Re: [tied] Re: Early PAlb Depalatisations of k', g' > k, g

From: Miguel Carrasquer
Message: 39072
Date: 2005-07-05

On Tue, 05 Jul 2005 14:11:58 +0000, alexandru_mg3
<alexandru_mg3@...> wrote:

>--- In cybalist@yahoogroups.com, "Patrick Ryan" <proto-language@...>
>wrote:
>>
>> ----- Original Message -----
>> From: Miguel Carrasquer<mailto:mcv@...>
>> To: cybalist@yahoogroups.com<mailto:cybalist@yahoogroups.com>
>> Sent: Monday, July 04, 2005 12:47 PM
>> Subject: Re: [tied] Re: Early PAlb Depalatisations of k', g' > k,
>g
>>
>>
>> <snip>
>>
>> PIE *e: simply does not yield *o: in Balto-Slavic.
>>
>> ***
>> Patrick:
>>
>> Nor does it in any IE language!
>
>
> Well, well ...once again "quick answers" for a "delicate story"
>(after all the posted messages here especially that ones posted by
>Sergei to come back to such a simple conclusion...sound very strange
>for me):
>
>I will try to resume here the situation:
>
> 1. we need here an e/i in the root because the depalatisation took
>place.

You mean: "we need here an e/i in the root because the
depalatisation DID NOT take place."

We don't.

> (=> Miguel's supposed form *k'low- will give Lith. klau- as in Lith
>klausyti)

And *k^leu- gives klau- as well, as in Lith. kláusti.

> 2. we don't have a SIMPLE e: here but *e:w from *ew(H)

What we have here is heterosyllabic e:u, just as we do in
lêviti (Derksen: *leh1u-) and plêva (Derksen: *ple:u-ah2).

> a) Kortlandt's derivation:
> --------------------------
> Miguel, Brian see below Kortlandt's explanation (even you like it
>or not, from the pdf that Sergei indicayted you):
>
> Kortlandt:
> "In the case of Lith. Slove: 'glory', OCS. slava, which cannot be
>separated from Latin clue:re 'be mentioned', I think that we have to
>start from a form *k'le:uH1,

"Start from" does not exclude the possibility of o-grade.

>with analogical lengthened grade
>as in Gothic qe:ns 'wife', PIE. *gwenH2, or Gr. e:par 'liver', PIE.
>*iekwr. "

You mean Grk. he:par. I don't think any of these lengthened
grades is "analogical".

> b) Derksen's derivation:
> ------------------------
> Following Kortlandt, Dersken has proposed *k'le:uH- too
>(->see on Leiden -> the Lithuanian dictionary).
> (Maybe Dersken has changed his opinion after the "the 3rd Level
>Contact with Miguel on Zagreb Airport" ...but "currently the Leiden
>site still wasn't informed" about this)
>
>On Leiden k'le:uH- is the reconstructed PIE, Miguel, not the PIE root
>so please don't create again confusions

If that was the case, Derksen would have told me.

>=> I have hoped that after you have read Kortlandt you have
>understood better that Dersken's has kept Kortlandt's derivation
>k'le:uH- -> so we have quite the reconstructed form on Leiden not a
>PIE root...and I'm convinced that you have understood this too,
>however you have continued to create confusions regarding the meaning
>on some fields that exists on Leiden.
>
> c) Sergei's posted rule ew > aw
> -------------------------------
> Please rememnber also, that Sergei correctly indicates you the rule
>ew > aw

Sure, as is well known.

> So based on Kortlandt (followed by Dersken), the original PIE
>contained *e:w from *ewH (and this finally gave a:w in Balto-Slavic)
>
>In fact I found that would be more appropriate to explain the long
>vowel in place of "an analogical lengthened grade" by considering
>here "a laryngeal metathesis" :
>
> *ewh1 > *eh1w > e:w
>
>and maybe also that the laryngeal wasn't h1 but h2 (but this second
>part is still only a pure supposition from my side needed to be
>further checked):
>
> *ewh2 > *eh2w > a:w
>
>Anyway in this context, PIE *e:w gave Balto-Slavic a:w (maybe similar
>with ew > aw indicated by Sergei)

Now you're completely off into ga-ga-land.

When Kortlandt says *e:uh1, he means *e:uh1, certainly not
*eh1u. The whole article *is* about long (non-laryngeal)
vowels in Balto-Slavic, for Christ's sake. Likewise, when
he says *h1, he means *h1. And when Derksen says /o:/ he
means /o:/, not /ah2/.

You're right that the Lithuanian form must come from PBS
*s^la:vija:, with an /a:/, but that /a:/ certainly has
nothing to do with *h2 or with the BS soundlaw *eu > *au
(*jau when not heterosyllabic). And certainly not with both
at the same time!

>and the derivation doesn't raise issues because next:
>
> BSl. a: > Lith. o
> BSl. a: > Sl. a
>
>
> Best Regards,
> Marius
>
>
>P.S.: What is wrong regarding Dersken reconstruction is the Balto-
>Slavic reconstruction that is s'lo:u- (Dersken himself put an (?)
>after his Balto-Slavic reconstruction).

No he didn't. He put a glottal stop there, an optional one.

> The correct Balto-Slavic reconstructed form is *c'la:u

Indeed, and I have explained why.

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