LINGUIST List 4.665

Fri 03 Sep 1993

Disc: Marginal utterances

Editor for this issue: <>

Directory

1. Bill Bennett, T

re:theoretical status of marginal utterances.

I have seen no reply to Mark Seidenberg's enquiry about our attitude to
ungrammatical utterances. The absence of any response to Mark would surely
appear impolite. This must not be.

 Mark asked whether uncertainly grammatical utterances were welcome for the
development of grammatical theory OR whether it was the marginality of such
utterances that was signalled by the grammar.

 I hope that Mark Seidenberg's enquiry is reasonably adequately represented,
in spite my brevity in reporting. I trust also that this draft reply will be
helpful.

 Fortunately, the task of seeking marginality is done for us by some kinds of
speakers. My theory of language behaviour includes the careful speaker/hearer
who is constantly matching intuition to introspection in language use. It was
these artefacts of my linguistic theory who belaboured me with their linguistic
worries at meeting written utterances of the following form

 That is the linguist whom we think seems obscure.

 In turning to me they assumed that I had a model-theoretical means for
explaining the structure. I was glad to have access to notions such as
"clause", "government" and "case". Local to my model-theory is the concept of
ECM. Out of this I have spun a structural account which should be featuring in
one of our journals fairly shortly.

 The structure appears to be restricted to British English, and this confirms
the well-known problem for the conventional linguistic theory: how to define a
language by sub-dividing a language? Another problem is the status of this new
structure within the synchrony of English, 1993? Yet, Saussure did not tie up
the linguist synchronically, as we too often believe, without leaving our
theory a push-down notion to handle language change. This case of syntactic
change is visible in -Usage and Abusage-, which appeared in the US in 1942 (my
copy is dated 1965), and in the best of the English newspapers at this end of
the century. Saussure did hazard 100 years as the time needed for an /etat de
langue. Where should that aspect of language use appear in our linguistic
theory?

Bill Bennett.