Re: ML4 (was Re: TECH: Testing again etc.)

Isaac Penzev wrote:

 > Stephen Mulraney scripsit:

 >>ObConlang/ObUTF-8test: Tonight I spent an hour or so trying to implement
 >>vowel harmony on my current language (ML4, for lack of a name).
 >>Some items of vocab:

 >> äldò    |ældø|     [{ld2]:       three [<<aldh]
 >> gäțeczr |gæθʲetʃɾ| [g{T_jetS4]:  watch-tower, (generally) tower [<<gabtesor]
 >> hròlțòw |xɾølθøv|  [x42lT2v]:    to go down, descend [<<rú-lhá(m)-teivú]
 >> laŭął   |lɑwɔl̴|    [lAwOl=e]:    city, [laufe-heile]
 >> mnițò   |mʲnʲiθ2|  [m_jn_jiT2]:  four [<<amnéth]
 >> cząd̦liëhązio |tʃɔðʲl̴ʲəxɔzʲu| [tSOD_jl_e_j@xOz_ju]: officer, commander;


 > A pretty thing. Just a couple of comments: why do you use such a
 > counterintuitive orthography? _ò_ for a [2] and _ą_ for a [O] seem rather
 > arbitrary.

Thanks for your remarks. I was afraid someone would pick up on the
orthography :). It's not finalised yet, and while it's not deliberately
perverse, I do want it to have a 'flavour', and to record (where it's
not terribly ambiguous) the evolution of the language. So for example,
[w] is *sometimes* represented by _f_, where it has derived from
Tetelgen _fu_, which was [\p].

Currently I'm using _ų_, _ǫ_, _ą_ and _ę_ for [y], [2], [O]
and [3\]. The main reason is that I want ogoneks sómewhere in the
orthography, mainly due to their visual effect. But I don't have any
nasal vowels - 14 oral vowels is enough for the moment! With this usage
they at least represent something consistent, namely rounding, though
[O] sometimes reduces to [V] and I'll bet that [3\] will never be
pronounced correctly. Given this, they're not terribly arbitrary - /y/
and /2/ are often represented by some manner of _u_ and _o_, _ę_ should
really be _ë_ with an ogonek :) since _ë_ is schwa (really, [3]);
and as for _ą_ for [O], perhaps I can refer to Scandinavian _å_ for
comparison.

(BTW, is _å_ in Scandinavian langs actually [O]? I know what it sounds
like, but I sometimes have some difficulties in mapping open and
half-open back vowels to the IPA)

I *am* thinking about using _ü_ and _ö_ at least for the first two of
these vowel (with the vowel harmony, it would look rather Turkish!),
so the issue isn't settled yet. In this case, I'd replace the other
two ogonek'd letters, but the ogoneks would move in somewhere else in
the orthography. (cf. section 1.3.2 of _ML4r2: Blueprint for Progress_
"Ogoneks! more ogoneks!" :)

Now as for the _ò_ for [2] you noticed... ah... well, you see, I have
room for a great deal of vowel glyphs, since there are 12 or 14 vowels,
some with alternate historical forms (e.g. /M/ is _ū_ usually, but _ō_
when it derives from [o], or when it has been arrived at through vowel
harmony), and in addition to that, there is a separate sequence of vowel
symbols indicating that the preceding vowel is narrow/palatalised/soft.
Currently, _ų_ is palatising /y/, and _ù_ is non-palatising /y/.
Similarly for _ǫ_ and _ò_. Alas, _ò_ seems to be commoner than _ǫ_,
which seems odd!

 > And I would recommend to substitute t-comma-below with t-cedilla _ţ_,
 > because t-comma-below has a bit dubious status (nobody knows for sure if
 > they are the same or not).

Very good point, thanks. Cedillas not being available on all the
consonants I might like them on, I was using comma-below as a quick fix,
but noticed that some applications mangle them with higher frequency
than other bits of unicode. Really, I should find precomposed characters
to replace all the undercommas, even if it's not the same feature
they've been composed with. The undercommas, not being precomposed, look
ugly, like in _cząd̦liëhązio_. (Though the fact that it's adjacent to
an ogonek doesn't help.)

Incidently, the undercomma shouldn't really have appeared in _gäţeczr_
(or _mniţò_): it represents lenition, but *all* (susceptible) intervocalic
consonants will have undergone lenition, so really _äte_ should be
sufficient to represent [{Te]. Where it's really needed is in situations
like _hròlţòw_ or _cząḑliëhązio_, where a vowel has been elided.

 > -- Yitzik

By the way, here's my current notes (fairly self-explanatory) on the
vowel orthography. FV, MV and BV refer to the environment (front, mid or
back) that a vowel is accommodating to in vowel harmony.

Look and tremble, ye mighty :)


 > IPA  +YOT     -YOT
 > y     ų {ï}     ù      {ï used where [y] derives from [i] rather than [e]}
 > i      i        ì
 > ø     ǫ {ā}     ò      {ā or á used where [ø] derives from [ɑ:]}
 > e      e        è
 > ʉ      iu       u
 > ɨ      y        ı
 > u      io       o
 > ɯ    iū {iō}   ū {ō}   {ō or ó used where [ɯ] derives from [o:] & not [u:]}
 > ɜ      ië       ë
 > æ      iä       ä
 > å   ią {iau}   ą {au}  {au used where [å~ʌ] derives from [au]}
 > ɑ   ia {iâ}    a {â}   {â used where [ɑ] derives from [a:]}

 > Vowels of more restricted distribution:
 > ɞ: MV harmonic of [ø], [å]: write as [ię] (+YOT) or [ę] (-YOT)
 > ʌ: write as [å]; or when BV harmonic of _e_, as [iĕ] (+YOT) or [ĕ] (-YOT)

As you can see, despite the front-back vowel harmony requirement, I
try to preserve both roundedness and palatising-ness (of course, since
that's really a feature of the preceding consonant).

--
Stephen Mulraney  ataltane@ataltane.net  http://livejournal.com/~ataltane
If a man does not keep pace with his companions, perhaps it is because he
hears a different drummer. Let him step to the music which he hears, how-
ever measured or far away.                        --  Henry David Thoreau