Re: Verbal nouns

On Tue, 3 Oct 2000, jesse stephen bangs wrote:

> > Chevraqis can treat verbs as nouns and nouns as verbs.  In the first
> > case, they decline like other nouns, and in the second case, you can
> > conjugate them.
> >
> > Example:
> > nabazu: to insult/be the enemy of, infinitive.
> >
> > Sjar naía nabazas: You insult me.  The conjugation -as is present reportive.
> >
> > If you wanted to say "Insulting people is bad," you'd have to decline nabazu:
> > nabaza (voluntary actor case)
>
> Excellent!  This sort of noun/verb conflation is very similar to what
> Yivrindil does, except that the ending is not a zero ending.  In the
<blink>  Zero ending?

I list all nouns in the voluntary actor case for convenience.  I guess
the "root" noun form would be nabaz- but it *never* occurs in Chevraqis.
(Unless Qenaren/Avren linguists are smarter than I am, which is very
possible.)

> time some noun class markers fused with the nouns to form the basis for
> current Yivrindil nouns, and these were taken as the bases from which all
> new verbs were derived.  Thus, in current Yivrindil the base lexicon per
> se contains only nouns, with all verbs being very simply derived from
> nouns.
Neat.  :-)  Theoretically I could just list the base infinitive verb,
since I'm using triconsonantal morphology and all the other infinitive
verbs, adjective and noun forms are derived using affixes.  But when you
have a base verb "rasranu" that means "to race/run" and its related forms
mean things like "wind," "horse," and "journey," it would be too
confusing to just list one verb form.  Linking together all the different
meanings of verb/noun/adjective forms is really fun because it gets me to
think about the concultures involved.

> Do you have trouble with inflected forms of nouns or verbs being
> ambiguous, so that morphologically it might be impossible to tell if a
> given word is being used nominally or verbally?
Almost.  -ad is the locative suffix, but in one of the conjugations--I
*think* it's present plain--verbs conjugate with -ad.  However,
verb-stems and noun-stems are usually distinguishable by the vowel affixes,
e.g.

rasjaru: infinitive/imperative
rasjoru: generic infinitive
resjaru: habitual infinitive
...
rasra: noun describing state (journey)
rasrana: noun describing actor (horse)
resra: noun describing tool (wind)

Now, if this didn't happen to have a mutation from "sj" to "s" you'd be
unable to distinguish from "resjarad" (habitual journey: verb in
nominalized locative) and resrad (wind: locative).  I don't have access
to my notes or I'd give a real example.

Basically, the habitual infinitive has pattern CeCaCu
and the noun-tool has pattern CeCaCa ("resara" is contracted to
"resra"--special case).  When you strip off the -u infinitive ending and
replace with -ad (locative suffix) and give the -ad conjugation instead
of the -a voluntary agentive case, the two are indistinguishable.

I figure context will do the job where I screwed up morphology.  :-p

> > This doesn't work or make much sense for every noun I have, but I'm
> > working on metaphorical or poetic usages...eventually.  :-)
>
> Yeah, I run into the same problem.  There are some verbs that don't seem
> to come from any noun, and some nouns that don't make any sense as a verb
> form.  With some work you can make the system fit, although I've found
> that it's impossible to have the lexicon be *completely* regular with it
> noun-verb correspondences.
<G>  I don't even try for complete regularity, but then again, the
conlang is supposed to be theoretically somewhat naturalistic.  All the
better for poetry....

YHL