--- In qalam@yahoogroups.com, "Peter T. Daniels" <grammatim@...> wrote:

> Richard Wordingham wrote:
>
> > > > > > > > And this is why a special name is wanted for the

syllabically organised

> > > > > > > > scripts where the option of further analysis is available.
> >
> > > > > > > What is an "option"?
> >
> > > > I don't understand why Peter Daniels is asking what he meant

when he

> > > > used the word 'option'.
> >
> > > You said "syllabically organi[z]ed scripts where the option of

further

> > > analysis is available," and I have no idea what you meant.
> >
> > PTD wrote:
> > But the Japanese child doesn't have the option the Tamil child does,
> > of internalizing the fact that all the consonants and all the vowels
> > are written separately and (well, except /u/) similarly.
> >
> > This may be generalised in the following steps:
> >
> > 1. The Tamil script is a script where a child has the option of
> > internalizing the fact that all the consonants and all the vowels are
> > written separately and (well, except /u/) similarly.
> >
> > 2. The Tamil script is a script where a child has the option of
> > further analysis.
> >
> > 3. The Tamil script is a script where the option of further analysis
> > is available.
> >
> > Is that clear enough now?
>
> Why are you saying that "the script has the option," when in the script
> it's not _optional_, it's simply _there_; but the child learning to read
> Tamil, especially if taught by syllable charts, has the option of
> discovering the structure of the syllable glyphs for themself? This is
> not a possibility in kana.

I don't think I am, but I won't pursue the matter as you seem to
understand what I intended to convey.

> > > > Yes. 'Alphasyllabary' as defined and interpreted by Bright is
> > > > probably inappropriate, as hPags-Pa is very clearly organised by
> > > > syllables. Sproat's concept seems more appropriate.
> > >
> > > Sproat has a concept??
> >
> > He expressed one.
>
> Which was ...?

... of scripts with a hierarchy of structure - roughly syllabic at one
level and roughly phonemic at a still lower level. He stated that the
term 'alphasyllabary' applied to such scripts.

> > > > So what word do you think should be used to capture the parallels
> > > > with, say, Tamil, which is not a typical Indian Brahmi script,

and,

> > > > say, Japanese kana? Tamil may have more in common with

Further Indian

> > > > Brahmi scripts than with other Indian scripts - and it's more

closely

> > > > related to the Further Indian scripts to boot!
> >
> > > Parallels of _what_ with Tamil and kana?
> >
> > Replace 'with' by 'between'.
>
> So what are the parallels between Tamil and kana? You've just recopied
> what I said about the difference.

The parallels are being composed of CV and C elements, with the CV
elements understood as units rather than being analysed. I think this
needs some heavy qualification - don't Tamils realise that a C element
is just Ca with a pulli on top, or the parallel forms of the syllables
with different preposed vowel symbols and no postposed elements?
(Tamil has actually been reducing the number of ligatures, so if it
was becoming a neosyllabary this process has been put into reverse!)

On the other hand, the parallel Tamil ee/e/ai and oo/o/au distinctions
might be viewed as an analogue of length marking in Japanese kana.

> > > More thought than is required to do the same for Hangul? Writing is
> > > conservative, and becomes more and more morphophonemic.

After reading King's article, my position is unchanged. The message
from that article is that morpheme boundaries take precedence over
syllable boundaries. In particular, vowel-initial suffixes are
written as vowel-initial syllables. In Hangul, the syllabification
effort is that of converting between orthographic and phonetic
syllabification. To me, the biggest problem seems to be agreeing on
the presence of morpheme boundaries, and is a writing issue. In Thai,
syllabification is a reading issue.

In the Thai script, the spelling rules and constraints work on
phonetic C((a)C)V(C) units - link syllables are disregarded. For 6 of
the 9 short vowels, the form of the vowel symbol depends on whether
there is a following consonant. For 3 short vowels in the high tone,
it also depends on the nature (stop or nasal) of the final consonant.
Indeed, /a/ has several special combinations with a following
consonant, which are *not* ligatures, namely those for /am/, /ai/,
/au/. There are some other similar combinations, but they may be
viewed as irregular for writing, as opposed to reading.

Whether a tone mark is required depends on the tone and the nature of
the following consonant. The tone mark goes on the last consonant
phonetically before the vowel, with the minor caveat that it may go on
a silent consonant.

Thus for writing Thai, the slightly-extended syllable is the basic unit.

> >I am talking of
> > syllabification at the level where it is related to word or morpheme
> > division, as part of the practical matter of reading with
> > comprehension. A great deal of the syllabification process in
> > reading Thai depends on recognising familiar, expected words, not
> > unlike interpreting an unfamiliar style of lettering, be it
> > handwriting or a font.
>
> You're talking about word recognition rather than syllabification
> specifically.

Thai does not mark consonant clusters, though a few three element
vowel symbols do reveal there presence. Another problem is that many
of the commoner vowels do not indicate whether they are followed by a
consonant. This makes actually identifying the syllables difficult
when reading Thai. To compound it all, Thai does not normally
separate the words of a phrase. (There are punctuation rules that do
cause some word boundaries to be marked within a phrase.) What word
recognition does is to reduce the size of the problem by bypassing the
syllabification stage and reducing the length of the strings to broken
into syllables. (Of course, sometimes what appears to be a word can
be just a contiguous subsequence of other words!)

Given all these issues, I wonder whether it does make sense to think
of Thai as an alphasyllabary. The syllables have to be treated as
units, but they do not leap out at the reader as separate units.

Richard.