Re: Quantity shift (was: Re: Native grammatical terms)

On Thu, 20 Nov 2003, Isidora Zamora wrote:

> With Trehelish, I would like to have a degemination process, but it would
> certainly be nice to have some natural model to get a sense of things.  I
> need to entirely eliminate all geminates in the modern language, but I
> think that I need them there in the proto-language to create some of the
> modern phonemes and phonotactic constraints (or lack thereof.)
Well, it's not uncommon for languages with length distinctions to develop
a redundant difference of quality (i.e. /i:/=[i:] but /i/=[I]). All you
need to do is drop the difference of quantity (which Dutch has done), and
you get rid of the difference. Alternatively, you could do what some
dialects of English have done* and diphthongise all your long vowels
(possibly after having a difference of quality).

* Except when /r/ interfered with the process and a couple of exceptions
(like 'broad'), all Middle English long vowels are dithpthongs in
Australian English; the long vowels I mention below come from borrowings,
de-diphthongisation of ME diphthongs, conditioning by surrounding
consonants, and one case of who-knows-what (i.e. an exception to the
rule).

> I'm curious about something.  Is there precedent for having geminate vowels
> in a language without having geminate consonants as well?  In this case, I
> am thinking about the parent language of Trehelish.  I assume that there
> would be no question about having geminate vowels and no geminate
> consonants in a language such as Nidirino, which allows only open syllables?
Of course. Australian English, which has a distinction of length, has
no geminate consonants. Though the distinction of length is relatively
young, so maybe it's supposed to come up later. And I think Middle English
might've.

--
Tristan

Hi, I'm a .sig virus! Copy me to your .sig file and help me propagate!