[tied] Re: Nominative: A hybrid view

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:

> On Fri, 06 Jun 2003 23:01:40 +0000, Jens Elmegård Rasmussen
> <jer@...> wrote:
>
> >If 'in the house' is *dé:m as we expect and Avestan confirms,
>
> Can also be *dó:m(i).

Why would it be such a thing?

>
> >then 'at the foot' should be *pé:d. I do not find that, but the
> >unexpected long vowel of *pe:dsu must come from somewhere in the
> >paradigm, and of course the obvious locus is the locative

singular

> >where we do expect to get the long vowel.
>
> I don't expect it, and I don't see it.

I did not expect *pe:dsu, but I find it anyway, therefore I have to
clear a spot for it to arise. The locative singular is just fine and
leads to a consistent analysis of the overall picture.

> In fact, there's another reason why
> the loc.sg. cannot be simply the lexical root. In the r/n-stems,

the

> loc.sg. ends in -n(i). The *-i is secondary, but the fact that *-

n does

> not become *-r proves that there was a vowel present (before zero

grade).

You already know my explanation of this fact. In (secondary) word-
final position /nt/ is reduced to /n/ which is what we have in the
locatives of r/nt-stems. Derivatives from them have -nt- (Latin -
mentum and all that, remember?). The nom.-acc. in -r reflects an
even older Auslautgesetz. Still, this is less certain than what it
is being used to support.

> So if the loc.sg. is **wa:dán-a > *wedén > *wed(e)n-i, then in the

case of

> "foot" it must be **pa:d-á > *pedé > *péd-i, which is what I

expect *and*

> find.

Then OIr. ís and Alb. përposh are just being disregarded? And then
what *is* Avestan daNm ? You said it could be *do:m or *de:m. Why
then are you making a set of rules where it can't be either? Surely
the weight of Skt. padí is light since it shows the same stem as all
the weak cases.

> >[...]
> >
> >> > Accepting all of that, I can now
> >> >utilize it to explain the o-type of suffixed neuters, such
> >> >as 'water'. If 'water' can have a young collective in húdo:r,
> >could
> >> >*wód-r perhaps be based on the old type of collective? It

would

> >look
> >> >like this: *wé:d-r-h2 > *wé::drh2 > *wó:drh2 > *wódrh2.
> >
> >
> >> But the lengthening worked _before_ zero grade, when the word

was

> >still (in
> >> your notation) *wé:d-or-h2...
> >
> >That's what I used to believe, but I have wised up since.

Actually,

> >the suffix only had a vowel in stems that would have ended in at
> >least three consonants without the suffix vowel, i.e. exactly

like

> >the i- and u-stem types in the distribution seen by Szemerényi

(and

> >explained by myself). Since the root of 'water' ended in a single
> >consonant, even adding -r-/-n- would not make the stem so long as

to

> >demand a vowel to be inserted.
>
> The vowel was part of the suffix. It was not inserted, but

deleted by zero

> grade in *wodr. < **wa:d-an.

I tried to prove just that; the result came out negative. So I am
taking the consequences.

> In the collective, it was lengthened, so that
> we get *wad-á:n-h2 > *udó:r. Didn't you just say that the loc.sg.

has the

> accent on the final vowel of the stem? So what is the stem in

Skt. loc.sg.

> udán?

That is /-en-/ taken from the type that did insert -e-. This is the
renewed form. The old regular form would have been *wéd-n-i; Avestan
gives us va{ng}ri 'in the spring' from *wé:s-r. The Slavic adj.
vesnInU 'of the spring, occurring in the spring' is taken to be
based on the locative meaning 'in the spring'; that could in fact be
*wésni.

>
> >> Does this explain all the neuters with *o/*e, such as:
> >>
> >> *wodr, *wednos "water"
> >> *smok^wr, *smek^nos "beard"
> >> *g^onu(r), *g^enwos "knee"
> >> *h2ost(Hi), *h2est(H)nos "bone"
> >> *pok^u(r), *pek^wos "livestock"
> >> *mostr(g), *mestnos "brain, marrow"
> >> *h1oudhr, *h1udhnos "udder"
> >> *k^ouh1r, *kuh1nos "hole"
> >> *h2ongl, *h2englos "charcoal"
> >> *k^okWr, *k^ekWnos "excrement" (also *sok^r, *sek^nos)
> >> *sókWt(Hi), *sekWtHnos "upper leg"
> >> *stomn, *stemnos "mouth"
> >> *wosr, *wesnos "spring"
> >> *h1osr, *h1esnos "autumn"
> >> *doru, *derwos "tree"
> >> *woh1r, *wehros "water" ?
> >
> >To the extent that they have been correctly reconstructed, I'd be
> >inclined to say yes. In *h2óst-h2/*h2ast- we even have the
> >collective marker sitting on the word in Skt. ásthi.
>
> So why is it not a trace of it seen in other forms?

What "other forms"? The collective marker was only present in the
nom.-acc., so it should not be there. And if the status of
collective has been lost, there was no need for it. You quote it
yourself in sákthi; you could add dádhi.

As far as I understand

> your 3-consonant rule, it doesn't make sense either for a lot of

these

> forms.

The rule should not be given this crude form, it is in need of some
refinement, but the facts are hard to get. It doe give the main
principle however as seen in the types of i- and u-stems. I am only
adding the same two types for r/n-stems, and along the same basic
principles. That would normally be regarded a progress.

>
> >One could
> >assume the same for há:rdi 'heart' which basically alternates

*k^érd-

> >/k^rd- and is neuter.
>
> It alternates *k^e:r(d), *k^r.d(i)- because the long vowel was

here **i:

> (*ki:rd, *ki:rdiás > *k^é:r, *k^r.di(y)és).

Where did all that come from?

> >> - the neuters are explained away as "old" collectives
> >
> >Only where they have taken the special form with -o- peculiar to

the

> >collective. Those with /-e:-/ are the corresponding non-

collectives.

> >
> >Why does it bother you less that húdo:r must also be derived from

a

> >collective form?
>
> Because it has the structure of a collective, and the lengthening

of o > o:

> caused by the collective marker *-h2.

But if I show that *wódr and *h2ósth2 also have the structure of
colectives as caused by the collective marker *-h2, then that is not
acceptable to you? Still, there is no difference, so I'll just go on
assuming it anyway, with or without your permission.

>
> >> On the other hand, if we take a fresh look at the data, we see
> >that the
> >> lengthening caused by *-s or *-h2 is already accounted for

(nom.

> >*po:ds
> >> with long *o: vs. acc. *podm with short *o, collective *udo:r

with

> >long *o
> >> vs. *wodr with short *o). It seems unnatural to apply the
> >lengthening
> >> twice. Moreover, why was it not applied twice in the HD type
> >(*p&2tér-z
> >> (1)-> *p&té:rz (2)-> *p&2té::r = p&2tór ?).
> >
> >Hey, this is a gross mistake, have you been thinking that way all
> >along, or is it just a sign of fatigue? The lengthening rule is

not

> >applied twice anywhere.
>
> OK, that was unfair: it is applied twice in my way of looking at

the

> evidence, not yours. In a form like *po:ds, I see lengthening

applied once

> by *-s (o > o:), so I can't use it anymore to explain the *o

itself, which

> as expected, is not long in the accusative.

Well, if you lengthening twice under your rules, and not under mine,
and you don't like that, choose mine.

> >Why did it not do that in a verb?

> It did. The perfect/stative shows the exact same distribution as

the noun,

> i.e. a few cases of e:/e, and a lot more of o/e.

Where do we find an o-type Narten ablaut opposing active sg. /ó/ to
active du./pl. and middle voice /é/? Especially some examples of
active /ó/ : middle /é/ would be welcome. I do not find such a type.
It ought to be the dominant type of Narten verbs if there were any
truth to the idea of /o/ being underlying.

> /o/ is the normal Ablaut
> grade of the perfect, but a few verbs show /e:/. The /e/-grade of

the

> (weak) plural is seen in Hittite. Elsewhere the situation is

rendered less

> clear by reduplication: the reduplication vowel has /e/-grade

(instead of

> /i/) and the root has zero-grade:
>
> **ti-tá:wd- > *ti-tówd-h2 + -a : Skt. tu-tód-a
> **tí:-tawd- > *te:-tud-m + -é > *te-tud-mé : Skt. tu-tud-má
>
> In the 3rd. plural, at least Indo-Iranian did initially not shift

the

> accent to the ending (*té:-tud-r.s -> tutudúr), and we find traces

of the

> reduplicative vowel *e:.

You are completely disregarding the strucures actually shown by the
IE verb: ó/zero is found in three reduplicated sets: perfect,
intensive, reduplicated aorist. These are all reduplicated, and the
weak form is zero in them all. An unreduplicated ó/é type does not
exist. I know that a grand theory assuming it anyway is spreading,
but the tonal facts of Balto-Slavic reflecting the reduplication in
precisely the verbs in question have just been disregarded; the
criticism cannot see such fine shades, so they're getting away with
it. We should know better.

Jens