Hobson Women as Property Owners in Roman Egypt

American Philological Association

Women as Property Owners in Roman Egypt Author(s): Deborah Hobson Source: Transactions of the American Philological Association (1974-), Vol. 113 (1983), pp. 311321 Published by: The Johns Hopkins University Press Stable URL: http://www.jstor.org/stable/284018 . Accessed: 25/07/2013 16:52 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

American Philological Association and The Johns Hopkins University Press are collaborating with JSTOR to digitize, preserve and extend access to Transactions of the American Philological Association (1974-).

http://www.jstor.org

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Transactionsof the American Philological Association 113 (1983) 311-321

WOMEN AS PROPERTY OWNERS IN ROMAN EGYPT* DEBORAH HOBSON YorkUniversity

thatwomenin RomanEgyptseemto It has longbeen acknowledgedl but thispotentially sighave owneda fairamountof privateproperty, the realmof hardlytranscended facthas, untilveryrecently,2 nificant Althoughnumerousscholarshave examinedand passingobservation.3 of theEgyptianpopulationduringthe discussedthe ethniccomposition * An earlier versionof this paper was presentedat the 1980 meetingof the American PhilologicalAssociationin New Orleans. Preliminaryworkon thissubject was carried out during a very enjoyable year spent at the Institutefor Advanced Study in 1978/9. My interestin discoveringand definingthe limits of female propertyownershipin Roman Egypt emerged from numerousentertainingand profitableconversationswith Professor J. F. Gilliam, to whom I dedicate thismodesteffortwithaffectionand gratitude.I should also like to express my appreciation to the two refereeswho read this article for TAPA, whose constructivecriticismsI have triedto take account of in what I presenthere. Papyrological abbreviationsused here are those found in Checklist of Editions of Greek Papyri and Ostraca, BASP Supplement 1 (1978). 1 See, e.g., A. C. Johnson,Roman Egypt to the Reign of Diocletian (Baltimore 1936) 28, 31, 227f.,682 note 5. 2 Since I delivered the firstversionof thispaper, S. Pomeroyhas published"Women in Roman Egypt: A PreliminaryStudy Based on Papyri,"Reflectionsof Women in Antiquity,ed. H. Foley (London 1981) 303-22. 3 See, e.g., C. Pr6aux, "Le Statutde la femme a l'6poque hellenistique,principalement en Egypte," Recueils de la societe Jean Bodin 11: La femme (1959) 174, and the references cited therein.Pr6aux's focus is, however,more juridical than sociological,and her studydoes not thereforeencompassthe concernsof the presentpaper. A hopefulindicationof a growingawareness of the ownershipof propertyby women in Egypt is perhaps suggestedby a comparisonbetween two treatmentsof one document, the firstpublishedthirtyyearsago and the second in 1978: in a discussionof P. Flor. I 71, a large fourth-century land registerfromthe Hermopolitenome, A. H. M. Jones,"Census Records of the Later Roman Empire,"JRS 43 (1953) 58-63 makes only one allusionto the fact that there were numerousfemale landownerson the list,when he-refers(58) to the quantityof land which "he or she" held. Elsewhere throughouthis article landownersare always referredto as "the man who . . ." On the contrary,in a recentrepublicationof this registertogetherwith P. Giss. Inv. 4 (= Zwei Landlisten aus dem Hermopolites,ed. P. J. Sijpesteijnand K. A. Worp [Amsterdam1978]), the sex of the landownersis noted,pp. 23f. (334 men and 57 women).

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

312

Deborah Hobson

Greco-Roman period in order to determine the extent to which the countrymay have been Hellenized or Romanized,4there has been remarkablylittleinterestin the question of how extensivelyit may have been "feminized."5The truthis that since women were able to derive propertyfromtheirfamiliesthroughdowry and inheritance,6 but were not liable to the poll tax nor to the same range of liturgicalfunctionsas men,7their capacity to acquire propertywould seem to have been in theoryat least as great as that of men, while their potentialliabilities may well have been considerablyfewer.8 As far as I know, no one has ever explored the implicationsof this situation by making a detailed study of the economic activities of women withina particularcommunity,althoughthe papyri provide us withan abundance of relevantevidence forthiskind of an investigation. The presentpaper representsan initial attemptto rectifythis omission, by a considerationof the economic positionof women propertyowners in the small village of Socnopaiou Nesos, a village at the westernborder of the Fayum oasis which existed from about 250 B.C. to A.D. 250. As source materialforthisinvestigationwe have about 950 publishedGreek papyri of the Roman period9emanatingfromor relatingto thissite,of 4 On this subject see D. H. Samuel, "Greeks and Romans at Socnopaiou Nesos," Proceedinigs of the XVI Congress of Papyrology = American Studies in Papyrology 23 (Chico, Calif. 1981) 389-403, esp. the referencescited at p. 390, note 5. 5 The basic literatureon the subject of women in Greco-Roman Egypt is cited by Pomeroy(above, note 2) 318, note 2. 6 See R. Taubenschlag, The Law of Greco-Roman Egypt in the Light of the Papyri, 332 BC-640 AD (New York 1944) 150f. 7 Oni the exemptionof women fromliturgicalservice see N. Lewis, "Exemptionsfrom Liturgyin Roman Egypt,"Actes du Xe Congres Internationalde Papyrologie,VarsovieCracovie, 3-9 Septembre 1961 (Warsaw 1964), 70 and note 5. See also BGU II 648.92-94, P. Tebt. 327 and P. Oxy. VI 899. 8 Thus Johnson(above, note 1) 28 remarksthat "therewas a tendencyforprivateproperty to pass into the hands of women, who could not be appointed for liturgiesnor be assigned land for forcedcultivation."On this point see Pomeroy (above, note 2) 313 and note 20. 9 The Ptolemaic papyri fromSocnopaiou Nesos are too few in numberto be taken into accouintforthe purposesof thispaper; Pomeroy(above, note2) 304ff.attemptsto establisha change in governmentpolicyaffectingwomen and theircapacity to own propertyfromthe Ptolemaicto the Roman era, but her argumentis (by her own admission)largelyex silentio and does not seem to me to be substantiatedby the small amount of evidence we do have fromthe Ptolemaic period. The fact that land is not included in the provisionsof dowries b)eforethe Roman period (as pointedout by Pomeroyat 304) is notsignificant in lightof the factthatwomen are given land in willsat least as earlyas 238 B.C. (see below, note 32). The phenomenonPomeroyis concernedwith(i.e. increasedownershipof privateland by women in the Roman period) would have been true of men as well, as Johnsonpointsout ([above, note 1] 27). However, the focusof the presentpaper is on the Roman period,and it is not withinthe scope of thisinvestigationto make any comparisonsbetweenthe Ptolemaicand Romaneras,particularlyin theabsence ofsufficient relevantevidence.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Women as PropertyOwners

313

which a total of 271 documentsprovide us with some kind of information about ownershipof propertywithinthe village.10From an analysis of thesetextswe can obtaina fairlyclear pictureof the role of women in the economic lifeof the village." In this paper I shall firstdiscussthe resultsobtained froman examination of the documentsfromSocnopaiou Nesos, and I shall then place these specific resultsfrom one small village into the larger contextof Egypt as a whole by consideringthe patternsof propertytransmission whichiare embodied in the provisionsof wills fromthe Roman period. In this way I hope to be able to clarifythe role of women property ownersin the economyof Roman Egypt. Socnopaiou Nesos was in many ways not a very typical Fayum village. Since it was the center of an importantcult of the crocodile god Souchos,therewas a large priestlyclass in the village,among whom the women membersfiguredprominently.Because the village was situated at the extremewesternedge of the cultivatedarea of the Fayum and at the end of the irrigationsystem,agriculturalconditions were poor, which almost certainlyexplains why it appears to have been of so little interestto the Greeks and Romans.12In fact our evidence indicatesthat the village had no privately-ownedagriculturalland at all,'3 although some residentsowned a certainamount of land in othervillages.14Since the village was at the edge of Lake Moeris and on the border of the desert,there was a customshouse located there which regulated commercial trafficpassingin and out of the nome, and the village seems to have provided transportationservices for goods passing through this customshouse.15 10 It is indeeda sobering experiencefora classicist, whofeelstemptedto reconstruct a societyfromwhatmightseem like the vastquantityof documentary evidencerepresentedby 950 Greekpapyri,to contemplate theenormous amountof materialavailable forsimilaranalysisto scholarsof laterperiods;see, forexample,David Herlihy,"Land, Familyand Womenin Continental Europe,701-1200,"Traditio18 (1962)89-120. 11 My use of theterm"economicrole"of womenrefers throughout to theirfunctioning in theworldbeyondtheirimmediatehouseholds. Cf. D. M. Schaps,EconomicRightsof Womenin AncientGreece(Edinburgh 1979)esp.p. 2. 12 See Samuel(above,note4) 400f. 1- In P. Gen. 16 = W. Chr.354 (A.D. 207),a groupof publicfarmers fromSocnopaiou Nesos,in petitioning againstan injustice theyhavesuffered at thehandsof someviolent usurpers of theircommunalland,specifically statethattheirvillagehas no land other thanthepubliclandalongtheshore(alylaAoq) .EVfE[aTOu]

7 'EXtvr K'vKIA7TV

,ur18e elbfav... a`XArnv

/.L74TEL18L[w]TcK' /.LA)rEf 8acr[Li\]cnv

(16f.) 14 On thispoint,see D. W. Hobson(Samuel),"The Villageof Apias in the Arsinoite Nome,"Aegyptus 62 (1982)80-123. 15 The importance of transportation servicesprovidedby residents of SocnopaiouNesos is reflected in thelargenumberof documents fromthevillagewhichrelateto ownership ofcamels;see below,note22.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

314

Deborah Hobson

Because of its particularlocation, therefore,the economy of Socnopaiou Nesos had a distinctivecharacteramong Fayum villages,in that it seems to have been based on commerce ratherthan agriculture.For this reason propertyownershipis not manifestedin the usual formof possessionof agriculturalland, as it would be for other villages, but ratherin termsof possessionof houses,camels,slaves,and capital.16 Thus when we set out to examine the extentto which women appeared as propertyowners in Socnopaiou Nesos, we are actually looking at the extentto which they owned houses,camels, slaves, or money,whereas, for example, in the case of Karanis we are able to evaluate women's propertyby calculatingtheirland holdingsas these are reflectedin the land tax paymentsrecordedin the village tax rolls(see below, p. 315). When one charts the extent to which women figureamong these various categories of property,one finds that they appear most frequentlyas ownersof housesor partsof houses.There are severaltypesof document which record house ownership,and in each of these women are stronglyrepresented.Among the 32 documentsrecordingsales of houses'7 there are 35 women principals as compared with 36 men. Among twelve documentsrecordingregistration of property,'8thereare eight women propertyowners and 17 men.19Of twelve extant census declarationsfromSocnopaiou Nesos20.only one (P. Rein. I 46) is thatof a woman householder,but the women listedas membersof the declarants' households in the other census returnsown an impressiveamount of property.A particularlynoteworthyexample is that found in P. Grenf. II 55, where the declarant is a 25-year-oldman who has a 13-year-old wife who owns 21/4houses, and another 12-year-oldgirl living in the same household (presumablythereforea relative) owns 21/2houses. Yet the male head of this householdlives in a house which he stateshe has inheritedfromhis mother,and he apparentlyowns no otherproperties. From these disparate pieces of evidence, spread as they are over a time period of some two hundred years,it is hard to arrive at an estimate of the percentageof houses owned by women in thisvillage; however, if one correlatesthe figuresfor home ownership,where almost as many women appear as men, with the figuresforpropertyregistrations, where about one-thirdof the registrantsare women, one can, I think, 16 Pomeroy (above, note 2) does not deal with these formsof propertyownershipand appears to regard land ownershipas the only significantindex of the economic statusof women in Roman Egypt. 17 For a listof these thirty-two documents,see Samuel (above, note 4) 392, note 9. 18 For a listof propertyregistrations fromSocnopaiou Nesos, see Samuel (above, note 4) 393, note 10. 19 These totalsreflectthe total numberof people, withoutregardforthe numberof individuals involvedin each document,or the distinctionbetween buyerand seller. 20 Socnopaiou Nesos census declarationsare listedat Samuel (above, note 4) 393, note 11.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Women as PropertyOwners

315

suppose that at least one-thirdof the village real estate may have been owned by women. A comparable figureis yielded by looking at the names of owners of contiguouspropertiesfound in the thirtyproperty sales fromSocnopaiou Nesos which contain such information:41 names are male and 16 are female. This figureof one-third,as a bare minimum and encompassinga long chronologicalspan, can be corroboratedby the evidence provided by nearby Karanis, a much larger and more flourishingvillage. The Karanis tax rolls fromthe years A.D. 171-74, published in P. Mich. IV, list people who paid taxes on privateland in thosethreeyears;although the textswere published withoutcommentary,and thereforeany interpretationof the data providedby them is necessarilysomewhathazardous, it is worthnotingthat almosttwo-fifths of the names which appear among those paying taxes for privately-ownedland are names of women. This statisticis usefulfortwo reasons:first,since the Karanis tax rolls constitutea comprehensive list from a restrictedtime period, they provide a more accurate estimate of proportionsof women property ownersthan any of the data fromSocnopaiou Nesos,given thatthe latter embrace such a large time period. Second, the numberof women land owners listed in the Karanis rolls points to the fact that Socnopaiou Nesos, with its apparently idiosyncraticeconomic conditions,was not atypical in containinga high number of female propertyowners,and furtherthat women's propertywas not limited to village house ownership, but that women owned agriculturalland to at least as great a degree as houses.2' Women's names occur oftenin connectionwithotherformsof property as well: among 60 identifiedcamel owners in Socnopaiou Nesos, twelve are women, which is a figureof one-fifthof the total.22Out of seventeen attested slave owners, eleven are women, which represents almost two-thirdsof the total in that category.23Thus the evidence for 21 Cf. 0. Montevecchi,"Ricerche di sociologia nei documentidell'Egitto greco-romano. III. I contrattidi compra-vendita,"Aegyptus 21 (1941) 144: in all contractsfor sales of housescollected and discussedhere, thereare 195 men and 133 women. Thus 40% of contractingparties in house sales are women, a figureentirelyconsistentwith the evidence presentedin thispaper. 22 There are three kinds of documents which provide informationabout ownershipof camels: sales of camels, receipts for payments of the camel tax, and registrationsof camels. A complete listof these threekindsof documentsfromSocnopaiou Nesos is given at Samuel (above, note 4) 394, note 15. The followingdocumentsfromthat list involve female camel owners:BGU I 87, BGU I 88, BGU II 416, P. Amh. 11 102, P. Lond. II 333 (p. 199), P. Grenf. II 45a, SB VI 8977, P. Lond. II 304 (p. 72), BGU III 869, BGU I 266, P. Monac. gr. inv. 26. 23 A complete list of Socnopaiou Nesos slave owners is given at Samuel (above, note 4) 394, note 14. The followingdocumentsfromthat listattestfemale slave owners:BGU III 805, SPP XXII 36, BGU III 855, P. Lond. II 311 (p. 219), SPP XXII 40, P. Lond. II 325a

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

316

Deborah Hobson

ownershipof houses,camels, and slaves all pointsto the factthatwomen in Socnopaiou Nesos owned a sizable amount of property,probablyat least a thirdof the total. The question is, what is the significanceof thisphenomenon?Given that men's liabilitiesto taxationand liturgywere greaterthan women's, we mightsuspectthat men put theirpossessionsinto theirwives' names in orderto reduce the size of theirown estates,a practicewhich is common in our own societytoday. Indeed, it may be an implicitassumption of this sort on the part of scholarswhich has deflectedinterestin the subject of women propertyowners,since it has perhaps been taken for grantedthatthe economic role of women in the life of the villagescould not have been of any real importance.For this reason it is critical to determinewhetherthe role of women is merelya mask forthe economic activitiesof the men in theirhouseholds,or whetherwomen really did functionin theirown capacities and in theirown financialinterests. The evidence of the papyri I have examined indicatesthatwomen's possessionswere indeed theirown, and that the basis of theireconomic positionwas theirrightto share in the propertyof theirfamiliesthrough dowryor inheritance.This is suggestedin the firstinstanceby two aspects of thestatistics derivedfromtheSocnopaiou Nesos documents.Firstof all, amongsales of houses,womenappear morefrequentlyas sellers(19 times) thanas buyers(11 times),whereasmen appear morefrequentlyas buyers (17 times) than as sellers (13 times). I thinkthis disparityindicatesthat women operate more conservativelyon the propertymarketthan men; theymaysell propertywhichtheyhave inherited,24 but theyare lesslikely thanmen to investin new properties.25 A furtherindicationof a more restrictedlevel of economic activity on the part of women than among men is that women occur much less frequentlyin documents involvingcapital than in those involving real estate. Among loan contractswe find24 women (9 as lenders,15 as borrowers),as contrastedwith 103 men (34 lenders, 69 borrowers).26 (p. 106), BGU II 467, P. Grenf. II 59, BGU II 630, SPP XX 30. There is in addition a female slave owner in P. Berol. inv. 6988 (ed. Sijpesteijn/Poethke)of A.D 139. 24 As, e.g., SB V 8950, P. Ryl. 11 162, SB X 10571, where the propertiesinvolvedare specificallystated to have been inheritedfroma motheror father,or P. Ryl. II 160a, BGU I 184, BGU XI 2095, where a group of siblingssells a jointly-ownedproperty,therefore presumablyone inheritedfroma parent. 25 Althoughsee, e.g., BGU III 854 and BGU I 350 where the woman buyeralso owns an adjoiningproperty. 26 The followingis a list of loans and loan repaymentsfromSocnopaiou Nesos fromthe Roman period. Documents involvingwomen lenders are indicated +, women borrowers are indicated *. It should be noted that the large figuresfor men derive fromdocuments where a number of men borrow or lend collectively,as, e.g., P. Lond. II 336 (p. 221) where five priestsfrom Socnopaiou Nesos take a loan together.The fact that women rarelyact collectivelywhereas men do quite oftenis another indicationthat women are

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Women as PropertyOwners

317

Furthermore,almost half of the documents in which women borrow money have the loan secured against a real estate holding,whereas this This mightindiformof securityis rarelyofferedby male borrowers.27 cate on the one hand that men were regardedas bettercreditrisksthan women, but on the other hand it also points to the extent to which women's economic role is directly linked to their ownership of real estate, which is the propertythey derive from their families through dowryor inheritance. If men tended to transfertheirholdingsto the names of theirwives or otherfemale dependents,we mightexpect to findmuch higherpercentages of women propertyowners,particularlyamong those buying houses and lending money,than we do. As it is, the numberswhich we do findare large enough to pointto the fact thatwomen reallydid own a significantamount of property,but not large enough to cause us to suspectany falsification of the identityof the real owner.28 Additional confirmationof the legitimacy and vitalityof female propertyownershipis supplied by an examinationof the maritalstatus of the variouswomen propertyownersat Socnopaiou Nesos; the identity of the KVplOS throughwhom a woman acts when she buys or sells propertyprovidesan indicationof her maritalstatus,since a marriedwoman will normallyhave her husband act forher in legal transactions, whereas a youngwoman not yet marriedmay have her fatheror a brother,and a widow will have her son or anothermale relative.If women's financial positionwere connected with marital status,then we should expect to findthat women's economic activitiesvaried in relationto theirmarital status.This does not seem to be the case. If we look, forexample, at the primarilydealing withtheirown inheritances,whereas men are perhaps involvedin more impersonalbusinessdealings. SB I 5244, BGU 1 189, SB I 5243, SB I 5245, BGU III 911*, P. Lond. II 277 (p. 217)*, P. Ryl. II 326 descr., PSI IX 1051*, P. Ryl. II 160c+*, BGU III 713, SB I 5110, BGU XI 2044*, SB XII 10804*, P. Amh. 111 10, SB V 8952*, BGU XIII 2330, BGU XIII 2331, P. Ryl. II 327 descr., SPP XXII 46+, P. Amh. 11 112, P. Amh. II 111, P. Monac. gr. inv. 32+*, P. Stras. IV 293+, SPP XXII 83, P. Ryl. II 174a, SPP XXII 78, P. Vindob. Worp 10+, SPP XXII 36*, P. Lond. II 308 (p. 218), BGU II 445+*, P. Lond. II 311 (p. 219)+*, CPR 1 15, P. Stras. V 383+*, BGU I 290, BGU XI 2043/SPP XXII 45*, P. Amh. 11 113*, CPR VI 3+*, CPR 1 16, CPR 1 14, P. Lond. II 322 (p. 209), P. Lond. II 336 (p. 221), SB VI 9369+, P. Flor. I 42, SPP XXII 69*, SPP XXII 76, BGU III 853, P. Ryl. II 334 descr.,SPP XXII 41, SB I 7, P. Ryl. II 337 descr. 27 Loans secured against propertywhere the borroweris a woman: P. Ryl. II 160c, SB V 8952, SPP XXII 36, BGU II 445, P. Lond. II 311 (p. 219), BGU XI 2043. Those where a man is the borrowerare P. Vindob. Worp 20, P. Lond. II 277 (p. 217), SPP XXII 41. 28 On this point see ChristineDoudna, New York Times Magazine Section (November 30, 1980) 55; in discussingthe question of whetherwomen in the moderncorporateworld use theirsexualityto advance themselvesprofessionally, she quotes Susan Meyer, executive directorof WorkingWomen's Institute,who commentsthat if you look at the small numberof women executives,you have to conclude that "eitherit's not workingor she's not doing it."

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

318

Deborah Hobson

eleven propertysales in which women are the sole buyers(that is, not acting jointlywith eithermale siblingsor husbands),we finda range of ages and statuses:in SB I 5117 the woman is probablynot married,in SB I 5108, P. Ryl. II 160b, P. Ryl. II 160c, P. Vindob. Tandem 24, P. Vindob. Tandem 26, BGU III 854 the buyer is most probably a widow, whereas in BGU I 350, P. Ryl. 11 162, P. Lond. II 334 (p. 211), P. Amh. II 97 she is a marriedwoman; in BGU I 854 and BGU I 350 the woman also owns an adjoining property.This shows quite clearly that women were capable of independent economic activity regardlessof maritalstatus.29 The conclusionwhich emergesfrommy analysisof thisone particular village, namely that women appear to have owned, in their own right,a considerableamount of the real estate in Roman Egypt,is clarified by the informationon inheritancepracticeswhich is provided by wills.30We have a large numberof wills fromGreek, Roman, and Byzantine Egypt-about 170 in all at thispoint-and an examinationof the provisionsin these servesto clarifythe extentto which women benefited fromtheir rightsof inheritance.31 For the presentpurpose I will limit 29 I do not share Pomeroy'sview, (above, note 2) 315f., that a woman's economic independence is seriouslycompromisedby her necessityto transactbusinessthrougha KV'pL0. I take as the significantindex of independence the clear identification of the woman herselfas the owner of the property.On thispointsee Preaux (above, note 3) 141ff.,esp. 143: "Si l'on fait abstractionde la presence du cvpLos, on peut dire que la femme grecque, comme la femme6gyptienne,a une capacite juridique tr&s6tendue." A good discussionof the economic authorityof the KVpLO is given by Schaps (above, note 11) ch. 4. 30 Althoughit is true that women derived some immovable propertyfrom dowries as well as fromwills,dowry documentsdo not afforda ground forcomparisonbetween the propertyacquired by men and thatacquired by women in a singledivision,and thusthey cannot be used to assess the relativeeconomic positionsof men and women with respect to familyproperty.In any case, it seems clear fromthe evidence thatwomen received far less immovable propertyfromdowries than fromwills: see 0. Montevecchi,"Ricerche di sociologia nei documentidell'Egitto greco-romano.II. I contrattidi matrimonioe gli atti di divorzio." Aegyptus 16 (1936) 3-83, esp. 49f. Among 57 contractsof marriageof the Roman period listed on pp. 4-6, only nine contain a giftof real propertyas part of the dowry. 31 The locus classicus on the subject of wills in Greco-Roman Egypt remains 0. Montevecchi,"Ricerche di sociologia nei documentidell'Egittogreco-romano.I. I testamenti," Aegyptus 15 (1935) 67-121. For the presentpurposesthe technicaldistinctionsbetween a will and a donatio mortiscausa will not be taken into account,and I use the general term "will" to include the donationes mortiscausa. For the basic literatureon the latter,see, in addition to Montevecchi, H. Kreller, Erbrechtliche Untersuchungen auf Grund der graeco-aegyptischenPapyrusurkunden(Leipzig and Berlin 1919) 215-23, R. Taubenschlag (above, note 6) 204-7, E. M. Husselman, "Donationes Mortis Causa from Tebtunis,"TAPA 88 (1957) 135-54. The followingis a listof wills and donationes mortiscausa (abbreviatedhere as dmc) published since the appearance of Montevecchi's article. Those where the testatoris a woman are marked +. SB X 10282 (176-70 B.C.), PSI Omaggio 5 (I B.C.-A.D. I), P. Alex inv. 352 (Roman), P. Vindob. Tandem 27 (dmc, I), P. Fouad I 33 (dmc, I)+, P. Mich. V

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Women as PropertyOwners

319

my discussionto those papyri which date fromthe Roman period,since thisis the period encompassedby the documentsfromSocnopaiou Nesos which formed my point of departure.However, I must point out that wills of all periodscontainample evidence of the factthat women could and did inheritmovable and immovable propertyof all sortsfromthe early Ptolemaic period throughthe Byzantineera.32Althoughlegal terminologyand detailsof distribution may varyfromone era to another,it is perfectlyclear that the female capacity to own propertywas not confinedto the Roman period, nor was it really more characteristicof one ethnic group than of another,since we find property(includingland) owned by Greek,Roman, and Egyptianwomen alike. When one firstlooks at the wills of the Roman period, one notes in the firstinstancethatamong the extantwills,63 are thoseof men and 33 those of women; thus the proportionof women transmitting property among the extantdocuments,which is about one-thirdof the total,is in itselfroughlycomparable to the proportionof female propertyowners we were able to identifyat Socnopaiou Nesos, and only somewhat smaller than the proportionof female landowners in the Karanis tax rolls. Among the 63 documentswhere the testatoris male, the variations are numerous,but there seem to be two common methodsof property distribution, one in which the propertyis divided equally among all the children,whetherthey are male or female (as, e.g., P. Oxy. III 651, P. Oxy. VII 1034, P. Hamb. I 73, SB I 4322, CPR VI 1), and another where the eldest son gets a double portionand the restis divided among the remainingsiblings (as in, e.g., BGU I 86 = MChr 306, P. Mich. 321 (division of property,A.D. 42), P. Mich. VII 437 (II), SB VIII 9642 # 6 (dmc, II), P. Wisc. 1 13 (II)+, P. Mich. IX 549 (A.D. 117/18)+, SB VIII 9642 #1 (dmc, A.D. 104)+, P. Mil. Vogl. IV 209 (A.D. 108), SB VIII 9642 #4 (dmc, Hadrian), SB VIII 9642 #2 (din, A l) 123), SB X 10572 (dmc, A.D. 126), SB VIII 9642 #3 (dmc, A.D. 125), SB X 10756 (A.D. 133)+, P. Oxy. XXXVIII 2857 (A.D. 134), SB VI 9377 (dmc, A.D. 138), SB VIII 9642 #5 (dmnc,Antoninus),P. Mich. VIII 439 (A.D. 147), P. Stras. IV 546 (ca. A.D. 155), P. Mert. III 105 (dnmc, A.D. 164)+, SB V 7816 (A.D. 166/7)+, P. Lund. VI 6 (A.D. 190/1)+, P. Mich. VII 453 (111),P. Oxy. XXVII 2474 (III), P. Stras. IV 277 (III), P. Lugd. Bat. XIII 14 (III), P. Coll. Youtie I 64 (A.D. 211), P. Oxy. XXII 2348 (A.D. 224), P. Princ. II 38 (A.D. 264)+, SB V 8265 (IV), P. Athen. 31 (V/VI), P. Oxy. XX 2283 (A.D. 586), P. Michael. 53 (VI), P. Ness. 115 (VI), P. Ness. 116 (VI), P. Ness. 117 (VI/VII). .32 The followingare wills of the Ptolemaic period in which property(including land, whert specified)was leftto women: P. Petr. III 1, col. ii (238 B.C.; plotsof land are leftto two differentwomen), P. Petr. III 7 (238/7 B.C.), P. Petr. III 6a.16-47 (237 B.C.), P. Petr. III 2 (237 1(,.), P. Petr. III 13a.1-19 (235 B.C.), P. Petr. III 13a.20-28 (235 B.C.), P. Petr. I 17.2 + P. Petr. III 17b (236/5 B.C.), P. Petr. 1 18.1, leftcol. (235/4 B.C.), P. Petr. 1 19.1531 (226/5 ic.c.),P. Petr. III 19c + d (225 B.C.), BGU III 993 (dmc, 127 B.C.), P. Cair. 10388 = APF I p. 62 (123 B.C.; 12 arouras of land). In most of these wills the testatorsimply leaIvesall of his property(7a v7rapXovra AOL IrafVra) to his wife.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

320

Deborah Hobson

V 321, P. Oxy. VI 907 = MChr 307).33 The practice of giving sons a largershare of the estate than daughtersis not limitedto male testators; among willsof women we findexamples of the same procedure(e.g., SB VIII 9642 #3, SB X 10756, P. Oxy. IV 837 descr.,P. Oxy. 1 104). The wife is not usually bequeathed any of her husband's real property,34 but provisionsare oftenmade forher use of the propertyand for her maintenance by the heirs so long as she remains alive and unmarried(e.g., BGU I 86 = MChr 306, P. Oxy. III 494 = MChr 305, CPR VI 1, P. Oxy. 1 105 = MChr 303). Since wills provide information only about the possessionsof the testatorhimself,we cannot know what propertythe wife may have had in her own name in each of these instances, but it seems likely that the husband did not ordinarily bequeath much propertyto his wife because she had her own property given in dowry or received throughinheritancefrom her family. A husband's primaryresponsibility would have been for the maintenance of his wife duringher lifetime,and thishe provided in the termsof his will. Althoughthere are many variationsamong the individual papyri, thereseem to be two basic principlesgoverningthe distribution of propertyin wills,and bothof thesehave implicationsforthe amountof property which passes into the hands of women. The firstprinciple is that every memberof the immediatefamilyshould be providedforas his or her circumstancesrequire.35For this reason a wife does not need to be providedforif she has an estateof her own, whereasa widowed daughter with dependent childrenmightneed a generousprovision(as is the case, one suspects,in P. Vindob. Tandem 27). Althoughthere is certainly a tendency to give real estate to male children and household furnishings to females,this is by no means always the case. Of the 63 wills of men, there are only five (SB VIII 9642 #5, SB X 10572, SB VI 9377, P. Oxy. VI 907 = MChr 317, P. Mich. V 321) where the female child gets a portionwhich is inferiorto thatof the male child, and there are at least nine where the female child (as in P. Oxy. VI 1034, P. Vindob. Tandem 27, BGU III 896, P. Oxy. I 105 = MChr 303, P. Berol. 7124) or in a few cases the wife (P. Oxy. III 493 = MChr 307, P. Oxy. III 489, P. Gron. 10, P. Oxy. XXXVIII 2857) getsthe bulk of the estate. In the remainingdocuments (where the text is sufficiently preserved to provide such information)eitherthe childrenare male, or the genderof the childrenis unspecifiedin the will. 33 On thispointsee Montevecchi(above, note 31) 102ff. 34 Althoughthereare a few instanceswhere the wife is the primaryheir (see below, note

35). 35 An interestingexample (though perhaps it is to the contrary)is found in the will of one AtureliaSerenilla,P. Princ. II 38, where her two sons are specificallydisinherited,and the estate is leftto the testator'smother.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions

Women as PropertyOwners

321

In sum, then,thoughit mighthave been a general practicethat,all other thingsbeing equal, the eldest son got the largest portionof the estate,and thatif everyfamilyconsistedof two sonsand a daughter,you mightfindthat women would inheritmuch less propertythan men, the factis thatnot everyfamilyhad male children,36 and the second general principle which seems to have governed distributionof propertywas thatthe propertyshould remainwithinthe immediatefamilyif possible. So if therewere no male childrenin a family,the femalesinheritedthe property,and this must have happened in a considerable number of cases. Furthermore,since women retainedcontrolover theirown propertyand disposedof it themselves,as can be seen fromthe factthatover one-thirdof the wills of the Roman period are those of women, their identitiesas propertyownersremainedintactthroughthe transmission of theirown properties. The applicationof these two general principles,that every member of the family be appropriatelyprovided for, and that propertybe retained withinthe immediate family,would have resultedin women coming into possessionof a significantamount of propertyin Roman Egypt. However, since they were given property-and particularlyreal estate-only secondarilyin many cases (that is, all other thingsbeing equal, the male child was somewhat more likely to inheritreal estate than the female),their-capacityto inheritlanded propertywas, in actual practice,not entirelyequal to that of men. This may be why we find that women seem to have owned only a thirdof the real estate in a village like Socnopaiou Nesos. So when we reconsiderour statisticsfrom this one small village in the lightof the broader spectrumof evidence provided by wills fromthe whole country,we obtain confirmationfor the hypothesisthat the economic role of women in Roman Egypt was indeed genuine,as far as it went,but that it derived quite directlyfrom the female rightto share in the propertyof her own familyby dowry and inheritance,and probablydid not extendmuch beyondthat. 36 On thispointsee E. A. Wrigley,"FertilityStrategyforthe Individual and the Group," in Historical Studies of Changing Fertility,ed. C. Tilly (Princeton1978) 137ff.,where it is convincinglydemonstrated(Table 3-1) that in a societywith a more or less stationary population (which is generally the case with preindustrializedsocieties like Roman Egypt), about 20% of familieswill have no heir at all, 20% will have at least one female but no male heir,and 60% will have at least one male heir.

This content downloaded from 71.172.230.227 on Thu, 25 Jul 2013 16:52:44 PM All use subject to JSTOR Terms and Conditions