Re: Hi, and Wow, and Eep!
- Do be intimidated. When I first joined this list ten years ago I didn't
understand half of what was posted. Now, ten years later, I still don't
understand half of the posts. But that doesn't prevent me from enjoying and
learning from the half I DO understand! :)
On Tue, Apr 9, 2013 at 1:47 AM, Jasyn Jones <jasynj@...> wrote:
> New to the list, and to conlang as an endeavor. And, I must say, just
> looking at the last few messages has me intimidated enough to slink away
> quietly into the night.
- Alex Fink, On 16/04/2013 03:09:
> On Fri, 12 Apr 2013 00:59:14 +0100, And Rosta<and.rosta@...> wrote:That's ingenious. So NURSE would have to be a vocalic /r/ (or /r/ a consonantal NURSE). The phonetic rationale is absent from nonrhotic accents, but it's still worth considering for nonrhotics too. And somehow it must counterfeed vowel reduction (so that _there, their_ with reduced vowel is not /T/) and syncope (so that _the red_ with syncope of schwa is not /T/ -- monosyllabic _the red_ is not homophonous with _thread_).
>> Alex Fink, On 11/04/2013 06:14:
>>> In native words:
>>> - initially, one has [T] in content words (like "thigh") and [D] in
>>> function words (like "thy").
>> [D] only in function words, but _thirty, thirteen_ mean that some function words have [T]. By "function word" I here understand "closed class".
> Your omission of _three_ points me to notice that for my lect, though probably less so for yours, a salvaging solution is to add a rule taking precedence over this one that only /#Tr/ occurs, never /#Dr/, whichever syllabicity the /r/ is: thus (modulo unimportant details of the rest of the transcription) /Trii, Trt-tiin, Trti/, as well as /Truu/ if that was a problem.
But it's not really worth going to these contorted analytical lengths. Even if we got your suggestion to work, we'd be saying that in morpheme-initial onsets there's just the one TH phoneme whose realization is conditioned by the morphological category of the form it's part of. Without your proposal, we say instead that even in morpheme-initial onsets /D/ is a phoneme in its own right, but it can occur only in closed-class morphemes. That's simpler, and it's less weird to have morphological category constrain phoneme incidence than phonetic realization, and furthermore it's more consistent with the fact that in TH-fronting dialects, #T > f, but #D doesn't become v (except in children's speech). (I realize you weren't arguing otherwise.)