax assessments emanated fromthe East over the next century. See Feissel, op. cit. (n. 32), 288.


8/37

90 KYLE HARPER

in the resent ime f the nscription,roundA.D. 371, he claimed tohave amost completeaccount of taxable iuga of thecivicclass.50 he letter mplies vigorousevaluation offiscal iabilitiesnAsia.51The censusinscriptions rom aria, however,would require nactbroader thanthe roconsulate fAsia. Indeed,the egal nd literaryvidence ointstoa full ensusthroughoutsiana andPontica inA.D. 37I.52 he legal vidence onsists f adossiergivento the raetorian refect, odestus. He was equippedwith broad authorityto renovate ax assessments,uditpublic liabilities, nd see that the ensus isstabilizedand argumentsre ended .53 e received andmark awssorting ecurionsfrom enatorsand determininghich colonishouldbe registeredith their andlord.54he itineraryndlegal brief fModestus inA.D. 37I show that a major reallocation f tax liability ascarried ut underhis supervision.55iterary videncefrom eighbouringonticavividlytestifieso this ensus.56Insum,while the nscriptionsustdate betweenc.A.D. 3io and 390,a strong ase canbemade forplacing them everaldecades laterthanthe traditional etrarchicdate. Thenumber f senators n theMagnesia inscription s a compelling rgument n favour f adate aftertheexpansionof the astern Senate in the350SA.D. If thesuggestion f theperiod aroundA.D.37I isright, he nscriptionshould be related o the entralization ftaxcollection whichmade the rchivalrecords site fcontention)nd thefullrevampingof thecensus inAsia. This date providesa motive and an administrativeontext.Ultimately, hough, his s a circumstantialase, and an earlier ate remains possibility.If theconjectureofValens reignis correct,however, the stonesbecome importantdocuments otonlyinthe istoryf fiscal ssessment,ut inthe olitics f taxation t thepointofcontactbetween mperialnd localpower.

III THE INSCRIPTIONS AS ECONOMIC DATA: LAND OWNERSHIP

Regardlessof their xactdate, the nscriptionsreserve aluable informationn landedwealth and the eploymentf labour nthefourth-centuryountryside.he stones llowquantitative insights nto the composition and scale of individual properties, the50A. Chastagnol, La l?gislation sur les biens des villes au IVe si?cle ? la lumi?re d une inscription d Eph?se , inAARC (1986), 77-104. His text is used. Lines 2?3: [Quod ex red]itibus fundorum iuris re[i publicae, quo]s intraAsiam diversis quibusque civitatibus ad instaurandfam mojenium faci[em pr]o certis | [partibu]shabita aestimatione concensimus. Lines 12-13: Ha(n)c sani (sic) quia ratione plenissima, quod intra Asiam rei

publica iuga esse videantur.51Valens asked for even more complete registers. Lines 19?20: sane quia rerum omnium integram cupimus haberenotitiam et ex industria nobis tuam expertam diligen[ti]a[m |confit]emur, plena te volimus (sic) ratione disquirereper omnem Asiam provinciam fundos iugationemque memoratam.52The geography of the stones would argue that Asia and Asiana were administered similarly (as does theircombination under a joint vicar-proconsul, see Feissel, op. cit. (n. 9)). Moreover, in a law of A.D. 377, levies were

applied at different schedules around the Empire. The law bundled the dioceses of Asiana and Pontica together asthough they also cohered in terms of theirmode of fiscal assessment (and itdid not need tomention the proconsulateseparately): CT 7.6.3.53 CT 13.10.7 (a.d. 371); CT 13.5.14 (a.d. 371); CT 13.10.7 (a.d. 371): exhibitis partibus secundum fidem rerumcoram cognoscant ac stabilitatem census finita altercatione component.54 CT 11.1.14 (a.d. 371); CT 12.1.74 (a.d. 371).55 PLRE IModestus 2, 606-7. The laws in the Theodosian Code often descend from specific administrativecontexts and were not, originally, general : J.Matthews, Laying Down the Law: A Study of the Theodosian Code(2000), 66?71. The itinerary ofModestus might thus argue that the census was limited toAsia, Asiana, and Pontica.Modestus was made consul in a.d. 372.56T. A. Kopecek, The Cappadocian Fathers and civic patriotism , Church History 43 (1974), 293-303, collectssome of the evidence, but does not acknowledge the role of the census behind the campaign. See Basil, Epistulae (Ed.Y. Courtonne (3 vols; 1957-66)), Letters 36, 37, 83, 88, 104, 284, 299, 309, 310, 312, 313, 315. In a.d. 371, as part ofthe reorganization, the province of Cappadocia was divided in two a well-known event that has not been situatedwithin the momentous reforms of a.d. 371. Basil struggled against this division (Letters 74, 73, 76). See W.-D.Hauschild, Basilius von Caesarea: Briefe I (1990), 138-41, 208-9; Gregory of Nazianzus, Or. 19 (PG 35, cols1044-64).


9/37

THE GREEK CENSUS INSCRIPTIONS OF LATE ANTIQUITY 9I

distributionf landedwealth, and the social profile f landownership.Four sets ofinscriptionsre particularlynformative:hera, Lesbos, Tralles, andMagnesia. However,the ata are recordedndifferentnits. he inscriptionsrom hera and Lesbos are primary registers,o that and nd labour re tabulated nrawtotals. he data from agnesiaandTralles, bycontrast, remore complex,for nthese nscriptionsherawnumbers avebeen converted o fiscal nits.The scheduleused toconvert hysical ugera nto iuga isunknown, nd a full nderstandingf the ata requires s toconsider he ossible rates fconversion. he Magnesia inscriptionoses specialproblems, ut it remains, espite theuncertainties,mong themost revealing ocumentsfor the structure fwealth inLateAntiquity.The inscriptionsrom hera andLesbos furnishirect videncefor he omposition ndscaleof several roperties. he sampleissmallbutstraightforwardnd establishes basisfor nterpretinghemore complexdocuments. n bothThera andLesbos, the registerswere organizedby individual roprietor. n owner's total roperty, ypically ragmentedinto umerous maller arms, as added up intermsf iugera f arable land, ugera ndervines, and the number of olive trees.57n individual wner's farmswere classifiedgeographically y district, illage,or rural pace. In the nscriptions,heproperties erenamed and locatedprimarilyn termsf xcpit,with somereferenceoKf7Ot, 6tOI,andaypoi. The Xxpiovwas the dominant geographicunit everywhere.58he inscriptionspresent fiscal eography, ot an attempt odescribethe ountrysidentraditionalermssuch as Kdogatl.s an overriding ureaucratic ategory, he termXcopiov as flexibleenough to encompass villages inwhich farmswere located,as well as independentnucleated settlements omprisedof singleestates.59 he inscriptions ive theoverallimpressionfhighly ragmentedandownership,ith occasionalglimpses fvillages, nthe nehand,anddispersed, state-based ettlementsn the ther.60The inscriptionsrom hera and Lesbos demonstrate hat roperties nthisregion ftheEmpire could reach considerableproportions.FromLesbos, four stones survive.Unfortunately,henames and headingsare lost,so it isunclearwhere each registrationbegins nd ends.We do not knowwho owned this and,nor therelationshipetweentheindividual tones. n the ssumption hat ach stonerepresented separate state,thereis informationbout four ifferentroperties:LESBOS ARABLE VINE OLIVE (# TREES) # PLOTSXII.z.76 15I4 I09 55II i6XII-z-77 267 20 Iz8I i8XII.z.78 52 I 5 248 5XII.z.79 594 19 2000 5

in iugera

57On Lesbos alone, land given over to pasture was included as a separate category.58 Except the inscriptions of Tralles: see Thonemann, op. cit. (n. i), 454.59Thonemann, op. cit. (n. 1), 454-7, has an extremely useful discussion of the terminology. See Kiourtzian, op. cit.(n. 2), 219-21. The place-names from Lesbos are analysed in N. Spencer, 'TO ITYPPAIf?N OPOS TOniTYilAES: an archaeological and epigraphical approach to a topographical problem', ZPE 12 (1996), 253-62.The structural difference between theEastern andWestern countryside is explained inWickham, op. cit. (n. 3), 442;Chavarria and Lewit, op. cit. (n. 4), 16-17.60The largest contiguous properties are attested atMagnesia (a 7$-iuga estate) and on Lesbos, where farms of 430,305, and 294 arable iugera are listed, probably all of the same owner: IG XII.2, no. 76. Archaeological and literaryevidence for rural villas and the intrusion ofWestern settlement patterns into parts of Greece: see J. J. Rossiter,'Roman villas of the Greek East and the villa inGregory ofNyssa Ep. 20', JRA 2 (1989), 101-10; Wickham, op. cit.(n. 3), 462-6; Sarris, op. cit. (n. 4), 121.


10/37

92 KYLE HARPER

The inscriptionsrom hera are similar nformat nd yieldcomparable results. t ispossible to reconstructhetotaldimensions f three istinct roperties.THERA ARABLE VINE* OLIVE TREES # PLOTSI4zA (Paregorius) 504 79 554 J0I4zB (Attalou) 6I4 i6z I4z0 I614zC 528 I20 586 I7?in iugera

The fragmentationf landholding s notable and has been emphasized by all commentators.61 ach property as built from mallerplots. It ispossible that land fromdifferent istricts as physically djacent, and it isequally possible that andwithin adistrict as furtherragmented,o that he nscriptions ay over- runderstate he ctualdegree f fragmentation.evertheless, he attern sbroadly ndubitable,nd it onfirmsthat the gyptianevidence forfragmentations not exceptional.The inscriptions romThera and Lesbos also providea valuablemeasure of elite land-allocation trategy.heestates nThera specialized nwineproduction, hile oliveswere an agriculturalriorityonLesbos.62 evertheless, he egree f specialization as limited,ndevenon these argeproperties,rable farmingas thedominant gricultural peration.This would suggestthat lite-drivenpecialization nstaplecrops,at least inthisregion, as a phenomenonthat ccurred long themarginsof a landscapedominatedbygrain ultivation.The appearance of a senator in a patronymic on Thera (I4zB) is the lone clue to thesocialprofile f these andholders,utcomparative ata allowus tocontextualize he caleof these roperties. neveryrespect, he gyptian andregisters rommid-fourth-centuryHermopolis are the ssential omparandafor he ensus nscriptions.hese papyri recordthe and-holdingsf hundreds f urban-based wners, nd they hus llow us to say howthe roprietors rom hera and Lesbos comparetotheirgyptian ounterparts.ith onlythe xception f the ncompletehird ropertynLesbos (no.78), the egean landownerswould have ranked mong theverywealthiestof theproprietors tHermopolis,whereholdings of i,ooo arourai (= I,095 iugera) were exceedingly rare and holdings of zooarouraior more (= 2i9 iugera)belonged only to the top3.6 per centof landowners.63Especially ifthe roprietors ttested nthe nscriptionseld landelsewhere, hey eserveto be classified s large-scale andowners. he small sample preservedfrom thesetwoislands sthus, t ppears,a snapshot f thehighest ier f landholding.This knowledge is valuable as we turn to thesecondaryregisters rom stypalaia,Tralles, andMagnesia. These documentsrecordedtax liabilityn terms f fiscal nits,iuga. The schedule used to convert raw iugera into iuga is unknown. Jones favoured alarge iugum, round ioo iugera f arable land,based on an apparent conversion' romphysicalto fiscal nitspreserved tThera; Duncan-Jones later dduced severalreasonswhy the iugummust be smaller, closer to iz iugera of arable.64 Thonemann has now provided arguments for a conversion rate similar to theone advocated by Jones. Thonemann'scase must be broken down into two separate claims, one about themethod of accounting,61Recently emphasized by Thonemann, op. cit. (n. i), 475. These figures generally follow Paton's tabulations forLesbos with Erxleben's additions to no. 77 and Thonemann's emendations for Thera.62Thonemann, op. cit. (n. 1), 467: on Thera land was allocated to arable/vine/olive in the range of 80/17/3 per cent.On Lesbos, the ratio was c. 88/6/6 per cent.63 See, for instance, Table 4B of Bowman, op. cit. (n. 6), 159. Other evidence from Egypt broadly confirms theHermopolite data: see the properties analysed by D. Kehoe, Management and Investment on Estates inRoman

Egypt during theEarly Empire (1992), especially 75; J.Rowlandson, Landowners and Tenants inRoman Egypt: TheSocial Relations of Agriculture in theOxyrhynchite Nome (1996), 123.64 Jones, op. cit. (n. 2), 50; Duncan-Jones, op. cit. (n. 2), 199-210.


11/37


the ther bout the pecific onversion chedule nuse throughoutsiana. Ina realbreakthrough, e has shown thatthe nscriptionseploya relativelyrestrictedange f fractions .. based on products fprimesno greater hanfive'.65ith this nsight, e arguesthat hefractionsttestednthe nscriptionshouldbe related othe onversion ates.66ethusworksout a 'satisfyinglytraightforwarderies f conversion ates': i iugum iooiugera rable= I5 iugera f vine= 300 olive trees. he interpretationf the ccountingmethod iselegant nd convincing, ut the specific onversion chedule, articularly he'large' rable iugum, resents nsurmountableifficulties.In the firstlace, there sno reasonwhy this pecific chedule ioo arable/I5vine/300trees)must be correct. t isbased entirelyn one of the nscriptions rom hera, whichmay preserve he alculation f raw intofiscal nits.67nthree laces, the tone n uestionreads? 0o1GvK(?xakXo)4(uya).ot only istheterm ephalozyga ompletelynattested,thedisposition f the calculations' n the tone ishard to reconcile ith thetheory hatthey re calculations.68t seemshighly lausible thatthese re not infact onversions.Moreover,Duncan-Jones's ase for small iugum snoteasilybrushed side.He citesthecontemporary ork ofEpiphanius,whose book onmeasurements laimedthat iugumwas equivalent o iz.5 iugera f arable land;this s a compellingource verlooked nmostaccounts.69 Duncan-Jones also invokes the letterofValens to theproconsul ofAsia, whichimplies rateof returnn estates, n terms f solidiper iugum, hatrequires helowerestimate; fthe iugumwere ioo iugera,these stateswere producingunbelievably owrevenues.70notherreasontodoubt this chedule ioo arable= i5 vine= 300 olive trees)is thatthe mplied axeson vines and olives are implausiblyigh:the o percentof landallottedtograpesandoliveswould bearover two-thirdsf thetotaltaxburden.71The smaller ugum lso allows for more persuasivereconstructionf landholding tAstypalaia,Tralles,andMagnesia. The registers romralles andAstypalaia preserve ivecomplete roperties.72ssumingfor implicityhatthe andwas onlyin rable,andusing

65Thonemann, op. cit. (n. i), 466. It isworth noting that there are, however, a number of exceptions.66 For instance, in light of the common fraction 1/300, he proposes that 300 olive trees = 1 iugum.67 IfThonemann's reading is correct (op. cit. (n. 1), 464, lines 4-5), since he reads k? rather than Kiourtzian's k?u,itwould allow for other interpretations, such as K?


12/37

94 KYLE HARPER

a conversion chedule f i iugum I1 iugera, he roperties omparewith the states nThera and Lesbos.73TRALLES ARABLE IUGERA ASTYPALAIA ARABLE IUGERATatianus 6i6 Heracleides I29Kritias 25ILatron 205Fulvius 39

We shouldnote that f he ugum ere closerto ioo iugera f arable, the andowners fTralles and Astypalaia far outclassed theproprietors f Thera and Lesbos. This isintrinsicallynlikely, orthe andownersfThera and Lesbos included, na rather mallsample, a senator'srelative nd someonewealthy enough to own I52 slaves.Moreover,there s no obvious reasonwhy thefew ttested andowners rom ralles andAstypalaiashouldbe offthe harts n omparison othe andowners f fourth-centuryermopolis, alarge ndwealthy town.74The inscription rom agnesia isalso a secondary egister,nd it sespecially aluablefor study fwealth inthe ateEmpire,for t reservesnformationn overninety arms.The farms re grouped lphabetically,y thenameof their ocation; nlypartsofA, B, E,and another etter urvive,o this s a random egmentfamuch largerist. t first lance,the nscription ivesthe mpressionf a broad base of urban landowners: ver sixty-fivedifferentwners areattested. welve of the andownersreattestedmore than nce,andfourteen ut-of-towners,ostlyfrom eighbouringitiessuchasEphesus,are registered.We thus eea regional ristocracyhosewealthierfamilieseld landupand down thefertileMeander valley, ut themajority f local landowners ere notofany specialcivicrank.

MAGNESIA IUGA OF PROPERTIESTHIS SIZE

75 Iz6 225 I

20-I 2I2-15 28-II 64-7 62-3 I5

1-1.5 I4< I i6

The sizeof the ugum eeply nfluencesowwe understand he ata from agnesia. If thelarge iugum c. ioo iugera rable)were correct, twould make thetop landowners t73 This conversion rate represents only an order ofmagnitude. I think i iugum - c. 10 iugera is equally possible.The number of iugera would be smaller ifwe accounted for grapes and olives, larger ifpasture/fallow. But the

inscriptions of Thera and Lesbos show how little land was actually allotted to vines and trees, and it is unlikely thatan inordinately large percentage of the tax burden (much less over half of it) fell on grapes and olives, so that thesimplifying assumption of arable iugera should not radically distort the scale of the properties recorded in iuga.74Given the (apparent) size of the estates on Cos and Samos (13, 4.56, and 27.3 iuga), the small iugum would makeall of the attested owners from Tralles, Astypalaia, Cos, and Samos large landowners, whereas the large iugumwould make them all exceptionally large landowners. Cf. Bowman, op. cit. (n. 6), Table IV.


13/37

THEGREEK ENSUSNSCRIPTIONSFLATE NTIQUITY 95Magnesia exceedingly rosperous,far richer than their ounterparts nEgypt or theAegean. The largest roperty, t 75 iuga,would constitute,s a singleplot,one of thelargest oldings nown from he astern editerranean.With a small ugum, his ropertywould stillrank mong thevery argest roperties nthe ermopolite landregister.hebottom end of the scale is also revealing. herewere numeroussmallproprietors tMagnesia, in the range of i-z iuga. These owners either closely resemble the small urbanproprietorstHermopolisor far utclass them. he medianholding nthe gyptian atawas aroundioarourai ii iugera).75he median holding tMagnesia was z iuga;this as,depending on the schedule we use, on the order of zo or zoo iugera. Surely, the medianlandholder atMagnesia was not the sort of richproprietor with zoo iugera. At the top andbottom f the cale,the rder fmagnitude mplied or ndividual oldings tronglyrguesfor the small iugum. The Magnesia register can also be used to analyse the stratification ofwealth inthisregion.

0.60.5

X=0.4u0.3-I0.2

0.10

DocileFIG. i. Land ownership by decile.

Duncan-Joneshas gathered number fdata-sets n the istributionf landedwealthacross Roman history.76The twomost stratified samples come fromHermopolis andMagnesia.77 But in the case ofHermopolis, Bagnall has noted a crucial caveat. The Egyptianland register recorded only urban landowners, exclusive of peasants and villagers. If thepropertyfEgyptian illagers ere included, hemeasureof inequalityould notappearnearly so drastic.78 This raises a fundamental question about theMagnesia register: doesit include theholdings of peasants and villagers, or is this only a listof urban proprietors?The answer to this question carries significant consequences for our understanding ofthe data. If the register originally recorded all taxable land in the territoryofMagnesia,then his nscriptioneflectshemost stratifiedociety nown from he omanworld.The

75 See Bowman, op. cit. (n. 6), 158-9. At a median of 200 iugera, these could hardly be family farms. A law of A.D.342 even suggests that in some cities a property of 25 iugera was sufficient to qualify for curial service: CT 12.1.33.76Duncan-Jones, op. cit. (n. 2), 121-42.77 I calculate a Gini coefficient of .677 atMagnesia. At Hermopolis, the raw figure is .815 extremely high:Bowman, op. cit. (n. 6), 150. This number should be adjusted downward (see below). R. Bagnall, Landholding inLate Roman Egypt: the distribution ofwealth , JRS 82 (1992), 128-49, at 131, fig. 2.78 Bagnall, op. cit. (n. 77), 133-5, especially 138; a corrected Gini of .560, including villagers. The largest owners,such as the senators, held property on a larger geographic scale, with estates in other cities and provinces. So thegraph is distorted in two directions: it understates the largest owners holdings, but overstates inequality ofdistribution in this delimited space by excluding rural proprietors.


14/37

96 KYLE HARPER

large conversionrate for the iugum ioo iugera)would require such a reading f theinscription, est the implied size of Magnesia territoryxceed plausible bounds. Thealternative cenario seemsmore coherent: small iugum, register hat included nlyurbanproprietors,nd severe ut credible nequalitiesfwealth.This case issupported yformal omparisonwithotherdocuments, uch as the ermopolite registers,hich onlyrecorded hetax liabilitiesf townresidents.79nEgypt, illageswere assignedthetask fcollecting axeson their nhabitantsnd renderinghe ums to the towns; sixth-centuryinscription romtheprovince fCaria shows thatvillages inAsiaMinor could also beliable for ollecting axes nd payingthem hrough he ity.80n administrativerounds,then, t sunlikely hatthe agnesia registeristed llproprietorsnthe ity sterritory.81The assumptions f a small iugum nd an urban-only egister lso allow us to projectplausible ordersofmagnitudeforthe opulation and overall civicterritoryfMagnesia.We have c. 8 per cent of the riginal list.The extantportion includes ixty-fiveniquelandowners: ourteenut-of-townersnd fifty-oneagnesians. Thiswould leadus toproject round6ooMagnesian proprietorsnthe ntire erritoryf the ity. he urbanpopulation fRomanMagnesia hasbeen estimated t I2,500souls.82his estimate squite possiblytoo large, nless the rbanizationratewas exceptionally igh; it iseasier tobelieveina city opulation intherange f5-i0,000,with a rural opulation at severaltimes hisnumber.83ixhundred ssurely oosmallas a number f total roprietorsntheterritory-especially ifthehigher stimates f the opulationare correct but itseemsplausibleas a number f urbanowners, nparticular fmanywere small-holdersround the istrictof the ity tself.84Similarorder-of-magnitudealculations an be carried ut for thetotal civic territoryofMagnesia, which has been estimated t 47i kmz f cultivable and.85fwe have c. 8 percentof the register, he 366.8 recorded ugaon the surviving ortionwould imply noriginal totalof 4,585 registereduga. If I iugum ioo iugera rable, then the totalregisteredroperty ould far xceed thesize ofMagnesian civicterritory.86f,however,the ugum ere around iz iugera, t implies hat rban landowners eld 55,ozo iugera ntotal.This would constitute oughly 9 per centof all taxable land in theterritoryfMagnesia. There are severalpossible sources ferror nthis alculation,not tomentionthatwe have noway ofknowinghowmuch landwas imperiallywned.But it is noteworthy that a ratio of urban/ruralwnership tMagnesia of 29/71 s almostpreciselyequivalentto theratio alculated inthe ermopolite nome, 30/70.87n sum,byassuminga small iugum nd an urban-only egister,heMagnesia inscription ieldsan imageofwealth inwhich both the caleof individual ropertiesnd the veralldistribution f landcan be closelycomparedwith contemporaryata from he thercensus inscriptionsndthe gyptian apyri.79 Jones, op. cit. (n. 2), 54-5, argued on this logic that theMagnesian list recorded only urban owners.80 For Egypt: Jones, op. cit. (n. 8), 454. For the Carian example, see D. Feissel, Un rescrit de Justinien d?couvert? Didymes , Chiron 34 (2004), 285-365.81There is another internal argument: virtually none of the smaller holdings have registered capita. Theconsequences of this distribution of labour are discussed in the next section, but the immediate conclusion must bethat these do not represent village or peasant households, which would have been liable for the head taxes of their

family members.82R. M?rchese, The Lower Maeander Flood Plain: A Regional Settlement Study (1986), 317.83M?rchese, op. cit. (n. 82), 317, employs a rank-size approach which implies very high rates of urbanization (over30 per cent). Cf. the population densities and urbanization rates in B. Frier, Demography , CAH2 (2000), 787-816.84The Hermopolite data, crucially, exclude the pagus nearest the city, but see Bowman, op. cit. (n. 6), Table VI,and the exponentially greater number of owners in the 5th, 6th, and 8th pagi, the next closest areas.85M?rchese, op. cit. (n. 82), 317.86 If the tax liability were only in arable, itwould imply 458,500 iugera, a number which could be reduced byfolding some of the liability into vine-land and olive trees (indeed, it is only by putting over half the tax burden onvines and olives, and by assuming that the register accounted for all land in the territory, that the scale becomesmanageable in the large-iugum scenario).87 See Bagnall, op. cit. (n. 77), 137.


15/37


Finally,the agnesia inscriptionrovides social profile f the andowners.CURIAL AND SENATORIAL LANDHOLDERS AT MAGNESIA

CURIAL IUGA CAPITA SENATORIAL IUGA CAPITAPaul 8.78 5-.5 Priscillianus z.65 Iz.z8Tuchichos 1.05 Aristocleia xMandrogenes 2.I 7.2 Farm C3 75.I5 5z.88Pollio 2I.I5 Capitolinus .74Phanius 4.07 5.02 Hermonactiane 3.I7Heracleides 9.5 I0.36 Eutyches I.I5Farm ii .37Farm i3 2.I5 I.58TOTAL 49.I7 29.4I TOTAL 8z.86 65.i6The rise fa newaristocracy asedon imperialervice s classicthemefLate Antiquity.This class gradually isplaced an older, locally-rooted unicipal gentryn the ast. Tosome xtent, his rocesswas a re-labellingfpre-existinglites, s the ntrenchedobilitywas able tomaintain itsstatusundernew circumstances.ut the mperial entrereconfigured hematrixof loyalties, rivileges, nd resources hrough hich theupper-classmaintained its status. he census inscriptionsrovidea static mage f this rocess in asub-regionf the mpirefor hich there slittle oncrete vidence. heMagnesia inscriptionshows that theimperial, enatorial tratum as intrudingn themunicipalaristocracy,though he atter etained ignificanteservesf economicclout.88 he inscriptioncannot tell us ifthe senatorial andowners ere new menwith newmoney or formerdecurions romotedto imperial tatus. itherway, the nscriptionastsvivid lightn theplight f thetown ouncilswhich is so audible inthe ources f the eriod.

IV LABOUR IN THE CENSUS INSCRIPTIONS: DENSITY AND STATUSThe census inscriptionslso allow quantitative nsightsntothedeploymentf labourinthe ate Roman countryside.he study f labourinRoman antiquity sbesetby intractableproblems n the mpirical ecord: hat evidence urvives stypically ague,unrepresentative, r ideologically oloured.89 he random,documentary ata of the censusinscriptionsan be a corrective, nd they eservewider notice thanthey ave received.Nevertheless, aution is inorder.Jonesdrew twoconclusions about labour from theinscriptions.e arguedfor lowdensityf labour n the andand anoverallratio f freemen to slavesaround 5:I. These claimsarenotmethodologically ound,and itwill be agoal of this ection toemphasizethe imits f thedata.Despite the onsiderable ncertainties, owever,solid evidence about theuse of labouron elite-owned and can berescued rom he tones.The density f labouron the and issurely lessurgent heme ow than n I953.Thebleak image of a landscape scarred by agri deserti is simply no longer tenable in light of

88As Thonemann, op. cit. (n. i), 473, points out, the only obvious 'absentee' landlord is at line a4, a farm ownedby Quadratus but declared by (an apparent slave manager) Syneros.89 See, e.g., U. Roth, 'Food rations inCato's De agri cultura and female slave labour', Ostraka 11 (2002), 195-213;W. Scheidel, 'The most silent women of Greece and Rome: rural labour and women's life in the ancient world',G R 42 (1995), 202-17 ana1 43 (1996)> 1-10; W. Jongman, 'Slavery and the growth of Rome. The transformationof Italy in the second and first centuries BCE', inC. Edwards and G. Woolf (eds), Rome the Cosmopolis (2003),100-22.


16/37

98 KYLE HARPER

the ver-accumulatingrchaeological vidence hich shows a denseandvibrant attern fsettlementnthe ateRoman East.90 ore profoundly,he ffortoextractmeasurementsof labour densityfromthesedocuments is flawed.The census inscriptionsecord nlyregisteredabour workersforwhose capitationtax the landownerwas liable.Jonesrecognized hat free, ontract-based and-leasing ould not be detectible, uthe nevertheless roceeded todiagnosea critical hortage f farm abour.His account shows thesubtle nfluencef the olonate,forhe seems to assumethat f abourwas not registered,it robablydid not exist. et revisionistork on the olonatehas demonstrated hat andleasingremained prominent lement f therural conomy ntothe ate Empire; the ossibilityfnon-registeredabour sentirelyealistic nd indeedfindsupport nthe nscriptions see below).Moreover, Jones'scase for shortage f labourwas inseparable romhis belief that the ugum as on the rder of ioo iugera. he downward revision f theiugum rastically lters the mpressionf labourdensity n the and.9'Ifwe begin the investigationxplicitly s a searchforpatterns f registered abour,interestingesults merge. he secondary egistersrom stypalaia,Tralles, andMagnesiaall offer ata about theratios fregisteredapita toregistereduga. ut aswith the ugum,theconversion cheduleused to convert ndividualhumans intounits of tax liability,capita, isuncertain.We seem to be fortuitouslyell-informed y a contemporary awabout the onversion chedule. constitution fA.D.386adjusted thenumber fmen andwomen in caput throughouteveral rovinces f the onticdiocese; it laimed that reviouslyoneman or twowomenhad equalledone unitof head-taxliability,he aput.92t isclear thatthings ere notquite so simple, ut as a bestestimate, he chedule f iman =zwomen = i caputseemsplausible.93his yields,for nstance, hedensity f labouron amid-sizedestate nAstypalaia:94

ASTYPALAIA LAND LABOURFiscal IO.8 iuga I3.I3 capitaPhysical 130 iugera I3-z6 workers

For the ensity easurements obemeaningful, e would have toknow the veragenumber ofworkersneededper iugerum.ur knowledge fmanning-ratiosnRoman farmingcomesfrom he griculturalriters nd landgrants. hese yielda range faverage abourdensities, omewhere etween5and 25 iugera farableperworkingmale, but these iguresmust havevaried significantlyy region, caleofestate, ature f cultivation, obilizationof female abour, tc.95t the east,they anprovidea rough deaof labourrequirements.With one worker (presumably ncluding omen) forevery5-IO iugera, thedensity f90A. Dunn, 'Continuity and change in theMacedonian countryside, from Gallienus to Justinian', inW. Bowden,L. Lavan and C. Machado (eds), Recent Research on the Late Antique Countryside (2004), 535-86; S. Kingsley andM. Decker (eds), Economy and Exchange in theEast Mediterranean during Late Antiquity (2001); Jameson, op. cit.(n. 3); S. Alcock, Graecia Capta: The Landscapes of Roman Greece (1993). Not tomention that agri deserti shouldbe interpreted in fiscal, not demographic, terms: C. R. Whittaker, 'Agri deserti', inM. I. Finley (ed.), Studies inRoman Property (1976), 137-65. Most recently, C. Grey, 'Revisiting the problem o? agri deserti in the Late Roman

Empire', JRA 20 (2007), 362-76.91As noted already by Duncan-Jones, op. cit. (n. 2), 207.92 CT 13.11.2 (a.d. 386), issued to the Prefect of the East. Equivalent fiscal schedules were apparently in use forAsiana and Pontica: CT 7.6.3 (a.d. 377). See in general Jones, op. cit. (n. 2), 50.93The Astypalaia inscription included a column of 'human capita'. But the figures recorded in this category includebizarre fractions like 1/10, 1/200, 1/300, and 1/750. See Thonemann, op. cit. (n. 1), 477-8.94 Imultiply the capita by 1-2, to reflect an all-adult male or all-adult female population. The truth should besomewhere in between, unless large numbers of children were included as small fractions.95M. Spurr, Arable Cultivation in Roman Italy, c. 200 B.C.?c. A.D. 100 (1986), 133-46; R. Duncan-Jones, TheRoman Economy: Quantitative Studies (1974), 327; Jones, op. cit. (n. 2), 56.


17/37


registeredabour nAstypalaia falls ithin therange f sufficiency.n this state there slittle eed to adduce thepossibility f extensive easing thatdid not include the landowner s responsibilityorhead taxes.Similardata can be extractedfrom ralles, although nthis ase there remore variables.The advantage f the ralles inscription,ike that tAstypalaia, is that e are ableto see the tax liability oran owner s entire roperty in this ase, four wners.ButunlikeatAstypalaia, the labelling f human capita isnot limpid. ome linestotal thecapita of animalsand slaves ,others thecapita of animals , and othersplain capita.Assuming, s did Jones, hat hefirstepresentslaves, the econd representsnimals, ndthe ast representsaroikoi,wewould see thefollowing esults.96

TRALLES SLAVES PAROIKoI LANDFulvius 0 4-8 39 iugeraTatianus 8-i6 56-iiz 6i6 iugeraKritias 2-4 9-I8 z5i iugeraLatron 4-8 0 205 iugera

The data from ralles yield interesting,funcertain,results.With theexceptionofLatron s property, ach landowner as suppliedwithenough abourtobe atornear sufficiency,ithout theneed forunregisteredabour.Latron, perhaps,rented omeof hisland, nd itiseasy to imagine hathis slavesmay havebeennotonlyworkers,but effectivelygents rmanagers.97atianus can claim tohave the statewith themostnon-slavelabourersnanyof theGreek censusinscriptions,ut itisnoteworthyhat venhe couldclaim arounda dozen slaves,perhapsaspartof a permanent oreworkforce a phenomenon attested lsewhere n ateAntiquity.98TheMagnesia inscriptionnce again represents challenging,etsignificantource finformation.his documentrecorded abour nly infiscal nits capita),withoutspecifying the tatus f theworkers,norwhether nimalswere included. t revealsmeasuresoflabourdensity hich at firstlanceseemtoreflect structuralhortage f labour.At thesame time, his ocumentreveals pattern hich offersmportantlues to thedynamicsof registeringabour in the ate Empire.AtMagnesia thetaxdueswere organizedgeographically, o we do not know thewhole tax liability orany individual wner.Theinscriptionsake itclear thatregisteredabourwas not spread evenly cross the land.99As we might expectwith scattered oldings, twas normal to concentrate abourers ncertain lotswhich functioneds domicilesfortheworkers.Jones alculated round263iuga ndzIz capita forthe artsof the nscriptionnwhichbothcategories revisible. 00This indicates lowerdensity f registeredabour than lsewhere. ossiblythis s a fluke(ifwe lackregisterediteswhere theworkforcewas domiciled), r possiblythere as ashortage f rural abouratMagnesia. But perhapssome landowners tMagnesia leasedland tovillagers ndpeasantswithoutassumingresponsibilityor heirapitation axes.

96 Some entries recorded ?o v K(eaA.ai), others oo?taov Kai ?obcov K(eaA.a?), others simply K(e


18/37

100 KYLE HARPER

Significantly,heMagnesia inscriptions the nly example inthe ntire ossierwherenumerouspetty landholders are recorded. The inscriptionsfromLesbos, Tralles,Astypalaia and Thera are heavilybiased towards largecurial-scaleproperties. erhapssmall andmiddling landownerswere themost likelyto lease their mallplots in thecountrysideovillagers ndpeasants. Ifsuchvillagers rpeasantsowned theirwn land,then he rban landlord as notresponsible or he apitationtaxes, nd the abourwouldnot be registerednderthe andlord s iability.101nsupport f this econstruction,iabilityfor apitationtaxes atMagnesia was stronglyoncentratedn three ypes f land: thelargestroperties, hefarms fcitizens ith high tatus, nd thefarmsf landowners romother towns, howerepresumably arge andowners.102hiswould arguethatfree, ontract tenancy nd the fiscalregistrationf dependentworkers co-existed,but on twoplanes. There was botha sector fmiddling landlords easing xtra fields o rural mallholders and a level f larger andowners ith their wn (fiscallydependent that is,registered)abour force.The inscriptionshat llowcomparison etweenregisteredorkers and registeredanddo not seemtoshowan acuteshortage f labour.103hat the ensus nscriptions eveal sa countrysideiven ver to intensiveolycultural xploitation n elite-ownedand,with alabourforce onstitutedf a mix of tenants, laves,and free ependents. f course, theregistrationf free ependents paroilkoi) inrelationshipshat ncluded he andlord sresponsibilityor apitationtaxes does appear as awidespread practice inthisregion.Most of the roperties ecordedrecurial-scale ortunes,ndonlyatMagnesia do we geta fuller icture thathelps us understandhow smallerowners could continue to usecontract-leasingvenas large wnersbecame increasingly esponsible orthe apitationtaxes f their orkers.This divergence elpsus tounderstand ow large andowners ayhave been able toexploitpublic lawsrestrictingiscalmovementfor heir wnbenefit.104InJones s nalysis,the hortage f labourwas one partof a story hosemain dramawas therise f the olonate.The colonatenotonlyledhimtounderestimatehe ossibilityof non-registeredabour, it seems to have encouragedhim to downplay the strikingevidencefor laverynthese ocuments.105he discoveryf thenewfragmentromhera,attestingver i50 slaveson a single state,demands thatwe paymore attention o thepositiverole of slavery n this ample.Moreover, theRoman colonatehas comeunder

101CT 11.I.14 (A.D. 371).102This pattern is difficult to quantify precisely, but the distribution is overwhelming. Among the twenty-onelargest registrations of capita (which account for nearly all the total capita), only three owners were apparently notsenatorial, curial, from other cities, or owners ofmultiple estates.103The only other place where Jones found a shortage of labour was no. 79 from Lesbos. In this inscription, fourof the properties included a personal name underneath the heading for each district. This is the only stone with thisformat, and other districts on this stone do not list names, making itprobable that all of these properties belongedto one owner. The named individuals were probably his actores, conductores, or tenants. This stone alone recordsland of different classes, a clue that the taxes on these particular farms needed to be carefully specified. The namesprobably signal a tax assignment separate from legal ownership ? cf. Bowman, op. cit. (n. 6), 142 not the totallabour on the farm.104See especially J. Banaji, Lavoratori liberi e residenza coatta: il colonato romano in prospettiva storica , in LoCascio, op. cit. (n. 11), 253-80; D. Eibach, Untersuchungen zum sp?tantiken Kolonat in der kaiserlichenGesetzgebung: unter besonderer Ber?cksichtigung der Terminologie (1980).105Jones, op. cit. (n. 2), 57, claimed that, excluding the gangs on Lesbos, slaves were 12-13 Per cent o? the ruralpopulation. Slavery at 12-13 Per cent of the rural population would be a remarkable figure and this excluded thetwo samples available to him where slavery was most prominent. Careful study of the Roman slave system hasdownsized the plausible extent of slavery, so the discovery of prominent agricultural slavery in this region of theEmpire (outside the traditional heartland) is immediately striking. See W. Scheidel, Quantifying the sources ofslaves in the Early Roman Empire , JRS 87 (1997), 156?69; W. Scheidel, The slave population of Roman Italy:speculation and constraints , Topoi 9 (1999), 129-44; Scheidel, op. cit. (n. 3); Jongman, op. cit. (n. 89), 100-22.


19/37

THE GREEK CENSUS INSCRIPTIONS OF LATE ANTIQUITY I0I

withering ritique s a paradigm for ate Roman labour relations.106na seminal rticle,Carrie showed thatthe egislationn coloniwas fiscal nnature and couldnotanchor anarrative f transitionfrom' lavery to' tiedtenancy s thedominant abour system.107His work allowed scholars to recognize hedeep continuities n land-leasing,nd it ledultimately o the striking ecognitionhat oloniwere not only tenants, ut oftenpaidestateworkers.'08fnoteveryone as agreedwith themore radicalversions f therevisionist iew,a generationf critical cholarshipn the olonatehas at least repared s for hediversityf the abour systemttested nthe nscriptions.The census inscriptions ecord iability orcapitation taxes on two classes of rurallabourers: laves nd paroikoi.Throughout have translated aroikoiwith the tudiouslyvagueword 'dependent',n rdertoavoidprejudging ow these abourers ereorganized,for nstance s tenants r estateemployees. n a strict ense, the nscriptions ellus onlythat hey ere of free egal tatus nd that he andowner as responsible or heirapitationtaxes.'09fthe nscriptionsate tothe ensusofA.D.37I, then t spossible to saywithcertaintyhatthesewere labourers ho ownedno landof theirwn and who were thusinscribed n thefiscal egisternder the andowner's state.The inscriptionsnlyrecordregisteredabour a fact hich immediatelyisqualifies he ffortfJones o extract heproportion f slaves nd tenantsntherural opulationfrom hese ocuments. oreover,it is imperative o rememberhat ur documents re biased towards theupper tiers flandedwealth.They are thusnot representativef thecountrysides awhole. Nevertheless,t ispossible toobtain a controlled mpressionf the rganization nd status flabouron elite-owned and,and thesefindingsorce s to reconsiderstablished eliefsabout therole f slave labour nthis eriodand region.Three of the ensus inscriptionsreserve ata about the egal status f the abourers:Tralles, Lesbos, andThera. 0 From Tralleswe have a small and extremelyroblematicsample,preserved n terms f fiscal nits.Jones rgued thattheratioof free enants oslavescould be extractedfrom hisdocument. he ratio is5:i, so that laveswould bearoundI7 percent f theregisteredabourforce. nfortunatelyhenumbersmay be compromisedby the nclusion f animals, nd, asThonemann has argued, t isnotclear thatthecategory f plain capita necessarilyrepresents reeparoikoi.The glimpsesof thecountrysidetTralles areonlyuseful ualitatively.he largefarm wnedbythe ecurionTatianuswaswell-equippedwithworkerswhowere likely ree, ut employed n the rderofeightto sixteen laves.Fulvius, priest, eld a smallfarm hichwasworkedbya few,apparently ree abourers. ritiashad perhaps two to four laves tocomplement largerstaff f freeworkers.Latron had fourto eightslaves but no freeworkers.This smallsample,with all of itsuncertainties,t leastpointstoa diverse aboursystem,ntermsfits wnership atterns nd the tatus f itsworkers.106 Selectively: Eibach, op. cit. (n. 104); J.-M. Carri?, 'Le colonat du Bas-empire : un mythe historiographique?',

Opus 1 (1982), 351-70; D. Vera, 'Forme e funzioni d?lia rendita fondiaria nella tarda antichit?', inA. Giardina (ed.),Societ? romana e impero tardoantico (1986), vol. 1, 367-477; D. Vera, 'Padroni, contadini, contratti: realia delcolonato tardoantico', in Lo Cascio, op. cit. (n. 11), 185-224; W. Scheidel, 'Slaves of the soil', JRA 13 (2000),727-32; D. Kehoe, Law and Rural Economy in theRoman Empire (2007), especially 163-91; Grey, op. cit. (n. 12);B. Sirks, 'The colonate in Justinian's reign', JRS 98 (2008), 120-43.107Carri?, op. cit. (n. 106).108For continuities in land leasing, see the work of Vera, op. cit. (n. 106). For the recognition that not all coloniwere tenants, see Banaji, op. cit. (n. 4), 209?11; Sarris, op. cit. (n. 4), 128-9; 0/ 11.48.19 (Anastasius).109 See Sarris, op. cit. (n. 4), 151; Kiourtzian, op. cit. (n. 1), 225. For a fourth-century definition, Basil, Homiliaesuper Psalmos 14.1 (PG 29, col. 252): TQarcfipo? Tc?poiKoi, ?Moxp?av ?K|xia0o?^i?voi yfjv, Ttpo? x? ?oOATina xo??K???cokoxo? YficopyoOai xf]v xc?pav.110There are other important clues that slavery was regionally significant. A proprietor atMagnesia, whoseproperties were named Barbaria and Barbariane, held little land but enormous capita liability. Thonemann, op. cit.(n. 1), 474, has made the suggestion that these were slaves used to quarry emery, an important local resource. InChios two census blocks survive. The second block listed the owner's liability in at least tendifferent villages. Undereach village the liabilities were listed in iuga and 7iap(o?K(ov) K(?a?uxi), but also in oou^(cov) K(?(|>aA


20/37

102 KYLE HARPER

With theevidencefrom esbos and Thera, recorded n rawhead counts,we are onfirmerround.The fragmentarytateof the vidence thwarts eat ratiosof free o slavelabour,but the tonesdemonstrate nequivocally hat lavery layeda structural oleonelite-owned and in thisregion. ll four f the tonesfrom esbos are incomplete,owelack the total liability or ny landowner. his iswhy Jones removed he videncefromLesbos in alculating he verall ratio f slaves tofree orkers.Though itis impossible osaywhat proportionof the labour forcewas composed of slaves, it is important oacknowledgethe oncrete estimonyorrural lavery nLesbos. Two of the four tonesconserve nformationbout the abourforce. he first tonerecords and insixteen lotsbutmentions labouron onlyone. 11The farm n that ocalewas sizeable:9i iugera farable land, zo iugera f vineyard, 52 olive trees, and forpasture, twenty xen, fiftysheep, nd twenty-twolaves.In the therdistrictsisted n this tone, the wnerhad notax liabilities or abourers, laveor free. t is likely hat omeofhis landwas worked bythe slaves, some letout to tenantsresponsiblefortheir wn personal taxes, or someworked by labourers ecorded n the ostpartof the tone.The only certaintybout thislandowner s abourforce s that t ncluded gang of twenty-twolaves.112Anotherstonefrom esbosmentionsproperties nfive ifferentlots. Inone of theseplaces, the wnerwas liable fortaxes n twenty-onelaves.113et thefarm nthis istrictincluded nly 5 iugera f arable land,somevines, nd 132olive trees. twas a smallplotwhichmayhavedone little ore than ouse the laves, ho presumablyorkedon the therfarms.fwe cannot xtract atios ffree oslave abour nLesbos,we shouldtryo stimate,as awayofassessing hether he laveswere integralrmarginalto griculturalxploitationinthis ample, hat proportion f theregisteredand the lavescould haveworked.Usingthemanning-ratiosfDuncan-Jones n thefarms ecorded n GXII.Z, no. 76:

LABOUR REQUIREMENTS, PROPERTY i, LESBOSRAW AMOUNT MANNING-RATIO REQUIRED LABOURERS15I4 iugerarable 25 iugeraerman 6o.6I19 iugera ine 8 iugeraerman I3.6IIO iugeralives 25 iugeraerman 4.478.6 labourers

There are a number f complications ith this xercise. t ispossible thatfallowarablewas countedin theregisterf taxableproperty. 14oreover, themanning-ratios ely ntreatises iscussing imaginarymonocultural farms.115olyculturewas thenorm forMediterranean farming, strategyhich reduced limatic ndmarketrisk ut alsomade111 IG XII.2, no. 76, c-d.112This stone presents another interesting problem. In four places, individual names, listed with animals, arerecorded between villages. Jones thought thesewere owners of animals grazed on a third party s land. But it ismorelikely that they were servile or dependent shepherds. Their names, Elpidephoros, Cyzicius, Philodespotos, andAristotle, are consistent with a group of slaves, though names are rarely probative. Aristotle had a patronymic, sopossibly he was free or freed and the others slaves. Philodespotos is as convincing as name evidence gets. Cf. Bagnall,op. cit. (n. 45), 126. Parallel evidence for slave shepherds in Late Antiquity is not lacking, e.g. Augustine, ContraCresconium 1.30.35 (Ed.M. Petschenig, CSEL 52.2 (1909), 355); Julian, Orationes 7.22 (Ed. Rochefort (1932-1964),vol. 2, 75). Archaeological evidence for estate-based pastoralism inNorth Africa: B. Hitchner, Image and reality:the changing face of pastoralism in theTunisian High Steppe , in J. Carlsen (ed.), Landuse in the Roman Empire(1994), 27-43.113 IG XII.2, no. 78, c114On the taxation of fallow land, Bagnall, op. cit. (n. 45), 116. The register from Lesbos, however, uniquelyrecorded pasture. On the relationship between fallowing and pasture, see Grey, op. cit. (n. 90), 369-70. Cf. CT9.42.7 (a.D. 369), a complete fiscal valuation of a confiscated estate included what was currently and had been

previously cultivated.115Duncan-Jones, op. cit. (n. 95), 327, is the best discussion.


21/37

THE GREEK CENSUS INSCRIPTIONS OF LATE ANTIQUITY I03

efficientse of the alendar.116inally,themanning-ratios ssumeworkingmen,whereasthe slave counts includedwomen. But it is likely hat slavewomen were worked hard,because slavestatus verrode ulturalnormsofgendered abour.This has been amajorsource f efficiencyor lavery.117actoring ll these onsiderations, hetwenty-twolavesof this andownermay have been able tohandle, roughly, quarter to a third f thenecessary ork on the isible registeredand.The calculations an be repeated or G XII.z.78:LABOUR REQUIREMENTS, ROPERTY 2,LESBOS

RAW AMOUNT MANNING-RATIO REQUIRED LABOURERS52 iugerarable 25 iugera erman z.o8I5 iugera vine 8 iugera perman I.885 iugera olives 25 iugera perman o.z4.i6 labourers

By any estimate, he twenty-onelaveson this roperty rovided far toomuch labourfor thepartof the wner'sholdingswhich arevisible.Most likely he wnerheld otherfarms hich were lost in thedamage to thestone, nd it is evenpossible that thesameownerheld all thefarmsn bothstones.We canonly saythat lave labour, na smallbutrandom ample, ppears structurallyntegral o the xploitation f elite-owned and onLesbos.The inscriptionsrom hera are thefinal etwhichoffers nformationbout the tatusof agricultural abourers. nhappily,these nscriptions efy uantificationn thesameway as thosefrom esbos: they re fragmentary,owe do notknow the total iabilityfany landowner xcepttheheirs fParegorius, ho had twoslaves,three ree ependents,and probably leased someof their and.Four othersignificantnscriptionsurvive, utaggregate atios f free oslave labour reunattainable.THERA LAND FREE SLAVEI4zb 6I4 iugera arable ?

i6z iugera vinei,4z0 olive trees

I42C 528 iugera arable ? ?I20 iugera vine586 olive treesI42d i6

New ? ? I52+

Although t is impossible o saywhat percentage f the total abour force nTherawascomposedofslaves,thenew inscriptions,byanyreckoning,emarkable.he attestationof over I5z slavesbelonging o a single wnerranks s themost concrete,redible rtefact116Duncan-Jones, op. cit. (n. 95), 36?7. Even in the U.S. South, cash-crop farms like cotton plantations allowed lessthan half the year to cotton production. See G. Wright, Slavery and American Economic Development (2006), 86.117Scheidel, op. cit. (n. 89), 213. On the efficiencies of using female slave labour in theNew World, see J.T. Toman,'The gang system and comparative advantage', Explorations inEconomic History 42 (2005), 310-23. Cf. Stobaeus,

quoting Hierocles (second century A.D.), Anthologium 4.28.21 (Ed. O. Hense and C. Wachsmuth (1894-1912), vol.5, 699): cocjxE \ir\xfj? xataxcjia? koivcove?v ii?vov xa?? 0?pa7caivai?, ?XXa Kai x v aXX v ?pycov x v?7tav?pox?pcov.


22/37

104 KYLE HARPER

of large-scale ural lavery rom he ntire oman Empire.This property usthavebeenlargerthan the farms ssociatedwith gangs of slaves on Lesbos, themixed estate ofTatianus atTralles,or the arge enatorial omainatMagnesia. The image f the ountryside from hera,where large andowners pecializedinwine production nd employedvastnumbers f slaves, dds evenmorediversityothe abour ystemttested nthe reekcensusinscriptions.Despite their imits, he ensus nscriptionsestify,ith a highdegree f credibility,hatslaverywas structurallymportant o theelite land-usestrategyf thefourth-centuryaristocracynthis egion. he presentationfJones, howas, tobe fair, orkingwithoutthe enefitf the ramaticnew fragment,as allowed someof themostvaluable evidenceforrural lavery n theRoman period tobemarginalized.Finley,Whittaker, and othershave long mphasized hat he iteraryvidencefor lavery n the ate Roman countrysideis as abundantand credible s for ny period of antiquity. 'The evidence for lavery nthe astern editerraneanhas beenparticularly eglected. variety fauthors, ncludingBasil ofCaesarea, Libanius,andJohn hrysostom, ake plausible statements o the ffectthat gricultural lavery as importantnthefourthentury.'19t the east,the nscriptions uggest hat he bservations f these uthorshad a basis inreality. s documentaryevidencefor gricultural lavery, here s little nRoman historythatcanmatch theseinscriptions.Ifonewanted tomarginalizethe videncefor lavery ntheGreek census inscriptions,itmight be possible to insist n thebias of the ample towards slands. t ispure accidentthattwoof the nscriptionsithgood data about labourare from slands, ut it is interesting. slands could foster hemicroeconomicparameters f a slaveestate.120rom theslavemanager's perspective,n islandmighthave reducedthe risk f flight. 9 km long,Thera was a natural cage for the slaves forcedtowork its land. Secondly,an islandenvironmentould aggravatethe risks f the pen labourmarket.With a limited atchmentarea,an island ouldhavean inflexibleupply fextra abour. lavery hus nsulatedthe andowner gainsttherisk nd transaction osts of findingabour.12'f course,thesefeaturesreemphases, otexceptional raits,eterminedyisland cology. editerraneanislands renotabnormalsocialhabitats, ndLesbos, after ll, is at i,6ookmznot a smallisland.122twould be unwarranted opressthe ase of island xceptionalism eryfar.Suchquibblesshouldnotdiminish hefact hat he nscriptionsnhance urknowledgeof how slavery ontributed oRoman agriculture.n theproperties ocumented, laverytook a variety f forms: few slaves on smallfarms,mid-sizedgangs in singleunits, apermanent ore ofworkerson a farm orkedmostly by free abour, nd truly ast slavebased estates. he prominence f slavery mphasizesthatthe ateRoman labour systemcannotbe described nterms f a struggleetween laverynd tenancy s two competingmodes of production.123his claim builds on Carrie's original insight hat thecolonate118C. Whittaker, 'Circe's pigs: from slavery to serfdom in the later Roman world', inM. Finley (ed.), Classical

Slavery (1987), 88-122; M. Finley, Ancient Slavery and Modern Ideology2- (1998; orig. 1980), 191-217.119John Chrysostom, InMatthaeum 63.4 (PG 58, col. 608); John Chrysostom, InMatthaeum 24.11 (PG 57, col.319); John Chrysostom, In acta apostolorum 32.2 (PG 60, col. 237); Basil of Caesarea, Homil?a in divites 2.2 (op.cit. (n. 98), 46?7); Basil, Homil?a in illud: Attende tibi ipsi 5 (Ed. S. Rudberg (1962), 31); Basil, Homil?a inmartyremJulittam 1 (PG 31, col 237); Libanius, Or. 62.46-8 (Ed. R. Foerster, Opera (1903-8), vol. 4, 370); Libanius, Or. 47.28(Ed. Foerster, vol. 3, 417-18); Libanius, Or. 14.45 (Ed. Foerster, vol. 2, 103).120cf. B. Shaw, ' AWolf by the Ears : M. I. Finley's Ancient Slavery and Modern Ideology in historical context',foreword toM. Finley, Ancient Slavery and Modern Ideology* (1998), 3-74, at 18.121For a vivid illustration of a desperate landowner hunting harvest labour (in the form of hired slaves) in fourthcentury Egypt, see P. Lips. m. Cf. Wright, op. cit. (n. 116), 117-19.122P. Horden and N. Purcell, The Corrupting Sea: A Study ofMediterranean History (2000), 224-30, 390.123On modes of production, see J. Banaji, 'Modes of production in a materialist conception of history', Capital andClass 2 (1977), 1-44. Cf. C. Wickham, 'The other transition: from the ancient world to feudalism', PP 103 (1984),3-36; C. Wick

harper - 2008 - the greek census inscriptions of late antiquity

Documents