Hittite cuneiform occasionally shows "plene" spellings, with extra vowel signs that might indicate vowel length, or show the height of back vowels, or distinguish homophones (like the French grave accent), or something else—it's not entirely clear.

However, from what I've read, Akkadian definitely had phonemic vowel length. Was this ever reliably/consistently distinguished in cuneiform, either by plene spelling, different glyphs, or any other method?

(Akkadian is the first language to come to mind that has both definitely-phonemic vowel length and a lot of attested cuneiform, but answers involving other languages are welcome too: I'm curious if any sort of cuneiform reliably distinguished vowel length, not just Akkadian.)

2 Answers 2


In Akkadian, Ca-a, Cu-u, Ci-i are often used to indicate Semitic long vowels, but this is not consistent. For example, dabābu “word” is usually written as da-ba-bu, but sometimes it appears as da-ba-a-bu. Use of the "plene" spelling is positive evidence for a long vowel, but non-use does not by any means prove that the vowel is short.

Double a-a for ay (Keelan's comment) is a different matter.


Vowel length is a fairly complex issue in Cuneiform.
To begin with, there are plenty of Cuneiform scribal traditions that do not use the same signs nor use them in the same way. So it is hard to make definitive statements on Cuneiform in general.
Some people claim that Archaic Akkadian in the 3rd millennium BC distinguished signs with short or long vowels. This may be true in my opinion, but this necessitates more work to be fully elucidated.
As for later Akkadian and especially Babylonian, we can infer that some vowels were long because they do not mute out. As a rule, a short vowel falls if it does not create an impossible cluster of consonants, for example C1aC2aC3u will change into C1aC2C3u. So if we have a word written C1a-C2a-C3u in cuneiform, we can infer it's C1aC2a:C3u with a second long vowel.
Some Hittite (or Hurrian) texts indeed resort to plene writing, like Ca-a- or Cu-u-. As we have no speaker to explain what that stands for, we can only hypothesize that it probably stands for vowel length triggered by some factor like stress. That's plausible but not certain.

  • 1
    I thought that occasionally doubled vowels can also indicate a glide, e.g. da-a-a-num for Old Babylonian dayyānum - so is this plene writing ambiguous?
    – Keelan
    Oct 15, 2019 at 11:11
  • 1
    Please see my answer.
    – fdb
    Oct 15, 2019 at 11:45
  • 1
    da-a-a-num is probably Old Assyrian, rather than Old Babylonian. Assyrian does not have the sign ia.
    – user23769
    Oct 15, 2019 at 16:36
  • @Keelan That could make a good question of its own! I know Hittite innovated a sign wi₅ for /wi/, but lacked anything for /we/.
    – Draconis
    Oct 15, 2019 at 18:03
  • 2
    @ArnaudFournet. Look at the references under dajānu in the CAD! The usual spelling in Old Assyrian, Old Babylonian and later stages is da-a-a-nu. The sequence a-a is a scribal convention for ajV. Some Assyriologists treat it as a single sign with the “Lautwert” aju, aji, aja. It is not an example of plene spelling of long vowels.
    – fdb
    Oct 15, 2019 at 19:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.