TURNTAKING IN SIGN LANGUAGE INTERACTION

1
TURNTAKING IN SIGN LANGUAGE INTERACTION

• Anne Baker

2
The importance of turn taking?
BACKGROUND

• Ever had the feeling of
• not getting a word in edgeways?
• that the other person would rather you stopped
  talking?
• that the other person wants you to do all the
  talking

3
What is turn taking?
BACKGROUND

• Universal pragmatic principle conversations
  involve different speakers who take the floor.
• The taking of turns is regulated
• - behaviour to hold the floor
• - behaviour to give the floor to another

4
Regulators of turn taking
BACKGROUND

• Transition Relevance Place (Sachs, Schlegloff
  Jefferson 1974)
• Verbal signals
• Vocal signals
• Somatic signals
• All for both turn holding and turn yielding.
• Feedback or backchannels

5
The form of the signals
CULTURAL DIFFERENCES

• Verbal signals
• questions, syntactic completeness
• Vocal signals
• intonation, speed of talking, vocalizations
• Somatic signals
• eye contact, head movement, body contact

6
The use of the signals
CULTURAL DIFFERENCES

• Verbal signals
• is a direct question polite?
• what are the indicators of completeness?
• Vocal signals
• intonation contours vary
• Somatic signals
• is eye contact polite?

7
Simultaneous talk/sign
CULTURAL DIFFERENCES

• Vocal/verbal feedback
• tolerance/requirement varies
• Interruptions
• tolerance varies
• Quick uptake
• amount of time between turns varies
• Floor sharing
• joint construction of talk

8
Variables
A STUDY OF TWO CULTURES

• Speed of talk
• Length of turn
• Length of pauses
• Turn at syntactic break
• Interruptions and overlap
• Feedback
• Type of feedback

9
Method(Baker Junefelt 2007)
A STUDY OF TWO CULTURES

• Conversation between 4 people
• 2 men 1 older, 1 younger
• 2 women 1 older, 1 younger
• Topic what is typical of your culture?
• Length 30 minutes

10
Results Dutch/Swedish comparison
A STUDY OF TWO CULTURES

• Swedish
• Slower
• Fewer and longer
• Longer
• More at break
• Fewer
• Fewer
• Fewer in total
• More vocal, less verbal

• Speed of talk
• Length of turn
• Length of pauses
• Turn at syntactic break
• Interruptions and overlap
• Feedback
• Type of feedback

11
Quantitative Results
A STUDY OF TWO CULTURES
12
Quantitative Results
A STUDY OF TWO CULTURES
13
Gender differences
A STUDY OF TWO CULTURES

• Men in both cultures had more interrruptions than
  the women
• Men had longer turns
• Older men more than younger.

14
The role of vision

• Somatic signals are mostly seen
• What happens in turn taking when you cannot see
  these?
• e.g. telephone conversations?
• in the dark?
• if you are blind?

15
Results
A STUDY OF TWO SUB-CULTURES

• Blind vs sighted
• Quicker
• Longer
• DU shorter SWlonger
• More
• Fewer
• DUmore SW fewer
• DUmore verbal
• SW less vocal
• No somatic

• Speed of talk
• Length of turn
• Length of pauses
• Turn at syntactic break
• Interruptions and Overlap
• Feedback
• Type of feedback

16
Explanations for the differences between blind
and sighted?
A STUDY OF TWO SUB-CULTURES

• Fewer and longer turns?
• Visual cues missing, so continue longer.
• Turn at syntactic break more often?
• More use of this non-visual cue.
• Fewer interruptions?
• Two conflicting auditory signals more confusing

17
Why the cultural differences?
A STUDY OF TWO SUB-CULTURES

• Dutch blind shorter pauses than sighted
• Swedish blind longer pauses than sighted?
• Dutch more overlap Swedish less?
• Dutch more verbal feedback Swedish more vocal.

18
Why the cultural differences?
A STUDY OF TWO SUB-CULTURES

• Dutch shorter pauses Swedish longer?
• Dutch more overlap Swedish less?
• Dutch more verbal feedback Swedish more vocal.
• Swedish lack of tolerance for simultaneous talk
  and tolerance of silence
• Dutch more pressure to grab floor.
• Both follow feedback patterns of own culture.
• What will happen in a blind-sighted conversation?
• What happens in children?

19
Conclusions
A STUDY OF TWO (SUB-)CULTURES

• Clear cultural differences
• In the absence of visual cues blind adults have
  learned to adapt to their cultural pattern
  leading to different behaviours.
• Blind children have to learn the pattern.

20
Visual attention in sign languages
TURNTAKING IN A SIGN LANGUAGE
A STUDY OF

• Signers focus on each others faces when signing
  in signing space.
• Manual signs are seen.
• Children have to learn to divide their attention
  between sign language and environment.

21
Strategies in turntaking
TURNTAKING IN A SIGN LANGUAGE

• Adults wait for eye contact before signing
  (Harris 1987, van den Bogaerde 2000, Loots
  Devisé 2003)
• In Child Directed Signing adults shift the
  signing space into visual field of child
• Waving or tapping used to attract attention or
  sometimes to signal desire to take turn

22
Strategies in turntaking (2)
TURNTAKING IN A SIGN LANGUAGE

• Collaborative floor (simultaneous signing) occurs
  easily in adult sign language interaction
  (Coates Sutton-Spence 2001)
• Overlap in adult-adult signing
• for feedback
• for feedback using repetition
• for clarification

23
Research Questions
TURNTAKING IN NGT ACQUISITION

• In early mother-child interaction
• Is visual attention to signing established at the
  beginning of utterances?
• How much overlap is found?
• What is the function of overlap?
• Are there differences between deaf and hearing
  children?

24
Method
TURNTAKING IN NGT ACQUISITION

• one deaf child
• one hearing child (brothers)
• at ages 20, 30 and 60
• in interaction with same deaf mother
• Sign Language of the Netherlands (NGT) and Dutch
  are used, plus combinations
• Five minutes of interaction analyzed per session
• Units of analysis
• turns, utterances and signs

25
Contribution of child
Results general measures

• Percentage of turns produced by the child in dyad
• 20 30 60
• Deaf-deaf 37 43 44
• Deaf-hearing 52 46 42
• Jonas (H) is more active at age 20 than Mark (D)
• Probably related to Jonas general language level
  (further in spoken Dutch than Mark in NGT)

26
MLU in signs
Results general measures

• Average number of signs per utterance
• 20 30 60
• Mother M 1.9 2.1 3.0
• Mark (D) 1.5 2.3 2.3
• Mother J 2.0 2.0 2.3 Jonas (H) 1.1 1.8 2.3
• Mother mostly ahead of child in MLU as expected
• Both children increase their MLU
• Jonas has a slower start in signs only 36 of
  utterances contain a sign at 20, but 78 at 60

27
How often is the beginning of the utterance seen
by addressee ()?
R Results

• 20 30 60
• Deaf-deaf (Mark)
• seen by mother 85 95 99
• seen by child 77 91 98
• Deaf-hearing (Jonas)
• seen by mother 44 49 67
• seen by child 72 61 62
• Jonas sees 80 of signs Mark 99

28
Percentage of overlapping utterances in dyad
Results

• 20 30 60
• Deaf-deaf 40 42 63
• Deaf-hearing 18 26 44
• Increase in overlap in both dyads
• Deaf-deaf dyad far more overlap collaborative
  floor
• Deaf-deaf dyad - long chains of overlaps not in
  Deaf-hearing

29
Percentage of child interruptions and
simultaneous starts
Results

• 20 30 60
• Child Interruptions
• Mark (D) 32 28 43
• Jonas (H) 58 14 34
• Simultaneous starts
• Deaf-deaf 10 28 17
• Deaf-hearing 10 14 17

30
Functions of overlap
Results

• Mother uses all functions at all ages
• Mark Jonas
• Feedback after 20 few
• Repetition few few
• Clarification small increase none
• Other most most

31
Mark with his mother (60)
32
Jonas with his mother (60)
33
Conclusions

• Visual attention at start of turn
• - mother at 20 waits for attention
• - deaf child learns to check for signing
• - hearing child looks when mother speaks
• Amount of overlap increases with age
• - Deaf-deaf dyad high percentage of overlap
• - Deaf-hearing dyad increase as Jonas signs
  more

34
Conclusions (2)

• Child Interruptions
• - Mark slight increase between 20 and 60,
• learning collaborative floor
• - Jonas overlaps with speech at 20, learns not
• to by 30 and sign overlap at 60
• Simultaneous start
• - Mark more active at 30 resulting in more
• - Jonas increases slightly

35
Conclusions (3)

• Functions
• - most overlap for children real interruption
• - Mark is learning functions of overlap
  (feedback and clarification)
• Deaf-deaf dyad moving towards collaborative
  floor
• Deaf-hearing dyad functions more as hearing,
  voice used by mother to gain attention/turn
• Fine-tuning in deaf-hearing dyad more complex due
  to mothers deafness

36
Effect of turntaking patterns

• Jonas as CODA makes different language choices
  than his mother and than Mark
• - more Dutch
• - more Dutch Base Language (code-blending)
• Jonas mother asks very few clarification
  questions compared to other mothers of CODAS.
• Effect on his language choice?

37
References

• Baker, A.E. B. v.d. Bogaerde (2006) Factors
  influencing child CODAs in early bilingual
  language acquisition. TISLR9, Brazil.
• Bogaerde, B. van den 2000 Input and interaction
  in deaf families, UvA. Utrecht Lot
  (wwwlot.let.uu.nl)
• Bogaerde, B. v.d. A. Baker 2002 Are deaf young
  children bilingual? In G.Morgan B.Woll,
  Directions in sign language research, Amsterdam
  Benjamins
• Coates J R. Sutton-Spence 2001 Turn-taking
  patterns in Deaf conversation. Journal of
  Sociolinguistics 5/4, 507-529
• Harris M.J. et al. 1987 Communication between
  deaf mothers and their deaf infants. Proceedings
  of CLS. In P.Griffith et al. (eds) Univ. of York
• Loots, G. I. Devisé (2003) The use of
  visual-tactile communication strategies by deaf
  and hearing mothers of deaf children. JDSDE 8,
  31-42.

38
CONTACT

• a.e.baker_at_uva.nl
• https//home.medewerker.uva.nl/a.e.baker
• ACLC
• University of Amsterdam
• Spuistraat 210
• 1012 VT Amsterdam
• The Netherlands

39
Amount of overlap