Re: New language under development

Hello!

On 5/31/05, Patrick Littell <puchitao@...> wrote:
> On 5/31/05, Julia Schnecki Simon < helicula@...> wrote:
> > So, if the harmony was based on frontness/backness, a stem containing
> > front vowels ("bibibe") would take affixes with back vowels
> > ("bibibebong") and vice versa ("bobbu" -> "bobbubing")? Fascinating. I
> > guess there's no feature we can come up with that's too crazy for
> > *all* natlangs...
>
> That's precisely it.  In one of the derivational suffixes, roots with a, e,
> and i, take -ul and roots with o and u take -il.  (Roots with @ might have
> already undergone lowering to a at this point in the derivation process.)  I
> can't recall which derivation suffix this is... it's probably a nominalizer,
> of which Itzaj has many.
Very cool, and definitely something to keep in mind. :-)

[snip]

> > What about consonant harmony, by the way? I'm not talking about the
> > normal, "simple", assimilation/dissimilation phenomena here either
> > (i.e. the familiar voicing, devoicing, deaspiration etc.), but about
> > something along the lines of "affix XYZ contains voiced plosive
> > archiphoneme, which is realized at the place of articulation of the
> > last consonant in the stem". With some invented stems and affixes,
> > this would look something like:
> >
> > affix 1 -aNa (where N = nasal archiphoneme)
> > affix 2 -FiB (where F = unvoiced fricative archiphoneme,
> >                    B = voiced plosive archiphoneme)
> >
> > stem 1 balas -> balasana, balassid
> > stem 2 fetep -> fetepama, fetepfib
> > stem 3 simuk -> simukaNa, simukxig
[snip]

> I think one finds alveolar/retroflex harmony in various languages of India.
I remember a lot of assimilation from the Sanskrit classes I took
<mumble> years ago, but no actual harmony. Do you happen to know if
the languages in question are Indo-Aryan or Dravidian?

> (A quick google mentions Finnish, in fact; it says that some Finns find it
> difficult to produce both /b/ and /p/ in the same word, and so will make a
> loan like "pubi" into "bubi" or "pupi".)
That's not really harmony. That's just a symptom of the almost
complete absence of a voicedness opposition in the Finnish phoneme
system (the only exception being /d/:/t/ in standard Finnish and in
some dialects AFAIK).

I once attended a class on Finnish pre-Christian religion and folk
beliefs here in Helsinki. The class was taught in English (it was even
included in the list of classes officially recommended to foreign
students interested in Finnish culture and history), but unfortunately
the lecturer, whose English was excellent otherwise, had a really hard
time with the English voiced:unvoiced opposition. It's hard to keep a
straight face when you're listening to someone who you *know* means
"pagan" when he says "bacon"... ;-)

(The class was fascinating, though, and they even let me write my term
paper on something linguistic.)

Hmm... I could have sworn that there's some Finnish dialect somewhere
that has other voiced-unvoiced plosive pairs, but I can't find
anything about this in the (admittedly small) book on Finnish dialects
I have here (Martti Rapola's "Johdatus Suomen murteisiin", in case
anyone's ever heard of that one ;) .

I did find some interesting sound (and morpheme) correspondences in
that book, though. For example, in some dialects [l] or [*] (alveolar
flap) appears for standard Finnish /d/ (which is technically just a
context-dependent variant of standard Finnish /t/, but it does have
phoneme status, since there are minimal pairs -- /t:/ becomes /t/ in
the same context where /t/ becomes /d/, just to make everybody's life
a little more interesting. _maton_ doesn't come from _mato_ "worm" but
from _matto_ "carpet"; the corresponding form of _mato_ is _madon_.
But I digress). This gave me the idea to maybe use liquids as
allophones for plosive phonemes in my language (intervocalic voiced
plosive reduced to homorganic flap or approximant, or something like
that).

> But a thoroughgoing, productive consonant harmony in affixes, I can't think
> of an example.  I could imagine this produced diachronically from
> reduplication, though... Say we indicated the plural by reduplication of the
> final syllable.  (balaba -> balaba-ba)  But then, for some reason the
> intensifier suffix -n gets added to all plurals (giving us balababan) and
> then, due to stress changes or something, the final "a" is elided (balababn)
> and the /bn/ mutates to /m/.  Provided this happens across all points of
> articulation, we now have a plural suffix -N.  (Archiphoneme N, of course,
> not "ng".)
>
> balaba -> balabam
> lomoto -> lomoton
> qapatj -> qapatjanj
> wilugu -> wilugung
>
> A half dozen funny reduplication processes like this could give us a number
> of interesting suffixes involving consonant mutation.
Sounds plausible.

Maybe I'll even create a morphology someday where this kind of process
can be seen "in action"... say, some feature is expressed by
reduplication of the final syllable, some other feature is expressed
by a (say) nasal suffix, and sometimes they occur together and the
nasal merges with the reduplicated syllable's consonant, keeping the
nasality of the one and the point of articulation of the other, as you
described. Or yet another feature is expressed by a suffix containing
a sufficiently neutral consonant (darn, I'm gonna need that glottal
stop after all!), which results in a "combination" suffix containing
either a lengthened or a glottalized version of the stem's final
consonant. All that while other forms of the same words have only
*either* reduplication *or* one of the suffixes, which therefore
retain their original form.

> Can't wait to see what you end up doing!
Me too. ;-)

Regards,
Julia

--
   Julia Simon (Schnecki) -- Sprachen-Freak vom Dienst
_@"  schnecki AT iki DOT fi / helicula AT gmail DOT com  "@_
si hortum in bybliotheca habes, deerit nihil
                                        (M. Tullius Cicero)