Re: GROUPLANG: Phonetics
|From:||Carlos Thompson <cthompso@...>|
|Date:||Thursday, October 15, 1998, 13:02|
De: Pablo Flores <fflores@...>
Fecha: Mi=E9rcoles 14 de Octubre de 1998 19:27
>Carlos Thompson wrote:
>>The phonetics, as propossed this far, is:
>>vouels: /a/, /e/, /i/, /o/, /u/, /y/
>>/p/ /b/ /t/ /d/ /k/ /g/ /q/ /q_voiced/
>>/f/ /v/ /s/ /z/ /x/ /G/ /h/
>> /m/ /n/ /N/
>> /w/ /r/ /l/ /j/
>Actually, we've discussed to drop /h/, mainly because it didn't
>quite fit into the system. We could do with /x/, variants [x] and [h].
Yes, I remember.
>Dropping <h> would also allow us to use all the h-digraphs without
>problems. Not only <kh>, <gh>, <qh> (which are invalid clusters, tho)
>but also <nh> for /N/. If we keep /h/ =3D <h>, we'll have to write <ng>
>or there'll be an ambiguity.
>Note that <ng> for /N/ is not ambiguous; /ng/ could be written <ngg>,
>because syllable-final nasals assimilate to the next place of
How about an N ending root combined by a velar inleaded suffix or composi=
lon + ka =3D> lonka ['lON.ka] or longka (explicit orthography)
lon + ge =3D> longe ['lON.ge] or longge.
but _longe_ could suggest ['loN.Ne]
With the nh diagraph, _longe_ would no need explicit orthography _lonhge_
after ng are two different sounds and asimilation wouls lead /ng/ to be
>>With the following orthography: all those sounds above, with lowercase
>>letters in IPA will use the IPA value, but "x": a, b, d, e, f, g, i, j,= k,
>>l, m, n, o, p, q, r, s, t, u, v, w, y, z
>>The other sounds are: /x/ as <kh> or <x>, /G/ as <gh>, /q_voiced/ as <q=h>
>><qg>, /N/ as <nh> or <ng>, /h_voiced/
>Let's decide about <h> and then I'll tell you :-)
>>No values this far for <c> and <x>.
><c> could be /q_voiced/, but it's not very intuitive...
><x> could be /x/, but I think it's better to keep <kh> and <gh>
>(they are more "mnemonic", and symmetrical).
Maybe <c> and <x> could be only used in borrowings with no phonetical val=
en grouplang or as alternative writing for /q_voiced/ and /x/
>>Because of the VV posibility in a syllabe, in composing (or in adding
>>affixes) there could rise V-V combinations. The posibilities are:
>>Avoid V-V combinations forcing C and F in every root.
>>Asimilate V-V as a VV diphthong (even ae, ao, oe, etc.)
>>Use something (both phonetically and orthographic to mark an hiatus, ma=ybe
>>dieresis or an apostrophe)
>>insert a sound, either a glide or a glotal stop.
>>I vote for a glide in root combination and asimilation fo V-V into VV f=or
>I vote for a glide in root combination AND for affixes.
>I don't think we should force the roots to have any special form.
I think affixes and possible roots too should have a flexible forms, and,=
we stick with voice harmony, affixes and roots should not differe in voic=
(at least in the places harmony will take place).
With your first proposals:
> frar- "dog"
> kjak- "bite, bit"
> wiv- "red"
> s- "I, me, first person"
> qaun- "hard, strong"
> Case markers:
> a- agent
> pe- patient
> di- predicate
> ys- undergoer
> mu- modifier
> Other markers:
> -o, -i (dummy gender markers)
> -ul past tense
> -e present tense
and a root like an- the prefixed root would be:
ajan- or awan-
with asimilated diphthong:
aan- [a?an] or [a:n] or [Aan] or [a&n] or what ever
pean- [p&:n] or [pean] or ?
dian- [djan] or [dIan]
muan- [mwan] or [mUan]
(note [ea] [Aa] or [Ia] should be diphthongs, maybe not easy ones)
with flexible prefixes
ajan- or aran- or anan-
pan- or pjan-
dan- or djan-
man- or mwan-
with glottal stops:
[a'?an], [pe'?an], [dI'?an], [y'san] and [mU'?an]
Endings fo a root like po-:
with assimilated diphtong
poo [po:] or [pOo] or ['po.?o]
poi [pOI] or [pOj]
poul [pOUl] or [pOwl]
with glotal stops
with flexible suffixes:
powo or pou /pow/ or pow
poi /poj/ or poji ['po.i] or ['po.ji]
poul or powl
with flexible root: (equal /w/ glide).
As I said I vote for flexible affixes, some flexible roots, and glide for
all other V-V combinations and unstressed <y> =3D [@] for Stop-C combinat=
-- Carlos Th