[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
--- In ceqli@yahoogroups.com, Jim Henry <jimhenry1973@g...> wrote:
>
> On 1/24/06, Jim Henry <jimhenry1973@g...> wrote:
>
> >.......V = vowel,
> > S = semivowel
> > L = liquid (l or r)
> > N = nasal ( m n q )
> > P = plosive/stop (p b t d k g)
> > F = fricative (s z f v h x) (there were no instances of
> > |y| = /Z/ in this version of the dictionary)
> > A = affricate (|c| or |j|, now written as |tc| and |dj|)
> >
> > Here's the list -- some analysis will follow later.
> >
> > 69 PVLV
> > 55 PVN
> .....
>
> More (regex) symbols for what follows:
> ^ means the beginning of a morpheme,
> $ means the end of a morpheme.
>
> Some tentative conclusions:
>
> 1. There are just a handful of words
> with an initial cluster of three consonants:
>
> CCCVC splin
> CCCVC spriq
> CCCV spru
> CCCVC straq
>
> So you could have a rule that the only
> allowed three-consonant clusters are:
> a. in initial position
> b. |s| + a plosive + a liquid
> c. followed by a vowel, NOT a semivowel
>
> But maybe it would be better to redo
> these four words and have a simple rule that
> no word has three or more consonants in
> a row.
>
> 2. Of the two-consonant initial clusters,
> I see these patterns:
>
> - a lot of ^FL, ^PL, and ^FP
> - some ^PF, mostly /ts/ with some /kv/
> (the affricates might count here too)
> - some ^FN (/sm/, /sn/, /fn/)
> - a little bit of ^FF (/sf/) and ^PN (/kn/)
> - one-off oddities like ^PP "pterna" (not in the current glossary)
>
>
> 3. It seems that initial affricates
> are always followed by either a vowel or
> (in a handful of morphemes) a semivowel.
> Good.
>
> 4. For medial clusters, I see
> - a lot of LN and NL (jaqlaq, pulmon, etc.)
> - a fair bit of NN (jonmo, etc.)
> - a little LL (berli, etc.)
>
> 5. Final clusters:
> - 18 instances of LN$ (e.g. "barn")
> - one instance of LL$ ("xulr", ?, not in the current glossary)
> - one instance of NL$ ("donr"?, not in the current glossary)
> - one NN$, ("daumn", not in the current glossary)
>
> It looks like you've eliminated all final clusters other than
> liquid + nasal since July. Good move! Now maybe
> you could prune some of those initial /str/ and /spl/
> clusters, too?.... And what about those medial LL and NN?
I'd like to get some more input about some of this. For me, NN internally seems pretty
easy to pronounce ? pomni, pinmo, ? I guess mn and nm, and maybe qm and qn, tho
probably not nq or mq, and likewise rl and lr.
And as for endings, yes, I'm eliminating the lr, mr, etc. that were my attempt to allow
syllabic n, r, l, m so as to import words like 'himmel' from German. Since the stress has
moved to initial, those can be brought in with a -e- ? 'himel,' 'culer,' etc.
And I kind of like spl, spr, str, skr, etc. Anybody else have preferences here? If the
concensus is to disallow any, I'll go with it.