Re: Verbal voice

amman iar has something similar in its Applicative voice. In Applicative
constructions the underlying oblique argument of a ditransitive predicate is
promoted to derived P-function and the underlyiing P-function argument is
expressed obliquely.

ACTIVE
adranne eleth an adriel ervannon antharren
Father hit mother with a stick.
        adran-e   eleth an  adriel-0  er-0-banno–o–n          anthar-en
        A=AGT:ERG       PAT P=PAT:ABS AGT:ACTV: :AGT/PAT:ACTN Obl=THM:DAT
        Father    did   to  mother    hit                     stick  with

APPLICATIVE
adranne anthar ertholbannien an adriellen eleth
Father with a stick hit mother.
        adran-e   anthar-0   er-tol-banno–ie–n         an  adriel-en   eleth
        A=AGT:ERG P=REF:ABS  AGT:APPL:   :AGT/THM:ACTN PAT Obl=PAT:DAT
        Father    with stick hit                       to  mother      did

Note that the active form above is the canonical form for all ditransitive
predicates, i.e. A=AGT/P=PAT/Obl=THM. Unlike English, the predicate must
undergo a voice transformation using the Applicative Voice to change this
argument structure to A=AGT/P=THM/Obl=PAT.

Both of these examples emphasize the "stick-hitting", i.e. the manner in
which mother was hit by father.  If the means by which he hit her were
important, the instrumental would be used.

Instrumental
adranne eleth an adriel ervannon antharrion
Father hit mother with a stick.
        adran-e   eleth an  adriel-0  er-0-banno–o–n          anthar-ion
        A=AGT:ERG       PAT P=PAT:ABS AGT:ACTV: :AGT/PAT:ACTN Obl=THM:INSTR
        Father    did   to  mother    hit                     stick  with

Unlike Dyirbal, amman iar does not allow an applicative of this form.

David

David E. Bell
The Gray Wizard
    dbell@graywizard.net
    www.graywizard.net

"Wisdom begins in wonder." - Socrates

> -----Original Message-----
> From: Constructed Languages List [mailto:CONLANG@LISTSERV.BROWN.EDU]On
> Behalf Of Tom Wier
> Sent: Wednesday, May 31, 2000 2:11 PM
> To: CONLANG@LISTSERV.BROWN.EDU
> Subject: Verbal voice
>
>
> I've been reading Dixon's _Ergativity_, and he has this to say
> about Dyirbal's "instrumentive" derivation (p. 170-171):
>
> "Now Dyirbal has a derivation -- which we can call 'instrumentive' --
> that places an underlying instrumental NP into derived O function,
> demoting the underlying O into dative case, leaving A as is, and
> marking the verb with the suffix -ma-l (which, like antipassive -Na-y,
> comes between verb root and final inflection).
>
> (65) yabu                   Numa-Ngu      balga-n            yugu-Ngu
>        mother+ABS(O) father-ERG(A) hit-NONFUT  stick-INST
>         father hit mother with a stick.
>
> Thus, from (65) is derived:
>
> (66) yugu       Numa-Ngu       balgal-ma-n                   yabu-gu
>         stick(O) father-ERG(A)  hit-INSTV-NONFUT  mother-DAT
>         father used a stick to hit mother"
>
> ("N" = [N])
>
> I don't think I've ever seen anything like this before.  Do any conlangs
> have this, or anything similar? Anybody know of any other natlangs
> like this?
>
> ===========================================
> Tom Wier <artabanos@...>
> "Cogito ergo sum, sed credo ergo ero."
> ===========================================
>