At 7:41 pm -0600 11/1/00, [log in to unmask] wrote:
[....]
>
>Ah, but how does one explain -amo<-amus,

Later to be replaced, like -emo & -imo, by -iamo  :)

>ma<magis, piu<plus, etc.? Shouldn't
>these have some remnant of the final palatal?

Like 'voi' <-- 'vos' and 'noi' from 'nos' ?

Yep - strange things are happening.

In fact 'magis' does give 'mai' in the sense of "ever", "never" - but I'll
admit the -i there probably has more to do with the medial -gi- then with
the final -s (cf. French 'jamais'  <-- ja(m) magis). In VL it appears to
have become *mais.  Which makes the form 'ma' even odder.

But I reminded that modern Greek for "but" is also 'ma' and this,
presumably, is a shortening of the ancient 'mallo(n)'.  Remembering that
Greek was widespread in southern Italy at one time, one wonders if there is
not some cross influence here: ma <-- magis ~ ma <-- mallon.  just
speculation.

'piu' could, of course, show well dissimilation from *pjuj.

The Italian treatment of final -s does appear to present some problems.
The 2nd pers. sing. ending must be derived ultimately from the Latin forms;
while I can see that -es and -is would become -i, one would've expected -as
then to become -e.  Even in the oldest texts IIRC the ending is uniformally
-i.  Of course one can argue analogically leveling - and this is probably
so.

But indeed why wasn't the 1st pers. plural -ami etc?
Loss of final -s was an early feature in some Italian Latin dialects as we
see in the early Roman poets.  We may have a derivation from an older -amU,
i.e. final -s lost before any post-ProtoRomance palatalizing of final -s.
After all the 2nd plural -te is clearly a survival from the ancient form
-te which throughout the Classical period was always a possible alternative
to -tis, especially for the poets.

Indeed, the plural verbal endings generally present their oddities.  Whence
came the 3rd pers. plural -no?  Certainly not from -nt.  It's generally
thought to be derived from an Italian dialect form -nunt, (e.g. danunt
"they give", instead of the Classical 'dant') - but I don't think this is
provable with certainty.

As I said, to answer all these queries one needs a time machine.

[...]
>
>AFAIK, razón came directly from Latin. I haven't seen dialect borrowing
>given as a reason for the inconsistency, but it certainly sounds possible.

Internal dialect borrowing is often the reason for inconsistency within a
language, e.g. some English dialects always pronounced -augh as /af/ and
others always as /O:/.  The standard language has borrowed forms from both,
hence 'laughter' but 'daughter'.

There was also considerable interborrowing between Romance languages at
different periods.

It all helps to confuse   :)
------------------------------------------------------------------
At 9:26 pm -0500 11/1/00, Nik Taylor wrote:
>Tom Wier wrote:
>> But what does that *mean*? The resident expert on these matters in
>> this group, Ray, said that one theory was that analogy had leveled out
>> the irregularity in one instance,
>
>What I meant was that it seemed unlikely to me that analogy would
>operate on "feminine plural" but not "masculine plural".

Esattamente, mi amico.

What Nik says is precisely what I have been arguing.

If 'amiche' /a'mike/ is, as I believe it is, derived from 'amicas'
/a'mikas/, then there is no leveling out of *amice /a'mitSe/ to be done.
You can't level out something that never existed   :)

Ray.


=========================================
A mind which thinks at its own expense
will always interfere with language.
                   [J.G. Hamann 1760]
=========================================