Hostname: page-component-5cf477f64f-n7lw4 Total loading time: 0 Render date: 2025-04-02T20:41:29.287Z Has data issue: false hasContentIssue false

LITERARY TEXTS FROM THE SIPPAR LIBRARY V: A HYMN IN PRAISE OF BABYLON AND THE BABYLONIANS

Published online by Cambridge University Press:  17 March 2025

Anmar A. Fadhil
Affiliation:
Department of Archaeology, College of Arts University of Baghdad, Iraq [email protected]
Enrique Jiménez
Affiliation:
Institut für Assyriologie und Hethitologie Ludwig-Maximilians-Universität München, Germany [email protected]
Rights & Permissions [Opens in a new window]

Abstract

The article presents a previously unknown hymn in praise of Marduk, the Esagil, Babylon and the Babylonians. It contains unparalleled descriptions of the healing powers of Marduk, the splendor of Babylon, the spring borne by the Euphrates to the city’s fields and the generosity of the Babylonians themselves. The text survives in 20 manuscripts, from the 7th to the 2nd/1st centuries BCE, and it can be shown that it was a fixture in the school curriculum of the time. The author of this highly accomplished piece immortalized his devotion to his city, gods, and people in words that resonated until the final decades of cuneiform culture.

نصوص أدبية من مكتبة سبار، الجزء الخامس: ترنيمة في مديح مدينة بابل والبابليين

أنمار عبدالاله فاضل — إنريكي خيمينيز

يعرض هذا المقال ترنيمة جديدة غير منشورة من قبل تتناول مديح الاله مردوخ وحارته الايساجيل، فضلا عن مدح لمدينة بابل وسكانها. يصف نص الترنيمة بصورة أدبية عالية لامثيل لها القوى الشفائية للاله مردوخ، روعة مدينة بابل، جمال نهر الفرات وما يقدمه من خيرات للحقول وللناس في موسم الربيع، كذلك كرم سكان مدينة بابل أنفسهم. عثر على نسخ من نص هذه الترنيمة مكتوبا على عشرين لوحا مسماريا ترجع للفترة بين القرن السابع والثاني قبل الميلاد، حيث كان يعتبر هذا النص أحد النصوص الأساسية التي تدرس في المناهج الدراسية المعتمدة انذلك. لقد خلد مؤلف هذا النص البديع إخلاصه لمدينته وآلهته وشعبه بكلمات بقيت تردد حتى العقود الأخيرة من عصور الكتابة المسمارية.

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - SA
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike licence (https://creativecommons.org/licenses/by-nc-sa/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the same Creative Commons licence is included and the original work is properly cited. The written permission of Cambridge University Press must be obtained for commercial re-use.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of The British Institute for the Study of Iraq

Introduction

Assyriology’s main objective is, as it was once put, “the recovery and reconstruction of a lost heritage” (George Reference George1997: 73a). Because of their profound impact on ancient literature and individuals, recovering the classics of ancient Babylonia – the texts that the literate population read, memorized, and cited – can significantly advance this goal: when a classic is recovered, not only is its text regained, but also the dense network of quotations, allusions and excerpts that show how the Babylonians responded to it.

The text published here for the first time can be added to the small cadre of compositions one may call “classics”: preserved in no fewer than 20 manuscripts, from the 7th to the 2nd/1st centuries BCE, it was a fixture in the school curriculum of the time.Footnote 1 Presumably literate Babylonians knew the text by heart, since school texts were wholly or partially committed to memory,Footnote 2 and indeed quotations of it can be found in other texts.Footnote 3 It contains unparalleled descriptions of the healing powers of Marduk, the splendor of Babylon, the spring borne by the Euphrates to the city’s fields and — most extraordinary of all — the generosity of the Babylonians themselves. The author of this highly accomplished piece immortalized his devotion to his city, gods, and people in words that resonated until the final decades of cuneiform culture.

1. Contents and Exegesis

About two-thirds of the original text, which may have been 250 lines long, have been completely or partially recovered.

§1. Opening Hymn (1–25). The poem opens with a hymn to Marduk, formulated in the third person, which lists epithets from the standard repertoire, along with some unique ones (such as “guardian spirit of Esagil,” l. 20), and ends with a stanza of the AA′B type (ll. 23-25).

§2. Hymn of God to Marduk (26–79). A formula introduces the speech of a god, who in all likelihood addresses Marduk. The speech in question is a long hymn, much more sophisticated than the previous one. Some of the lines of the text are reminiscent of the “great hymns” to Marduk, and indeed l. 39 appears verbatim in ‘Marduk 2’. The hymn traverses several realms in which Marduk’s help is providential, such as financial loss (50f.) and provision of food and shelter through vegetation (69–72). The final segment of the hymn presents the waters, whose dominion Marduk inherits from his father Ea, as the nourishers of all creatures; with these waters are mixed (summuḫu) fire and air (75f.). Why these three elements are mentioned only becomes somewhat clear later. Before, in an unparalleled passage, the “gods of the land” (i.e. the lower gods), are presented as standing as “their servants” (dāgil pānīšunu), i.e. as the servants of the waters or else of the three elements mentioned (l. 77). In the awesome presence of Marduk they do not dare to speak (l. 78f.); they “take counsel with each other” (l. 80f.). Then the action for which the assembly has been convened takes place: the three great gods (Anu, Enlil, and Ea) and their consorts “bless” (ikarrabū, note the present tense) Marduk. This event reveals the importance of the mention of the three elements water, fire, and air earlier: according to a Babylonian doctrine, these are the three primordial elements. In the syncretistic hymn to Marduk ‘Eriš šummi’, water and fire (“air” is mentioned earlier) “support” (ukallū) life, i.e., they make it possible.Footnote 4 In the Babylonian view, the triad Anu-Enlil-Ea represents these three elements: thus, in the commentary 1881,0204.419 ll. 6′–8′ (eBL transliteration: Stadhouders):Footnote 5

The three basic constituents of the universe (fire, air, and water) are, therefore, associated with Anu, Enlil, and Ea. The three elements, to which the “gods of the land” stand in service, are the cosmogonic counterparts of Anu, Enlil, and Ea.

§3. Hymn to Esagil (86–99). The blessing to Marduk is followed by a hymn to his temple, Esagil, which is “beautiful” (or “built,” banû, s. commentary). A number of epithets are showered on Esagil, some of them etymologically derived from the name of the temple and known in other treatises devoted to it. Esagil is “Eridu” (for that is the name of the neighborhood where it is located), it is the gateway to the Apsû and the underworld, and it is built with mysterious and subtle artistry.

§4. Hymn to Babylon (100–124). The literary caliber of the text increases when it reaches the section dedicated to Babylon, the city whose “ordinances are perfect” (l. 100). After identifying Eridu with Babylon (“It is called Eridu, Babylon is its name”), the city is described as a hoard of precious stones of all kinds (ll. 105–108). Babylon “flourishes in her charms” like a fruit garden (l. 109). After this description, the different components of the city are introduced. First, its star, the “star of Marduk” (i.e. Nēberu); then its “gate” and its wall, Imgur-Enlil; and, most arrestingly, its king, Alulu (l. 115). In Mesopotamian tradition, Alulu was the first king, and he reigned in the city of Eridu. That he should be described as king of Babylon in our text is not altogether surprising: according to Berossos, “Aloros (῎Αλωρος), a Chaldaean from Babylon, was the first king and he reigned for ten saroi” (De Breucker Reference De Breucker2012: 232). Our text thus shows that the tradition recorded by Berossus is also present in cuneiform texts.

The next element of Babylon to which the poet directs his attention, its river, triggers a lyrical torrent describing the arrival of spring – a gift of the Euphrates – to the fields of Babylon (ll. 116–124). The water makes the fields bloom, the grain sprout, and the flocks graze on it.Footnote 6 Mesopotamian literature is sparing in its descriptions of natural phenomena; in particular, only one other lengthy description of the coming of spring seems to exist, in the cosmogonic prologue to ‘Ox and Horse’ (s. below, Genre).

§5. Hymn to the Babylonians (125–158). The poet now turns his attention to his fellow citizens, “the clan of Lugal-abzu,” the “progeny of Alalgar” (125, 127, Alalgar being the successor of Alulu as king of Eridu in the Mesopotamian tradition). The “free citizens” (ṣābū kidinni) of Babylon are, for the author of this hymn, in essence the priests of Babylon — an indication that its author probably was a member of this class. The Babylonians are fair, protect the orphan and the humble (ll. 136–138), follow the divine precepts and keep justice (l. 141); in particular they abide by “the original stele, the ancient law,” perhaps a reference to an actual stele, such as Hammurapi’s. They respect one another, please each other (ll. 146f.) and –strikingly– respect the foreigners who live among them. The concept of respecting the foreigners has, of course, Biblical connotations,Footnote 7 although in our text the foreigners referred to are specifically the foreign priests living in Babylon.

After the Babylonian men, the text pays attention to the Babylonian women, whose quintessence are the Babylonian priestesses. The passage has great importance for understanding the roles played by the various classes of priestesses: ugbakkātu, nadâtu, and qašdātu. The priestesses are particularly virtuous but, in contrast to the active role of men in protecting the helpless, the main virtue praised in women is devotion and discretion. The hymn to the Babylonians ends with the doxology: “These are the ones freed by Marduk” (šunū-ma šubarrû ša marūtuk, l. 159).

§6. Broken section (159–end). About one hundred lines of the ending are missing or mutilated; it is difficult to ascertain what they might have contained. Since the end of the third column (ll. 191–198) refers to goods granted to a multiplicity of persons, it is possible that the hymn to the Babylonians continued for some 40 more lines. The beginning of the fourth column (ll. 204–211), by contrast, seems to describe the awe-inspiring appearance of a warrior god, perhaps Marduk, and his steeds.

2. Genre

The incipit of the text, not yet fully recovered, reads: […] nagbi šamê u erṣ[eti marūtuk (?)]. The title appears to be absent from catalogues and tablet inventories. It is possible, however, that this title appears in the catchline of the main manuscript of the ‘Hymn to Šamaš’, K.3182+ iv 34 (NinNA1 in Rozzi Reference Rozzi2021): [o o o o o o o o o o a]n-e u ki-t[ì o o o]. Although the phrase “heaven and earth” is, of course, very common, both the incipit of the present text and the catchline have erṣetu as the penultimate word, which makes the identification plausible. At least some of the great hymns were linked to each other by means of catchlines, forming a series: thus, the catchline of several tablets of ‘Marduk 1’ links to the ‘Hymn to Šamaš.’Footnote 8 If correctly identified, the sequence in the series would be ‘Marduk 1’ → ‘Hymn to Šamaš’ → our text.

In any case, the text is clearly hymnic in character, like the other “great hymns” with which it is perhaps linked. One feature of the text, however, is not typical of hymns: after the initial praise, the text contains a speech introduction formula (ll. 26f.: [… pâšu] īpuš-ma iqabbi | [… šam]ê u erṣeti amāta izakkar), followed by a speech of one god, at the end of which the great gods bless Marduk. These introduction formulae are known almost exclusively in narrative poetry and in parodies thereof.Footnote 9 They also appear, however, in the mythological sections of two Old Babylonian hymns.Footnote 10 The occurrence of the formula suggests that the present text also represents a sort of hybrid between hymn and mythological narrative composition.

Other elements of the text suggest that the events that appear in it are situated not in the atemporal plane of the hymns, but rather in illo tempore, in the mythological time immediately prior to the beginning of history.Footnote 11 On the one hand, human kings appear, albeit they are the first two kings of the Mesopotamian tradition, Alulu and Alalgar (ll. 115 and 126). On the other, each of the sections begins with a reference to creation: Esagil is “created” (banû, l. 86, s. commentary ad loc.), and “was created” (ibbanû) in the city of Babylon (l. 100), alongside which the Babylonians were also “created” (ibbanû, l. 127). Thus, the text seems situated at the dawn of history, in the moments following creation. As in the historiola ‘The Founding of Eridu’,Footnote 12 the text seems to assume that the first entities to be created in the universe were Eridu/Babylon and Esagil. As in other Mesopotamian texts, the dawn of the story takes place at the dawn of the year, in the spring,Footnote 13 brought to Babylon by the waters of the Euphrates.

Hymns to cities and temples are, as noted by A. R. George (Reference George1992: 3), much less popular in Akkadian literature than they are in Sumerian. Although the number of manuscripts of the present text shows that at least this hymn gained considerable popularity, few other Akkadian texts are known which have such a detailed description of a city, its temples, and its gods. Perhaps the most notable parallel is constituted by a hymn to Borsippa known on two three-column tablets, which has certain parallels with the present text (see comments on ll. 52 and 136).Footnote 14

3. Date

The precise dating of this text, as is often the case with literary works, remains elusive. The terminus ante quem is marked by the earliest surviving manuscript (AssSchNA1), a tablet from the Assur school devoid of precise archaeological context but datable to around the 7th century BCE. Moreover, its adoption as a school text implies a prior period of circulation preceding its incorporation into the curriculum. As for the terminus post quem, the lack of Old and Middle Babylonian manuscripts, if this silence is indeed indicative, hints at a composition date sometime after the first half of the second millennium BCE. The text resonates with the ideology characteristic of the “Rise of Marduk,” placing it potentially within the same timeframe as other works incorporated into the “Marduk Syllabus,”Footnote 15 such as ‘Enūma eliš’ or ‘Ludlul’, i.e. in the second half of the second millennium BCE. Yet, the paucity of reliable chronological anchor points makes a precise dating of these texts no trivial task either.Footnote 16

As stated above, the text is set at the beginning of history, so few elements can be extracted from it that would allow its dating. The fact that it mentions Nippurean and Susian priests living in Babylon (l. 134) could suggest that the text was composed in the Kassite period, during which time expatriates from both places are documented living in Babylon.Footnote 17 The description of the kind treatment of Elamite expatriates by the Babylonians would seem unlikely after the beginning of the Elamite invasions of Babylon during the 13th century BCE.

4. The Manuscripts

Although none of the colophons of the manuscripts of this text are fully preserved and the provenance of most of them can be determined only approximately, two clear groups can be established on the basis of the paleography and orthography of the manuscripts: Neo-Babylonian and Late Babylonian manuscripts. The excerpts from the text on school tablets constitute a third group, more miscellaneous than the other two.

Group 1: Neo-Babylonian Manuscripts

The first group consists of tablets datable to the ‘long sixth century’, i.e. between the rise of the Neo-Babylonian dynasty in 626 BCE and the ‘End of Archives’ in 484 BCE. The Sippar tablets of this period compete in quality and reliability with the often-praised manuscripts from Nineveh. MSS SipNB1, SipNB2, SipNB3, and SipNB4 come from Sippar; they all divide the text into 2 columns per side. The most important manuscript of the text, that of the Sippar Library (MS SipNB1), consists of two fragments found, according to the excavation record, in two different niches: IM.132512, from niche 7 D (1), and IM.132667, from niche 6 D (1).Footnote 18 The Sippar Library tablet divides the lines into two halves, which generally correspond to the two hemistiches of the verse: the vertical line thus marks the caesura.Footnote 19 This “finesse of the scribal art” (Lambert Reference Lambert1960a: 66 fn. 1) is found only in a few Neo-Babylonian copies of literary texts.Footnote 20 The tablet also includes decimal markers every ten lines, and twice every five lines (ll. 85 and 95). MS SipNB2 and SipNB4, both belong to the British Museum’s “Sippar Collection” and derive in all likelihood from that city.Footnote 21 SipNB2 seems to derive from the same archetype as SipNB1: the spellings are very similar, and very few variants exist. SipNB3, in the Istanbul Sippar collection, is similar to other fine two-column Neo-Babylonian literary manuscripts that derive from J.-V. Scheil’s excavations at Sippar, such as Si.15 (‘Hymn to Šamaš’, Lambert Reference Lambert1960a, pls. 33 and 36) and Si.851 (‘Marduk 2’, CTL 1, 100).

Of the Neo-Babylonian tablets from Babylon, BabNB1 is a two-column tablet written in a particularly small script. Slightly larger, but still relatively small is the script of BabNB2, a fragment that appears to belong to a single-column tablet, but whose surface has sustained extensive damaged. The obverse of the tablet, where only a few signs can be identified with certainty, does not seem to match the known text, its reverse ends abruptly at l. 123, followed by a ruling and by what appears to be a colophon: either the manuscript is part of an edition that divides the text into several tablets, or it is an excerpt on a school tablet. Even more damaged is the obverse of BabNB3, which seems to contain the beginning of the text, without this identification being certain. Its reverse, better preserved, cannot be matched with the known text, and the fact that its l. r 3′ mentions Ea, Šamaš, and d asar-lú (apparently not -ḫi) suggests that it may contain an entirely different composition. The consignment of BabNB1 and BabNB2 (80-11-12) comes almost entirely from Babylon, that of BabNB3 (81-7-1) contains many tablets from the same city (Reade apud Leichty Reference Leichty1986: xxx).

Group 2: Late Babylonian Manuscripts

Three tablets (BabLB1 to BabLB3) can be dated to the last centuries of cuneiform culture, perhaps specifically to the second or first centuries BCE. The criteria for dating the tablets to this period are, first, paleographic: the manuscripts exhibit the sign forms that are normally associated with the terminal phase of cuneiform, such as ku, lu, lagab, and ezen bereft of an upper horizontal wedge;Footnote 22 lul as dumu+pap (80 BabLB1); and az and uk with za and ud, respectively, added after the sign, rather than inscribed (38 and 46 BabLB2). The orthography of the manuscripts is also indicative of a late date: BabLB2 uses the sign ma k (ma5 = ka×éš) syllabically, which is attested only in a handful of late manuscripts (Frazer Reference Frazer2020). The late period also develops some ligatures that are absent from earlier manuscripts, most relevantly those of dumu×sal (on which see Jiménez Reference Jiménez2020: 242) and zu×ab (in l. 88, MS BabLB1). In addition, BabLB1–3 are the only ones to use bàd as determinative for the Imgur-Enlil wall in l. 114, a feature reminiscent of the use in Hellenistic and Parthian manuscripts of redundant logograms, such as šeš and ìr (in earlier periods simply šeš and ìr, see Jiménez Reference Jiménez2017: 346). The consignment of the tablets is compatible with a late date, since both the Spartali collection (BabLB2 and BabLB3) and 81-7-1 and 81-7-6 (BabLB1) contain many tablets, especially astronomical ones, from the Hellenistic and Parthian periods, many of them from Esagil.Footnote 23 Interestingly, BabLB2 and 3 share the odd spelling kurun.nam instead of gurrunu in l. 121 (BabLB1 is broken here), suggesting that they derive from the same archetype.

These three manuscripts testify to the popularity of the hymn in the waning decades of cuneiform, when Babylon, once a gem-laden mountain (ll. 105ff.), had faded into the semi-deserted city famously described by Pliny the Elder (Naturalis Historia VI 122). In this stark setting, Babylonians faithfully copied the hymn that spoke of the vanished glories of their city, the first of all creation.

Group 3: School Tablets

The only manuscript from Assyria (AššNASch1) is a school tablet with a selection of long excerpts from different texts (our text, ‘Hymn to Šamaš’, ‘Marduk Hymn 2’, ‘Erra Epic’). Although its first editor “saw nothing impossible in the inclusion of the four excerpts consecutively” in his edition of the ‘Erra Epic’ (Lambert Reference Lambert1980: 95), most subsequent scholars have noted that the first excerpt belonged to a text in honor of Babylon, and have described it as a text whose lines “praise the ideal of the šubarrê ša Marduk (= citizen of Babylon)” (Landsberger/Jacobsen Reference Landsberger and Jacobsen1955: 21 fn. 26), as a “Preislied auf Babylon” (Borger Reference Borger1967: 102), as “a hymn in praise of Babylon, probably of Middle Babylonian origin” (Lambert Reference Lambert and Haas1992: 143), and a description of “die gerechten Babylonier” (Maul/Manasterska Reference Maul and Manasterska2023, no. 33). The tablet’s high-quality writing suggests the work of an advanced scribe. The presence of a colophon, a rare feature in school tablets, further supports this interpretation.Footnote 24 The tablet contains, in this and the other excerpts, a few Assyrianisms,Footnote 25 and in the ‘Marduk 2’ excerpt in particular, quite a few corruptions.

School tablets with excerpts are, due to the trying character of their writing, more difficult to classify palaeographically than “library” manuscripts. Those from Babylonia are classified in the list of manuscripts according to the museum collection to which they belong, into Neo-Babylonian tablets from Sippar (SipNBSch1-3) and from Babylon (BabNBSch1-7); some of the latter, e.g. BabNBSch6, may also be Late Babylonian.

5. List of Manuscripts

Fig. 1. SipNB1 obverse. Copy by Anmar A. Fadhil

Fig. 2. SipNB1 reverse. Copy by Anmar A. Fadhil

Fig. 3. SipNB1 obverse. After conservation by C. Gütschow in November 2021

Fig. 4. SipNB1 reverse. After conservation by C. Gütschow in November 2021

Fig. 5. SipNB1 obverse (IM.132667 only). Shortly after excavation. No old photograph of the reverse seems to have survived

Fig. 6. SipNB1 (IM.132512 only). Shortly after excavation

Fig. 7. SipNB2, SipNB3, and SipNB4. Copies by E. Jiménez

Fig. 8. BabNB1, BabNB2, and BabNB3. Copies by E. Jiménez

Fig. 9. BabLB1. Copy by E. Jiménez

Fig. 10. BabLB2, BabNBSch2, and BabNBSch3. Copies by E. Jiménez

Fig. 11. BabNBSch6, BabLB3, and SipNBSch1. Copies by E. Jiménez

Fig. 12. SipNBSch2 and SipNBSch3. Copies by E. Jiménez

6. Fragments Potentially Belonging to Unrecovered SectionsFootnote 30

BM.38076, BM.39161+, BM.39252, BM.39269, BM.40328, BM.43372, BM.46177, BM.46197, BM.46823, BM.72061 a, BM.72135, BM.76021, BM.76244, BM.76769, BM.76996, BM.114741

7. Edition

8. Commentary

  1. 1. As noted above (§4 The Manuscripts), the deteriorated condition of the surface of BabNB3 makes it difficult to confirm its association with this composition. If true, MS BabLB1 would contain a bound form in -i (nagbi, on these forms, see George Reference George2003: 432f.), while BabNB3 would contain the normal form (nagab). It is unclear how many signs are missing after erṣeti, but is seems clear that there must be at least one word missing, since otherwise BM.42723 would join BM.45986 directly. Moreover, érṣetu is not acceptable as the last word in a line of poetry, since it would produce a non-trochaic ending. The reconstruction of Marduk’s name at the end is hypothetical;Footnote 42 alternatively one could read [(d)]⸢mes⸣ at the beginning, which would exclude the adopted reconstruction at the end. As noted in the introduction, the title is perhaps the catchline that appears in the large Nineveh MS of the ‘Hymn to Šamaš’ (K.3182+, MS NinNA1 in Rozzi Reference Rozzi2021): [o o o o o o a]n -e u ki-t[ì o o].

  2. 15. The restoration at the beginning is inspired by ‘Maqlû’ I 138 and 192 ((girra) nūr ilī kayyānu). If correctly restored, kayyān could be a nominalized adjective acting as the regens (as reflected in the translation), or else an adjective agreeing with nūru that interrupts the bound chain (**nūr ilī šamê u erṣeti kayyānu).

  3. 16. MS BabLB1 uses gír.tab for ád also in l. 92, so the suspicion arises that it may be a late convention rather than an error.

  4. 18. The restoration is inspired by ‘Marduk 1’ 10//12 (Fadhil/Jiménez Reference Fadhil and Jiménez2019: 167): ṭāb nasḫurka, “your attention is sweet.”

  5. 20. The confusion of lamassu and lamaš(t)u (BabLB1) also occurs in a late manuscript of ‘Marduk 1’ l. 176: lū atrat la-maš-šá-áš-šú (other MSS: la-mas-sa-šú, d lamma-šú) el [ i ša qadmi], “May his good fortune surpass that of before!” This orthography, together with the ušātu of l. 24 and šimat in l. 86 in BabLB1, may perhaps reflect a shift in the phonetic status of the sibilants in the terminal phases of Akkadian.Footnote 43

  6. 23–24. The phrase bēl usāti is particularly common in Akkadian onomastics (Stamm Reference Stamm1939: 212). Outside of proper names, the phrase is attested only rarely, e.g. in the ‘Dialogue of Pessimism’ 78 (Fadhil Reference Fadhil2022): ayyu bēl lemuttim-ma ayyu bēl usāti, “Which was the doer of evil, and which was the doer of good deeds?” As already noted by Montgomery (Reference Montgomery1908), the phrase also appears in Aramaic incantation bowls, as mry ʾswʾtʾ.

  7. 26ff. The addressee of the speech is probably Marduk. bēl mātāti (l. 28) is, of course, a traditional epithet of Enlil, but one that the god cedes to Marduk in ‘Enūma eliš’ VII 156. The restoration in l. 27 is inspired by the šuʾila ‘Marduk 1’ (Si.7+ // K.3505.B // K.17421+) l. 8: bēl(en) mātāti (kur.kur) šar (lugal) šamê (an-e) u erṣeti(ki -tì).

  8. 37. Words such as dannatu are usually not written logographically in library manuscripts of literary texts, so the decipherment may be incorrect.

  9. 38. The restoration at the beginning is based on ‘Ludlul’ V 82 (Hätinen Reference Hätinen2022): [ap]âtu mala bašâ marduk dullā, “[Tee]ming humankind, as many as they be, give praise to Marduk!”

  10. 39. The line appears verbatim in ‘Marduk 2’ (l. a+14; Lambert Reference Lambert1960b: 65): tattanašši lā lēʾâm-ma tereʾʾi ulāl[a]. Since ‘Marduk 2’ contains several verbatim quotations of other texts,Footnote 44 the present text is probably the lender and not the borrower.

  11. 43. ‘Ludlul’ V 56 (Hätinen Reference Hätinen2022): ušamḫir erba ṭaʾta igisê etandūti, “An offering, a gift, sundry donations I presented.”

  12. 46.gíl ! -la !⸣-tu 4 is a virtual emendation of the traces in MS SipNB1.

  13. 48. Compare the incantation in AO.17656 o 4 (Nougayrol Reference Nougayrol1947: 31): an-ḫu dal-pu šu-nu-ḫu a-me-lu.

  14. 49. Compare in the acrostic DT.83 r 3′ (eBL edition): ⸢zi⸣-⸢kirša[p*-ti-šú] ⸢ki⸣-ma làl-la-ri ugu ab-ra-a-ti li-šá-ṭib, “May he make his speech as pleasant as honey to humankind.” gabbu is normally not a literary word; in particular its use as a noun is very rare in literature (see AHw. 272a s.v. gabbu I 1).

  15. 51. Compare the line ‘Ninurta as Savior’ 52 (Mitto Reference Mitto2022b), a line that has in the various manuscripts, all of them school tablets, a slightly different shape: manīt mīšari iddekkâššum-ma (var. tašâhšum-ma, var. tadekkâššum-ma) ša īṣa (var. ina īṣi, var. šattu) uḫalliqu irâbšu (var. urābšu) māda ?(lal), “That he will (var. ‘you will raise’ and ‘you will blow’) have a propitious breeze spring up for him and compensate him amply for every bit he had lost (var.: ‘for what he lost in a year’).” The difficulty of interpreting the line, particularly its last word, contrasts with the clarity of the opposition šattu : ūmakkal in the present hymn, and suggests considering our text as the lender and the Ninurta hymn (a Middle Babylonian composition, see Mitto Reference Mitto2022c) as the borrower.

    The spacing in BabNBSch2 and in SipNB1 suggests that at least one word intervenes between ṭābu and tašâ]ḫšum-ma (?). MS BabNBSch2 could conceivably be emended to read ta-ziq ! -qa !, and šāru ṭābu then taken as a predicative complement instead of a direct object, but this solution is far from satisfactory. The sign after the second ta in BabNBSch2 could be š[aḫ.

  16. 52. Compare in the Hymn to Borsippa BM.61625+ ii 30 (eBL transliteration): [o] x x x x (x) ⸢ana ?is-qí-šú-nu ú-x [o o (o)].

  17. 53. ipru is, like gabbu in l. 49, a word of poor literary pedigree.

  18. 54.ki-i-ni looks less likely.

  19. 59. Compare perhaps l. 114: šad(u) kīni (and commentary ad loc.).

  20. 64. Compare in the ‘Syncretistic Hymn to Gula’ A+54: tâmta ušraqqam ina nagbi mīlī ugappa[š], “The ocean she empties, in the deep she makes the floods huge” (Bennett Reference Bennett2023).

  21. 66. The verb at the end appears to be šakānu D, which is very poorly attested.

  22. 68. tu-šar-si is best interpreted as a hitherto unattested Š stem of the verb recorded in the dictionaries as russû (AHw. 996a: “etwa ‚(durch Wasser) aufweichen‘”; CAD R 425b: “to sully”). The meaning of the Š stem is perhaps similar to its D. As argued by Schwemer (Reference Schwemer2007: 9f.; Abusch/Schwemer Reference Abusch and Schwemer2011: 385), “to bind” (a meaning that its frequent parallelism with šuknušu seems to allow) appears to be the most common meaning of russû, since it translates Sumerian lá in bilingual texts and appears together with verbs such as kamû and kasû in magic texts. On the merism “the hill and the flatlands” (i.e., “everywhere”), compare e.g. ‘Erra’ IV 87: mūlâ u mušpāla kī aḫāmiš tagmur, “You have destroyed the hill and the flatlands alike.”

  23. 70. “Your plants” and “your wood” probably refer to that which Marduk is said to supply in the preceding lines. šuḫnu, “warmth,” is here attested for the first time outside of the lexical corpus.

  24. 71. In spite of the writing -ra-a of all Sippar MSS, berû must be the subject of šebû (an intransitive verb), like kaṣû in the following line.

  25. 72. On the use of a morphological Gtn stem (lištaḫḫan) with the meaning of the Gt stem in the preterite and precative of certain verbs, see Mayer (Reference Mayer1993: 337 ad 112; Reference Mayer1994: 115) and Streck (Reference Streck2003: 10–13).

  26. 73. mitḫāriš seems to govern balāṭi, as in ‘Theodicy’ 18 (nišī mitḫāriš apât[i], “the people, all mortals”) and 258 (lipit qāt aruru mitḫāriš napišti, “all living creatures, the handiwork of the birth-goddess”). The line in SipNB2 begins with a kúr, a particle that sometimes marks textual problems.

  27. 75f. As described in the introduction, these lines mention the Babylonian ‘three elements’: water, fire, and air.

  28. 81. ḫitmuṭū appears to be stative Gt, a rare stem known mainly in the adverb ḫitmuṭiš: according to Kouwenberg Reference Kouwenberg2010: 372f., ḫamāṭu Gt it is simply a literary use and has the same meaning of G and no detransitive value. It is possible that BabLB1, emended here, has a different verb (perhaps itmudu < itʾudu, naʾādu Gt). The use of the form bbl for wbl is very rare: see GAG §103j, AHw. 92a, and CAD A/1 10b). An erased decimal marker appears at the beginning of SipNB1.

  29. 82–85. Note the use of -bi for the plural -bū in MSS SipNB1 and BabLB1, perhaps resulting from contamination of nominal and verbal endings. See Mayer (Reference Mayer1992: 38 fn. 18) for a collection of NB and LB texts in which -i is used instead of the expected /ā.

  30. 86. ba-nu-ú is best interpreted as a preposed adjective (banû I, “well-formed”), since it cannot be a participle (the expected form would be **bānû bītīšu)Footnote 45 or a stative (**bani). banû I, however, is apparently only here predicated of a building: it is normally used for people, words, or animals. Moreover, in the hymn several of the transitions between the sections have references to the “creation”: see in particular l. 100 and 127 (ibbanû). It seems possible, therefore, to interpret the adjective as deriving from banû A = IV, “to build.” Perhaps both senses of the word are intended at the same time, as reflected in the translation. Compare the etymology of the name of Esagil as bītu bānû napḫar il[ī], “the house that built all the gods,” as [sa7 = ban]û, kìl = napḫaru, and ìl = ilu (VAT.17115 ll. 7f. = George Reference George1992: 80 no. 5).

  31. 87. Since a reading pi-ta-at seems to yield no sense, it is assumed that the word written in the two Sippar manuscripts as wa -ta-at is the word booked in the dictionaries as itûtu A (CAD I/J 317a), itûtu I (AHw. 407b) and utā/âtu (AHw. 1443b). As noted in Jiménez Reference Jiménez2016: 223, the MSS of ‘Bullussa-rabi’s Gula Hymn’ 93 (itût kūn libbi ellil) attest to the readings ⸢e-ta⸣-[at], i-tu-ut (MSS Ashm-1937.620 and BM.62744) and e-ta-at (Sm.1036, see Földi Reference Földi2021a), so the word is attested as (w)e/itī/ū/ātu (the variant with u- is an Assyrianism).Footnote 46 At the end, compare ág = narām in VAT.17115 ll. 3f. (George Reference George1992: 80 no. 5).

  32. 88. The description of Esagil as a “replica of Apsû” and “counterpart of Ešarra” (the cosmic abodes of Ea and Enlil, respectively) is also encountered in ‘Enūma eliš’ V 120 and VI 62 (see George Reference George1992: 296f.). Compare also the similar line in an inscription of Esarhaddon: é-sag-gíl é.gal dingir meš | ma-aṭ-lat abzutam⸣-šil | é-šár-ra mé-⸢eḫ-ret⸣ | šu-bat d é-atam-šil⸣ | mul aš.iku, “Esagil, the palace of the gods, an image of the Apsû, a replica of Ešarra, a likeness of the abode of the god Ea, a replica of Pegasus” (RINAP 4 Esarhaddon 104 iii 47–51).

    The line in MS BabLB1 is cited in CAD T 149a: according to it (ibid. 148a), this would be the only instance of a morphologically feminine plural form of tamšīlu. maṭṭalātu appears to be always a plural:Footnote 47 perhaps tamšīlu is built here analogically, or perhaps it is contaminated by tašīltu, “joy,” a word normally used in the plural.Footnote 48

  33. 89. It seems likely that the logogram gaba.ri, which normally stands for gab(a) or meḫru, but also for māḫiru and maḫāru (see Mayer apud Deller/Mayer Reference Deller and Mayer1984: 108) should stand in this line for meḫertu, “copy,” which is the word normally used as regens of temple names, most relevantly in ‘Enūma eliš’ V 120: mé-eḫ-ret é-šár-ra. In ‘Tintir’ IV 2 (George Reference George1992: 58f. 296f.), Etemenanki is the meḫret ešarra. There is no space at the end for a possessive suffix, so “the splendor of its aura” does not seem possible. The second half of the line is therefore perhaps best interpreted as an accusative of respect qualifying the first half, “an equal to Ešarra with respect to splendor an aura.” –mat in the three MSS that write šá-lum-mat should probably be interpreted as -mata.

  34. 92. Note the spelling šu-un-š[ú] in MS BabLB1: on the shift -- > -nš-, see GAG §31f.

  35. 93. Esagil is probably synecdochically called Eridu, as that is the name of the quarter in which it was located. According to ‘Tintir’ IV 3 (George Reference George1992: 58f. 300–303), Ekarzagina, the sanctuary of Ea in Esagil complex, is the “Gate of Apsû” (bāb apsî; abul arallî in l. 94 seems to be a synonymous phrase). bīt pirišti is normally understood as the “sacristy,” i.e. a room to store the garments of priests and statues of gods (so Doty Reference Doty1993); if taken literally, the phrase may refer to Ekarzagina in relation to Esagil, or else to Esagil in relation to the Apsû. Alternatively, it may be another etymology of the name of Esagil, based on the common equation sag/zag = pirištu.

  36. 94. markas šamê rabûti is another etymology of the name of Esagil in VAT.17115 l. 25 (George Reference George1992: 80 no. 5): [sa = marka]su, an = šamû, gíl = rabû.

  37. 95. The line, whose meaning is less than satisfactory, probably contains an etymology of the name of Esagil. Note (á-)áⓖ = têrtu, ìl = ilu, sa = milku (as in VAT.17115 l. 18 = George Reference George1992: 80 no. 5).

  38. 97. The line is probably an etymology of the name of Esagil, note: zag (i.e. sag) = eširtu, zag = tāmītu, an (from (pa.)an) = pelludû, an = išpikku (attested in the commentary on ‘Enūma eliš VII 65, see Heinrich Reference Heinrich2021; the origin of the equation is unknown), and sag/zag = pirištu (as in l. 93).The upper Winkelhaken of ta in MS SipNB1 (the only MS to preserve the word) is very weak, so one could conceivably read uš-ziz(zu), “sanctuary that established the rites.”

  39. 98. The couplet 97f. is “unbalanced,” i.e. its two halves are not grammatically independent (Jiménez Reference Jiménez2017: 74). The three units uṣrāti, šīmāti, and kullat nēmeqi niṣirt[i] appear to be appositive nouns to pirišti.

  40. 100. Compare ‘Tintir’ I 10 (George Reference George1992: 38): uru me-bi kal-laki = kimin (scil. bābilu) ālu ša parṣūšu šūqurū.

  41. 103. On the restoration at the end, see the note on the next line. ⸢d+e[n] is of course also possible, but there seems to be space for one more sign.

  42. 104. urukugû, “pure city” is a byname of Babylon according to ‘Tintir’ I 49 (George Reference George1992: 40f. 266), where it is explained as ālu ellu in the Akkadian column. The phonetic complement -ú in the present text suggests normalizing it not as ālu ellu, as ‘Tintir’ does, but as urukugû. Note that Nebuchadnezzar’s bilingual ‘Seed of Kingship’ (Frame Reference Frame1995: 29 B.2.4.9 ll. 9–11) also seems to contain the Akkadianized version: uru kù–ga || uru.kù.ga. In l. 126, Urukugû is assigned to Enzag, the god restored here in l. 125; the restoration of Lugal-abzu, very uncertain epigraphically, is inspired by l. 125.

  43. 106. The phrase semer tamlî is known also in an inscription of Sargon II: ḫar meš tam-le-e tulīmānuš arkus-[ma], “I fastened inlaid bracelets on his two wrists” (e.g. in Frame Reference Frame2020: 364 ‘Sargon II 84’ v 58′).

  44. 107. The sequence is similar in inscriptions of Sargon II: na4 na4 za.gìn na4 babbar.dili na4 aš.gì.gì na4 ugu.aš.gì.gì (e.g. in Frame Reference Frame2020: 150 ‘Sargon II 7’ l. 142). Babylon is famously called “mountain of obsidian” (šadû ša ṣurri) in the Middle Babylonian ‘Games Text’ (HS.1893 o 1; Kilmer Reference Kilmer1991; Zomer Reference Zomer2019, no. 4); but since ṣurru appears as the first word of the line, it is unlikely that it should also be restored as the last one.

  45. 108. Jasper is called the “stone of kingship” also in the commentary BM.54312 l. 19 (George Reference George2006: 181), as well as in an inscription of Nabonidus (cited by George Reference George2006: 183f.).

  46. 110. It is possible to take inbī lalîša as a genitive chain, but MS SipNB1 places the caesura between the two words. Therefore, lalîša is taken as relational accusative.

  47. 111. The grammar of the line is difficult. The manuscripts seem to take kīma as a preposition, not a conjunction, since edê is a genitive; moreover, emūqā()šu is in the nominative, and should therefore be the subject of (w)abālu (the masc. ending it-ta-nab-ba-lu in SipNB1, the only MS to preserve the verb, can be disregarded, since the use of -ū for the fem. pl. in verbs is much more frequent than would be the use of -u before suffix in emūqātū-šu for the acc./gen. plural). The object of the verb can therefore only be dumuqšu, of which kullu would then be an attributive adjective. The meaning would be, “like a wave, (Babylon’s) strength brings (Babylon’s) goodness attached,” i.e. its strength is the cause of its beauty. Alternatively, as suggested by an anonymous reviewer, one would take kul-lu as a stative kullā, “Like a wave, her strength brings her gooddness, provides (it).”

  48. 112. The line is cited in an Ashurbanipal hymn from found in Sippar (CBS.733+, eBL transliteration): mul d utu.è.a ù d utu.šú.a ṣa-⸢a⸣-[a-ḫ]u šam-šu <šu>-q[u-ru]. As seen by T. Mitto (privatim), the awkward phrase ṣayyāḫu šamšu is an etymological translation of the name of Marduk, where amar = zur = ṣâḫu and utu = šamšu. It is possible that šūquru (i.e. kal) is also part of the etymology, since another well-known etymology of Marduk, the “flood of a weapon” (abūb kakki, i.e. a-ma-ru tukul) also ends in -v l (on which see Lambert Reference Lambert2013: 165, who considers “doubtful whether the final l of tukul is amissable”).

  49. 113. The line may refer to the Šamaš Gate in Babylon. It would, however, be strange that only this gate should be mentioned; so, alternatively, it may refer to an archetypical gate in Babylon, whose size is such that they occupy “all that the sun (covers).” It is, however, strange that the gate should be mentioned before the wall makes its appearance (l. 114); so one may take the gate to refer instead to Marduk’s Star, Nēberu, mentioned in the previous line, whose “gate” would be located “where the sun (is).” Two manuscripts read an -e in the oblique case, although the sentence appears to be nominal. Other cases of writing an -e for the nominative are known in NB manuscripts, e.g. an-e ù ki -tì irūbū, “heaven and earth shake,” in all MSS of ‘Eriš šummi’ 31 (Fadhil/Jiménez Reference Fadhil and Jiménez2022: 235).

  50. 114. šad(u) kīni appears to be an etymology of the Imgur-Enlil, where Imgur < magāru = gin, and Enlil = šadû. Note the excerpt VAT.13234 (Bab.36574; Bartelmus Reference Bartelmus2016: 310):

    ⸢bàd⸣-bi im-gur-d+en-líl-le m[u-pà]-⸢da?⸣-bi d+en-líl še-še-ga

    du-ur-šu im-gur-d+⸢en-lílana zi-[kir ? šu ?-mi ?]-⸢šu ?im-[t]a-na-⸢x x⸣ d+en-líl

    lú tìl-la šà-⸢ga⸣-a-né ⸢ki-tuš⸣ x x mi-[ni]-in-tuš-a

    a ? x (x) x x x x⸣ ma ? ⸢x⸣ [(x)] ⸢x (x) x (x) x⸣ ab [x (x)]

    lú kar?-ra sa6-sa6-ga-bi? dumu? dnu-dím-⸢mud⸣

    ⸢x (x) mi x x x (x) x⸣ ma-⸢a ?-ar ?⸣ [d]nu-di-mud

  51. 115. On the appearance of Alulu, see “§1 Contents and Exegesis.” The epithet he receives, abi nišī aḫrâti, is etymological, since a = abu and (a-za-)lu-lu = nišū.

  52. 116. šikittu denotes here, as it normally does, “das Ergebnis des ‘Setzens’, nicht die Tätigkeit selbst” (Mayer Reference Mayer2017: 224). In other Akkadian texts it is Marduk who establishes Tigris and Euphrates, although the adscription to Ea is of course not surprising. On the mythological origins of Tigris and Euphrates, see in general Blaschke Reference Blaschke2018: 227–231.

  53. 117. Note in MS BabLB2 the use of the sign mak (ka×éš), on which see Frazer Reference Frazer2020. bamāti was interpreted by Landsberger “wie ṣuṣû normaliter Weideland” (Landsberger Reference Landsberger1949: 277 fn. 91), and specifically as the terrain between the river bed and the plateau, i.e. the river terraces (so also Leemans Reference Leemans1991: 119f.).

  54. 118. tâmati is interpreted as a singular form with anaptytic vowel because of the parallelism with ayabi (on forms of the napšatu type, see Jiménez Reference Jiménez2017: 77f., with further literature). The writing ⸢ta-a-⸢tu 4⸣ of MS BabLB3, if not simply a mistake, could reflect the syncopation resulting from tâmatu > tâwati > tâati > tâti

  55. 119. The poetic tone of the text does not seem to allow an interpretation of the verb as parû, “to vomit,” so it seems advisable to take the verb iptanarrâ (in the NB MSS) or iptarrâ (LB, “gnomic” preterite) as related to the verbs booked in the dictionaries as parāʾu II (“to sprout”; AHw. 833, CAD P 182, EDA P0382: u/u) and parāḫu I (“to ferment”; AHw. 827, CAD P 145, EDA P0403: a/?), which are usually understood as lexical variants of each other (e.g. Stol Reference Stol2008: 351; EDA P0403). dīša u šamma would then be relational accusatives. Since both verbs have complementary distributions (parāḫu I is only attested lexically and in OB; parāʾu II only in NA), and since parû “to vomit” (u/u), attested almost exclusively in medical texts, has no convincing etymology (EDA P0379),Footnote 49 one could perhaps understand that the three forms are variants of the same verb, whose etymological meaning (“to sprout”) would have acquired a more specific meaning (“to vomit”) in medical texts, perhaps by lexical transfer (compare German spritzen vs. sprießen) or euphemism.

  56. 120. Lit. “Having its meadows been made resplendent, barley sprouts.” The reading adopted in the reconstructed text, šunmurā, is preserved only in MS SipNB1 (the sign in question resembles šu rather than ku, and the -n- supports the reading šunmurā over kunmurā, as šunmurā is the regular morphographemic writing of šummurā, whereas kunmurā would involve an unusual dissimilation from kummurā).

  57. 121. The strange variant of the two LB MSS, kurunnu, written kurun.nam, is probably a corruption of gurrunu, although one that makes sense in the context. Note that writing gu-ru-nu in MS SipNB1 could be interpreted as q/gurunnu, “heap,” a noun phonetically closer to kurunnu and which would then be in apposition to karê (the latter in bound form, < karā- < Sum. kara6, see Attinger Reference Attinger2021: 657). MS BabNB2, however, does not admit this interpretation.

  58. 122. gipāru appears to be used here metonymically not for the “pastureland” (on this meaning of gipāru, see Held Reference Held1976: 232 fn. 14), but for the “cattle”; gipāru u laḫru would then be a poetic equivalent of the more prosaic lâtu((áb.)gu4 ḫi-a) u ṣēnū(u8.udu ḫi-a), “herds and flocks.” aburriš rabāṣu expresses “not primarily the idea of safety, but rather stresses exuberance and delight. Water-meadows, recently emerged from the river, will soon have been covered by young grass (…) which together with the immediate vicinity of water for drinking and bathing made (…) an ideal pasture for cattle, a proverbial pleasure resort” (Veenhof Reference Veenhof1973: 374).

  59. 123. The phrase simat baʾūlāti is apparently only attested at the beginning of ‘Poor Man of Nippur’ (l. 5, Heinrich Reference Heinrich2022a), probably an allusion to our text: ul īši kaspa simat nišīšu | ḫurāṣa ul īšâ simat baʾūlāti, “He had no silver, as befits his people, | He had no gold, as befits humankind.”

  60. 124. ab-⸢ḫu⸣ fits the traces in SipNB1 better than ab-⸢ri⸣ (< apāru?). Several parsings of the signs ab–ḫu⸣ seem possible, none of them without difficulties: (1) an irregular form of ebēḫu, “to girt” (for other examples of phonemic a written as e in NB texts, see Woodington Reference Woodington1982: 20); (2) a form of the rare verb (w)abāʾu, “to be full of weeds, to grow wild” (CAD U/W 397b; but cf. AHw. 1454: “verunkrauten”), hitherto poorly attested in first-millennium texts, and otherwise written with w- and aleph; (3) a form of the verb abāḫu (AHw. 56a. 1544a: “behaftet sind”), which is apparently otherwise known only from a passage in an acrostic hymn to Nabû (K.8204): šá šul-ḫa-a u mi-iq-ti ab -ḫu ú-qa-a-ú ka-a-[šá], “he who is ab -ḫu by … and sickness expects you.” This last line is understood by AHw. 56a as “die mit … behaftet sind (?)”; CAD N/1 270a and Š/3 240a read èz-ḫu, from ezēḫu, “to gird on” (a verb of which no other metaphorical use appears to be known). It seems more likely that both the acrostic hymn to Nabû and the present text should be read as ès-ḫu (< esēḫu, “to assign”), despite the fact that the reading ès of ab is relatively rare.

  61. 125. ummatu means “descendant” (Finkel Reference Finkel1988: 149 fn. 57) or “(Angehöriger einer) Gruppe von Kultpersonal” (Jursa 2001/Reference Jursa2002: 84a) in some contexts. Since both ummatu and nišūtu are feminines, tukkulu is interpreted as an epithet of Lugal-abzu (i.e. Ea, Krebernik Reference Krebernik1987/1990a). Ninazu is usually identified with Ninurta, more rarely with Nergal (Wiggermann 1997: 33–35; Reference Wiggermann1998/2000: 333a).

  62. 126. On urukugû, see the comment on l. 104. uru.kù.ga-ú may also be taken as a substantivized nisbe form, “Urukageans,” i.e. “Babylonians.” Lugal-asal is usually identified with Nergal (Krebernik Reference Krebernik1987/1990b), which in the context makes little sense. A diverging tradition makes Lugal-asal the father of Zarpanitu, i.e., Marduk’s father-in-law. Thus, a hymn to Zarpanitu reconstructed by T. Mitto:

    mu-de-e [… šamê(an-e)] ⸢uerṣeti(ki)-tì mi-lik dur-an-k[i]

    ši-i-mai⸣-[lat] i-lá-a-ti mārat(dumu.munus) šar (lugal) ilī(dingir meš) d lugal-giš ása[l]

    ši-i-ma [be-l]et be-le-e-ti kal-lat bēl(en) ilī(dingir meš) d lugal-abzu

    [š]i-⸢i-mašar-rat šar-ra-a-ti ḫi-rat šar (lugal) ilī(dingir meš) d asar-lú-ḫi

    d iš-tar d iš-tar meš ši-i-ma um-mi e-tel ilī(dingir meš) d en-zag

    Well versed in [… of heaven] and earth, the advice of Duranki, Cf. 126

    She is goddess among goddesses, daughter of the king of the gods, Lugal-asal, Cf. 126

    She is lady among ladies, daughter-in-law of the lord of the gods, Lugal-abzu, Cf. 104. 125

    She is queen among queens, wife of the king of the gods, Asalluḫi,

    Ištar among Ištars, she is the mother of the foremost of the gods, Enzag. Cf. 103. 126

    K.3031 o 11′–r 1 // Sm.1719 r 1–4 // BM.31749 r 1–2 // BM.35923 r 1–5 // BM.38468+ r 1–5

    Enzag is a god of Dilmun, identified with Nabû in Mesopotamia (Pomponio Reference Pomponio1978: 175–176; Nashef Reference Nashef1984: 8–10).

  63. 127. The line refers to the antediluvian king Alalgar i.e. to the second ruler of Eridu after kingship came down from heaven according to the ‘Sumerian King List’. The name of the king is written as e||á||a-làl-gar in the ‘Sumerian King List’ (s. Jacobsen Reference Jacobsen1939: 70; and George Reference George2011: 199 no. 96 o 3. 201 no. 97 o 3. 202 no. 98 i 5), as m a-lá-al-gar in the ‘Uruk List of Sages and Scholars’ (IM.65056 o 2 = W.20030/7, BagM Beih. 2, 89) and as ᾽Αλάπαρος by Berossus (De Breucker Reference De Breucker2012: 232).

  64. 128. aburrūtu is a hapax legomenon. The word is probably related to abrātu, a poetic designation for “people.”Footnote 50 The Assyrian word (a)gu(r)ru/atu, “ewe,” is probably irrelevant here.

  65. 130. Note the strange ending -ḫa in MSS SipNB1 and BabNBSch3 for what should be the masculine plural . An emendation to a fem. pl. -ḫa-<> in both MSS seems unlikely.

  66. 132. As interpreted here, šarru ana šarri is a distributive expression, like šattu ana šatti and arḫu ana arḫi (see the references to such expressions collected by Mayer Reference Mayer1989: 163 fn. 20). One may also consider restoring ana lugal dingir m[ šu-ba-a]r-šú-nu, although the spelling šubaršunu instead of the regular šubarrâšunu (< Sum. šu-bar-ra) would be surprising.Footnote 51 ana could also be interpreted as “in favor of,” i.e. the king establishes their freedom so that they can devote themselves to the service of the king (of the gods).

  67. 133. ramkūtu is just as common as ramkū as the plural of ramku (CAD R 127a: see also Still Reference Still2019: 194 with cases in which ramku designates a “priest” in general, not a specific caterogy). Alternatively, one may restore [… uru ak]-ka-di and assume that Akkad is here a poetic byname of Sippar, just like Eridu is of Babylon.Footnote 52 No other occurrence of ebbu as a type of priest is known,Footnote 53 so the term may just be an adjective: “pure ones.” The priests of Ištaran are presumably from Der, those of Šamaš probably from Sippar.

  68. 134. The word b/puḫl(a) was previously attested only in an inscription of Assurbanipal: adi sangugê/šangê bu-uḫ-la-le-e ašlula ana māt aššur, “(the Elamite gods) along with the šangû Footnote 54 and buḫlalû-priests I took captive to Assyria” (RINAP Ashurbanipal 9 v 33. 11 vi 46. 96 r i′ 2′). Vallat (Reference Vallat2001) suggested that the word is a composite of puḫu, “child,” and lar/lal, “priest,” so the meaning would be “seminarist.” Elamites resident in Babylonia during the Middle Babylonian period were generally of lower classes (Zadok Reference Zadok1987: 15; Sassmannshausen Reference Sassmannshausen2001: 133), but those mentioned in the text are priests of Šušinak (or: of Susa). Parallelism with buḫlû would suggest taking nibru ki-ú not as the gentilic (“Nippurean”) but rather as a designation of a group of priests (“priests of Enlil”?). The same usage may perhaps be attested in a late copy of an inscription of Kurigalzu II (MS 3210; George Reference George2011, no. 61 // George Reference George2012), according to which certain “Nippureans” (dumu meš nibru ki) were massacred in the courtyard of the temple Esaⓖdiⓖirene, the only known temple of such name being in Dūr-Kurigalzu; one of the possible explanations is that these were Nippureans living in Dūr-Kurigalzu (so Clayden Reference Clayden2017: 448 fn. 39); the place where they were massacred suggests perhaps that they were priests (so Bartelmus Reference Bartelmus2017: 254).

  69. 135. The verb uš-ba-áš-šú is interpreted as ušbaššū, i.e. as bâšu Š. Only a doubtful occurrence of the Š stem of bâšu is booked in the dictionaries (AHw. 1547b), in an Old Babylonian letter: matīma anāku ana bīt rāmānīya uš-bi-iš, “Have I ever made him ashamed of my own house?” in (Ashm-1923.342 = OECT 3, 74 = AbB 4, 152 l. 19).

  70. 136. Cf. BM.65653 vi 19′ (eBL transliteration, Hymn to Borsippa): šá a-bi-ri-i ḫa-áš-šú ḫa-at-nu la ma-gi-ri i-du-uš-šú ri-x [(o)]. ḫaššâ is probably the same word written as ḫa-aš-ša-a-ú in OB Lu A (MSL 12, 160):

    The word appears probably also in the ‘Hymn to Ninurta as Savior’ 38: ḫaš-⸢šá-mu⸣-ú ana emūqīšu u anāku akû adallal kâšu, “the weak one for his strength, and I, the helpless, praise you!”Footnote 55 The existence of the word ābirû, of uncertain etymology and not booked in the dictionaries, was first detected by Lambert (Reference Lambert1982: 282–283).

  71. 138. mār mīti for “orphan” or “heir” is attested only in Old Assyrian sources and in Nuzi (CAD M/2 140b), and literary manuscripts generally restrict the use of logograms to the most common words only; but no other parsing of dumu úš seems convincing.

  72. 139. As interpreted here, the Babylonians’ generosity to their captives is such that they would be willing to pay a talent of silver, an exhorbitant amount, to set them free.

  73. 140. AššNASch1 áš-pi has been read as [ša] našpi, “they (the people of Babylon) distribute rations of našpu-beer” (CAD Z 144a) and [šá k]a ?-áš-pi, “von (ihrem eigenen) Silber stellen sie einen Anteil bereit” (Maul/Manasterska Reference Maul and Manasterska2023: 112). In view of MS SipNB1, it should probably be read as (w)ašbu: (Neo-Assyrian texts occassionally write /p/ with b-signs and vice versa; see GAG3 §27d*; Deller Reference Deller1959: 234–242 §47; Parpola Reference Parpola1983: 255 fn. 457; Luukko Reference Luukko2004: 72f.; De Ridder Reference De Ridder2018: 127–130). When the Babylonians divide an inheritance with someone who is absent, they perform a pious act toward someone who would not have noticed their lack of piety. In doing so, they defy the cynical wisdom of the proverb BM.38297+ ii 8–10 (partially Lambert Reference Lambert1960a: 268): lú al-ti-la áš;-a-na-ne (x) aš-ak-ab [l]ú nu ti-la «ab» [e]me-sig-šè dug4-ga-ab | ša áš-bi e-pu-uš ṣi-bu-ti ša la áš-[bi] a-⸢kul⸣ ⸢kar*⸣-[ṣi-šú], “of the present, fulfill his desires; of the absent, say slander.” The term lā ašbu may have a more concrete meaning in the context, perhaps similar to Old Assyrian laššuʾu, “absent,” i.e. living in Aššur and not in the colony, as opposed to wašbu, “present” (Dercksen Reference Dercksen2004: 127–130).

  74. 141. The signs ik-⸢ke-e⸣ are relatively clear on the tablet. CAD K 351b parses the word as ki-i-ik-ke-e palḫi<š>, connecting it with the adverb kīkî, “how?” A rhetorical question or exclamation seems out of place in the context. Mayer Reference Mayer2009: 439 normalizes the word as kī ikkē palḫi and translates it as “in ehrfurchtig-gestimmter Weise.” This seems the best interpretation in view of the similar expressions kī ikkim “in launischer Weise, aus Laune,” and specially kī ikki rīqi ū kimilti, “in leichtfertiger Laune oder im Zorn” (Mayer Reference Mayer2009: 431, on the latter cf. also Paulus Reference Paulus2014: 438), albeit the additional -e seems difficult to justify.

  75. 144. MS SipNB1 could also be read - !, but the writing would be very strange; moreover, as reflected in AššNASch1, the line “is hardly correct as it stands” (Landsberger/Jacobsen Reference Landsberger and Jacobsen1955: 21 fn. 26). emūqāt, “the hosts,” enables a more satisfactory understanding of the line. As interpreted here, mišaru is a variant of išaru Footnote 56 (an interpretation as mīšari, “Ningirsu of righteousness,” i.e. “righteous Ningirsu,” is of course also possible, although poetry tends to avoid complex genitive chains). The variant mešrâ for mīšaru is attested also in two parallel lines of the šuʾila prayer ‘Ištar 1’, the first of which has imnuk mi-šá-ri, the second šá imnukki mešrâ (see also Mayer Reference Mayer1992: 40 fn. 22; Zgoll Reference Zgoll2003: 194. 196 ll. 17. 32). The variant is reminiscent of the writing of the word mīšaru as meš-šá-r v, which is attested occasionally (Jiménez Reference Jiménez2022: 64 ad 39) and may have originated as an erroneous parsing of it.

  76. 149. Since ugbabātu-priestesses were famously celibate –in ‘Atramḫasīs’ III they are among the professions established to to keep the population’s birth rate under control– the “partners” (ḫāmerū) mentioned in this line are in all likelihood their divine or spiritual spouses, i.e. the gods to which they are devoted (so also Stol Reference Stol2000b: 461f.).

  77. 150. The line seems to imply that the nadītu plays a role as a midwife (see already Harris Reference Harris and Biggs1964: 135; Lambert Reference Lambert and Haas1992: 144; Stol Reference Stol2000a: 172f.). The “midwife” (šabsūtu) is elsewhere associated with the qadištu (von Soden Reference von Soden1957: 119f.; Stol Reference Stol2000a: 173).

  78. 151. qašdātu are normally associated with wet-nursing: see Lambert Reference Lambert and Haas1992: 144f.; Stol Reference Stol2000a: 186–188. The word šuḫtu is a hapax legomenon, it is a purs-form of šaḫātu IV, “to wash.”

  79. 152–155. These lines could refer to the women of Babylon alone, or to all Babylonians (the ending in the verbs would then refer to “the people,” nišū).

  80. 153. The reading r[a-šá-a] follows Maul/Manasterska Reference Maul and Manasterska2023: 112; cf. Ebeling Reference Ebeling1925: 8: i ?-[na t]e-me-qí.

  81. 154. The word nakdu (on the reading with -k- instead of -q- see Stol Reference Stol2010: 43f.) is equated with palḫu in the commentary on ‘Theodicy’ l. 22 (Heinrich Reference Heinrich2022b): nak-di : pa[l-ḫu].

  82. 157. The restoration at the beginning is inspired by ‘Gilgameš’ MB Ug1 13 (George Reference George2022): šū rīmšina šina arḫātu, “(Gilgameš lets no young bride go free to her husband,) he is their (fem.) wild bull, they (fem.) are (his) cows,” see also ‘Gilgameš’ SB I 71). kul-lat bābili has been interpreted as “of all Babylon” (Foster 2005³: 878), “von ganz Babylon” (Ebeling 1925: 9; Reference Ebeling and Gressman1926: 216) and “von Babylon in seiner Gesamtheit” (Maul/Manasterska Reference Maul and Manasterska2023: 114). Although the expression seems solecystic (CAD K 505b gives no other example of kullat GN, but cf. tin.tir ki gab-bi e.g. in SAA 17, 21 r 3. 5), no better alternative suggests itself. Collation reveals that the sign after ⸢ká⸣.dingir.ra ki is su, not la, as read in previous editions. The word gained is sug/kullu, “herd.” The collation of the divine name at the end of the line allows the awkward reading found in earlier editions of the excerpt (an-da-ḫaṣ) to be dismissed, thereby eliminating the need to introduce a speaker into the context.

  83. 159. The readings adopted, t[a and i-⸢bat !⸣-[t]aq, look possible on the tablet, but not certain. In ḫu-ub-ba-t[a-šú-nu, one may either excise ba (ḫu-ubba»-t[a-šú-nu, i.e. ḫubt[ašunu) or take it as an anaptyptic vowel, although the expected vowel would be /bu/ (see Parpola Reference Parpola1983: 47).

  84. 191–198. The verbs in this section seem to be preterites and statives, not precatives.

  85. 192. Compare the apodosis attested e.g. in ‘Šumma Ālu’ XXIII 20′ (Freedman Reference Freedman2006: 204; also 1881,0727.59 r 10; eBL transliteration): bēl bīti šuāti ḫengalla uštabarra, “the owner of that house will enjoy permanent prosperity.”

  86. 193. Cf. ‘Enūma eliš’ I 13 (Heinrich Reference Heinrich2021): urrikū ūmī uṣṣibū šanāti, “Lengthy were they of days, added years to years.”

  87. 205. The word nagalmušu, equated with gitmālu, “perfect”; šaqû, “lofty”; and nabû, “shining” in ‘Malku’ I 68f. and IV 178 (Hrůša Reference Hrůša2010: 201. 305. 389), is elsewhere attested only in a fragment of a Neo-Assyrian prayer (VAT.11666 = KAL 9, 15 o (?) 2′) and in an Old Babylonian text that mentions Narām-Sîn (BM.120003 l. 37, see Lambert Reference Lambert1973: 361. 363).

  88. 206. Compare ‘Eriš šummi’ 14 (Fadhil/Jiménez Reference Fadhil and Jiménez2022: 233): šarru ekdu ša rāšuššu agîš etpuru (var.: šitpuru) burummī ellūti, “Fierce king, who wears on his head the pure heavens like a tiara.”

  89. 209. Compare BM.76692 6′ (eBL transliteration): […] x x x ub-bat dūr(bàd) [abni (?) …].

Acknowledgements

Thanks are expressed to M. Frazer and an anonymous reviewer for the careful revision of the manuscript. The text was read at the Keilschriftwerkstatt of LMU Munich in 2023, and ideas provided by its participants are acknowledged in the commentary. The tablet in the Iraq Museum is published with the kind permission of the College of Arts (University of Baghdad) and the State Board of Antiquities and Heritage; those in the British Museum are presented here by permission of the Trustees of the British Museum. Thanks are expressed to J. Eule (Vorderasiatisches Museum) for facilitating the collation of MS AššNASch1, to J. Taylor for providing access to the British Museum’s collection of formerly unnumbered fragments (MS SipNB4) and to K. Simkó for some last-minute collations. The article was written under the auspices of a Gerda Henkel “Patrimonies” Project; the tablet from the Sippar Library was conserved in the framework of the project “Cuneiform Artefacts from Iraq in Context” (Bayerische Akademie der Wissenschaften).

Without the eBL platform (https://www.ebl.lmu.de/) and the work of the entire eBL team, it would be impossible for this edition to use 21 manuscripts, consisting of 31 fragments, most of them previously published only on eBL. A debt of gratitude is owed to the whole team, especially to those who have contributed digital (Zs. J. Földi, A. Hätinen, T. Mitto, J. Peterson, L. Sáenz, H. Stadhouders, J. Taniguchi) or manuscript (A. R. George, W. G. Lambert (†), W. R. Mayer) transliterations that served as the basis for the identification and edition of the manuscripts of this text.

The article is dedicated to the honoree of the ‘Approaches to Cuneiform Literature’ conference, in deep gratitude for his mentorship and support over the years.

Footnotes

1 The eleven excerpts of the text that have been found so far put it on a par with texts such as ‘Enūma eliš’ (74 excerpts in the edition of Heinrich Reference Heinrich2021), ‘Marduk 2’, ‘Marduk 1’, and the ‘Aluzinnu Text’ (24, 13, and 19 excerpts, respectively, in Jiménez’s forthcoming edition), the ‘Hymn to Šamaš’ (21 in the edition of Rozzi Reference Rozzi2021), ‘Ludlul’ (16 excerpts in the edition of Hätinen Reference Hätinen2022), ‘Marduk’s Address to the Demons’ (16 excerpts in the edition of Peterson Reference Peterson2020), and the ‘Counsels of Wisdom’ (13 excerpts in the edition of Földi Reference Földi2022).

2 See Jiménez Reference Jiménez2022: 21–25.

3 See commentary on ll. 39 and 51.

4 Fadhil/Jiménez Reference Fadhil and Jiménez2022: 239, 245, 252f. l. 64.

5 Restored after the apparent quotation of these lines in Cutha.1 (‘Esoteric Babylonian Commentary’ l. 17f., ed. Biggs Reference Biggs1968: 54): d gíra : d ānu(60) : išātu(izi) : ul-la-nu : d ēa(40) : mu-ú | tu15 tu .ḫur.sag : d+en-líl : šá-a-ri : šu-ut pî(ka) šá ṣa-a-tú e-du-tú, “Gira = Anu = Fire; ‘Then’ = Ea = Water; ‘Mountain Wind’ = Enlil = Wind; prominent oral teachings of lexical lists.” Scurlock and al Rawi (Reference Scurlock and al Rawi2006: 371f.) suggest that the “Esoteric Babylonian Commentary” is a treatise that tries to find Mesopotamian equivalents to the Hellenistic four elements, but the parallel in 1881,0204.419 means, on the one hand, that it speaks of three elements, not four; on the other, that it is a Mesopotamian tradition, not an attempt to fit Hellenistic philosophy into it.

6 The arrival of spring in the arid Babylonian lands has often surprised travelers as a sort of oasis in the middle of the desert. Compare the poetic description of the almost unexpected spring experienced by W. König in Babylon in 1931 (König Reference König1940: 163f.)

7 Lev. 19: 32f.: “And if a stranger sojourn with thee in your land, ye shall not vex him (w e ḵî-yāḡûr ʾitt e ḵā gēr b ə -ʾarṣ e ḵem lōʾ ṯônû ʾōṯô). | But the stranger that dwelleth with you shall be unto you as one born among you, and thou shalt love him as thyself; for ye were strangers in the land of Egypt” (KJV).

8 The catchline to the ‘Hymn to Šamaš’ appears in manuscripts of diverse provenance: Nineveh, Neo-Babylonian Sippar and Late Babylonian Babylon. See provisionally Fadhil/Jiménez Reference Fadhil and Jiménez2019: 171.

9 See Jiménez Reference Jiménez2017: 92–94.

10 ‘Agušaya’ A r vi 14′-16′ and B r v 5-6 (Pohl Reference Pohl2022: 124-175), as well as in BM.93828 (CT 15, 3f. i 7, Pohl Reference Pohl2022: 96-101), the latter a hymn to Adad often classified as an epic.

11 Some of these ideas were developed in conversation with A. Zgoll, to whom thanks are expressed.

12 Edited by Lambert Reference Lambert2013: 366–375. A new manuscript of the text from the Sippar Library (IM.132602), reported in al-Rawi/George 1990: 149 fn. 1, preserves some ten new lines of the text.

13 Creation and spring co-occur in the only other extensive description of spring known in Mesopotamian literature, the prologue of the disputation poem ‘Ox and Horse’, constituted by a cosmogony in which the waters “turned the wastelands into damp soil” (Jiménez Reference Jiménez2021 A+17).

14 BM.61625+ and BM.65653. A draft transliteration can be found on the eBL Fragmentarium, based on the transliteration kindly made available for the project by A.R. George. BM.38779 // BM.43372 and BM.83029 may represent additional pieces of the text.

15 On the “Marduk Syllabus,” see provisionally Jiménez Reference Jiménez2022: 6f.

17 See below the comment on l. 134.

18 The join was recorded by Hilgert Reference Hilgert2004: 529 no. 247. As noted by Hilgert (Reference Hilgert2004: 234–241), the fact that both fragments were found in different niches is not uncommon: of the 19 joined manuscripts from the Sippar Library, only 3 are recorded in the excavation journal as having both their joining fragments found in the same niche. Hilgert assumes as a working hypothesis that the information in the excavation report is incorrect; but, as will be discussed in a future study, most of the excavation numbers recorded in the museum catalogue are confirmed by the partial catalogue that A.R. George made at the Sippar excavation house in the spring of 1989 at the invitation of Abdulillah Fadhil, which George has generously made available to the authors. It must be assumed, therefore, that most of the fragments that can be joined were found in separate niches, i.e. that the collection was disturbed in antiquity. An anonymous reviewer has noted that this observation is compatible with the archaeological evidence, since the room in which the tablets were found (355) was blocked in antiquity by “briques rangées en hâte et sans ordre pour fermer cette chambre” (al-Jadir Reference al-Jadir1998: 713), and since only about 15% of the capacity of the niches was utilized (Hilgert Reference Hilgert2013: 143). This fact suggests that the room was robbed and then closed by a wall.

19 Manuscripts which mark scansion also use the ruling to divide between two verses written on a single line, as does MS SipNB1 in l. 51.

20 It is uncertain whether this feature of the manuscripts is what is described as ṣullupu, “sliced,” in colophons, as advocated by Lambert (Reference Lambert1960a: 66) and, following him, by Hunger (Reference Hunger1968: 7) and Livingstone (Reference Livingstone2010: 593): of the two known occurrences of the term, one appears on a tablet with this format (BM.93016 [CT 13, pl. 14f.] = ‘Enūma eliš’ IV), while the other appears in a tāmītu-prayer (Rm.601 = Lambert Reference Lambert2007, pls. 47f.) that is neither metrically divided nor seems possible to so divide.

21 SipNB2 belongs to the 82-9-18 consignment, which comes mostly from Abu Habba (Reade apud Leichty Reference Leichty1986, xxxiii); SipNB4 to the 84-2-11 consignment, which contains pieces from Sippar, Babylon, and other northern Babylonian cities (Leichty/Finkelstein [e. a.] 1988: xiii).

22 As first described by George Reference George2000: 260; Reference George2003: 408; Frame/George Reference Frame and George2005: 266. Note in 108 (BabLB1 and BabLB3), bàd in 114 (2× BabLB1), lagab in 20 (BabLB1), šar in 110 (BabLB1), nag in 1 and 15 (BabLB1) and mes in 95 (BabLB1). Although the absence of upper wedge is a diagnostic feature of the late period, its presence does not seem to have diagnostic value: thus, signs of this type in BabLB2 have an upper horizontal wedge (lu in 46 and 115, engur in 116, ma5 in 117), as also occurs in other tablets of the 2nd/1st century. Cuneiform paleography of this period is, however, in desperate need of further study.

23 On the late tablets from the Spartali collection, see Britton 1991/Reference Britton1993: 71–73; George Reference George2000: 260. To the 81-7-6 consignment belong the two latest dated non-astrological tablets known, BM.45746 (CTL 1, 85), dated to October 35 BCE; and BM.46070(+), dated February/March 22 BCE (Jiménez/Mitto Reference Jiménez and Mitto2024).

24 See also Maul/Manasterska (Reference Maul and Manasterska2023: 20a), who include only this tablet as a product of the final rung of the education ladder.

25 E.g. 146: plural masc. -ātu; 150: na-da-te; 151: tēlilte; o 13: mé-e-tu (mītu); r 3: -te (2×).

26 BabNB1 and BabNB2 do not seem to be part of the same tablet, since the script in MS BabNB1 is considerably smaller.

27 BM.45986+ BM.46063+ BM.46121+ BM.46166 were joined by I. L. Finkel in 1984. BM.42720 (81-7-1, 484), BM.42739 (81-7-1, 503), and BM.43087 (81-7-1, 851) may be part of the same tablet, but no overlap with the known sections could be found. Not part of the same tablet, although thematically possibly related, are BM.46101 and BM.42865. The colophon fragment BM.42856, dated 179 [SE/AE], i.e. 133/132 or 69/68 BCE, and BM.46118, written in very late script, could be part of the tablet but it is very far from certain.

28 Several fragments are suspected to be part of the same manuscript, but no overlap can be detected: BM.40017 (unlikely), BM.38736 (possible, but no overlap), and BM.39420 (likely, but no overlap).

29 BabNBSch1 was recognized as a duplicate of BabLB1 and published by George (Reference George1992: 230 no. 46); Foster (20053: 878) compares it with KAR 321 (AššNASch1), which can now be shown to be part of this same text.

30 Editions of all these texts, and photographs of most of them, can be found in the eBL Fragmentarium (https://www.ebl.lmu.de/fragmentarium). See also above fn. 27.

31 MS BabNBSch7, a school tablet, reads: “you let the herd rest.”

32 Scil., with the waters (not “the creatures”).

33 Scil., of the waters, or else of water, fire, and air.

34 Scil. Marduk.

35 Either the gate of Babylon or the celestial gate of Marduk’s star: see commentary.

36 MS BabNB2 reads “its (grain-)heaped meadows.” MSS BabLB2 and BabLB3 read “its prepared meadows.”

37 MSS BabLB2 and BabLB3 read “Gathered (or: poured) thanks to it are the ale (and) the sheaves.”

38 MS AššNASch1 reads “they piously pray.”

39 MS AššNASch1 reads “(Their) strength is Ningirsu, (their) righteousness the sweet Alala”; see commentary.

40 MS AššNASch1 reads “merriments.”

41 MS BabNBSch4 reads: “was [placed] upon them.”

42 The tablet BM.39042 (eBL: J. Peterson) has a rubric (or catchline) ending […-t]ì d amar.utu, but seems to be otherwise irrelevant for the present text.

43 Greek transcriptions of Akkadian use a single sign (σ) for the three Akkadian sibilants s, , and š, which is explicable by the inability of the Greek alphabet to reflect all Akkadian sounds (Knudsen Reference Knudsen1990: 152; Geller Reference Geller1997: 65; Westenholz Reference Westenholz2007: 279).

44 ‘Marduk 2’ 55–57 = ‘Counsels of Wisdom’ A+99–A+101 (Földi Reference Földi2022, probably a quotation from ‘Marduk 2’ in ‘Counsels’); ‘Marduk 2’ 80 and 82 = ‘Marduk 1’ 5 and 7 (see Fadhil/Jiménez Reference Fadhil and Jiménez2019: 173); ‘Marduk 2’ a+9f. = ‘Ninurta as Savior’ 1f. (Mitto Reference Mitto2022b); ‘Marduk 2’ a+45 = ‘Hymn to Šamaš’ 198 (Rozzi Reference Rozzi2021).

45 Verbal rection of the participle (i.e., bānû bīssu with bītu in accusative) is in theory possible, but very rare: see the few examples collected in GAG³ §148c*, Groneberg Reference Groneberg1987: I 87f. II 39, and Guichard Reference Guichard2014: 28 ad i 6. In any case, this interpretation would result in an anacoluthon, since the object of the description in the following lines is not Marduk (the “builder”) but Esagil.

46 The etymological w- is attested occasionally in the first millennium with the verb (w)atû(m) (see Jiménez Reference Jiménez2016: 223; the word a-na i-tu-ti-ia in the Old Babylonian letter BM.80160 l. 7 [CT 45, 60], ascribed to itûtu in AHw. 1564b, is no doubt from itû, “neighbor”; cf. Charpin Reference Charpin1986: 130: “dans mon voisinage”). The variation between -ītu and -ūtu is known in nouns derived from III-w roots, e.g. šapû I, whose feminine form is either šapītu or šapūtu (CAD Š/1 487b, AHw. 1177a), or šūpû, attested in feminine as šūpūtu and šūpītu (on the latter, see Beaulieu Reference Beaulieu1995: 194 l. 6). The form with -a- is either a plural or a pirsat- form.

47 The word is booked as maṭṭalātu in CAD M/1 428a, following its occurrence in ‘Aa’ I/6 261 (MSL 14, 233) as maṭ-ṭa-la-tu 4. The spelling maṭṭaltu in AHw. 635b is probably motivated by Landsberger’s comparison of the word with maṣṣartu (apud Moran Reference Moran1959: 265), a word whose poetic bound form ends in -at (maṣṣarat), and which is apocopated as maṣrat in a MS of ‘Enūma eliš’ V 46 (with which compare maṭlat in the present text and in the Esarhaddon passage mentioned). The syncope of the second syllable in maṭ(ṭa)lātu is, though elsewhere attested, a rare phenomenon in Akkadian morphology (see GAG §12f).

48 Contaminations between tašīltu and tamšīltu – more specifically, writings of tašīltu with tam- – are attested elsewhere. Compare in ‘Bullussa-rabi’s Gula Hymn’ 166: ina erēbīya ta-ši-la-a-tu 4 (MA BabLB2) vs tam-ši-la-a-ti (MS BabLB1, Földi Reference Földi2021a) and LKA 32 o 8 (SAA 3, 8): uru tam-ši-la-a-ti, “city of delights” (in parallelism with uru nigûti and uru isinnāti, among others).

49 M. Krebernik (privatim), however, considers the possibility of comparing parû, “to vomit” to Ar. frġ, “to empty” (Biberstein-Kazimirski Reference Biberstein-Kazimirski1850, II 581a X.3: “Vomir, avoir des vomissements”), and thus to regard it as a word with a different etymon from parāḫu I and parāʾu II (EDA P0403).

50 Although the etymology of abrātu is unknown, according to CAD T 30b, tabrātu would be a byform of abrātu (but cf. Kogan/Krebernik [e. a.] 2020: 206).

51 One may, however, compare in the ‘Binning Tablet 1’ l. 13′ (Walker/Kramer Reference Walker and Kramer1982: 70; Frame Reference Frame1995: 158): ⸢šu-bar-šú-nu (or šu.bar- šú-nu?) iškun. Note the loanword kisurrû (< Sum. ki-sur-ra), of which the bound form kisur- (wr. ki-sur-šu and ki-su-ur|úr-šu) is attested in MB and NB inscriptions (CAD K 434, AHw. 488a). Alternatively, one may read [šu-b]ar ! -ri-šú-nu, less convincing epigraphically, since the bound form šubarrî- (a plural?) is attested in the Nippur copy of ‘Advice to a Prince’ l. 27 (IM.77087; Cole Reference Cole1996: 268; Mitto Reference Mitto2022a): šu-ba-ri-šú-nu (DT 1: šu-ba-ra-šú-nu) ukinnū.

52 It has been noted that there is a close connection between Sippar and Akkad in the Middle and Neo-Babylonian periods (McEwan Reference McEwan1982: 12f.; Sommerfeld Reference Sommerfeld2014: 157f.), and some times both terms appear to be interchangeable, e.g. in the text BM.60381 (Nbn. 662; cited by McEwan Reference McEwan1982: 13a and Jursa Reference Jursa2010: 112 fn. 629), which mentions at the beginning garments from the temple of the “Lady of Sippar” (l. 2) while the summary at the end represents them as “from Akkad” (l. 16: ta a-kad ki).

53 The Old Babylonian profession ebbu, “controller” (on which see Földi Reference Földi2021b: 206 with further literature) is probably irrelevant here. Compare also the title ubbubūti, “cleared (‘of claims’),” given to the Babylonians in the inscription Si.4+ (Frazer/Adalı Reference Frazer and Adalı2021: 234. 244).

54 One manuscript reads sangugê, which according to Borger Reference Borger1996: 54 is the best reading.

55 Mayer Reference Mayer1992: 25; Mitto Reference Mitto2022b. Mayer (Reference Mayer1992: 40), following a suggestion by W. G. Lambert, parses the word as a byform of ḫiššāʾum, ḫiššamû, and translates: “Der Kraftvolle nach seiner Starke, und auch ich, der Hilflose, preise dich!”

56 On mišaru as a variant of išaru, “righteous,” see Deller/Mayer Reference Deller and Mayer1984: 117.

References

Abusch, T. and Schwemer, D. 2011. Corpus of Mesopotamian Anti-Witchcraft Rituals. Volume One. Ancient Magic and Divination 8/1. Leiden: Brill.CrossRefGoogle Scholar
al-Jadir, W. 1998. “Découverte d’une bibliothèque dans le temple de la ville de Sippar (Abu Habbah)” in: H. Erkanal [e. a.], ed. XXXIVème Rencontre Assyriologique Internationale - XXXIV. Uluslararasi Assiriyoloji Kongresi. Ankara, pp. 707–715.Google Scholar
al Rawi, F. N. H. George, A. R. 1990. “Tablets from the Sippar Library II. Tablet II of the Babylonian Creation Epic”. Iraq 52: 149157 (DOI: https://www.doi.org/10.2307/4200325).CrossRefGoogle Scholar
Attinger, P. 2021. Glossaire sumérien–français. Principalement des textes littéraires paléobabyloniens. Wiesbaden: Harrassowitz.CrossRefGoogle Scholar
Bartelmus, A. 2016. Fragmente einer großen Sprache. Sumerisch im Kontext der Schreiberausbildung des kassitenzeitlichen Babylonien. Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie 12. Berlin: De Gruyter.Google Scholar
Bartelmus, A. 2017. “Die Götter der Kassitenzeit. Eine Analyse ihres Vorkommens in zeitgenössischen Textquellen” in: A. Bartelmus/K. Sternitzke, ed. Karduniaš. Babylonia under the Kassites. Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie 11. Boston/Berlin: De Gruyter, pp. 245–312.Google Scholar
Beaulieu, P.-A. 1995. “Theological and philological speculations on the names of the Goddess Antu”. Orientalia 64: 187213.Google Scholar
Bennett, E. A. 2023. “Syncretistic Hymn to Gula”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/3/12).Google Scholar
Biberstein-Kazimirski, A. d. 1850. Dictionnaire arabe-français. Paris: Barrois.Google Scholar
Biggs, R. D. 1968. “An Esoteric Babylonian Commentary”. Revue d’assyriologie et d’archéologie orientale 62: 5157.Google Scholar
Blaschke, T. 2018. Euphrat und Tigris im Alten Orient. Leipziger Altorientalistische Studien 9. Wiesbaden: Harrassowitz.CrossRefGoogle Scholar
Borger, R. 1967. Handbuch der Keilschriftliteratur. Band I. Repertorium der sumerischen und akkadischen Texte. Berlin/New York: de Gruyter.CrossRefGoogle Scholar
Borger, R. 1996. Beiträge zum Inschriftenwerk Assurbanipals. Die Prismenklassen A, B, C = K, D, E, F, G, H, J und T sowie andere Inschriften. Wiesbaden: Harrassowitz.Google Scholar
Britton, J. P. 1991/1993. “A Table of 4th Powers and Related Texts From Seleucid Babylon”. Journal of Cuneiform Studies 43/45: 71–87 (DOI: https://www.doi.org/10.2307/1359847).CrossRefGoogle Scholar
Charpin, D. 1986. “Transmission des titres de propriété et constitution des archives privées en Babylonie ancienne” in: K. R. Veenhof, ed. Cuneiform Archives and Libraries. Papers read at the 30e Rencontre Assyriologique Internationale. Leiden, pp. 121–140.Google Scholar
Clayden, T. 2017. “Dūr-Kurigalzu: New Perspectives” in: A. Bartelmus/K. Sternitzke, ed. Karduniaš. Babylonia under the Kassites. Untersuchungen zur Assyriologie und Vorderasiatischen Archäologie 11. Boston/Berlin: De Gruyter, pp. 437–478.Google Scholar
Cole, S. W. 1996. Nippur IV. The Early Neo-Babylonian Governor’s Archive from Nippur. Oriental Institute Publications 114. Chicago: The Oriental Institute.Google Scholar
De Breucker, G. 2012. De Babyloniaca van Berossos van Babylon. Inleiding, editie en commentaar. PhD thesis: Groningen.Google Scholar
De Ridder, J. J. 2018. Descriptive Grammar of Middle Assyrian. Leipziger Altorientalistische Studien 8. Wiesbaden: Harrassowitz.Google Scholar
Deller, K. 1959. Lautlehre des Neuassyrischen. PhD thesis: Wien.Google Scholar
Deller, K. and Mayer, W. R. 1984. “Akkadische Lexikographie: CAD M”. Orientalia 53: 72–124.Google Scholar
Dercksen, J. G. 2004. Old Assyrian Institutions. MOS Studies 4. Leiden: Nederlands Instituut voor het Nabije Oosten.Google Scholar
Doty, L. T. 1993. “Akkadian bīt pirišti” in: M. E. Cohen [e. a.], ed. The Tablet and the Scroll. Near Eastern Studies in Honour of W.W. Hallo. Bethesda MD: CDL, pp. 87–89.Google Scholar
Ebeling, E. 1925. Der akkadische Mythus vom Pestgotte Era. Berliner Beiträge zur Keilschriftforschung. Berlin: Selbstverlag.Google Scholar
Ebeling, E. 2 1926. “Babylonisch-assyrische Texte” in: Gressman, H., ed. Altorientalische Texte und Bilder zum alten Testamente. Zweite Auflage. Berlin/Leipzig: De Gruyter, pp. 108–449.Google Scholar
Ermidoro, S. 2013. “Babilonesi allo specchio: un esercizio di autorappresentazione” in: G. Baldacci [e. a.], ed. Percorsi identitari tra Mediterraneo e Vicino Oriente Antico. Contributi del Dottorato in Storia Antica e Archeologia. Quaderni del Dipartimento di Scienze dell’Antichità e del Vicino Oriente 9. Padova, pp. 89–106.Google Scholar
Fadhil, A. A. 2022. “Dialogue of Pessimism. With contributions by Z. J. Földi, A. Hätinen, E. Jiménez, T. D. N. Mitto and G. Rozzi. Translated by Benjamin R. Foster”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/4).Google Scholar
Fadhil, A. A. and Jiménez, E. 2019. “Literary Texts from the Sippar Library I: Two Babylonian Classics”. Zeitschrift für Assyriologie und Vorderasiatische Archäologie 109: 155176 (DOI: https://www.doi.org/10.1515/za-2019-0012).CrossRefGoogle Scholar
Fadhil, A. A. and Jiménez, E. 2022. “Literary Texts from the Sippar Library III: Eriš šummi, a Syncretistic Hymn to Marduk”. Zeitschrift für Assyriologie und Vorderasiatische Archäologie 112: 229274 (DOI: https://www.doi.org/10.1515/za-2022-0004).CrossRefGoogle Scholar
Finkel, I. L. 1988. “Adad-apla-iddina, Esagil-kīn-apli, and the series SA.GIG” in: E. Leichty [e. a.], ed. A Scientific Humanist: Studies in Memory of Abraham Sachs. Occasional Publications of the Samuel Noah Kramer Fund 9. Philadelphia: University Museum, pp. 143–159.Google Scholar
Földi, Z. J. 2021a. “Bullussa-rabi’s Hymn to Gula. With Contributions by A.C. Heinrich and E. Jiménez”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/3/7).Google Scholar
Földi, Z. J. 2021b. “New Fragments of the Counsels of Wisdom ”. KASKAL. Rivista di storia, ambienti e culture del Vicino Oriente Antico 18: 189213.Google Scholar
Földi, Z. J. 2022. “Counsels of Wisdom. With contributions by A. C. Heinrich, E. Jiménez and T. D. N. Mitto. Translated by Benjamin R. Foster”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/3).Google Scholar
Foster, B. R. 2005³. Before the Muses. An Anthology of Akkadian Literature. Bethesda MD: CDL Press.Google Scholar
Frame, G. 1995. Rulers of Babylonia from the Second Dynasty of Isin to the End of Assyrian Domination (1157-612 B.C.). The Royal Inscriptions of Mesopotamia. Babylonian Periods 2. Toronto/Buffalo/London: University of Toronto Press.Google Scholar
Frame, G. 2020. The Royal Inscriptions of Sargon II, King of Assyria (721-705 BC). Royal Inscriptions of the Neo-Assyrian Period 2. University Park PA: Eisenbrauns.Google Scholar
Frame, G. and George, A. R. 2005. “The Royal Libraries of Nineveh: New Evidence for King Ashurbanipal’s Tablet Collecting”. Iraq 67: 265284 (DOI: https://www.doi.org/10.1017/S0021088900001388).CrossRefGoogle Scholar
Frazer, M. 2020. “The Sign maK”. Nouvelles Assyriologiques Brèves et Utilitaires 2020/127.Google Scholar
Frazer, M. and Adalı, S. F. 2021. ““The Just Judgements that Ḫammu-rāpi, a Former King, Rendered”: A New Royal Inscription in the Istanbul Archaeological Museums”. Zeitschrift für Assyriologie und Vorderasiatische Archäologie 111: 231262.CrossRefGoogle Scholar
Freedman, S. M. 2006. If a City Is Set On a Height. The Akkadian Omen Series šumma ālu ina mēlê šakin. Volume 2: Tablets 22-40. Occasional Publications of the Samuel Noah Kramer Fund 19. Philadelphia: University Museum.Google Scholar
Geller, M. J. 1997. “The Last Wedge”. Zeitschrift für Assyriologie und Vorderasiatische Archäologie 87: 4395 (DOI: https://www.doi.org/10.1515/zava.1997.87.1.43).CrossRefGoogle Scholar
George, A. R. 1992. Babylonian Topographical Texts. Orientalia Lovaniensia Analecta 40. Leuven: Peeters.Google Scholar
George, A. R. 1997. “Assyria and the Western World” in: S. Parpola/R. M. Whiting, ed. Assyria, 1995: Proceedings of the 10th Anniversary Symposium of the Neo-Assyrian Text Corpus Project Helsinki, September 7-11, 1995. Helsinki: Helsinki University Press, pp. 69–75.Google Scholar
George, A. R. 2000. “Four Temple Rituals from Babylon” in: A. R. George/I. L. Finkel, ed. Wisdom, Gods and Literature. Studies in Assyriology in Honour of W.G. Lambert. Winona Lake IN: Eisenbrauns, pp. 259299.Google Scholar
George, A. R. 2003. The Babylonian Gilgamesh Epic. Introduction, Critical Edition and Cuneiform Texts. Oxford/New York: Oxford University Press.Google Scholar
George, A. R. 2006. “Babylonian texts from the Folios of Sidney Smith, Part Three. A Commentary on a Ritual of the Month Nisan” in: A. K. Guinan [e. a.], ed. If a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty. Cuneiform Monographs 31. Leiden: Brill, pp. 173–185.Google Scholar
George, A. R. 2011. Cuneiform Royal Inscriptions and Related Texts in the Schøyen Collection. Cornell University Studies in Assyriology and Sumerology 17. Bethesda, MD: CDL Press.Google Scholar
George, A. R. 2012. “CUSAS 17 no. 61”. Nouvelles Assyriologiques Brèves et Utilitaires 2012/72.Google Scholar
George, A. R. 2022. “Poem of Gilgameš Chapter Middle Babylonian Ug1. With contributions by E. Jiménez and G. Rozzi”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/1/4).Google Scholar
Gesche, P. D. 2001. Schulunterricht in Babylonien im ersten Jahrtausend v. Chr. Alter Orient und Altes Testament 275. Münster: Ugarit-Verlag.Google Scholar
Groneberg, B. 1987. Syntax, Morphologie und Stil der jungbabylonischen „hymnischen“ Literatur. Freiburger Altorientalische Studien 14. Stuttgart: Steiner-Verlag-Wiesbaden.Google Scholar
Guichard, M. 2014. L’épopée de Zimrī-Lîm. Florilegium Marianum 14. Paris: SEPOA.Google Scholar
Harris, R. 1964. “The nadītu Woman” in: Biggs, R. D., ed. Studies Presented to A. Leo Oppenheim. Chicago: The University of Chicago Press, pp. 106135.Google Scholar
Hätinen, A. 2022. “Righteous Sufferer (Ludlul bēl nēmeqi). With contributions by Zs.J. Földi, A. C. Heinrich, E. Jiménez and T. D. N. Mitto. Translated by Benjamin R. Foster”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/2).Google Scholar
Heinrich, A. C. 2021. “Poem of Creation (Enūma eliš). With contributions by Zs. J. Földi, A. Hätinen, E. Jiménez and T. Mitto. Translated by Benjamin R. Foster.”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/1/2).Google Scholar
Heinrich, A. C. 2022a. “The Poor Man of Nippur. With contributions by E. Jiménez and T. D. N. Mitto. Translated by Benjamin R. Foster”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/1/11).Google Scholar
Heinrich, A. C. 2022b. “Theodicy. With contributions by E. Jiménez and T. Mitto. Translated by Benjamin R. Foster.”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/1).Google Scholar
Held, M. 1976. “Two Philological Notes on Enūma Eliš” in: B. L. Eichler, ed. Kramer Anniversary Volume. Cuneiform studies in honor of Samuel Noah Kramer. Alter Orient und Altes Testament 25. Kevelaer: Neukirchener Verlag, pp. 231–239.Google Scholar
Hilgert, M. 2004. Bestand, Systematik und soziokultureller Kontext einer neubabylonischen „Tempelbibliothek“. Ein Beitrag zur altorientalischen Textsammlungstypologie. Habilitationsschrift: Friedrich-Schiller-Universität Jena.Google Scholar
Hilgert, M. 2013. “„Tempelbibliothek“ oder „Tafeldepot“? Zum rezeptionspraktischen Kontext der „Sippar-Bibliothek“” in: K. Kaniuth [e. a.], ed. Tempel im Alten Orient. 7. Internationales Colloquium der Deutschen Orient-Gesellschaft, 11.-13. Oktober 2009. Wiesbaden: Harrassowitz, pp. 137–150.Google Scholar
Hrůša, I. 2010. Die akkadische Synonymenliste malku = šarru. Eine Textedition mit Übersetzung und Kommentar. Alter Orient und Altes Testament 50. Münster: Ugarit-Verlag.Google Scholar
Hunger, H. 1968. Babylonische und assyrische Kolophone. Alter Orient und Altes Testament 2. Kevelaer/Neukirchen-Vluyn: Neukirchener Verlag.Google Scholar
Jacobsen, T. 1939. The Sumerian King List. Assyriological Studies 11. Chicago: The University of Chicago Press.Google Scholar
Jiménez, E. 2016. “Loose Threads of Tradition. Two Late Hemerological Compilations”. Journal of Cuneiform Studies 68: 197227 (DOI: https://www.doi.org/10.5615/jcunestud.68.2016.0197).CrossRefGoogle Scholar
Jiménez, E. 2017. The Babylonian Disputation Poems. With Editions of the Series of the Poplar, Palm and Vine, the Series of the Spider, and the Story of the Poor, Forlorn Wren. Culture and History of the Ancient Near East 87. Leiden/Boston: Brill (DOI: https://www.doi.org/10.1163/9789004336261).CrossRefGoogle Scholar
Jiménez, E. 2020. “New Fragments of Gilgameš Chiefly from the ‘Babylon Collection’”. KASKAL. Rivista di storia, ambienti e culture del Vicino Oriente Antico 17: 236246.Google Scholar
Jiménez, E. 2021. “Palm and Vine. Translated by E. Jiménez”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/12).Google Scholar
Jiménez, E. 2022. Middle and Neo-Babylonian Literary Texts in the Frau Professor Hilprecht Collection, Jena. Texte und Materialien der Frau Professor Hilprecht Collection 13. Wiesbaden: Harrassowitz (DOI: https://www.doi.org/10.13173/9783447118811).CrossRefGoogle Scholar
Jiménez, E. 2024. “Marduk and the Battle with the Sea: On the Dating of Enuma Elish” in: J. Haubold [e. a.], ed. Enuma Elish. The Babylonian Epic of Creation. The Library of Babylonian Literature 1. New York: Bloomsbury.CrossRefGoogle Scholar
Jiménez, E. and Mitto, T. 2024. “The Latest Dated Non-Astrological Tablet”. Nouvelles Assyriologiques Brèves et Utilitaires 2024/47.Google Scholar
Jursa, M. 2001/2002. “Göttliche Gärtner? Eine bemerkenswerte Liste”. Archiv für Orientforschung 48/49: 76–89.Google Scholar
Jursa, M. 2010. Aspects of the Economic History of Babylonia in the First Millennium BC. Economic Geography, Economic Mentalities, Agriculture, the Use of Money and the Problem of Economic Growth. Alter Orient und Altes Testament 377. Münster: Ugarit-Verlag.Google Scholar
Kilmer, A. D. 1991. “An Oration on Babylon”. Altorientalische Forschungen 18: 922.Google Scholar
Knudsen, E. E. 1990. “On Akkadian Texts in Greek Orthography” in: E. Keck [e. a.], ed. Living Waters: Scandinavian Orientalistic Studies Presented to Frede Løkkegaard on His Seventy-Fifth Birthday, January 27 th 1990. Copenhagen: Museum Tusculanum Press, pp. 147–161.Google Scholar
Kogan, L., Krebernik, M., Linkohr, O., Nurullin, R. and Sęk, O. I. 2020. Etymological Dictionary of Akkadian Volume 1/1. Volume 1/1: Roots beginning with P and B. Boston/Berlin: De Gruyter.Google Scholar
König, W. 1940. Neun Jahre Irak. Brünn/München/Wien: Rohrer.Google Scholar
Kouwenberg, N. J. C. 2010. The Akkadian Verb and its Semitic Background. Languages of the Ancient Near East 2. Winona Lake IN: Eisenbrauns.Google Scholar
Krebernik, M. 1987/1990a. “Lugal-abzu”. Reallexikon der Assyriologie und Vorderasiatischen Archäologie 7: 110a.Google Scholar
Krebernik, M. 1987/1990b. “Lugal-asal”. Reallexikon der Assyriologie und Vorderasiatischen Archäologie 7: 115116.Google Scholar
Lambert, W. G. 1960a. Babylonian Wisdom Literature. Oxford: Clarendon Press.Google Scholar
Lambert, W. G. 1960b. “Three Literary Prayers of the Babylonians”. Archiv für Orientforschung 19: 4766, pls. VIII–XXIII.Google Scholar
Lambert, W. G. 1973. “Studies in Nergal [Review of von Weiher AOAT 11]”. Bibliotheca Orientalis 30: 355363.Google Scholar
Lambert, W. G. 1980. “Review of Cagni Poem of Erra (SANE 1/3)”. Journal of Semitic Studies 25: 9496.Google Scholar
Lambert, W. G. 1982. “Review of AHw 15-16 [T-Z]”. Journal of Semitic Studies 27: 281286.CrossRefGoogle Scholar
Lambert, W. G. 1992. “Prostitution” in: Haas, V., ed. Außenseiter und Randgruppen Beiträge zu einer Sozialgeschichte des Alten Orients. Xenia 32. Konstanz: Univ.-Verl. Konstanz, pp. 127–157.Google Scholar
Lambert, W. G. 2007. Babylonian Oracle Questions. Mesopotamian Civilizations 13. Winona Lake IN: Eisenbrauns.Google Scholar
Lambert, W. G. 2013. Babylonian Creation Myths. Mesopotamian Civilizations 16. Winona Lake IN: Eisenbrauns.Google Scholar
Landsberger, B. 1949. “Jahreszeiten im Sumerisch-Akkadischen”. Journal of Near Eastern Studies 8: 248272, 273–297.CrossRefGoogle Scholar
Landsberger, B. Jacobsen, T. 1955. “An Old Babylonian Charm Against Merḫu ”. Journal of Near Eastern Studies 14: 121.CrossRefGoogle Scholar
Leemans, W. F. 1991. “Review of Kraus Königliche Verfügungen in altbabylonischer Zeit”. Journal of the Economic and Social History of the Orient 34: 116122.CrossRefGoogle Scholar
Leichty, E. 1986. Catalogue of the Babylonian Tablets in the British Museum. Volume VI: Tablets from Sippar 1. London: British Museum Publications.Google Scholar
Leichty, E., Finkelstein, J. J., Walker, C. B. F. 1988. Catalogue of the Babylonian Tablets in the British Museum. Volume VIII: Tablets from Sippar 3. London: British Museum Publications.Google Scholar
Livingstone, A. 2010. “Review of Lambert, Babylonian Oracle Questions ”. Journal of Semitic Studies 55: 591593.CrossRefGoogle Scholar
Luukko, M. 2004. Grammatical Variation in Neo-Assyrian. State Archives of Assyria Studies 16. Helsinki: The Neo-Assyrian Text Corpus Project.Google Scholar
Maul, S. M. and Manasterska, S. 2023. Schreiberübungen aus neuassyrischer Zeit. Mit Beiträgen von Anmar A. Fadhil und einer internationalen Forschergruppe. Keilschrifttexte aus Assur literarischen Inhalts 15. Wiesbaden: Harrassowitz.Google Scholar
Mayer, W. R. 1989. “Die Verwendung der Negation im Akkadischen zur Bildung von Indefinit- bzw. Totalitätsausdrücken”. Orientalia 58: 145170.Google Scholar
Mayer, W. R. 1992. “Ein Hymnus auf Ninurta als Helfer in der Not”. Orientalia 61: 1757.Google Scholar
Mayer, W. R. 1993. “Das Ritual BMS 12 mit dem Gebet Marduk 5”. Orientalia 62: 313337.Google Scholar
Mayer, W. R. 1994. “Akkadische Lexikographie: CAD Š1”. Orientalia 63: 111120.Google Scholar
Mayer, W. R. 2009. “Akkadische Lexikographie: CAD T und ṭ”. Orientalia 78: 423439.Google Scholar
Mayer, W. R. 2017. “Zum akkadischen Wörterbuch: ṣ-Z”. Orientalia 86: 202252.Google Scholar
McEwan, G. J. P. 1982. “Agade after the Gutian Destruction: The Afterlife of a Mesopotamian City” in: H. Hirsch/H. Hunger, ed. Vorträge gehalten auf der 28. Rencontre Assyriologique Internationale in Wien 6.-10. Juli 1981. Archiv für Orientforschung Beiheft 19. pp. 8–15.Google Scholar
Mitto, T. 2022a. “Advice to a Prince. With contributions by Zs. J. Földi, A. C. Heinrich, A. Hätinen and E. Jiménez. Translated by Benjamin R. Foster”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/2/5).Google Scholar
Mitto, T. 2022b. “Hymn to Ninurta as Savior. With Contributions by A. Hätinen, A.C. Heinrich, E. Jiménez, and G. Rozzi. Translated by T. Mitto”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/3/10).Google Scholar
Mitto, T. 2022c. “The Nippur Scholar Rīmūt-Gula and His Oeuvre”. KASKAL. Rivista di storia, ambienti e culture del Vicino Oriente Antico 19: 106109.Google Scholar
Montgomery, J. A. 1908. “Review of Radau Babylonian Expedition 17/1”. Old Penn Weekley Review 17 October 1908: 4.Google Scholar
Moran, W. L. 1959. “A New Fragment of DIN.TIR.KI = Bābilu and Enūma Eliš VI 61-66”. Analecta Biblica 12: 257265.Google Scholar
Nashef, K. 1984. “The Deities of Dilmun”. Akkadica 38: 133.Google Scholar
Nougayrol, J. 1947. “Textes et documents figurés”. Revue d’assyriologie et d’archéologie orientale 41: 2353.Google Scholar
Parpola, S. 1983. Letters from Assyrian Scholars to the Kings Esarhaddon and Assurbanipal. Part II: Commentary and Appendices. Alter Orient und Altes Testament 5/2. Kevelaer/Neukirchen-Vluyn: Butzon & Bercker.Google Scholar
Paulus, S. 2014. Die babylonischen Kudurru-Inschriften von der kassitischen bis zur frühneubabylonischen Zeit. Untersucht unter besonderer Berücksichtigung gesellschafts- und rechtshistorischer Fragestellungen. Alter Orient und Altes Testament 51. Münster: Ugarit-Verlag.CrossRefGoogle Scholar
Peterson, J. 2020. “Marduk’s Address to the Demons. With Contributions by Zs.J. Földi, A.C. Heinrich, E. Jiménez, T.D.N. Mitto. Translated by W.G. Lambert”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/3/3).Google Scholar
Pohl, A. 2022. Die akkadischen Hymnen der altbabylonischen Zeit. Grammatik, Stilistik, Editionen . Leipziger Altorientalistische Studien 13. Wiesbaden: Harrassowitz.Google Scholar
Pomponio, F. 1978. Nabû. Il culto e la figura di un dio del Pantheon babilonese ed assiro. Studi Semitici 51. Roma: Istituto di Studi del Vicino Oriente.Google Scholar
Rozzi, G. 2021. “Great Prayer to Šamaš. With Contributions by A.C. Heinrich, E. Jiménez, and T. Mitto”. electronic Babylonian Library (DOI: https://www.doi.org/10.5282/ebl/l/3/4).Google Scholar
Sassmannshausen, L. 2001. Beiträge zur Verwaltung und Gesellschaft Babyloniens in der Kassitenzeit. Baghdader Forschungen 21. Mainz: Von Zabern.Google Scholar
Schwemer, D. 2007. Abwehrzauber und Behexung. Studien zum Schadenzauberglauben im alten Mesopotamien. Wiesbaden: Harrassowitz.Google Scholar
Scurlock, J. and al Rawi, F. N. H. 2006. “A Weakness for Hellenism” in: A. K. Guinan [e. a.], ed. If a Man Builds a Joyful House: Assyriological Studies in Honor of Erle Verdun Leichty. Cuneiform Monographs 31. Leiden: Brill, pp. 357–382.Google Scholar
Sommerfeld, W. 2014. “Die Lage von Akkade und die Dokumentation des 3. Jahrtausends” in: N. Ziegler/E. Cancik-Kirschbaum, ed. Entre les fleuves – II. D’Aššur à Mari et au-delà. Berliner Beiträge zum Vorderen Orient 24. Gladbeck: PeWe, pp. 151–228.Google Scholar
Stamm, J. J. 1939. Die Akkadische Namengebung. Mitteilungen der Vorderasiatisch-Aegyptischen Gesellschaft 44. Leipzig: Hinrich.Google Scholar
Still, B. 2019. The Social World of the Babylonian Priest. Culture and History of the Ancient Near East 103. Leiden/Boston: Brill.CrossRefGoogle Scholar
Stol, M. 2000a. Birth in Babylonia and the Bible. Its Mediterranean Setting. Cuneiform Monographs 14. Groningen: Styx.CrossRefGoogle Scholar
Stol, M. 2000b. “Titel altbabylonischen Klosterfrauen” in: J. Marzahn/H. Neumann, ed. Assyriologica et Semitica. Festschrift für Joachim Oelsner anläßlich seines 65. Geburstages am 18. Februar 1997. Alter Orient und Altes Testament 252. Münster: Ugarit-Verlag, pp. 457–466.Google Scholar
Stol, M. 2008. “Review of CAD P”. Orientalistische Literaturzeitung 103: 348355.Google Scholar
Stol, M. 2010. “«To be ill» in akkadian the verb salāʾu and the substantive siliʾtu” in: A. Attia [e. a.], ed. Advances in Mesopotamian Medicine: From Hammurabi to Hippocrates. Cuneiform Monographs 37. Leiden: Brill, pp. 29–46.Google Scholar
Streck, M. P. 2003. Die akkadischen Verbalstämme mit ta-Infix. Alter Orient und Altes Testament 303. Münster: Ugarit-Verlag.Google Scholar
Vallat, F. 2001. “Les puhlalê d’Assurbanipal”. Nouvelles Assyriologiques Brèves et Utilitaires 2001/65.Google Scholar
Veenhof, K. R. 1973. “An Old Babylonian Deed of Purchase of Land in the de Liagre Böhl Collection” in: M. A. Beek [e. a.], ed. Symbolae Biblicae et Mesopotamicae Francisco Mario Theodoro de Liagre Bohl Dedicatae. Leiden: Brill, pp. 359–379.Google Scholar
von Soden, W. 1957. “Die Hebamme in Babylonien und Assyrien”. Archiv für Orientforschung 18: 119121.Google Scholar
Walker, C. B. F. and Kramer, S. N. 1982. “Cuneiform Tablets in the Collection of Lord Binning”. Iraq 44: 71–86.CrossRefGoogle Scholar
Westenholz, A. 2007. “The Graeco-Babyloniaca Once Again”. Zeitschrift für Assyriologie und Vorderasiatische Archäologie 97: 262313 (DOI: https://www.doi.org/10.1515/ZA.2007.014).CrossRefGoogle Scholar
Wiggermann, F. A. M. 1997. “Transtigridian Snake Gods” in: I. L. Finkel/M. J. Geller, ed. Sumerian gods and their representations. Cuneiform Monographs 7. Groningen: Styx, pp. 33–55.Google Scholar
Wiggermann, F. A. M. 1998/2000. “Nin-azu”. Reallexikon der Assyriologie und Vorderasiatischen Archäologie 9: 329335.Google Scholar
Woodington, N. R. 1982. A Grammar of the Neo-Babylonian Letters of the Kuyunjik Collection. PhD thesis: Yale University.Google Scholar
Zadok, R. 1987. “Peoples From the Iranian Plateau in Babylonia During the Second Millenium B.C.”. Iran 25: 126.CrossRefGoogle Scholar
Zgoll, A. 2003. Die Kunst des Betens. Form und Funktion, Theologie und Psychagogik in babylonisch-assyrischen Handerhebungsgebeten an Ištar. Alter Orient und Altes Testament 308. Münster: Ugarit-Verlag.Google Scholar
Zomer, E. 2019. Middle Babylonian Literary Texts from the Frau Professor Hilprecht Collection, Jena. Texte und Materialien der Frau Professor Hilprecht Collection 12. Wiesbaden: Harrassowitz.CrossRefGoogle Scholar
Figure 0

Fig. 1. SipNB1 obverse. Copy by Anmar A. Fadhil

Figure 1

Fig. 2. SipNB1 reverse. Copy by Anmar A. Fadhil

Figure 2

Fig. 3. SipNB1 obverse. After conservation by C. Gütschow in November 2021

Figure 3

Fig. 4. SipNB1 reverse. After conservation by C. Gütschow in November 2021

Figure 4

Fig. 5. SipNB1 obverse (IM.132667 only). Shortly after excavation. No old photograph of the reverse seems to have survived

Figure 5

Fig. 6. SipNB1 (IM.132512 only). Shortly after excavation

Figure 6

Fig. 7. SipNB2, SipNB3, and SipNB4. Copies by E. Jiménez

Figure 7

Fig. 8. BabNB1, BabNB2, and BabNB3. Copies by E. Jiménez

Figure 8

Fig. 9. BabLB1. Copy by E. Jiménez

Figure 9

Fig. 10. BabLB2, BabNBSch2, and BabNBSch3. Copies by E. Jiménez

Figure 10

Fig. 11. BabNBSch6, BabLB3, and SipNBSch1. Copies by E. Jiménez

Figure 11

Fig. 12. SipNBSch2 and SipNBSch3. Copies by E. Jiménez