Re: Hmong and semi-syllabic writing

dirk writes:
>>>>>>
On Wed, 13 Oct 1999, Paul Bennett wrote:

> There's another natlang that does that in romanisation, I can't think of the
> name of it off the top of my head, (Hmoob-something?).  There are [quite a
few]
> tones, all of which are romanised by following the syllable with a seemingly
> random letter of the alphabet.  The script writes two glyphs per syllable,
IIRC
> it's the CV component in the first glyph, with tone and (post?)nasalisation in
> the second glyph.  Someone with the very excellent [Daniels 1996] should be
able
> to clarify this a bit better.  I have a copy, but it's not at work.
The language is Hmong. The script was devised by a Shong Lue
Yang, a Hmong peasant who was reportedly illiterate. Each
syllable is represented by two characters: the first represents
the rhyme and includes tone (of which there are eight), and the
second represents the onset which may contain a cluster. Note
the orthographic order of the characters; the rhyme precedes the
onset in writing, although the syllables proceed from left to
right across the page (lines of text going from top to bottom).
There's a fascinating book by William Smalley called _Mother of
Writing_ which discusses the development of the Hmong script.
Worth reading if you can find it.
<<<<<<


I went and burrowed in The Book last night, and I found it a far more
fascinating script than I originally thought, and it does indeed have
interesting things for conlangers to consider.

For anyone who's interested, here's a functional summary, culled and
concentrated from The Book (Daniels & Bright 1996).  I've taken the liberty of
reorganising the character orders in an attempt to increase readability.



<B>The Pahawh Hmong Script</B>
Pahawh Hmong ({phajhauj hmoob} /p_ha(32) hau(32) hmON(3)/) is
an alphabet used for writing the Hmong language of Southeast
Asia.  The script was invented by a previously illiterate peasant in
1959, in the borderlands of Laos and Vietnam.

The script is romanised using the "Romanised Popular Alphabet"
or RPA.

The language has 61 consonants, 13 vowels (2 vowels include a
following /N/) and eight tones (2 tones include etxra phonetic
characteristics).  There is a strictly CV syllable structure.  The
script is written left-to-right, but each syllable, while phonetically
/CV/, is written "VC".  The fit between the symbol and phoneme
inventories is exact.  The sounds {k-} and {-au(2)} are not normally
written.  <CVV> is read as if it were {CV kV}, and <CCV> is read
{Cau(2) CV}.  There is a "redundant" {-au(2) k-} glyph which is used
to distinguish {Cau(2) kV} from {CV} where it's ambiguous.

60 of the 61 consonants are represented by using 20 glyphs
combined with three diacritic forms.  See above regarding the
representation of /k/.  The consonants are assigned to each
glyph+diacritic combination on a fairly, but not completely, irregular
basis.

The vowels of RPA are:

a    /a/
ai   /ai/
au   /au/
aw   /ai-/
e    /e/
ee   /3N/
i    /i/
ia   /ia/
o    /O/
oo   /ON/
u    /u/
ua   /ua/
w    /i-/

(/3/ = Open-mid unrounded front vowel, /O/ = Open-mid rounded
back vowel, /i-/ = i-bar)

Tones are represented by the following final consonants:

1-3 = low-high

s    1
d    12
<>   2    (No final cons used in RPA)
v    23
b    3
j    32
g    21B
m    1A

A = Advanced Tongue Root
B = Breathy-voiced

Onset consonants are represented by the following RPA sequences:

p    ph   np   nph
pl   plh  npl  nphl
(k)  kh   nk   nkh
t    th   nt   nth
tx   txh  ntx  ntxh
ts   tsh  nts  ntsh
r    rh   nr   nrh
l    hl   nl   hnl
d    dh   ndl  ndlh
c    ch   nc   nch
q    qh   nq   nqh
ny   nyh
n    hn
m    hm
z    s
v    f
x    xy
g
h
y
<>   '

The above is roughly IPA, except:
s = /S/
z = /Z/
x = /s/
xy = /C/ (c-cedilla)
r = retroflex /t/
d = glottalised /d/ (transcribed /?d/)
<> = /?/ (not transliterated into RPA)
' = 0 (no sound)
g = /N/
Preconsonantal <n> assmimilates to the following POA.
nl = /ml/
(k) does not have a glyph in Pahawh orthography

There's a decimal numeral system, plus ideograms for (10) and
(100).  There are two additional symbols, one to show reduplication
of the previous syllable, and the other to show a sentance which
should be chanted rather than read.

I've got scans of the alphabet with RPA transcriptions, and I'm in
the process of putting together a TTF, if anyone's interested.





*************************************************************
This email and any files transmitted with it are confidential
and intended solely for the use of the individual or entity
to whom they are addressed.
If you have received this email in error please notify the
sender. This footnote also confirms that this email message
has been scanned for the presence of computer viruses.
*************************************************************