On Mon, 11 Jun 2001, Mangiat wrote:

>Padraic wrote:
>> More grist please!
>> I thought I'd put in the Kerno list while I'm at it. Lots of nice
>> irregulars here - usually of the split declension sort which is
>> very common in Kerno. I-mutation plural nominatives are heard
>> less frequently than the full (correct) form. [coll. = collective
>> noun]
>> ascendere       ascender
>> barba           la barba                y varbān        (also la barbá)
>Does this barba mean 'beard' or 'uncle' (as in many Gallo-Italic varieties)?
>The plural -ān suggests the latter hypothesis. -ani is a quite common plural
>suffix in some characteristic alpine dialects; in some valleys it is always
>used with surnames.

Means beard. Though the meaning "uncle" is quite interesting! How
did (presumably) Latin barba come to mean uncle? Did all uncles
have beards or something?

Especially in the last couple centuries, Kerno has exhibited the
curious phenomenon of nominal declension shifting based on the
final consonant of the root; after which, other similar nouns
sometimes follow by analogy even if that consonant isn't part of
the word in question.

For example, the K word "la Couba" means Cuba, the country. One
majour export of the place is "y cigarillas coubanas" (< Span.);
which got shortened in K to "y choubannes", literally 'Cubans'.
The -n element gets reanalysed as the stem, and this word migrates
from the -a declension to the -n declension. At times other words
get reanalysed as well, even if there was never an -n in there
(like barba), and it also ends up as an -n stem. As you can see,
la barba is halfway in the -a declension (the singular) and halfway
in the -n declension (the plural).

Kerno seems to be in the middle of a great process of shift away
from the old "Latin" declension system and into something new. As
the language stood in the mid to late 19th century (the Revival),
the process was well underway; currently its travelling at full
steam and the best efforts of the various Language Bureaux can't
seem to stop it.

The process is this: In the oldest levels of the language, there were
the same five vowel declensions Latin had plus three special consonant
declensions (-n, -t, -r/-s) which were largely declined like -i stems.
By the 15th century or so, the -o and -u stems are largely coalescent;
by the late 19th, the current state of affairs is found where there's
essentially one declension. What's happening is that -r, -n, -s, and
-t are becomming reanalysed into (official) plural terminations.

It should be noted that what facilitates this is the loss of the case
endings in casual speech:

WRITTEN                         SPOKEN

il murs         y vurores       /Il murs        i vUror/
li muri         lis murib
le mmurre       y vurores       /le~ mur        i vUrors/

[You don't get the dative in daily speech. I'm also sure that [o]
isn't the right IPA for the second vowel in the plural, but you
get the idea.]

Probably what's going to happen is _all_ nouns will end up in one of
these four new declensions, with reworking of the case endings.

Not quite the answer you were looking for, but it does explain why and
how la barba gets its -n!

Ah... by the way: in Kerno, uncle is "patrecko (al laddes la mmatre)"
for maternal uncle; and "brater peitron" for paternal uncle.