Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.duke.edu!news-feed-1.peachnet.edu!insosf1.infonet.net!internet.spss.com!markrose
From: markrose@spss.com (Mark Rosenfelder)
Subject: Re: Semantic classifiers and plurality
Message-ID: <Cz0Mw2.3w7@spss.com>
Sender: news@spss.com
Organization: SPSS Inc
References: <39p0bp$e3e@cherokee.advtech.uswest.com>
Distribution: na
Date: Wed, 9 Nov 1994 19:45:36 GMT
Lines: 22

In article <39p0bp$e3e@cherokee.advtech.uswest.com>,
Becky Root <rroot@advtech.uswest.com> wrote:
>I'm hoping someone here can help me save face as a
>cocktail-party linguist.  I have vague memories of there
>being languages which have different plural markers based
>on the semantic class of the plural noun in question.  The
>example I was thinking of dealt with categories of shape,
>e.g. different plural markers for long things vs. spherical
>etc, or something like this.  I couldn't/can't recall an
>example language and now I'm wondering if I was just thinking
>of languages like Japanese which mark a quantity term, like
>a numeral, differently according to the class of things being
>counted.  So can anyone tell me of a language which actually
>marks plural differently based on semantic class?  Many thanks!

You may be thinking of Bantu languages such as Swahili, which have nine or
more gender classes, with plural markers different in each (e.g. _kiti
'chair', _viti 'chairs'; _mtu_ 'man', _watu_ 'men'; _shamba_ 'field',
_mashamba_ 'fields').  The gender classes are not exactly "semantic
classes", but some of them do have pretty strong semantic content:
e.g. the n class contains most animal names; the m class contains many
nouns referring to humans; the u class contains mostly abstract nouns.
