'town'

[David Costa]

Well, offhand I can't give any reasons why that *couldn't* have happened, 
but it definitely would have had to happen at the Proto-Algonquian level, no
later. Then the question arises as to whether the Proto-Algonquians were
geographically anywhere near the Proto-Siouans. I tend to think not, but
what do I know?

If Proto-Algonquian borrowed it, it very likely didn't borrow it as
*/ohti/, since Proto-Algonquian had preaspirated obstruents, and this
etymon for sure has a plain */t/ in Algonquian.

Dave


> One quick observation before I go off to ponder this:  the competing form for
> *htuNwa(N) 'group of associated people' (waxing a little abstract here) - or
> at least for derivatives of that - as a 'town' word in Siouan is *o-hti 'dwell
> in a place' (= IN-dwell).  What if Algonquian borrowed *ohti and massaged it
> into o:te:-weni et al., whereupon Siouan borrowed it back as *ohtuNwaN, cut it
> down to htuNwa(N) by analogy and put it back to work in various derivations
> from that?

>> These Algonquian 'town' forms are all nominalizations of */o:te:-/. Moreover,
>> as Ives has pointed out, there is a corresponding medial */-o:te:-/ which can
>> be seen in constructions like Miami /minooteeni/ 'town', Menominee
>> /meno:tE:w/ 'he goes partners with someone' and Fox /meno:tani/ 'the enemy
>> side'. (PA */men-/ = 'grouped apart') ... In defense of the idea of it being
>> a Siouan -> Algonquian loan, word-initial */o:/ is rather uncommon in
>> Proto-Algonquian.