Re: Fun with prenasalized stops.txt

From: tgpedersen
Message: 47227
Date: 2007-02-03

--- In cybalist@yahoogroups.com, "Richard Wordingham" <richard@...> wrote:
>
> --- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@> wrote:
> >
> > --- In cybalist@yahoogroups.com, "C. Darwin Goranson"
> > <cdog_squirrel@> wrote:
> > >
> > > I like this idea of prenasalization. The m/w thing seems a nice
> > > touch, and your theory has some interesting parts ot it, but I
> > > do have a few questions about it.
> > > 1) How would the /b/ turn into a /w/ around an /m/? Where would
> > > the lip-rounding come from?
>
> The lip-rounding is there in /b/. Lenited /m/ has become [w] in
> Irish, so there is no problem there.

Of course, I forgot. Also in Scotch Gaelic, says my "Gaelic without
Tears". Which reminds (I'll put some big quotes here, since I'm not
completely convinced about my idea):

Sihler: New Comparative Grammar of Greek and Latin, p. 255
"
16. Nominative/Accusative dual. The evidence for the dual case-markers
in general is of poor quality. Many IE languages have no dual category
at all; and even when they do, they provide fragmentary, ambiguous, or
otherwise unsatisfactory evidence.
The best-attested case is the nom./acc.m./f., but its forms lead to
incompatable reconstructions:
The vowel stems show lengthening alone: so o-stems in G lúko:, Ved.
devá: 'two gods'; i-stem inflection in Ved. páti: 'two lords', OCS
gosti 'two guests'; u-stem in Ved. su:nú: 'two sons' = OCS syny. A
suffix *-e is indicated in consonant stems: G patére, pherónte; the
same is found in OLith. (du z^mune 'two persons'), and OIr. attests an
ending consisting of a front vowel: caraitL /kar&d'/ 'two friends'
continues *karont- plus *-e: or *-i:. In InIr. the o-stem form has
invaded the consonant stems, so Ved. pa:da: 'the two feet'. Evidence
for an earlier state of affairs has been seen in the dvandva com-pound
ma:tara-pitara: 'parents' (Pa:n.ini). Now, this is neither a Vedic
dvandva, that is with both components marked with the usual dual cases
(as in RV ma:tára:-pitára: 'parents'), nor is it either of the two
later types—with the first element in stem form (bhra:tr.-bhagini:
'brother and sister') or in the nom.sg. (pita:-putrá:u 'father and
son'). Since the first element of a dvandva was originally in an
overtly dual case, it is probable that ma:tara- here is the original
nom./acc.du. that was supplanted by the a-stem ending -a:(u).
The Ved. -a thus slenderly attested, the -e of OLith., and the front
vowel of OIr. can be gathered up under PIE *-e. On the other hand, the
lengthening of the vocalic stems is most easily accounted for by a
final laryngeal. The G forms in -e, which might reflect either *-e or
*-h.1 are ambiguous. Likewise ambiguous is the o-stem dual, as either
*-oh1 or **-oe will give the same outcome in all IE languages (cf.
nom.pl. *-o:s < **-oes, 260.1). The Indic, Lith., and OIr. cons. stem
forms cannot reflect *-h1, nor can the i- and u-stem forms readily
continue **-ie, **-ue. Nevertheless, each of the two possibilities has
its proponents among modern scholars as the sole etymon.
a. For what it is worth, an ending *-e would be unique among nominal
case markers in containing no consonant at all, which might support
amalgamating the two candidates for the ending into ablaut variants of
a single ending, *-h1e and *-h1.
b. Indic evidence for original -a is perhaps slightly stronger than it
first appears: the importation of the a-stem ending -a: into the cons.
stems would have been facilitated if the original ending were -a <
*-(h1)e, rather than *-i < *-h.1 as it would mean nothing more
complicated than lengthening a vowel.
c. There is some slender evidence (370a) that the marker of the
o-stem and pronominal dual was actually *-h3, not *-h1. If accepted,
this would of course forestall the speculation floated in note a, above.
"

Burrow: The Sanskrit Language, p. 240-41
"
Nom. Voc. Acc. Dual. This case was made by various formatives in IE,
according to the type of stem. The ending -au, -a: of Sanskrit was
originally, from the evidence of the related languages, confined to
the thematic stems (Gk. lúko:, Lith. vilkù, O. SI. vlUka) from which
in Sanskrit it has been extended to other types of stem (pá:dau,
pitárau, etc.). In these latter classes Greek and Lithuanian have an
ending -e (me:tére, áuguse). It has been suggested that this
termination, elsewhere replaced by -au in Sanskrit, is preserved in
the dual dvandva ma:tara-pitarau ' parents ' which the grammarians
quote as a northern form. The termination is regularly -au in
classical Sanskrit, but in the Vedic language it varies between -au
and -a:. As a general rule -au is used before vowels, becoming -a:v,
elsewhere -a:. Some such variation must go back to the IE period, and
it is the latter form which has been generalised in the related languages.
The feminines in -a: and the neuters take the termination -i:. This
agreement is another sign of the close relations existing between the
feminine a:-stems and the neuters. Examples from neuter consonantal
stems are : vácasi:, cáks.us.i:, ná:mni:, br.hatí:. In the thematic
neuters and the feminines in -a: this -i: combines with the vowel of
the stem to form the diphthong -e : yugé, s´r´.n.ge ; ás´ve, séne.
This -i: is found also in Old Slavonic, though rarely (imeni, têlesi),
and Slavonic shows the same diphthong in o- and a:-stems (selê, ro,cê)
a form which then spreads to consonantal stems (imenê 'two names ').
The i- and u-stems make the dual by lengthening the vowel of the stem
: páti:, su:nú:, and with them are to be classed the feminines in -i:
of the deví: type in the Vedic declension (du. deví:, later devyaù).
This type is ancient, appearing also in Avestan (gairi, mainyu),
Slavonic (O. Sl. po,ti, syny) and Lithuanian (nakti, sunu).
"

Thurneyson: A Grammar of Old Irish, p. 149
"
[Nasalization takes place] after the nom. voc. acc. gen. neuter dual,
and after the dat. (all genders) of the numeral ' two ' ; but
nasalization is not found after the dat. dual of nominal forms.
"

This Old Irish nasalization takes place after former previous nasal,
and its surprising it should happen here. The only ending that looks
promising for postulating an ending in a nasal is *-au/*-a:, which
would then be *-onW/-o:, and perhaps *-inW/*-unW for the -i and
-u-stem's *-i: and *-u:. Why it is limited to the neuter in Old Irish,
I don't know. The Dsg of "two" is Skt. dvabhyam.

Now, if this NVAsgDu *-nW was true, the condition that the o-variant
of the thematic vowel appears before voiced consonant could be
strengthened to 'before nasal', except for the troublesome Nsgm *-os.
Jens has proposed that it was actually *-oz, with a unique /z/, but
how about *-oNs?


Torsten