Date: Wed, 6 Nov 1996 17:28:30 -0500 (EST)
Message-Id: <199611062228.RAA01021@sarah.albany.edu>
In-Reply-To: <01IBHRE0G2EQ9BXKIK@EMUVAX.EMICH.EDU>
To: Linguistics Conference <LINCONF@tamvm1.tamu.edu>
From: George Aaron Broadwell <gb661@csc.albany.edu>
Subject: Defining the scope of binding theory (Reply to Seely)
In my view, binding theory is a very general formal system that may apply to
a variety of distinct
empirical domains, including (perhaps)
pronouns
switch-reference markers
negative polarity items
traces (e.g. in approaches like Aoun's generalized binding)
We tend to think of binding theory as the part of grammar that treats
pronouns, but that is just an accident of the fact that we discovered
binding theory while looking at pronouns. It is probably a mistake to
assume that selected chunks of grammar (like "pronouns") will turn out to be
completely accounted for by single components of our theory.
When we say that things "are anaphors", I believe that that is only
short-hand for "the distribution of this item is described by the binding
theory." I doubt that it is possible to define a priori the elements that
binding theory applies to by semantic or morphological criteria.
In my view, binding theory is a formal tool in language's tool bag.
Determing how this tool is used is a matter for empirical investigation.
There may be close correlations between certain semantic and morphological
criteria and the formal devices that a languages uses, but this is to be
demonstrated, not assumed.
To appeal to our favorite source of authority, biological analogy, think of
the role of chemical messengers in the body (e.g. estrogen, testosterone,
adrenalin, etc.). These hormones have effects in several different organs
of the body. To determine whether, say, adrenalin affects the kidney or the
brain, we don't use some prior notion "brain chemicals" or "kidney
chemicals" to decide the question -- we do the tests and see how it comes
out. Similarly, if we want to know how the kidney works, we don't assume
that all the relevant principles are unique to kidneys.
To decide if binding theory is operating with negative polarity items, we
can't rely on assumptions about what the scope of binding theory is --
instead, we have to do the tests (for things like command and locality) and
use those results to construct the most parsimonious theory we can.
Similarly, if we want to understand how reflexive pronouns work, we can't
assume that a structurally based binding theory is always the appropriate
formal device for every language. We may well discover that other
components of grammar interact with or
even supercede binding theory.
--------------------------------------------------------------------------
George Aaron Broadwell, g.broadwell@albany.edu
Anthropology; Linguistics and Cognitive Science,
University at Albany, SUNY, Albany, NY 12222 | 518-442-4711
--------------------------------------------------------------------------
"Hsun Tze, who lived about seventy years after Mencius, in his chapter on
the Rectification of the Names, repeats the recommendation found in the Li
Ki, Book III, iv, 16, that those who introduce new terms or make
unauthorized distinctions should be put to death."
--- I.A. Richards, *Mencius on the mind* (London: Kegan Paul. 1932.)