form and function

[Edith A Moravcsik]

This is in response to Jon Aske's interesting message on form and
function. I agree with each of his four main points.

First, Jon pointed out that the description of function should not be
left out of integral accounts of functional objects since function serves
(or may serve) to explain form, such as in the case of knives.
I believe that this point is uncontroversial: both functionalists and
formalists would subscribe to it, with the exception that
functionalists believe that MORE of form is explainable in terms of
meaning than formalists do.

Jon's second point is that, once function is included in descriptions
of functional objects, the description of function should not be seen as
secondary to that of form - a kind of add-on, or afterthought. I agree
that function should not be seen as second to form _in significance_.
As far as _logical_ ranking is concerned, it depends on what exactly
one wants to do. Here are two scenarios:

   (a) If the immediate goal is to _describe the relationship between form
and function_, then form and function are logically on a par - that is,
neither is secondary to the other: the descriptions of both are logically
primary to (i.e., presupposed by) statements about the relationship
between the two.

   (b) If the goal is to _explain form in terms of function_, then
function is logically secondary to form since form needs to be described
before any explanation can be looked for it. But "logically secondary"
does not mean "second in importance".

Thirdly, Jon argued that there was no reason to describe form separately
from function, rather than describing the two together, without a clear
separation of the two concepts. I basically agree on this, too:
traditional descriptive grammars often proceed in this manner. Describing
form and function as two distinct entities is imperative ONLY if our goal
is to study the relationships between the two; such as how form conveys
meaning and how meaning explains form. If the goal of linguistic
description is formulated differently so that "form" and "function" do not
figure as terms, then separate descriptions of form and meaning are not
called for.

I think, however, that one can argue for the usefulness of accounting for
meaning-form relations in language chosen as the goal of descriptive
linguistics. First, there are some interesting linguistic patterns that we
could not talk about otherwise. Such are intra-language and cross-language
synonymy and ambiguity and the existence of formally well-formed but
semantically ill-formed sentences, or formally ill-formed but semantically
well-formed ones. Second, by looking at sentences as relating form
and meaning, we also manage to link language to many extralinguistic
phenomena. These  include other semiotic objects which, by definition,
provide a link between some form and some meaning (such as body language,
road signs etc.). More broadly, they also include things that share form
properties with language even if there is no meaning attached (such as any
temporal activity whether it conveys meaning or not; or for that matter,
any complex object that consists of classifiable parts).

Fourthly, Jon raised the question of why there should be different people
to describe form and to describe function. As he said, if you describe
form, why would you stop there and leave "the interesting stuff" to
others? I'd say linguists should be free to focus on one or more aspects
of the total endeavor.

Edith



   ************************************************************************
                         Edith A. Moravcsik
                         Department of Foreign Languages and Linguistics
                         University of Wisconsin-Milwaukee
                         Milwaukee, WI 53201-0413
                         USA

                         E-mail: edith at uwm.edu
                         Telephone: (414) 229-6794 /office/
                                    (414) 332-0141 /home/
                         Fax: (414) 229-2741