Verb "To swim"

[Koontz John E]

On Wed, 2 May 2001, Lance Foster wrote:
> I'm looking at some comparisons/usage of the verb "to swim".
>
> In his dictionary JGT has:
>
> to swim: nyiwan; nyinwan; hira
> to bathe: pidha; pitha; hira; pira
> hira/pira "to take a bath"
>
> Any thoughts on usages distinctions in comparison with other Siouan
> languages?

I thought for once I'd look first and answer second.

Omaha-Ponca has:

hi=dha'  (Pop Orth:  hitha<accent>) 'to bathe' The second syllable accent
seems to be anomalous, which may suggest interesting length issues.

niN'=uwaN (Pop. Orth:  niNuwoN) 'to swim' (Swetland), and niN=(w)aN
(Dorsey) The w might be epenthetic or a reduction of the u.

Both of these more or less closely resemble the IO forms, except the pi (=
phi?) alternative, which may be influenced by phi 'good'.

And, since knowing the stem is only a small part of the battle, I looked
into the inflection of these.

I believe hi=dha' is doubly inflected:  ahi'=bdha, dhahi'=na, hi=dha'(=i),
aNhi'=dha(=i), deduced from the following derivative's first upon third
personal form, and consistent with the Osage pattern (in OP form) given by
LaFlesche.  (Osage has dhe 'to go' appended, too, so it is triply
inflected.)

A derivative is i'=hi=dha 'to bathe by means of', e.g., 'to bathe by means
of the rocks used in a sweatbath'.  This is idha'hi=bdha, i'dhahi=na,
etc., cf. i'wihi=bdha 'I will bathe by means of you' addressed to the
stones.

Another derivative is hidha'=khidhe 'to cause to bathe'.

And we might mention hi=dha'=i 'Saturday' or 'bath night', lit. 'they
bathe'.

Actually, the word for '(to take a) sweat bath' is iN=u'ppe, which looks
like it might be 'to crawl into stone(s)'.  The inflection seems to be
iN=u'(w)appe, iN=u'dhappe, iN=u'ppe, iN=aN'waNppe, deduced from the third
and first persons.

The inflection of niN'=waN in Dorsey is obscure, as he provides only
niN'=aNwaN 'we swam' in text.  Unfortunately I can't check his slip file
without a major expedition yet.  So this could perhaps be:  niN'=maN,
niN'=z^aN, ..., or niN'=awaN, niN'=dhawaN, etc.  It's also not clear if
the -w- and/or -uw- are epenthetic w before aN or reflect u- 'in' or a
stem waN or what.  It's possible that both uwaN or u(w)aN and waN or aN
are reflected, i.e., 'to swim in' and 'to swim'.  What LaFlesche gives for
Osage is aniN'=moN, dhaniN'=moN, niN'=moN, oNniN'=moN, which is a
prefixing pattern, contrary to the OP inclusive in Dorsey's texts.

Winnebago has niNiNp, inflected prefixally, which doesn't fit the set
easily, though a similar waffling between *w and *p afflicts the 'to
see/watch' set, which is tuNwaN in Dakotan and *toNpe in Dhegiha.
Winnebago does have, however, niNiN=waN' 'to paddle', inflected
niN=aNwaN', etc., i.e., niN=n^a'waN (?), niN=iN'waN= (?), etc., which more
or less fits the pattern.

Dakotan nuNwaN' is inflected prefixally:  wanuNwaN, etc..  Across the
Missouri from Decatur, NE, lies Onawa, IA, which presumably reflects
Dakotan onuN'waN 'to swim in'.

It looks like some sort of cline from niN(w)aN to nuNwaN exists for the
'swim' set, at least potentially analyzable as niN=(w)aN or n(iN)=o(w)aN,
but maybe not.  In general we expect Siouan verbs to develop from infixing
patterns to prefixing patterns, via various interesting intermediate
possibilities, if they don't start and remain prefixing, but I wouldn't
care to guarantee the reverse never occurs.  Still, it does look, from the
OP and Wi (paddle) forms as if this stem may originally have been
infixing, which tends to support the assumption that the first part of the
stem is n(iN) 'water'.  However, awkwardly, Dakotan doesn't attest niN
like the other languages, but mniN, reflecting original *wriN or *priN
(which I like for Proto-Mississippi Valley), and this stem has #n, not #mn
in Dakotan.  It might be borrowed, but that seems awkward with a verb like
'swim' in the absence of evidence of pervasive borrowing.

Incidentally, Santee has yuz^a'z^a and spaye'ya for 'to bathe'.  Teton
makes the first of these 'to wash clothes' and the second 'to cause to be
wet'.  OP has kkigdhi'z^a 'to clean oneself', which seems to involve the
same root as the first.  OP and Osage have a root -sihi that seems to
involve ritual or metaphorical cleanness.

It looks from the IO, OP, Osage, etc., definitions and cross references as
if the Mississippi Valley Siouan languages employ some of the same
metaphors and ambiguities as English with bathing, swimming, cleaning,
etc.