Grzegorz Jagodzinski wrote:
>>[Miguel:] So, as Piotr says, it's not [w], which by definition is a
>>labiovelar approximant.
>
>
> But I have shown that IT IS NOT TRUE - see the last my sentence? Piotr is
> simply wrong (more precisely: his view is not in concordance with Hannusch
> who is a far more reliable source), and I just quote him, nothing more.
> Can't you see quotation marks? Please do not snip my posts if you have
> troubles with understanding them.
I'm not a "source". The source on which I base my opinion is Hinc
(Heinz) Schuster-S^ewc (1968, _Gramatika hornjoserbskeje re^c^e. I.
Fonematika a morfologija_; English translation [by Gary H. Toops] 1996,
_Grammar of the Upper Sorbian language. Phonology and morphology_). It's
a standard reference book written by a linguist whose native language is
Upper Sorbian, so I don't think it's any less reliable than Hannusch.
Schuster-S^ewc is a former student of Lehr-Spl/awin'ski at the
Jagiellonian University and the author of the monumental etymological
dictionary of Upper and Lower Sorbian. He explicitly describes the
articulation of the Upper Sorbian reflex of *v and delateralised *l
(spelt <w>, <l/>) as different from that of Polish /w/ (<l/>). The X-ray
tracings of its main allophone show the tongue surface in the neutral
position, just slightly convex, without any distinct raising of the
dorsum -- in short, the sound is a purely labial [B]. Polish /w/, by
contrast, is a true labial velar: the tongue is retracted and bunched up
into a ball-like shape, and the dorso-velar constriction is rather close.
The allophonic weakening and vocalisation of Sorbian <w> in coda
positions is a fact, but a pretty trivial one. Interestingly, the
Sorbian languages also have labio-dental /v/ as a marginal phoneme
occurring only in loanwords.
Bilabial reflexes of Slavic *v are quite common in West Slavic. They are
found in Slovak (with the allophone [w] occurring in syllable codas, as
in Slovene, Ukrainian and Belarusian), and in the central dialects of
Kashubian. At the other end of the spectrum we have the Polish and Czech
realisations of /v/, which are very distinctly labiodental. In Czech, in
particular, the constriction for the fricative is so narrow that /v/ may
be realised almost like a labio-dental stop.
Piotr