My original plan was not to allow geminates but found that it lost something
and I have been using them now, so that wouldn't be a problem. The 'd'
sounds quite real to me as well.
Guess I have to keep saying the phrase to see what happens with my mouth.
One of the most unfortunate parts of conlanging is that one rarely gets to
hear his/her language spoken unless it is recorded and played back. For me I
want to enunciate everything perfectly so I miss out on this "real" way of
From: Constructed Languages List [mailto:[log in to unmask]] On
Behalf Of R A Brown
Sent: Monday, October 29, 2007 12:16 AM
To: [log in to unmask]
Subject: Re: elision
Scotto Hlad wrote:
> Recently, in response to another thread, I posted a Regimonti idiom:
> Les buves se en rapoti = the cows have hurried (themselves)
> I have omitted the grave accents in "se" and in "en". If you enunciate
> each word the phrase is pronounced as follows:
> 1. /les/ /'bu.ves/ /sE/ /En/ /r`a.'pO.ti/
> (hoping to the nearest conlang deity that I got all the xsampa right)
> Sadly trying to spit that out at normal speed, one would end up with
> an unattractive glottal stop between the two /E/s
> 2. /les/ /'bu.ves/ /sE?En/ /r`a.'pO.ti/
> (which sorta sounds like the speaker is clearing his/her throat)
Or like a Britisher saying 'setten' :)
> If one says that at normal speed, the "se" and "en" would no doubt run
> together so that it would be said,
> 3. /les/ /'bu.ves/ /sEn/ /r`a.'pO.ti/
> The problem for me arises in that following an /n/ with an /r'/
> requires a bit of oral gymnastics and it would seem to me that either
> the /n/ or the /r`/ would disappear in the process:
Mark J. Reed wrote:
> I would guess that the E would be nasalized and the [n] would drop.
That is indeed one possibility, adopted in many natlangs.
In Latin in similar situations the /n/ becomes assimilated so that /n/+/r/
--> /rr/. That is possible, of course, only if a language allows geminate
The other 'solution' often adopted in natlangs is to insert an intrusive /d/
between the two consonants. We find that often in ancient Greek, e.g.
_aneer_ (man, adult male) has accusative _andra_ <-- *anra.
Entia non sunt multiplicanda