More to Tell About Billa!

: Despite being one of the most prominent and best-documented northern client states of the Ur III state, the city of Asimānum/Šimānum remained unlocalized. Here, we demonstrate through both historical-geographical and philo-logical argumentation and archaeological evidence that it must correspond to later Middle and Neo-Assyrian Ši baniba, modern Tell Billa, northern Iraq. Building upon this finding, we propose a new history of Tell Billa during the Early and Middle Bronze Age.


Introduction
The site of Tell Billa (more properly Tall Billā in Arabic), near modern Baˁšīqā, Iraq, is best known for its Middle and Neo-Assyrian (MA and NA) occupation layers, from which a sizeable textual corpus emerged during the excavations of Speiser in the 1930's.1 This, in turn has permitted Billa's identification as MA Šib/manibe (and NA Šibaniba).2 Yet, the expedition's (unfortunately only scantily published) archaeological findings indicate that this large mound was already occupied from the Late Chalcolithic (LC) period onwards, which poses the question as to the site's identity in cuneiform sources prior to the Middle Assyrian period.
In turn, recent advances in the understanding of the historical geography of northern Iraq, most prominently the identification of the site of Bassetki (Arabic: Bāsitka) with MA Mardamā(n)3 (earlier Maridabān; Pfälzner/Faist 2020), have raised once more the question of the localization of the important Ur III-period city of Šimānum (also frequently attested as Simānum, Ašimānum, and Asimānum), very likely to be equated with the Sargonic-era (OAkk) city of Asimānum.
Here, we contend that an identification of Asimānum/ Šimānum with later Šibaniba, modern Tell Billa, simultaneously resolves both issues, providing the archaeological evidence prior to the Middle Assyrian period at Billa with a historical context, and the Early Bronze Age (EBA) historical city a suitable archaeological site. This paper first presents a historical-geographical assessment of the data and an argument for its localization at Billa, before presenting pertinent surviving archaeological evidence from the tell. The final part of the paper provides a proposed new history of Tell Billa and the surrounding region from the Early to Middle Bronze Age (MBA) in light of these findings.
Turkey, Šināmum is now certainly located at nearby Üçtepe, permitting closer scrutiny of this argument.9 Other commentators such as Lafont (2009, 3, fig. 2) and, most recently, Michalowski (2020, 719) have located Šimānum somewhat north of Cizre, Turkey, perhaps to reconcile the northern and southern interpretations. 10 With Mardamān now incontrovertibly at Bassetki,11 and Talmuš almost certainly at Gir-e Pān,12 there is new information with which to consider the question of Ur III Šimānum's location. From his map, Patterson (2018, 263, map 5) recently seems to have tentatively selected Basorin Höyük, a sizeable multi-period mound on the Iraqi-Turkish border,13 a notion apparently followed since by Garfinkle.14 The historical-geographical evidence will now be reviewed, and then the suggestions as to its localization reconsidered.
While the main body of evidence for Šimānum hails from the Ur III period, it has periodically been equated with the city of Asimānum known from the Akkadian period. During the famous 'Great Rebellion' against Narām-Suen which would ultimately provoke his dramatic self-deification, Asimānum served as the northern base of operations for the rebellious ruler Amar-Girid of Uruk while he unsuccessfully wrote to the lords of Subir for aid.15 Besides its relative proximity to Subir,16 the only geographical indication of its location is that Amar-Girid marches from here to the Tigris, crossing it at a place called Šišil before marching westwards to Basar, the modern Ǧabal Bišrī (not featured on Fig. 1), crossing, in turn, the Euphrates on the way.17 Šišil most likely corresponds to 16 On Subir's location during the third and early second millennia, see esp. Steinkeller (1998) and Michalowski (1999). Regardless of exact placement, this implies that Asimānum was well north of the Akkadian core. 17 Note also the parallel dynamic evident in the intervention of the Amorite factions of the Tidnum and Yaˀmadium in the Šimānum rebellion, and Amar-Girid's recourse to Basar, glossed in the text as the later Šišlu (also written Šišil), a riverine settlement north of Aššur appearing as a polity in its own right during the era of the Third Dynasty of Ur;18 this, hence, necessitates a 'Mountain of the Amorites'; both Asimānum and Šimānum would appear to have alliances with the Amorites. 18 See the very comprehensive discussion of this toponym by Llop/ George (2001, 16 f., fn. 110). We concur that Middle Assyrian Šaṣilu and its variations is a topographic red herring. Parpola/Porter (2001) suggest a localization of Šišlu at modern al-Qayyāra, although Reade's (1978b, 170) suggestion that the latter rather corresponds to Ubasē on the strength of its bitumen sources is quite cogent; Šišil may well be located at a ford somewhere in the vicinity or even much further north, should its association with Ḫariḫumba, not far from Kalḫu (Nimrūd), be recalled (Bagg 2017, 572). location east of the Tigris for Sargonic Asimānum. This, in turn, already all but rules out Sallaberger's identification of Šimānum with Šināmum (Üçtepe) as it is on the Tigridian rive droite.
The later polity of Šimānum was a prominent Hurrian city and kingdom on the Ur III state's northern border; in a memorable episode (revisited in our conclusions), an Ur III princess named Kunšī-mātum was sent here as a ward for an eventual diplomatic marriage with one of its ruler's sons, prior to a rebellion of the state in conjunction with nearby Ḫābūrā, the surrounding districts, and, interestingly enough, the Amorite groups of Tidnum and Ya'madium. This was crushed by Šu-Suen, who conquered the state and undertook deportations (RIME 3/2.1.4.1; see also Michalowski 1975;Studevent-Hickman 2018, 46-48). Something of the city's political significance is demonstrated by Šu-Suen's creation of a settlement of the same name near Nippur within which to house the deportees, but of which he was tutelary deity.
Besides Ḫābūrā (Zaxo Plain), the polities mentioned most prominently with Šimānum during this period are Mardamān (Bassetki), Talmuš (Gir-e Pān), and Ninūa (Nineveh). Like Šimānum, onomastics demonstrate all of these sites to have been culturally Hurrian; in turn, it can be noted that each site is roughly equidistant from the next, implying similarly sized respective spheres of influence. It would hence be difficult to fit as important a kingdom as Šimānum between any of these, meaning that it must lie somewhere beyond this riverine swath of land.
Our first piece of evidence is the much-discussed tablet BCT 1, 4 which seems to describe tribute or tax from soldiers at Hābūrā, Ḫiliš of Talmuš, the wife of Rimānum the Amorite, and soldiers at Ninūa and Ura'u following Šimānum's rebellion (see Owen 1988, 112 f.;Horowitz/ Watson 1991, 409;Maeda 1992, 137). While Ura'u remains elusive, and an Amorite (or his wife) could be residing almost anywhere,19 the remaining toponyms are close to one another, and the context must be of the securing of territory near Šimānum by the Ur III state in the wake of the uprising.20 19 Frayne (1997, 288) suggests an emendation of line 6 to [m a] -r i -⟨d a⟩ -m a -n u -u m , but this seems unlikely from the traces present on the cdli photograph (P105106). 20 Sallaberger (2007, 443 f., fn. 128) argues very convincingly that BCT 1, 4's transactions occurred immediately following the campaign, just as the year had been named. Steinkeller (2007, 15 f., fn. 12) argues that this was g ú n m a -d a tax whereas Sallaberger considers it campaign tribute; thankfully the point may remain moot in our present analysis.
An earlier tablet, UDT 92, presents a list of dignitaries (including Šimānum) of a similar configuration, featuring Naplānum the Amorite, Arib-atal, son of Pusam, of Šimānum, Arib-ḫubbi of Talmuš, Guzuzu of Mardamān, and a man (or men) of N i -k a b . Again, the final entry vexes and the wily Naplānum is known to have been very mobile, but otherwise the geographical context is quite tight.
Telling in this light is the provisioning at Ur of five barley-fed sheep and five barley-fed goats for the kitchen "on the behalf of the man (or men) of Šimānum and man (or men) of Ninūa";21 this shared order suggests that these two or more gentlemen were acting in concert in some way, and suggests, in turn, the relative proximity of Ninūa to Šimānum. This suspicion is further confirmed by evidence from two tablets of a remarkable diplomatic mission consisting of Tiš-atal of Nineveh and a large Ninevite entourage travelling to Nippur by way of Ešnunna in order to swear a treaty with the Third Dynasty of Ur immediately following the defeat of Šimānum.22 As argued by Steinkeller (2007, 15), the logic of such an undertaking must be that Nineveh had previously been subordinate to Šimānum, and that the necessity of maintaining a northern buffer demanded that Nineveh be immediately sworn in as a new vassal of the first rank. Certainly, the timing implies that Nineveh's fealty was crucial following the Šimānum crisis, and hence that the two were both politically and geographically close.
In turn, as noted by Patterson (2018, 257), Šimānum occurs in concert with Aššur (Qalˁat Širqāṭ) in the figure of Zarriqum, e n s i 2 of Aššur, and, indeed, one envoy of Šimānum even bears the name Puzur-Aššur. Both of these points imply some proximity to Aššur.
A further, perhaps revealing point has been made by Studevent-Hickman, who notes the prominent field of G a -m a r -d a in the Nippurian countryside near Šu-Suen's 'new' Šimānum, and the similarity of its latter element to that of the toponym Mardamān, which he considers to have participated in the Šimānum insurrection (Studevent-Hickman 2018, 48). Should this hold,23 then it would 21 m u l ú š i -m a -n u -u m ù l ú n i -n u -a -š è (BCT 1, 68: 5-6, see collation by Horowitz/Watson 1991, 409). 22 As 31-T.615 (edition by Whiting 1976); 6 NT 559 (edition by Zettler 2006), see discussion by Sallaberger (2007, 443 f.) and Steinkeller (2007). 23 The issue of Mardamān's etymology cannot be discussed here in detail, but a parsing of *Marda-/w(e)/-ān (i. e. Marda with the Hurrian genitive suffix -we common in toponyms and often rendered in Sumerian and Akkadian as -ba/e, followed by the Amorite suffix -ān) would seem at least preliminarily cogent, particularly in light of the serve further to underline the propinquity of Šimānum to Mardamān.
A final piece of information is an itinerary noted by Pitts (2015, 171) wherein an embassy to Simānum (suggested by her to be an escort) stops at a river called Titin by the settlement of Kiniḫumma to make offerings. Neither of these toponyms are presently identifiable, although the latter name has something of a Hurrian flavor.
It should be noted at this juncture that none of the available information contradicts the identity of Sargonic Asimānum and Ur III Šimānum; indeed, there is much evidence to support such an assumption.24 Firstly, both are evidently in northern Mesopotamia. Secondly, both play an important role in this region. Thirdly, and most decisively, Ur III Šimānum is also frequently written as Simānum, Asimānum, and Ašimānum.25 This alternation between s and š points to the Semitic sibilant transliterated ś, i. e. (A)śimānum, which would neatly correspond to the Sargonic orthography a-si-ma-num, whereby si can render śi, thus *Aśimānum.
In light of these points, it would be strange to presume a location for Šimānum as far afield as the Upper Tigris Valley; rather, the 'southerly' solution is to be preferred. Asimānum/Šimānum must hence be located at a sizeable tell with monumental EBA remains and a large hinterland east of the Tigris somewhere between Talmuš (Gir-e Pān) and Šišil (perhaps somewhere between al-Qayyāra and Nimrūd) within reasonable distance of both the Zaxo Plain (Ḫābūrā) and Nineveh, in a culturally Hurrian area.
Considering these parameters, an outstanding candidate for Šimānum is the Old Assyrian (OA) city of Šibānum personal names of the Amorites M a r -d a -b a -a n -u m in OIP 115, 175 (i. e. *Marda-/w(e)/-ān/-um), and M a r -d a -b u -u m in BIN 3, 546 o. 6. 12 (i. e. *Marda-/w(e)/-um). An argument could be made for similar 'hybrid toponyms' elsewhere in the region. 24 First suggested by Foster (1990, 36). Though Sallaberger remains cautious, this is on account of his placement of Šimānum in the region of the Upper Tigris. 25 It is somewhat ironically the city's own fall, immortalized in a year name of Šu-Suen, which has furnished so many writings from which variants might be identified (Edzard/Farber 1974, 165 f.; discussion by Garfinkle 2000, 298); the alternation between s and š in the Ur III writings can be explained as an alternation between Sargonic and later spelling conventions for the Semitic sibilant ś (see Gelb 1961, 34-36 for the former). The appearance of variant toponyms with and without initial vowels on the fringes of the Zagros is well attested in other periods, such as NA Lapsia and Allapsia (Fales 1989), MA Pakute and NA Apuqu (MacGinnis/Skuldbøl/Colantoni 2020, 107), or MA Uqumānu and NA Qumēnu (Bagg 2017, 495). Whether this initial vowel reflects an abbreviated vocalization of ālu(m) or the workings of local dialect or even substrate language cannot presently be divined. attested from the reign of Šamšī-Adad I. This belonged to the short-lived larger polity of Nurrugum,26 and was not far from the city of Ninêt, as the crown prince Išme-Dagān was able to take both simultaneously during Šamšī-Adad I's larger campaign to Nurrugum and Qabrā, as appears in a much-discussed letter from Māri, ARM 1, 124: 7-10:27 i-na⟨-an-na⟩ ši-ba-na-am ki ni-né-et ki ù ma-a-tam ka-la-ša aṣ-ṣaba-at "I have now taken Šibānum, Ninêt, and all of the region." As has been convincingly argued, Ninêt is none other than later Ninūa, Nineveh,28 which has prompted an equivalence between Šibānum and later MA Šibanibe, NA Šibaniba, modern Tell Billa (Durand 1987, 224). The ending -be (later -ba) reflects the Hurrian genitive -we, otherwise realized as -pa, -pe, and even -me in toponyms from Akkadian-language texts; its retention within Sumerian and Akkadian texts often occurs among Northern Mesopotamian and Zagrine toponyms regardless of period, so that one might encounter, for example, both Gutium and Gutibum29 or Lullum and Lullube.30 The factors underlying this phenomenon in various toponyms have not been comprehensively investigated, but likely the scribe or speaker's own cultural and linguistic background played a role.31 That a Hurrianized version of the toponym survived into the Neo-Assyrian period may well have been due to the survival of Hurrian cultural tradition in the hinterland north of Nineveh, as is demonstrated by not only the name Šibaniba itself, but also the names of other adjacent settlements32 and local onomastics.33 Hence, Šibanibe (as 26 The precise dimensions of Nurrugum remain unclear, although there is now enough evidence to support it also extending westwards of the Tigris; the Ǧabal Baˁšīqā may well have marked its easternmost extent. Its eponymous capital remains unlocalized. 27 See also collations by von Soden (1952, 83) and Durand (1987, 224). 28 First proposed by Durand (1987, 224), further confirmed by Wu (1994) and Ziegler (2004). 29 See the variant writing g u -t e -b u -u m ki in RIME3/2, 1.4.13: 18. 30 See discussion and references from Taş/Adalı (2016, esp. fn. 2). 31 Much in the manner in which the archaeologist today tends to make free use of the definite article al when referring to modern Arabic toponyms in English, as one might casually mention 'Jawf', 'the Jawf', or 'al-Jawf' for al-Ǧawf (also Dūmat al-Ǧandal), Saudi Arabia. 32 Maganuba's ending clearly distinguishes it as Hurrianizing at the very least and nearby Ḫalaḫḫu is clearly Hurrian. NA Gingiliniš seems Hurrianizing in its final shin. Zadok (1995, 262) further suggests Išparirra and Ḫih ̮ i-ḫina as being Hurrian in origin. 33 While Zadok (1995, 278 f.) remarks on the disappearance of the once-burgeoning Hurrian culture at Arrapḫa, he also calculates that some 14.28 % of names from the rural environs of 9 th century BCE Šibaniba are non-Semitic in origin, presumably Hurrian. it was presumably pronounced in Akkadian)34 might be effortlessly philologically equated with Šibānum.
That OA Šibānum (and hence MA Šibanibe) might, in turn, be identified with Ur III Šimānum presents little in the way of difficulty, as m and b are often interchangeable in Upper Mesopotamian and Zagrine toponymy.35 Indeed, as the MA gināˀu lists demonstrate, an alternate spelling ši-ma-ni-ba exists;36 this is attested enough elsewhere within the corpus that Postgate even conventionally refers to "Šib/manibe" within his 2013 work 'Bronze Age Bureaucracy'.37 In such a light, it is surprising that this neat solution to the localization of Šimānum has not yet been proposed; EBA-MBA Billa's poor state of publication may well be the cause.38 Another may be that the drama and detail of Šimānum's uprising has led to the assumption that it was annihilated, Astour (1987, 49) stating for example that it was "devastated and depopulated so thoroughly that [that] very name was wiped out". This hardly tallies with the narrative of the conquest; while the mass deportations do seem to have been realized,39 Kunšī-mātum is returned to her residence there (RIME 3/2, 1.4.1 iv 26-33), and her father-in-law Pušam is mentioned as late as Šu-Suen's fifth year (Michalowski 1975, 718). In turn, the potency of b a -ḫ u l 'destroyed' in the Ur III period is often less 34 We retain the traditional spelling of Šibaniba here both for ease and as Hurrian lacks long vowels. 35 Perhaps the best example of m to b is Šināmum becoming Sinābu; the reverse is also apparent from Maridabān's transformation to Mardamān, or Lullume, a by-form of Lullube. This phenomenon may well be related to the commonplace interchangeability of m and b between the Babylonian and Assyrian dialects of Akkadian (see Soden 1995, 26, § 21d). It must further be noted that the cursive signs ba and ma look near-identical on ARM 1, 124, this necessitating a justification of the reading -ba-by Durand, and adding to the ambiguity. 36 See the tables by Postgate (2013, 94, tab. 4.1) and Rosa (2010, 341, tab. 1). 37 Consider also MARV 4, 119: 12 hal-zi uru ši-ma-ni-[ba] (read by Freydank 2001, 120 as uru Ši-ba ! -ni- [be], and correctly with ma by Postgate 2013, 25). It is possible that the Šimanibe was the more archaic (and hence prestigious) writing for the province, which might explain its appearance as such in the gināˀu lists as compared to the Šibanibe of more pedestrian administrative texts. De Ridder (2015) has argued that many more attestations should be read with -ma-than previously thought, although MTT II/2 (Cancik-Kirschbaum/Hess 2022, 125f.) remains hesitant in this regard. 38 Steinkeller (1991, 31, fig. 6) places Šimānum north of Nineveh, roughly in the region of modern Tell Billa on his map, but does not clarify this decision within the text itself. 39 Consider the many hundreds (if not thousands) of men of Šimānum appearing as laborers in the archive from near Nippur recently published by Studevent-Hickman (2018), further discussed in our conclusion. than catastrophic. Hence, the continued existence of (a presumably much reduced) Šimānum following the Ur III period is entirely plausible. Finally, Nineveh and Tell Billa are quite close to one another, which may have dissuaded some historical geographers from such an identification. Here, Steinkeller's suggestion that Nineveh was subordinate to Šimānum goes a long way to resolving this issue. It is worthwhile at this juncture to summarize the historical-geographical arguments presented, and to consider these against the alternate proposals made hitherto.
The core historical arguments which we present are hence as follows: Firstly, there is an important northern Mesopotamian city east of the Tigris called Asimānum in the Old Akkadian period which corresponds to an influential Hurrian city-state called (A)Š/Simānum in the Ur III period; this polity lost its significance following this period due to its conquest and deportation at the hands of Šu-Suen, and became a city subordinate to Nineveh called Šibānum (m and b being interchangeable within toponyms in this region), and then an MA/NA provincial capital Šib/ manibe/a (Tell Billa), the ending -be/a simply being a Hurrian genitive suffix frozen in Akkadian. Secondly, a remarkable consistency in toponymy is evident over time, which may be related to the endurance of Hurrian culture within the region of Tell Billa. Thirdly, all of the historical geographical indications point to a location east of the Tigris in a Hurrian-speaking area accessible to Amorite tribes within reasonable distance of Zaxo, Bassetki, Gir-e Pān, Qalˁat Širqāṭ, and a ford somewhere between Nimrūd and al-Qayyāra respectively, and close proximity to Nineveh, for which Tell Billa is ideally located. Finally, as will be considered in the following section, this tell presents an archaeological sequence compatible with all of these factors.
We would argue, in turn, that prior suggestions as to the localization of Šimānum are now unworkable. While the proximity of Ḫābūrā to Šimānum is correct, Astour's suggestion can be dismissed as Bassetki is now known to be Mardamān. In turn, the northerly hypothesis suggested most prominently by Sallaberger and Frayne has lost much of its plausibility, as Mardamān is now much further south than previously assumed.40 In turn, the metathesis of consonants necessitated by Sallaberger's equivalence of Ur III Šimānum with OA Šināmum seems to us unparalleled.41 Indeed, not only was Šināmum called such prior 40 Sallaberger was influenced by van Koppen's (2004, 26-28) placement of Mardamān in the Cizre Plain, which rendered a location north of the Ṭūr ˁAḇdīn highly attractive. 41 OA Šināmum's equivalence to MA Šināmum and NA Sinābu (Üçtepe, Turkey) is unambiguous not only on grounds of its asso-to the era of Šamšī-Adad, as a toponym Šinām is attested, for example in the Ebla corpus (Bonechi 1993, 296), but a kingdom of Šinām(um) continues to appear in Old Babylonian documentation (which would seem to contradict the proposed metathesis). To our knowledge, there is no suitable archaeological site at Sinanköy to justify Frayne's suggestion.
Forlanini's suggestion of Tall Ḥamūqār is interesting, and is close enough to Ḫābūrā, Mardamān, and Talmuš that it can be seriously considered as a candidate for Šimānum. Broadly bespeaking such a suggestion is the very large size of the tell, and archaeological evidence pointing to a violent destruction ending its settlement at some point between the 21 st and 20 th centuries BCE (Colantoni/Ur 2011). Nonetheless, such an ascription presents difficulties. Firstly, an ascription of Ur III Šimānum to Ḥamūqār would exclude an equivalence between this city and Sargonic Asimānum, as the latter must be east of the Tigris according to Amar-Girid's itinerary. While the phenomenon of toponymie en miroir is known for this period, this would nonetheless result in two Asimānums disconcertingly close to one another.42 The next point to be noted is that the total abandonment and destruction of Ḥamūqār would not seem to match up to the conquest mentioned in the Ur III sources, whereby, as noted, it is implied that the deportations did not completely end settlement there, as nance, but also because of its associations with Tušḫum, Elaḫut, and Burundum, all of which are unequivocally to be located in the Upper Tigris Basin. 42 Considering the unbroken period of occupation at Ḥamūqār between the mid-to late third millennium BCE, it would have to be another Asimānum parallel to the Sargonic Asimānum east of the Tigris near Šišil.
Kunšī-mātum is reinstalled in her residence;43 any association with OA Šibānum and MA-NA Šim/baniba/e would also have to be rejected.
Patterson's localization at Basorin (if this is to be inferred from his map), benefits from a large mound and proximity to Mardamān and Ḫābūrā, but the geographical context would be restrictively tight for a polity of this importance, resulting in a historically implausible clustering of toponyms about the modern Zaxo Plain. 44 We would hence argue that the weight of the historical-geographical evidence is very heavily in favor of equating Sargonic Asimānum and Ur III Šimānum with OA Šibānum and MA-NA Šib/manibe/a. Indeed, doing so would even resolve an additional issue, this being the identity of another very sizeable tell in the Transtigris, namely modern Tall Gōmel, the largest and likely most important site in the nearby Navkūr basin during the EBA, and even a conceivable candidate for Šimānum.45 Recorded as Gōgemal by Thomas of Margā, and perhaps corresponding to NA Gammagara,46 a series of recent studies by Marciak and collaborators have demonstrated its undoubted identity with Classical Gaugamēla, site of the decisive clash between Alexander and Darius in 331 BCE (Marciak et al. 2020;2021 Hurrian-speaking kingdom on Ur III periphery, is associated with Ḫābūrā (Zaxo Plain), Talmuš (Gir-e Pān), Mardamān (Bassetki), Aššur, and Nineveh, the latter of which is probably its vassal.
OB/OA Šibānum Possession of Nurrugum, close proximity to Nineveh according to letter, probably of less importance than the latter site.
Šimānum occurs immediately following a polity called Gigimni in another very broken list from Drehem (Arabic: Durayhim), wherein they are both handled by the same m a š k i m . 47 We would suggest that this is identical to the settlement otherwise known as Gigibni48 or Gigibini49 within the Ur III corpus, and perhaps the OB/ OA Kigamnum at which a man of Ḫābūrā delivers wine destined for Tall ar-Rimāḥ (OBTR 251: 6-7, see also OBTR 259: 12), or indeed an uru ki-ga-am-li-[im] from the same corpus (Dalley/Walker/Hawkins 1976, 97).50 Gigib/m(i)n/ li is clearly part of the same Ḫābūrā/Mardamān/Talmuš/ Šimānum complex observable during this period, and the same m/b alternation notable for Maridāban/Mardamān and (as we propose) Šimānum/Šibānum is plain. This ascription would explain the considerable EBA material found at Gōmel, and present a forerunner to the (unetymologized) toponym Gaugamēla.51 To conclude, the historical-geographical information supports an identification of Asimānum and Šimānum with later Šibānum and Šib/manibe/a; indeed, the recent localizations of Mardamān and Talmuš leave scarce other viable locations. This would provide a more or less continuous historical attestation for Tell Billa from the Akkadian to Neo-Assyrian periods with a progression from Asimānum to Šimānum to Šibānum to Šib/manibe/a to Šibaniba.52 47 ⸢g i⸣ -g i -i m -⸢n i⸣ [ ki ] (OIP 121,458,8). The traces of -⸢n i⸣ are clear from the cdli photograph P124188). 48 g i -g i -i b -n i ki (Owen 1997, 389). 49 g i -g i -b i -n i -u m ki (OLP 8, 6 o. 12), see also Steinkeller (1980, 3), preceded by Mardamān and Hābūrā. 50 This name originates from the Akkadian word gi/ugamlu 'paddock' (Old Assyrian ki/ukamlu), potentially of substrate origin. Gigib/m(i)n/li may well be a Sumerian rendering of the same toponym/ word. Note the potential of reading ni as lí, and the frequent ambiguities between n and l in Semitic renderings of Hurrian toponyms, e. g. N/Lullû, Nuḫašše and later Lu'aš/Luḫutu, or Mardamān and Ugaritic mrdml. 51 Consider also a somewhat damaged Sargonic text from Kiš mentioning very large tracts of land (perhaps relating to a conquest?) near various otherwise unknown settlements, one of which is termed gi-gi-ni ki and another si-ba-ni ki (MAD 5, 12). Astour (1987, 32) already argued for an identification of this Sibāni with later Šibanibe, but kept it separate from Šimānum. A detailed historical geographical study of Gigibni and Gaugamēla is being prepared by Edmonds. 52 Considering Billa's identity as OB/OA Šibānum and a potential Sargonic form Sibāni, it must be asked in passing as to whether the mysterious polity of Šabbunum conquered by (a pre-deified) Narām-Suen (TMH 5, 37, see also Westenholz 1975, 29;Such-Gutiérrez 2020, 19) is merely another strange by-form by which this city could go. While such a transformation or by-name is philologically unappealing, Šabbunum is presently a hapax, and Narām-Suen did subjugate the neighboring polities of Talmuš and Maridaban, as recorded in other year names. The question must remain presently open, though we remain skeptical of such an equivalence. This thesis might now be considered in light of the extant documentation of Tell Billa's early archaeological record.

Archaeological Evidence of the Early and Middle Bronze Age at Tell Billa
Archaeological interest in the large mound at Tell Billa was sparked by Layard's soundings at the site in the mid-19 th century, though they did not produce the same monumental art as Nineveh or Nimrud.53 It would be nearly a century before it would be excavated between 1930 and 1937 as a joint venture by the University of Pennsylvania Museum of Archaeology and Anthropology (Penn Museum) and ASOR. Starting at the northeast corner of the mound in the autumn of 1930, E.A. Speiser and his colleagues investigated five areas, expanded to seven by the end of the campaign (Bache 1933, see fig. 2). The goal of the first season was to identify the span of occupation for the site; in Area III, they excavated down to virgin soil after identifying at least seven distinct phases of occupation, preliminarily assigned titles such as "Hurrian" and "Assyrian". Though excavations at the site were never comprehensively published,54 it is possible to outline the occupation history of the site by revisiting the letters written as monthly briefings and pairing them with the remaining field records held in the Penn Museum Archives.55 The site's occupation likely begins at the very end of the Chalcolithic period, c. 3000 BCE. At the close of the first season, the team had cleared down to the seventh 53 Layard (1849, vol. 1, 52; 1853, 133 f.) wrote (in the latter): "It is a vast mound, little inferior in size to Nimroud, irregular in shape, uneven in level, and furrowed by deep ravines worn by the winter rains. Standing, as it does, near abundant quarries of the favorite sculpture-material of the Assyrians, and resembling the platforms of Kouyunjik or Khorsabad, there was every probability that it contained the remains of an edifice like those ruins. … It is remarkable that no remains of more interest have been discovered in this mound, which must contain a monument of considerable size and antiquity." Though the full extent of the site remains unknown (and we cannot presently answer whether Billa possessed a lower mound or not), just the mound itself measures an impressive 34 ha. 54 Indeed, remaining archival records are only partial, see Creamer (2021a) for more details. 55 The discussion of archaeological material that follows is gathered from the correspondence and context notecards contained within the Penn Museum Archives. The only comprehensive field report was published on the 1932-33 (3 rd ) season by the field director Bache. Brief publications by Speiser (1933a;1933b) inform the discussion of ceramic chronology and small finds. stratum of the site, which rested on virgin soil. This deepest stratum was only reached at the northeastern edge of the mound in Area III, and therefore little of it was exposed. Small finds and ceramics were overall scarce in this stratum, yet the examples published by Speiser (1933) can clearly be assigned to painted Ninevite V ware, known from both nearby Nineveh (naturally) and other sites in northern Mesopotamia. Concerted efforts in the last decades to further understand the Ninevite V sequence have demonstrated that the painted variety coexisted with incised types, albeit the former was more popular in the phase's earlier centuries (Roaf/ Killick 1987;Schwartz 1985;Smogorzewska 2016). When paired with the incised Ninevite V ceramics of the sixth stratum, it is clear that the site remained well-populated throughout the early third millennium. Tell Billa's use as an agricultural depot even this early in its existence can be inferred from the presence of a number of "silos" uncovered in this first occupation phase, apparently used for storing grain in deep walls of hard-packed clay. The team exposed five of these silos, which contained mainly discarded ceramics.
Further indicators of occupation during this period are seen in several burials uncovered as part of the seventh and sixth strata. Burial types varied between earthen pit graves and tombs of stone and mudbrick material. Grave goods included squat bowls with broad rims and small bowls serving as lids to larger ones. Often, burials contained a cylinder seal, usually of ivory and with geometric patterns, similar to others found in the area and illustrating the site's connections to the regional sphere of administration practices. One particular burial, Tomb 2, points to the presence of social stratification at EBA Billa; the tomb consisted of a square chamber of brick upon a foundation of stone, covered with a mudbrick "tent" roof. It contained two large painted jars (Ninevite V) in the corners, with a bronze bowl serving as a lid. In the middle of the tomb were three fine bowls with painted rims, and a silver pin lay near the head of the deceased. A silver ring rested around one of the departed's fingers. The evident wealth of this burial in comparison to its contemporaries signals some amount of wealth accumulation at Tell Billa in the early third millennium, lending credence to its importance and, therefore, proposed association with Asimānum.
While continuing investigations in Areas II, III, and IV, a trial trench was also dug at the base of the eastern slope. Here, a "Cyclopean" retaining wall was revealed, identifiable in some areas as reaching a height of 4.5 meters, with a core of mudbrick and stone facing (Fig. 3). This later proved to be part of the fifth stratum at Tell Billa, which can be dated to the latter half of the third millennium. The wall's thickness was roughly five meters in some places, and extended at least 50 meters from roughly north to south. Just inside the retaining wall was a deep stone well, found to descend at least 10 meters before work was necessarily halted. The monumental construction activities at the site in the latter half of the third millennium evidenced by such a structure were accompanied by additional smaller buildings, including a square chamber made of stone blocks entered via a stone staircase. This was suggested to have been used for storage before eventually being utilized as a space for three burials.
Directly outside of the monumental wall were found numerous weapons, including a lance head with a ridged blade and octagonal bulb at the haft, an axe-head, an adze, a mallet, three barbed flint arrowheads, and (most strikingly) a bronze lance-butt incised with what seems to be an Early Dynastic cuneiform sign.56 The massive size of the walls, the association of these artifacts of war with the wall's perimeter, and the ability to draw water from inside all demonstrate a preoccupation with the site's defensibility during the mid-third millennium. The involvement of Asimānum/Šimānum in conflicts both during the Sargonic and Ur III dynasties certainly creates a tempting parallel. Most notably, Tomb 22 in the fifth stratum contained two further impressive bronzes: a semi-circular bronze battleax, and a short bronze sword (50 cm long; Fig. 4). The 56 Speiser 1933a; the sign seems to resemble ne (= i z i 'fire'?) from the photograph (see Fig. 5). tomb was constructed with "low rubble" walls and a cover of mudbrick, with further mudbrick surrounding the tomb to a height of one meter.
The fourth stratum breaks from the monumental architecture of the previous level. This phase of occupation contained a number of buildings, likely domestic residences, and is securely dated to the MBA by the presence of Khabur Ware (Speiser 1933b, pl. LIX). The structures of the fourth stratum were mainly built with stone foundations, undoubtedly contributing to the longevity of occupation during this period. A street ran through the middle of the excavated area, following the bend of the mound, appearing to run southeast-northwest. Two stone wells were dug near the street, with the deepest one reached by a flight of steps constructed of baked brick. A continuation of significant social stratification in the residential population is seen in several rich tombs, including one constructed of stone containing a lapis amulet, numerous other stone beads, multiple bronze implements, gold earrings, and a semi-circular bronze battleax.
In sum, Strata 7-4 at Tell Billa illustrate the site's continuous occupation from the very beginning of the third millennium through the mid-second millennium. The later strata of the mound (Strata 3, 2, and 1) can be securely dated to Hurrian, Middle Assyrian, and Neo-As-syrian cultural horizons, respectively.57 While investigations were confined to the northeast corner of the mound in the lower levels, excavations were still able to identify large-scale storage facilities as part of Strata 7 and 6, monumental architecture and further storage in the fifth stratum (coinciding with the latter half of the third millennium and, therefore, lending credence to the site's importance during the Sargonic and Ur III dynasties), and part of a residential neighborhood in the fourth stratum. This continued occupation supports the identification of Tell Billa as the historical Asimānum/Šimānum. We might even consider the abrupt shift from monumental architecture and defense in Stratum 5 to the domestic residences of Stratum 4 as resulting from the conquest and subsequent deportations and "destruction" claimed by Šu-Suen, and the end of Šimānum as an autonomous power.58 A reorganization of spatial use between occupation levels such as this could easily denote an event-based cause; the "destruction" wrought by Šu-Suen (to whatever degree it was actually carried out) might have been one 57 See Creamer 2021a, for discussion of Stratum I, and 2021b for Strata II/IA. 58 No destruction layer is mentioned by the excavators, but there is also no need for such an event considering Kunšī-mātum's reinstatement in Šimānum and the survival of Pušam and Ipḫuḫa. such event -if not from the actual destruction itself, then from the victor's demands stipulating the dismantling of defensive architecture.59

Conclusions: A New History of Tell Billa
Between the sites of Tepe Gawra, Tell Billa, and, of course, Nineveh itself, the Nineveh Plain presents a flourishing LC landscape. As the titular ceramic horizon suggests, Ninevite V ware represented a growing population in this region in regular contact, organically occupying the landscape at key locales (Roaf/Killick 1987;Smogorzewska 2016). As mentioned, Tell Billa occupies a desirable location, oriented so as to enjoy the rich agricultural potential of the Nineveh Plain, while also situated along the excellent pastoral lands provided by the Ǧabal Baˁšīqā and its surrounding foothills. Furthermore, both Billa and Gawra likely benefitted from lying athwart the passes leading between the plain of Nineveh and the likewise productive Navkūr Plain, where we now know both Tall Gōmel and Asīngeran to have been occupied in this period, these latter two being well-situated to take advantage of a trans-Baˁšīqā route (Morandi Bonacossi /Iamoni 2015).60 This network of large sites in the Late Chalcolithic undoubtedly encouraged the expansion of a broad, interregional communication network towards polities further north and south, only growing further during the onset of the EBA, as is supported by the ubiquity of Ninevite V ware among sites of the Greater Ḫābūr and Upper Tigris regions. Tell Billa would have profited from its particularly strategic position within this interregional network.
With the expansion of Akkade into Northern Mesopotamia, the settlement first becomes historically visible as an important site from which Amar-Girid could make appeals for insurrection among the kings of Subir. Asimānum's location would here have been vital; in the event of a sudden disaster, he could have climbed the Ǧabal Baˁšīqā for safety. In turn, the city oversaw the important Tigridine communication route to Northern Mesopotamia and South-Eastern Anatolia via Talmuš and Mardamān. When help was not forthcoming, he marched to the Ǧabal Bišrī, presumably to seek support from the Amorite groups 59 Further archaeological investigation of Billa would hence provide an excellent potential case study for the impact of deportation upon urban planning in the Early to Middle Bronze Age. 60 For LC Gōmel, see Morandi Bonacossi et. al. (2018, 73, 132-138), and for Asīngeran, see Iamoni/Qasim (2019). there, crossing the Tigris at Šišil, and marching via the Wādī aṯ-Ṯarṯar to Basar. It is interesting that Amorite groups also played a role in the rebellion against Kunšī-mātum; perhaps a traditional alliance existed between the inhabitants of Tell Billa and such mobile factions.61 Indeed, while it is difficult to assess the strength of the Hurrian states of the north, it is striking that Narām-Suen fought and won a battle with Talmuš far afield in Lulluba (Such-Gutiérrez 2020, 20), implying the capacity of these polities to field expeditionary armies and that a complex web of alliances may have existed between these different piedmont states. Certainly, the elite graves, "Cyclopean" wall with a well and storage, and weaponry strewn before it all imply that Billa was an important (and likely formidable) polity during this epoch.
An interesting dynamic between Tell Billa and Nineveh is evident from the very outset; the Hurrian goddess Šauša of Nineveh (later syncretized with Ištar) was a patron deity of interregional significance, and widespread popularity in the region is intimated by later veneration of Ištar at Talmuš and Šibaniba, and the existence of a Dūr-Ištar in NA Šibaniba's hinterland;62 indeed, mention of her as Šaušga still occurs as late as the reign of Šarru-ukīn.63 Following the conquest of Nurrugum, her temple was restored by Šamšī-Adad I, who claimed that Man-ištūšu of Akkade was its original builder (RIMA.0.39.2). While dismissed by Westenholz (2004), who notes the apparently autonomous nature of the local material and the lack of Akkadian archaeological remains and argues that this was a fiction of Šamšī-Adad I who sought to emulate the Akkadian Dynasty, the question remains as to why he would choose Man-ištūšu rather than his more famous father.64 Considering his successor Narām-Suen's conquest of Talmuš 61 Consider the aforementioned connection between Amar-Girid and Basar. The level of interplay between Amorite nomads and these ostensibly Hurrian cities remains somewhat obscure, although it is interesting to note a figure called m a r -d a -b a -a n -u m is described as a m a r -t u in OIP 115, 175, and one m a r -d a -b u -u m as such in BIN 3, 546 o. 6. 12. 62 Note also the prophecy of Ištar of Ninêt related to Zimri-Līm (ARM 26/1,192), and that a bronze vessel dedicated to her is mentioned as far afield as Šemšārra (SH 894: 9). 63 d ša-uš-ka ra-ši-bat nina ki "Šauška, the awe inspiring one of Nineveh" (RINAP 2, 43: 54;Frame 2021, 229 suggests that this might be an error for āšibat 'inhabitant'). 64 Considering the text's frankly obsessive and repetitive mention of the preservation of Man-ištūšu's inscriptions during the renovation works, it seems unlikely that Šamšī-Adad I would have confabulated this, particularly as the inscription evidently expects a later ruler to read its earlier counterpart and thus attribute the survival of Maništūšu's inscriptions to Šamšī-Adad I's own respectfulness. early in his reign and that Amar-Girid's insurrection was hosted in part here, it may well be that temple construction by Man-ištūšu at Nineveh was an earlier attempt at winning the 'hearts and minds' of Upper Mesopotamia.65 Moreover, it has been suggested that later building projects of the Neo-Assyrian period may have destroyed any remains of Akkadian architecture; further excavations along non-central areas of the site, like the "gully cut" of the UC Berkeley excavations, may reveal further Akkadian presence.66 65 Note also the deposition of a copper spearhead in the temple of Ištar at Aššur by Azuzu, likely the local ruler, on Man-ištūšu's behalf (RIMA.0.1002(RIMA.0. .2001. Maridabān is mentioned as a target of Šarrum-kīn's campaigning in literary sources (Westenholz 1997, 85, no. 7 i 16ʹ), so Akkadian interests in the region may well have predated Narām-Suen. The mention in the same passage of Tidnum (as Didnum) is perhaps also telling. 66 Westenholz's argument can be put to rest when we consider typical pottery forms of the era found at Nineveh which were also recovered at Akkadian-occupied sites, such as Tall Brāk (Nagar/Nawar), Tall Asmar (Ešnunna, not on fig. 1), and Yorġān Tepe (Nuzi, not on fig. 1) (Gibson/McMahon 1995, 16;Zettler 2006, 508). New information from the German and Italian teams returning to investigate Nineveh will likely clarify this further.
It remains unknown as to whether Asimānum was punished by Narām-Suen for harboring Amar-Girid's army, but it reached its floruit, nonetheless, during the apparently peaceful Ur III period.67 Šimānum commanded enough interregional significance that a diplomatic marriage was planned between the Dynasty of Ur and the local Hurrian dynasty.68 Probably late in the reign of Šulgi, while Šu-Suen was still governor of Uruk,69 a daughter of his was sent to Pušam, king of Šimānum, presumably as a ward considering the length of her unmarried sojourn; her name, Kunšī-mātum ('Submit, land!'), may be less than coincidence (Weiershäuser 2008, 264). The presumed aim 67 Argued by Sallaberger (2007, 433), if a tad ex silentio. That Ḫābūrā and surrounding lands rose in support of Šimānum's revolt implies that the latter wielded considerable influence within the broader region. 68 The frequent use of diplomatic marriages by the Ur III state to influence its northern allies is well attested, e. g. Weiershäuser (2008, 263 f.). 69 For an overview of references to Pušam, see the tables compiled by Klein (1990). In Šu-Suen's third regnal year, Pušam was ousted and Kunšī-mātum expulsed from her residence by an insurrection said to be composed of Šimānum, Ḫābūrā, and the surrounding districts. Moreover, marauding Amorite groups including the Tidnum, against whom Šu-Suen was engaged in building a defensive wall,72 and the Yaˀma-70 The reconstruction here centers upon the interpretation of the Sumerian word é -g i 4 -a , which might be translated as either "bride" or "daughter-in-law". Michalowski (1975) translates the latter consistently in his reconstruction, as does Weiershäuser (2008, 263); we follow this interpretation. Stol (2016, 486) presents a completely different interpretation, considering Arib-atal to have been her husband, and for him to have become king, which seems less likely considering the political events. 71 As noted by Michalowski (1975), Kunšī-mātum is described both once as the é -g i 4 -a of Arib-Atal and once as the é -g i 4 -a of Ipḫuḫa; was there some indecision as to which son would marry her? 72 See the synthesis by Frayne (1997, 290-292). dites73 entered the region in support of the rebellion. The timing of this event very early in the reign of Šu-Suen is not circumstantial; with the accession of Šu-Suen after Amar-Suen's celebrated and untimely "death from the bite of a shoe", a princess who previously would have belonged to a mere cadet branch of the Third Dynasty of Ur was now the daughter of the reigning god-king. It may well be that Arib-atal incited this rebellion as she had been conclusively promised to his brother Ipḫuḫa,74 presumably drawing upon neighboring polities' disenchantment with the Ur III state's northern policies. If so, then he made the mistake of failing to take Kunšī-mātum hostage.75 Šu-Suen first defeated the nomads and their kings, perhaps even campaigning into the Greater Ḫābūr in the process (Frayne 1997, 290), and then took his revenge upon Šimānum, deporting thousands of rebels to a purpose-built colony on the Nippurian frontier, named by him 'Šimānum' after their city of origin, but of which he was the patron god.76 The regional power vacuum was filled by Tiš-atal following his oath of fealty to the Third Dynasty of Ur; Nineveh's star was now ascendant, and its importance would only grow in the Old Babylonian era.77 Meanwhile, Kunšī-mātum was re-established in her residence, now with Šimānum, Ḫābūrā, and the surrounding districts "assigned to her service".78 There is no mention 73 Likely the progenitors of the subsequent kingdom of Yamḫad despite the unappealing phonological changes necessary. 74 We assume this on the basis that Arib-atal is not mentioned after the events of the rebellion, while Ipḫuḫa is still attested. 75 Considering the sparing of family members, another interpretation might be that the uprising was popular in nature and coordinated with the Amorite factions, and Arib-atal was merely a coerced figurehead, rather than the episode reflecting a dynastic dispute, the reconstruction ultimately preferred by Michalowski (1975). 76 See the references to deported troops of Šimānum in the environs of Nippur and Girsu collected by Sallaberger (2007, 443), to which might now be added a recently edited corpus featuring thousands of unnamed men from Šimānum laboring in a settlement near the city of Nippur (Studevent-Hickman 2018, esp. 42-48). It would not surprise us if this vast manpower was ultimately used to build Šū-Suen's wall against the Amorites. 77 The suggestion that a.ab.ba-bašti, a minor wife of Šū-Suen, was Ninevite in origin (and thus could have influenced events) remains presently unproven (Sharlach 2017, 283 f.). Although little is archaeologically known of OB/OA Nineveh, some official correspondence and literary compositions found there would imply a certain importance (Dalley 2001). 78 Note the aforementioned delivery of BCT 1, 4. While, previously, a further rebellion by Ḫābūrā and Mardamān against Šū-Suen in his seventh regnal year might have been reconstructed (RIME 3/2, 1.4.3 vi 15-18), collation by Krebernik (2002, 133) has rendered the reading of Ḫābūrā impossible and Mardamān rather unlikely. Sallaberger (2007, 443) discards Ḫābūrā but continues to consider Mardamān plausi- of Arib-atal after the rebellion, while there is of Ipḫuḫa of Šimānum as late as the year of the accession of his presumed brother-in-law Ibbi-Suen (Michalowski 1975, 718). Thus, it might be assumed that the royal couple evidently lived as happily ever after as was possible in a somewhat depopulated Šimānum in the closing years of the Ur III state.
Šibānum emerges once more in the era of Šamšī-Adad I as a possession of the polity of Nurrugum, the eponymous capital of which remains to be localized. Šibānum would now appear to be of much lesser significance, being taken by Yasmaḫ-Adad in tandem with his grander siege of Ninêt but otherwise presently unmentioned within the historical record.79 While the archaeological record from Stratum 4 displays an affluent urban environment, the lack of the previous period's monumental defensive installations implies a loss of independence.
Just such a dynamic would follow for the remainder of Assyrian history, with the holy city of Nineveh as an important metropolis and ultimately a sprawling imperial capital,80 and Šibaniba as a smaller, albeit prosperous provincial center during the Middle and Neo-Assyrian periods.81 This is all the more striking considering Steinkeller's arguments that Ninūa had been subordinate to Šimānum prior to the Kunšī-mātum crisis: Would Jonah have been sent to Nineveh at all if things had gone differently for a Sumerian princess?
Regardless, Šibaniba and the region of the Ǧabal Ba'šīqā would retain their Hurrian identity well into the Neo-Assyrian period where Nineveh did not. Indeed, when Šarru-ukīn began his famous and ill-fated city-building project of Dūr-Šarru-ukīn (modern Ḫorsābād) nearby, he chose the Hurrian village of Maganuba for his site, the name of which has ironically lasted much longer than that of his own city, surviving in that of the Ǧabal Maqlūb.82 ble. We would rather forgo both readings until proven otherwise, as retaining the now-doubtful Mardamān as only half of the doublet raises far more questions than it answers. 79 By means of comparison, Mardamān, Ḫābūrātum, and (more distantly) Kumme all enjoyed interregional importance during this era. 80 The cultural importance of Nineveh already in the subsequent period is demonstrated by the (perhaps fictional) Hurrian king Pizigarra of Nineveh's destruction of the city of Ebla at the command of Teššub of Kumme in the Hurrian "Song of Release", the appeal to Ištar of Nineveh at the close of the Tale of Appu, and the popularity of Ištar of Nineveh in the Hurro-Hittite world (Beckman 1998). 81 See Finkelstein 1953;Creamer 2021a; note Šibaniba's fame for its livestock according to the moniker of its eponymous gate at Nineveh (Reade 1978a, 52). 82 Consider also the variant forms ma-ag-ga-nu-ub-ba and ma-gani-ib (Bagg 2017, 380 f.), and the aforementioned confusion between l In turn, the Syriac name for the mountain, ˀAlp̄ap̄, likely derives from the name of the Hurrian village of Lipapan once ensconced within this range.83 That our Assyrian king termed this mountain Muṣri in his inscriptions (derived from the Semitic root mṣr, 'border') could point to it having being a frontier between Semitic and Hurrian speakers at some point during the Early and Middle Bronze Ages.84 Perhaps a similar dynamic of cultural longevity is exemplified in Billa's surroundings today by the Arabic-speaking Yazīdī populations of Baˁšīqā and Baḥzānī or the Syriac Orthodox Christians of Mērkī and their many centuries of both fruitful and tragic interaction with the surrounding plain. The striking interplay between landscape, culture, and political formations become at once apparent when such longue durée reconstructions from a historical and archaeological perspective are possible. Regardless, we hope to have provided a new piece of the larger history that Billa has still to tell.