¹ The famously problematic incipit of the OB text has been rendered in various ways, without a consensus emerging. Most scholars opt for either “When the gods were man” or “When the gods like man”, though other translations have also been tried (for histories of the argument see Pettinato, G., Or Ant 9 (1970), p. 76 Google Scholar; Jacobsen, T., Finkelstein Mem. Vol., p. 113 Google Scholar; more recently note Oberhuber, K., Zikir Šumim, p. 281 Google Scholar: “als (die) Götter an Menschen Statt”). The line caused problems for ancient scholars, too: see the note below on SB I 1. We take awīlum as a metaphor, following Moran, W. L., Biblica 52 (1971), p. 592Google Scholar. For examples of comparison expressed by the construction noun + complement as predicate, see now Groneberg, B., AfO 26 (1978–1979), p. 20 Google Scholar, who translates the incipit as “als die Götter wie Menschen waren”. Seux, M.-J., RA 75 (1981), pp. 190–1Google Scholar, has pointed out a compelling parallel in OB Atram-ḫasīs 193 // 95, as understood by Borger, R., HKL 2, p. 158 Google Scholar: be-lí (95: ^d en-líl) bi-nu bu-nu-ka, “My lord/Enlil, your features are a tamarisk” (i.e., as pale as its wood when cut: cf. the simile kīma nikis bīni in similar contexts). There, as in the incipit, a plural subject is complemented by a singular noun as predicate, without -ma, to yield a vivid figurative expression.

² Lambert, W. G. and Millard, A. R., Babylonian Literary Texts (CT 46; London, 1965), Nos. 1–15Google Scholar; eid., Atra-ḫasīs. The Babylonian Story of the Flood (Oxford, 1969)Google Scholar. Tablet I has been re-edited by von Soden, W., “Die erste Tafel des altbabylonischen Atramḫasīs-Mythus. ‘Haupttext’ und Parallelversionen”, ZA 68 (1978), pp. 50–94 Google Scholar. The most recent treatments of the whole myth are the translations by Bottéro, , “La grande Genese babylonienne”, in Bottéro, J. and Kramer, S. N., Lorsque les dieux faisaient l'homme. Mythologie mésopotamienne (Paris, 1989), pp. 526–64Google Scholar; by Dalley, S., Myths from Mesopotamia (Oxford, 1989), pp. 1–38 Google Scholar; by Foster, B. R., Before the Muses (Bethesda, Md, 1993), pp. 158–201 Google Scholar (with full bibliography), repeated with minor alterations in idem, From Distant Days (Bethesda, Md, 1995), pp. 52–77; and by von Soden, W., “Der altbabylonische Atramchasis-Mythos”, in Hecker, K. et al. (eds.), Mythen und Epen II (TUAT III/4; Gütersloh, 1994), pp. 612–45Google Scholar. The whole text is the subject of an unpublished edition by Jesús García Recio, “Inūma ilū awīlum”, whose kindness in sending us a draft copy we acknowledge.

³ Lambert, W. G., “New evidence for the first line of Atraḫasīs ”, Or NS 38 (1969), pp. 533–8Google Scholar; “New fragments of Babylonian epics. Atra-ḫasīs”, AfO 27 (1980), pp. 71–6Google Scholar; “Three new pieces of Atra-ḫasīs”, in Mélanges Garelli, pp. 411–14. Note also the new copy of VAT 17099 (BE 36669, Lambert and Millard's MS y) by van Dijk, J., published as VAS 24 93 Google Scholar (cf. ibid., p. 13).

⁴ Groneberg, B., “Atramhasis, Tafel II iv֊v”, in Mélanges Garelli, pp. 397–410 Google Scholar, with a copy of HE 529 by J.-M. Durand. A further fragment, as yet unplaced, seems to be Bo 809/z (KBo 36 26; cf. Siegelová, J., ArOr 38 (1970), p. 138 Google Scholar).

⁵ Iraq 49 (1987), p. 249 Google Scholar.

⁶ Iraq 52 (1990), pp. 1–13 CrossRefGoogle Scholar; ibid., pp. 149–57; Iraq 56 (1994), pp. 135–48Google Scholar; Iraq 57 (1995), pp. 199–223 Google Scholar; ibid., pp. 225–8. The Sippar tablets are published by kind permission of the University of Baghdad and the Iraqi Directorate General of Antiquities. Work on the tablets was once again supported by the British School of Archaeology in Iraq, to whom all thanks are due. A draft of this article was read by W. G. Lambert with customary acumen, and his patient corrections are gratefully acknowledged. Where we are still in error is our fault alone.

⁷ Or Nūr-Aya; the reading of the scribe's name is disputed: see Soden, von, ZA 68, p. 501 Google Scholar; Lambert, , AfO 27, p. 711Google Scholar.

⁸ As suspected in Atra-ḫasīs, p. 35: “L obverse has a ruling and some damaged signs just above it that seem to be colophonic. If this is a correct understanding of the traces, L is derived from a series of which one (presumably the first) tablet ended with 1. 110.” As we can now see, the traces above the line are of ab-bāb (see fn. 12) and so belong to the line that corresponds to OB 1110, but after the ruling, before da-num, the word that opens the line corresponding to OB I 111, there is a trace of extra material that is likely to part of a brief colophon (see further below, the note on II 1).

⁹ Cf. von Soden, W. and Röllig, W., AnOr 424, p. xxxv Google Scholar.

¹⁰ See von Soden, W., ZA 41 (1933), pp. 134–6, 148–51Google Scholar.

¹¹ For “hymno-epic” features in the OB text see Lambert, and Millard, , Atra-ḫasīs, p. 30 Google Scholar.

¹² In the manuscripts from Nineveh and Babylon the only diagnostic phrases are: a) showing detached prepositions: MS L rev. 2' (= SB II 57): ana š[amāmī]; MS P obv. 11 (// OB I 249): a-na bīt; obv. 20: a-na šumēli; K 17752 = AfO 27, p. 75, 2' (// OB I 290): ina b[īt]; MS Q rev. 5 (// OB II ii 5'): ana ḫalāqi; rev. 16' (= SB V 8): in māti; MS x rev. i 3 (= SB V 50): ina ḫubūrišin; i 18 (= SB V 65): a-na [miṭrati]; i 24.25 (= SB V 71՜ 72): i-na pūt; i 34 (= SB V 80): i-na [kār]; ii 6: a-na nišī; ii 13: ana nišī; ii 15: ana mār; ii 31: a-na qurādi; ii 45: [a]-na puḫur; ii 46: a-na māmītu; ii 47: i-na pānī; MS W, 16': ina qaqqari; and b) showing attached prepositions: MS L obv. 5' (= SB I 99'): ab-bā[b]; K 17853 = AfO 27, p. 74, 5'(= SB II 89): ew-warḫi; MS x rev. i 40 (= SB V 85): an-nâši. This makes 22 instances in the manuscripts not from Sippar of ina and ana before nouns beginning with a consonant, of which only three are attached. However, two factors should caution us against leaping to the conclusion that the attached preposition is a peculiarity of the copying tradition represented by the Sippar library manuscripts. First, 12 instances of detached prepositions in the manuscripts not from Sippar are from a single manuscript, MS x; second, 13 are from SB V, where even the Sippar manuscripts show a preponderance of detached prepositions. In SB I–IV the score in the manuscripts not from Sippar is at present 5 : 2, which from so small a sample is not conclusive.

¹³ See SB 12, 5, 53', II 97, 103–4, 109, V 6, 16//32//38, 61.

¹⁴ On this god in general see Lambert, W. G. in Alster, B. (ed.), Death in Mesopotamia (CRRA 26; Copenhagen, 1980), pp. 63–4Google Scholar.

¹⁵ Edited with duplicates by Pettinato, G., Das altorientalische Menschenbild und die sumerischen und akkadischen Schöpfungsmythen (Heidelberg, 1971), p. 25 Google Scholar.

¹⁶ See Lambert, loc. cit.; the two cultic texts are now republished by Livingstone, A., Mystical and Mythological Explanatory Works … (Oxford, 1986), pp. 194–9Google Scholar.

¹⁷ Wê: Pettinato, G., Or Ant 9 (1970), p. 80 Google Scholar; Labat, R., Les religions du Proche-Orient asiatique (Paris, 1970), p. 29 Google Scholar; Jacobsen, T., The Treasures of Darkness (New Haven, 1976), p. 118 Google Scholar; Bottéro, , Lorsque les dieux, p. 537 Google Scholar; Moran, W. L., Studies Reiner, p. 24915Google Scholar; Geštu'e: von Soden, passim; Dalley, , Myths from Mesopotamia, p. 15 Google Scholar: Wê-ila: Lambert, and Millard, , Atraḫasīs, p. 153 Google Scholar; We-ilu: Foster, , Before the Muses I, p. 166 Google Scholar; etc.

¹⁸ Collated by Lambert in Atra-ḫasīs, PI. 11, No. 1, 47.

¹⁹ See Kilmer, A. D., Or NS 41 (1972), p. 164 Google Scholar; Abuser, T., in van der Toorn, K. et al. (eds.), Dictionary of Deities and Demons in the Bible (Leiden, 1995), 588 Google Scholar. The proposal of Moran, W. L., Biblica 52 (1970), pp. 53–5Google Scholar, that man's eṭemmu is the ghost of the dead god, offers a different slant on the passage but does not disallow the word-play. The dissenting view of von Soden, W., Symbolae Böhl, pp. 350 ff.Google Scholar, that a word (w)edimmu < Sum. id im is meant, was answered by Bottéro, J., whose article on “La création de l'homme et sa nature dans le poème d'Atraḫasîs”, Studies Diakonoff, pp. 24–32 Google Scholar, explores the significance of the word-plays to the full. However, the argument eṭemmu vs. (w)edimmu continues: see Soden, von, “Die Igigu-Götter in altbabylonischer Zeit und Edimmu im Atramḫasīs-Mythos”, in Cagni, L. and Müller, H.-P. (eds.), Aus Sprache, Geschichte und Religion Babyloniens (Naples, 1989), pp. 339–49Google Scholar; idem, “Der Urmensch im Atramḫasīs-Mythos”, in L. De Meyer and H. Gasche (eds.), Mésopotamie et Élam (CRRA 36; Ghent, 1991), pp. 47–51; cf. Groneberg, B., AoF 17 (1990), pp. 252–3Google Scholar.

²⁰ So now Foster, who in his revised translation of the myth notes the variant Alia and has changed the name from We-ilu to Aw-ilu (From Distant Days, p. 59). For the sign WA with the value aw in the early second millennium, see Gelb, I. J., “WA = aw, iw, uw in cuneiform writing”, JNES 20 (1961), pp. 194–6Google Scholar; idem, Or NS 39 (1970), p. 539.

²¹ This would be a position close to that adopted by Matouš and Oberhuber, for whom the signs in question were not a proper noun but a compound of two common nouns, i.e., “god-man”: Matouš, L., ArOr 35 (1967), p. 740Google Scholar: ila–(a)wēla; Oberhuber, K., Zikir Šumim, p. 280 Google Scholar: Ham aw-e-i-la, with the comment “daß hier eine Art ‘Krypto’-graphie zum Zwecke einer ‘Aitio’-logie beabsichtigt ist”.

²² The method is best exemplified by the well-known commentaries on the names of Marduk ( Bottéro, J., Finkelstein Mem. Vol., pp. 5–28 Google Scholar; Lambert, W. G., “Etymology, Ancient Near Eastern”, in Coggins, R. J. and Houlden, J. L. (eds.), A Dictionary of Biblical Interpretation (London and Philadelphia, 1990), pp. 214–16Google Scholar), and by abnormal spellings of Sumerian temple names and city epithets in the expository lists collected in George, A. R., Babylonian Topographical Texts (OLA 40; Leuven, 1992)Google Scholar. For speculative spellings in these lists see in particular No. 1: Tintir I 4–7 Google Scholar; No. 5: E-sagil Commentary, passim; No. 18: Nippur Compendium, §6; No. 19: Nippur Temple List, 1'–10'.

²³ On the development amar utu-(ak) > Marūtuk etc., see Sommerfeld, W., Der Aufstieg Marduks (AOAT 213; Kevelaer and Neukirchen-Vluyn, 1982), pp. 7–12 Google Scholar; Lambert, W. G., BSOAS 47 (1984), pp. 7–8 CrossRefGoogle Scholar.

²⁴ The OB MS A rarely spreads a line of poetry over two lines, especially in its first few columns. In those of the first 75 lines that are fully preserved, the only line of poetry so arranged is ll. 66–7. The damaged section at the bottom of MS A col. i, ll. 49 ff., held either six lines or eight, depending on whether the column was 54 or 56 lines long (the total on the lower edge is broken). It may thus have contained as many as four couplets (so Lambert and Millard's edition, ll. 49–56), but no fewer than two. If four couplets, the column contained 56 lines of poetry; if two, 52. Accordingly, we can be sure that OB I 1–75 represents between 70 and 74 lines of poetry.

²⁵ Or, “the six hundred gods fashioned the soil-basket.” The line is a corruption of OB “they bore the work and carried the soil-basket.”

²⁶ This line was left out by mistake, and was later added on the left edge of the tablet (Fig. 6).

²⁷ This line occurs in MS J only.

²⁸ The three lines of tablet represented by SB 149'–54' are restored as three couplets after OB I 57–62.

²⁹ Corrupt. OB I 61 has: “Now then, shout for war!”

³⁰ Corrupt. OB 1214 has: “So that in future days we may hear the drum,” i.e., the heart.

³¹ Text in disorder, see the note on this line.

³² Text in disorder, probably corrupt; see the note.

³³ Corrupt. The OB text probably had: “In future [days] they [heard the drum].”

³⁴ See further the note on this line.

³⁵ Here we agree with Foster, , Before the Muses I, p. 1743Google Scholar, in viewing the eventual release of the sea's produce (miširtu) as a salvation, not as “a calamity”, pace CAD M/2, p. 124.

³⁶ SB Etana II 135; IV R ² 22 No. 2 = Maul, S. M., Herzberuhigungsklagen, p. 332 Google Scholar, 11'; Ugaritica V 162, 6'Google Scholar; Ludlul II 7 Google Scholar. See further Lambert, , BWL, pp. 288–9Google Scholar; Oppenheim, , Dreams, pp. 221–2Google Scholar; Römer, W. H. Ph., JAOS 86 (1966), p. 145 Google Scholar, on CT 15 5–6 Google Scholar, vi 2'; and CAD Š/1, pp. 111–12.

³⁷ Ludlul II 7 Google Scholar: maš-šak-ku = sur-qí-nu šá ^lúšā'ili(ensi).

³⁸ CAD M/2, p. 279; cf. Š/1, p. 111: “this incense, called ma//uššakku, served as offering or in libanomancy.” AHw, p. 684, repeats the explanation of the ancient commentator, but also assumes that the incense is a fumigant: “Räucheropfer des Traumdeuters”.

³⁹ PSD A/1, p. 115, reads so, but does not translate: “(meaning unknown)”. Thorkild Jacobsen translates the complex as “on fresh waters” (The Harps that Once …, p. 412), while in Angim 172 Jerrold S. Cooper reads a.mir and associates it with a.ma.ru, “flood”, and mir, “storm” ( AnOr 52, p. 135 Google Scholar), though he did not see that meaning as appropriate for Gudea Cyl. A xx 6. Instead one is tempted to read in both passages a.aga (or ^aaga), as a variant of aga and a.gi₆.a, “river-wave, current”.

⁴⁰ Following Falkenstein, A. in La divination en Mésopotamie ancienne (CRRA 14), p. 55 Google Scholar.

⁴¹ AO 3112: Nougayrol, J., “Aleuromancie babylonienne”, Or NS 32 (1963), pp. 381–6Google Scholar.

⁴² So CAD Q, p. 4, s.v. qablītu 1.d.

⁴³ Rabelais III/25: “aleuromantie, meslant du froment avecques de la farine”. Aleuromancy was practised in classical antiquity, of course, but Greek and Latin sources are reticent about the details: see Bouché-Leclercq, A., Histoire de la divination dans l'antiquité 1 (Paris, 1879), p. 182 Google Scholar; Halliday, W. R., Greek Divination (London, 1913), p. 185 Google Scholar.

⁴⁴ This incident is a further demonstration that according to ancient belief, rivers eventually unloaded their waters into the source from which they took them, the cosmic domain of Ea. In terms of practical geography this belief was of course grounded in the very evident fact that the waters of the Tigris and Euphrates flowed into the marshes and not directly into the sea.

⁴⁵ This is clearer from the fragmentary OB source, which states that Atram-ḫasīs was at this time paying special attention to dreams (OB II iii 7–10), than it is from the new text.

⁴⁶ Corrupt; OB: inside your house.

⁴⁷ Literally, “the middle earth”, here meaning the cosmic level between the heavens and the domain of Ea; see Lambert, and Millard, , Atra-ḫasīs, p. 166 Google Scholar.

⁴⁸ Apparently corrupt; OB: in the morning.

⁴⁹ I.e., he arose at midnight.

⁵⁰ MS x in two lines: “The man who by day(?) […]/ this [……]”

⁵¹ Correct the printer's error on p. 220, 1. 599: for ˹rì˺ read ˹ì˺.