LINGUIST List 12.1975

Mon Aug 6 2001

Disc: Semantics and Pragmatics of to-contraction

Editor for this issue: Karen Milligan <karenlinguistlist.org>

Directory

1. rtroike, http://www.linguistlist.org/issues/12/12-1971.html
2. Mills, Carl (MILLSCR), RE: 12.1964, Disc: New: Semantics and Pragmatics of to-contractio n
3. Dan Everett, reading carefully: response to to-contraction
4. Dick Hudson, Re: 12.1971, Disc: Semantics and Pragmatics of to-contraction

Message 1: http://www.linguistlist.org/issues/12/12-1971.html

There may be some US/British differences for some respondents, but for me, 
as a Texan, 

	1a. I gotta go. (= I must go.)

is derivable from an uncontracted:

	1b. I have got to go. (or, alternatively, I have to go/hafta go.)

	2a. I got to go. (= I was allowed/able to go.)

is the past tense of 

	2b. I get to go. 

which has a possible present perfect (though rare)

	2c. I have ('ve) gotten to go. 

As these are different lexical verbs (the first is arguably a small "v"),
it is entirely reasonable to suppose that, as suggested, they take 
different complements and thus have different syntactic behaviors. The 
apparent puzzle may be no more than between different interpretations of 
"They decided on the boat."

	Rudy Troike
 

Message 2: RE: 12.1964, Disc: New: Semantics and Pragmatics of to-contractio n

With apologies to Dan and the discussants who have contributed so much
to this topic, I must say that I find all or nearly all the starred
examples fully acceptable. In fact, I have said "I getta go" on more
than one occasion. And I have heard occurrences of starred 'wanna'
numerous times. OK, so, as has been remarked, there is quite a lot of
idiolectal and dialectal variation in this area. This then raises an
important question: Should important points of syntactic theory be
decided on the basis of such unclear cases? Last week at LACUS,
Stefan Gries gave a good paper on acceptability judgments as
linguistic evidence. Among his finds, if I am not twisting his
conclusions, was the observation that acceptability judgments of naive
native speakers tend to mirror facts discoverable through corpus
studies. But the judgments of trained linguists tend to vary among
themselves and to be at variance with corpus evidence.

Carl Mills
Linguistics Program
Director of Undergraduate Studies
Department of English and Comparative Literature
University of Cincinnati

k

Lofti didn't read my posting carefully. Notice that in my original
posting I did not say that Chomsky's proposal on to-contraction was
designed to handle the facts I raised, nor that it had a mistaken
account of them. The claim is pretty clear and perhaps even innocuous:
the Chomskyan account does not extend to the cases I raised. Two
questions arise in this connection. First, is there a general account
of to-contraction not based on traces, since the trace-based account
does not extend to what look to be quite similar cases? The next
question is how this NEW account might itself extend to the
trace-blocking effects Chomsky mentions (and the original proposal by
Chomsky uses data originally presented by George Lakoff in a
presentation at the University of Michigan around 1969. Some might
even say that trace-theory was inspired by the paper by Lakoff - as a
reaction against it).

I proposed a solution. There are others imaginable. But saying that
Chomsky didn't intend to account for these new cases is hardly an
answer. It merely reinforces the novelty of the examples.

Let me be clear that I have no stake in the outcome. In fact, the best
account so far may be the alternative based on frequency just
suggested, based on Joan Bybee's model. Whatever the answer is, it is
not going to be based on traces or tree-structure (unless the latter
is just made isomorphic with the semantics, which removes its very
motivation).

- Dan Everett

Message 4: Re: 12.1971, Disc: Semantics and Pragmatics of to-contraction

Dan Everett:
>On the other hand, there are various other proposals which do not involve
>movement, e.g. the proposal made by Zwicky and Pullum, I believe, that
>'wanna' is simply a newly emerging modal, i.e. a single word. This
>suggestion might apply to all such cases, in fact, as a few on the FUNKNET
>discussion list have suggested.
## As I'm sure others have pointed out before, this proposal doesn't look
promising because 
"wanna" has exactly the same distribution as "want to":
A. It's bad after 3rd singular subjects which require "wants": at least for
me, *"He wanna go" is out, whereas "I/you/we/they wanna go" are all ok. 
B. "Do you wanna go?" is fine, just like "Do you want to go?", whereas no
modal verb can occur with "do" (*Do you can go?). 
Maybe a modal verb analysis lies ahead, but it's still a long way ahead.
We're still in the stage where "wanna" is an alternative to "want to". 

Dick

Richard (= Dick) Hudson

Phonetics and Linguistics, University College London, 
Gower Street, London WC1E 6BT.
+44(0)20 7679 3152; fax +44(0)20 7383 4108;
http://www.phon.ucl.ac.uk/home/dick/home.htm