The Handshape Parameter in
Kenyan Sign Language
Hope E. Morgan
Rachel Mayberry
U.C. SAN DIEGO
TISLR 10, Purdue University, Indiana
Sept. 30 – Oct. 2, 2010
Goals
• Provide a description of the phonological system of
under-studied sign language, Kenyan Sign Language.
• Show that KSL conforms to phonological restrictions
governing two-handed signs (Symmetry & Dominance
Conditions) in most ways, with a few exceptions.
• Provide evidence for a new kind of handshape
restriction: on the dominant hand in two-handed
signs with handshapes that don’t match.
PART 1
• History & background of KSL
• Handshape parameter in Kenyan Sign Language
PART 2
• Symmetry & Dominance Condition in KSL
• Explanations for KSL signs that don’t conform to the
Dominance Condition
Background of Kenyan Sign Language
• Origin in 1960s with 2 deaf schools in the west;
spread during the1970s-80s, with standardization
over that time. (Okombo & Akach 1997)
• 46 primary schools, 4 secondary schools, 35 units
(serving 8300 + students) (U.S. Peace Corps Survey 2007)
• Evidence for some influence of ASL and/or Signed
Exact English in the lexicon (Hochgesang 2007; Roberts 2009;
Morgan, et al., in preparation)
• No more than 20% full cognates with ASL; “not a
creole of ASL” (Roberts 2009)
Background of Kenyan Sign Language
• Origin in 1960s with 2 deaf schools in the west;
spread during the1970s-80s, with standardization
over that time. (Okombo & Akach 1997)
• 46 primary schools, 35 units, 4 secondary schools
(serving 8300 + students) (U.S. Peace Corps Survey 2007)
• Evidence for some influence of ASL and/or Signed
Exact English in the lexicon (Hochgesang 2007; Roberts 2009;
Morgan, et al., in preparation)
• No more than 20% full cognates with ASL; “not a
creole of ASL” (Roberts 2009)
Deaf Primary Schools & Units in Kenya
Background of Kenyan Sign Language
• Origin in 1960s with 2 deaf schools in the west;
spread during the1970s-80s, with standardization
over that time. (Okombo & Akach 1997)
• 46 primary schools, 4 secondary schools, 35 units
(serving 8300 + students) (U.S. Peace Corps Survey 2007)
• Evidence for some (limited) influence of ASL and/or
Signed Exact English in the lexicon (Hochgesang 2007;
Roberts 2009; Morgan, Gilchrist, & Burichani, in prep)
• No more than 20% full cognates with ASL; “not a
creole of ASL.” (Roberts 2009)
Data set
• Interactive video dictionary of 991 QuickTime movies.
• Joint project of the KSL Research Project (U. of Nairobi) & U.S.
Peace Corps volunteers (2004) .
• Design: a tool for families with deaf members to learn KSL.
• Female signer in her 30s from Central Province, Kenya.
• 958 lexical entries, after 33 fingerspellings, duplicates, &
homophones, removed.
• Data coded in a in a FileMaker Pro database:
Data set
• Interactive video dictionary of 991 QuickTime movies.
• Produced by KSL Research Project (U. of Nairobi) & U.S. Peace
Corps volunteers (2004) .
• Design: a tool for families with deaf members to learn KSL.
• Female signer in her 30s from Central Province, Kenya.
• 958 lexical entries, after 33 fingerspellings, duplicates, &
homophones, removed.
• Data coded in a in a FileMaker Pro database:
FIELDS:
HANDSHAPE
HANDEDNESS (1or2 hands)
MOVEMENT TYPE
SIGN TYPE (Battison 1978)
Phonological parameters in KSL
• The three major phonological parameters in signed
languages are handshape, location, and movement.
• Minimal pairs: two signs that vary by only one
parameter
show that each parameter is phonemic. That is, a change
in only a single handshape, a single location, or a single
movement can change the meaning of the sign.
(near) Minimal Pairs in KSL
HANDSHAPE LOCATION MOVEMENT
GITHERI (beans & rice dish)
LUO (name of tribe)
(near) Minimal Pairs in KSL
HANDSHAPE LOCATION MOVEMENT
GLASSES
A.I.D.S
(near) Minimal Pairs in KSL
HANDSHAPE LOCATION MOVEMENT
PORRIDGE
IGNORE
Phonetic Inventory of 52 KSL Handshapes
Using Hamburg Notation System (Prillwitz, et al. 1989)
0
50
100
150
200
flat
196
1
146
53
open
23%
16%
47
A
35
T
bent
S
302927
claw
X
flat-o
242421
C
F
curved
191717
A-thumb
H
bent-1
V
cupped
1515141312
open-curved
L
complex
11 1110
small-C
bent-L
i
bent-V
9 9 7 6
closed-G
spray
open-E
E
B
flat-tense
W
WOOD
Y
open-bent
G
O
bent-L-claw
mid-bend
K
5 5 5 4 4 3 3 3 3 3 2 2 2 2 2
teeny-C
Distribution of KSL Handshapes
closed-claw
open-G
bent-i
V-claw
uganda
N
R
4
flat-thumb-out
NAMIBIA
Thumb-in-fist
1 1 1 1 1 1 1 1 1 1 0 0
middle+thumb
Distribution in other sign languages
Rozelle (2003)
American Sign Language
New Zealand SL
SL of the Netherlands (NGT)
Finnish Sign Language Korean Sign Language
Italian Sign Language (LIS)
(SVK)
Data & images from Rozelle (2003)
Symmetry & Dominance Conditions
(Battison 1978)
• Govern combinatory possibilities of the two hands in
two-handed signs:
When both hands involved in a sign, what combinations of
handshape, movement, & location are possible for each hand?
• Constrain phonological complexity of signs.
• Generally have held up in cross-linguistic studies.
The Symmetry Condition
(a) if both hands of a sign move independently during its
articulation, then
(b) Both hands must be specified for the same location, the same
handshape, the same movement (whether performed
simultaneously or alternatingly)
“Type 1 signs” (Battison 1978)
The Dominance Condition
(a) if the hands of a two-handed sign do not share the same
specification for handshape (i.e. they are different), then
(b) one hand must be passive while the active hand articulates
the movement and
(c) the specification of the passive handshape is restricted to be
one of a small set: A, S, B, 5, G/1, C, O. [unmarked set of
handshapes]
“Type 3 signs” (Battison 1978)
Two criteria for the non-dominant hand:
• passive (not moving)
• shape is restricted
KSL Sign Types (Battison’s typology)
Sign Type
Description
In KSL Dictionary:
Type 0
1-handed; neutral space
148
39%
1-handed
Type X
1-handed; contact body
221
2-handed; handshape &
Type 1 movement matched
310
2-handed; handshape matched, 48%
Type 2
movement unmatched
65
2-handed
2-handed; handshape &
Type 3
movement unmatched
91
80
13%
Compounds
[mixed]
131
compounds
958 Total
Symmetry Condition in KSL
All two handed signs in which both hands move
independently have the same handshape, except:
Two lexical entries that violate both conditions:
START
PROBLEM
Dominance Condition in KSL
Two criteria for the non-dominant hand:
• passive (not moving)
• shape is restricted
FIRST CRITERION:
All two handed signs with the unmatched handshapes
have a passive non-dominant hand, except one:
START
Dominance Condition in KSL
Two criteria for the non-dominant hand:
• passive (not moving)
• shape is restricted
SECOND CRITERION:
Approaches to handshape restriction:
1. Battison’s set of 7 handshapes will apply to all languages
2. Language-specific sets (Rozelle 2003; Eccarius & Brentari 2007: 1178)
3. A universal unmarked set: B 5 1 s a (Sandler/Lillo-Martin
2006, Rozelle 2003)
4. Markedness across both hands, not just non-dominant
hand (Eccarius & Brentari 2007)
Type 3 signs in KSL
Handshapes on the non-dominant hand in Type 3 signs:
Battison:
Picture
Name
# Signs
%
� B flat/B
� �
Universal � S S
� �
Unmarked
� 5 5
� �
Set
� 1 1 � �
� A A
� �
t thumb-T
� �
complex
� �
� C C
� �
claw
� �
v V
� �
(includes signs in compounds)
What do these odd cases tell us?
• ASL also has cases that violate handshape restriction:
Eccarius & Brentari (2007: 1180) – 4.1% of Type 3 signs
Napoli & Wu (2003: 128) – 3.6% of Type 3 signs
e.g., THEN , SKIP - CLASS , CHOOSE , etc.
• Three hypotheses for the KSL cases:
1. These handshapes are in a KSL-specific unmarked set
2. Conform to featural constraints across both hands
3. Another phonologically explanation?
Rare cases with no pattern/generalization
HYPOTHESIS 1:
Language-specific “unmarked set” ?
• Rozelle 2003; Eccarius &
Brentari 2007
� � �
�
• Are the handshapes on the
b flat/B
� �
non-dominant hand in Type 3
signs “unmarked” in KSL?
s A S/A
� �
• Frequency as measure
5 open/5
� �
o f markedness (Greenberg 2005)
1 1
� �
t thumb-T
� �
[complex ]
� �
claw
� �
CONCLUSION: A language-
specific set does not explain v V
� �
these handshape. c C
� �
HYPOTHESIS 2:
Featural constraints on both hands?
• Is the restriction on the complexity across both
hands, not just the non-dominant hand?
• Eccarius & Brentari (2007) :
Markedness score on each hand for selected fingers and
joint specification.
Maximum possible = 4 marked features.
Constraint: of two marked features across both hands
• Results: All Type 3 signs have a score of 2 or less.
CONCLUSION: featural constraints
across the hands account for all of
the Type 3 signs in KSL.
HYPOTHESIS 3:
Another phonological generalization
• In Type 3 signs, the 1 handshape becomes the most
frequent on the dominant hand:
ALL SIGNS
TYPE 3 SIGNS
Type 3
Base
Universal Frequency Type 3
Handshape
unmarked
(H1)
Count
Frequency
B � 1 � �
1 � B � �
A S � A S � �
(showing only the most common handshapes)
HYPOTHESIS 3:
Another phonological generalization
The seven signs show a pattern on the dominant hand:
� � �
Hand interaction effect?
1. RUSSIA
1 c
When H2 is marked
2. POTATO
1 (infrequent), the H1 must
3. START
1 v be a 1 handshape ( 1 ).
4. CONDUCTOR
1
5. HOW-MANY
1
6. CLITORIS
t
7. FEMALE-
CIRCUMCISION
1 t
HYPOTHESIS 3:
Another phonological generalization
Battison (1978: 36)
“the reduction of from approximately 45 handshapes to a mere 7 greatly
reduces the complexity of the sign and increases the redundancy, since a
specification of one hand from among seven possibilities requires less
information than a specification among 45 possibilities.”
Information structure constraint?
Prefer the most common (least marked)
handshape in a complex sign (e.g., Type 3).
When that is not possible, choose the next
most common (least marked) handshape.
Summary of Findings
• KSL has sub-lexical structure similar to other SLs:
A phonetic inventory of approx. 52 handshapes.
The frequency distribution of handshapes in the lexicon in an
exponential decay curve, similar to other sign languages.
Handshape is constrained in two-handed signs (Symmetry &
Dominance Condition).
• KSL has a preference for the 1 handshape on the
dominant hand in Type 3 signs.
• When non-dominant hand (H2) is marked, the
dominant hand (H1) will surface as a 1 handshape.
Thank you!
ACKNOWLEDGMENTS:
• Language informants: Isabella Mugure, Evans Burichani, Jared
Osome, Carolyne Obwango.
• Commentary & support: Comparative Language Acquisition Lab
at UCSD; Dr. Sharon Rose; Dr. Eric Bakovic; Dr. Victoria Nyst;
UCSD Division of Social Sciences.
REFERENCES:
Battison, Robbin. 1978. Lexical borrowing in American Sign Language. Silver Spring, MD: Linstok Press.
Eccarius, Petra & Diane Brentari (2007). Symmetry and dominance: A cross-linguistic study of signs and classifier
constructions. Lingua 117. 1169-1201.
Greenberg, Joseph H. (2005). Language universals: with special reference to feature hierarchies. Berlin: Mouton de
Gruyter.
Hendriks, Hermina Berndina. 2008. Jordanian Sign Language: aspects of grammar from a cross-linguistic
perspective. Utrecht, the Netherlands: LOT.
Hochgesang, Julie (2007). Exploring the language contact situation between deaf and hearing in Kenya. Poster
presented at Theoretical Issues in Sign Language Research 9 Meeting. Florianoipolis, Brazil.
Hulst, Harry van der (1996). On the other hand. Lingua 98. 121-143.
Okombo, D. Okoth & Philemon O. Akach (1997). Language convergence and wave phenomena in the growth of
a national sign language in Kenya. International Journal of the Sociology of Language 125. 131-144.
Prillwitz, Siegmund et al (1989). HamNoSys, version 2.0; Hamburg Notation System for Sign Languages: an
introductory guide. International Studies on Sign Language and Communication of the Deaf 5. Hamburg: Signum.
Roberts, Page (2009). Is ASL a colonial language in Kenya?: a lexical comparison of ASL and KSL. Unpublished
manuscript, Gallaudet University.
Rozelle, Lorna (2003). The structure of sign language lexicons: inventory and distribution of handshape and location.
PhD dissertation. University of Washington.
van der Hulst & Channon (2009). http://www.ldc.upenn.edu/signtyp/. Downloaded May 2009.
U.S. Peace Corps, Kenya (2007). Building a brighter future: Peace Corps deaf school survey. Nairobi, Kenya: U.S. Peace
Corps.
