On Fri, 21 Sep 2007 17:19:48 -0700, uaxuctum@[EMAIL PROTECTED]
wrote:
>On 22 sep, 00:55, "Logan Kearsley" <chrono.sur...@[EMAIL PROTECTED]
> wrote:
>
>> <uaxuc...@[EMAIL PROTECTED]
> wrote:
>>
>> > I find it very odd that you chose to represent these consonants
>> > using letters that are primarily associated with voiced sounds.
>> > Obstruent consonants are typically voiceless (whereas sonorant
>> > consonants and vowels are typically voiced).
>>
>> Change the symbol list to p, t, k, c/x, f, & s, then. Which set of
symbols
>> you write it with isn't as im****tant as what the sounds are.
>> Note, I intentionally left out distinctions based on voicing- unvoiced
and
>> voiced forms of each are considered allophones of a single phoneme.
>
>Yes, but using <d>, <z>, etc. seems to imply that [d], [z], etc.
>would be the primary allophones of those phonemes (the ones
>representing their "ideal" sounds out of context) while [t], [s], etc.
Don't get stuck equating the language's choice of Latin character with
a particular IPA symbol!
>would seem to be relegated to secondary allophones; when it
>should be the other way round, because the unmarked, "basic"
>phones in the obstruent category are the voiceless ones.
>
>
>> > > n- nasal
>>
>> > Why did you leave out "m"?
>>
>> I didn't. [m] is a nasal.
>
>Do you mean you would conflate [n] and [m] into a single
>phoneme /n/? Now _that_ would be extremely odd.
That's exactly what they did. There is a "nasal sound" that has two
phomenes. Or phonemes, depending. So, no distinction between "nene"
and "meme".
>Nasals
>are frequently conflated into an |N| archiphoneme (phonetically
>realized as a homorganic nasal according to the following
>sound), but *only* in coda position (i.e., frequently a language
>has the syllable-initial phonemes /m/ and /n/ and the coda nasal
>archiphoneme |N|; but not a single /n/ phoneme everywhere
>that might be realized as [m] syllable initially so that [mama]
>and [nana] would be allophones). The distinction /m/ vs. /n/
>syllable-initially is about as basic as the /p/ vs. /t/ distinction.
>Only a very few oddball languages lack /m/ (I can only think
>of reconstructed ancient Basque, and a few Native American
>languages that seem to lack labial consonants altogether;
>but these are the exceptions, not the rule, and in them the
>sound [m] either does not occur at all or occurs as allophone
>not of /n/ but of /w/, /b/ or /nb/).
>
>
>> > I'm afraid I think it currently doesn't look very optimal as
>> > a "universally pronounceable" phonology. But here you are
>> > an idea for a quite minimal phonology (ten phonemes in all)
>> > that is not likely to cause much trouble:
>>
>> Can you say what in particular is wrong with it? What and for whom
would be
>> difficult to pronounce?
>
>If you are trying to be "universally pronounceable", look only
>for the most essential kinds of articulations and distinctions.
>Get rid of labiodentals altogether (historically they have been
>repeatedly troublesome for speakers of several languages,
>leading to the frequent evolution [f] > [h] > zero), and get
>rid of the "s" vs. "sh" distinction (troublesome for Spanish,
>Indonesian and Tagalog speakers, for example, and it didn't
>exist in Latin or Greek either; one sibilant is enough for
>a basic phonemic repertoire), and make it clear that the
>voiceless allophones would be the primary allophones of
>the obstruents.
I get the impression that this language doesn't have "primary
allophones". The impression is illusory in any event, because we
associate "b" with a voiced sound in both our normal spelling system
and in the IPA.
Padraic
--
Posted via a free Usenet account from http://www.teranews.com
|
