The civilization of Babylonia and Assyria

Its remains, language, history, religion, commerce, law, art, and literature

by Morris Jastrow | 1915 | 168,585 words

This work attempts to present a study of the unprecedented civilizations that flourished in the Tigris-Euphrates Valley many thousands of years ago. Spreading northward into present-day Turkey and Iran, the land known by the Greeks as Mesopotamia flourished until just before the Christian era....

It was by the roundabout way of the old Persian inscriptions that the approach to the decipherment of the cuneiform material, found in such (abundance through the excavations conducted during the past seventy years in the mounds covering Babylonian and Assyrian cities, was made. Even before Botta's finds at Khorsabad arrived at the Louvre, it was apparent from the few specimens of cuneiform inscriptions from the Euphrates Valley brought to London by Rich early in the nineteenth century, [1] that the script was of the same style as Class III on the trilingual inscriptions from Perseopolis and neighboring districts.

It was some time, however, before the obvious conclusion was drawn that Class III represented the script and language used in Babylonia and Assyria. Grotefend began, in 1814, to publish the results of his study of the writing of Class III on the few inscriptions that had come from Babylonia, but it was not until 1818 that he recognized the identity of the two and explained the variations as merely incidental modifications of the same order as the differences in handwritings.

He also recognized the much larger number of combinations of wedges forming the signs in Class III and in the inscriptions from Babylonia. Indeed, by the year 1819 he had distinguished no less than 287 signs. For all that, he assumed that the Babylonian cuneiform also represented, like the Persian variety, an alphabetic script, though with this modification, that the sign varied according to the vowel accompanying a consonant. In this way he hoped to account for the larger number of signs in Class III as against Class I.

As more and more inscriptions were brought out of the mounds at Khorsabad, at Nimrud, at Kouyunjik and Kaleh-Shergat and subsequently from the mounds in the south, the identity of both the Assyrian and Babylonian varieties of cuneiform characters with Class III of the Persepolitan inscriptions was confirmed, and at the same time the nature of the variations entirely secondary in character came to be better understood. There was no longer any doubt that Class III represented the variety of cuneiform writing used in Babylonia and Assyria and, therefore, also the language spoken in these countries.

Because of the accident that the first large finds were made at northern mounds, i.e., on the site of Assyrian cities, the language of Class III was designated as Assyrian, and the science that grew up out of the discoveries in the northern and southern mounds, Assyriology. Since, however, the north owes her civilization, her literature and art largely to the -south, it is more proper to speak of the language -as Babylonian or as Babylonian-Assyrian.

That Persian kings should commemorate themselves and their deeds in the language of Babylonia and Assyria in addition to doing so in the official language of the kingdom was quite natural, seeing that the old realm of Babylonia and Assyria formed one of the most important of the lands conquered by Cyrus and retained by his successors, just as the third variety of cuneiform script on the monuments of Persian rulers Class II proved to be [2] the language of the large district of Elam within which the Persian kings had established their capital cities, Persepolis, Susa and Ecbatana.

The Babylonian on all the trilingual inscriptions from ancient Persia was evidently a translation. The inscription in old Persian as the official language of the kingdom was the original occupying, therefore, always the -first or most prominent position of the three of which those in the two other varieties of script were translations. The decipherment of Class I, therefore, served as a vantage point for attacking Class III, old Persian cuneiform furnishing the same aid in deciphering Babylonian cuneiform script as the Greek translation on the Rosetta stone served in laying the foundation for the reading of the Egyptian hieroglyphics.

In both cases, as indeed in the decipherment of any unknown script, the beginning was made with proper names, which H could be picked out through a comparison of their relative position in inscriptions of Class I and Class III, or in the case of the Rosetta Stone, by a study of their relative position in the Greek and hieroglyphic texts. "When proper names occurred more than once in any inscription or occurred in several inscriptions, as was the case in the old Persian monuments, most of which dated from Darius and Xerxes, there could of course be no difficulty in picking out in Class III the series of signs corresponding to these names in Class I.

The case was somewhat more difficult when a proper name occurred only once, since Class III did not have the very convenient diagonal wedge separating words from one another, but, on the other hand, with a large number of names, both of persons and places revealed through the large inscription at Behistun, the constant occurrence of the same signs in a variety of names that could be read in Class I furnished a control in picking out the series of signs in Class III, corresponding to any particular name in Class I.

The early decipherers like Lowenstern, [3] Longperier [4] and Saulcy [5] floundered about considerably. Botta, the successful explorer of Khorsabad, alone made a really valuable contribution by his careful study of the numerous inscriptions found by him and which showed a large number of words evidently identical and yet written in different ways. By making a list of these variants he paved the way for the discovery made by Hincks that the Babylonian-Assyrian script was not alphabetical but syllabic and ideographic, that is to say, that words were written by means of signs, each having a syllabic value, or by means of a single sign standing for the word.

This discovery was announced by Hincks, in 1847, [6] and threw an entirely new light on the character of the third variety of cuneiform script. With signs expressing syllables or standing for entire words, it at once became clear why there should be so many signs in this variety of script. The variant ways of writing the same word, as shown by Botta, also became clear. Not only could a word be written by a single sign used ideographically or by a series of signs, each having a syllabic value, but since syllables were of three kinds, (a) consonant and vowel, (b) vowel and consonant, (c) consonant, vowel and consonant, it was possible to write a word syllabically in various ways.

Thus the word for god, ilu, could be written by a single sign or it could be spelled out and written by two signs t and lu; and if, e.g., the syllable lab formed part of a word, it might be written by one sign having that value or further subdivided into la and ab and written with two signs. Until this discovery was made by Hincks there was no certainty even as to the reading of the proper names that could be picked out in Class III through comparison with Class I.

The signs constituting the names could be chosen, but since the number of signs forming a name in Class III did not agree with the number in Class I, it was evidently impossible to determine the value of each sign. Guess followed upon guess, conjecture upon conjecture, until Hincks definitely demonstrated the general character of the script of Class III, which represented the same language as that found on the monuments of Assyria and Babylonia.

It was now possible by a comparison between proper names in Classes I and III of the Persepolitan inscriptions to read the syllabic equivalents in Class III for the alphabetical signs in Class I. Thus, e.g., the seven signs representing the name Xerxes in Class I

Kh.sh(a).y.a.r.sh.a

corresponded to six signs in Class III, to be read

Khi-shi-'i-ar-shi-i.

Similarly the seven signs in Class I for the name Darius

D.a.r.h.e.u.sh

corresponded to five signs in Class III, the correct reading of which turned out to be

Da-ri-'i-a-mush

but which might also be written with six signs

Da-ri-'i-a-a-mush.

The word designating these rulers as Achaemenians appeared in Class I as

H(a).kh.a.m(a).n.i.sh.i.y(a) .

These nine signs were represented in Class III by seven signs to be read

A-kha-ma-an-nish-shi-'i

or by eight signs

A-kha-ma-an-ni-ish-shi-'i

since nish could be written either by one sign or by two (ni and ish). An important result of Hincks' investigations, which greatly facilitated the picking out of proper names, both in Class III and in the unilingual inscriptions of Assyria and Babylonia, was the observation that names of persons were preceded by a vertical wedge, names of gods by a sign which signified "heaven", while names of countries and of cities also had special "determinatives", as they were called.

It was thus possible to be quite certain as to the beginning of names at least, whether in the trilingual or in the unilingual inscriptions. A secure basis for determining the correct reading of signs occurring in proper names was obtained upon the recognition of the fact that the vowels alone represented the alphabetic element in the Assyrian and Babylonian cuneiform script.

It was now possible also to proceed with a greater feeling of assurance to the reading of ordinary words in Class III, such as "son", "king", "country", "father", "god", "heaven", "earth", which occurred with sufficient frequency to enable scholars, by a comparison with Class I, to pick out the series of signs or the single sign with which they were written. At this juncture, in 1851, [7] Henry Rawlinson again appeared on the scene with the publication of the Babylonian section (i.e., Class III) of the great Behistun inscription. [8]

Accepting Hincks' principle of the syllabism of the Babylonian cuneiform, he was enabled through the comparison of the several hundred names of persons and places occurring in Class I with the corresponding manner of writing these names in Class III now rendered comparatively simple through the observation of the determinatives preceding names of persons, cities and countries, to settle the value of a very large number of the signs, in fact over 200 of them.

This marked a great progress indeed. Rawlinson was also enabled to add to the number of ideographic writings that could be regarded as certain, including the signs designating son, father, great, lord and more the like. Hincks followed up his own researches by preparing lists of the Assyro-Babylonian characters and by 1855 he had fixed the value of 252 combinations of wedges.

The investigations of Hincks and Rawlinson had shown beyond possible doubt another fact which at first sight seemed very strange, that a single sign could have more than one syllabic value. To this feature the name "polyphony" was given; and though the proposition encountered opposition, it was not long before it replaced the supposed "homophony", proposed by Lowenstern, according to which different signs were supposed to have the same value ; and that in this way the existence of so many signs was to be accounted for.

Jules Oppert, who in 1855 gave a survey of the stage reached in the decipherment of Babylonian cuneiform, came to the support of Hincks and Rawlinson and showed that even a larger number of signs than Hincks had suspected had more than one sound, and it subsequently turned out that outside of the vowels the number that had only one syllabic value was very limited.

We owe to Oppert also the utilization of an important help for determining the various syllabic values for a sign and for proving that corresponding to "polyphony" we have also the phenomenon of "polyideography" in Babylonian cuneiform, that is to say, the circumstance that the same sign may also stand for several words, though usually in some logical connection with each other. Among the tablets of Ashurbanapal's library brought to the British Museum by Layard, [9] were long lists of signs arranged in columns.

Oppert went to the British Museum to study these lists and found that they formed part of a large text-book literature prepared by Babylonian and Assyrian scribes to facilitate instruction in cuneiform writing. The lists were of various kinds, consisting usually of three columns with a single sign in the central column and a series of signs in each of the two other columns. Hincks had recognized that in some of these lists the signs in the right hand column were intended to indicate the syllabic value of the sign in the central column.

Thus a certain sign MW was repeated three times in the central column, while the right hand column contained different signs, as follows:

The value of the signs of the right hand column having been determined from a comparison of proper names of Class III with those of Class I of the Persepolitan inscriptions, Hincks could interpret the lines as follows:

line 1: The sign  has the syllabic value of li-ib
line 2: ""            ""            ""            ""              da-an
line 3: ""            ""            ""            ""              ka-al

That is to say the sign in question may be read in the composition of words as lib, dan or kal, as the case may be. It will readily be seen how with long lists of such signs, the principle of syllabism and polyphony was not only definitely confirmed, but the syllabic values of the signs were ascertained with equal definiteness.

It is one of the many merits of Oppert to have demonstrated the full significance of these syllabaries (as the lists were called) in the further progress of decipherment. The right hand column in some of these syllabaries contained a series of signs which furnished in syllabic writing the words which a sign represented, or in other words the ideographic values, while the left hand column furnished the syllabic values. Thus in the case of a sign $& we had a series as follows:

ki-i                             it-tu, "side"
""                   ""                ash-ru, "place"
""                   ""                ir-si-tu, "earth"

which meant that the sign in question as an ideograph could be read ittu, ashru or irsitu with the meanings "side," "place" and" 'earth" respectively, while when used as a syllable entering in the composition, written syllabically it had the value ~ki, so that a word kirib meaning "within" could be written by the sign ^M' (i.e., ki) and the sign ^lt which, in addition to the syllabic values above [10] pointed out, has also the value rib, Again, a sign ^D appeared in a list as follows :

shi-ti                             me-nu-tu, "number"
ag                     ""                 it-ku "strong"
la-ag                 ""                 kir-ba-an-nu, "offering"

i.e., the sign in question as an ideograph could be read in one of the three fashions indicated, while corresponding to the three ideographic values, there were also three syllabic values.

An invaluable aid had thus been secured for the reading of the Assyrian and Babylonian inscriptions and an aid, moreover, whose authenticity could not be doubted, since we had before us the syllabic and ideographic values of the signs that the schoolmasters of ancient Mesopotamia had themselves compiled as a help towards reading the inscriptions on the monuments and with a view of initiating their pupils into the method of writing cuneiform, as well as reading it.

Now to be sure the existence of various syllabic and ideographic values for one and the same sign added to the difficulty of reading inscriptions of Class III, but it was not long before, through the combined efforts again of Hincks, Rawlinson and Oppert, it was found that the Babylonian and Assyrian scribes used certain devices to simplify the cumbersomeness of their cuneiform script. In case of a sign X which could be read lib, dan, kal, rib, etc., the final consonant was frequently repeated by following up the sign in question with a sign beginning with that consonant.

Thus, if after the sign in question, a sign Y was written which had the value bi, it was an indication that the preceding sign was lib; if the next sign, however, was li, it was an indication that in that particular instance the sign X was to be read kal; if the next sign was nin, it indicated that the sign was to be read dan. Besides, when once the character of the Babylonian language was ascertained, it was possible, in a large number of cases, to decide without difficulty which of several values attached to a sign should be chosen in order to produce a word which could represent either a possible verbal form or a noun formation in the language.

This brings us to the important question as to the language of Class III and of the Babylonian-Assyrian inscriptions, and how it was possible, after spelling out the words of an Assyrian or Babylonian inscription to determine to what class of languages the words belonged. At first, scholars were entirely at the mercy of their individual guesses.

Grotefend, who was the first to call the language Assyrian, refrained from committing himself beyond expressing his opinion that he could find no indications that the language belonged to the Semitic class. Gesenius, one of the most eminent Semitic scholars of his day, declared that it belonged to the Medo-Persian group. Philoxenus Luzatto, the son of a distinguished Hebrew scholar, published two monographs in 1849, [11] in which he proposed the thesis that the Assyrian was allied to Sanskrit.

There were others, however, who, starting from the Biblical tradition, [12] which placed Ashur (or the Assyrians) among the sons of Shem, conjectured that the language was Semitic. The question could not be definitely decided until it was possible to reach a degree of certainty as to the exact way in which proper names and the most common words of the Assyrian inscriptions could be read.

With the recognition of the syllabism of the Babylonian-Assyrian cuneiform a beginning in this direction was made ; and when, by following the method introduced by Hincks 1 and his successors, such words as a-bu ("father"), ra-bu-u ("great"), shar-ru ("king"), a-na-ku ("I") and verbal forms like i-zan-nan ("he beautifies"), i-kash-shad ("he conquers"), were spelled out, the meanings of which were settled by a comparison with the corresponding words in inscriptions of Class I or through the context the comparison with the common Semitic noun ab for "father", with the Semitic stem rob ("great"), with the Hebrew sar ("prince"), with the Hebrew pronoun of the first person anoki and with the common Semitic verbs zandnu and kashddu in the sense of "adorning" and "conquer", the indications pointed unmistakably towards Assyrian as one of the group of languages known as Semitic. [13]

The peculiarities of the Semitic languages are so marked that one cannot long be in doubt in the case of a new language discovered whether it belongs to the group or not. The forms or moods of the Semitic verb are also of a stereotyped character, and a Semitist can tell almost at a glance whether any given verbal form is a possible one in a Semitic language. Furthermore the agreement in vocabulary among the Semitic languages is also considerable, though this varies naturally among the subdivisions of the group.

Step by step, little by little, the difficulties were overcome, one problem after the other was solved until, in 1857, a test was made which showed that the decipherment of Assyrian rested on a firm basis. At the suggestion of H. Fox Talbot, who was among the early students of Assyriology, an Assyrian historical text was chosen and four scholars Hincks, Rawlinson, Oppert and Talbot himself agreed to send to the Royal Asiatic Society a translation independently made.

No translation of the inscription had ever been published. The plan was carried out, and the commission appointed to compare the four translations [14] found the agreement to be so complete in all essentials as to carry conviction even to those who had hitherto questioned the soundness of the method pursued. But the skeptics were not all silenced, and even when a few years later another remarkable confirmation of the correctness of the decipherment was quite accidentally furnished, many scholars among them distinguished investigators like Ernest Renan in France and Alfred Gutschmid in Germany continued to query the results reached.

The reason for the doubt still existing in the minds of such scholars as Renan and Gutschmid [15] was due largely to the difficulty of accounting for the polyphonic character of the signs and to the puzzling complications in the writing of native Assyrian and Babylonian names of persons and places, in consequence of their being written in part ideographically and in part syllabically.

It was natural to raise the question, since writing is a medium of expressing facts and ideas, why a people should have developed a script so confusing that each sign might have one of several values, and furthermore how could one ever be sure in the case of signs used ideographically that any proposed reading was the correct one, since a sign could stand for a number of words, even though there was an association of ideas between the words so represented?

The answer to these questions could not be furnished until some light had been thrown on the origin of cuneiform writing. . That the wedge-shaped signs represented originally pictures and were modifications of hieroglyphics was to be concluded from the fact that a sign could stand for an entire word. In the case of some of the signs, the pictorial origin was, moreover, quite apparent.

Thus, the sign for "god" and "heaven" which had ordinarily the form , in older inscriptions, particularly in those found in mounds of the south, had a form like another form of heaven in cuneiform and it was an obvious conclusion that this represented a star.

A sign Cuneiform sign for Hand signifying "hand" showed even in this late form its origin from a picture of the fingers of the hand ; nor was it difficult to recognize in the form Cuneiform word for House, standing for "house", its development from the picture of some kind of construction, especially when one compared the late form with a more elaborate one Cuneiform symbol for construction, found in some inscriptions of an older period or which imitated the older forms of the script.

Oppert, as far back as 1856, had shown that the sign Cuneiform symbol for Fish signifying "fish" had been evolved from the picture of a fish, the outlines of which head, body, tail and fins could still be distinguished in a more archaic form Archaic symbol for Body, found on Babylonian monuments.

As a means of facilitating the reading of signs used ideographically, Oppert and others had also pointed out the use of a sign intended to be read syllabically and placed after an ideograph to indicate the final syllable of the word designated. By means of this phonetic complement it was possible to feel certain, e.g., that the sign for "god" and "heaven" when followed by a sign having the value tu was to be read elitu, "upper" ; a sign that could stand for umu (day), urru (light) and shamshu (sun) was to be read as urn if followed by mu, whereas if "sun" was intended, it was accompanied by a phonetic complement shu or shi or ash, which indicated that it was to be read shamshu (nominative case), shamshi(gen.), or shamash (construct state).

All this was of some help, but uncertainty still existed in very many cases, and even the explanation of the hieroglyphic origin of the wedges did not account for the many values that a sign used phonetically might have, for there seemed to be no connection between the syllabic and ideographic values.

It was again the ingenuity of Hincks that suggested the solution. In a paper read before the British Association for the Advancement of Science, in 1850, [16] Hincks threw out the hint that while the oldest cuneiform writing that of Class III and the Assyrian-Babylonian inscriptions was Semitic, the origin of the script was not Semitic. He based this view upon the insufficiency of the cuneiform syllabary for distinguishing between softer and harder palatals and dentals that form an ingredient of the consonantal system in the Semitic languages, and that in other respects it was not suitable for writing words belonging to a language of the Semitic group.

He drew the inference that the writing had been adopted by the Babylonians and Assyrians from some Indo-European people which had conquered the country; he expressed the further belief that this people had relations with Egypt from which the cuneiform script was ultimately drawn. Rawlinson at first also accepted the Egyptian origin of the cuneiform script, but afterwards advanced the view that the people who conquered Babylonia and imposed their script on the country were Scythians a view that was modified by Oppert to the extent of designating the language of the inventors as Casdo-Scythian, and who compared it to some of the languages of the Turanian group of Russia.

On the assumption of a foreign origin for the cuneiform script, it was possible to explain the circumstance that there was no agreement between the ideographic and the syllabic values of a sign. The syllabic values represented the non-Semitic words which were the equivalents in the language of the inventors to the ideographic values of the sign in the Semitic idiom of Babylonia.

Thus, if in the class of three-columned syllabaries above referred to, [17] we find the sign Cuneiform 'an' meaning God in the middle column, explained as follows:

an Cuneiform 'an' meaning God ilu

This meant that an was the equivalent in Casdo-Scythian for the Semitic ilu, ' ' god. ' ' The Babylonians, when adopting the foreign script, conceived the idea of using the non-Semitic word an as a syllable with which to write words particularly verbal forms and inflected nouns which could not well be expressed ideographically. Thus the non-Semitic word an would be used syllabically to write a Semitic word ending in an like dan-an.

The theory assumed that the inventors of the script used it as an ideographic medium, and that the borrowers took the forward step of converting it into a mixed ideographic and syllabic script. In this way the various syllabic values of a sign admitted of a reasonable explanation, while the various ideographic values could in most cases be accounted for by association of ideas. The case would be analogous if the French had adopted a form of sign-writing from the English, and at the same time used the English sounds of the signs to spell words in their own language, while the same sign when standing for a word would of course be read as a French word.

Thus the French word del would be written with the sign, which would be read "heaven" in English, or it would be written syllabically ci + el, in which case the sign which in English designated "sea" would be used because it had the same sound as the first syllable of the French word for heaven, while the second syllable would be written by the English sign for "ell", because the sound of the English word fitted the case. In the same way, the Babylonians wrote their words in non-Semitic form but pronounced them as Semitic.

The designation Scythian or Casdo-Scythian was vehemently contested by various scholars. Rawlinson himself abandoned it in 1855 in favor of Akkadian, because of the frequency with which the name Akkadian occurring as Akkad also in Gen. 10, 10 was mentioned in the Babylonian and Assyrian inscriptions.

In 1869, [18] Oppert, basing his arguments on the occurrence of the title, "king of Sumer [19] and Akkad" in the inscriptions of very ancient rulers, proposed the term Sumerians for the non-Semitic settlers of the Euphrates Valley, and Akkadians for the Semitic population. This view, after a long controversy with many changes of front on the part of scholars, has been finally demonstrated to he the correct one.

But who were these Sumerians? Where did they come from ? And what was the nature of the language which they spoke? Before taking up this question a few words need to be said about a long and animated controversy regarding Sumerian and the Sumerians which began in 1874, and which has continued down to the present time.

While the theory of the non-Semitic origin and character of the cuneiform script seemed to furnish an explanation for some of the problems involved in so complicated and comprehensive a form of writing as the Babylonians developed and passed on to the Assyrians, new difficulties arose as more material was brought out of the mounds, difficulties that did not appear to be met by the Sumerian theory as we may briefly call it. In the first place it was observed that many of the syllabic values of the signs were portions of a Semitic word for which the sign stood.

So a sign Cuneiform 'reshu' meaning Head which, both in syllabaries and in texts, stood for the word reshu, "head", has as its syllabic values sag and risk. The former was the non-Semitic word for head, according to the Sumerian theory, but the other value, risk, evidently stood in some relationship to the Semitic equivalent of the sign used as an ideograph. Again, if among the syllabic values of a sign which stands for the Semitic dannu, "strong", we find dan, it was evident that this value was an abbreviation of the Semitic word.

Such instances began to multiply and when it was found that at least one hundred syllabic values had all the appearance of representing pails of Semitic words, the conclusion was forced upon scholars that the Babylonian-Assyrian syllabary was in part at least Semitic. To account for this the adherents of the Sumerian theory maintained that the Babylonians after adopting the non-Semitic mode of writing and taking the step of converting it from an ideographic to a mixed ideographic and syllabic script, continued to develop cuneiform writing and added to the Sumerian words employed as syllabic values, parts of the Semitic words for which the signs stood, but used likewise as syllabic values.

Meanwhile, cuneiform texts of the older period were coming to light from mounds in the south, from which it became clear that the Assyrian civilization was merely an offshoot of the culture that arose in the south, in the Euphrates Valley. It was therefore in the south that the solution of the problem as to the origin of the culture and the script was to be sought. Now, as one proceeded backwards, the texts appeared to be more and more ideographic in character.

Ere long texts were found which seemed to be entirely ideographic, and such texts increased largely in numbers with the unearthing of the ancient city of Shirpurla (or Lagash) through de Sarzec. [20] The inscriptions on the many statues and votive offerings of Gudea and of other rulers were written in the older style, which scholars now began to regard as Sumerian; and yet even on these monuments Semitic words appeared and again some of the oldest inscriptions of the south were clearly Semitic and not Sumerian.

What did all this mean? If the Sumerians originated the Sumerian culture and were the inventors of the script, we should expect to find the oldest inscriptions to be in Sumerian and, what is more, in pure Sumerian; and it ought also to be possible to reconstruct the original language of the cuneiform script in such a way as to place the language in some definite group, as the Babylonian and the Persian cuneiform had been. Various attempts of this kind to find affiliations between Sumerian and Turkish or between Sumerian and some Ural-Altaic groups failed.

It was therefore natural that a doubt should have arisen whether the Sumerian represented a real language or whether the Sumerians, if they existed, were the originators of the culture and the inventors of the script.

The Sumerian theory manifested at first such weaknesses that one of the most eminent Semitists of his day, Joseph Halevy, was led to put forward the thesis that what scholars regarded as the Sumerian language was nothing but an older ideographic method of writing the Semitic Akkadian or Babylonian, which, in the course of its evolution, had adopted many more or less artificial devices for expressing niceties of thought and grammatical complications. The thesis carried with it the Semitic origin of the Euphratean culture and practically eliminated the Sumerians altogether. Sumer and Akkad as they appeared on the tablets of early rulers in the Euphrates Valley were purely geographical designations of the southern and northern portions of the valley respectively. [21]

Even the opponents of Halevy were obliged to admit that he had revealed weak points in the Sumerian theory and it is due to him that Assyriology was deflected from the erroneous direction into which it had turned. It is now admitted that many of the hymns and incantations which scholars had been accustomed to regard as Sumerian are comparatively late compositions, or that they have come down to us in a late revised form betraying Semitic influences.

It is also generally admitted to a larger extent than was formerly the case that the Semitic settlers of Babylonia had a large share in perfecting the cuneiform syllabary, that many texts which are written ideographically are in reality Semitic compositions and are to be read as such, and that even in genuine Sumerian texts Semitic influence is apparent ; but for all that, evidence sufficient in both quantity and quality has been brought forward to show that the early population of the Euphrates Valley was mixed in character, that by the side of Semites we find a Turanian race clearly depicted on the monuments and demarcated by their physiognomies and by differences of costume from the Semitic population.

We owe to Eduard Meyer [22] the definite establishment of this thesis. On the linguistic side, evidence for the existence of a Sumerian language has recently been brought forward which does not rest upon guesswork or on pure conjecture, but is made conclusive by the study of the oldest texts of Babylonia. As long as Sumerian was simply to be deduced from the ideographic values of the signs, one was justified in doubting whether we were in the presence of a real language, for since ideographs could be read as Semitic as well as Sumerian, it was indeed possible to regard a "Sumerian" inscription as merely another form of writing Babylonian a very artificial form to be sure and yet, since all writing is a more or less artificial device, a possible form.

When, however, the proof was furnished from the texts that Sumerian words could be written phonetically as well as ideographically, that even in Sumerian the device existed of writing a word as in Babylonian either by a single sign representing the word or by signs representing the syllables of which it is composed, there could no longer be any question as to the genuine linguistic character of Sumerian. In addition to the evidence for phonetic writing, which became more and more abundant as scholars penetrated deeper into the study of the oldest texts from ancient

Babylonian centres, [23] the proof of a fixed grammatical structure for nouns and verbal forms was furnished in a manner to carry conviction to the minds of those who had hitherto maintained a skeptical or non-committal attitude towards the linguistic evidence.

Taking up now the question who these Sumerians were, an impartial verdict must confess that the problem still remains obscure. We know that they were not Semites; their features as depicted on the monuments reveal a Turanian type, but the term Turanian is too vague to furnish any definite clue. Various indications point to their having come from a mountainous region. They brought the worship of their native gods with them, and the nature of these deities suggests their having had their original seats on the tops of mountains.

It is to the Sumerians that we owe the construction of the stage-towers of which remains have been found in all the important centres of Babylonia and Assyria. Built in imitation of mountains with an imitation of a mountain road leading to the sanctuary at the top, it is reasonable to conclude that the thought of housing the gods in this way arose in the minds of a people accustomed to the worship of gods whose seats were on mountain peaks. There is other evidence pointing in the same direction of an original mountain home whence the Sumerians came at a remote period to settle in the Euphrates Valley.

Now there are mountains to the east and north-east of Babylonia, and it is therefore possible that the Sumerians entered the Valley from this side perhaps under pressure of other mountain hordes coming from the north. But they may also have come, as has been recently maintained, from mountainous districts to the northwest of Mesopotamia.

Whether the Sumerians already found the Semites in possession of Babylonia and then conquered them, or whether the Sumerians were the earliest settlers and founded the culture in that district is another question that has not been definitely decided, with the evidence, however, in favor of the view that the Semites were the first on the ground and that they had already made some advance in culture when the Sumerians swept down on them and imposed their rule and such culture as they brought with them on the older settlers.

Footnotes and references:

[1]:

Above, p. 14.

[2]:

See below, p. 107 seq.

[3]:

Essai de dechiffrement de l'ecriture assyrienne pour servir a l'explication du monument de Khorsdbad (Paris, 1845). Lowenstern attempted to explain the Khorsabad inscriptions without having recourse to the Persian inscriptions, with results that were naturally disastrous. He did guess correctly that the Assyrians spoke a Semitic language, but this led him to make the absurd attempt to explain the cuneiform characters as modifications of Hebrew letters.

[4]:

Several articles in the Revue Arckeologique for 1847 in reply to Lowenstern's second work, Expose des elements constitutifs du systeme de la troisieme ecriture cuneiforme de Persepolis (Paris, 1847).

[5]:

Recherches sur l'ecriture cuneiforme du systeme Assyrien (Paris, 1849). Both Longperier and Saulcy made some correct and ingenious guesses, by the side, however, of so many errors that their work did not mark any real progress.

[6]:

On the Third Persepolitan Writing, etc. (Transactions of the Royal Irish Academy, vol. xxi, Part II, pp. 249-256.)

[7]:

Analysis of the Babylonian Text at Behistun (Journal of the Royal Asiatic Society, vol. xiv, pp. i-civ).

[8]:

See above, p. 83.

[9]:

See above, p. 22, seq.

[10]:

Page 93.

[11]:

Le Sanscritisme de la Langue Assyrienne, etc. (Padua, 1849).

[12]:

Gen. 10, 22. The grouping of nations in this chapter as sons of Shem, Ham and Japhet has of course no scientific value, though the list is remarkable as an indication of the knowledge of the day and because of the traditions that it embodies. The division appears to be into three zones. The peoples living in the northern zone are grouped as sons of Japhet, those in the middle as sons of Shem and those in the south as sons of Ham. Babylonia is placed in the southern zone, Assyria in the northern. The chapter is composite in character and full of late insertions and glosses. The system is abandoned in the case of the Canaanites, who are placed among the sons of Ham because of the hostile feelings of the Hebrews towards them.

[13]:

It is now customary to range the Semitic languages into two groups: (1) Northern Semitic to which Hebrew, PhcEnieian and the various dialects of Aramaic and Syriac and Babylonian-Assyrian belong, and (2) Southern Arabic, Himyaritic, Ethiopic with their various dialects. Other scholars prefer a division into eastern and western. See on these divisions Brockelmann's Grundriss der vergleichenden Grammatik der Semitischen Sprachen (Leipzig, 1908-1913), I, p. 5, seq.

[14]:

Published by the Royal Asiatic Society in 1857. The text chosen was an inscription of Tiglath-pileser I, King of Assyria, who ruled c. 1130-1100 B.C.

[15]:

Renan voiced his doubts in an elaborate criticism of Oppert's Expedition scientifique en Mesopotamie, published in the Journal des Savants for 1859, pp. 165-186 ; 244-260 ; 360-68 ; Gutschmid in Neue Beitrage zur Oeschichte des alien Orients (Leipzig, 1876).

[16]:

On the Language and the Mode of Writing of the Ancient Assyrians (Transactions of the twentieth meeting of the British Association for the Advancement of Science, p. 140, seq.).

[17]:

Above, p. 92, seg.

[18]:

Observations sur l'origine des Chaldeens, in the Comptes-Eendus de la Societe franchise de Numismatique et d'Archeologie, I, pp. 73-76.

[19]:

Sumer is represented in the Old Testament as Shinar, e.g., in Gen. 11, 2, where mankind is described as congregated in the "valley of Shinar"; in Gen. 14, 1, Amraphel, who is Hammurapi, is designated as "King of Shinar". See above, p. 4, note 3.

[20]:

Above, p. 39 scq.

[21]:

It is not possible to present more than a bare outline of Halevy's thesis, which has many ramifications. He has written voluminously and always with critical acumen on the subject. For details the reader is referred to Halevy's articles in the Revue Semitique edited by him. An epitome of his theory will be found in his recent work, Precis d'Allographie Assyro-Babylonienne (Paris, 1912). A summary of the controversy up to 1898 will be found in F. H. Weissbach's Die Sumerische Frage (Leipzig, 1898).

[22]:

Sumerier und 8 emit en in Babylonien (Berlin, 1906).

[23]:

We owe largely to F. Thureau-Dangin the progress made during the past decade in the interpretation of these texts. See especially this author's Les Inscriptions de Sumer et d'Akkad(P&ris, 1905) ; also in German translation, Die Sumerischen und Akkadischen Konigsinschriften (Leipzig, 1907). See now, for an exposition of Sumerian grammar, Delitzsch's Qrundzuge der Sumerischen Grammatik (Leipzig, 1914) and Dr. Arno Poebel's volume of Sumerian grammatical texts in the publication above referred to (page 46) and which represents a further advance on Delitzsch's investigations.

Like what you read? Consider supporting this website: