Stefanie Schmidt

University of Zurich / FU Berlin

By 6422All dates are given as AD / CE if not stated otherwise.Byzantine Egypt had been conquered by a Muslim army, which hereby further extended the territory of a growing Muslim Empire. No coeval source offers any information on the political aspiration of this emerging power. We do not even know how the new rulers of formerly Byzantine terrain designated this early stage of territorial formation. The absence of any contemporary statement gave rise to a long discussion among scholars about whether the Muslim power of the 7th century could be labelled a “state” or an “empire” respectively. Some scholars denied an early concept of centrality and characterised the Rashidun(632–661) and the Umayyad (661–750) administration of the 7th century as “a very loose tributary state,” “a loose confederation of Arab tribes,” or “politically independent communities of mu’minūn” (believers).3Johns 2003, 418; Robinson 2000, 166; Noth and Conrad 1994, 56; Sharon 1991, 124. All discussed in Hoyland 2006, 398.Others embraced the idea of early political institutions such as “(a) a governing group, (b) army and police,(c) a judiciary, (d) a tax administration, and possibly (e) institutions to implement state policies other than taxation, adjudication, and maintenance of control by the elite” that had come into being at least by the time of caliph ?Abd al-Malik(685–705).4Donner 1986, 283; 293: “This brief survey suggests that, on the basis of our definition, a state did exist from the time of ?Abd al-Malik (A.D. 685–705), and that it probably existed back into the time of Mu?āwiya b. Abī Sufyān(A.D. 661–680), that is, back about A.H. 50.”The question of change and continuities of Byzantine institutions under Muslim rule, and more recently also of how existing structures influenced what should become the highly bureaucratic Muslim Empire, is one of the most widely discussed issues among scholars of this transition period.5See, for instance, (in alphabetical, then chronological order) Berkes 2017; Bruning 2018; Foss 2009a; 2009b; Gascou 1983; Gascou and Worp 1982; Gonis 2009; 2004b; 2001; Hoyland 2006;1997; Legendre 2016; Morelli 2017; 2016; CPR XXX, 15–22; Morimoto 1981; Papaconstantinou 2010; Richter 2009; Sijpesteijn 2013; 2007a; 2007b; Simonsen 1988. Moreover, in recent years two research projects addressed the topic of change and continuity in Egypt under Muslim rule: The ERC Starting Grant project “The Formation of Islam: The View from Below” (P. Sijpesteijn) in Leiden,see, inter alia, Sijpesteijn 2013; Bruning, De Jong and Sijpesteijn forthcoming. And the SNF project“Change and Continuities. From a Christian to a Muslim Society – Egyptian Society and Economy in the 6th to 8th Centuries” (S. R. Huebner), see Huebner et al. 2020.In general it is assumed that earlier administrative structures were maintained more or less intact for about half a century after the conquest.6Sijpesteijn 2013, ch. 2., esp. 88.Former high Byzantine officials as,for instance, the pagarch7A pagarch was the head of a subdistrict that comprised an eponymous town, its hinterland, and villages.and the duke, were left in charge by the conquerors as is demonstrated, for instance, by the archive of Senouthios,anystēs, head of the northern district of the Hermopolis pagarchy.8The published texts are discussed by F. Morelli in CPR XXII and XXX, cf. also TM Arch ID 418. A substantial part of the archive of Senouthios (anystēs and notarios) is still unpublished.The papyri show, moreover, that military commanders (?amīr, pl.?umarā) as representatives of al-Fus?ā? were placed at the side of the pagarch and the duke ensuring hereby a tight control system of supervision.9CPR XXX, esp. pp. 16–18; Legendre 2016.When exactly this coexistence ended and the formation of a Muslim bureaucracy that penetrated these structurs began, is difficult to say.Some researchers assume that major changes had already been implemented following the first Muslim civil war (656–661), others do not suspect that such measures were undertaken before the end of the 7th century / beginning of the 8th century.10See, for instance, Legendre 2016, 12–13 who describes an integration of the office of the duke into an emerging Muslim political order. According to Legendre, first indications of an increase in the duke’s scope of responsibility are visible under the duke Jordanes (660/661–669 or 669–675/676),who is said to have been ‘amīr at the same time. The equation of the titles duke and ‘amīr was first suggested by R. Rémondon in P.Apoll. 1, intro, 3; 9, 27, l. 2 and also assumed by Gascou and Worp 1982, 85. However, more recently Morelli 2016 raised objections to this by offering a new interpretation of P.Apoll. 9, the only example for this equation. Changes in personel and the structure of the duke’s office can be observed in the course of the 2nd half of the 7th century, when he became responsible for Arcadia and the Thebais. By the end of the 7th century the office saw, moreover, an incumbent with an Arab or Arabized name, cf. the archive of Flavius Atias (‘A?iyya b. Ju‘ayd) in CPR VIII 72–84, CPR XIX 17, P.Ross. Georg III 23, SB III 7240, SB XXIV 16219, and the Coptic texts discussed by Cromwell 2013. At what time exactly the unification of Arcadia and the Thebais under one duke took place is, however, difficult to say: P.Mert II 100 (669, Arsinoite) and P.Apoll. 9 (675/676 or 660/661, Apollonopolis Magna) are often taken as evidence that it started to be jointly administered already under the duke Jordanes, cf. Legendre 2016, 13; Gascou and Worp 1982, 90, no. 3. But see also Foss 2009, 262, n. 18 who notes that, it cannot be excluded that Jordanes held the positions of duke of Arcadia and that of the Thebais in succession. Palme 2011, 5, who also refers to P.Mert II 100,2 (669, Arsinoite), assumes that changes took place between 655 (CPR XIV 32) and 669.Integrating oneself into the administrative structures of a conquered country is one way to lay the grounds for an empire. But before this, an even more urgent matter needs to be faced by the invaders – the supply of the army and the covering of conquest expenses. If an invader’s only objective is quick booty, he starts plundering the inhabitants. If he wants to make the conquest“pay” for a longer time he needs a strategy that reduces resistence among the inhabitants and guarantees the economy’s continued productivity. The aim of this paper is thus to explore how the conquerers mobilized economic resources, where they took over already existing levy systems, and whether they set incentives to stabilize the economy.

The levy of imposts in kind

Egypt’s papyri provide good insights into how the invaders began to exploit local sources. Payment orders, demand notes, tax receipts, and acknowledgements shed light on what they levied and the infrastructure they made use of. Apart from mobilizing capital, which will be discussed below, the new rulers demanded imposts in kind. In receipts issued for the delivery of goods, these requisitions were calleddianomai. Although a termdianomēoccurs in Roman Egypt from at least the 2nd until at least the 4th century,11BGU IX 1894, col. XI, 121 (158–159, Theadelphia); P.Col. V 1, verso 1a, col. III, 38 (159–160,Theadelphia); SPP XX 14, 10–11 (188, Arsinoite); SB XII 10893, 10 (II, unknown), perhaps P.Oxy.XII 1490 recto 1 (after 320, Oxyrhynchos), for the date BL X, 142; CPR VII 18, 10 (364/379/394,Hermopolis).the majority of texts referring todianomaias requisitions dates from after the Arab conquest.12P.Lond. I 113 (10), 13, p. 222 (639–640, Arsinoite) (see the discussion below); CPR XXX 32, col.I, 1 (643–644, Hermopolite); P.Amh. II 153, 20 corr. ex. [[dianomēn]] (ca. 643–644, Hermopolite?);154, 2; 9 (643–644?, Hermopolite); SPP III2 106, 2 (2nd half 7th c., Arsinoe (?)); 108A, 6 (2nd half 7th c., Arsinoe); 119, 3 (656/671, Arsinoe); 121, 4 (2nd half 7th c., Arsinoe); 180, 4 (2nd half 7th c.,Arsinoe); 204, 7 (2nd half 7th c., Arsinoite?); 665, 2 (7th c., Arsinoe); 685, 2 (7th c., Arsinoe); maybe also 55 (2nd half 7th c., Arsinoe?)), cf. comm. on p. 98; SB XVIII 13267, 1 (7th c., Arsinoite); I 4864,2; 3 (7th–8th c., Arsinoite); XXVI 16664, 1 (7th–8th c., Hermopolite?); P.Poethke 26, col. II, 1 (7th–8th c., Arsinoite); P.Lond. V 1763v, col III, 25 (7th–8th c., Hermopolite?), for the location, cf. CPR XXX,p. 113; IV 1447, 138; 144; 188; 192 (685–705, Aphrodito); 1441, 50 (706, Aphrodito); 1433, 23; 98;140; 150; 177; 187; 198; 225; 235; etc. (707, Aphrodito); 1440, 5 (ca. 709, Aphrodito); 1346 verso,2 (710, Aphrodito); 1369, 6 (710, Aphrodito); 1400+1364, 9 (710, Aphrodito); 1449, 44; 46 (711,Aphrodito); 1413, 7; 8; 23; 24; 35; 36 etc.; (723, Aphrodito); 1416B, 13 (732–733, Aphrodito); 1414,11; 65; 68; 111; 144, etc. (8th c., Aphrodito); Berkes 2017 (= P.IFAO Edfou Jarre inv. 205 + 218, frgs.A&B), 5 (660–670, Apollonopolis Magna); and the discussion in Morelli 2010a, 155, n. 65.Requisitions consisted of animals,13P.Lond. I 113 (10), 15 (p. 222) (639–640, Arsinoite); P.Amh. II 154, 3 (643–644?, Hermopolite).textiles/mats,14SPP III2 106, 1 (2nd half 7th c., Arsinoite?); 108A, 5 (2nd half 7th c., Arsinoite?); 121 (2nd half 7th c.,Arsinoite); SB XVIII 13267, 1 (7th c., Arsinoite); XXVI 16664, 3–4; 10 (7th–8th c., Hermopolite).leathers,15P.Lond. I 113 (10), 15 (p. 222) (639–640, Arsinoite).foodstuff,16P.Lond. I 113 (10), 15–16 (p. 222) (639–640, Arsinoite); CPR XXX 32, 1 (643–644, Hermopolite).supplies for the cavalry (saddles, perhaps horse trappings, fodder),17P.Lond. I 113 (10), 16 (p. 222) (639–640, Arsinoite); SPP III2 204 (2nd half 7th c., Arsinoite),perhaps also the camel saddles in SPP III2 119 (656/671, Arsinoite).and the navy,18Berkes 2017 (= P.IFAO Edfou Jarre inv. 205 + 218, frgs. A&B) (660–670, Apollonopolis Magna).(i.e.dianomai tou Klysmatos19SB I 4864, 2–3 (7th–8th c., Arsinoite); P.Lond. IV 1441, 63 (706, Aphrodito); 1433, 140 (707,Aphrodito); 1440, 5 (709, Aphrodito); 1346 verso, 2 (710, Aphrodito); 1449, verso 1, 46 (711,Aphrodito).or thekaraboiin the island of Babylon20P.Lond. IV 1433, 23; 98; 150 (707, Aphrodito).).Dianomaiwere levied from towns, villages,21P.Lond. I 113 (10) (p. 222) (639–640, Arsinoite); CPR XXX 32 (643–644, Hermopolite); SPP III2180 (2nd half 7th c., Arsinoite); P.Lond. IV 1346 (710, Aphrodito).and smaller districts (hyper dianomēs lauras Perseas),22SPP III 665, 2 (7th c., Arsinoite). See also SPP III 685, 2 (7th c., Arsinoite).from hospitals23P.Amh. II 154 (643–644?, Hermopolite), a discussion of the text can be found in CPR XXII 2, p. 27.and monasteries24SB XXVI 16664, 1 (7th–8th c., Hermopolite); maybe also P.Bal. II 346, 14 (6th–8th c., Bala’izeh).. If contributers were unable to come up with the required goods, they could substitute them with money-payments(apargyrismos),25P.Lond. IV 1413, 7–8; 23–24; 35–36; 46–47; 57–58; 69–70 etc. (723, Aphrodito); 1414, 11 (8th c., Aphrodito); probably also P.Poethke 26 (7th–8th c., Arsinoite); P.Lond. V 1763, v3, 25 (7th–8th c., unknown). Perhaps Gascou 2010, 384, no. 19 (= O.IFAO inv. 146), but see the commentary on l. 2. Official orders to pay in money instead of in kind are transmitted in P.Lond. IV 1407 (709,Aphrodito): for embolē; 1408 (709, Aphrodito): for wages instead of producing nails; 1410 (709,Aphrodito): wages instead of workmen.which was a common practice for payments in kind also in Byzantine times.

Several texts imply that some of these deliveries were remunerated in some way. In a fragmentary receipt from the Arsinoite nome (Fayyūm), for instance,anepistatēsof linen weavers handed over the price of three shirts (kamisia)26For prices of kamisia from the 6th to 8th century, cf. Morelli 2019, 24–25.of that year’sdianomēto a certain Neilammon,27See the re-edition SPP III2 108A, 1 (comm. ed.).probably the weaver in charge of manufacturing the shirts.28SPP III2 108A (2nd half 7th c., Arsinoe).The amount is lost, but in a governmental purchase of 70kamisiafrom 8th century Aphrodito, the price was set at ?nomismaperkamision.29P.Lond. IV 1352 (710, Aphrodito) and the discussion in Morelli 2019, 25.Similarly, in the early 640s a delivery of at least 710,500 bricks30CPR XXX 1 (643–644, Hermopolite), comm. on p. 60 for the quantity. A part of the papyrus is lost,so the amount could have been higher.in the Hermopolite nome was apparently reimbursed in some fashion: In a letter accompanying the sample of a brick it is stated that an amount of money – which is lost in the text – should be handed over to farmers, whom the editor of the text suggests may have furnished raw materials for the production of bricks.31CPR XXX 2, 13 (643–644, Hermopolite) with comm. on p. 141.

In search of the first published example ofdianomaiin the 7th century, one inevitably comes across a receipt from the Arsinoite nome dating from 639Booth 2013, where he discusses the possibility that a unit of the Muslim army came from the Red Sea coast. Early Muslim presence South of the Fayyūm is now supported by an Arabic receipt from Asyūt that was recently re-dated to 641, cf. Tillier and Vanthieghem 2019, no. 1. In this, ?Abd Allāh b.?Umays acknowledged to have paid three nomismata to Isidoros and his wife (see also the discussion at the end of this article).–640.32P.Lond. I 113 (10) (p. 222) (639–640, Arsinoite). See Morelli 2010a, 155, n. 65 for a discussion of this document: “probabilmente si tratta solo di una lacuna della nostra documentazione, ma quest’uso del termine διανομ? lo conosciamo solo per il periodo arabo.”It is a declaration from the villagers of Kaminoi confirming to the pagarch of Arsinoe that they received the price of what they delivered asdianomaiaccording to the order of the most holy and honoured by God (hagiotatos kai theotimētos)papa33For papa = ecclesiastic title of a priest or bishop, cf. F?rster 2002, s.v.Kyros.34Most probably Kyros, patriarch of Alexandria (630/632 – March 642), on him most recently Booth 2016.The requisitions consisted of hides (byrsaria),wool, hay, sheep and sheepskins, olives, cabbage, andstrōmata.35For strōma, cf. Morelli 2019, 33: “strōma indica tutto ciò che può essere steso, e cioè un materasso,un tappeto, una coperta.”Ulrich Wilcken recognized an imperial date at the beginning of the document, including references to the emperors Heraclius and Heraclius (Neos) Constantine.36W.Chr. 8, comm. ll. 1–4.In the past, it was thus primarily discussed in the context of Byzantine preparations against the invading Muslim army.37Booth 2016, 511; Trombley 2013, 23–25; Papaconstantinou 2010, 66–67.The early date of the receipt seemed to support this assumption since it is difficult to reconcile it with the traditional conquest narrative that places the Muslim conquest of the Fayyūm after that of the Delta.38However, see already Morelli 2010a, 155, n. 65, who rendered a Muslim context likely and recommended a re-consideration of the traditional conquest chronology.However, based on Phil Booth’s recent study of John of Nikiou’s conquest account, in which he argued for an initial Muslim approach from the South, an early occupation of the Fayyūm cannot be entirely ruled out.39Nor would the use of the Byzantine dating formula necessarily be a contradiction to a date after the Muslims conquest. As other documents show, dynastic dating formulas could take some time to reflect real circumstances.40Gonis 2008, 201 with n. 8.Particularly in this early stage of the conquest it was certainly not obvious to the parties involved that the invading army would become the next dominant power. However, it is ultimately impossible to decide whether remunerated requisitions from the villagers of Kaminoi were intended to supply a Muslim or a Byzantine army. The absence of any other published examples ofdianomaiin the 6th or 7th century before the Muslim conquest is certainly not an argument against a Byzantine precursor. Thus, the receipt from the Fayyūm may either be the first evidence ofdianomaiunder Muslim rule or indicates that a Byzantine term found entrance into the post-conquest tax vocabulary.

Apart fromdianomai, there were other regular payments in cash and kind – a delimination between them is often difficult to make: The Greek termrouzikon,for instance, referred to the army’s regular payment in kind41P.Apoll. 49, recto 5; 94, 6; 95 frB, 2 (all 2nd half 7th c., Apollonopolis Magna); P.Lond. IV 1335,5 (709, Aphrodito); 1407, 2–3 (709, Aphrodito); 1404, 7 (709–714, Aphrodito); 1434, 165 (716,Aphrodito); 1435, 122 (716, Aphrodito).and is generally taken as transcription of the Arabicrizq.42See, for instance, Mayerson 1994, 126 and Sijpesteijn 2010, 255, n. 21.However, the termrizqthat served likewise to denote regular payments in kind (and money),43In kind: P.Ness. 60 (b), 5–7 (674; Nessana); 61, 6–7 (675, Nessana); 62, 6–8 (675, Nessana);perhaps P.Gascou 27b, 6 (695, al-Fus?ā?/Antinoe?) with comm. l. 6; P.RagibQurra 1, v2 (709,Arsinoite?). In money: P.BeckerNPAF 12, 4 with 9 (Dietrich 8; 13) (709, Aphrodito); P.David-WeillLouvre 26, 7 (8th c., unknown).is translateddapanē44This tax was used to cover staff expenses while performing service for the government, cf. SPP III2465, comm. p. 26; Gonis 2009, 201. Apparently, also sailors could benefit from this, since in P.Apoll.83, 8–9 (2nd half 7th c., Apollonopolis Magna) bread was bought on behalf of the account of dapanēs nautokarabōn.in a Qurra45Qurra b. Sharīk was governor of Egypt from 709 to 714.letter, notrouzikon.46P.Lond. IV 1375, 1 (710, Aphrodito): fī thaman rizq (Arabic) and 5–6: hyper timēs tōn hypotetagmenōn eidōn dapanēs (Greek).J?rgen B?k Simonsen, who already pointed this out, suggested to takerouzikonas a translation of the Arabic?arībat al-?a?ām,but the latter is found translated withembolēat least in Aphrodito.47SB I 5638, 5; 11 (709, Aphrodito); P.Heid. Arab. I (Appendix) a (= SB I 5644), 6; 12; c (= SB I 5646), 6; 11; e (= SB I 5648), 2; 7; g, 6; 11 (all (709–710, Aphrodito)). Unfortunately, Simonsen 1988, 110 gives no reference for his statement.Another term that referred to supplying troops was the Greekrhoga,48BGU I 304, 11 (647, Herakleopolite); CPR VIII 74, 3 (698, Arsinoite); P.Lond. IV 1433, 19 (707,Aphrodito); CPR XXII 50, 15 (8th c., unknown). In Byzantine papyri: P.Lond. V 1660, 9–10 (553,Aphrodito); P.Oxy. XVI 1913, 60 (555?, Oxyrhynchos); P.Cair. Masp. 1 67076, 4 (6th c., Aphrodito);P.Oxy. XVI 2010, 2 (618, Oxyrhynchos); P.Oxy. LVIII 3960, 38 (621, Oxyrhynchos) for the maintenance of the Sasanians. In P.Lond. IV 1335, n. 5, Bell defines rhoga as money-allowances, but observes also frequent exceptions since it could also be paid in kind, cf. P.Lond. IV 1349, n. 15; 1433,comm. l. 17.probably deriving from Latinerogare.49Kramer 2011, 284: “Das aus dem lateinischen Milit?rterminus ērogāre ins Griechische entlehnte?ογε?ειν wirkte in der Zweisprachigkeitssituation des ostr?mischen Milit?rs auf das Lateinische zurück, so dass man dort ērogāre zu Gunsten von rogāre aufgab. Dazu wurde im ?stlichen Milit?rlatein die suffixlose substantivische Ableitung roga gebildet, die wiederum als terminus technicus für ‘Sold’ als ??γα im Griechischen erscheint.”From a letter of governor Qurra b. Sharīk (709–714) we get the information that therhogafor themuhājirūn50Mōagaritai is believed to mean muhājirūn = migrants, cf. Legendre 2013, 99; Hoyland 1997,547–48 with ns. 13 and 15.of al-Fus?ā? was paid from the gold tax (chrysika dēmosia) that was collected in money.51P.Lond. IV 1349, 10–17 (710, Aphrodito).Contributions torhogamust, however, also have consisted in kind as is demonstrated by an acknowledgement issued by a baker. He confirms to have received a certain amount of wheat on behalf of the account oftēs rhogaswhich he will return as bread.52BGU I 304 (647, Herakleopolite).The case of the baker is a fitting example for cooperation and compliance with the new rulers at a level where economic power was generated. In fact,skilled workers could have refused to perform services what would have raised transaction costs for the new rulers. The negative effects of passive resistence in office are best illustrated by the example of Basileios, pagarch of Aphrodito.In his letters, Qurra constantly threatened Basileios with punishment and death unless he improved his administration of the pagarchy.53See the examples given by Papaconstantinou 2015.The accusations against Basileios included tax evasion, trafficking, lying to the authorities, and hiding fugitives.54Strictly speaking, they represent more blunt corruption than pure resistence against rulers who were no longer new. See ibid., 279–280, who pointed out that it was probably non-cooperative behaviour like that of Basileios that finally lead to a replacement of long-established Christian elites by Muslim officials who had no social backing in the district they administered.Non-cooperative behaviour, hostile workforce and sabotage must have been among the myriad issues the new rulers had to face, but as the papyri show they relied nonethless on entire communities as workmen.55In one case, the pagarch Basileios should distribute iron (provided by the Muslim authorities)among certain chōria and take care that the inhabitants made nails out of it, cf. P.Lond. IV 1369 (710,Aphrodito).The outsourcing of production processes is always a risky undertaking and demands a great deal of trust on the side of the principal – or effective measures to supress opportunistic behaviour on part of the agent. A suitable means of increasing the agent’s cooperative behaviour is to offer incentives. Apart from the threat of violence this could be, for instance, a reimbursement of his work as mentioned above with regard to thekamisiaand the bricks. Due to the scarcity of evidence it is not possible to postulate any rule in this matter, but we observe that in some cases,raw material for production was provided by the Muslim treasury – for instance,iron for manifacturing nails56See n. 55.and the wheat of the baker. In other cases – like the bricks and the shirts – it seemed as if at least production costs were borne by the financial administration.57For considerations on the quota between production costs and raw material, see most recently Morelli 2019, 64.Reimbursements were certainly an appropriate means of keeping resistance low, and the documents show that compensation was already offered at the very beginning of Muslim rule in Egypt. In 643, the governor ?Amr b. al-?ā? ordered the pagarch of Herakleopolis, for instance, to supply the soldiers of ?Amer b. (al-?)Asla? with fodder for twonomismataand to take from ?Amerto logar(ion), which Federico Morelli translated with “l(fā)a somma di denaro.”58SB XX 14443 (643, Hermopolite). Morelli 2013, 167, n. 8: “nient’altro, prendendo da lui la somma di denaro (?) – Stephanus ThGL VI 348, E. Trapp, Lex.Byz.Gr. p. 944; cfr. anche l’?λ?γον λογ?ριν di P. Ness. III 53, l. 9?” Grohmann 1952, 115–116 translates “taking from him a receipt,” and Harrauer 2001, 80, no. 12 (= SB XXVI 16354 (643–644, Hermopolite)) interprets it as “Abrechnung.”In a Greek order, theamīr?Abdullāh instructed themeizones(village headmen)59For meizones, cf. Berkes 2017, 53–83.of Psōphtheōs to sell (thelēsate pōlēsai) to the son of Asla? fodder for threesolidiat a rate of twoartabaepersolidusand the maintenance of three attendants.60SB VI 9578 (642, Heracleopolite?) with Bell 1933, 78, n. 1 (no. 3) who suggested several new readings to Grohmann’s first edition; for the location, see Morelli in CPR XXII, p. 48, l. 5 comm. A similar example to SB VI 9578 is SB VIII 9753, 2 (643, Herakleopolis): paraske(uason) prathēnai Yazīd, son of Saraēl.Registers with prices for products to be acquired for the Muslims,likeP.Laur.IV 192, are also indicators that these compulsory purchases had been reimbursed by the Muslim administration.61For P.Laur. IV 192, cf. Morelli 1995 and 2019, 62.But certainly not at any price. Thus,it is legitimate to ask whether the monetary value of such reimbursements was comparable to market prices. In a recent study, Federico Morelli compared prices for single commodities from the Byzantine to the Arab period. In cases, where contemporary sources could be compared, he concluded that government prices indeed were below the market price, but not significantly.62See, for instance, the comparison between prices for kamisia in ibid., 24–25.

Some prices for governmental purchases were thus, in all likelihood, fixed.63It certainly depended on the kind of product or purchase and could have varied also among purchasers, as F. Morelli pointed out to me by personal comment.However, nonetheless, we may assume a free market with functioning price building mechanisms. This can be demonstrated using the example of exceptional circumstances. In times of a shortage of wheat, the governor Qurra b. Sharīk(709–714) ordered that taxes on trade (mukūs) should be suspended in order to set incentives for the merchants to sell their corn with a perspective to make profit.64P.Cair. Arab. III 147 (710, Aphrodito); P.Heid. Arab. I 2 (710, Aphrodito).Moreover, he ordered that all merchants in the country should sell their corn immediately. He thereby anticipated protectional behaviour of merchants whom he expected to withdraw their corn, in order to drive up prices. In contrast to the Romans in a similar situation, the Muslim government did apparently not make use of selling at governmentally set prices.65Wallace 1938, 22–23 on forced sales of crops (pyros / krithē synagorastikos): “One of the methods used by the Roman government to meet increased demands was the forced sale of grain termed pyros / krithē synagorastikos. Certain cultivators were compelled to sell a part of their crop remaining (epigenēma) after the payment of taxes at a price set by the government. Such grain was delivered by the cultivators to the state granary, where a separate account of it was kept by the sitologi. […] It is probable that the government resorted to the forced sale of grain whenever the price of grain rose in the open market to such an extent that the soldiers stationed in Egypt could not afford to purchase it. The great majority of references to forced sale of grain come from the reigns of Antoninus Pius, Marcus Aurelius, and Commodus.”By creating incentives to sell the wheat, namely by reducing taxes on trade, Qurra’s actions are politically and economically rational since he accepted a general loss in taxes on behalf of state revenues in order to ensure political and social stability.

Nonetheless, the struggles pagarchs may have gone through in finding suppliers who sold goods at often predetermined prices must have been substantial. It is best reflected in a letter from Plato, pagarch of Latopolis (Isnā), to Papas, pagarch of Apollonopolis Magna (Edfū) in the 2nd half of the 7th century. Reporting about repeated compulsory purchases (peri agorasias pollōn pragmatōn) by order of theamīr, Plato sputters, with regard to the latter, “geusasthō hydatos,”fittingly translated by Roger Rémondon with “Qu’il go?te à l’eau du Nil!”66P.Apoll. 37, 11 (2nd half of 7th c., Apollonopolis Magna), p. 88.

Payments in money

Besides requisitions in kind, which were at least partly reimbursed, the treasury received a regular influx of money taxes. Under Muslim rule we find again a taxation of the person. The tax is referred to in a letter of the pagarch of Hermopolis (al-Ushmūnayn) probably dating from 644.67CPR XXII 1 (644? Hermopolite).Herein, the pagarch ordered measures to be taken against tax evasion by flight related to a tax calledandrismos, one of two terms for the poll-tax under Muslim rule. In modern research, the discussion of how thisandrismosdiffered from the other term for poll-tax,diagraphon,68For instance, SPP VIII 740, 2 (645? Arsinoite); P.Rainer Cent. 144, 2 (648/663, Arsinoite?); SPP III 694, 2 (652, Arsinoite). In documents from Nessana, the term for poll-tax was epikephaleion, cf.P.Ness. III 59, 8 (684, Nessana); 70, 6 (685? Nessana); 76, 1 (687–689? Nessana); 77, 1 (685–690,Nessana).and whether or not the Muslim poll-tax was related to a tax of similar wording levied in Byzantine times, is still ongoing.69Discussed most recently in Schmidt 2019 with further literature.It is generally accepted thatandrismosanddiagraphonwere identical taxes, but that the termandrismoswas used in the Hermopolite nome, whilediagraphonwas used everywhere else in Egypt.70Clackson 2009, 8; Poll 1999, 252, n. 84; Simonsen 1988, 184, n. 53; Gascou 1983, 102; Hussein 1982, 59; Casson 1938, 275.In fact, the majority of texts mentioningandrismoscomes from the Hermopolite nome. However, we also find the termdiagraphonin the Hermopolite71P.Laur. III 112, 7 (7th–8th c., Hermopolis); 113, 9 (7th–8th c., Hermopolis); 118, 8 (7th–8th c., Hermopolis); P.Monts. Roca IV 74, 2 (7th–8th c., Hermopolis); P.Lond. V 1744, 2 (7th–8th c.,Hermopolite?); and perhaps P.Wuerzb. 20, 3 (7th–8th c., Hermopolis).andandrismosin Edfū,72P.Apoll. 24, 6 (2nd half 7th c., Apollonopolis Magna).Aphrodito,73P.Lond. IV 1421, 10; 12; 17; 20; 28; 29 etc. (705, Aphrodito); 1420, 7; 25; 26; 48; 49; etc (706,Aphrodito); 1422, 83; 84; 87 (707–708, Aphrodito); 1573, 5 (709–710, Aphrodito); 1424, 3; 35(714?, Aphrodito); 1435, 152 (716, Aphrodito); 1419, 1382–1389 (716–717, Aphrodito); P.Heid. X 454, verso, 1; 5 (7th–8th c., Aphrodito?); P.Lond. IV 1423, 12–17; 21; 23; 31; 38 (8th c., Aphrodito);1442E, 50–51 (8th c., Aphrodito); 1446, 15, 26 (8th c., Aphrodito); 1578, frg. 1, 2 (8th c., Aphrodito).and Djēme.74P.Bal. II 130App. (p. 541–542), 4 (724, Djēme).Occasionally, we finddiagraphonandandrismosin the same document,whereupondiagraphoncould possibly be used to refer to the collective sum(thediagraphonof the monks) andandrismosto the individual contribution (hisandrismos).75P.Bal. II 303B, 16; 24 (7th–8th c., Deir el-Bala’izah). This collective / individual character of diagraphon / andrismos is found in other early Muslim documents. In a receipt, Qays b. al-?ārith,amīr of Herakleopolis, confirms the collective payment of diagraphon to the inhabitants of the chōrion Phys, cf. SB VIII 9756 (653, Herakleopolis), with BL VIII, 353. In another receipt issued to those from the village of Pkmo, 12 individual payments for andrismos were confirmed (hyper onomatōn 12), cf. P.Lond. V 1750, 1–2 (642?, Hermopolite). A joint mention also appears in some Aphrodito papyri such as P.Lond. IV 1338, 22 (709, Aphrodito) and 1339, 5–6 (709, Aphrodito)although the meaning is not altogether clear.

Additionally, the origin of the tax could not yet be sufficiently clarified.76A principal question is certainly why Muslims, in creating a new tax in Egypt, did not make use of a graecised form of an Arabic word like, for instance, the jizyah introduced later.It is conceivable that in this case as well, the conquerors adapted an already existing form of impost such as, for instance, the Byzantinediagraphon,77Written to(u) diagraph(ou) in P.Hamb. I 56, col. VII, 1 (6th–7th c., Aphrodito); tou diagraphou in P.Oxy. I 127, 2 (late VI); to(u) diagrapho(u) in P.Cair.Masp. I 67058, col. VIII, 8 (538?, Aphrodito)and probably tou d??g.r.?ph?[u] in P.K?ln. VII 317, 28 (6th c., Hermopolite?).sometimes calleddiagraphē.78For instance, tēs diagraphēs in P.Cair.Masp. I 67058, col. VII, 23 (538?, Aphrodito); perhaps also in P.Mert. II 95, 2 (V, Arsinoite?).However, in general it is assumed that the Muslimdiagraphonshared no common ground with this Byzantine tax. Two explanations are usually deployed for this: 1) After the abandonement of the last true capitation tax, the Romanlaographia, by theConstitutio Antoniniana(212) there is no unambiguous evidence for a “head-tax” before the Muslim period;79The term kephalē, used in the 4th century for computing tax rates, gave rise to assume a new capitatio, but as Bagnall 1980 convincingly argued, it is more persuasive to interpret it as an abstract measure of landed property.2) the Byzantine tax was probably imposed on landed property, not by head.80Gascou 1983, 102; Rémondon 1965, 412–413.

With regard to the latter argument we observe indeed that land played a role in the payment of the Byzantine tax. Documents of the Dioskoros archive (6th century)81TM Arch ID 72.show thatdiagraphaordiagraphai82The use of the genitive plural in Dioskoros’ document, by the editor given with circumflex(diagraphōn), is not indicative of whether he paid diagraphai or diagrapha. From a 6th century document from Aphrodito, in which both forms occur in the same account, it appears as if they were synonymous, but it could also be a scribal error, cf. P.Cair.Masp. I 67058, col. VII, 23 and col. VIII, 8(538?, Aphrodito).accrued together with other taxes on behalf ofastikē synteleiawhich had fallen upon Dioskoros due to landed property of his father.83P.Lond. V 1686, 22–26 (565, Aphrodito) and intro p. 81 discussed by Rémondon 1965, 412–413.A payment eis log(on) diagra[ph(ēs)] by Dioskoros is likewise found in P.Cair.Masp. III 67325, col.VIIIv, 1–13 (6th c., Aphrodito). Apparently, he paid for land he had inherited, cf. Fournet 2000, 238.For the synteleia, cf. Mirkovi? 2008, 196–201.Procopius, writing in the 6th century, underlines a property-based character regardingdiagraphai84Procopius also used the gentive plural diagraphōn, but takes the matter up again a few sentences later using tautas to refer to the previously mentioned diagraphai.when he states they were paid by landowners proportionally to what taxes they usually paidkata logon tēs egkeimenēs hekastōi phoras.85Procop., Anecdota 23, 17–21. Both its assessment key and whether or not this impost was a regular payment remain unclear in Procopius.It seems as if both, Dioskoros’diagraph( ) and Procopius’diagraphai,86See n. 84.were the same taxes which accrued in an unspecified connection to landed property, but due to the ambiguity of Dioskoros’ impost, we cannot be certain about that.

Diagraphaiare also referred to in the legal texts. TheNovellaedefine them as extraordinary taxes from which, for instance, church property was exempted.87Nov. 131, caput V, 20; 30 (545) (Kroll and Sch?ll 1912).However, they also give the impression that the term was used for imposts unrelated to any land or property. In one case it was prohibited for high officials,who travelled from one city to another, to levydiagraphai, since they should bear the costs out of the funds available to them for expenses.88Nov. 128, caput XXII, 10 (545) (Kroll and Sch?ll 1912).In another it was prohibited for both landowners (Greekktētores/ Latinpossessores) of cities and inhabitants (Greekoikētores/ Latinhabitatores) to deduce money from funds set aside for a city’s public work, wheat funds, or aqueducts for payments ofdiagraphai89Since diagraphai in the feminine is used in all other instances in the Novellae, it can be assumed that also in caput XVI the genitive plural refers to diagraphai.orsportulae.90Nov. 128, caput XVI, 1–10 (545) (Kroll and Sch?ll 1912). Alla mēde tous ktētoras tōn poleōn ē oikētoras tolman hoiōidēpotoun tropōi aphaireisthai ti ex autōn tōn chrēmatōn ē onomati diagraphōn ē hyper sportoulōn didonai ti ek toutōn ē analiskein. The passage is difficult to understand, but the key may lay in the meaning of onomati of which the LSJ provides a further translation in the sense of a “pretext.” It could imply that tax payers tried to circumvent payments for public funds under the pretext that they had already paid for diagraphai. The imperial order could thus mean that it was not allowed to rededicate money intended for public funds to be used for payments of diagraphai. I thank Ulrike Babusiaux and Anne-Sophie Meyer for sharing their thoughts with me on that issue. The other scenario, a direct physical deduction from public funds, is only hard to envisage since it would raise the question of how property-owners and inhabitants could have accessed such funds in the first place.The old German edition of theNovellae(1833)translatesdiagraphai91For the feminine, cf. n. 89.as “Auflagen” (imposts), that would have been levied, for instance, on the occasion of announcements in the provinces or when the image of the emperor was shown;92Otto, Schilling and Sintenis 1833, vol. 7, 635, n. 26. See also Cod. Iust. 12.64 (383) where with regard to descriptiones it is ordered that announcements should be carried out without inappropriate payments.the new English translation gives “l(fā)evies” without explanation.93Miller and Sarris 2018, 852.As Avshalom Laniado made plausible, the juxtaposition ofktētoresandoikētoresrefers to the entire population, which meansdiagraphaicould be paid by both landowners and those without landed property.94Laniado 2002, 180–191. Ktētores and oikētores are also mentioned in a papyrus dated from the 6th century that listed money payments, inter alia, payments on account of to(u) diagraph(ou), cf.P.Hamb. I 56, col. 1, 1; 20 and col. 7, 1 (6th c., Aphrodito), for the date, cf. Fournet 2008, 328. It was thoroughly discussed by Rémondon 1965, 414, but in the end he noted: “dans le cas de P.Hamb. I 56,il est impossible de dire si le diagraphon a été levé sur les seuls propriétaires, ou si les associations professionelles y ont aussi contribué.”

In the 7th century, after the Muslim conquest of Egypt, we find the same generic group (landowners and those without landed property) paying an impost called in Latindiagrafa. TheLiber Pontificalisrecords that while he was in Sicily (after July 663),95For the date, cf. Zuckerman 2005, 80–81.Constans II imposedafflictionesin the form ofdiagrafa seu capita atque nauticationeonhabitatoresandpossessoresof the provinces of Calabria, Sicily, Africa, and Sardinia.96Duchesne 1892, LXXVIII, IV (p. 344): […] tales afflictiones posuit populo seu habitatoribus vel possessoribus provinciarum Calabriae, Siciliae, Africae vel Sardiniae per diagrafa seu capita atque nauticatione per annos plurimos. Davis 1989, 72 translates capita as “poll-taxes” (cited by Zuckerman 2005, 81, n. 6), but see also Zuckerman 2005, 82 who interprets it as element in the fiscal system of capitatio and iugatio.Constantin Zuckerman suggested that the source of inspiriation for this tax was the Muslim poll-tax (Greekdiagraphon, pl.diagrapha) that had been introduced in Egypt shortly before.97Ibid., 83: “Constans II imported into the empire the poll tax freshly created by the Arabs.” The first evidence for diagraphon in Egypt is possibly SPP VIII 740, 2 (645? Arsinoite), followed by SB VIII 9756 (653, Herakleopolis), with BL VIII, 353.However, as theNovellaeillustrate,diagraphaithat were paid byhabitatoresandpossessoreswere already known in 545. Both thediagraphaiof theNovellaeand Constans’ IIdiagrafawere levied fromhabitatoresandpossessores(oikētoresandktētores), from those with and without land. It is thus likewise possible that Constans’ IIdiagrafawere de facto based on the Byzantinediagraphai. A change in the Byzantine tax terminology or a misunderstanding caused by its translation into Latin, is perhaps easier to envisage than that Constans II took over a newly created tax of Muslim origin. However, more research is necessary, both on the sources of theLiber Pontificaliswith regard to its use of juristic terminology (and whether or not we may perhaps face an anachronism with regard todiagrafa)and the differences between the Byzantine terms used in the papyri and in theNovellae, which is beyond the scope of this article. A Byzantine impost that was not only levied on landed property, but on all inhabitants could in fact have been a suitable precursor of the Muslimdiagraphon,98In research, the thesis has gained ground that in post-conquest Egypt the term diagraphon prevailed over diagraphē, cf. Gonis 2003, 150, ns. 2–3.which is considered a genuine poll-tax under the new Muslim rulers of Egypt. Two early Islamic tax receipts that mention adiagraph(on) Sarakēnōnwould in fact speak in favour of a Byzantine predecessor from which thediagraphonof the Muslims needed to be distinguished.99SPP III 666 (7th–8th c., Arsinoe); VIII 707 (7th c., Arsinoe). If so, the date of the texts could probably be limited to the beginning of Muslim rule.But it remains ultimately unclear whether thediagraphonof the Saracens referred to a different kind ofdiagraphonor the fact that the Byzantinediagraphonwas now levied by Muslims.

Comparing the Byzantine and post-conquest period is, moreover, impeded by the fact that one cannot clearly distinguish the Byzantine from the postconquest evidence since several papyri, which were dated by their editor to the 6th or 6th/7th century, date more likely from the post-conquest period.100For instance, SPP III 657 (6th c., Arsinoe); 663r (6th c., Arsinoe); 664 (6th c., Arsinoe); 765 (6th c.,Hermopolite?). I thank Sophie Kovarik for this information.Some of them, mostly from Arsinoe, mention a unit calledlaura(street or district)101Worp 2004.as basis of assessment ofdiagraphon(hyper/apomerismou)diagraphou laurasxy).102SPP III 652–656; 675 (all 6th–7th c., Arsinoe); 668; 670; 677; 682; 684; 686; 689 (all 7th c.,Arsinoe); 694 (652, Arsinoe); 695; 697; 698; VIII 703–704; 717; 719; 720; 741 (all 7th c., Arsinoe);SB I 5128 (4th–7th c., Arsinoe); CPR XXII 60, 41 (7th–8th c., Arsinoite); BGU II 679 (7th–8th c.,unknown); III 739, 3 (7th–8th c., unknown); P.Lond. I 116a (p. 221) (7th–8th c., Arsinoite), for this date, cf. Schmidt 2018, 611–612; CPR XXII 8, 1–2 (729–730, Herakleopolis).The amounts paid differ from somekeratiato severalnomismatawhich is generally spoken rather unusual for a poll-tax.103See, for instance, P.Lond. I 116a (p. 221) (7th–8th c., Arsinoite): 6 1/2 1/4 keratia, for the date, cf.Schmidt 2018, 611–612; BGU II 679 (7th–8th c., unknown): 7 1/4 keratia; CPR XXII 8, 4 (729–730,Herakleopolis): 1/2 nomisma; SB XXVIII 16871 (7th c., Arsinoite): 2 nomismata 20 keratia. An exhaustive study of the rates is beyond the scope of this article, the reader finds a list with amounts in Poll 1999, tabs. 4–5.2.The Muslim poll-tax was paid in two installments, but even if we assume partial payments, one would expect a certain regularity in the amounts paid.104According to Poll 1999, 240 the Coptic tax receipts and those on ostraca show a higher degree of regular payments than the Greek papyri. He admits, however, that the evidence of 17 Greek receipts on ostraca is too small to draw universal conclusions. The amounts in the Coptic receipts are also very erratic and differ between 1/6 nom., 1/2 nom., 2/3 nom., 5/6 nom., 1 nom, 1 1/3 nom., 1 2/3 nom.and 2 nom. (tab. 5.1).That is why the idea of shares imposed on thelauraas a tax unit, brought forward by Karl Wessely, gained ground in research.105Worp 2004, 246; Gascou 1983, 102; Casson 1938, 276; Wessely 1887, 261–264.Hereafter,diagraphonwas based on the number of persons living in a tax unit and paid per household / association / group of people.106Ibid., 263 refers to SB I 5131 (7th/8th c., Arsinoite) as a possible list of a tax allocation within a laura, but no tax is mentioned. Apart from the laura, there were also other units to calculate the tax quote, such as, for instance, a unit called onoma, cf. CPR XXII, p. 122.However,there may have been a further distinction which has not received a lot of attention in research so far. A yet little understood list oflauraiandepoikiarecords alsoktētoresandousiaiwhich were registered alternately with thelaurai.107SB XXVIII 17063 (7th–8th c., Arsinoite?), published and discussed in Worp 2004, 244–247.This raises the question of whether those listed among alaurawere – in contrast toktētores– persons without landed property, offering again a juxtaposition ofoikētoresandktētores. Could thelaurabe the tax unit for those without land while the quota forktētorestook also their respective landed property into account? This is just a hypothesis for which I cannot provide an answer here.A certain distinction between those who also paid landtax (dēmosia gēs) and those who had no land (ateleis) can also be observed in some fiscal documents in which theateleiswere separated from the others and paid onlydiagraphonanddapanē(maintainance tax).108P.Ross. Georg IV 23 (early 8th c., Aphrodito) in which those paying for dēmosia, computed on the amount of arourae, are separated from the ateleis, who paid diagraphon and dapanē. For a discussion and more examples, see CPR XXII 24 (8th c., unknown) with pp. 122–123.Nonetheless, it should not be concluded thatateleispaid the poll-tax while landowners did not. The letter of the pagarch Athanasios cited above made it clear thatandrismoswas to be paid by both theemporosin the city and thepaganoswhich had probably the same inclusive meaning asktētoresandoikētoresin theNovellaeand just meant “everybody.”109CPR XXII 1, 7 (644?, Hermopolite). Nov. 128, caput XVI, 1–10 (Kroll and Sch?ll 1912).

The discussion above demonstrates that the true nature of the Muslimdiagraphonand its potential origins are yet not fully explored. A further, in my opinion yet too easily dismissed issue in research pertains to the relation of the Muslimdiagraphonandandrismos. At least in 8th century Aphrodito, we experience some inconsistencies with the assumption that they were equivalents.In two letters, Qurra ordered the pagarch Basileios to bring himkatagrapha katachōrion, registers of villages / tax units, including theandrismosprevailing in a place (?) (tou ontos andrismou en autōi). He further wishes to knowei ti esti di’autōn diagraphon kai ei ti hyparchei hekastōi en gēidiois apo te en ampelōi kai en sporimōi gēs.110P.Lond. IV 1339, 5–7 (709, Aphrodito).Not only doandrismosanddiagraphonseem to have different meanings in this document – the “ei” raises, moreover, the question of whetherdiagraphonwas not necessarily paid by everybody.111However, see also P.Lond. IV 1338, 22–23 (709, Aphrodito) where “ei” is missing.The phenomenon that some people paiddiagraphonwhile others did not can also be observed in a tax list of unknown origin from the 8th century.112CPR XXII 24 (8th c., unknown).While all people paid the tax, at least one person bears a “0” for not payingdiagraphonbehind his name.113CPR XXII 24, 6.Also in this regard, more detailed research is needed in order to set out the differences betweenandrismosanddiagraphon, whether or notdiagraphonhad any connection to landed property for its assessment, and whether thediagraphonof the 8th century differed from that of the early conquest period taking into account that the Arabic terminology started to use a term of its own for poll-tax calledjizyah.

The taxes discussed above give only a small idea of what was levied from Egyptians under early Muslim rule. We may add taxes for pursuing a profession,114For instance, the weaving-tax in P.Bal. II 132 (after 740, Deir el-Bala’izah); 133; 134 (both 7th–8th c., Deir el-Bala’izah).imposts for the maintenance of the Muslim administration,115For instance, probata in SPP III2 24B (7th c., Herakleopolite?); III2 88 (2nd half 7th c., Arsinoite),cf. also Morelli 1996, 112.a hospitality tax for the caliph (apo xeniou tou amir al-moumnin),116For instance, P.Pintaudi 27, 3 (7th–8th c., Hermopolis); P.Lond. IV 1433, 20; 94; 148; 195 etc. (707,Aphrodito); O.CrumST 417 (8th c., Thebais); O.Medin. Habu 281, 4–5 (8th c., Djeme).custom dues,117CPR II 228 verso (8th c., Babylon). For customs, cf. also P.Berl.Frisk. 6, 17 (710, Aphrodito);P.Lond. V 1754, 1 (7th–8th c., Babylon).and imposts fordēmosiafrom which the maintenance of soldiers and sailors was paid, among others.118For instance, SB VIII 9750, 5 (642/657, Herakleopolis); CPR XXII 2, 4–5 (643–644, Hermopolis);SB XXVI 16354, 2 (643–644, Hermopolis); SPP VIII 1192b, 1–2 (651/666, Arsinoite); SB XVIII 13771, 11 (677/707, Herakleopolite). For payment of soldiers and sailors: P.Apoll. 28, 1–3; 6 (2nd half 7th c., Apollonopolis Magna); P.Lond. IV 1349, 15–16 (710, Aphrodito); perhaps also SB VI 9576, verso 1 (643, Herakleopolis (?)), but only if the dapanē mentioned on the recto was classified among the dēmosia on the verso.The first dateable evidence ofdēmosiain Muslim Egypt is an acknowledgement of tax reduction for a hospital in 643/644,119CPR XXII 2, 5 (643–644, Hermopolis).but a debt acknowledgement that records a credit loan for the payment ofdēmosiamay already date from 642.120SB VIII 9750, 5 (25.2.642/657, Herakleopolis), for the early date, cf. Morelli 2017, 127 with n. 26.It would thus be contemporaneous with the early Muslim receipt SB VIII 9749 (26.01.–24.02.642),mentioning an order of the amīr Harigatos. Grohmann 1957, 12, l. 2 comm., referring to Karabacek,identified him with Khāri?a b. ?udhāfa al-‘Adawī al-Qura?ī whom Legendre 2013, 106, n. 140 related to the conqueror of Akhmīm.Sincedēmosiahappened to be paid by those with121For instance, P.Ross. Georg. IV 23, 15 (early 8th c., Aphrodito); SB XVI 12481, 19 (668, Arsinoe);SPP VIII 812, 2–3 (7th c., Arsinoite); CPR VIII 80, 2; 4 (7th–8th c., unknown); in several texts, the land is called dēmosia gēs as, for instance, in CPR XIX 26, 2–3 (718/733, Herakleopolis). As a letter from Qurra indicates, it could be paid from money farmers received by selling their harvest, cf. P.Lond.IV 1380, 4; 12–19 (710, Aphrodito). For a general discussion, see Bell in P.Lond. IV, intro §3, passim.and those without landed property,122Like, for instance, the ergasia of paktones-constructors, cf. SPP VIII 878 (7th c., Arsinoite); see also SPP VIII 842 (7th c., unknown).the opinion gained ground in research that the term could be used for both the money tax on land (dēmosionordēmosia gēs)and generally all taxes in money (chrysika dēmosia).123Gonis 2004a, 215–216. See, moreover, the Coptic tax receipt CPR IV 12, 2–5 (7th c., unknown):“Ich habe empfangen und bin befriedigt worden (plēroun) (3) von dir für die Steuern (dēmosion),die du mir zahlst nach (pros) Ma?gabe (4) deines Verkaufs (prasis), embolē, chrysikon, hexakeraton,kurz (haplōs) (5) alles für das 5. Indiktionsjahr.” (trans. Till).

The individual pressure of taxes caused many to flight, a reaction that is known very well from the Byzantine period where outstanding taxes were redistributed among remaining inhabitants of a tax unit.124For the anachōrēsis in Graeco-Roman Egypt, see J?rdens 2009, 304–330. An example from the Muslim period is P.Lond. IV 1343 (709, Aphrodito).However, despite the high amount of taxes and imposts that were demanded from the inhabitants it seems as if the Muslim administration was cautious in acting fair and accurate. This is demonstrated, for instance, by the example of the pagarch Rashid ibn Khaled under whose administration of the Hermopolite nome (724–731) an overpayment of money was returned to the taxpayers.125Schenke 2014.Corrupt officials and injustice in collecting taxes were obstructive to maintaining a balance between exploiting ressources and keeping up the inhabitants’ willingness to obey to this system which explains why we repeatedly find admonitions from Muslim authorities to treat tax payers with fairness.126See, for instance, P.Lond. IV 1345 (709, Aphrodito); 1356 (710, Aphrodito); 1380, 25–28 (710,Aphrodito).

The levy of money taxes and the reimbursements for the delivery of goods right at the beginning of Muslim rule indicate that at least in Egypt, the Muslim army did not rely only onad hocbooty and requisitions in the early years. We may assume a complex supply system that was partly built upon already existing tax collection structures. The influx of money was used in various ways. Money that was not transported to the Muslim capital was spent in Egypt – either to maintain the troops and administration or to fund local infrastructure projects. The supply of troops took the form of stipends known as?a?ā?. It was distributed to all members being recorded in the so-calleddīwān al-jund, a register, that contained eligible soldiers and their dependents.127It probably differed from province to province; in Egypt it is said to have been established by the first Muslim governor ?Amr b. al-‘ā? (first office 642–645; second 658–663). For the jund, see the discussion in Sijpesteijn 2010, 255 who published an important document from the 7th/8th century that gives insight into the receiver’s perspective of this stipend. Under Qurra, dependents were still eligible to receive a pension as is documented in P.Heid. Arab. I 1, 8–9 (710, Aphrodito). There is a dissertation on the jund of Egypt, but as to my knowledge it is still unpublished: Bouderbala 2008.See, moreover, Morimoto 2012; Kennedy 1995; Puin 1970.Although the soldiers were apparently well supplied with food and commodities,128In Nessana, for instance, they received monthly rations, cf. P.Ness. III 60–63 (674–675), discussed in Sijpesteijn 2010, 256.it is to be assumed that they invested the money they received in consumer products from Egyptian markets.129The Muslim historian Ibn ?Abd al-?akam (Hilloowala), 56, writing in the 9th century, stressed the importance of markets for the invaders when he reports that already during the invasion, ?Amr demanded that markets be built between Alexandria and al-Fus?ā?. As Bruning 2018, 58–86 pointed out in his thesis, the market of Alexandria retained its importance for Copts, Muslims and foreign trade business in the early years of Muslim rule.The earliest evidence of Muslim private financial transactions in Egypt – and the earliest document at all for Muslim presence so far south of Egypt!130See n. 39.– is an Arabic acknowledgement from modern Asyūt that was re-dated recently to 641.In this, ?Abd Allāh b. ?Umays acknowledged to have paid threenomismatato Isidoros and his wife.131Tillier and Vanthieghem 2019, 168–172, no. 1.The formula leaves some room for the question of whether it could be a receipt for the reimbursement of requisitions. However, as discussed above, at that time Muslims handled the acquisition of goods and their reimbursement usually with higher officials who forwarded the money to the contributors/producers. Moreover, it is noteworthy that the receipt was written in Arabic while early documents issued for locals were either bilingual or written in Greek. This supports the editors’ assumption that the document was an individual debt acknowledgement intended for the internal Muslim bureaucracy.132Ibid., 149–150.As Mathieu Tillier and Na?m Vanthieghem demonstrated, the administration in al-Fus?ā? kept debt registers that documented loans Muslim stipend-holders took up privately until they were repaid from their yearly pensions.133Ibid., 151–153. Examples of such Arabic registers of debt acknowledgement are ibid., nos. 2(676–677, al-Fus?ā??) and 3 (667–668, al-Fus?ā??), the latter lists a Christian name among Muslims.The document issued to Isidoros could thus have been a receipt for the repayment of a debt which was kept in a central financial office in al-Fus?ā?.134See also Rā?ib 2007, 194–201. A further, but Greek example of a Muslim-Christian debt relationship, may be a notification of the amīr ‘Ubayd to a certain Kalomēnas. The latter should furnish 6 nomismata to ‘Ubayd, but at the same time there is a passus regarding the return of money(kalōs poieis parechōn moi ta 6 nomismata, haper echete m[oi], hina en(i) en chrei[ai] palin dōsō hymin), cf. SB VIII 9748, 2–3 (middle of 7th c., Arsinoite). Until now this communication has been interpreted as order to hand over money, but the expression about the return of money in times of need is also reminiscent of debt formulas.

To sum up: From its very beginning, Muslim administration firmly depended on a sufficient supply with money and goods and a strong local economy. Tribute and plunder were one means to exploit resources of a conquered country. The papyri and ostraca show, however, a more sophisticated, even though not less exploitive approach by the Muslim invaders. Several examples show how Muslims took over the existing tax infrastructure, even retaining the terminology,and adapted it to their needs. Taxes were used to finance the military and new administration, but also to buy goods and services in the country – which kept productivity at a high level. A familiar tax system and the continuity of market mechanisms that ensured work and pay certainly increased the willingness among the new subjects to cooperate. Economic collaboration and a well-tempered carrot-and-stick policy laid thus the ground for the economy of the embryonic Muslim Empire.

Bibliography

Bagnall, R. S. 1980.

“P.Oxy. XVI 1905, SB V 7756 and Fourth-Century Taxation.”Zeitschrift für Papyrologie und Epigraphik37: 185–196.

Bell, H. I. 1933.

“Bibliography: Graeco-Roman Egypt A. Papyri (1931–1932). Ch. 3: Publications of Non-literary Texts.”The Journal of Egyptian Archaeology19/1–2: 67–93.

Ben Shemesh, A. (ed.). 1967.

Ya?yā b. ādam, Kitāb al-kharāj. Taxation in Islam. Leiden: Brill.

Berkes, L. 2017a.

Dorfverwaltung und Dorfgemeinschaft in ?gypten von Diokletian zu den Abbasiden. Wiesbaden: Harrassowitz.—— 2017b.

“1. Letter from a Muslim Official about Requisitions for the Fleet.” Part of:A. Boud’hors et al. ‘Un noveau depart pour les archoves de Paps Papyrus coptes et grecs de la jarred d’Edfou’.”Bulletin de l’Institut Fran?ais d’Archéologie Oriental117: 87–124, here: 90–93.

Booth, Ph. 2013.

“The Muslim Conquest Reconsidered.” In: C. Zuckermann (ed.),Constructing the Seventh Century. Paris: Association des Amis du Centre d’Histoire et Civilisation de Byzance, 639–670.—— 2016.

“The Last Years of Cyrus, Patriarch of Alexandria (? 642).” In: J.-L. Fournet and A. Papaconstantinou (eds.),Mélanges J. Gascou (P.Gascou). Paris: Association des Amis du Centre d’Histoire et Civilisation de Byzance, 509–558.

Bouderbala, S. 2008.

?und Mi?r: étude de l’administration militaire dans l’égypte des débuts de l’Islam 21/642–218/833. PhD-thesis: Université de Paris 1 – Panthéon-Sorbonne,Paris.

Bruning, J. 2018.

The Rise of a Capital: Al-Fus?ā? and its Hinterland, 18/639–132/750. Leiden:Brill.

Bruning, J., De Jong, J. and Sijpesteijn, P. M. forthcoming.

Incorporating Egypt. From Constantinople to Baghdad, 500–1000 CE.Cambridge: Cambridge University Press.

Casson, L. 1938.

“Tax-collection Problems in Early Arab Egypt.”Transactions and Proceedings of the American Philological Association69: 274–291.

Clackson, S. J. 2009.

It is our Father Who Writes. Orders from the Monastery of Apollo at Bawīt(P.Bawit Clackson). Cincinnati: American Society of Papyrologists.

Cromwell, J. 2013.

“Coptic Texts in the Archive of Flavius Atias.”Zeitschrift für Papyrologie und Epigraphik184: 280–288.

Davis, R. 1989.

The Book of Pontiffs (Liber Pontificalis). The Ancient Biographies of the First Ninety Roman Bishops to AD 715.Liverpool: Liverpool University Press.

Donner, F. 1986.

“The Formation of the Islamic State.”Journal of the American Oriental Society106/2: 283–296.

Duchesne, L. 1892.

Le Liber Pontificalis.Paris: Libraire des écoles Fran?aises d’Athènes et de Rome, du Collège de France et de l’école Normale Superieure.

F?rster, H. 2002.

W?rterbuch der griechischen W?rter in den koptischen dokumentarischen Texten.Berlin: De Gruyter.

Foss, C. 2009a.

“Egypt under Mu?āwiya. Part I: Middle Egypt, Fus?ā? and Alexandria.”Bulletin of the School of Oriental and African Studies72: 1–24.—— 2009b.

“Egypt under Mu?āwiya. Part II: Middle Egypt, Fus?ā? and Alexandria.”Bulletin of the School of Oriental and African Studies72: 259–278.

Fournet, J.-L. 2008.

“Liste des papyrus édités de l’Aphrodité byzantine.” In: id. (ed.),Les archives de Dioscore d’Aphrodité cent ans après leur découverte. Histoire et culture dans l’égypte byzantine. Actes du colloque de Strasbourg (8–10 décembre 2005).Paris: De Boccard, 307–343.

Gascou, J. 1983.

“De Byzance à l’Islam. Les imp?ts en Egypte après la conquête arabe.”Journal of the Economic and Social History of the Orient26/1: 97–109.—— 2010.

“Ostraca byzantins d’Edfou et d’autres provenances.” In:Mélanges C. Morisson.Travaux et mémoires. Collège de France, Centre de Recherche d’Histoire et Civilisation de Byzance 16. Paris: De Boccard, 359–385.

Gascou, J. and Worp, K. A. 1982.

“Problèmes de documentation Apollinopolite.”Zeitschrift für Papyrologie und Epigraphik49: 83–95.

Gonis, N. 2001.

“Reconsidering Some Fiscal Documents from Early Islamic Egypt.”Zeitschrift für Papyrologie und Epigraphik137: 225–228.—— 2003.

“Five Tax Receipts from Early Islamic Egypt.”Zeitschrift für Papyrologie und Epigraphik143: 149–157.—— 2004a.

“Arabs, Monks, and Taxes. Notes on Documents from Deir el-Bala’izah.”Zeitschrift für Papyrologie und Epigraphik148: 213–224.—— 2004b.

“Reconsidering Some Fiscal Documents from Early Islamic Egypt II.”Zeitschrift für Papyrologie und Epigraphik150: 187–193.—— 2008.

“SB VI 8986 and Heraclius’ Sons.”Zeitschrift für Papyrologie und Epigraphik166: 199–202.—— 2009.

“Reconsidering Some Fiscal Documents from Early Islamic Egypt III.”Zeitschrift für Papyrologie und Epigraphik169: 197–208.

Gonis, N. and Morelli, F. 2002.

“Two ‘Entagia’ in Search for an Author.”Bulletin of the American Society of Papyrologists39/1–4: 17–25.

Grohmann, A. 1952.

From the World of Arabic Papyri.Cairo: Al-Maaref.—— 1957.

“Greek Papyri of the Early Islamic Period in the Collection of Archduke Rainer.”études de Papyrology8: 5–40.

Harrauer, H. 2001.

“Neue Protokometen-Papyri. Mit einer Dokumentation der Protokometen.”Aegyptus81: 47–159.Hilloowala, Y. 1969.

The History of the Conquest of Egypt, Being a Partial Translation of Ibn ‘Abd al-?akam’s“Futuh Misr”and an Analysis of this Translation.PhD-thesis:University of Arizona, accessed under: http://hdl.handle.net/10150/282810(12.02.2020).

Hoyland, R. 1997.

Seeing Islam as Other Saw it. A Survey and Evaluation of Christian, Jewish and Zoroastrian Writings on Early Islam. Princeton: The Darwin Press.—— 2006.

“New Documentary Texts and the Early Islamic State.”Bulletin of the School of Oriental and African Studies69/3: 395–416.

Huebner, S. et al. (eds.). 2020.

Living the End of Antiquity. Individual Histories from Byzantine to Islamic Egypt.Berlin: De Gruyter.

Hussein, F. 1982.

Das Steuersystem in ?gypten von der arabischen Eroberung bis zur Machtergreifung der ?ūlūniden 19–254/639–868 mit besonderer Berücksichtigung der Papyrusurkunden. Frankfurt a.M. et al.: Peter Lang.

Johns, J. 2003.

“Archaeology and the History of Early Islam. The First Seventy Years.”Journal of the Economic and Social History of the Orient46/4: 411–436.

Kennedy, H. N. 1995.

“The Financing of the Military in the Early Islamic State.”The Byzantine and Early Islamic Near East3: 361–378.

Khan, G. 2019.

“The Opening Formula and Witness Clauses in Arabic Legal Documents from the Early Islamic Period.”Journal of the American Oriental Society139/1: 23–40.

Kramer, J. 2011.

Von der Papyrologie zur Romanistik.Archiv für Papyrusforschung Beihefte 30.Berlin: De Gruyter.

Kroll, W. and Sch?ll, R. 1912.

Corpus Iuris Civilis. Novellae. Berlin: Weidmann.

Laniado, A. 2002.

Recherches sur les notables municipaux dans l’Empire protobyzantin.Monographies du Centre de recherche d’Histoire et Civilisation de Byzance 13.Paris: Association des Amis du Centre d’Histoire et Civilisation de Byzance.

Legendre, M. 2013.

Pouvoir et territoire. L’administration islamique en Moyenne-égypte pré-?ūlūnide (642–868).PhD-thesis: Leiden University.—— 2016.

“Neither Byzantine nor Islamic? The Duke of the Thebaid and the Formation of the Umayyad State.”Historical Research89/243: 3–18.

Mayerson, P. 1994.

“?ουζικ?ν and ?ογ? in the Post-Conquest Papyri.”Zeitschrift für Papyrologie und Epigraphik100: 126–128.

Miller, D. J. D. and Sarris, P. 2018.

The Novels of Justinian. A Complete Annotated English Translation. Cambridge:Cambridge University Press.

Mirkovi?, M. 2008.

“Les ktêtores, les syntelestai et l’imp?t.” In: J.-L. Fournet (ed.),Les archives de Dioscore d’Aphrodité cent ans après leur découverte. Histoire et culture dans l’éypte byzantine. Actes du colloque de Strasbourg (8–10 décembre 2005). Paris:De Boccard, 191–202.

Morelli, F. 1995.

“P.Laur. IV 192. Rilettura dells γν?σι?.”Analecta Papyrologica7: 165–171.—— 1996.

Olio e retribuzioni nell’Egitto tardo (V–VIII d. C.). Florence: Istituto papirologico G. Vitelli.—— 2010a.

“‘Amr e Martina. La reggenza di un’imperatrice o l’amministrazione araba d’egitto.”Zeitschrift für Papyrologie und Epigraphik173: 136–157.—— 2010b.

“Consiglieri e comandanti. I titoli del governatore arabo d’Egitto symboulos e am?r.”Zeitschrift für Papyrologie und Epigraphik173: 158–166.—— 2013.

“Egitto arabo, papiri e papirologia greci.” In: T. Derda, A. ?ajtar and J. Urbanik(eds.),Papyrology AD 2013. 27th international Congress of Papyrology=Journal of Juristic Papyrology43: 163–186.—— 2016.

“Duchi ed emiri. Il gioco delle skatole cinesi in PSI XII 1266/P.Apoll. 9.” In:A. Casanova, G. Messeri and R. Pintaudi (eds.),E sè d’amici pieno. Omaggio di studiosi italiani a G. Bastianini per il suo settantesimo compleanno.Florence: Edizioni Gonnelli, 267–282.—— 2017.

“Per o contro il nemico? Le razioni di un carpentiere e la cronologia dell’invasione araba secondo P.Vindob. G 39718 = PERF 550.”Tyche32: 119–129.—— 2019.

I prezzi dei materiali e prodotti artigianali nei documenti tardoantichi e del primo periodo Arabo (IV ex. – VIII d. C.). Mitteilungen aus der Papyrussammlung der ?sterreichischen Nationalbibliothek (Papyrus Erzherzog Rainer) N.S. XXXIII.Berlin: De Gruyter.

Morimoto, K. 2012.

“The D?wans as Registers of the Arab Stipendiaries in Early Islamic Egypt.” In:F. M. Donner (ed.),The Articulation of Early Islamic State Structures. Farnham:Ashgate Variorum, 227–239.

—— 1981.

The Fiscal Administration of Egypt in the Early Islamic Period. Kyoto: Dohosha Publisher Inc.

Noth, A. and Conrad, L. I. 1994.

The Early Arabic Historical Tradition. A Source-Critical Study. Princeton: The Darwin Press.

Otto, C. E., Schilling, B. and Sintenis, F. F. 1833.

Das Corpus Juris Civilis in’s Deutsche übersetzt von einem Verein Rechtsgelehrter. vol. 7. Leipzig: Carl Focke.

Palme, B. 2011.

“The Administration of Egypt in Late Antiquity.”Imperium and Officium Working Papers (IOWP)May, accessed under: https://iowp.univie.ac.at/sites/default/files/IOWP_palme_administration02.pdf (30.06.2020).

Papaconstantinou, A. 2010.

“Administering the Early Islamic Empire. Insights from the Papyri.” In: J. Haldon(ed.),Money, Power and Politics in Early Islamic Syria. Ashgate: Ashgate Publishing Company, 57–74.—— 2015.

“The Rhetoric of Power and the Voice of Reason. Tensions between Central and Local in the Correspondence of Qurra ibn Sharīk.” In: S. Procházka, L. Reinfandt and S. Tost (eds.),Official Epistolography and the Language(s) of Power.Proceedings of the First International Conference of the Research Network Imperium & Officium (Pap. Vind. 8).Vienna: Verlag der ?sterreichischen Akademie der Wissenschaften, 267–281.

Pharr, C. 1952.

The Theodosian Code and Novels and the Sirmondian Constitutions. New York:Greenwood Press.

Poll, I. 1999.

“Die δι?γραφον-Steuer im sp?tbyzantinischen und früharabischen ?gypten.”Tyche14: 237–274.Puin, G. R. 1970.

Der Dīwān von ?Umar ibn al-?a??āb: Ein Beitrag zur frühislamischen Verwaltungsgeschichte. Bonn: Rheinische Friedrich-Wilhelms-Universit?t.

Rā?ib, Y. 2007.

“Une ère inconnue d’égypte musulmane: L’ère de la juridiction des croyants.”Annales Islamologiques41: 187–207.

Rémondon, R. 1965.

“P.Hamb. 56 et P.Lond. 1419 (notes sur les finances d’Aphrodito du VIe sècle au VIIIe).”Chronique d’égypte79: 401–430.

Richter, T. S. 2009.

“Greek, Coptic, and the ‘Language of the Hijra’. Rise and Decline of the Coptic Language in Late Antique and Medieval Egypt.” In: H. Cotton et al. (eds.),From Hellenism to Islam. Cultural and Linguistic Change in the Roman Near East.Cambridge: Cambridge University Press, 402–446.

Robinson, C. 2000.

Empires and Elites after the Muslim Conquest. Cambridge: Cambridge University Press.Schenke, G. 2014.

“Rashid ibn Chaled and the Return of Overpayments.”Chronique d’égypte177:202–209.

Schmidt, S. 2019.

“Adopting and Adapting. Zur Kopfsteuer im frühislamischen ?gypten.” In:A. Nodar and S. Torallas Tovar (eds.),Proceedings of the 28th Congress of Papyrology. Barcelona 1–6 August 2016. Barcelona: Publicacions de l’Abadia de Montserrat, 609–616.—— forthcoming.

“Border Trade between Egypt and Nubia in the 8th Century and Some Considerations on the baq?.” In: S. Esders, S. Polla and T. S. Richter (eds.),The 8th Century. Patterns of Transition in Economy and Trade Throughout the Late

Antique, Early Medieval and Islamicate Mediterranean in Multidisciplinary Perspectives.Berlin: De Gruyter.

Sharon, M. 1991.

“The Umayyads as Ahl al-Bayt.”Jerusalem Studies in Arabic and Islam14:115–152.

Sijpesteijn, P. M. 2007a.

“The Arab Conquest of Egypt and the Beginning of Muslim Rule.” In: R. S. Bagnall(ed.),Egypt and the Byzantine World 300–700. Cambridge: Cambridge University Press, 437–459.—— 2007b.

“New Rule Over Old Structures: Egypt after the Muslim Conquest.” In: H. Crawford(ed.),Regime Change in the Ancient Near East and Egypt: From Sargon of Agade to Saddam Hussein. Proceedings of the British Academy 136. London: British Academy, 183–200.—— 2010.

“Army Economics. An Early Papyrus Letter Related to ?a?ā? Payments.” In:M. E. Roxani, A. Sabra and P. M. Sijpesteijn (eds.),Histories of the Middle East.Studies in Middle Eastern Society, Economy and Law in Honour of A. L. Udovitch.Leiden: Brill, 245–268.—— 2013.

Shaping a Muslim State. The World of a Mid-Eighth-Century Egyptian Official.Oxford: Oxford University Press.

Simonsen, J. B. 1988.

Studies in the Genesis and Early Development of the Caliphal Taxation System.Copenhagen: Akademisk Forlag.

Tillier, M. and Vanthieghem, N. 2019.

“Recording Debts in Sufyānid Fus?ā?: A Reexamination of the Procedures and Calendar in Use in the First/Seventh Century.” In: J. Tolan (ed.),Geneses: A Comparative Study of the Historiographies of the Rise of Christianity, Rabbinic Judaism and Islam. London: Routledge, 148–188.

Trombley, F. R. 2013.

“Fiscal Documents from the Muslim Conquest of Egypt: Military Supplies and Administrative Dislocation, ca. 639–644.”Revue des études byzantines71: 5–38.

Wallace, S. L. 1938.

Taxation in Egypt from Augustus to Diocletian. New York: Greenwood Press.

Wessely, K. 1887.

“Stra?enverzeichnisse aus Arsinoe.”Mitteilungen aus der Sammlung der Papyrus Erzherzog Rainer3: 261–264.

Worp, K. A. 2004.

“Town Quarters in Greek, Roman, Byzantine and Early Arab Egypt.” In:P. M. Sijpesteijn and L. Sundelin (eds.),Papyrology and thr History of Early Islamic Egypt. Leiden & Boston: Brill, 227–248.

Zuckerman, C. 2005.

“Learning from the Enemy and More. Studies in ‘Dark Centuries’ Byzantium.”Millenium. Jahrbuch zu Kultur und Geschichte des ersten Jahrtausends n.Chr.2:79–135.