Szemerenyi's reductionism creates more problems that it solves and so
is untenable. On the other hand, extreme laryngeal proliferation (as
in the well-known ten-term system proposed by Martinet) is even
worse. I take *h1 to be a glottal sound (a breathy or voiceless
glide /h/ or a glottal stop /?/ -- it's really difficult to be more
precise given the available evidence), *h2 to be a velar or (perhaps
more likely) uvular fricative /x/, and *h3 to be a labialised
counterpart of *h2, IF *h3 should be reconstructed at all. I have
promised to make my position clearer and I will do so, but please
have some patience. I've got a lot to do these days.
The question "How many laryngeals?" is often resolved as in this old
limerick:
There was an old party of Lyme
Who married three wives at one time.
When asked, "Why the third?"
He replied, "One's absurd,
And bigamy, sir, is a crime."
Piotr
Piotr
--- In cybalist@egroups.com, "Mark Odegard" <markodegard@...> wrote:
> OK. Could y'all define as well as can be what you understand the
> laryngeals to be? I gather MCW and PG regard H3 as Xw; presumably
H1
> and H2 are glottal and pharyngeal?
>
> Szemerenyi, as I recall, suggests they all represent plain old h.
>
> Mark.