Re: Abugidas (was: Chinese writing systems)

Roger Mills wrote:
> If the language
> perchance allows word-initial V syllables, then you would need either 5
> distinct "initial-V" symbols (like Devanagari IIRC), or a dummy
> vowel-carrier symbol (which by itself would represent the inherent V)
> modifiable with the V diacritics (like Buginese/Makassarese whose V-carrier
> by itself represents /a/)
It seems to me that another possibility exists, to use a certain
consonant (perhaps /h/) for V syllables, so that, for example, _h(i)k_
could be either _hika_ or _ika_.  Since many writing systems permit a
certain degree of ambiguity, that seems naturalistic enough, especially
if there were few, if any, perfect minimal pairs, so that a knowledge of
the language would tell you if _h(i)k_ was "hika" or "ika"

> Adapting such a system to a
> language that allows more complex syllables (like English!) produces a lot
> of problems.
Common Kassi (the ancestor of Uatakassi and the now-extinct Northern
Kassi) had a pretty simple (C)V(C) syllable structure, (where the
initial C was *almost* obligatory, and the final C could not be a stop),
with 6 vowel - /i e a o u @/ (/@/ is transcribed ë), and only CV
characters.  The modern language has become
(C)(l,w,y)V(length)(s,f,v,z,l,n,*) with three vowels /i a u/.  The old
/i/ and /e/ merged, as did /o/ and /u/.  /@/, for the most part, was
lost, but sometimes merged with /a/.  The modern /Ci/ characters are
derived from /Ci/ and /Ce/, with any given consonant, one or the other
was chosen, thus /fi/ was kept and /fe/ lost, while /ke/ was kept, with
/ki/ lost (I think; the exact examples might be wrong, but the principle
holds), likewise for /Cu/ from /Cu/ and /Co/.  /Ca/ comes entirely from
the old /Ca/ syllables.  Most of the /C@/ syllables were lost.  There
were also two consonants in CK which do not exist in Classical
Uatakassi, /q/ and /h/.  The /q/ syllables were lost, and the /h/
syllables gave rise to the pure vowel characters (even in CK it could
ambiguously represent hV or pure V)

First, a note on transcriptions: when describing the characters, capital
letters (e.g., KA or PLI) represent the characters themselves, while
lower case letters (e.g., s or l) represent diacritics, * represents the
gemination diacritic, and (lv) represents the length-marker, and hyphens
clarify where a new character begins.

Now, in the original writing system, CVC was written with CVC@.  CVC@
was written the same way, causing ambiguities.  Scribes invented a way
of showing CVC as distinct from CVC@, they would write the C@ part
underneat the CV part for CVC.  Thus, _kas_ would be written KAsë, while
KA-SË represented "kasë".  Thus, the /C@/ characters came to be used as
diacritics, as they were simplified in subordinate position.  The long
vowel marker in Classical Uatakassi is derived from _kë_, as /k/, /g/,
and /q/ syllable finally (which became possible only after the loss of
/@/) became /x/, /G/, and /X/, and still later were lost, making the
previous vowel long.  Thus, /tak@/ -> /tak/ -> /tax/ -> /ta:/ (in the
characters themselves, TA-KË -> TAkë -> TA(lv)), since the /k/-derived
forms were most common, that diacritic was chosen over the /g@/ and /q@/
derived forms.  Gemination is derived from the /r@/ character (later,
/r/ became /l/), syllable-final /r/ assimilated to the following
consonant creating geminates (e.g., karta -> katta, KA*-TA).  The
non-diactriticized /r@/ became the character (*not* diacritic) for
syllable-final /l/ in the classical script.  (E.g., /karta/ -> /katta/,
but /kar@ta/ -> /kalta/, KA-L-TA)

CLV syllables are derived from ligatures of C@ and LV syllables, which
reflects the origin of that syllable type (/k@ro/ -> /kro/ -> /klu/),
after the initial schwa loss, people came to write the /C@/ part above
the LV part, which later merged into a new character.  Thus, one can see
patterns in the CLV characters.

Finally, CyV and CwV syllables are written simply with Ci and Cu
followed by the V, e.g., /kwa/ is KU-A.  However, it's not uncommon to
see the V characters written as diacritics of the Ci/Cu characters
(e.g., /kwa/ can be KU-A or KUa), likewise syllable-final L character is
sometimes diacriticized.  There's still a strong tendency to write
syllables as single glyphs, hangul-style.

There are also a special set of fairly-recent characters, whose
romanizations are t'i, d'i, k'i and gi, which represent secondary
palatizations of /t/, /d/, /k/, and /g/ (long ago, /gi/ became /ji/ and
then /i/, and thus the /g/ row traditionally had only /ga/, /gu/, /gli/,
/gla/, /glu/).  They are formed by the ta, da, ka, and ga with a
diacriticized /i/, and are realized as [ti] or [tsi], [di] or [dzi],
[ki] and [gi].  The original /ti/, /di/, /ki/, and /gi/ became
respectively /tSi/, /dZi/, /Ci/ and /i/.  The "proper" way or writing,
e.g., t'ia ([tja] or [tsa]) is TAi-A, but is sometimes written TAia.
"Tia" ([tSa]) is written TI-A, but sometimes TIa.

Added to this complexity are the diphthongs, which are properly written
as separate characters, thus "tai" = TA-I, but the second element is
sometimes diacriticized, e.g., TAi.  So, "t'iais" ([tsajS] or [tjajS],
for example, may be written TAi-A-Is, or TAia-Is or TAiais, with a
whopping *four* diacritics, two of them the same.  Actually, now that I
think about it, *five* diacritics is possible, as diphthongs become long
in monosyllables, thus, a words like "t'iaais" ([tsa:jS] or [tja:jS]) is
theoretically possible, and would be written in one of the following
ways: TAi-A(lv)-Is, TAia(lv)-IS, TAia(lv)is

--
"There's no such thing as 'cool'.  Everyone's just a big dork or nerd,
you just have to find people who are dorky the same way you are." -
overheard
ICQ: 18656696
AIM Screen-Name: NikTaylor42