From: Miguel Carrasquer Vidal
Message: 47262
Date: 2007-02-05
>Because the merger a/o is Balto-Slavic.
>--- Miguel Carrasquer Vidal <miguelc@...> wrote:
>
>> On Sun, 4 Feb 2007 15:36:33 -0800 (PST), Sean Whalen
>> <stlatos@...> wrote:
>
>> > There is no reason to assume oi/ai had merged
>> >at the time final V() were shifting.
>>
>> Yes there is.
>
> They merge. Why before this rule, not after?
>> >The changes in word-final combos:But it's silly.
>> >
>> >is is. ix i: i
>>
>> But it is in fact -I
>>
>> >us us. ux y: y
>>
>> But it is in fact -U
>
> The nom. forms become -ix/ux but analogy causes
>-ux > -us, etc., after -os > -us. I already wrote
>this out.
> The reason I know -is > ix > i: isNo. This clearly has nothing to do with -oi vs. -ai.
>-tis > ti (no place for ana.).
>
>> >os os us u U
>> > as æs æ E
>>
>> There seem to be no cases of PIE *-as attested in
>> Slavic.
>
> That's why the PIE spot is blank. Final *-wus
>dis. > *-was; etc.
>
>> >ei ei i: i: i
>> >oi oi ui i: i
>>
>> It is in fact -ê (and -oj > -i)
>
> You started this thinking about why -oi/ai become
>different sounds;
>I say it's because they hadn'tBut no room for *wos either.
>merged before -oi > -ui.
>
>> > ai æi æ: E
>> >oyi o:i u:i u: u
>> >oyis o:is u:s y: y
>>
>> The ins.pl. can't have been -o:is, as that would
>> have given
>> Lith. *-uis instead of actual -ais.
>
> Long dip. > short before C
>
> o:i > u:
>
>> > The change of I>E occurred word-final after n
>> >(and morpheme final (or sim.) in gnEzdo).
>> > The change of U>E occurred word-final after w
>> >so, final -wos>wU>vE.
>>
>> Nsg. *-wos gives plain -vU.
>
> Again, this is simple analogy. The dual > vE
>is evidence enough, no room for ana. there.
>> > After os>us analogy would allow u-stem nom. -uxDoes it?
>> >to become -us, etc. Similarly as jus>jux
>>
>> What form is that?
>
> Since *nos > *nus > ny
>there must have been analogy
>nus/jux > nux/jux in pl. pronouns before *jux was
>replaced. The existence of second person plural
>*jux in PIE is fairly certain.
>
>> >so nos
>> >>nus>nux>nu:>ny. With no direct analogy (its
>> >origin as a nom. forgotten) -tis>tix>ti:>ti.
>> >
>> > Threre is optional contamination of -si > -sis
>> >explaining -six>si:>si.
>>
>> What forms are those?
>
> You're the one who wrote them in the context
>of your theory:
>
>> >--- Miguel Carrasquer Vidal <miguelc@...>
>> wrote:
>> >
>> >> Not usually considered in this context, but worth
>> >> taking a
>> >> look at:
>> >>
>> >> - the 2sg. athematic ending -si (if from *-saj)
>> >> - the infinitive ending -ti (if from *-taj)
>
> I take them as -sis and -tis; there is no reason
>to derive -ti from *-tai when Baltic -tis exists.
>> > The Greek evidence is probably nothing more thanWhy would, say, the locative and the nominative plural
>> >sandhi (oi#V > oj#V, etc.) since the limitation is
>> >relatively late.
>>
>> The point is that _some_ -oi, -ai endings count as
>> short,
>> others as long. So it's not sandhi.
>
> Why? There are completely certain cases of
>languages with sandhi different before C and V
>that have analogy so
>
>oi#a oi#t
>oy#a oi#t
>
>with this variation the tone would be different
>depending on the following word. One set of
>forms retains one tonal pattern, the other the
>other. It may simply depend on which are most
>often found before V (or any other word without
>a pause).