Road to Translation
For centuries, travelers on the road to the ancient city of Persepolis, in modern-day Iran, noticed strange carved inscriptions riddling the ruins, words that remained wondrous and yet mystifying. Arab and Persian historians of the medieval Islamic empire often attempted to decipher the much older Persian writings, but remained largely unsuccessful. Not until the Englishman Sir Thomas Herbert in his 1634 travel book, “A Relation of some yearers travaile”, did the decipherment of the ancient cuneiform and the long road to translation begin.
Herbert noted in the travel book that on one of the carved walls of Persopolis (the Capital of the Persian Empire), the wall contained, “a dozen lines of strange characters…consisting of figures, obelisk, triangular and pyramidal.” Originally Herbert thought the signs resembled Greek. In a later edition of the book penned in 1664, Herbert reproduced some of the signs for readers; the man commented that the signs appeared legible and intelligible, and therefore potentially decipherable. The Englishman also guessed, correctly, that the cuneiform signs did not represent letters of hieroglyphs but words and syllables, read left to right.
Archaic Era Sumerian cuneiform
Then Cartsen Niebuhr brought the first accurate copies of the inscriptions to Europe and Bishop Frederic Munter of Copenhagen found the inscriptions divided by an oblique wedge and later dated the writings to Cyrus the Great of the Persian Empire. The bishop correctly deciphered one of the words that reoccurred at the beginning of each of the inscriptions without variation as “King”. In 1802, Georg Friedrich Grotefend determined that two king’s names existed in the text as Darius and Xerxes, and assigned appropriate alphabetic values to the corresponding cuneiform characters.
Professor Christian Lassen of Bonn published a work in 1836 called “The Old Persian Cuneiform Inscriptions of Persepolis.” Christian, along with pupil Burnouf, identified the names of the Persian satrapies (regional governors of king Darius), and thereby fixed the values of the Persian characters. Christian succeeded in fixing the true values of nearly all the letters in the Persian alphabet and proved that the language did not represent Zend, but strongly related.
A month later in 1836, the eminent French scholar, Eugene Burnouf, the pupil of Christian, published an alphabet of thirty letters translated from the cuneiform, mostly correct in translation.
In 1835 Henry Rawlinson, a British East India Company army officer, paid a visit to the Behistun Inscriptions in Persia. Rawlinson found that the carved signs dated to the reign of King Darius (522–486 BC) and consisted of identical texts in the three official languages of the empire: Old Persian, Babylonian, and Elamite. Effectively, the Behistun inscription represented the “Rosetta Stone” of cuneiform.
Rawlinson correctly deduced that the Old Persian consisted of phonetic script. In 1837 the Englishman sent a copy of the first translated paragraph of the Behistun inscription to the Royal Asiatic Society. However, Rawlinson consulted a number of corrections and innovations in decipherment produced by Lassen and Burnouf earlier, and accordingly revised the paper. Then in 1847, after many delays, the Royal Asiatic Society finally published the first part of Rawlinson's Memoir; the second part did not appear till 1849. A large part of deciphering the Persian cuneiform texts now appeared virtually complete.
After translating the Persian, Rawlinson and the Irish Assyriologist Edward Hincks working independently of each other, started translating the other scripts. The techniques used in actually deciphering the Akkadian remain unpublished. A number of scholars speculate that, given Hinck’s readily given testimony of used techniques and Rawlinson’s apparent silence, Rawlinson actually copied Hinck’s work. No one can prove the argument though. In 1842, Paul Emile Botta’s discovery of the city of Nineveh and the subsequent uncovering of the great library of Assurbanipal greatly accelerated the pair’s translation efforts; the royal archive lay complete with tens of thousands of intact and extant baked clay tablets covered in cuneiform.
By 1851, Hincks and Rawlinson purported to be able to read 200 Babylonian signs. Young German-born scholar Julius Oppert and British Orientalist William Henry Fox Talbot joined the other two in translating the languages. In 1857 the four men met in London and took part in a famous experiment to test the accuracy of the group’s abilities to decipher. Edwin Norris, the secretary of the Royal Asiatic Society, gave each of the men a copy of a recently discovered inscription from the reign of the Assyrian emperor Tiglath-Pileser I to translate. A panel of expert jurors presided over the examination of the resulting translations to assess the group’s accuracy. For the most part, the four translations strongly coincided except for a few discrepancies. The inexperienced Talbot’s translation contained a number of mistakes, and Oppert's translation contained a few doubtful passages, which the jury attributed to the German’s misunderstandings of English words. But Hincks' and Rawlinson's versions corresponded remarkably closely in many respects. According to the jury, the basic decipherment of Akkadian cuneiform now reached the status of "a fait accompli".
The early days of cuneiform decipherment presented great difficulties in translating proper names. However, now, due to a better understanding of the principles behind formation and the pronunciation of the thousands of names found in historical records, business documents, votive inscriptions and literary productions, the problem is markedly minimal. The difficulty initially lay in characteristic use of old Sumerian non-phonetic logograms in other languages simultaneously possessing different pronunciations for the same symbols. Until scholars identified the exact phonetic reading of many names through parallel passages or explanatory lists, scholars remained in doubt and recoursed to conjectural or provisional readings. Fortunately, variant readings of inscriptions, the same name being written phonetically (in whole or in part) in one instance, and logographically in another, provided valuable insights into translation.
History of Cuneiform
The cuneiform writing of ancient Mesopotamia represents one of the earliest known forms of writing in the world. Scientists contend that the script emerged fully formed in Sumer around the 30th century BC, yet with predecessors reaching back into the late 4th millennium BC, the Uruk IV period, where cuneiform writing initially began as a series of progressively more complex pictographs. In over 3000 years of evolution, the script eventually simplified into more abstract and smaller characters, about 1000 unique characters in the Early Bronze Age to just about 400 near the Late Bronze Age with Hittite cuneiform.
Scribes wrote in cuneiform by pressing characters into wet clay slabs with a blunt reed also called a stylus. The wedged shaped impressions formed the characters, and gave the writing its name (Cuneiform comes from the Latin cuneus, which means “wedge”).
The script characters found adaptations in the many Mesopotamian languages, including Akkadian, Eblaite, Elamite, Hittite, Luwian, Hattic, Hurrian, and Urartian. The system also inspired the Ugaritic and Old Persian alphabets. Cuneiform eventually died out being replaced by the much simpler and easier to use Phoenician alphabet during the Assyrian Empire. By the 2nd century AD, the script largely became an extinct and antiquated form of writing.
For this reason, the cuneiform writing system remained undeciphered until the 19th century; most scholars date the decipherment to 1857. The system consists of a combination of logophonetic, consonantal alphabetic and syllabic signs. The cuneiform script underwent considerable changes over a period of more than two millennia, as the progression of signs of SAG meaning “head” below demonstrate (see row 7 of the table directly below).
Evolution of Script
The first sign is a pictogram prevalent around 3000 BC. Then, at stage 2, the pictogram rotated counterclockwise around 2800 BC. The abstracted form of the SAG character took root around 2600 BC on archaic monumental inscriptions, and saw use in clay at stage 4 around the same period. Stage 5 represents the late 3rd millennium, and stage 6 represents Old Assyrian ductus of the early 2nd millennium, as adopted into Hittite. Stage 7 is the simplified sign as written by Assyrian scribes in the early 1st millennium, and until the script's extinction.
Mesopotamia's "proto-literate" period spanned from about the 35th to 32nd centuries. The first documents unequivocally written in the Sumerian language date to 31st century, uncovered at Jemdet Nasr.
Around 8000 BC, the Sumerians originally used clay tokens to count agricultural and manufactured goods. Later, the Sumerians placed the tokens in sealed hollow clay containers. On the outside, the people started to impress pictograms of each of the tokens sealed in the container. After some time, the people dispensed of the tokens and instead relied on symbols drawn on clay surfaces. To avoid making a picture for each instance of the same object (for example: 100 pictures of a hat to represent 100 hats), small marks “counted” the number of object inside. The Sumerians invented the system around 3300 BC.
In the beginning, people drew pictograms either on clay tablets in vertical columns with a sharpened reed stylus, or incised symbols on stone. The early style lacked the characteristic wedge shape of the strokes of later times. Certain signs to indicate names of gods, countries, cities, vessels, birds, trees, etc., are known as determinants; the signs essentially served as a guide for the reader in “determining” the meaning of certain phrases. Proper names continued to be usually written in purely "logographic" fashion.
Around 2900 BC, many pictographs began to lose their original function, and a given sign could have various meanings depending on context or determinatives. The sign inventory system shrank in size from some 1,500 signs to some 600 signs; as a result, the writing became increasingly phonological, that is to say, a matter of sounds rather than logograms. Writers actually re-introduced determinative signs to avoid ambiguity. The development is thought to directly parallel similar changes in Egyptian hieroglyphics around the same time.
In the mid-3rd millennium BC, the writing direction of the cuneiform changed to left to right in horizontal rows, subsequently rotating all of the pictograms 90° counter-clockwise in the process. A new innovate wedge-tipped stylus replaced the cruder stylus of the past; the two developments allowed writing to be quicker and easier for scribes. By adjusting the relative position of the tablet to the stylus, the writer used a single tool to produce a wide variety of impressions.
Scribes fired clay tablets in kilns to harden the clay, if people required permanent record. If not, the tablet may be molded and recycled into a new tablet. Ironically, many clay tablets recovered by archaeologists actually remained preserved for posterity due to attacking armies burning the city, which fired and hardened the tablets during the invasion. Besides clay tablets, the rulers also commissioned scribes to inscribe the script on commemorative stelae and carved reliefs on monuments designed to honor the king or queen.
The spoken language actually consisted of many similar sounds and in many cases, cuneiform symbols might denote an idea and sound. For example, in the beginning scribes used the same symbol for the words "Life" [ti] and "Arrow" [til]. After the Semites conquered Southern Mesopotamia, signs gradually changed from being pictograms to syllabograms, probably out of need to simply the system. In that way the sign for the word "Arrow" became the sign for the sound "ti".
Gradually, as the writing spread to more and more people, the cuneiform grew more and more complex. Still, a sound might represent different words, thus the writer must write each with a different sign. For example, the syllable “gu” has over fourteen symbols. If on the other hand, a symbol might possess similar meanings but different sounds, the writers often used the same symbol for each. For instance, scribes often used the same symbol for “voice” [gu] as that of the words "tooth" [zu] and "mouth" [ka]. To clarify the specific meaning of each, the writers added signs or combined two signs to define the meaning. In such cases scribes used a geometrical pattern or another cuneiform sign.
As time went by the cuneiform got very complex and the difference between a pictogram and syllabogram grew increasingly vague. Several symbols became too “overloaded” to be clear. The writers decided to continue to refine and combine symbols to produce a more concise script. The word "Raven" [UGA] had the same logogram as the words "soap" [NAGA] "name of a city" [ERESH] and "the patron goddess of Eresh" [NISABA]. Two phonetic compliments defined the word of raven with [u] in front of the symbol and [gu] behind. To be sure no confusion occurred over the interpretation of the word scribes further added the determinative symbol for "bird" [MUSHEN] behind the three symbols. Although the written cuneiform survived until the 1st century AD, the spoken language largely died out around the 18th century BC, around the time of the Sumerian collapse. The closely related successor of Sumer, Babylon, spoke Akkadian, a language that often mixed with Sumerian in centuries prior.
Semitic Simplification: Akkadian Influence
The Akkadians adopted the archaic cuneiform script ca. 2500 BC, and by 2000 BC, the script evolved into Old Assyrian cuneiform. Semitic equivalents for many signs became distorted or abbreviated to form new "phonetic" values, because the syllabic nature of the script as refined by the Sumerians remained unintuitive to Semitic speakers.
At this stage, the Akkadians reduced the former pictograms to basic abstractions, signs composed of five basic wedge shapes: horizontal, vertical, two diagonals and the Winkelhaken impressed vertically by the tip of the stylus.
Except for the Winkelhaken which is tail-less, the length of the wedges' tails could vary as required for sign composition.
Signs tilted by 45 degrees are called tenû in Akkadian, thus DIŠ is a vertical wedge and DIŠ tenû a diagonal one. Signs modified with additional wedges are called gunû, and signs crosshatched with additional Winkelhaken are called šešig.
Usually, typical signs consist of about five to ten wedges, while more complex ones contain over twenty. Scientists still debate whether a number of strokes might represent distinct, new signs. The ligature of KAxGUR7, for example, consists of 31 strokes.
Although Akkadian script changed many parts of the earlier Sumerian script, the new writing still included phonetic symbols from the Sumerian syllabary and logograms read as whole words. Quite a few signs tend to be polyvalent, much like the Egyptian hieroglyphs, possessing both a syllabic and logographic meaning. The complexity of the system also bears a striking resemblance to old Japanese written in Chinese-derived script, which used Sinograms as logograms and others as phonetic characters.
Translation and Abbreviation in Western Alphabets
Transliteration of cuneiform conforms to a specific format. Because of the script's polyvalence (one sign has several, unrelated readings), transliteration often contains more information than the original document. For example, the sign DINGIR in a Hittite text may represent either the Hittite syllable an or may be part of an Akkadian phrase, representing the syllable il. Or the word may be a Sumerogram, representing the original Sumerian meaning, 'god' or the determinative for a deity. Depending on the present context, the rendition of the same glyph varies greatly.
Therefore, a text containing DINGIR and MU together might be construed as the
words "ana", "ila", god + "a", god + water, or a
divine name "A" or just Water. A scholar, through transliteration of the
signs, might separate the signs with dashes "il-a", "an-a",
"DINGIR-a" or "a". The result is easier to
read than the original cuneiform and now the reader can trace the sounds back
to the original signs.
There are differing conventions for transliterating Sumerian, Akkadian (Babylonian) and Hittite (and Luwian) cuneiform texts. One convention in wide use across the different fields is the use of acute and grave accents as an abbreviation for differentiating symbols that appear the same but possess different meaings. Thus, u is equivalent to u1, the first glyph expressing phonetic u. An acute accent, ú, is equivalent to the second, u2, and a grave accent ù to the third, u3 glyph in the series.
In Sumerian transliteration, a multiplication sign 'x' is used to indicate ligatures (composited symbols or signs). As shown above, signs represented as ligatures appear in capital letters, while the specific reading selected in the transliteration is represented in small letters. Thus, capital letters can be used to indicate a so-called Diri compound - a sign sequence that has, in combination, a reading different from the sum of the individual constituent signs (for example, the compound IGI.A - "water" + "eye" - reads imhur, meaning "foam"). In a Diri compound, the individual signs are separated with dots in transliteration. Capital letters may also be used to indicate a Sumerogram (for example, KUG.BABBAR - Sumerian for "silver" - being used with the intended Akkadian reading kaspum, "silver"), an Akkadogram, or simply a sign sequence of whose reading the editor is uncertain. Naturally, the "real" reading, if it is clear, will be presented in small letters in the transliteration: IGI.A will be rendered as imhur4.
The Sumerian language has only been widely known and studied by scholars for approximately a century, and thus changes in the accepted reading of Sumerian names still occur from time to time. For example, the name of a king of Ur, read Ur-Bau at one time, later read as Ur-Engur, and is now read as Ur-Nammu or Ur-Namma; some read Lugal-zaggisi, a king of Uruk as Ungal-zaggisi. In other cases, scholars remain confused on the distinction between Sumerian or Semite names. In the past, doubt lay over whether the signs composing a Semite's name represented a phonetic reading or a logographic compound. Thus, inscriptions of a Semitic ruler of Kish with a name written Uru-mu-ush, first deciphered rendered a name first taken to be logographic because uru mu-ush might read simply, "he founded a city" in Sumerian; scholars accordingly retranslated the name back to the original Semitic as Alu-usharshid. Even later still, the URU sign also could be read as ri, and thus formed the name of an Akkadian King named Rimush.
The Sumerians used a numerical system based on 1, 10 and 60. As such, the Sumerians might mark the number 70 by writing the sign for 60 and then the sign for 10.
Number of Signs: Not Your Usual Alphabet
Altogether, the Sumerian cuneiform script contained on the order of 1,000 unique signs (or about 1,500 if variants are included). The number shrank to about 600 by the 24th century BC and the beginning of Akkadian records. Not all Sumerian signs are used in Akkadian texts, and not all Akkadian signs are used in Hittite.
Falkenstein (1936) lists 939 signs used in the earliest period (late Uruk, 34th to 31st centuries). With an emphasis on Sumerian forms, Diemel (1922) lists 870 signs used in the Early Dynastic II period (28th century, "LAK") and for the Early Dynastic IIIa period (26th century, "ŠL"). Rosengarten (1967) lists 468 signs used in Sumerian (pre-Sargonian) of the city of Lagash, and Mittermayer ("aBZL", 2006) list 480 Sumerian forms, written in Isin-Larsa and Old Babylonian times. Regarding Akkadian forms, Borger ("ABZ", 1981) represented the standard handbook for many years with 598 signs used in Assyrian/Babylonian writing, but recently, Borger ("MesZL", 2004) supersedes the earlier version with an expansion to 907 signs, an extension of Sumerian readings and a new numbering scheme.
Forrer (1922), Friedrich (1960) and the HZL (Rüster and Neu 1989) list signs used in Hittite cuneiform. The HZL lists a total of 375 signs, many with variants (for example, 12 variants are given for number 123 EGIR)