Re: [tied] *h3 (More deja-vu)

From: Miguel Carrasquer
Message: 15831
Date: 2002-10-01

On Tue, 1 Oct 2002 02:12:53 +0200 (MET DST), Jens Elmegaard Rasmussen
<jer@...> wrote:

>> >[MCV:]
>> >> Put differently: Hittite zero-grade forms like as-, ad- (*&1s-, *&1d-)
>> >> force me
>> >> to accept that PIE had no words starting with a vowel, but they do not
>> >> force me
>> >> to accept something as unnatural as that the initial phoneme in all
>> such
>> >> cases
>> >> was /h-/.
>> >
>> >I do not see how Hitt. as-, ad- can force one to any such inference.
>> The inference that these roots began with *h1-, I presume.
>No, the inference that there were no words in the language beginning with
>a plain vowel. The fact that *these* words had /H1C-/ does not exclude
>others from having /VC-/.

I mean that they show that initial *h1- is not a chimera, and make it at least
very likely that PIE was a language like Semitic, where words cannot begin with
a vowel.

>> I was talking about *h1t, not *h2t.  OK, let's make that: "and I accept
>> that in cases like *h1t > *th, *h1 must have been /h/, given what is
>> otherwise
>> known about *h1 (e.g. that it wasn't something like /x/ or /X/)."
>If /H2/ was any of the last-mentioned, i.e. [x] or [X] (velar fricatives
>of different degrees of backness), and underlying *-H1t- and *-H2t- *both*
>produce PIE *-th-, then this cannot be reason enough to determine /H1/ as
>not a velar fricative.

No, but given what is *otherwise* known about *h1, we can confidently say it was
not a velar fricative.

>I do it on the basis of its disappearance, its
>syllabic reflex /e/ in Greek, its aspiration, and its distinct lack of
>breaking (after high vowels in Greek, Armenian and Tocharian).

We agree completely. However, given the two hypotheses on the table here about
the nature of *h1 (/h/ or /?/), only the aspiration caused by *h1 in *some*
cases is an argument for /h/. And for instance in Latin -ginti/-ginta we not
only do not have aspiration (I mean elsewhere, of course, not in Latin itself),
but we have *d > *h1 and *h1k > g, which is more readily explained by /?/. I
see no other option than to assume that *h1 was *both* /h/ and /?/, and in most
cases we simply cannot tell which.

Miguel Carrasquer Vidal