We are searching data for your request:
Upon completion, a link will appear to access the found materials.
The Epic of Gilgamish A Fragment of the Gilgamish Legend in Old-Babylonian Cuneiform
In the year 1914 the University Museum secured by purchase a large six column tablet nearly complete, carrying originally, according to the scribal note, 240 lines of text. The contents supply the South Babylonian version of the second book of the epic &scarona nagba imuru, &ldquoHe who has seen all things,&rdquo commonly referred to as the Epic of Gilgamish. The tablet is said to have been found at Senkere, ancient Larsa near Warka, modern Arabic name for and vulgar descendant of the ancient name Uruk, the Biblical Erech mentioned in Genesis X. 10. This fact makes the new text the more interesting since the legend of Gilgamish is said to have originated at Erech and the hero in fact figures as one of the prehistoric Sumerian rulers of that ancient city. The dynastic list preserved on a Nippur tablet mentions him as the fifth king of a legendary line of rulers at Erech, who succeeded the dynasty of Kish, a city in North Babylonia near the more famous but more recent city Babylon. The list at Erech contains the names of two well known Sumerian deities, Lugalbanda and Tammuz. The reign of the former is given at 1,200 years and that of Tammuz at 100 years. Gilgamish ruled 126 years. We have to do here with a confusion of myth and history in which the real facts are disengaged only by conjecture.
The prehistoric Sumerian dynasties were all transformed into the realm of myth and legend. Nevertheless these rulers, although appearing in the pretentious nomenclature as gods, appear to have been real historic personages. The name Gilgamish was originally written dGi-bil-aga-mi&scaron, and means &ldquoThe fire god (Gibil) is a commander,&rdquo abbreviated to dGi-bil-ga-mi&scaron, and dGi(&scaron)-bil-ga-mi&scaron, a form which by full labialization of b to uÌ¯ was finally contracted to dGi-il-ga-mi&scaron. Throughout the new text the name is written with the abbreviation dGi(&scaron), whereas the standard Assyrian text has consistently the writing dGI&Scaron-á¹¬U-BAR. The latter method of writing the name is apparently cryptographic for dGi&scaron-bar-aga-(mi&scaron) the fire god Gibil has also the title Gi&scaron-bar.
A fragment of the South Babylonian version of the tenth book was published in 1902, a text from the period of Hammurapi, which showed that the Babylonian epic differed very much from the Assyrian in diction, but not in content. The new tablet, which belongs to the same period, also differs radically from the diction of the Ninevite text in the few lines where they duplicate each other. The first line of the new tablet corresponds to Tablet I, Col. V 25 of the Assyrian text, where Gilgamish begins to relate his dreams to his mother Ninsun.
The last line of Col. I corresponds to the Assyrian version Book I, Col. VI 29. From this point onward the new tablet takes up a hitherto unknown portion of the epic, henceforth to be assigned to the second book.
At the end of Book I in the Assyrian text and at the end of Col. I of Book II in the new text, the situation in the legend is as follows. The harlot halts outside the city of Erech with the enamoured Enkidu, while she relates to him the two dreams of the king, Gilgamish.
Cuneiform Tablet Listing the Names of Old Babylonian Kings - History
The use of sexagesimal numbers for astronomy by the Babylonians in the last centuries bc gave them a great advantage over contemporary Greek astronomers who had no convenient mathematical notation. As a result many Babylonian astronomical calcu- lations were used by the ancient Greek and medieval Arab astronomers long after knowledge of cuneiform writing was lost. Our present base 60 system of counting sixty seconds in a minute, sixty minutes in an hour, and three hundred and sixty degrees in a circle is a survival of Babylonian mathematics.
Click image for larger view .
There is a famous story of numerical manipulation in Assyrian history. Sennacherib had sacked Babylon in 689 bc. His son Esarhaddon on succeeding to the throne in 680 bc in a dramatic shift of policy decided to embark on its restoration, and justified it by announcing that whereas the god Marduk had decreed that the city should remain desolate for seventy years he had now relented and turned the number upside down. So seventy became eleven:
CUNEIFORM SCRIPT, the conventional name for a system of writing ultimately derived from the pictographic script developed by the Sumerians in southern Mesopotamia (Uruk) around 3000 B.C.E. Cuneiform was written with a reed stylus, which left wedge-shaped impressions on soft clay tablets the tablets were then dried in the sun or baked in a kiln. The term &ldquocuneiform&rdquo was introduced by Thomas Hyde (1700), although he disputed the claim that the wedge-shaped symbols had functioned as a writing system Engelbert Kaempfer also used the term, in 1712.
Historical background. In the 3rd millennium B.C.E. cuneiform script had already spread far beyond the region of its origin and was widely used in the Near East. Rather early the signs lost their original picto­graphic character and became abstract vertical, hori­zontal, and oblique lines continued to be used, but there were no more curves. Cuneiform had been invented as a system of word signs (logograms), but in the course of time a phonic system was developed, in which syllabic signs denoted vowels (a, e, i, u) and vowel-consonant combinations of various kinds (CV [consonant-vowel]: ba, bi, etc. VC: ab, ib, etc. CVC: tam, tim, etc.) these signs were sometimes combined with logograms that had remained unchanged and also with a number of so-called &ldquodeterminatives,&rdquo which functioned mainly as markers for noun classes. The signs were written from left to right, usually without separation between words. This writing system was adopted by the Akkadians, that is, the Babylonians and later the Assyrians, and transmitted to a great many neighboring peoples: the Elamites, the Lullubi, the Eblaites, the Kassites, the Hurrians, the Hittites, the Luwians, the Urartians, and others. By the Late Bronze Age (q.v.) many peoples of the Near East, between the Aegean Sea and the Zagros mountains and between the Black Sea and Egypt, used cuneiform. Through simplification and conventionalization the number of signs, originally about 2,000, was eventually reduced to about 800. The Babylonians used only about 570 signs, fewer still with any regularity. Such reduction did, however, lead to ambiguity or polyphony of many signs. Following the expansion of Aramaic-speaking peoples into Mesopotamia, beginning in the 7th cen­tury B.C.E., cuneiform was gradually displaced by the Aramaic consonantal script.
The classical Greek authors (such as Herodotus and Ctesias) knew little more of cuneiform than the mere fact that a distinctive system of writing was in use in the Achaemenid empire they called it &ldquoAssyrian letters&rdquo (Assyria [or Syria or Persikà] grámmata), by which term, however, they also sometimes referred to the Aramaic script (e.g., Thucydides, 4.50.2). The latest known texts written in cuneiform are from the 1st century C.E. then it sank into oblivion, and its exist­ence became known in Europe only in the 17th century through travelers like the Italian Pietro della Valle, who had seen it at Persepolis in 1621. A great many scholars and amateurs worked on deciphering cunei­form script(s) during the first half of the 19th century the most prominent among them were Georg Friedrich Grotefend, Rasmus Rask (see below), Henry Creswicke Rawlinson (whose bilingual method opened the way for deciphering the Elamite and Babylonian versions of the inscriptions at Bīsotūn), Edwin Norris, Edward Hincks, and Jules Oppert.
Cuneiform scripts on Persian territory. Rather than a general survey of cuneiform scripts, an outline of the evidence for cuneiform writing on Persian territory, mainly western Persia, will suffice here. Particularly important are two Akkadian inscriptions of Anubanini, king of the Lullubi tribes (20th century B.C.E., slightly later than the Third Dynasty of Ur see Edzard, 1973, p. 75a), on the rock at Sar-e Pol-e Ḏohāb near the so-­called Gates of Asia in the Zagros mountains of Kurdistan. The first of these inscriptions is located next to a relief depicting the king in a triumphant pose facing the goddess Inanna the text is far from com­plete and does not permit identification of the event depicted. Still less can be said about the second inscription, which was found about 200 m away and shows traces of the same royal name it must thus be almost contemporary (Edzard, 1973).
Several other approximately contemporary Akkadian inscriptions in the Zagros are known, for example, one accompanying a rock relief near Ḵūrīn &Scaronayḵ Ḵān (northwest of Sar-e Pol, near the Iraqi border), which is somewhat provincial in character (Farber). Other­wise, inscriptional evidence for cuneiform writing among the Lullubi, Gutian, Kassite, and other popula­tions of pre-Achaemenid Iran is very scanty (except for the Elamites see below). Several dozen of the many &ldquobronzes of Luristan&rdquo bear cuneiform inscriptions these objects range in date from the 3rd millennium to the 7th century B.C.E., though most are concentrated in the 10th century B.C.E., for example, a Babylonian inscription of a certain &Scaronilisruḫ (?) on both sides of a bronze plaque found near Hamadān (Diakonoff). The texts, some of which are undoubt­edly from western Persia, range from owners&rsquo marks to votive inscriptions, but the exact circumstances in which they were inscribed are not indicated (see Calmeyer, pp. 161-74).
Within the broad range of cuneiform cultures Elam constitutes a distinct province, for there the essentially pictographic-logographic system of Proto-Elamite and Linear Elamite can be traced back at least to 3000 B.C.E. (see elam iv, v), whereas Akkadian cuneiform is attested only from the 23rd century B.C.E. the earliest text in the latter is the so-called Treaty of Narâm-Sin. During a short period in which both writing systems existed side by side, cuneiform gradu­ally supplanted the older Elamite script, though it was not well suited to the Elamite language. After about 2200 B.C.E. it was the only script used by the Elamites.
Although only a few texts survive from the Old Elamite period (23rd-15th century B.C.E.), there is a rich array of inscriptional evidence (building inscrip­tions, dedications, etc.) of the Elamite kings between the 13th and 11th centuries B.C.E. (the Middle Elamite period). Inscriptions of Unta&scaron-Napiri&scarona and &Scaronilhak-­In&scaronu&scaroninak I from Susa and several other sites (includ­ing Liyan [Līān] and Čoḡā Zanbīl) constitute the most reliable basis for grammatical analysis of the Elamite language (see elam vi). After an interval of about 400 years, from which no record survives in Elamite script and language, inscriptions are once more plentiful from the Neo-Elamite period (8th-early 6th centuries B.C.E.), including specimens of eco­nomic and literary character (e.g., a tablet with astrological omens). Rock reliefs and inscriptions of a local ruler called Hanni from the Mālamīr plain (at Kūl-e Farah and &Scaronekaft-e Salmān) east of Susa show that knowledge of cuneiform writing had in the meantime spread far from the urban centers. The unbroken Elamite language tradition of Susa and Elam was continued with the inclusion of Elamite cuneiform in the mostly trilingual Achaemenid royal inscriptions (where it takes second place to Old Persian) and the administrative tablets of Persepolis (Royal Achaemenid Elamite period, 6th-4th centuries B.C.E).
When borrowing cuneiform script from the Akkadians, the Elamites made a selection from extant signs but did not immediately change their forms and values. In the course of time, however, some modifi­cation of the cuneiform syllabary did occur, so that a kind of provincial scribal tradition developed (Steve, 1992). Special traits of Elamite cuneiform script (beginning in the Middle Elamite period) are the rela­tively small number of determinatives (used only for gods, personal names [including personal pronouns], place names, and wooden objects) and word signs (logograms) and the use of the Akkadian plural marker ME&Scaron as a postpositional indicator of logograms (which are sometimes simply Elamite words or abbrevia­tions). By the Middle Elamite and especially the Achaemenid periods a completely differentiated and considerably simplified variant of Akkadian syllabic cuneiform had been developed it was characterized by an almost complete absence of polyphony or ho­mophony, only a small number of complex syllabic signs of the CVC type, and so-called &ldquobroken writing&rdquo (e.g., -nu-i&scaron- instead of -nu-u&scaron- for/-nus-/). A list of the signs used in Royal Achaemenid Elamite has been provided by R. T. Hallock (1969, pp. 82-86).
Urartu, which included parts of modern Persia (west­ern Azerbaijan), also constituted a distinct province among cuneiform cultures. Writing seems to have been unknown there until the 9th century b.c.e. (whether or not the so-called &ldquoUrartian hieroglyphics&rdquo found mainly on vessels and objects functioned as a system of writing is still in dispute). Only under King Sardure I were the Assyrian language and cuneiform script introduced for inscriptions. Under his son I&scaronpuini (ca. 830-20 b.c.e.) the first texts in the Urartian language began to appear they were also written in a variant of Neo-Assyrian cuneiform. One of the most important texts is on a stele from the Kal-e &Scaronīn pass (on the border between Persia and Iraq southwest of O&scaronnūya), a bilingual inscription in Assyrian and Urartian referring to the joint foundation of a temple by King I&scaronpuini and his son King Minua. As there are close parallels with the inscriptions, formulas, titles, epithets, and the like of the Neo-Assyrian kings and especially of A&scaron&scaronunarṣirpal II, it must be concluded that the Urartian writing system came directly from Assyria, perhaps from a northern provincial town (Wilhelm). One departure from Assyrian usage is the scribes&rsquo tendency (which recurs in Old Persian see below) to avoid crossing a horizontal and a vertical wedge.
The surviving Urartian inscriptions are almost exclusively monumental those carved on carefully smoothed rock faces range in date from the reign of King Minua to the fall of the empire. The texts, which are often formulaic and repetitive, include not only building inscriptions but also foundation deeds and war records at the height of the empire, under Argi&scaronti I and Sardure II, these annals were very extensive, containing hundreds of lines.
The only evidence that writing was known to the Medes is found on an inscribed silver fragment from a hoard excavated at Tepe Nū&scaron-e Jān (Brinkman) only parts of two cuneiform signs can be recognized, however, so that it is impossible to identify the species of cuneiform script or even the language of the document. It can be assumed by analogy with other known cul­tural links between Assyria and Media that Assyrian cuneiform script had been adopted by the Medes, an assumption that conforms to the evidence about the origin of the Old Persian cuneiform created and used by the Achaemenid kings (Sancisi-Weerdenburg, pp. 213-14). Some phonological differences between the Old Persian and Babylonian versions of the Bīsotūn inscription also suggest that the latter is closer to the Median dialect and was perhaps even copied by Me­dian scribes.
Old Persian cuneiform script. The term &ldquocuneiform&rdquo in its broader sense also encompasses scripts that cannot be regarded as continuing the rather complex Akkadian cuneiform tradition but instead resemble it only superficially, owing to the wedge and angle forms of the single elements in the signs. One of these &ldquodifferent&rdquo cuneiform scripts is that in which Old Persian was written, a simplified version invented in the 6th century B.C.E. It was the official script adopted by the Achaemenid kings (from Darius I to Artaxerxes III) for writing their mother tongue, which was essentially the southwestern Iranian dialect of Persis (modern Fārs) but in its attested form also shows some characteristics (foreign words, archaisms, etc.) of an artificial literary language. Old Persian cuneiform (as well as the language itself) was &ldquoconfined to royal prestige purposes&rdquo (Gershevitch, p. 122), particularly monumental inscriptions (mostly trilingual in Old Per­sian, Elamite, and Babylonian), which in large part could not even have been intended to be read, for they were either engraved too high on rock faces or encased in foundation walls. It was a &ldquosplendid&rdquo script suitable only for hard surfaces (stone, metal, and occasionally clay tablets but not parchment, papyrus, etc.) and limited almost entirely to the central lands of the empire: Persis, Elam, and Media. It was thus clearly not intended for use in everyday life. There is not yet a complete corpus of surviving inscriptions in Old Persian language and cuneiform script (for the present, see Kent, Old Persian, pp. 107-57 Mayrhofer, 1978, esp. pp. 37-47 and Schmitt, 1989, p. 58 par. 184.108.40.206 for a new edition of the Bīsotūn text, see Schmitt, 1991).
Old Persian cuneiform was not a continuation of the Mesopotamian system (with the exception of the sign for l see below) but was, on the contrary, an independent creation, resembling Aramaic in that it reflects a tendency to equate one sign with one sound. There is a total of thirty-six phonic signs, which may be classed in four groups: three pure vowel signs (a, i, u) twenty­-two neutral consonant signs, that is, either with no inherent vowel (occurring only before consonants or at the ends of words) or with inherent a: b a , c a , ç a , d a , j a , g a , h a , j a , k a , l a , m a , n a , p a , r a , s a , &scaron a , t a , &theta a , v a , x a , y a , z a four with inherent i: d i , j i , m i , v i and seven with inherent u: d u , g u , k u , m u , n u , r u , t u . There are also eight logograms, which are not obligatory and not used consistently, two word dividers (in the form of oblique wedges), and several numerals (for the full list, see Kent, Old Persian, p. 215). The phonic signs consist of a maximum of five elements each (horizontal or vertical wedges and angles, called in German Winkelhaken), which never cross (including v i , against earlier opinions see Hoffmann, p. 621). The logograms, which denote &ldquoking,&rdquo &ldquogod,&rdquo &ldquoland,&rdquo &ldquoearth,&rdquo and Ahura Mazdā re­spectively, have much more complex shapes, how­ever, with up to twelve elements and even horizontal wedges or angles placed above angles. For a synopsis of attested signs (excluding numerals), see Figure 23.
Only in a few instances can development in the shapes of single signs be observed: In the Bīsotūn text, which is the first known Old Persian inscription, the oblique wedge (not an angle) functioning as word divider (see Hinz, 1973, p. 24) is only half the height of the line, whereas in all other inscriptions it stands the full height. Similarly, the first (vertical) wedge of the y sign is only half the height of the line at Bīsotūn, in contrast to later usage. The logogram for &ldquoland&rdquo is normally DH1, whereas DH2 is restricted to a few instances in later inscriptions.
The decipherment of Old Persian cuneiform, which had to be based on the texts themselves, became possible only after 1778, when Carsten Niebuhr pub­lished for the first time more precise copies of some Achaemenid inscriptions (mainly from Persepolis pp. 139, 158, pls. 23, 24, 31). The process has often been described, in varying degrees of detail (e.g., Kent, Old Persian, pp. 10-11 Weissbach Borger, 1975-78 see also above). Niebuhr recognized at once that the script was written from left to right. In 1798 O. G. Tychsen identified the oblique wedge as the word divider (pp. 24-25), and in 1802 Friedrich Münter attributed these texts to the Achaemenid kings (pp. 124ff.). The actual decipherment of the script was initiated later that same year by Grotefend, who succeeded in ascertaining the approximate phonic values of about ten signs (see Meyer). He assumed that the inscriptions had been sponsored by Persians and that, following the pattern of Sasanian royal inscriptions, which had first been deciphered only a few years earlier, the names, titles, and genealogies of some of the Achaemenid kings would be mentioned in them. On the sound founda­tions laid by Grotefend other scholars have built step by step. Rask (1826, pp. 27ff.) identified the n and m signs in the genitive plural ending -ānām (correspond­ing to Av. -aną&hellipm) in 1836 Eugène Burnouf and Christian Lassen made systematic comparisons with Avestan (q.v.), which had by then become better known and had turned out to be closely related to Old Persian and in 1845 Lassen recognized that the consonant signs have inherent vowel components, as in the old writing systems of India. The process was almost completed when Rawlinson, in 1846-47, published, translated, and interpreted the entire text of the large Bīsotūn inscription. Finally, in 1851, Oppert deci­phered the last of the phonic signs, the l sign (1851, p. 76), which is attested only in four foreign names, for example, l-b-n-a-n- = Lab(a)nāna- &ldquoLebanon&rdquo (the phoneme /l/ being nonexistent in Old Persian proper) this sign seems to be the only one borrowed from the contemporary Elamite or Neo-Babylonian cuneiform syllabary (Paper).
Old Persian cuneiform is neither phonemic nor pho­netic, as is apparent from the ambiguity of most signs (e.g., the entire second group: b or b a , etc.), as well as from the inconsistent and asymmetrical structure of the inventory as a whole (e.g., d a , d i , d u , t a , and t u but not t i ). The thirty-six phonic signs could therefore not be used without certain &ldquoorthographic conventions&rdquo for rendering particular phonemic sequences. It is sufficient to mention only the most important of these conventions, those that are attested with certainty or linguistically probable.
First, long vowels are not distinguished from short vowels, except for /ā/ in medial position.
Second, proto-Iranian final short -ă is written -C a -a /-ā/ (and apparently lengthened in speech see Hoffmann, pp. 633-35).
Third, the vowels /i, u/ are written with the corresponding vowel signs and, in medial position, with an additional preceding C i , C u , when such signs are avail­able, or C a , when not.
Fourth, final /-ī/ĭ, -ūÂ¦/ are written with an additional semivowel -i-y, -u-v (traditionally considered to be purely graphic, but see Hoffmann, p. 635).
Fifth, short diphthongs are written -C a ,-i-, -C a -u-, so that they are only partly distinguishable from simple vowels (m a -i- = /maḭ-/ and m i -i- = /mī/ĭ-/ occur side by side, but, e.g., /taḭ-/ and /tī/ĭ-/ are written t a -i-).
Sixth, long diphthongs are written -C a -a-i-, -C a -a-u ­but are not distinguished from short diphthongs in initial position.
Seventh, /ṛ/, in all probability pronounced [ər], is written a-r a - = /ṛ-/ in initial position (and thus cannot be distinguished from /ā/ă-/) but C a -r a -C x = /CṛC/ in medial position.
Eighth, nasal consonants /m, n/ are written before consonants only in exceptional cases (e.g., /mn/ in k a -m a -n a - = /kamna-/).
Ninth, in contrast to final /-n/, final /-m/ is com­monly written (a-b a -r a -m a = /abaram/ &ldquoI brought&rdquo but a-b a -r a = /abaran/ &ldquothey brought&rdquo).
Tenth, the only written final consonants are -m, -r, and -&scaron.
Eleventh, postconsonantal *ḭ, *ṷ (proto-Iran. Cḭ, Cṷ) are regularly written C i/a -i-y, C u/a -u-v, often (and especially in personal, geographical, and other names) /Ciḭ, Cuṷ/ (and presumably pronounced Ciḭ, Cuṷ see Hoffmann, pp. 636-37).
Twelfth, Iranian h (from Indo-Iran. s) before Old Persian ūÂ¦, m, r was omitted in writing, as it was pronounced weakly or not at all in the underlying Old Persian dialect (also the group proto-Iran. hṷ is repre­sented as OPers. u-v a /uṷ/).
Finally, Old Persian /ī/ĭ/ is often omitted after h in writing.
From this survey it is clear that it is impossible to associate a single phoneme unequivocally with each sign. On the contrary, there are rather important deficiencies in the writing system, including the lack of graphic distinction between /tī/ĭ/ and /taḭ/ (so that, e.g., the verbal endings /-ti/ and /-taḭ/ of 3rd pers. sing. pres. act. and med.-pass. cannot be distinguished) and the omission of nasal consonants before consonants (so that, e.g., the verbal endings /-ti, -tu/ of 3rd pers. sing. and /-nti, -ntu/ of 3rd pers. plur. cannot be distinguished).
Altogether these orthographic conventions give rise to several possible interpretations of nearly every attested word, requiring in each instance philological and linguistic analysis or both to achieve a correct reading. In order to illustrate the problems, an oft­ repeated example may be cited. The expression a-s a -t a -i-y a &ldquohe/she/it is&rdquo can be read, according to established orthographic conventions, in seventy-two different ways: ā/ă¦(n)-s a(n) -t a -ī/ĭ-y a .&rdquo That the correct reading must be astiy = /asti/ can be ascertained only by comparing Avestan asti, Vedic ásti, Middle Persian and New Persian ast, and the like. A reading can be found when a word is cited in its Old Persian form in the Elamite or Akkadian versions (e.g., a-r-j-n-m &ldquoor­namentation,&rdquo compared with the Elamite ha-ra-an­-za-na-um, should be read as āranjanam, as already conjectured before the Elamite form was known). Many such difficulties remain unresolved, however, and perhaps will never be resolved. In using existing editions, grammars, manuals, and the like, it must be kept in mind that the readings of the words and gram­matical forms represent only interpretations of the highly ambiguous graphemes. In publishing inscrip­tions and discussing Old Persian words and grammar, both a graphemic transliteration and a phonemic tran­scription should be provided.
The origin of the Old Persian script. The genesis and introduction of Old Persian cuneiform are among the most controversial problems in Old Iranian studies since the 1960s they have been treated repeatedly and from different points of view, without achieving gen­eral agreement (see Schmitt, 1980, pp. 17-20). The following general outline is based on the testimony about the invention of the script given in par. 70 of Darius&rsquo major Bīsotūn inscription (DB 4.88-92), where a new style of writing &ldquoin Aryan&rdquo devised by the king is mentioned (for the full text, see bīsotūn iii) on various archeological and stylistic observations that permit the delineation of several stages in the genesis and development of the Bīsotūn monument on inscriptions supposed to predate Darius and, finally, on analysis of the writing system itself. It is likely that development of a new script in which to write the Achaemenids&rsquo mother tongue had already begun in the reign of Cyrus II, but the oldest attested ex­amples of the new type of cuneiform are the major and minor inscriptions of Darius I at Bīsotūn. Claims that some surviving inscriptions are older can be con­clusively refuted: Two inscriptions on gold tablets from Hamadān composed in the names of Ariaramnes (AmH) and Arsames (AsH), respectively great-grand­father and grandfather of Darius I, appear from their rather faulty language, which resembles that of the latest Achaemenid texts, not to be authentic. Furthermore, two small fragments from the Pasargadae inscriptions (CMb, CMc), which have been assigned to Cyrus II, appear in fact to belong to Darius (Mayrhofer, 1978, pp. 11-13), whereas a third (CMa) was written first in Elamite and Babylonian only, to which an Old Persian translation was added later (in the time of Darius) on uninscribed portions of the surface (Nylander).
Nevertheless, the conclusion that Old Persian cunei­form began to evolve under Cyrus finds some support in the observations and arguments of Hallock (1970), who focused on the signs k u and r u , which are necessary in writing the name Kuru&scaron: These signs must have been among the oldest, for their shapes are quite simple (with two and three wedges respectively), even though the frequency of the phonemic sequences /ku/ and /ru/ is rather low. That there must have been a chronological gap between invention of the script and its actual use is confirmed by some striking features, which have been brought out most clearly by Karl Hoffmann (1976), who also tried to explain them. The inconsistent structure of the inventory of signs seems best explained by assuming that originally the aim was a consistent and unambiguous system of marking vow­els and diphthongs by means of three signs planned for each consonantal set plus the pure vowel signs (e.g., *da = /da/, *da-a = dā/, *da-i = /daḭ, *da-a-i = /dāj/, *da-u /daṷ/, *da-a-u = /dāṷ/, *di = /di/, *di-i = /dī/, *du = /du/, *du-u = /dū/) but that at some point a less clear system triumphed instead. This system includes many ambiguities (see above), especially as most consonantal sets contain not three but two or even single signs. The decisive factor in discarding the original concept must have been a desire for simplicity, as is attested by such formal and stylistic features as a tendency to avoid complex signs of more than five elements and signs involving crossed wedges. Indeed, the latter feature, along with the preference for rock inscriptions, the royal title &ldquoking of kings&rdquo (OPers. x&scaronāya&thetaiya x&scaronāya&thetaiyānām), and several specific formu­las (va&scaronnā Auramazdāha &ldquoby the favor of Ahura Mazdā,&rdquo &thetaātiy NN x&scaronāya&thetaiya &ldquosays NN, the king&rdquo), may be ascribed to Urartian influence (see above). Only an assumption that the script was introduced in haste can explain this astonishing concentration on ease in writing instead of on ease and clarity in reading.
The decisive argument for the introduction of the script in connection with the inscription beneath the Bīsotūn relief undoubtedly lies in the history and the genesis of that monument itself, for the Old Persian inscriptions accompanying the single figures in the relief and in the major inscription are later additions to the original design of the monument (see, e.g., Hinz, 1976, pp. 21-37 bīsotuᵛn iii). The initial adoption of only pre-Achaemenid languages must be interpreted as evidence that Old Persian had not yet come into use for written records. Old Persian cuneiform was obvi­ously created specifically for writing the Old Persian language, rather than for some other Old Iranian dia­lect like Median (Hoffmann, pp. 620-21): The shape of the ç sign is quite simple and was thus probably among the initial signs but represents the phoneme (< OIran. *&thetar, phonetically close to [s]), which is characteristic, as far as is known, only of Old Persian and is foreign to the Median phonological system, which retained &thetar.
R. Borger, Assyrisch-babylonische Zeichenliste, Kevelaer and Neukirchen-Vluyn, 1978.
Idem, &ldquoDie Entzifferungsgeschichte der altpersischen Keilschrift nach Grotefends ersten Erfolgen,&rdquo Persica 7, 1975-78, pp. 1-5.
J. A. Brinkman, apud A. D. H. Bivar, &ldquoA Hoard of Ingot-Currency of the Median Period from Nūsh-i Jān, near Malayir,&rdquo Iran 9, 1971, pp. 97-111, esp. pp. 102, 107.
E. Burnouf, Mémoires sur deux inscriptions cunéiformes, Paris, 1836.
P. Calmeyer, Datierbare Bronzen aus Luristan und Kirmanshah, Berlin, 1969.
I. M. Diakonoff, &ldquoA Cuneiform Charter from Western Iran,&rdquo Festschrift Lubor Matou&scaron I, Budapest, 1978, pp. 51-68.
D. O. Edzard, &ldquoZwei Inschriften am Felsen von Sar-i-Pūl­i-Zohāb Anubanini 1 und 2,&rdquo Archiv für Orientforschung 24, 1973, pp. 73-77.
Idem, &ldquoKeilschrift,&rdquo RIA V, 1980, pp. 544-68 (especially on technique, origin of the signs, and paleography).
W. Farber, &ldquoZur Datierung der Felsinschrift von &Scaronaiḫ-ḫān,&rdquo AMI, N.F. 8, 1975, pp. 47-50.
J. Friedrich, Entzifferung verschollener Schriften und Sprachen, 2nd ed., Ber­lin, 1966, esp. pp. 44-57.
I. Gershevitch, &ldquoThe Alloglottography of Old Persian,&rdquo TPS, 1979, pp. 114-90.
R. T. Hallock, Persepolis Fortification Tablets, Chicago, 1969.
Idem, &ldquoOn the Old Persian Signs,&rdquo JNES 29, 1970, pp. 52-55.
E. Hincks, &ldquoOn the First and Second Kinds of Pesepolitan Writing,&rdquo Transac­tions of the Royal Irish Academy 21/2, 1848, pp. 114­-31 tr. as &ldquoUeber die erste und zweite Gattung der Persepolitanischen Schrift,&rdquo Zeitschrift für die Kunde des Morgenlandes (Bonn) 7, 1850, pp. 201ff.
W. Hinz, Neue Wege im Altpersischen, Wiesbaden, 1973.
Idem, Darius und die Perser. Eine Kulturgeschichte der Achämeniden I, Baden-Baden, 1976.
Idem and H. Koch, Elamisches Wörterbuch, 2 parts, Berlin, 1987 (pp. 1332-68, for a full bibliography).
K. Hoffmann, &ldquoZur altpersischen Schrift,&rdquo in Aufsätze zur Indoiranistik II, Wiesbaden, 1976, pp. 620-45.
T. Hyde, Veterum Persarum et Parthorum et Medorum Religionis Historia, Oxford, 1700 2nd ed., Oxford, 1760.
E. Kaempfer, Am&oelignitatum Exoticarum Politico-Physico-Medicarum Fasciculi v., Lemgo, 1712.
C. Lassen, Die altpersischen Keil­inschriften von Persepolis. Entzifferung des Alpha­bets und Erklärung des Inhalts . . . , Bonn, 1836.
Idem, &ldquoDie altpersischen Keilinschriften nach Hrn. N. L. Westergaard&rsquos Mittheilungen,&rdquo Zeitschrift für die Kunde des Morgenlandes 6, 1845, pp. 1-188, 467-580.
M. Mayrhofer, Supplement zur Sammlung der altpersischen Inschriften, Vienna, 1978.
Idem, &ldquoÜberlegungen zur Entstehung der altpersischen Keilschrift,&rdquo BSOAS 42, 1979, pp. 290-96.
Idem, &ldquoÜber die Verschriftung des Altpersischen,&rdquo ZVS 102, 1989, pp. 174-86.
W. Meyer, &ldquoG. Fr. Grotefend&rsquos erste Nachricht von seiner Entzifferung der Keilschrift,&rdquo Nachrichten der königlichen Gesell­schaft der Wissenschaften in Göttingen 14, 1893, pp. 571-616 separate ed., Darmstadt, 1972.
F. Münter, Versuch über die keilförmigen Inschriften zu Persepolis, Copenhagen, 1802.
C. Niebuhr, Reisebeschreibung nach Arabien und andern umliegenden Ländern II, Copenhagen, 1778.
E. Norris, &ldquoMemoir on the Scythic Version of the Behistun Inscription,&rdquo JRAS 15, 1851, pp. 1-213.
C. Nylander, &ldquoWho Wrote the Inscriptions at Pasar­gadae?&rdquo Orientalia Suecana 16, 1967, pp. 135-80.
J. Oppert, Études sur les inscriptions des Achéménides conçues dans l&rsquoidiome des anciens Perses, Paris, 1851 (from JA).
H. H. Paper, &ldquoThe Old Persian /L/ Phoneme,&rdquo JAOS 76, 1956, pp. 24-26.
M. Pope, The Story of Decipherment. From Egyptian Hieroglyphic to Linear B, London, 1975, esp. pp. 85-122.
R. Rask, On the Age and Genuineness of the Zend Language and the Zendavesta, Madras, 1821 tr. R. Rask as Om Zendsprogets og Zendavestas Ælde of Ægthed, Copenhagen, 1826 tr. F. H. von der Hagen as Über das Alter und die Echtheit der Zend-Sprache und des Zend-Avesta, und Herstellung des Zend-Alpha­bets . . . , Berlin, 1826.
H. C. Rawlinson, The Persian Cuneiform Inscription at Behistun, Decyphered and Translated, JRAS 10, 1846-47.
E. Reiner, &ldquoThe Elamite Language,&rdquo in HO I/II/I-II/2. Altklein­asiatische Sprachen, pp. 54-118.
H. Sancisi­-Weerdenburg, &ldquoMeden en Perzen,&rdquo Lampas 12, 1979, pp. 208-22.
R. Schmitt, &ldquoAltpersisch-Forschung in den Siebzigerjahren,&rdquo Kratylos 25, 1980, pp. 1-66.
Idem, &ldquoDänische Forscher bei der Erschliessung der Achaimeniden-Inschriften,&rdquo Acta Orientalia 47, 1986, pp. 13-26.
Idem, &ldquoAltpersisch,&rdquo in R. Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989, pp. 56-85.
Idem, ed., The Bisitun Inscriptions of Darius the Great. Old Persian Text, Corpus Inscr. Iran., London, 1991.
M.-J. Steve, Syllabaire élamite. Histoire et paléographie, Neuchâtel and Paris, 1992.
O. G. Tychsen, De Cuneatis Inscriptionibus Persepolitanis Lucubratio, Rostock, 1798.
F. H. Weissbach, &ldquoDie altpersischen Inschriften. III. Geschichte der Entzifferung und Erklärung der Inschriften,&rdquo in Grundriss II, pp. 64-74.
Die Welt des Alten Orients. Keilschrift&mdashGrabungen&mdashGelehrte, Göttingen, 1975, esp. pp. 15-18 (W. Hinz), 155-184 (R. Borger).
G. Wilhelm, &ldquoUrartu als Region der Keilschrift-Kultur,&rdquo in Das Reich Urartu, ed. V. Haas, Constance, 1986, pp. 95-116.
Figure 23. Attested cuneiform signs in Old Persian. After R. Schmitt, ed., Compendium Linguarum Iranicarum, Wiesbaden, 1989, p. 63.
In order to generate our datasets, we collected transliterated texts from the Achemenet website, based on data prepared by F. Joannès and coworkers in the framework of the Achemenet Program (National Center for Scientific Research [CNRS], Nanterre, France) (http://www.achemenet.com/fr/tree/?/sources-textuelles/textes-par-langues-et-ecritures/babylonien). We designed a tokenization method for Akkadian transliterations, as detailed in Materials and Methods. We trained a LSTM recurrent network and a n-gram baseline model on this dataset (see Datasets S1–S3 for model and training details).
Results for both models are in Table 1. Loss refers to mean negative log-likelihood and perplexity is two to the power of the entropy (in both cases, lower is better).
Loss and perplexity while training the model on Achemenet dataset
As expected, the RNN greatly outperforms the n-gram baseline, and despite the limitations of the dataset, it does not suffer from severe over-fitting.
Completing Random Missing Tokens.
In order to evaluate our models’ ability to complete missing tokens, we took random sentences from the test corpus, removed the middle token and tried to predict it using the rest of the sentence. Our model returns a ranking of probable tokens and we report the mean reciprocal rank (MRR). The MRR is the average over the dataset of the reciprocal of the predicted rank of the correct token. It is a very common and useful measure for information retrieval as it is highly biased toward the top ranks, which is what the user is mostly interested in. We also evaluate the “[email protected],” which measures the percentage of sentences where the correct completion is in the top k suggestions. For evaluation, we used all test sentences 10 or more tokens in length that contain no breaks, which yielded a total of 520 sentences.
We compared two variations of our model, one that finds the optimal completion based only on the tokens that precede the missing token, denoted “LSTM (start),” and one that takes the full sentence into account, denoted “LSTM (full).” As the “LSTM (full)” model needs to run separately for each candidate for the missing token, we first picked the top 100 candidates using “LSTM (start).” We then generated 100 sentences, one for each possible completion, and reranked them based on the full sentence log-likelihood. If the right completion was not in the top 100, we took the reciprocal rank to be zero.
For comparison, we used two simple 2-gram baselines: one that takes into account only the previous token, denoted “2-Gram (start),” and one that takes into account both the previous and the next token denoted “2-Gram (full).” While this is a relatively weak model, we found it to work surprisingly well, although it was still significantly inferior to the LSTM model in the accuracy ([email protected]) metric.
To further investigate our model’s ability to complete various numbers of missing tokens in various locations, we removed up to three tokens in random locations. We ranked possible completions using our model and beam search and show the results in Table 3.
It is clear from the results in Tables 2 and 3 that our algorithm can be of great help in completing a missing token, with an almost 85% chance of completing the token correctly and a 94% chance of including the correct token in the top 10 suggestions. However, as expected, the task becomes much harder and performance is degraded when more tokens are missing. We note that even with two or three missing tokens, however, the model is still useful as the correct completion is present in the top 1 (two missing) or 10 (three missing) completions almost half of the time.
Completing missing fifth token in sentences
Completing various number of tokens
Designed Completion Test.
We designed another experiment in order to evaluate our completion algorithm and understand its strengths and weaknesses. We generated a set of 52 multiple choice questions in which the model is presented with a sentence missing one word and four possible completions, and the goal was to select the correct one. Of the three wrong answers, the first was designed to be wrong semantically, the second wrong syntactically, and the third both. This allowed us to track the types of mistakes the algorithm makes. The assumption is that the learning algorithm would be more likely than a human to make semantic mistakes but should be better than a nonexpert in grammar. If this is the case, then the effectiveness of our approach as a way to assist humans should rise, as the strengths of human and machine complement each other.
When we used our model to rank four possible restorations for each of the missing words in the 52 random sentences, it achieved 88.5% accuracy in selecting the one with the highest likelihood (see Dataset S4 for the complete list of questions and answers). Looking at the six failed completions—questions 18, 26, 32, 35, 45, and 50—we see that four are semantically incorrect, one is syntactically incorrect, and one is both, which agrees with our hypothesis.
The Assyrian and Babylonian Chronicles are historiographical texts from ancient Mesopotamia. Although they contain references to the earliest times, they deal especially with the second half of the second and the entire first millennium down to the first century BCE.
Cuneiform texts look complex and seem hard to read, and, frankly, they are complex and are hard to read. Yet, there are degrees of complexity and even a layman can make sense of a cuneiform text. For example, the Persian script is alphabetic and often used in clearly legible rock inscriptions. Because the texts are highly stereotypical, you can start to recognize the names of kings fairly soon.
Reading the Assyrian-Babylonian cuneiform characters, however, is a difficult job, even to specialists, and both the layman and the professional scholar have to settle for a critical edition made by someone who has meticulously studied the tablet. In fact, the same holds for Greek and Latin texts. Only a few classicists actually study the medieval manuscripts.
The difference is that cuneiform writing is not alphabetical but a mixture of ideograms (one sign is one word) and syllabic. If a tablet is only slightly damaged, complete words are illegible, and a surprisingly large part of modern scholarly literature is devoted to simple questions as "what is this or that sign?" (The advantage of the system is that small tablets can contain large texts.)
To make things worse, Babylonian cuneiform is based on an older system, Sumerian. One part of the inheritance is the use of Sumerian signs to indicate well-known words. For example, the word for king could be written with two signs, shar-ru, but in Sumerian cuneiform, only one sign is needed to write lugal. This is easier to write, and not incomparable to English abbreviations (e.g., AD, PS, NB) which we hardly recognize as renderings of Latin words. In a critical edition, sumerograms are indicated by CAPITALS.
This is, as noted, easier to write, and an experienced reader will not have found it difficult. To us, it is rather confusing. Now let's make things more complex and look at the signs to write variants of the word "king". For example, kingship, sharrutu, can be spelled as shar-ru-tu, but it was more common to write LUGAL-u 2 -tu: first, a sumerogram, then two signs to describe u and tu. (See below for the small 2.) Now, the advantage has gone: a word with three syllables needs three characters, and the use of a sumerogram is just the result of oversystematization (or pedantry).
Unfortunately, the same signs are used in different ways, as ideograms and as syllables. For example, KU can be read as TUKUL = sumerogram for "weapon", or syllabic as ku in Si-lu-ku = Seleucus. There is a damaged tablet that contains the signs
GAZ means "to kill". If we reconstruct Si-lu in the lacuna and read ku as a syllable, we have "He killed Seleucus". Alternatively, we can reconstruct "ina TUKUL GAZ": "He killed with a weapon". This makes Babylonian cuneiform an entertaining puzzle. It must be noted that this is not without parallel in our own writing system think of the song by The Artist Formerly Known As Prince, Nothing compares 2 u. (Just imagine Hamlet saying 2B or not 2B.)
Add to this that there are about 600 signs, and things become really complex. Babylonian scribes found the system (or lack of system) puzzling too, and introduced determinatives to make reading easier. (The same happened in Egyptian hieroglyphic.) In modern editions, they are indicated in superscript . For example:
- LU 2 ("human"): to indicate people, occupations, nations
- an added vertical wedge indicates a man (in modern editions, this is translated with an m )
- the sign SAL indicates a woman (in modern editions an f )
Yet, this is not entirely without problems, because a vertical wedge is not only the determinative "man", but also the sign for the number one and the word ana, "to".
Subscript is used by modern scholars to indicate differences between sounds that may once have been distinct, but had later become almost identical. (Cf., in the Roman age, ancient Greek had several signs to describe the i, even though e, ê, ei and i had once indicated distinct sounds.) In Babylonian and Assyrian, there are several u-like sounds, indicated like u, u 2 , u 3 , u 4 (or, often, like u, ú, ù, u 4 ). Although these signs indicate almost identical vowels, they are employed in specific contexts. Only u and ù can be used to describe our word "and" ú is only used to lengthen verbs u 4 is the only sign to spell ud, "day".
Because of the variant spellings of more or less identical sounds, an easy transcription system cannot be created. It may be more correct to spell a word like á-ki-ti-še-gur 10 -ku 5, but Akitu (the name of a festival) is a lot easier.
One final remark: the letter -m- was only written, never pronounced.
So, it is not really easy - yet, it is not entirely without system either, and you can make sense of it without too much knowledge of the Babylonian or Assyrian languages.
Ancient World History
Mari was a circular city (1.2 miles in diameter), excavated first by André Parrot (from 1933) and later by Jean Margueron (from 1979). Excavations reveal a series of palaces from the Early Dynastic II–III Periods (early third millennium b.c.e.) to the Old Babylonian Period (early second millennium b.c.e.), each palace built upon the ruins of the preceding one.
The latest palace is one of the best preserved and most impressive of the entire Bronze Age. It was exceptional for its time period, because it incorporated various religious shrines together with the royal residence.
More than 20,000 cuneiform tablets were uncovered at Mari, most dated to the Old Babylonian Period. Although the language of most texts is Akkadian (east Semitic), northwest Semitic grammar and syntax show up in proper names and in various constructions.
The archive consists mostly of palatial and provincial administrative texts, letters, and treaties, demonstrating the political value of writing in this period. It is one of the major sources of information on how the great Assyrian king Shamshi-Adad I organized his empire in northern Mesopotamia. In addition, Mari has the largest number of Mesopotamian prophetic texts. These were letters from prophets, often to the king, claiming direct messages from deities.
The city of Mari likely originated from the very start of the Early Dynastic I Period (beginning of third millennium b.c.e.). It prospered rapidly due to its strategic location along the trade route connecting Mesopotamia with Syria.
The archaeological evidence found for the Early Dynastic III Period (c. 2600 b.c.e.) shows Mari’s indebtedness to much of Sumerian culture. Short inscriptions from this period refer to Ansud as the king of Mari, a name that may also appear in the Sumerian King List.
During c. 2250 b.c.e., the title shakkanakku (Akkadian for "governor") was used for the rulers of Mari, a term that may allude to a time of foreign control, when Mari’s rulers were the deputies of other kings. This seems to have been a period of great power, when the city underwent extensive renovation.
The final century before Mari’s destruction is much clarified by the written record. Yahdun-Lim, who derived from the Sim’alite stock of the Amorites, ruled as king over Mari.
Cuneiform Tablet Listing the Names of Old Babylonian Kings - History
Ancient clay tablets that went on public display for the first time this week provide a rare glimpse of how Jews lived during their exile in Babylon over two millennia ago, reports said.
The approximately 200 tablets, which date to 572–477 B.C., the time of the exile under King Nebuchadnezzar, were discovered in modern-day Iraq, possibly during the 1970s. David Sofer, a London-based Israeli collector, owns 110 of them, or about half of the collection, which is known as the Al-Yahudu archive, Artnet News said.
The tablets are being shown at the Bible Lands Museum in Jerusalem.
Written in Akkadian Cuneiform script, an extinct Semitic language, the palm-sized tablets document the lives of members of the Jewish communities in villages in the Fertile Crescent, between the Tigris and Euphrates rivers, the report said.
Each tablet is inscribed with a date ranging from 572 – 477 B.C., with the earliest written about 15 years after Nebuchadnezzar destroyed the temple, and the last written about 60 years after the first wave of exiles returned from Babylon, the Christian Examiner said.
Filip Vukosavovic, an ancient Babylon, Sumeria, and Assyria expert who curated the museum’s “By the Rivers of Babylon” exhibition, told Reuters that the tablets’ discovery “fills in a critical gap in understanding of what was going on in the life of Judeans in Babylonia more than 2, 500 years ago” Previously, little was known about Jewish life during the time of exile. “It was like hitting the jackpot, ” he added, Artnet News said.
Each clay tablet, like an ancient iPad, reveals details about taxes, payments, trading, property leases, and local trading of fruit and other items. The family of one Jewish patriarch, Samak-Yama, can be traced over five generations through the documents. “We even know the details of the inheritance made to the five great-grandchildren, ” said Vukosavovic, according to the report. “On the one hand it’s boring details, but on the other you learn so much about who these exiled people were and how they lived.”
The tablets also have Judaic names recorded on them that may be central characters of the biblical narrative surrounding the capture of Jerusalem and 70-year exile, the Examiner said.
Though the exile was fairly short-lived for many, as Persia’s King Cyrus enabled the Jews to return to the Holy Land in 538 B.C., a Jewish community some 80, 000 people strong remained in the region for over two millennia. “The descendants of those Jews only returned to Israel in the 1950s, ” Vukosavovic said, according to Artnet News.
Wayne Horowitz, one of the archaeologists who studied the tablets, says this is the most important ancient Jewish archive since the discovery of the Dead Sea Scrolls, according to Haaretz digital. Until now very little had been known about the life of the Judean community in Babylon.
Uruk King List
Uruk King List: historiographical document from ancient Babylonia, mentioning the length of the reigns of several kings from Kandalanu (r.647-627) to the Seleucid king Seleucus II Callinicus (r.246-226/225).
The Uruk King List (also known as "King List 5" and ANET 3 566) is an important historiographical document from ancient Babylonia. It mentions the length of the reigns of several kings, beginning with Kandalanu (r.647-627) and continuing to the Seleucid king Seleucus II Callinicus (r.246-226/225). Together the Babylonian King List of the Hellenistic Period, the Uruk King list is a useful text for those who are reconstructing the chronology of Babylonia in the late fourth to mid-second centuries.
Unless it was stolen when the museum was looted in April 2003, the cuneiform tablet (IM 65066) is in the Bagdad Museum. On this website, you will find a slightly adapted transcription by A.K. Grayson, from the Reallexikon der Assyriologie, s.v. "Königslisten und Chroniken".
Chronological notes have been added the right-hand column is a modern approximation of regnal dates.
Description of the tablet
This list of kings of Babylonia and their regnal years, which appears on a fragment from the middle of a small tablet found at Uruk, covers in its preserved portion the period (obverse) from Kandalanu (r.647-627 BCE) to Darius I (522-486 BCE) and from (reverse) Darius III (r.335-331 BCE) to Seleucus II (r.246-226/225 BCE). The script is late Babylonian and the tablet was obviously inscribed some time after the reign of Seleucus II.
The Earliest Known Dictionaries
The oldest known dictionaries are cuneiform tablets from the Akkadian empire with biliingual wordlists in Sumerian and Akkadian discovered in Ebla in modern Syria.
The Urra=hubullu glossary, a major Babylonian glossary or encyclopedia from the second millenium BCE, preserved in the Louvre, is an outstanding example of this early form of wordlist.
"The canonical version extends to 24 tablets. The conventional title is the first gloss, ur5-ra and ḫubullu meaning "interest-bearing debt" in Sumerian and Akkadian, respectively. One bilingual version from Ugarit [RS2.(23)+] is Sumerian/Hurrian rather than Sumerian/Akkadian.
"Tablets 4 and 5 list naval and terrestrial vehicles, respectively. Tablets 13 to 15 contain a systematic enumeration of animal names, tablet 16 lists stones and tablet 17 plants. Tablet 22 lists star names.
"The bulk of the collection was compiled in the Old Babylonian period (early 2nd millennium BC), with pre-canonical forerunner documents extending into the later 3rd millennium" (Wikipedia article on Urra=hubullu, accessed 05-08-2009).
Watch the video: Θέλεις tablet. Θέλεις Public! (May 2022).