Re: Active case-marking natlangs (was Re: What is needed in anconlang classificatory system?)

Matt Pearson wrote:

>Tokana is another conlang with an active case-marking system.  Tokana is
>strictly dependent-marking in this regard, inasmuch as there is no agreement
>morphology on verbs.  The breakdown of case forms works more-or-less like
>this:
>
>ABSOLUTIVE:  patients/themes (including subjects of non-eventive verbs)
>NOMINATIVE:  volitional agents of eventive verbs (both transitive and
>intransitive)
What do you mean by "eventive verbs"?

>DATIVE ("to/at/in"):  recipients, experiencers of verbs of
>perception/emotion/cognition, goals, locations, possessors
>ABLATIVE ("from/of/out of"):  sources, non-volitional agents of eventive
>verbs,
>standards of comparison, supersets in partitive relations, reasons/motives,
>substances
>INSTRUMENTAL ("with/by/via/through"):  instruments and inanimate actors,
>measurements/durations, degrees of comparison
While this system makes distinctions not commonly found in accusative and
ergative languages, I don't see why you call it active.

If by "eventive verb" you mean, roughly, that it denotes an event, then
this is not the distinction that active languages are based on. Clearly,
this does not hold true for the Muskogean and Iroquoian langs, and probably
not the Soiuan langs either, though I haven't looked at them very much.

My reluctance to call this active is quite similar to my reluctance to call
Georgian active. The treatment of Georgian case that I have read suggests
that the pattern is correlated with the unergative/unaccusative
distinction. If this is correct, then I would not classify it as active. Or
maybe we need to distinguish two (or more) different types of "active"
languages.



Marcus Smith

"Sit down before fact as a little child,
be prepared to give up every preconceived notion,
follow humbly wherever and to whatsoever abysses Nature leads,
or you shall learn nothing."
                -- Thomas Huxley