Acquisition of Perspective and Role Shift in Deaf Children:Evidence from British Sign LanguageKearsy Cormier* & Sandra Smith^*Deafness Cognition and Language Research Centre, University College London^Centre for Deaf Studies, University of Bristol

BackgroundSign languages such as British Sign Language (BSL) make regular use of a strategy known asrole shift (aka referential shift, constructed action, constructed dialogue), which involvesenactment - use of the signer’s head, face, hands, and/or body to describe a referent’s actions,thoughts or feelings. Role shift is a way for a signer to take on the perspective of a referent, e.g.a character within a narrative. Fluent signers are able to seamlessly switch between differentroles/perspectives (including that of the signer himself, or an additional role of narrator), whilestill maintaining referential & discourse cohesion. Figure 1 shows two examples of role shift.Generally signers establish a referent nominally then shift into the role of that referent.Discourse markers such as lexical signs THEN, KNOW and deictic points (functioning asdeterminers, pronouns, or adverbials) may also be used to aid in discourse cohesion throughoutnarratives in which role shift is used.

Figure 1: Two examples of role shift; a) depicts a bear which is about to attack a person, and b)depicts the person who is about to be attacked pretending to be dead.

a) b)

Role shift can be accompanied by classifier constructions, which describe location and/or motionof referents, as in Figure 2. Role shift can also be accompanied by lexical signs (nouns, verbs,adjectives, etc), as in Figure 3.

Acquisition of role shift and classifier constructions in deaf children begins at about 3 years ofage but progresses slowly. Even by ages 9-10, deaf children struggle with various discourse andpragmatic functions of role shift (Reilly 2000, Slobin et al. 2003).

MethodsParticipants consisted of 15 severely/profoundly deaf children in the following categories:

Methods (continued)The table below shows the alias name for each child participant, their ages in years;months, andtheir preferred language as reported in parental questionnaire.

TaskChildren were asked to watch short film clips from a Wallace & Gromit film and a Pink Panthercartoon and to describe each to a Deaf adult native signer. Instructions/clarifications wereprovided in the child’s preferred language.

Figure 4: Stills from the two stimulus clips

a) b) c) d)

Coding/analysisSigned productions were coded for any form of enactment, including which role(s) the child wastaking on and which body part(s) were representing each role. Signed productions were alsocoded for any BSL lexical signs and classifier constructions or equivalent (specifically so-called“entity” classifiers where the hand represents all or part of an entity in order to describe locationand/or motion of that entity).

ResultsAll but 2 of the children (Isabel and Millie) used some form of enactment in their narratives.Note that 2 other children (Oliver and Hunter) only produced one of the two narratives.

Figure 5a: Tokens of enactment across all children, both narratives

Figure 6a: Use of lexical discourse markers (e.g. THEN, KNOW, deictic points)

Figure 7a: Percentage of enactment tokens (from Fig. 5a) which were accompanied by BSLlexical sign(s) or entity classifier(s) (or equivalent)

Figure 8a: Lexical establishment of reference (visually) before first enactment for each referent(Note: Two of the DH-TC/Oral children, Maya and Andre, gave their narratives in spoken English and established

initial reference for all characters verbally. Maya additionally used some BSL lexical signs as well; Andre did not.)

Results indicate that, overall, the deaf children who are native BSL signers (DD) used moreinstances of enactment (Fig. 5a) and included more cues for referential cohesion through lexicalestablishment of reference prior to enactment (Fig. 8a) and use of discourse markers (Fig. 6a).Use of enactment with lexical signs and/or entity classifiers (Fig. 7a) occurred slightly more inDD and DH-bi children than DH-TC/Oral, suggesting that providing information about locationand/or motion (and/or lexical information), at the same time as enactment, may be somethingthat requires very early experience with a signed language. It is not clear why native signerPenny patterned differently from the other DD children; the second and third stages of thecurrent longitudinal study may help determine this.

DiscussionThese findings suggest an effect of sign language experience, consistent with other studiesidentifying age of acquisition effects in linguistic abilities of native vs. non-native signers (e.g.Newport 1990). Additionally we suggest that late acquisition of Theory of Mind in non-nativesigning deaf children (Woolfe, Want & Siegal 2002; Morgan & Kegl 2006) may be related tolate role shift acquisition, given the pragmatic difficulties of shifting perspective (Courtin 2000).

AcknowledgementsWe thank Kate Rowley for coding much of the data discussed here. Figures 1-3 are from Woll, B., Sutton-Spence,R., & Waters, D. (2004). ECHO data set for British Sign Language (BSL). London: City University.http://www.let.ru.nl/sign-lang/echo.

ReferencesCourtin, C. (2000) The impact of sign language on the cognitive development of deaf children: The case of theories of mind. Journal of Deaf Studies and Deaf Education, 5(3), 266-276.Morgan, G., & Kegl, J. (2006). Nicaraguan Sign Language and Theory of Mind: The issue of critical periods and abilities. Journal of Child Psychology and Psychiatry, 47(8), 811-819.Newport, E. L. (1990). Maturational constraints on language learning. Cognitive Science, 14, 11-28.Reilly, J. (2000). Bringing affective expression into the service of language: Acquiring perspective marking in narratives. In Emmorey & Lane (Eds.), Signs of language revisited (pp. 415-433). Mahwah, NJ: LEA.Slobin, D. I., et al. (2003). A cognitive/functional perspective on the acquisition of "classifiers". In Emmorey (Ed.), Perspectives on classifier constructions in sign languages (271-296 ed.). Mahwah, NJ: LEA. Woolfe, T., Want, S. C., & Siegal, M. (2002). Signposts to development: Theory of mind in deaf children. Child Development, 73, 768-778.

EnglishEnglishEnglishEnglishNot knownPref lang

BSLBSLBSLBSLBSLPref lang

BSLBSLBSLBSLBSLPref lang

DH-TC/OralDH-biDD

6;76;86;117;07;3Age

Kendra (O)Andre (O)Millie (TC)Maya (TC)Ben (TC)Alias

6;36;56;57;07;4Age

SallyIsabelKyleConnorHunterAlias

5;1Oliver6;4Penny7;0Gretchen7;0Tom7;5RachelAgeAlias

Figure 5b: Exampleenactment token

(Gretchen, see Fig. 4a)

Figure 2: Role shift as person beingattacked, with entity classifierdepicting approach of a bear

Figure 3: Role shift as bear lookingdown, with lexical sign LOOK

Figure 6b:Example discourse

marker (Kyle, THEN)

Project funded by:

Figure 7b:Example

enactment (man)+ entity CLs(man & wall)

(Tom,see Fig. 4b)

Figure 7c:Example

enactment (dog)+ sign

NOTHING(Tom, see Fig.

4d)

Figure 8b: Example of establishment ofreference at first mention, MAN (with

deictic point as discourse marker),followed by enactment of man starting

fire (Rachel, see Fig. 4c)

Deaf children from hearing families with minimal exposure to signlanguage (in schools using either oral methods or Total Communication)

DH-TC/Oral

Deaf children from hearing families who have acquired BSL from age 5(in bilingual schools where both BSL and English is used)

DH-bi

Deaf children from deaf families who have acquired BSL natively sincebirth

DD