Re: Evolution of Applicatives

From:    John Cowan <cowan@...>
> Thomas R. Wier scripsit:
>
> > Ah, well, we *were* talking about linguistics on a rather formal
> > level, so I just assumed...
>
> Nah, it's just that the polysemy of "topic" escaped me for a moment.
Actually, it's not clear to me whether it's true polysemy, or that
the two uses are distinguished only by two separate thematic roles.

> > but I still can't get any reading other
> > with Tuesday as the patient, and not as the day.  I'm pretty sure
> > that most English speakers will agree with me in this respect.
>
> You astound me.  If I wrote a poem called "Ode to Tuesday", you really
> can't describe this as "writing on Tuesday", in exactly the same way
> that if I wrote a paper about the _Critique of Pure Reason_, you could
> describe this as "writing on Kant"?
You know, now that I think about it I think I can get the topic reading.
I still can't get the temporal reading.  Yesterday I asked around the
department and sent out a questionnaire to our departmental mailing list,
and all of them, of the 10 or so responses without exception, found the
raising-from-temporal-adjunct reading impossible. (I gave them exactly the
same info that I gave y'all, so they were operating under the same
conditions.)

> > Thus, the fact that someone or something named 'Tuesday' might exist
> > in the world does not change the fact that the day of the week by that
> > name cannot be passivized for any reason from an adjunct.
>
> Fair enough, though I still deny "the fact", given the polysemy of "on".
See above. If they are in fact distinguished entirely or mostly
by two separate thematic roles, that would explain why some people
can get any reading at all out of the passivized form, since passives
in English seek out certain kinds of arguments.

> > I have to disagree... what you say here is tantamount to denying
> > formal structure any role in providing grammaticality judgements.
> > (I respect you, John, but you must admit that this is a rather
> > extreme functionalist position you're taking.)
>
> Well, my real position is that the concept "grammaticality judgment"
> is simply a category mistake.  Grammaticality is defined within a
> given formal structure, and within that structure it is an absolute
> notion, as much so as validity in formal logic: judgment just doesn't
> enter into it.  What we actually have in all these Chomskyan self-tests
> are acceptability judgments, which are quite a different thing.
So... then, as an empirical science, how does one go about
getting discovering grammaticality?  I'm certainly not one of those
people who discount statistical surveys, but one has to be very
careful using them because they in fact aren't referring to precisely
the same entity (the language as conceived by individuals speakers,
and the abstraction of 'Language' of a community of speakers).

In other words, it's not clear to me not only how one would identify
what structures are absolutely grammatical, but also not clear how
to falsify the claim of absolute grammaticality itself.  We have no
way right now of looking into speaker's minds to "see" the grammar,
so we have to make do with judgements about that grammar.

(There are also too many clear examples of gradient grammaticality
for me to accept the view easily.)

> > Again, this isn't relevant. *"The NEA was given money to by
> > liberal activists" is grossly ungrammatical, and that's the analogous
> > structure you're invoking.
>
> Unfortunately, the more I think about this one, the more acceptable it
> becomes.
 [...]

> Sometimes I wonder.   See the Harvard Law of Animal Behavior:
>
>        In a well-controlled experimental situation, a well-trained
>        experimental animal will do whatever it damned well wants to.
Well, this is a general problem in linguistics:  how do we distinguish
between judgements based on the general behavior of speakers, and items
that they have been trained to use but would never have used otherwise?
There have been a number of experiments showing that if you bombard
speakers with ungrammatical sentences time after time, they will end up
saying what was unquestionably ungrammatical to them before ended up
being grammatical.  Clearly, though, we are not (or should not be)
interested in the latter set of judgements except in terms of what they
show about their nonlinguistic cognitive origins -- *unless*, that is,
they start producing analogous structures in their normal speech, in
which case it is a linguistic situation.

 =========================================================================
Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637