[YG Conlang Archives] > [romconlang group] > messages [Date Index] [Thread Index] >


[Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [romconlang] Declension trouble



Sorry for late answer. I had a test and a deadline...

Peter Collier skrev:
> A brief summary first:
>
> The language is a High German development of Latin. After
> those sound changes have nicely messed up the Latin
> declensions I have three cases (direct, genetive and
> dative), two genders (masc & fem) and two numbers
> (sing/plural).
>
> The definite article may come form ILLE, or maybe from
> IPSE (I haven't decided yet) but that doesn't really
> affect the declension. I end up with the following (based
> on ILLE, change the /l/ to /s/ for IPSE):
>
>|            Ms      Fs      Mp      Fp
>|Dir        /le:/      /la:/    /li:/        /le:/
>|Gen      /lai/      /lai/    /lu:re/    /la:r/
>|Dat       /li:/      /li:/      /li:/        /li:/
>
> No problems with the direct or genetive cases - except
> perhaps Dir. Mp - but, as you can see the dative all ends
> up in the same place, regardless of gender or number, due
> to a loss of final /s/ at some point earlier in the
> development.

Would the Vulgar Latin dative singular forms m. ILLUI and f.
ILLAEI develop differently in your GMP? I could imagine LUI
> lüü > löü > loi (_leu_). LAEI might fall together with
your genitive, but seeing how the genitive and dative
singular were identical in the first noun declension already
in Latin that doesn't strike me as any great calamity. The
resulting great similarity to Rhodrese pronouns is a bit
embarrassing, tho not unlikely when starting from the same
raw material:

<AFMOCL>
|               m.sg.         f.sg.         pl.
|
| nom.          él            elle /'el\`I/ il
| acc.          lo            la            le
| dat.          leu /ly/      lai           lor
| gen.          		so/sa              lor

The accents in the nominative forms are ostensibly in order
to distinguish them from the definite article forms el and
il. If so it is notable that pronominal la is not marked to
distinguish it from the feminine article. In reality of
course the nominative pronoun forms are always stressed
as Rhodrese is a pro-drop language, while the accusative
forms are always unstressed. The spelling lá exists but
is of course ILLAC 'there'. Moreover the accusative
forms lose their l when following a consonant, becoming
[U @ I]. To write e.g. "vairt'o" instead of "vairt lo"
for VIDET ILLUM is a popular way of indicating spoken
language in writing, but the pronunciation in question
is hardly provincial or substandard, but just the normal
spoken forms.
</AFMOCL>

> I am minded to lose the gender distinction for the
> plurals, for no other reason than German has done so and
> if I do the same it simplifies the problem a little. I can
> also use the original accusative rather than nominative
> form for the direct plural, which leaves me here:
>
>|            M       F       P
>|Dir        /le:/      /la:/    /lu:/
>|Gen      /lai/      /lai/    /lu:re/
>|Dat       /li:/      /li:/      /li:/
>
> Any idea how I could mark that dative plural? In some
> declensions it will be apparent in the noun itself due to
> the original -IBVS ending which collapses to /-p/, but not
> all. Extend -IBVS to all declensons by analogy

There is the -âbus sometimes used in classical Latin,
although that is probably overkill, though it could be
useful in cases where you don't want -INA in words for
female beings.

# The dative and ablative plural of dea, goddess, filia,
# daughter, end in an older form -abus (deabus, filiabus) to
# distinguish them from the corresponding cases of deus,
# god, and filius, son ( deis, filiis). So rarely with other
# words, as, liberta, freed-woman; mula, she-mule; equa,
# mare. But, except when the two sexes are mentioned
# together (as in formulas, documents, etc.), the form in
# -is is preferred in all but dea and filia.

> (not ideal, as I really don't like the way that word final
> <b> looks when I do have it!)?

Why not let it become -f/-v like in Franconian
(Rhenisch) dialects?

> Use something other than ILLÍS (what?), add some kind of
> clitic to the noun to ensure it has a distinctive ending
> in all declensions (I added SVVM to all gen. sing early on
> for similar reasons)
> - what though?


It is interesting to see how in German all forms of the
definite article occur in more than one slot in the
declension, der in particular being ambiguous, so that it is
in practice the combination of endings on the article and
noun, combined with lexical knowledge what gender the noun
is, which enables the listener/reader to determine which
case and number a given instance of article + noun
represents.  A certain amount of vagueness and ambiguity
is a part of language after all -- indeed what punsters
and poets make profit off.

/BP 8^)>
--
Benct Philip Jonsson -- melroch atte melroch dotte se
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"Sometimes you can't make it
 The best you can do is to fake it" (U2)