--- In
cybalist@yahoogroups.com, "alexandru_mg3" <alexandru_mg3@...>
wrote:
>
> --- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@> wrote:
> >
> > On 2007-03-21 21:09, alexandru_mg3 wrote:
> >
> > > Thanks for this observation. It was my misunderatnding due to
> the
> > > notation.
> > >
> > > 1. Do you know a real language where xH is present?
> > >
> > > 2. Could this presence go together with x and xW in the same
> > > language?
> >
> > Aspirated fricatives are extremely rare. Burmese has a three-way
> > contrast of /s/ : /sH/ : /z/, and the non-tense sibilant in
Korean
> (very
> > similar to Burmese /sH/) may also be described as aspirated, but
in
> both
> > cases the sounds in question are aspirated coronal sibilants (s-
> like
> > fricatives). I don't know of any language with a contrastively
> aspirated
> > back fricative.
> >
> > Piotr
> >
>
> 1. If so, we should better consider a sequence of Voiceless velar
> fricatives?
>
> x - plain velar fricative
> x^ - palatal-velar fricative
> xW - labialized velar fricative
>
> to have a similar sequence with k, k^, kW
>
>
> In this case:
>
> I) h3=xW
>
> we would have
> *pip-h3e-ti /*pip-xWe-ti/ > [maybe voiced Only in this context?] >
> *pip-be-ti > *pi-be-ti
>
> with p/xW > p/b > b
>
>
> and:
> *h3e,a,o > xW/e,a,o > xo
>
> I think h3=xW is good enough
>
>
>
> II) h2=x^
>
> I don't see how a simple x can transform h2e /xe/ in h2a /xa/
>
> But x^e > x^a or ex^ > ax^ seems possible
>
> *ph2ter /*p&x^-ter/
>
> The most problematic is this h2 that could be either:
> x^ - palatal
> x' - ejective
> etc...
>
> But if the other series were really:
> k k^ kW
> g(H) g^(H) gW(H)
>
> We need to select a palatal-velar fricative /x^/ here
>
>
> III) h1=x
> x is neuter enough not to color anything
> It Could be ok.
> h1esmi /xes-mi/
>
>
> So x, x^, xW (unvoiced)
> as k, k^, kW?
>
>
> Marius
I would question your phonetic assignment for h1 and h2.I would think
that h1 would be further forward than h2, for the simple reason
that /e/ is more frontal and higher than /a/. I think /x/ for h2 is
not improbable, perhaps alternating between velar and uvular -
although regarding the value of Hittite /h/, Egyptian records of
Hittite names record this using a velar, not uvular sound; had it
been uvular, it would have been notated using another Egyptian symbol
so uvular is less likely.
That being said, I can never agree to rules that allow single
exceptions. There is a general consensus that h3 was voiced, although
I agree that it was labialized too. In addition, h1 was easily lost;
unless it turned into "s" in certain environments, it is unlikely to
have been a palatal fricative, as those are anything but quiet and
thus not easy to lose. The fact that in Anatolian *h1e became a low
front vowel (/ae/) suggests that it had the potential to work some
very weak a-colouring, which would speak further against a palatal
identification. It seems more likely that h1 was pure aspiration -
indeed, the fact that p/t/k (especially t) + h1/h2 = ph/th/kh in Indo-
Iranian seems to speak for that.
Your approach may have merit in grouping the laryngeals together by
class. So far, we have h1 = aspiration (possibly alternating with a
glottal stop?) and h3 = Gw (labialized voiced velar fricative). As
discussed earlier, it is likely that h2 = x (voiceless velar
fricative) Thus, in the grouping of the different kinds of back
consonants, we have satisfied the following:
.................................................................
. . Palatal . Velar . Labiovelar . (Glottal) .
.................................................................
. Voiceless . . h2 . . .
. Voiced . . . h3 . .
.(Voiced) aspirated . . . . h1 .
.................................................................
While admittedly this arrangement leaves an empty palatal spot, it
does manage to fill out the 3 voicing qualities nicely. Likewise, the
resultant syllabic "schwas" are not too unfitting: h1 matches the
traditional schwa sound exactly in being a mid central lax vowel, h2
a lowered version of that (i.e., closer to back /a/), and h3 is a
rounded version of h2 (although the position is less certain, as the
main feature of it is rounding).
................
. i
.
. e
.
. ae