Dorsey's Breves; e vs. i in OP (Re: Biloxi/Ofo)

[Koontz John E]

On Tue, 30 Nov 2004, R. Rankin wrote:
> No, Dorsey always writes the length distinction as if it were a quality
> distinction.  No one has had the courage to undertake a complete
> analysis of JOD's use of the breve and other vowel diacritics (it would
> be a massive undertaking).

Agreed.  I did look at this to some extent in the OP texts and came to the
conclusion that e-breve is pretty closely associated with the context
Ch__#, e.g., =the, =khe, =dhiNkhe 'the', but also e'be 'who' (first e),
e'di 'there (demonstrative)', edi' 'there was (existential predicative)',
e'gidhe 'it came to pass that ...; as expected ...', he' 'female
declarative'.  Various longer forms have -<e-breve>, like t?e=dhe 'kill',
we'ahide 'far away', and so on. I-breve was in variation with i and also
mostly final.

Typical cases of i-breve in Dorsey are kki 'if, when', -z^i NEG, Kki
(khi?) 'And ...', =s^ti 'too', s^i 'also', and so on.

Typical cases of a-breve are =ga 'male imperative', =a 'female
imperative', =a 'interrogative', ha' 'well', ha' 'male declarative'.

Typical cases of u-breve are exclamations, huN, wu, wuNh.

It looks to me as if it could be argued that e-breve, i-breve, and a-breve
are indications of a short vowel, but that lack of a breve would not be
conclusive evidence of a long vowel.

This reminds me that there are a number of OP words that Dorsey writes
with final i, but LaFlesche with final e.

The only form I have been able to come up with quickly is:

ppaN'ghi 'parsnip' (JOD 90:653.11) :  ppaNghe 'radish' (Swetland 91:144)

- ppaN'ghe is the form in in Ks and Os
- there's no breve on this form in Dorsey

Possibly relevant are cases like

dhe'ze 'tongue', but cf. Da c^e'z^i', Wi reezi'

In this case PSi seems to have -i, but Dh has -e.

wathaN'zi 'corn plant' (all sources), but cf. Qu wathaN'se
idha'di 'father' (all sources), but cf. Qu ida'tte
ine'gi 'mother's brother' (all sources), but cf. Qu itte'ke
sagi' 'hard' (all sources), but cf. Ks  da'sage 'harden in wind'
-ppu'kki instrumental sound root, but cf. Ks -ppokki ~ -ppokke ditto

In these cases some Dh has -e, but most -i.

Not participating in this pattern are forms that seem to end in hi from
PDh *hu in the sense of 'plant', e.g., bu(u)'de hi 'oak' or hazi 'grape'
(probably from haz-hi); body parts like nu(u)'si 'armpits', ppi'zi 'gall',
i'kki 'chin'; kinterms like itti'mi 'father's sister'; verbs like uaN'si
'jump', c^hi' 'to copulate', tte'xi 'difficult'.  Also ma(a)'si 'hail'.

A number of these OP -i are from *-u, but not all, e.g., not ppi'zi or
itti'mi.