Re: Stative passive

Quoting Carsten Becker <post@...>:

> A second question: In German, we've got a "Vorgangspassiv"
> (actional/active?! passive) and a "Zustandspassiv" (stative passive). That's
> the difference between "The door is closed" (done by somebody) and "The door
> is closed" (state of being). Is it really necessary to think about this in
> conlanging or is one passive enough to express both.
This would be the difference between _die Tür wird geschlossen_ and _die Tür
is geschlossen_? My books call them "werden passive" and "sein passive", but I
can't think what else it could refer to.

Well, as others have alredy pointed out, English appears to be doing fine with
only one passive. Moreover, only accusative languages, or so I'm told, are
prone to have passives at all. Ergative langs frequently simply drops the
ergative argument - OTOH, they frequently demand the absolutive one to be
present, giving rise to the antipassive voice, which kills off the old
absolutive argument and promotes (demotes?) the old ergative to the absolutive
argument slot.

I am not aware of any anadewism for this, but my conlang Tairezazh, which is
perfectly accusative, avoids the need for a passive voice by simply allowing
nominative arguments (subjects) to be dropped.

Swedish, OTOH, musters no less than three versions for "the door is closed";
_dörren stängs_, _dörren blir stängd_ and _dörren är stängd_. The last
corresponds to the German "sein passive", while the former two to the "werden
passive". I'm hard pressed to point to any difference in meaning between them -
 possibly the second more immediately assumes an agent that the first. (The
ending -s, seen on the first, besides passives also forms reflexives, in
addition to appearing on certain verbs for no reason known to man.)

> And what about "The soup cooks"? It does not cook itself, and it is not
> cooked by anyone, it just cooks. Because I'm working on a trigger language
> at the moment (oh wonder! but hey, it was me who started the
> trigger-language-boom again, actually!): There wouldn't be an agent here,
> right? There is no cook, or is "the soup" the agent and a patient is
> missing?
Spontaneously, I'd say it's an undergoer.

But a heretical part of my mind is telling me that case assignments have alot
more to do with analogy and bad habits than with any semantic universals
like "agent", "patient" or "undergoer". There simply happens to be an English
verb that takes what is cooking s as the nominative argument.

                                                          Andreas