form without meaning

[Jon Aske]

I wanted to add something to what I said yesterday. I have received one
private response to that posting, in which Edith's rosary analogy was
claimed to be instrumental in explaining the need to separate form from
function, and  I would like to explain why I do not think that that
analogy is valid.

If someone unfamiliar with rosaries came across one, s/he could describe
its form, but would be missing a big part of the picture if the
description omitted the function of the rosary.  It would be a very
limited description, one which we would accept if we had no other
choice, but not one that I would be satisfied with.  Furthermore, and
this is crucial, the rosary, as an artifact, has a formal (physical)
existence in the world apart from its function, but I don't think that
linguistic constructions do.  I believer that constructions exist in the
*minds* of speakers and that they are learned and represented as forms
with functions attached to them.  Thus, to the extent that we are trying
to describe linguistic 'competence' (in addition to 'performance' and
the relation between the two) I think that we have to describe
constructions in their totality.

I do not believe, for instance, that in Basque, and in other languages
in which the order of constituents in asserted clauses depends primarily
on pragmatic characteristics (roles and statuses) of the ideas
represented by those constituents, one can describe such ordering
without resorting to those pragmatic categories and statuses.  I believe
that this has been recognized even by formalists, who have resorted to
(formal?, functional?) categories such as [+FOCUS] or [+TENSE] or
[+INFL] to account for the facts (that is what David was talking about
the other day).  That is a significant step forward I believe, but the
way it is implemented seems to me to be more of an attempt to salvage a
faulty model of linguistic units/systems than anything else.

One more thing.  I believe that all so-called structural units in
language are cognitive units.  What holds things together in any
construction are semantic and pragmatic (informational) 'forces'
'binding' those elements together.  What some people call a VP (a
syntactic unit), for instance, exists only to the extent that there are
semantic and pragmatic reasons for holding some elements of the clause
together, while excluding others (the clause itself is a cognitive unit
par excellence).  But, unless we view those 'forces' as being functional
(semantic and pragmatic) in nature and having a variable nature
(non-referential objects, for instance, are bound to verbs more strongly
than referential and topical ones), and as being overridden under
certain circumstances (some clauses do not have topics), we will end up
believing silly things, such as that some languages have VPs while
others don't.  And it just isn't that simple.

Anyway, I'm going back to lurking too, just like everyone else.  Unless
someone gets me going again, of course.

Have a great weekend,

Jon

--
Jon Aske
jaske at abacus.bates.edu
http://www.bates.edu/~jaske/
--
Hiru belarritan igaren hitz isila, orotan lasterka dabila
"A secret that has been through three ears, won't remain a secret much
longer."