transitivity, locative prefixes & the pronomin. argument hypothesis.

[Koontz John E]

On Fri, 27 Sep 2002 rlarson at unlnotes01.unl.edu wrote:
> > I don't know how you would say something like 'Since I
> > don't have you, my life isn't worth living.'

> Would the first clause be something like:  Dhi aNdhiN'ge (egaN')...  ?

No idea, really, though I think not, as I've never run into any verbs with
multiple patient inflection, and the closest thing to an independent
pronominal without pronominal agreement I can think of is aNgutta as the
alienable possessive (cf. aNgu 'us' as independent inclusive).  But all of
the alienable forms tend to look like that in OP, and this is kind of off
topic, so ...

> The word dhiNge' shouldn't be thought of as 'to lack', which is
> transitive in English, but rather as 'to be gone' or 'to be lacking'.
> You can then have a subject which is missing with respect to me.
> This is normally in the third person, so we don't get an affixed
> pronominal as a visible parameter to the verb.  So if the emphatic
> pronoun dhi is a free argument here, does it need a corresponding
> affixed pronoun, or can it be treated the same as any other free
> nominal argument?

I like to gloss verbs with English verbs that have similar argument
structures, too, but this is an area in which it gets a bit difficult at
times.  I think the morphological argument with dhiNge is always in
patient form, and always signifies the person experiencing the lack.  The
person/thing lacked has to be a third person reference (I'm not sure it
can be a person), and coincidentally the verb, of course, doesn't agree
with it.

I'll suggest somewhat uncertainly that this is true across at least
Mississippi Valley Siouan, hoping to hear some confirmations or
contradictions.

I think there are several other verbs like this in Dhegiha languages, and
also in other Mississippi Valley Siouan languages, and that we tend to
overlook them in classifying Siouan verbs.

The general syntax is

NP arg-verb

where

NP is a nominal (always third person)
arg is a patient (or dative-patient) prefix on the verb

arg experiences something involving NP

I expect that it must be possible to echo arg with an independent
pronominal for focus purposes, as usual, though I don't have any specific
examples.

Other verbs like this may be 'to lose/drop' and 'to have one's own die'
(with dative patient).  However, there are others involving bodily
sensations and states.  We've discussed them before as dative-subject
verbs.  The NP is often incorporated, leading to infixing verb stems.

> aNska'.                 I am (colored) white.
> hi ska'.                The teeth are white.
> hi aNska'.              My teeth are white.  (= the teeth are white with
>                                                 respect to me. )
>
> This sequence, if correct, would mean that a stative verb can take two
> arguments, to one of which the quality is ascribed, and the other of
> which is sort of the indirect object of that relationship.

One could interpret this as something different, involving possessor
raising, and it is different to the extent that dhiNge plainly always
involves two arguments, while ska doesn't.  I don't think it is really
possible to use dhiNge in cases like 'you were missing'.

However, I think examples like this behavior of ska are really comparable,
anyway.

There's a small difficulty is knowing whether hi ska is 'his teeth are
white' or, getting fancy 'he experiences white teeth'.  But it doesn't
strike me as particularly problematic if this middle case - middle in the
list above, I mean! - is either ambiguous or an alternate argument
structure.  Clearly there are some cases that parallel dhiNge in argument
structure.  We should be used to verbs that have multiple argument
structures, some intransitive and some transitive, from English, e.g.,
'roll' in 'it rolls' and 'I roll it', or 'eat' in 'I ate' and 'I ate it',
or even 'it eats like chicken'.  Or 'feel (OK)' with 'it feels OK (to me)'
and 'I feel OK', where the last two cases involve intransitive structures
with different kinds of argument.

The thing to notice, though, is that in the two argument cases in OP, the
verb marks the experiencer and uses the same pronominal series it uses
when it's marking the subject in one-argument cases.  I've switched to
saying one-argument and two-argument, because it's clear that there are
difficulties in calling the two-argument cases transitive in the Siouan
context.  I don't want to get hung up on terminology.  I think people who
agreed on the facts of these forms might disagree on whether to call them
transitive, both for theoretical reasons and because these cases are
somewhat intermediate.

This is essentially the point Rory goes on to make.

> We might conceive it differently, though, if we suppose that the
> nominal argument can function adverbially as a qualifier of the verb;
> i.e. hi aNska' could mean "I am white teeth-wise".  Then hi ska'
> could be interpreted in either of two different ways.  But if this
> were true, "ShaN'ge aNdhiN'ge" would have to mean: "I am missing
> horse-wise", which doesn't seem to make much sense.  I think I favor
> the first possibility.
>
> Interesting issue...

Yep!