squash

[David Costa]

> Could we try to put our assumptions on the table regarding the direction of
> loan words?  If I'm understanding correctly, it seems to be agreed that the
> word for 'squash' reconstructs phonologically throughout (eastern?) Algonquian

No, just plain 'Algonquian'. It's found throughout the family, from Micmac
to Cree to Arapaho to Shawnee to Delaware and almost all points in between.
The only decently-documented part of Algonquian where it's NOT found are the
Southern New England languages (the languages of Massachusetts, Rhode
Island, 
and Connecticut), which are a lexically innovative group in general.

Also keep in mind that it means 'spoon' -- not 'squash' -- in every
Algonquian language except Miami.

> as *eemehkwaan-. 

Very likely *eemehkwaana. Animate gender.

> Siouan, on the other hand, does not reconstruct so easily, though similarities
> are apparent with each other and with the Algonquian term.

> Model 1.  The term originates in proto-Algonquian and is adopted variously by
> different Siouan daughter languages, which means that the term will not be
> consistent within Siouan.

Coming from a position of being mostly ignorant about Siouan, that seems
most 
likely to me, esp. since it seems to match the results of some other
apparent Algonquian -> Siouan loans.

> Model 2.  The term originates from wherever as an international term and
> spreads variously to different languages in eastern North America.  Siouan is
> already differentiated into its daughter languages; Algonquian is not.  Thus,
> proto-Algonquian, proto-Dakotan, proto-Dhegihan, proto-Chiwere, etc., might be
> contemporaries, each adopting a technical term that comes from one of the
> Siouan daughter languages or, perhaps more likely, Muskogean or some other
> southeastern language.

The problem is with this scenario is that Algonquian is a few thousand years
old, and since this word reconstructs very cleanly in Algonquian (and is
present in every corner of the family), Occam's razor would lean heavily
toward saying it was present in Proto-Algonquian, or, at the very least, a
VERY OLD loan in Algonquian (almost the same thing). By the time this word
was present in Proto-Algonquian, Proto-Siouan obviously hadn't split off
into its daughter languages yet, either. If this etymon was present in some
Siouan language that far back, that's almost equivalent to saying this word
was present in Proto-Siouan. And if I understand Bob correctly, this word
will not reconstruct in Proto-Siouan. Right?

BTW, Bob, what is the geographic distribution of this word in Siouan? All
over, or only in the Mississippi Valley languages, or what?

(Far as I know, the word doesn't reconstruct in Muskogean, either. The word
obviously has a special status in Algonquian.)

> Either of these models suggests that proto-Siouan is earlier than
> proto-Algonquian,

You lost me here. Why would your first model suggest that?

> and that the spread of the term between the two language groups takes place
> after the divergence of proto-Siouan.

It would help to know whether Proto-Algonquian was older than Proto-Siouan
or vice versa, but offhand I couldn't offer an answer to that.

> The proper level of comparison for internal consistency would be the Siouan
> daughter branches with each other and Algonquian, not all of Siouan with
> Algonquian.
> 
> To choose what language group the term originated from, we should probably
> look for corresponding verb roots that really make solid sense for deriving
> the noun (not fanciful constructions that might have been chosen to chime with
> a foreign term).

Well, in this regard, within proto-Algonquian, this word is totally
reconstructible but cannot be segmented. The root is unrecognizable. That
might speak to it being a loan into Algonquian EXTREMELY long ago. Or
perhaps it simply shows an old root that dropped out of use everywhere
except this noun.

> In this light, exactly what is the reason for supposing that the 'squash' term
> passed from Algonquian to Siouan (or vice versa)?

Well, if the word is not reconstructible within Siouan, and if it has a
limited distribution within Siouan, that's a big hint that it's not that old
within Siouan. And nothing argues against it being old within Algonquian. If
the word was borrowed in Algonquian and old in Siouan, you'd expect it to
(a) reconstruct cleanly in Siouan and (b) to not reconstruct well in
Algonquian, and (c) to be missing from big swathes of Algonquian. The latter
phenomena are not seen.

If I understand Bob and John correctly, the Algonquian loans into Siouan are
usually concentrated in Dakotan, Dhegiha, and Chiwere -- that is, the
languages with the closest proximity to Algonquian. If this word is likewise
missing from, say, Tutelo, Biloxi, and Crow/Hidatsa, there'd be no way to
explain that if this word was present in Proto-Siouan. But if it's a loan in
Siouan, we'd expect that distribution.

David C


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/siouan/attachments/20060827/34f30d48/attachment.html>