Dear Yong Peng,

I don't know enough about the old Pali grammarians' treatment of
grammatical number to be of much help. But it is certainly a subject
worthy of interest and study. Grammatical number is treated in
suttas 663-671 of the Suttamaalaa (Smith's ed. pp. 735-7) which is
the third part of the Saddaniiti. These suttas are explained in the
accompanying glosses and examples are given. To give you some idea of
the scope and sophistication of treatment, I have included the 9
suttas (without the glosses) below. I haven't been able to find
anything similar in the Kaccaayanabyaakara.na but perhaps one can
find something in its many commentaries.

663 sa.mkhaa-li"ngatthaavikara.nattham uppatti vibhattiina.m.
664 ekamhi ekavacana.m.
665 ekamhi viya bahumhi pi.
666 bahumhi bahuvacana.m.
667 bahumhi viya ekamhi pi.
668 samudaaya-jaati-nissay'-ekattalakkha.nesv ekavacana.m.
669 bahumhi samudaaye bahuvacana.m.
670 kvaci jaati-atta-garusu ca.
671 apariccheda-maatikaanusandhinaya-pucchaanusandhinaya-
pucchaasabhaaga-puthucittasamaayoga-puthuaaramma.na-
tannivaasa-ta.mputt'-ekaabhidhaana-tannissitaapekkh'-
aaramma.nakiccabhedesu ca.

I think "samudaaya" is a reference to collective nouns, and "jaati"
seems to relate to class or genus. Suttas 669-671 deal with
exceptions to the general rule -- a plural where one would expect a
singular.

I hope this helps a bit.

Best wishes,
Jim

> I have a feeling that Pali _may not_ have a clearly defined concept

of

> countable noun. If it is true, I assume that in Pali, most nouns

occur

> in both singular and plural, with the exceptions of a few, e.g.

odana

> (rice), which are always treated in singular. These may in turn be

the

> predecessors of non-countable nouns in later languages. What do you

think?

>
> I would love to hear the opinions of other members on the list.
>
>
> metta,
> Yong Peng.
>