Re: Umlaut

Henrik Theiling <theiling@...> writes:

> Hi!
>
> "Mark J. Reed" <markjreed@...> writes:
> > Quick terminology question:
> >
> > I thought umlaut was a specific variety of ablaut - a consistent featural
> > change.
>
> Hmm, ablaut is a quite arbitrary shift of the stem vowel (at least
> today).  I don't know what motivated it, but probably very ancient
> phonology rules that inserted two different parenthetic vowels like in
> Old Greek (e vs. o vs. consonantal stem variants).
In fact, the alternation in Old Greek between /e/, /o/ and zero,
and the Germanic ablaut are one and the same thing:
an alternation inherited from Proto-Indo-European.
The origin of PIE ablaut is somewhat uncertain, but one of the
conditioning factors seems to have been accent position.
Old Greek preserves the PIE state of affairs quite nicely,
while in Germanic it is more obscure due to later vowel changes.

> Umlaut is newer.  Some kind of phonological harmony.  E.g. the German
> umlaut is an i-umlaut technically as it shifted all preceding vowels
> of the word to the front in front of an /i/ in an appended ending.
Exactly.

> > But it doesn't seem to be; in German, for instance, while umlaut
> > does always move a vowel from back to front,
>
> Right.  In ancient times this happend because there was an /i/
> following the umlauted vowels.
>
> > it has an inconsistent effect on the height: ä [e] is higher than a
> > [A]; ö [9] is lower than o [o],
In some dialects perhaps, but not in the standard pronunciation.

> >    while ü [y] is the same height as u.
> > Is there some other consistency here that I'm missing?
>
> In Modern German the umlaut effect is lexicalised since the former
> endings that contained the trigger /i/ have vanished or changed to
> different vowels.  Further, I suspect that the vowels have shifted a
> bit after umlauts first occured.  (BTW, the umlaut in German is
> only uneven in /a/:
>
> Regular umlaut by simple fronting:
>      /u:/ -> /y:/
>      /U/  -> /Y/
>      /o:/ -> /2:/
>      /O/  -> /9/
>
> Shifted in Modern German:
>      /a:/ -> /E:/
>      /a/  -> /E/
> )
>
> If you have a look at Icelandic, the effect is more overt (more /i/s
> are still visible), while many things are also lexicalised now.  To
> see the effect, you have to look at Proto-(West?)Nordic (the
> reconstructed ancester of Old-Westnordic).  (I don't know when this
> effect happened to German).
It was an opinion held for a long time that it happened at the end
of the Old High German period because it doesn't appear in OHG
orthography, but the current opinion is that it happened earlier,
namely in Common Germanic, and that it doesn't show up in
written OHG because the umlauted vowels were merely allophones
of /a/, /o/ and /u/ rather than separate phonemes.  They became
true phonemes only when the conditioning factors were lost
at the end of the OHG period.  (Some umlauted /a/s appear as
|e| already in OHG because the fronted allophone of /a/ merged
with /e/ in some positions.)

> > Is the effect perhaps consistent for a given vowel across languages
> > that exhibit umlaut?
>
> Well, i-umlaut occurs in quite the some form in Old Westnordic (Old
> Icelandic).
It seems to be a Common Germanic phenomenon (as I wrote above).
It shows up in Norse, in German, and also in English
(with unrounding of /ö/ and /ü/; e.g. foot:feet, which were
/fo:t/ and /fe:t/ < */fö:t/ in Old English).  There is no indication of it
in Gothic, but that might be for the same reason as in Old High German.

>       Old Westnordic also has u-umlauts that result on a change
> of /a/ to /2/ (at least) in Modern Icelandic.  For whatever reason. :-)
This u-umlaut seems to be a specifically Norse phenomenon.

ObConlang: I am going to use both i-umlaut and u-umlaut in my
Q ("Elvish") languages; possibly also a-umlaut (which lowers
high vowels).

Jörg.

____________________________________________________________________________
Jetzt bei WEB.DE FreeMail anmelden = 1qm Regenwald schuetzen! Helfen
Sie mit! Nutzen Sie den Serien-Testsieger. http://user.web.de/Regenwald