Reflexives as subjects?

[Miriam.Butt at UNI-KONSTANZ.DE]

Balthasar Bickel wrote:

> South Asian languages usually allow reflexives controlled by agents in
> the passive (especially in the 'ability' passive); see Lust et al.
> [eds.] "Lexical Anaphors and Pronouns in Selected South Asian Languages"
> Berlin: Mouton de Gruyter 2000. A close parallel to the German case
> would be the following Gujarati example (Mistry, p. 352):

I would maintain that the `ability-passive' in Hindi/Urdu is not a passive
but in fact a complex predicate formed with the light verb `go'.  There
are some indications that the instrumental agent in this case should be
analyzed as a subject.  See

    http://www.ling.uni-konstanz.de/pages/home/butt/index.html

for an extensive handout on

   Aspectual Complex Predicates, Passives and Disposition/Ability

which includes an examination of a dispositional semantics (English
example of the type of semantics involved: `My pet toad will eat flies.'
(taken from Lawler)) that this and another similar construction carry,
which are not usually associated with passives.

Some private discussion with Tara Mohanan, who has very carefully worked
out a list of subject tests for Hindi (Mohanan 1994), has confirmed my
initial idea that the instrumental agents are indeed subjects.

> Raaj-thi potaane  naa vakhaaNaayo
> R-INSTR  REFL:DAT not praised:PASS
> 'Raaj could not praise himself'
>
> This is yet another instance of what seems to be a subject reflexive.
> (The reflexive here is in the dative/accusative case, but this is a
> general possibility of passive subjects in Indo-Aryan.)

I don't know very much about Marathi, on the other hand, but I don't
see why the reflexive here has to be a subject.  Why not the
instrumental?  Judging from the gloss, which carries a kind of modal
meaning, this could also be a different kind of construction that
just looks like a run-of-the-mill passive at first glance.

In any case, I don't think this Gujarati example is a close parallel to
the constructions Frans was inquiring about:  Gujarati and Hindi/Urdu have
no expletives, which is precisely what makes those German examples
so hard to grasp.

Miriam

*********************************************************
Miriam Butt			Tel: +49-(0)7531 88 29 28
Universitaet Konstanz		Fax: +49-(0)7531 88 30 95
FG Sprachwissenschaft           miriam.butt at uni-konstanz.de
Fach D186
78457 Konstanz
Germany

  http://ling.sprachwiss.uni-konstanz.de/home/pages/butt/
*********************************************************

>
> Frans Plank wrote:
> >
> > I've once (in Studies in Lg 17, 1993) been puzzling over reflexives (if
> > this is what they are) in such passive constructions as these, permissible
> > for (many) speakers of German (and a few other languages):
> >
> > Wird sich taeglich gewaschen?
> > is.3SG REFL daily washed
> > 'Does one wash oneself daily?'
> >
> > Hier wird sich nur von mir die Haende gewaschen.
> > here is.3SG.SBJ REFL (3rd person) only by me the hands (ACC=NOM) washed
> > 'It's only myself who washes his hands here'
> >
> > I found them odd with respect to control:  only subjects are supposed to
> > control reflexives, while here the (semantic) controllers would seem to be
> > the agents (usually indefinite and omitted).  But then grammatical
> > relations aren't so obvious in such passives in the first place (as shown
> > by uncertainties in verb agreement and case marking).
> >
> > Maybe the reflexives themselves are the grammatical subjects of such
> > passives, instead of being morphosyntactically controlled by the elusive
> > dummy subject ES 'it'.
>