Re: DECAL: Examples #2: Phonotactics
From: | Benct Philip Jonsson <bpj@...> |
Date: | Tuesday, January 18, 2005, 10:03 |
First off an important addition to #1 Q2 (Allophones): /h/ has an
utterance-final allophone [?].
Sai Emrys wrote:
> Same deal as last time.
>
> Q1: What are your allowable syllable structures?
(S)CrVCC
Where V is any vowel, S is /s s\ z/, C is any consonant and r is /r/.
Only V is oblicatory.
>
> Q2: Onset clusters?
pr, br, fr, tr, dr, kr, gr, xr, (s\r*)
(* where |s\r| is probably only a bogus spelling for /r_0/ or [s`]!)
In addition the Kidilib dialect has the onsets
sp, spr, sf, sfr, st, str, sk, skr. These are lost as onsets in
Classical Sohlob and Heleb through the development of a prothetic
vowel (/i/ or /&/ depending on vowel harmony).
>
> Q3: Codas?
In practice the possible CC clusters are
restricted due to the fact that the protolanguage (Kedjeb) only
allowed _s, n/m/N, r, t/d_ as syllable final consonants. Due
to loss of final vowels a variety of final CC clusters arose:
/s s\ z/ followed by any other consonant.(*)
/m n J N/ followed by a homorganic stop
/r l/ followed by any other consonant
Any other non-stop consonant followed by /t d/(*), through
the metathesis of most t/d+C clusters, although e.g. */dn/
gives /zn/.
(*) Fricative+stop and s/s\ +fricative clusters must
agree in voicing. There are no stop+stop clusters.
>
> Q4: Any changes depending on place in word, etc.?
Yes. Apart from what already has been described in
the context of allophones there is above all vowel harmony.
In Classical Sohlob vowel harmony stipulates that
the vowels of a root word must be either all high or
all low. Thus vowels in non-first syllables will
change their height if necessary to agree with the
vowel of the first syllable. Heleb also has palatal
harmony, whereby vowels must be all front or all back
in addition to all high or all low. The elements of
compound words need not agree with each other, thus
_zurqnansj_ /zurGnans\/ "driver, charioteer, lit.
wagon-man" is a well-formed compound.
>
> Q5: Motivation / reasoning / goals behind this?
My aesthetic preferences, mostly. I wanted a
restricted yet naturalistic inventory of clusters.
I also decided at an early stage to have vowel
harmony, but disliked ordinary bland palatal
harmony, so I came up with height harmony. It
turns out Old and Middle Korean has something similar.
> Thanks,
>
> - Sai
/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se
Solitudinem faciunt pacem appellant!
(Tacitus)
Reply