[YG Conlang Archives] > [romconlang group] > messages [Date Index] [Thread Index] >
Hi!
Peter Collier writes:
>...
> Problem is a lot of vowels are disappearing due to a-umlaut early on, then
> i- and j-umlaut later, and the Hochdeutsch phonology is eventually
> levelling final vowels to <e>, which is messing up the Latin endings (good),
> but not in any readily useable way (not so good) . My ideal would be to be
> able to use umlaut morphologically, as with German - but the triggers for
> umlaut exist in all all forms in some declensions, only some forms in
> others, and not at all in others.
>
> Henrik, you must have had similar problems - although I guess Icelandic kept
> the differentiation among its final vowels, which must help a lot.
Well, yes, but Icelandic keeps more vowels in its endings that German,
so it's even! :-)
But I also had those problems. Some things collapse badly in the
Icelandic sound changes applied to Latin. In Icelandic it is usually
the syncopes that let you bang your head. E.g. distinctions like
filia vs. filius are just impossible to be kept (in all but nominative
singular). It drove me crazy, too.
But then remember that the plural in Latin had some collapsed forms
already (in 'filiis') and the 3rd declension not not m/f gender
distinctions altogether. So I did not fear such a collapse and went
for innovations to resolve this if necessary, e.g. filius vs. filiola
or sometimes -ina endings. Latin already did the same thing in
agricola/agricolina IIRC.
Moreover, this happened in Germanic, too. German noun endings are
reduced to a few consonants (let's say -er, -es, -en, -em), with only
very few of all the possible endings occuring in the same declension.
And the (i-)umlaut that today applies to plural forms is a vast
simplification of history, in earlier times occuring in singular and
plural forms depending on case.
I compared the original Germanic changes to those of my conlang quite
often to those that happened in Icelandic and found that the forms
were quite similar, and also shifted and simplified similarly.
E.g. compare:
Lat. Þrj. Isl. PG/PN
sg. nom. gallus > göllur fjörður < ferþuz
acc. gallum > göll fjörð < ferþu(n)
dat. gallo: > galla firði < ferþiu
gen. galli > gelli fjarðar < ferþo:z
pl. nom. galli > gelli firðir < ferþiuz
acc. gallo:s > gallar firði (*) < ferþunz
dat. galli:s > gellir fjörðum < ferþum(i)z
gen. gallorum > göllur fjarða < ferþo:(n)
(*) seems to have been fjörðu in ON.
Both Isl<PN and Þrj<Lat mostly follow regular sound shifts without
much unexpected surprises here and the two look quite similarly
complex, I think. Even the interesting -um ending in Isl. (while all
other endings are -{a,i,u}(r)) has an equivalent in Þrjótrunn, just
not visibly here: the -ibus endings become -(i)fur, breaking the
strict -{a,i,u}(r) pattern in one of the eight forms, just like in
Isl.
The above applied to NHG might look like this:
sg. nom/acc/dat: Gall
gen. Galler
pl. nom/acc/gen: Gälle
dat. Gäller
It would need more thinking whether it would be sensible to keep case
distinctions here -- I just kept gen in sg and dat in pl for symmetry
with MHG here. But I think the above is already quite funny.
One of the reasons I will definitely try to run a project like
Þrjótrunn for High German is that I like the resulting words. When I
decided to use 'cru:s' for 'leg' in Þrjótrunn ('krýr', IIRC), I
imagined how this would become 'Krauer' in that High-German-shifted
conlang. 'Krauer' - beautiful isn't it? :-)
> It's a conundrum.I could "cheat" and make some drastic changes to
> the phonological development, but wheres the fun in that? Also
> moves it all away from the OHG, which I want to keep as much as
> possible.
Some changes were dramatic in Germanic. What do you mean by moving
away?
**Henrik