Re: [tied] a:/o: merger

On Mon, 01 Oct 2001 21:39:03 -0000, "Sergejus Tarasovas"
<S.Tarasovas@...> wrote:

>--- In cybalist@..., Miguel Carrasquer Vidal <mcv@...> wrote:
>> I've been doing a preliminary inventarisation of Slavic Auslauts in
>> -Vs, -VN (i.e. -Vm or -Vn[t]) and -Vns [mainly nominal forms], and I
>> think the rules can be formulated as follows:
>>...
>
>I've practiced rule-set creations for Slavic auslaut since my teens.
>Every result I got was full of stretches and exceptions to such an
>extent that the rules were losing their explanatory power completely.
>Your try is not an exception, you can easily point out the weak
>points yourself.

Indeed. On the other hand, I've written computer programs far more
ugly and complex than what this amounts to (and they worked...).

>To add some fresh blood, how would you interpret
>Krivichian (non-standard ORuss) o-stems N. sg. ending -e (not -U)?

Unfortunately, I have misplaced your exposé on Krivichian. Before I
can formulate a hypothesis, I need more information: does this -e
palatalize a preceding consonant, and if so, how? What is its
counterpart in the jo-stems?

>> [6] normal development (D sg. o-stems): -o:i > -uo [> -u]
>
>Interestingly enough, if your explanation of the origin of o-stems D
>sg. is accepted, this example will clearly point out different
>interpretations of *o: and *a: in some positions in PSlavic:
>
>a:-stems D. sg. *-a:i > *e^
>o-stems D. sg. *-o:i > *u.
>
>;)

Well, mergers are practically never complete... Given that the
development must have been /o:i/ > /ui/ [as in Lith.] > /u@/ > /uo/,
the development of jo-stem Ins.pl. *-jo:is (> -ji), may be easier
formulated as:

-juis > -jiis > -ji:s > -ji,

without influence from the -s (but then the problem becomes D.sg.
-ju).