I was recently reminded (by this question) that intervocalic single -s- turned into -r- by rhotacism, and later new instances of intervocalic -s- were produced from -ss-. If the vowel preceding -ss- is short, then the transition -ss->-s- transforms the syllable from long to short (heavy to light). This is significant especially for metric poetry. I would expect this shortening to be compensated by elongating the preceding vowel. Although many -s- are preceded by a long vowel, there some examples of a short vowel followed by a single -s-: posui, casa, asinus.

Are there any examples where compensation is known to have happened? Or is the intervocalic transition -ss->-s- something that was allowed to freely change syllable length (weight)? Are the words I listed (and others like them) somehow exceptional while most words did have compensation?

1 Answer 1


The general tendency was to dissolve a geminate ss in writing, and also presumably in speech, only after a long vowel or diphthong. If this weren't the case, then there would be virtually no instances where ss still occurred in text.

“There are several processes at work in Latin which simplify geminates to single consonants. The most important of these is the simplification which takes place after long vowels. This simplification is clearly based on syllable patterns; for example, the heavy syllable provided by the long root vowel of Lachmann’s Law forms like cāsus ‘fallen’ (< cad+tos) allows the simplification of the putative intermediate form *cās-sus. This process contrasts with that seen in forms such as missus ‘sent’, where the root vowel is short and the syllable weight is preserved by the geminate” (Baldi, P., The Foundations of Latin. Berlin – New York 1999: 298)

So no, this intervocalic transition did not allow to freely change syllable weight. This would be exceptional anyway: Latin is very 'aware' of its moraic structure and will not allow random morae to disappear or resurface without a general rule, such as Brevis Brevians (iambic shortening).

I have no specific ideas for the examples you've given, but there may be multiple reasons why the rhotacism didn't occur as you expect. The word may be a loanword that is more recent than the period of productive rhotacism, or Brevis Brevians may actually occur after degemination, etc.


Wiktionary on asinus:

Recent loanword, as shown by intact intervocalic -s- which otherwise rhotacized to -r- in native words.

  • Thanks! This makes sense. It's fairly common to see short vowel +ss or long vowel +s in Latin. Do you happen to know how recent the donkey is?
    – Joonas Ilmavirta
    Mar 17, 2017 at 8:42
  • 2
    There's a strict terminus ante quem of 184BC, because then is when Plautus, the writer of the Asinaria, died. For a terminus post quem, an acceptable ballpark figure for the end of productive application of the rhotacism rule seems to be about 350BC. But since the source language seems to be unknown, the loanword theory may have to be considered speculative.
    – blagae
    Mar 17, 2017 at 8:54

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.