Re: GROUPLANG: optional features and case

Herman wrote :
> >ERG Ergative
> >PAT Patientive =3D 'Patient', 'Accusative' and 'Avoiding'
> >ABS Absolutive =3D 'Undergoer'
> >AGE Agentive =3D 'Copulative'
> >ATT Attributive =3D 'Genitive', 'Modifier', and 'Dative'
> >I also vote for CAUS Causative because it's easy and saves time and =
> dificult verbal suffixes.
> >Let me know what you want to change or keep.
>
> Well, the only case I remember suggesting was Genitive anyway, which =
> falls
> under Attributive, so this is fine with me.
Attributive in agent/patient system usually shows an absolute feature :
att-he death = he dies = pat/abs-he dying
att-he head = he features a head
att-he illness = he is ill = pat-he illness
Compare absolute attribute : 'att-he home' = he has a home' = 'he lives somewhere' and
relative attribute 'att-he house' = he has a house' = 'he is a landlord'.
Compare also 'att-he clothes' = 'he wears clothes' (absolute) or 'he owns
clothes' (relative).
Genitive is a case of nominative system to show indifferently possession, relative and
absolute attribution and even origin like ab-lative.

>The term "agentive" is a bit
> confusing,
but we'll have our own "grouplang" words for the cases

Yes, case names are confusing here because they are originally designed to work
within only one of the 3 definite frames : nominative/accusative,
ergative/absolutive and agentive/passive systems. Mixing the 3 systems as we do
here shuffles everything awrong. We need coin new names to show reference in
this system.

For example : 'agentive' is either 'nominative' or 'instrumental' in a nom/acc
system and 'absolutive' in an ergative system.

I would say :
ergative = farmer (he who *grows* the world)
patientive = prey
agentive = mask (to play the *role* of a boss, a hammer, a desease, etc.)
absolutive = sun (the *rising* sun)
attributive = colour (*feature* of the world)
causative = smither (the *Maker*)

> eventually. (BTW, what should we call the lang itself?)
the Tongue :-)

> I also think that dative doesn't really belong under the attributive =
> case,
> but I'm not sure what would be a better place for it. Absolutive, =
> perhaps?
Dative is a case in nom/acc system usually crammed with benefactive, final and many
other obviate cases as remote as genitive (Hebrew : I have a car = a car is to
me).
So in this mixed system, you would find it a bit everywhere. I've crammed in
'agentive' and 'attributive' all issues some of us will certainly discuss again
like 'modifier' (which is actually 'integration' - not a case), 'copulative'
(which is actually 'equative') and 'dative'/'avoiding' (which are actually the
benefactive and obviate patientive in 'I protect the child (ben) from the wolf
(pat)').
>
I'd rather tag on the predicate whether it's =
> verb or noun-rooted, then
> >> >cases would be understood from context.
> It sounds essentially equivalent to my suggestion to use specific
> derivational affixes. Making it optional would allow for brevity when the
> meaning is obvious.
>
Yes, exactly like Japanese verbal suffixes. That's what Pablo means.




-----
See the original message at http://www.egroups.com/list/conlang/?start=17508
--
Free e-mail group hosting at http://www.eGroups.com/