Re: articles
From: | Chris Bates <chris.maths_student@...> |
Date: | Tuesday, February 1, 2005, 20:12 |
>
> I've already wondered about this. IIRC, _ang_ indicates the
> following argument being the agent and _ng_ indicates that
> it's a patient. But what happens if you've got an
> indefinite amount that is the agent resp. a definite amount
> that is the patient?
>
"ang" doesn't mark the agent. It marks the trigger, which is often the
patient, beneficiary, location etc. Basically, it's similar but more
complicated than the English passive:
I was hit by the man
here I is the subject, but I is *not* the agent of the verb to hit.
Similarly, the trigger marks the "core" argument, but that argument may
not be the actor depending on the marking on the verb. "ng" on the other
hand (pronounced nang even though written "ng" btw) marks pretty much
all other arguments except locatives, which take "sa".
So definiteness doesn't feature (or rather isn't grammaticalized) in
this system, apart from the fact that the choice of trigger heavily
depends on which argument is most focused. Since new information is more
likely to be focused than old information, there may be some
correlations between definiteness and the choice of trigger although
this does mean "ang" is in any way an article in the sense or English,
German, French etc.