[Fair warning: More theory-mongering. Potentially amusing for linguists
but no classroom application]

Herb,

I think speakers *do* use syntactic structure to mark assertional
status; the process just operates in complex way. We can, for example,
consider the finite operator in declaratives as a necessary but not
sufficient condition for assertion, and the finite operator is a
grammatical element. Likewise, choice of a subordinator is both a
grammatical and a semantic phenomenon, and it affects assertional status
as well. 

There is, of course, an issue of theoretical stance here; I'm from a
tradition that does not want to draw strict boundaries between syntax
and semantics, and my analyses are biased (as are all analyses). But
considering the "I think S" construction only as a simple matrix clause,
with the hedging feature operating only at a semantic level, does not, I
believe, account nicely for the odd postpositional usage of the
expression or, I *suspect*, for the frequency with which primary
sentential stress falls in the nominal clause. There is an unexamined
assumption in much grammatical work that an unambiguous sentence can
have only one syntactic structure, hence the analyst is expected to
choose whether "I think that X is Y" is underlyingly "I think S" or "(I
think) X is Y," with the former winning out on the basis of analogy with
other matrix constructions. I would question that assumption at this
point.  

Bill

-----Original Message-----
From: Assembly for the Teaching of English Grammar
[mailto:[log in to unmask]] On Behalf Of Stahlke, Herbert F.W.
Sent: Wednesday, October 12, 2005 10:35 AM
To: [log in to unmask]
Subject: Re: A logical thought

Bill,

I'd probably quibble with a couple my examples too, but I think the
point stands.  However, you point to another important observation, that
the assertional or presuppositional character of a clause isn't
determined by syntactic structure, although syntax can certainly enhance
it.  In 

I think that taxes are too low.

"that taxes are too low" is clearly the noun clause DO of "think".  That
is, it is grammatically an embedded, subordinate clause.  However, the
pragmatic function of the higher clause "I think S" is not to assert
what I am doing but to hedge the assertion that taxes are too low.  In
this sense it is sort of adverbial, although I prefer that term for
things that modify verbs, sentences, or adjectives.  Rather, it's
function is at discourse level.  But the sentence is a case in which the
assertion is in a subordinate clause and the main clause modifies the
assertion.

Herb


Subject: Re: A logical thought

Herb,

I agree, although I would quibble (trivially, in the final analysis)
with a couple of the sentences -- I think "No it doesn't" might well
work as a contradiction for the extraposed clause example you give, and
I can't quite get "wonder that" to work right in my head. Of course,
that might just be because I'm short on coffee right now.

Not only is there probably a scale, but we might consider particular
constructions as potentially occupying more than one spot on it.
Depending on what you choose to contradict, it could be argued that "I
think" constitutes a matrix clause or that it is acting more like an
adverbial element:

[I think that] taxes are too low.	No they aren't!
I think [that taxes are too low].	No you don't!

I decided that taxes are too low.	?No they aren't!
I decided that taxes are too low	No you didn't! 

The kind of dual contradictablity of "I think" constructions may be tied
to their status as projecting verbs, using Halliday's term -- these are,
for example, the ones that allow postposition in direct quotes and, in
some cases, indirect ones:

Taxes are too low, I think.

Taxes were too low, he thought, and therefore he took action.

There's something interesting in the fact that the postposed ones are
always that-less, but I can't quite figure out where I want to go with
that.

It may be that we assign double structures to these constructions, and
then focus on one or another of them in context. Contradicting the
putative matrix foregrounds its interpretation *as* a matrix, while
contradicting the putative complement clause establishes it as the main
assertion. In speech, of course, intonation could similarly tilt the
structure one way or another.

Some other constructions, such as "The fact is, X" seem to act even more
strongly as if what looks like a matrix clause is actually a complex
adverbial element (in that example, roughly like "Actually..."; the
evidence includes the ability of speakers to leave off the "the" and to
double the copula without anyone other than intrusive linguists
noticing).

Bill Spruiell
Dept. of English
Central Michigan University

-----Original Message-----
From: Assembly for the Teaching of English Grammar
[mailto:[log in to unmask]] On Behalf Of Stahlke, Herbert F.W.
Sent: Tuesday, October 11, 2005 3:24 PM
To: [log in to unmask]
Subject: Re: A logical thought

Bill,

That's an interesting proposal.  How about considering the test to
reveal a scale rather than a dichotomy.  Extraposed clauses like 

It appears that Bush stole the Florida vote.

Allow "No, he didn't", and "No, it doesn't" would be odd.  In

I think Bush stole the Florida vote.

"No, he didn't" works well and "No, you don't" doesn't deny the
complement at all.  Of course, the head verb and the conjunction factor
in too, as in 

I wonder if Bush stole the Florida vote.
I wonder that Bush stole the Florida vote.

With "if", no presupposition is made.  With "that", the presupposition
is that he stole it, and "No, he didn't" doesn't work as well.  
Then with

The fact that Bush stole the Florida vote meant that the 2004 election
would be at risk too.

"No, he didn't" doesn't work at all.  I suspect that this continuum
could be extended and fleshed out further, both with more structures and
perhaps with more tests.

Herb


Herb,

It occurs to me that not only do activities like the ones you described
perform the function of uniting grammar (broadly construed) with wider
issues of language awareness, they also provide an additional way to get
at a kind of heuristic students can use: 

"If it's a statement that you've punctuated as a sentence, and you can't
possibly contradict it, it's a fragment."

I'm fairly sure this won't yield false negatives, but I'm still trying
to sort out whether it can yield false positives. From informally
polling native English-speaking students, I've noticed some potential
variation -- for some reason, fewer people object to contradicting
material in a because-clause than in a when-clause:

Bjorn was in the kitchen when Brunnhild murdered Bjarki.	*No she
didn't!

Bjorn was rather put out because Brunnhild murdered Bjarki	?No she
didn't

That second one doesn't sound very good to me, but a number of my
students were not bothered by it. However, moving the subordinate clause
to the beginning caused everyone to reject the contradiction. Something
is going on with end-rheme, I think, but I haven't dived into the
research on this at all (yet).

So, I'm not sure the proposed heuristic would enable students to find
*all* fragments (and it does not give students any way of thinking about
intended fragments), but it should work for a large number of cases.

Bill Spruiell
Dept. of English
Central Michigan 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/

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/

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/

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/

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/