Re: Chinese Dialect Question

JS Bangs wrote:
> True. However, singular examples of gemination aren't unknown, and can be
> phonologically useful. This actually poses a question: In those dialects
> of Spanish that have a vowel quality difference in closed syllables, does
> the first syllable of /perro/ behave differently from the one in /pero/?
Non-native speaker says, no, and suspects the word is syllabified pe-rro;
both would have the more open [E] allophone that occurs before a rhotic.
>
> Not necessarily. I'd look at it this way: there's a single phoneme /r/,
> pronounced [r] initially and finally and [4] medially, and of course when
> two /r/'s collide they fuse to [r].
This is mildly troublesome to me-- all the exs. of -rr- with a known Romance
history had, IIRC, -rr- already in Latin/Proto Rom. I'm not sure, but
theoretically Italian ought to retain the same contrast /r/:/rr/.  Also, a
lot of the exs. of -rr- in Span. are of non-Romance origin (or at least,
unique to Spain).