Movement Research in ASL

[Rachel Channon]

Dear Kathy,
Since you are at Purdue, I'm sure you are aware of Ronnie Wilbur's and Diane
Brentari's work  related to movement. In addition, my dissertation, as I
recently mentioned on the list, is available on line, and if you can get
copies of Els van der Kooij's and Onno Crasborn's dissertations, I think you
will find them very very valuable.
I think the representation of movement is a central issue for sign language
phonology.  There are two possible approaches (simplifying a bit and leaving
out some complexities to keep this short):
A. Represent the line of motion by specifying its two end points. Example:
IDEA has a begin point at the forehead and an end point in space a short
distance away from the forehead.  All current forms of this approach use
dynamic features to represent some characteristics of the motion, but at
least theoretically, dynamic features are redundant, because the line can
alway be subdivided into a sufficient number of smaller segments, with begin
and end points.
B. Specify one point and use dynamic features such as direction to produce a
line that has a fuzzy end or begin point. Example: IDEA has a specified
begin point on the forehead, a dynamic feature "move out", and an
unspecified end point somewhere out from the forehead.
Almost all researchers have used some version of approach A. Corina (1993)
and I are the only ones that I know of who have used approach B.
Approach A is much more powerful - it can describe the trajectory of ANY
motion. Approach B only works if the set of motions are limited in a variety
of ways. My claim is that this less powerful approach does handle the data,
and is therefore preferable precisely because it  is less powerful - that
is, it generates all and only the kinds of motions that are attested in sign
languages.

I'd be interested in hearing about your work. Have you have already
published or written anything?
Rachel Channon
University of Maryland
rchannon at speakeasy.net