Re: NATLANG: Chinese parts of speech (or lack thereof)

On Tue, 10 Aug 2004 14:59:33 +0200, Andreas Johansson <andjo@...>
wrote:

> Quoting Paul Bennett <paul-bennett@...>:
>
>> On Mon, 09 Aug 2004 13:59:43 -0400, John Cowan
>> <jcowan@...>
>> wrote:
>>
>> > In the meantime, various other systems have been devised more or less
>> > independently: these turn out to always be syllabic.
>>
>> Although arguably in the less-independent column, there are Sorang
>> Sompeng
>> (an abugida), Ol Cemet' (an alphabet), and Varang Kshiti (which is a
>> defective abugida), all invented for minority languages in India.
>
> "Defective" abugida? How does that work?
The unmarked vowel may be any of "a", "e" or "o", even though the script
has characters for all three vowels, as well as the other vowels needed
for the underlying language (called "Ho"). For all I know, that might be
structurally sound according to the needs of the Ho language, but it's not
the normal behaviour in abugidas, and is an underspecification rather than
an overspecification, so I called it defective.





Paul

Reply