Date sent:      	Fri, 14 Jul 2000 13:31:00 +0100
Subject:        	RE: subjects and objects
From:           	A Rosta <[log in to unmask]>
To:             	[log in to unmask]
Send reply to:  	[log in to unmask]

> Jasp:
> > WGers,
> >
> > What are the syntactic properties of subjects and objects? And has
> > suggested some already, in connection with adjectives, but I
> > wonder how many others we can come up with.
> 
> A question I've been pondering this year is similar: what makes it
> necessary to recognize the category Subject, and what are the
> distinctive properties of subjecthood? I can't give you an answer
> yet, though... But anyway, my notion of subjecthood is different
> from WG's, so the answer might't be relevant.
> 
> Object seems to me a slightly tougher one, because the case for
> recognizing the category Object is far weaker than for Subject.
> If "object" just means "(nonpredicative) nominal complement",
> then I would argue that there is no need to recognize such a
> category. But there is a small amount of evidence for an
> 'object position', just after the verb, just as the subject
> position is just before the verb.
> 
> I think "subject/object" is a kind of folk-dichotomy/taxonomy,
> like Animal/Vegetable/Mineral -- i.e. more folk-linguistic
> notions than theoretical constructs.

You may be right, but then if 'folk' (ie language users) think of 
subjects and objects as different (from each other, as well as from 
other kinds of dependent), then you'd expect them to have different 
properties all the same.

> 
> > I have the following already (you may have something to say about
> > some of
> > these).
> >
> > Properties of subjects in (some versions of) English:
> >
> > a. Pronoun subjects have a marked form (_I/*me_ etc.).
> 
> Except for "between you and I" dialects, in quasi-standard English
> the rule is that _I_ must be a subject, but not that _me_ can't
> be a subject. For subjects of things other than finite verbs there
> is huge idiolectal variation in I/me choice.
> 
> > b. Subjects precede their parents.
> 
> IMO, without exception, but in WG, subjects of auxiliary verbs can
> follow their parent.
> 
> > c. The subject is obligatory in a relative clause (_The burglar *(who)
> > stole my mink_).
> 
> In Std E in kosher relatives.
> 
> > d. The subject of the main verb is obligatorily construed as the
> > subject
> > of a (preceding) subjectless adverbial (_Shouting wildly, they
> > charged
> > into the breach_).
> 
> True for some idiolects (including mine) but not for many or most
> (including Dick's, IIRC).
> 
> > e. Subjects cannot be iterated.
> 
> In my view, "cannot be iterated" = "is not an adjunct".
> 
I think so too. I suppose this property can only help identify 
subjects: it cannot be iterated, so it's not an adjunct; ot precedes 
its head, so it must be a subject then.

> > f. Sharers (xcomps) (almost always) share their subject with their
> > parent...
> 
> Too many exceptions to be a worthwhile generalization.

what do you mean? Except in a very few cases (see below) the 
shared dependent is the subject of the sharer, so if a word has two 
parents you can be sure it is the subject (or object, see below) of 
the lowest of its parents.
> 

> >  ...and may share their parent's subject.
> >
> > g. err...
> >
> > Properties of objects in (some versions of) English (all robbed from
> > Hudson 1992):
> >
> > a. The usual object of a verb shows up as the subject of the
> > corresponding
> > passive.
> 
> I'm not sure how useful this is. Not all subjects of passives are
> promoted objects, and not all of what EWG takes to be objects can
> promote to passive subject.
> 
Indeed
> > b. Objects follow their parents...
> >  ...immediately,
> >   except under 'heavy NP shift'.
> >   and (for non-pronouns) where the verb also has a particle.
> >  ...though they do permit extraction.
> >
> > c. Objects are selected by the parent (DINE/EAT/DEVOUR) (and
> > their
> > semantic role is fixed by the parent).
> 
> In other words, objects are complements.
> 
Yes.
> > d. Objects can't be iterated either.
> >
> > e. Many idioms involve verb/object pairs (_break wind_, _kick the
> > bucket_,
> > etc, etc, etc.)
> 
> There is a kind of hierarchy of susceptibility to idiomaticity,
> with direct objects in second place after particles, but I'm not sure what
> further conclusions to draw from this.
> 
> > f. Sharers (xcomps) may share their parent's object (in a very few
> > cases,
> > sharers share their object with their parent).
> 
> ??? Examples please.
> 
Errr... oh yes, there are those like this that we discussed (well, I 
just read about them actually, you all discussed them!) earlier.

The car needs cleaning [object of sharer, subject of its parent 
shared]
I want the car cleaning [object of sharer, object of its parent shared]

Jasp
>  --And.
> 
> 
> > That's it.
> >
> > Jasp
> >
> >  ---------------------------
> >
> > Rijksuniversiteit Groningen
> > Oude Kijk in 't Jatstraat 26
> > Postbus 716
> > 9700 AS Groningen
> > The Netherlands
> > +50 363 5843
> > fax:
> > 




%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%