Consider the words sūs and sŭŭs. The former has one long u, the latter has two short ones in two syllables. For another similar pair with a different vowel, consider īmus and ĭĭmus. I wonder how easy it was for the Romans to confuse a long vowel with two short ones of the same quality, like ū vs. ŭŭ.

To make the question more concrete, I would like to know (some of) the following:

  • Are there examples of word plays with ū/ŭŭ or other such combinations? I can imagine how confusing sūs and sŭŭs could make a decent joke.
  • Are there misspellings that indicate that it might have been difficult to make the distinction?
  • Is the distinction considered by ancient grammarians?

For me personally it is much easier to confuse ū with ŭŭ than with , although I can distinguish all three.

  • 5
    Interesting question! It is possible, by the way -- though this is a debated point -- that the difference between (at least some) short and long vowels was not solely one of quantity but also of quality: e.g. that short i was [ɪ] while long i was [i:]
    – TKR
    Dec 10, 2016 at 20:50
  • @TKR, interesting. Then the question arises whether or not the difference in quality helped make the distinction between a long and a double short vowel.
    – Joonas Ilmavirta
    Dec 10, 2016 at 20:53

2 Answers 2


Well, there is some fairly simple evidence that a sequence of two identical short vowels could in some cases be treated as equivalent to a single long vowel, namely that the former can contract into the latter: e.g. ĭĭt ~ īt, nĭhĭl ~ nīl.

This does not necessarily imply that the pronunciations were identical, of course, but it does show that the two could be easily conflated. That said, it's not clear if this was a generally productive process: for example, I don't know of any cases where suus is written or scanned as a monosyllable.

  • Now that I reread your answer, I realize I have never seen īt. Can you give an example? I am familiar with nīl.
    – Joonas Ilmavirta
    Dec 13, 2016 at 15:37
  • 1
    Check Lindsay §152 for more examples of vowel-contraction. This does not mean they were confused, though.
    – cmw
    Dec 13, 2016 at 16:46
  • 1
    @JoonasIlmavirta, examples of it: latin.packhum.org/search?q=%23it%23
    – TKR
    Dec 13, 2016 at 17:55

sūs has a long vowel, but the other cases (suis etc.) have a short vowel in the stem, so I suppose the genitive plural suum would look and sound just like the acc. sing. m. and nom./acc. s. n. of suus. But otherwise sūs and suus sound quite different.

  • 1
    sūs and suus sound quite different -- what makes you say that? It seems to me that they would most likely be phonetically identical in normal speech, unless there was an inserted glide between the two vowels of suus (I don't know if there's any evidence for or against that possibility).
    – TKR
    Dec 10, 2016 at 20:54
  • 2
    @TKR Finnish allows that in compound words and otherwise at word boundaries, and the distinction is clear and easy for a Finn. However, Finnish does not have truly word-internal double shot vowels. As fdb writes, long vowels are written as two vowels but articulated as a single long one.
    – Joonas Ilmavirta
    Dec 10, 2016 at 21:35
  • 4
    As in "Joonas" (for Ιωνας) of course.
    – fdb
    Dec 10, 2016 at 21:54
  • 3
    @TKR I have a contrast of /kre:/ /kre.e/ and /kre.e:/ in my native language (créent/craie, créé and créée), as well as /ni.il/ and /ni:l/ (nihil and the Nile). Glide insertion is possible but rare and stigmatized (at least in sequences of identical vowels) and I've never heard even a sign of ongoing merger. Dec 11, 2016 at 12:33
  • 3
    @TKR I'm Belgian, yes. Your initial statement was a bit stronger than that, so I wanted to offer an example of a similar contrast that was synchronically stable. Based on my experience, I don't think the number of ambiguous tokens would be that high, although there's always the danger I might hear what I'm expecting to hear rather than what's actually being said Dec 13, 2016 at 21:35

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.