I  find Bill's last post interesting about SFG (systemic functional
grammar).  Remember that this discussion is about the following observation:

> My point, I guess, is that we run into trouble if we think of=20
>complements as dictated by the nature of the verb (like indirect object=20
>or beneficiary, which only certain verbs will support) and as occurring=20
>in the predicate, precisely because that doesn't allow us to acknowledge=20
>the flexibility available to us as we construct meaningful discourse.


Bill's example from Molly Ivins can be accounted for in other ways.

>We need at least $10 billion in new taxes to fix this without harming
>the schools. The alternative is a $2 billion fix patch on the old system
>that will further decay the schools.
>
>Now, 'decay' isn't really supposed to be a transitive verb, but she's
>using it as one, and it flows rather naturally (creating a rather nice
>parallel structure).
>
Notice the specially pleading here:  "it flows rather naturally."
There is a large body of research on
causatives (in other words, indicating that X caused Y).  In fact little
kids, have difficulty in figuring it out.
For example, such statements are not at all unusual in first language
learners of English.

            1)  Mommy, I felled the milk.  (I caused the milk to fall.)


The example from Ivins  is along this line.  If little kids want to
create causative constructions, then we should not
be surprised that adults do the same.  However, this example does not
show that discourse "dictated" this particular
construction.

I have no idea what this means without example.

>If the argument structure of a clause is determined by
>its verb, it's hard to account for anomalous argument structures
>(shouldn't it cause a problem? A kind of thetastrophe?).
>


One has to be very careful about overgeneralizing from very particular
examples.

>Halliday has been very careful to link his
>distinctions among process types to specific characteristics of
>structure. For example, what in traditional grammar are called direct
>objects of a number of cognition and locution verbs behave differently
>than the direct objects of physical action verbs:
>
>1.a        I ate a sandwich.
>   b       ?A sandwich, I ate.
>
>2.a        I said, "Hello."
>   b       "Hello," I said.
>
>While 1b is certainly possible, it's far more marked than is 2b, which
>is actually a rather common pattern in narrative dialog. This is one of
>the features upon which Halliday bases his distinction between "material
>processes" and "mental processes" - the latter are realized by V X
>constructions in which X is a frontable complement which is usually
>clausal; the former  are realized by V X constructions in which X is a
>non-frontable complement which usually isn't clausal.
>
I am bothered by phrases such as the following "rather common pattern."
There are some very interesting constraints
on when (1b) is possible. (See the discussion of Birner, Ward, and
Huddelston in The Cambridge Grammar of English).

Halliday's distinction as reported by Bill is not quite accurate.  The
complement of the "mental process" verbs is restricted to the
actual speech or thought.  The preposing the reported speech or thought
is as decidedly odd  as 1b.

     3a   Throckmorton said that he is coming tomorrow.
       b)  ?That he is coming tomorrow, Throckmorton said.

This is the claim supporters of SFG make.

> The system is, however, useful to many analysts, and it's
>internally consistent.
>
Several years ago Jim Kenkel and I looked at this claim and found that
SFG is not internally consistent.  Send me a note off-line and I can send an
electronic version of it.

Bob Yates, Central Missouri State University

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/