5. Mari and the Amorite Age: The City and Its Music

The kinnāru is next attested in the eighteenth century at Mari (Tell Harīri) on the middle Euphrates. The city’s massive archive makes it a type-site for the political dynamics and economic complexities of the period. There is rich evidence for an ‘international’ music-culture, much like that of Ebla or Ur under Shulgi, but currently known in much greater detail. [1] I would stress that, while the larger scope of this study justifies my focus on the kinnāru material, these lyres were but one element of Mari’s diverse instrumentarium. All the same, the kinnāru’s linguistic association with the West gives it a special position vis-à-vis the OB city. For this was the so-called Amorite Age, when dynasts of western extraction held power in many Mesopotamian cities—most famously Hammurabi, who ultimately destroyed Mari. [2]
Mari was apparently subject to significant Sumerian musical influence in pre-Sargonic times, to judge from the famous statue of Ur-Nanshe who bore, in addition to the title NAR, both a Sumerian name and professional garb like that worn by singers on the ‘Standard of Ur’ and elsewhere. [3] This same Ur-Nanshe, earlier in his career, may be among a group of visiting Mariot singers attested at Ebla, of whom at least nine have names that are recognizably Sumerian, presumably adopted as being professionally appropriate. [4]
The sparse administrative texts of the so-called Shakkanakku period (ca. 2100–1850), which record isolated distributions to singers (NAR), chief singers (NAR.GAL), and songstresses (NAR.MÍ), are clear traces of a lively and stratified music-culture. [5]
But it is the reigns of two later kings—the interloper Yasmah-Addu (ca. 1790–1776) and the restored Zimri-Lim (ca. 1775–1761)—for which we have much material detailing the royal management of music. The texts, which yield abundant evidence for artisan mobility generally, [6] include numerous musical contacts with Karkemish, Babylon, Aleppo, Qatna, Hazor, and elsewhere. The city’s musical affairs were directed by a Chief Singer (Akk. nargallum), typically a foremost confidant of the king and often appearing among other high officials in the economic documents. [7] Based in a sort of conservatory (mummum), his duties included recruiting and training harem musicians (often from war captives), [8] supervising the construction and repair of instruments, [9] organizing musical ensembles and events, and undertaking sensitive diplomatic missions like arranging royal marriages. [10] Zimri-Lim even left his Chief Singer, Warad-Ilishu, in charge of the city while taking the field against Eshnunna. [11] (Compare the unnamed singer whom Agamemnon left in charge of Klytaimnestra at Mycenae. [12] ) Enough Chief Singers are attested for other states in contact with Mari for us to conclude that such officials were quite typical of this period. [13] Mari’s musical apparatus, if not identical to that of other states, must have been compatible for all practical purposes.
Although systematic records for male musicians are lacking, there are examples of them receiving land allotments from the king, and other indications of esteem. [14] The management of female musicians, however, may be reconstructed in considerable detail from a series of administrative texts, which, though not completely continuous, span many years. This was a relatively stable, self-sustaining system, with singers maintaining their careers in the face of dynastic change, although naturally the individual was ever vulnerable to royal whim. Thus, while Ilshu-Ibbishu, chief musical instructor under Yasmah-Addu, does not reappear in the records of Zimri-Lim, [15] Rishiya, the Chief Singer under Yasmah-Addu, continued for a time in this office after Zimri-Lim’s restoration. [16] Similarly, the same three female music-teachers were apparently active under both kings training harem-musiciennes. [17] This helps explain why, despite numerous demotions when Yasmah-Addu’s harem was integrated into that of Zimri-Lim, four young girls of the previous regime emerged as full-fledged musiciennes in the new. [18] Evidently they had not only come to sexual maturity, but completed their musical training, for which they were duly rewarded.
It is clear that an ‘international style’ of music was deliberately cultivated, with the foreign and exotic carefully recorded as though important for an accurate inventory. [19] In an age without sound recording, the craving for musical variety was satisfied through the mechanism of royal gift-exchange. [20] In practice this involved the acquisition, training, and trading of players. One set of texts deals with a heavily armed caravan, supervised by Zimri-Lim’s Chief Singer, which escorted a group of ‘Benjaminite’ musiciennes to the king of Aleppo. [21] Another tablet refers to the integration of a group of Elamite musiciennes into the harem. [22] The need to have a ready stockpile of these ‘commodities’ accounts for the surprisingly high numbers maintained by the palace—at least 200 in the reign of Zimri-Lim, managed by several dozen Senior Musiciennes. [23] Music was thus one of the ‘household industries’ that contributed to a larger interpalatial economy. Women might even be trained in a specific foreign style: Zimri-Lim committed captives from Ashlakka to a ‘Subarian’ musical education. From the contemporary Mariot perspective, this probably means Hurrian, so that the Ashlakkans would be cultivating their own traditions for the enrichment of musical life at Mari. [24]

The Kinnāru at Mari

This cosmopolitan musical environment is important for fully appreciating the position at Mari of the kinnāru, relatively well attested in the city’s administrative texts, some of which deal with the building, decoration, and maintenance of musical instruments. A letter to Zimri-Lim from one of his officials is a status-report on a royal order for five kinnāru-lyres; two were ready for delivery to the king, while another three were behind schedule. [25] Another text records an allocation of a kind of varnish for two kinnāru-lyres. [26] A third, gold given to adorn several instruments, including a kinnāru. [27] A fourth mentions a kind of skin or leather allotted for various cult-objects, with several instruments; here two kinnāru-lyres are found alongside instruments of foreign provenance or associations. [28] This further attests an openness, at the highest social levels, to the blending or juxtaposition of different music-streams, whether in concerted or consecutive performances.
Among the musiciennes known to have served members of the nobility outside the royal palace, some are mentioned as playing the kinnāru. [29] This is earliest textual evidence for a recurring pattern, from the Levantine lyre-girls of NK Egypt to the female lyre-ensembles of the Cypro-Phoenician symposium bowls, the kinnōr-playing ‘harlot’ of Isaiah, and beyond. [30] Vis-à-vis Mari itself, one may note an unprovenanced terracotta plaque of the OB period which shows a naked woman standing on a pedestal and playing a box lyre with curving arms in nearly horizontal position; before her a man dances and plays a frame-drum (Figure 4.1i). [31]
This plaque’s lyre has clear morphological affinities with instruments of Syria and the Levant. Nor is it alone. There are five further OB representations of such instruments, and others from the Ur III and even OAkk. periods. [32] While such lyres must have been current in Amorite traditions, [33] this was probably not an exclusive association; the chronological spread of the material indicates a more general, long-term musical interaction between Syria and southern Mesopotamia. Kinnāru(m) must have been one name by which these instruments were known. [34] We saw that the glossing of BALAĜ as kinnārum at ED Ebla is mirrored by the roughly contemporary presence of kinnāru in Mesopotamian lexical tradition. [35] While the cognate zannāru evidently prevailed in southern Mesopotamia during the OAkk. and Ur III periods, P. Michalowski has plausibly suggested that the currency of kinnāru at OB Mari may not be due solely to Amorite influence, but represent the same larger regional usage that accounts for kinnārum at Ebla. [36]
While kinnāru has not yet appeared in the extant portions of any OB lexical text, it most probably was present in some strands of the scribal tradition in this period. This is the readiest explanation for a pair of parallel passages in MB (later second millennium [37] ) exemplars of two distinct lexical series, found in areas outside of Babylonia itself. [38] From HAR.ra=hubullu (Ḫh), [39] as it was known at the Syrian site of Emar on the upper Euphrates in the fourteenth or thirteenth century, comes the following triad of equations:
Ḫh, Emar [40]    
za-anMÙŠ za-na-ru: qà-an tá-bi-tum [41]
za-anMÙŠ ki-in-na-ru  
za-anMÙŠ ti-in-du-u  
MÙŠ functions here as a logogram, one way of designating the goddess Inanna; the signs ZA.AN are plausibly taken by M. Gantzert as a phonetic gloss of the underlying pronunciation (the first part of zannāru). [42] Each of the three entries here was thus considered a variety of ‘Inanna-instrument’, and closely comparable or akin. One may note here the OAkk. seal, which shows a bull-lyre played before the goddess Inanna/Ishtar. [43] Further permutations of ‘Inanna-instrument’ (gišza.dInanna, gišzà.mí dInanna, etc.) are known from other Mesopotamian lexical texts, and lyres are commonly so described in Hittite sources. [44] Note that Zannaru also occurs as one of the names by which Ishtar was known in a passage of the Hymn to the Queen of Nippur (a MB cento of earlier sources); as Zannaru she was “the wise/skillful goddess” and “honored by Dagan,” the latter phrase suggesting a special connection with the middle Euphrates and North Syria. [45]
Now the scribal traditions of Emar and other cities of the so-called western periphery—including Ugarit—are known to derive from those of OB southern Mesopotamia. [46] But because they also contain a number of local innovations and expansions, it was natural, when our passage of Emar Ḫh was viewed in isolation, to see kinnāru as a secondary elaboration of the entry for zannāru. [47] But the same sequence has now turned up in a tablet from Assur (Assyria), part of the series Diri = (w)atru (a difficult collection of compound logograms) which also derives from an OB tradition of the south: [48]
Diri, Assur [49]    
[z]a-an-na-ru GIŠ.ZA.MÙŠ zannāru
[ ] [GIŠ.ZA].MÙŠ kinnāru
[ ] [GIŠ.ZA].MÙŠ tindû
Here too, presumably, ZA is a phonetic gloss, and GIŠ the determinative ‘wood’. The agreement of Ḫh and Diri is so close that the passages should be considered duplicates. [50] Although such correspondences are not extensive in OB exemplars of Ḫh and Diri, there are enough “to believe that the two compositions influenced each other” (the direction of influence is unclear in any given case). [51] The simplest conclusion to be drawn from this material is that kinnāru, if it did not persist from the third millennium in Mesopotamian lexical tradition, re-entered two or more branches in the OB period thanks to the instrument’s currency in the Amorite age. It then ‘returned’ to the western peripheral cities as part of the scribal tradition, and passed independently northward into Assyria. Whatever the explanation, the probable OB scribal currency of kinnāru has interesting implications for the treatment of the Divine Kinnaru in the pantheon texts of Ugarit. [52]

The Amorite Connection

One must in any case assume, on general historical grounds, some integration of Amorite musical traditions into a wider Mesopotamian music-stream. We should therefore examine the broader cultural phenomenon in more detail, as it may further illuminate the position of the kinnāru in this time and place, and ultimately perhaps help account for its divinization at Ugarit and elsewhere in the West.
Increasing numbers of Amorite names in Ur III texts indicate a gradual process of infiltration and integration into Mesopotamian society during the late third millennium. [53] In the first centuries of the second, many Mesopotamian cities were controlled by dynasties of Amorite descent. The whole complex process is reflected in several compositions dealing with the god Martu, a Sumerocentric eponym for the Amorite parvenus. [54] In the so-called Marriage of Martu, he is presented as a powerful but crude barbarian seeking a bride among the local gentry. [55] His people are mocked for their nomadic ways, eating raw meat, ignorance of the gods and their temples, and the inability to recite prayers. [56] The text’s conclusion is broken, but presumably Martu succeeded in his suit, and so at last won a place in polite society. There are also two hymns to Martu, much like those for other Sumerian gods, which seem to show his integration into the Mesopotamian divine order. While keeping his savage power, he is accepted as a son of An, favored by Enlil, and ranked among the great gods. [57] As such he now enjoys the perquisites of civilization, including a ‘normal’ cult:
In holy songs musicians sing of him—the dearly cherished one, the god, the man of the hills, renowned everywhere—and promote his name gloriously. Martu, son of An, it is sweet to praise you! [58]
These texts make it reasonable to suppose the Amorite adoption or adaptation of Mesopotamian liturgical practices in the course of their acculturation. This inference is supported by the case of Ishme-Dagan, the fourth king of Isin, who, while bearing an Amorite name, promoted a late flowering of Sumerianizing literary activity, including more than twenty royal hymns. [59] In one of these is found a “direct imitation” [60] of Shulgi’s boasts of musical prowess—including expert command of the zannāru:
I have devoted myself to the art of singing, and know the occasions when praise songs are to be sung. That I am eminent in the performance style for … songs; that I know how to intersperse appropriate words with the accompaniment of the fingers and instruments; that I have mastered the drumsticks, the sa-eš, the sabitum, the ḫarḫar and the zanaru instruments; that I have completely mastered the developed aspects of the art of singing and the recondite points of … songs—all these things the scholars and the composers of my … songs have put in my great songs and have declared in my hymns. [61]
If such posturing was largely symbolic, it was so within a Neo-Sumerian ideological framework; the important point is its purposeful resurrection by Ishme-Dagan, which he grounded in a real cultural program. [62]
Similarly the king of Mari continued to promote Sumerian cult-music. We find a music-instructor requesting a gift from the king for teaching his students balang-compositions, while an Ishtar ritual, discussed below, contains cues for a number of Sumerian lamentation-songs. [63]
The Amorites’ own ancestral culture, however, is practically invisible in the archaeological and written record. With the exception of PNs, their language went largely undocumented, since textual production continued to be in Akkadian and literary Sumerian. Nevertheless, the Amorites and their kings evidently maintained a sense of distinct ethnic identity. This may be inferred first from a large corpus of PNs, often theophoric (and so giving limited theological information), or containing words relating to social structure, tribal ancestries, and a semi-nomadic cultural background. [64] A relatively high proportion of Mariot scribes, carriers of literate Mesopotamian culture, nevertheless bore Amorite names. Constant political relations with the West probably further encouraged the preservation of inherited traditions, and perhaps the mother tongue, for some part of the OB period. [65] The case of Zimri-Lim is suggestive: as a young exile he went west to Aleppo and the court of Yarim-Lim, the powerful king of Yamhad who became his father-in-law through interdynastic marriage to the princess Shibtu.
It would hardly be surprising if, in the cosmopolitan musical environment that is clearly seen in the administrative texts, the kings of Mari equally cultivated their Amorite musical heritage. An event of special interest here is the arrival at Mari of a caravan from Hazor carrying three Amorite musicians, for whom Zimri-Lim exchanged three of his own musiciennes. [66] Whether ‘Amorite’ here has an ethno-linguistic or only a geographical sense is unclear, but it is likely enough to be both. [67] The text is equally valuable for tying Canaan, otherwise poorly documented in MBA texts, into a larger, cosmopolitan musical world. From the earlier reign of Yasmah-Addu, a letter from the chief musical instructor complains about the quality of Amorite musiciennes who had been brought back from a westward military expedition in support of Qatna (“They are all truly cold and old. There’s not a single woman among them!”). [68] They were six in number, a small subset of the ninety-four women whose training he managed. [69] The ‘Benjaminite’ musiciennes sent by Zimri-Lim to Aleppo, mentioned above, are also probably relevant, as the Banu-Yamina was an Amorite tribal group in the area of Mari. [70] These texts indicate an awareness of Amorite music as something identifiable across a considerable geographical range. It was a distinct strand within the complex web of musical traditions that were brought together and elaborated by the royal courts of Mari, Qatna, Aleppo, and elsewhere.
The fairly prominent position of the kinnāru at Mari thus becomes more intelligible. The presence of these lyres among other cult-objects indicates that they were being constructed not solely for novelty entertainment in the harem, but for more lofty roles.
We saw that the Bible, in crediting its kinnōr’s invention to Jubal, not only traces it to the distant past, but makes it part of a semi-nomadic lifestyle; and while in isolation this may seem a romantic anachronism, the portrait is generally confirmed by the Beni-Hassan tomb-painting. [71] A comparable intersection of music-making and a semi-legendary nomadic past may also be inferred behind king-lists relating to Amorite dynasts of Babylon and Assyria.
The Assyrian King List famously lists seventeen ancestral “kings who dwelled in tents.” [72] This section agrees partially with the Genealogy of the Hammurabi Dynasty (GHD), where several pairs of cognate rhyming names reveal that the two texts derive from a common heritage of myth-making, and suggest “that this segment of the tradition was originally preserved as some kind of desert chant—perhaps as part of oral epic of early tribal heroes.” [73] The ancient ‘kings’ of these texts are in reality a variety of tribal eponyms—names and relationships that were probably gradually recomposed over the generations to reflect shifting political and social patterns in relations between various Amorite groups and the urban states with which they interacted. [74] One of the names (Ditānu/Didānu) resurfaces at Ugarit as a quasi-deity and seemingly an ancestor of the royal line. [75]
The fluidity of this material would indeed accord with a derivation from some form of oral epic tradition. Numerous ethnographic analogies show that musical accompaniment is often involved in such narrative singing; most common are stringed instruments, which provide in a single convenient package both tonal material and the rhythm essential for structuring verse. Note that the performance medium for the preservation of the GHD was royal ancestor cult, internal evidence showing that it was used in the course of a kispu ritual, when food and drink were offered to the ghosts of kings past. [76] We encountered something of the sort at Ebla, where I argued for the involvement of kinnārum-threnody; similarly at Ugarit I will explore the symbolic importance of the kinnāru in royal mortuary cult.

Divine Instruments and the Amorite World

The third-millennium practice of divinizing instruments continued strong in the OB period. Years named after the dedication of lilissu-drums are known for Immerum of Sippar (ca. 1845), Itur-Šamaš of Kisurra (ca. 2138), and Iter-Piša of Isin (ca. 1833–1831). [77] An inscription of Warad-Sin of Larsa (ca. 1834–1823) commemorated a lilissu (and perhaps also a BALAĜ) “dedicated for his own life and for the life of his father.” [78] Years were also named from the dedication of ala-drums by two early nineteenth-century kings, one in the temple of Zababa by Iawium of Kish, [79] the other in that of Nanna by Manana of Urum/Ilip. [80]
An administrative text from Mari records a large consignment of silver and gold sent to Tuttul, in northern Syria, for decorating ‘Ninigizibara’. [81] We saw this name earlier applied to an instrument built and dedicated by Ibbi-Sin of Ur. [82] Ninigizibara is otherwise well known from Ur III and OB god-lists, where it can appear as a “GU4.BALAĜ advisor of Inanna.” [83] Animal offerings to Ninigizibara are also attested at Uruk, once “on the occasion of a lamentation rite which accompanied a circumambulation starting from the gate of the Ĝipar,” and in a number of other ritual contexts. [84]
Most striking of all, as a comparandum for Kinyras, is a detail in the OB balang-composition (or ‘oratorio’) Uru’amma’irabi (That City Which Has Been Pillaged), in which Inanna lamented the destruction of her city, temple, and the balang Ninigizibara itself, as well as Dumuzi’s infidelity and death. [85] In this work, Ninigizibara seems to be treated as Inanna’s husband or lover, “shar[ing] Inanna’s bedroom as an intimate partner of the goddess.” [86] Other laments show that it was a trope of these compositions to include the balang, or its hall, among what has been destroyed in a city or temple. [87]
Uru’amma’irabi also featured in a ritual performed at Mari itself during the reign of Yasmah-Addu, which focused on the balang Ninigizibara. This text gives us our most detailed glimpse of how divinized instruments might serve in complex ceremonies, in this case a sequence of lamentation rites involving the king himself. [88] On the last day of an unnamed month, the goddess’s temple was purified and the instrument set up before her. The king and other participants, including lamentation-priests and ensembles of male and female musicians, were carefully arranged around and facing Ninigizibara, which was itself flanked by various cultic symbols. [89] An elaborate series of rites then unfolded, punctuated by lamentation-singing; the structure of this ritual seems to be informed by the sequence of elements in Uru’amma’irabi itself. [90] Laments were somehow conjoined with prophecy by an ‘ecstatic’ (muhhûm), although the precise relationship between the two practices is unclear. [91]
It may be that Ninigizibara was not actually played in these performances, but was the object of song “as a representation of Inanna herself … in her aspect as a lamenting goddess.” [92] This interpretation would fit nicely with the idea, attested in both Greek and ANE sources, that lamentable situations like war and royal deaths are times when lyre-music should be stilled. [93] At the same time, the participating ensembles make it quite certain that instrumental music, including strings, was indeed heard. Thus, while Ninigizibara itself might represent the stilling of music, the ritual as a whole will have used music to reverse the divine mood—the normal function of lamentation singing. [94] Perhaps sympathetic vibration played a role here, since an unplayed Ninigizibara would still murmur in response to the music of others. Was this seen as evidence that the instrument was indeed alive, had its own voice, and was itself lamenting?
The Mari tablet, which records silver and gold for Ninigizibara, makes it quite possible that, with other texts involving precious metals for musical instruments, we are again dealing with divinized specimens. We saw above that the kinnāru is attested in just such a context. Indirect support for the deduction comes from the urzababîtum included in the same transaction, which transpired under Zimri-Lim. For another Mari text, this time from the reign of Yasmah-Addu, in referring to a group of instruments ready for royal ‘audition,’ mentions the incomplete status of an urzababîtum whose name incorporates that of the king himself: Samsi-Yasmah-Addu, ‘Yasmah-Addu is my Sun’. [95] This must be a divinized instrument. The name, to judge from the parallels, seems to place Yasmah-Addu in the position of a master-god who will be served by the urzababîtum. [96] We encountered another such ‘King of Kish instrument’ in the texts of Shulgi, and one appeared in a god-list as servant of Ninurta; the complex model of ‘musical cognition’ this implies brings together king, past king, and the divine through the medium of royal music. [97] To find yet another urzababîtum in the service of Mari’s monarch, and even bearing his own name, raises interesting questions about the intersection of musical ideology and the tradition—that is, the handing down—of royal power. [98] It may even be that Yasmah-Addu’s urzababîtum is the very one which Zimri-Lim—his successor from a rival dynasty—caused to be adorned.
Finally one must note a Babylonian royal inscription relating to the fortieth year of Hammurabi’s reign. The king marked his defeat of Zimri-Lim and the destruction of Mari by dedicating two musical instruments and a standard in the Emeslam, a temple of Nergal in Kutha (a day’s ride northeast of Babylon):
Eternal seed of kingship, mighty king, king of Babylon, king of all the Amorite land, king of Sumer and Akkad, when he captured Mari and its villages, destroyed its wall, and turned the land into ru[bble heaps (and) ru]ins, he set up a BALAĜ and a bronze kettledrum (for) holy songs, which please the heart, etc. [99]
We saw two earlier dedications of a BALAĜ, one by Gudea of Lagash, the other by Ibbi-Sin, last of the Ur III emperors. These events were of sufficient political importance to give their names to the year in question. These parallels underscore the gravity of Hammurabi’s action, and show this self-consciously Amorite king to be equally a pious perpetuator of Mesopotamian cult practice (“king of all the Amorite land, king of Sumer and Akkad”). [100] The context also permits reasonable guesses about the significance of the dedication. That this was considered the right gesture to punctuate the king’s final triumph over Zimri-Lim is indicated, first, by the text’s immediate juxtaposition of the two events. Furthermore, the god Nergal, to whom the instruments are devoted, has already been invoked as “the terrifying king who [goes] at the head of the troops, who annihilates the enemy lands.” With Mari’s defeat, Hammurabi’s long and careful expansionist career reached a successful climax, giving him unrivalled control over the Babylonian heartland and the eastern stretches of the Amorite cultural sphere. [101] The instruments may therefore be seen as a gesture of thanksgiving to Nergal, on the one hand, and a symbol of Hammurabi’s New World Order on the other—with the vanquished enemy ushering in an age of peaceful, festive music, and the end of lamentation.
Three key points remain uncertain. First, we cannot be sure that the BALAĜ of this text means ‘lyre’, for this is the period in which it seems to have made its transition to a kind of drum. [102] It may also be that BALAĜ functions merely as a determinative, qualifying the lilissu in some way. [103] Nor can we be certain that this BALAĜ was divinized, although the dedicatory context makes this probable. Finally, it is possible that BALAĜ may conceal some more properly Amorite instrument name, just as it was glossed as kinnārum by the scribes of Ebla and in the ED Practical Vocabulary. [104] If one could infer such an equation here, it might be that Hammurabi was expressing his triumph over a rival Amorite king with a gesture that was not only devout from a traditional Mesopotamian perspective, but also symbolically potent within the Amorite cultural continuum.


Amorite integration in Mesopotamia during the Ur III period and OB periods, combined with the continuing sense of Amorite identity across a wide geographical range, together provide a favorable environment for the emergence of a Divine Kinnaru. [105] The Martu texts hint at the assimilation of Amorite cult to Mesopotamian liturgical practices. Mari, where we can most clearly document Amorite traditions surrounded by ancient Mesopotamian cult practices—including divinized instruments—should be considered a type-site in this respect too. The presence of an ecstatic prophet within the Ishtar/Ninigizibara ritual above is a suggestive case of West-meets-East. [106] So too the lexical equation of kinnāru and zannāru (probably of OB date, but certainly MB), where their definition as ‘Inanna-instrument’ surely implies some theological interpretation of the lyre—for typically divine balangs reflected and embodied various facets of their master gods. [107] It should be noted that Byblos, one traditional home of Kinyras, was within the cultural orbit of the Ur III emperors (Chapter 19). The practice of divinizing lyres and other cult-objects may have been generally adopted among the Amorites and other Syro-Levantine kings of the late third and early second millennium as part of a conscious emulation of the ideological and cultural models of contemporary Babylonia. The kinnāru’s importance in royal cult—implicit at Ebla and Mari, and more clearly demonstrable elsewhere from the LBA onwards as the following chapters will show—provides a plausible motivation, with the instrument’s deep antiquity being matched by that of the royal cults themselves.


[ back ] 1. This material has been well analyzed in several recent studies by N. Ziegler: FM 4; Ziegler 2006a; Ziegler 2006b; FM 9. Other relevant discussions include: Williamson 1969; von Soden 1988; Malamat 1999; Malamat 2003. For the ED period, see also Tonietti 1998.
[ back ] 2. See further below. I use ‘Amorite’ advisedly, well aware of the current debate about the nature of Amorite ethnicity, the degree to which those so described in Mesopotamia (Sum. MAR.TU/Akk. amurrum) identified themselves as a coherent group, and so on (recently Miglio 2013:189–197). Nonetheless, the term remains a useful shorthand for discussing cultural and demographic patterns of the period.
[ back ] 3. Ur-Nanshe: Damascus S 2071, Parrot 1967:88, fig. 127–131 and pl. 45–46; Gelb and Kienast 1990:13–14; Braun-Holzinger 1991:249; FM 9:7–9; RIME 1, 10.12.3. Parallels: MgB 2/2:44–45 (Standard of Ur), 48–49 (fig. 11–12, 17), etc.
[ back ] 4. Mander 1988:482; Steinkeller 1993:237–238, 240; Tonietti 1988:88–89; Archi et al. 1988:283; Tonietti 1998:91; Archi 2006:98.
[ back ] 5. Limet 1976:7–9, 36 (dating), 28 (singers), and text-references in index.
[ back ] 6. Sasson 1968; Durand 1992 passim; Zaccagnini 1983b.
[ back ] 7. See generally FM 9:7–12, 83–201. As a boy, Yasmah-Addu seems to have been tutored by Ibbi-Ilabrat, the Chief Singer of his father Shamshi-Addu, and, once king of Mari, he appointed his friend Rishiya to the office (FM 9:83–88, 148–149 and n109–110).
[ back ] 8. FM 9:10–11, 42–43, 168–169, 180–189; Ziegler 2010:119–126. The term ‘harem’, despite its orientalist connotations, is both convenient and appropriate; for a defense of its use vis-à-vis OB Mari, see FM 4:5–8; for Achaemenid Persia and other ANE contexts, see Llewellyn-Jones 2013:97–102. The distribution lists do not distinguish sharply among queen, secondary wives, royal princesses, musiciennes, and a variety of domestic staff. The ‘harem’ thus comprised all female residents of a palace, not all of whom served as concubines.
[ back ] 9. FM 4:30n173; Ziegler 2006a:345n6, 348, and n31; FM 9:86, 170, 190–193.
[ back ] 10. Ziegler 2006a:348; FM 9:11, 86–89, 149–151, 171–175.
[ back ] 11. FM 9:200–201; cf. 176–179 for a synopsis of Warad-Ilishu’s attested activities, which equally show his high standing.
[ back ] 12. Homer Odyssey 3.267–272.
[ back ] 13. Ziegler 2006a:347n21, citing examples from Shubat-Enlil, Ekallatum, and Karana; FM 9:10.
[ back ] 14. Ziegler 2006a:347; FM 9:20–31.
[ back ] 15. FM 9:205.
[ back ] 16. Durand 1988:95–117; FM 4:96; Ziegler 2006a:348n33; FM 9:83, 97 et passim.
[ back ] 17. FM 4:82–83; FM 9:15–16.
[ back ] 18. FM 4:12, 35–38.
[ back ] 19. Ziegler 2006a; FM 9:19–20, 37–41 et passim.
[ back ] 20. This point was stressed by R. Eichmann in discussion at the 2006 meeting of the International Study Group on Music Archaeology in Berlin.
[ back ] 21. FM 9:172–175, 194–199 (no. 45–46). For the sense of ‘Benjaminite’, see p82.
[ back ] 22. FM 9:120–122 (no. 21).
[ back ] 23. FM 4:71–72, 78–79, 81–82, 94–96, and n527, 116–118, 120–122; Ziegler 2006a:346–347; Ziegler 2006b:36. Between thirty-two and thirty-five Senior Musiciennes (MUNUS.NAR.GAL) are always listed immediately after the royal family, an indication of their high social standing. A main body of Junior Musiciennes (MUNUS.NAR.TUR) follows. Next comes a certain Izamu, principal second wife of Yasmah-Addu, who once enjoyed the title ‘Servant of the King’ and led a group of former wives and singers of the deposed ruler. Some of these were probably Zimri-Lim’s own sisters or cousins, whom Shamshi-Addu had committed to musical training after his acquisition of Mari: FM 4:12n59, 70, 76–79. Ziegler suggests that the giving away of harem musiciennes was further motivated by the need to avoid incest (FM 9:15).
[ back ] 24. FM 9:18, 20, 168–169; Ziegler 2010:16. It seems clear that Ashlakka was in the area known to the Mariots as Subartu; and that while ‘Subartu’ itself underwent several semantic shifts in different periods (Michalowski 1986; Michalowski 1999), it included for the Mariots the Hurrian zone of the upper Habur triangle: see Finkelstein 1955:2–3.
[ back ] 25. ARM 13 20.5, 7, 11, 16 = J. Bottéro in Dossin et al. 1964:39, with brief comment on 162; FM 9:72–73 (no. 11); cf. Ellermeier 1970:77; Krebernik 1983:21; HKm:97 and n196. For royal ‘audition’ of instruments, cf. FM 9:220–221 (no. 53), 6.
[ back ] 26. ARM 23 180.12 = Bardet et al. 1984:174–175.
[ back ] 27. ARM 25 547.9 = Limet 1986:171–172 with FM 9:72 and n247. Among the several other instruments mentioned here is an urzababîtum, for which see p35–36. For gold and silver adornment, see further Heimpel, “Balang-Gods,” 23g, 34c, 42c.
[ back ] 28. ARM 21 298, lines 16, 20 = Durand 1983:370–371, cf. comments on 367–368, including the mysterious material šinûntum. This text is largely duplicated by ARM 23 213 = Bardet et al. 1984:189–191 (kinnāru in line 31–32), cf. 140. See also FM 9:71–72n245. The tilmuttu is the ‘instrument from Dilmun’ (Baurain, or the east coast of the Persian Gulf: Howard-Carter 1987). The paraḫsitu or parašitu is ‘the instrument from Marhashi’, an area of the Iranian plateau. These instruments have been identified as lyres (or other chordophones) partly on the basis of their adjacency to the kinnāru here (they also appear in lexical lists alongside other instruments): Durand 1983:368; cf. Stauder 1970:217; von Soden 1988; FM 4:70n465; Ziegler 2006a:352. Steinkeller 2006:7–10 assembles the textual evidence for the parašitu and proposes an identification with the horizontal harps commonly depicted in pre-Iranian art of the region. But note that they appear to be an exclusively female instrument at Mari: FM 9:49.
[ back ] 29. FM 4:70n465, 221–222 (no. 42.4–5); Ziegler 2006a:347n18; FM 9:41, 50.
[ back ] 30. See p61, 105–111, 245, 250, 258–272, 302.
[ back ] 31. Berlin, Vorderasiatisches Museum, VA 7224: MgB 2/2:76–77 fig. 59; DCPIL:44 fig. 1(i); AMEL:60 fig. 21. A suggested connection with Inanna-Dumuzi cult is doubtful: MgB 2/2:76 (following Moortgat); embraced by MAIP:76.
[ back ] 32. OAkk. cylinder seal: RlA 6:581 (Collon, *Leier B, §2 II.5.a); MgB 2/2:64–65 (fig. 43). Ur III figurine: MgB 2/2:66–67 (fig. 47–48); AMEL:38–39 and fig. 14; DCPIL:44 fig. 1(h). OB material: MgB 2/2:76–77 (fig. 59), 90–91 (fig. 76–77, 80); AMEL:59–60.
[ back ] 33. HMI:79; Stauder 1961:12–19; AMEL:38–39.
[ back ] 34. This is implicit in Lawergren’s analysis, who in DCPIL includes them among his ‘thin lyres’, which he connects with kinnāru (58–59).
[ back ] 35. See above p54, 79 and Heimpel, “Balang-Gods,” 4c.
[ back ] 36. Michalowski 2010b:122.
[ back ] 37. In this usage ‘Middle Babylonian’ is chronological, not geographical.
[ back ] 38. I thank †Joan Westenholz and Yoram Cohen for confirming the likelihood of this point (communication, November 2011 and July 2013, respectively).
[ back ] 39. HAR.ra is now also read/known as ur5-ra: see e.g. Veldhuis 1997:46.
[ back ] 40. Arnaud 1987 no. 545, lines 391–393’ (p. 76); I print the text as it appears in Gantzert 2008, I:102 (Ḫh VI.10–12), also I:118 and II:65 (composite text, entry 4253a–c), where the first two signs are interpreted as a phonetic gloss (see below). Note that the wood determinative is written only at beginning and end of the column: Civil 1989:14. Dating: Pentiuc 2001:10; Civil 2004:5.
[ back ] 41. Cohen 2010:825–826, has recently reread this ‘second gloss’ as qà-an tá-bi-tum (see CAD s.v. timbuttu: stringed instrument), thus correcting ka-[a]n-da-bi-tu4 in the texts of Arnaud and Gantzert.
[ back ] 42. I owe this reference to Sam Mirelman (communication, June 2013).
[ back ] 43. MgB 2/2:64–65, fig. 42. See further Heimpel, “Balang Gods,” p574.
[ back ] 44. For Hittite, see p89–96. Material collected by CAD s.v. zannaru; discussions in Sjöberg 1965:64–65; MSL 6:119, 123, n81, 142; Lambert 1982:213; RlA 6:573 (Kilmer, *Leier A); Lawergren and Gurney 1987:41; Krispijn 1990:12; HKm:98.
[ back ] 45. Lambert 1982 for text and commentary: Zannaru at III.65–68, cf. IV.28, where Inanna is “offspring of Dagan.” The most important cult-site of Dagan was at Tuttul, at the confluence of the Euphrates and Balikh rivers; he received state worship at Ebla, and theophoric PNs show that he was venerated throughout northern Syria, including Mari and as far north as Tell Beydar, from at least late pre-Sargonic times onwards: Feliu 2003:8–41. An equation of Zannaru here with the zannāru instrument—which could also accord with her being named by and “beloved of” Ea/Enki—is assumed by CAD s.v.; MSL 6:119, 123, and n81; Sjöberg 1965:64–65; questioned by Lambert 1982:213. I thank †Joan Westenholz for feedback here (2/2011).
[ back ] 46. Veldhuis 1997:67, 70–71, stressing that the relationships between OB traditions and their MB descendants on the western periphery cannot be precisely determined before the lexical texts from Ugarit are fully published. For the latter, see van Soldt 1995:171–175.
[ back ] 47. Pentiuc 2001:98; DCPIL:59; Michalowski 2010b:122.
[ back ] 48. For Diri generally, see Veldhuis 1997:56, 117; Civil 2004:4–6.
[ back ] 49. Diri III.043–045: I follow the mise-en-page of Civil 2004:138. His text as presented is, like a Greek or Latin edition, an editorial composite from several exemplars. See his textual notes on 139 for the reading ki]n-na-ru in Assur 11884 (exemplar E: key on 134); cf. 6, “the significant sources for [Diri III] are all M[iddle] A[ssyrian].” The text is in the so-called 1-2-4 format (Civil 2004:4), where the first column is a ‘reading gloss’ or phonological ‘description’ of the compound ideogram in column 2; the final column contained the Akk. translation. For lexical text formats, see generally CANE:2305–2314 (Civil).
[ back ] 50. Cf. Veldhuis 1997:118 (on two other parallels from OB texts of Ḫh and Proto-Diri): “Since the item is repeated three times both in ur5-ra [i.e. Ḫh] and in Proto-Diri we may safely assume that the sections duplicate.”
[ back ] 51. See Veldhuis 1997:118–120.
[ back ] 52. See p121–122.
[ back ] 53. Buccellati 1966, with the problems raised by Michalowski 2011:82–121.
[ back ] 54. See generally Klein 1997; Pongratz-Leisten 2011:93–94; other texts, in which Amorites are characterized in similar terms as the portrait of Martu to be discussed, are collected in Buccellati 1966:89–95, cf. 330–332.
[ back ] 55. Ironically, the bride herself seems to have an Amorite name—dAdgarudu < Ashratu, cognate with Athirat: Cross 1973:57; Smith and Pitard 2009:377.
[ back ] 56. Marriage of Martu (ETCSL 1.7.1), 126–141. Römer 1989; Kramer 1990; Klein 1993; Klein 1996.
[ back ] 57. Martu A/B (ETCSL 4.12.1, 4.12.2).
[ back ] 58. Martu A, 57–59, translation ETCSL; cf. Falkenstein 1959:120–140.
[ back ] 59. Römer 1965; Klein 1990:65–67.
[ back ] 60. Klein 1990:67.
[ back ] 61. Ishme-Dagan A + V, 10–20 (ETCSL, 367–377). For the joining of A and V, Ludwig 1990:161–162 with Frayne 1998:7, 9.
[ back ] 62. For a detailed comparison with Shulgi, and other parallels, see Ludwig 1990:189–200; Klein 1990:72–79; Frayne 1998:20–23. But note also Michalowski 2005:201, et passim on innovations in the hymnography of Ishbi-Erra, first king of Isin.
[ back ] 63. FM 9:237–238 (no. 59); cf. Durand 1992:127; Ziegler 2006a:346, 348, 352; Ziegler 2010:122–123.
[ back ] 64. Huffmon 1965; Buccellati 1966; Gelb 1980.
[ back ] 65. Durand 1992:123–126; Durand 1997 1:39–40.
[ back ] 66. For this event, Bonechi 1992; Malamat 1999; Malamat 2003; Ziegler 2006a:4; FM 9:19–20.
[ back ] 67. Cf. FM 9:19–20.
[ back ] 68. FM 9:217–220 (no. 52.8’–9’), following her translation. For the political background, Charpin and Ziegler 2003:101–102, 124–125.
[ back ] 69. FM 4 37; cf. FM 9:20, 85n14.
[ back ] 70. For the Banu-Yamina, see e.g. CANE:1238 (Whiting).
[ back ] 71. See p44–45.
[ back ] 72. ANET:564; CS 1 no. 1.135 (here p. 463).
[ back ] 73. Finkelstein 1966:112.
[ back ] 74. See Michalowski 1983:243–246.
[ back ] 75. Lipiński 1978; HUS:613; RCU:113–114n124.
[ back ] 76. Finkelstein 1966:113–116; cf. CANE:1239 (Whiting).
[ back ] 77. See PHG:99–100, with reference to the issue of semantic shift and performance contexts of BALAĜ (see p531, 573 ); cf. Gabbay 2014 §4.
[ back ] 78. RIME 4 2.13.1002, iii: 4’-9’; PHG:99 (quotation), noting: “Another possibility is that the sequence balaĝ li-li-ìs (zabar) is to be understood as ‘balaĝ and (bronze) lilissu drum’, perhaps indicating that in this period the balaĝ stringed instrument was still used in cult together with the lilissu drum.” (The same interpretive issue arises with the inscription of Hammurabi discussed on p86–87.) See also Gabbay 2014 §4; Heimpel, “Balang-Gods,” 44.
[ back ] 79. Simmons 1960:83 (ii); Charpin 1978:28n56, chronology on 40; cf. Mirelman 2014.
[ back ] 80. Simmons 1960:76–77 (kk, ll); Charpin 1978:28 (e), chronology on 40; cf. Mirelman 2014.
[ back ] 81. The text is ARM 25 566 = Joannès 1985:111–112 no. 10; cf. FM 3:47, discussing this goddess’s relationship to Dagan at Tuttul; PHG:106; RlA 9:382 (Heimpel, *Ninigizibara); Heimpel, “Balang-Gods,” 23g2.
[ back ] 82. See p22.
[ back ] 83. Heimpel 1998:10–11; RlA 9:382–384 (Heimpel, *Ninigizibara); Gabbay 2014 §10 and n23 (quotation); PHG:106–107; Heimpel, “Balang-Gods,” 23.
[ back ] 84. FM 3:47; RlA 9:382–384 (Heimpel, *Ninigizibara, quotation 383), noting several contexts in which Ninigizibara is known from Umma (see Heimpel, “Balang-Gods,” 23).
[ back ] 85. CLAM:536–603; Heimpel, “Balang-Gods,” 47.
[ back ] 86. Volk 2006:94, line 14 with PHG:112–113 (quotation), raising the question of Ninigizibara’s gender; also Heimpel, “Balang-Gods,” Section 2d and 23f.
[ back ] 87. See for example Uruhulake of Gula (She of the Ruined City), a+45, 48 (CLAM:256, 262) and Abzu Pelam (The Defiled Apsu), 86 (CLAM:55, 60); I thank U. Gabbay for these references (communication, July 2012). See further Heimpel, “Balang-Gods,” 49.
[ back ] 88. OB Ishtar ritual from Mari: Dossin 1938; FM 3 2, the name appearing as dNingizippara at i.8’, 10’ (see comments on 47), also in “Le rituel d’Eštar d’Irradân,” FM 3 3 i.21’ (cf. p62); FM 9:55–64; Nissinen et al. 2003:80–82 (nos. 51–52), with further literature; cf. PHG:106; Ziegler 2010:126–127.
[ back ] 89. Cf. FM 3:48, with the illustration of FM 9:56.
[ back ] 90. See FM 3:49–50; FM 9:61 (on ii.19’); PHG:182–183.
[ back ] 91. FM 3 2 ii.19’–27’, and the comparable 3 iii.4’–13’, is unfortunately lacunose. See remarks of Durand 1988:386–387; FM 3:50; Nissinen et al. 2003:82n a; FM 9:61, 63–64. Stökl 2012:211–214 rejects a direct link between music and ecstatic prophecy. But key readings are quite uncertain, and the immediate conjunction of musicians and ecstatics in both texts must be somehow significant. The instructions may be elliptical, not fully elaborating the stages of ‘collaboration’ between musicians and ecstatics.
[ back ] 92. U. Gabbay (communication, July 2012); cf. also Ziegler’s observations on the instrument in FM 9:60 (on i 8’), 62 (on iii, 12, 14, 18, 28). Heimpel, “Balang-Gods,” Section 4a, suggests that the instrument was indeed played—an “illogical element.”
[ back ] 93. See p41, 291–303.
[ back ] 94. See p23–29.
[ back ] 95. See FM 9:221–222 (no. 53.7–8), arguing against an alternative interpretation of this name as belonging to a musician. So too U. Gabbay (communication, March 2010): “An urzababîtum instrument in Mari is called Samsi-Yasmah-Addu … surely referring to the king of Mari of this name … which again shows the connection of this instrument to kings.” Cf. Heimpel, “Balang-Gods,” p627.
[ back ] 96. Is the sun god also somehow invoked?
[ back ] 97. See p35–36.
[ back ] 98. Cf. p134–141.
[ back ] 99. Sollberger and Walker 1985; but I follow the text and translation of RIME 4 3.6.11, who read BALAĜ in line 31 (the passage quoted here is lines 23–34).
[ back ] 100. Hammurabi bears the title ‘king of all the Amorite land’ again in RIME 4 3.6.10, 8.
[ back ] 101. See generally Kuhrt 1995:95–109; van De Mieroop 2005:64–79. RIME 4:344, follows Stol in suggesting that Hammurabi had already assumed the title “King of All the Amorite Land” in his thirty-fourth regnal year.
[ back ] 102. See p531, 573.
[ back ] 103. See PHG:99, and above n78.
[ back ] 104. See p54, 65–67, 79.
[ back ] 105. Liverani 1971:61 writes of this period: “l’omogeneità delle popolazioni stanziate in Mesopotamia e in Siria (gli Amorrei) e gli ampli rapporti politici e commerciali tra le due aree (come sono esemplarmente documentati dai testi di Mari) rendevano particolarmente agevole il trapasso di idiologie e di procedimenti politico-sociali e religiosi.”
[ back ] 106. Cf. FM 3:50: “Cette intervention d’un prophète occidental au sein d’une grande liturgie sumérienne, pour déconcertante qu’elle soit, montre bien à quel degré de syncrétisme on en était arrivé dans la région d’Akkad.” For the Amorite dimension of the Mari prophetic texts, see Lemaire 1996.
[ back ] 107. See Heimpel, “Balang-Gods,” passim.