Locative Postpositions

[Koontz John E]

On Thu, 28 Oct 1999, Robert L. Rankin wrote:
> In Dhegiha there is a distinction between {?e:}, the demonstrative, and
> {he} 'to be', so I'd expect all the e's in DH to be of the demonstrative
> variety.  You find the {-he} 'be' compounded with positional verb roots to
> form auxiliaries (reduced biclausal constructions).  ...

...
> I don't think so.  I think they are historically different.  They just
> look alike in Dakotan because they've become homophones and thereby
> subject to a lot of confusion. Because of their short (sometimes
> identical) phonological forms, we must be very careful in analyzing
> them in order to determine which we're dealing with.  In this case though,
> Dhegiha preserves the evidence (my previous post) that 'to be' had an
> initial /h/ that has apparently been lost in Dakotan for whatever
> reason.  Is there evidence for the *h anywhere in Dakotan?  The place to
> look would be frozen compounds.  How many morphemes is echa
> historically?  If it's related to ekta, then could -ha be ablauted *he.


I definitely agree that *(?)e and *he are different, and that the Dhegiha
postverbal he (second person s^e) forms are the *he root.  I just think
that the Dakotan forms cited so far don't represent this *he, but rather
*e, albeit used predicatively.  There are actually similar usages in
Dhegiha, for what this is worth, clearly involving the *e form:

JOD1890:439.20  Dhe'=e ha  this is it
JOD1890:497.17  S^u'kka=maNdhiN e'=e ha  it is S.
JOD1890:597.1-2 Xdhabe' dhe=the maNghe idhabattu=tte     ehe=   e
ha
                tree    this the sky   it will extend to I said that DECm
                What I said was let this tree extend to the sky.
             or It is that I said let this tree extend to the sky.

JOD1890:647.4   Hu'ttaNga wamaNdhaN=dhaNkha  we'bahaN      e'=e       ha
                Winn.     thieves   the (PL) they know him that it is DECm
                It is this:  he knows who the Winnebago [horse]thieves
are.
     or maybe   It is that he says that he knows who the Winnebago thieves
are.

I notice also that 'dead' is often rendered with t?e'e 'it is that he is
dead' or perhaps 'he is one who is dead'.

JOD1890:487.17 Frank wa?u' miNgdhaN, e'=de t?e'=e ha
               F.    woman he married but she died
       ???     A woman whom F. married has died.

JOD1890:512.4  I'kkuhabi=s^ti t?e'=e     ha
               I.        too  he is dead DECm

The =e can also be used in questions.


JOD1890:567.4  A'wa=the=e    a?
               where    that QUESTION
               Where is it (of two places)?


Examples with pronominals:

JOD1890:77.9  wi'=e=bdhiN 'it is I'
JOD1890:22262.15:  dhi'=e=hniN=de 'since it is you'
JOD1890:113.7:  e'=e he 'it is he DECf'

Of course, the third person is just what we've been seeing.

The example from p.  487, with ede seems to me to involve an Omaha-Ponca
analog of Dakotan ec^ha (indefinite topic).  Dorsey invariably renders
this (or =de) as 'but', but the Dakotan form is reported to be associated
with unexpectedness (Rood & Taylor, if I recall properly), so this seems
reasonable.  Even definite relative clauses were not quite recognized as
such in Dorsey's work, though he seems to have understood their practical
import well enough.

We can also find forms with edi 'at it/that; there' as a predicate:

JOD1890:337.18  e=d=e=di=akha he is here
JOD1890:425.10  ihaN' ugi'ne               j^uba e=d=e=di=dhaN ama
                mother they seek their own some  of the company there are
there
                some of that lot of nurslings are there

The double article dhaN ama is used in partitives.

JOD1890:443.14  s^aaN'=ama e=d=e=di ama    dhaNz^a
                Sioux  the there are there though

JOD1890:697.3   AN'phaN=wadaNbe=dhiNkhe e=d=e=di=   kki
                Elk-gazer       the     he is there whether

All these constructions amount to:

NP [article] e=d=e=di [article]
(the NP) is there

The articles agree, if both are present.  The final article occurs when
the clause is a main clause.  The first article occurs with the NP, if
any.  The e=d=e=di is either e=di e=di 'there (is) there' with contraction
(would be el etu in Dakotan), or e=de e=di 'one (is) there' (would be
ec^ha etu in Dakotan).  (I don't imply that the Dakotan comparisons
actually exist.)  The contraction makes it impossible to tell.

There are some more existential constructions without the ed=:

JOD1890:88.21  ni'as^iNga j^uba e=di ama ha
               people     some  there were
JOD1890:63.8 AN'phaN he'ga=s^te=aN=z^i e=di=m=ama
             elk     not a few         there were
JOD1890:736.3 e=di'=ma=s^e 'you who are there'

JEK