Re: Sound Change With Labialized Consonant Codas

Hi!

David J. Peterson wrote:
> I thought of something today, and wanted to see if it actually
> existed in natural languages.
>
> Has there ever been a case where labialized consonant codas
> produced front rounded vowels?  Something like the following:
>
> i > y / _C_w
>
> So...
>
> *kik_we > kyk_we
> *mek_wt_wa > m2k_wt_wa
>
> It feels like a natural change to me, but I've never run across an
> instance of it.
Very natural indeed and I think Old Norse had this:

Old Norse u-umlaut did this, e.g. Old Norse _lyng_ 'heather, ling' is
from Proto-Germanic *lengwa-. (Or something similar.  I don't know why
-y- and not -ø-, but my sources give -e- for the Germanic stem vowel.)

Other examples from Köbler's Old Norse lexicon (ON=Old Norse,
PN=Proto-Norse, PG=ProtofGermanic):

   ON         PG/PN
   søkkva   < PG *sinkwan/*senkwan           'to sink'
   syngva   < PG *singwan/*sengwan           'to sing'
   fyrvar   < PN *firwo:R < PG *ferhwio:z    'men'
   nykr     < PN *nikwR   < PG *nekwiz       'water daemon'

In _sökkva_, the -ö- may be due to regular lowering before -kk-, so
the pattern PG -e- > PN -i- > ON -y- might be the same for all those
examples.  Dunno what causes PG -e- > PN -i- here.  An -e- before -w
will usually break into -ja- and then be u-umlauted to -jQ-, today
-jö-, e.g.:

   mjQl     < *melwa-                        'flour (the powder)'

This _Q_ is usually written o with ogonek, probably /Q/ or /O/.

Unfortunately, that ON -y- is /i/ in Modern Icelandic again, but the
spelling still represents the umlauted vowel.  Further, _ø_ and _Q_
have merged into _ö_ /2/, making things a bit blurry in MIS.

**Henrik