site, now positively identified as Urkeš, located in Syria, on the Khabur plains, near the headwaters of Wadi Dar῾a, one of the streams feeding the Khabur River, just south of the Mardin pass leading to the Anatolian plateau (37°02′ N, 41°0′ E). The site, with its commanding position, was the major gateway to the copper mines in the north and was a hub in the communication network to the south, along the Khabur, and on the east-west road that flanks the Taurus range. [See Taurus Mountains.]

The central High Mound measures 18 ha (45 acres) and stands 25 m above virgin soil. The Outer City extends for 400 m, for a total surface of 135 ha (334 acres). Tell Mozan has been identified as the site of ancient Urkeš, the mythical city of Hurrian mythology and one of the major third-millennium capitals that had so far escaped identification. It is described in later Hurrian myths as the residence of Kumarbi, the major god of the Hurrian pantheon. As such it plays a prominent role in the mythical geography of ancient Syria. The famous bronze lions of King Tish-atal of Urkeš, which had been sold on the antiquities market and are now exhibited in the Louvre and the Metropolitan Museum, are thus to be reinterpreted as having come from Tell Mozan. [See Hurrians.]

Tell Mozan was first surveyed and briefly excavated by Max Mallowan In 1934. He dug three trenches (R0, G0, H0) and noted some surface remains, identified as N0. Mallowan first assumed that the site was Roman, which was one reason he chose Chaghar Bazar for his excavations. (The most detailed report is to be found in his wife's autobiography: Agatha Christie, Come Tell Me How You Live, New York, 1977, p. 72ff.). [See the biography of Mallowan.]

Regular excavations, under the direction of Giorgio Buccellati and Marilyn Kelly-Buccellati, began In 1984; by 1995 eight seasons were completed. The long-term research strategies, favored by the fact that the site has almost no modern occupation, have taken three major directions: extensive surveys (topographic, geophysical, and artifactual) of the entire area; distinct operations on the High Mound, with a view toward obtaining an understanding of different functional units (the city wall in KW; the temple, BA; the royal storehouse, AK; a private house, F1) and different periods (the stepped trench, AS; the deep sounding, S2; the later occupations in BH and C1); and, preliminary soundings in the Outer City, indicating that it was utilized both for burials (OA4, OB1) and domestic structures (OD2, OE1).

The chronological distribution of the occupation of Mozan began in the Halaf period, on the basis of ceramic evidence found on the surface of the Outer City and from the lowest levels of deep sounding S2, just above virgin soil. [See Ḥalaf, Tell.] No early third-millennium strata have yet been excavated, but in deep sounding S2 three small pointed-base Ninevite V cups were found in stratum A12 next to a carbon sample for which the calibrated date 2920 ± 170 BCE was obtained. The surface ceramic survey suggested that early-third-millennium sherds were more prevalent on the northern portion of the mound.

The major occupational strata at Mozan and the largest component in the buildup of the high mound can be dated to the second half of the third millennium (a calibrated C-14 date of 2435 BCE ± 60 has been obtained for phase A1 of temple BA). It is remarkable to find structures from this period not only at lower elevations (city wall KW, storehouse AK), but also at middle (private house F1) and topmost elevations (temple BA). This suggests that the city's skyline in the mid-third millennium was already similar in its general outline to the profile of the tell today.

From the initial ceramic survey it appeared that second- millennium BCE occupation was centered on the highest portion of the mound. This was verified in a small sounding (Q1) and in the upper strata of the step trench AS, where traces of a large building have been partially uncovered. The latest levels above temple BA dated to the Khabur period, but no building remains were found. On a lower portion of the high mound, area C1, a single Khabur-period grave was excavated. The latest occupation dates to the Nuzi period and is characterized by private houses (AS, BH) containing very few small finds. Even the ceramics are of the coarser variety, with only portions of two painted goblets found in the excavations.

Mozan, Tell

MOZAN, TELL. Figure 1. Plan of temple BA, phase B1A. (Courtesy G. Buccellati)

view larger image

The earliest phase of temple BA is the best preserved (see figure 1). It had a large interior space (9 × 16.5 m) with a monumental entrance accessed by a long (8 m) stone ramp. The foundations were constructed of large and roughly hewn limestone blocks with mud-brick walls resting immediately on stone foundations. The plan and the exterior dimensions are similar to those of temple G at Ebla and the “Aussenbau” at Chuera (which, however, have an access in antis, as opposed to the Mozan bent-axis approach); at Ebla the walls are twice as thick, which means that the interior space is considerably smaller. [See Ebla; Chuera, Tell.] A thick, cementlike pavement covered the entire interior space; it is assumed that, in spite of its large size, this room was roofed and that the roof was pitched.

Mozan, Tell

MOZAN, TELL. Figure 2. Plan of storehouse AK, stratum B12. (Courtesy G. Buccellati)

view larger image

At the base of a stepped trench on the western side of the tell, portions of a large storehouse have been excavated. Judging from the layout of the excavated portion, this is the southwestern corner of a building that, once excavated, may be three times as large as the present exposure. The plan (see figure 2) consists of an accession suite, a large hall with a closet or vault, and probably a courtyard and an interior storage area behind. The sectors to the east appear to have a plan that is the mirror image of the western half. The lower courses of the walls are stone, up to about 1 m high, with mud brick on top, but they are not plastered. The vault in sector B is a small closet (1.8 × 1.3 m) with thick walls and an anteroom. The notching, or rabbeting, in the doorways is for the most part structural, in that it must have held a door panel—the location of the door coincides in many cases with a strongly marked threshold.

More than six hundred seal impressions, all dated to the early Akkadian period, came from a floor deposit associated with the vault in sector B (stratum B12). More than 150 of these were inscribed: these legends give the name of the king (endan in Hurrian) of Urkeš and of the queen, plus other members of the royal court. It appears therefore that this was a royal storehouse, where goods were stored in boxes, baskets, and jars on which the seals were rolled; there were only two door sealings. Since the majority of the seals belonged to the queen, the western wing of AK must have been reserved for goods belonging to her. The iconography and style of the seal impressions are, for the most part, in striking contrast with the more formal early Akkadian glyptic of the south. The seals carved in Urkeš have a unique stylistic vigor that can be traced to the production of different workshops. Distinctive aspects include: the close correlation between the scenes and the individuals mentioned in the legends; the fact that several different seals were used by the same person; various depictions of the royal family with the king, the queen, the crown prince and a smaller child seated on the lap of the queen; the realistic rendering of animal figures, which include a lion crouching under the throne of the king; the extremely fine and naturalistic definition of the figures, resulting in what appears to be portraitlike representations of the queen and the prince.

Two administrative cuneiform tablets dating to the Late Sargonic period (c. 2250–2000 BCE) were found in F1; they are written in Old Akkadian, but with several Hurrian personal names. A school tablet from AK contains an excerpt of six lines from a list of professions also known from Abu Salabikh and Ebla. [See Akkadian; Abu Salabikh.] With it were found one lentoid tablet and fragments of some thirty additional administrative tablets, all from the Early Sargonic period. About eighty seal impressions from AK bore cuneiform legends from the same period. [See Seals; Cuneiform.]

More than forty seal impressions were found in KW—mostly door sealings dating to the Early Dynastic III period, some possibly even Early Dynastic II. While some motifs continue into the Early Akkadian period, there are no clearly recognizable Akkadian sealings in the deposit. Prominent among the designs are the nude hero motifs and animal and human combat scenes.

Typical Early Dynastic III ceramics include medium jars with grooved rims, Simple-ware conical cups, and Metallic-ware jars with rounded or ring bases. Akkadian ceramics are characterized by a later variety of metallic ware and a continuation of the Simple-ware tradition. One tomb with more than fifty vessels excavated in the Outer City (OB1) exhibits the transition between the Late Ninevite 5 tradition and the use of early Metallic ware. During this transition, painted Scarlet-ware stands were utilized at Mozan; they appear to have been manufactured locally because both painted and unpainted examples were found at the site.

Mozan, Tell

MOZAN, TELL. Figure 3. Small statue of a lion found in temple BA. (Courtesy G. Buccellati)

view larger image

From third-millennium contexts a large number of metal objects have been excavated and analyzed—mostly points and pins, but also spears, daggers, and one scraper. A range of metal alloys was used—relatively pure copper, copper alloyed with arsenic, or low-tin bronzes. Some recycling of metals is indicated. The majority of these objects were cast, annealed and finished by cold working.


  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. Mozan, vol. 1, The Soundings of the First Two Seasons. Bibliotheca Mesopotamica, 20. Malibu, 1988. The first documentary report on the current excavations, with introductory information on the earlier explorations.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “Tell Mozan.” Syrian Archaeology Bulletin 2 (1990): 4–7. Preliminary report on the fifth and sixth seasons.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “Tell Mozan.” Les Dossiers d'Archéologie, no. 155 (1990): 18–23. Introductory overview of the major finds.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “Mozan.” American Journal of Archaeology 95 (1991): 712–714; 98 (1994): 131–133. Preliminary reports on the sixth and seventh seasons of excavation.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “Mozan, Tall.” In Reallexikon der Assyriologie und Vorderasiatischen Archäologie, Berlin, forthcoming. Preliminary report.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “The Seventh Season of Excavations at Tell Mozan, 1992.” Annales Archéologiques Arabes Syriennes (forthcoming). Preliminary report.
  • Buccellati, Giorgio, and Marilyn Kelly- Buccellati. “The Eighth Season of Excavations at Tell Mozan, 1993.” Annales Archéologiques Arabes Syriennes (forthcoming).
  • Kelly-Buccellati, Marilyn. “A New Third Millennium Sculpture from Mozan.” In Essays in Ancient Civilization Presented to Helene J. Kantor, edited by Albert Leonard, Jr., and Bruce B. Williams, pp. 149–154, pl. 26. Chicago, 1989. Critical analysis of the plowman and herd stelae from temple BA.
  • Kelly-Buccellati, Marilyn. “Three Seasons of Excavation at Tell Mozan.” In Tall al-Ḥamīdīyā 2, by Seyarre Eichler et al., pp. 119–132. Orbis Biblicus et Orientalis, Series Archaeologica, 6. Göttingen, 1990. Preliminary report.
  • Milano, Lucio, et al. Mozan, vol. 2, The Epigraphic Finds of the Sixth Season. Syro-Mesopotamian Studies, 5.1. Malibu, 1991. Documentary publication of the first find of stratified third- millennium cuneiform texts from Mozan.

Giorgio Buccellati and Marilyn Kelly- Buccellati