Re: THEORY: transitivity

From:    "J. 'Mach' Wust" <j_mach_wust@...>
> > Paul Hopper and Sandra Thompson wrote an excellent article in the early
> > 80s (Language, Vol. 56, no. 2) entitled "Transitivity in Grammar
> > and Discourse". Therein they list 10 (!) different criteria that
> > languages use to encode transitivity, and they show that transitivity
> > is really more of a cline than a discrete proposition.
[...]
> > So, if we've decided to lump things as transitive or intransitive,
> > I would say 'folgen' is transitive.
>
> So you prefer underlying structure to morphology, semantic transitivity to
> overtly encoded transitivity?
The point is that it is not clear what a 'transitive' construction
really is outside of such criteria.  The fact that a verb has two arguments
does not automatically imply that one is a subject and one an object,
since in some languages various kinds of tests (passivization, e.g.)
show the second one to be some kind of oblique.  Georgian has a whole
class of verbs which are syntactically and morphologically intransitive,
but they obligatorily take an oblique dative-marked argument (the
verb for 'to hit' it like this).

> I like better sticking to the surface,
> because otherwise I'm getting the feeling that the analysis of all those
> funny forms and their functions becomes very pointless and that I'd better
> study Logics than Linguistics.
The problem is the surface can be deceiving.  How would you handle
so-called pro-drop languages? In a language like Georgian, one need not
have any overt noun phrase at all, ever. This, despite the fact that
there are no overt markers of transitivity at all.  A surface-based
analysis would say these are zero-place predicates like weather
verbs.  The only reason we know they are not is that the mapping to
semantics is different. One simply cannot have a wholly syntactocentric
view in linguistics; it becomes incoherent very quickly.

Anyways, as a science that presumes to be both descriptive and explanatory,
you cannot posit theoretical phenomena (e.g., transitivity) on the basis
of other theoretical phenomena (e.g., phrase structure) until you've
actually grounded that theoretical phenomenon in something unaxiomatic,
something descriptive like such criteria.

 =========================================================================
Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637

Reply