Re: Latin -idus as from dH- too

On Sat, 15 Mar 2008 20:23:06 +0100, Piotr Gasiorowski
<gpiotr@...> wrote:

>On 2008-03-15 19:10, alexandru_mg3 wrote:
>
>> To make clear the situation for anybody here:
>> ==============================================
>> 1. Both Piotr and Miguel's abandonned (implicitly) 2 days ago Mrs.
>> Olsen's theory (as it was presented by Mrs. Olsen) => after both said
>> for years 'how impressed they are about this theory' => you can find
>> their old assertation on this forum too
>>
>> 2. But none of them wrote this explicitly here...as usual...
>
>I can't speak for Miguel, but as for me, I haven't changed my mind about
>Olsen's theory. I still like it.

And I like it more than I used to, now that I've taken a
closer look at it. There is no other way to explain the
constellation of e:-verbs with nouns in -or and adjectives
in -idus in Latin. I wasn't happy with all the participles
in -a:tus being secondary restorations (although the
materials for such a restoration are admittedly there), but
perhaps I've found a solution for that (if a form stressed
*-áh2tos is not affected by Olsen's law).

>> 3. Then : they have tried to improve this theory (in fact only
>> Miguel) with a new version of it => trying to eliminate the cases
>> that didn't fit with the rule by adding different other rules:
>> - vocalisation & non-vocalisation of laryngeals (that was at Olsen
>> too: and I was waiting for them to arrive there also...)
>> - stress position (Miguel's supposition)
>
>Improving and fine-tuning earlier models is precisely what science is
>mostly about. What Miguel has suggested is an additional structural
>constraint on Olsen's rule.
>
>> the intention is to eliminate all the cases that Mrs Olsen theory
>> cannot cover at all
>
>I plead guilty as charged: I like Miguel's proposal since it improves an
>already attractive theory.
>
>> 4. Next I showed them that:
>> ph2te'r and dHug2te'r , accented on the last syllables, are in
>> contradiction with Miguel supposition
>
>... as you (mis)understood it.
>
>> 5. but they invoked that the vocalisation of laryngeal didn't
>> trigerred the pre-aspiration
>
>It was part of Olsen's original formulation of her rule. Miguel and I
>didn't conspire to invent it.
>
>> 6. Against this I showe them that in:
>> -> in PIE dialectal times a vocalized laryngeal /h2/ has induced
>> an aspiration in dHugh2ter (g>gH)
>
>You have shown nothing of the kind. The aspiration is not "dialectal
>PIE" but strictly Indo-Iranian. If you claim otherwise, the burden of
>the proof is on you.
>
>> -> and that even without this I cannot see any difference
>> between the vocalized h2 in p&x-ter or dHhu-g&s-ter and the non-
>> vocalized h2 of max-ter regarding ' a supposed metathesis' x-t>t-x
>> because this vocalisation was donme by adding a prop. vowel (in these
>> 2 contexts BEFORE the laryngeal)
>
>Your failure to see the difference is your own problem. Even if you
>swear on the Bible that the only possible realisation of vocalised *h2
>is [&x], it's just your personal prejudice.

Worse, it's demonstrably incorrect. Vowel + laryngeal gives
/V:/ in Indo-European, while "schwa indogermanicum" yields a
short vowel (/i/ in Indo-Iranian, /a/~/e/~/i/ in Greek,
mostly /a/ elsewhere). A vocalized laryngeal also doesn't
trigger Hirt's law in Balto-Slavic, whereas all combinations
of vowel + laryngeal do. As you have stated, the Vedic
metre shows that it was something like [hI] in Indo-Iranian.
Whatever the realization of laryngeal /&2/ was in
Proto-Indo-European, the one thing it almost certainly
couldn't have been is [&x].

>> 7. Here hey both started to talk about all the languages in these
>> world other than PIE
>
>How do you know what PIE sounded like? How do you now that *&2 was not
>an actual syllabic fricative or, say, a vowel with a non-modal
>phonation? For your information: there are IE languages with glottalised
>and pharyngealised vowels too.
>
>> and Piotr placed the aspiration of dHugh2ster 'very late'
>
>You mean *dHugh2ter- > *dHughItar- > duhitar-? Did I really say the
>aspiration was "very late"? Where? I said it was Indo-Iranian (which
>could mean, say, 2000 BC). It's older than Grassmann's Law in
>Indo-Aryan, at any rate. But it's still branch-specific, not PIE.

Now there is one group of examples where "schwa
indogermanicum" yields a long vowel (i:) in Vedic. According
to Starostin and Nikolaev, this happens in recessive roots
with a recessive suffix (what I call "non-inherently
stressed roots with a non-inherently stressed suffix"). A
nice example is Ved. stári:ma, which corresponds to Greek
strôma. Here I would expect PIE *stérh3mn., G. *str.h3mnós,
which would have yielded Grk. *stéroma, stro:mat-, and Skt.
*stárima, *stri:m(a)n-. Apparently, Greek generalized the
oblique root (stro:ma, stro:mat-), while Sanskrit
generalized the NA root, but with lengthening of the vocalic
laryngeal, after the oblique forms (stari:ma, stari:m(a)n-).


=======================
Miguel Carrasquer Vidal
miguelc@...