The Whorf Hypothesis

[Koontz John E]

On Wed, 18 Dec 2002, David Kaufman wrote:
> I guess what I'm curious to know here is whether there would be an
> alternative in Hidatsa/Crow/Mojave and any other language that expresses
> "(The) man talls" construct to distinguish between the state of BEING tall
> and the process of GROWING TALLER.  Unfortunately all I have is the
> dictionary from 1886 by Matthew Washington in Hidatsa, and nothing from
> these other languages, and we haven't heard from our Hidatsa expert yet,
> John Boyle, who would probably know best from his research on Hidatsa.

[There's a good place to look for the example "getting taller" in stories
in which a spell is uttered that causes a tree in which the hero is
located to get taller.]

The Omaha-Ponca construction is to use snede' 'tall' or maN's^i 'high'
plus a verb of motion, perhaps combined with a positional, perhaps
causativized, e.g., 'arrive high', 'send high', etc.  In the texts these
constructions are sometimes translated '(suddenly) become', or, with other
verbs '(suddenly) begin/start', but the construction is precisely what
I've called 'suddenly' auxiliaries or "aorist"/inceptive/iterative
auxiliaries.  (Someone once told me a Latinate term for "suddenly" verbs,
but I've lost it!) These forms are particularly well developed and
documented in Dhegiha and Ioway-Otoe, but the situation seems similar in
Winnebago and there are a reduced set of obviously cognate forms in
Dakotan, too.  Crow has some auxiliaries (punctual? I forget) that look
like highly rediuced forms of some of these.

Note that OP snede' 'tall' is a stative verb, but maN's^i 'high' is
something different.  It appears by itself as 'high', 'high in the air',
'on high', etc., but also combines with postpositions to produce at least
maN's^iadi and maN's^iatta.  There's also maNs^ia'ha.  This is basically a
directional adverb or maybe directional would suffice.  There are some
comparable forms, I think, with other senses, though I admit I have never
tried to nail down this impression of mine.  The bare form and the
postposition-augmented forms function as predicates, though they are never
inflected.

Also, the construction 'that high' (and similar ones) is ga=thaN' that +
thaN, where thaN behaves morphosyntactically like a postposition, but is
not at all locative.  There are a small set of similar enclitics,
including =naN 'many', as in ga'=naN 'that many', and so on. I can't
explain the difference in accentuation.  The =naN is obviously the cognate
of the demonstrative plural marker in Dakotan.

Here's a fun OP verb I noticed while looking into this:  u..hi' 'to have
grown, been raised' The first person is ua'hi 'I grew' (active), but the
second is udhi'hi 'you grew, were raised' (stative).  This is the only
mixed active/stative verb I can recall.  I may have managed to forget some
examples provided by Dorsey.

Incidentally, the gloss 'grown' (probably 'mature') is attached by Dorsey
to the OP stem naN.

> If there are alternate methods depending on whether one is referring
> to a state or process, then that would correlate with IE languages
> (i.e., he is tall, he is getting taller).

Returning to David's point, as far as OP is concerned there are alternate
methods, though the processual alternative requires an auxiliary.

> If not, then I'm still left wondering why this attribute would be
> stated only in a verb form.  I realize one could say it's because they
> don't use adjectives, but why choose a verb form instead, unless an
> adjective has much in common with a verb (?).

For me the explanation is that adjectives are functionally intermediate
between nouns and verbs, so that particular concepts in this intermediate
range may in some languages that distinguish all three categories fall in
a category other than the one in which similar concepts fall in English.
In extreme cases, very few concepts may be realized as adjectives.  In
fact, a language may eliminate the intermediate category adjective
completely, and transfer all modifier concepts to noun and verbs.  The
verbs of this sort may be inflected objectively, in which case you end up
with what is called an active/stative language.  Essentially all Siouan
languages are of this type.

Of course, this is just a classificatory or typological approach to
analysis, founded on an imprecise "intermediate between noun and verb"
characterization of the underlying mechanics. Obviously logical categories
like "static/state", "entity", and "process" are lurking in the
background.

JEK