On Mon, Apr 05, 2004 at 04:19:01PM -0400, Trebor Jung wrote:
> What are those characters? They aren't readable by me.

Which is one of the reasons I'm reluctant to include them in the
orthography.  I'd like to stick to Latin-1 if possible.

The characters are edh, eng, thorn, and theta, respectively.  At
least edh and thorn are in Latin-1, so these should be legible:
<�>, <�>.  I might end up admitting them, but I'm reluctant to
introduce unfamiliar letters.

> In any case, you could use edh for /D/, n~ for /N/ (Tolkien did this in
> older versions of Quenya), and thorn for /T/.  Or: dh, ng, th.

Yeah, like I said, I'm reluctant to introduce new letters, and I'd
rather avoid digraphs, too.  Which pretty much leaves diacritics;
I was already considering using <�> for /N/.

> What are you using <q> and <x> for?

Nothing.  Wouldn't mind using 'em, but they don't map to the sounds in
any intuitive way.

> <q> could represent /N/;

Ah, no.  Too much of a leap for me.  If I were going to use <q> for one
of the sounds, I'd prefer [T] over [N]; <q> at least sort of looks like
a backwards decapitated thorn. :)

> if you used <x> for /S/ (as
> in Maltese, Basque, and some Mesoamerican languages), you could use <c> for
> /T/ (as in Castilian). And maybe you could use <dd> for /D/, as in Welsh.

Castilian or not, <c> for [T] doesn't feel natural enough.  And if I
were going to use a digraph for [D], I'd use <dh> rather than <dd>.

> "that really indicates quality rather than quantity: /e/ = [E], /e:/ = [e],
> /i/ = [I], /i:/ = [i], /o/ = [O], /o:/ = [o], /u/ = [U], /u:/ = [u].
> Interesting system.

Not really.  Pretty much identical to Classical Latin.

> <a> could represent [V]

Which is again in line with one interpretation of Classical Latin
pronunciation (the other being that short and long <a> differ only in
length and not in quality).

> and ae-ligature could represent [&].

Consistent with IPA.  I like it, but a distinction between <ae> and
<�> might be a little too subtle, especially in email where some
Exchange servers substitute one for the other in passing (don't get me
started . . . ).  But I'm not sure the morphology allows <ae>; I'll have
to check.

> Albanian writes it as e-trema. Unless you want as few diacritics as
> possible.

I don't mind diacritics per se, though using trema for schwa would mean
it was unavailable as a Latin-1 substitute for the long-vowel macron.  I
guess the circumflex would do for that, though.  However, <�> doesn't
look very schwalike to me, because (despite its usual symbol), the schwa
sound doesn't feel e-like to me; it's too far back.  I would consider
<�> possible, because to me it's suggestive of [I\] or [i\], which are
allophones of the schwa in some environments in some 'lects of English.
I would also consider oe-ligature a plausible schwa, but it's not in
Latin-1; and its often-equivalent <�> doesn't work for me because the
schwa is no more o-like than it is e-like; it's in between.

So if I adopt the suggestions as indicated above, I'm left with this:

[V]     <a>
[&]     <�>
[D]     <�>
[N]     <�>
[S]     <c>
[T]     <�>
[Z]     <j>
[@]     <�>
[j]     <y>

Let's try a sample text.  Other than the above, letters have their CXS
values.  Vowels in dipthongs are generally long but unmarked for length.
There is stress-emphasis; except where marked explicitly with an acute
accent, it falls on the penultimate schwaless syllable.

Proposed:       Ail s�j�l u� hoz��,
CXS:            ail 'sa.Z@l UN 'hO.z@N

Proposed:       nou wel �oun d� ��?lg.
CXS:            nou wEl Noun du 'aN.@lg

Proposed:       Nou wel p�mg�n p�m ���.
CXS:            nou wEl 'pum,g@n p&m TiN

Proposed:       Nou wel yur d� b�� hul �� ulc, gvi�k a� uk u� u� hoz��.
CXS:            nou wEl jUr du b&N hUl aN UlS, gvITk VD Uk UD UN 'hO,z@N

Proposed:       Buz �� ail s�g sel k�gou,
CXS:            bUz &T ail sig sEl k@'gou

Proposed:       a�g selb�z �� ail �u��,
CXS:            VNg sEl'bUz &T ail TUND,

Proposed:       gvi�k a� j� haz selbuz�� jei� h� haz �u�g �bo��k ��.
CXS:            gvITk VD ju hVz ,sEl'bUz.@N ZeiD hi hVz TUNg @'bONTk &T

Proposed:       A�g gei�k rok �� w�rg k� kontkouf��,
CXS:            VNg geiNk rOk &T wurg ki ,kOnt'kou.f@N

Proposed:       g�k g�ruz�l �� sl�n j� �zur y��,
CXS:            g&k g@rUz@l &T sl&n Zu uzUr j&N

Proposed:       sel wel� u� j� p�mg�n u�g j� tai�l u�g j� brel� selo��l.
CXS:            sEl wElD UD Z@ pumg@n VNg Z@ 'tai.@l VNg Z@ brElu [log in to unmask]