from a review...

[Ash Asudeh]

Dear George and other list members,

With respect to George's message, the relevant part of which is given
below:

If what Chomsky is saying is that the set of convergent derivations should
not be identified with the set of grammatical sentences, because there are
extra-grammatical factors that influence the acceptance of sentences as
grammatical by native speakers (which is what your message seems to imply;
I don't have the MP book handy to look up the footnote), then I submit
that this is a confusion between grammaticality and acceptability, which I
find surprising on C's part.

Degrees of grammaticality, as Keller (2001) rather carefully argues, are
nevertheless claims about GRAMMATICALITY, not acceptability.

If it's not a confusion between grammaticality and acceptability, then I
submit that the theory, sorry program, he's adopting is problematic,
because it generates convergent derivations that are ungrammatical and
considers this unproblematic.

I also reject the contention that the "set of grammatical sentences" is
not a well-defined class -- you just defined it. But I do accept the
contention of Pullum and Scholz, building on work in Model Theoretic
Syntax (Blackburn, Rogers, and many others), that thinking about grammars
as generative is a) wrong, and b) leads to confusion about the set of
grammatical sentences and other such things.

However, all this aside, I think what we've seen in Montalbetti's review
is a somewhat sloppy reading of Chomsky, or at least a somewhat sloppy
presentation of C. If Chomsksy says what you claim he says, then it falls
well short of saying "there's no such thing as grammaticality", which is
what M seemed to be saying.

I do not think that the world is ready for the movement of Linguistic
Nihilism. I certainly would vote down any grant proposals (once I hold
such authority, if ever) that seemed to imply there's no such thing as
grammaticality, unless they contained a theory of acceptability.

Yours faithfully,
Ash

On Thu, 11 Apr 2002, G Tsoulas wrote:

>
> It should perhaps be pointed out that Chomsky actually said that it
> would be pointless to identify the set of sentences with convergent
> derivations with the set of `grammatical' or `well formed' sentences,
> simply because the latter are not well defined classes (see for example
> the discussion around the notion "degrees of grammaticalness" in Aspects)
> . Now, does anyone really disagree with this?