From: Patrick Ryan
Message: 41028
Date: 2005-10-04
----- Original Message -----
From: "Miguel Carrasquer" <mcv@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, October 02, 2005 2:48 PM
Subject: Re: [tied] Re: Gypsies again
> On Sat, 01 Oct 2005 22:07:22 -0500, Patrick Ryan
> <proto-language@...> wrote:
I think it is about time to change the label of this thread.
<snip>
> >> Otherwise, he makes valuable points about PIE vowel-grades.
> >>
> >> In all Proto-European (abstracting Proto-European from
> >> Proto-Anatolo-Indo-European) languages, we have abundant evidence of a
> >> phoneme which I prefer to call the Ablaut-phoneme, and indicate by *A;
> >> it
> >> has the expressions in real languages of <e>, <o>, and <Ø>.
> >>
> >> Desperate as the attempts have been to identify it, there are _no_
> >> traces
> >> of this Ablaut-phoneme in Anatolian or Indian.
>
> There are traces of it in both sub-groups (in the shape of
> a ~ a/a: ~ 0 and e/i ~ a ~ 0, respectively).
***
Patrick:
I should have stated more precisely that there are _no_ traces of the
_vocalic_ expressions of this Ablaut-phoneme in Anatolian or Indian. No one
can doubt a *V/*Ø variation based on the stress-accent occurs in Anatolian
and Indian.
I am not sure what the notation above is intended to indicate since it
corresponds to no phenomena that I can recognize occurring between PIE and
Anatolo-Indian.
So far as I know, the equation is simplicity itself: PIE *e, *a, *o > AI
<a>; PIE *e:, *a:, *o > AI <a:>.
I realize you would like to assert that PIE *o > AI (IIr, probably, for
you) <a:>.
I just do not think that you (or Brugmann) proved the relationship. So many
fine scholars have lined up on both sides of this issue that I am not sure
what _I_ really can contribute except to say that the argument has not been
made satisfactorily for me.
The example you gave to prove your point: cakara/caka:ra has a long,
checkered history of disputation. My best guess at present is that the
lengthened vowel was simply introduced to provide a means of differentiating
1st and 3rd persons.
> >> Therefore, we must consider the Ablaut-phoneme strictly a
> >> Proto-European
> >> phenomenon, developed _solely_ by the Proto-European languages.
> >>
> >> Most of us, I hope, do not doubt that an earlier stage of the language
> >> which was the basis for Proto-Anatolo-Indo-European (and others) had
> >> three
> >> vowels: *i, *a, and *u.
> >>
> >> I do not know how many list-members would agree with me (and, I
> >> presume,
> >> Miguel) in reconstructing a phase, which I term Pontic (what do you
> >> call
> >> it, Miguel?)
>
> Just "pre-PIE".
***
Patrick:
How would you distinguish it from the preceding phase, during which the
vowel inventory was *i, *a, *u?
***
<snip>
> >> We can say that there was a reconstructable Proto-Anatolo-Indo-European
> >> language through the Pontic period but after *Y palatalized dorsals and
> >> *W
> >> velarized coronals, the ways parted for Proto-Anatolo-Indian and
> >> Proto-European. Proto-Anatolo-Indian retained <a> (and <Ø>) while
> >> Proto-European changed <a> to *A and also retained zero-grade (<Ø>).
>
> The ablaut e ~ o ~ zero is pan-IE, so I cannot agree with
> that.
***
Patrick:
Perhaps we will just have to agree to disagree.
The whole point of our discussion, I thought, was whether *e~*o~*Ø was
_pan_-IE. If AI (or IIr) did not have a separate reflex for PIE *o, then we
have to scrub the 'pan'.
***
> e ~ 0 ablaut is easy to explain, as the PIE accent for the
> most part still reflects the acting cause behind the
> phenomenon: most stressed vowels become é, most unstressed
> vowels are reduced to zero.
***
Patrick:
Actually, I think it is a little more involved than that. I believe only a
_formerly_ stressed *é becomes *o; *e remains *e; *é remains *é.
***
>
> o-grade is more complicated. My current thinking is that we
> find o-grade in the following circumstances:
>
> 1) "lexical" /o/, from pre-PIE stressed long vowels **a: and
> **u:. The ablaut is o ~ e (if from **a:, e.g. pód-/péd-),
> or o ~ zero (if from *u:, e.g. *póntH-/*pn.tH-). [**i: gives
> /e:/ ~ /0/)
>
> 2) "vrddhi" /o/, from pre-PIE grammatically lengthened **a,
> **u (e.g. the perfect, or the collective). [lengthened **i
> gives /e:/].
>
> 3) "svarita" /o/, from pre-PIE posttonic lengthened **a, **u
> after a light stressed syllable [svarita-lengthened *i gives
> /e/].
>
> 4) "lengthened schwa", in the nom.sg., with
> "Szemerényi-lengthening" of reduced grade **-&C-s > -o:Cs,
> after a heavy stressed syllable (e.g. *pónt-o:h2-s, acc.
> *pónt-h2-m < **pónt-&h2-s, **pónt-&h2-m).
>
> 5) thematic /o/: the development of the thematic vowel
> before a voiced consonant (probably by lengthening).
>
> 6) "Rasmussen" or "infix" /o/, e.g. in the
> causative-iterative. The ablaut is /o:/ ~ /o/ ("full-grade"
> /o:/ is rare, because the environment is always a zero grade
> environment: we have /o:/ where lexical **a: was reduced to
> **/e/, and then (**/Oe/ > */o:/).
***
Patrick:
After mostly agreement above, I must disagree with all six points just
above.
Rather than give a cursory treatment of all six, I will only comment on 6).
If you believe a successful argument can be made for any of the remaining
five, make it and we can discuss it if you think that might be productive.
So far as I know, there is only one 'infix' in PIE, namely -*n-, and it is
not really an infix proper but a euphonic metathesis of a suffix, -*n-.
Infixes per se fly in the face of everything we think we know about PIE
structure. They are positively against the genius of the language.
There is no euphonic reason why a prefixed *o- should be incorporated into
the root syllable; and there is nothing other than speculation to propose
such a non-attested *o-prefix.
As an aside, I notice you refer to pre-PIE *u above; but you cann have both
*pre-PIE *u and *Wa as you seemed to be saying earlier unless you regard my
Pontic phase as a final phase of pre-PIE. We should be able terminlogically
to distinguish a period of *u/*i as opposed to *Wa/*Ya, do you not agree?
>
> =======================
> Miguel Carrasquer Vidal
> mcv@...
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>