Re: Patient marking in active languages
From: | Carsten Becker <carbeck@...> |
Date: | Sunday, February 19, 2006, 13:29 |
From: "caeruleancentaur" <caeruleancentaur@...>
Sent: Sunday, February 19, 2006 2:35 AM
> This may be so with the other verbs you mentioned, but my
> personal experience is that riding a horse is not at all
> a passive activity! Nor is riding a bicycle as I see it!
> :-)
Agreed, but I think he rather means how Active languages
handle passives.
So first, consider the following:
[I] [ride] [a horse]
A V P
Is an active sentence with A = agent, V = verb, P = patient.
In English, the _passive_ form of this sentence is
[A horse] [is ridden] [(by me)]
P V A
where the A argument can be omitted. In ergative languages,
there is something called _antipassive_, but I don't know
which construction this would produce, unfortunately --
maybe the absolutive object can be omitted? However, in
accusative-language-style passives, the object of a sentence
is made the subject, while the original subject becomes
oblique, i.e. it is not crucial to form a valid sentence.
This is called "valency-decreasing" if I remember correctly.
So, I am not a pro in Linguistics, but I know that purely
accusative languages arrange their cases as [S A] [P], while
purely ergative ones have [S] [A P], where S is the single
argument of an intransitive clause, and A and P as above.
Active languages have both arrangements, i.e. [S A] [S P],
depending on whether a verb is classified as active or not
(stative, passive, I don't know how you best call it).
According to Daniel Andréasson, there exist three
distinctions that this difference in classification can be
based on: whether it's an event (±E), whether the action is
performed, effected or instigated (±P/E/I) (whatever that
means) or whether it's controlled or not (±C). Languages
can mix these three criteria.
Accordingly, active languages (i.e. fluid-S ones) may either
work like English and German and such and have a passive
voice, or they have an antipassive voice, or have both
depending on the classification of the verb, or have none.
Additional to agents and patients, I have got experiencers
in Tarsyanian, which would exactly match the subject of
passive sentences.
I hope this helped at least a little.
Yours,
Carsten Becker