The CDLI has moved to a new server

Dear user, the CDLI is now located at https://cdli.earth. Please update your bookmarks and use this new domain name in print.

NCBS 135: a re-cut seal

CDLN 2024:3

Cuneiform Digital Library Notes (ISSN: 1546-6566)

Published on 2024-12-31

© The Authors

Creative Commons License This work is licensed under a Creative Commons Attribution 4.0 International License except when noted otherwise and in the case of artifact images which follow the CDLI terms of use.

Jacob L. Dahl ORCID logo

jacob.dahl@ames.ox.ac.uk

University of Oxford

Agnete Lassen ORCID logo

agnete.lassen@yale.edu

Yale University

In 1934 H.H. von der Osten published E.T. Newell’s collection of seals, today part of the Yale Babylonian Collection. In this brief study we discuss one of the seals originally published by Osten as number 135, NCBS 135 (P498000). All of the seals in Newell’s collection were acquired on the antiquities market, with notable contributions from the former collections of W.H. Ward and O.N. Rood (Osten 1934, vii). Osten (1934) did not indicate clearly the date of the seal, but in Buchanan (1981, 449) NCBS 135 (P498000) is included among the seals dated to the Ur III period (as number 596) although it was not shown there. NCBS 135 (P498000) is a rock crystal seal with an introduction scene and a two-column inscription. The seal was re-cut at least once in antiquity and the inscription likely relates at least one instance of the seal to the periphery of the Ur III state.

Figure 1: Impression in polymer clay of NCBS 135. Photography by K. Wagensonner.
Figure 1: Impression in polymer clay of NCBS 135. Photography by K. Wagensonner

The inscription of NCBS 135 (P498000) can be reconstructed as follows:[1]

dšul-gi / da-num2 / lugal ⸢uri5⸣ki // nu-ur2-i3-li2 / ⸢aga3-us2⸣ / dumu SI.A-a / šagina

Šulgi, strong one, king of Ur. Nur-Ili, soldier, child of SI.A-a, general.

Osten (1934) did not include a transliteration and translation of the seal and the name of the owner of the seal was mistyped as nu-ur2-i-li in W. Hallo, in Buchanan (1981, 449). Hallo did not read the title of Nur-Ili, here read aga3-us2.[2] The name Nur-Ili is attested in more than 500 Ur III texts, including in some instances with the title aga3-us2, aga3-us2 gal, or aga3-us2 lugal (soldier, chief soldier, or soldier of the king). The distribution of the name Nur-Ili suggests that at least two, and perhaps four, different people by that name were employed in the military and/or courier service of the Ur III state:[3]

  • A “solider” (or “chief soldier”) (lu2 geštukul(-gu-la)) is attested in eight texts from Girsu.
  • A “runner” (lu2 kas4) is attested in 14 texts from Girsu.
  • An “envoy” (sukkal) attested in 14 texts from Girsu, Nippur and Irisagrig.
  • A “royal messenger” (lu2 kin-gi4-a lugal) in 42 texts primarily from Irisagrig dating to AS 7 to IS 2.

As such, the most prominent, identifiable, persons with the name Nur-Ili in the Ur III texts are all, in one way or another, associated with the eastern periphery of the Ur III kingdom. Although there is no evidence of an ethnic conflict between Sumerian and Akkadian speakers in the Ur III period, a slight preponderance of messengers and army staff held Akkadian names.[4]

The inscription which is partly erased is highly unusual. Firstly, any Ur III seal inscription mentioning the king, but failing to list the owner as “your slave” (ARAD2-zu) is extremely rare.[5] Seals with an inscription listing the owner as the slave of the person in the first line, are sometimes referred to as Royal Servant Seals (Tsouparopoulou 2015, 38). Apart from the so-called in-na-ba seals, which includes the line ARAD2-da-ni-ir (“to his slave (he gave)”, see most recently R.H. Mayr and Owen (2004)), and a number of seals, originally belonging to members of the royal family, where terms of family relationship replaces statements of servitude,[6] the vast majority of Ur III Royal Servant Seals included the line ARAD2-zu. Only eight seals that mention the king have been published without the information “your slave” (ARAD2-zu).[7]. Two of these warrant a closer look:[8]

  • CDLI Seals 006154 (composite) (P458864) found on a seal impression from Ur (see Legrain (1951, 31) number 407 = U. 13617  = P200950, see also < http://www.ur-online.org/subject/13920/>), with the impression published as: dšul-gi / nita kal-ga / lugal uri5ki-ma / lugal an-ub-da limmu2-ba // dnanna-ki-ag2 / dub-sar / dumu igi-an-na-ke4-zu / [u-ku-il ni?]. This term, u-ku-il ni is indexed under “Uku-íl … or nam-tùn?” on p. 4 of Legrain (1951). It is not clear what Legrain meant with Uku-íl (perhaps ug3-il2, “carrier” a lowly title, unlikely to be found at the end of this seal of a devotee of Šulgi). It is also not clear what Legrain meant with nam-tun3. A better reconstruction of column 2 line 4 may be [ARAD2-zu].
  • CDLI Seals 000170 (composite) (P430269) rolled on PDT 1, 516 (P125932), a tablet from Drehem today in Istanbul, for which no image is available and where the impression is published as: dšul-gi / nita kal-ga / lugal uri5ki-ma // lugal an-ub-da limmu2-ba / lu2-dnanna / dub-sar / dumu […]. Gomi (1988, 117) suggested to read the final line as dumu-ni 2-zu> comparing the seal to Legrain (1951, 32) seal impression number 422 (U. 13612 = P466612 = P200960, the object itself has “crumbled away” see < http://www.ur-online.org/subject/13914/>), another seal of Lu-Nanna, which does include the line ARAD2-zu. However, Tsouparopoulou (2015) number 160, did not include a transliteration beyond the final line dumu […], but indicated in her commentary that the seal likely had more than the seven preserved lines. In addition, Tsouparopoulou classified the seal as a Royal Servant Seal, which implicitly in her catalog indicates the presence ARAD2-zu at the end of the seal inscription. The next seal in her catalog (n 161) has 8 lines, the last reading ARAD2-zu.

Secondly, the Akkadian language title given to Šulgi in Newell 135 (P498000), da-num2, is found in very few Ur III texts as a title for the king. In Ur III Sumerian royal inscriptions da-num2 is an obvious Akkadian calc on the Sumerian title lugal or nita kal-ga, “strong king or man”. However, the title lugal da-num2 is only found in Ešnunna and in one fragmentary seal from Irigsagrig (see below). Alone, da-num2 is simply translated as “(the) strong (one)”. Most of the examples of the use of Akkadian da-num2 as a title of the Ur III king are found in the periphery of the Ur III kingdom, in areas where Akkadian was presumably the main language or the best-known Mesopotamian language. The use of LUM for num2 writing /danum/ suggests a continued preference for the Old Akkadian syllabic values (Gelb 1952, 115). We list here all the attestations of Akkadian da-num2 in Ur III royal texts, first for Šulgi:

  • RIME 3/2.01.02.023 composite (P432188): with one witness RIME 3/2.01.02.023, ex. 01 (P226977) (Ur III, Nineveh?) (lines 1-2): šul-gi / da-num2
  • RIME 3/2.01.02.027 composite (P432191), with one witness: RIME 3/2.01.02.027, ex. 01 (P226511) (Ur III, Eshnunna), (lines 1-2): šul-gi / da-num2
  • RIME 3/2.01.02.029 composite (P432193), with one witness: RIME 3/2.01.02.029, ex. 01 (P226602) (Ur III, Tell ed-Dēr) (lines 4-5): šul-gi / da-num2
  • RIME 3/2.01.02.033 composite (P432197): with 15 witnesses (Ur III, Kani Joni near Kuhdasht in Luristan, Iran) (lines 1-3): dšul-gi / dingir ma-ti-šu / da-num2
  • RIME 3/2.01.02.065 composite (P432224): with one witness RIME 3/2.01.02.065, ex. 01 (P227094) (a possible modern fake) (lines 3-4): [d?]šul-gi / [da-num2]

The ruler of Mari, Apil-kin, a contemporary of Ur-Nammu of Ur, also used the title da-num2:

  • RIME 3/2.04.01.01 composite (P432366) with two witnesses (P227498 and P227499) (lines 1-2): a-pil3-gi / da-num2

Ibbi-Suen is given the title da-num2 in one or two seals from Irisagrig:

  • CDLI Seals 014143 (composite) (P519510) rolled on the Akkadian language text CUSAS 40, 0247 (P515031) from Irisagrig: di-bi2-dsuen / da-num2 / lugal / uri5ki-ma // be-li2-kal / dumu ⸢zi⸣-[zi] / ARAD2-zu
  • A fragmentary seal impression is found on CUSAS 40, 1170 (P515954) a tablet from Irisagrig, giving the title LUGAL da-num2 for Ibbi-Suen: di-bi2-dsuen / lugal da-num2 / lugal uri5ki-ma // (illegible)

In three seals from Ešnunna dating to the Ur III period, the god Tišpak is given the title lugal da-num2:

  • RIME 3/2.03.01.001 (composite) (Pnnnnnn), rolled on As 31-630 and As 31-670: (lines 1-2): dtišpak / ⸢lugal⸣ da-num2
  • CDLI Seals 005997 (composite) = RIME 3/2.03.01.2002 (P458707), rolled on OIP 043, 144 TA 1931 663 (P200403): [9] dšu-i3-li2-a / na-ra-am dtišpak / dnin-te-ra-ba-an / dnin-šuk-nir / lugal da-num2 // lugal ma-at wa-ri-im / at-ta-a-a / x-x-x / ARAD2-zu
  • CDLI Seals 005677 (composite) = RIME 3/2.03.01.2003 (P458387), rolled on JAOS 097, 177 8a (P200394): dšu-i3-[li2-a] / ⸢dumu⸣ dtišpak [lugal da-num2] / ⸢lugal⸣ ma-at [wa-ri-im] / ⸢lu2⸣ KA-[...] / ⸢dumu⸣ ARAD2-mu ⸢ARAD2⸣-[zu]

Finally, a Neo-Babylonian copy of an Akkadian language Ur III text also has the title da-num2 for Šulgi.

  • RIME 3/2.01.02.025 composite (P432190) with one witness (BM 078681 + BM 139969): (lines 3-4) [d?šul]-gi / [da]-num2

The usage of the Akkadian language title da-num2 for Šulgi in NCBS 135 (P498000) and the lack of the servitude statement (ARAD2-zu) makes this a very rare inscription among seals attested from impressions on dated Ur III tablets.

Figure 2: Digital unwrapping of the seal’s surface and photos of the seal from multiple angles. Photography by K. Wagensonner.
Figure 2: Digital unwrapping of the seal’s surface and photos of the seal from multiple angles. Photography by K. Wagensonner

The seal bears strong evidence of being recarved at least once in antiquity. The area with the inscription is hollowed out and the signs are much shallower than the imagery, suggesting that an inscription was erased at least once and possibly multiple times. The final iteration of the inscription was either unsuccessfully erased (it is still legible although quite vague) or just engraved very lightly. The imagery has problems as well that suggest it too was recarved or that the seal carver did a poor job of planning and executing the designs. This is most clearly visible in the interceding goddess, who has two sets of arms: one set clasped on the chest and another where one hand grasps the hand of the worshipper and the other extended in front of her towards the seated deity. Additionally, one may note that the worshipper is more crudely cut than the other two figures, particularly the face and hand. The worshipper is also placed lower in the image field than the interceding goddess and the seated deity. A possible recarving history of the seal imagery could include a first step in which a figure with clasped hands stands in front of a seated god. In the second step, the figure was changed into an interceding goddess and a third figure, a worshipper, was added behind the interceding goddess. It is unclear if the erasure and possible modification of the inscription was done along with the recarvings of the image scene and it is not possible to date the two proposed carving steps. Both image scenes are attested in the Ur III and early Old Babylonian periods.

As pointed out already by Delaporte (1910, 70-72), worshippers on Ur III presentation scene seals found impressed on tablets always have a bald head. Changing the hairdo of a figure or adding a hat is, however, not very difficult and many Ur III seals may have been re-cut in the Old Babylonian period, slightly altering the worshiper, and erasing the original inscription (see already Feingold (2014) and Dahl (2024)). The head of the worshiper on NCBS 135 (P498000) has a poorly cut hat similar to those seen on Old Babylonian seals. Additionally, the crescent moon and the rosette is unlike those found on most Ur III seals attested impressed on tablets (see R. Mayr (n.d.) and Fischer (1992) and (1997) and Tsouparopoulou (2015) for corpora of seals impressed on dated Ur III period tablets).

Figure 3: Drawing of the seal’s imagery. Drawing by A. Lassen.
Figure 3: Drawing of the seal’s imagery. Drawing by A. Lassen

We must therefore conclude that we only know parts of the story of this fascinating object, which at one time, perhaps immediately before it disappeared from circulation, was in the possession of a person, Nur-Ili, who was loosely connected to the Ur III state, and likely a member of a local elite on the eastern fringes of the Ur III state. How he got hold of the seal, and whether he was the one who had it recut is unclear (see for example the seal of Bilalama, Reichel (2003)). It is also unclear whether the seal belonged to someone after Nur-Ili who erased, or attempted to erase the name and titles of Nur-Ili, and who altered the scene. In that case, this final owner did not leave his name inscribed on the object.

Notes

[1] The transliteration of the inscription of cylinder seals is given with / separating lines and // separating columns. Given the brevity of the inscriptions no line numbering is provided. Transliterations follow the CDLI, except that we indicate Akkadian language switch with italics.

[2] Thanks are due to Klaus Wagensonner for inspecting the original in YBC and providing photos of the impression and the seal.

[3] Since a majority of the texts are ‘messenger texts’ which are rarely dated to the year, we do not here attempt to give a range of operation for the use of the different titles, except for ‘royal messenger’ for which there is sufficient attestations.

[4] A Nur-Ili is attested as the brother of the lukur in OIM A00824 (P209895), possibly highlighting this Nur-Ili’s tenuous links to the elite (Dahl 2007, 147-155).

[5] See R. Mayr (n.d.) p 103-104 for the suggestion that ARAD2-zu should be read ARAD2.ZU and interpreted as a direct ‘logographic’ borrowing from Old Akkadian seals where this was the standard writing for “his slave” (Akkadian warassu).

[6] dumu-ni, his child, see e.g. P458624 or P430485; nin9-a-ni, his sister, see e.g. P465344; lukur-a-ni, his concubine, see e.g. P458401. A number of other seals are dedicated to deities on behalf of the life of the king and the syntax does not allow for the inclusion of the usual statement of servitude, see e.g., P226946, P226633, P226811, P226975.

[7] In two of these cases ARAD2-zu can easily be reconstructed at the end of the inscription (CDLI Seals 004265 (composite) (P456975) rolled on RTC 417 (P128570); and CDLI Seals 005704 (composite) (P458414) rolled on MVN 03, 271 (P113831)). Another four from ancient Irisagrig, were published without sufficient photographic evidence to prove whether a final line indicating servitude was present (CDLI Seals 006092 (composite) (P458802) rolled on Nisaba 15, 604 (P453949); CDLI Seals 005912 (composite) (P458622) rolled on Nisaba 15, 0054 (P387870) and Nisaba 15, 0064 (P412151); CDLI Seals 006087 (composite) (P458797) rolled on Nisaba 15, 0983 (P412138); and CDLI Seals 006089 (composite) (P458799) rolled on Nisaba 15, 0372 (P453802)).

[8] Occasionally, publishers or editors of texts have assumed, with or without comments, the presence of ARAD2-zu at the end of a seal inscription, those choices have been accepted here (see for example the seal rolled on ITT 3, 5395 (P111179) edited by Lafont (1994), who added a final line reading [ARAD2-zu], and the seal rolled on UTI 3, 2232 (P140251) where R. Mayr (n.d., 218) number 251 added [ARAD2-zu] at the end of the inscription.

[9] The syntax of the inscription is complex and possibly emended and the relationship between the title and the various persons mentioned in the seal unclear, see also Reichel (2003) for Ešnunna royal seals.

Bibliography

  1. Buchanan, Briggs. 1981. Early Near Eastern seals in the Yale Babylonian Collection. Edited by William W. Hallo and Ulla Kasten. New Haven: Yale University Press, 1981.
  2. Dahl, Jacob L. 2024. “Where Have All the Ur III Seals Gone?” Avar: An Interdisciplinary Journal of Life and Society in the Ancient Near East 3 (2): 195–252.
  3. Dahl, Jacob L. 2007. The ruling family of Ur III Umma : a prosopographical analysis of an elite family in Southern Iraq 4000 years ago. Vol. 108 Uitgaven van het Nederlands Instituut voor het Nabije Oosten te Leiden. Leiden: Nederlands Instituut voor het Nabije Oosten.
  4. Delaporte, Louis. 1910. Catalogue sommaire des manuscrits coptes de la Bibliothèque nationale de Paris. Paris: August Picard.
  5. Feingold, Rony. 2014. Engraved on stone : Mesopotamian cylinder seals and seal inscriptions in the old Babylonian period. Vol. 7 Gorgias studies in the ancient Near East. Piscataway, NJ: Gorgias Press.
  6. Fischer, Claudia. 1992. "Siegelabrollungen im British Museum auf neusumerischen Tontafeln aus der Provinz Lagaš." Zeitschrift für Assyriologie und Vorderasiatisches Archäologie 82 (1): 60-91. https://doi.org/doi:10.1515/zava.1992.82.1.60.
  7. ---. 1997. "Siegelabrollungen im British Museum auf Ur-III-Zeitlichen Texten aus der Provinz Lagas." Baghdader Mitteilungen 28: 97-184.
  8. Gelb, I. J. 1952. Old Akkadian writing and grammar. Chicago, Ill.
  9. Gomi, Tohru. 1988. "Neue Kollationen zu den Texten PDT, I." Orient, Report of the Society for Near Eastern Studies in Japan 24: 108-123.
  10. Hilgert, Markus. 2003. Cuneiform texts from Ur III period in the Oriental Institute. Oriental Institute publications 121. Chicago: The Oriental Institute of the University of Chicago.
  11. Lafont, Bertrand. 1994. "L’avènement de Shu-Sîn." Revue d’Assyriologie et d’Archaeologie Orientale 88, no. 2: 97-119.
  12. Legrain, Leon. 1951. Seal cylinders. Publications of the Joint Expedition of the British Museum and of the University Museum, University of Pennsylvania, Philadelphia, to Mesopotamia : Ur excavations ; v.10. London: Published for the Trustees of the two museums.
  13. Mayr, Rudi. n.d. Seal Impressions on Tablets from Umma. Edited by David I. Owen. Vol. 7. Cornell University Studies in Assyriology and Sumerology. Pennsylvania: Penn State University Press.
  14. Mayr, Rudolf H., and David I. Owen. 2004. "The royal gift seal in the Ur III period." In Von Sumer nach Ebla und zurück. Festschrift. Giovanni Pettinato zum 2 7. September 1999 gewidmet von Freunden, Kollegen und Schülern, edited by Hartmut Waetzoldt, In Heidelberger Studien zum alten Orient, 145-174. Heidelberg: Heidelberger Orientverlag.
  15. Osten, Hans Henning von der. 1934. Ancient oriental seals in the collection of Mr. Edward T. Newell. University of Chicago Oriental Institute publications; 22. Chicago: University of Chicago Press.
  16. Reichel, Clemens. 2003. "A Modern Crime and a Ancient Mystery: the seal of Bilalama." In Festschrift für Burkhart Kienast zu seinem 70. Geburtstage dargebracht von Freunden, Schülern und Kollegen., edited by Gebhard Selz, In AOAT, 355-389.
  17. Tsouparopoulou, Christina. 2015. The Ur III seals impressed on documents from Puzriš-Dagān (Drehem). Vol. Band 16 Heidelberger Studien zum alten Orient. Heidelberg: Heidelberger Orientverlag.
Cite this Article
BibTeX RIS
Dahl, Jacob L., and Agnete Lassen. 2024. “NCBS 135: A Re-Cut Seal.” Cuneiform Digital Library Notes 2024 (3). https://cdli.earth/articles/cdln/2024-3.
Dahl, J. L., & Lassen, A. (2024). NCBS 135: a re-cut seal. Cuneiform Digital Library Notes, 2024(3). https://cdli.earth/articles/cdln/2024-3
Dahl, J. L. and Lassen, A. (2024) “NCBS 135: a re-cut seal,” Cuneiform Digital Library Notes, 2024(3). Available at: https://cdli.earth/articles/cdln/2024-3 (Accessed: April 16, 2025).
@article{Dahl2024NCBS,
	note = {[Online; accessed 2025-04-16]},
	address = {Oxford; Berlin; Los Angeles},
	author = {Dahl, Jacob L. and Lassen, Agnete},
	journal = {Cuneiform Digital Library Notes},
	issn = {1546-6566},
	number = {3},
	year = {2024},
	publisher = {Cuneiform Digital Library Initiative},
	title = {NCBS 135: a re-cut seal},
	url = {https://cdli.earth/articles/cdln/2024-3},
	volume = {2024},
}


TY  - JOUR
AU  - Dahl, Jacob L.
AU  - Lassen, Agnete
DA  - 2024///
PY  - 2024
ET  - 2024/12/31/
ID  - cdln-2024-3
IS  - 3
J2  - CDLN
SN  - 1546-6566
T2  - Cuneiform Digital Library Notes
TI  - NCBS 135: a re-cut seal
UR  - https://cdli.earth/articles/cdln/2024-3
VL  - 2024
Y2  - 2025/4/16/
ER  -
This website uses essential cookies that are necessary for it to work properly. These cookies are enabled by default.