Omaha-Ponca bi vs. i with "egaN"

[rlarson at unlnotes01.unl.edu]

This is another discussion of OP egaN, in reply to two of the
messages posted by John Koontz on September 22.

First, I'd like to summarize John's model as I understand it:

  Two different conjunctions exist, both pronounced egaN.

  The first means "having".  It can be accented on either
  the first syllable or the second, depending on where it
  would best fit to extend an alternating accent pattern
  from the last accented syllable of the preceding verb.
  If the preceding clause is third-person proximate, an
  -i/-bi particle is added between the verb and the conjunction.
  In the case of the "having" egaN, the particle chosen is -bi.
  Here, the [i] is generally elided, giving us b=egaN.
  Then, if the last syllable of the verb is accented, the
  accent will be b=egaN'; but if the penultimate syllable
  of the verb is accented, we will have b=e'gaN.

  The second form of egaN means "so", "as" or "because".
  It is always accented on the first syllable.  If the
  preceding clause is third-person proximate, the -i/-bi
  particle chosen is -i.

John, is this a fair statement of your position?


My model is as follows:

  There is just one word egaN, which is a compound of
  e, "that", or "the preceding", and gaN, "so", "thus",
  "like", or "in such manner".  The first element captures
  the preceding idea and feeds it into the second, which
  makes it an abstraction if it is a noun or a verb, or an
  affirmation of the whole idea under discussion if the
  preceding is discourse.

  Used as a conjunction, egaN ranges in meaning between
  the sequentiality implied in our English word "having",
  and the causation implied in our English words "so",
  "as" and "because".  It tends to mean that the foregoing
  clause is/was a prerequisite for the following clause.
  It does not necessarily mean that the foregoing clause
  is *the* necessary and sufficient explanation of the
  following clause, but it does imply that the following
  clause would not or could not have happened had the
  foregoing clause not been in place first.

  Used as a conjunction, the accent on egaN is normally,
  but not always, on the second syllable; otherwise it
  normally falls on the first syllable.

  Use of the particles -bi or -i depends entirely upon the
  semantics of the preceding clause; this usage is
  completely independent of the conjunction egaN that
  follows the clause.  The particle -bi signals that the
  foregoing clause is based on hearsay, while the particle
  -i may be used when the speaker is making a personal
  assertion.  The latter is the normal mode of speech,
  and sometimes appears in short stretches even where -bi
  ought properly to be used.  Since a myth is based on
  hearsay, narrative clauses are normally qualified by -bi.
  Dialogue clauses, on the other hand, usually represent
  the character's personal assertion, and therefore usually
  use -i when appropriate.  This is why I stress that we
  need to partition the narrative statements from the
  dialogue statements when analysing this issue.

  I am not yet convinced (though open to being so) of the
  existence or utility of the proximate/obviative dichotomy
  in OP.  (John was explaining it to me here a couple of
  months ago.  He offered a formula and a made-up example
  to illustrate it, with the suggestion that I try it and
  see if it didn't correspond to what I found in Dorsey.
  I've tried it and I still don't get it.  The formula
  seems not to cover large areas of the language, the number
  of possible obviative statements seems minuscule, and when
  I do find a third person statement with an "obviative"
  subject that takes dhiNkhe' or thaN or whatever, it seems
  as likely to take a clause-final -bi as any other
  narrative statement.  I asked for the practical semantic
  difference between "The Omaha saw the Pawnee <proximate>"
  and "The Omaha saw the Pawnee <obviative>", and was
  referred to Ardis, who did not take the bait.)  Until I
  am shown some compelling evidence in Dorsey for the
  proximate/obviative distinction, I can't take this model
  seriously as a factor in the current analysis.


I had offered some examples of egaN used as a conjunction,
with and without preceding -bi, from the short story "How
the Rabbit Killed the Giant", pp. 22-25.  I pointed out
that egaN in the narrative statements were preceded by -bi
at the end of the preceding clause, while in two dialogue
statements that used conjunctive egaN there was no preceding
-bi.  John critiqued my argument last month, offering his
interpretation of the -bi-less cases.

>> Rory:
>> Finding compound third-person dialogue statements is
>> difficult, but I found two of them in the story,
>> "How the Rabbit Killed a Giant", pages 22 - 25.  The
>> first is on page 23, line 10-11.  When the giant demands
>> to know which of them had had the audacity to cut up the
>> deer they had shot, the two frightened men admit that the
>> Rabbit made them do it:
>>
>>     She' akha' MashtshiN'ge-iN' akha' pa'de wa'gazhi egaN'
>>     aNpa'dai ha -- "That one, the Rabbit, bade us cut it up,
>>     so that's why we cut it up".
>>
>> Here the accusation arises from personal experience, and
>> the men do not precede egaN' with -bi.

> John:
> The first example is 'having', and should have =bi if it is
> proximate, but it is evidently (consistent with the context)
> obviative and has neither =i nor =bi.

I had understood before that proximate statements were marked
by =i or =bi at the end, and subjects that took akha' or ama'
as articles; while obviative statements had no =i or =bi at
the end, and had subjects that did not take akha' or ama' as
articles, but rather such "object" articles as dhiNkhe' or
thaN.  Here we have a clause that has no =i or =bi at the
end, but has a subject that takes akha' as its article.  Am I
misunderstanding the criteria for distinguishing proximate
from obviative?


> Note also that Rory has reglossed 'having' as 'so' in
> [this] case.

Dorsey glosses this egaN' as 'having' in the interlinear;
I offered a free translation without including the interlinear.
Dorsey's own free translation on page 25 is as follows:

    "That one, the Rabbit, commanded us to cut it up,
     and so we cut it up," said the two.

This uses 'so', the causative translation of egaN, as I did.

And in fact, 'having' really doesn't work very well in this
case.  If you have been disobeying your supervisor's standing
orders at work, at the insistence of your charismatic co-worker
Steve, and have just been called on the carpet by your
fire-breathing boss, do you explain yourself:

    "Steve having insisted that we do it his way, that's
     what we did"

which would sound like a history professor lecturing on the
progress of Caesar through Gaul?  Or would you not rather say:

    "Steve insisted that we do it his way, so that's why
     we did it that way"

which clearly places the blame where it belongs?

In the story, the men are trying to justify themselves to an
outraged authority figure, not to give an impartial history
of their actions.  Despite the interlinear gloss of 'having',
any sensible English translation needs to use a conjunction
of causation, not of sequentiality.

But if we accept that egaN' in this case should really be
translated as 'so' or 'because' rather than 'having', the
accent on the second syllable of egaN' seems to contradict
the criteria given for the egaN of causation:

> But i before e'=gaN 'as, because, so'.
>
> This conjunction always has initial stress.

unless we assume that the rule for initial stress on the
'as, because, so' conjunction egaN works only if the
preceding clause is proximate.

>> Rory:
>> The second is on page 23, line 17-18.  As the giant
>> proceeds to maul him, the Rabbit declares the difference
>> between himself and the craven men:
>>
>>   Dhe'ama naN'dhiphai' egaN' a'dhikhi'dha-bazhi'-hnaN'-i;
>>   wi' naN'wipha ma'zhi egaN' a'wikhi'bdha ta' miNkhe. --
>>   "These ones fear you, so they don't attack you;
>>   I fear you not, so I will attack you".
>>
>> Here again we have no -bi in front of egaN' in either
>> of the two places it appears.  The first one has -i,
>> which can be construed as the plural particle.  The
>> second has only the first person negator ma'zhi, but
>> can't be counted in this test since its subject is not
>> third person.

> John:
> The second example has two intances of egaN which were
> glossed 'because' in the original, cf. 'as', fairly
> reglossed as 'so', As we are dealing with the 'as',
> and the first case is a third person plural it predictably
> has =i, while the second case is a first person singular
> and so lacks both =i and =bi.

Here we have two instances of egaN', accented on the second
syllable, glossed in the interlinear as 'because', glossed
as 'so' in Dorsey's translation, and accepted by John as
the 'as' version of egaN.  This definitely contradicts the
rule that this version of egaN is always accented on the
first syllable, unless that rule was meant to apply only
in case the preceding was third-person singular proximate.

But if egaN 'having' can be accented on either syllable,
and egaN 'as, because, so' can also be accented on either
syllable, then we have no phonological basis left for
distinguishing the two.  We are left with only a presumed
semantic bifurcation that depends on Dorsey's glosses.
In other words, Dorsey's choice of a suitable English
equivalent for egaN in various contexts makes the OP word
egaN into two distinct words to separately match the
functionality of the English morphemes we depend on to
translate it.  This is as if a German linguist analyzing
English were to conclude that the English word 'but' is
actually three distinct homophonous words: one signifying
'aber'; another meaning 'doch'; and yet another that
equates to 'sondern'.


> In dealing with Dorsey's texts one has to be cautious
> about his glosses, but one ignores them at one's peril.
> Thus, he spuriously glosses many =bi markers as 'they say',
> working from a false conclusion as to the relation
> of =bi=ama to the gloss 'they said', but a careful
> consideration of the evidence suggests that it may
> provisionally be taken as an error.  I have not yet
> regretted making that provisional assumption, and so
> I stick with it.  On the other hand the consistent
> pattern of 'having' vs. 'as' does reflect something
> very real, if easily overlooked, since it corresonds to
> something real in meaning, morphology, and phonology.
> Dorsey's very tendency to consistency in glossing makes
> the one practice a probable error and the other an
> important distinction.

There is nothing spurious about Dorsey's gloss of the
=bi markers as 'they say'.  I am fairly confident by
now that every active, non-fossilized instance of =bi
in Dorsey carries the functional meaning: "The foregoing
is based upon something that has been said; it is not my
current personal assertion."  We have no good English
equivalent of this morpheme.  Our word 'suppose' comes
close, as in: 'supposing X', 'supposedly X' or
'supposed to do X', though even this doesn't capture the
sense of =bi that the supposition is based on something
said.  In the narrative statements of a myth, the
meaning of =bi and ama', 'they say', are almost
equivalent.  They do not fill the same syntactic slots,
and they are functionally different otherwise, but at
that point their meaning is so close that they tend to
merge into a single word.  When =bi comes alone at the
end of a narrative clause to let the listener know that
the statement is based on hearsay, rather than being an
assertion upon which the narrator is staking his personal
honor, what better English gloss can we find than
'they say'?

I agree that one needs to deal judiciously with Dorsey's
glosses, especially when beginning to learn the language.
But I also think that following the glosses slavishly
without learning to grasp the meaning directly from the
Omaha-Ponca language behind them can lead us into errors
of our own making.  In particular, I object to the
assumption that seems to be buried in some of John's
arguments, that there must be some sort of one-to-one
match between English words and Omaha words semantically.
It seems if we find two different English glosses for
the same Omaha word, then either the Omaha word is
really two separate words to match our English functional
paradigm, or else Dorsey has made an error.  This is
unfair to Dorsey, who was a magnificent scholar, and it
needlessly throws sand in our own eyes.  The functional
morphemes in Omaha need not translate precisely into
corresponding functional morphemes in English.  The range
of a particular Omaha word might overlap the territory of
several different English words, depending on context, or
it might have no good English translation at all.  In
this case, why would the glossing of that word not be
"inconsistent"?  That "inconsistency" would not be
Dorsey's fault; it would simply be inherent in the
nature of translation.

Rory