Just a Series of Misunderstandings? Assyria and Bīt-Zamāni, Ḫadi-/Iḫtadi-libbušu, and Aramaic in the early Neo-Assyrian State

The region of the Upper Tigris serves as a key case study in understanding the early expansion of the Neo-Assyrian Empire. Nevertheless, various aspects of its incorporation within the Neo-Assyrian pale remain obscure, particularly the date and nature of the establishment of the province of Amēdu or Na’iri, previously the Aramean polity of Bīt-Zamāni. After a summary of prior arguments and an investigation of the polity’s Middle Assyrian past, two overlapping and complimentary histories are written, one of the political interactions between Assyria and Bīt-Zamāni, and another of Assyria’s provincialisation of the Upper Tigris. The former finds that Bīt-Zamāni was remarkably resilient in the face of Assyrian aggression, while the latter argues that an early Assyrian presence at Damdammusa was replaced in 879 BC by the provinces of Sinābu/Na’iri and Tušḫan. These two histories are then supplemented by a prosopographical investigation of the Assyrian eponym of 849 BC, the first attested governor of Na’iri, one Ḫadi-libbušu or Iḫtadi-libbušu. It is demonstrated that the two contemporaneous variants of his name within the Assyrian textual corpus may be explained as an ambiguity in translating the Aramaic personal name *ḥdhlbbh into Akkadian for use as an eponym date. It is hence likely that Ḫadi-/Iḫtadi-libbušu was an indigenous potentate made governor, and thus that the polity of Bīt-Zamāni serves as a previously unrecognised example of the Postgatian ‘transitional case’ within the Early Neo-Assyrian Empire analogously to Bīt-Baḫiāni/Gūzāna. Indeed, it is argued that a similar phenomenon of translating the transitional ruler/governor’s name into Akkadian for limmu dating may here be attested for Gūzāna’s two initial governors. In light of these findings, their broader implications for the use of Aramaic in correspondence or record-keeping within 9th century Assyria are considered, and it is suggested that Ḫadi-/Iḫtadi-libbušu’s correspondence was conducted in Aramaic, whence scribes must have had recourse in spelling this potentate’s name. This would mark the earliest use of Aramaic within the Neo-Assyrian bureaucracy presently known. It is then finally concluded that the threat of Urarṭu in the last years of Aššur-nāṣir-apli II’s reign may well have compelled him to enter in a manner of compact with BītZamāni, and that the indigenous rulers were thereafter made Assyrian governors, only to be unseated in favour of Ninurta-kibsī-uṣur, šāqiu rabiu to Salmānu-ašarēd III just prior to Amēdu’s rebellion in the succession war of 826-820 BC, after 74 Alexander Johannes Edmonds which it was conclusively incorporated. Some insufficiencies of present theories of Neo-Assyrian imperialism in explaining this complex historical scenario are finally highlighted.

Abstract. The region of the Upper Tigris serves as a key case study in understanding the early expansion of the Neo-Assyrian Empire. Nevertheless, various aspects of its incorporation within the Neo-Assyrian pale remain obscure, particularly the date and nature of the establishment of the province of Amēdu or Na'iri, previously the Aramean polity of Bīt-Zamāni. After a summary of prior arguments and an investigation of the polity's Middle Assyrian past, two overlapping and complimentary histories are written, one of the political interactions between Assyria and Bīt-Zamāni, and another of Assyria's provincialisation of the Upper Tigris. The former finds that Bīt-Zamāni was remarkably resilient in the face of Assyrian aggression, while the latter argues that an early Assyrian presence at Damdammusa was replaced in 879 BC by the provinces of Sinābu/Na'iri and Tušḫan. These two histories are then supplemented by a prosopographical investigation of the Assyrian eponym of 849 BC, the first attested governor of Na'iri, one Ḫadi-libbušu or Iḫtadi-libbušu. It is demonstrated that the two contemporaneous variants of his name within the Assyrian textual corpus may be explained as an ambiguity in translating the Aramaic personal name *ḥdhlbbh into Akkadian for use as an eponym date. It is hence likely that Ḫadi-/Iḫtadi-libbušu was an indigenous potentate made governor, and thus that the polity of Bīt-Zamāni serves as a previously unrecognised example of the Postgatian 'transitional case' within the Early Neo-Assyrian Empire analogously to Bīt-Baḫiāni/Gūzāna. Indeed, it is argued that a similar phenomenon of translating the transitional ruler/governor's name into Akkadian for limmu dating may here be attested for Gūzāna's two initial governors. In light of these findings, their broader implications for the use of Aramaic in correspondence or record-keeping within 9 th century Assyria are considered, and it is suggested that Ḫadi-/Iḫtadi-libbušu's correspondence was conducted in Aramaic, whence scribes must have had recourse in spelling this potentate's name. This would mark the earliest use of Aramaic within the Neo-Assyrian bureaucracy presently known. It is then finally concluded that the threat of Urarṭu in the last years of Aššur-nāṣir-apli II's reign may well have compelled him to enter in a manner of compact with Bīt-Zamāni, and that the indigenous rulers were thereafter made Assyrian governors, only to be unseated in favour of Ninurta-kibsī-uṣur, šāqiu rabiu to Salmānu-ašarēd III just prior to Amēdu's rebellion in the succession war of 826-820 BC, after

INTRODUCTION
The region of the Upper Tigris presents a vital scenario for investigating Neo-Assyria's expansion, particularly considering the wealth of archaeological and philological information which might be brought to bear. 1 Separated from the Fertile Crescent by the Ṭūr 'Abdīn, the ancient Kāšiāri, its enclosed basin witnessed Assyrian intervention, conquest, settlement, and provincialisation both in the Middle and Neo-Assyrian eras, rendering it a fascinating counterpoint to contemporaneous historical phenomena on the Upper Ḫābūr. Like the latter region, an archaeological image of excellent resolution has emerged over the past decades, the most exemplary thereof being the Assyrian provincial capital of Tušḫan, 2 where a plethora of information on the late Neo-Assyrian settlement including monumental architecture, cuneiform documents, and floral and faunal remains has been evinced. In turn, extensive surveying and further investigations at companion sites have provided a fuller picture of the Assyrian settlement pattern. 3 Rich and informative as this history of the Assyrian Upper Tigris basin has become, archaeological and philological knowledge of the important Aramean polity of Bīt-Zamāni and its capital at Amēdu 4 has remained scant, and often Assyria's interactions with it have been portrayed as little more than a prelude to the construction of the Assyrian province of Tušḫan. 5 Despite the difficulty and apparent contradictions of many of the textual sources available, it is nonetheless the present author's conviction that the heavily disputed issue of the date and nature of Bīt-Zamāni's annexation can be satisfactorily solved, and a new history of this historical scenario written. This is accomplished through the undertaking of two parallel but interconnected histories of Assyria and Bīt-Zamāni in the early Iron Age. The first of these focuses upon Assyria's interactions with this Aramean polity and Machtpolitik within the region. The second of these examines Assyria's early provincialisation of the Upper Tigris. These two histories overlapping in time and space demonstrate that the actual history of Bīt-Zamāni/Amēdu's integration into the Upper Tigris region was far more complicated and nuanced than has previously been understood. From these, the personage of the first governor of Amēdu attested today as both Ḫadi-libbušu and Iḫtadi-libbušu within Assyrian dating formulae is investigated, and the puzzling alternate versions of his name within the Assyrian textual record explained. The evidence collected is then compared to the 'transitional case' of Bīt-Baḫiāni/Gūzāna. Before a concluding reconstruction is presented of the annexation of Bīt-Zamāni/Amēdu, the repercussions of the onomastic findings on Ḫadi-/Iḫtadi-libbušu are briefly discussed in relation to the history of the use of Aramaic in the Neo-Assyrian Empire. Firstly, however, the problem of the annexation of Bīt-Zamāni/Amēdu must be considered.

THE PROBLEM OF BĪT-ZAMĀNI'S ANNEXATION
The writing of the history of the annexation of territories to the māt Aššur generally follows a series of historical inferences. On the one hand, the eponym (or limmu) lists and chronicles and administrative texts hailing from the established empire are exhaustively scoured for the names of governors of provinces, presenting a diachronic array of first attestations of various provinces, and, on the other, annalistic accounts of campaigning by Assyrian kings are studied to identify expansion. These two sets of information are then harmonised. Archaeological evidence is generally employed to support these claims where necessary. Such intricate work demands constant refinement as new information becomes available. 6 Two provinces are attested within the Upper Tigris during the later stages of the Assyrian Empire, namely the aforementioned province of Tušḫan, inaugurated 879 BC by Aššur-nāṣir-apli II, and a province known as Na 'iri,Sinābu,Amēdu, The date of the second province's inauguration generally hinges upon the reconstruction of a period of seventeen years between the failure of Aššur-nāṣir-apli II to capture Amēdu in 866 BC and the appearance of the limmu year, or eponym, of one Ḫadi-libbušu or Iḫtadi-libbušu, governor of the land of Na'iri, in 849 BC. The only extant occurrence punctuating this is a brief and uneventful account of Salmānuašarēd III's march through the land of Bīt-Zamāni in 856 BC. 8 Following Ḫadi-/Iḫtadi-libbušu, the eponym of another governor is known, one Ninurta-kibsī-uṣur, from 838 BC, although he still governed Raṣappa at the time, and must only have later been posted to Na'iri. Finally, Amēdu rebelled during the succession war which rocked Assyria between 826 and 820 BC 9 but was brought once more to heel soon thereafter. 10 The point at which Bīt-Zamāni became an Assyrian province has long been debated. Views generally fall within one of two camps, namely those proposing an annexation between 866 and 849 BC, 11 and those championing a later incorporation during the last quarter of the 9 th century. 12 Both views rest upon complex inferences.
The 866-849 camp adopts an argument rooted in the eponym lists and the Stelenreihen of Aššur; while very convincingly armed with Ḫadi-/Iḫtadi-libbušu's limmu in 849 BC, the major stumbling block of the argument is that the geographical extent of his province is unknown, and must be inferred from Aššur Stelae nos. 39 and 47, which belong to the later governors Marduk-šimanni (eponym for 799 BC) and Ninurta-kibsī-uṣur (eponym for 838 BC) respectively. Amēdu does not appear on these stelae, but rather 'Andi', which must hence be taken as a byname for the city. 13 6 The most comprehensive recent study is Radner 2006b. 7 Radner 2006b: 49-50. Na'iri was the Assyrians' traditional moniker for the broader region, Sinābu was the name of the most prominent 'Assyrian' city within its extent, previously its Middle Assyrian provincial capital. 8 'Moving on from the city Kār-Salmānu-ašarēd, I crossed over Mount [Ḫa]sumu and went down to the land of Bīt-Zamāni. Moving on from the city of Bīt-Zamāni, I crossed over Mounts Namdānu and Merḫisu. ' (A.0.102.1 ii 40-41 = Grayson 1996: 19). 9 'When Aššur-da''in-apla, at the time of Salmānu-ašarēd (III), his father, acted treacherously by inciting insurrection, uprising, and criminal acts, caused the land to rebel and prepared for battle; (at that time) the people of Assyria, above and below, he won over to his side, and made them take binding oaths. He caused the cities to revolt and made ready to wage battle and war. The cities Nineveh, Adia, Šibaniba, Imgur-Enlil, Iššabri, Bīt-Šašširia, Šīmu, Šibḫiniš, Tamnuna, Kipšūna, Kurbail, Tīdu, Nabulu, Kaḫat, Aššur, Urakka, Raqmat, Huzirīna, Dūr-balāṭi, Dariga, Zaban, Lubdu, Arrapḫa, and Arbail, together with the cities Amēdu, Tīl-Abnē, and Hindānu, -altogether twenty-seven towns with their fortresses which had rebelled against Salmānu-ašarēd (III), king of the four quarters, my father, sided with Aššur-da''in-apla. By the command of the great gods, my lords, I subdued (them).' (A.0.103.1 i 39-53 = Grayson 1996: 183). 10 For a list of known governors of Amēdu, see Radner and Schachner (2001: 770-772). 11 Kessler 1980: 100-102;Radner 2006b: 49. 12 Forrer 1920Lipiński 2000: 160-161;Younger Jr. 2016: 306. 13 A correction of an-di to ti ! -di, i.e. the Assyrian settlement of Tīdu, seems unlikely considering that Andi appears both in Stelae nos. 37 & 49. A reduction of Amēdu to Andi suffers from the fact that the toponym's middle vowel has otherwise weathered the ravages Those suggesting a later annexation generally focus their argument upon the nondescript accounts of the campaigns of 856 and 830 BC which could be taken to imply an enduring vassaldom on Bīt-Zamāni's part, and the description of the succession war of 826-820 BC, which may intimate that Amēdu was still a client state at the time, rather than a province. 14 Nonetheless, the proponent of Amēdu's clientship up to the last quarter of the 9 th century must contend with the unequivocal attestation of two governors of Na'iri prior to Amēdu's rebellion.
Having introduced the basic argumentation, the main historical investigations of this paper may begin in earnest; prior to this, however, a brief study of the preceding Middle Assyrian period must be undertaken.

BĪT-ZAMĀNI IN THE MIDDLE ASSYRIAN PERIOD
The first attestation of Bīt-Zamāni hails from the 13 th century, referring to a ḫassiḫlu of the ḫalzu of Bīt-Zamāni. 15 This startlingly early reference would locate it firmly within the era in which Assyrian kings conquered much of the Upper Tigris during their Mitannian campaigns. 16 While the actual extent of their conquests remains difficult to judge, a reference to Šināmu 17 in an administrative document detailing the dispatch of hurādu-troops for work there during the reign of Salmānu-ašarēd I 18 heavily implies the establishment of an Assyrian presence in the region focused upon this city. 19 This would have been accompanied by the creation of dunnu-settlements as typified by the site of Giricano and other hallmarks of Middle Assyrian provincialisation. 20 In considering Bīt-Zamāni itself, its designation as a ḫalzu is key; while this term's precise use and semantics remain disputed, 21 a defensive connotation seems likely. 22 In turn, the title ḫassiḫlu may imply a culturally Hurrian origin for the region's administration. 23 Nonetheless, its marcher lord's solidly Assyrian name of Aššur-kāšid son of Bēl-qarrād demonstrates that this region was under Assyrian control. 24 A similar practice may also be evidenced for a region southwest of Šināmu, Eluḫat, conquered by Salmānu-ašarēd I and colonised by Assyrians, later termed Halziluḫa in Neo-Assyrian sources. 25 Quite evidently, the Assyrian settlements founded on the wide plains of the Upper Tigris were protected by ḫalzu-districts such as the two evidenced, presumably appended to the province.
Aramean troubles of the 12 th and 11 th centuries seem to have heavily undermined the Middle Assyrian state's authority in the region and may have led to the abandonment of the lowland settlements. 26 Despite the chaos intimated by mention in Broken Obelisk of Aššur-bēl-kala of a battle with Arameans at Dunnu-ša-Libur-zānin-Aššur 27 in the pāḫatu of Šināmu, the White Obelisk of Aššur-nāṣir-apli I 28 describes the king pursing an enemy north of the Kāšiāri. 29 It is probably apt to consider the Upper Tigris to have been in a state of political flux. 30 As will be further discussed, this instability would seem to have given both the relict Assyrians of the region and the polity of Bīt-Zamāni a necessity for self-reliance and sense of independence which would hamper Assyrian efforts in the region during the Iron Age. What must also be stressed is that Bīt-Zamāni's past as a fringe military holding of the Middle Assyrian state may well explain the unusual and invasive relationship which Assyria had with its vassal in the 9 th century, and its later history.

BĪT-ZAMĀNI IN THE ASSYRIAN (RE-)CONQUEST OF THE UPPER TIGRIS
While no campaigns to the Upper Tigris region can yet be attested for the reign of Aššur-dān II, his early reimposition of vassaldom upon the kingdom of Katmuḫu on the north of the modern Cizre plain was a necessary precondition for such. 31 His successor Adad-nārārī II's explosive campaigning featured an early concentration upon the Upper Tigris, to which he ventured four times. 32 This likely began in 908 BC, and was certainly concluded in 21 Cf. Postgate 1995: 1-2. That pāḫatu 'province' and ḫalzu were not conceived of as exactly interchangeable is demonstrated by MARV 4, 119 (cf. Llop 2012: 93). 22 Cf. Jakob 2003 This raises the fraught question of Bīt-Zamāni's etymology, and hence ethnic composition; Zadok posits a Hurrian etymology to explain the earliness of this attestation (1991: 113), while Lipiński elects for a conventional Semitic interpretation (2000: 135-136). The latter is probably to be preferred, but for want of more information this must remain open. 24 Note the same's appearance, along with his son, at Šibaniba, although it remains unclear as to whether this was coterminous with his posting there or not (cf. Machinist 1982: 22-23). 25 A.0.101.1 i 101-103 (= Grayson 1991: 200). Cf. Liverani 1992: 99;Postgate 1995: 1. 26 The end of the Giricano archive around 1068 BC being a case in point (Radner 2004: 115). 27 Note Fales' recent emendation of this toponym (2012: 103). 28 The present author considers the White Obelisk an inscription of Aššur-nāṣir-apli I; this is founded not only in the incompatibility of its account with the early reign Aššur-nāṣir-apli II, but also in its affinities with the other texts presently attributed to Aššur-nāṣirapli I (cf. discussion in Frahm 2009: 117-123). 29 A.0.101.18 '18-'33 (= Grayson 1991: 256). 30 Brown 2013, cf. Roaf andSchachner 2005. 31 A.0.98.1. 33-41; A.0.98.2 17 '-22' (= Grayson 1991: 133-134;137). See Radner 2006a on passage through the Ṭūr 'Abdīn. Generally overlooked are the eastern routes into the Upper Tigris region, such as those offered via the plains of Şırnak and Siirt. While arduous for a large army such as Xenophon's Ten Thousand, they would not have posed exceptional difficulties to messengers or small caravans. The course of the Tigris itself can also be followed; Layard did so, heading southwards by way of Çelikköy and descending into the plain of Cizre at Fındık (Layard 1853: 50-51). See also Comfort and Marciak 2018: 34-41. 32 A.0.99.2 30 (= Grayson 1991: 148). 903 BC. 33 The first two and probably the fourth of his campaigns are fragmentarily extant, 34 describing chevauchées into the Upper Tigris. At some point during these campaigns, he annexed three formerly Assyrian towns which had fallen to Šubria. 35 It is only come Tukultī-Ninurta II that Bīt-Zamāni swims into focus; the close of a campaign in 887 BC presents a unique occurrence within the annals, a miniature campaign of sorts conducted by the son of Ammi-ba'lī against one Bialasi in Udu of the land of Nirdun on Assyria's behalf, forwarding the spoils to Assyria. 36 That a campaign by a local polity ostensibly performed on the Assyrian king's behest would make it into the annals is astounding enough, but is easily surpassed by the events of the following year.
In 886 BC, a figure with a fragmentary name and title 37 wrote to Tukultī-Ninurta II stating that Bīt-Zamāni had further aspirations in the region, seeking, indeed, to march through the Kāšiāri. Tukultī-Ninurta II pre-empted this and ravaged the countryside around the city of Patiškun belonging to Bīt-Zamāni, depriving the region of grain and putting Ammi-ba'lī's sons to the sword. This harrying must have driven Bīt-Zamāni to negotiate; Ammiba'lī submitted, and Assyria took spoils from his palace. Most striking, however, is that Tukultī-Ninurta II had its ruler Ammi-ba'lī swear that Bīt-Zamāni would no longer sell horses to any power but Assyria. 38 This singular agreement was accompanied by two other unusual developments, namely that Assyrian officials were stationed within Bīt-Zamāni, and the displaced population of Bīt-Zamāni was peacefully internally resettled by Tukultī-Ninurta II. 39 The significance of this reference to horses cannot be stressed enough. 40 Assyria's anxieties that rival polities may gain a superiority in horses likely fuelled her endless campaigns to the Zagros even during the first half of the 8 th century BC. 41 While Ammi-ba'lī delivered tribute in 882, 42 he was assassinated in a putsch in 879 BC by the nobles of Bīt-Zamāni and one Bur-Rammān, which prompted Aššur-nāṣir-apli II to return to the region, have Bur-Rammān flayed at Sinābu, confiscate the polity's considerable wealth, impose a much higher tribute, and deport some 1500 Ahlamean soldiers in Ammi-ba'lī's pay to Assyria. 43 Ilānu, Ammi-ba'lī's brother, 44 was installed as the new client ruler.
Following the leanly attested mid-period of the king's reign, during which he likely fought inconclusive skirmishes with the transeuphratine polity of Bīt-Adini and then undertook his much-lauded 'March to the Sea', Aššur-nāṣir-apli II returned to the region for the last time, burning his way along the Euphrates and annexing the 33 This analysis, and that of Tukultī-Ninurta II's campaigns which follows is the result of the present author's ascription of five unattributed royal inscriptions published by Eckart Frahm (2009) to Adad-nārārī II and Tukultī-Ninurta II, as outlined in the appendix of the present author's unpublished doctorate (Edmonds 2018). A discrete publication of these findings is in preparation.  20-24 (= Grayson 1991: 171-172). This exceptional situation strangely blends vassalage and provincialisation, with Assyria seeking to influence Bīt-Zamāni internally far more than was usual with a client kingdom, see the following section. 40 Sworn before the local divinity of Adad, this may be the earliest attestation for the later commonplace adê oath. It should be recalled that Tukultī-Ninurta II's reign also witnesses the first evidence of the use of cavalry by the Assyrian army (A.0.100.5 37 = Grayson 1991: 173). 41 The efforts made by the Assyrians to maintain a superiority in horses are striking cf. Āl-sūsānī, i.e. 'horse trainer-town' (Bagg 2017: 26). Other evidence of the value placed upon horses includes ritual activity to protect them within the Assyrian army (Maul 2013). 42 A.0.101.1 ii 12 (= Grayson 1991: 202). 43 See Edmonds 2019a for discussion of Ahlameans in the Iron Age. 44 The ambiguous wording of the annals has led many to consider Ilānu Bur-Rammān's brother despite the counter intuitiveness of this arrangement, cf. Sano (2015). region of Mallānu by the modern Karacadağ. 45 Damdammusa had defected to Bīt-Zamāni and it was forced to surrender Ilānu's men, whom Aššur-nāṣir-apli II proceeded to impale before the city of Amēdu. Despite this shock and awe, 46 Aššur-nāṣir-apli II's final assault on Amēdu proved all but fruitless, save for the orchards that he hacked down. 47 On the route home, the king pillaged the most likely unsuspecting settlement of Udu nestled in the Ṭūr 'Abdīn, 48 presumably to collect some token loot both to placate his army and to gild an otherwise lukewarm triumphal return to Assyria. The absence of any further annalistic accounts of campaigns for the remainder of this monarch's reign perhaps intimates to the scrappy nature of his final years, likely preoccupied with the outbreak of war with Urarṭu. 49 Ten years later and some 175 miles or so to the southwest, Salmānu-ašarēd III departed from Tīl-Barsip, 50 marching up the Euphrates to strike at the very heart of Urarṭu by way of the Upper Tigris, a deed immortalised in a royal epic still read in the last years of the Assyrian Empire. 51 His passage through Bīt-Zamāni was uneventful, as was that of his turtānu Dayyān-Aššur in 830 BC, some 19 years after the first attestation of a governor there.
What this survey of the interactions of Assyria with Bīt-Zamāni and their occasional hostilities demonstrates is the striking dynamism of this supertigridine Aramean polity. Not only did it undertake its own campaigning, but it even threatened to march over the Kāšiāri. In turn, it successfully enticed Damdammusa to its side, and made other territorial gains. 52 Its wealth is apparent from its ability to field chariotry and its stores of metals. In turn, its nobility seemed repeatedly keen to rebel from Assyria, perhaps more from confidence than desperation. This is demonstrated by the absence of the city of Amēdu from Assyrian sources prior to 866 BC; its strategic position was easily defensible 53 and all Assyria could do was to ravage the countryside around. It seems unlikely that Aššur-nāṣir-apli II succeeded in conquering this difficult city during the remainder of his reign without leaving an account, and the absence of more than passing mention of the polity in Salmānu-ašarēd III's annals is also telling. Most likely, a diplomatic solution of some sort was undertaken by Assyria between 866 and 856 BC.

BĪT-ZAMĀNI IN THE PROVINCIALISATION OF THE UPPER TIGRIS
That some manner of Assyrian 'pre-provincial' 54 structure was already in place in the Upper Tigris prior to the inauguration of Tušḫan in 879 BC is evidenced from various intimations of such within the textual record. Firstly, Adad-nārārī II's early annexation of three formerly Assyrian towns which had fallen to Šubria 55 must be 45 A.0.101.1 iii 101 (= Grayson 1991: 220). 46 The city's resistance is a remarkable example of the potential ineffectiveness of Aššur-nāṣir-apli II's 'calculated frightfulness' (cf. Olmstead 1918). 47 A.0.101.1 iii 109 (=Grayson 1991: 220). See Cole 1997 for discussion of this common Assyrian tactic. 48 Already the target of Bīt-Zamāni's campaigning in 887 BC. 49 As has been convincingly argued by de Filippi, the variance between Aššur-nāṣir-apli II's geographical summaries stating 'to Nērbi' and 'to Urarṭu' clearly demonstrates a significant later campaign to Urarṭu by this king (1977). Cf. Grayson's commentary to A.0.101.1 iii 122 (Grayson 1991: 221 considered. 56 Secondly, Tukultī-Ninurta II was warned of Bīt-Zamāni's imminent march across the Ṭūr 'Abdīn in 886 BC by a figure who may have been a governor. 57 The candidates for such an individual within the 9 th century Upper Hābūr are thin, 58 and a 'man on the spot' in the Upper Tigris seems more likely. 59 In turn, following his confrontation of Bīt-Zamāni during the same campaign and swearing of oaths, Tukultī-Ninurta II installed officials to supervise the polity; 60 it seems unlikely that such an implementation would have been effective were there not an existing Assyrian administration in the region to support them.
Proceeding from these inferences, concrete candidates for a location for this might be considered. Within Aššurnāṣir-apli II's Kurḫ Monolith, the tribute gathering of 879 BC is described as having been centred upon four cities, Tušḫan, Sinābu, Tīdu, and Damdammusa; 61 the first of these is qualified as having just been rebuilt after a period of decline, 62 while Sinābu and Tīdu are stated to have been reclaimed from the 'Arameans' the same year. 63 It must hence be concluded that Damdammusa was the only significant Assyrian-held settlement in the Upper Tigris prior to this episode, and hence the focal point of Assyrian control; indeed, Damdammusa is termed an āl šarrūtīya 'city of my kingship' in Aššur-nāṣir-apli II's annals, implying that it contained an Assyrian royal residence. 64 In turn, the Kurḫ Monolith not only relates the establishment of Tušḫan, but also the provincialisation of regions of Na'iri. 65 These would logically have been assigned to Sinābu considering its later synonymy with the province of Na'iri. 66 It hence seems apparent that both Sinābu/Na'iri and Tušḫan were formally established in 879 BC, and that this constellation replaced a previous situation in which Damdammusa was the key Assyrian possession in the region, irrespective of whether or not it is to be considered an early province; the explanation for this is to be found in the rebellion of Ḫūlāya.
In 882 BC, Aššur-nāṣir-apli II received a report that one Ḫūlāya, lord of the relict Assyrians of Ḫalziluḫa had been courting the city of Damdammusa. The inhabitants of Ḫalziluḫa, the descendants of Assyrians settled by Salmānu-ašarēd I, 67 would have been largely left to fend for themselves and co-operate with neighbouring groups such as Hurrians, Arameans, or ḫabḫu-folk, creating a manner of independent 'frontier spirit'. 68 Something of the relative egalitarianism usually exhibited by offshoot societies may well be reflected in the name of the leader of the 56 The nearest province to which they could have been appended at this early date would have been Katmuḫu on the other side of the Kāšiāri. It is conceivable that 'pre-provincial' Assyrian territory was notionally a direct possession of the king, which could explain the frequent use of ana ramānīya aṣbat in the annals and the assignment of rough border zones to members of the royal court. 57 A.0.100.5 24-25 (= Grayson 1991: 171-172). Grayson reads GÌR ? .ARAD ? , i.e. šakkanakku. Should this reading be correct, then it would be highly unusual; this highly antiquated title was restricted to the Assyrian king's titulary during the Neo-Assyrian period.
wayward Assyrians, Ḫūlāya, 'The one of the road' or 'Roadling', possibly a name for a foundling. 69 They most likely resented the return of external Assyrian power to the region. The close of this rebellion saw it quashed and Ḫūlāya flayed at Damdammusa. The two major flayings of this period recorded, Ḫūlāya at Damdammusa in 882 and Bur-Rammān at Sinābu in 879 BC, both occurred at key locations; that the king meted justice over Bīt-Zamāni from Sinābu is telling considering both the Middle Assyrian past and that these territories would be amalgamated but decades in the future into a single province.
Aššur-nāṣir-apli II's creation of a new provincial structure in the aftermath of this insurrection sought to end the power monopoly of this less than reliable Assyrian city. This, in turn, neatly explains Damdammusa's defection to Bīt-Zamāni around 866 BC; having lost its previously privileged status in the region, the city sided with its wealthy neighbour. In the wake of this provincial reorganisation and Aššur-nāṣir-apli II's failure before Amēdu in 866, it may be presumed that they did not undergo any further territorial changes until Sinābu absorbed Bīt-Zamāni.
856 BC witnessed Salmānu-ašarēd III's march through Bīt-Zamāni as detailed, and the first known governor of Na'iri is attested thereafter as the eponym for 849 BC, one Ḫadi-libbušu or Iḫtadi-libbušu. For many scholars, this serves as the terminus ante quem for Bīt-Zamāni's annexation. The next attested governor of the region is Ninurta-kibsī-uṣur, eponym for 838 BC albeit still governor of Raṣappa then. 70 His governance of Amēdu is suggested rather by Aššur Stele no. 47, wherein he is styled šāqiu rabiu and ascribed the holdings of Andi, Sinābu, Suḫna, Mallānu and Alzu; 71 that these constituted the province of Amēdu is demonstrated by the Aššur stele of the next known incumbent, one Marduk-šimanni, eponym for 799, which is near-identical in its toponymy. 72 Ninurta-kibsī-uṣur's stint in charge of Na'iri presents some interesting features. The title of šāqiu rabiu is first attested in the reign of Aššur-nāṣir-apli II and appears to have existed parallel to the better-known office of rab šāqê, or cupbearer. 73 Considering that Mulissu-mukannišat-Nīnua, queen of Aššur-nāṣir-apli II and Salmānuašarēd III, was the daughter of the earlier šāqiu rabiu Aššur-nīrka-da''in, it might be assumed that this was an influential position. In turn, the province of Raṣappa had grown to become a vast and distended province by Salmānu-ašarēd III's time, encompassing the lower course of the Ḫābūr, the Middle Euphrates down to Sūḫu, and the Sinğār by this period, the province effectively dividing the extreme west of Assyria's realm from the heartland. 74 It may well be that Ninurta-kibsī-uṣur's appointment to this post and the assignment of the province of Na'iri's territory was an attempt by the king to alter the balance of power between his various magnates and to sever Ninurta-kibsī-uṣur's connection with his powerbase. 75 Another potential thesis for Ninurta-kibsī-uṣur's posting is that Salmānu-ašarēd III may have been experimenting with the creation of marcher provinces along Assyria's wilder borders, as is far better known from the late 69 Consider the analogous foundling names Suqā'a or Šulâ/Šulâya 'One of the street, Streetling' common in Babylonia, the latter also borne by a hapless messenger to the Zagros in a likely portion of the Na'id-ŠīḪU Epic (Edmonds 2019b: 329-330). Ḫūlāya would be the logical equivalent of such a name within a more rural setting. The assignment of distinct names to foundlings is frequently attested within ancient cultures; besides the mythical example of Oedipus, 'swollen-foot' , in reference to his laming on his abandonment, the infamous case of the assignment of 'copronyms' to infants found on dunghills in Hellenistic Egypt serves as a particularly striking example (cf. Pomeroy 1986). 70 Finkel and Reade wish to amend the entry for 838 BC which displays r]a-ṣap-pa to Na'iri on grounds of Ninurta-kibsī-uṣur appearing as the governor of Na'iri on Aššur Stele no. 47 (= Andrae 1913: 53-54, cf. Millard 1994 and suspect that dittography is responsible for this writing (Finkel and Reade 1998: 248), but this is unnecessary. 71 Andrae 1913: 53-54. 72 Andrae 1913: 49. 73 Mattila 2000: 47-48. 74 See Radner 2006b: 52-53 and recently Parpola 2017. After reaching its zenith under Pālil-ereš, who may even have sought to incorporate Sūḫu into Assyria under his own initiative, the province was subdivided. While Laqû had become a discrete province by 736 BC at the latest, Ḫalzi-adbāri's mention in Tukultī-apil-Ešarra III's inscription at Mila Mergi from 739 BC would imply an even earlier division, should this province be localised in the Sinğār, formerly Raṣappa's north-easternmost reaches, a proposition which is, however, still uncertain, as Cizre is probably the better source of basalt (adbāru) within the vicinity of ancient Ulluba. 75 Indeed, the latter half of Salmānu-ašarēd III's reign is already characterised by the emergence of powerful officials jockeying for positions, cf. esp. Fuchs 2008; Grayson 1994. Neo-Assyrian period. 76 In 830 BC, Salmānu-ašarēd III's turtānu Dayyān-Aššur passed through Bīt-Zamāni once more on campaign to Urarṭu, ultimately trouncing Sarduri I. 77 Regardless of when precisely Ninurta-kibsī-uṣur's Amidine incumbency had begun, it was likely brought to a close with the outbreak of the succession war in 826 which engulfed Assyria in Salmānu-ašarēd III's terminal years.
An interesting archaeological correlate in this context is a perceived reduction in activity at Ziyaret Tepe following the reign of Salmānu-ašarēd III, 78 perhaps due to the chaos of the rebellion, or perhaps from the region bracing itself for a coming Urarṭian onslaught. To the Assyrians' credit, it would seem that this defensive strategy largely succeeded; save a foray by Minua, the Urarṭians ultimately circumvented the region and extended their influence down to the Syro-Aramaic polities of the west by way of the western bank of the Euphrates until Tukultī-apil-Ešarra III dramatically broke their hold west of the Euphrates in the opening years of his reign. With 745 BC and the advent of empire proper, the present history of the region may conclude. A final interesting point is the survival of the name Bīt-Zamāni into the later stages of the empire, not only in the eponym lists and imperial correspondence, 79 but also in an Aramaic missal. 80 The later profusion of names for this province, i.e. Na'iri, Amēdu, Sinābu, and Bīt-Zamāni is striking. It may well be that this phenomenon genuinely belies a battle of precedence between Sinābu/Na'iri, the senior settlement, erstwhile capital, and 'Assyrian' face of the province, and Amēdu/Bīt-Zamāni, the natural city from which to govern, 81 but also an Aramean addition to the original province. Having presented these two histories, the figure of Ḫadi-/Iḫtadi-libbušu may now be investigated.

ḪADI-LIBBUŠU OR IḪTADI-LIBBUŠU, THE FIRST GOVERNOR OF NA'IRI
Little ink has been spilt on Ḫadi-libbušu or Iḫtadi-libbušu, 82 and, quite surprisingly, the peculiar variation in his name has not yet been explicitly discussed within academic literature. It is first expedient to consider the various attestations of this figure and their details.
As might be noted, these attestations are all associated with his eponymate. Outside of date formulae, nothing is known presently to have been written of him by Assyrian hand. Nonetheless, something of his origins and career might be inferred; his name serves as a reasonable point of departure.
Firstly, it should be noted that the individual Ḫadi-libbušu (Akk. 'His heart is joyful') also occurs as Iḫtadilibbušu (Akk. 'His heart rejoiced') within the Neo-Assyrian textual record; these variants present differing forms of the same verb in Akkadian, ḫadû, 'to rejoice'. While to the present author's best knowledge otherwise unattested, both are plausible and grammatically correct Akkadian personal names. In turn, despite these two names' clear semantic propinquity, they are both phonologically and graphically distinct enough that arbitrary confusion between them seems unlikely.
While it is common to misremember an individual's name for one perhaps better known, 83 it must be recalled that Ḫadi-/Iḫtadi-libbušu's name appears solely in the context of his eponymate. His name would have been used 76 Support for this notion is lent by the die of the masennu Aya-ḫālu (or Yaḫālu), mentioning his governorship of the difficult westernmost flank of the Zagros (cf. Millard 1994: 8-front.). 77 A.0.102.14 141-146;A.0.102.16 228'-267' (= Grayson 1996: 69, 81). 78 Cf. Köroğlu 2016. 79 For example, Nasḫir-bēl, governor of Amēdu in Šarru-ukīn's reign, himself refers to his province as such in correspondence. 80 Mirrored in an attestation of bny zmn in an Aramaic document from Tall Šiūḫ Fawqānī (cf. Fales et al. 2005: 609, fn. 101). 81 On a clear day, one can still see across the entire Upper Tigris basin from Diyarbakır's citadel today. 82 See Ambos 2000. Remarkably, Lipiński's otherwise exhaustive discussion of Bīt-Zamāni omits any mention of Ḫadi-/Iḫtadi-libbušu (Lipiński 2000: 160-161). This name is to be distinguished from the otherwise attested and near homophonous apotropaicon Ḫādêlīpušū 'Let the malevolent do (as they wish)!' , perhaps to be read Ḫādê-lipūšū 'May they scorn the ill-wishers!' . 83 Correspondence postal, telephonic, and electronic addressed to one 'Mister Edwards' is a strange and irksome constant in the present author's life. universally within the Assyrian administration not only during the year of his incumbency, but also for years thereafter, 84 this rendering the confusion between these two forms even stranger; sheer legal and administrative imperative would dictate a modicum of uniformity.
The present author should like to propose the following solution in light of these remarks, namely that this name had been translated from Aramaic, and that an ambiguity in the writing of the original name provoked these variant appellations; indeed, should the names Ḫadi-libbušu and Iḫtadi-libbušu be translated into Old Aramaic as a thought experiment, 85 then the results are as follows:

Akkadian
Old Aramaic Aramaic orthography Ḫadi-libbušu 86 *ḥādɛ-libbeh 87 *ḥdhlbbh Iḫtadi-libbušu 88 *ḥadā-libbeh 89 *ḥdhlbbh As is evident, while both Aramaic forms of the name are easily distinguished aurally, they are nonetheless graphically identical. Without foreknowledge or context, a reader cannot deduce the pronunciation or semantics of this name due to its ambiguous writing, then as now. While Assyrian royal annals and private and administrative documents alike can sport strange manglings of Aramaic personal names, this individual's name had been chosen to date a year, and hence its precise rendering would have been necessary. Accordingly, the present author suggests 84 That Iḫtadi-libbušu was merely misremembered as Ḫadi-libbušu by the compilers of the eponym lists and chronicles perhaps centuries later is effectively discounted by the Billa tablet, composed only four years after the eponymate, when the year name would have remained in the collective memory. 85 The Old Aramaic vocalisation in Folmer (2011)  that the scribes of the Assyrian administration resolved to translate this individual's name into good Akkadian; 90 unfortunately, however, there were two competing schools of thought, and both translations became common within the administration before a single rendering could be standardised. 91 As is immediately apparent, underlying this interpretation of an otherwise perplexing state-wide administrative contradiction is the crucial premise that the name was known only known to the scribes of the Assyrian bureaucracy in Aramaic writing and not orally. Nevertheless, Ḫadi-/Iḫtadi-libbušu's position as governor of an Aramaic-speaking province on the far edge of Assyria's extent renders this not implausible. In turn, it is no great step to assume that Ḫadi-/Iḫtadi-libbušu hailed from this very region, and, indeed, that he was, in fact, also the local ruler following Ilānu. This inference is rendered all the more plausible when the similar situation within another Neo-Assyrian province of the 9 th century BC is considered, Bīt-Baḫiāni/Gūzāna, a so-called 'transitional case'.

BĪT-ZAMĀNI AS A TRANSITIONAL CASE
J. Nicholas Postgate's seminal article 'The Land of Assur and the Yoke of Assur' 92 first introduced what he termed 'transitional cases', client states exhibiting some but not all of the hallmarks of incorporation into the Neo-Assyrian Empire during the 9 th century. 93 While various examples of this may be found within the Neo-Assyrian textual and archaeological record, 94 the present author should like to focus upon the prominent example of Bīt-Baḫiāni/Gūzāna.
The Aramean polity of Bīt-Baḫiāni coalesced within the radius of the newly founded Iron Age settlement of Gūzāna 95 in the 11 th to 10 th centuries BC in a region with a long history of imperial endeavour, most saliently embodied in the nearby settlement of Sikānu, formerly the Mitannian capital of Waššukkan(n)i, which had also played host to a Middle Assyrian administration before finally adopting a primarily cultic function during the early Iron Age. 96 Gūzāna swiftly became affluent should the somewhat outlandish monumental structure of Kapara be anything to go by.
In stark contrast to its more easterly neighbours, Naṣibīna and Gidāra/Raqmat, Bīt-Baḫiāni did not war with Assyria in the 9 th century BC, but rather remained circumspect, rendering tribute when approached by Adad-nārārī II, but allowing him access solely to Sikānu. 97 Aššur-nāṣir-apli II's annals also present an image of docile clientdom. 98 90 The translation of Aramaic personal names into Akkadian is a well attested phenomenon, the most famous example being the later Assyrian queen Naqī'a/Zakūtu. 91 The implications that this bears for the workings of the Assyrian administration in the 9 th century are discussed at this article's close. Various solutions present themselves for the peculiar writing ḫa-di-i beyond mere scribal mistake, none of which is particularly enticing. Firstly, considering that a break ensues, it is plausible that this is, in fact, a hypochoristicon, i.e. Ḫadî, although a cursory perusal of the eponym lists did not yield any other examples of commensurate informality in year-naming. Another solution is that a scribe took ḫa-di as a nominal form within an omitted anticipatory genitive construction, i.e (Ša-)ḫadî-libbušu 'The joy (of ) his heart' , and thus spelt it plene as ḫa-di-i. Finally, at a very considerable stretch, ḫa-di-i could be taken as an Assyrianised 2. f. s. imp. D, thus Ḫaddî-libbušu 'Make him happy!' , the obvious downside of this being the male addressee. Of these, the second option is perhaps the least miserable. 92 Postgate 1992. 93 'This was not the same as the later practice of attaching an Assyrian agent to a local court, best attested in the Phoenician ports. Rather, it entailed replacing the local ruler by, or converting him into, a 'governor' answerable to the king, but probably not the incorporation of his territory or local administration into the Assyrian system proper, as they do not appear to have been provincial governors of the regular variety' (Postgate 1992: 257). No mention is made of the polity within annalistic accounts of Salmānu-ašarēd III's campaigning, and it is assumed to have been incorporated into the empire during this period. Indeed, its first governor, Šamaš-nūrī, is attested as eponym for 866 BC, the next, Adad-rēmanni, for 841 BC. Finally, an Assyrian campaign to the city is known from 808 BC, the region having evidently gone its own way during the chaos of the latter quarter of the 9 th century.
A tidy end to Bīt-Baḫiāni's sovereignty in 866 BC at the very latest would be a done deal were it not for a celebrated bilingual inscription of Adad-it'i/Adda-yis'ī, son of Šamaš-nūrī, in Aramaic and Akkadian discovered at Tall Faḫḫārīya. 99 As is now well known, the inscription styles both of them as governor, šaknu, in the Akkadian, and as king, malk, within the Aramaic. It is this ambiguity between languages and the lack of any concomitant political events which renders Bīt-Baḫiāni a transitional case within the expansion of the early Neo-Assyrian Empire; the local rulers had been dubbed governors within the Neo-Assyrian 'co-prosperity sphere', but evidently continued to depict themselves as Aramean kings to their own populace.
The question of translation and ambiguity runs even deeper in this case, and becomes at once highly pertinent to the present study inasmuch as the common Assyrian name Adad-rēmanni 'Adad have mercy on me!' of 841 BC's limmu is not dissimilar to the Aramaic name of the Faḫḫārīya inscription's commissioner, Adda-yis'ī, 'Adda is my help'. This has prompted the suggestion that Adad-it'i/Adda-yis'ī's name had been translated into Akkadian 100 for the sake of eponym dating. Beyond a chronological congruency, 101 this is supported in the present author's mind by a crucial point, namely that while Adda-yis'ī's name is rendered Adad-it'i in Akkadian, this is in the cuneiform of his own, local inscription; it cannot be proven that the Assyrian central bureaucracy at all referred to him as such. 102 In turn, the same inscription reveals Šamaš-nūrī to have been Sās-nūrī (ssnwry) all along. 103 The uprising ending in 808 BC, after which Gūzāna would have found itself under the governance of the obviously Assyrian Mannu-kī-Aššur, thus spelt the end of a local dynasty. Even then, Gūzāna would rebel again between 759 and 758 BC, and Bīt-Baḫiāni is even referenced in the Book of Isaiah, this demonstrating an enduring local identity. 104 The parallels to Bīt-Zamāni's fate are at once apparent, and it might reasonably be suggested that Amēdu was an Upper Tigridian example of such a 'transitional case'. Both cases evidence affluent and powerful Aramean polities with traditions rooted in the Late Bronze Age. In both cases, no forceful takeover can be evidenced, and, in turn, the governors are most likely local rulers in Assyrian guise whose names were translated into Akkadian for the eponym list. 105 This last point must be examined against the backdrop of the history of Aramaic within early Neo-Assyrian administration. 99 See Dušek and Mynářová 2016 for a recent edition of this inscription and discussion. 100 'This quite common type of Akkadian proper names may have been used as an Assyrian adaptation of Hadd-yiṯ'i's Aramaic name. Incidentally, no other eponym of that period bears a name with the theophorous element Hadd/Adad.' (Lipiński 2000: 129). Dornauer concurs with this assessment (2010: 57). 101 Adad-it'i must otherwise be shoehorned in between his father in 866 and Adad-remanni in 841; it must be hypothesised thereby that the junking of the local dynasty did not spell any unrest worthy of mention in extant Assyrian sources, but that a rebellion occurred come 808 BC for reasons unrelated. By means of comparison, the endurance of the local ruling family until their final replacement by Mannu-kī-Aššur raises little in the way of further difficulties, and is hence to be preferred for want of evidence to the contrary. 102 Note Younger's erroneous objection on grounds that '[t]he Akkadian scribes were entirely capable of translating the Aramaic name into Akkadian: Adad-it'i' (Younger Jr. 2016: 265). 'Adad-it'i' is not a translation of 'Adda-yis'ī' , but rather a partial 'Akkadisation' , as 'it'i' is merely a phonetic rendering of 'yis'ī' without semantic import. 103 Due to the closeness and occasional mutual comprehensibility of these languages, the distinctions between a phonetic rendering of an Aramaic name by means of cuneiform, the 'Akkadianisation' of an Aramaic name, and its outright 'translation' are often ambiguous; whether Sās-nūrī himself would have considered 'Šamaš-nūrī' an Akkadisation or a translation of his name is unclear. 104 Demsky 2008. 105 While there are eponyms with Aramaic names already attested in the 9 th century, such as Il-milkī and Aya-ḫālu the masennu, and perhaps Yarî, there is no evidence that these had previously been local rulers. Indeed, should they have belonged to the Assyrian court, as Aya-ḫalu must have, then an entirely different milieu from the governors of 'transitional cases' could be assumed for these individuals; current scarcity of evidence precludes judgement.

ON THE USE OF ARAMAIC IN THE EARLY NEO-ASSYRIAN STATE
A final issue raised by the thesis of this paper is that of the use of Aramaic and the Aramaic script within the 9 th century Neo-Assyrian sphere. As has been contended, the confusion between Ḫadi-libbušu and Iḫtadi-libbušu could only have occurred if the bureaucracy of the Neo-Assyrian heartland was in possession of the first governor of Amēdu's name in Aramaic script but were unsure as to its vocalisation in Aramaic, and hence created two competing Akkadian translations of his name. This implies that Ḫadi-/Iḫtadi-libbušu's correspondence as governor would have been written in Aramaic, and whatever of it reached the Assyrian heartland's bureaucracy was not concertedly, centrally, or systematically translated into Akkadian, but rather consulted on the fly by scribes, and presumably translated as required, if at all; moreover, were there to have been a 'card index' of Assyrian potentates then the first governor of Na'iri had evidently been either omitted or double filed. This finding has considerable bearing both on the history of the development of the Aramaic script, and on its use within Neo-Assyrian administration.
The precise dissemination of the Aramaic alphabet eastwards from Phoenicia to the Upper Ḫābūr and finally the Assyrian world is difficult to trace. 106 Something of a milestone for present purposes is an inscribed altar from Tall Ḥalaf which dates to the early 9 th century, 107 a few decades at most before the creation of the Faḫḫārīya Inscription. 108 By means of contrast, Kapara's inscriptions of the previous century had been in crude Akkadian. In the case of Bīt-Zamāni, Tukultī-Ninurta II's annals exceptionally appear to quote verbatim a letter from Ammiba'lī's son to the Assyrian king; 109 unfortunately, its brevity and broken state permit scarce judgement on whether this had been written in Akkadian with Aramaicisms or in Aramaic and subsequently translated for the annals. 110 Turning to Assyria, intimations as to the advent and degree of adoption of Aramaic within the Neo-Assyrian state apparatus are scant. 111 Perhaps the earliest traces of Aramaic within an institutional context are the bricklayer's marks found at Fort Shalmaneser at Kalḫu, 112 albeit these were for the benefit of workmen, not scribes. 113 Come the era of Adad-nārārī III, the Nimrud Wine Lists attest for the first time to Aramean scribes within a palatial setting. 114 In turn, the first depiction of the thereafter common trope upon reliefs of two scribes counting spoil, one with tablet and stylus, the other with pen and scroll, hails from Tukultī-apil-Ešarra III's Central Palace at Kalḫu. 115 The tight proximity of this pair in all attested examples implies some manner of institutionalised tandem documentation. With the famous letter from Šarru-ukīn demanding letters be written in Akkadian rather than Aramaic, 116 the floodgates were well and truly open. 117 Rare Aramaic texts such as the Aššur Ostracon imply that colleagues within the administration might tend to write each other missals on other media in between com-106 See discussion in Gzella 2015: 57-63. 107 KAI 309, cf. Dankwarth and Müller 1988. 108 Cf. linguistic discussion in Gzella 2015: 63-67. 109 A.0.100.5 5-6 (= Grayson 1991: 171). 110 nota bene, however, Lipiński's interesting reconstruction of a phrase within this passage as ištu Udu ana Šūaru ana siliḫi irtedīma 'he went from Udu to Šūaru within this javelin's throw' which he takes as an Aramaicism analogous to a Hebrew phrase found twice in the Book of Job (Lipiński 2000: 139-140). There is no Akkadian word siliḫu, and this would have to have come from Aramaic šlaḥ, itself only weakly attested. 111 Elsen-Novák and Novák (2020: 146-147) present an interesting, albeit highly tenuous equivalence of the script of Taymā' as mentioned by Yariri of Carchemish with the Temanite Arameans of the Upper Ḫābūr and hence the Aramaic script. Were such an admittedly conjectural thesis proven to hold, then it would imply an earlier dispersal of the Aramaic script eastwards into the Assyrian sphere than otherwise thought. 112 Millard 2008: 268. 113 Considering that the bricks formed a cuneiform inscription when assembled correctly together, however, it may well be that a scribe literate in cuneiform had been the one who had patiently annotated these bricks with Aramaic letters prior to their transport to the building site so as to prevent any mix-ups. 114 CTN 1, no. 9, rev. 20 (=Wilson 1972: 138). 115 See discussion in Reade 2012: 706. 116 SAA 17, 3 (= Dietrich 2003: 5-6). 117 Cf. Beaulieu 2006. posing tablets, 118 while themes known later from Aramaic 'romances' may already have permeated Assyrian literary compositions. 119 Considering the present overview, the correspondence from Ḫadi-/Iḫtadi-libbušu, although not extant, would nonetheless serve as the earliest present attestation for the use or at least reception of Aramaic within the Neo-Assyrian state's administration.

CONCLUSION
The elements are now all present to present a new portrait of Bīt-Zamāni's incorporation within the māt Aššur. In the first historical section, it was demonstrated that Bīt-Zamāni was a more powerful force in the Upper Tigris than is usually depicted. Not only was it affluent, but it also pursued an active foreign policy, campaigning within the Upper Tigris basin and even planning to march over the Kāšiāri against Assyria herself. In the following section on the provincial history of the region, it has been established that the political reality in the Upper Tigris was more complicated than the pristine establishment of the provinces of Tušḫan in 879 BC and Na'iri in 856 BC respectively often advanced. Rather, an Assyrian administration at Damdammusa existing perhaps already in the reign of Adad-nārārī II was replaced in 879 BC with two provinces, Na'iri/Sinābu and Tušḫan, in response to difficulties with the local Assyrian population. While Na'iri/Sinābu did not encompass Bīt-Zamāni/ Amēdu, the meting of justice to Bur-Rammān on Sinābu's walls already presaged a provincial relationship akin to that during the Middle Assyrian period. In turn, although Aššur-nāṣir-apli II's invasion of Bīt-Zamāni due to Damdammusa's insurrection proved futile, the Aramean polity was once more under heel come 856, and had a governor by 849 BC.
These two overlapping histories converge in this very figure, Ḫadi-/Iḫtadi-libbušu, who, it has been argued, possessed an Aramaic name which had been translated in conflicting manners into Akkadian by Assyria's central bureaucracy. This, in turn, suggests that he was an indigenous ruler promoted to governor, an example of the Postgatian 'transitional case'. The clear parallels to Bīt-Baḫiāni/Gūzāna, another Aramean polity on the edge of the Assyrian pale, would seem to support this. Indeed, it has been argued that the names of its first two governors, whose incumbencies coincide with that of Ḫadi-/Iḫtadi-libbušu, also had their names translated into Akkadian for the benefit of the eponym systems. This translation implies that their correspondence would have been in Aramaic, and that Assyrian scribes translated this ad hoc and as required within their bureaucracy.
From these findings, the date and nature of Bīt-Zamāni/Amēdu's annexation might finally be resolved. With hostilities with Urarṭu on the horizon, Aššur-nāṣir-apli II made a startling volte face and reconciled with Ilānu at some point in the years following his failed siege of 866 BC, making him governor of the province of Na'iri/ Sinābu/Amēdu/Bīt-Zamāni and thus amalgamating the Assyrian and Aramean holdings into an uneasy single territory. 120 While a compromise, it ensured the region's safety, much as the ḫalzu of Bīt-Zamāni had in the Middle Assyrian era, and access to the equines vital to Assyria's military. This situation continued with Ḫadi-/Iḫtadilibbušu, who was even 'honoured' with an eponymy, and the indigenous dynasty was only unseated with Salmānuašarēd III's installation of the powerful Ninurta-kibsī-uṣur at some point after 838 BC, most likely in 830 BC when Dayyān-Aššur passed through the region. This situation likely provoked Amēdu's subsequent revolt during the succession war of 826-820 BC, 121 after which it was likely placed under Assyrian governorship again, Mardukšimanni appearing as its governor and eponym in 799 BC. As a 'transitional case' between perhaps already 865 and 118 Cf. Fales 2010. 119 Edmonds 2019b: 344. 120 This could potentially explain the broken line of the Salmānu-ašarēd III Epic wherein it is stated 'Aššur-nāṣir-apli harnessed and mobilized the land of Na'iri [ …]' (SAA 3 17, 16 = Livingstone 1989: 44). 121 Note that this presents a potential harmonisation of the rebellion of Amēdu during the succession war of 826-820 BC: Its stint as a 'transitional case' had only just concluded and the Šamšī-Adad V's scribes still considered it such in their list of rebellious cities. Or, it could be presumed that Ninurta-kibsī-uṣur never actually arrived at his new posting prior to the rebellion's outbreak. 830 BC, the date of its actual annexation depends entirely on one's own definition. This finding naturally bears repercussions for the further investigation of Assyria's expansions, especially during the 9 th century.
As has been demonstrated, contradictions and ambiguities in the governance of the early Neo-Assyrian realm are to be understood as precisely such. A coherent doctrine of Assyrian expansion as known from the reign of Tukultī-apil-Ešarra III onwards with its irreversible three-step program of initial contact to client state to Assyrian province struggles to explain the complexities of the 9 th century, and has yielded controversy over Bīt-Zamāni's annexation. Diachronic and thematic studies of the early Neo-Assyrian period grounded in an understanding of Middle Assyrian precedent 122 are the only corrective to the counterproductive imposition of the paradigmatic strictures of late Neo-Assyrian expansion upon these fragmentary and byzantine historical scenarios.
These findings are also a vital qualification to the study of Assyrian imperial 'discourse', essentially the most recent iteration of the 'propaganda' school in Assyrian historiography, demonstrating the potential degree of misunderstanding between the Assyrian heartland and its own peripheries; if Ḫadi-/Iḫtadi-libbušu ever consumed the stuff of Assyrian annals, a doubtful prospect, then it was probably in Aramaic. Indeed, while it must be presumed that this governor of the province of Na'iri visited Aššur to hobnob with the king and his other magnates at New Year's, dragomen presumably in tow, it can only be speculated as to the contents of his limmu stele, should it ever be found; were an inscription of his to be unearthed at the citadel of Diyarbakır, however, then the present author would wager it reading ḥdhlbbh mlk 'md.