Re: Q Re: Sound Change (On Glide)

En réponse à David J. Peterson :

>Now, I've never actually been taught this in any linguistics class (hence,
>why I'm asking), but in Spanish, it appears that Latin long mid vowels,
>when stressed, became diphthongs with an on-glide.  Examples:
>
>sentar "to sit" (stress on the /a/)
>
>nosotros sentamos "we sit" (stress on the /a/)
>yo siento "I sit" (stress on the /e/)
>
>dormir "to sleep" (stress on the /i/)
>
>nosotros dormimos "we sleep" (stress on the/i/)
>yo duermo "I sleep" (stress on the /e/)
>
>Note: I'm imagining that there was an secondary stage with "dormir" where
>it was actually /dwormo/, at one time, and then the vowel fronted, for
>reasons unknown.
Indeed. In italian, it stays at the /wo/ stage, which explains Italian
"buono" where Spanish has "bueno". Some linguists propose an influence from
Basque to explain the change /wo/ -> /we/. I don't know the details of
their hypothesis though.

>My question: Does anyone have any examples of this happening in other
>languages (nat or non)?  I don't believe I've ever seen another example of
>this (though I've seen countless where obstruents become palatalized
>before front vowels and labialized before back vowels, resulting *kind of*
>in diphthongs with on-glides), so I was wondering how it worked, and if it
>could work another way.
Well, nearly all Romance languages, in some way or the other, have had this
diphtongation stage. At least Portuguese, French and Italian have had it (I
don't know about Romanian and Sardinian). I think Occitan has it too, and
I'm pretty sure Catalan also has it. In French, you have (stress is always
on the last syllable, so I won't indicate it):

venir:

nous venons
je viens

In French though, plenty of other phenomena happened, which means that this
diphtongation is not the principal thing that happened and only a few verbs
still show it. In French, the principal phenomena were changes of vowels
like Spanish "pedir: nosotros pedimos, yo pido". For instance:

mourir /muRiR/:

nous mourons /muRO~/
je meurs /m9R/

And even then, French underwent a stage of massive analogical changes which
destroyed most of the alternations that had appeared earlier. For instance,
"aimer" had a a-ai alternation in Old French (a when unstressed, ai /E/
when stressed). But it was levelled out and only "ai" survived (except in
the noun "amant" which was originally a present participle but changed
meaning to "(male) lover" and became disconnected to the verb, which
recreated a present participle "aimant").

My Narbonósc has also a few diphtongations. But like in French, most of
them were levelled out. So it kept them in verbs like "venîre" (same
pattern as its French and Spanish cognates), but exhibits more frequently
verbs with weakening (like Spanish pedir or French mourir. Indeed,
Narbonósc "mourîre" has the same weakening as its French cognate). In this
case, it kept them more than in French ("amâre" features the a-ai
alternation it lost in French). There are also isolated cases like "bouem"
where French has "bon" and Spanish "bueno", as well as "douem" and "douêne"
(from Latin DOMINUS and DOMINA) which exist aside "dom" and "dône", of same
origin but without diphtongation (and with a different meaning. "Dom" and
"dône" have taken the meanings of "man" and "woman", while "douem" and
"douêne" are a more direct continuation of the original meanings of DOMINUS
and DOMINA and mean approximately "sir" and "madam" - or "lord" and "lady" -).

Christophe Grandsire.

http://rainbow.conlang.free.fr

You need a straight mind to invent a twisted conlang.