Re: ergative? I don't know...
| From: | Sally Caves <scaves@...> |
|---|
| Date: | Sunday, October 25, 1998, 0:11 |
|---|
On Fri, 23 Oct 1998, David G. Durand wrote:
> Clinton is right, as the agent patient system is often called an "active"
> system (but I don't think I used that term).
>
> The question that I have is whether Clinton's experiencer is really the S
> role, or some sort on indirect case. If it's always used for S (subject of
> intransitive sentence), then we have an active, or agent/patient system. If
> not, it still might be used in the S role to differentiate level of
> agenthood in intransitive sentences. I thought that this was Sally's
> suggestion.
My suggestion, or rather question, was whether Teonaht could be considered
as at least incorporating characteristics of an active language within a
basically nominative/accusative language: in other words, having either a
nominative with two partitions (one that made a distinction between
volitional and non-volitional agent) or two separate cases: nominative
volitional and nominative non-volitional. Along with a strongly
accusative case. I think it more logical to see Teonaht as doing the
former. If it does, then what kind of linguistic system does it fit into,
and are there other natural languages that do this? From the overall
silence (overwhelming, in fact) that greets this question that I've posed
numerous times, I'll have to conclude that there aren't, and that what
I've concocted here is unique.
As usual, I've been answering these posts out of sequence: I find out how
many hundreds of new posts I've accrued by jumping to the bottom of the
file and working my way up. So I haven't even answered David's fine
explanations of the differences between these language types and shall do
so now.
Sally.
>
> There are a few ways that an extra case could be used to supplement such a
> standard system.
>
> If the other case used for S in intransitive sentences is Agent, it's
> essentially a nominative/accusative system that uses an indirect case to
> modify the agentive status of intransitive verb arguments. I assume that a
> nom/acc system would use Patient for agentive instransitive clauses, and
> the indirect case as a de-agentive variant.
>
> NOM=AGT, ACC=PAT:
>
> he-NOM kill you-ACC.
> David-NOM fidgets. (actively as agent)
> door-EXP closes. (not actively)
>
> If the Patient is the default, you'd have an essentially ergative system,
> with the same kind of distinction. An erg/abs system might use the indirect
> case to express "agentivization" of the normal absolutive argument, but use
> of the ergative seems more natural for this function.But there's no reason
> that things have to be reversed. The absolutive could still be thought of
> as not an agent, but as still more agentive than an "experiencer", so that
> the indirect EXP case, would be a de-agentivized absolutive or Patient case.
>
> ERG=AGT, ABS=PAT:
>
> he-ERG kill you-ABS
> David-ABS fidgets. (does it himself, actively)
> door-EXP closes. (closes, but not actively)
> door-ABS closes. (ooh spooky Halloween, the door closed itself!)
>
> The interaction of such systems with passivization and other argument
> moving processes is really interesting...
>
> -- David
>
>
>
> _________________________________________
> David Durand dgd@cs.bu.edu \ david@dynamicDiagrams.com
> Boston University Computer Science \ Sr. Analyst
>
http://www.cs.bu.edu/students/grads/dgd/ \ Dynamic Diagrams
> --------------------------------------------\
http://www.dynamicDiagrams.com/
> MAPA: mapping for the WWW \__________________________
>