¹ The basis for subsequent discussion was an article by Lees, G. M. and Falcon, N. L. entitled “The Geographical History of the Mesopotamian Plains” in the Geographical Journal Vol. CXVIII Part I (03 1952)Google Scholar. Note also that in the September number of the same Journal certain objections were formulated by Professor Sidney Smith and Mr. M. G. Ionides to whom Lees and Falcon in turn replied.

² Parrot, André, Déluge et Arche de Noé, 1953 Google Scholar; Heidel, Alexander, The Gilgamesh Spie and Old Testament Parallels, 1945 Google Scholar.

³ That the Mesopotamian Flood Story and the Gilgamesh Epic reached Syria and Palestine is proved by the discovery of a fragment of a cuneiform tablet at Megiddo and by recent finds at Ras Shamra. My colleague C. F. A. Schaefier has kindly shown me a letter dated 12th May 1964 from Monsieur Jean Nougayrol in which it appears that a fragment of the Flood Story has been found at Ugarit, and that one line repeats exactly that of the neo-Assyrian version in tablet XI ‘Hear Oh Wall! Hear me reed-hut’ cf. Nougayrol, J. in C.R.A.I.B.L. 1960, p. 170 Google Scholar. A second fragment may possibly name Gilgamesh and perhaps be concerned with the exploits of this hero in youth. These two texts are due for publication in Ugarilica V, No. 167 (=R.S.22.421) and No. 268 (=R.S.22.219 + 22.398) respectively, and together with a fragment of the Gilgamesh epic found at Sultan Tepe are of interest as showing the wide distribution of the story which has been found on tablets in Mesopotamia, Anatolia, Syria and Palestine.

⁴ This name, which apparently means “the exceedingly wise”, as Heidel has noted, is another designation for Utnapishtim.

⁵ A convenient summary of early Flood legends in various parts of the world will be found in Peake, H., The Flood (1930)Google Scholar, Chapter II, based mainly on the work of Sir James Frazer, Folklore in the Old Testament. It is interesting that tribal legends in Africa when obviously related to the Old Testament story can be ascribed to the teaching of Christian missionaries and are relatively modern. There is no reason to doubt that Deucalion's Flood may have been founded on what the Meteorologists call a rainfall maximum which would have had a disastrous effect on the Copais lake land, cf. Myres, J. L. Who were the Greeks, p. 363 Google Scholar, but also J.N.E.S. XXIII (1964), p. 200 Google Scholar, Michael C. Astour, “Greek Names in the Semitic World and Semitic Names in the Greek World”—Thêbai or Thêba is alleged to correspond exactly with Hebrew têbā, ‘ark’. This loan word is attested by Greek lexicographers. A Boeotian myth states that Ogygos saved himself from the flood in a chest and landed on the spot where Thebes was built by Cadmos an immigrant from Phoenicia whence the Greek Flood myths may have originated. There is however evidence that in the thirteenth century B.C. sea-going ships of up to 500 tons were built for work in the Mediterranean; See Nougayrol, Jean, “Nouveaux Textes Accadiens de Ras Shamra” in the Compte Rendu for 1960, Académie des Inscriptions et Belles-Lettres (published 10 1961) pp. 160–171 Google Scholar.

⁶ A version from Nippur, said to be the oldest of the series and written down about the time of Hammurabi. See Heidel, op. cit., p. 102.

⁷ See Heidel, op. cit., p. 247.

⁸ Heidel, op. cit., p. 240, insists that Akkadian abubu should be rendered “rainstorm” or “rain-flood”, and that there is no evidence to suggest that it was ever employed to denote an inundation caused by the river unaccompanied by heavy rains.

⁹ Salonen, A, Die Wasserfahrzeuge in Babylonien, p. 160 Google Scholar and Parrot, A., Déluge et Arche de Noé, p. 46 Google Scholar. See also n. 5 above.

¹⁰ Unfortunately it has to be admitted that the so-called “boat-shrine” built by the Neo-Babylonians under the lee of the Ziggurrat has only a weak claim to be identified by this name, see A.J., V (10 1925), No. 4, p. 359 Google Scholar, for the plan, and p. 362. The shrine about 10 metres in length might have been a support for the sacred boat of the moon—so L. Legrain surmised—but proof was lacking and there is no warrant for suggesting any connection with Noah's Ark, but the proximity to the Ziggurrat is interesting. See also U.E., V, “The Ziggurat and its Surroundings” (1939), p. 67 f.Google Scholar, where Sir Leonard Woolley did not press the identification.

¹¹ See Heidel, loc. cit., p. 250, where references are given. In this connection it is interesting that no evidence has been observed between Tigris and Euphrates, north of Babylonia, of ancient Flood strata; but in 1931 I noted at Nineveh deep down in the mound, a series of wet levels, perhaps pluvial, which may have indicated a climatic change. These strata occurred after the end of the Halaf period, between it and Ninevite 3 which perhaps corresponded to the later phases of ‘Ubaid and therefore had nothing to do with the Babylonian Flood. See AAA, XX (1933), p. 134 Google Scholar and plate LXXIII.

¹² Jacobsen, T., “Formative Tendencies in Sumerian Religion“, in The Bible and the Ancient Near East, p. 276 Google Scholar, (ed. by G. E. Wright, Essays in Honour of W. F. Albright).

¹³ Poebel, A., Babylonian Legal and Business Documents from the time of the First Dynasty of Babylon (chiefly from Nippur), No. 38a, p. 66 Google Scholar; and for Ibbi-Sin Flood at Ur, see Gadd, C. J. in C.A.H., I, (1962), Ch. XIII, p. 19 Google Scholar.

¹⁴ Millard, A. R. in Iraq XXVI (1964), Pt. 1, p. 23 Google Scholar, article entitled “Another Babylonian Chronicle Text”.

¹⁵ Heidel, op, cit., p. 106 f.

¹⁶ See Kramer, S. N., The Sumerians, (1963) p. 46 Google Scholar who discusses the Tummal texts from Nippur which would make Mesannipada earlier than Gilgamesh. For the original publication of this text see also the same author and Bernhardt, Inez, Sumerische Literarische Texte Aus Nippur p. 14 fGoogle Scholar. Akademie-Verlag (1961). See however note 18 below for another version which reverses the order of succession and appears to be historically preferable.

¹⁷ U.E. I, p. 61 Google Scholar and U.E. II, pp. 312–313 Google Scholar.

¹⁸ Gadd, C. J., C.A.H. Vol. I Google Scholar, Ch. XIII, end table. Kramer, op. cit., p. 50, and Roux, G., Ancient Iraq (1964)Google Scholar, end table II.

¹⁹ The archaeological evidence that appears to confirm a relatively high date for Gilgamesh is that, from the Tummal texts, and in the Sumerian King Lists, it may be deduced that he was a slightly younger contemporary of En-me-baragesi whose name appears to have occurred in an Early Dynastic II level on one of the Diyala sites, see Rowton, M. B. in C.A.H. I Ch. VI, pp. 55 and 66 Google Scholar: it may be convenient to take the latter monarch as marking the end of E.D. II and Gilgamesh the beginning of E.D.III. S. N. Kramer in op. cit. p. 46 pointed out that before the decipherment of the Tummal text from Nippur practically all scholars believed that Mesannipada lived considerably later than Gilgamesh. But it is probable that this scribe erred, for in a text from Ur, (see Sollberger, E. in J. C. S. XVI (1962), p. 40)Google Scholar, also recording repairs to the sanctuary in Nippur, Mesannipada comes later than Gilgamesh, as in the Sumerian King List, and Sollberger has adduced sound reasons for preferring this sequence. However that may be, all the weight of the evidence goes to show that the Kings in question were not very far removed in time from one another. The difficulty in fixing dates for the earliest Kings in the Sumerian King List is, that since we now know that many of them were in fact contemporary, we have little ground for confidence in calculating the dates of the very earliest kings in the list. Although most authorities do not now differ by much more than about a century for the reign of Gilgamesh, it becomes increasingly difficult to guess the dates of his predecessors.

²⁰ Edzard, D. O. in Z.A., 53 (1959), pp. 9–26 Google Scholar. He was father of Agga who fought against Gilgamesh, and in the Sumerian King Lists is alleged to have ruled for 900 years. He despoiled Elam. The name also occurs on a fragment of a bowl alleged to have been found in an E.D.II level of the Temple Oval at Khafajah; see also Rowton, M. B. in C.A.H. I (1962), Ch. VI. p. 55 Google Scholar.

²¹ Jacobsen, T., The Sumerian King List, p. 75, note 32, and p. 76 note 34Google Scholar. An Epic fragment K.11624 has him after Ubar-tu-tu(k) as son of that ruler. See also Gadd, , C.A.H., loc. cit., p. 18 Google Scholar; and Kramer, S. N., “Literary Texts from Ur” in Iraq XXV (1963), p. 175 Google Scholar where the filiation is written as follows: Shuruppak, son of Ubar-tu-tu, instructed his son Ziusudra. The most recent evidence on this subject is to be found in J.C.S. XVII No. 2 (1963)Google Scholar by J. J. Finkelstein on The Antediluvian Kings, p. 39 f.

^21a In the Sumerian King List 23 rulers were recorded in succession as composing the First Dynasty of Kish after the Flood—the last two were En-me-baragesi, and Agga the contemporary of Gilgamesh. But as Jacobsen has observed, we cannot regard these 23 rulers as successive, for it is clear from the Etana Epic that this monarch was officially recognised as the First King of this dynasty. The so-called ‘animal kings’ bearing Akkadian names, who preceded Etana, and the Etana Kings form two separate units, which have been joined together. On this reckoning the number of successive Kings of the First Kish dynasty after the Flood need not amount to more than 11 in all, from Etana to Agga (the successor of En-me-baragesi). It would therefore appear that we need not allocate more than 200 years to cover the total of 23 names in the list because less than half of them were successive. For discussion of this problem see Jacobsen, T., The Sumerian King List, p. 152 fGoogle Scholar.

²² Fara tablets, see Gadd, C.A.H., loc. cit, pp. 3–4 Google Scholar, and Driver, G. R., Semitic Writing, Schweich Lectures 1944, pp. 4, 5 Google Scholar. Fara texts appear to belong to the period known as Early Dynastic III, about the time of the early graves in the Royal Cemetery of Ur, see Burrows, E. in U.E. II p. 311 fGoogle Scholar. Some texts of this type were found at Kish, apparently in debris above the earliest flood levels and below the latest, (E.D.II— III but the account of the stratification is not satisfactory—see Watelin, and Langdon, , Kish IV p. 35 Google Scholar. Although the scribes of the period were certainly capable of recording the Flood, and may have done so, it seems unlikely that a full-length account of it was written down at so early a period. My friend S. N. Kramer, whom I have consulted, kindly writes as follows: “On common sense “guessing” it seems rather unlikely that writing was sufficiently developed at the time of Gilgamesh to justify the assumption that the forerunner of the Flood story had existed in his days in written form. Even in ‘Fara’ days the script does not seem ‘ripe’ for a lengthy myth, although as you know there are some proverbs among the Fara texts. My feeling is that it was in oral form that the early prototypes of the later myths and epic tales were composed—in this the nar (‘minstrel’) played the leading role; the nar was an important person in those days (witness the statue of the nar of Ur-Nanshe, found at Mari), and at court banquets and celebrations it was he who composed and recited (to the accompaniment of the harp or lyre) the heroic forms that his audience craved. It seems unlikely to me that the epics and myths were written down earlier than the 25th century B.C., but I could be wrong.”

²³ A.J., IX, No. 4 (10 1929), pp. 327, 329 Google Scholar, and illustrated section opp. p. 330.

²⁴ U.E., IV, p. 15 Google Scholar. Pits Y and Z and plate 83 for table of pits.

²⁵ Frankfort, H., Cylinder Seals, pp. 38–57 Google Scholar.

²⁶ O.I.P., LXIII, Pottery from the Diyala Region.

²⁷ O.I.P., LVIII, Pre-Sargonid Temples in the Diyala Region, pp. 40–78 Google Scholar.

²⁸ U.E., IV, plate 73 for illustration of the section showing the successive levels of occupation above and below the Flood Level in Pit F at Ur; on p. 58 these wet levels are said to occur in stratum B; in A.J. X p. 340 Google Scholar they are reported as between 14·00 and 15·50 m. above sea-level and compared with one of the flood-deposits at Kish. In U.E. IV p. 59 Google Scholar it is said that archaic account tablets, older than Sargon, later than S.I.S.4–5 were found here. We may infer that some of these were probably Fara texts of E.D.III period. See also Burrows, E. U.E. Archaic Texts II p. 23 Google Scholar, but this group of tablets was not satisfactorily recorded with reference to the stratification. Flood Pit F is also shown on Plate XVI of this article.

²⁹ U.E., IV, p. 15 Google Scholar.

³⁰ L'Anthropologie XLI (1931), p. 271 Google Scholar.

³¹ U.E., IV, Appendix VID, “The Flood Deposit”, pp. 165–6.

³² U.E., IV, p. 165 Google Scholar.

³³ Atlas of Mesopotamia, ed. by Rowley, H. H., (Nelson, 1962), p. 12 Google Scholar.

³⁴ First published in Journal of the Institute of Petroleum Technology, Vol. 22, No. 149, p. 180 Google Scholar, and later reproduced in U.E., IV, Appendix VI, p. 160 f.

³⁵ In a letter to me, dated 20 July 1964, W. A. Macfadyen restated the problem as follows:

“Until 1952 it had been tacitly assumed and accepted that the Mesopotamian plain was purely a delta which had been filling up the northern end of the Persian Gulf from time immemorial. Then Lees and Falcon in 1952 and Hudson, Eames and Wilkins in 1957 proved the presence of Recent fossiliferous marine deposits where only freshwater silts and clays had hitherto been supposed. These marine deposits could only be dated as ‘Recent’ geologically, i.e., not older than about 8000 B.C., but may possibly be very appreciably later. Before one can be certain of the real nature of the ‘Flood’ deposits found by the archaeologists they must be re-examined to see whether marine microfossils are present or not.

“That there is no inkling in the ancient records and stories of a flood coming up from the sea is by no means surprising in the circumstances. For the constancy of sea and land levels was no doubt assumed as absolute and unchanging in the old days as it was in England, certainly by some of the first geologists who considered the matter, e.g. J. A. de Luc, F.R.S., about the year 1800. If as I think probable the sinking of the land relative to sea level was there a slow process, of perhaps not greatly more than a foot or two per century, it might well pass unnoticed, but would help materially to emphasize the effect of river flooding in Lower Mesopotamia. The Shatt al Arab is nowadays tidal, I believe, as far as Qurna or thereabouts, and marine microfossils are found to-day at Basra. May not similar conditions have existed at Ur at the date of Noah's Flood? Perhaps not very dissimilar conditions may have been adequate to produce the Hammar Formation.”

³⁶ L' Anthropologie, XLI, p. 271 Google Scholar, “Analyse des Limons de Kish et d'Ur“.

³⁷ L'Anthropologie loc. cit., p. 270.

³⁸ I have to thank Professor Seton Lloyd for a reference to a quotation from Thompson, R. Campbell in Archaeologia, Vol. LXX, p. 106 Google Scholar: “Hilprecht's deep valley does not exist: the sandstone ridge drops abruptly merely 10 or 15 feet on the Abu Shahrain (Eridu) side at six metres from the mound and the intervening space is flat desert containing a broad and shallow depression.” My own recollection is that in travelling from Ur to Eridu there was a gradual rise up to the ridge, and although Eridu may appear to be in a hollow there may not have been a great discrepancy between the latest ‘Ubaid levels at Ur and the latest ‘Ubaid levels at Eridu because the latter rested on a much greater depth of accumulated debris which may have compensated for the difference above sea level of the two sites. But these comments remain purely speculative in the absence of precise information about sea level at Eridu. See also note 39 below.

³⁹ A contour survey of the site of Ur was made by Richardson, F. L. W. and published in A.J., XII (10 1932), No. 4 Google Scholar, pl. LVIII, but no reference to sea level is made therein. In U.E. IV, however, where the Flood Pit K is described, heights above sea level are given and could easily be correlated e.g. with Eridu Temple VI (end of the ‘Ubaid period) which was probably founded a little before the occurrence of the Ur “Flood” deposit. In Sumer IV (09 1948), p. 116 Google Scholar, it is stated that the foundations of Temple VII, the penultimate temple of the ‘Ubaid period at Eridu, were almost 15 metres above the level of the plain. In A.J., X (10 1930), p. 334 Google Scholar, Woolley recorded the top of the clean water-laid sand in the great Flood Pit at about 4·50 m. above sea level; the depth of this deposit varied and is here recorded as not more than 3 metres thick. Beneath it were the Al ‘Ubaid settlements which went down almost to sea level and then occupied a depth of about 1·50 m. of accumulation. The total accumulation of ‘Ubaid debris at Eridu thus appears to have been very much greater, although it is possible that some early ‘Ubaid strata remain to be found at Ur below present water level in soil alleged to be virgin. Heights above sea level of Ur (106 feet) and Abu Shaharain, (Eridu) are given on sheet No. H–38 K of the ¼ of an inch to a mile map of Iraq published by the War Office, G.S.G.S. 3919, 3rd edition, 1947. But as no specific reference points are given it is not possible to correlate the relative levels of the excavated ‘Ubaid settlements at Ur and Eridu. The datum from which all heights on the map are measured is mean sea level at Fao on the Persian Gulf.

⁴⁰ U.E., IV, p. 15 Google Scholar.

⁴¹ There is no evidence so far for Frankfort's ascription of a Flood to Warka, as in his “Archaeology and the Sumerian Problem”, O.I.C., No. 4, Table opposite, p. 72. The caption under Erech, Clay Deposit 1·55 should be deleted, and this has been confirmed for me by Dr. H. Lenzen.

⁴² The position of Eridu with reference to Ur is shown in the sketch map of Jacobsen, T. in his article entitled “The Waters of Ur”, Iraq XXII (1962)Google Scholar, Plate XXVIII, opp. p. 175. At the time of the Flood the city of Ur lay on the left or east bank of the Euphrates (which has subsequently shifted its course some 10 miles eastwards), see U.E. IV p. 15 Google Scholar, while Eridu, about 12 miles S.W. of Ur, lay on a canal which branched off from the Euphrates and was thus in any case not on the main flood stream. We cannot yet be certain whether the Eridu canal already existed in the ‘Ubaid period though Jacobsen (loc. cit. p. 181) believes that it may have been there as early as the third millennium B.c. “The course of the canal crosses the flat depression of Eridu from north-west to south-east and its nearest point to Eridu is about 3 kilometres from the south-west of that site,” (p. 180). See also, for the relative position of Eridu and Ur, Campbell-Thompson's, R. sketch map in Archaeologia, LXX, p. 102 Google Scholar, Fig. 1, which also marks the line of the sandstone ridge in between Ur and Eridu, running more or less parallel with the present course of the Euphrates.

⁴³ These considerations and Woolley's argument mentioned above are, in my opinion, on the basis of the present evidence a sufficient answer to the criticism made by H. W. F. Saggs in the footnote on p. 34 of his excellent book The Greatness that was Babylon. I do not think it is fair to say: “The often asserted claim that Sir Leonard Woolley discovered evidence of the Biblical Flood at Ur goes wildly beyond the facts.” I agree that this was not the “Biblical Flood” which was cosmic mythology remotely based on the prehistoric Mesopotamian Flood. Moreover, as Woolley has explained, we cannot expect that a similar flood layer would necessarily have existed at a corresponding level in every Sumerian city. The presence or absence of flood-traces depends entirely on the topography of each individual city. The continuity of civilisation in Mesopotamia is attested by archaeological discovery which has also demonstrated that no flood was ever of sufficient magnitude to prevent some men and some cities from being salvaged from the wreckage.

⁴⁴ The cera mie overlap between the ‘Ubaid and Uruk periods has been clearly noted by Joan Oates in her article entitled “Ur and Eridu, The Prehistory” in Iraq, XXII (1962), Pt. 1, p. 41 fGoogle Scholar.

⁴⁵ S. Langdon wrote a letter to The Times on 16 March 1929 claiming that the authentic Flood had been found at Kish before its discovery at Ur. See also Woolley's criticism and comment in U.E. IV, p. 15 Google Scholar and note 3.

⁴⁶ Watelin, L.Ch., Excavations at Kish, IV, 1934, p. 53 Google Scholar.

⁴⁷ L'Anthropologie, Vol. XLI, p. 269 Google Scholar, but in Kish IV, p. 40 the average is recorded as 30 cm. thick; Langdon's statement on p. 66 that this stratum lay at 2 to 2·5 m. below plain level appears to be inaccurate. We may accept the figures given by Watelin, the excavator.

⁴⁸ Kish IV, p. 41 Google Scholar.

⁴⁹ Watelin, , Rapport sur les Fouilles de Kish p. 4–5 Google Scholar (Extrait du Journal Asiatique 1929).

⁵⁰ Kish IV (1934), p. 42 Google Scholar, a recantation of the opinion previously expressed in Rapport sur les Fouilles de Kish (1929), p. 14 Google Scholar.

⁵¹ Frankfort, H., Cylinder Seals, pp. 62–7Google Scholar, while casting doubt on the often accepted identifications of Gilgamesh and Enkidu on Early Dynastic seals, pointed to one cylinder of E.D.II period, loc. cit. pl. XI (m) which perhaps portrayed Gilgamesh, holding the herb of life, sitting opposite Utnapishtim in a ‘gondola’ ferried by the boatman Urshanabi.

⁵² Kisb, IV, p. 53 Google Scholar. Flood I in the table.

⁵³ Kish, IV, p. 43 Google Scholar.

⁵⁴ J.C.S. XV (1961), p. 105 fGoogle Scholar. (A. Goetze, “Early Kings of Kish”). I differ from him in assigning these Akkadian Royal names to E.D.III rather than E.D.II; palaeographically the relevant records agree with the Fara tablets which are now usually assigned to E.D.III.

⁵⁵ See the discussion of this subject by Falkenstein, A. in Cahiers D'Histoire Mondiale, Vol. I, No. 4, (Avril 1954), p. 804 fGoogle Scholar. (article entitled La Cité-Temple Sumerienne).

⁵⁶ See the Museum Journal, Museum of the University of Pennsylvania, Vol. XXII, Nos. 3–4 (1931)Google Scholar, article entitled “Excavations at Fara, 1931”. by Erich Schmidt, pp. 193–217, and particularly p. 200 and section on p. 201. For the location of Fara see map opposite p. 193 and topographical description on p. 195. “To the northwest beyond the tell of Abu Hatab lies Ebrah, the nearest town, 23 kilometres away … 100 kilometres further is Diwaniyah.” We may also note that Fara lies about 18 miles north-west of Uruk. See map, Plate XVIII.

⁵⁷ Fara tablets. See note 22, above, Schmidt himself found 85 “valuable Fara tablets of the archaic type” loc. cit. p. 206, in stratum 2 which lay above the Jamdat Nasr levels, and see also loc. cit. p. 216.

⁵⁸ When considering the Mesopotamian Flood it is worth examining the evidence of Indian floods in the Indus Valley where in the second millennium B.c. considerable disruption of urban life occurred, probably from similar causes—see S. R. Rad, Excavations at Rangpur and other Explorations in Gujarat(Ancient India Nos. 18 and 19). Formidable fluviatile deposits have been observed, one about forty feet high, at Budh-Takkar in Sind and another three to four feet thick of river sand over Harappan occupation-layers at Koth near Lothal which suggest an ancient flood of long duration in Kathiawar. In general, there is reason for believing that the Harappan cities suffered very serious floods, notably in about 2000 B.c. and of even greater violence round about 1500 B.C. It is thought that these occurrences caused a shifting of the population on a considerable scale; the first emigrants then returned, but none of the architectural precautions taken at places such as Mohenjo-daro and other sites sufficed to meet the flooding that recurred, and came to a climax in about 1500 B.C. These continuing floods may finally have been at least partly responsible for a decline in the Harappan civilisation which degenerated into a series of smaller refugee settlements sometimes in closer proximity to the hills. It would appear, therefore, that the disruption of life in India in the second millennium B.C. was even more severe than that which occurred in Mesopotamia in the third millennium. Investigators who examine this subject in the future may do well to bear in mind the possiblity of relating the collateral evidence and seeing whether it has any bearing on the long distance trade which, in these times, was carried on between these two remote quarters of the Ancient East.