We often need or want to write about cuneiform without using the signs themselves. Since the 1840s Assyriologists have developed a method of representing individual cuneiform signs using the letters of the western alphabet. There are several variants of the transliteration system in use today (and it has also evolved since it was first invented). But all versions enable a reader to see exactly which signs were used in the cuneiform text while showing the reader which value of each sign was meant. Transliteration is not merely mechanical transformation but an act of interpretation too.
Assyriologists distinguish between transliteration, which is an alphabetic representation of cuneiform signs, and transcription or normalisation (these words are synonymous), which is an alphabetic representation of the language that does not give any information about the signs used to write the original text.
In transliterations of Akkadian texts, words are separated by spaces. The syllables of a word are written in lower-case letters (often in italics) and separated by hyphens. Logograms are written in upper-case (or small upper-case) letters and separated by full stops. Determinatives are written in superscript lower-case, with no connecting punctuation. Phonetic complements are written in lower-case, often in superscript, and are connected to the rest of the word with a hyphen. Sometimes scribes wrote explanatory glosses to help with the reading of difficult signs; we write these in superscript too.
In alphabetic normalisation, we write the Akkadian words just like any language: no hyphens or full stops or superscripts. But we usually put the Akkadian in italics, just as we might do for words of any foreign language. We mark long vowels and doubled consonants as necessary, even when they are not written explicitly in the original. Normalisation allows us to focus on the language, filtering out the writing system. It is also useful when we want to talk about words without reference to a particular text—in dictionaries and grammar books, for instance.
For example, the sequence
๐ ๐ฎ ๐ ๐ญ ๐ ๐ ๐ฒ ๐
can be transliterated as A.ŠA3-el dUTU-e-แนญi-ir and normalised as eqel Šamaš-eแนญir, "Šamaš-eแนญir's field". The cuneiform itself does not distinguish between different types of sign, and often does not show word boundaries. The fact that transliteration does all this makes it much easier to read than cuneiform, even for experienced Assyriologists. But transliteration also gives further help.
We have seen that most cuneiform signs have more than one possible reading (polyphony). So transliteration has to do more than represent the modern name of each sign: it must give the correct reading of it too, the one that makes most sense in the context. For instance, the sign ๐ KAL has the readings kal, dan, rib, and lak. We must transliterate the sequence
๐ ๐ ๐ ๐ ๐ก ๐
as kal-bu-um dan-nu-um (for kalbum dannum "strong dog") in order for it to make sense. Here the two identical KAL signs take different readings in each of the two words.
We have also seen that the same word or syllable can be written several different ways (heterography). We distinguish between identical readings of different signs with so-called diacritics. These are unobtrusive markings that tell us which sign the scribe used. We mark diacritics with subscript numerals. Often the second and third values are written instead with an acute and grave accent respectively. For instance:
So the transliterations u4-mu-u2 u3 mu-šu-u or u4-mu-ú ù mu-šu-u (for ลซmลซ u mลซšลซ "days and nights") tells us that the signs used can only have been ๐ ๐ฌ ๐ ๐ ๐ฌ ๐ ๐.
Fortunately there are standard sign lists which systematically catalogue all known cuneiform signs, together with their different syllabic and logographic readings. That means that Assyriologists can work with agreed conventions and don't have to memorise every single sign and all its possible interpretations.
Many cuneiform texts contain numbers, weights and measures. Assyriologists don't normally try to normalise these, as they are rarely written syllabically. Even the transliteration of numerical and metrological signs is complicated, as scribes used several different notation systems, depending on the context.
Most normal numbers were written in base 60 (just as we count modern minutes and seconds), with vertical strokes for units 1–9 (๐น ๐ซ ๐ ๐ผ ๐ ๐ ๐ ๐ ๐) and Winkelhakens (corner wedges) for 10–50 (๐ ๐๐ ๐ ๐ ๐). Base 60 numbers are transliterated with spaces, full stops or commas between the sexagesimal places, e.g., ๐น ๐ ๐น as 1 41 (1 sixty and 41 = 101).
Some systems of weights and measures used the normal base 60 numbers with separate signs for the units. Others had special signs which combined the number and unit together. A good way to represent this type of sign is to write the numerical value followed immediately by the unit in parentheses. For instance, the capacity units ๐ ๐ ๐ ๐ ๐, can be transliterated as 1(BÁN), 2(BÁN), 3(BÁN), 4(BÁN), and 5(BÁN) respectively, but other conventions are also used.
For more details of cuneiform numbers, weights and measures, see the Digital Corpus of Cuneiform Mathematical Texts [https://oracc.museum.upenn.edu/dccmt/].
Many ancient cuneiform texts are damaged or otherwise difficult to read in places, and it is important to show this in transliteration. The standard editorial marks are as follows:
You may find these conventions useful when transliterating the laws of Hammurabi in the Texts to read section of the site.
Content last modified on 02 Apr 2024.
Eleanor Robson
Eleanor Robson, 'Transliteration: Representing cuneiform alphabetically', Knowledge and Power, Higher Education Academy, 2024 [http://oracc.museum.upenn.edu/saao/knpp/cuneiformrevealed/aboutcuneiform/transliteration/]