Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!canoe.uoregon.edu!hammer.uoregon.edu!news.hawaii.edu!not-for-mail
From: "Bart Mathias" <math...@hawaii.edu>
Newsgroups: sci.lang.japan
Subject: Re: Watashi v boku
Date: 15 Jan 2002 20:31:59 GMT
Organization: University of Hawaii
Lines: 56
Message-ID: <i0.32.bH2q8ZZceBa9kha.NR@hawaii.edu>
References: <K6P0KMIkTgO8Ewa3@xerez.nospam.co.uk> <a1d8e4$qa5$1@news5.svr.pol.co.uk> <7jbk3u8645aarirjf6gircc10aums9dqdj@4ax.com> <3C3A3BF1.4AFAAFA7@yahoo.com> <2jik3uodm6qsgc042qcd5vhba7latiprfb@4ax.com> <a1dj67$q4mtg$1@ID-105084.news.dfncis.de> <3c41fe6b.16597016@enews.newsguy.com> <pmo08.5082$kn.615613@news1.rdc1.ky.home.ne.jp> <a8544ucv94koeevt54mvl5ack8daa2cm7t@4ax.com> <nAr08.5087$kn.616265@news1.rdc1.ky.home.ne.jp>
NNTP-Posting-Host: usrns14.dialup.hawaii.edu
Mime-Version: 1.0
X-Trace: news.hawaii.edu 1011126719 28968 128.171.66.14 (15 Jan 2002 20:31:59 GMT)
X-Complaints-To: usenet@hawaii.edu
NNTP-Posting-Date: 15 Jan 2002 20:31:59 GMT
X-Newsreader: Amiga NewsRog 1.8b (7.10.99)

"Sho" writes:

> I'm sure Bart will give us an accurate account of this, but one way
> of looking at the so-called Japanese pronous is that they are not
> as distinctly separate from common nouns as their counterparts are
> observed in most western languages. In fact, I find it very
> difficult to state their syntactic characteristics that give them a
> separate status from nouns in general. ...

I can't find my accuracy shoes this morning, but I certainly agree
with you that there is nothing to establish daimeishi as a separate
category from meishi; there are some syntactic phenomena (such as
which clause of a complex or compound sentence can "kare" substitute
for "hondakun" if "hondakun" is used in the sentence), but otherwise
they are just a subset (?) of nouns that refer to people (which for
morphological reasons are a subset of nouns referring to animate
beings, a subset of general nouns).

The late John Hinds published a paper about 25 years ago that still
bothers my conscience.  It claimed to give seven or eight reasons
Japanese pronouns were distinct from nouns, and he had sent me as
usual a manuscript version asking for comments.  I was busy at the
time (moving to Hawaii, I think) and never responded that the reasons
he gave weren't true.

I can't remember all of his criteria any more, but they included
non-modifiability  (Japanese "pronouns" are easily modified with the
same words and forms that modify other nouns; contrast English where
some pronouns can't be modified at all, and most others are
resisttant to adjectives and determiners and usually require relative
clauses) and pluralizability (*all* single-person-referent nouns in
Japanese can be made to refer to groups of people by adding
suffixes).

I'm pretty sure the only issue is substitutability, as in the
difference between 

     hondakun-ni atte kare-no hon-o kaesita.  (kare can be hondakun)

and

     kare-ni atte hondakun-no hon-o kaesita.  (kare can't be
     hondakun)

This kind of thing applies to regular nouns in English as well:

     I met Honda, and gave the idiot's books back to him.  (the idiot
     could be Honda)

     I met the idiot and gave Honda's books back to him.  (Honda's
     not the idiot)

This is the reason I questioned the need to establish a subset for
daimeishi above.  Would we establish a subset for nouns like "idiot"?

Bart