On Tue, 23 Oct 2001 17:16:09 -0000,
markodegard@... wrote:
>I've encountered the term 'singulative'. I'm not entirely sure what it
>means, nor if it's germane to this group, but here goes.
>
>AHD4 gives this:
>http://bartleby.com/61/87/S0428725.html
>
>Salikoko Mufwene gives interesting if not entirely clear discussion of
>English as a 'singulative language'. I don't understand what he means
>by "numeral classifying languages".
>http://babel.ling.upenn.edu/~nagy/nwav/WWWabs/Mufwene.html
Numerical classifying languages, like Chinese, use a number of
words/particles (between 20 - 200 pieces of them) which are mandatory
after numerals and sometimes after pronouns.
For instance, in Chinese, one cannot say:
Zhei4 shi4 yi4 shu1 - "This is a book", but one must say:
Zhei4 shi4 (yi4)ben3 shu1 - This is [one piece of] book.
Similarly:
Nei4-ge xue2sheng shi4 Zhong1guo2 ren2 : "That-GE student is China
person" -> "That student is Chinese", where <ge> is the numerical
classifier that goes with xue2sheng (just like <ben3> goes with <shu1>
"book"). A handful of words do not require a classifier, such as
<nian2> "year" (one could say they have a zero classifier).
Since numerical classifiers are also required with the numeral "1",
I'm not so sure what the relationship with plurality or singularity
(or with "singulative langauges") is.
>Someplace in my reading I remember seeing a comment that some
>langauges treat their nouns as intrinsically plural (or collective),
>unless modified in some way to make them explicitly singular. This
>seems no more and no less logical that making nouns singular in their
>base form.
We have singulatives in e.g. Welsh, where simplex words like <pysgod>
"fish(es)", <plant> "children", <sêr> "stars" have suffixed singular
counterparts <pysgodyn> "a fish", <seren> "a star", <plentyn> "a
child". Something similar in the Slavic ethnonymic suffix -an-inU pl.
-ane (e.g. SlovjaninU "a Slav", Slovjane "Slavs"). But I have never
seen a language where singulatives (or pluralia tantum) form anything
but a minority of the lexicon. As far as natural languages are
concerned I think it *is* more "logical" to have the noun
intrinsically singular (if number is marked at all).