'totally novel sentence'

[Johanna Rubba]

I always understood 'totally novel sentence' to mean a totally new
_combination_ of previously existing units -- the fundamental insight
behind this being that grammr provides us with underspecified patterns
into which existing forms can be plugged, enabling the creativity
which is one element distinguishing human language from animal
communication
systems. That this creativity is theoretically infinite was, I have been
trained to believe, an important insight of generative linguistics.

I find this nontrivial. But this great insight does have a downside -- I
believe it helped drive a wedge between semantics and syntax by giving the
impression that the patterns were the core of the grammar, and you could
plug any old thing into them and get a 'grammatical' sentence, such as the
turnip doozie contributed by John Myhill. 'Grammatical, but semantically
deviant' has been with us ever since. Why exclude the potential for making
sense of a particular concatenation of lexical items from the purview of
'grammaticality'?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Johanna Rubba   Assistant Professor, Linguistics              ~
English Department, California Polytechnic State University   ~
San Luis Obispo, CA 93407                                     ~
Tel. (805)-756-2184     Fax: (805)-756-6374                   ~
E-mail: jrubba at polymail.calpoly.edu                           ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~