Cuneiform Digital Library Journal
§1. In his publication of the texts from the archive of the Ur III merchant Tūram-ilī, Marc van de Mieroop (1986, 3) presented a list of 13 Sumerian month names attested in the archive. Due to the recent and excellent treatment of several new texts from this archive by Steven Garfinkle (2000), this list may now be further completed. The new texts published by Garfinkle add the two months šeš-da-gu7 and tam2-hi-ru to the previously known month names, offering the distribution of month names in table 1.
§2. As can be seen from the table, the texts 92 and 99 are the only documents from the archive offering any direct clues concerning the order of the months in the calendar(s). The absence of more than one year formula in these two loan documents implies that the loans began and expired within one single year (i.e. ŠS 8 and ŠS 9 respectively). Text 92 demonstrates that the month kir11-si-ak (when the contract was drawn up) occurred before giš-apin (when the loan was to be repaid) while text 99 suggests, following the same logic, that the month nig2-e-ga should be placed before the repayment month giš-apin. In any case, we may be reasonably confident that the three months kir11,-si-ak, nig2-e-ga and giš-apin all belonged to the same (see below) calendar in the Tūram-ilī archive.
Table 1. The different month names attested in the archive of Tūram-ilī according to Garfinkle 2000.
§3. The large number of month names attested in the texts has led to the logical conclusion that the Tūram-ilī archive employed more than one calendrical system. It appears that a similar practice can also be observed in the private archive of the entrepreneurial shepherd SI.A-a as well as in the small village of Išān Mizyad situated approximately 4 kilometers north of ancient Kiš. All tablets of Tūram-ilī are supposed to belong to a single archive in the, still unidentified, city or center where Tūram-ilī was active. The alleged employment of different calendars by Tūram-ilī’s scribe(s) is puzzling and appears to be highly impractical from an administrative point of view. One explanation could be that Tūram-ilī’s scribe(s) employed foreign calendars depending on the origins of the different clients in the documents and/or in which cities the contracts were drawn up.
§4. A problem with this explanation is that it does not make sense that the private scribes and archivists of Tūram-ilī should be interested in adjusting their administration to the various calendars of Tūram-ilī’s clients. Tūram-ilī’s business interests have been believed to reach out and cover all of Babylonia and would thus include a number of different calendars. In some cases (i.e. the texts with ezen-dŠul-gi or še-KIN-ku5), identical month names refer to entirely different months depending on the local calendar from whence they derive. To administer an archive with texts dated according to all these calendars would be most difficult, if not completely impossible.
§5. Moreover, the texts in the archive itself speak against this explanation: in three texts from IS 2 and 3 (Garfinkle 2000, 116, 123, 129), the merchant Nūr-Adad (identified by his seal in all the cases) is receiving silver from Tūram-ilī. If Tūram-ilī adjusted his calendar to his clients, it is difficult to explain why the first text has been dated with a month specific for the Ur/Puzriš-Dagan calendar (ezen-mah), the second with a month name only attested in Lagaš (ezen-dLi9-si4), while the third text uses a month that is only found in the texts from the archives of SI.A-a and Tūram-ilī (gi-sig-ga). In fact, these three texts strongly suggest that the months ezen-mah, ezen-dLi9-si4 and gi-sig-ga belonged to one (hitherto unknown) calendar used in the Tūram-ilī archive. A similar connection can be found between the Ur/Puzriš-Dagan month ezen-an-na and the Nippur month giš-apin in two texts from IS 1 and 2 where Tūram-ilī is delivering building materials to NE.NE (Garfinkle 2000, 105 and 114).
§6. Finally, in the few cases when the locations of the transactions are specified in the texts, the month names used do not fit the local calendars of these places: In Garfinkle 2000, 95, a certain Ilī-arani receives peas from Tūram-ilī in E-sagdanaki, which has been thought to refer to the toponym in Nippur or Lagaš. However, the month name in the text (ezen-an-na) is specific for the calendar used in the provinces of Puzriš-Dagan and Ur. On the other hand, in Garfinkle 2000, 85, Šu-Mama is receiving silver from Tūram-ilī in the city of Ur, but the contract is dated with a month only used in the city of Lagaš (ezen-dLi9-si4). Both nos. 109 and 128 of Garfinkle’s texts suggest that Tūram-ilī was active in Uruk but only the latter text is dated by month. The text contains the month name ezen-mah, which is the most common month in the Tūram-ilī archive(s).
Table 2. The month names in Garfinkle’s texts and their occurrences in other cities or archives.
The Ur III calendar of Uruk remains somewhat uncertain but if Mark Cohen’s reconstruction is correct (1993, 208-210), the month ezen-mah was not used in this city. In fact, if we do not count the Tūram-ilī texts, the month ezen-mah is (again) only found in the Ur/Puzriš-Dagan calendar.
§7. The attestations in other calendars of the month names (or the festivals on which the months were based) in the Tūram-ilī calendar(s) can be demonstrated as shown in table 2 above.
§8. There are alternative explanations for the apparent use of more than one calendrical system in the archive of Tūram-ilī.
§8.a.1. We have to consider the fact that the personal name Tūram-ilī was rather common in the Ur III period. In fact, we find officials from all over the state from a variety of professions called Tūram-ilī and many texts (and therefore also month names) in the “Tūram-ilī archive” may in reality derive from archives belonging to different people. Our Tūram-ilī can be identified as a merchant involved in the loan business of (mainly) silver during the final stages of the Ur III state (see note 28) in northern Babylonia. As an example of a text with a different merchant called Tūram-ilī, one can mention AUCT 1, 757 from Šulgi 40 dated with the tenth month in the Puzriš-Dagan calendar (eleventh in Ur):
§8.a.2. The text shows that another merchant called Tūram-ilī was active in Uruk, at least during the end of the reign of Šulgi. This, in turn, raises the question if the above-mentioned texts Garfinkle 2000, 109 and 128, where a certain Tūram-ilī is delivering barley connected to the bala in Uruk, could rather be referring to this Uruk based individual. In the former text from IS 2, our Tūram-ilī and two of his colleagues (all identified through their seals) are receiving the barley balance of the bala in Uruk (si-i3-tum še bala a3 Unugki-ga) from this unidentified Tūram-ilī. It appears highly plausible that this Tūram-ilī refers to the same individual as we find in text 128 where Šu-Ninšubur receives barley on account of the bala (mu bala-a-še3) in the same city. The main problem with the identification of the merchant in AUCT 1, 757 and this official in Uruk is that it would require that he remained in business for at least 28 years. Therefore, the merchant in AUCT 1, 757 may perhaps rather be referring to an earlier predecessor stationed in the city.
§8.a.3. The second highly interesting aspect of AUCT 1, 757 is that the transaction in the text takes place in the E-sagdana of Nippur. Not only does this imply that the E-sagdanaki in the already mentioned Garfinkle 2000, 95 refers to the toponym in Nippur, but it also shows that Uruk had business interests in this institution. This should perhaps be taken as an indication that Tūram-ilī in the rather uncharacteristic text Garfinkle 2000, 95 may also have been an Uruk merchant. With all these different individuals named Tūram-ilī in mind, it may appear somewhat injudicious to add to the calendar the month šeš-da-gu7 (otherwise only found in Ur/Puzriš-Dagan) from one single text (Garfinkle 2000, 89) that, apart from the appearance of a Tūram-ilī who delivers silver, shows no prosoprographical or structural connection to other texts from the archive.
§8.b.1. Even if we can securely identify Tūram-ilī in the texts, we cannot presuppose that all texts mentioning a certain individual were archived in one single archive belonging to that individual. This may appear to be an obvious remark but the fact is that when it comes to the private archive(s) of Tūram-ilī (and indeed also of SI.A-a), the main emphasis has been placed on the presence in the text of the personal names rather than the context in which the names occur. However, texts are archived by the party that has acquired a specific right. Only when one or several individuals (usually identified by his/their seal(s)) is/are receiving (šu ba-ti) silver (or, in a few cases, some other products) from Tūram-ilī (ki Tūram-ilī-ta) do we have any reason to assume that the texts were stored in Tūram-ilī’s loan-archive. Texts where Tūram-ilī is supplying products in other contexts (sale contracts, bala deliveries, etc.) may certainly also have been archived by the administrators of Tūram-ilī, but it is by no means impossible that these transactions were filed separately from his loan contracts.
§8.b.2. Thus, while Tūram-ilī may have used multiple calendrical systems, we cannot presuppose that these systems were used together in one single archive. More importantly, if Tūram-ilī is merely mentioned (e.g. Garfinkle 2000, 133) or appears as the receiver, thus being the debtor rather than the creditor (e.g. Garfinkle 2000, 77, 109, 113, 124, 131), the texts were certainly not archived by Tūram-ilī’s organization. Obviously, this has to be taken into account when we try to reconstruct the calendar used in Tūram-ilī’s loan-archive. For example, in Garfinkle 2000, 131, our Tūram-ilī (identified by his seal) is borrowing flour from another official named Tūram-ilī. Since the texts are dated with the Adab month šu-gar-ra it seems rather likely that this flour-lending Tūram-ilī came from this city. This makes it highly plausible that the similar flour-loan from the same year and month (Garfinkle 2000, 130) also should be attributed to this Adab official. This would mean that the month šu-gar-ra is not attested in the Tūram-ilī archive, which in turn rather significantly would attenuate the Adab connection to the archive.
§8.c. The many month names found in Tūram-ilī’s texts may be the result of a sudden change or modification of the calendar. There are many examples in the Ur III period of calendars that for various reasons were altered. Thus, while Tūram-ilī possibly made use of more than one calendrical system, we cannot presuppose that the different systems were used at the same time. According to the texts collected by Garfinkle, Tūram-ilī’s texts span altogether 18 years including the reigns of the last three kings of the Ur III state. During this politically dynamic period, the Ur III state went from being the major political, cultural and military power in Mesopotamia to an unimportant petty state in southern Babylonia. Already from the very beginning of the reign of Ibbi-Suen, we have to assume that the power in northern Babylonia shifted from the south to the major Amorite settlements in the northern and central parts of Babylonia. One possible example of a new element in the calendar that may be ascribed to this political development is the archive’s only Semitic month tamhiru attested in one single text dated to IS 1 (Garfinkle 2000, 102).
§9. There can be no doubt that the use of more than one calendar within one single archive at the same time must have been not only highly impractical but also perfectly pointless. Nevertheless, previous studies of the Tūram-ilī archive have produced more month names than one calendar could possibly need, and it has therefore been assumed that the archive employed multiple calendars. However, earlier discussions of the Tūram-ilī archive display some methodological problems. First of all, scholars have been too impetuous in identifying any individual called Tūram-ilī with our particular man. Once Tūram-ilī has been identified, it has been assumed that every time he is mentioned in a text—regardless of the context—that particular text was filed in the Tūram-ilī archive. Moreover, two further aspects need to be considered before any attempts of reconstructing the calendar can be undertaken: 1) Tūram-ilī could (and probably did) keep more than one archive in his organization; 2) one or several month names in the calendar may have been replaced by other names during the time span of the archive.
§10. In section 8(a-c) I have shown that if these aspects are accounted for, the number of different months attested in Tūram-ilī’s loan-archive should be reduced. The single text with the Ur/Puzriš-Dagan month šeš-da-gu7 (Garfinkle 2000, 89) as well as the two texts with the Adab month šu-gar-ra (nos. 130 and 131) were not archived by our Tūram-ilī and should be attributed to the archives of other officials called Tūram-ilī. As for the single attestation from IS 1 of the Old Babylonian month tamhiru (Garfinkle 2000, 102), it seems plausible that this month did not represent a regular element in our calendar. The occurrence of this Semitic month in the otherwise Sumerian calendar of Tūram-ilī may perhaps be explained by the increased influence of the Amorites in the region during the reign of Ibbi-Suen.
§11. Thus, we end up with one single Sumerian calendar, or rather twelve different month names, used by Tūram-ilī’s administration. While the sequence of the months remains uncertain, several of the month names can be connected to each other within one single calendar. In addition to the direct connections of the months kir11-si-ak, nig2-e-ga and giš-apin that are found in the above-mentioned texts 92 and 99 (see §2), I have tried to show (see §5 and note 10) that several other months can be connected on the basis of structural similarities of the texts and/or prosoprography (i.e. 119?- 122; 105- 114; 123- 116-129). This new calendar appears to be hitherto unique, although seven of its twelve months also appear in the texts from the SI.A-a archive(s) (table 3).
Table 3. The different month names in the archive of Tūram-ilī and their different connections to each other.
Mark Cohen, The Cultic Calendars of the Ancient Near East (Bethesda 1993)
Steven Garfinkle, Private Enterprise in Babylonia at the End of the Third Millennium BC [Ph.D. dissertation, Columbia University] (Ann Arbor 2000).
Samuel Greengus, “The Akkadian Calendar at Sippar,” Journal of the American Oriental Society 107/2 (1987) 209-229.
Antonella lo Castro, “Erra e Mama e l’archivio di SI.A-a,” Nouvelles Assyriologiques Brèves et Utilitaires 1999/63.
Nicholas Postgate, “Inscriptions from Tell Al-Wilayah,” Sumer 32 (1976) pp. 77-100
Walther Sallaberger, Der kultische Kalender der Ur III-Zeit, vol. 1 (Berlin - New York 1993).
Tonia Sharlach, Bala: Economic Exchange Between Center and Provinces in the Ur III State [Ph.D. dissertation, Harvard University] (Ann Arbor 1999).
Piotr Steinkeller, Sale Documents of the Ur-III-Period (Stuttgart 1989).
Marc van de Mieroop, “Tūram-ilī: An Ur III Merchant,” Journal of Cuneiform Studies 38/1 (1986) pp. 1-80.
Claus Wilcke, “É-sag-da-na Nibruki. An Early Administrative Center of the Ur III Empire,” in M. deJong Ellis, ed., Nippur at the Centennial. Papers Read at the 35e Rencontre Assyriologique Internationale (Philadelphia 1992) 311-324.
Zhi Yang, Sargonic Inscriptions from Adab (Changchun 1989).
Version: 20 February 2003