Re: Question about transitivity/intransitivity

Quoting "Thomas R. Wier" <trwier@...>:

> Quoting Andreas Johansson <andjo@...>:
>
> > Quoting "Thomas R. Wier" <trwier@...>:
> >
> > > Quoting Andreas Johansson <andjo@...>:
> > >
> > > > Quoting Rob Haden <magwich78@...>:
> > > >
> > > > > Thanks for your reply, Mathias.  The sentences you list above
> are
> > > > > examples of what I would call "English preposition-omitting
> > > > > ambiguity." Prepositions indicating oblique relationships are
> often
> > > > > omitted in casual speech in English, giving rise to sentences
> like
> > > > > "I give John the dog." While such sentences can be sorted out
> by
> > > > > fluent English-speakers via context, they may be difficult for
> > > > > others to sort through.
> > > >
> > > > Sure this has anything to do with dropped prepositions? "I give
> > > > John the dog" looks EXTREMELY much like the dative constructions
> > > > found in other Germanic languages.
> > >
> > > Depends on your theory of morphosyntax.  Mark Baker's theory of
> > > incorporation holds that N's are not the only lexical heads
> > > that may incorporate into verbs.  Prepositions and various kinds
> > > of null categories may also incorporate, deriving applicative,
> > > causative, antipassive, passive, and generally any valence
> > > changing construction.   In the case of the dative-shift
> > > construction that you mention, Baker claims that there is a
> > > null preposition that governs the NP "John" at D-structure,
> > > which incorporates into the verb, forcing "John" to raise to
> > > get abstract case.  (Baker really likes these null categories;
> > > I myself am rather allergic to them, but that's his argument.)
> > > In support of this argument, Baker provides evidence from some
> > > Inuit language (West Greenlandic, IIRC) which only allows this
> > > kind of dative shift construction when an overt morpheme is
> > > present to show the argument structure has changed.  I don't
> > > think he would agree that the preposition is "dropped", however;
> > > it's just never pronounced at all.
> >
> > Either I'm misunderstanding you, or this is a complex way of saying
> that
> > English has a null realization of the Germanic dative marker.
> > This is assuming that there's no deep difference between a null case
> ending
> > and a null preposition, and that dropping a null element isn't any
> > different from keeping it in.
>
> Well, in most brands of Chomskyan syntax, case is an abstract
> property or force. It is not so much simply a requirement of the
> verb as the result of specific structural relationships holding
> between a head and its specifier or complements. Prepositions,
> verbs, nouns, etc. *are* those heads, specifiers or complements.
> You might think of case as being the skeletal relation in a
> tree-structure, and nouns, prepositions, etc. are the flesh of
> the sentence.
You've lost me (and jolly surprised we are too!). I mean, precisely _because_
case is an abstract property, it ought to be synchronically immaterial if we
analyze "John" as being a zero proposition plus noun or noun plus zero case
ending - either fleshes out the dative bone. (More economically, one'd argue
that WO itself does it.) Unless we take case = that which is indicated by case
endings, I can't see the difference between a null preposition indicative
dativity and a null case ending doing the same.


> So, for Chomsky (and adherents of his theory like Baker), there
> is a very big difference between case-assigners, like prepositions,
> and the cases that they assign.
I'm no Chomskyite (I don't know enough about it to accept or reject it), but
that seems very sensible to me. Problem is, I'd see it as supporting my
interpretation.

> > But I was thinking diachronically; I'd be very surprised to learn
> > that constructions like _I give John the dog_ are reformed from
> > things like _I give the dog to John_ rather than cognate to things
> > like _Ich gebe ihm den Hund_.
>
> Remember that, for Chomsky and his ilk, diachronic facts are
> historical curiosities of no fundamental importance.  He is
> only interested in the internal workings of a speaker's grammar
> as a window into the human mind.
Well, bad for them.

Then I invoke Ockham's razor. Its more straightforward to assume that SVxO
syntax by itself indicates x to be a beneficirary than that "to" has an
optional zero allomorph, which moves the prepositional phrase to between verb
and direct object. We're not seeing any null endings nor null prepositions,
and refuse to believe in them until they show up!

I'm confidently waiting to be shot down ...

                                                         Andreas

Reply