The Phonetics of Fingerspelling

114 Pages·1992·24.052 MB·English

by  Sherman Wilcox

THE PHONETICS OF FINGERSPELLING
STUDIES IN
SPEECH PATHOLOGY AND CLINICAL LINGUISTICS
SERIES EDITORS
Martin J. Ball Raymond D. Kent
Dept. of Communication Dept. of Communicative
Univ. of Ulster at Jordanstown Disorders
Shore Road University of Wisconsin
Newtownabbey BT37 OQB 1975 Willow Drive
UK - Northern Ireland Madison
Wisconsin 53706, USA
EDITORIAL BOARD
Christopher Code (University of Sydney, Australia)
Alvirda Farmer (San Jose State University, USA)
John H.V. Gilbert (University of British Columbia, Can.)
Yvan Lebrun (Vrije Universiteit, Belgium)
Lise Menn (University of Colorado at Boulder, USA)
AIMS AND SCOPE
The establishment of this series reflects the growth of both interest and research into
disorders of speech and language. It is intended that the series will provide a plat
form for the development of academic debate and enquiry into the related fields of
speech pathology and clinical linguistics.
To this end, the series will publish book length studies or collections of papers on
aspects of disordered communication, and the relation between language theory and
language pathology.
Volume 4
Sherman Wilcox
The Phonetics of Fingerspelling
THE PHONETICS
OF
FINGERSPELLING
SHERMAN WILCOX
University of New Mexico
JOHN BENJAMINS PUBLISHING COMPANY
AMSTERDAM/PHILADELPHIA
1992
Library of Congress Cataloging-in-Publication Data
Wilcox, Sherman.
The phonetics of fingerspelling / Sherman Wilcox.
p. cm. -- (Studies in speech pathology and clinical linguistics, ISSN 0927-1813;
v. 4)
Includes bibliographical references and index.
1. Finger spelling. 2. Grammar, General and comparative-Phonetics. I. Title. II.
Series.
HV2477.W55 1992
419--dc20 92-8926
ISBN 90 272 4334 4 (Eur.)/l-55619-390-4 (US)(alk. paper) CIP
® Copyright 1992 - John Benjamins B.V.
No part of this book may be reproduced in any form, by print, photoprint, microfilm, or
any other means, without written permission from the publisher.
John Benjamins Publishing Co. · P.O. Box 75577 · 1070 AN Amsterdam · The Netherlands
John Benjamins North America · 821 Bethlehem Pike · Philadelphia, PA 19118 · USA
TABLE OF CONTENTS
Chapter 1: A Linguistic Approach to Fingerspelling 1
The Structure of Signed Languages 1
Signs as Unsegmentable Wholes 2
Signs as Simultaneously Segmentable 3
Signs as Sequentially Segmentable 4
Fingerspelling 9
Lexical Borrowing 13
Research on Fingerspelling 14
Fingerspelling as a Research Tool 15
Models of Fingerspelling 16
The Production and Perception of Fingerspelling 19
Learning to Fingerspell 21
Fingerspelling Fluency 23
Coarticulation in Fingerspelling 24
Kinematics, Dynamics, and Articulatory Movements 25
Organization of the Studies 27
Learning to Fingerspell 27
The Phonetics of Fingerspelling 27
Chapter 2: Learning to Fingerspell 29
Adult Acquisition of Fingerspelling 29
Method 31
Subjects 31
Materials 31
Procedures 31
Results 31
Chapter 3: The Measurement of Fingerspelling Movement 37
Background 37
Hardware 37
Software 39
Sample Data Analysis 43
Methods and Procedures for Motion Analysis Studies 48
Subjects 48
Stimulus Materials 48
Room and Camera Arrangement 49
vi PHONETICS OF FINGERSPELLING
Marker Placement and Attachment 52
Sampling Times and Rates 52
Calibration 53
Chapter 4: Targets and Transitions 55
Models Revisited 55
The Salience of Targets and Transitions 59
Dynamic Modeling of Phonetic Structure 62
Chapter 5: Towards a Dynamics of Fingerspelling 65
From Kinematics to Dynamics 65
Fingerspelling Speed 65
Cooperativity in Fingerspelling Production 69
Discussion 74
Chapter 6: Future Directions in Signed Language Research 79
The Linguistic Study of Fingerspelling 79
Limitations and Suggestions for Further Study 81
Motion Analysis of Signed Language 81
Learning to Fingerspell 82
Models of Fingerspelling 83
Sign Science and Speech Science: The Search for Unity 85
Speech as Gestures 86
Signing as Gestures 88
Is Speech Special? 90
Notes 93
References 97
Index 107
Chapter 1
A Linguistic Approach
to Fingerspelling
The Structure of Signed Languages
"A language is a set of principles relating meanings and phonetic sequences" (Lan-
gacker, 1972:1). One type of phonetic medium is sound; one way of conveying lan
guage is through speech. Until recently, however, most linguists and researchers in
the language sciences would have gone beyond the characterization of sound as a
possible phonetic medium and would have insisted that more than being "merely
one way of conveying language, the sounds of speech are, instead, its common and
privileged carriers" (Liberman, Cooper, Shankweiler, and Studdert-Kennedy,
1967:431).
It is certainly true that in approaching the study of language, the data most
readily at hand come from speech. Spoken languages are all around us; sounds are
clearly a common way of representing meanings. But are they privileged? We must
be careful about making claims concerning language in general on the basis of a par
ticular language. If we examined only English, for example, we might conclude that
relatively strict word order is a property of all languages. Cross-linguistic data are
necessary in order to get a full picture of the structure of human language.
Just as important are cross-modal data. Linguists now recognize another medi
um through which languages can be conveyed. Numerous natural languages exist
which are represented in the visual-gestural modality. Research has established that
these signed languages1 are fully independent of spoken languages. The most exten
sive research to date has examined American Sign Language (ASL), the signed lan
guage used by deaf people in America (Bellugi and Studdert-Kennedy, 1980; Fischer
& Siple, 1990; Grosjean and Lane, 1980; Klima and Bellugi, 1979; Siple, 1978; Wilbur,
1987).
2 PHONETICS OF FINGERSPELLING
For many years, signed languages were thought to be dependent on spoken lan
guages. In particular, ASL was considered to be merely a derivative of English. One
explanation for this misconception can be found in the distinction between languag
es and modalities (Table 1).
Table 1: The Language/Mode Distinction
Language Mode
English
Speech (Spoken Language)
Spanish
Cochiti
French Writing (Written Language)
American Sign Language
French Sign Language
Signing (Signed Language)
British Sign Language
Speech is the primary means of representation for most natural languages.
Written languages are usually characterized as secondary systems used to represent
spoken languages in a written form, for particular purposes such as overcoming the
temporal or distance limitations of speech (Gelb, 1963).
Table 1 can be used to explore the relationships among languages and their
modes of representation. English, Spanish, and French, for example, are both spoken
and written. Cochiti, like many of the world's languages, has no conventional written
form. American Sign Language has no conventional spoken or written form; its only
means of expression is through the signed modality.2
Since speech — spoken language — was for so many years viewed as primary
and the written modality as secondary, it was perhaps natural for linguists to suppose
that the signed modality was likewise used as only a secondary medium for the rep
resentation of spoken languages. What must be recognized is that the signed modal
ity can serve as the primary medium for a set of natural languages.
Signs as Unsegmentable Wholes
As for our understanding of the phonological structure of signed languages, the com
mon belief for many years was that the words of signed languages, unlike the words
LINGUISTIC APPROACH 3
of spoken languages, had no internal structure. Signs (words) were regarded as unan-
alyzable wholes. Thus, a word such as LIKE (Figure 1) was regarded as a wholistic
item, an unanalyzable gesture, with no internal structure.
Figure 1. The ASL Word LIKE
Signs as Simultaneously Segmentahle
It was not until the publication of the classic monograph, "Sign Language Structure"
(Stokoe, 1960), that this conception was refuted. Stokoe noted that signs, like words,
do have an internal structure by demonstrating that signs must be described in terms
of three parameters: dez (hand configuration), tab (location), and sig (movement).
Stokoe originally coined the term chereme for these sublexical units, but later re
searchers pointed out that what was being described was the phonological structure
of ASL and preferred to call them phonemes. Working within Stokoe's framework,
researchers began in earnest to study the phonology of signed languages (Battison,
1973,1974; Battison, Markowicz, and Woodward, 1975; Friedman, 1976; McIntire,
1977;Wilbur, 1985).
As an example of how Stokoe's three parameters can be used to describe the
phonology of ASL, it is instructive to consider minimal pairs. Varying only the hand-
shape parameter, it is possible to sign the minimal pair ONION and CHINESE.
Varying only the location parameter, we can sign the minimal pair CHINESE and
BORING. Varying only the movement parameter yields the minimal pair NAME
and SHORT. A fourth parameter, orientation, was suggested by Battison (1973; Bat
tison, Markowicz, and Woodward, 1975). Varying only the orientation parameter
yields the minimal pair TRAIN and SHORT.
As research proceeded under this view of sign structure, linguists began to note
that the structure of signed and spoken words differed in one remarkable way. Unlike
4 PHONETICS OF FINGERSPELLING
spoken words, in which phonemes are arranged sequentially, the phonemes of signs
are simultaneously present. As Stokoe noted (1960:37):
The sign morpheme, however, unlike the morpheme or word of a spo
ken language, is seen as simultaneously, not sequentially, produced. Analy
sis of the sign morpheme then cannot be segmentation in time but must be
aspectual.
The different sublexical structure of signs and words — simultaneous versus se
quential — was taken as evidence for how languages are differentially structured by
their modality (Bellugi and Studdert-Kennedy, 1980). Klima and Bellugi (1979:39)
succinctly present the argument:
Thus the lexical items of ASL and all other primary sign languages we
know of appear to be constituted in a different way from those of spoken
languages: the organization of signs is primarily simultaneous rather than
sequential. ASL uses a spatial medium; and this may crucially influence its
organization.
Signs as Sequentially Segmentahle
Recent phonological research on American Sign Language is suggesting that signs are
not only analyzable simultaneously but also sequentially (Liddell, 1984a, 1984b,
1990; Liddell and Johnson, 1986, 1989; Padden & Perlmutter, 1987; Perlmutter,
1990; Sandler, 1986, 1989, 1990; Wilbur, 1990). Liddell, for example, argues that
signs can be sequentially analyzed into Movements (M) and Holds (H). Under this
analysis, the sign LIKE consists of three segments (HMH), as in Figure 2.
The initial H segment is defined by several features; for the sake of simplicity,
only a few are depicted in Figure 2 and only the handshape parameter is given values.
In Liddell's analysis, autosegmental feature spreading would account for the M-seg-
ment features.
LiddelTs model of ASL phonology has been developed in great detail. Present
ing evidence from word formation and compounds, and drawing on work in spoken
language autosegmental phonology, he convincingly demonstrates that the phono
logical structure of signs is more similar to that of spoken words than was previously
thought.
Sandler (1986) has proposed an alternative autosegmental model of ASL pho
nology. In her model, the opposition is not between movements and holds but be
tween movements and locations, with handshape on a separate autosegmental tier;