Tuesday, April 16, 2013


T. M. Hickey, Wine, Wealth, and the State in Late Antique Egypt: The House of Apion at Oxyrhynchus. New texts from ancient cultures. Ann Arbor: University of Michigan Press, 2012. Pp. xvii, 229. ISBN 9780472118120. $70.00.

Reviewed by Brian Muhs, University of Chicago (bpmuhs@uchicago.edu)

Version at BMCR home site


The title of this book accurately describes its contents, even as it profoundly understates its implications for the late antique economy. The great estate of the Flavii Apiones at Oxyrhynchus in Egypt in the late fifth, sixth and early seventh centuries CE is relatively well documented by papyrological sources, and has long served as the model of aristocratic landholding in Byzantine Egypt, and indeed the Byzantine East. This book argues strongly for revising that model, and by extension the reconstructions of late antique economy and society that have been built upon it.

Traditionally, the Apion estate was considered to be enormous, and was seen as evidence for local nobility gradually taking away responsibilities and authority from the central state. In 1964, Arnold Jones used tax payments to calculate the size of the Apion estates in the Oxyrhynchite and Kynopolite nomes as 112,000 arouras of land, or 75,000 acres, or 120 square miles.1 He noted that in P. Oxy. I 127, the Apion estate contributed 140,618.5 artabas of grain for the grain tax of these nomes, and that P. Cair.Masp. I 67057, from Aphrodito in the Antaeopolite nome, indicated a tax rate of 1.25 artabas per aroura. Assuming that this tax rate was universal, it gives an estate of 112,495 arouras.

In 1985, however, Jean Gascou proposed that the Apion estate proper may have been considerably smaller than its grain tax payments suggest, because the estate may have acted as an imperial agent responsible for collecting the taxes of others besides its own tenants.2 In 2001, Todd Hickey presented supporting evidence in his dissertation, 3 and in 2008 he argued that P. Oxy. XVIII 2196 verso also confirmed Gascou's hypothesis.4 The papyrus gives the revenues of the Apion estate proper as 108,816.5 artabas of grain, plus over 53,000 artabas from villages, for a total of 162,313+ artabas. He observed that the grain tax of 140,618.5 artabas paid by the Apion estate in the previously mentioned P. Oxy. I 127 must have included the taxes of other properties, because the estate proper did not produce that much grain in P. Oxy. XVIII 2196 verso. Furthermore, he notes that if the Apion estate paid a higher tax rate than that known from Aphrodito, 108,816.5 artabas could have been produced from as little as 18,000 to 22,000 arouras.

In the meantime, however, Peter Sarris had dismissed the Gascou hypothesis, and essentially ignored Hickey's dissertation, though he cited it. Instead Sarris argued that the Apion estate was perhaps even larger than Jones had calculated, and was profit oriented, relying on bound wage labor to directly cultivate much of its lands to produce surpluses for the market.5 He suggested that the estate consisted of two parts, the ktemata and the autourgia. Tenant farmers bound to the estate cultivated the ktemata, which produced little marketable surplus and revenue but provided wage labor that the estate used to directly cultivate the autourgia, which produced the bulk of the estate's surpluses for the market and thus its revenues. Sarris hypothesized that the surviving documentation for the estate refers disproportionately to the ktemata, and hence most of the estate's lands, surpluses, and revenues are invisible in the documentation.

Hickey therefore revised his dissertation to produce this book. In it, he attempts once again to calculate the size of the Apion estate, independently from the sources used by Jones and by himself in 2008. Basically, he proposes to calculate how much of one crop the Apion estate produced, and from this to calculate how much land was needed to produce it, and thus the size of the Apion estate. Furthermore he proposes to use a cash crop, in order to determine whether the Apion estate was oriented toward profit-driven production for the market, as Sarris argued. Grains like wheat and barley were the commodities most frequently mentioned in the estate documentation, but Hickey concludes that the estate produced just enough to pay the grain taxes and meet the estate's own needs, including seed. Wine was the commodity most frequently mentioned after grain, however, and was the only commodity that the estate sold on the market in significant quantities, and thus it is the subject of this book.

Hickey first examines the terminology for vineyards in the Apion documents. These include property owned by the Apiones, imperial property administered by the Apiones, and others' properties from which the Apiones collected taxes for the state. Hickey notes that even critics of Gascou admit that the Apiones were responsible for collecting taxes from some private properties that they did not own, and this has been recently been confirmed by Hickey's analysis of P. Oxy. XVIII 2196 verso mentioned previously. In addition, P. Oxy. XVI 1915 indicates that the Apiones also collected revenues from imperial property, which had been assigned for cultivation to individuals living near the Apion estate, lending further support to Gascou's view of the Apiones as imperial agents rather than adversaries.

Hickey then examines the evidence for labor in the Apion documents. The debate about direct cultivation versus tenancy on the estate has been influenced by references to enapographoi georgoi or coloni adscripticii in the estate documentation. The Roman law codes define coloni adscripticii as agricultural laborers registered on the census under their employers, who paid their capitation taxes in return for labor. Consequently, Sarris has argued that they provided the Apion estate with wage labor for direct cultivation. However, Hickey notes that in the Apion documentation, enapographoi georgoi are usually mentioned in connection with irrigation works, rather than direct cultivation. Furthermore, he points out that virtually all of the explicit evidence for cash wages concerns irrigation works, not viticulture. Instead, Hickey argues that tenant farmers cultivated much of the estate, and that some paid their rents in cash, while others paid a share of the harvest, either in grain or wine.

In support of this, he notes that the estate was divided into at least sixteen districts known as protasiai. Each protasia contained several villages (komai) and hamlets (epoikia). A protasia was overseen by a pronoetes, whose accounts include revenues (lemmata) and expenditures (analomata) in both cash and grain. Wine revenues and expenditures, however, were the responsibility of a separate wine steward (oinocheiristes). Hickey argues that these revenues represent the rents of tenant farmers. No vineyard leases from the Apion estate have survived, but contemporary vineyard leases from elsewhere in Egypt specify rent in cash as well as in kind, and require tenants to provide or pay for labor, transport, and sometimes wine vessels, especially when the tenants pay in cash. Significantly, one account (P. Oxy. XVI 1916 verso) from the Apion estate indicates payments in wine for rent as well as for transport, suggesting that these were rental payments by share-cropping tenants.

Finally, Hickey examines the evidence for wine production and consumption in the Apion documents. He argues that one wine steward (oinocheiristes) at a time was responsible for estate wine revenues and expenditures, and consequently their accounts indicate the total amount of wine produced and consumed. Three such accounts survive (P. Oxy. XXVI 2480, LVIII 3960, and PSI VIII 953), and reveal that estate wine revenues came from rents in kind from tenants, from direct production from the autourgia, and occasionally from purchases. Estate wine expenditures included pious donations to churches, monasteries, widows and invalids; perquisites for officials; and distributions to soldiers stationed in the area, to employees on the estate, and to the hippodrome; and sales of wine.

The wine expenditures of the Apion estate are almost the same as the wine revenues of the estate in two accounts (P. Oxy. XXVII 2480 and P. Oxy. LVIII 3960), while the third account (PSI VIII 953) records some surpluses, but also some purchases. The estate therefore does not appear to have produced regular wine surpluses for the market, and the potential cash revenues from the surpluses that they did produce were small compared to the cash revenues from rents. This in turn suggests that the Apion family favored less profitable but lower risk cash rents from tenant farmers, over potentially more profitable but riskier direct cultivation and marketing of surpluses, since wine was the only crop with which the Apion estate could have significantly engaged with the market.

Furthermore, the wine distributions of P. Oxy. LVIII 3960 suggest that at that time there were 572 permanent employees on the estate, which is too few to directly cultivate an estate of the size suggested by Jones. Likewise, the largest volume of wine recorded in the accounts of estate revenues would require only 175 arouras of vines in the autourgia. To these can be added the vineyards cultivated by tenant farmers, who paid cash rents. These can be very roughly estimated at 650 arouras, giving a total of 825 arouras under vines on the estate. This is a ridiculously small figure for an estate of the size suggested by Jones (112,000 arouras), and fits much better with an estate of the size previously suggested by Hickey (18,000 to 22,000 arouras). It also fits well with one of the estate's accounts (P. Oxy. XVI 1918 recto), which records arrears in both the ktemata and the autourgia, and that the autourgia was responsible for about 15% of the estate's arrears. Furthermore, another account of arrears in the autourgia (P. Oxy. XVI 2032) refers only to grain and fodder, rather than to market crops like wine.

Hickey thus shows, once again, that a close reading of the primary documentation for the Apion estate strongly supports the Gascou hypothesis of a smaller estate collecting taxes from its neighbors. Furthermore, this reading suggests that the Apion estate generated most of its revenues through rents from tenant cultivators in cash and kind, and renders it improbable that the marketing of surpluses from direct cultivation of large but undocumented parts of the estate accounted for a major portion of its revenues. Reconstructions of late antique economy and society based on other models of aristocratic landholding in Egypt and the Byzantine Empire, such as that proposed by Sarris, will therefore need to be seriously reconsidered, if not rejected outright.


1.   A.H.M. Jones, The later Roman Empire, 284-602: A social, economic, and administrative survey. Norman: University of Oklahoma Press, 1964.
2.   J. Gascou, "Les grands domaines, la cité et l'État en Égypte byzantine: Recherches d'histoire agraire, fiscale et administrative," TandMByz 9 (1985), p. 1-90.
3.   T.M. Hickey, A public "house" but closed: "Fiscal participation" and economic decision-making on the Oxyrhynchite estate of the Flavii Apiones. PhD thesis: University of Chicago, 2001.
4.   T.M. Hickey, "An Inconvenient Truth? P.Oxy. 18.2196 verso, the Apion Estate, and fiscalité in the Late Antique Oxyrhynchite," BASP 45 (2008), p. 87-100.
5.   P. Sarris, Economy and Society in the Age of Justinian. Cambridge: University Press, 2006.

No comments:

Post a Comment