LINGUIST List 11.269

Tue Feb 8 2000

Disc: Newmeyer: Language Form & Language Function

Editor for this issue: Anthony Rodrigues Aristar <aristarlinguistlist.org>

Directory

1. Laurence Horn, Re: 11.191, Disc: Newmeyer: Language Form & Language Function
2. Phil Gaines, Re: 11.191, Disc: Newmeyer: Language Form & Language Function
3. Martha McGinnis, Re: 11.191, Disc: Newmeyer: Language Form & Language Function

Message 1: Re: 11.191, Disc: Newmeyer: Language Form & Language Function

In LINGUIST 11.191. Sun Jan 30 2000,
Subject: 11.191, Disc: Newmeyer: Language Form & Language Function,
Thomas Egan <Thomas.Eganluh.hihm.no> writes about
"The rat that the cat that the dog bit chased ran":
>
>As far as I am concerned the important question is not whether it is
>theoretically possible to contextualise the sentence but whether in
>fact it, and sentences like it, ever are encountered in spontaneous
>speech. Nor would one single reported occurrence be enough to warrant
>our labelling it as "English" or "grammatical". The fact that
>linguistics students can understand it is neither here nor there.
>This merely shows that it bears sufficient resemblance to grammatical
>English constructions to be susceptible to analogous interpretation.
>"to Cat the on mat sat the" is too far removed from grammatical
>constructions to afford us a starting point for analogy. A better
>example would be a sentence like "some people would have went mad"
>(from the BNC). Regardless of whether or not this sentence is
>grammatical in the dialect of the speaker in question, there is no
>doubt that it could be understood and explained by linguistics
>students whose idiolects do not contain "went" as past participle. But
>would that be sufficient to warrant our labelling it as grammatical in
>Standard English? In my view "some people would have went" is closer
>to grammatical English than "The rat that the cat that the dog bit
>chased ran". Indeed I suspect it would require the authority of a
>lecturer in lingusitics to convince students that the latter is
>grammatical (as opposed to understandable)!
>

Here's one real example in honor of the recently completed National
Football League season:

	"It's ironic that I'm here,
	 where the man [the trophy [I won =D8] is named after =D8] coached."

- Attested quote from Tony DeGrate, winner of the Vince Lombardi trophy as
a college senior at the U. of Texas, upon being cut from the Green Bay
Packers professional football team, which had won five championships under
their legendary coach Vince Lombardi.

Here are two more constructed examples fashioned from the same mold:

	The man [(that) the woman [I love =D8] is married to =D8] is insanely
jealous.

	The difficulty [(that) someone [I know =D8] is having =D8 with syntax]
is...

In each case the inner-most embedding is anchored to the speech situation;
replacing the first-person subject by a third-person one would render the
sentences harder to process, as would increasing the syntactic complexity
of that layer (even by inserting a complementizer THAT before the I).

Larry Horn

Message 2: Re: 11.191, Disc: Newmeyer: Language Form & Language Function

Tom Egan wrote (LINGUIST 11.91):

> As far as I am concerned the important question is not whether it is
> theoretically possible to contextualise the sentence but whether in
> fact it, and sentences like it, ever are encountered in spontaneous
> speech. Nor would one single reported occurrence be enough to warrant
> our labelling it as "English" or "grammatical". The fact that
> linguistics students can understand it is neither here nor there.
> This merely shows that it bears sufficient resemblance to grammatical
> English constructions to be susceptible to analogous interpretation.
> "to Cat the on mat sat the" is too far removed from grammatical
> constructions to afford us a starting point for analogy. A better
> example would be a sentence like "some people would have went mad"
> (from the BNC). Regardless of whether or not this sentence is
> grammatical in the dialect of the speaker in question, there is no
> doubt that it could be understood and explained by linguistics
> students whose idiolects do not contain "went" as past participle. But
> would that be sufficient to warrant our labelling it as grammatical in
> Standard English? In my view "some people would have went" is closer
> to grammatical English than "The rat that the cat that the dog bit
> chased ran". Indeed I suspect it would require the authority of a
> lecturer in lingusitics to convince students that the latter is
> grammatical (as opposed to understandable)!

 From the standpoint of generative grammar, "Some people would have went mad"
is grammatical regardless of the dialect. This particular dialectal use of
"went" as the past participle does not come into the discussion of whether
the phrase structure is possible according to the parameters of English.
Same thing for "The rat..." Not the same thing for "Cat the..."

When Tom says,

>In my view "some people would have went" is closer
> to grammatical English than "The rat that the cat that the dog bit
> chased ran". Indeed I suspect it would require the authority of a
> lecturer in lingusitics to convince students that the latter is
> grammatical (as opposed to understandable)!

he seems to be not discriminating carefully in his use of "grammatical".
For the generativist, grammaticality has nothing to do with
understandability. If I may drag out the well-worn but still useful chess
analogy: Most average chess players would not understand a chess game that
Gary Kasparov might strategize, yet if he conforms to the rules it is a
grammatical game. The fact that most English speakers would not understand
"The rat..." is not an interesting point from the standpoint of the theory.
This segues me into Tom's reference to "spontaneous speech". What is that?
Does it have to be something that a speaker does not think about in advance?
Does it have to have been already said somewhere by someone? Does it have
to be spoken? Right now, I'm re-reading Ulysses, wherein Joyce famously
does delightful acrobatics with grammar. One of his games in a long
narrative section is to separate the verb from the subject by as much as 10
lines of text. Two or three careful re-readings of such sentences are
necessary to parse them. Now, if this is not "spontaneous" speech, then I
would say that such a notion is on a continuum that is not particularly
useful.

Finally, if you're going to argue that understandability is important in
determining "grammaticality" (in Tom's quasi-prescriptivist sense) then why
isn't "Cat the on mat sat the" grammatical? My point is that once we get
into this kind of discussion, we're not talking about principles and
parameters anymore, and the conversation goes nowhere. The question, to me,
is: Do generativists have good cause to talk about an underlying notion of
grammaticality regardless of performance factors? If not, the argument
against that notion should be marshalled on grounds other than such things
as gradations of understandability.

Phil Gaines

Message 3: Re: 11.191, Disc: Newmeyer: Language Form & Language Function

Marc Hamannn wrote (LINGUIST 11.191):

>Kevin R. Gregg wrote (LINGUIST 11.169):
>
> >the whole community of
> >English-speakers could with one voice reject 'The rat that etc.' as
> >unacceptable, without--simply in virtue of that unanimity-- impugning in
> >the least the veridicality of the theory that marks it grammatical.
>
>Essentially you have just said that a theory of grammar:
>
>a) cannot be tested by empirical means
>b) is completely arbitrary to the whims of the grammar theorist
>c) or else, only exists as a Platonic Ideal

Not true. Gregg's remark just reminds us that we can't decide a priori
what constitutes evidence for a theory of grammar. In general,
linguists base linguistic theories on judgements of acceptability and
interpretation, but other kinds of evidence also seem to be relevant,
like language acquisition, historical changes, patterns of deficits in
aphasia, neural activity during language processing, and so forth.

If linguists can't be sure that _only_ acceptability judgements are
relevant to linguistic theory, we also can't be sure that _all_
acceptability judgements are. Chomsky's claim about center-embedded
sentences (e.g. in Aspects) is that they're unacceptable, not because
they're ungrammatical, but because they're difficult to process in
real time. As you know, that was just a promissory note until someone
could come up with a coherent theory of sentence processing.

Since then, people like Ted Gibson have offered a unified account of
various kinds of garden-path and other difficult-(or
impossible-)to-process sentences, including center-embedded
sentences. His approach supposes that the unacceptability of some
sentences can be reduced to independent constraints on working memory
- unlike other cases of unacceptability, for which we seem to need an
autonomous linguistic theory. These issues remain open, of course;
for example, Colin Phillips has argued that the parser _is_ the
grammar. Nothing has been finally resolved, including the empirical
scope of the theory, but that's just normal science.

_________________________________________________________
Dr. Martha McGinnis
Assistant Professor of Linguistics, University of Calgary
2500 University Dr. NW, Calgary, AB T2N 1N4 CANADA
phone: (403) 220-6119, fax: (403) 282-3880
http://www.acs.ucalgary.ca/~mcginnis/