Is there a particular sound change that would explain changing a word-initial [ʒu] (or alternatively [dʒu]) to [iʒ] before a stressed syllable? Or might this be best explained as dropping the [u] by syncope and adding a [i] by prothesis?

I found this in a couple words of the São Vicente dialect of Cape Verdean Creole, which is related to both Portuguese and other dialects of Cape Verdean Creole:

| Portuguese        | Sotavento Creoles | São Vicente |
| jogar   [ʒug'aɾ]  | ['dʒugɐ]          | [iʒ'ga]     |
| ajudar [ɐʒudˈaɾ]  | ['dʒudɐ]          | [iʒ'da]     |

This dialect often elides unstressed [i] and [u] sounds in words, compared to other dialects or languages. So a change from [ʒu-] → [ʒ-] (or [dʒu-] → [ʒu-] → [ʒ-]) would make some sense. In other languages I'm familiar with, [ʒda] or [ʒga] would be seen as "unpronounceable", so adding an initial vowel could make sense.

However, this particular dialect elided many (most?) of the initial unstressed vowels present in Portuguese. Many words start with [ʃt-], [ʃk-], [ʃp-], etc. The dialect seems very comfortable with consonant clusters at the start of words: commonly-used words include [tʃga], [pʰta], [fka], and [zbi]. There are other cases of vowel prothesis, however.

Is there another sound change I should be looking at for a possible explanation? Or is ([dʒu-] →) [ʒu-] → [ʒ-] → [iʒ-] the best bet?

(I started to wonder more about this when I heard a speaker of this dialect pronounce a name [iʃ'sɛ] instead of the standard [ʒu'zɛ]. This speaker has both speech and hearing disorders, so isn't representative of all speakers, but the sound change was so similar that I wondered if this is a pattern seen in other languages.)

A couple notes about other prothetic vowels from @KennyLau's questions:

Words that come from Portuguese words starting with "es" plus a consonant generally have no initial vowel (as is common in Portugal), across the dialects of Cape Verdean Creole.

| Portuguese | Sotavento Creoles | São Vicente |
| escola     | ['skɔlɐ]          | ['ʃkɔlɐ]    |

Another example of vowel prothesis in the São Vicente dialect is the addition of an initial /a/ in certain words:

| European Portuguese | Sotavento Creoles | São Vicente |
| mulher [muʎˈɛɾ]     | [mu'dʒɛɾ]         | [am'dʒɛʀ]   |
| melhor [mɘʎˈɔɾ]     | [mi'dʒoɾ]         | [am'dʒoʀ]   |

However, this might not be consistent for all speakers of the São Vicente dialect, as Wikipedia seems to say that these start with a syllabic m, instead of [am].

  • Semirelated: In Mirandese (the other official language of Portugal), some unstressed CV(C) syllables have become syllabic fricatives (the most obvious being z- and ç- for Romance des-). These evolved from the elided vowels that, in modern Portuguese, are frequently pronounced voiceless. Your progression of [ʒu-] → [ʒ-] → [iʒ-] seems, at least, incredibly plausible. When you cite [tʃ'ga] / [p'ta], is the [tʃ] and [p] fully syllabic? Jan 5, 2016 at 9:28
  • @guifa to be honest the only one of those 4 examples that I'm certain is 2 syllables is [d'bɔʃ]. The others might not be. Is there an easy way to tell the difference?
    – Dan Getz
    Jan 5, 2016 at 10:45
  • that can sometimes be a bit harder to peg down. In writing you certainly couldn't and in speech can be tricky because it happens on unstressed syllables, but in a song, it'd ought to be easier: if [tʃ] gets one note, and [ga] another, then it's probably syllabic. If they cram [tʃga] into a single note, then it's likely not. Do you have any clips? Jan 5, 2016 at 18:21
  • 1
    @KennyLau yeah, that's the thing. These prothetic vowels are apparently not consistent across all the speakers of this dialect, if that article doesn't mention them. But you can see "amdjer" in these song lyrics. Haven't found a source for [iʒ'da] yet.
    – Dan Getz
    Apr 16, 2016 at 16:41
  • 1
    @DanGetz I really can't say anything if I only have 4 examples. The most I could say is that [ʒ] and [m] initials undergo prothesis.
    – Kenny Lau
    Apr 17, 2016 at 3:34

1 Answer 1


Without historical data on the dialect, I'd think that the second hypothesis (Or might this be best explained as dropping the [u] by syncope and adding a [i] by prothesis?) sounds natural and plausible.

But maybe, you can dig up historical records shedding more light on the evolution of the São Vicente dialect of Cape Verdean Creole.

  • Just looking at those few examples, how about initial syllable metathesis followed by fronting of the vowel via unstressed /u/ -> [ʉ] -> [i]? In the absence of more data (especially, historical data) it's all just guesswork. Mar 7, 2016 at 20:56

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.