Re: [tied] The Penultimate Accent Rule

From: P&G
Message: 11037
Date: 2001-11-06

matrs but pitrn
As far as I know, all the feminines in -r take this form of the acc pl in -s
rather than -n. The declension consists mostly of words in a masculine
agentive -tr suffix. The only feminines in -tr are ma:tr, duhitr and ya:tr.
There are, however, a few other words not in -tr, eg svasr, nana:ndr, usr
and the feminine numerals tisr and catasr, plus a few masculines. The
grammarians give neuter forms as well, but they are not found in the Vedas.

There is one example in the Vedas of the feminine acc plural ma:tr:n,
instead of ma:tr:s, but it is in apposition with masculine nouns (RV X.35.2)

[Further notes: usr "dawn" has acc plural usras; and the nominative plural
forms (pitaras ma:taras etc) can be used as acc plural in the epics.]

The feminine forms undoubtedly come from the declensions in -i and -u.

Peter

----- Original Message -----
From: "Miguel Carrasquer Vidal" <mcv@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, November 04, 2001 2:35 PM
Subject: Re: [tied] The Penultimate Accent Rule


On Sun, 04 Nov 2001 13:57:29 -0000, "Guillaume JACQUES"
<xiang@...> wrote:

>
>> while "father" has, in Old Indic, the G.sg. or the A.pl. from the
>> "mother" word (Skt: pitúr after matúr < *máh2trs, pitr.:n after
>> ma:tr.:n < *máh2trns): the two paradigms were originally one.
>
>Sans entrer dans le debat sur l'etymologie de ph2te:'r (je n'ai pas
>reussit moi non plus a trouver de changement semantique feeder >
>pere) , l'accusatif pluriel de maatr est maatrrs et non maatrrn.
>C'est un acc. irregulier (je ne sais pas s'il y a d'autres cas comme
>ca en skrt), et il est probable qu'il reflete un etat ancien -- le -s
>a l'accusatif pluriel comme dans les feminins en -i et -u. En tout
>cas, l'acc. pl. de pitr n'est pas copie sur celui de matr.

You're right, it is (f.) ma:tr.:s vs. (m.) pitr.:n.
This is the same masc./fem. distinction in the Acc.pl. as in the i-
and (dev)i:-stems -i:n / -i:s, u-stems -u:n / -u:s. The masc. forms
have sandhi variants -m.s/-m.r, showing -n derives from -ns. Je ne
crois pas que cela reflete un etat ancien: il s'agit vraisemblablement
de l'influence des formes regulieres en -i:s (< *ih2-ns) du paradigme
de <de:vi:>.

=======================
Miguel Carrasquer Vidal
mcv@...





Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/