Re: Latin -idus as from dH- too

On 2008-03-13 20:55, alexandru_mg3 wrote:

> I wanted to say that "Today => Nobody knows exactly the nature of
> laryngeals...."

Speak for yourself. I think we have enough evidence to reconstruct the
approximate pronunciation of the three laryngeals.

> A very general remark that served to nothing.
>
> More General for you : we could have C3_1 + C3_2 + C_3_3 = 3 + 3 + 1
> = 7 distinct classes in total
>
> These classes are: h1 h2 h3 h1 h2 h2 h3 h1 h3 h1 h2 h3
>
> Now remains to choose who are the two-s against the third?

If you take /t/, /z/ and /n/, for example, you can group two of them
against the third in three different ways: voiced /z, n/ against
voiceless /t/, nasal /n/ against oral /t, z/ (this could also be
sonorant vs. obstruent), or continuant /z/ against noncontinuant /t, n/.
All these contrasts are equally real and may be relevant as factors
conditioning real-world sound changes. Likewise the laryngals: for some
purposes *h1 and *h2 may form a natural class versus *h3, which doesn't
mean that other groupings are excluded.

> Why not all three? And why not each by its own?

How funny. But yes, they also contrasted with each other. That's why we
reconstruct them as different phonemes. They also seem to have formed a
class together, all three of them.

> *h1 (probably just a glottal glide) is always
> lost.
>
> 1) "probably just" IS THIS SCIENCE?

What's unscientific about "probably"? All science is probable (at best
highly probable) rather than absolutely certain.

> So to resume: - "If h3..." and then - "h1 and h2 might..." and - "if
> h1 probably just ..." THAN => "we have a metathesis for h1/h2-t but
> not h3-t"
>
> Bravo, Piotr!

Your fine irony is misdirected. I wasn't even arguing for segmental
metathesis, so you're attacking a straw man.

> Next, I refuse to consider Latin barba:tus 'a recent formation' or 'a
> non-dateable one' when I have Lithuanian and Slavic counterpart all
> of them reflecting *bHar(z)dH-eh2-to

We have lots of such derivatives, some of them very recent.

> As for your information: Olsen herself treats barba:tus as an old-
> formation and proposed a 'morphological restauration'=> better to use
> her argumentation in this case, I think...

I agree. When I say that individual -a:tus (-atU, -otas, -ed) adjectives
may be of any age, I don't question the antiquity of the _formation_.
The restoration of *-to- for *-tHo- (and a long vowel on the analogy of
the verb forms in question) is of course a likely explanation.

> the dHeh1- inside SOLIDUS => CRIES BY ITSELF

Poor *dHeh1-. Please, don't cry.

Piotr