Tell Leilan Utilan phoinxtaphed from the wesi Seen from a h flam - the modem village- which ti bull! on top of the Acropoln. probably looks much at the uw did in antiquity on the Habur Plains of Syria B) H.wn Wnu ^^^^^ell Leilan is certainly one I of the more imposing sites I in northern Mesopotamia. •X> Situated on the left bank of the Wadi larrah, in the heart of the fertile Habur Plains of northeastern Syria, the massive extant walls rise more than 15 meters above the level of the plain, and enclose an area of some 90 hectares {900,000 square meters), making it one of the largest ancient sites in northern Mesopotamia, even larger than Ebla (56 hectares), Ashur |30 hectares), and Tell Brak |43 hectares] The gates of the city were on the north, south, and cast, while on the west the ancient river probably provided a protective shoulder The site is dominated by a 15-hcctarc Acropolis, which probably featured large public buildings in its northern section and a "ziggurat" to the south. In 1978, with the cooperation of the Directorate-General of Antiquities in Damascus, Yale University began its work at Tell Leilan with a topographic survey of the site. In association with the Metropolitan Museum of An in New York, three full seasons of excavation (1979,1980, and 1982) have since been conducted. These excavations have tested four areas of the site. The Acropolis-northeast has been the focus of horizontal excavations, while three stratigraphic soundings have also been undertaken: Operation 1, a 4.5-metcr-wide step trench, now almost 16 meters deep, which goes down the northwest slope of the Acropolis; a small sounding RIHIH \I Km IIMOUX,)st MAttCII IWS a Ibpuxraphical map of Tell Leilan showing areas of excavation M of the 1982 season [designated 57F02) in the Lower Town; and Operation 2, a small sounding at the City Wall. In the first pan of this paper I shall briefly present some results of the excavation of the Acropolis-northeast, and then discuss what these suggest lor our understanding of the site during the early second millennium r.c. In the second part I shall summarize what we have learned in the three soundings, and consider what this may tell us about northern Mesopotamia in the third millennium b.c. Tell Leilan in the Second Millennium b.c: Excavations on the Acropolis-northeast Three seasons of excavation on the Leilan Acropolis now provide new data for the significance of Leilan, its ancient name, and its role on the Habur Plains of the early second millennium b.c. The topography of the Acropolis suggests that large public buildings arc situated within the northeast quadrant. For the purposes of establishing the chronology of settlement within the site and its Acropolis, as well as testing loci that might provide evidence for the site's historical role, this area has become one of the central research loci of the Tell Leilan Project. Initial explorations in 1979, barely scratching its surface, allayed all previous fears that the Leilan Acropolis was capped by a Roman-period fortress. At 50 centimeters down, the trained excavator is able to articulate the tops of massive, sun-dried mudbrick walls erected some 4,000 years ago. Three building levels of such collapsed structures have now been identified within our excavations on the Acropolis-northeast. Building Level 1. Immediately under the surface, Building Level I comprises the remains of a mudbrick platform or paving, now only a few courses high in some places. This surface and its brickwork were set against the collapsed southern facade of an earlier building level. Building Level II (see stratigraphic section). Later surfaces related to the Building-Level-I brick platform have also been identified elsewhere within the collapsed walls of Building Level II, and associated with these surfaces are potsherds of the "Habur ware" variety that is securely dated to the nineteenth century b.c. These same kinds of ceramics also comprise the assemblages of Building Level U and Building Level III. This then is the temiirial occupation on the Acropolis, perhaps representing scrappy, insubstantial habitations, possibly of squatters or temporary settlers who were seeking shelter within the ruins of large, recently collapsed buildings. These ruins are now known to be the remains of a major sccond-millennium-b.c. temple. Building Level II. Thirteen hundred square meters of the Building-Level-II temple have now been retrieved, with an equivalent area probably remaining to be excavated. The northern facade of the temple presented an imposing configuration of niches and engaged columns arranged in panels, alternately spiral and plain-continued on page 12 INBI.K AL APniMIIUK.M MAIM 11 I'M', 8 HIBUCAl AftCHAlOLOGISTMARťH 4--»« To date 1,300 square meters of the Btiilding-LeveUH temple have been retrieved. left: View ofexcawtion urea 45 PIQI2 of ihr Building Levelll temple of the AcropoIls^ northern taken from the north Below: Plan of the Level II of the Acropolis-northeast The areas in darker color indicate secondary wall connections of a slighilv later date, these weir done with a whiter, coarser, and more hagtle mudbtwk than was used in building the original walls -\- «40J -f- the trunk of the palm This opcratton generally kills any suciet bud» from the tree |Do«f*on 1921.26| The mudbnek columns of the two lr i Ian temples provide, so tar. iacadc decoration using four different types of engaged or "half" columns |I| a palm trunk column, with dumond-tthaped frond scars, surrounded by braided columns (Building Level III), |2| a palm trunk column with petal) ike imbricated jovw-lappingl fronds [Building Level n. south facade); (3) columns of mud-brick spirals twisting in alternate direction* (Building Level 0, north facade); and (4) plain-faced columns either twisted (Building Level HI) or straight (Building Level ill What arc palm-tree columns doing m nonheasternmost Syria? Palm trees certainly arc not at home m northern Mesopotamia Indeed, they arc rarely found further upstream than modem Abu Remal, on the Euphrates near ancient Man, at the border of Syria and Iraq. But they ore and were at home in southern Mesopotamia, and apparently were an architectural convention for the decoration of public building facades from at least as early as the Uruk penod when the pillars and engaged columns ol the Uruk temples were emblazoned with cone-mosaic designs imitating the trunks ol palm trees (Buren 194S: 29] Brandea 1968). At Al-Uhaid. in the laic Early Dynastic penod, palm trunks were used as the cores for mosaic and sheathed columns (Hall and Woolley 1927 100; Howard Carter 1983:65). Large mudbnek date palm columns with diamond-shaped frond scars decorated the gateway into the so-called Bastion of WaradSin at Ut m the nineteenth century ac (Woolley 1936). Contemporary Man, however, provides the most context* for palm-tree decoration within public buildings: three for palaces and one for a temple A much discussed chamber within the Palace of Zimn-Lim later occupied by Shamshi-Adads son YaimakliAdad. was known as the in BUH l( »1 AK( IUI OliK.ISTHAHC 11 l«*S Dahil Abbas, a tncmn wirkman from the leilan village, axcmxttci one of the BuildtngrLevelll ■.' ■: j j i r i ■ of the north facade of the ttimple. on the Acropollt northeast Palm Trees and Spiral Columns Sizes of Syro-Mesopotamian Sites During the Third and Early Second Millennia b.c. Ancient Area in Site Name Hectare* Chucra 100 Ťava 100 Leilan Shubat Enlil? 90 Hamoukar 90 Khoshi 90 HaJh.nl 90 Mardikh EbU 56 Qarat Sherqat Ashur SO Brak Nilabshinu? 43 McskenťBalis E mar 37 Bi'a Tuttul? 36 Tbuqan Urshuf 28 U mi.id Katana? 28 Hjmmam ct-Turkman Zalpahí 25 Bam Kahai 23 Bdla Shibaniba 15 Gcrmayir IS Chagar Bazar 13 Arbit 13 Allun U Yorgan Tepe Nuzi 4 Cawra Si/esnl I Other Mcsopuiamian Sites Mishnfe Qatna 100 Fara, ED m Shuruppak 100 Hariri Man 54 Inghara, ED III Ktsh 50 Mizyad, ED in Agade- 48 Each year the dead and dying outer frond* (Aiabic wiifl are cut fruni ihc palm H about a foot from It* trunk .. When the palm » about . i i i i yean, old the woody «id expanded bate u of the fronds (Arabic karib am cut away close to the trunk of the palm This operation generally -ill* any sucker bud* from the tree {Duwson 1921:16) ryhc mudbnek column.* uf the two X LeUan temples provide, so far, facade decoration using four different types of engaged or "half' columns* |■ t a palm trunk column, with diamond-shaped frond scars, surrounded by braided columns (Building Level HIV |2| a palm trunk column with pctdlhkc irnhncaied |ovcr-lappingl fronds (Building Level 11, south t.u i-ic (3) columns of mud' buck spirals twisting in alternate directions (Building Level I), north facade)- and (4) plain-faced columns cither twisted (Building Level Hit or straight (Building level Il| What arc palm-tree columns doing in nurtheastemmost Syria i Palm trees certainly arc not at home in northern Mesopotamia Indeed, they are tarely found further upstream than modem Abu Kcraal, on the Euphrates near ancient Man, it the border of Syria and Iraq But they arc and were at home in southern Mesopotamia, and apparently were an architectuial convention for the decoration of public building facades from at least as early as the Uruk period when the pillars and engaged columns of the Uruk temples were emblazoned with cone-mosaic designs imitating the trunks nf palm trees (Buren 1945: 29, Brandcs 1968). At Al-Ubaid, in the late Early Dynastic period, palm trunks were used as the cores fot mosaic and sheathed columns (Hall and Woolley 1927: 100; Howard Carter 1983:65). Large roudbnek date-palm columns with diamond' shaped frond scars decorated the gate -way into the so-called Bastion of Warad-Sin at Ur in the nineteenth century m: (Woolley 1936). Contemporary Man. however, provide* the most contexts for palm-tree decoration within public buildings, three fot palaces and one for a temple A much discussed chamber within the Palace of Zimri-Lim, later occupied by Shamshi-AdadS son Yasmakh Adad, was known as the M Hint i* si skouiouk.im «AM h im* "Date Palm Court'(Al -Khalcsi 1978), while the famous "Investiture" wall paintings oi the palace depict palm trees with fronds trimmed in the •diamond" fashion, like the mudbrick columns of Leilun Building Level I!! and the Bastion of Warad-Sin at Ur (Pan.1958: plates 10-13). Less well known, but very imiiguing, is the reference to a 'Palm Tree" Palace in Shamshi-Adad's letter of reprimand to Yasmakh-Adad, quoted in the sidebar to the present article entitled "The Search for Shamshi-Adad's Capital City." (Might this be referring to yet another Man palace'| A stone-column base from Mari cut in imitation of palm scales suggests that columns rcsemhling palm-tree trunks would have been quite at home here (Parrot 1939: plate V. 2| And lastly, it did not escape the notice of Andre Parrot that the left side of the doorway into the Dagan Temple at Mari "semble avoir etc decorc de truncs de palm ids' iParrot 1938: 211. In southern Mesopotamia, palm trees arc also mentioned in association with the Shamash temple at Larsa a major contemporary city on the Euphrates. Cungunum, king of Ursa from 1932 to 1906 b.c, went so far as to name a year "The year he brought two bronze date palms into the temple of Shamash'' [Ungnad 1938: 155] The E BABBAR Shamash temple at Larsa has, for several years, been under excavation by the Univer- sity ■1' Parts team directed by Professor f.-L. Huol, but bronze palms have not been retrieved. However, a set of beautifully constructed courtyards have been exposed The interior walls of one of these, Courtyard I, were decorated with spiral columns very similar to the spiral columns used as exterior facade decoration in Leilan Building Level II (Calvet and others 1976; Hunt and others 1983). A very intriguing patallcl tar the use of columns, both palmlike and spiral, is available at the contemporary temple of Tell a!*Rimah, just across the border near Tell Afar, Iraq. The Rimah temple features spiral columns similar to those of Leilan, as well as two kinds of palmtlke col-umns, a "scale" pattern, and the diamond-shaped pattern The petallike imbricated pattern of Leilan Building Level II is not in evidence here, but may have been used in the still unexcavatcd portions of the temple. Two carved stone blocks depicting deities standing between palm trees have recently been published from ihc excavations at Tell al-Rimah iHoward-Cartcr 19831. One of these presents u goddess standing between palm trees with fronds trimmed with "compass-like scale patterns." A second hlock features a bullman between palms with trunks decorated "with a herring-bone pattern* (Howard-Carter 1983:67, plate LHA}. The "herringbone pattern* here precisely replicates the spiral pattern presented by the spiral columns at Rimah. Indeed, the spiral columns at Rimah, Leilan, and Larsa, accompanying other palm tree columns, probably also represented palm trees whose frond imbrications could be perceived and represented as diagonal cuts along the palm trunk. In southern Mesopotamia the annual fertilization of the female palm resulted in a bounty of dates and date byproducts. Hence, the palm tree was a symbol of agricultural fertility, even in northern Mesopotamia. On the treeless Habur Plains, and across northern Mesopotamia, the mudbrick palmlike spiral colurnns of Leilan and Rimah probably reflect, as well, the practical use oi palm timbers in building construction. As Shomshi-Adad himself wrote to Yasmakh-Adad The palms, cypresses and myrtles that have been brought from ihe town of 1 ir-:i" l)C 0t i I; tent Ul the town ol Subrum Send Mashiya and a lew officials with him to Subrum. where ibey *hall divide the palim. cypresses, and myrtle* ami three lot*. Send one-third of the palms, cypresses, and my tile* to Ekallalum. one third to Nineveh, and one third to SbuKii Enid. . , _ Thai which ymi tend to Shubat Enlil is to be transported by ship to the town oi Sagsawum. then from Sag-caratum to Qattunan. From Qat-tunan let the men of Qaitunan lake it in vmiv m , ■ and let them bring tt to Shubat Enlil (ARM I. 7;4-31) A small portion of the southern facade of the Building Level-II temple ha* ihm with niches and engaged columns but was not an well preserved. One mudhrit 'dressed'palm tree heen excavated. Like the northern facade it was decorated >lnmn was clearly sculpted to resemble the trunk of a ■aUCAJ KKBAMOUXmnUaKB loss II faced arrangements, across a distance of more dun SO meters. The western portion of this facade apparently extends across a massive mudhnck platform, still only pan tally ex-cavated, that seems to antedate the construction of the temple, and against which it was built. Portions of this facade still stand to heights of 3 meters( to tudgc from the thickness of its walls, the facade may have stood as high as 6 or 7 meters in antiquity. Looming over the plain, more than 20 meters below, this array of mudbrick architectural power would have imposed itself as a formidable vision upon the merchants and mule caravanners trekking along the great cast-west "trans-Mcsnpotamian" trade route that passes alongside Tell Lei Ian On the Acropolis interior, and looking southward toward the zig-gurat, the southern facade of this temple also featured niches and engaged columns. Only 9 meters of this facade have been excavated so Dannx the 1V79 MOda an earlier tempt* was found. BuiMinx Uwllllofihe Acropolis nofthta.il S*t aetnmpamint text for an explanation of the plea. _ -I- North + c, -1JF "t 1 ■ f ■ I Jjl JTrffilljTt SIP-' ^ A ■ II * i i * i + far, but the niches here surround the badly eroded surface of a mudbnek column coated with thick mud-plaster and sculpted to resemble the trunk of a "dressed"palm tree. Indeed, that is what the spiral columns of the north facade may have represented to the observer in the eighteenth century tut (See the accompanying sidebar, "Palm Trees and Spiral Columns") Building Level III. Immediately south of the south facade of Building The temple in Building Level III was rebuilt in Building Level II. Level II our excavations have retrieved portions of what appears to be an earlier temple, more than likely a larger temple whose restoration or reconstruction in Building Level II resulted in its foreshortening. That is, Building Level n seems to be a rebuilding of Building Level III, but without a southern courtyard with side rooms. The extant plan of this structure reveals a large central courtyard (A) on the south that is Hanked by narrow rooms (B) on the east, and probably the west as well. SOU!!) 11 III BIII'Al ABTHAI010GISTMAWH1«*V Example* of Habur wan dating to the nineteenth tenuity nc. hum Building Uveh II and III of the Acmpolis northeast _ Detail ol the northern facade of the Building* tsvelIII temple on the Acropolis-northeast. An engaged, mudhnck column that is sculpted ui resemble the trunk ol a palm tree it the focal point of this section of the facade, ills location is indicated by the letter "C'on the accompanying plan l Left edge fragment of an inscribed stele ma finegrained black stone With parts of three lines of Old Babylonian style "monumental' script, this fragment, which is obviously only a small portion of a very large none monument, was retrieved within wall-collapse strata of room 3 of the Building-Uveil I temple The northern face of ihe cast west wall that closes the northern rooms was decorated with stepped mches symmetrically set against a central, engaged mudbnek column c The late of this column was heavily coated with mudplastcr, and then sculpted to resemble the trunk of a palm tree. (See sidebar! The floors that are set against this facade were rclaid three'times, their extension to the north underlies the slightly later constructions of Building Level II (see the stratigraphic section!. Second-millennium-temple artifacts. The floors of the Building-Lcvcl-II temple were littered with thousands of potsherds, as well as animal bones and carbonized wheat, barley, and other seeds - the refuse of daily cooking and eating, from which we hope to reconstruct not only the range of comestibles consumed within the temple but also the crops and agricultural practices that characterized the Habur Plains during the second millennium ».c Cuneiform tablets were also retrieved within several rooms; most are economic documents, recording the receipt of various commodities ■■■IIC W Mil HUilliM.lN] W\H\ H mi Cities, by definition, arc functional center. serving j dependent hinterland When titles first emerged in southern Mesopotamia, j means of recording (he transact tons thai maintained this new social and economic system became a necessity The transactions were complex and involved a multitude of groups, individuals, and instuutKins cittern and villages, classes of administrators and laborers, and officials regulating and recording the transfer of goods and services Two devices evolved and were regularly employed to facilitate these exchanges. One was writing, and the second was cylinder scaling Writing was. of course, used to record the details of transactions; but some means was needed (o insure ihe veracity of the inscription, or in cases where only the goods were to be transported or received, the integmy ol the shipment Ancient Near Eastern officials, therefore, sealed tablets a* well as containers and even storerooms with cylinders bearing then names and titles, much the way post offices stamp telegrams, or customs off icials bind and seal international shipments. Abosr: Cylinder seal impret won /LS3-10S) found on the floor In the southern part of room 13 ol the BuildingUvel II temple, Its inscription read\. 'iun Adad. son of Zidnva tenant of Shamshi Adad' Middle: Thirteen cylinder teal impression* Inund in the Building Level 11 temple bore the second Sun Adad uucnp turn 'Adad tonal inspector ofthegod.... and thegod..,. Hurt Adad. the .'Seven were found on the floor of room I2IL80-176, - ISO. -186. -190, -191. -194. and - l9Sf-. three were found on the floor of room I3IL82-1I8. -119. and -1201. and three were discovered in the second arr blockage between rooms 8 and 12ILB2-I23. -126. -127) The scene depicted on this teal it a ttandard. Old Batnhnian representation ol the ~god with mace'm front ot the 'suppliant rpddess' Below: Alio scattered among the rubbish of room 8 were 227sealimpressions in various stages of preservation bearing, this ascription ~Beh-emiM)i. i*minr of Khayaabum. ser vant of the god Adad' The ttandard Old Babylonian style glyptic design, the "god with mace'and 'suppliant goddet*.' It here supplemented with a "winged-lanwssu" demon standing behind the goddess A ctenent-uai and a monkey are used at filler between the god and goddess important fox the temple economy. Systematic sieving of the temple floors also nude it possible to retrieve numerous inscribed cylinder seal impressions From the southern part of room 13, one seal impression bears the inscription of Sun Adad, son of Zidnya, servant of Shamshi Adad thereby conclusively proving the occupation and use of this temple duringShamshiAdad'sreign. Seven impressions of another seal of (the same") "Suri-Adad" were also found on the floor of room 12 and three more were found on the floor of room 13: Adad . . canal inspector of the god ..and the god ..., Sun-Adad. the... But sometime, probably not too long after the initial use of these floors, three alterations were made to this building, each utilizing a characteristic mudbnek that was whiter, coarser, and more fragile than that of the structure's original walls. The relationship of the alterations to the temple's original walls can be observed in the plan of Build ing Level II. where the alterations are mdicated in dark shading. A por tion of room 12 was walled off to become a door less room 13, with a north-facing window; the long central cclla, which probably had a mudbnek altar set squarely in front of its northern wall, would have then ceased to serve as the carefully planned focus of cultic activity. The floor of room 13 is the last living surface in this room. This floor passes under the enclosure wallj a similar situation obtains to the west of the enclosure in room 12. The face of the eastern wall of room 13 extends below this last floor. The two sets of rooms 15-16 and 8-9, which are essentially parallel arrangements, each had one of their two entrances sealed with a curtain wall. Removing the secondary blockage of the doorway between rooms 8 and 12, three additional clay sea] mill 11 At ABI luroiocivi MAM II mi Above: Iwo imprettmns from a cylinder teat 1182 - 74 and - 7Sf were found among the organic rubbish of roam 8 in the Building-Level II temple Theii inscription reads as follows "Apll-ilixhu. ton of Altbanithu. senvnt ofThrumnatki' These cylinder seal tmpressUim are derived from a cylinder seal with an apparently unique design A 'hero" holds the tails of a cow and a lion Other 'hemes' appear to lump over the back% of these animals On each side of a mythological bird, in the lower refiner there n a guilkrche Some paralleli fat dm seats dextgm occur in contemporary Anatolia and m southern Metopotamia during the thud millennium «c. Below: This macrophotogiaph of a lor ttopper /measuring approximately 40 millimeter, wide, found in room 8 xhirwt the seal imprestion of 'Apil-dlshu. son of Ali-banisha servant ot Thrumnatki.' impressions ot the second Suri-Adad inscription were retrieved from the interstices of the brickwork. These were probably lying on the floor when a mason swept them up to fill cracks in his sloppy construction of the secondary wall. After the construction of this wall, a deposit of ash and trash built up against it upon the floor of room 8. Within this organic rubbish 229 additional seal impressions were tossed as jars of commodities were opened. Two of these bore the inscription of: Apil ilt-hu son of Ali-banishu, servant of Tunnn-iutkl, while 227 (complete and fragmentary) bore the inscription of: Beli-emuqi, servant of Khaya-abum. servant of the god Adad. Comparison with other temples. The Building-Level 11 temple at Tell Leilan. apart from its historically fascinating floor debris, remains an artifact, an expression of personal and social styles identifiable in space and time. As such, it is worthy of comparison to other, similar, monumental architecture, even though its plan is not yet complete. We have speculated that the original plan of the temple will be available in Building Level III, with Building Level II only representing a partial rebuild of that temple. If this suggestion proves correct, the Leilan temple may have been one of the largest constructed during this period, for it would then be approximately 6,000 square meters, or about twice the size of the Sin-Shamash temple at Ashur and the temple at Tell al-Rimah, and the equal of the Ischali temple and the Ashur Temple at Ashur. This, however, is not too surprising because there does seem to be a gross correlation between the size of a city and the size of its public buildings. A "langraunT-temple? The spe cific plan of this building is, however, rather surprising. (Note that the isometric plan of the Building-LevcMI temple does not include the building's secondary wall construe- Rim It m \ih HUOUX.IM MtlKHim U Two leilan village smrkraen aevt floor debris of the Building level II temple with millimeter tcmn* Supervising the u farouA lunatl. then a graduate undent arid now a proteuot of iithient Sear Eastern languages at the UnivtmtY of Aleppo Excavation is not tor the unit of hcan. Thin- is a daring kind of brinkmanship, a continuous tension, between the need to excavate and remove, and the need to preserve and isolate, while the clock ticks away, workmen stand by waiting, and precious research funds dwindle. In a building such as the Leilan temple, massive brick collapse is first removed, and wall faces of mudplastei arc then carefully picked with hand tools 10 is not to create" walls but to define them against the matrix of virtually identical rnudbrick collapse. Following wall faces down to their floors can be nerve-racking There u the ever-present danger of missing the floor, following the wall-face down to its 141011001 foundations or to an earlier floor and thereby mixing the strati-graphic deposition that provides the temporal framework for archaeological reconstruction. Delicately tracing with handpicks the "break" between collapse and wall-face down to the first centimeter-sized patch of "break,' which indicates the stamped, sometimes lightly plastered floor, is an anxiety-ftiled process. There i- no second chance. Unique among research disciplines, archaeology destroy* pan of its data, the archaeological context, as that data is retrieved and then removed m the excavation of still earlier deposits. When floors are located, student supervisors and pickmen call out for fine one-millimeter screens The lli»n deposits provide the crucial evidence lor activities that can be securely dated, as opposed to postoccupatinn collapse deposits. Sieving assures uniform retrieval: No artifacts, however small, will be passed over as the debris resting immediately upon the floor surfaces is cleared._| Hons.) Here it is possible to see the almost symmetric arrangement of side rooms (rooms 4, 5,8, 14, 15, and 16) around a long central cella (room 12|, which itself is, apparently, preceded by a wide antccella (room 101, only tragmcntarily defined in the areas excavated to date. With the addition of the secondary blockage walls, access to the cella would have been impossible except through the antecella and, presumably, a doorway to the south through the south facade. Have we then a "langraum"-, or long-room, temple, the classic Assyrian temple-form of the first millennium b.c. which always features the lineal arrangement of "doorway' - "wide-room" antccella -"long-room' cella' If the Building* LcvcI-II temple at Leilan is "langraum" it may be the earliest temple of this type. Some archaeologists have ar gucd that "langraum"-temples do appear in the early second millennium b.c. at Ischali and Tell Harmal (Strommengcr 1962: 416j Amiet 1980: 535; |. Oates 1979: 79). These temples, however, do not really have the room arrangement characteristic of "langraum'-templcs, and teem to be examples of the period's characteristic "Babylonian" temple with a "brcitraum" ("wide-room"l cella (Hrouda 1971: 152, Hcinrich 1982: 189). The earliest "langraum" known .it present is that of the Sin-Shamosh temple at the Assyrian capital of Ashur, constructed by Ashur niran I in the sixteenth century »x. The next oldest is the famous Innin Temple of Karaindash at Warka, which dates to the fifteenth century p i (Heinnch 1982). If the Leilan temple is of the "langraum'-typc it is no longer necessary to hypothesize cxtra-Mcsopotamian, possibly "Kassitc," origins for this temple-type as WM argued in the past |Martiny 1936, lantz I960; Matthiae 1975). Concurrently, however, the Leilan temple raises new questions: Why is this temple-form appearing at Leilan at this time, and what are its origins' One hypothesis that might now be entertained is that the "Assyrian langraum" temple-plan actually was a Shamshi-Adad, or Shamshi-Adad period, innovation subsequently adopted or copied by later Assyrian royalty. In much the same way that Shamshi-Adad mimicked the royal titulature of the Akkadian dynasts, so later Assyrian monarchy perpetuated many Shamshi-Adad innovations. Two outstanding examples of this are his name, which was subsequently adopted by four other Assyrian kings, and his Ashur inscriptions, whose style and dulcet were imitated by Middle Assyrian kings in their royal annals |Lacssoc 1963: 95). Is the "langraum* temple then an innovation of Shamshi-Adad' If this were the case, we would expect the temple constructed by Shamshi-Adad at Ashur to be "langraum." Unfortunately, the excavation of this structure does not allow us to make definitive statements about the temple's plan in the bib i if At «K HUOUM.ivt VUH H IW Isotnetnc plan of the Building U\rl II temple of the ■!...*;«,:.:.'. n bean This shows th* budding's plan prior to the construction of secondary blockage walls Note the central cella (room 12) surrounded by an almost symmetrical attangtment of ode rooms (rooms 4, 5. 8. 'd. 15. and 161 Room 10 it an anteceUa days of Shamshi-Adad but, as Anion Moortgat noted, the foundations of this structure leave open the possi-bility that the temple's cella was originally "langraum" |Moortgat 1969: 76). Wry intriguing, as well, arc the engaged columns that once decorated the exterior of this temple (Hallcr 1955: 33, figure 8j Heinrich 1982: 198 - 99). V* do not know how they were decorated. They might have been spirals or have been decorated with one of the other palm-frond motifs. For the moment, however, we must refrain from absolute statements about origins and explanations, for definitive evidence is not available - nor, in the archaeological world, is it ever likely to be. The typology of temple-plans seems to allow for the categorization of the Building Lcvcl-D temple, but only in so far as it has been excavated until now This last caveat is necessary because the Leilan temple has not yet defined itself conclusively as "langraum." A bent-axis temple? Recall the note above concerning the disposition of the temple doorway. We have assumed that the main doorway into the temple lies directly in line with the doorway into the long cella because when the secondary blockage was in place there would have been no other access into the building. At present, therefore, we anticipate finding a magnificent doorway along the facade where wc have already located a palm-tree column. And if the doorway is not there-* MRUCAl 4JMHMOLOClST«UM'H l«M I' «PI! 45PU|4SQ11 * <»*» Limit ol Eicovolian Mete.s T/w stratigiaphic section of Building Lewis I. II. and III on the Acropolis northeast is shown abow and is continued on the following paxes. The entire section documents sixty meien of stratigraphy across the Acropolis. Note Building Lewi II in squates 45P1I, 45Q1I. and 4SRU and the foundation trench for the south facade, of Building Level II identified as stratum 4 in square4SRII The exterior surface for Building Lewi II Is identified in square 45RII as floor-stratum 2. Buildint Level I. the low platform and pavinx that was sei against the ruins of Building Level II. ts also visible in 4SR1I abow floor 2. _ Another characteristic of the Leilan temple's decoration is the use of "reveals," or doorjamb insets, to accent important passageways. On the isometric plan of the temple, a "cookie-crumb trail" ol reveals defines first the corners of the long cclla, and then leads the worshipper out, not through the hypothesized *langraum"-type doorway, but to the west, along a bent axis, or "knich-achse," past two side rooms and then into room 19, at which our excavation has halted. Quite simply, an important route has here been defined that, in spite of its eventual blockage, once featured prominently in the traffic pattern of the building. If our next excavation season shows that there was no doorway out of the building directly in front of the long cella, this revcai-decorated "bent-axis" route must have provided access to the cella. Such a "bent-axis" type temple-plan takes us back to the Diyala excavations of the Oriental Institute at Khafajah where the famous Sin temple sequence for the Early Dynastic period is dominated by "bent-axis* temples. These temples mark a clear dis-juncture with those of preceding periods in the south; their existence in the Diyala, east of the Tigris, has led some archaeologists to see the type as an "osttigridische Erfindung" to be associated with the third-millennium-b.c. Hurrians of north Mesopotamia and north Syria (Lcnzen 1955: 17, Hrouda 1984: 65). Far from being a Shamshi-Adad-period innovation, then, the Building-Level-II temple plan may harken back to the still earlier, third-millennium, urban roots of Tell Leilan (sec the section below on Tell Leilan in the third-millennium b.c.|. Tell Leilan and Shubat Enlil. Does the deposition of seal impressions of Suri-Adad, Turum-natki, and Khaya-abum within the Buiidtng-Level-II temple allow us to equate Tell Leilan with Shubat Enlil through the documentation for the city's last days? (See the accompanying sidebar, "The Search for Shamshi-Adad's Capital City") Such a suggestion would be bold, if not rash. The deposition of seal impressions inscribed "Suri-Adad servant of Shamshi-Adad" cer- tainly, however, occurred prior to those of Turum-natki and Khaya-ahum, and these rulers only figure in the Mari documentation for Shubat Enlil after the death of Shamshi-Adad. To be sure, we have no Tell Leilan documentation as yet for Kunnam the Elamite and Atamrum of An-dariq, the other rulers of the city. Nor do the impressions of "Suri-Adad, servant of Shamshi-Adad" by themselves require that Tell Leilan be considered the seat of Shamshi-Adad's power, for such seal impressions are known from other sites across the Habur Plains and northern Iraq such as Chagar Bazar, Till Taya, and Tell al-Rimah, and even AcemhOyuk on the Anatolian plateau (Loretz 1969: no. 13; Postgate 1973i 173-75, Hawkins 1976, Ozguc 1980: 99). There remains, too, the conundrum of 227 seal impressions and fragments inscribed "Khaya-abum of Apum." In most circumstances such would be taken as prima facie evidence for identifying Tell Leilan with Apum, a city near Shubat Enlil that also has yet to be ii lllHiirM ARCHUOUK.ISTM-VHCH 1»S* Umí' ol Etwvalcn identified on the Habur Plains. At this time, it seems safe to answer our questions only with additional questions. In consideration oi Tell Lcilan's location, size, morphology, and terminal occupational history, if the site is not Shub.it Enlil, what is it' Apum" But Apum is not known to have existed in the third millennium ».c, which is when our excavations indicate that Tell Leilan first became a large city |see the second half of this paper!. Similarly, Shubat Enlil is not known as a city name prior to the reign of Shamshi Adad. If Till Leilan is Shubat Enlil. what was its name in the third millennium? In the early second millennium >£. lei! Leilan was clearly one center of regional power on the Habur Plains. The sequence of Acropolis building levels, their artifacts and inscriptions, and their debris, litter, and collapse provide an arena for historical investigation, just as they dramatically draw attention to the actions of individual personalities who represented the contending interests of villages, cities, regions, and empires in the early second millennium. Whether Tell Leilan was Shubat Enlil, or another documented large city such as Apum, remains to be determined and adds another, if tangential, problem for resolution. Sites such as Tell Leilan do not draw their inherent archaeological significance from their correlation with historically documented settlements. On the contrary, it is the settlement itself that is of signili In 1800 B.C. Tell Leilan was a center of power on the Habur Plains. cance because of the role that it played within a region. A useful example of this name-site relationship is Jell Mardikh |ancicnt Eblaj. Prior to the recovery of the third-millennium s.<: palace at Mardikh, Ebla was simply one of several west Syrian toponyms known from south-em Mesopotamian documents to have been destroyed or conquered by Sargon and Naram Sin. The archaeological recovery of Tell Mardikh, however, now informs us of Ebla's role in Syrian history. Similarly, Lcilan's size and geographical position inform us of its general role within the region. His- torical references to a city named Shubat Enlil inform us of that city's significance in the region. If the two kinds of evidence pertain to each other a sencs of well-established historical problems can be defined. If they do not, an entirely new set of problems may emerge. Regardless, therefore, of Tell Leilan's name in the second millennium B.c, the details of its historical and regional role remain to be examined. The imperial and local dramas of the early second millennium on the Habur Plains were not without precedent, however. Nor was it simply fortune that situated this very large second-millennium occupation at Tell Leilan. Tell Leilan in the Third Millennium n.c: Soundings at the Acropolis-northwest, Lower Town, and City Wall In order to establish a framework for problem-specific investigations of the site, a preliminary series of three, deep stratigraphic soundings were undertaken in I9B0. These soundings-designated Operation I, Operation 57F02, and Operation 2 —retrieved the ceramics associated with each stratum of occupation, as well as radiocarbon samples and floral and botanical remains thai BWMt m MfCHAlOITH.IVT M « M H Square* 4SS1I and 4577? show the continuation of Building-Level-l paving. Underneath the paving in 4SS11 and 4ST1I the continuation of floor stratum 2 of Building U\rl II can also be seen. Underneath that, however, are several strata of bricky wall collapse derived from wall A la 45771. Below those strata of wall collapse, numbered 6 through 12. the last of three Period-Ill floors can be seen. These plastered floors abut the plastered face of wall A. The extension of Building Level 111 to the south can be seen in the remainder of 4577/ and4SVll. Relative Chronology i.e. Northwestern Syria Hobur Plains Northern Iraq Southern Iraq 1600 Old Syrian Period Lcilanl Old Assyrian Old Babylonian 1900 |rVUrdikhülA-B) .in i ■ in ■n Late Protosyrian TayaVI Is in - Larsa (Mardikh IIB2| Leilan lib UrlTJ Guti Mature Protosyrian Akkadian 'tun (Mardikh tlBIl Leüanlla TayalX Early Dynastic III Leilan tJIc Early Dynastic n Early Protosyrian Leilan lllh Ninevitc V (Mardikh HA) (Amuq H) Leilan Ola Early Dynastic I AmuqG Leilan IV LateUruk Late Uruk won [Mardikh I) 4100 Amuq F Leilan V Early Uruk Early Urak Uilan Via Late Northern Ubaid Ubaid 4 5000 AmuqE Leilan VIb Early Northern Ubaid Ubaid 3 Amuq D Halai Halai Ubaid 2 5500 Ubaid 1 allow for the initial occupational sketch of the site as far back as the fifth millennium b.c. (For the precise locations of the soundings, sec the topographical map at the beginning of this paper.) The stratigraphic sequence of ceramics has now been statistically analyzed, and allows us to charac-tcrize each occupation floor by the presence or absence of specific kinds of pottery and, still more importantly, the relative frequency of each pottery-type within the sample for each stratum. This kind of quantitative analysis, a prerequisite for eventually establishing smaller pcriod-izations and linking occupations at different loci to each other, also makes possible an "objective" lumping of strata to form ceramic periods. Judging from the relative frequency of ceramic types, strata more similar to each other than to other strata can be statistically defined as a ceramic "period." Sets of radiocarbon samples retrieved from these soundings have augmented the periodization available from the ceramic analyses. In a ■ llllll l( Al AfiCIIAIOllH.IST.MAW "It I9SS 45T11 45V11 region as sparsely explored as the Habur Plains, these radiocarbon samples mark the beginning of the resolution of fundamental chronological problems, including some that have still not been resolved in adjacent regions where archaeologists have worked for many years. To facilitate the resolution of some basic chronological problems, we haw attempted to process a large number of samples from individual contexts, thereby providing for the reduction, through weighted averaging, of the standard deviation thai accompanies each determination. The first stratigraphic sounding at Leilan, which we have called Operation 1, was actually started briefly in 1979 but became a major research effort in 1980 |scc Schwartz 1982). This sounding is now 16 meters deep and presently has reached to the Ubaid period (see the strati-graphic section of the Acropolis-northwest; see also the Tell Leilan ceramic periodizations). Virgin soil, probably under several strata of Halaf-period settlements, is likely to be another 10 meters below. Above the Ubaid-pcriod strata (period VI|, which comprise (he remains of domestic structures, arc several strata with similar ceramic shapes but few painted vessels (period V|. These may be contemporary with the 'Early Uruk" period in the south. Four distinct strata then follow with sherds from beveled-rim bowls that characterize the "Laie Uruk" period in southern Mesopotamia. Immediately following these are some 25 strata (period ID) with painted and incised "Ninevite V ceramics and residential remains. This sequence of Ninevite-V-period strata is perhaps the longest yet retrieved. It appears to span the enigmatic gap between the northern equivalents of the south's Uruk period and the Early Dynastic III period (Schwartz 1982; Weiss 1981-1982, 1983). Startling, however, is the occupational history that can be reconstructed from the stratigraphic evidence of Operation I and from additional tests on the Lower Town (Operation 57F02) and at the City Wall (Operation 2). Cumulatively these tests indicate that at the end of the Leilan IuVNincvitc V period and at the beginning of the Leilan 11 period a major transformation of settlement occurred on the Habur Plains. Operation 2, we thought, might prove thai the City Wall was first built in the time of Shamshi-Adad. Who else would have been able to muster and control the labor required for the construction of a mudbrick wall 3.5 kilometers long, at least 15 meters thick, and at least 15 meters high? In the last days of the 1980excavation season, however, it was with considerable shock that we found ourselves against the City Wall excavating surfaces much earlier than those littered with "Habur ware" and tramped upon in the days of Shamshi-Adad. These earlier surfaces and City Wall construction phases are characterized by ceramics of the period that we designate Leilan II or the "Leilan" period, because it is the period when the site emerged to regional prominence. The ceramics associated with the first interior floors set against this wall, visible in the section drawing of Operation 2, are illustrated here. Operation 57F02 revealed precisely the same ceramic-stratigraphic phenomenon: The fust Lower Town occupaiion, set on virgin soil, was associated with the early Pcriod-h" ceramics. Through the Leilan III/Ninevitc V period, therefore, settlement at Leilan had not extended beyond the area of the 15-hcctare Acropolis, and conceivably was still smaller. Suddenly, however, at a time when Ninevite V ceramics had passed RlfHICAL APCHAEOIOGISI MARCH iw 21 from use and Leilan 11 ceramics had lust begun to be used, the settlement expanded sixfold, (mm 15 to 90 hectares, and the enormous City Wall was constructed. This kind of alteration in settlement is unlikely to have been a unique event. As geographers have long observed, "Cities do not grow up of themselves. Countrysides set them up to do tasks that must be performed in central places* (Jefferson 1931:435). The new information provided by the soundings at the City Wall and Lower Town sets the stage, therefore, for an examination of the regional forces and conditions behind this development. Two ques tions immediately require attention When did this expansion of settlement and circumvallation occur? What other developments, historical, demographic, or economic, might have occurred at this same period? When did the expansion of (he city occur? Two sets of data allow us to begin to clarify the relative and absolute date for the construction of the City Wall, when the rapid expansion of settlement took place. Four radiocarbon dates derived (mm a very large deposit of charred gram retrieved in stratum 20 of Operation 1 have now been analyzed. Two of these samples were sent to a laboratory in Florida and two were sent to a laboratory in Tokyo; the dates determined by these laboratories arc indicated in the chart of Leilan radiocarbon dates. Because these dates are derived from one large sample, they can be averaged m a fashion that allows us i 1,'v'mii.iii /. north sei r inn. Acmpahs northwest Operation I, which ts now 16 meters deep* was the tirst sttatinraphtc sounding made at Leilan At present the lowest esxasvted strata date to the Ubaid period mm no. oil miiuiH.iM mok k m. 5 Examples of Leiian III iNmevtu V; potury from Ml Leilan Operation I. Incised ware trim and body sherdii and painted ware 'complete vessels) The date of these intricate and very beautiful ceramic vessels has been a mystery for decodes. The retrieval of twenty-five successive strata characterized by such ceramics within Operation 1 now permit* us to date them to the period immediately preceding the anumvallation of Tell Leilan and unmediately after the Late Uruk period m northern Mesopotamia Reproduced from Clem v Schwanr. From Prrhiwory to History on ihc Hibur Plains (1982) Tell 1 a'il.iii Radiocarbon Samples UK Number Provenience Coniril Material before prevent ■x. t 2» UM18I6 L8045Q12 IOC-Uno. 1 Acropolisnortheast Building Level 0 wood 3895 i 80 2760 -2143 N-3900 L79 45V1022C14nc5 Acropoli* noriheast Building Level III wood 3330 i 80 1885 -1415 N-3901 L79 45TM I6C14na8 Acropulivnnrtheast Building Level III contaminated rejected NsW02 L7945TII8C14no.3 Acropolis-northeast Building Level III contaminated rejected UM-3101 L79 45VI013CI4no2 Acropolis northeast Building Level III contaminated rejected UM-1818 L80Op2 67C-|4na2 ■City VVkll- ptusc E wood 4320 ± 90 3355 -2665 UM 3098 L80Opl4IC-I4na9 Op J Stratum 19 gram 287Q x 130 I4IÜ- 790 N-3896 L790p 126C 14no 2 Op 1 Stratum 19 wixid 4980 i 80 3935- -3565 N-3897 L79Op]40C-I4naS Op 1 Stratum 20 gram 3970 t 85 2865- -2190 N-3898 L79Opl40C-14no.6 0p 1 Mi.Knill 20 grain 4070 t 70 2885- 2415 UM-1777 L790p 1 40C-14 no A Op 1 Stratum 20 grain 4090 i 70 2895- 2420 UM 3099 LSOOp I 40C-I4na 2 Op 1 Stratum 20 grain 4060 i 60 2880- 2410 L7VOpI45 c Ii m. ' Op 1 Stratum 34 grain 4210 t 85 3150- 2555 UM 1814 LSOOp 194C Una6 Op 1 Stratum 34 grain 4890 i 70 3875- 3395 UM 1815 L80Opl96CI4na7 Op 1 Stratum 35 gram • wood 4625 i 85 3655- 3055 UM-18131 L80OplA6C-14no6 Op 1 Stratum 38 grain 4735 i 110 3783- 3193 UM-1812 L80OplA40C14no.fi Op I Stratum 44 grain 4705 i 85 3775- 3173 UM-1817 LSOOp IC35C-I4no. 1 Op I Stratum 58 grain • wood 6580 i 100 5785- 5240 N«-~ TV vm ,«ht h*n4culurnn.«ml«. HWW Jcmum SwmbmNW Mjm. um Im.«4UM-3099 wt,ita ymplara wUbnMnrtinFkttbmd ~R*w> h< «ulm lwo vampir« bona *rnt id tch pint BIBLICAL AKCllAlOLlHitST MARCH ml Abort: North itrattgraphic section of Operation 2 Strata I and 2 are surfaces littered with Habur ware, while urate 3 throuth S are characterized by Period-II ceramics The foundation trench for the City Wall U itratum ° Below: Representative wheel-made pottery of Leilan Penod II iclrca 2$00-2DO0t-c, from arata 3 throuth & In Operatton 2. to reduce the standard deviation (the plus-minus figure that accompanies a radiocarbon "date"). This ^righted average date is 2673 b.c. ± 70, which means that the date of the original sample (short-lived grain) is 85 per cent certain to fall within 2755 and 2595 KC This date for stratum 20 in Operation 1 provides us with a re rm in us post quern (that is, the point after whichl for the construction of the City Wall |Weiss 19S3). But it seems clear that the extant surface upon which the City Wall now rests in Operation 1 was not the last surface deposited there. This area had been scraped and levelled prior to the City Wall's construction. How many intervening strata were removed cannot now be known. Probably, however, strata with ceramics similar to those now retrieved at Tell Mohammad Arab, across the border near Eski Mosul in Iraq, are to be situated between the last pre-wall strata and the construction of the wall in Operation 1 |Weiss 1985b). The date of the City Wall's construction, therefore, might be around 2500 b.c. A second set of dating evidence is comprised of the ceramics associated with the construction and first use of the City Wall (sec the section drawing for Operation 2, north sea ion, and the illustration of representative pottery). It is now quite certain that these ceramics are the same as those recently retrieved at Tell Brak. Tell Brak is a large, 43-hectarc site, located 51 kilometers southwest of Leilan, alongside the lagh-lagh River, another of the effluents of the Habur that |oin together near Hasseke to form the 'triangle," as the Habur Plains are sometimes called. Brak was first excavated by Sir Max Mallowan in 1936 and 1937, and until recently those excavations have served as the maior guide to the archaeology of the Habur Plains. Sir Max was fortunate in the time that he spent at Brak to uncover a very large mudbrick fortress, almost one SlBLIt U MM MMDil-.IM MiRdlINi Chronological Relationships Tell Brak i IU,I CH, ST, and Lcilan Op. 1 ■ Other Soundings K'riod 1 Habur ware 1 1-12 1 "Ur HI" í 1 2 "interregnum" 13 1 3-4 Late Agade II 14 5 reconstruction 15 destruction, levelling 6-late ED HI" building city wall gap* |levclling| Ilk 16-20 wtoo Illb 21-34 3100 Ninevite V Ilia 35-40 Laic Uruk IV 41-44 ■ I 45 Early Uruk V 52 ■ 52a ■ vib 57 58 Via Ubaid 61 -WOO 5200 , Halaf Tell Leilan Tell lays Acropolis-nonheast B-L- I-III 5TF02 1-3 Í 4 16 virgin soil Op. 2 1-2 9 city wall virgin soil Acropolis m-iv v-vi VII destruction VIII destruction IX Ninevite V [surface I -i V ilup' ihe ume as IcwU M Uuuugh R at Muhammid Aiah hectare in size, with bricks bearing the stamped inscription of Naram-Sin, Sargon's imperial grandson. Within the partially excavated fill of the fortress, Mallowan also retrieved a fragmentary votive inscription bearing the name of Rimush, Sargon's son. It is possible, therefore, that the fortress was even constructed prior to Naram-Sin. This imposing structure has been taken as unequivocal evidence for Sargonic control of the Habur Plains (Mallowan I947|. More recently, David Oaies, successor to Mallowan at Tell Brak, has retrieved portions of a large building adjacent to, but strati-graphically below, the Naram-Sin fortress, and he assigned it to the "late Early Dynastic" period. The excavators also suggest that this structure "had some official — politi- cal or military—character, and was not simply an indigenous phase in the continuous occupation of the city as a whole" |Oates 1982a: 67). This building was, in turn, de* strayed, and then rebuilt, prior to the foundation of the Naram-Sin fortress. It is entirely possible, therefore, that this building was destroyed by Sargon (Oates 1982b: I97|. The ceramic assemblage associated with BiniKAL A HC' HA tOLO(i 1ST'MARC H 19S5 2» Map ol the Habur Plaint, with modem isohytts lima that connect points of equal rainfall* drawn in Figures are in millimeter* this building is remarkably similar to the early Leilan-pcriod-II ceramic assemblage, the assemblage associated with the sixfold expansion of Leilan and the construction of the City Wall |I. Dates 1982; Weiss 19831 A pre-Naram-Sin date, and possibly prc-Sargon date for the City Wall at Tell Leilan is, therefore, now suggested by the Leilan radiocarbon dates, the relative ceramic chronology of Leilan ceramics and Mohammad Arab ceramics, and the building sequence at Tell Brak. If correct, this date may alter considerably our understanding of the origins of cities and civilization in Syria and Mesopotamia. Subir in the late third millennium B.c. After its probable date, the most significant feature of Leilan's cir-cumvallation, and the most important feature for understanding its genesis, is the observation that Leilan was not unique. Surface collections made by the Tell Leilan Project in 1979 at lell Hamoukar, 46 kilometers southeast, indicate that this 90-hectare settlement was also occupied during the early Leilan n period, and in fact was already a very large settlement in late Uruk times. Similarly, surface collections at Tell Mozan, 43 kilometers northwest of Tell Leilan, indicate that this site too, was probably a large early Leilan-n settlement, which continued to be occupied in Leilan-I times. Hence the circuravallationof—that is, the City Wall construction around— Leilan allows it to be understood as a regional phenomenon, within a specific portion of the Habur Plains: the extremely fertile area of the plains that receives more than 400 millimeters of rainfall per annum. Similar sites appear across the border in Iraq, south of the Jebel Sinjar and near Tell Afar. Another site in the region, Tell Brak, appears to be a different kind of settlement, however. Tell Brak might be understood as one of a class of settlements, occurring in a variety of historical and geographical contexts, sometimes labelled "gateway cities." Such settlements characteristically control the entrance into a region, command the connections between that region and the "outside world," and are often located eccentrically at one end of the region, sometimes at the border between regions defined by different kinds of agricultural production (Burghardt 1971), These characteristics hi the geographical, climatic, and cultural situation of Tell Brak, as we know it, quite well. Brak is located at the southern extremity of the Habur Plains, quite distant from its most productive centers. A glance at the map displaying mean annual rainfall on the Habur Plains shows that Brak M BIMJCAL ARCMAEOUK.IST'MARCII IWSB The Search for Shamshi-Adaďs Capital City receives only 289 millimeters oi rainfall per annum, iust enough rain to generate a dry-farming |that is, (arming th;it depends on rainfall and doesn't utilize irrigation! cereal crop. This location is markedly distinguished from that of the three large-walled settlements (Tell Leilan. Tell Mozin, and Tell Hamoukarl that are each much larger than Brak, and situated almost equidistant from each other within the dry-farming belt at points that apparently maximize access to cultivable flatland. Local tonally, Brak controls the entrance into the Habur Plains provided by the Habur River itself as it passes through the "gates" of the kbel Abd al-Azziz and the lebel Siniar This situates Brak directly between the area of high-rainfall dry (arming that characterizes the area of gently rolling plains around Leilan, Mozan, and Hamoukar and the irrigation-dependent regions of the south. The cultural inventory of Tell Brak in the late prehistoric and early historic periods may also be understood in terms of the settle mcnt's "gateway" status, since it seems to have mam* elements of southern culture, while also apparently preserving indigenous elements foreign to southern Mesopotamia (Amiet 1983: 51). The Habur Plains, entered through Tell Brak, were known to the third-millennium dynasts of southern Mesopotamia as the land of Subir Later, in the second and first millennia »c. the region was called Subartu, and came to include much of northern Mesopotamia |Gelb 1944; Edzard, Farber, and Soli-berger 1977: 146- 47). Beginning in the late Early Dynastic period and continuing through the Satgonic period, southern Mesopotamian rulers repeatedly claim to have conquered or subiugated Subir, a claim that until now has lacked historical meaning But the evidence tor large third-millenrnum-B.c. cities on the Hahur Plains changes our evaluation of these sources. Cities such as Rising from stdl unidentified roots, perhaps among recently sedentar tzed Amonte speaking peoples from the Habur Plains. Shamshi-Adad (whose name means *My sun is the god Adad") briefly transformed the political and economic landscape of north-cm Mesopotamia in the last years of the nineteenth century St., just prior to the accession of Hammurabi in Babylon. Id on unexplained flash of historical stardom. Shamshi-Adad managed tosubtugate the towns and cities of the northern plains and extend his imperial bold across all ol northern Mesopotamia from the Zagros Mountains re the Euphrates Rivei Quickly seizing control of the upper Tigris River area, including Arthur itself, he deposed local dynasties at nodal control points (Ekallatum on the Tigris and Mori on the Euphrates), and then installed a son at each city as ruler Shamshi-Adad then established a new capital at a place that he called Shubat Enlil ("The Residence of Enlill Thereafter, dynastic alliances were created with distant city rulers, tribute and gifts were extracted hum subiect kings, long-distance trade relationships were reestablished across Mesopotamia and into Anatolia, and a hierarchy of regional control, descending from Shamsht Adad. was extended across the nonhcrn dry-farming plams. No city ruler could successfully challenge the armed forces of Shamshi-Adad within this region during his reign of less than thmy-five yeara|18l3-l782 sr.). In spite of his apparent administrative and organizational capabilities and the strength of armed forces loyal to him for still unknown reasons, the disintegrative and centrifugal forces that characterized the plains of northeastern Syria dnd northern Iraq eventually proved too fractious for the bonds that tied Shamshi-Adad's empire The difficulties included independent and widely spaced cities with extensive tracts of cultivated plams. large seasonally migrant forces of pastoral nomads moving between the irrigated tracts along the Euphrates and the rain-fed Habur Plains, and persistent challenges from the centralized r v.-i. ■'< southern Mesopotamia. Par tieularly vulnerable were the outposts of the empire, such as Mori, where the incompetence of Shamshi Adads son, Yosmakh-Adad, only made matters worse. In the ancient Near East, as in more recent Europe, diplomacy was sealed by marriage. Yasmakh Adads personal affairs, however, teem to have made it difficult for Shamshi-Adad to preserve his imperial alliances. Hence this lettct from Shamshi-Adad to Yasmukli-Adad Did not tin hirmcr king* . esiab Iish their tpouscs in ihe palace' YakhdunLim. (howcverl. honored his consult - placed his wtfr to the side, and moved her into the desen Perhaps, m the same way, you are planning to place the daughter of tihl Adad tin lunx of (Jjtnj m the BIBIIC4L AlCHAFOinGIYt'MARCH l*»S 1" desert Hci laihct will be pawly disturbed by this Thi» U not good' There are many rooms m the 'Palm Tree* Palace Ui i hem chooac a room tor her inert, and let hct stay in that nxim Do n make hci dwell tn the desert. (AO IMfl In Sasson 1973: 76) While ('.. ■ Man archives relate the detail* of imperial rule across the northern landscape, we still have vet to understand ihc origins ol Shamshi-Adads rule, ita developmeni, and its eventual crash The crash, however, was dramatic and conclusive The death of Shamshi-Adad was a mam Mcsopoumian event even celebrated as the name ol (he year after which it occurred. (Pot a discussion ol the Mcsopotamian calendar, particularly that of Mart, tec Sasson 1984: 249- The i .1-1 Days ui Shubat Enlil In the tumultuous two decades (hat iollowcd (he death ol Shaimhi-Adad, (he princes and kings ol the city stales on the Habur Plains ransacked and pillaged Shuhat Enlil and fought with each other over its spoils Some of these postmortem activities can be Iollowcd quile clearly in ihe graphic, detailed documentation provided by numerous letters within the Man palace archives and two letters from the palace at Tell alRimah. The chart shown here is one ordering ol the available documentation, and although u cannot presume (o be totally accurate, u allows us to follow some ol (he movement ol armies back and forth across the Habur Plains for almost two decades. After the death ol Sharmhi-Adad one ol his sons, Ishmc-Dagan. was able (o briefly preserve the northern empire and hold off armies from the south-cast, from along the Diyaia River (Eshnunna) and southwestern Iran (Elam) Bui Ishmc-Dagan was shortly defeated, and (he northern capital of Shuhat tnlil was seized by a former Shornshi-Adad officer. Turum-natki. (he ruler of an unmentioned but probably dose-by cily, allied himself with (he forces of Zimri-Lim who had regained the Man throne and decided to establish his own order on the fertile Habur Plains {tee A on (he chart). Zimri-Lims vassal. Yassi-Dagan, now controlled Shubai Enlil, bui a threat from Qami-Lim was already perceived n QarruLim. ruler of the nearby town of Andanq, apparently beat Zimri-Lim to Shuhat Enid, and was able (oplunder the grain of thecity (C). Qami-Lim (hen wined forces with (he "man of Eshnunna." and established himscll at the city of Apum with Turum-natki. According to (his docu-mcnt the son of Turum-natki was (hen appointed ruler of Shuhat Enlil, hui (he document was one of the earliest retrieved from the excavations at Man and was never hilly published |D1 The ruler of Eshnunna (Ibal-pi-El) then apparently lurncd his attention (owards Zimri-Lim The latter sought (he help of yet another ruler, Khatnu rapi ol Karana. who recaptured Shuhat Enlil from the king ol Eshnunna, pd-lagcd (he city a second time, and walked off with his booty without sharing any with Zimn Lim (E and Fl From Tell al-Riroah, a small kingdom east of the febel Sin tar in northern Iraq, the following letter records the jealous exchanges among (he looters of (he fallen capital: Speak to ■>•■■-." i i (hus Bullu-Isbtar your brother "You art bring mg "ui Zimn-LimV »hare from the spod that you are ukinjc rrom Shuhat Enlil, but why are you still keeping hw share' Will he iu« liHik oof" (Halley and oihen lv7ft: number 5) A temporary coalition of other wise contending iorccs (Eshnunna and Elam lo the southeast and lshmc-Dagan at Ashur) (hen attempted to defeat Zimri-Lims ally, Razama |G|. At some later point, Zimri-Ltm regained control ol Shubat Enhl and installed an Elamilc by the name ol Kunnam (or Kunnamal. as the city's governor. At the same time Zimn-Lim had apparently already organized a (lercd system of control, such that Kunnam was actually liable to the king of Apum, Khayaabum. who in turn was liable to Zimri-Lim. Railing against this vassalage. Kunnam protested to Zimri-Lim for status equal to that enioyed by Khayaabum (H and I Shortly thercaltci yet anothei local ruler gained control of Shubat Enlil. A um rum. who succeeded Qami-Lim as the ruler ol Andanq (| through Nl A military officer and emissary of Aumrum (Larwala Addu eventually took charge of Shubat Enid and from this base proceeded with 3,000 soldiers (o attack Khayaabum ol Apum, .'rni-Lims vassal in the Habur () Thereafter, there is no record of Shuhat Enlil In time, it was forgotten Where h Shubat Enlil' It was only with (he recovery of the Mari archives centuries later that (he existence of Shuhat Enlil was once again known and scholar* began suggesting sites as candidates for the ancient capital. The distinguished As syriologist Francois Thureau-Dangin with Georges Dossin, the future doyen of Man studies, proposed that Shubat Enlil was simply anothei name used in the Man (cxts for the city of Ashur (Thurrau Dangin 1937). Although this identification had its long-term, vocal supporters, such as lulius Lewy (1953), it was challenged early on by the redoubtable Benno Landsbcrget, who suggested that the site of Chagar Bazar is ancient Shubat Enlil jLandsbcrgcr and Balkan 1950|( Undsbcrgcr was followed in (his by Albrcchl Goetze of Yale Uniwrsity (Goetze 1953). The issue was one of several, substantive as well as personal, which divided the leading Assyriologists of the (ime. Chagar Baur. Before (he outbreak of the Second World War, Max Mallowan, who had been Sir Leonard WoollcyS assistant at Ur, was forced to abandon the temtory that is now Iraq in the fact ol the political and cultural inroads thai German political agents were cuuing within Iraqi official circles (Up to that (ime British agents literally controlled archaeology in Iraq I Stdl wishing, however, to pursue his archaeological research, Mallowan retreated across the border onto the Habur Plains in the French mandate of Syria and proceeded to undertake his now famous excavations at Tell Brak and Chagai Bazar (Mallowan 1947). His colleague C. |. Gadd from the In addition to thou dud In the than, the following source* were consulted fn ptepanng thu information Anbar 1978 and 1981. Sd.son I97& and Sauren 1971. mam \i ku mimi».iw «i"i ii ins Shubat Enlil Afler the Death of Shamshi-Adad Oate'Chronologt Urn mm nt.it of Shubat tnlil_ 1781 ».c. Death of ShamshrAddd's son Ishnic-Dagiin boasts to hi* ShamshiAdad brother Yasmakh-Adad that he holds Elom and _^_ Eshnuruu |ARM IV 20) _ 17?2i-C- ZimnLim year 1 1761 Mt Aumrum. last regnal year D Zlmn Lim orders Turum natki of i; city) and Khaya-Sumu of llonsura to join forces with Sima-ila-khancm of Numkhs to liberate Shubat Enlil hi -in Samlya, (renegade'] servant of Shamshi-Adad, who hold* the city It), Zlmrl Lim ha* ordered spies into the city but thty have not returned. (ARM X.5I Yassi-Dagan holds Shubit Enlil for Zimri-Lun but Qami-Um of Andanq is "rumored to be passing through to Shubat Enlil "(ARM II IM\\ Qami-Llm of Andanq. plunders the grain of Shubat Enid. [ARM XIY109) Qarni-Lim and the 'nun of Eshnunna* |lbal pi Elf*| art m Shubat Enlil Qarni-Lim and Turum-natki arc entrenched at Apum. Tummrutki» son is appointed the ruler of Shubat EnlU \t\. (lean 1938] Ibal-pi El of Eshnunna moves from Shubat Enlil toward* Zimri-Lim's territory at Mari. Zimri-Lun requests help from Khatnu-rapi ot Karana. (Dallcy and others 1976. Rimah letter 21 Khatnu-rapi retake* Shubat Enlil from IbalpiEl, take* booty remaining from the first pillage by I'm. i" I ! and Qarni-Lim, and doesn't share with ZimnLim. (Dalley and others 1976, Rimah letter 5) ______ Eshnunna. 1 lam and Ishmc-Dagan Kim forces to defeat Razama (ARM VI.27, D.2S) H. (Elamitc* ukc control ot Shubat Enid.]_ L Kunnam, the "man of Elam ," wntes to his lord Zlmr I I mii 'Khaya-abum |of Apuml i* the 'son* of ZimnLim, but I, I am not his |Khaya-abum'*l 'son' I want to meet with my lather ' (ARM XW1021 Aumrum of Andanq plots to raid Zimn Lim's territory when Zimn-Lim marches lo help Raiama .(ARM VLSI) K. Ar .n: i n ii i ■ wants to enter Shubat Enlil, but Kuniunu won't leave (ARM XIV. 1011 'The city is the city of the sukkaJ |Kunnonu'|' (ARM XIV104,_ 1760 B,c. Hammurabi conquers Man M.r Shubram is the shjpnum official of Shubat Enid under ZimnLim (ARM H. 109 and X 84) N. Atamrum controls Shubat Enlil His Qutu-troops .in within the city. ; ARM II -11 Riiuault 1970:48,77)___ Q. LawalaAddu.theigbiamuttxm-commander (and emissary of Aumrum), leads 3,000 troops from Shubat Enlil to attack Khova-ahum |ARM!LI3SI _ _ British Museum published a preliminary analysis of the cuneiform ublets retrieved from both sites shortly after the conclusion ol the excavation* (Gadd 1940|. Codds report included mention of a document recording grain shipments to Shubat Enlil Hence Landsberger'* proposal thai Chagar Bazar is Shubat Enlil, a notion thai persists to this day (Kupper 1973-45). Scant attention was paid to Sidney Smith, the eminent British Assynolo-gist, who observed that other place names as well occur among the documents mentioning Shubat Enid, thus making it unlikely ihat it is the an cient name of Chagar Bazar (Smith 1956* 36). In his memoirs, published only a few years before hi* death, even Mallowan felt obliged to emphasize the obvious with regard to the Chagar Bazar identification: But in my opinion thi* (identities-non! is wroryr. because irne ublet records the dispatch ol supplies to Shubai Enlil - not received by it. and moreover our sue teem* insufficiently massive and important and not Mraicjpcally placed for the i.ri capital which probably lir« somewhere in the district noi tar off (Malt-wan \979: I2_| Tell Lelum. It was the Assynologist Margaretc Falkner who picked up Emil rbrrcrV and ultimately Max Freiherr von Oppcnheim's and Hormuzd Ras-sam's, mention of Tell Leilan (see accompanying sidebar nn 'Rediscovering Tell Leilan") and fust connected the site with the missing capital of Shubat Enhl (rtlkner 1957:371. At almost the same time Rarthel Hrouda. who was then a young archaeologist working with Anton Moortgai ol Berlin and who was able to assess the significance of surface archaeological observations, also suggested that Leilan could be the missing capital (Hrouda 1958). When new documentary evidence was brought forward with the cuneiform "itineraries" they too were found to prt*ent routes that matched the avid-able archaeological facts suggesting the identification of Tell Lctlan with Shubat EnlU (Hallo 1964). Tell Brak. Over the years other sugges tions lot the location of Shubat Enlil have been made. Tell Brak - a all imposing site of 43 hecures, whose an "imn M AKCHAlOIiH.IVT MARCH 10*5 H In 1978 the Yale expedition began its work at Jell LeUan by sutwying the site In the fore-of this photograph, taken from the west. Mark Kross of the surveying team Is seen king The Lallan Acropolis it visible in the background academic world its provenience was said, by its dealer, to be the site of Amuda, west of Kamishh |van Liere 1957). The site of Amuda has been identified with Urkish in the archaeological literature ever since. Two surveys of the sue by the Tell Lcilan project, however, have failed to retrieve sherds of Lcilan periods III 11, or 1, although nearby Tell Mozan, now being excavated by M. Buccel-lati seems to have each of these. Still a regional center in Zimn Lim s struggle for control of the Habur Plains after Shamshi-Adads death, Urkish was located just three caravan stops west of Shubat Enlil iSasson 1973: 74, Hallo 1964: 651. Assynologjsts have reasoned that the "royal titulature" of the Human rulers, referring to the cities of Urkish and Nawar, "groups two cities distant from each other in order to designate the entirety of the land of" Subir (Sollbergcr and Kupper 1971: 128). Dependent, therefore, upon where one locates Nawar, the land of Subir controlled by late-third-millennium Humans may have been quite extensive [Hallo 1978: 17). It remains unlikely, however, that Nawar could be as distant This 'Human" foundation peg with a cast bronze lion served as a temple foundation deposit tor Ttth-atal of Urkish The date of the lion, and it* 'sister' in the Louvre, has (vrn much debated but certainly falls wtthm the hit quarter of the third millennium nc It is II. 7 centimeters high and 7 9 centimeters wide Courtesy of the Metropolitan Museum of Art. Purchase, foseph PuUizer Bequest. I*M -M ISO from Urkish as the Jcbel Hamrin or the Zagros Mountains, and a location upon the Habur Plains is probable jARM 2:57). Historical geographical problems will always plague ancient Near Eastern research to lesser or greater degrees. Very substantial gains seem close by, however, in a region that until recently, and in spite of years of research, was virtually unknown. But another, and perhaps more substantial, contribution remains to be made by archaeological research on the Habur Plains for the genesis of third-millennium urbanism here, and its trafectory through the early part of the second millennium, remains to be delineated and analyzed. Postwar archaeological research is now entering its second research phase on the plains of Syria and Mesopotamia with research horizons considerably more extensive than those of its predecessors. The dry-farming plains of northwestern Syria, extending from the Amanus range south lo Aleppo, Tell Mardikh, Hama, Horns, and Qatna, present themselves as one region of high rainfall and high agricultural production with its own developmental history coming into conflict with the irrigation-agriculture southern regions around Man and Sumer in the late third millennium. Similarly, the Habur Plains, long known from thtrd-millcnniurn documents recording the conquests of southern dynasts, and famous as the most productive cereal agriculture region in Syria and Mesopotamia, apparently also experienced sudden urbanization in the third millennium. The inevitable conflict with southern forces, however, may have curtailed this development, as it did in the northwest. The cuneiform record for late-third-millennium developments in this region is sadly laconic, and the extensive archaeological exploration of such settlements is iusi beginning at Tell Lcilan and other sites. Ill Bill U AWHAICHOt.tf.T MAM H l«M 11 The renewed attempt by the forces represented by Shamshi-Adad to centralize control of the Habur Plains may indicate that the region's productive strengths and organizational potentialities were not diminished, continued to emerge and dominate the plains at permissible junctures, and again threatened the irrigation-agriculture centers of the south. This may explain why Shubat Enlil was no longer occupied and "Shamshi-Adad" was just a name on little pieces of mud when Hammurabi returned to Babylon from his last campaigns against Subartu. Conclusion Archaeological and historical documents are by their very nature partisan sources that must be evaluated in the light of our own intellectual biases, as well as the biases of the sources themselves. It has long been recognized that the history of Mesopotamia that we have been retrieving, recording, and interpreting is mostly the history of southern Mesopotamia observed through excavations at southern sites. At Tell Leilan, however, we have before us another source for the early history of the ancient Near East: an important city in the heartland of Subartu, the "other Mesopotamia." For the years ahead, the Tell Leilan project has now set the stage for the investigation of a formidable array of historical problems: the origins of cities and civilizations on the Habur Plains, the ancient history of Sumcr's rival Subir, the interaction between pastoral nomads and city-based powers, and the history of Shubat Enlil and Shamshi-Adad's northern empire. Archaeology, perhaps the only discipline to presume to study the long-term history of human societies, will be put to the test. Acknowledgments The Tell Leilan Project's held epigraphcrs, Dr. B. R. Foster |Yale University', Dr. Daniel Sncll (Uni- versity of Oklahoma) and Dr. Robert Whiting (University of Chicago), copied the seal impression inscriptions from Leilan 1979, 1980, and 1982. Ms. Lorraine Ferguson (Yale University) chief draftsperson for Leilan 1979 and 1980, drew the ceramics presented here. Dr. Nancy Leinwand drew the seal impression designs from field drawings and macrophotographs. The Tel! Leilan Project, sponsored by Yale University and the Metropolitan Museum of Art, has also been supported by grants from the National Science Foundation, the National Endowment for the Humanities and the Hagop Kevorkian Foundation, and the generosity of Barbara Clay Dcbevoise, Betty Starr Curnmin, Roger and Barbara Brown, lonathan Rosen, and Daniel Foster. Note Pans of this article are revised and expanded from two aniclcs by the author that appeared in the Yale Alumni Magazine and journal (Weiss 1981 and 1984). These are used with permission of the editors. This article also includes materials presented in Weiss 1985a, b. All excavation photographs are by the author. Bibliography Al-Khalcsi, Y M 1978 The Court of the Palms: A Functional Interpretation of the Man Palace. Scries: Bibliotheca Mesopotamia s Malibu, CA Undena Publications. Amid, P. 1980 Art of ihe Ancient Neat East. New York. Abrams. Amiet. P., editor 1983 Aupaysde Baal etd'Astarte. Paris: Muscc du Petit Palais. Anbar, M. 1978 Review at The Old Babylonian i.ir ■'■ I ■ from Tellal-Rimah by S. Dalley and others. Bibliotheca OrUntahsiSIS-4:208-16. 1981 I ■■■.<■■'.: ion ,i mre Zlmri-Lim, roi dc Man. et Qarni-Ltm, mi d'Andariq loumal of Cuneiform Studies 33/1: 48-51. ARM 1942- Archives toyales de Man Putt Vau] Ceuthnci and Imptimcnc nauonalc Bra noes, M 1968 Uatetsuchunxen :ur komposition der sttftmosaiken in der Pfetlerballe der Schicht IVa in Uruk-Warka. Scncs: Baghdader Mitteilungen, Beiheft 1. Berlin: Mann. Buren, E. D van 1945 Symbols of the Cods in Mesopo- tamian Art. Series. Analects Ori- entalia 23. Rome: Pontifical Biblical Institute. Burghudt, A P. 1971 A Hypothesis About Gateway Ones. Anna's of the Association of American Geographers 61:269-85. Calvct, Y. and others 1976 Liisj rapport pre liminaire sur U sixicmc Campagne de fouilles. Syria 53: 1-45, Charpm. D 1983 Temples a dccouvrir en Sync du nord. Xra mdon Warminster Goctxc, A. 1953 Ad Olj Babyloňan Ittncrary hurnal of Cunelfoím Studia 7: 51 - 72. Hall.H R. and Woalley, L 191' AI*Ubaid UrExcawtiony Volume 1 Oxtord Oxlafd Univcrsitv Píru Halin, A. 1955 Die Heihgttimei des Gottes Astut und der Su+Samaliempel tn Assur. Seite*. Wissenschaftliche Vcroncnt-hchungcn der Deutschen Orient Ceaellschah 67 Berlin Mann Hallo. W W 1964 The Road io Emai loutnal of Cune- tfotm Studiei 18. 57-88 1978 Simurnim and thc Hurrun Fromicr Revue Hnuté et Auamque 36: 71-83 Hawfcui*. I D 1976 Thc Insciibcd Seal ImpressKwis. Pp. 247 - 55 m Thc OJaf Kabvhman Tableti trom Teil al Finnin by S Dallcy and ocher» London British School ot Archacouxry in u■.; Heinrich. E 1982 Die Tempel und Heiligtümer im Alten Mesopotamien Serie» DcnkmÜcr Antikri Arehiickiur 14. Dem sehe» Archäologische* Institut Berlin: de Cruyier Howard-Carter. T 1983 An torcrpietat**! ol ihr Sculptural DeconrUon at thc Sccond Mdlen ".um "itrnplt *t Teil al Rimah Iroq 45/1:64-72. Hmuda.B 1958 WaišukaniU. Urkii Subai Enhl Mitteilungen der Deutschen Orient- gesellschatťK: 22- 45. 1971 Vorderasien. I Munlch; Beck 1984 Die Kunst und Kultur der Hurritcr in neuerer Sicht Pp 6.1-69 in Mededelmgen van de Ktsntnkliike Academie mor Wetenuhappen Lei teren en Schone Künsten van Belgie. Klane der Lettcrcn. Uanranjt 46. I. Huot. |.L.. andothers 1981 Larsa ei Ouetlt. Paru: Edition* techerche sui les civil uation*. Memoire na 26. laruz, K. i960 DicKuhunc«edcrK**oicn Anrhrofxu 55: 17-84 kan, C F 1938 Ezcerpu de la Corrc*pondance de Man Revue de* tludet temttiaues 1938: 128-32. 1939 E*cerpta de La Corit>p«>txLif>cc de Mari Revue des etudes semitu/uet 1939:62-69. Mtersnn. M. 1931 The Distribution oi thc World» Citv Folks A Study In Compaiativc Civilization Geographical Renew 21 446-65 Kammcnhubcr, A. 1976 Hi*iori*ch-Gcographischc Nachrichtcn am dei alihuni*chen LfbciliclriuiiK . . vor dem Einlatl dcr Gutter Pp 157 - 247 in Wm tchatt und Gesellschaft tm alien .-i ■. edited by I Harmatu ■in.!«. Kornoroczy Budapest Kudu 1977 Die Atki irn Vbrderen Onent und die hutnritchcn VVohnsitze dcr Hurnttt Ottentalia46/1: 129-44. Kupper, |.-R. 1973 Northern Me*opotamia and Syria Pp 1-41 in The Cambridge Ancient History, 2>'l. third edition. Cam-hndjee Cambridge University Pre** LacssOe, | 196.» tvoplr of Ancient Assyria. London Routlcdrt and Kcgan Paul Landkhcrprr B. and Balkan. K 1950 Die Inschnh dc* AMynscben Kimi** luium Belleten. Turk Tonka Kummu 14 219-68. Layard. A H 1849 Smewh and its Remains London Win Murray Lcnzcn. H. | 1955 McMipotainlochc TcmpclanUgcn von det Fnihicit hi* rum zweitcn Jahr laufccnd Zeiischnh Hit Assynologie, Ncuc Mr 17: 1-36- Lewy.l 1953 SubatEnhl Annuaue de Tlnstituie de Phtlologte et d'Histout onentales at slave* 13: 293-321. toreti, O 1969 leitc au* Chagar Bazar und Tell Rrak. Tcil 1. Series: Alter Onent und Altc* Testament: 3/1. Neukirchen-Vluyn: Bution and Berckcr KcvcUer MalUwan, M 1947 Excavation*at Brak and Chagai Bazar Iraq 9 1979 Malkiwan Memoirs London CoUuu Martiny.C 1936 und astynschen lempelbau. Scncs Ahhandiunaen tur die Kundc dc> Mofuj'iil in : ■ 21J. Lcipzix Dcut*chr MiifpnUndische GeselUchait Matihiac, P. 1975 Untie ct devellopemcni du temple dan* la Syne du bronze moyen. m Le Temple et le Culte. ^•edcrland* HittoriKhAnhacoloRisch lnuituut ic Isunhul. 37. Moon gat A 1969 The Art of Mesopotamia New York F'haidon Oatcs.D 1967 The Excavation* at Tell al-Rimah. 1966 Ira,/ 29/2: 70-96. 1982a TeUrlni.Pp.62-7luiftrfVBVtfncrf Mesopotamian Discovery, edited by I Cunis. London British School 01 Aichaeology in lia^ 1982b Escavatiim* at Tell Brak, 1978-1981. Iraq 44:187 - 204 Oaies, ]. 1979 Babylon. London Thame* and Hudson 1982 Some Late Early Dynamic m Potter? from Tell Brak Iraq 44: 205-19. Oppenheim. M. F. son 1899- Mwn Mntelmeet nun Persischen 1900 Golf. Berlin Reimer OT8UC, N. 1980 Seal Impression* hum Accmhot-uk. Pp. 61 -99 in Ancient Art in Seals, edited by E. Pmada Princeton: Piinceton Universliy Pre** Pardee. D. and Glass, J.T 1984 Literary Source* lor the History ol Palestine and Syria The Man Archives. BiMtcal Archaeologist 47 88-99. Parrot, A 19.W Les kiuilWs dc Man. quatneme wmpaane |hiver 1936-37) Syria 19. 1-29. 1939 Les fouille*de Man. cinquKme Campagne. Syria 20: 1-22. 1958 lepaiais Ptintures murales Serie»: Mission archColnftiiiuc de Man 2 InMiiut Franca» dAichfolocic dc Beyrouth Bihlu«heque arche-ülopque et hiMoriquc 69 Pan* l*aul Getrthner Parrot. A. and Nougayrol. I 1948 UndocurnemdckmdationhourTitc Revue dAssynologle 42VI -1 1 - 20. Poidebard, A. 1934 La Trace de Rome dans le dtsen de Syrle- Paris: Geuihnci hmeate, N. 1973 Appendix 1 Tell Taya Tablet*. 1972-73 /1*135 173-7S Ranam, H 1897 Asshur and the Land ofNimrod. New sbrk: Eaton and Main* Rouault.O 1970 Andanq ct Atamrum Revue dAssynologle et dArcheohgie OnentaleM 107-18. Sasson, |. 1973 Riogiaphical Notice* on Some Hm-.il Ladies from Man lournal of Cune Ihm Studies 25: 59- 78. Sauren. H. 1971 lantronisation du roi cn luael a la lunuere ilune lettre de Man Onentalia Lovaniensia Prnodtca 1l 5-12. Vhwanz. G-M. 1981 From Prehistory to History on the Habur Plains The Operation I Sounding at Tell Uilan Doctoral dissertation, Yale University. Pit" ii m ksk HAlOIIM.Ivr HARTH ims it Smith, S. 1956 Uriu «Dil i .Anatolian Studies 6:35-43. Sollbcríer. E., and Küpper, l.-R. 1971 Inscriptions loyales sumertennes et akkadienet. Parts. Lcs editions du Cert. Siromrnenger, E. 1962 fünf*lahrtausende Mesopotamien. Munich: Hirmer. Thureau-Dangin, F. 1937 lasmab'Adad. Revue d'Assyriologu et dArchéologie Orientale 34: 135-39-Ungnad, A. 1938 Dawnlisten. Rrallexikon der Assyrtologie. 2: 131-96 van Liere, W 1957 Urkish, centre religleux hourrlte. Annates archéologiques de Sýrie 7: 91 -94, 1963 Capitals and Citadels of Bronze-Iron Age Syru. Annates archéologiques syrUnnes 13:107-22. van Liere. W, and Laufray, I. 1954- Nouvelle»prospectlon»dan>la 1955 Haute lezireh syhenne Annates archeologiquei synennes4-5'. 129 -48. '...n:' L., and Pardee, D. 1984 Literary Sources for the History of Palestine and Syria: The Ebla Tablets. Biblical Archaeologist 47: 6-16. Weiss. H 1981 Capita! City of a Lost Empire. Yale Alumni Magazine and foumal 45: 20-26. 1981- Tell Leilan. Archiv für Orient- 1982 • . - 226-29. 1983 Excavations at Tell Leilan and the Origins of Nonh Mcsopoumun Cities in the Third Millennium ic. fti/eonem 9/2:39-52. 1984 History from Bits of Clay Yale Alumni Magazine and foumal 47: 22-26. 1985a lei) Leilan and Sbubat Enid. Mori: Annales de Recherche* Inurdiscipli-naires 4. (CoUoque internationale du Centre nationale de recherche seien-tifaque, no 620, Strasbourg, 7/83.) 1985b Review of Firry rears of Mesopotamian Discovery, edited by |. Curtis.. ■ ■■..* r j .■1 of the American Oriental Society 105/2. Wilhelm, C. 1982 Grundzage der Geschichte und KulturderHurriten. Darmstadt: Wissenschaftliche Buchgcsellschaft Woollcy, L 1939 t/r Excavations, volume V: The Ziggurat and Its Surroundings. New York: British Museum and University Museum. Philadelphia. 1984 Back Issues of Biblical Archaeologist tVdinUi Biblical Archaeologist •"■ 1 ■ I - —■ ■! 'I *-- BMcal Archaeologist lune M V — '■ I SI Biblical Arclmeolopst Biblical Archaeologist 'When reading the bibliographic r»uv by Viganc. keep In mind the new cuneiform diicovrric* reported by Matthiai presented here lot the Inn time to English readers* [Match 8d. Ebla\ 'Although everyone bad told Zimrl-Llro to ignore the prophecy, he kept returning toll. Addu-dun waa competent and not prone to hysteric*, and the drum ahe reported wis bloodcurdling * l/une 84 Man) Coocemratin* on the synaapeue, rcuthiocum. and Christian chapel of DuiaEuiopoa, Gates illustrates how ihe an and architecture oj diHeient religious faith* share a common heritage1 \Stpttmber M. Dura Eumpoi) 'rounded In 1900. the American Schools of Oriental Research has participated In most ol the important archaeological events in the Middle Eait. Including thedlKOveryandpublicatiunuf the Dead Sea Scrolls.' (December 84. ASOR at 85) Plr**c send check or money order, payable to the American Schools of Oriental Research, for 16.00 postpaid |U$l tor each issue ordered Be certain to tpecify which tsauci you want Please allow 4-6 week* tor delivery ASOR Publication! P O. Bos HM.. Duke Station Durham, NC 27706 Iii Iii K M AKCHAtOlOGIST MARCH 198$