A Late Ptolemaic Grapheion Archive in Berkeley
Skip other details (including permanent urls, DOI, citation information)
This work is protected by copyright and may be linked to without seeking permission. Permission must be received for subsequent distribution in print or electronically. Please contact email@example.com for more information. :
For more information, read Michigan Publishing's access and usage policy.
In the winter of 1899 to 1900, Grenfell and Hunt excavated a number of crocodile mummies at Tebtunis. These crocodile mummies were wrapped in papyri, which are now housed at the Berkeley Center for Tebtunis Papyri. One group of crocodile mummies produced the famous papyri of Menches, the village scribe of Kerkeosiris in the late 2nd century BC. Another group of crocodile mummies, the so-called "First Batch," has produced both Greek and Demotic papyri from the village of Theogonis in the Fayum from the early 1st century BC. At least some of the papyri from the "First Batch" of crocodile mummies appear to come from the archive of a bilingual grapheion. This paper will first describe the papyri from the grapheion archive. Then it will compare the contents of these papyri to those in the early Roman grapheion archive from Tebtunis, and to Ptolemaic papyri concerned with the registration of Demotic and Greek contracts. Finally, a model describing the evolution of contract registration procedures and the development of the institution of grapheia will be proposed.
Source of the Grapheion Papyri
There are approximately 47 papyri from the apparent bilingual grapheion archive. The papyri derive from two crocodile mummies (crocodiles 1 and 19) found together in a single tomb (crocodile tomb "a") in the cemetery of Tebtunis. These papyri were studied at Oxford, but Grenfell and Hunt published only three of them as descripta in P.Tebt. I in 1902, probably because the majority of the texts were in Demotic. The papyri were then shipped in 1938 to the University of California, Berkeley. Since then, only the Demotic text of one of the descripta has received full publication by Richard Parker in REgypt 24 in 1972.
Date of the Grapheion Papyri
The texts can be broadly dated to the late 2nd or early 1st century BC, based on the references to large sums of inflated copper money. They include specific mention of regnal years 13, 14, 15, 17, and 18. These are presumably sequential regnal years, because the Demotic texts are all written by the same hand and were found together in the same crocodile mummies. Within the late 2nd and early 1st centuries BC, the sequence of regnal years 13–18 is only attested for Ptolemy XII (r. 81–51 BC), and would correspond to years 69 to 63 BC.
Contents of the Grapheion Papyri
The 47 papyri from the posited bilingual grapheion archive are written predominantly in Demotic, and sometimes include Greek summaries at the bottom of the text or on the verso. The Demotic texts are all written with a fine reed pen in identical or very similar cursive hands. Although the papyri are quite fragmentary and are sometimes difficult to identify, most seem to fall into one of three categories:
- abstracts of contracts: 23 papyri including
- P.Tebt. I 227 descr. = REgypt 24 (1972) 129–136.
- Tebt. frags. 6422–6423, 6451–6455, 6473–6479, 6491, 6501, 6531–6532.
- registers of contracts in daybook format: 7 papyri including
- P.Tebt. I 228 descr.
- Tebt. frag. 6503.
- accounts of money payments in daybook format: 10 papyri including
- P.Tebt. I 140 descr.
- UC 1731, 2489–2490.
- Tebt. frag. 8141.
Category 1: Abstracts of Contracts
The largest group consists of approximately 23 fragmentary abstracts of Demotic contracts. These usually begin with the year, month and day, but without the names or epithets of the king or the eponymous priests. The date is then followed by the phrase: "a document of type X, which Contractor A made it" (wꜥ sẖ X r.ir s PN A). This formula is radically different from the usual formula of Demotic contracts, which begins with the dating and proceeds with "has said Contractor A to Contractor B" (ḏd Contractor A n Contractor B). Richard Parker suggested that one of these texts, P.Tebt. I 227, was an abstract of a contract. This suggestion now seems to have been confirmed by an early Ptolemaic register of contracts among the Lille papyri, published by Françoise De Cenival in 1987. The register is organized in daybook format, often with multiple entries recorded under each day. The entries often begin "a document of type X, which Contractor A made it" (wꜥ sẖ X r.ir s PN A), which is exactly the formula used in the Tebtunis abstracts. Three examples will illustrate the formula which recurs in the Tebtunis texts:
P.Tebt. I 227 descr. = REgypt 24 (1972) 129–136, dated to year 18, Phamenoth 18, lines 1–4: "A document of sale and quitclaim which the farmer and servant of Souchos, Paesis son of Paesis, his mother is Kolluthes, made it, concerning his half share of his house ..." (wꜥ sẖ ḏbꜣ ḥḏ wꜣy | ir s wyꜥ bꜣk Sbk Pa-is.t (sꜣ) Pa-is.t | mw.t(=f) Ḳlwḏ ḥr tꜣy=f tny.t pšy.t n | pꜣy=f ꜥ.wy).
Tebt. frag. 6501, dated to [Year lost], Hathyr 2, lines 1–4: "A document of sale which the farmer and servant of Souchos, Petechonsis? son of Marres, his mother is Ta-...?, made it, [concerning his] one share in five shares of his house ..." (wꜥ sẖ ḏbꜣ ḥḏ | r.ir s wyꜥ bꜣk Sbk Pꜣ-ti-ḫnsw? | Mꜣꜥ-rꜥ, mw.t=f Tꜣ-n.t-...? | ... tny.t wꜥ.t ẖn tny.t 5.t n pꜣy=f ꜥ.wy).
Tebt. frag. 6473, dated to Year 15, Hathyr 6, lines 2–3: "A lease? which the cavalryman Ptolemaios ... made it [concerning] his vineyard which measures 4+ arourae ... " (wꜥ sḥn? r.ir s rmt ḥtr ... Ptrw[m]y[s] ... | pꜣy=f ꜣḥ ꜣlly nt ir 4.t ? ꜣḥ ...).
Category 2: Registers of Contracts
Perhaps not coincidently, a second group of texts consists of seven fragments of what appear to be bilingual registers of contracts, some clearly in daybook format, others clearly recording money payments. These documents list a series of days of the month, each followed by one or more entries beginning: "a document of type X, of Contractor A" (wꜥ sẖ X n PN A). Curiously, the formula is different from that of the Berkeley abstracts and the entries in the Lille register of contracts, which begin "a document of type X, which Contractor A made it." The entries may be followed by a sum of money. The entries for the last day of the month are followed by the total of the sums of money for the month. There may also be a Greek label on the otherwise blank verso, giving the date and nature of the text.
P.Tebt. I 228 descr., dated to a Year 13, Thoth 30, several entries begin: [Demotic] "the annuity contract of PN son of PN ...". (pꜣ sẖ n sꜥnḫ n PN <sꜣ> PN ...). Register ends, lines 13–14: [Demotic] "completes the money of the labor of writing, 144 deben ..." (r ḥḏ hy n sẖ | ḥḏ 144 n ...). Label on verso, lines 1–3: [Greek] "The surplus for the royal treasury, one copper talent 615 (drachmas), Year 13, Thoth 30, in Demotic" ([πε]ρ̣ισσεύει εἰς τ̣ὰ̣ βασι|λ[ι]κὰ χα(λκοῦ) (τάλαντον) α χιε | (ἔτους) ιγ Θωυτ λ ἐ̣ν̣χ̣ωρίοις).
Category 3: Accounts of Money Payments
The third group of texts consists of ten fragmentary accounts of money payments in daybook format. Most of the accounts are written in Demotic with Greek summaries, and at least one account is entirely in Greek. These texts list a series of days of the month, each followed by the names of one or more commodities, followed by their price. Sometimes also a note about the purpose of the commodities is made, such as "for me," and the pattern is repeated for each day of the month. In the demotic accounts, the entries for the last day of the month are followed by the mention, in Greek, of the total for the entire month. There may also be a Greek label on the otherwise blank verso, giving the date and nature of the text. Here follow three examples of the particularities of this category of text.
The account P.Tebt. I 140 descr., dated to Year 10+?, Thoth 28, begins, lines 1–3: [Greek] "Year 10[+?], Thoth 28. Account of the royal revenues and of expenses and of the price of unwritten rolls of the contract-writers' offices of Kerkethoeris and Theogonis" (Ἔτους ι[.] Θωυθ κη̅. Λόγος βασιλικῶν καὶ δαπά|[ν]ης καὶ τειμῆς ἀγράφων συναλλαγματογραφι|ῶ̣ν̣ Κερκεθοήρεως καὶ Θεογονίδος).
UC 1731, dated to Year 14, entries for Phamenoth 28 and 29 in Demotic for the price of cheese and dates, and for the price of sesame and cabbage "for me". Account ends, lines 10–15: [Greek] "makes until day 30, 665 (drachmas), makes the total expenses of Phamenoth of Year 14, one copper talent 4665 (drachmas), makes one talent 4665 (drachmas)", [Demotic] "for Phamenoth, makes a total of 533 deben 2 1/2 kite", [Greek] "Year 14, Pharmouthi 2" ((γίνεται) ἕως λ χξε | (γίνεται) ἡ πᾶσα δαπάνη τοῦ Φαμενὼθ | τοῦ ιδ (ἔτους) χα(λκοῦ) (τάλαντον) α ᾽Δχξε | (γίνεται) (τάλαντον) α ᾽Δχξε | r ibt 3 pr.t r dmḏ ḥḏ 533 ḳt 2 1/2 | ἔ̣τ̣ο̣υς ιδ̅ Φ̣αρμοῦτι β̅).
UC 2489, dated to Year 17, entries for Epeiph 25?, 26, 29? and 30? in Demotic for the price of sesame. Account ends, lines 8–10: [Greek] "makes the total capital, one talent 3155 (drachmas), Year 17, Mesore 1 morning" ((γίνεται) τὸ πᾶν κεφάλαιον | (τάλαντον) α ᾽Γρνε (ἔτους) ιζ | Μεσορὴι α π̣ρω̣ί̣). Label on verso, lines 1–3: [Greek] "Account of the grammatikon and of the expenses and of the other costs of Epeiph of Year 17 in Demotic writing, Year 17, Mesore 1 morning" (λόγος γρα[μ]ματ̣ι̣κῶν καὶ δαπάνης | καὶ ἄλλων ἀνηλωμάτων τοῦ Ἐπεὶφ τοῦ ιζ (ἔτους) | ἐνχωρίοις γράμμασιν (ἔτους) ιζ Μεσορὴ α πρω̣ί̣).
Identification as a Grapheion Archive
The identification of the 47 Berkeley papyri as a grapheion archive is based on a comparison with the early Roman grapheion archive from Tebtunis, now housed in Ann Arbor and in Florence. The early Roman Tebtunis grapheion archive consists of about 210 papyri dating from 7 AD to 56 AD. Most of the papyri comprised in it fall into one of four categories, three of which correspond to those in the Berkeley papyri:
- The early Roman Tebtunis archive contains a number of εἰρόμενον-registers, whose long entries include abstracts of contracts. These seem to correspond functionally to the separate abstracts among the Berkeley papyri. (P.Mich. II 121 recto; V 241; and P. Mich. inv. 940, 949, 950, and 939).
It includes several ἀναγραφή-registers, whose shorter entries give only the type of contract and the name of the first contractor. These seem to correspond to the registers of contracts in the Berkeley papyri.
Some of these ἀναγραφή-registers serve as indexes to the εἰρόμενον-registers (P.Mich. II 121 verso). Some serve as accounts of the payment of the γραμματικόν, the fee paid by the contracting parties to the grapheion to draw up and register the contracts (P.Mich. II 125.10–28; II 126.1–13; II 128, cols. i(b), ii.1–27, iii; V 237–240). Others serve as accounts of the payment of the διαγραφή, the fee paid by the notary officials to lease the grapheion (P.Mich. II 123 recto, cols. ii–xxii; II 124 recto, cols. i–ii, and verso).
- It also includes a number of other accounts of revenues and daily expenses of the grapheion, which seem to correspond to the accounts of money payments in the Berkeley papyri (P.Mich. II 123 recto, col. i, and verso cols. ii–xii; II 124 recto, col. iii; II 125.1–9; and II 128, cols. i(a), ii.28–45).
- Lastly, the large number of incomplete versions and copies of Greek and bilingual contracts found in the early Roman grapheion archive is seemingly without parallels in the Berkeley papyri (P.Mich. V 242–356; X 586–587; XI 621; XII 632–634; PSI VIII 901–918; SB VI 9109–9110; XVI 12539; XX 14313–14315; P.Bingen 59).
Proposed Relationship Between the Ptolemaic and Roman Grapheia
The similarities between the categories of documents in the early Roman grapheion archive from Tebtunis, and the late Ptolemaic Berkeley papyri, strongly suggest that the latter also belonged to a grapheion archive. Furthermore, the similarities raise the possibility that the late Ptolemaic grapheia are the institutional ancestor of the early Roman grapheia. The Ptolemaic ancestry of the early Roman grapheia can perhaps also be seen in a Greek offer to lease a grapheion from Soknopaiou Nesos, dated to 46 AD. The offer includes a promise to register contracts in a pasted together roll (τόμος συγκολλήσιμος), in an εἰρόμενον-register, and in an ἀναγραφή-register. This text shows that the early Roman grapheia were leased concessions, like the earlier Ptolemaic monopolies. And indeed, the accounts of money payments in the Berkeley papyri, with their monthly summaries in Greek, obey the requirements of P. Revenue Laws that lessees of tax farming contracts and monopolies balance their accounts every month. The concessionary nature of the early Roman grapheia was thus probably inherited from their Ptolemaic predecessors.
Function of the Ptolemaic Grapheia
The proposed Ptolemaic ancestry of the early Roman grapheia raises questions, however, about the function of the Ptolemaic grapheia. Early Roman grapheia served as notary offices and to some extent as registration offices of both Demotic and Greek contracts. They filled a niche created by the disappearance of Demotic temple notaries and Greek six-witness contracts shortly after the Roman conquest. In the Ptolemaic period, however, Demotic contracts were usually written by temple notaries, and Greek six-witness contracts could be drawn up by anyone who could find six witnesses to authenticate them, and a syngraphophylax to preserve them. I propose, therefore, that the Ptolemaic grapheia were originally introduced primarily to register and collect the tax on Demotic and Greek contracts written outside the grapheia, and that their role as notaries was a secondary development. Evidence that Ptolemaic grapheia played a role in registration can be seen in Greek registration subscriptions on both Demotic and Greek contracts. Many of these registration subscriptions state that "it (the contract) has been entered into the ἀναγραφή-register" (τέτακται or πέπτωκεν εἰς ἀναγραφήν), and some registration subscriptions explicitly state that this registration took place at the grapheion office. Evidence that Ptolemaic grapheia played a role in taxation can be seen in the registers of contracts among the Berkeley papyri, which refer to sums of money and to the γραμματικόν. Note that Demotic contracts were also taxed by the temple notaries who wrote them, resulting in two layers of taxation.
Origin of the Ptolemaic Grapheia
Another issue raised by the proposed Ptolemaic ancestry of the early Roman grapheia concerns the date when the Ptolemaic grapheia were introduced. The term grapheion does not become common until the middle of the 2nd century BC, although a registration requirement for Greek contracts existed already from the reign of Ptolemy II onwards. A Greek legal manual for Alexandria dated to the 3rd century BC (P.Hal.) states that a sales tax of five percent of the purchase price must be paid to a steward (ταμίας) whenever land, houses or other immovables were sold, and that the steward must register the names of buyer and seller and the date on which the sales tax was paid. Outside of Alexandria, a fragmentary Greek papyrus dealing with tax laws from the reign of Ptolemy II states that sales and other transfers of slaves were required to be registered at the agoranomion or "market regulation office," and a tax on the sale or transfer that had to be paid at the royal bank. Registers of Greek contracts provide further evidence for a registration requirement in the 3rd century BC. Registers with shorter entries, and registers with longer entries or abstracts are both attested. A registration requirement probably also existed for Demotic contracts from the reign of Ptolemy II onwards. Some Demotic contracts bear Greek registration subscriptions already from his reign, and a few of these also name tax officials, implying a relationship between registration and taxation. The absence of Greek subscriptions from some Demotic contracts may simply indicate that the subscriptions were optional. The registration itself was probably mandatory, however, given the association with taxation, whether or not the contracts received subscriptions. A single register of Demotic contracts with shorter entries provides further evidence for a registration requirement in the 3rd century BC.
Given the similarities between the functions of the early Ptolemaic agoranomia and the late Ptolemaic grapheia, I am inclined to follow the suggestion of Bärbel Kramer, that the agoranomion and the grapheion were different names or different aspects of the same offices. The appearance of the term grapheion in the middle of the 2nd century BC may be associated with a reform of the registration procedures for Demotic contracts. In 145 BC, a letter to a Greek official in Thebes describes in considerable detail the procedure for registering Demotic contracts written by temple notaries.  The letter says that the procedure is first to make an abstract of the Demotic contract, then to enter the information into the register of contracts, and finally to make a registration subscription on the Demotic contract, stating that it has been entered into the register of contracts. The letter stresses the importance of knowing the correct procedure, which suggests that the procedure had recently been revised. The revisions were probably that the Greek registration subscription was now made mandatory, and that a separate abstract was now necessary. The introduction of the abstract brought the registration procedures for Demotic contracts much closer to those for Greek contracts. This revised procedure is the one seen in the posited Berkeley grapheion archive. The appearance of the term grapheion may also be associated with the growing use of Greek "agoranomic" contracts, in which the agoranomos served as notary. This secondary notary function of the agoranomion may have given rise to the new name grapheion or "writing office," and may have enabled the late Ptolemaic grapheion to become a unified notary and registry in the early Roman period, when the traditional Greek six-witness contracts and Demotic temple notary contracts disappeared.
In conclusion, the agoranomion was probably introduced under Ptolemy II, to create a unified registration for Demotic temple notary contracts and Greek six-witness contracts. Ptolemy II and his fiscal advisors knew that to tax something properly, one first had to register it, whether it be people, real property, or sales. In the course of the Ptolemaic period, the agoranomion also began to serve as a notary itself, particularly for Greek contracts. Perhaps for this reason, it became known as the grapheion. Finally, in the early Roman period, as Demotic temple notaries and Greek six-witness contracts were eliminated, and the grapheia became a unified notary and registry, for both Greek and bilingual contracts.
I wish to thank Todd Hickey of the University of California, Berkeley, Center for the Tebtunis Papyri, for permission to publish these papyri, and for assistance with digital images. I would also like to thank Arthur Verhoogt for calling my attention to these papyri, Brigit Flannery for sharing her research on them with me, Cisca Hoogendijk for helping me with the Greek texts, and the participants of the Michigan Papyrological Congress for their many helpful comments. Finally, I would like to thank the Gratama Stichting for its generous financial support of my research.
Cleopatra III and Ptolemy IX are only attested for years 1–11 (117–106 BC), Cleopatra III and Ptolemy X use double year dates (107–100 BC), Ptolemy X alone is only attested for years 14–27 (101–87 BC), Ptolemy IX alone is only attested for years 29–37 (88–80 BC), and Cleopatra VII is probably too late (51–30 BC).
R.A. Parker reads line 2 as "which ... made for me" (r.ir n=y ...), but the parallels suggest a reading "which ... made it" (r.ir s ...). Palaeographically both are possible, but the latter seems to make better sense.
Published as P.Mich. II 121–128; V 226–356; X 586–587; XI 621; XII 632–634; PSI VIII 901–918; SB VI 9109–9110; XVI 12539; XX 14313–14315; P.Bingen 59. See E. Husselman, "Procedures of the Record Office of Tebtunis in the First Century AD," in D.H. Samuel (ed.), Proceedings of the Twelfth International Congress of Papyrology. Am.Stud.Pap. VII (Toronto 1970) 223–238.
B.P. Muhs, "The Grapheion and the Disappearance of Demotic Contracts in Early Roman Tebtynis and Soknopaiou Nesos," in S. Lippert and M. Schentuleit (eds.), Tebtynis und Soknopaiou Nesos – Leben im römerzeitlichen Fajum. Akten des Internationalen Symposions vom 11. bis 13. Dezember 2003 in Sommerhausen bei Würzburg (Wiesbaden 2005) 93–104.
Cf. UPZ II 175b.4, from Thebes, dated to 145 BC; P.Tor.Choach. 13.10, from Thebes, dated to 126 BC; P.Ran.Cent. 4, from Soknopaiou Nesos, dated to 110 BC; P.IFAO II 1 (= SB XII 10843.4) dated to 105 BC; SB VI 9612.9, from Theogonis, dated to 88/87 BC; UPZ I 118.9, from Memphis, dated to 83 BC; UPZ I 135, from Memphis, dated to 78 BC; UPZ I 136, from Memphis, dated to 74 BC; P.Ashm. D 14 + 15, from Hawara dated to 72/71 BC; UPZ I 139, from Memphis, dated to 64 BC; UPZ I 140, from Memphis, dated to 64 BC; and UPZ I 141, dated to 63 BC.
Cf. UPZ II 175b.4: ὁ πρὸς τῶι γραφίωι τοῦ Περὶ Θήβας μετείληφα εἰς ἀναγραφήν, "I, the one in charge of the grapheion of the district around Thebes, have accepted (it) into the ἀναγραφή-register"; P.Tor.Choach. 13.10: ὁ πρὸς τῶι γραφίωι κεχρη(μάτικα), "I, the one in charge of the grapheion, have registered (it)"; P.Rain.Cent. 4: ἐνετάγηι [ὑπὸ] PN1 καὶ διὰ PN2 τοῦ πρὸς τῶι γρ(αφείωι), "It has been registered by PN1 (= the Demotic notary) and by PN2 the one in charge of the grapheion office"; SB VI 9612.9: ἐπιτετάχαμεν τῶι συναλλαγματογράφωι καὶ τοῖς μάρτυσι καὶ τῶι πρὸς τῶι γραφίωι γράφειν, "We have ordered the contract scribe and the witnesses and the one in charge of the grapheion to write"; UPZ I 118.9: κατὰ συγγραφὴν τροφῖτιν τὴν ἀναγραφεῖσαν διὰ τοῦ γραφίου, "according to the annuity contract registered through the grapheion"; P.Ashm. D 14 + 15: ἀναγέγρ(απται) διὰ τοῦ ἐν Πτ(ολεμαίδι) Εὐ(εργετίδι) γρα(φείου), "It has been registered through the grapheion office in Ptolemais Euergetis." This same contract also says: ἀναγέγρ(απται) ἐν τῆι β(ιβλιο)θή(κηι), "It has been registered in the bibliotheke."
S.P. Vleeming, "The Tithe of Scribes (and) Representatives," in J.H. Johnson (ed.), Life in a Multi-Cultural Society: Egypt from Cambyses to Constantine and Beyond. SAOC 51 (Chicago 1991) 343–350; and M. Depauw, The Archive of Teos and Thabis from Early Ptolemaic Thebes. Monographies Reine Élisabeth 8 (Turnhout 2000) 56–63.
P.Tebt. III.2 969 descr., from Tebtunis, dated to 235 BC, register of Greek sales contracts with enkyklion-tax; CPR XVIII 1–34, from Theogonis, dated to 231–232 BC, register with longer entries or abstracts of Greek contracts with syngraphophylax; P.Tebt. III.1 815, from Tebtunis, dated to 228–221 BC, register with shorter entries of Greek contracts with syngraphophylax.
See N.J. Reich, "The Greek Deposit-Notes of the Record Office on the Demotic Contracts of the Papyrus-Archive in the University Museum," Mizraim 9 (1938) 19–32; and B.P. Muhs, Tax Receipts, Taxpayers and Taxes in Early Ptolemaic Thebes. Oriental Institute Publications 126 (Chicago 2005) 19–20.
P. Cairo JE 89367 (Phil. dem. 14), from Thebes, dated to 263 BC: τελώνης, "tax-farmer"; P. BM 10829 (Andrews 18), from Thebes, dated to 209 BC: ir=w pꜣ gy ḏrꜥ md pꜣ sẖ nty ḥry, sẖ PN1 pꜣ rt PN2 nty ḥr pꜣ ꜣggryn n pꜣ tš n Pr-Ḥwt-ḥr, "the registration was made for the aforementioned document, has written PN1 the agent of PN2 who is in charge of the enkyklion of the Pathyrite nome."
P.W. Pestman, "Registration of Demotic Contracts in Egypt, P. Par. 65; 2nd Cent. BC," in J.A. Ankum, J.E. Spruit, and F.B.J. Wubbe (eds.), Satura Roberto Feenstra, sexagesimum quintum annum aetatis complenti ab alumnis collegis amicis oblata (Fribourg 1985) 17–25.