animate wa-

[Wablenica]

Hello John:
Monday, January 5, 2004, 10:30:18 AM, you wrote:

KJE> On Sat, 3 Jan 2004, Wablenica wrote:
>> I would like you to explain me one thing: Buechel has about 60
>> derivatives with wicha'- prefixed, and 30 entries with wicha- infixed,
>> almost _all_ of them are nouns. At the same time there are 2600 words
>> starting with wa-, 1600 of them are verbs. So how can we talk about
>> wicha- being productive with animate verbs, if we have only a couple
>> or two of them, thiwichakte and wichak'u?

KJE> As I understand it, the anomaly arises from the bulk of the productive
KJE> instances being essentially verbal in nature, and predictable from the
KJE> stem.  Wa and wic^ha forms that are fixed nominal expressions or seem
KJE> idiomatic in use are listed in the dictionaries, but those that are
KJE> predictable "detransitivized" forms derived from transitive verbs are not
KJE> listed.

--I agree that the Buechel's dictionary cannot be considered a
reliable source for derivation type frequency
estimations.

However I haven't yet seen the strong evidence for wicha- as a
detransitivizer. Let's recall Regina's examples once again:

(1)     okichize el ota      wicha-kte-pi
        war        in  many  WICHA-kill-PL
        'many were killed in the war'

(2)     *okichize el  ota       wa-kte-pi
        war           in  many  WA-kill-PL
        'many were killed in the war' "

IMHO, the first sentence has an explicit PAT, a pronoun "ota", and wicha- here is
just a regular pleonastic 3d person plural pronominal affix. Ota is
"non-specific", not wicha :-)!
This exlains the ungrammaticality of the second sentence: wa- cannot
plug the valence already "filled" with ota. I would even translate the
sent.#2 as "many made a killing in the war", if detransitivized wakte could be freely
used as an independent verb, not an adverbial (wakte glipi).

To my mind, we could talk about some additional function of wicha- if
it could occur
a) in sentences without "ota", "huNx", etc.:

okichize el wicha-kte-pi
-with non-specific meaning, not anaphoric ("people were killed in the
war", not "(they) were killed in the war")

b) and, more important, with non-specific _singular_ patiens implied:
? John the-wicha-xila.
? "John fell in love".

However I doubt that such ambiguous usage of wicha- may easily spring
forth.

Regina wrote: "I hope that with ota 'many', I have created a PAT that
is non-specific enough to "deserve" being
cross-referenced by wa-, at least theoretically."

--But I guess that the usage of wa- is triggered by formal grammatical
conditions, not semantical ones. Likewise, we can say:
wichasha wan ktepi, "they killed some man"
tuwa ktepi, "they killed somebody"
wanji ktepi, "they killed (some)one"
--but wa- cannot be used here too.

Finally, talking about the productivity, I'd like to suggest the
following thesis:
While stand-alone verbs with a wa-plug almost always have some
idiomatic shades of meaning, _most_ transitive verbs may take wa- in
some compounds, like those V+V cases described in Regina's great
article:

wa-khute okihi - he's able to shoot
wa-khute chin - he wants to shoot
wa-khul wayuphike - he's an expert in shooting, etc.

Then what is the meaning of
wicha-khute okihi
wicha-khute chin
wicha-khul wayuphike ?

Best wishes,
Constantine Chmielnicki