The presence of the puzzling lines, with which the Sixth Column of our text opens, was not explained by Dr. Poebel; indeed, they would be difficult to reconcile with his assumption that our text is an epic pure and simple. But if, as is suggested above, we are dealing with a myth in magical employment, they are quite capable of explanation. The problem these lines present will best be stated by giving a translation of the extant portion of the column, where they will be seen with their immediate context in relation to what follows them:

The first two lines of the column are probably part of the speech of some deity, who urges the necessity of invoking or conjuring Anu and Enlil "by the Soul of Heaven, by the Soul of Earth", in order to secure their support or approval. Now Anu and Enlil are the two great gods who had determined on mankind's destruction, and whose wrath at his own escape from death Ziusudu must placate. It is an obvious inference that conjuring "by the Soul of Heaven" and "by the Soul of Earth" is either the method by which Ziusudu has already succeeded in appeasing their anger, or the means by which he is here enjoined to attain that end. Against the latter alternative it is to be noted that the god is addressing more than one person; and, further, at Ziusudu is evidently already pardoned, for, so far from following the deity's advice, he immediately prostrates himself before Anu and Enlil and receives immortality. We may conjecture that at the close of the Fifth Column Ziusudu had already performed the invocation and thereby had appeased the divine wrath; and that the lines at the beginning of the Sixth Column point the moral of the story by enjoining on Ziusudu and his descendants, in other words on mankind, the advisability of employing this powerful incantation at their need. The speaker may perhaps have been one of Ziusudu's divine helpers—the Sun-god to whom he had sacrificed, or Enki who had saved him from the Flood. But it seems to me more probable that the words are uttered by Anu and Enlil themselves.(1) For thereby they would be represented as giving their own sanction to the formula, and as guaranteeing its magical efficacy. That the incantation, as addressed to Anu and Enlil, would be appropriate is obvious, since each would be magically approached through his own sphere of control.

It is significant that at another critical point of the story we have already met with a reference to conjuring "by the Name of Heaven and Earth", the phrase occurring at the close of the Third Column after the reference to the dream or dreams. There, as we saw, we might possibly explain the passage as illustrating one aspect of Ziusudu's piety: he may have been represented as continually practising this class of divination, and in that case it would be natural enough that in the final crisis of the story he should have propitiated the gods he conjured by the same means. Or, as a more probable alternative, it was suggested that we might connect the line with Enki's warning, and assume that Ziusudu interpreted the dream-revelation of Anu and Enlil's purpose by means of the magical incantation which was peculiarly associated with them. On either alternative the phrase fits into the story itself, and there is no need to suppose that the narrative is interrupted, either in the Third or in the Sixth Column, by an address to the hearers of the myth, urging them to make the invocation on their own behalf.

On the other hand, it seems improbable that the lines in question formed part of the original myth; they may have been inserted to weld the myth more closely to the magic. Both incantation and epic may have originally existed independently, and, if so, their combination would have been suggested by their contents. For while the former is addressed to Anu and Enlil, in the latter these same gods play the dominant parts: they are the two chief creators, it is they who send the Flood, and it is their anger that must be appeased. If once combined, the further step of making the incantation the actual means by which Ziusudu achieved his own rescue and immortality would be a natural development. It may be added that the words would have been an equally appropriate addition if the incantation had not existed independently, but had been suggested by, and developed from, the myth.

In the third and eleventh lines of the column we have further references to the mysterious object, the creation of which appears to have been recorded in the First Column of the text between man's creation and that of animals. The second sign of the group composing its name was not recognized by Dr. Poebel, but it is quite clearly written in two of the passages, and has been correctly identified by Professor Barton.(1) The Sumerian word is, in fact, to be read nig-gil-ma,(2) which, when preceded by the determinative for "pot", "jar", or "bowl", is given in a later syllabary as the equivalent of the Semitic word mashkhalu. Evidence that the word mashkhalu was actually employed to denote a jar or vessel of some sort is furnished by one of the Tel el-Amarna letters which refers to "one silver mashkhalu" and "one (or two) stone mashkhalu".(3) In our text the determinative is absent, and it is possible that the word is used in another sense. Professor Barton, in both passages in the Sixth Column, gives it the meaning "curse"; he interprets the lines as referring to the removal of a curse from the earth after the Flood, and he compares Gen. viii. 21, where Yahweh declares he will not again "curse the ground for man's sake". But this translation ignores the occurrence of the word in the First Column, where the creation of the niggilma is apparently recorded; and his rendering "the seed that was cursed" in l. 11 is not supported by the photographic reproduction of the text, which suggests that the first sign in the line is not that for "seed", but is the sign for "name", as correctly read by Dr. Poebel. In that passage the niggilma appears to be given by Ziusudu the name "Preserver of the Seed of Mankind", which we have already compared to the title bestowed on Uta-napishtim's ship, "Preserver of Life". Like the ship, it must have played an important part in man's preservation, which would account not only for the honorific title but for the special record of its creation.

It we may connect the word with the magical colouring of the myth, we might perhaps retain its known meaning, "jar" or "bowl", and regard it as employed in the magical ceremony which must have formed part of the invocation "by the Soul of Heaven, by the Soul of Earth". But the accompanying references to the ground, to its production from the ground, and to its springing up, if the phrases may be so rendered, suggest rather some kind of plant;(1) and this, from its employment in magical rites, may also have given its name to a bowl or vessel which held it. A very similar plant was that found and lost by Gilgamesh, after his sojourn with Ut-napishtim; it too had potent magical power and bore a title descriptive of its peculiar virtue of transforming old age to youth. Should this suggestion prove to be correct, the three passages mentioning the niggilma must be classed with those in which the invocation is referred to, as ensuring the sanction of the myth to further elements in the magic. In accordance with this view, the fifth line in the Sixth Column is probably to be included in the divine speech, where a reference to the object employed in the ritual would not be out of place. But it is to be hoped that light will be thrown on this puzzling word by further study, and perhaps by new fragments of the text; meanwhile it would be hazardous to suggest a more definite rendering.

With the sixth line of the column it is clear that the original narrative of the myth is resumed.(1) Ziusudu, the king, prostrates himself before Anu and Enlil, who bestow immortality upon him and cause him to dwell in a land, or mountain, the name of which may perhaps be read as Dilmun. The close parallelism between this portion of the text and the end of the myth in the Gilgamesh Epic will be seen from the following extracts,(2) the magical portions being omitted from the Sumerian Version:

The Sumerian Version thus apparently concludes with the familiar ending of the legend which we find in the Gilgamesh Epic and in Berossus, though it here occurs in an abbreviated form and with some variations in detail. In all three versions the prostration of the Deluge hero before the god is followed by the bestowal of immortality upon him, a fate which, according to Berossus, he shared with his wife, his daughter, and the steersman. The Gilgamesh Epic perhaps implies that Ut-napishtim's wife shared in his immortality, but the Sumerian Version mentions Ziusudu alone. In the Gilgamesh Epic Ut-napishtim is settled by the gods at the mouth of the rivers, that is to say at the head of the Persian Gulf, while according to a possible rendering of the Sumerian Version he is made to dwell on Dilmun, an island in the Gulf itself. The fact that Gilgamesh in the Epic has to cross the sea to reach Ut-napishtim may be cited in favour of the reading "Dilmun"; and the description of the sea as "the Waters of Death", if it implies more than the great danger of their passage, was probably a later development associated with Ut-napishtim's immortality. It may be added that in neither Hebrew version do we find any parallel to the concluding details of the original story, the Hebrew narratives being brought to an end with the blessing of Noah and the divine promise to, or covenant with, mankind.

Such then are the contents of our Sumerian document, and from the details which have been given it will have been seen that its story, so far as concerns the Deluge, is in essentials the same as that we already find in the Gilgamesh Epic. It is true that this earlier version has reached us in a magical setting, and to some extent in an abbreviated form. In the next lecture I shall have occasion to refer to another early mythological text from Nippur, which was thought by its first interpreter to include a second Sumerian Version of the Deluge legend. That suggestion has not been substantiated, though we shall see that the contents of the document are of a very interesting character. But in view of the discussion that has taken place in the United States over the interpretation of the second text, and of the doubts that have subsequently been expressed in some quarters as to the recent discovery of any new form of the Deluge legend, it may be well to formulate briefly the proof that in the inscription published by Dr. Poebel an early Sumerian Version of the Deluge story has actually been recovered. Any one who has followed the detailed analysis of the new text which has been attempted in the preceding paragraphs will, I venture to think, agree that the following conclusions may be drawn:

(i) The points of general resemblance presented by the narrative to that in the Gilgamesh Epic are sufficiently close in themselves to show that we are dealing with a Sumerian Version of that story. And this conclusion is further supported (a) by the occurrence throughout the text of the attested Sumerian equivalent of the Semitic word, employed in the Babylonian Versions, for the "Flood" or "Deluge", and (b) by the use of precisely the same term for the hero's "great boat", which is already familiar to us from an early Babylonian Version.

(ii) The close correspondence in language between portions of the Sumerian legend and the Gilgamesh Epic suggest that the one version was ultimately derived from the other. And this conclusion in its turn is confirmed (a) by the identity in meaning of the Sumerian and Babylonian names for the Deluge hero, which are actually found equated in a late explanatory text, and (b) by small points of difference in the Babylonian form of the story which correspond to later political and religious developments and suggest the work of Semitic redactors.

The cumulative effect of such general and detailed evidence is overwhelming, and we may dismiss all doubts as to the validity of Dr. Poebel's claim. We have indeed recovered a very early, and in some of its features a very primitive, form of the Deluge narrative which till now has reached us only in Semitic and Greek renderings; and the stream of tradition has been tapped at a point far above any at which we have hitherto approached it. What evidence, we may ask, does this early Sumerian Version offer with regard to the origin and literary history of the Hebrew Versions?

The general dependence of the biblical Versions upon the Babylonian legend as a whole has long been recognized, and needs no further demonstration; and it has already been observed that the parallelisms with the version in the Gilgamesh Epic are on the whole more detailed and striking in the earlier than in the later Hebrew Version.(1) In the course of our analysis of the Sumerian text its more striking points of agreement or divergence, in relation to the Hebrew Versions, were noted under the different sections of its narrative. It was also obvious that, in many features in which the Hebrew Versions differ from the Gilgamesh Epic, the latter finds Sumerian support. These facts confirm the conclusion, which we should naturally base on grounds of historical probability, that while the Semitic-Babylonian Versions were derived from Sumer, the Hebrew accounts were equally clearly derived from Babylon. But there are one or two pieces of evidence which are apparently at variance with this conclusion, and these call for some explanation.

Not too much significance should be attached to the apparent omission of the episode of the birds from the Sumerian narrative, in which it would agree with the later as against the earlier Hebrew Version; for, apart from its epitomized character, there is so much missing from the text that the absence of this episode cannot be regarded as established with certainty. And in any case it could be balanced by the Sumerian order of Creation of men before animals, which agrees with the earlier Hebrew Version against the later. But there is one very striking point in which our new Sumerian text agrees with both the Hebrew Versions as against the Gilgamesh Epic and Berossus; and that is in the character of Ziusudu, which presents so close a parallel to the piety of Noah. As we have already seen, the latter is due to no Hebrew idealization of the story, but represents a genuine strand of the original tradition, which is completely absent from the Babylonian Versions. But the Babylonian Versions are the media through which it has generally been assumed that the tradition of the Deluge reached the Hebrews. What explanation have we of this fact?

This grouping of Sumerian and Hebrew authorities, against the extant sources from Babylon, is emphasized by the general framework of the Sumerian story. For the literary connexion which we have in Genesis between the Creation and the Deluge narratives has hitherto found no parallel in the cuneiform texts. In Babylon and Assyria the myth of Creation and the Deluge legend have been divorced. From the one a complete epic has been evolved in accordance with the tenets of Babylonian theology, the Creation myth being combined in the process with other myths of a somewhat analogous character. The Deluge legend has survived as an isolated story in more than one setting, the principal Semitic Version being recounted to the national hero Gilgamesh, towards the close of the composite epic of his adventures which grew up around the nucleus of his name. It is one of the chief surprises of the newly discovered Sumerian Version that the Hebrew connexion of the narratives is seen to be on the lines of very primitive tradition. Noah's reputation for piety does not stand alone. His line of descent from Adam, and the thread of narrative connecting the creation of the world with its partial destruction by the Deluge, already appear in Sumerian form at a time when the city of Babylon itself had not secured its later power. How then are we to account for this correspondence of Sumerian and Hebrew traditions, on points completely wanting in our intermediate authorities, from which, however, other evidence suggests that the Hebrew narratives were derived?

At the risk of anticipating some of the conclusions to be drawn in the next lecture, it may be well to define an answer now. It is possible that those who still accept the traditional authorship of the Pentateuch may be inclined to see in this correspondence of Hebrew and Sumerian ideas a confirmation of their own hypothesis. But it should be pointed out at once that this is not an inevitable deduction from the evidence. Indeed, it is directly contradicted by the rest of the evidence we have summarized, while it would leave completely unexplained some significant features of the problem. It is true that certain important details of the Sumerian tradition, while not affecting Babylon and Assyria, have left their stamp upon the Hebrew narratives; but that is not an exhaustive statement of the case. For we have also seen that a more complete survival of Sumerian tradition has taken place in the history of Berossus. There we traced the same general framework of the narratives, with a far closer correspondence in detail. The kingly rank of Ziusudu is in complete harmony with the Berossian conception of a series of supreme Antediluvian rulers, and the names of two of the Antediluvian cites are among those of their newly recovered Sumerian prototypes. There can thus be no suggestion that the Greek reproductions of the Sumerian tradition were in their turn due to Hebrew influence. On the contrary we have in them a parallel case of survival in a far more complete form.

The inference we may obviously draw is that the Sumerian narrative continued in existence, in a literary form that closely resembled the original version, into the later historical periods. In this there would be nothing to surprise us, when we recall the careful preservation and study of ancient Sumerian religious texts by the later Semitic priesthood of the country. Each ancient cult-centre in Babylonia continued to cling to its own local traditions, and the Sumerian desire for their preservation, which was inherited by their Semitic guardians, was in great measure unaffected by political occurrences elsewhere. Hence it was that Ashur-bani-pal, when forming his library at Nineveh, was able to draw upon so rich a store of the more ancient literary texts of Babylonia. The Sumerian Version of the Deluge and of Antediluvian history may well have survived in a less epitomized form than that in which we have recovered it; and, like other ancient texts, it was probably provided with a Semitic translation. Indeed its literary study and reproduction may have continued without interruption in Babylon itself. But even if Sumerian tradition died out in the capital under the influence of the Babylonian priesthood, its re-introduction may well have taken place in Neo-Babylonian times. Perhaps the antiquarian researches of Nabonidus were characteristic of his period; and in any case the collection of his country's gods into the capital must have been accompanied by a renewed interest in the more ancient versions of the past with which their cults were peculiarly associated. In the extant summary from Berossus we may possibly see evidence of a subsequent attempt to combine with these more ancient traditions the continued religious dominance of Marduk and of Babylon.

Our conclusion, that the Sumerian form of the tradition did not die out, leaves the question as to the periods during which Babylonian influence may have acted upon Hebrew tradition in great measure unaffected; and we may therefore postpone its further consideration to the next lecture. To-day the only question that remains to be considered concerns the effect of our new evidence upon the wider problem of Deluge stories as a whole. What light does it throw on the general character of Deluge stories and their suggested Egyptian origin?

One thing that strikes me forcibly in reading this early text is the complete absence of any trace or indication of astrological motif. It is true that Ziusudu sacrifices to the Sun-god; but the episode is inherent in the story, the appearance of the Sun after the storm following the natural sequence of events and furnishing assurance to the king of his eventual survival. To identify the worshipper with his god and to transfer Ziusudu's material craft to the heavens is surely without justification from the simple narrative. We have here no prototype of Ra sailing the heavenly ocean. And the destructive flood itself is not only of an equally material and mundane character, but is in complete harmony with its Babylonian setting.

In the matter of floods the Tigris and Euphrates present a striking contrast to the Nile. It is true that the life-blood of each country is its river-water, but the conditions of its use are very different, and in Mesopotamia it becomes a curse when out of control. In both countries the river-water must be used for maturing the crops. But while the rains of Abyssinia cause the Nile to rise between August and October, thus securing both summer and winter crops, the melting snows of Armenia and the Taurus flood the Mesopotamian rivers between March and May. In Egypt the Nile flood is gentle; it is never abrupt, and the river gives ample warning of its rise and fall. It contains just enough sediment to enrich the land without choking the canals; and the water, after filling its historic basins, may when necessary be discharged into the falling river in November. Thus Egypt receives a full and regular supply of water, and there is no difficulty in disposing of any surplus. The growth in such a country of a legend of world-wide destruction by flood is inconceivable.

In Mesopotamia, on the other hand, the floods, which come too late for the winter crops, are followed by the rainless summer months; and not only must the flood-water be controlled, but some portion of it must be detained artificially, if it is to be of use during the burning months of July, August, and September, when the rivers are at their lowest. Moreover, heavy rain in April and a warm south wind melting the snow in the hills may bring down such floods that the channels cannot contain them; the dams are then breached and the country is laid waste. Here there is first too much water and then too little.

The great danger from flood in Babylonia, both in its range of action and in its destructive effect, is due to the strangely flat character of the Tigris and Euphrates delta.(1) Hence after a severe breach in the Tigris or Euphrates, the river after inundating the country may make itself a new channel miles away from the old one. To mitigate the danger, the floods may be dealt with in two ways—by a multiplication of canals to spread the water, and by providing escapes for it into depressions in the surrounding desert, which in their turn become centres of fertility. Both methods were employed in antiquity; and it may be added that in any scheme for the future prosperity of the country they must be employed again, of course with the increased efficiency of modern apparatus.(2) But while the Babylonians succeeded in controlling the Euphrates, the Tigris was never really tamed,(3) and whenever it burst its right bank the southern plains were devastated. We could not have more suitable soil for the growth of a Deluge story.

It was only by constant and unremitting attention that disaster from flood could be averted; and the difficulties of the problem were and are increased by the fact that the flood-water of the Mesopotamian rivers contains five times as much sediment as the Nile. In fact, one of the most pressing of the problems the Sumerian and early Babylonian engineers had to solve was the keeping of the canals free from silt.(1) What the floods, if left unchecked, may do in Mesopotamia, is well illustrated by the decay of the ancient canal-system, which has been the immediate cause of the country's present state of sordid desolation. That the decay was gradual was not the fault of the rivers, but was due to the sound principles on which the old system of control had been evolved through many centuries of labour. At the time of the Moslem conquest the system had already begun to fail. In the fifth century there had been bad floods; but worse came in A.D. 629, when both rivers burst their banks and played havoc with the dikes and embankments. It is related that the Sassanian king Parwiz, the contemporary of Mohammed, crucified in one day forty canal-workers at a certain breach, and yet was unable to master the flood.(2) All repairs were suspended during the anarchy of the Moslem invasion. As a consequence the Tigris left its old bed for the Shatt el-Hai at KÛt, and pouring its own and its tributaries' waters into the Euphrates formed the Great Euphrates Swamp, two hundred miles long and fifty broad. But even then what was left of the old system was sufficient to support the splendour of the Eastern Caliphate.

The second great blow to the system followed the Mongol conquest, when the NahrwÂn Canal, to the east of the Tigris, had its head swept away by flood and the area it had irrigated became desert. Then, in about the fifteenth century, the Tigris returned to its old course; the Shatt el-Hai shrank, and much of the Great Swamp dried up into the desert it is to-day.(1) Things became worse during the centuries of Turkish misrule. But the silting up of the Hillah, or main, branch of the Euphrates about 1865, and the transference of a great part of its stream into the HindÎyah Canal, caused even the Turks to take action. They constructed the old HindÎyah Barrage in 1890, but it gave way in 1903 and the state of things was even worse than before; for the Hillah branch then dried entirely.(2)

From this brief sketch of progressive disaster during the later historical period, the inevitable effect of neglected silt and flood, it will be gathered that the two great rivers of Mesopotamia present a very strong contrast to the Nile. For during the same period of misgovernment and neglect in Egypt the Nile did not turn its valley and delta into a desert. On the Tigris and Euphrates, during ages when the earliest dwellers on their banks were struggling to make effective their first efforts at control, the waters must often have regained the upper hand. Under such conditions the story of a great flood in the past would not be likely to die out in the future; the tradition would tend to gather illustrative detail suggested by later experience. Our new text reveals the Deluge tradition in Mesopotamia at an early stage of its development, and incidentally shows us that there is no need to postulate for its origin any convulsion of nature or even a series of seismic shocks accompanied by cyclone in the Persian Gulf.

If this had been the only version of the story that had come down to us, we should hardly have regarded it as a record of world-wide catastrophe. It is true the gods' intention is to destroy mankind, but the scene throughout is laid in Southern Babylonia. After seven days' storm, the Sun comes out, and the vessel with the pious priest-king and his domestic animals on board grounds, apparently still in Babylonia, and not on any distant mountain, such as Mt. Nisir or the great mass of Ararat in Armenia. These are obviously details which tellers of the story have added as it passed down to later generations. When it was carried still farther afield, into the area of the Eastern Mediterranean, it was again adapted to local conditions. Thus Apollodorus makes Deucalion land upon Parnassus,(1) and the pseudo-Lucian relates how he founded the temple of Derketo at Hierapolis in Syria beside the hole in the earth which swallowed up the Flood.(2) To the Sumerians who first told the story, the great Flood appeared to have destroyed mankind, for Southern Babylonia was for them the world. Later peoples who heard it have fitted the story to their own geographical horizon, and in all good faith and by a purely logical process the mountain-tops are represented as submerged, and the ship, or ark, or chest, is made to come to ground on the highest peak known to the story-teller and his hearers. But in its early Sumerian form it is just a simple tradition of some great inundation, which overwhelmed the plain of Southern Babylonia and was peculiarly disastrous in its effects. And so its memory survived in the picture of Ziusudu's solitary coracle upon the face of the waters, which, seen through the mists of the Deluge tradition, has given us the Noah's ark of our nursery days.

Thus the Babylonian, Hebrew, and Greek Deluge stories resolve themselves, not into a nature myth, but into an early legend, which has the basis of historical fact in the Euphrates Valley. And it is probable that we may explain after a similar fashion the occurrence of tales of a like character at least in some other parts of the world. Among races dwelling in low-lying or well-watered districts it would be surprising if we did not find independent stories of past floods from which few inhabitants of the land escaped. It is only in hilly countries such as Palestine, where for the great part of the year water is scarce and precious, that we are forced to deduce borrowing; and there is no doubt that both the Babylonian and the biblical stories have been responsible for some at any rate of the scattered tales. But there is no need to adopt the theory of a single source for all of them, whether in Babylonia or, still less, in Egypt.(1)

I should like to add, with regard to this reading of our new evidence, that I am very glad to know Sir James Frazer holds a very similar opinion. For, as you are doubtless all aware, Sir James is at present collecting Flood stories from all over the world, and is supplementing from a wider range the collections already made by Lenormant, Andree, Winternitz, and Gerland. When his work is complete it will be possible to conjecture with far greater confidence how particular traditions or groups of tradition arose, and to what extent transmission has taken place. Meanwhile, in his recent Huxley Memorial Lecture,(1) he has suggested a third possibility as to the way Deluge stories may have arisen.

Stated briefly, it is that a Deluge story may arise as a popular explanation of some striking natural feature in a country, although to the scientific eye the feature in question is due to causes other than catastrophic flood. And he worked out the suggestion in the case of the Greek traditions of a great deluge, associated with the names of Deucalion and Dardanus. Deucalion's deluge, in its later forms at any rate, is obviously coloured by Semitic tradition; but both Greek stories, in their origin, Sir James Frazer would trace to local conditions—the one suggested by the Gorge of Tempe in Thessaly, the other explaining the existence of the Bosphorus and Dardanelles. As he pointed out, they would be instances, not of genuine historical traditions, but of what Sir James Tyler calls "observation myths". A third story of a great flood, regarded in Greek tradition as the earliest of the three, he would explain by an extraordinary inundation of the Copaic Lake in Boeotia, which to this day is liable to great fluctuations of level. His new theory applies only to the other two traditions. For in them no historical kernel is presupposed, though gradual erosion by water is not excluded as a cause of the surface features which may have suggested the myths.

This valuable theory thus opens up a third possibility for our analysis. It may also, of course, be used in combination, if in any particular instance we have reason to believe that transmission, in some vague form, may already have taken place. And it would with all deference suggest the possibility that, in view of other evidence, this may have occurred in the case of the Greek traditions. With regard to the theory itself we may confidently expect that further examples will be found in its illustration and support. Meanwhile in the new Sumerian Version I think we may conclude that we have recovered beyond any doubt the origin of the Babylonian and Hebrew traditions and of the large group of stories to which they in their turn have given rise.