Re: USAGE: Garden paths

Matt Pearson wrote:

>Hmph.  Parasitic gaps are generally characterized as grammatical, but it's
true that
>they're often rather marginal.
Nowadays they are.  In the older literature they were often called
ungrammatical.  (At least that is how it was taught to me, I haven't actually
read the older literature on the topic.  I'm trying to find my notes on the
topic, but can't seem to.)

  For those of you who are unfamiliar with this term, a
>"parasitic gap" is a gap in an embedded sentence which is licensed by a
gap in
a main
>sentence.
Right.  So we have a single constituent binding two gaps.  Under a movement
based theory of language, it's hard to see how that is possible.  I remember
the structural configurations that are supposed to hold in parasitic gaps, but
I cannot recall ever being given a reasonable explanation of why they are
possible. I guess there would have to be a silent or erased element somewhere,
but I personally find that kind of reasoning sloppy and hard to prove.

As I mentioned before, when this phenomenon was taught to me (three or four
months ago) the prof expressed the opinion that they were ungrammatical, but
acceptable.  I presume the acceptibility is due to the fact that they are
completely unambiguous.  That is, since English generally does not allow
arguments to drop out of a sentence, you can safely assume that the "moved"
constitutent originated in both gaps.

Marcus