Principles of reconstruction.

[Justïn]

Right!....um...I was just about to say that...

--- In gothic-l at yahoogroups.com, "llama_nom" <600cell at ...> wrote:
>
> 
> --- In gothic-l at yahoogroups.com, Justïn <justinelf@> wrote:
> >
> > Is number 2. not triu?  I stumbled across it in the Wright link 
> > actually, but I may be mistaken.  Would 6. be aul?  This is 
really 
> > embarrassing...I'm working on the rest, at work right now so far 
away 
> > from my Bennett text...
> > Again, this is my first attempt, so as embarrassing as it is 
> > hopefully I'll learn...
> 
> Ah, no need for embarrassment!  I was just thinking of it as an
> excercise in phonetic reconstruction.  In other words, what would a
> Gothic cognate of (word descended from the same Proto-Germanic
> ancestor as) Modern English 'wood' look like?  But you're 
right 'triu'
> does mean "a tree" or "a stick".  I guess that illustrates another
> issue in reconstruction: where a word already exists in the same
> semantic field, how might that have related to the meaning of a
> hypothetical, reconstructed Gothic cognate?  Does that make sense?
> 
> In this case, it seems that the better attested early Germanic
> languages did have a few partly overlapping words in this senantic
> field, e.g. Old Norse has a word 'viðr' which is cognate 
with 'wood',
> as well as a word 'tré' congate with English 'tree'.  So there's
> nothing improbably about supposing Gothic had cognates for both, 
even
> though only one, 'triu', is recorded.
> 
> Number 6 is probably the sneakiest one; um, sorry about that... 
> ('bath' is much simpler!)  A few different derivatives exist in
> Germanic from the same (probbaly onomatapoeic) root, but working 
back
> from Modern English 'owl', Old Saxon 'ûuuila', Old High 
German 'ûwila'
> you could reconstruct Proto-Germanic *'uwwilôn', or for English
> perhaps *'uwwalôn', since it doesn't show i-mutation (=i-umlaut) of
> the root vowel.  The Scandinavian forms also lack i-mutation, and 
thus
> might also come from a form without 'i' in the suffix.  
Alternatively
> the lack of i-mutation in OE and ON might be due to an early loss of
> the medial (middle syllable) vowel -i- in the suffix.  So I'd
> tentatively go with Proto-Germanic *'uwwilôn' (weak noun, feminine
> on-stem), but keep an open mind that a Gothic derivative might have
> existed with a different vowel (namely a back vowel) in the suffix 
in
> place of 'i'.  The clue to the root of the hypothetical Gothic form
> though is in Old Norse 'ugla' and especially (Old) Swedish 'uggla'. 
> These show a sound change common to North and East Germanic whereby
> 'ww' > 'ggw', as described by Wright.  In Old Norse, the medial 
vowel
> of the suffix has been lost, which is normal, but presumably it 
would
> have been present in Gothic (compare 'mawilo' "little girl"), and 
the
> 'w' has been dropped, as always between two consonants.  So, I'd
> reconstruct Gothic *'uggwilo': weak noun, feminine on-stem, i.e.
> declined like 'mawilo', 'tuggo', etc.  One last clue is the Catalan
> word 'òliba', which it's been suggested may be derived from the 
Gothic
> word for owl [
> http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0406&L=gothic-
l&P=284
> ].  If so, it's obviously been through some rearrangement, so we
> probably can't read too much into this, but still, it does have 'i'
> for its medial vowel.  I wonder if any earlier forms of this word 
are
> recorded.  Anyway, that's what I'd do.
> 
> LN
> 
>  
> > --- In gothic-l at yahoogroups.com, "llama_nom" <600cell@> wrote:
> > >
> > > 
> > > Clue for Number 6: see Wright § 151.
> > > 
> > > http://lexicon.ff.cuni.cz/png/goth_wright/b0071.png
> > > http://lexicon.ff.cuni.cz/png/goth_wright/b0072.png
> > > 
> > > Clue for Number 4: compare the Modern English verb 'to lay' and 
> > cognates.
> > > 
> > > > 1. bath
> > > > 2. wood
> > > > 3. dwarf
> > > > 4. edge
> > > > 5. elbow
> > > > 6. owl
>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/gothic-l/attachments/20080208/51c06b92/attachment.htm>